Citrus Processing Quality Control and Technology PDF
Citrus Processing Quality Control and Technology PDF
Citrus Processing Quality Control and Technology PDF
Dan Kimball
Director of Qua/ity Control, Research and Development
AII rigl1ts reserved. No part ofthis book covered by the copyright hereon may he reproduced or used in allY form Of by any means--graphic,
electronic, Of mechanical, including photocopying, recording, tap ing, or informaticn storage and retrieval systems-~without the written
permission ofthe puhlisher.
2 3 4 5 6 7 8 9 XXX 01 00 99 98 97
Kimball, Dan A.
Citrus processing: quality control and technology / Dan A. Kimball.
p. cm.
Includes bibliographical references and index.
ISBN 978-94-010-5645-8 ISBN 978-94-011-3700-3 (eBook)
DOI 10.1007/978-94-011-3700-3
1. Citrus juices. 2. Citrus fruits--By-products. t Title.
TP562.K55 1991 90-25153
663' .63--dc20 CIP
Contents
Preface v
Chapter 1 Introduction
Unit 1 Citrus Juice Characteristics
Chapter 2 Brix and Soluble Solids 7
Chapter 3 Acids in Citrus Juices 34
Chapter 4 The Brixl Acid Ratio 55
Chapter 5 Testing of Fruit Samples 66
Chapter 6 Citrus Oils, Aromas, and Essences 73
Chapter 7 Citrus Juice Pulp 102
Chapter 8 Juice Cloud 117
Chapter 9 Color of Citrus Juices 126
Chapter 10 Bitterness in Citrus Juices 136
Chapter 11 Nutritional Content of Citrus Juices 162
Chapter 12 Citrus Rheology 175
Chapter 13 Citrus Processing Varieties 180
Unit 2 Citrus Juice Sanitation
Chapter 14 Inspections 213
Chapter 15 Citrus Microbiology 226
Chapter 16 Insects, Rodents, and Birds 244
Chapter 17 Physical and Chemical Contamination 254
Chapter 18 Processing Contamination 258
Chapter 19 Juice Adulteration 279
Unit 3 Citrus Juice By-Products
Chapter 20 Food-Grade Nonjuice Products 303
Chapter 21 Animal Feed and Fuel By-Products 337
Chapter 22 Wastes from Citrus Plants 353
iii
iv CONTENTS
Citrus juices constitute the majority of the fruit juices consumed in the United
States and around the world. Along with the rest of the fruit juice industry, they
playa major role in the entire food industry as well. In spite of this prominence,
few texts have been written on quality control technology; and most of the texts
have been written by researchers who may possess great technical skill but
generally are less familiar with daily routine quality control problems and con-
cerns than quality control technologists are. On the other hand, quality control
technologists and managers generally do not have the time and/or the talent to
write books or communicate through scientific literature.
The author recognized the need for an updated, comprehensive, and easily
understood text on citrus quality control. This text has been designed to be used
not only by processors, bottlers, canners, and others involved in the citrus in-
dustry, but it can be of value to instructors and students of citrus technology.
Researchers also can find value in the foundations laid down by the text, es-
pecially in regard to the needs and concerns of the processing industry. Also,
consultants and marketing personnel will be greatly helped by understanding
the concepts of this volume. Persons in related industries also will find many
applications that can be easily adapted to their needs.
It may be impossible to foresee all problems and situations that can occur in
various plants at various times. It has been said that the only thing that does
not change is change itself. Constant revision of procedures, policies, and tech-
nical information should be an integral part of every quality control program.
All quality control personnel should be receiving constant training', and proce-
dures should be under continued scrutiny and development. Each company
should have a quality control manual to supplement this text. The manual should
outline procedures and policies specific to the individual plant's operation and
facilities. Reference material should be orderly, readily available, and easy to
understand.
A great lack of uniformity plagues a portion of the citrus industry, stemming
v
vi PREFACE
Introduction
The commercialization of fruit juices has a long and colorful history. The first
commercial fruit juice was produced in 1869 by Welch, who began bottling
unfermented grape juice in Vineland, New Jersey. However, juice that could
be preserved for long periods of time did not emerge until after Welch intro-
duced the principles of heat sterilization. By the 1930s, flash pasteurization was
developed, and it was during this time that fruit juices gained public notice as
a significant source of vitamin C and began to increase in popularity. World
War II made new demands on the juice industry, resulting in the development
of dehydrated fruit juices and frozen concentrates. Prior to this time, California
dominated the citrus industry with its fresh markets. With the advent of frozen
concentrated orange juice and its increase in popularity, the citrus industry
moved east to Florida.
Today, Florida grows approximately twice as much citrus as California, with
about 90% of its fruit going to the processed markets. Florida's hot and humid
climate induces thin peels that are easily scarred and develop poor color, and
these cosmetic effects generally render Florida fruit inferior in the fresh mar-
kets. However, the juice flavor from such fruit is rated as superior. In Califor-
nia, the fresh markets still rule supreme, generating about ten times as much
profit for citrus growers as processed fruit. The drier Mediterranean climate of
California produces thick richly colored peels, which are more durable to scar-
ring than Florida fruit, and deep-colored juices. Culls that do not make fresh
fruit quality because of scarring, color, size, or freeze damage are sent to pro-
cessing plants in California. Thus, in spite of the fact that one-third of the U. S.
citrus is grown there, California is responsible for only about 10 to 15 % of the
juice market. Another factor that has affected the difference between California
and Florida markets is the fact that in California a wider variety of agricultural
products is produced over a much larger and more diverse geographical region
than Florida. In Florida, citrus is king, a situation that led to the development
of the Florida Department of Citrus, which has exerted great political influence
in protecting a very important industry, second only to the tourist industry.
One Florida venture was to initiate a citrus industry in Brazil in order to avoid
the economic damage of freezes in Florida. This budding industry was later
sold to the Brazilians, who have, in recent times, emerged as a major contender
in the international citrus juice market. With a few devastating freezes in Flor-
ida, it did not take long for the Brazilian citrus empire to establish markets in
the United States and usurp Florida's previous position as king of the mountain
in the citrus industry. Brazil has the largest juice processing plants in the world
and is a leader in citrus technology in many areas.
Citrus processing consists mainly of the receiving and storage of fruit, fruit
washing and grading, juice extraction, juice finishing or screening, heat treat-
ment, packaging, and storing. Blending of previously prepared or purchased
juices is very common, along with the manufacture of by-products such as pulp
wash juices, oils and aromas, animal feeds, pectins, marmalades and jellies,
drinks, flavors, wines, vinegars, fuels, and many more. Some areas grow citrus
purely for oils or fragrances to be used in perfumes. All these products require
a degree of quality control to ensure their marketability and consumer accep-
tance. However, even though there exists a wide diversity of citrus products,
the majority of citrus products consist of 100% juices.
The chapters in this book treat the main parameters of citrus juices and some
of the main by-products and other quality-control-related topics. New emerging
technologies such as aseptic processing, debittering, reverse osmosis, and per-
haps even supercritical extraction undoubtedly will improve the quality of citrus
products but probably will have little impact on the basic principles of quality
control outlined in this book. As new technologies and research advance, greater
light is expected to be shed on citrus quality control and citrus processing. New
concepts already are slipping over the horizon, such as those regarding the po-
tential anti carcinogenic attributes of some components of citrus juices and the
potential for debittering wastes to be used as antifeedants for some agricultural
pests. Quality control personnel should be serious students of citrus research
efforts and perhaps even participate in them when possible. On the other hand,
researchers should be serious students of the industry and quality control tech-
nology and keep informed on what is happening in the industry because the
industry is the prime benefactor of their research.
Even though approximately 70% of all juices consumed in the United States
are citrus juices, it would be appropriate to mention that a significant portion
of the juice industry is comprised of noncitrus juices. Citrus juices are preferred
in an opaque form most closely resembling the natural juice. Other juices, such
as apple, grape, and berry juices, are preferred in a clarified form, involving
the use of filter presses, centrifuges, enzymatic treatments, and filters in various
stages and engineering schemes. Except for this major difference between clar-
ified and citrus juices, the quality control concepts of both are very similar, and
the principles outlined in this text can be of great benefit to all. Even persons
INTRODUCTION 3
in the sugar industry can utilize much of this information in the crystallization
of granular sugar as well as the manufacture of syrups and related products.
Bottlers and canners of any type of juice will find much of this information
valuable. Production, marketing, and other managers involved in fruit juice
processing often need a reference source regarding related technology, a role
that this text can fill.
The mission of a quality control department in a citrus juice processing fa-
cility can be divided into three main parts. The first is the measurement and
control of juice characteristics. The procedures outlined in this text can serve
as a complete guide to both the latest and traditionally accepted methods com-
monly used in the industry. Quality control goes beyond mere measurement of
these parameters. It includes comparing these characteristics to customer or in-
house specifications in determining compliance with the goals and policies of
the company. Many operations await the decision of the quality control de-
partment before proceeding. This gives quality control the ability literally to
control the quality of the products. Quality assurance is defined by some as
being even more inclusive than this, with the added responsibility of ensuring
that the products achieve a certain quality standard. This goes beyond mere
measurement and reporting of discrepancies.
The second portion of the mission of a quality control department is to ensure
the sanitation of the product. Good industrial hygiene involves a host of con-
siderations, as detailed in Unit Two of the text. Modem lifestyles demand san-
itary foods that are guaranteed to be free from any type of contaminants. Con-
tamination can come from many places-the environment, a dirty plant,
improper processing, inadequate sanitary procedures or enforcement, or even
intentional contamination. Quality control personnel are responsible for ensur-
ing a clean and healthy environment for food processing and safe, authentic,
and palatable products. In fact, quality control personnel often are held crimi-
nally negligent in a court of law if sanitation problems reach litigation.
The third portion of the quality control mission involves management-man-
agement of people, procedures, policies, specifications, inventories, and infor-
mation. Decisions have to be made on the spot, and problems must be quickly
analyzed and corrected. Consequently, information must be readily available,
and communication must be timely and thorough. The day will soon come when
quality control decisions and functions will require the use of computers. Com-
puters and programmable calculators already are widely used, and computer
applications are an intrinsic part of this text. Citrus quality control software in
GWBASIC for IBM-compatible personal computers entitled "CITQUIC" is
available from the author, including not only those programs referred to in the
text, but integration of these programs into a complete inventory management
system and quality control program. One of the advantages of such software is
that the GWBASIC programs can easily be modified to fit any commercial need.
4 INTRODUCTION
Citrus juices contain a wide variety of chemical compounds, but none as prev-
alent as sugars or carbohydrates. Carbohydrates make up better than 80% of
the soluble material in citrus juices, and of these soluble carbohydrates, half are
in the form of sucrose. The sucrose molecule consists of one molecule of glu-
cose and one molecule of fructose, as shown in Fig. 2-1. The other half of the
carbohydrates in citrus juices consist of relatively even amounts of glucose and
fructose, which result from natural enzymatic breakdown of the sucrose. Other
carbohydrates playa minor role in overall juice composition. Because sucrose
consists of one molecule of glucose and one molecule of fructose, the densities
of aqueous solutions of sucrose mixed with equal portions of glucose and fruc-
tose are similar to the densities of 100 % sucrose solutions.
Juice density is one of the most important quality control parameters in the
juice industry. Juice densities are used in making conversions, back and forth,
between volume and weight parameters in performing a host of important cal-
culations in predicting juice blends and formulations and juice concentration,
standardizing laboratory results, and managing inventories and marketing. Be-
cause carbohydrates occur in such high levels in citrus juices, juice density is
detennined by using methods and scales that apply to pure sugar solutions. The
soluble material is not all in the pure form of the major carbohydrate compo-
nents; so the soluble material, including some noncarbohydrates, is referred to
as soluble solids rather than sugars. However, the soluble solids are treated as
sugars in regard to density and other quality control measurements. Also, the
soluble solids differ from the insoluble solids, such as cloud and pulp material,
which contribute little to density measurements. Organic acids and their salts
also occur in significant amounts in citrus juices and contribute to the soluble
solids content. Corrections usually are applied to density measurements in order
to account for these noncarbohydrate soluble solids as if they were carbohy-
drates, to facilitate the use of carbohydrate scales and tables.
Several scales have been developed to relate the measured specific gravity or
density to the concentration of various solutions. The API scale, selected in
H 0 4 C H ,0o H
HO OH AOH H
H OH
GLUCOSE ° 0 CH 2
OH
CH,OH H
FRUCTOSE
Fig. 2-1. A sucrose molecule. composed of a molecule of glucose and a molecule of fructose.
1921 by the American Petroleum Institute, the U.S. Bureau of Mines, and the
National Bureau of Standards, is used for petroleum products. The Balling scale
is used by the tanning and tanning extract industries. The Baume scale was
proposed in 1768 by Antoine Baume, a French chemist, to measure the con-
centrations of acids and syrups that were lighter and heavier than water by using
a different scale for each type of solution. The Quevenne scale is an abbreviated
specific gravity scale used primarily by the milk industry. The Richter, Sikes,
and Tralles scales were developed to measure the alcohol content in water via
density, and the Twaddle scale is an attempt to simplify the density measure-
ments of liquids heavier than water.
One of the best-known scales relating the concentration of sucrose solutions
to solution density was published by Balling (Balling n.d.) and served as a basis
for the development of the more complete and expanded table established by
the German mathematician Adolf Ferdinand Wenceslaus Brix (1798-1870) in
1854. The original Brix density table, used exclusively by the sugar industry
for many years, was based on the density of a sucrose solution at a standard
temperature of 17.5 ° e. This scale based on 17.5 ° e still is used by the fresh
citrus fruit industry for maturity tests. Domke (1912) proposed a table of dens-
ities of sucrose solutions according to concentration based on sucrose solution
densities at 20 o e, which has since become the standard for the processed juice
industry. At least by 1941 the Brix scale was being used by the fruit juice
industry in determining the sucrose equivalent of soluble solids; the term "Brix"
or "degrees Brix" was being used interchangeably with the % sucrose or the
% soluble solids by weight in fruit juices and was determined by using density
measurements. This usage led to the Brix scale's becoming the standard for the
measurement of juice concentration in the citrus and related industries.
BRIX BY HYDROMETER
The most economical commercial method of measuring the Brix of citrus juices
uses a weighted spindle or hydrometer, illustrated in Fig. 2-2. The buoyancy
of the hydrometer is directly proportional to the density of the solution, and the
BRIX AND SOLUBLE SOLIDS 9
Juice Level
•••••••'~;;;;;;;;;;;;;I:l~;;;;;;~.. • • • • • • • •• t>
Hydrometer
. - Cylinder
....- I f - - Hydrometer
Thermometer
Temperature
Ind.f.:'I.--II--- Correction
Scale
degree that the hydrometer sinks into the juice can be calibrated to a Brix scale
on the neck of the hydrometer, which can be read at the level where the juice
surface strikes the hydrometer neck.
Dissolved gases affect this buoyancy, so deaeration of the juice aids in ac-
curate Brix determinations by hydrometer. Air is incorporated into the juice
during processing through agitation, mixing, and pumping.
Temperature also affects the density of the solution; therefore, many Brix
hydrometers have a built-in thermometer accompanied by a temperature correc-
10 CITRUS JUICE CHARACTERISTICS
tion scale that can be used to correct the Brix reading. In the absence of such a
scale, a thermometer and Tables 2-1 and 2-2 can be used for hydrometers based
on 17.5 °C and 20.0°C, respectively, to make the Brix correction for tempera-
ture. Hydrometers usually are used to measure the Brix of single-strength juices
where acid corrections are small. Consequently, acid corrections generally are
not used with hydrometers.
5 .36.47 .56 .65 .73 .80 .86 .91 .97 1.01 1.05 1.08 1.10 1.14
10 .32 .38 .43 .48 .52 .57 .60 .64 .67 .70 .72 .74 .75 .77
11 .31 .35 .40 .44 .48 .51 .55 .58 .60 .63 .65 .66 .68 .70
12 .29 .32 .36 .40 .43 .46 .50 .52 .54 .56 .58 .59 .60 .62
13 .26 .29 .32 .35 .38 .41 .44 .46 .48 .49 .51 .52 .53 .55
14 .24 .26 .29 .31 .34 .36 .38 .40 .41 .42 .44 .45 .46 .47
15 .20 .22 .24 .26 .28 .30 .32 .33 .34 .36 .36 .37 .38 .39
16 .17 .18 .20 .22 .23 .25 .26 .27 .28 .28 .29 .30 .31 .32
17 .13 .14 .15 .16 .18 .19 .20 .20 .21 .21 .22 .23 .23 .24
18 .09 .10 .10 .11 .12 .13 .13 .14 .14 .14 .15 .15 .15 .16
19 .05 .05 .05 .06 .06 .06 .07 .07 .07 .07 .08 .08 .08 .08
17.5 .11 .12 .12 .14 .15 .16 .16 .17 .17 .18 .18 .19 .19 .20
Add to Brix
21 .04 .05 .06 .06 .06 .07 .07 .07 .07 .08 .08 .08 .08 .09
22 .10 .10 .11 .12 .12 .13 .14 .14 .15 .15 .16 .16 .16 .16
23 .16 .16 .17 .17 .19 .20 .21 .21 .22 .23 .24 .24 .24 .24
24 .21 .22 .23 .24 .26 .27 .28 .29 .30 .31 .32 .32 .32 .32
25 .27 .28 .30 .31 .32 .34 .35 .36 .38 .38 .39 .39 .40 .39
26 .33 .34 .36 .37 .40 .40 .42 .44 .46 .47 .47 .48 .48 .48
27 .40 .41 .42 .44 .46 .48 .50 .52 .54 .54 .55 .56 .56 .56
28 .46 .47 .49 .51 .54 .56 .58 .60 .61 .62 .63 .64 .64 .64
29 .54 .55 .56 .59 .61 .63 .66 .68 .70 .70 .71 .72 .72 .72
30 .61 .62 .63 .66 .68 .71 .73 .76 .70 .78 .79 .80 .80 .81
35 .99 1.01 1.02 1.06 1.10 1.13 1.16 1.18 1.20 1.21 1.22 1.22 1.23 1.22
40 1.42 1.45 1.47 1.51 1.54 1.58 1.60 1.62 1.64 1.65 1.65 1.65 1.66 1.65
45 1.91 1.94 1.96 2.00 2.03 2.05 2.07 2.09 2.10 2.10 2.10 2.10 2.10 2.08
50 2.46 2.48 2.50 2.53 2.56 2.57 2.58 2.59 2.59 2.58 2.58 2.57 2.56 2.52
55 3.05 3.07 3.09 3.12 3.12 3.12 3.12 3.11 3.10 3.08 3.07 3.05 3.03 2.97
60 3.69 3.72 3.73 3.73 3.72 3.70 3.67 3.65 3.62 3.60 3.57 3.54 3.50 3.43
(Calculated from data on thennal expansion of sugar solutions by Plato and assumed that the hydrometer is of
lens 16 glass. Table should be used with caution and only for approximate results when the temperature differs
much from the standard temperature or from the temperature of the surrounding air.)
11
12 CITRUS JUICE CHARACTERISTICS
500 ml
Vacuum
Flask
Procedure
1. Extract enough juice to fill the hydrometer cylinder, and place it in the
juice beaker shown in Fig. 2-3.
2. Evacuate the vacuum flask with clamps closed. Open the clamp to the
juice beaker, and draw the juice into the flask.
3. Turn off the vacuum source, and break the vacuum by leaving the juice
beaker clamp open.
4. Open the second clamp, and allow juice to flow into the deaerated juice
beaker.
5. Fill the hydrometer cylinder with the de aerated juice, and insert the hy-
drometer into the cylinder so that the juice overflows the cylinder. This
is best done in a sink or outside.
6. Read the Brix where the level of the juice meets the Brix scale in the neck
of the hydrometer, as shown in Fig. 2-2.
7. Using the thermometer, measure the temperature, and add or subtract the
temperature correction to the Brix according to the scale in the hydrom-
eter, or use Tables 2-1 or 2-2 if no correction scale is available.
BRIX AND SOLUBLE SOLIDS 13
BRIX BY REFRACTOMETER
Even though refractometers are more expensive than hydrometers, they require
only 2 to 3 ml of sample, compared to about 200 ml or more required by the
hydrometer, and deaeration generally is not necessary. Brix measurements can
be performed much faster, and the Brix range of refractometers can be much
broader, up to 0 to 70oBrix, which is important in processing plants that man-
ufacture concentrated juices. These advantages quickly justify the use of re-
fractometers in processing plants in spite of their cost. For an instrument that
is used as often as a refractometer, with quality control, inventories, packaging,
and marketing depending on the results, the cost of the instrument is trivial; the
refractometer literally becomes the foundation of any citrus juice processing
procedure. In fact, it is wise to have at least two refractometers so that if one
needs to be repaired, the operations of the plant will not be hampered. Having
two refractometers also provides a means of checking and comparing results to
ensure accuracy.
The refractive principle that forms the basis of this type of measurement lies
in the variability of the speed of light through two mediums of different density.
To illustrate this principle, a simple analogy can be used. Suppose that a person
is drowning offshore, as shown in Fig. 2-4, and a lifeguard is trying to deter-
mine which is the fastest way to reach her. Even though the shortest distance
between two points is a straight line, it is not necessarily the fastest; in this
case, the lifeguard probably can run faster than he can swim. Thus he would
want to spend as much time as possible running and the least amount of time
swimming. The speed that he can run and the speed that he can swim are de-
termined by the density of the mediums through which he must pass, and de-
termine the ideal angle that he should take to reach the drowning victim in the
shortest possible time. It is a phenomenon of nature that light "knows" the
ideal angle that will allow it to arrive at a certain point, going from one medium
to another of different density, in the shortest possible time. As the densities of
one or both of the mediums change, so does this angle. Thus a change in the
angle of refraction permits the measurement of the density of citrus juices, as
light passes through the juice at one density and then through a glass prism at
another density, which can be translated into the concentration as degrees Brix.
The wavelength of light used with a refractometer has been shown to affect
the angle of refraction, and, for this reason, a sodium vapor lamp has been
suggested for use as a light source. However, any soft yellow lamp is sufficient
for industrial purposes. The light first enters a fogged or ground glass lens that
evenly scatters the light in all directions, as shown in Fig. 2-5. The scattered
light then enters the sample and is refracted at the surface of the prism on which
the sample lies. The critical ray is a ray that travels parallel to the surface of
the prism and represents the minimum angle at which the scattered light coming
14 CITRUS JUICE CHARACTERISTICS
-"'--
---
~
I
Shortest I
I
----i-
I
~I
i
Fastest
----------- : .....
': . '"
~ ".
. .'
" ,'.
,':" . -', .....
,',: ,'.
Fig. 2-4. Illustration suggesting why light refracts when going from one medium into another
medium of greater density, Light has a "nose" for the fastest route and bends or refracts accord-
ingly ,
through the sample can strike the prism, Of course, a ray perfectly parallel to
the surface of the prism never will strike the prism, If it changed its angle
slightly, however, it would be refracted, defining a boundary for a dark zone
or area, as shown in the figure. As perceived from the side, as illustrated, no
scattered light coming from the surface of the prism can enter the dark zone,
This dark area boundary can easily be calibrated to a Brix scale because both
the Brix and the angle of refraction are determined by the sample density,
Because the standard error for Brix readings is ±O . l°Brix, the law of sig-
nificant figures dictates that Brix values should be expressed only to the nearest
tenth of a Brix (see Chapter 23). Brix values expressed to more decimal places
than this imply deceptive accuracy. Acid corrections and temperature correc-
tions often are expressed to two decimal places . When these corrections are
applied, the final Brix value should be rounded to the nearest tenth of a Brix.
Several factors contribute to the errors in Brix readings. One of the greatest
sources of errors is the fact that citrus juices, especially concentrates, produce
an indistinct shadow or boundary between the dark and light zones of the re-
fractometer due to insoluble matter in the juice. This fuzzy shadow engenders
a wide range of opinions as to where the Brix value should be read. Whether
BRIX AND SOLUBLE SOLIDS 15
Light Source
Fogged Prism
t
Prism Brix
Scale
Fig. 2-5. Illustration of the formation of the shadow in a refractometer. No light can enter the
dark zone because of refraction as the light passes from the sample into the prism at a different
density.
the shadow is adjusted to the intersection of two cross hairs or used directly on
a Brix scale itself, the Brix should be read in the middle of this fuzzy region,
where it is estimated that a perfect shadow line would appear if the fuzzy line
were clear and focused. Calibration of shadow readings can be achieved by
having lab technicians measure the Brix of several samples and comparing the
results of one lab technician to another. Those that are reading too high or too
low will easily be detected and thus can be corrected. This type of comparison
should be done as often as needed to ensure consistency and confidence in re-
sults. In fact, such multitechnician calibrations should be performed for all lab-
oratory procedures, regularly or as often as needed.
Another source of Brix reading error is the refractometer itself. Most refrac-
tometers have an adjustment screw that permits calibration of the shadow. This
can be done with standard sucrose solutions, but the use of distilled water to
calibrate the refractometer according to the refractive index is sufficient in most
cases. The refractive index is the ratio of the sine of the angle between the
incident ray and the line perpendicular to the surface of the prism to the sine of
the angle between the refracted ray and the same perpendicular line. The pro-
cedure for the calibration of a refractometer using distilled water is as follows.
16 CITRUS JUICE CHARACTERISTICS
Calibration of a Refractometer
Procedure
1. Measure the temperature of the water if the refractometer does not have
a means of doing so.
2. Place a few drops of the distilled water onto the prism of the refractometer
using a dropper or plastic stirring rod, being careful not to scratch the
prism. If the temperature is to be measured by the refractometer, let the
sample sit for a few minutes for the temperature to equilibrate. Otherwise,
read the Brix right away.
3. Using Table 2-3, look up the refractive index for water at the temperature
determined above.
4. Switch the refractometer to measure the refractive index instead of the
Brix, and adjust the refractometer to read the refractive index determined
from Table 2-3 regardless of where the shadow occurs.
5. Using the wrench or screwdriver, adjust the calibration screw until the
shadow is aligned with the cross hairs or agrees with the refractometer
reading from Table 2-3.
6. Switch the refractometer back to the Brix mode.
(The instructions may vary slightly from one instrument to another. The
manufacturer's operational manual should be consulted for details.)
Temperature Corrections
Temperature fluctuations also affect the Brix reading for refractometers, and
temperature corrections are required, as with the hydrometer. However, be-
cause the Brix is measured by using a different principle, the correction tables
are different, as can be seen by comparing Tables 2-1 and 2-2 with Table 2-4.
Many modem refractometers make this correction automatically. By applying
a double-nested least squares analysis to the data in Table 2-4, the following
equation can be constructed, which can be used in computer applications instead
of Table 2-4.
The correction calculated above from the Brix (B) and the centigrade temper-
ature (T) can be added to, if positive, or subtracted from, if negative, the Brix
reading to give a temperature-corrected Brix.
Acid Corrections
Table 2-4. Temperature corrections to the Brix reading using a refractometer based on a standard of co
temperature of 20°C (J./ntern. Sugar 1937). n
:::j
21
Subtract from the Brix c:
(II
Brix a 5 10 15 20 25 30 35 40 45 50 55 60 65 70 c-
c:
·c 0m
10 .50 .54 .58 .61 .64 .66 .68 .70 .72 .73 .74 .75 .76 .78 .79
11 .46 .49 .53 .55 .58 .60 .62 .64 .65 .66 .67 .68 .69 .70 .71
n
J:
12 .42 .45 .48 .50 .52 .54 .56 .57 .58 .59 .60 .61 .61 .63 .63 lo
21
13 .37 .40 .42 .44 .46 .48 .49 .50 .51 .52 .53 .54 .54 .55 .55 lo
14
n-I
.33 .35 .37 .39 .40 .41 .42 .43 .44 .45 .45 .46 .46 .47 .48 m
15 .27 .29 .31 .33 .34 .34 .35 .36 .37 .37 .38 .39 .39 .40 .40 21
r;;
-I
16 .22 .24 .25 .26 .27 .28 .28 .29 .30 .30 .30 .31 .31 .32 .32 0
(II
17 .17 .18 .19 .20 .21 .21 .21 .22 .22 .23 .23 .23 .23 .24 .24
18 .12 .13 .13 .14 .14 .14 .14 .15 .15 .15 .15 .16 .16 .16 .16
19 .06 .06 .06 .. 07 .07 .07 .07 .08 .08 .08 .08 .08 .08 .08 .08
26 .40 .42 .43 .44 .45 .46 .47 .48 .48 .48 .48 .48 .48 .48 .48
27 .48 .50 .52 .53 .54 .55 .55 .56 .56 .56 .56 .56 .56 .56 .56
28 .56 .57 .60 .61 .62 .63 .63 .64 .64 .64 .64 .64 .64 .64 .64
29 .64 .66 .68 .69 .71 .72 .72 .73 .73 .73 .73 .73 .73 .73 .73
30 .72. .74 .77 .78 .80 .80 .81 .81 .81 .81 .81 .81 .81 .81 .81
BRIX AND SOLUBLE SOLIDS 19
where A is the % titratable acidity. The federal code utilizes the equation:
which is slightly less accurate in the acid level ranges of commercial lemon and
lime concentrations (CFR, Title 21, 146.132(a)(2) 1983).
The procedure for measuring the Brix level by refractometer is as follows.
• Refractometer.
• Thermometer if refractometer does not provide for the measurement of the
temperature.
• Table 2-4 if the refractometer does not automatically correct for the tem-
perature.
• Appendix A and a % titratable acid value in order to make acid corrections
to the Brix.
• Dropper or plastic stirring rod.
• Water and cleaning tissue for prism.
Procedure
1. Using the water and cleaning tissue, clean and dry the prism and the sur-
face of the fogged glass used to scatter the light.
2. Prepare the sample by stirring with the plastic stirring rod and/or swirling,
and, using the plastic stirring rod or sample applicator, place a few drops
of the sample onto the prism. Care should be taken not to use metal ap-
plicators or other devices that may scratch the surface of the prism. Bub-
bles or foam also will distort the Brix reading.
3. Close the refractometer by lowering the fogged glass onto the sample, and
make sure that it is securely in place.
4. Position the light source to shine through the fogged glass, adjusting the
shadow to the cross hairs, and read the Brix, or read the Brix directly if
the shadow falls directly on the Brix scale itself. Cold samples may need
to sit a few minutes so that the temperatures can equilibrate.
5. Using Table 2-4 or Equation 2-1 and Appendix A or Equation 2-2, make
the necessary acid and temperature corrections, and round off the cor-
rected Brix to the nearest tenth of Brix. The standard Brix error is
±O.I°Brix. The Brix can be compared to the USDA grade standards de-
picted in Table 2-5.
20 CITRUS JUICE CHARACTERISTICS
unsw sw unsw sw
Frozen concentrated orange 41.8 42.0 41.8 42.0
juice
Reconstituted frozen concen- 11.8* 11.8*
trated orange juice
Canned orange juice 10.5 10.5 10.0 10.5
Canned concentrated orange 41.8 42.0 41.8 42.0
juice
Reconstituted canned concen- 11.8* 11.8* 11.8* 11.8*
trated orange juice
Grapefruit Juice (48FR 9/12/83, 52.1221-52.1230)
Grade A Grade B
Type of Juice unsw sw unsw sw
Grapefruit juice 9.0 11.5 9.0 11.5
Grapefruit juice from concentrate 10.0 11.5 10.0 11.5
Frozen concentrated grapefruit 38.0 38.0 38.0 38.0
juice
Reconstituted frozen concentrated 10.6 10.6* 10.6 10.6*
grapefruit juice
Reconstituted dehydrated grape- 10.0 11.5 10.0 11.5
fruit juice
Reconstituted concentrated grape-
fruit juice for manufacturing 10.5 10.5
Ws = BW/IOO (2-4 )
The density in grams of solution per milliliter (d) can be used to calculate D:
d = 0.524484e(B+330.872)2/170435 (2-7)
22 CITRUS JUICE CHARACTERISTICS
which can be used in Equations 2-5 and 2-6 to give the following:
SPG = 0.0437691Be(B+330.872)2/170435 (2-8)
D = 4.37691e(B+330.872)2/ 170435 (2-9)
For example, if we had 1000 gallons of 65.0 o Brix concentrate and wanted to
dilute it to 60.00Brix, the water needed would be:
Vw = 109 gal
A particularly irritating problem occurs when a tank is full, and the Brix is
too high or too low. As there is no room for volume adjustments, some of the
juice needs to be removed and replaced with concentrate of a higher Brix or
water. The volumes involved can be calculated as explained in the following
paragraphs.
If the Brix is too low in a tank filled to capacity, and we need to add a higher
Brix concentrate to raise the Brix, we can use the following:
(2-13 )
BRIX AND SOLUBLE SOLIDS 23
where VH is the volume of the juice that we want to replace in the tank, V is
the volume desired or the tank capacity, SPG is calculated from the desired Brix
using Equation 2-8, SPGL is calculated from the original low Brix level in the
tank, and SPGH is calculated from the Brix of the concentrate that will be used
to bring the Brix level up. For example, if we have a WOO-gallon tank full of
60.0 oBrix concentrate that we want to raise to 62.0oBrix using 65.0 oBrix con-
centrate, using Equation 2-13 we would get:
where 6.712 is the SPG desired (62.0oBrix) calculated from B, 6.436 is the
SPGL calculated from the existing low Brix (60.0 oBrix), and 7.135 is the SPGH
calculated from the 65.0 oBrix concentrate used for the adjustment. In other
words, in this example, we would replace 395 gallons of 60 0 Brix concentrate
with 65°Brix concentrate to get a full tank of exactly 62°Brix concentrate.
If the Brix is too high, and we want to add water to reduce the concentration,
we can use the following:
(2-14 )
with the SPG values calculated from the Brix values by using Equation 2-8 as
before. In other words, we would need to replace 59 gallons of the 65°Brix
with water to get a full tank of exactly 62°Brix concentrate.
We can easily see that involved computations are often required to account
for one of the many parameters of citrus juice. Not only are these and the
computations found hereafter complex, but they are often routine and must be
performed over and over again. Such situations are ideally suited for computer
applications. Computing devices range from hand-held programmable calcula-
tors to sophisticated multiterminal mainframe computers. Programming lan-
guages also differ considerably. However, the basic logic of how these equa-
24 CITRUS JUICE CHARACTERISTICS
tions can be used varies little and can be represented in the form of flow charts.
These flow charts can be used as is or modified to fit any industrial need in the
writing of programs for any computing device.
The flow charts in Fig. 2-6 through Fig. 2-9, and throughout the remainder
of this text, illustrate some computer logic that can be used in routine citrus
EQ 2-14 EQ 2-13
Vw EQ 2-18
VH
(n=2)
V,orVw
Fig_ 2-6. Flow chart for a computer program that will determine the Brix adjustment in citrus
juices.
BRIX AND SOLUBLE SOLIDS 25
EQ 2-15
Vj
Vc = Vd - Vj
Fig. 2-7. Flow chart for programming a computer to cal-
culate the volume of juice and high Brix concentrate to
make a juice product.
quality control. There are four different box shapes, which imply specific func-
tions. A rectangle with a curved or tear-sheet bottom represents input of data
or the output of results. Within these boxes you will find the data that needs to
be inputted or the variable that represents the final targeted output or answer.
Diamond-shaped boxes represent decisions that have to be made by the com-
puter or the operator. Circles represent the use of subroutines that have been
found useful in more than one program or more than once in the same program.
Plain rectangles represent computations that are unique to the program and do
not justify the formation of a subroutine.
Figure 2-6 shows a flow chart that can be used to program a computer to
calculate the amount of Brix adjustment needed in a tank with a Brix that is too
high or too low, as discussed previously. The input data includes the Brix of
the tank, the desired Brix, and the volume of the juice in the tank. The computer
26 CITRUS JUICE CHARACTERISTICS
Bf=10xSPG
B = Bf
n
Z = L: Vi (SPGi)
i=1
n
V=L: Vi
i=1
no
no
B = Bf = average Brix
EQ 2-16
SPG = ZN
Fig. 2-8. Flow chart that can be used to program computers to calculate the average Brix of a
blend.
then must detennine if the Brix on hand is too high or too low compared to the
desired Brix. It then must know if the volume of the juice can vary. If the tank
is full, the volume cannot vary. If there is room in the tank to adjust the Brix
level, then the volume can vary. Subroutines using Equation 2-8 can then be
used to detennine the SPG values needed, and Equation 2-13, 2-14, or 2-18
BRIX AND SOLUBLE SOLIDS 27
n
Z = l:
Vi (SPGi)
n=2
n
V = l: Vi
n=2
no
EQ 2 -18
V 1 = V (SPG) - Z
SPG1 -SPG Fig. 2-9. Flow chart that can be used to pro-
gram computers to calculate the volume of a
single blend component needed in a blend to
achieve a desired Brix.
can be used to detennine the final volume of water or high Brix juice needed to
achieve the desired Brix.
An example of the use of this flow chart using GWBASIC can be found in
Appendix B. A similar program for the HP-41C programmable hand-held cal-
culator can be found in Appendix C.
28 CITRUS JUICE CHARACTERISTICS
Because important volatile flavor components of the juice are lost during evap-
oration, a common, though obsolete, practice is to restore the lost flavor com-
ponents by diluting high Brix concentrates with freshly extracted juice (cutback
juice). This can be done by using the SPGc calculated from the Brix of the
concentrate in Equation 2-8, the SPGj calculated from the Brix of the single-
strength juice, the SPG calculated from the desired final Brix, and the corre-
sponding volumes (Vo lj, and V):
(2-15 )
For example, if we wanted to make 1000 gallons of 41. 8 °Brix concentrate from
12.0oBrix freshly extracted juice and 60.0 oBrix concentrate, using Equation
2-15 we would get:
To find the gallons of concentrate that would be needed, you can subtract 427
gallons of the single-strength juice from the needed 1000 gallons to get 573
gallons of concentrate ( Vc ). This method of using cutback juice has largely been
replaced by the addition of aromas recovered from commercial evaporator con-
densates and single-strength or folded citrus oils recovered from citrus peel.
In Fig. 2-7, the volume of cutback juice and concentrate needed to make a
concentrate blend can be determined. The four input values must be entered,
the subroutine using Equation 2-8 must be used to calculate the SPG values,
and Equation 2-15 can be used to calculate the final cutback juice and concen-
trate volumes. An example of a GWBASIC program using this flow chart can
be found in Appendix B.
Brix Blending
The previous example of using cutback juice to blend with high Brix concen-
trates illustrates general juice blending using only two blend components. The
sophisticated blending common in the industry today is the blending of multiple
juice components in order to achieve a desired final product. One of the main
specifications desired in the final product is the Brix value. To blend more than
one juice component to achieve a desire Brix level or range, the following equa-
tion can be used:
where SPG and V are the final calculated values after blending n components.
Once the final SPG has been calculated, the corresponding final Brix can be
calculated by making successive approximations using the following equation:
Again, the exponential expression can be replaced with the Taylor series ex-
pansion in Equation 2-10 if it is necessary to facilitate computer programming.
In using Equation 2-17, a first guess for the B value can be made by multiplying
the SPG value by 10. The BI value then can be calculated and compared to the
original B value used. If they differ by more than 0.0001 °Brix, then the BI value
can become the new B value to calculate a new BI value. The differences be-
tween the two Brix values should quickly converge to identical values, which
then can be used as the final calculated exact Brix of the blend. Again, tables
can be and have been used to convert back and forth between SPG and Brix
values, but the use of these equations makes it possible to completely automate
blend calculations.
The following example illustrates the use of these equations. Suppose that
we want to calculate the final average Brix of a blend consisting of five lots (see
tabulation below). The volumes of the given lots can be expressed in gallons
or in 52-gallon drums, and the calculation is valid as long as the same units of
volume are used throughout the calculation.
The SPG values can be calculated from the given Brix values by using Equa-
tion 2-8 (or 2-10), or found from a table. They are given here for convenience,
to illustrate the calculation procedure for the blend. Using Equation 2-16 we
get:
SPG = _3(-,-6_.2_7_3~)_+_2-,(_6._62_9-,-}_+_1-,-(7_.1_7_8~)_+_1_1....:....{6_.4_9_1-,--}_+_8---,(,-6_.4_50-"..}
25
= 6.490
This SPG value can be used to determine the corresponding Brix by means of
tables or Equation 2-17. A first guess for B in Equation 2-17 could be lOx
6.490 or 64.9°Brix, which converges to 60.4°Brix quickly via successive ap-
proximations, which becomes the exact calculated Brix of the final blend.
30 CITRUS JUICE CHARACTERISTICS
In Fig. 2-8 the average Brix can be calculated for a blend. The Brix and the
volume of the first component are entered, the subroutine using Equation 2-8
is used to obtain the SPG value, and summations are begun to obtain the inter-
mediary values of Z and V. The computer must determine if this is the last
record. If it is not, then the computer will prompt the operator is to input the
data of the next component. This continues generating the summation values of
Z and V until the data of the last blend component is entered. Equation 2-16
then can be used to calculate the average SPG value. The first guess in the
successive approximations in Equation 2-17 is made by multiplying the SPG
value by 10, and Equation 2-17 is used until the two Brix values are within a
preset range. When the two Brix values are within range, they represent the
output as the average Brix of the blend, the desired result. In Appendix B is a
GWBASIC program example of the use of this flow chart. After entering the
Brix and volume of the last component, you enter' '0" when asked for the next
Brix. The program will list the input data and calculate and display the average
Brix and total volume. A similar program that can be used with an HP-41C
programmable hand-held calculator can be found in Appendix C.
Many procedures require that the blend target a specific Brix or Brix range.
In order to achieve the desired targeted Brix, the blend components need to be
altered or rearranged. The final Brix can be fine-tuned to a specific target, usu-
ally by adjusting the volume or the Brix of a single component. Algebraic rear-
rangement of Equation 2-16 can be used to automate this procedure.
(2-18 )
The SPG value can be determined from the desired or target Brix, and the exact
amount of component 1 needed to achieve this Brix can be calculated. If a
negative number is obtained, then you will know that you must vary another
blend component in order to achieve the desired Brix. To illustrate the use of
Equation 2-18, suppose that we wanted to know how many drums of blend
component 1 in the previous example we would need to achieve exactly
60.0 o Brix. Using the data given and Equation 2-18, we obtain:
The 3 drums of lot #1 given in the example would give a final Brix of 60.4,
but it would take 11 drums to bring the Brix down to exactly 60.0. Equation
BRIX AND SOLUBLE SOLIDS 31
2-16 can be rearranged in several different ways to enable the varying of the
Brix of a component rather than the volume, or the varying of serveral com-
ponents. The more components you vary, the more complex the calculation
becomes.
Figure 2-9 shows the flow chart that can be used to determine the volume
needed of one blend component in order to achieve a targeted or desired Brix
in a blend. The input data is, again, the Brix and volumes of the known blend
components (all except blend component 1 in this case), as well as the final
desired target Brix. The Brix value of the component whose volume will vary
(component 1) needs to be entered as well. The Equation 2-8 subroutine is used
to calculate the needed SPG values, and the intermediary summations are per-
formed. If this blend component is not the last blend component, then the data
from the next blend component is entered until all the blend components are
entered and the proper summations performed. Equation 2-18 can then be used
as shown in the flow chart to determine how much of blend component 1 is
needed to achieve the desired Brix. An example of the use of this flow chart is
illustrated in the GWBASIC program found in Appendix B.
The SPG values used above are based on the density of the juice at 20.0°C.
These values are generally accurate enough for industrial purposes. Chen's ta-
ble also lists densities as a function of temperature (O-50°C) at various Brix
levels. Because most juices and concentrates are processed at much colder tem-
peratures (-20 to 5°C), accounting for the effects of temperature on juice
density may be in order, provided that sufficient computer resources are avail-
able. Making manual calculations to account for these temperature effects gen-
erally is not worth the trouble involved. By performing least squares regression
analyses on Chen's data, the following equation was found to be accurate within
the range of his data (0-50°C).
QUESTIONS
PROBLEMS
8. What would be the final Brix after blending the following blend components? Could
this product be used to make Grade A Frozen Concentrated Orange Juice according
to USDA standards?
9. How much 11.9°Brix cutback juice would be needed to blend with 267 gallons of
60.0 o Brix juice and 472 gallons of 64.8°Brix juice in order to make FeOJ at
41.8°Brix?
10. Suppose you want to make a concentrate with a Brix value of 41.9°Brix using 8
drums of 59.7°Brix concentrate A, 6 drums of 64.6°Brix concentrate B, and cut-
back juice at 13.3°Brix. How many gallons of cutback juice do you need? Suppose
that concentrate A is at 19 ° F, concentrate B is at 3 ° F, and the cutback juice is at
34°F. How many gallons of cutback juice do you now need in order to make the
same FCOJ at 15°F? (52 gal/drum, °C = (OF - 32)5/9)
REFERENCES
Acids playa vital role in the quality of citrus juices, second only to the Brix in
importance. They give the characteristic tartness or sourness of citrus products
and have been acclaimed for their effectiveness as thirst quenchers. These acids
and their salts replace many of the acids and salts lost by the body through
vigorous exercise.
Acids in citrus fruits are formed from the energy-releasing citric acid cycle
common to all life forms. This respiratory process, also known as the Krebs
cycle or tricarboxylic acid cycle, breaks down stored carbohydrates to carbon
dioxide and various organic acids, as shown in Fig. 3-1. This energy-generating
process occurs in the mitochondria of the juice cell, as shown in Fig. 3-2. The
juice cell is elongated and, when fitted together with other juice cells, forms
juice vesicles that are visible in the fruit and can be manually separated from
each other. Juice vesicles are organized into citrus sections in the fruit. As the
juice cell grows, carbohydrate-laden fluid from the sap of the tree flows into
the fruit and into the juice cell. Cell vacuoles, which have the function of stor-
ing food for the cell, absorb the aqueous carbohydrate fluid. As the fruit ma-
tures, the vacuoles grow so that they dominate the cell volume. The fluid in the
vacuoles becomes the juice of the citrus fruit. As this accumulation is taking
place, the mitochondria, the organelles next door to the vacuoles, are generating
acids in the citric acid cycle. Carbohydrates are broken down into pyruvic acid
in the membrane of the mitochondria. Once inside the mitochondria, the py-
ruvic acid enters into the citric acid cycle, generating the various acids.
As can be seen in Fig. 3-1, citric acid is the first acid formed in the cycle.
Even though much of the way that citric acid gets into the vacuole is not under-
stood, one of the most logical explanations appears to be that as soon as the
citric acid is manufactured inside the mitochondria, it migrates through the
membranes of the organelles into the juice vacuole before it can proceed with
the rest of the citric acid cycle. There is evidence that mitochondria are very
active throughout the lifetime of the fruit, indicating that not all of the citric
acid escapes the mitochondria, but some proceeds in the cycle to provide energy
34
Pyruvate
CoA--~~
~_~
L • CO 2
NAOH. Acetyl-CoA (2C)
~COA
NAOH 70xalacetiC~
,
Acid (4C) Ci~ Acid (6C)
FAO:
~;n;c Add (4C) NAOH
otA
Fig. 3-1. Krebs or citric acid cycle that occurs in the mitochondria of the juice cell. Acids pro-
duced give citrus a characteristic tart flavor.
for fruit growth and maturation. Some observers have suggested that the aco-
nitase enzyme that converts citric acid to aconitic acid may be dormant in the
early stages of fruit maturity. It is postulated that this inactivity results in the
early accumulation of citric acid in the fruit. As the fruit matures, and the aco-
nitase enzyme becomes more active, the citric acid continues in the cycle, pro-
viding much-needed energy for fruit growth resulting in a slowing or end to
citric acid accumulation in the juice cell vacuole. However, carbohydrates and
water continue to accumulate in the vacuole, causing a dilution of acid concen-
tration that gives a characteristic decrease of acid or tartness with fruit maturity.
The biological function of citric acid accumulation in citrus fruit is not only
interesting to contemplate, but it may have a direct bearing on the prediction
and understanding of citrus juice quality. There are two processes that govern
citrus growth or the growth of any plant: photosynthesis, which generates car-
bohydrates from sunlight, and respiration in the mitochondria, as already men-
tioned. The first process forms the fuel, and the second bums it. The rates of
these two processes, however, are not the same. At cooler temperatures, pho-
36 CITRUS JUICE CHARACTERISTICS
_ _ _ veSiCles in Citrus
fruit sections
Juice Cell
Juice Cells
Mitochondria
+-i-t--- Vacuole containing juice
Other organelle
+1--- Cell membrane
Fig. 3-2. Juice cell anatomy and location within the fruit. Juice accumulates in the vacuole and
occupies the majority of the volume of the mature cell.
Rate of Photosynthesis
...o ~
iu
:::s
'0
~
D..
>-
~
II)
c Rate of Respiration
W
Temperature - .
Fig. 3-3. Temperatures above the compensation point, where respiration exceeds photosynthesis,
cause a high demand for citric acid, which is obtained from juice cell vacuoles. This causes a
sudden drop in juice acidity.
1.4
NAVEL
1977·78 •
1978.79 0
1.2 1979.80 x
1980.81 6
1981.82 a
1982·83 •
"0 1.0 1983·84 •
()
<i
~
0
0.8
0.6
10 12 14 16 18 20 22 24
Accumulated Temperatures from July 1st x10·3
Fig. 3-4. Changes in % acid with accumulated average temperatures from July I just after bloom
for navel orange juice. Smooth curves suggest a dilution effect in acid levels as the fruit matures
(Kimball (984). (Reprint from the Proceedings o/the Florida State Horticultural Society.)
38 CITRUS JUICE CHARACTERISTICS
1.2
VALENCIA
1.1
1.0
"0
Q
«
~
0
0.9
0
•1977-78
1978-79 o 1981-82
0.8 x 1979-80 • 1982-83
6 1980-81 • 1983-84
ACID MEASUREMENTS
There are two basic ways that the acidity can be measured. Methods that
determine the pH measure only the free hydrogen ions in the solution. Acid
titrations with sodium hydroxide (NaOH) standards measure the total amount
of acid hydrogen, whether free or undissociated. The flavor of citrus juices
would be more closely associated with pH measurements because it is the free
hydrogen ions that interact with the taste receptors on the tongue. However,
maturity tests and Brix corrections preferably are determined by means of acid
titrations, which better reflect the actual citric acid present-the reason being
that pH values change very little in the acid range citrus juices and are difficult
to quantify or relate to taste and maturity differences. Acid titration, the method
used by the industry, provides a more stable and detailed scale for such mea-
surements, although the results of pH and acid titrations are of course related.
Most organic acids are weak acids that change pH when titrated with a strong
base, as illustrated in Fig. 3-6. The dashed curve represents the titration of a
pure aqueous citrus acid solution identical in amount and concentration to its
juice counterpart, which is illustrated with a solid line. The buffering action of
the weak acid can be seen by the small change of pH during the early stages of
the titrations. As the number of equivalents of base added approaches the num-
ber of equivalents of acid in the sample, or when neutralization occurs, a sharp
inflection point emerges that signals the endpoint of the titration. Because citric
12
5.9449 of CA soln made
from 3.8089 CA-H 2 0 ",. _ - - ------
in 1 00 mls water "
(3.46% aCid)-+,'
10 - I
I
.......................
8.2 I
5.9449 of 59.6 Brix,
II' 17.1 B/A ratio
8
Valencia Conc.
I
a. (3.48% acid)
6
differential analysis
4 +-- of endpoint
o
o 5 10 15 20 25 30 35 40
acid is a triprotic acid, three separate inflection points would be expected, one
for each hydrogen as it comes off the molecule. However, the equilibrium con-
stants for these three dissociations are all within an order of magnitude; so a
single inflection point is observed, as seen in the figure. The equilibrium con-
stants for both of the dissociations of the diprotic malic acid are also close to
those of citric acid. Thus, malic acid becomes indistinguishable from citric acid
in these acid titrations.
There has been some controversy about what pH should be taken as an end-
point for an acid titration. The Association of Official Analytical Chemists sug-
gests a pH endpoint of 8.1 (AOAC 1980). The inflection point is somewhat
broad and slightly skewed, so that slight variations in what is taken as the end-
point can cause a significant error in the acid detennination. However, as shown
in Fig. 3-6, the pure citric acid solution gave a standard unifonn inflection point
that intersected the juice titration curve at a pH of 8.2. This value is used by
the USDA in detennining grade standards for citrus juices and concentrates,
and its use was supported by Sinclair and Bartholomew (1945). Accordingly,
this pH endpoint should be and generally is used in citrus juice acid detenni-
nations. Phenolphthalein indicator endpoints generally occur in this pH range
and can be used if a pH meter is not available.
Even though the concentration of the sodium hydroxide used in the titration
does not affect the final results, it does affect the way that the results are cal-
culated. The Association of Official Analytical Chemists (AOAC) recommends
a concentration of O.lN (N = nonnality = number of moles of H+ or OH-
ions per liter of solution) (AOAC 1980). The fresh fruit industry uses 0.1562N
NaOH because all one must do is divide the milliliters titrated by the weight of
the sample in order to obtain the % acid as citric acid. Another popular con-
centration is twice 0.1562N or 0.3125N NaOH, which is convenient in titrating
high acid juices such as lemon or lime and is used in Florida for orange juice
as well. The method used in Florida involves multiplying the nonnality (0.3125)
by a factor of 0.064, followed by mUltiplying by 100% and dividing by the
volume of the sample. The result is a w Iv percentage of acid as citric acid.
The full calculation is:
Such a w I v % cannot be used to calculate the Brix I acid ratio, however. This
value must be divided by the juice density to get the w Iw % citric acid needed
to calculate the Brix I acid ratio. The density of single-strength juice remains
ACIDS IN CITRUS JUICES 41
essentially unchanged between 11.00Brix and 13.0 o Brix at 1.04 to 1.05 gj ml,
and the w jv % can be divided by 1.04 to get the w Iw % citric acid. When
one is using the California method to determine the acid in a large number of
single-strength juice samples using a 10 ml pipette, the standard weight of 10.4
or 10.5 grams of sample can be used in calculating the acid level. If the NaOH
solution is at a concentration other than the convenient 0.1562N, the following
equations can be used to calculate the % acid:
Procedure
1. Clean and dry the 50-gallon container. Empty the 50 one-gallon con-
tainers of distilled water into the 50-gallon container, discarding the used
caps.
2. Add about 1180 g of the NaOH pellets to the water and agitate until the
pellets are completely dissolved. This amount of NaOH should actually
produce a concentration of about 0.1590N. However, the excess NaOH
is needed to compensate for atmospheric carbon dioxide neutralization
during agitation.
3. Weigh between 0.4 and 0.5 g of benzoic acid into a beaker and record
the exact weight, or pipette 25 ml of the standard 0.1562N HCI solution
into the beaker. If an acid other than benzoic acid is used, the 121.13
g/mole molecular weight for benzoic acid should be replaced by the
molecular weight of the acid used, divided by the number of acid hydro-
gens per molecule of acid.
4. If benzoic acid is used, add about 50 ms of ethanol to solubilize the acid.
5. Rinse the buret with the NaOH solution from the tank, discard the rinse,
fill the buret with the NaOH solution, and zero the buret or adjust the
level to 0 ml.
6. Add 5 drops of the phenolphthalein indicator to the acid in the beaker
and/or rinse the electrode from the pH meter, carefully dry the electrode,
and insert it carefully into the acid solution so that the solution level is
above the reference wire in the electrode. These electrodes should be
kept in solution at all times when not in use, and the pH meter should
be calibrated from time to time according to the manufacturer's instruc-
tions.
7. Titrate the acid solution with the NaOH solution while stirring or swirl-
ing it until a persistent pink color of the indicator is observed, or to a
pH of 8.2 if using a pH meter.
8. Using Fig. 3-7, determine the milliliters of titrant from the buret, and
calculate the normality as follows: The Agricultural Commissioner's Of-
fice in California uses standard 0.1562N HCI to standardize the basic
solution and allows a deviation of ±0.1O ml from the exact 25.00 ml of
ACIDS IN CITRUS JUICES 43
Fig. 3-7. How to read a buret properly: Place a white card with an even heavy black line behind
the buret and about,'o inch below the meniscus. The dark line of the meniscus then can be compared
to the figure.
or:
9. If the normality is too high, use the following to determine the amount
of water to add where N is the high normality:
For example:
10. If the normality is too low, use the following to determine the grams of
NaOH pellets needed to add to the solution where N is the low normality:
12. With the solution at the proper concentration, using the spigot at the
bottom of the 50-gallon container, refill the one-gallon containers, snap
on new sealing lids, and apply a label to each container according to the
instructions of the inspector.
(3-7)
The carbonic acid then will proceed to neutralize the NaOH solution, changing
the concentration. For this reason it is important to isolate NaOH solutions from
the atmosphere. When one is using one-gallon containers, excessive opening
and closing of the container as well as excessive agitation in the presence of air
will result in a decrease in the NaOH normality. Syphon dispensing systems
ACIDS IN CITRUS JUICES 45
thus are recommended, as shown in Fig. 3-8. As air is drawn into the NaOH
reservoir during dispensing, it passes through ascarite CO 2 absorbent and Drier-
ite moisture absorbent. Ascarite turns white upon saturation, and Drierite turns
from blue to rose red and should be changed when needed.
Both manual titration and automated titration are used in citrus processing.
The advantages of auto-titration, which uses a motor-driven buret that stops at
a preset pH, include faster analyses and the need for less effort by lab techni-
cians. Manual titrations, as depicted in Fig. 3-8, are nearly as rapid as auto-
+- latex tubing
................... pH meter
~ pH electrode
pH reference wire
waste
."--150-200 ml beaker containing
and acid or juice sample
~06 stirring bar
~ magnetic stirer
mated methods and cost much less. Either method is suitable for industrial pur-
poses. The following titration procedure assumes a manual titration.
Acid Titration
Procedure
The USDA grade standards for acid content are generally incorporated in the
Brix/acid ratios, discussed in Chapter 4. Only canned orange juice (47FR 12/
10/82,52.1557) and lemon juice products have acid standards (CFR, Title 21,
146.114, 146.120, 146.121.). In California and Arizona sweetened or unsweet-
ened canned orange juice must have an acid level of 0.70 to 1.40 % acid for
Grade A, and 0.60 to 1.40% acid for Grade A in other areas. Grade B sweet-
ened canned orange juice requires 0.60 to 1.60% acid in any location, and
unsweetened canned orange juice requires 0.55 to 1.55 % acid for the same
grade. Lemon juice requires at least 0.70 % acid in reconstituted lemonade for-
mulations in order to meet the Standards of Identity for Frozen Concentrate for
Lemonade and Frozen Concentrate for Artificially Sweetened Lemonade. The
Standards of Identity for Lemon Juice require at least 4.5 % acid.
Because lemon and lime juices contain a higher percentage of organic acids by
weight than carbohydrates, lemon and lime juice concentrations are determined
as the weight of acid as anhydrous citric acid (ACA) per unit volume of juice-
grams ACA/liter or GPL (grams per liter). Instead of the sucrose equivalent
weight of soluble solids being the standard for inventories and marketing, the
weight of acid is the standard unit of measure. Even so, Brix measurements are
necessary in order to obtain densities in quality control calculations and in the
making of concentrates, as in-line refractometers are convenient to use. There
is no accurate in-line citric acid detector available for lemon evaporators, and
there are no accurate density scales available based on citric acid. Also, there
are few evaporators dedicated solely to lemon processing. The use of refrac-
tometers and the Brix scale makes it possible to process other types of citrus
fruit. The lemon or lime GPL values can be calculated from Brix data and acid
measurements.
Most lemon concentrates are evaporated to 400 or 500 ± 5 GPL. To accom-
plish this, the uncorrected observed Brix that is desired must be determined so
that the evaporator operator will know at what Brix level to operate the evap-
orator. Hence, the acid and the temperature-corrected Brix of the inbound sin-
gle-strength lemon juice must be known along with the % acid. First, the SPGj
of the single-strength juice must be found from the Brix by using Equation 2-8
(or 2-10) and the density d from Equation 2-7. The following equation then can
be used to calculate the SPGc of the final desired lemon concentrate:
or:
where A is the % acid of the single-strength lemon juice. The Brix correspond-
ing to SPGc can be found by using Equation 2-17 or equivalent density tables.
This is the acid-corrected Brix, Be- Because the evaporator operator needs to
know the uncorrected observable Brix, B o ' the acid correction for this Brix must
be determined and subtracted from Bc. This can be done by finding the density,
d c from Be by using Equation 2-7, which enables the calculation of the final
acid concentration, A c ' according to:
or:
(3-14 )
which gives 57.0 oBrix by using Equation 2-17. If we use Equation 2-7,
57.0 oBrix corresponds to a density of 1.2680 g/ml, which in tum is used in
Equation 3-14 to give:
Using this value in Equation 2-2 or Appendix A gives an acid correction to the
Brix of 5. rBrix. The corrected Brix (57 .0oBrix) minus the correction
(5.7°Brix) is 51.3°Brix, which the evaporator operator should adjust the evap-
orator to produce according to an in-line refractometer. If the temperature of
the concentrate is determined to be 28°C, the temperature correction to the Brix
ofO.64°Brix also should be subtracted from the 51.3°Brix, to give 50.7°Brix,
which should be run on the evaporator according to Equation 2-1 or Table 2-4.
ACIDS IN CITRUS JUICES 49
As with the other quality control calculations discussed thus far, the above
calculation lends itself to computer application. A flow chart outlining the steps
in such a program is shown in Fig. 3-9. A suggested GWBASIC program using
Fig. 3-9 can be found in Appendix B. A similar program, which can be used
with an HP-41C programmable hand-held calculator, can be found in Appendix
C.
no
EO 3-14
Aeor
EO 2-1
EO 3-12
SPGe
Teor
Bo = B - Aear - Tear
Bf = 10 X SPGe
Fig. 3-9. Flow chart that can be used to program computers to calculate the uncorrected Brix
needed from an evaporator when processing lemon concentrate.
50 CITRUS JUICE CHARACTERISTICS
where A is the % acid, and d is the density calculated from the corrected Brix
by using Equation 2-7. For example, if we were to make a 56.2°Brix lemon
concentrate with an acid content of 31.50%, Equation 3-15 would become:
which is within the standard 400 ± 5 GPL specification. Figure 3-10 shows a
flow chart that can be used to write a computer program for such a calculation,
such as the sample GWBASIC program found in Appendix B. A similar pro-
gram that can be used with a hand-held HP-41 C programmable calculator can
be found in Appendix C.
Since GPL and SPG are similar parameters, the blend equations (2-12 through
2-16) for other citrus juices can be used for lemon juice blending simply by
replacing all SPG values with GPL values. For example, suppose we want to
dilute 1000 gallons of 420 GPL concentrate to 400 GPL. The water needed can
be calculated from Equation 2-12:
Thus 167 gallons of the 500 GPL concentrate would be needed to replace 167
gallons of the 380 GPL concentrate in order to bring the concentration up to the
needed 400 GPL.
Similarly, if we have a full l000-gallon tank of 418 GPL lemon concentrate
and want to dilute it to 400 GPL, using Equation 2-14 we obtain:
Brix (B)
% Acid (A)
EQ3-15
GPL
This is the amount of the 418 GPL concentrate that would have to be replaced
with water to get 1000 gallons of exactly 400 GPL lemon concentrate.
In blending several lemon concentrates, the final average GPLj can be cal-
culated as follows, using the data given in the tabulation and Equation 2-16:
A recent advance of citrus processing is the use of weak anionic exchange resins
to remove organic acids from citrus juices (Varsel 1980). This technique is
advantageous in situations where blending may not be so practical in reducing
acid levels. However, the use of ion exchange may be permitted only for juices
that allow such in the standards of identity, such as Reduced Acid Orange Juice.
Nineteen different types of ion exchange resins have been approved by the Food
and Drug Administration (FDA) to date for use in food processing (CFR, Title
21, 173.5-173.25.), and many more are expected to emerge. Weakly basic
resins favor the removal of stronger organic acids, such as citric acid, which is
important in citrus processing where retention of the weaker nutritional ascorbic
and folic acids is preferred. The presence of larger amounts of citric acid also
favors its removal during ion exchange. The reaction that takes place during
this process is:
(3-16)
The positive unit of the resin (R +) exchanges its hydroxide ion to neutralize
the juice acidity. Overall, citrate is removed, and water is formed. As the resin
becomes saturated with citrate, the resin efficiency declines, and regeneration
with a basic solution is required.
Two types of columns containing these resins have been considered. An up-
flow fluidized bed approach has the advantages of treating juices with normal
pulp levels and of enhancing efficiency by more vigorous and thorough contact
between the juice and the resin. The main disadvantage is the difficulty in mak-
ing resins of sufficient particle size or density that they will not be carried out
with the juice. A down-flow packed column approach has the advantage of
increased surface availability of the smaller resin, which enhances contact ef-
ficiency as well as the efficiency of regeneration. The main disadvantage is that
the juice needed to have the pulp removed to a level of at least 2 % or less in
order to avoid plugging the column. The high cost of depulping centrifuges or
UF membranes adds to the capital costs. Backwashing the column to remove
entrapped pulp may determine how often regeneration needs to be done, rather
than saturation of the resin with citrates. Overall, the down-flow process has
proved to be the preferred industrial technique. The larger the particle size is,
the less the depulping required. Insoluble solids removed in depulping can be
restored after column treatment to replenish the original character of the juice.
As explained in Chapter 15, the acids in citrus juices serve to guard against
pathogenic organisms that flourish in less acid foods. For this reason, care should
be taken in manufacturing reduced acid juices so that the pH does not remain
above 4.6 for any length of time. Timely blending of reduced acid juices with
ACIDS IN CITRUS JUICES 53
other juices is a convenient way to achieve this. The standards of identity spec-
ify that reduced acid concentrate must have a Brix/acid ratio of 21-26 (CFR,
Title 21,146.148.).
Ion exchange and its cousin, adsorption techniques, are being used to remove
bitter principles in citrus juice, and often are used in conjunction with acid
removal. A single resin can be used for both. If only debittering is desired,
saturation of the resin with commercial acids as a last step in regeneration will
prevent significant citric acid loss during resin treatment.
QUESTIONS
PROBLEMS
1. What would be the % acid of a lemon concentrate sample weighing 1.684 g using
10.62 ml of a 0.781ON NaOH solution?
2. In the standardization of a 5-gallon NaOH solution, the meniscus on the buret is flat
on the 25.00 ml mark with just the tip of the meniscus below the mark, and 25 ml
of a 0.1562N Hel solution was being titrated. How much NaOH or water must be
added to get exactly 0.1562N NaOH?
3. Exactly 28.12 ml of 0.3125N NaOH solution was used to titrate 10 ml of single-
strength lemon juice. What is the % acid?
4. A 9.462 g sample of orange concentrate was titrated with a 0.1562N NaOH solution
with the meniscus on the buret just touching the 45.60 mark. What is the % acid?
5. How much water do you need to add to a 20-gallon NaOH solution standardized with
0.436 g benzoic acid using 22.42 ml of NaOH solution in the titration?
54 CITRUS JUICE CHARACTERISTICS
6. What is the GPL of a lemon concentrate containing 32.46% acid with a Brix of 51.2?
7. At what Brix would the evaporator operator run the evaporator when concentrating
an 8.9°Brix single-strength lemon juice with an acid content of 4.98% in order to
get a 500 GPL concentrate at 18°C?
8. How many gallons of water do you need for 1216 gallons of 496 GPL lemon con-
centrate in order to make 400 GPL concentrate?
9. If you have a full 3264-gallon tank of 408 GPL lime concentrate and want to change
it to a full tank of 400 GPL lime concentrate, what do you do?
10. How many gallons of 398 GPL lime concentrate do you need to blend with the
following in order to get a final 500 GPL product? How many gallons do you have
total? (Hint: See Equation 2-18.)
REFERENCES
The empirical Brix / acid ratio, found by dividing the acid-corrected and tem-
perature-corrected Brix by the % titratable acidity w / w as citric acid (B / A
ratio), is one of the most commonly used indicators of juice quality as well as
fruit maturity. In California, the fruit harvested for the fresh fruit markets needs
a B / A ratio of at least 8: 1 or 8, whereas the fruit harvested for juice in Florida
must have a B / A ratio of at least 10: 1 or 10. Even through fruit destined for
juice in Florida requires a B / A ratio of 10, commercial Florida juices must
have a B / A ratio of at least 13, which can be achieved through blending. Con-
sumers of citrus juices generally prefer a B / A ratio of 15 to 18, depending on
the product and individual tastes.
It is generally agreed that taste is a four-dimensional phenomenon consisting
of sweet, sour, salty, and bitter. Taste is sensed by receptors about 60 microns
in diameter, which are budlike structures having numerous microvilli about 2
microns long that project into the taste pores on the tongue. Taste is detected
by a disturbance of the electrical charges on the receptor due to contact with a
food substance. These receptors are very sensitive to abuse, and are replaced
every 12 to 17 days. The receptors respond to more than one taste, in various
combinations and intensities.
In citrus products, the sourness of the organic acids and the sweetness of the
sugars compete for the same receptor sites on the tongue. Thus the actual amount
of sugar or acid is of less importance in the taste of citrus products than the
ratio of the two; and this is why the B / A ratio (or simply just the ratio) is so
important as a flavor quality indicator in citrus juices. It also means that an
overabundance of acid sourness or sugar sweetness can be blended out by using
various combinations of the high- and low-ratio juices. This is done by using
the volume (Vn ), the pounds soluble solids/gallon (SPGn ) found from the Brix
55
and Equation 2-8 and the B / A ratio (Rn) for each component of the blend of n
components, in the following equation:
If the Brix values of all the components are essentially the same, the more
compact equation below can be used instead of Equation 4-1:
For example, if we wanted to calculate the final ratio of the following compo-
nents, Equation 4-1 would be used as shown below.
R = 12 + 9 + 18 + 6 + 4 = 144
12/13.0 + 9/16.7 + 18/12.3 + 6/21.8 + 4/18.9 .
Each equation gives a slightly different answer because the Brix values are
slightly different. The wider the range of Brix values, the greater the difference
will be in the results of the two equations. Figure 4-1 shows a flow chart that
can be used to program a computer to calcuate the average B / A ratio. Appendix
B contains a sample GWBASIC program that utilizes the flow chart in Fig. 4-1.
THE BRIX/ACID RATIO 57
n
Z' = 2: Vi (SPGi)
i=1
n
Z" = 2: Vi (SPGi)/Ri
i=1
no
EO 4-1
R = Z'/Z"
Fig. 4-1. Flow chart that can be used to program a computer to calculate the average Brix/acid
ratio in a blend.
58 CITRUS JUICE CHARACTERISTICS
Suppose that you want to know how much of a certain component you will
need to in order to achieve a certain desired ratio. Equations 4-1 and 4-2 can
be rearranged to the following:
(4-3)
and:
(4-4)
For example, if we wanted to calculate how much of Lot #1 we would need to
achieve the 14.3 ratio calculated in the previous example, using Equation 4-5
and 4-6 we would obtain with R = 14.3:
VI = 11 drums
or:
Here, the two equations again give different answers because the latter equation
does not take into account the higher Brix of the first component and thus is
less accurate than the first equation. Also, 11 instead of the original 12 drums
were calculated because of rounding-off errors. Apparently 11 or 12 drums will
give us essentially the desired 14.3 overall B / A ratio. Figure 4-2 gives a flow
chart that can be used in programming the above calculations. A sample of a
GWBASIC program using this flow chart can be found in Appendix B.
In reviewing the last three chapters and the given examples of computer ap-
THE BRIX/ACID RATIO 59
n
Z = L Vi (SPGi) ( -R 1. )
1 -R
i=2 I
no
EQ4-3
Z
V1 = - - - - - - -
(1/R1 - 1/R)SPG1
---~
Fig. 4-2. Flow chart that can be used to program a computer to calculate the volume needed of
one blend component to achieve a desired B / A ratio.
60 CITRUS JUICE CHARACTERISTICS
B = Bi + Acor + Teor
BAR = BfA
Fig. 4-3. Flow chart that can be used to program a computer to calculate the B / A ratio from
laboratory data.
THE BRIX/ACID RATIO 61
Bf = 10 x SPG
B = Bf
n
Z =? Vi (SPGi)
1=1
n
Zr = 2: Vi (SPGi)fRi
i=1
n
V = 2: Vi
i=1 no
no
B = Bf = average
Brix
EQ 4-1
avg BfA ratio= Z/Zr
EQ 2-16
SPG =ZN
Fig. 4-4. Flow chart that can be used to program computers to calculate the average Brix and the
Brix/acid ratio of a blend simultaneously.
Table 4-1. USDA standards for Brix/acid ratios.
Grade A Grade B
Type of Juice unsw sw unsw sw
62
THE BRIX/ACID RATIO 63
plications, one may notice that the programs can be consolidated and used to
calculate the Brix, % acid, B / A ratio, and water needed to adjust the Brix if
necessary. Figure 4-3 is a flow chart that can be used to construct such a pro-
gram along with the sample GWBASIC program found in Appendix B.
In blending, the average Brix and the B / A ratio can be calculated by com-
bining the flow charts of Fig. 2-8 and Fig. 4-1, to create the flow chart shown
in Fig. 4-4. This flow chart can be used to generate a program such as the one
found in Appendix Busing GWBASIC language. In this program, after adding
the data of the components, you merely need to press "enter" and the results
will appear. You also can add more data after a calculation is made. A similar
program for HP-41 C programmable calculators can be found in Appendix C.
Because the standard error for B / A ratios is ± O. I, these ratios should be
expressed or rounded off to the nearest tenth of a ratio point. Table 4-1 gives
the USDA B / A ratio grade standards for citrus products.
In the fresh fruit industry, once the fruit is mature enough for harvesting,
little attention is paid to the B / A ratio. However, juice processors are concerned
about the B / A ratio throughout the season. The rate of increase is linked to the
rate of changes in the Brix and acid levels in the fruit, and of these two param-
eters, the change in the acid levels generally has the greater effect on the B / A
ratio. Brix levels range from about 9 to 15, whereas acid levels may range from
0.5% to 1.5%, a threefold increase. Smooth acid changes, as depicted in the
last chapter for Washington navel oranges, have been shown to generate linear
relationships with heat accumulation, as shown in Fig. 4-5. These linear rela-
tionships have been empirically correlated with such parameters as crop size ( C
in tons / acre), tree growth or age (Y), maximum accumulated temperatures
between May and August just after bloom (M), and accumulated heat during
the growing season (H), according to the following equation:
QUESTIONS
1. What are the minimum B / A ratios required for harvesting oranges in the
United States?
64 CITRUS JUICE CHARACTERISTICS
22
NAVEL
20
18
o
......
16
ca
a:
~
CD 14 1977·78 •
1978.79 0
1979·80 lC
1980·81 ~
1981.82 0
12 1982·83 &
1983·84 •
10
12 14 16 18 20 22
PROBLEMS
Blend A Blend B
Lot # # Drums Brix BfA Ratio Lot # # Drums Brix BfA Ratio
1 4 60.1 14.6 1 16 66.4 10.2
2 10 59.5 21.6 2 3 58.7 14.6
3 7 60.0 18.7 3 11 62.8 19.1
4 24 60.4 13.9 4 8 60.5 16.3
5 9 59.9 15.1 5 5 64.0 9.9
1. What is the average B / A ratio of blend A, and which equation would be best to use?
2. What is the average B / A ratio of blend B, and which equation would be best to use?
3. How many drums of 59.8°Brix 20.4 B / A ratio are needed to bring up the ratio in
blend B to 14.0?
4. How many drums of 60.1 °Brix 22.3 B / A ratio juice are needed to bring up the B / A
ratio in blend A to 16.4?
5. Suppose that it is September and you want to know if navel oranges will be mature
enough for harvesting in California by November 1. You know that the summation
of the maximum summer temperatures from May through August gave 10,879 degree
days, that the estimated crop load is about 13.5 tons/acre, and the estimated sum-
mation of daily temperatures from July 1 up until November 1 is 10,702 degree days
for the upcoming season (Y = 10). Based on this infonnation and that found in the
chapter, is it likely that navel oranges will be mature enough to harvest by November
I?
Chapter 5
66
Procedure
The yield can be reported as gallons of juice per ton of fruit, but the most
popular yield expression is pounds of soluble solids per ton of fruit. This can
be calculated from the fruit sample data by use of the following equation:
where W; is the net weight of the juice, Wf is the net weight of the fruit, B is
the corrected Brix, and F is a factor applied to account for fruit loss during
68 CITRUS JUICE CHARACTERISTICS
grade-out in normal processing. The generally accepted value of this factor for
oranges, grapefruit, and tangerines is 0.85, allowing for a 15% loss of fruit,
whereas for lemons and limes the accepted factor is 0.78.
The gallons of juice per ton of fruit can be found by first obtaining the density
D (lb / gal) from the Brix by using Equation 2-9 or 2-11, and then using the
following:
This value is helpful in estimating the size and uses of processing equipment.
Also of value in estimating the size and uses of the equipment employed in
peel processing, as well as in crediting fruit growers for the resulting profits in
the sale of peel products, is calculation of the pounds of peel per ton of fruit.
This can be done by using fruit sample data, as follows:
Each of the three above parameters can be used with the total net weight of the
fruit to give the total weight of solids, gallons, and weight of peel from a given
load of fruit. For example, if 20,172 lb of fruit were brought into a processing
plant with a random fruit sample weighing 34.11b (including a 3.8-lb container)
that yielded 14.2 lb of juice (including a 1.1-lb bucket) with a corrected Brix
of 12.6, Equations 5-1,5-2, and 5-3 would give the following results:
As with other quality control calculations, fruit sample data processing can
be performed most efficiently on a computer. Figure 5-1 is a flow chart that can
be used to write the appropriate programs. A suggested GWBASIC program
using such a flow chart can be found in Appendix B, with a similar program
for HP-41 C programmable calculators in Appendix C.
Even though fruit sample analysis is used to credit fruit owners for the fruit
processed, processors generally are responsible for ensuring that the results of
such tests are sufficiently accurate. Also, the processors need to monitor actual
yields to ensure that they compare to the theoretical yields of the fruit samples,
so that they will not shortchange the fruit owners or themselves. For example,
if, in a given week, a processor recovers 100,000 lb of soluble solids but credits
the fruit owners with 110,000 lb of soluble solids, he will have shortchanged
himself by 10,000 lb of soluble solids, which he will have to pay for from some
TESTING OF FRUIT SAMPLES 69
B == Bi + Acor + Tcor
R == B/A
Fig. 5·1. Flow chart that can be used to program a computer to calculate the fruit sample data
needed to credit a fruit owner for his or her fruit that is processed.
other funds. Conversely, if he credits the fruit owners with only 60,000 lb of
soluble solids, he will have cheated the fruit owners out of 40,000 lb of soluble
solids. This is why it is important to obtain as representative fruit samples as
possible, with results that match actual industrial yields as closely as possible.
Production efficiency can easily be measured by using:
to 11 0 %, which gives the processor the freedom and the economic means to
make slight adjustments in returns to fruit owners and to remain financially
solvent. If the % efficiency is outside the ideal range and/or fruit sample results
appear out of line, the following checks should be made:
Lemons (and limes) require a slightly different approach to fruit sample analysis
because their juice concentration is based on the weight of acid per unit volume.
The procedure for extraction of the juice from lemon fruit samples is the same
as with oranges, and the juice is tested in the same manner except that the NaOH
solutions used in the acid titration may be more concentrated here. Once the
fruit has been tested, the calculations of the results differ from those performed
for other citrus fruit. First, the density, D, needs to be calculated from the Brix
by means of Equation 2-9 (or density tables), and used in the yield equations
below:
where llj is the net weight of the lemon juice, Wf is the net weight of the lemon
fruit, and F is the production loss factor of 0.78 for lemons and limes. For
example, a lemon fruit sample weighing 31.2 net Ib yielding 11.61b net weight
of juice with a corrected Brix of 9.7 and a % acidity of 5.37 % would give:
The lb peel/ton is calculated the same way as for orange fruit samples, men-
tioned previously. Figure 5-2 shows a flow chart that can be used in program-
ming computers to perform these calculations, and a sample GWBASIC pro-
gram can be found in Appendix B. A similar program for the HP-41C
programmable calculator can be found in Appendix C.
TESTING OF FRUIT SAMPLES 71
B = Bi + Acor + Tcor
EO 5-5 EO 5-6
Ibs acid/ton gallons/ton
F = 0.78 F = 0.78
Fig. 5-2. Flow chart that can be used to program computers to calculate data from lemon or lime
fruit sample testing.
QUESTIONS
1. Why are actual yields not used to credit fruit owners for juice and by-
product yields?
2. What is the purpose of the fruit receipt or ticket?
3. Why and how often should actual yields be determined?
4. Why is a different factor, F, used with lemon fruit samples compared to
other citrus fruit?
5. What would probably be the cause of an 80% level of efficiency in pro-
cessing tangerines?
72 CITRUS JUICE CHARACTERISTICS
PROBLEMS
1. What would be the yield in sol/ton, gal/ton, and peel/ton of a fruit sample with a
weight of 27.6 Ib (including a 3.4-lb container) that yielded 13.4 Ib (including a
1. l-Ib container) of juice that had a corrected Brix of l3.4?
2. If grower A sent 19.6 tons of fruit into a processing plant that gave a yield of 106.2
Ib sol/ton, and grower B sent in 22.4 tons of fruit with a yield of 92.2 Ib sol/ton,
who would receive more money for the fruit?
3. Is the efficiency of a plant that obtains 315.6 tons of oranges in one week with a total
solids of 31,150 Ib credited to fruit growers or owners, and which produces 5073
gallons of 60.0 o Brix concentrate from that fruit, within an acceptable range?
4. What would be the yield in lb acid/ton, gal/ton, and peel/ton of a lemon fruit
sample with a net weight of 28.6 Ib yielding a net weight of 9.8 lb of juice with a
corrected Brix of 9.2 and 6.12 % acid?
5. If the fruit sample in problem 4 were grapefruit instead of lemon, what would the
fruit sample yields be?
Chapter 6
The sweetness of carbohydrates and the sourness of organic acids impart about
the same flavor characteristics to all citrus juices. However, the distinctive fla-
vor of an individual variety can be attributed to varying components in the oils,
aromas, and essences found within the juice. Although some compounds con-
tribute more to flavor than others, exact citrus flavor duplication would require
a delicate balance of a host of contributors. About 87% of the fresh squeezed
flavor of orange juice has been restored to freshly evaporated concentrate by
using d-limonene, ethyl butyrate, citral, and acetaldehyde (Ahmed, Dennison,
and Shaw 1978). It is well known that oxygenated hydrocarbons are major con-
tributors to citrus flavor, with d-limonene acting more as a carrier of these fla-
vors than an actual contributor itself. The excess d-limonene levels, however,
produce the characteristic "oil bum" taste that can irritate the skin or eyes. The
d-limonene in the peel will bum the edges of the mouth when one is eating fresh
fruit, and d-limonene fumes will bum the eyes during juice extraction in un-
ventilated areas. Natural variations of the delicate balance of flavor-producing
compounds are primarily responsible for vintage and off years. Processing also
has an effect on these compounds and can make or break the quality of a par-
ticular citrus product.
Not only do citrus oils, aromas, and essences contribute to juice quality, but
they can be commercially recovered and used to enhance juices and other food
products as well as be used in other industries. Because these oil products were
extracted from natural sources, the citrus fruit itself, they can be added back to
juices without violating federal standards of identity. This method of restoring
citrus flavor to concentrates generally has replaced the older method of adding
freshly squeezed cutback juice to high-Brix concentrates. Citrus oils also have
been used in the manufacture of textiles, chemicals, and other commodities.
Their toxicity to certain insects, such as houseflies, fire ants, and fleas, as well
as their herbicidal and flammable nature, suggest other promising uses.
73
D. Kimball, Citrus Processing
© Van Nostrand Reinhold, New York, NY 1991
74 CITRUS JUICE CHARACTERISTICS
OIL COMPOSITION
Citrus oils are found primarily in oval-shaped sacs in the flavedo or colored
portion of the peel, and they act as a natural toxic barrier to many microorga-
nisms and insects. Citrus oils are composed primarily (over 90%) of
d-limonene, a sesquiterpene, with other monoterperenes and sesquiterpenes
being found in trace amounts.
In orange oils, 111 volatile constituents have been found (Shaw 1977), in-
cluding 5 acids, 26 alcohols, 25 aldehydes, 16 esters, 6 ketones, and 31 hydro-
carbons. Nonvolatile constituents comprise about 1.5% of the orange oils, in-
cluding waxes, coumarins, flavonoids, carotenoids, tocopherols, fatty acids,
and sterols.
Grapefruit oil is characterized by the presence of nootkatone, which gives a
grapefruit flavor and aroma. It consists of 20 alcohols, 14 aldehydes, 13 esters,
3 ketones, and 14 monoterpenes and sesquiterpenes. The nootkatone increases
with fruit maturity. The nonvolatile portion of grapefruit oil comprises about
7 % of the oil, including coumarins, flavonoids, tocopherols, and waxes.
Tangerine oils contain the distinctive nonvolatile tangeretin in significant
amounts (4 %), along with 24 alcohols, 11 aldehydes, 4 esters, 2 ketones, 7
acids, 24 hydrocarbons, and 2 ethers.
Lemon oils are believed to contain about 2 % nonvolatiles, primarily in the
form of coumarins, along with 18 alcohols, 16 aldehydes, 11 esters, 3 ketones,
4 acids, and 23 hydrocarbons.
Lime oils have been known to contain 12 alcohols, 7 aldehydes, 4 esters, 1
ketone, and 22 hydrocarbons, with 7 % nonvolatiles, mostly in the form of cou-
marins.
oil from the product. The Scott method is much easier and faster, with greater
accuracy and precision, than the Clevenger method. Scott oil values are higher
then Clevenger levels, suggesting a more efficient analysis. The Scott method
involves bromine addition to the two double bonds in d-limonene, the mea-
surement of d-limonene as a representation of the oil level having become stan-
dard in the industry. Citral, the cis-trans mixture of neral and geranial that
occurs in lemon and lime oils in quantities up to 4 %, reacts with half as much
bromine as d-limonene. Also a-pinene reacts with the same amount of bromine
as d-limonene, even though a-pinene has only one double bond. Scott attributed
this phenomenon to some sort of secondary reactions or molecular rearrange-
ment of the a-pinene.
Because bromine is very volatile in pure form or in solution, straight standard
bromine solutions are unreliable in concentration. Scott used a bromide-bro-
mate salt solution to titrate the acidified alcoholic distillate from a juice sample,
as this salt solution was found to have a very stable concentration. As soon as
the bromide-bromate solution comes into contact with the acidified distillate,
an acid-catalyzed oxidation-reduction reaction takes place between the bromide
and the bromate to form bromine, which can react immediately with the double
bonds of d-limonene, according to the following equation:
The EO value for the above reaction is +0.43 V, which is indicative of the
thermodynamic favorability of the reaction. Once the bromine has been gen-
erated, it proceeds to react with d-limonene, and the other minor constituents
mentioned above, according to the following reaction:
Br
2Br2 +
bromine
(6-2 )
brominated
d-limonene
d-limonene
What makes this basic chemistry effective as a tool in routine industrial qual-
ity control is detection of the endpoint by means of methyl orange, a common
acid-base indicator. Bromine reacts preferentially with d-limonene until the
76 CITRUS JUICE CHARACTERISTICS
unbrominated terpene is exhausted. It then reacts with the pink or violet methyl
orange according to the following reaction:
Methyl Orange
(6-3 )
water
out
Condenser
Boiling Flask
~;~C:=-"'~I-- water
Juice and in
Alcohol ---~..
Fig. 6-1.
Heater
g.~-- Distillate
+- 5 Gallon Resevoir
Fig. 6-2. Setup for titration of distillate in Scott method oil detennination.
Procedure
distilled water to the boiling flask, and heat the mixture to boiling using
the setup in Fig. 6-1.
3. Distill until moisture can be seen inside the distillation apparatus in the
form of water beads. Even though 25 ml of alcohol was added to the
sample, more than 25 ml should be collected in the distillation because of
the azeotropic nature of water/alcohol mixtures. Premature termination of
the distillation will result in artificially low oil levels.
4. Using an auto-dispenser or a graduated cylinder, add about 10 ml of the
diluted acid and 1 drop of methyl orange indicator to the distillate. An
additional 25 ml of isopropyl alcohol can be added to the distillate to
provide more volume for mechanical stirring but is not necessary for the
titration.
5. Titrate the distillate, while stirring, with the bromide-bromate to a col-
orless endpoint.
6. Oil levels are reported as the % by volume in 11.8 °Brix juice. This is
found as follows:
where the factor is found from the above standardization. For example,
if 1.5 ml of bormide-bromate were titrated, Equation 6-4 would become:
The USDA grade standards for citrus products can be found in Table 6-1.
Even through grade A orange juice is allowed up to 0.035 % oil, oil bum can
be detected below 0.020 %, depending on the juice. Commercial orange juices
generally contain about 0.015 to 0.025% oil.
It is sometimes of interest to determine the final oil in a blend so that one can
predetermine the amount of additional oil or oil-based flavors without exceeding
any maximum oil level specifications. For example, if an oil specification re-
quires the oil level to be 0.020% or lower, and you want to add as much oil-
based flavor enhancer as possible, the average oil content of the components in
a blend must be determined prior to any further oil addition. Once the oil level
of the blended components has been calculated, the amount of flavor enhancer
that can be added without exceeding oil juice specifications can be calculated.
It is important to determine these parameters prior to blending because of the
difficulty of mixing the immiscible oil with the aqueous juice. Oil addition is
80 CITRUS JUICE CHARACTERISTICS
Table 6-1. Maximum oil levels (% v /v) allowed under USDA grade
standards
Grade A Grade B
best done as the components are being added. The average oil of the blend
components can be found from the following equation:
( 6-5)
where volumes (V), % oils (0), and the pounds of solids per gallon (S) (cal-
culated from the Brix using Equations 2-8 or 2-10) of the blend components are
used to calculate the final average oil of the blend (0a). If the Brix values of
the blend components are similar, the S values may be omitted. The amount of
oil needed to raise the oil level to a certain level can be calculated by using the
following equation:
(3785.306 ml/gal)SV
mloil = (Of- 0;) 100%(1.046 sol/gal) = (Of- 0;) (36.19)(SV)
(6-6)
CITRUS OILS, AROMAS, AND ESSENCES 81
where the final desired oil level (Of), the initial oil level (0;), the pounds of
solids per gallon (S) (found from the Brix using Equation 2-8 or 2-10), and the
juice volume (V) can be used to calculate the milliliters of oil that need to be
added. In a blend, the three latter values should be the calculated average values
for the final blend. If the calculated average oil level of a blend is too high to
allow you to add as much oil or flavor enhancer as you would like, one of the
blend components can be exchanged for one with less oil to maintain the same
overall volume, or a different volume of one of the blend components can be
used. In the former case, the following equation can be used:
Both of the above equations are rearrangements of Equation 6-5. Again, if the
Brix values of the components are similar, the S values can be omitted.
Let us take an example to illustrate the use of the above equations. Suppose
that we had the following blend components:
Also, suppose that we want to add as much flavor enhancer as possible, not to
exceed 0.020% oil. First, Equation 6-5 would be used to determine the average
oil in the blend as follows:
As you can see, the units of volume can be anything as long as the same units
are used throughout the calculation. In order to raise the oil level from 0.012 %
82 CITRUS JUICE CHARACTERISTICS
to the desired 0.020%, Equation 2-16 can be used to find the average pounds
of solids/gallon:
SPG = _18-....:('--7_.1_35-'.)_+_12--.:('--6_.7_26....:.)_+_1_6(-,-6_.4_0_9),-+_7--'.(_6._5_18-,-)_+_7_(,-7_.1_0-,-7)
18 + 12 + 16 + 7 + 7
= 6.784
Next, Equation 6-6 can be used to calculate the milliliters of oil needed to raise
the oil level from the average of 0.012 % to 0.020%:
This means that we need to reduce the 0.012 % average oil level in the blend to
0.010% by changing the blend components. Suppose that we wanted to ex-
change lot 3 for a lot with a lower oil content with a Brix of 60.1 (S = 6.450).
Equation 6-7 would become:
= 0.013% oil
which represents the maximum oil level that the replacement lot could contain.
If a lot with such an oil level is difficult to find, the number of drums of lot 3
CITRUS OilS. AROMAS. AND ESSENCES 83
can be changed instead of the oil level, using Equation 6-8 as follows:
During certain times of the harvesting season, freshly extracted juice reaches a
B / A ratio that is suitable for commercial juice as is without blending. This juice
can be pasteurized to stabilize the enzyme activity or packaged as is, depending
on the standard of identity employed. Freshly extracted juice, however, con-
tains high levels of citrus oils that exceed the USDA grade standard of 0.035 %.
These oils generally require deoiling using a heated vacuum technique. This
deoiling also serves as pasteurization and enzyme stabilization. New extraction
techniques are emerging that minimize the oil levels of freshly extracted juice
that may eliminate the need for deoiling in the future. During deoiling opera-
tions, the oil levels should be constantly monitored. If the oil levels are too
high, more steam or slower flow rates should be used, and vice versa. Slight
concentration of the juice during deoiling is commonly observed. The aqueous
distillate containing aroma components is commonly returned to the deoiled
juice in order to restore some of the flavor lost during deoiling as well as to
counter the slight concentration effect.
Citrus oils and aromas are generally the major by-products of citrus processing
and are used in a variety of products. Eight types of oils and aromas are com-
monly manufactured, and their general characteristics, standards, uses, and
method of processing will be discussed hereafter. Generally speaking, the tenn
"aromas" refers to the aqueous phase separated from the oil phase in evapo-
84 CITRUS JUICE CHARACTERISTICS
EO 6-5
Oa = Zl/Z2
EO 6-9 EO 2-16
EO 2-16 Oe or max. Oi
SPGavg = Z2Nt allowed per SPGavg = Z2M
specification
EO 6-6
mls needed to get Of
Fig. 6-3. Flow chart that can be used to program computers to calculate the milliliters of oil
needed in a blend or the maximum allowable milliliters of oil-based flavor enhancer that can be
added per specification.
CITRUS OILS, AROMAS, AND ESSENCES 85
rator and pasteurized condensates, which contains important volatile flavor and
aroma components.
Cold pressed oils are recovered from the flavedo of citrus peel by rasping or
rupturing the oil sacs close to the surface of the fruit just prior to or during juice
extraction. Because heat easily removes the valuable volatile oil components,
no heat is applied to the oil. As the oil is extracted, water is used to wash it
from the peel to form a slUrry. The use of too much water during this step causes
excess peel material to get into the slurry, which can absorb important oil con-
stituents, such as aldehydes, that are removed with the peel material. Excessive
pressure in oil sac rasping or rupturing also will cause excess peel material to
get into the slurry. The peel then is washed with water, and the washings are
conveyed to a finishing press that separates the oil slurry from pieces of peel.
Excessive finishing pressure will cause excessive insolubles to get into the oil
emulsion, making it more difficult to break during polishing.
The oil slurry is centrifuged in a desludging centrifuge (8,000-10,000 rpm)
that separates the slurry into an oil-rich emulsion, an aqueous discharge, and a
semisolid sludge material. The sludge usually is discarded, but the aqueous
discharge may be filtered and returned to wash the peel to form additional oil
slurries. In this way, any small amounts of oil remaining in the aqueous dis-
charge may be recovered through a recycle of the water. The oil-rich emulsion
may be treated with enzymes for several hours, depending on the enzyme, or
it may be placed in a freezer for about 30 days in order to help it break up for
increased oil recovery. The emulsion then is centrifuged in a two-way separator
or oil polisher (16,000-18,000 rpm) that centrifuges it into a discarded heavy
phase and a light clear oil phase. Some plants polish the emulsion immediately
after desludging. The product then can be treated enzymatically or stored in a
freezer for about 30 days to facilitate the precipitation of waxes dissolved in
the oil. After such "winterizing," the oil can be decanted and filtered and the
waxy precipitates discarded. The oil should be stored in closed-head glass, tin-
dipped, aluminum, wall-galvanized, or suitably lined drums under nitrogen or
carbon dioxide. Citrus oils absorb readily into most plastics, so plastic con-
tainers should be avoided. Refrigeration and air exclusion in the headspace of
the oil drums minimize oxidation. However, the precipitation of waxes may
continue, as precipitation slows with time but never really stops at low temper-
atures. Storage temperatures of 60 to 75°F generally are recommended to keep
the waxes in solution. Also, protection from light is advised because photo-
chemical reactions with some oil components may detract from the quality of
the oil.
86 CITRUS JUICE CHARACTERISTICS
Oil yields generally increase with maturity of the fruit but decrease right after
periods of rainfall. Soft fruit generally gives poorer yields than firm fruit be-
cause of less efficient rasping or rupturing of the oil sacs. Grapefruit oils are
best extracted from February through April and need to be cured after dew axing
for about 6 months (60-70°F). The oil starts out orange-like in nature and then
undergoes a chemical change. Waxes take longer to precipitate in grapefruit
oils than with orange oils even though 95 % of the waxes are removed in the
first two months of storage. However, waxes may continue to precipitate for up
to 2 years. The lower the temperature is, the faster the precipitation.
The % recovery should be above 90%. The heavy aqueous discharge may be
filtered and reused to wash more oil from the peel into the oil slurry in order to
reclaim a portion of the oil remaining in the aqueous material.
The oil coming from the oil-rich emulsion from the desludger also is a pa-
rameter ofthe oil recovery efficiency. It can be measured by weighing 1-2 grams
of the emulsion into a 100 ml volumetric flask and filling the flask to the mark
with isopropyl alcohol. The weight of the sample must be recorded exactly.
Then 1 m1 of this dilution is added to 25 ml of isopropyl alcohol and titrated
CITRUS OILS, AROMAS, AND ESSENCES 87
by the Scott method. The % oil in the emulsion can be calculated by using:
where Oe is the % oil in the emulsion, and d is the density of the oil. For orange
oil the density is 0.842 giml, and that for lemon oil is 0.853 giml. The oil
level in the emulsion should be between 70 and 80 % and can be changed by
varying the flow rate through the desludging centrifuge, the length of time be-
tween bowl flushings, and the length of time the bowl remains open during the
flush cycle. The oil emulsion then should be treated enzymatically and stored
in cold storage, or it may feed into the polisher immediately, depending on the
oil recoveries desired. Immediate treatment of the oil emulsion may result in
some oil loss during polishing because the emulsion will be more difficult to
break. The rate of flow into the polisher varies with equipment but should be
about 1 to 1.5 gallons per minute. The heavy discharge from the polisher can
be analyzed for oil by the Scott method in a manner similar to that for the oil
emulsion; and it should not contain more than 5 to 7 % oil. In larger operations
more than one polisher may be required to minimize oil losses. A 100-mesh
screen sometimes is used to remove extraneous debris from dewaxed or pol-
ished oils. Aroma oil is made from the distillation of the aqueous discharge
from polishers.
Folded Oils
Some cold pressed oils are further distilled in order to remove d-limonene while
concentrating the less volatile oxygenated flavor compounds. These concen-
trated oils, referred to as folded oils, are valuable materials that may be added
back to citrus juices to enhance their flavor without increasing oil levels as much
as untreated cold pressed oils do. Some oils are stripped of all the terpenes and
are referred to as terpeneless oil-a product that is very rich in flavor. Aromas
from the aqueous layer in these steam distillations also may be recovered.
The natural fresh flavors and aromas characteristics of freshly extracted juice
are largely removed during commercial evaporation. Essence oils are recovered
from the condensate from these evaporators and differ from other citrus oils in
the content of valencene, a sesquiterpene (0.5-2.0%), which is not found ap-
preciably in other oils. Excess heat treatment during the recovery of essence oil
will result in chemical conversion from valencene to nootkatone, a character-
istic component of grapefruit oil, as well as breakdown of the esters, under the
acidic conditions of the juice, to alcohols and acids, to give an off aroma com-
monly called a "wet dog" aroma.
There are basically two types of essence oil recovery systems. One uses a
fractionation column to concentrate the volatile materials from the first effect
or first vaporization in commercial juice evaporators, and the second fraction-
ates the volatiles from both the first effect and the second effect; and then the
final products from the two systems are mixed together. The aqueous phase, or
aroma, primarily consists of alcohols and aldehydes (13 % ethanol) (Johnson
and Vora 1983). Yields are generally in the neighborhood of 0.20% of the
inbound juice (11.8°Brix) feeding into the evaporator for aroma and 0.014%
for essence oil.
Juice Oils
The term "juice oils" really has two meanings. Citrus oils not only are found
in oil sacs in the flavedo of the fruit peels, but they occur in the juice cells
themselves. About 0.005 % of the oil in fresh single-strength juice comes from
the juice cell. The juice cell oil is not exactly the same as that found in the fruit
peel, as the juice oil is higher in esters and lower in aldehydes than cold pressed
oil. The oil recovered from single-strength juice in deoiling operations also is
called juice oil even though most of this oil comes from the peel. The latter
definition is the one more commonly used in the industry. This juice oil is
CITRUS OilS. AROMAS. AND ESSENCES 89
The quality of citrus oils is centered primarily around their flavor attributes,
which arises from a multitude of oil components in varying concentrations.
These delicate flavor balances are little understood; so quality control or quality
monitoring other than subjective organoleptic methods has been very difficult.
However, other methods that are easily used in routine quality control have
served as a means of estimating citrus oil quality, especially in regard to adul-
teration and oxidative tendency. Standards have been set by the United States
Pharmacopeial Convention, Inc. (USP), which first met on January 1, 1820 and
met every ten years thereafter until 1970, when the decision was made to update
its information every five years. The information given below is based on the
convention held in July, 1975. The Food Chemicals Codex (FCC), published
by the National Research Council and officially recognized by the Food and
Drug Administration, contains standards for citrus oil quality. The USP and
FCC standards are summarized in Table 6-2. For more detailed descriptions of
these standards the reader is referred to the original texts (The United States
Pharmacopeia 19th Revision 1975; Food Chemicals Codex 1981).
Optical Rotation
Many chemical compounds occur in pairs that differ from one another only in
the fact that they are mirror images and cannot be superimposed upon each
other. These pairs are referred to as enantiomers, and their difference sometimes
affects their chemical reactivity. If such a pair contains a chiral carbon (a carbon
bonded to four different entities), they will rotate plane polarized light as shown
in Fig. 6-4. If equal amounts of both enantiomers exist together, no net rotation
of polarized light will be observed, as one enantiomer rotates the light to the
right (dextrorotatory) and the other rotates the light to the left (levorotatory).
The degree that the light is rotated as it passes through a solution containing
the enantiomers is directly proportional to the concentration difference of one
enantiomer compared to the other, according to:
R = ale ( 6-12)
90 CITRUS JUICE CHARACTERISTICS
Polarizer 1 Polarizer 2
Monochromatic
Light
.~ .. -'~-l>
EB CD
Unpolarized Plane Polarized
(])
Rotated
Light Light Light
Fig. 6-4. Measurement of the optical rotation of citrus oils. The first polarizer polarizes the light,
and the oil rotates the light. The second polarizer can be rotated to detect the degree of rotation.
Procedure
Refractive Index
same thing, the estimated d-limonene content of the oil. However, it would be
very difficult to adulterate citrus oils with a component that had both the same
optical rotation properties and the same optical density properties as d-limo-
nene. Both tests are relatively quick and easy, so both are worthwhile if adul-
teration is a concern. Pure d-limonene has a refractive index of 1.4727 (20°C),
and the refractive index of citrus oils ranges from 1.4720 to 1.4740 (20°C).
Most processing plants already possess a refractometer for Brix measurements,
making the refractive index of citrus oils an even easier parameter to define. In
dry years the refractive index will increase with maturity. The following pro-
cedure can be used.
Procedure
1. Carefully clean the prism of the refractometer with water or alcohol, and
dry it with tissue paper.
2. Using the dropper or plastic stirring rod, apply a few milliliters of the
sample to the prism so that it is completely covered. Care should be taken
that no hard or abrasive object comes into contact with the prism.
3. For indirect scales, adjust the shadow to the center ofthe cross hairs, and
read the refractive index on the corresponding scale. For direct scales,
read the scale where it intersects the shadow.
4. It is recommended that all samples be measured at 20°C using a temper-
ature bath and thennometer. However, if corrections are necessary, adjust
the refractive index for orange oil by adding or subtracting 0.00045/ ° C,
and for lemon oil by adding or subtracting 0.00046/ °C if the temperature
is above or below 20°C (Kesterson, Hendrickson, and Braddock 1971).
Specific Gravity
The specific gravity is the density of the material divided by the density of water
at the same temperature. This test also is redundant if used with the refractive
index and the optical rotation in estimating the d-limonene content of citrus oils
CITRUS OILS, AROMAS, AND ESSENCES 95
and the general composition. Again, however, the greater the variety of tests
perfonned, the greater the chance is of detecting adulteration. The specific grav-
ity of d-limonene is 0.8419 (21°C), compared to that of citrus oils which ranges
from 0.842 to 0.881. In determining the specific gravity, a pycnometer com-
monly is used to measure a certain volume of the oil that, along with the oil
net weight, can be used to determine the specific gravity of the oil. The specific
gravity generally will increase with fruit maturity.
Procedure
1. Rinse the pycnometer with a few portions of the oil that is to be analyzed,
and fill it with the oil without putting the cap on the pycnometer; or place
about 50 ml of the oil into a beaker.
2. Place the pycnometer or beaker in a temperature bath, and warm it to
20°C.
3. Place the cap on the pycnometer and wipe the excess oil from the exterior
until it is clean and dry; or pipette 10 or 25 ml of the oil into a clean dry
beaker. Weigh the pycnometer or beaker immediately, using the lab bal-
ance to ± 1 mg.
4. Discard the sample, and clean and dry the pycnometer or beaker and re-
weigh it. Subtract the tare weight to get the sample net weight.
5. Repeat the procedure using distilled water. Divide the density of the sam-
ple (gjml) by the density of distilled water (gjml) to get the specific
gravity. The density of the water need not be determined each time. A
standard value of 0.998203 (200C) can be used instead for the density of
water, based on a water density of exactly 1 at 3.98°C.
6. Temperature corrections, if necessary, can be made by adding or subtract-
ing 0.00078 (0.00077 for lemon oil) for each centigrade degree above or
below 20°C (Kesterson, Hendrickson, and Braddock, 1971).
Aldehydes
Unlike the optical rotation, refractive index, and specific gravity, measurement
of the aldehyde level in citrus oils is a direct measure of the flavor quality of
96 CITRUS JUICE CHARACTERISTICS
those oils. For this reason, aldehyde measurements probably are the major qual-
ity parameter used in the marketing of these oils. Even though other oil con-
stituents contribute significantly to the organoleptic characteristics, the alde-
hydes are considered the major contributor. In lemon oils the most abundant
aldehyde is citral, which is actually the cis-trans mixture of neral and geranial
and occurs at levels of about 2 to 4 % in the oil. The most abundant aldehyde
in other citrus oils is decanal, with levels ranging about 0.8 to 2.0% in the oil.
Aldehyde analyses thus are based upon the citral or decanal equivalent of al-
dehydes.
There are several ways to measure the aldehyde content of citrus oils, in-
cluding the Kebler method (Kebler 1921), the Kirsten modification of the Ke-
bIer method (Kirsten 1955), and the N-hydroxy benzene sulfonamide (or HBS)
colorimetric method (Petrus, Dougherty, and Wolford 1970), as well as several
forms of the hydroxylamine hydrochloride method (J. o/the AOAC 1953; Carter
1981). The fastest, safest, and easiest method is that of Petrus, Dougherty, and
Wolford (1970). Hydroxylamine hydrochloride is added to the oil and reacts
with the aldehyde groups as follows:
0 N-OH
II II
(NH 3 OH)+Cl- + RCH ~ RCH + HCI (6-13 )
hydroxylamine an aldehyde an oxime hydrochloric
hydrochloride acid
Procedure
1. Weigh about 5 g of the oil to be analyzed into the beaker shown in Fig.
3-8. Record the exact weight.
2. Add 35 ml of the 0.5N hydroxylamine hydrochloride solution to the flask
using the graduated cylinder. Stir the mixture for exactly 30 minutes.
3. Using the setup in Fig. 3-8, titrate with 0.1562N NaOH solution to a pH
endpoint of 3.5.
4. Calculate the % aldehyde using:
where M is the molecular weight of the citral for lemon oils (152.23
g/mole) or for decanal for other citrus oils (156.27 g/mole).
Evaporative Residue
The residue left after evaporation of the volatile components of oil products has
a tendency to stabilize the oil toward oxidation, one of the main causes of oil
quality deterioration. For this reason, the amount of evaporative residue can
serve as an estimate of the oxidative stability of the oil, which may be of oc-
casional interest. The evaporative residue will generally increase with fruit ma-
turity as well as with storage time of the fruit before processing. The following
procedure can be used.
Procedure
1. Using the bunsen burner or steam bath, heat the evaporating dish; then
dry and cool it in a desiccator for about 30 minutes.
98 CITRUS JUICE CHARACTERISTICS
Ultraviolet Absorption
Determination of UV Absorption
Procedure
1. Add 0.250 g of oil to the 100 ml volumetric flask, fill it to the mark with
isopropyl alcohol, and mix the solution.
CITRUS OILS. AROMAS. AND ESSENCES 99
2. Rinse and fill the 1 cm cell of the spectrophotometer, and measure the
absorbance between 260 nm and 400 nm at intervals of at least 5 nm to
within 12 nm of the peak. Then read the absorbance at intervals of at least
3 nm three times, and then at 1 nm intervals until 5 nm beyond the peak
maximum. Thereafter absorbance readings should be made at least every
10 nm.
3. Draw a tangent to the areas of minimum absorbance. Drop a vertical line
from the main peak maximum to the tangent line, and determine the dif-
ference in absorbance that this CD line represents (absorbance at C at the
peak minus absorbance at D or the tangent line).
4. If a sample is used that is not exactly 0.250 g, a weighted ratio can be
applied to the CD value. For example, if a 0.255 g sample is used, you
should make the following calculation:
QUESTIONS
1. What imparts the majority of the flavor to citrus juices next to sugars
and acids?
2. What is the major constituent of citrus oils and at about what percent?
3. What compounds characterize grapefruit oil and tangerine oil?
4. What terpene (besides d-limonene) is found in lemon and lime oils?
5. What contribution does d-limonene make to citrus flavor?
6. What four compounds have been used to duplicate up to 87 % of fresh
orange juice flavor?
7. What are the basic principles behind the Scott method of oil analysis?
8. Why is the methyl orange indicator effective in the Scott method?
9. Why is a bromide-bromate salt solution used in the Scott method rather
than a straight bromine solution?
10. Can any condenser be used in the Scott method? Why?
11. How do you determine the end of the distillation in the Scott method,
and why?
12. What is the purpose of the HCI in the Scott method?
13. What are the industrial ranges of oil in commercial orange juices?
14. What is the difference between citrus oils, aromas, and folded oils?
15. What is "winterizing," and why is it performed?
16. Is it possible to directly distill citrus oils out of aqueous solutions?
17. What declaration must appear on citrus juice labels if citrus oils or aro-
mas are added?
18. What is the purpose of determining the optical rotation, refractive index,
and specific gravity of citrus oils?
100 CITRUS JUICE CHARACTERISTICS
19. What are the purposes of measuring the evaporative residue, ultraviolet
absorption, and aldehydes in citrus oils?
20. What is the best way to determine the quality of citrus oils and aromas?
PROBLEMS
1. If you titrate the distillate of several juice samples using the Scott method, what are
the % oils if the following quantities of the bromide-bromate solution are used?
ml bromide-bromate solution
1.67
2.72
3.11
4.76
6.01
2. What is the average oil in a blend consisting of the following blend components?
3. Suppose that you wanted to increase the oil level in the above blend to 0.020% oil.
How much oil would you need to add?
4. Suppose that in the above blend you want to add 2.5 ml of oil flavor enhancer per
gallon of concentrate. How much would you want to add?
5. In the above blend, suppose that you wanted to add the flavor enhancer determined
in the previous problem but not to exceed 0.020% in the final blend according to
specification. What average oil level must the blend have before the addition of the
oil?
6. In comparing the answers in problems 2 and 5, we see that one of the blend com-
ponents must be changed in order to meet the 0.020% oil specification and allow
for the flavor enhancer. Suppose we could replace blend component 3 with the same
amount of 64.9°Brix concentrate with an oil level of 0.003 %. Would this solve the
problem?
7. Suppose that in the above blend you eliminate lot 5 and increase the volume of lot
2 to 3000 gallons. Would this solve the problem stated in problem 6?
8. What would be the oil extraction efficiency in the manufacture of cold pressed oils
if you titrated 4.40 ml for a heavy discharge from the desludger sample, using the
Scott method and the procedure outlined in the chapter, and 25.52 ml for a sample
from the inbound oil slurry? Would this efficiency be acceptable according to in-
formation in the chapter?
CITRUS OILS, AROMAS, AND ESSENCES 101
9. What % oil would there be in the oil emulsion discharge from the de-
sludger in a cold pressed orange oil operation if 1.984 g of oil were analyzed with
18.46 ml titrated, using the Scott method according to the procedures explained in
the chapter? According to information in the chapter, would this be an acceptable
level of efficiency?
10. In the aldehyde determination for lemon oil, suppose you weighed 4.568 g of the
oil and titrated it with 6.16 ml of O.1562N NaOH according to the procedure in the
chapter. What would the % aldehyde be, and would this be good or poor quality
oil?
REFERENCES
Ahmed, E. M., Dennison, R. A., and Shaw, P. E. 1978. Effect of selected oil and essence volatile
components on flavor quality of pumpout orange juice, 1. Agric. Food Chern., 26, 368-372.
Carter, B. 1981. Private communication. Ventura Coastal Corp.
Food Chemicals Codex, 1981. National Academy Press, Washington, D.C., 140, 168, 169, 170,
172,209,210,319.
Johnson, J. D. and Vora, J. D. 1983. Natural citrus essences, Food Tech., 12,92-93.
1953.1. of the AOAC, 119.
Kebler, 1921. 1. of the A.O.A.C., 4,474.
Kesterson, J. W., Hendrickson, R., and Braddock, R. J. 1971. Florida Citrus Oils, Bulletin 749
(technical). University of Florida, Gainesville, Fla., 24-27, 114-127.
Kirsten, 1955.1. of the A.O.A.C., 38, 738.
Petrus, D. R., Dougherty, M. H., and Wolford, R. W. 1970. A quantitative total aldehydes test
useful in evaluating and blending citrus essences and concentrated citrus products, 1. Agric. Food
Chem., 18,908-910.
Shaw, P. E. 1977. Essential oils. In Citrus Science and Technology Vol. I, Steven Nagy, Philip
Shaw, and Matthew Veldhuis, eds. The AVI Publishing Company, Inc., Westport, Conn., 430-
435.
Scott, W. C. and Veldhuis, M. K. 1966. Rapid estimation of recoverable oil in citrus juices by
bromate titration. J. of the A. O.A. c., 49, 628-633.
The United States Pharmacopeia 19th Revision, 1975. United States Pharmacopeial Convention,
Inc., Rockville, Md., 560, 563.
Chapter 7
Most juices, such as apple, grape, and berry juices, are preferred by consumers
in a filtered and clarified form. However, citrus juices are preferred in a pulpy
and opaque form. Most of the opaque nature of citrus juices is attributed to the
colloidal cloud material, which will be described in Chapter 8. The solid par-
ticles of the fruit that will eventually settle out, which are primarily juice sac
and membrane material, impart a turbidity or mouth-feel that characterizes cit-
rus juices as well as giving a natural appearance. This citrus pulp can be divided
into two main groups-sinking, or spindown, pulp and floating pulp.
SINKING PULP
102
affect the results. In order to see how these parameters affect a particular juice,
it is recommended that tests be done that vary the speed, duration, and temper-
ature used in the procedure.
• Lab centrifuge.
• Tachometer.
• Two 50 ml conical centrifuge tubes. (Clear plastic tubes are the safest and
easiest to use.)
• Temperature bath and thermometer.
Procedure
1. Using the temperature bath and thermometer, bring the 11.8 °Brix juice
sample to 78°F (26°C).
2. Pour the temperature-adjusted juice into a clean and dry 50 ml centrifuge
tube with mixing to the 50 ml mark.
3. Put the tube into the centrifuge, along with another tube filled with another
sample or water to 50 ml on the opposite side to balance the load. More
than two samples can be analyzed at once, depending on the capacity of
the centrifuge. However, samples should be properly balanced in the cen-
trifuge. Place the tubes so that the graduations are facing the direction of
rotation. This will allow an average reading of the pulp directly on the
graduations if the surface of the pulp is uneven.
4. Using the tachometer, bring the rpm to 1500 for a centrifuge measuring
II! inches from the bottom of one centrifuge tube to the bottom of the
other on the opposite side when both tubes are extended in the horizontal
position. For centrifuges with diameters that are different from II! inches,
the following equation can be used to determine the rpm needed to exert
the same centrifugal force on the sample:
6. After the centrifuge comes to a complete stop, read the pulp level on the
graduation on the centrifuge tube halfway between the highest and lowest
levels of the separated pulp sediment. If the graduations are hard to see,
use a felt-tip pen to color the raised portions of the graduations.
7. The % pulp is found from:
The USDA grade standards can be found in Table 7-1. The industrial range
for orange juices is generally 8 to 12 % pulp. Grapefruit juice usually has about
2 % less pulp than orange juices have.
Pulp levels can be controlled to some extent through control of the finishing
pressure used just after extraction. Higher finisher pressures will generally re-
sult in higher pulp levels. Finishing pressures can be monitored by % pulp
measurements. If the % pulp is consistently running high, the pressure can be
lowered, and vice versa. The chapter on statistics (Chapter 23) can be used to
define how many consecutive high or low pulp levels determine a nonrandom
effect. High finisher pressures often mean higher juice yields, but juice quality
may suffer, especially in the early part of the season. Excess loss of juice yields
can be estimated by examining the moisture content of the pulp expelled by the
finisher. If it is too wet, too much juice is being lost with the pulp, and the
finisher pressure should be increased, and vice versa. A more exact way to
determine the juice content in expelled pulp is to perform a quick fiber test on
it. This test is, again, somewhat empirical but less so than the spindown pulp
procedure.
Procedure
1. Weigh 200 g of well-mixed pulp into a I-liter beaker using a large serving
spoon. Add 200 g of water and stir for I minute, let the mixture sit for 3
minutes, and then stir it for another 1 minute.
CITRUS JUICE PULP 105
Table 7-1. USDA grade standards for maximum free and suspended
pulp levels in citrus juices.
Grade A Grade B
Grapefruit Juice (48FR 9/12/8321,52.1226-52.1227)
Grapefruit juice 10 15
Grapefruit juice from concentrate 10 15
Frozen concentrated grapefruit
juice (sweetened) 10 15
Grapefruit and Orange Juice (6FR 11/1172, 52.1289) 12 18
2. Pour the contents of the beaker into the shaker screen and shake it for 3
minutes.
3. Weigh the liquid from the drain pan. The grams of liquid is a measure of
the juice content of the pulp. Optimum values depend on the commercial
machinery used and the manufacturing objective. High plant juice yields
mean high pulp levels and perhaps some loss in flavor quality, and vice
versa.
Industrial ranges are classified into three categories: tight ( < 150 g liquid),
moderate (150-180 g liquid), and loose (180-200 g liquid). In California it is
difficult to get quick fiber pulp measurements of less than 180, with 160 being
the standard target value. In Florida the quick fiber usually ranges from 90 to
140 g liquid.
The physical condition of the fruit and the variety will also have an effect on
the pulp level. Fruit has been known not only to break down and go soft after
peak maturity, but to do so during certain times of the early and mid seasons
as well. This often causes sudden shifts in pulp levels. If more than one variety
is being run, the varieties will most likely be at different stages of maturity, in
addition to differing in physical composition. Therefore, it is necessary to closely
monitor and adjust finisher pressures. For all these reasons, quality control per-
sonnel need to monitor pulp levels, juice yield, and juice quality in maintaining
the ideal finisher pressure. It should be noted that if pulp washing is being done
and added back on-line in accordance with federal standards of identity for
100% juice products, juice yields may become less important because juice lost
to the pulp expelled by finishers will be recovered in the pulp wash operation.
Another consideration that is important to the pulp quality of citrus juices con-
cerns the changes that take place in the pulp during processing. The spindown
106 CITRUS JUICE CHARACTERISTICS
pulp of the juice leaving the finisher needs to be about 12 to 20% for it to be 8
to 12 % in the final 60 0 Brix concentrate. The reason for this is that the heat
applied in pasteurization and evaporation operations, as well as the chopping
action of pumps and the juice itself, induces a degree of breakdown in the pulp
particles that results in lower pulp levels after processing. Such breakdown can
be illustrated by considering the change in the pulp level in freshly extracted
juice before and after evaporation to 60 to 65°Brix. A rule of thumb is that for
each effect in the evaporator (E), the final pulp level in the concentrate (PI)
can be found from the pulp level of the freshly extracted juice (P), by using
the following equation:
If the juice were reconstituted and reevaporated, E would be twice the number
of effects used in the evaporation, and so on. Citrus juices are rarely reconsti-
tuted and reevaporated more than once, if at all. This step is sometimes taken
to remove contaminates such as potassium citrate crystals or burnt pulp parti-
cles, to produce low-pulp products, or to debitter citrus concentrates.
It should be kept in mind that the above equation gives only approximate
results. The nature of the pulp and the empirical nature of the procedure for
measuring the pulp make it difficult to predict precise results. For example, if
the pulp level of the juice coming out of the finisher is 17 %, and it is concen-
trated to 60 0 Brix using a three-effect evaporator, the expected approximate pulp
level in the final concentrate is:
If this concentrate were for some reason reconstituted to lZOBrix and reevap-
orated to the same concentration in the same evaporator, the pulp level in the
second concentrate would be:
lOW-Pulp Juices
The pulp in citrus juices sometimes presents a problem for citrus processors and
their customers. The pulpy nature of the juices requires greater space between
plate heat exchangers in pasteurizers, chillers, and evaporators, not to mention
CITRUS JUICE PULP 107
nozzles in fillers, filters, and other machinery having restricted spaces that is
commonly used to process juice. The use of natural juices in various drinks is
increasing in popularity, with efforts to increase the health appeal of many bev-
erages. Also, low-pulp juices can be evaporated to higher Brix levels because
of lower viscosities that can save on storing and shipping costs. Moreover,
many dairies and other nonjuice processors are processing juices with equip-
ment designed for milk and other nonpulp products. The plugging of such
equipment has prompted many users to seek citrus juices with lower pulp levels.
Once a piece of machinery becomes a site for pulp collection, microbial spoil-
age can easily set in, as well as pulp burning in pasteurizers, not to mention the
restriction or stoppage of product flow. The increased use of ion exchange and
adsorption resins for acid reduction and debittering also has increased the de-
mand for depulping technology.
Where pulp specifications exist, predictions of final pulp levels are necessary
to control the operation of the industrial centrifuge. By using the pulp levels
desired in the final concentrate, the pulp levels required before evaporation can
be estimated, and from these values the required pulp levels before centrifu-
gation can be determined. If freshly extracted juice were to be centrifuged and
concentrated to make low-pulp products, Equation 7-3 could be used to deter-
mine the needed P value from the PI specification. However, if reconstituted
juice were to be used, Equation 7-3 would pose problems because centrifugation
takes place between the first and the second evaporations. To estimate the cen-
trifuge pulp levels needed before evaporation based on final pulp specifications,
the following alternate form of Equation 7-3 can be used:
(7-4 )
For freshly extracted juice a factor of 1.619 for k can be used, and for re-
constituted juice a factor of 1.189 has been used for a three-effect evaporator.
It should be noted that the number of effects of the evaporator is not used in
Equation 7-4 for lack of data. Because each processing system is different, good
record keeping should help you to determine the exact factors for your system.
For example, if a final pulp specification in 60 0 Brix concentrate were 2.0 to
3.0% pulp, we would want to shoot for 2.5% pulp. Using Equation 7-4, we
would want to shoot for 4.0% pulp coming out of the centrifuge for freshly
extracted juice and 3.0% pulp for reconstituted juice. Again, flow rates can be
adjusted to target these levels. Because centrifugation is a fairly rapid process,
pulp tests are best done continuously during depulping. Part of the operation of
most centrifuges is the periodic opening of the inner bowl in order to discharge
the collected heavy layer, or pulp in this case. The length of time that the bowl
is open and the length of time between bowl openings affect not only the pulp
level of the outbound juice but the amount of juice lost with the expulsion of
108 CITRUS JUICE CHARACTERISTICS
the pulp. Optimum flush cycles should be determined in order to have no more
than 10% loss of juice soluble solids, and thereafter the pulp levels are best
controlled by adjusting the flow rates through the centrifuge. The expelled pulp
can be added back downstream, collected, or added to other products in order
to minimize the loss of juice soluble solids.
In light ofthe fact that juice pulp is heterogeneous and amorphous in nature,
difficulty in predicting the exact nature and level of pulp in citrus juices should
be expected. The control of pulp in juices has been described as more an art
than a science, and periodic deviations in expected results should be considered
normal. Depulped juice is characteristically less viscous and can be concen-
trated to higher levels than conventional juices.
where P is the final average pulp level and Vn, P n, and Sn are the volumes, pulp
levels, and the pounds of solids per gallon of the various components of the
blend. As before, if the Brix levels of all the components are similar, the S
values can be eliminated from the equation. If the final pulp level is too high
or too low, a lot with a different pulp level can be substituted, or the amount
of a particular lot can be varied to get the desired pulp level. If the first lot is
substituted, the required pulp level of the new component needed to achieve a
certain final pulp level can be calculated by using the following:
(7-6)
pulp particles that may cause problems for low-pulp users even though the
spindown pulp levels are within specifications.
As an example of pulp blending, suppose that we wanted to blend the fol-
lowing components:
By using Equation 7-5, the final average pulp of this blend can be calculated as
follows:
This would fit the 2.0 to 3.0% pulp specification mentioned in the last section.
Suppose, however, that 2.8 % pulp was the desired pulp level. In order to
meet this specification, lot 1 could be substituted for a lot with a pulp level
determined as follows by using Equation 7-6:
If, instead of exchanging lot 1 for another lot, you wanted to change the amount
of lot 1 to achieve the 2.8 % final pulp level in the blend, Equation 7-7 could
be used as follows:
used for both. A flow chart that can be used to adjust pulp levels in citrus juice
blends is shown in Fig. 7-1. An example of the use of this flow chart can be
found in the GWBASIC program found in Appendix B. A similar HP-41C pro-
gram can be found in Appendix C.
Juice sacs refer to the coarse membrane material that is screened from the juice
during juice finishing operations. Citrus juices generally go through two stages
of finishing prior to evaporation. The reason for this is to prevent coarse pulp
from sticking to heated surfaces in the evaporator with resultant burning or the
formation of brown or black flakes in the juice. Coarse pulp particles also may
clog or become entrapped in the evaporator, thus restricting flow and inducing
the same type of burning. The pulp that is removed can be washed with water
in order to leach out additional juice solids to make pulp wash juice. Instead of
being washed, the juice sacs can be frozen or dried. The latter products usually
are referred to as commercial juice sacs and can be added back to juices after
evaporation to provide floating pulp or a fresh juice appearance and mouth-feel.
Juice sac material also can be used as an extender or texturizer in other food
products.
Dried juice sacs can be made from washed pulp by using pulp expelled in pulp
wash operations as well as from pulp expelled from juice finishers. Dried juice
sacs are not usually added back to citrus juices because such actions generally
will violate standards of identity for 100% juice products. Juice sac material is
believed to be responsible for much of the air pollution in feedmill operations.
CITRUS JUICE PULP 111
no
n
Z3 = ~ ViSPGi (Pavg - Pi)
i=2
EO 7-7
EO 7-8
Z3 PI = Pf + Z3N1 SPG 1
Fig. 7-1. A flow chart that can be used to program computers to calculate the average pulp and
pulp level adjustments in citrus juice blending.
If the juice sac material were dried separately, this pollution would be reduced
considerably. Unwashed juice sacs can be dried in foam mats or freeze-dried
under a vacuum. The latter method is qualitatively best but is the costliest to
perform. The sugar content of unwashed juice sacs produces a darker cara-
melized color than that of washed juice sacs upon drying. Washed juice sacs
112 CITRUS JUICE CHARACTERISTICS
can be dried in drum dryers or direct fire dryers. The final moisture content of
the dried sacs must be below 10 %. (See Chapter 21 for moisture tests on dried
citrus peels.) A third method of drying juice sacs involves the leaching of the
juice sacs with acetone. The resulting juice sacs are lighter in color and fluffier,
looking more like individual juice cells when reconstituted with water, com-
pared to juice sacs produced by the other drying methods.
Dried juice sacs can be milled to any consistency down to fine flour. The
common types are grits, flakes, and flour. Grits are milled to a mesh of -16 to
+30. Flakes are milled to pass through a -4-mesh sieve, and flour is milled
to pass through a 100-mesh screen.
The color of dried juice sacs may fade but is not an important quality param-
eter. Juice sac composition is illustrated in Table 7-2. Antioxidants may have
to be added to prevent off flavor development in the dried sacs. Ambient tem-
perature storage for longer than one year may result in weevil infestation.
The measurement of floating pulp in juices is a little less precise than the
measurement of spindown pulp. Screening of the pulp followed by weighing of
the larger screened pulp particles can provide for an estimate of the floating
pulp level. This method measures the floating pulp only and is somewhat com-
parable to the quick fiber test mentioned previously, but requires less costly
equipment, is more rapid, and thus is better suited to routine quality control.
The procedure is outlined below.
• 20-mesh sieve.
• Laboratory balance or triple beam balance.
• I-liter volumetric or Erlenmeyer flask.
• Stirring rod or spatula.
Procedure
1. Weigh 1 liter of well-mixed 11.8 °Brix juice into a I-liter flask while
swirling it so that the juice contains a representative amount of pulp.
2. While swirling, pour the juice through the 20-mesh sieve over a sink and
shake the screen until the pulp retained by the screen "balls up" and is
free of excess juice.
3. Carefully tum the ball with the spatula or stirring rod into one mass and
then onto the balance pan and weigh it. Disregard any slight amount of
pulp that may cling to the screen.
4. Divide the weight of the recovered pulp by the weight of the juice, and
multiply by 100% to get the % floating pulp. In Florida the floating pulp
is reported by use of the following:
Pulp levels of 0.75 to 0.80% are ideal in most situations. This is equivalent to
roughly 4 gallons of juice sac material per 1000 gallons of single-strength juice.
PULP WASH
The pulp expelled from primary and secondary finishers contains about 80 %
juice. These juice solids can be recovered by washing and refinishing the pulp
several times. The resulting juice is commonly referred to as pulp wash or
water-extracted soluble orange or grapefruit solids (WESOS or WESGS). In the
first stage, or the first time that the juice pulp is washed and refinished, about
50% of the available juice from the expelled pulp can be recovered. If two
stages are used, 63 % of the available juice can be recovered, and if three stages
are used, 75 % of the available juice in the pulp can be recovered. A four-stage
system can recover up to 80% of the available juice from the pulp. More stages
than this generally do not recover enough juice to be economically feasible.
The efficiency of a pulp wash system depends on a balance of parameters in-
cluding the water-to-pulp ratio and the number of stages. Figure 7-2 shows the
results of a study performed by FMC using its equipment. As can be seen in
the figure, the ideal water-to-pulp ratio is between 1.5 and 2.0, with the number
of stages exceeding 3 to 4 contributing little to the % recovery of juice from
the pulp. If 100 lb of pulp can be produced per ton of fruit processed, a pulp
wash system can increase the juice yield by 5 to 8 lb of soluble solids/ton of
fruit. Usually 80 to 120 lb of soluble solids/ton can be obtained in normal juice
processing without pulp washing, depending on the time of year and the variety
114 CITRUS JUICE CHARACTERISTICS
R = 2.0
>-
a:
w 75
100
:
R = 1.0
>
0
0
w
a: 50
eft
R = 0.5
25
R = WATER
PULP
2 3 4 5 6 7
NUMBER OF STAGES
Fig. 7-2. The efficiency of an FMC pulp wash system as a function of the water-to-pulp ratio and
the number of stages (Ballentine and Ferguson 1989).
of fruit. This increase in yield due to pulp washing can represent a significant
increase in juice yield, especially when juice is in high demand.
Pulp wash juice generally is considered inferior to extracted juice in both
color and flavor. The quality control of pulp wash juice generally consists of
the measurement of limonin (see Chapter 10). The limonin content in pulp wash
juices is responsible for the major effect of pulp wash on citrus juice quality,
as pulp wash can contain as much as twice the limonin found in conventional
juices. Concerns over juice quality and excessive juice supplies have led to
restrictions on the use of pulp wash in 100% juice products. The federal stan-
dards of identity permit the adding of pulp wash on-line to freshly extracted
juice, but Florida law prohibits any adding of pulp wash to 100% juice prod-
ucts. In order to enforce this law, Florida requires the addition of a sodium
benzoate tracer to all pulp wash products made in the state so that detection of
their illegal use is easier. Florida also requires that drums of pulp wash be
clearly labeled and encircled by a 5-inch yellow band.
Pulp wash juice contains a greater amount of high molecular weight com-
pounds, such as pectin, which increase the viscosity of resulting concentrates.
The added viscosity can pose problems in conveyance during evaporation and
chilling, so enzymes sometimes are added during the first stage of pulp washing
to help break down the high molecular weight compounds. Thus these pectinase
or polygalactonase enzymes have a maximum amount of time to work to de-
crease the viscosity. Without enzyme treatment, concentrate levels of over
400Brix may present difficulty in the concentration of pure pulp wash juices.
An alternative enzyme treatment consists of enzyme addition to the final pulp
wash and holding the juice for about an hour at 115°F. The pulp wash may be
CITRUS JUICE PULP 115
CORE WASH
Core wash is the same as pulp wash except that it is obtained from the core
material expelled from the orifice tubes in the FMC extractor. This core mate-
rial consists of much of the rag or membrane material and seeds formerly con-
tained inside the fruit. The rag and seed material is very high in limonin, and
successive stages of washing and finishing of this material can result in a juice
extremely high in limonin (up to three to five times more than in conventional
juices, depending on the variety and the time of the season). Core wash juice
is very opaque and is used as a clouding agent in drink bases. Core washing
methods generally employ gentle washing using fewer stages than in pulp wash-
ing and gentle treatment to avoid breaking the seeds and releasing large amounts
of limonin. Commercial debittering systems have successfully debittered such
juices, as explained in Chapter lO, with other characteristics remaining similar
to those of pulp wash concentrates.
QUESTIONS
PROBLEMS
1. What speed, in rpm, should a lab centrifuge use if the distance from bottom to bottom
between two centrifuge tubes in the horizontal position in a lab centrifuge is 131
inches?
2. What is the % pulp in each of the following spindown pulp tests according to the
procedure in the text?
Ml Pulp in Tube
1.6
2.5
4.8
3.6
5.1
3. If the pulp level of freshly extracted juice is 19.3 %, what would be the expected
spindown pulp level after evaporation using a seven-effect evaporator? What would
be the expected pulp level if this same concentrate were reconstituted and reevapor-
ated a second time in the same evaporator?
4. If a customer wanted juice with a pulp level of 2.8 to 3.0%, what pulp level would
you want to shoot for coming out of the centrifuge using freshly extracted juice?
5. In blending pulp levels, what would be the final average pulp level using the follow-
ing blend components?
6. In the above blend, suppose you want to exchange lot 5 for a lot with a pulp level
that would give a final pulp level of 2.0%. What is the required pulp level for this
new blend component, assuming the same Brix and volume for lot 5?
7. If the number of drums in lot 5 in problem 5 are to be changed to meet the 2.0
specification in the previous problem rather than substituting a new lot, what is the
required number of drums of lot 5?
8. What would be the % floating pulp in 1045 g of single-strength juice that yielded
8.6 g of screened pulp?
Chapter 8
Juice Cloud
PECTIN
Even though pectin comprises a small portion of the cloud material, and al-
though juice cloud stability can be achieved in its absence, the effect of pectin
on juice cloud stability is dramatic. The term "pectin" refers to a class of high
molecular weight compounds, with molecular weights of 100,000 to 200,000,
consisting of 150 to 1500 galacturonic acid units linked together via a (1 --+ 4)
glycoside bonds with side chains of rhamnose, arabinans, galactans, xylose,
117
o
I II
OCOH ••- - - - galacturonic
acid
Q~ OH 0 Co-CH 3
~W _ _ methoxy
group
OH~~O
~II
Fig. 8-1. Segment of a pectin mole-
cule showing the characteristic galac- OH~
turonic acid and esterified methoxy
groups. OH
and fucose (Stevens 1941). Many of the carboxyl groups are esterified with
methanol to form methoxy groups, as shown in Fig. 8-1. These methoxy groups
block many reactions, including polymerization or gelation; so the degree of
esterification (D. E.) is a measure of the gelling ability of the pectin or the grade
of pectin, and ranges from 1 to 100%. The grade also may be determined ac-
cording to the percentage of methoxy groups, from 0 to a maximum of 16.32 %.
See Chapter 20 for a more detailed description of commercial pectin recovery
and characteristics.
There are two basic types of reactions that result in gelation in citrus concen-
trates or cloud loss in single-strength citrus juices. One reaction is the reversible
acid- or base-catalyzed esterification of the carboxyl groups of the galacturonic
acid components of pectins, according to:
o o
II H+ orOH- II
CH 3 + R-COH • ~ R-COCH 3 + H 20 + Heat (8-1 )
H+ or OH-
The equilibrium of this reaction depends on many factors, including pH, water
content, heat, and secondary reactions. In the presence of sugars, under the
proper conditions, polymerization can occur in citrus concentrates, resulting in
gelation. Even though the methoxy groups themselves prevent such polymer-
ization, the reverse reaction in equation 8-1 allows it to occur. Such polymer-
ization is not common in citrus concentrates.
JUICE CLOUD 119
IOH
I
o
~
f:':\
OH
Q
II OH
OCO a ~ -08°~ OH
OH 0
II
~ -oc~
co-
OH~O~ 8
OOH
OH 0 OH
II
-oc O~
OH
O co - f:':\c
2+
OH~~ oII OH
OH 0 0
O ~O - f:':\c
2+ -oc I
OH~O\l ~ 0
Fig. 8-2. Calcium pectate structure
~OH
responsible for most gelation and cloud
loss in citrus juices.
A more common form of gelation occurs when there are sufficient gal act-
uronic acid groups to form pectates, which link together via divalent cations
such as calcium, as shown in Fig. 8-2. This can occur only if there are enough
deesterified carboxyl groups available. In freshly extracted citrus juices, the
D.E. values are usually over 50%. Levels below about 20% may be sufficient
to cause the formation of calcium pectates, which in tum results in gelation in
concentrates and cloud loss in single-strength juices. This precipitation in sin-
gle-strength juices seems to entrap or affect other cloud components as well,
resulting in total cloud loss. Also, the length of the pectic chain or the molecular
weight of the pectin affects the cloud stability. Juices with low molecular weight
pectins generally will not gel. If too much heat is used during lemon and lime
juice processing, the acid hydrolysis described in the reverse reaction of Equa-
tion 8-1 will occur (because these juices contain more acid than other citrus
juices), which will result in polymerization and/or calcium pectate formation
or gelation.
PECTINASE ENZYMES
Another type of deesterification that can lead to cloud loss and/or gelation
through calcium pectate formation results from the pectinase enzymes that nat-
urally occur in citrus juices. These enzymes, also referred to as pectin esterases
or pectin methylesterases, are the primary cause of gelation and cloud loss in
commercial juices. In order to prevent the removal of the methoxy groups by
this enzyme, heat treatment is used to deactivate or destroy the enzyme. In most
120 CITRUS JUICE CHARACTERISTICS
o 0
II ~;i~~~e II (8-2)
R-COCH 3 + H 20--R-COH + CH3 0H
acid
The rate of this reaction can be expressed in the form of a first-order rate law:
where the rate R is expressed in terms of the acid concentration and the rate
constant k. The concentrations of other species, assuming a negligible effect on
the reaction rate, are incorporated in the rate constant. This equation suggests
that a measurement of the change of acid concentration or pH can be used as
an estimate of the reaction rate or the enzymatic activity. The absolute value of
R usually is expressed in standard scientific units. However, the enzymatic ac-
tivity, a relative value, is of interest here; so convenient units can be used to
calculate a final reaction rate, which can be expressed in arbitrary units estab-
lished by standard industrial methodology.
The change in pH of a citrus juice sample is difficult to detect with the small
amount of pectin naturally present in the juice. In order to speed up the reaction
JUICE CLOUD 121
PEU Test
• pH meter
• 10 ml volumetric pipette
• 100 ml beaker
• 250 ml beaker
• 50 ml graduated cylinder
• Temperature bath
• Two disposable Pasteur pipettes
• Graduated 1 ml pipette
• Stopwatch or timer
• Magnetic stirrer
• 1 % pectin-salt solution made by mixing 10 g of pectin and 15.3 g NaCI
in a one-liter container and filling it with distilled water. (Heat may be
needed to get the pectin into solution.)
• 2.0N NaOH (20 g NaOH in 250 ml distilled water) protected from at-
mospheric carbon dioxide and moisture. (See section on acid titrations in
Chapter 3.)
• 0.05N NaOH (32.0 m1 of 0.1562N NaOH mixed with 100 ml distilled
water) protected in the same way as the above basic solution.
122 CITRlJS JUICE CHARACTERISTICS
Procedure
Thermome ter ~
I
250 ml beaker -+ - -
'"
L~ :(f)" :6':
pH meter
- ...
~
~ pH probe
100 ml beaker
water -e:.
F" I
pectin/juice sample
, I ,
where the minutes refers to the time it took to regain the pH recorded in
step 7. For example:
Most juices processed in modem evaporators and pasteurizers have PEU val-
ues of 1 x 10- 6 to 1 X 10- 4 . Those with levels much higher than this are
susceptible to gelation and/or cloud loss, depending on pH, temperature, and
other related conditions. In routine operations, rather than having a maximum
allowable PEU level, it may be more convenient to have a minimum time in-
terval for the pH recorded in step 7 to be regained, such as 2.5 minutes. If the
recorded pH is regained in 2.5 minutes or less, the cloud should be considered
unstable, and further heat treatment should be applied. In the above process it
should be noted that with stirring of the sample during the timing of the pH
change, the incorporation of atmospheric CO 2 into the sample will result in
some drop in pH. However, this should not prevent the determination of a stable
or an unstable cloud test. It should be kept in mind, however, that even with
no enzyme activity the pH will drop because of this CO 2 effect.
It is possible to have a gelled or clarified citrus juice with low enzyme activity
because deesterification can take place prior to heat treatment. Hence, one
should be careful not to allow excess time between extraction and heat treat-
ment. Sometimes when a juice concentrate gels, measurement of the enzyme
activity is of little value because deesterification could have taken place prior
to heat treatment. In such a situation, direct measurement of gelation is in order.
For frozen concentrated orange juice (FCOJ) in 6 oz cans, the following guide
can be used in grading the degree of gel when it is poured out onto a tray at
room temperature:
VISCOSITY
Cloud and pulp material contribute strongly to the viscosity of citrus juices,
especially at high Brix levels. The cold temperatures required for storage and
124 CITRUS JUICE CHARACTERISTICS
QUESTIONS
1. What is the source of most of the cloud material found in citrus cloud?
2. What is the relationship between cloud material and pulp?
3. What is the main cause of cloud loss or gelation in citrus products?
4. What role do pectinase enzymes play in cloud stability, and how can ge-
lation be prevented commercially?
5. Why is excess pectin added to juice samples in the PEU test?
PROBLEMS
1. Suppose that you open a drum of concentrate obtained from another processing plant
and notice a high viscosity. You take a sample and discover that the Brix is not
excessively high, but gel lumps are found. What is the gel number? Does this mean
that the concentrate must have a high PEU value? What can be done to correct the
situation?
2. What would be the PEU value, according to the procedure in this chapter, for a
sample that took 2 minutes and 23 seconds to return to the original pH? Would this
be considered cloud-stable juice?
3. What would be the PEU value of a juice sample that took 6 minutes and 10 seconds
to return to the original pH? Is this juice cloud stable?
4. Suppose you used 1.0 ml of 0.01562N NaOH in the procedure for 10 ml of sample,
and it took 28 minutes and 53 seconds to return to the original pH. What would be
the PEU value, and would this be considered stable juice cloud?
JUICE CLOUD 125
5. What would be the PEU value per gram of soluble solids per second using 0.5 ml of
0.0312NNaOH for 15 ml of sample (12.0oBrix) that took 25 minutes and 16 seconds
to return to the original pH? Would this be considered stable juice cloud?
REFERENCES
Baker, R. A. and Bruemer, J. H. 1971. Proc. Florida St. Hort. Soc., 84, 197.
Bennett, R. D. 1987. From presentation to the Citrus Products Technical Committee at the USDA
Fruit and Vegetable Laboratory, Pasadena, Calif., March 13, 1987.
Berk, Z. 1964. Food Tech., 18, 1811.
Rouse, A. H., Atkins, C. D., and Huggart, R. L. 1954. Effect of pulp quantity on chemical and
physical properties of citrus juices and concentrates, Food Tech., 8, 431-435.
Stevens, J. W. 1941. U.S. Patent 2,267,050, December 23, 1941.
Chapter 9
The natural bright color of citrus juices has long been regarded as one of their
major qualitative advantages over other food products. Their carotenoidal pig-
mentation has been associated with the color of the sun, producing a bright and
cheerful effect that complements the sweet and tart flavors and pleasant aromas
of the juices. The yellow-to-orange-to-pink-to-red colors have been used exten-
sively in marketing through transparent packaging and the use of sliced fruit in
advertising. The USDA holds color equal to flavor in its quality scoring system,
an indication of the importance of color in the quality of commercial citrus
products.
The main carotenoids responsible for the orange color of orange and tanger-
ine juice are a-carotene, /3-carotene, zeta-antheraxanthin (yellowish), violax-
anthin (yellowish), /3-citraurin (reddish orange), and /3-cryptoxanthin (orange)
(Stewart 1980), with the red or pink color of the pigmented grapefruit juice
varieties due to the presence oflycopene (Khan and MacKinney 1953). The red
color in blood oranges and related varieties is due to the presence of anthocy-
anins (Chandler 1958).
In dry cooler Mediterranean climates, as in California, fruit pigmentation is
well developed. In hotter and/or more humid areas, such as Florida, the color-
ation appears more dilute. For this reason the measurement of juice color plays
a greater role in juice processing in Florida than it does in California.
Color enhancement can occur only through blending with juices of higher
coloration. The federal code permits the blending of up to 10% of tangerine
juice (Citrus reticulata) and its hybrids with orange juice while still allowing
the product to be called 100% orange juice, in an effort to enhance juices with
weak color.
Off colors due to Maillard oxidation (as a result of scorching or excessive
heat applied to concentrates over too long a time) detract from the color quality
of the juice and can result in lower color scores. Such browning reactions dull
the natural citrus color. However, off flavors from these reactions develop be-
fore the visible effect occurs and further render the juice inferior. Also, white
126
hesperidin flakes or brown or black flakes from evaporator bum are readily
visible in citrus juices and mar their appearance. See Chapter 18.
The USDA has established a standard for the grading of the colors of orange
juice using six plastic color tubes, 011 (lightest) to 016 (darkest). One-inch-
diameter plastic tubes matching these color standards can be obtained from a
licensed manufacturer (such as Fiper/Magnuson, P.O. Box 11427, Reno, Nev.
89510, 1-800-648-4737) or by contacting:
Procedure
Gray Background
Fig. 9-1. Proper rack arrangement for direct comparison of orange juice color with USDA stan-
dard color tubes.
3. Place the sample tube between the standard color tubes that are just lighter
than and darker than the sample tube .
4. Use Table 9-1 to determine the USDA color score. Table 9-2 gives the
corresponding USDA grade standards for citrus juices.
TRISTIMULUS TEST
The standard method of color scoring in the food industry recommended by the
International Committee of Illumination is based on the "standard observer"
or simulated standard eye and consists of three primary colors or filters, referred
to as X (amber), Y (green), and Z (blue). Table 9-3 correlates these parameters
with the wavelength of visible light usually associated with color.
In 1952 Hunter came up with a scheme that measured redness (a), yellow-
COLOR OF CITRUS JUICES 129
Table 9-1. USDA color scores from comparison with plastic USDA
color tube standards (C juice color). =
Frozen cone. Pasteurized Canned orange
orange juice, orange juice and juice and cone.
reduced acid, and orange juice from orange juice for Dehydrated
canned conc. concentrate manufacturing orange
orange juice juice
C> OJ! 40 40 40 40
012 < C < OJ! 40 40 40 40
C = 012 40 40 40 40
013 < C < 012 39 39 39 39
C = 013 39 39 39 39
014 < C < 013 38 38 38 38
C = or
slightly > 014 37 37 38 37
OJ5 < C < OJ4 36 36 37 36
C = OJ5 36* 36 37 36
016 < C < OJ5 35 36* 36 35
C =016 34 35 36* 34*
C < OJ6 33 or less 34 or less 35 or less 33 or less
*Limits for Grade A.
ness (b), and lightness (L), which can be depicted as shown in Fig. 9-2 (Hunter
1958). These Hunter parameters can be related to the tristimulus parameters
according to (Kramer and Twigg 1970):
The Hunter citrus colorimeter further simplifies the color measurement by mea-
suring the citrus redness (CR) and the citrus yellowness (CY), which are related
to the tristimulus parameters as follows:
The Florida Department of Citrus adopted the Hunter D45 colorimeter as the
official method of color measurement, using the following procedure.
130 CITRUS JUICE CHARACTERISTICS
Table 9-2. USDA color grades for citrus juices. (No color standards
exist for concentrated grapefruit juice for manufacturing, dehydrated
grapefruit juice. tangerine juice or its hybrids, or any lemon or lime
juices.)
Grade A Grade B
'> OJ6.
'*Means color representative of juice from mature. well-ripened grapefruit; the juice may show fading and lack
luster.
"*Means juice may be slightly. but not materially, affected by scorching, oxidation, or caramelization.
L
100-white
yellow
b red
blue
-30 o 30
a
0- black
Fig. 9-2. Illustration of Hunter's color scheme based on the parameters a, b, and L.
Procedure
VISUAL
50
COLOR
POINTS
OJ2/40/~
45 OJ3/39 /~ ,-40.S
~'
/
Q __ / 38 / • 1 - 39.5
w 40 / I
a:
en
~
a: 35
OJ4/37
/
/~ /
,{38.S
t: /
0
OJS/36'~
/ /
,-37.S
30 / HCC
/ !. 36.5
, /
EQUIVALENT
25
~ " COLOR SCORE
70 85 90 95
Fig. 9-3. Nomograph for converting CR and CY values to USDA color scores.
134 CITRUS JUICE CHARACTERISTICS
Table 9-4 can be used to convert these color numbers into USDA color scores.
There are no color grade standards for grapefruit juices or other citrus juices.
The Texas Citrus Exchange (U .S. Expressway 83 at Mayberry, Mission, Tex.
78572) has established a grapefruit color scoring system that may be of value
to other grapefruit juice processors. Pigmented citrus juices containing lycopene
(pink grapefruit) and anthocyanins (blood oranges) have a tendency to undergo
a color change during processing to a brownish or dull color. This color change
generally renders the juices inferior as processed products even through pig-
mented citrus fruits are said to be the most delicious. Decolorization of these
pigmented juices is the subject of ongoing research.
COLOR OF CITRUS JUICES 135
QUESTIONS
1. What main carotenoids are responsible for the colors observed in citrus
juices?
2. What is the difference between California and Florida orange juice color,
and why?
3. How much tangerine juice can be added to orange juice while it still may
be called 100% orange juice, and why is this done?
4. What processing problems can occur to mar or degrade the color appear-
ance of citrus juices?
5. What colorimeters are approved for measurement of the color of citrus
juices?
PROBLEMS
1. What would be the L, a, and b Hunter color parameters for orange juice reflecting
average daylight at a wavelength of 620 nm, which represents pure orange color?
2. What would be the CR and CY values for the light in problem 1, and how does this
compare to CR and CY values found from orange juice measurements?
3. What would be the USDA color score for FCOJ with a CR value of 34 and a CY
value of 78? What would be the USDA color score for COJFM with the same CR
and CY values?
4. What USDA score and grade would FCOJ have if it were slightly lighter in color
than an OJ5 USDA standard color tube?
5. What would be the color number and corresponding USDA color score and grade for
concentrated orange juice for manufacturing using the Minolta colorimeter if the
tristimulus values were X = 555, Y = 321, and Z = 148?
REFERENCES
Chandler, B. V. 1958. Anthocyanins of blood oranges, Nature, 182, 993.
Hunter, R. S. 1958. Photoelectric color difference meter, 1. Opt. Soc. Am., 48(12), 985-995.
Khan, M. U. D. and MacKinney. 1953. Carotenoids of grapefruit, Plant Physiol., 28, 550-552.
Kramer, A. and Twigg, B. A. 1970. Quality Control for the Food Industry Vol. I. The AVI Pub-
lishing Company, Inc., Westport, Conn., 28, 32.
Stewart, 1. 1980. Color as related to quality in citrus. In Citrus Nutrition and Quality, S. Nagy
and J. A. Attaway, eds. The American Chemical Society, Washington, D.C., 129-150.
Chapter 10
NARINGIN
One class of compounds associated with citrus juices is the flavonoids, which
have the following characteristic molecular structure:
136
R-O OH
OH 0
These flavonoids are differentiated from each other by the carbohydrate linkup,
R. In sweet oranges (Citrus sinensis (L.) Osbeck), as well as in most other
citrus species, the predominant flavonoid is the tasteless and odorless hesperi-
din. In grapefruit varieties (Citrus paradisi Macf.) the primary flavonoid is
naringin, a bitter principle. As with other flavonoids, naringin is primarily found
in the membranes and albedo of the fruit and contributes to the bitterness of
fresh fruit as well as to the juice. The taste threshold of naringin is about 50
ppm even though levels of about 500 ppm are considered ideal, and levels of
over 900 ppm can occur.
The naringin content can be measured in several ways (Ting and Rouseff
1986). However, the spectrophotometric Davis test is the method most com-
monly used. The Davis test involves the reaction of naringin, or any flavonoid,
with diethylene glycol in dilute alkali to the corresponding chalcone, which is
yellow in color. The reaction is as follows:
CH,oH CH,oH
~ ~~O
0 OH +HOCCOCCOHNaOH.I~ ~~OH
HO~ DEG H~
~ I~? OOH
H~
~~
OHO H~OHO
~~
NARINGIN CHALCONE
(yellow)
• 4N sodium hydroxide.
• Graph paper and/or calculator.
• 200 mg naringin for standards in distilled water.
• 50 ml graduated cylinder.
Procedure
1. Centrifuge the juice sample, using the procedure for determining the %
pulp.
2. Add 10 ml of 90% diethylene glycol to a test tube along with 0.1 ml of
the centrifuged juice serum. Add 0.1 ml of the NaOH solution, mix the
solution, and let it stand for 10 minutes.
3. Measure the sample's absorbance on the spectrophotometer at 420 nm,
comparing it to a distilled water blank. It may be easier to measure the %
transmittance and convert it to absorbance by using:
Absorbance = kC (10-2 )
5. The standard graph or the factor k can be found from the following cali-
bration or standardization.
Calibration
• Isocratic or single pump HPLC system with UV detector and chart re-
corder. Gradient or multipump systems are useful when one is analyzing
for more than one flavonoid at one time. Integrators or computers linked
to the HPLC are helpful but not necessary.
• 20: 80 acetonitrile-water mobile phase.
• Lab centrifuge.
• Syringe filter (2 I-/m).
• 50 1-/1 HPLC syringe.
• Same naringin standards used in the Davis test.
Procedure
There are no USDA standards for naringin content in citrus juices. Naringin
bitterness is accounted for in flavor scores. However, Florida law (Official Rules
Affecting the Florida Citrus Industry 1975) requires grade A grapefruit juice to
have less than 600 ppm naringin and grade B juice to have less than 750 ppm
naringin.
LIMON IN
Limonin, which is the principal bitter component in most citrus juices, includ-
ing grapefruit juice, is a product of a series of reactions that originate in the
trunk of citrus trees (Hasegawa et al. 1986). These reactions generate a class
of compounds known as limonoids, which are similar to each other in chemical
structure and characteristics. The limonoid nomilin is formed in the trunk of the
tree and translocated to the fruit, where it is converted into several limonoids
including the tasteless limoninoate A-ring lactone, the open ring structure of
limonin. In most citrus varieties, this limonin precursor concentrates in the seeds
within mature citrus fruit. However, in some orange varieties, such as navel
oranges (Citrus sinensis (L.) Osbeck), the A-ring lactone remains in the neutral
environment of the juice cell cytoplasm or membrane. Upon the rupture of this
membrane during juice processing, the A-ring lactone encounters the acid en-
vironment of the juice, which gradually catalyzes the closing of the ring to form
limonin according to the following reversible reaction:
~o ~O
OH
COO-
The rate of this reaction is primarily heat-dependent, with some effect due to
the juice pH. At cold temperatures viscosity can also affect this reaction in high-
Brix concentrates. The rate of this reaction is slow enough that it does not affect
the quality of fresh fruit, which generally is consumed soon after it is cut.
However, navel juice that is stored overnight in a refrigerator will tum bitter if
it is from early or midseason fruit. Even though navel juice is considered substan-
dard because of this bitterness problem, the navel orange is excellent as a fresh
fruit. It has an early growing season, outstanding fresh flavor, and a low juice
content that makes it easy to eat as a fresh fruit. Because more money can be
made in the fresh markets than with processed products (up to 10 times as
much), both fruit variety and production are dominated by the fresh markets in
areas such as California where navel oranges can be grown of satisfactory qual-
ity. Such fresh market domination has left juice processors in such areas with
the dilemma of either selling the resulting bitter juice at a lower price or turning
to commercial debittering techniques.
Another problem associated with the delayed development of limonin has to
do with premature testing for limonin before development is complete. The rate
of limonin formation is affected primarily by acidity and heat, as mentioned
previously. Also, it has been shown that limonin solubilities range from 1 to
18 ppm in aqueous solutions containing 0 to 10% sucrose (Chandler 1971).
Pectin and other complex components of citrus juices increase this solubility,
but it is clear that high limonin levels (up to 30 ppm) can easily be affected by
the solubility equilibria of limonin in citrus juices. It also is safe to assume that
upon juice extraction not all the limonin in the juice cell membranes is leached
out into the juice at one time. The leaching has its own rate, which can affect
detectable limonin amounts during storage. Also, it is possible that continued
biochemical reactions may interfere with limonin levels, including formation
from precursors, limonin metabolism, and eqUilibria with aqueous-phase li-
monoid glucosides.
In the absence of a more thorough knowledge of these effects, a good rule of
thumb is that you should boil freshly extracted juice four minutes for every
expected ppm; reconstituted juices require about one minute of boiling for every
expected ppm. Overheating of juice has from time to time resulted in sudden
drops of limonin as well as Maillard abuse of the sample, and should be avoided.
Figure 10-1 shows some general behavior of the limonin levels in freshly ex-
tracted California Washington navel juice. From the figure it can be seen that
a significant amount of heat is required to completely develop limonin in freshly
extracted juices. Many observers have assumed that the heat applied during
normal juice processing is sufficient to fully develop the limonin in citrus juices.
However, it is common in California for early and midseason navel concen-
trates to develop significant amounts of limonin even after complete heat treat-
ment in evaporation and relatively long storage periods.
As mentioned previously, carbohydrates increase limonin solubility. This ef-
142 CITRUS JUICE CHARACTERISTICS
13
•
12 •
11
-
10
-
E
c. 9
C.
c: 8
c:
0
E 7
:J
6
5
4
3
2
fect is a result of hydrogen bonding. Such complexing reactions also have been
shown to affect the ability of the tongue to detect limonin. Guadagni et al.
showed that increasing the sweetness of the juice decreases one's ability to taste
the limonin bitterness (Guadagni, Maier, and Turnbaugh 1974). Organic acids
also interfere with the detection of limonin by taste, primarily through pH ef-
fects. A pH of 3.8 has been shown to be the ideal pH for suppressing limonin
detection (Tatum and Berry 1973) (Maier et al. 1977).
Several methods have been developed to measure limonin levels in citrus
juices, including spectroscopy (Wilson and Crutchfield 1968), thin layer chro-
matography (TLC) (Tatum and Berry 1973), gas chromatography (GC) (Kruger
and Colter 1972), radio immunoassay (RIA) (Weiler and Mansell 1980), en-
zyme-linked immunoassay (EIA) (Jourdon et al. 1984), and high performance
liquid chromatography (HPLC) (Rouseff and Fisher 1980). The two most widely
used in commercial plants are EIA and HPLC; the EIA method has the advan-
tage of being able to analyze several samples simultaneously and thus is faster,
whereas the HPLC has the advantage of being more accurate and reliable.
BITTERNESS IN CITRUS JUICES 143
Enzyme-linked Immunoassay
bility conditions, which can result in significant errors. The lower the Brix is,
the less soluble the limonin; and once the limonin crystallizes, it cannot be
detected by the Idetek method. In order to avoid these solubility problems, all
limonin analyses should be done with samples of at least 11.8°Brix. Higher-
Brix samples also upset the solubility equilibria and should be avoided as well.
When using standard solutions of limonin for calibration, one should check the
effect of the organic solvent used in the standard on the procedure. This can be
done by adding similar amounts of the pure solvent without limonin to a juice
sample and observing the calibration curve followed by using the standard with
the limonin. Generally, the more solvent that is used, the more distorted the
calibration will become. It is also recommended that when one is calibrating
the Idetek system, a standard should be added to a juice sample before analysis
rather than used directly. Instructions for the use of the Idetek kit are included
with it.
HPLC has undergone dramatic development in the past several decades, having
become a major analytical tool in the food industry. It is possible to vary col-
umns, detectors, and mobile phase, and to make slight modification of the pro-
cedures, so that few compounds exist that cannot be detected by this method.
HPLC analysis can be divided into four main sections: sample preparation,
compound separation, detection, and calculation or display. Because orange
juice contains many insoluble and colloidal macromolecules and much partic-
ulate matter, sample preparation usually is mandatory to prevent column plug-
ging. This preparation can range from simple filtering to elaborate extractions
and preliminary separations.
Limonin is readily soluble in organic solvents, so the use of such solvents is
common in preliminary separations. The elimination of aqueous material sim-
plifies the HPLC procedure and provides for a cleaner chromatograph, void of
extraneous interfering peaks. Three organic solvents have a high solubilizing
power for limonin-acetone, acetonitrile, and chloroform; and chloroform is
the solvent of choice for several reasons. First, its immiscibility with aqueous
solutions makes it a strong extracter, whereas acetone and acetonitrile dissolve
in juice samples and are much weaker in pulling the limonin from the aqueous
solution. Of the three, chloroform is the most volatile, a property that facilitates
vacuum evaporation when chloroform extracts are concentrated to raise the li-
monin level so that it is high enough to detect. Also, acetonitrile and acetone
sometimes extract other compounds that interfere with limonin peaks. There are
three different ways in which the limonin can be extracted from the juice: one
way is to use a simple chloroform extraction; another is to use special diato-
maceous earth in a 60 ml coarse-porosity fritted disc funnel to extract the li-
BITTERNESS IN CITRUS JUICES 145
Heat
Source
Fig. 10-2. Setup for sample preparation for normal phase HPLC limonin determination.
• 15 ml screw-cap test tube and 2 /-tm filter syringe (optional for the normal
phase) or a 0.45 /-tm nylon 66 Millipore filter for the reverse phase.
• Limonin chloroform standard (25 mg /100 ml).
• 1 ml graduated pipette or 0.1 ml autopipette.
• 25 ml pipette and extra 100 ml volumetric flask for reverse phase separa-
tion.
material that may interfere with the limonin peak. If there is no interfer-
ence, this step can be ignored.
3. Without removing the stirring rod, fill the funnel with chloroform, stir it,
and vacuum-extract the chloroform into the rotary evaporator, which is
already in motion with a heat source applied. A three-way stopcock, as
shown in Fig. 10-2 should be open during the extraction and closed when
one is adding new solve At and stirring.
4. Repeat step 3 four more times while concentrating the chloroform in the
rotary evaporator. Continue concentrating it until less than 2 ml of chlo-
roform remains in the evaporation flask. The rate of evaporation can be
enhanced by stronger vacuum, greater heat, and/or the use of refrigerated
coolant in the condenser. It is recommended that one use moderate heat,
compressed air, aspirated vacuum, and tap water in the condenser. Too
much heat results in hot glassware that is difficult to handle.
5. Remove the evaporation flask, rinse it with four or five 2 ml portions of
chloroform into a 10 ml graduated cylinder, and record the volume (ml
GC). Be careful not to lose any of the limonin-rich chloroform rinses.
6. Immediately filter the rinses with a syringe filter, place the chloroform
solution in a screw-cap test tube, and seal it.
7. Rinse the 20 1-'1 syringe several times with the chloroform extract, and
inject 151-'1 of the solution into the HPLC injector using the normal phase
mobile phase described above. Start the chart recorder (10 mV 5
mm/min), and inject the sample at the same time. The pump speed should
be 1.5 ml/min.
S. The peak height (P) can be combined with the ml GC, the density of the
sample, d (g/ml), and the pounds soluble solids/gal, S, in the following
equation:
o 2 4 6 8 10
Minutes
Fig. 10-3. HPLC chromatograph of limonin and nomilin peaks using the normal phase method.
Calibration
1. Add 0.1, 0.3, 0.5, 0.8, and 1.0 ml of the standard to 10 ml each of 5
identical 11.8 °Brix juice samples for normal or reverse phase calibration,
and proceed with the analysis as explained above.
2. The pg limonin per unit of peak height h can be found as follows:
where OP is the difference in the peak height of the un spiked juice sample
compared to the spiked juice sample.
150 CITRUS JUICE CHARACTERISTICS
3. The average /1-g/h from the five standards can be used to calculate the
factor F used in Equation 10-5:
5. In order to check the calibration, the factor calculated above can be used
in Equation 10-5 to determine the expected limonin levels of each of the
standard tests. This is done by first calculating the limonin level of the
unspikedjuice sample (L), using the average Fvalue from Equations 10-6
and 10-5. Then the expected limonin levels of the spiked samples can be
calculated:
7.0 ppm limonin for Grade B (Florida Dept. of Citrus). These grade standards
apply to concentrate (not finished products) processed between December 2 and
July 31. They also apply to finished products processed from concentrate or
bulk single-strength juice between December 2 and July 31, provided that any
juice or concentrate blended with it that is processed between August 1 and
December 1 meets the permissible limits of limonin for the desired grade.
Procedure
1. Filter the juice, using the syringe filter, and spot 25 J.l.1 of filtered single-
strength or reconstituted juice (11.8°Brix), using the 50 J.l.1 syringe, about
1 cm from the bottom of the TLC plate. Multiple samples can be ana-
lyzed at one time on wider plates. Spots should be about 1 cm from the
edge of the plate and/or 2 cm apart.
2. Dry the spot with a wann air stream, respot with another 25 J.l.1 sample
of the juice over the first spot, and redry the plate.
3. Using the 10 J.l.1 syringe, spot 1, 2, 3, 4, and 5 J.l.1 of the limonin standard
solution about 1 cm from the bottom and 2 cm apart on the TLC plate.
For estimated limonin levels over 10 ppm, spot 6, 8, 10, 12, and 14 J.l.1
of the standard instead.
4. Dry the TLC plate again with wann air for at least 1 minute. Incomplete
drying will result in streaked spots.
5. Place the plate in the solvent tank containing acetone with the liquid level
about halfway between the bottom of the plate and the spot. Elute until
the liquid level travels about 3 cm, remove the plate, and dry it. Uneven
liquid levels on the plate indicate insufficient drying of the sample spot.
6. Place the plate in a paper-lined solvent tank containing the benzene so-
lution, and elute to the top of the plate.
BITTERNESS IN CITRUS JUICES 153
7. Remove the plate, dry it, and repeat the elution two more times.
8. Spray the plate with the sulfuric acid solution, using the chromist sprayer,
until it is lightly wet.
9. Incubate the plate in the oven (125 °C) for 6 minutes, and then remove
and cool it.
10. The limonin spot will appear dark in color and rectangular in shape if it
was dried properly during sample preparation.
11. The unknown sample can be compared to the standards visually by view-
ing them against a light source from the back or glass side of the TLC
plate. They also can be viewed by using an ultraviolet light source under
the plate, in which case the spots appear reddish brown.
12. The limonin level can be determined by comparison with the closest
standard and multiplying the microliters of the standard by 2.
PROCESSING CONSIDERATIONS
Even though juice bitterness is induced naturally, processing can have an im-
portant effect on it. For example, it has been shown that ethylene gas, com-
monly used to degreen early-season fruit for the fresh markets, can trigger a
mechanism that reduces limonin levels (Maier, Brewster, and Hsu 1973). It has
been suggested that leaving fruit so treated in storage bins for as long as possible
will help to decrease the bitterness. Even so, early-season navel juices are usu-
ally too bitter for this precaution to have a significant effect on final juice qual-
ity.
Because both naringin and limonin enter the juice primarily from membrane
material in the fruit, extraction and finishing pressures and the length of time
that the juice is in contact with the membrane material can affect the degree of
bitterness of the resulting juice. Such variations are small, becoming important
only when the limonin and naringin levels are reduced through natural matu-
ration to levels close to the taste thresholds. Most juice extractors can be ad-
justed to give a soft or a hard squeeze, and finishers usually can be adjusted to
give a soft or a hard finish. In either case, hard extraction or finishing usually
results in greater bitterness and greater juice yield. Therefore, in the early sea-
son, extraction and finishing pressures should be set for hard squeezing, as it
gives greater yields, and the bitterness is too high for a soft squeeze to yield
good-quality juice anyway. Later in the season, when bitter taste thresholds are
approached, the squeeze should be changed to a softer squeeze in an effort to
produce a lesser or a nonbitter juice. Lesser or nonbitter juice products usually
bring in more than enough money to offset losses of yield due to softer squeezes.
Also, immediate centrifuging of the juice to remove naringin-Iaden pulp has
been shown to reduce naringin levels in grapefruit juices by as much as 10 to
20% (Berry and Tatum 1986).
1 54 CITRUS JUICE CHARACTERISTICS
The lb / gal can be found from the Brix by using Equation 2-8 or 2-10 or sucrose
density tables. We can find the final weighted average limonin level ofthe blend
by using the following:
( 10-8)
= 9.7 ppm
This value is well over the 7 ppm taste threshold. Suppose you wanted to reduce
the number of drums of lot 2 so that the final blend would be at or below the
threshold. The number of drums of lot 2 needed for this could be calculated as
follows:
( 10-9)
Therefore, if you use I drum of lot 2 instead of 10 drums, you will have a final
weighted average limon in level in the blend of exactly 7.0 ppm.
Rather than change the number of drums of lot 2, you may wish to exchange
lot 2 for the same number of drums with a lower limonin level in order to
achieve 7.0 ppm limonin in the blend. The limonin level needed in such a lot
can be calculated as follows:
(10-10)
Thus if the 10 drums of lot 2 at 16.2 ppm were exchanged for 10 drums of a
lot with 7.9 ppm and the same Brix, the weighted average limonin level in the
final blend would be the desired 7.0 ppm.
It must be remembered that the juices from all citrus varieties contain limonin
to some degree, and it cannot be assumed that they contain no limonin even if
they come from what is considered a nonbitter juice such as Valencia. It is
interesting to note that Valencia orange juice, the most popular orange juice in
the world, is considered a nonbitter juice even though early-season California
Valencia juice contains limonin levels as high as 15 ppm. However, the acid
content of this early season juice is so high that by the time the acid levels are
blended to commercial levels, the limonin levels are blended to below taste
threshold values. Hence, limonin levels, even in traditionally nonbitter juices,
should not be treated lightly in blending.
1 56 CITRUS JUICE CHARACTERISTICS
As with other citrus juice parameters, the adjustment of limonin levels lends
itself to computer application. Limonin levels usually are not determined as
often as other juice parameters, so blending or adjusting citrus juices to a spec-
ified limonin level does not have the same import as adjusting the Brix, ratio,
oil, or pulp levels. However, where debittering is not done, the use ofthe above
equations may be warranted. Figure 10-4 shows a flow chart that can be used
to program computers to make the above calculations. Also, a sample
GWBASIC program can be found in Appendix B.
Methods to debitter citrus juices have been under investigation for many years,
including: ethylene gas treatment (Maier, Brewster, and Hsu 1973); the use of
EO 10-8 yes
Lavg = Z1/Z2
yes
EO 10-9 EO 10-10
Lavg
V1 L1
Fig. 10-4. Flow chart that can be used to program computers to adjust the limonin level in citrus
blends.
BITTERNESS IN CITRUS JUICES 157
the nonbitter flavone neodiosmin, which competes with limon in for receptor
sites on the tongue (Guadagni, Maier, and Turnbaugh 1976); the use of im-
mobilized bacteria and enzymes (Hasegawa and Maier 1983); the use of super-
critical carbon dioxide (Kimball 1987); the use of auxin sprays on trees, which
interfere with the biosynthesis of limon in in the fruit (Hasegawa et al. 1986);
and the use of adsorption and ion exchange resins (Shaw and Buslig 1986;
Maeda et al. 1984; Johnson and Chandler 1985). The last has proved to be the
only commercially feasible method (Kimball and Norman 1990a,b; Norman
and Kimball 1990). Ion exchange resins not only have the capacity to remove
limonin and naringin, but they can remove substantial quantities of organic
acids as well. Early-season bitter juice may not need to be debittered to below
the taste threshold because of the high acid content of the juice; as with the
Valencia juice mentioned above, acid blending will dilute out some of the li-
monin content. If deacidification and debittering take place simultaneously,
however, limonin levels probably will need to be reduced to below taste thresh-
old values.
When one is using absorption or ion exchange resins, saturation of the resin
eventually requires its regeneration. In commercial debittering the entrapment
of pulp particles results in pressure buildup in the column, which necessitates
regeneration before the resin becomes saturated with limonin or naringin. Pulp
and cloud clarification by means of ultrafiltration reduces the concern for col-
umn clogging, so that regeneration can be less frequent. Regeneration generally
consists of a series of backflushes of basic solutions, clear water rinses, air
scrubs, and, if acid reduction is to be minimized, a final acid treatment. Because
of the difficulty of disposing of sodium hydroxide, potassium hydroxide or am-
monium hydroxide generally is preferred. Also, sulfuric acid is considered to
be the acid of choice as far as waste disposal is concerned. If regeneration
wastes are to be disposed of through irrigation, then sodium bases, nitric acid,
and hydrochloric acid should be avoided, as they are considered detrimental to
agricultural soils.
QUESTIONS
1. When tasting juice for bitterness, what should one look for?
2. What two compounds cause citrus juice bitterness, and in what varieties
do they occur?
3. What is the taste threshold of naringin in grapefruit juice, and what is
the ideal level?
4. What are the USDA and Florida grape standards for naringin and li-
monin?
5. Why is it that limonin bitterness does not affect the fresh fruit industry
but naringin bitterness does?
6. What affects the rate of limonin development?
7. How can you be sure that the limonin in a juice sample is completely
developed?
8. What are the advantages and disadvantages of the limonin EIA, HPLC,
and TLC methods of analysis?
9. What is the commercially accepted means of debittering citrus juices,
and why has this method been successful?
10. What are some ways of reducing bitterness in citrus juices during pro-
cessing?
PROBLEMS
1. In the HPLC nonnal phase limonin detennination, finish the calibration table (p.
159), assuming that the sample is 11.8°Brix. What is the factor F?
BITTERNESS IN CITRUS JUICES 159
2. Using the factor calculated above, what is the limonin level in a 12.1 °Brix juice
sample with a peak height of 1.56 and a GC value of 9.6 ml?
3. What would be the expected limonin level after blending the following components?
Would this be considered bitter juice?
4. Supposed you wanted to make the same amount of juice in the blend in problem 5.
What would be the maximum limonin level of a 62.1 °Brix lot be if you exchanged
it with lot 1 to get a final limonin level of 7.0 ppm?
5. Suppose you had no other lot to exchange with the components in problems 3 and 4,
but you wanted to reduce the limonin level to 7.0 ppm. How many gallons of lot 1
would you need to pull from the blend?
6. What would be the expected naringin level after blending the following components?
What would be the Florida grade?
7. Suppose you wanted to make the same amount of juice, exchanging lot 2 above for
another lot of the same Brix. What would be the maximum naringin content allowed
in order to get grade A juice by Florida standards?
8. Suppose you wanted to keep the number of drums of lots 1 and 3 and use one of the
other lots to adjust the naringin content to the maximum level for grade A in the state
of Florida. How many drums of lots 2, 4, or 5 would you need?
160 CITRUS JUICE CHARACTERISTICS
REFERENCES
Belitz, H. D., Chen, H., Jugel, H., Treleano, R., Wieser, H., Gasteiger, J., and Marsili, M.
1979. Sweet and bitter compounds: Structure and taste relationship. In Food Taste Chemistry,
ACS Symposium Series 115, J. C. Boudreau, ed. The American Chemical Society, Washington,
D.C., 125.
Berry, R. E. and Tatum, J. H. 1986. Bitterness and immature flavor in grapefruit: analyses and
improvement in quality, J. Food Sci., 51,1368-1369.
Chandler, B. V. 1971. Some solubility relationships of limonin. Their importance in orange juice
bitterness, CSIRO Food Res. Q., 31, 36-40.
Chandler, B. V. and Kefford, J. F. 1966. The chemical assay of limonin, the bitter principle of
oranges,J. Sci. Food Agric. , 17,193.
Davis, W. B. 1947. Determination offlavanones in citrus fruits, Anal. Chem., 19,476-478.
Fisher, J. F. and Wheaton, T. A. 1976. A high pressure liquid chromatographic method for the
resolution and quantitation of naringin and naringenin rutinoside in grapefruit juice, J. Agric.
Food Chern., 24, 898-899.
Florida Dept. of Citrus, Official Rules Affecting the Citrus Industry, Lakeland, Fla., 20-64.03(5),
20-64.09(7) .
Guadagni, D. G., Maier, V. P., and Turnbaugh, J. G. 1976. Effect of neodiosmin on threshold
and bitterness of limonin in water and orange juice. J. Food Sci., 41, 681-684.
Guadagni, D. G., Maier, V. P., and Turnbaugh, J. G. 1974. Some factors affecting sensory thresh-
olds and relative bitterness of limonin and naringin, J. Sci. Food Agric., 25, 1199-1205.
Hasegawa, S. 1980. Private communication. Biosynthesis of limonoids in Citrus: sites and trans-
location.
Hasegawa, S. and Maier, V. P. 1983. Solutions to the limonin bitterness problem of citrus juices,
Food. Tech., 37(6), 73-77.
Hasegawa, S., Herman, Z., Orme, E. D., and Ou. P. 1986. Phytochemistry, 25, 542,1323,2783.
Hasegawa, S., Patel, M. N., and Snyder, R. C. 1982. Reduction of limonin bitterness in navel
orange juice serum with bacterial cells immobilized in acrylamide gel, J. Agric. Food. Chem.,
30, 509-511.
Johnson, R. L. and Chandler, B. V. 1985. Debittering and deacidification of fruit juices, Food
Tech. in Australia, 38, 294-297.
Jourdon, P. S., Mansell, R. L., Oliver, D. G., and Weiler, E. W. 1984. Competitive solid phase
enzyme-linked immunoassay for the quantification of limonin in citrus, Anal. Biochem., 138,
19-24.
Kimball, D. A. 1987. Debittering of citrus juices using supercritical carbon dioxide, J. Food Sci.,
52, 481-482.
Kimball, D. A. and Norman, S. I. 1990a. Changes in California navel orange juice during com-
mercial debittering, J. Food Sci., 55, 273.
Kimball, D. A. and Norman, S. I. 1990b. Processing effects during commercial debittering of
California navel orange juice, J. Agric. Food Chern., 38, 1396-1400.
Kruger, A. J. and Colter, C. E. 1972. Gas chromatographic identification oflimonin in citrus juice,
Proc. Fla. State Hort. Soc., 85,206-210.
Maeda, H., Takahashi, Y., Miyake, M., and Ifuku, Y. 1984. Studies on the quality improvement
of citrus juices and utilization of peels with ion exchange resins. 1. Removal of bitterness and
reduction of acidity in Hassaku (Citrus hassaku hort. ex. Tanaka) juice with ion exchange resins
and adsorbents, Nippon Shokuin Kogyo Gakkaishi, 31, 413-420.
Maier, V. P., Bennett, R. D., and Hasegawa, S. 1977. Limonin and other limonoids. In Citrus
Science and Technology Vol. I, S. Nagy, P. E. Shaw, and M. K. Veldhuis eds. The AVI Pub-
lishing Company, Inc., Westport, Conn., 379, 382.
BITTERNESS IN CITRUS JUICES 161
Maier, V. P., Brewster, L. C., and Hsu, A. C. 1973. Debittering citrus juices, Citrograph, 58,
403-404.
Norman, S. I. and Kimball, D. A. 1990. A commercial citrus debittering system, Proc. o/the 36th
Citrus Eng. Conf., Florida section of the ASME, Lakeland, Fla., March 29, 1990, 1-31.
Official Rules Affecting the Florida Citrus Industry, January 1, 1975, State of Florida, Dept. of
Citrus, 20-65.01 to 20-65.05.
Rouseff, R. L. and Fisher, J. F. 1980. Determination of limonin and related limonoids in citrus
juices by high performance liquid chromatography, Anal. Chern., 52,1228-1233.
Shaw, P. E. and Buslig, B. S. 1986. Selective removal of bitter compounds from grapefruit juice
and from aqueous solution with cyc\odextrin polymers and with Amberlite XAD-4, J. Agric.
Food Chern., 34, 837-840.
Shaw, P. E. and Wilson, C. W. 1984. A rapid method for determination oflimonin in citrus juices
by high performance liquid chromatography, J. Food Sci., 49, 1216-1218.
Tatum, J. H. and Berry, R. E. 1973. Method for estimating limonin content of citrus juices, J.
Food Sci., 38, 1244-1246.
Ting, S. V. and Rouseff, R. L. 1986. Citrus Fruits and Their Products. Marcel Dekker, Inc., New
York, 108-113.
Weiler, E. W. and Mansell, R. L. 1980. Radioimmunoassay of limonin using a titrated tracer, J.
Agric. Food Chem., 28, 543.
Wilson, K. W. and Crutchfield, C. A. 1968. Spectrophotometric determination of limonin in or-
ange juice, J. Agric. Food Chem., 16, 118-124.
Chapter 11
Natural foods, especially citrus products, have always been highly regarded as
excellent sources of human nutrition. Nutrition is the ability to engender growth,
and many of the components of citrus products contribute to the growth and
well-being of the human body. In order to grow, the human body requires cer-
tain dietary components, and this need has given rise to the regulation of nu-
tritional labeling by many governments. Such labeling, which provides nutri-
tional information to consumers, is of great value, especially for persons who
must monitor their nutritional intake for medical or dietary purposes. Nutri-
tional labeling also gives an aura of professionalism and quality to a product,
especially one that contains significant amounts of nutrients.
There are two basic types of nutrients. The TrUlcronutrients are those that
occur as major components of the food product, such as sugars or carbohy-
drates, salts, acids, protein, and fats. These nutrients provide energy for the
body in addition to regulating the absorption rates of various body chemicals
and water retention in the body. Proteins act as catalysts for certain life-giving
chemical reactions and as building blocks for many parts of the body. The
micronutrients are those that occur in trace amounts. They either regulate spe-
cific chemical reactions, like the proteins, or they form important organome-
tallic complexes such as the iron complexes that make up the blood.
To establish a uniform way of reporting nutritional information, laws regu-
lating labeling have been passed in various countries. In the United States, the
Food and Drug Administration (FDA) first published nutritional labeling reg-
ulations in the Federal Register in March of 1973 (U .S. Food and Drug Admin-
istration 1973). Since that time, changes in the regulations have occurred, and
they continue to occur because of constant research and new understanding
about human nutrition. The current FDA regulations can be found in the Code
of Federal Regulations (CFR), Title 21, 101.9-101.13. For natural products
where no nutrients have been added, nutritional labeling is optional. About two-
thirds of the 100% citrus products on the market do not include any nutritional
162
labeling. If nutritional labeling is used, the rules and fonnat outlined in the
federal code must be followed.
With respect to labeling, there are two classes of nutrients. Group I nutrients
are those added to the product, which must be present at least at the levels
declared on the label. Group II nutrients are those that occur naturally, which
must be present in at least 80 % of the levels declared on the label. Excess
amounts of nutrients are allowed within good manufacturing practice except that
the calorie, carbohydrate, sodium, and fat levels cannot exceed 20% of the
value declared on the label. Official compliance with declared label values is
detennined by taking 12 samples, selected from 12 separate shipping cases at
random, and analyzing them by AOAC or other suitable methods.
Even though FDA is reviewing the commerical nutritional labeling fonnat
that will most likely contain changes; the fonnat that the nutritional infonnation
must presently follow is illustrated in Table 11-1. All nutritional levels are ex-
pressed on a per-serving basis. A serving of citrus juice usually is designated
as 6 fluid single-strength ounces (177 ml) with 113 g designated for citrus
sections. The servings per container is the number of reconstituted or consum-
able servings rounded off to the nearest whole number.
REPORTING OF MACRONUTRIENTS
Calories
I g protein x 4 = 4 calories
21 g carbohydrates X 4 = 84 calories
o g fat x 9 = 0 calories
Total = 88 calories
Calorie values then are rounded off to the nearest even number up to 20 calories,
to the nearest increment of 5 from 20 to 50 calories, and to the nearest increment
of 10 from there on up. Because, in the above sample, the number of calories
is greater than 50, the calorie value is rounded off to the nearest 10, or to 90
calories, which appears on most nutrition labels of 100% orange juice.
Protein
Proteins are enzymes or biological catalysts that govern the biochemical reac-
tions occurring in the body, and provide energy like the carbohydrates. The
human body manufacturers many of the proteins it needs. However, it requires
an outside source for some of them. Some parts of the body, such as hair, are
themselves made up of proteins. Proteins are comprised of long amino acid
polymeric chains that are cross-linked in various ways. For the purposes of
nutritional labeling, the protein level is calculated by multiplying the weight
percent of total nitrogen by 6.25. The nitrogen content in citrus juices has been
reported to be 60 to 120 mg per 100 ml of single-strength juice (Ting 1967),
with higher values reported in California orange juices (Tockland 1961). The
nutritional quality of proteins should aslo be taken into consideration. This qual-
ity is dependent on the protein efficiency ratio (PER). A PER value of at least
20 % of casein is required for a food to be considered a significant source of
protein. Citrus products generally fall short of this and thus are considered to
be insignificant or poor sources of protein.
NUTRITIONAL CONTENT OF CITRUS JUICES 165
The Kjeldahl procedure is generally used to measure the total nitrogen level
in citrus juices, but it is a lengthy and involved procedure for routine quality
control work and may account for only 11 to 20% of the total juice protein
(Attaway et al. 1972). Also, at least 1 gram of protein is required for nutritional
labeling, and citrus juices just meet this requirement. Given these circum-
stances, protein analysis usually is not performed on citrus products in quality
control laboratories. When protein is reported on nutritional labels, the value
should be rounded off to the nearest gram per serving, which is usually 1 gram.
The USRDA for protein depends on the gender and age of the consumer, rang-
ing from 23 grams for children 1 to 3 years old to 56 grams for adult males 23
to 50 years old.
Carbohydrates
Rounding this off to the nearest gram gives 22 grams, which would be reported
on the nutritional label. If nutritional sweeteners are added to make sweetened
juice, they must be included in the carbohydrate level in nutritional labeling.
Fat
Fats are lipids (natural compounds that are soluble in hydrocarbons but not in
water), and are generally considered to be detrimental to health. In citrus juices
there are essentially no fats, and that can generally be declared on nutritional
labels. If fats were to exist in significant quantities, their level would be reported
to the nearest gram per serving.
Sodium and potassium labeling are optional under the present federal code.
When reported, the sodium or potassium level is given to the nearest milligram
per serving. Sodium generally is considered to be detrimental to health, whereas
potassium is considered to be advantageous. Citrus juices generally contain
insignificant amounts of sodium and can be so labeled. However, high levels
of potassium are common; levels of 200 mg to 2000 mg per serving can be
found in commercial juices, depending on the geographical location (Harding,
Winston, and Fisher 1940). Also, legally added water in reconstituted products
may be rich in potassium and/or sodium and contribute to the presence of these
nutrients. Again, as mentioned earlier, naturally occurring nutrients can legally
exist in levels of at least 80% of the declared value, with no maximum limit
for potassium. The natural sodium or potassium levels in water used to legally
reconstitute citrus products fall in this category. Because sodium levels are low
in citrus products, routine quality control of those levels is seldom performed.
Also, because potassium is plentiful in many foods, its presence in citrus prod-
ucts does not constitute an anomaly and rarely justifies routine monitoring. On
those infrequent occasions when it is required, professional laboratories, using
ion chromatography or atomic absorption, can be used.
When nutritional labeling is used, seven vitamins and minerals as well as pro-
tein must be mentioned on the label: vitamin A, ascorbic acid (vitamin C),
thiamin (vitamin B 1), riboflavin (vitamin B2 ), niacin (vitamin B4), calcium, and
NUTRITIONAL CONTENT OF CITRUS JUICES 167
Vitamin C
Vitamin C, also known as ascorbic acid, hexuronic acid, cevitamic acid, and
antiscorbutic acid, has long been associated with citrus nutrition. Limes once
were used extensively to stave off scurvy attacks on board seagoing vessels, so
much so that British sailors were labeled "Limeys." Even though other fruits
and vegetables have higher levels of vitamin C than citrus juices, few are as
attractive in color, taste, and thus popularity as citrus. A single 6-ounce serving
(177 ml) provides more than 100% of the USRDA of vitamin C; so foods
containing higher levels of the vitamin offer no real nutritional advantage over
citrus. Also, vitamin C is very stable in citrus juices and degrades very little
with storage, another nutritional advantage.
L-Ascorbic acid has the following chemical structure:
OH
H-C-H
I
H-C-OH
r C- H
I C-OH
o II
L~-OH
II
o
168 CITRUS JUICE CHARACTERISTICS
OH
H-C-H
I
H-C-OH
,C-H
I C=O
o
L C=O
C=O
L-Dehydroascorbic acid is less stable than L-ascorbic acid, but the two have
nearly the same nutritional effect. Vitamin C contributes to: iron absorption;
cold tolerance; maintenance of the adrenal cortex; antioxidizing activity; me-
tabolism oftryptophan, phenylalanine, and tyrosine; body growth; wound heal-
ing; synthesis of polysaccharides and collagen; formation of cartilage, dentine,
bones, and teeth; and capillary maintenance. Humans cannot synthesize vitamin
C, and must depend on an external source for its supply.
The vitamin C level in citrus decreases with maturity. Florida oranges go
from about 50 mg /100 ml of single-strength juice at the beginning ofthe season
to about 30 mg/IOO ml by the end of the season (Harding, Winston, and Fisher
1940). During an average marketing period, the loss of vitamin C amounts to
less than 10%, which is indicative of the stability of vitamin C in citrus juices.
Atmospheric oxygen· is responsible for most vitamin C loss during long-term
storage. Many polymeric containers readily admit oxygen, which degrades vi-
tamin C as well as contributing to the development of off colors and flavors.
Grapefruit juice generally contains slightly less vitamin C ( 45 mg /100 ml) than
orange juice (50 mg/100 ml) and lemons (60 mg/IOO mI). Tangerines and
limes contain even less vitamin C (about 30 mg/IOO ml). These vitamin C
levels represent approximate averages and indicate general relative compari-
sons. Actual vitamin C levels may vary widely, by as much as 50%.
Even though vitamin C is a major nutrient in citrus juices and is considered
to be an important quality parameter, its stability and generally consistent levels
render routine analysis unimportant. Industrially, vitamin C levels are not con-
sidered an important parameter. However, vitamin C levels can be of interest
on occasion, especially in establishing nutritional labeling; and the following
two methods of analysis can be used.
NUTRITIONAL CONTENT OF CITRUS JUICES 169
Standardization Procedure
For example:
Titration Procedure
For example:
The following HPLC procedure uses ion pairing and can be used to determine
both the L-ascorbic acid level and the dehydroascorbic acid level. Dehydroas-
corbic acid is derivatized with 1,2-phenylenediamine in order to increase its
ultraviolet absorbance.
Procedure
Nearly 75% of the vitamin C in oranges and over 80% of the vitamin C in
grapefruit can be found in the peel. The USRDA for vitamin C is 60 mg.
Thiamin
This vitamin is also known as vitamin B\, aneurin, antineuritic factor, antiber-
iberi factor, and oryzamin. The metabolically acitve form of thiamin is thiamin
Y
pyrophosphate.
NyNH2
o 0
I ~/s cc-o-r-O-r- OH
N~ ~ "';/ OH OH
~ ""C-N
+
This molecule acts as a coenzyme which participates in decarboxylation of
a-keto acids in the initial phases of the citric acid cycle. Thiamin deficiency is
responsible for the development of beriberi.
172 CITRUS JUICE CHARACTERISTICS
Orange juice has the highest level of thiamin ( 170 I-tg /6 oz serving or 11 %
of the USRDA), followed by tangerine juice (100 I-tg/6 oz serving or 7% of
the USRDA) and grapefruit juice (70 I-tg/6 oz serving or 5% of the USRDA).
Unlike vitamin C, thiamin levels increase with fruit maturity, expanding from
as low as 45 I-tg/ 100 g of single-strength juice to 90 I-tg/100 g of juice through-
out a given season. The USRDA for thiamin is 1.5 mg.
Thiamin determinations are commonly performed by using a combination of
enzymes, packed open columns, and fluorimeters (Ting and Rouseff 1986). This
procedure is generally too tedious and complex for routine quality control anal-
ysis. Because thiamin determinations are performed infrequently, professional
laboratories usually are consulted.
Calcium
Even though calcium occurs in very small amounts naturally in citrus juices,
calcium-added products have increased in popularity. Any nutrient that is added
must be declared on the nutritional label, and nutritional labeling becomes man-
datory. As mentioned earlier, most minerals are determined by using ion ex-
change or atomic absorption. This is usually best done in professional labora-
tories because of the infrequent need for such determinations and the cost of
the instrumentation. The USRDA for calcium is 1 gram. Calcium hydroxide is
easily added to citrus juices and neutralizes the acidity, producing a sweeter
juice at the same time. However, it should be remembered that such calcium
additions violate the standards of identity for 100% juice products, rendering
them "drinks" rather than "juices."
Even though there are no labeling requirements or formats in the federal code
regarding pectins and flavonoids, it should be noted that these compounds, found
abundantly in citrus products, are believed to be linked to the reduction of serum
cholesterol, a rare and important characteristic in foods. Like many natural
products, the nutritional label, as outlined in the federal code, falls far short of
completely describing the nutritional advantages of citrus products. Many such
advantages have yet to be discovered and understood. Even though nutritional
analysis plays a minor role in overall citrus quality control, it should be remem-
bered that the nutritional value of citrus products is one of their major selling
points, and quality control personnel should properly represent it as such to the
industry and to the consumer.
NUTRITIONAL CONTENT OF CITRUS JUICES 173
QUESTIONS
PROBLEMS
1. Suppose that super citrus scientist Dr. VanOhjay has discovered a new orange variety
that produced juice of a higher nutritional quality. This new juice contains 11 grams
of protein, 8 grams of carbohydrates, and 1 gram of fat per 6 ounces (177 ml). What
calorie level would be declared on the label?
2. In the indophenol titration for vitamin C, suppose that you titrated 19.64 ml in the
standardization and 11.46 ml in the sample titration. What would be the dye concen-
tration, the vitamin C level (mg/lOO ml juice), and the highest declarable % US-
RDA value on the label?
3. How many grams of carbohydrates would be declared on a label of product whose
serving size is 100 ml of single-strength juice at 12.0oBrix?
4. What would be the maximum declarable amounts on a nutritional label (serving 6
ounces or 177 ml) for a juice with the following analysis results. Assume USRDA
levels for an adult male.
• 0.501 g protein
• 20.4 g carbohydrates
• 35.6 mg/100 ml vitamin C
• 52.3 J.(g/l00 g thiamin
• 63. 1 mg / serving calcium added to the juice
5. If the peak height of a standard ascorbic acid solution, using the HPLC method, were
1.46 and the dehydroascorbic acid peak height 0.32, what would be the resulting
levels of the two if the juice sample peaks were 2.23 for ascorbic acid and 0.36 for
dehydroascorbic acid? What would be declared on the label?
174 CITRUS JUICE CHARACTERISTICS
REFERENCES
Attaway, J. A. et al. 1972. Some new analytical indicators of processed orange juice quality, Proc.
Fla. State Hort. Soc., 85, 192-203.
AOAC. 1980. Official Methods of Analysis, 13th Edition, 10.031, Association of Official Analyt-
ical Chemists, Washington, D.C.
Harding, P. L., Winston, J. R., and Fisher, D. F. 1940. Seasonal changes in Florida oranges.
USDA Tech. Bulletin, 753.
Keating, R. W. and Haddad, P. R. 1982. Simultaneous determination of ascorbic acid and dehy-
droascorbic acid by reverse phase ion-pair high performance liquid chromatography with pre-
column derivatization, J. Chromatography, 245, 249-255.
Ting, S. V. 1967. Nitrogen content of Florida orange juice and Florida orange concentrate, Proc.
Fla. State Hort. Soc., 80, 257-261.
Ting, S. V. and Rouseff, R. L. 1986. Citrus Fruits and Their Products. Marcel Dekker, Inc., New
York, 128-129.
Tockland, L. B. 1961. Nitrogenous constituents. In The Orange: Its Biochemistry and Physiology,
W. B. Sinclair, ed. University of California Press, Berkeley, Calif.
USDA. 1955. Energy Values of Foods-Basis and Derivation, A. L. Merrill and B. K. Watt, eds.
Government Printing Office, Washington, D.C.
U.S. Food and Drug Administration. 1973. Food: nutritional labeling, Federal Register, 38, No.
49, 6959-6961. U.S. Government Printing Office, Washington, D.C.
Chapter 12
Citrus Rheology
"Rheo" comes from the Greek word meaning "to flow," thus rheology is the
study of flow or, in the citrus industry, the study of the flow of citrus juices and
concentrates. Fluids are divided into two basic types in rheology. Newtonian
fluids are chemically pure homogenous fluids whose viscosity, or resistance to
deformation, does not change with shear rate. (Shear rate can be described as
the flow rate, the stirring rate, or the rate at which one surface passes over
another.) Non-Newtonian fluids have different apparent viscosities at different
shear rates. The latter description generally characterizes heterogeneous solu-
tions such as citrus juices and concentrates. The non-Newtonian nature of citrus
juices is primarily due to pulp and cloud material within the juice. Removal of
these components renders the resulting serum Newtonian in nature. The term
"consistency" or "apparent viscosity" is generally used with non-Newtonian
fluids, while "viscosity" is generally used with Newtonian fluids.
Citrus juices at concentrations above 200Brix are considered to be pseudo-
plastic; that is, their apparent viscosity decreases with an increase in shear rate.
In other words, as the concentrate moves more rapidly through a pipe or during
mixing in a tank, its apparent viscosity decreases. In order to start flowing,
citrus concentrates require an extra amount of energy, which is called the yield
stress.
Citrus concentrates are also thixotropic: when flow begins just after the yield
stress is applied, there is a breakup of the forces that hold the concentrate to-
gether. When the flow is stopped, these forces are restored. In rheology this
behavior is called going from a gel to a sol and then back to a gel condition.
The opposites of pseudoplastic and thixotropic are dilatent and rheopectic, re-
spectively.
In summary, citrus concentrates are non-Newtonian, pseudoplastic, thixo-
tropic fluids.
175
Citrus juice and concentrate consistency plays an important role in the quality
of citrus products. It has an important impact on the operation of every piece
of equipment used in citrus processing, as well as on consumer acceptance and
satisfaction. Even though routine measurements of the apparent viscosity gen-
erally are not necessary, consistency becomes an important parameter in de-
signing equipment for new products. Although this usually is a problem for
engineers, quality control personnel often are consulted about product consis-
tency, especially in small processing plants.
Equipment Sizing
Citrus processing plants are often referred to as just a bunch of tanks, pipes,
and pumps. Even though there may be a little more to processing plants than
that characterization implies, juice conveyance and storage are involved in nearly
every step of citrus processing. Thus flow rates, which are a function of pump-
ing power and pipe or orifice size, are of vital interest in integrating all the
processing steps and in transporting juice products in a timely manner.
Citrus concentrates exist at a wide range of temperatures, concentrations, and
compositions during their processing, and fluctuations in their apparent viscos-
ities can be dramatic. A pump, pipe, or other piece of equipment that is too
small can become a lethal bottleneck in processing, blending, packaging, and
shipping. A pump, pipe, or other piece of equipment that is too large can turn
out to be an excessive and costly expenditure or even an electrical tapeworm,
consuming large amounts of unnecessary energy. Single-strength juices gener-
ally have the same apparent viscosity as water for engineering purposes, or
about 18 centipoise. For 42°Brix frozen concentrated orange juice (FCOJ), the
apparent viscosities range from 430 to 2749 centipoise at 26.7°C, with grape-
fruit concentrate ranging from 150 to 1885 centipoise (Ezell 1959). For 60 to
65 °Brix concentrates the apparent viscosities range from 400 to 1000 centipoise
at 25°C and 3000 to 7000 centipoise at -10°C, all with a shear rate of 115.5
sec-I (Crandall, Chen, and Carter 1982). Bulk storage and shipping viscosities
generally are close to the latter extreme.
The apparent viscosity increases dramatically at commercial freezer temper-
atures, so care should be taken not to drop the temperature of bulk freezers or
tank farms much below -10°C (15°F). Quality control personnel generally
have to monitor freezer temperatures. As mentioned in Chapter 18, the increase
in fluidity that occurs with increased temperatures, compared to drum freezer
temperatures of about -20°C or about OaF, is responsible for the formation of
potassium citrate crystals in lower Brix / acid ratio concentrates.
CITRUS RHEOLOGY 177
Processing Considerations
Consumer Considerations
Rheology Measurement
Many instruments and methods are used to measure viscosities and apparent
viscosities. To compare results from one method or instrument with another, it
must be remembered that for non-Newtonian fluids the apparent viscosity is
characterized by the temperature and the shear rate. The apparent viscosity gen-
erally is expressed as centipoise ( 100 g / sec cm) and shear rate (sec - , ). Among
the methods used are procedures that measure the flow through a capillary tube
or orifice, the rate of a falling weight through the sample, the power consumed
in stirring, the penetration into a test sample, the rate and degree of spread on
a flat surface, ultrasonic vibration, and radioactive density measurements. For
non-Newtonian fluids, the most applicable and most commonly used method is
the rotation of a spindle or a cylinder within a coaxial cylindrical tube. These
instruments generally provide a consistent shear rate or rotational speed for
selected spindles, from which the torque needed to maintain that speed is mea-
sured and correlated with the apparent viscosity. The shear rate in such mea-
surements can be found by using:
( 12-1)
where R, is the radius of the cylindrical spindle, R2 is the radius of the cylin-
drical tube containing the sample, and a is the angular velocity of the spindle
(rpm) (Loncin and Merson 1979). The procedure outlined by the manufacturer
of each instrument should be followed in making apparent viscosity measure-
ments.
Newtonian methods of measuring apparent viscosities, such as measuring the
rate of fall of a heavy object through a sample or the rate of flow through a tube
or an orifice, generally involve varying shear rates throughout a single mea-
surement. For Newtonian fluids where the viscosity does not change with shear
rate, these methods are acceptable, but with non-Newtonian solutions, such as
citrus concentrates, the varying shear rates can give misleading results. Never-
theless, Newtonian methods can have value when applied to non-Newtonian
solutions in some cases. For example, if all that is desired is a pass/fail result,
Newtonian methods may suffice. Also, analysts developing in-house standards
can employ Newtonian methods if the product composition remains sufficiently
consistent.
3. How do pulp levels, Brixj acid ratio, pectinase enzyme actIvity, and temperature
affect the apparent viscosity of citrus concentrates?
4. Which piece of equipment is not generally affected by juice consistency: pipes, evap-
orators, extractors, chillers, bulk storage, or pumps?
5. What would be the shear rate of a rotary viscometer using a spindle with a radius of
1.0 cm and a sample container with a radius of 1.2 cm with an angular velocity of
100 rpm?
REFERENCES
Crandall, P. G., Chen, C. S., and Carter, R. D. 1982. Models for predicting viscosity of orange
juice concentrate, Food Tech., 36(5), 245-251.
Crandall, P. G., Davis, K. C., Carter, R. D., and Sadler, G. D. 1988. Viscosity reduction by
homogenization of orange juice concentrate in a pilot plant taste evaporator, J. Food Sci., 53(5),
1477-1481.
Ezell, G. H. 1959. Viscosity of concentrated orange and grapefruit juices, Food Tech., 13,9-13.
Kimball, D. A. 1986. Volumetric variations in sucrose solutions and equations that can be used to
replace specific gravity tables, J. Food Sci., 51 (2), 529-530.
Loncin, M. and Merson, R. L. 1979. Food Engineering, Principles and Select Applications. Ac-
ademic Press, New York, 34.
Chapter 13
Many factors affect the quality of citrus juices, including industrial practices,
fruit maturity, climate, soil conditions, rootstock, and fruit variety. Industrial
practices and fruit maturity have been discussed in detail in this text. Climate,
soil conditions, and rootstock are generally beyond the control of the processing
quality control department and/or processing plant. The variety of inbound juices
used in blending and/or product manufacture, however, is of interest and is
controlled to some degree by the processor, especially in regard to juices re-
ceived from outside the local area. Therefore, this chapter has been included to
give the processor a global picture of citrus varieties. Taxonomy that is related
to rootstocks, ornamentals, or other areas not directly relevant to citrus pro-
cessing has been omitted. Only those taxonomical parameters that concern cit-
rus products are considered.
For many years the diversity of citrus species and varieties was unknown.
Citrus species always have been and always will be undergoing natural and
manmade changes, which make classification an ongoing and controversial task.
Also, the effort to compare varieties grown in distant geographical areas and to
separate the effects of local conditions from varietal differences added to the
confusion. In addition, the separation of hybrids from true species remains
largely a matter of opinion and accepted practice. In 1967 Swingle and Reece
proposed 16 species of the Citrus subgenus that have generally been accepted
by the industry (Swingle and Reece 1967). Tanaka proposed 159 species of the
Citrus subgenus (Tanaka 1954). Tanaka's species may be of benefit to the hor-
ticulturist, but it is considered too complex for industrial use, especially to pro-
cessors. The author of this text also utilizes extensively the opinions and infor-
mation supplied by Hodgson (1967), Rangama, Govindarajan, and Raman
(1983), and Considine (1982).
Table 13-1 presents a taxonomical description ofthe commercially important
citrus species. The order in which the species appear represents general impor-
tance by groups of species that are similar. The commercial importance is a
relative and seasonal term that may vary from time to time. Other members of
180
the Rutaceae family are sometimes used for food products, such as the kumquats
of the Fortunella genus in the Citreae tribe. However, these products are of
minor importance compared to those found in the Citrus L. genus.
This species constitutes by far the most important class of commercial citrus,
with about two-thirds of the citrus produced worldwide falling into this cate-
gory. Sweet oranges are known by different names around the world, including
"naranja" in Spain, "aranico" in Italy, "laranja" in Portugal, "malta" in
India, and "kan" in Japan. Sweet oranges rank behind only sour oranges and
mandarins for hardiness, except that mandarins are more susceptible to frost
injury to the fruit. Sweet oranges are used extensively for fresh fruit, but in the
United States and Brazil they are used primarily for juice products. Sweet or-
anges can be classified into four categories, including the common orange, na-
vel orange, blood orange, and acidless or sugar orange.
Common Oranges
About two-thirds of all sweet oranges fall into the common orange category.
Common oranges are referred to as white or blond oranges. In Spain they are
called' 'blanca," in Italy' 'biondo," and in France' 'blonde. " The predominant
variety in this group is the Valencia orange, which constitutes about half of all
the oranges grown in the United States. Valencia oranges are also the predom-
inant commercial variety of any type of citrus, one reason for this dominance
being their long growing season and their resulting ability to adapt to a wide
range of growing conditions. Therefore, this orange is grown in every major
orange-producing country in the world. Contrary to popular belief, the Valencia
orange came from the Azores Islands and perhaps from Portugal, not from the
well-known city of Valencia in southern Spain, which is also known for its
citrus production. This variety should not be confused with the Valencia Tem-
prana variety in Spain, which matures earlier than the Valencia and is a smaller
fruit.
Table 13-2 lists the major varieties found in the common orange group. Main
budlines and related strains are generally classified together under one variety,
especially when the literature describes them as indistinguishable. Synonyms,
as well as countries where the literature has reported that the variety is of com-
mercial importance, are listed alphabetically, along with subjective and relative
descriptions of flavor and color. This relative rating system attempts to provide
the processor with a comparison of varieties within a given species according
to the literature and should not be construed as absolute. Also, some varieties
may be of local use and interest only, and so are not listed here.
CITRUS PROCESSING VARIETIES 183
The literature cited early in this chapter also includes some information that
may be of interest to the processor. One can learn, for example, that the Bel-
ladonna variety may be the same as the Shamouti orange. The Calabrese variety
is the most important common orange in Italy. The Hamlin orange is the world's
principal early-season common orange. The Homosasa is one of Florida's old-
est varieties. The Macetera is considered excellent for processing because of its
high juice content and unique flavor. The Maltaise blonde closely resembles
the Shamouti and Shamouti Masry oranges. The Parson Brown is the earliest-
season common orange. The Sathgudi is the principal common orange in India.
186 CITRUS JUICE CHARACTERISTICS
Because the Shamouti orange is the most common sweet orange in North Africa
and the Near East, it sometimes is simply referred to as the "beledi" or com-
mon orange. This designation probably is misleading because other common
oranges are grown in the area as well. The Shamouti orange generally is not
considered to be a good processing orange because it lacks juice and has a
relatively poor color. The development of bitter compounds has also been re-
ported in the Shamouti oranges (Levi et al. 1974), as well as in the Balta,
Mosambi, and Pakistan Valencia. The varieties described as early-season in-
clude the Balta, Biondo Commune, Hamlin, Marrs, and Mosambi, with the
Parson Brown described as very early. Early to midseason fruits include the
Premier, Vicieda, and Westin. Midseason fruits include the Khettmali, Mace-
tera, Maltaise Blonde, Pera Rio, Pera Coroa, Pineapple, Queen, and Sathgudi.
The late mid season varieties include the Maltaise Ovale and the Seleta. The
late-season fruits include the Calabrese, Pera, Pera Natal, and Valencia.
Navel Oranges
Blood Oranges
favored flavor and is considered the most delicious of all citrus juices. However,
the anthocyan ins have a tendency to fade during processing and storage that
gives the juice an undesirable muddy color. Activated charcoal has been used
commercially in Italy to remove the anthocyanins, restoring a healthy color to
the juice. This activated charcoal also removes limonin, which imparts a de-
layed bitterness to many blood varieties similar to that of navel juice. However,
the activated charcoal also removes significant amounts of vitamin C, which
cannot be replenished in the United States in 100% juice products without vi-
olating federal standards of identity. The color development in blood oranges
is directly proportional to the heat applied during the season. High temperatures
deepen the pigmentation.
Table 13-4 lists the main blood orange varieties. Blood oranges can be clas-
sified into three regional groups. The oldest group is the ordinary blood orange,
which is similar to the common orange in every way except the pigmentation.
It includes the Sanguinello Commune, Maltaise Sanguine, and other light blood
varieties such as the younger Moro and Tarocco varieties. The Doblefina group
includes the Doblefina of Spain, Entrefina, Spanish Sanguinelli, and Doblefina
Amelioree. The smallest group is the Shamouti or Palestine Jaffa Blood Orange
group, including the Shamouti Mawadri and the Mawadri Beledi. The latter is
the earliest-season blood orange. The Maltaise Sanguine and the Sanguinello
Commune are similar to the Sanguinello Moscato, the latter being the most
important commercial blood orange, which is grown primarily in Sicily. The
oldest blood orange is the Sanguigno Semplice. The Moro blood orange matures
very early, with midseason blood oranges including the Ruby, Sanguigno
Semplice, Sanguinello Commune, Sanguinello Moscato, Tarocco, and To-
mango. Late midseason blood oranges include the Doblefina, Doblefina Ame-
lioree, Maltaise Sanguine, Mutera, and Spanish Sanguinelli.
Acidless Oranges
189
190 CITRUS JUICE CHARACTERISTICS
Sour oranges are also referred to as bitter or Seville oranges. In Spain they are
called "naranja agria" or "amarga," in Italy "melangolo" or "arancio
amaro," in France "bigarade" or "orange amere," in Israel "khuskhash," in
West Pakistan "khatta," and in Japan "daidai." Sour oranges are generally
too bitter and sour to be processed into juice, and the oils are generally strong
and disagreeable. However, the trees are resistant to excess soil moisture, frost,
and neglect and thus are grown in areas prohibitive to other species. Sour or-
anges are used to make marmalades and, to some extent, essential oils. By
British standards, the quality of marmalades made from sour oranges surpasses
the quality of marmalades made from sweet oranges.
There are three basic types of sour oranges, the main varieties of which are
listed in Table 13-6. Common bitter oranges consist mainly of the Sevillano
variety, which is used extensively for the production of marmalades, essential
oils, and bitter juices used in drink bases in Spain and Australia, and is a late-
season fruit. The bittersweet group contains the essentially synonymous vari-
eties of the Bittersweet of Florida and the Paraguay or Apepu. This group is
grown mainly for essential oils. The variant bitter oranges are grown primarily
for perfumery and ornamentals.
Hybrids or species that closely resemble the sour orange include the Gunter
(c. maderaspatana Tan.), Bergamot (c. bergamia Risso), Sanbo (c. sulcata
Takahashi), and Myrtleleaf (c. myrtifolia Rafinesque). The Gunter is of im-
portance primarily in South India. Bergamot oil is used in Italy as a base for
cologne water. Sanbo remains popular in the Wakayama Prefecture of Japan
and the Myrtleleaf (Large Chinotto variety) is used for candied peel (marma-
lade) or crystallized whole in Italy.
CITRUS PROCESSING VARIETIES 191
Grapefruit and pummelos (shaddocks) are closely related. In fact, many people
believe that the grapefruit is a hybrid or a variety of the pummelo. The grape-
fruit (so named, it is believed, because the fruit grows in clusters like grapes)
is recognized as the separate species Citrus paradisi Macfadyen. Grapefruit has
a highly distinctive flavor with a lighter color, whereas the flavor and the color
of pummelos vary widely and give rise to the three pummelo species-Citrus
grandis (L.) Osbeck, the most common, Citrus maxima (Burm.), and Citrus
decumana L. In France the pummelos are referred to as "pamplemousse," in
Italy as "pompelmo," in Spain as "pampelmus," and in Japan as "buntan"
or "zabon." Most pummelos are very thick-skinned and are worthless as a fresh
fruit or for juice. However, there are some that are of importance commercially
in the Orient. Pummelos are less cold-tolerant than grapefruit, and some of the
major varieties do not have a bitter flavor like that of grapefruit.
Tables 13-7 and 13-8 briefly describe commercially important grapefruit and
pummelo varieties. In Spain grapefruit are referred to as "toronja." They rank
third behind oranges and mandarins in world production. In processed products
the glucoside naringin imparts an immediate bitter flavor as well as delayed
192 CITRUS JUICE CHARACTERISTICS
bitterness from limonin. In Florida, a little less than half of the grapefruit that
is grown is processed. Grapefruit peel is used in the production of pectin and
essential oils. There are two basic types of grapefruit-common or white and
pigmented or pink. The late-season Marsh variety is by far the most popular
common grapefruit. The mid-early-season Duncan grapefruit is considered to
CITRUS PROCESSING VARIETIES 193
be the best flavored, setting the standard for all other varieties. It is a common
trend among varieties that the seedier ones (such as the Duncan) have a richer
flavor than the less seedy or seedless varieties (such as the Marsh). The Triumph
Grapefruit is another early-midseason grapefruit.
Pink grapefruit varieties are characterized by the presence of the pink to red
pigment of the carotenoid lycopene. Although attractive in a fresh fruit, this
pigmentation has a tendency to fade during processing and storage similar to
that of the anthocyanins in blood oranges. For this reason, pigmented grapefruit
juice is not popular. However, pink grapefruit juice cocktail contains pink
grapefruit juice along with sweeteners, colorants, and other ingredients de-
signed to give a pink grapefruit juice appearance without this color fading ef-
194 CITRUS JUICE CHARACTERISTICS
MANDARINS OR TANGERINES
Orangelos
Chironja Puerto Rico mild, lacks yellow or-
grapefruit ange
bitterness
Tangelos
Hassaku (c. has- Hassaku Mikan Japan good, turns light yellow
saku Hort. ex. Hassaku Zubon bitter
Tanaka)
K-early Florida rather acid yellowish
orange
Melogold California good buff-chamois
Natsudaidai (c. Natsumikan Japan too acid for yellowish
natsudaidai Natsukan most orange
Hayata)
Orobanco California good buff-chamois
Poorman New Zealand Australia pleasantly reddish
Grapefruit New Zealand subacid,
Poorman Orange trace of
bitterness
Smooth Seville Smooth Flat Australia pleasantly reddish
Seville subacid, orange
trace of
bitterness
Wheeny Wheeny Australia good but straw color
Grapefruit New Zealand acid
darins rank second behind oranges in global importance and are more cold-
resistant than other types of citrus, except that freeze damage to the fruit can
be severe because of the loose nature of the rind. Increased heat during the
latter part of the season results in lower acid levels and milder juice. In Japan
about 15 % of the mandarins harvested are processed. Elsewhere, processed
mandarin juice is of less importance and is used primarily to enhance color in
light-colored orange juices. In the United States, up to 10 % mandarin juice can
be added to orange juice without declaration or violation of federal standards
of identity. There are four basic types of mandarins, which have been assigned
their own separate species classification.
The common mandarins, described in Table 13-10, are the most important
group. They are characterized by a tight rind. The Beauty variety is similar to
the Dancy, and the Campeona closely resembles the King variety in the King
196 CITRUS JUICE CHARACTERISTICS
mandarin group. The Lee, Robinson, Osceola, and Nova varieties are all sim-
ilar and are actually hybrids of the Clementine mandarin and the Orlando tan-
gelo. The Kinnow and Wilking varieties are hybrids of the King mandarin and
the Willow leaf. The Kinnow variety in West Pakistan is known to exhibit de-
layed limonin bitterness. The mandalina of Lebanon resembles the Dancy va-
riety. In the United States the Dancy tangerine is predominant with the Ponkan
variety dominating the Oriental markets, especially in China. In India the Pon-
kan is referred to as the "nagpur santra" and is the leading mandarin variety.
The very early-season fruits include the Cravo, Imperial, and Nova varieties.
Early-season fruits include the Clementine, Fewtrell, and Robinson varieties.
Early to midseason fruits include the Emperor, Lee, Osceola, Ponkan, and Sun-
burst. Midseason fruits include the Beauty, Dancy, and Wilking. Late midsea-
son fruits include the Ellendale, Emperor, Ortanique, and Tankan. Late-season
fruits include the Campeona and Murcott. Very late-season fruit includes the
Malvaiso variety.
The Satsuma mandarin varieties are listed in Table 13-11, and, as can be seen,
dominate the Japanese citrus industry. Satsuma mandarins are also found in
198 CITRUS JUICE CHARACTERISTICS
significant quantities in many other parts of the world, and are considered the
hardiest of mandarins as well as the most cold-tolerant of citrus. Satsuma tan-
gerines generally are divided into five categories. The Wase Satsumas include
the early-season varieties (September-October) and represent one-fifth of all
Japanese mandarins. The oldest variety is the Aoe Wase. The Zairai group
includes old native varieties indigenous to Japan. The Owari group, the most
important, is grown in the Owari Province and is especially suited for the can-
ning of fruit sections. The Ikeda and Ikiri groups are primarily of local use and
interest only. Early-season varieties include the Aoe Wase, Iseki Wase, Mat-
suyama Wase, Miho Wase, Miyagawa Wase, Okitsu Unshiu, Okitsu Wase, and
Silverhill. The medium-early varieties include the Dobashi Beni and the Owari
Unshiu. The midseason varieties include the Nankan-4 and the Yonezawa. The
late-season varieties include the Ishikawa and the Sugiyama Unshiu.
CITRUS PROCESSING VARIETIES 199
This group of mandarins, so named because they are primarily grown in the
Mediterranean basin, is sometimes referred to as the Willow leaf mandarins (see
Table 13-12). This group contains the broadest distribution of varieties, being
referred to as: "Ba Ahmed" in Morocco; "Blida," "Boufarik," or "Bougie"
in Algeria; "Bodrum" in Turkey; "Paterno," "Palemo," or "Avana Spe-
ciale" in Italy; "Nice" or "Provence" in France; "Valencia" in Spain; "Se-
tubal" or "Galego" in Portugal; "Baladi," "Effendi," or "Yousef" in Egypt
and the Near East; "Thorny" in Australia; "Mexirica" or "Do Rio" in Brazil;
"Chino" or Amarillo" in Mexico; and "Koina" in Greece. This group is the
second most important mandarin variety behind the common mandarin but is
being rapidly replaced by the Satsuma mandarins as well as the Clementine
mandarin. Some Chinese varieties have been reported to resemble Mediterra-
nean mandarins, including the Suhoikan variety. The Mediterranean variety is
considered a late-season fruit.
Mediterranean Mandarins
Bergamota Argentina
Malaguina Argentina
Mediterranean Mediterranean N. Africa distinctly mild light
Common Algeria and plea- orange
Willow leaf Argentina santly aro-
Brazil matic,
Egypt sweet
Greece
Spain
Suhoikan China
King Mandarins
King Indo-China California rich deep orange
Mandarin China
Camboge Florida
Mandarin Indo-
Kunembo China
Malaysia
Okinawa
Taiwan
200 CITRUS JUICE CHARACTERISTICS
Hybrids
Several important tangor and tangelo varieties that most closely resemble man-
darins are listed in Table 13-13. The most important hybrid in the United States
is the Temple tangor (c. sinensis X C. reticulata), which has been assigned
the species C. temple Hort. ex. Tan. and is considered a medium late-season
fruit. The Iyo mikan of Japan has been assigned the species C. iyo Hort. ex.
Tan. and is a mid season fruit. The Umatilla tangor is a cross between the Sat-
suma tangerine and the Ruby orange. The Minneola, Seminole, and Orlahdo
tangelos are all crosses of the Duncan grapefruit and the Dancy tangerine. The
Iyo tangor is known to exhibit delayed limonin bitterness. Of this group, the
Orlando tangelo is considered an early-season fruit, the Thorton tangelo is con-
sidered a mid season fruit, the Minneola is a medium late-season fruit, and the
Ugli is a late-season fruit. Other mandarin-like hybrids include the Fallgo
(Bower x Temple), an early-season fruit, and the Sudachi (c. ichangensis x
C. reticulata), the latter constituting the species C. sudachi Hort. Shirai.
Lemons are characterized by their light yellow color and their excessively high
acid content. Lemons once were used in the manufacture of citric acid but have
been replaced by more inexpensive methods. Lemons are referred to as "li-
mone" in Italy, "limon" in Spain, and "citron" in France, but are different
from the Mediterranean limetta (c. limeua Risso). They contain the widest
divergence of all citrus species. Lemons are primarily processed into juice and
lemonade, as well as pectin and lemon oil. Table 13-14 lists the main lemon
varieties. The Bema lemon is much like the Lisbon lemon, and the Interdonato
is the earliest-season lemon in Italy. The main varieties by far are the Eureka
and the Lisbon, with the latter generally replacing the former. The best-pro-
cessing lemon is said to be the Mesero, with the Monachello being resistant to
the mal secco disease. Lemons may have more than one crop per year and
sometimes several crops. The Bearss and Mesero are considered winter fruits,
with the Interdonato included as an early-season fruit. The Eureka lemon ma-
tures in the late winter, spring, and early summer, with the Lisbon lemon ma-
turing in the winter and early spring. The Femminello Ovale is a late winter
fruit. The sweet lemon variety dorshapo is of minor importance except in Ara-
bian countries, where it is highly favored. The related species limetta (c. li-
metta Risso) resembles the Dorshapo sweet and includes the single important
variety of Millsweet. Some evidence suggests that the elderly prefer the higher-
acidity drinks made from lemons.
LIMES
Limes are similar to lemons except that limes have a greener fruit and flesh
color than lemons and have a distinct and unique flavor and aroma. In Italy and
Spain limes are referred to as "lima," and in French-speaking countries they
are called "lime." In Arabian countries both lemons and limes are called "li-
mun," and in the Orient they are referred to as "nimbu" or "limbu." Limes
are the most tropical of all citrus and are generally preferred to lemons. India
is the largest producer of limes, which in the United States are grown almost
202 CITRUS JUICE CHARACTERISTICS
entirely in southern Florida. Limes have the highest acid content of all citrus
but are lower in vitamin C and other nutrients compared to lemons. Limes are
used mainly as additives to alcoholic beverages, in limeade, and to produce
lime oil. Table 13-15 lists the main lime varieties.
Limes can be classified into three species. The small-fruited group (c. au-
rantifolia Swing.) consists of a single commercial variety, West Indian, which
is the most commercially important of all lime varieties. The large-fruited group
CITRUS PROCESSING VARIETIES 203
(c. [atifolia Tan.) is second in importance, with two of the varieties, Bearss
and Tahiti, closely resembling each other. The Sahesli lime of Tunisian and the
Pond lime of Hawaii are very similar to the Bearss lime described in the table.
The sweet lime (c. limettoides Tan.) is considered too insipidly sweet in the
United States but is popular in India and the Near East. The single variety,
Indian or Palestine, is referred to as "mitha nimbu" in India and "limun helou"
or "sucarri" in Egypt. In California the dessert and coastal fruit of this variety
exhibit marked differences. Sweet limes are often used to prevent fevers and
liver complaints. The West Indian and Tahiti limes are mainly winter fruits,
along with the Meyer hybrid. Some lemon and lime-like hybrids also are de-
scribed in Table 13-15.
QUESTIONS
REFERENCES
Considine, G. D. 1982. Foods and Food Production Encyclopedia. Van Nostrand Reinhold Com-
pany, New York, 410-414.
Hashinaga, F. and Itoo, S. 1983. Seasonal changes in limonoids in Hassaku and pummelo fruits.
l. lap. Soc. Hort. Sci., 51 (4), 485-492.
Hodgson, R. W. 1967. Horticultural varieties of citrus. In The Citrus Industry I, W. Rether, H. J.
Webber, and L. D. Batchelor, eds. Division of Ag. Sciences, University of California, Berkeley,
Calif., 431-580.
Iwata, T. and Ogata, K. 1976. l. lap. Soc. Hort. Sci., 45(2),187-191.
210 CITRUS JUICE CHARACTERISTICS
Levi, A. et al. 1974. The bitter principle in Shamouti orange juice. 1. Seasonal changes and dis-
tribution in different parts of the fruit, Lebensm.-Wiss. Techno!', 7,234-235.
Rangama, S., Govindarajan, V. S., and Raman, K. V. R. 1983. Citrus fruit varieties, chemistry,
technology, quality evaluation, and analytical chemistry II (Review), Chemical Rubber Company
Critical Reviews in Food Science and Nutrition, 18(4),313-386.
Swingle, W. T. and Reece, P. C. 1967. The botany of citrus and its wild relatives. In The Citrus
Industry. Division of Ag. Sciences, University of California, Berkeley, Calif., publication 4012.
Tanaka, T. 1954. Species problems in citrus (Revisio Aurantiacearium, IX), Japan Soc. Prom.
Sci., Ueno, Tokyo, Japan.
Chapter 14
Inspections
213
The United States Food and Drug Administration (FDA) has the responsibility
for regulating and ensuring the sanitation and safety of all foods and drugs
produced or imported into the United States according to the Federal Food,
Drug, and Cosmetic Act enacted on June 25, 1938. This Act has since been
modified and incorporated into the Code of Federal Regulations (CFR). In this
code, Title 21 part 110 refers to sanitation and defines the "good manufacturing
practice" that is required of all food processors. Some of the main features of
good manufacturing practice as outlined in the code are the following:
10. There should be an ample potable water supply for proper cleaning, of
sufficient quality that it will not contaminate the food.
11. Sewage and refuse disposal should be done in an adequate manner to
avoid food contamination.
12. Doors to restrooms should be self-closing and not open directly into pro-
cessing areas. Double doors should be used if necessary.
13. All equipment and utensils used in processing should be cleaned suffi-
ciently and often enough that they will not contaminate the food, and
they should be stored in a sanitary fashion.
14. All ingredients or raw materials should be inspected to ensure that they
are clean, wholesome, and fit for human consumption.
15. The food processing should be monitored in such a way as to ensure the
sanitation of the product, by procedures including, but not limited to,
container inspections, temperature monitoring, sanitary inspections, and
microbial monitoring.
16. Food products should be so labeled as to allow proper product identifi-
cation, and records concerning product quality should be retained for at
least the shelf life of the product, or no more than two years. The ad-
dition of any defective lot renders the final product defective, regardless
of the degree of the defect.
17. Storage and shipment of food products should be done in such a way as
not to contaminate the food.
FDA Inspections
Preinspection Phase
When an FDA representative appears at the plant, the first thing one should do
is arrange for a preinspection conference. Those persons who should attend
include the quality control manager, the chief executive officer of the company
(CEO), and the FDA inspector. FDA inspectors are required to inspect plants
at reasonable times and not during weekends, holidays, or off shifts. Each com-
pany should maintain an FDA file. The inspector should be required to show
216 CITRUS JUICE SANITATION
his or her credentials, and these credentials should be recorded in the file along
with the inspector's name and the date and time of the inspection. FDA in-
spectors do not have unlimited authority to access company records except for
shipping documents of interstate commerce under section 703 of the Act and
processing records of low acid foods. They also are not granted authority to
take photographs. It must be remembered that volunteered information can be
used against a processor in a court oflaw. However, the best policy is to always
be prepared for an FDA inspection through a proper sanitation program. Then
full cooperation with FDA inspectors can be a pleasant experience for all con-
cerned, perhaps even a time to show off a little.
The preinspection conference is a good time for processors to review the
results of the last inspection and clarify any questions they or the inspector may
have. The FDA file should include information on all cleaners and additives
used in the plant as well as all paperwork from previous inspections. Also, the
file may include informative pamphlets and brochures such as "FDA Inspection
Authority" (NFPA Bulletin 39-L, NFPA, 1133-20th Street, Northwest, Wash-
ington, D.C. 20036) or "Good Manufacturing Practice in Manufacturing, Pro-
cessing, Packing, or Holding Human Food" (U.S. Department of Health, Ed-
ucation, and Welfare, Public Health Service, Food and Drug Administration,
HFF-326, 200 C Street, Southwest, Washington, D.C. 20204). Some compa-
nies have a form letter that they give to the inspector on arrival that outlines
company policy about what the inspector will or will not be allowed to do
beyond his or her inherent authority. The preinspection conference presents a
good opportunity for those involved to go over these policies.
Inspection Phase
FDA inspectors should always be accompanied while inspecting the plant, pref-
erably by the quality control manager. The names and titles of those persons
who accompany the inspector should be recorded in the file. If a USDA in-
spector is present, he or she may wish to accompany the inspector as well. The
FDA inspector has authority to inspect all areas of the plant without warrant
and should be allowed to do so. During the inspection the inspector will fill out
form FD-482, shown in Fig. 14-1, a copy of which should be kept with the
FDA file. If the inspector takes any samples, he or she should issue a receipt
for them, which also is kept in the file. It is recommended that the company
take duplicate samples. The form used by the FDA inspector as a sample receipt
is shown in Fig. 14-2. If any problems are found, the inspector is required to
notify the company promptly by using FD-483, shown in Fig. 14-3.
Postinspection Phase
After the inspection, a postinspection conference should be held by the same
people who attended the preinspection conference. If any of the above forms
INSPECTION 217
Notice of Ins on it hereby given pursuant to Section 704{a) & (e' of the Federal Food~ Orug~ and Cosmetic Act [21
U.S.C. 374(01 8. (olJ'and/or Part F or G, Title III of the Public Hoalth S.",ice Act [42 U.S.C. 262·264)2
purpose, may dUring all reasonable hours enter and Inspect any estab·
Iishment for the propagation or manufacture and preparation of any
VirUS, serum, tOXin, antllOXtn, vaccme, blood. blood component ordenv-
lSec. 704. (a) FOf" purposes of enforcement of this Act, officers or em- atlve. allergenic product or other product aforesaId for sale. barter, or
plovees dul.,. designated bV the Secretary, upon presenting approp- exchange m the D,stTict of Columbia. or to be sent, carTied. or brought
r1"le credentials and wfltten notice to the owner. ope rat Of. or agent In from any State or possession Into any other State or posesslon or Into
charge, are iluthorlzed (1) to enter, at reasonable times. any lactory any foreign country, or from any foreign country Into any State or
warehouse. or establishment In which luod. drugs. devices, or cosmet posseSSion
ICS are manufactured, processed. packed. or held. for Introduction Into
mterstate commerce or after such introduction. or to enter any vehicle
being used to transport or hold sucb food. drugs, devices, or cosmetiCs In Part F - •• " Control of Radiation
Interstate commerce. and (2) to Inspect. at reasonable limes and wlthm
reasonable tlmlts and In a reasonable manner. such factory, ware· Sec 360 A(eI) '"If the Secretary finds for good cause that tbe methods,
house, establishment. or vehicle and all perTment eqUipment. finished tests, or programs related to electroniC product radiation safey In a
onel unfinished millerlals containers, and labeling therein In thp ril<;e particular factory, warehouse, or establishment In whIch electronic
of any factory, warehouse, establishment, or consulting laboratory in products are manufactured or held. may not be adequate or reliable,
which prescription drugs or restricted devices are manufactured, proc- officers or employees duly designated by the Secretary, upon presenting
essed, packed, or held, inspection shall extend to all things therein approprlBte credentials and a written notice to the owner. operator, or
(including records, files, papers, processes, controls, and fiICilities) agent III cbarge, are thereafter authOrized (1) to enter. at reasonable
bearing on whether prescription drugs or restricted devices which are times any area In such factory, warehouse. or establishment In which
adulterated or misbranded within the meaning of this Act, or which the manufacturers tests (Of resting programs) reqUired by sectIon
may not be manufactured, introduc:ed into interstate commerce, or 358(h) are carTied out and (2) to Inspect. at reasonable times and WIthIn
sold, or offered for sale by reason of any provision of this Act, have reasonable limits and In a reasonable manner. the facilities and proce·
been or are being manufactured, processed, packed. transported, or dures wllilin such area wblch are related to electroniC product radla1lon
held in an.,. such place, or otherwise bearing on violation of this Act. safety Each such Inspection sh;]11 be commenced and completed With
No inspection authorized by the preceding sentence shall extend to reasonable promptness In addition to otber grounds upon whIch good
financial data, sales data other than shipment data, pricing data, cause may be found for purposes of thiS subsection. good cause Will be
personnel data (other than data as to qulllHications of technical and conSidered to eXist In any case wllere the manufacturer has Introduced
professional personnel performing functions subject to this Act), and Into commerce any electroniC product which does not comply With an
research data (other than data, relating to new drugs and antibiotic applicable standard prescnbed under thiS subpart and With respect to
drugs and devices anc., subject to reporting IJnd inspection under regu- which no exempllon Irom the notificatIOn requirements has been
lations lawfully issued pursuant to section 505(i) or (j), section SOl(d} granted by Ole Secretary under sectIOn 359(a~2) or 359(e) ..
or (g), sections 512(1} or (m), section 519, or 520(g}, and data relating
to other drugs or devices which in the case of a new drug would be (b) Every manufacturer of electroniC products shall establish and
subject to reporting or inspection under lawful regulations issued pur- maintain such records (mc/udmg restmg records). make such reports.
suant to section 505(j)). A separate notice shall be given for each such and proVide such mlormatlon, as the Secretary may reasonably requITe
inspection, but a notice shall not be required for each entry made to enable him to determine whether such manufacturer has acted or IS
during the period covered by the inspection. Each such inspection shall acting In compliance With thiS subpart and standards preSCribed pursu'
be commenced and completed with reasonable promptness. ant to thiS subpart and shall. upon request of an officer or employee duly
deSignated by the Secretary, perml! sucb officer or employee!O Inspect
Sec 704(e) Every person reqUired under section 519 or 520(g) to main·
appropriate books, papers, records and documents relevant to deter
lam records and every person who IS In charge or custody of such
ITllnlng wllether such manufacturer has acted or IS actIng In complIance
recorris shall upon request of an officer or employee deSIgnated by the
With standards prescTlbed pursuant to section 359(a)
Secretary. permit such officer or employee at all reasonable times to
have access to and to copy and verify, such records
(f) , The Secretary may by regulation (1) require dealers and dlstflb·
2Applicable sections of Pans F and G of Title III Public Health Service
utors of electroniC products to which there are applicable standards
Act f42 U.S.C. 262-264J are quoted below:
prescnbed under 011S subpart and the relall pnces of whlcb IS not less
Pari F LicenSing - Biological Products and Clinical Laboratories than $50, to furnish manufacturers of such products SUCll InformatIOn
and"'" •• as may be necessary to Identify and locate. for purposes of seclJon 359,
Ihe first purchasers of such products for purposes other than resale. and
Sec 351(c) "Any officer, agent, or employee of the Department of (2) reqUire manufacturers lapreserve such informatIOn Any regulation
Health, EducatIOn, and Welfare, autllorlzed by the Secretary for the establlsblng a reqUirement pursuant to clause (1) of the preceding
Fig. 14-1. Form FD-482, filled out by the FDA inspector during an FDA inspection.
218 CITRUS JUICE SANITATION
Thr follo ... !n" ",.,....,,1 ...... ,,.. (,ClII"'iI'"lf'~1 I» th!l' F"oCod IoI\I! DIU, ,ir, dllllll'll.tnlllliOft.1Id Ii!CiI!.pt ,. Iltul'br 1o["lu!.o ... lf"dacd pu r tu. "110
8,.".110/'0 :0. ((') o f 11lI,. Pri .. , .. , PIDold , Orull:. MId CO.fM'lIcA~1 CZI U.S.C. l7"Ccl:: .nJ oilorP.rt ..... Sub P.rt J. $rclllJlll'~(bJ ",f
'rh .. Put;.ll~ H.-01111'" $,.,,,j.:,. A C~ - .1' U.S.C. ~6J..J- Io"d/OI' 2 1 COd .. <Ol F.-It ... ,., R,. ... h.U_. !CF R j IJ07.0'1. EltClupn or Ih,u e
. , .... ..al<l'·d ot\ \hit 'f'WIU' of Ih ... (<0_.
Fig. 14-2. FDA sample receipt, used if samples are taken during an FDA inspection.
INSPECTION 219
) V (ifl.......W/'f)' :J::&...C'fJ.~
'&l., ~ '1'110 -
&.r
1C.
97.A.i>1..<',Vl(it<
NAM e; 0" INDIVIDUAl. T O WH O ,.. "IiPOR' ' .... ulEe CAT I' 0" Ult ... eCT ION' fl . N U MDCR
1' 1" 1'14 NAM" NAME 0" '1ft ... , BRANCH OR UNIT IN.~I[CTI.D
(' I. /~
.TAIlCTlAODAI; ••
,'in (,j., " jl, rvrI ' f',)
"'fAIEET AOORE.• • OF" PRE."U&E8 IN&PECTt:D
~£F. P.f.l'&R51.
or fillS PAC;'
", .11'
~1!~Tlzk:£
( •• ,
ri;;." ~ ~
, ;~T~' ii.
' 0,
J:..¥1-, •
';, ,-p:. \
A
. I < "P~C<:'::' nj7r7~
11)1\: '" ri 1~ ' ... _... . ~. 7 ~
T'r ....
"' • "r.~" ,
INSPECT,O ...... l OBSER V " T' 0 ... S
Fig. 14..3. Fonn FD-483, filled out by the FDA inspector during an FDA inspection.
220 CITRUS JUICE SANITATION
have not been filled out, this is a good time to complete them. Any problems
found during the inspection are discussed, and the FDA inspector outlines a
plan for correcting the problems. Notes should be taken on these conversations
and any decisions made, and included in the FDA file.
After the inspector leaves, plant management should meet to decide how any
requested corrections will be made, in accordance with the FDA inspector's
instructions including specific assignments.
Imported Products
Whenever citrus products are imported into the United States, both customs and
FDA clearances must be obtained before any product can be further processed
or sold. Custom inspectors are primarily concerned with proper representation
of the product, contraband, disease, and contamination brought in from foreign
countries. FDA officials are concerned with adulteration and proper identifica-
tion of products. Most citrus importers identify the products they import into
the United States by using the U.S. federal standards of identity found in Chap-
ter 19. These standards of identity sufficiently describe imported products for
identification by the FDA officials, who then issue form FD-702, shown in Fig.
14-4.
USDA INSPECTIONS
1. Weeds, trash, rubbish, rodent and insect harborages around the plant,
and protection from such in food storage areas.
2. Offensive odors, dust, soot, or inadequate drainage.
INSPECTION 221
o
ENTRY DATA TAKEN FROM HEALTH AND HUMAN SERVICES
Number FOOD AND DRUG AOMtNIST"ATfON
IOAd.I"""
Notice
o Mlnifet!
39-M10127-0
0 ..006/14/88 439-0010127-0 06/14/88
D
I
IT Ad"anct
=~[ ENTRY NO. AND DATE
N01;ee
BII...L OF" LADING NO. POAT OF LADING COUNTRY OR ORIGIN
Mexico
PORT OF UNLOADING
Hidalgo, Texas
I Hidalgo, Tx
PORT OF ENTRY
FOR THE 'Ntme. A~r",. IMPORTER IName, Add,e... , MANUFACTVAER/SH'''ER (NIIM., AddrtSs,
.,
ACCOUNT OF I CONiI..- ZOp Codol OF RECORD Zip Code) ZIp Code)
Tropical Bf"Verllgea Inc Smne Ju~uera Veracruzana SA
700 Jackson Ave, Apt 134 Kill 297 Carrt Mexico-TUltpa
HcAllen, Texas 78501 POM Rica, Ver .• H exlco
SROKER fllltOt.."..,Ibo.,.' LOCAT ION OF LOT (FOr" FOA DATE
.1tM"I1".,'otI) AVAIL
N._
J1IIIny Santoa Inc
J1,=.1
T~~~ Citrua Excha~ .e
~~~i I I same
mw~. LUD.-u II I I PHONE NO. 5B5 4503
MAY PROCEED
Without FDA Extmlnltlon 216 Dltu1l\s Frozen cn-"nge Juice
Concentrate
This notlQe does not Pf't'cludt tction should
the mereh.ndi .. In!t be found ....iol .. tive •
• " lLlDONLY,FSlGNED
I!
2. loatlon in 10001.re, of the Ihipment on that dlte.
FORM FDA 102(6/85) p.ey,ous EDITION MAY 8! USED. MAY PROCEED NOTICE
Fig. 14-4. Fonn 702, used by the FDA to clear an imported product.
222 CITRUS JUICE SANITATION
FORM FV-41o_3
15-76)
U.S. DEPARTMENT OF AGRICULTURE
AGRICULTURAL MARKETING SERVICE
PREMISES
<1) O,Mrs Le-?+ "pelll - ;VO-t ;·n~4 ~G ""5 ....
1. Outside areas \I V~"
A
2. Waste disposal fMI V IV ~ peel +'rItSftI ".., jr"u",j
3.
RECEIVING AREA
1. Unloading pit ./ V- i/'
2. Grading V VIV
B
3.
4.
Conveyors and chutes
Bins
V
M:r V V
VV'
B:> .fIles SWl1vw-'JI\.J /':"'.5 - +# iJ -f.o yeW\", IJ
5.
6.
7.
PROCESSING AREAS
1. Washing and grading V (fiN II' (i) c./" loy: \1\ ..."ibl-" vu"'f W DvK •' : ;
2. Sizers and leads
3. Extractors V .... V
/.-.
4. Troughs and lines V' '-""' II.
5. Finishers ./ V
6. Valves and lines V v
7. Blend and storage tanks l/ V
C 8. Disposal screw conveyors V' ...... V'
9. Fillers & closing machines V L; V
10. Stabilizer V V V
11. Floors, gutters & walls V'-"
12. Waste storaQe ..... V
13.
14.
15.
16.
WAREHOUSE OR COLD STORAGE
1. General housekeeping V- V L..
V 1/ L-
....
2. Sugar storage
D 3. Can storage V V
4.
5.
6.
REST ROOMS
1. Supplies V frV\ V ~ No 5bo..{/ .for- Wtl-ski~ ha..ds
2. Wash basins L- V. ...
E
3. T oi lets & urinals V' V
4. Floors & walls V 0/ 20
5.
PERSONNEL
1. Cleanliness V V ~
2. Head covering V V V
F
3. Smokil]9 V V ......
4. I" )" Iv V '1/
5. "'" ..l"
Fig. 14-5. Inspection fonn used by USDA inspectors during continuous inspection.
224 CITRUS JUICE SANITATION
unsatisfactory condition persists for three successive days or three days in one
week, the inspector is required to contact his or her supervisor and/or other
USDA officials as needed. Again, failure to resolve sanitation problems can
result in the withdrawal of grading services.
IN-HOUSE INSPECTIONS
Some good guidelines for in-house inspections can be found in the FDA and
USDA inspections. Performing FDA-like and USDA-like inspections is the best
way to prepare for such inspections. A good pamphlet that can be used as an
additional guide is "Do Your Own Establishment Inspection-A Guide to Self
Inspection for the Smaller Food Processor and Warehouse" (U.S. Department
of Health and Human Services, Public Health Service, Food and Drug Admin-
istration, 200 C Street, Southwest, Washington, D.C. 20204).
Even though the legal requirements of sanitation must be kept in mind, san-
itation programs in citrus processing plants should be based on product quality.
If proper care is taken to ensure the quality of the products in-house, no serious
problems should ever occur with FDA or USDA inspections. A loss of repu-
tation and profits is usually much more devastating than conflicts with regUla-
tory agencies. In-house personnel also are more knowledgeable about equip-
ment and procedures and can implement a more effective sanitation program
than outside inspectors. This knowledge of products and equipment should be
constantly used in designing and redesigning sanitation programs so that pro-
duction time and cleaning materials will not be wasted during the production
of high-quality products.
There are two types of in-house inspections. One is based on solving a spe-
cific problem that has occurred. For example, the presence of a large number
of Drosophila fruit flies may require the periodic inspection of peel residue
accumulations in certain areas of the plant. Once these accumulations are being
removed on a regular basis, or when the weather gets cooler and the flies be-
come less active, the procedure can be modified or eliminated. If a particular
procedure or inspection system does not solve the problem, new procedures
should be tried until the problem is solved. At least on a weekly basis, the entire
plant should be shut down for a complete cleanup, and an FDA- or USDA-like
inspection should be performed. It is important that those individuals doing the
inspecting are not those doing the cleaning. Otherwise, shortcuts will be taken
in the inspection. Also, management should always be aware of and supportive
of monitoring the sanitation condition of the plant, especially in construction
and reconstruction within the plant. Quality control personnel should always be
consulted in such modifications.
Inspections form just the foundation for a good sanitation program. The sub-
sequent chapters go into greater detail about the many facets of citrus sanitation.
INSPECTION 225
QUESTIONS
1. What are some of the factors that affect sanitation in processing or bot-
tling plants?
2. What is the foundation of any sanitation program?
3. What are some of the reasons why sanitation is important?
4. What is the law regarding chewing tobacco in processing areas?
5. What is the law regarding restroom doors that open directly into pro-
cessing areas?
6. Who should attend the preinspection conference when an FDA inspector
appears at the plant?
7. What forms will an FDA inspector fill out, and why?
8. What records does an FDA inspector have the authority to access?
9. Can a processor go to jail for failing a USDA inspection?
10. Why is a survey taken prior to USDA continuous inspection?
11. Give examples of minor, major, and critical USDA sanitation infrac-
tions.
12. What should be the basis of in-house sanitation programs?
13. What are the two types of basic in-house inspections?
14. How should FDA and USDA inspectors be treated, and why?
15. When is it okay to leave an FDA inspector alone in the plant?
Chapter 15
Citrus Microbiology
BACTERIA
Single-cell organisms are divided into two types according to cellular structure.
One type is the eucaryotes ("eu" means true; "karyo" is the combining fonn
for nucleus), or microorganisms with a true nucleus. The other category is the
procaryotes, which are microbes that lack a true nucleus and usually have their
DNA in a single molecule. Procaryotes include bacteria and bluegreen algae,
whereas eucaryotes contain all other algae, molds, yeasts, and protozoans. There
are three types of bacteria, which differ in their shape or morphology. Bacillus
bacteria are elongated or rod-shaped, coccus bacteria are spherical, and spirillus
bacteria are spiral or corkscrew-shaped. Bacteria are classified by many param-
eters, including growth conditions, size, and wastes produced. Even though
other acid-producing bacteria have been isolated from citrus juices from the
genera of Aerobacter and Xanthomonas (Faville and Hill 1951a), only lactic
acid bacteria from the genera Lactobacillus and Leuconostoc have been found
to significantly affect citrus products.
226
Lactobacillus
Lactobacillus bacteria (Buchanan and Gibbons 1974) are gram positive rod-
shaped bacteria from the Lactobacillaceae family, which derives its name from
the fact that at least half of the carbon produced from carbohydrate metabolism
is in the form of lactate. Other products from Lactobacillus growth are diacetyl,
acetate, formate, succinate, carbon dioxide, ethanol, and other one- or
two-carbon acids. Diacetyl (2,3-butanedione), which is synthesized by the bac-
teria from citric acid, imparts an undesirable buttermilk-like flavor and odor to
citrus juices and is the major cause of product rejection of single-strength juices
due to spoilage.
These nonpathogenic organisms grow at temperatures of 5 to 53°C (41-
12rF) with optimal temperatures generally 30 to 40°C (86-104°F). They are
very acid-tolerant, with an optimal pH range of usually 5.5 to 5.8 with signif-
icant growth below this pH range. Lactobacillus bacteria also are very sensitive
to osmotic pressure or juice concentration, with slow growth at 35 to 38°Brix
and no growth over 45°Brix. The main species found in citrus juices are Lac-
tobacillus plantarum and Lactobacillus brevis (Hays and Riester 1952). L.
planarum has an optimal temperature range of 30 to 35°C (86-95°F), with
growth at 15°C (59°F) but generally not at 45°C (l13°F). L. brevis has an
optimal growth temperature of about 30°C (86°F), with growth at 15°C (59°F)
but no growth at 45°C (113°F). Some strains are pigmeted from orange to red,
but most are white in color.
Leuconostoc
Leuconostoc (Buchanan and Gibbons 1974) are gram positive spherical coc-
coids, usually in pairs or in chains. Leuconostoc is a nonpathogenic member of
the Streptococcaceae family, which normally produces lactic acid, ethanol, car-
bon dioxide, and diacetyl with the same buttermilk-like off flavors and odors as
produced with Lactobacillus bacteria. Optimal temperatures for growth are in
the range of 20 to 30°C (68-86°F), which is cooler than the range for Lacto-
bacillus bacteria. Most species will grow in media with a pH of 5.5 to 6.5. One
species, Leuconostoc oenos, will grow in the range of 4.2 to 4.8 and even lower
but probably not higher. Generally Leuconostoc bacteria grow very slowly at
pH levels of around 3.6 compared to Lactobacillus, but they grow faster at a
pH of 4.0 and above. Most species are facultative anaerobes, which means they
can grow in the presence or absence of air. The main species found in citrus
are Leuconostoc mesenteroids and Leuconostoc dextranicum (Hays and Riester
1952). L. mesenteroids grows in the 10 to 37°C (50-99°F) range, with an
optimal range of 20 to 30°C (68-86°F), but these bacteria cannot withstand
228 CITRUS JUICE SANITATION
55°C (131°F) for 30 minutes. Some strains, however, may withstand temper-
atures of 80 to 85°C (176-185°F) for shorter periods of time. L. mesenteroids
forms a slime of dextran metabolized from sucrose and/or levano polymers from
fructose. L. dextranicum also produces slimy dextran growth, though not as
actively as L. mesenteroids. The growth temperature range and the optimal
range for L. dextranicum are the same as for L. mesenteroids.
YEASTS
Yeast cells are generally larger than bacteria, as illustrated in Fig. 15-1 by
bacteria and yeasts that have been isolated from citrus juices. Generally yeasts
grow much more slowly than bacteria, with optimum temperatures for growth
of 20 to 30°C (68-86°F), the same as for Leuconostoc bacteria. Yeasts gen-
erally are more tolerant of high temperatures (65-70°C or 149-158°F) than
CITRUS MICROBIOLOGY 229
bacteria or molds. They also have a much higher tolerance for preservatives,
cold temperatures, and osmotic pressures than bacteria. In fact, yeasts are some-
times referred to as osmiophilic or osmiotolerant. Within a microorganism there
is a delicate balance of aqueous life-giving solutions. If the cell wall comes into
contact with a medium that is more concentrated in soluble material than the
cell is, the water from these life-giving solutions will migrate out of the cell
into the more concentrated solution in an effort to equalize the two concentra-
tions. This osmotic effect is generally lethal to bacteria and is one reason for
the use of concentrated syrups in the canning of fruits. However, yeasts are
more resistant to osmotic pressures than bacteria and can easily survive in citrus
concentrates of 58 to 65°Brix. Yeasts have been shown to remain more viable
at 50 0 Brix than at 12 to 30 0 Brix after heating at 60°C (140°F) for up to 8
minutes (Juven, Kanner, and Weisslovicz 1978), so they are of primary concern
in the microbiology of citrus concentrates. If yeasts are predominant in the juice
flora, the juice most likely came from reconstituted concentrate. If bacteria
dominate the flora, the juice most likely came from single-strength juice or was
contaminated in a single-strength form. Yeasts also grow in media with a pH
as low as 1.5, well within the acidity range that is common to citrus juices.
It is interesting to observe that even though yeast can be found in nearly every
concentrate that has not been aseptically treated, fermentation occurs incon-
sistently in lots of citrus concentrates. This phenomenon may be partially due
to the number of viable microbes prior to spoilage, but that does not account
for such fermentation in samples with lower microbial counts. In the fermen-
tation process, there is oxidation by a microbe in the absence of an electron
230 CITRUS JUICE SANITATION
receptor such as oxygen. This results in the production of more oxidized cabron
dioxide gas and more reduced ethanol, as follows:
Not all of the 57 kilocalories thus produced is released as heat. Some of this
energy is stored in high-energy phosphate bonds in ATP within the microbe.
One reason for the inconsistent fermentation in citrus concentrates is that not
all species that have been found in citrus juices are capable of fermentation. As
seen in Table 15-1, the six most commonly found species in one study will not
ferment at all (Rhodotorula and Cryptococcus species). The next eight species
ferment only glucose, which comprises only one-fourth of the carbohydrates in
citrus juices. The next three species (Trichosporan) will not ferment at all,
whereas two others that were among the least common yeasts in this study are
capable of fermenting both glucose and sucrose. (Note: The eight species at the
bottom of the table were isolated from citrus concentrates in separate studies;
thus their percent of incidence in relation to the others is unknown.) It is evident
that the species present may significantly affect whether or not fermentation will
occur. A more recent study documents that the most prominent yeast in pas-
teurized citrus juices is probably Sacchromyces cerevisiae, followed by Rho-
dotorula species (commonly called pink yeast) and Zygosaccharomyces species
(an osmophile) (Parish 1988). Rhodotorula species are not thought to be spoil-
age yeasts and may indicate post-pasteurization contamination.
Mold
Most of the molds of concern to the food industry belong to the same class as
yeasts-Ascomycetes (sac fungi). The major contributing factor to mold growth
is the availability of moisture. Molds have been known to grow in any location
where moisture is found, ranging from air with a relative humidity of 70% to
water reservoirs. In addition to moisture, air availability is important. Also,
molds generally require an environment free from agitation or disturbance. Idle
concentrate tanks exposed to mold spore contamination from the air have been
known to accumulate mold growth if not cleaned out or flushed out with fresh
concentrate. Mold colonies found in standard plate counts generally indicate
that there has been undisturbed concentrate exposed to the air for a period of
time up to several days, or that the plate has been exposed in the air. An upper
temperature limit for mold growth would be about 60°C (140°F), with a pH
tolerance down to as low as zero.
Molds that get into citrus juices from the surfaces of fruit generally are ho-
CITRUS MICROBIOLOGY 231
mogenized and destroyed by the juice acidity, and usually are not likely to affect
the quality of the juice, except perhaps for carton juice. Freshly squeezed juice
obtained from the processing of moldy fruit may also contain high levels of
mold, most of which can be destroyed through proper pasteurization. Black
mold growing on the surfaces of moist processing equipment or buildings may
break off and create an unsightly cosmetic contamination. However, molds
found in citrus juices have not been known to be pathogenic or produce off
232 CITRUS JUICE SANITATION
flavor or odors in citrus products under nonnal conditions. Some of the molds
that have been isolated from citrus juices include, in order of prevalence, Au-
reobasidium pullulans, Aspergillus niger, Botrytis species, Phoma species, Mu-
cor species, Aspergillus Jumigatus, Cladosporium sphaerospermum, and Pen-
icillium species (Kobatake, Kurata, and Komagata 1978). A more recent study
found the following in commercial pasteurized orange juices: Aureobasidium
pullulans (a black yeast), Penicillium species (black yeasts), and Cladosporium
species. Also the following were found in unpasteurized laboratory samples:
Fusarium species, Geotrichum species, and Penicillium species (Parish and
Higgins 1989).
DETERMINATIVE TECHNIQUES
Diacetyl
+HN=C
NH2
~-6H
I 2
oII
C-OH
06
----.~ Dye (15-2)
40% KOH
Diacetyl CH 3
Creatine
constant rate throughout the distillation in the procedure given below. Acetyl-
methylcarbinol also reacts with the creatine to give the same color as produced
with diacetyl; so a correction must be applied to the diacetyl distillate to account
for the acetylmethylcarbinol. Reconsituted concentrates result in much less dye
formation than freshly extracted juices, because of volatile losses during evap-
oration.
Procedure
6. After 5 minutes of contact time with the creatine solution, measure the %
light transmission or absorbance, using a spectrophotometer set at 530 nm
(Abs = log (lOO/%Trans).
7. Subtract the absorbance of the third 25 ml portion from that of the first,
which contains most of the diacetyl. This step corrects for the acetyl-
methylcarbinol in the first diacetyl portion. Compare the corrected ab-
sorbance with the absorbance of diacetyl standards.
Diacetyl values can be plotted on a control chart, and when those values
increase dramatically, a cleanup should be performed immediately. A range of
0.6 to 6 ppm has been suggested for the maximum diacetyllevel, depending on
the particular operation and desired specifications.
Standard plate counts comprise the primary method used in monitoring micro-
bial growth in citrus products. Plate counts account only for viable organisms
but take 2 to 3 days to get results. The plate count is performed by mixing a
juice sample with an agar medium and incubating the resulting gel. Each or-
ganism in the sample reproduces to form a visible colony that can be easily
counted. The location in the plate, shape, color, and texture of the colonies
generally can be used to identify the type of organism that created the colony.
Lactobacillus organisms form small white to red circular colonies about the size
of a pinhead and can be embedded throughout the agar, not on the surface.
Leuconostoc colonies are large domelike circular colonies, appearing only on
the surface ofthe agar, with a smooth translucent color and texture. One species
of Leuconostoc produces colonies that appear as a thin film on the bottom of
the medium under anaerobic conditions. Yeast colonies generally are recog-
nized by their star shape, formed by two flat circular colonies intersecting at
right angles. These colonies are generally between Lactobacillus and Leucon-
ostoc in size and grow throughout the medium. Surface yeast colonies may
resemble Leuconostoc colonies but usually are flatter on the surface. Mold col-
onies are always on the surface and consist of a very fibrous or cottonlike ma-
terial. Some mold colonies may result from air contamination rather than the
sample. A blank plate containing no juice sample aids in the monitoring of
sterilization procedures as well as air contamination.
Plate counts usually are expressed as the number of colony forming units
(cfu) per milliliters of single-strength juice ( 11.8 °Brix). This allows the com-
parison of juice products that may vary in concentration. When the number of
colonies per plate exceeds 300, a dilution should be performed until the number
of colonies per plate is less than 300. Trying to count over 300 colonies may
result in counting errors, and too many colonies on a plate may result in the
CITRUS MICROBIOLOGY 235
Sterilization
Procedure
where C is the count / ml SS] or the number of colony forming units (cfu) ,
P is the number of colonies counted on the plate, D is the dilution or 100
in the procedure above, and S is the pounds soluble solids/gallon accord-
ing to the Brix of the original sample, using Equation 2-8 or 2-10. If
greater dilution is required in order to keep the number of colonies counted
below 300, then 1 ml from the first dilution bottle can be added to a second
99 ml dilution bottle to give a dilution of 10,000 instead of 100. This
value then can be used for D above.
238 CITRUS JUICE SANITATION
Automated Methods
PROCESSING CONSIDERATIONS
Fruit
lation. Chlorine levels are easily determined by using locally available swim-
ming pool test kits, which involve the yellow color development of chlorine
with o-tolidine. Dilution of water samples may be required in order to keep
within the chlorine concentration range on the color comparison chart included
with such a kit. Fruit grading, washing, and the use of chlorinated water is
generally more important in warmer humid climates or seasons. It should be
remembered that fruit that has undergone spoilage prior to chlorinated washings
may still impart off flavors and odors to the resultant juice.
Pasteurization
Cooling Effects
Both bacteria and yeasts have difficulty growing in concentrates of 60 0 Brix and
higher; so the careful control of water addition to concentrate lines and tanks
has been found to be of great importance in controlling the growth of micro-
organisms in citrus concentrates. Partial rinsing of a concentrate line or tank is
one of the major causes of high microbial plate counts in concentrate lots be-
cause the microbes grow much more rapidly under the more dilute conditions.
Also, a constant flow of fresh concentrate through the lines and tanks minimizes
the time available for growth and reduces contamination. The only growth that
can occur is on the inner surface of the pipes where yeast may adhere and feed
on the flow of juice passing by, and this is why light yellowish yeast slime can
be found on the inner surfaces of contaminated lines. This buildup should be
cleaned immediately. During busy times of the year when fresh concentrate
constantly flows through the lines and the weather is cooler, the number of
cleanups can be reduced. Hot summer temperatures and infrequent processing
usually require a frequent-cleanup program. Cleanups may occur at various in-
tervals, from daily to monthly. Incomplete cleanups run the risk of inducing
microbial growth through dilution of the concentrate; therefore, partial cleaning
is worse than no cleaning at all for citrus concentrates.
Contamination of the lines from other sources also may cause spoilage. For
example, condensate may form inside the concentrate tank, diluting the con-
centrate and then spoiling it when left for a long period of time such as over
the weekend. Another source is blending with contaminated product, which
contaminates the entire blend. Another is blockage in plate heat exchangers or
lines, which provides a site for the accumulation of proudct that is not removed
during cleanups, and then spoils when left for long periods of time in what is
mistakenly thought to be clean equipment. Entrapped debris also provides a site
for microbial growth, and dead spaces in product lines present another hazard
where microbial growth may thrive. Such dead spaces should be taken apart
and checked regularly or removed altogether. A buildup of pressure from car-
bon dioxide production during fermentation literally can cause product hoses,
lines, containers, and even steel drums to explode, creating a safety as well as
a quality hazard.
Juice lines, tanks, evaporators, extractors, fillers, and packagers can all be
cleaned by using automated CIP (Cleaning In Place) systems. They also can be
CITRUS MICROBIOLOGY 241
taken apart manually and cleaned. Manual cleaning is the easiest and surest
method to inspect, but permanent lines are difficult or impossible to take apart
and inspect completely. However, most processing plants have enough equip-
ment to justify the use of an automated CIP system for at least part of it. Prop-
erly programmed CIP systems provide a more efficient and consistent cleanup
even though they are more difficult to inspect than manual cleaning systems.
Because citrus juices are acidic, 2 % caustic solutions are often used for clean-
ups. After the juice is removed, it is common practice to use a chlorine or
iodophor sanitizer rinse to reduce surface contamination of the equipment.
However, a rule of thumb may be that if a piece of equipment can be totally
inspected (all surfaces can be seen), such as a tank, and it looks clean to the
eye and there are no off odors, it can be considered clean. Equipment that can-
not be totally inspected, such as process lines, must be inspected by indirect
methods such as observance of clear water rinses and the presence of off odors.
Laboratory plate counts, however, are the surest means of determining the level
of product microbial contamination.
QUESTIONS
1. What is the main by-product of lactic acid bacteria that degrades citrus
juices?
2. What other genera of bacteria have been found in citrus juices besides
Lactobacillus and Leuconostoc?
3. What microorganisms are considered pathogenic to humans that are found
in citrus juices?
4. How can you recognize yeast cells under the microscope?
5. Why do yeasts survive better in citrus concentrates than bacteria?
6. To what eucaryotic class of microorganisms do the yeasts and molds
commonly found in citrus juices belong?
7. How can mold contaminate citrus juices?
8. What can be deduced from each of the following statements about sam-
ples?
• "A" has a high level of bacteria.
• "B" has a high level of yeasts.
• "C" has a high level of mold.
9. What is the minimum plate count before off flavors and odors will de-
velop from microbial spoilage?
10. Cite evidence from the chapter that the growth of microorganisms found
in citrus products is inhibited by light.
11. What is the maximum allowable % broken and % decayed fruit leaving
the grading table under USDA inspection?
242 CITRUS JUICE SANITATION
12. How does the condition of fruit correlate with the microbial plate counts
of the resulting processed juice?
13. What is the minimum temperature that has been shown to adequately
control microbial growth?
14. About how long does a concentrate juice lot with a high microbial count
need to be stored in a commercial freezer before the counts are reduced
to acceptable levels?
15. What are some of the most likely causes of high plate counts in citrus
concentrates?
PROBLEMS
1. What is the cfu of a 65°Brix sample that resulted in 316 colonies appearing on the
plate?
2. What is the cfu of a 60 Brix sample yielding 146 colonies of yeasts, 4 colonies of
0
What would be the diacetyl value of a sample with an absorbance of 0.7 for the
sample and 0.5 for the blank?
5. Identify the following plates:
,
1.
.. & D • 0 0 1 iPS \C'tc=n rmnl
REFERENCES
Berry, R. E. and Ve1dhuis, M. K. 1977. Processing of oranges, grapefruit and tangerines. In Citrus
Science and Technology II, S. Nagy, P. E. Shaw, and M. K. Velduis, eds. The AVI Publishing
Company, Inc., Westport, Conn., 202.
CITRUS MICROBIOLOGY 243
Buchanan, R. E. and Gibbons, N. E., eds. 1974. Bergey's Manual of Determinative Bacteriology.
The Williams & Wilkins Company, Baltimore, Md., 510-513, 576-593.
Faville, L. W. and Hill, E. C. 1951a. Recent investigations pertaining to acid-tolerant bacteria in
citrus juices. Second Annual Citrus Processors Meeting, Citrus Experiment Station, Lake Alfred,
Fla., October 9, 1951.
FaviUe, L. W. and Hill, E. C. 1951b. Incidence and significance of microorganisms in citrus
juices, Food Tech., 10, 423-425.
Hays, G. L., Riester, D. W. 1052. The control of "off odor" spoilage in frozen concentrated
orange juice, Food Tech., 6(8),386-389.
Hill, E. C., Wenzel, F. W., and Barreto, A. 1954. Colormetric method for detection of micro-
biological spoilage in citrus juices, Food Tech., 3, 168-171.
Juven, B. J., Kanner, J., and Weisslowicz, H. 1978. Influence of orange juice composition on the
thermal resistance of spoilage yeasts, J. Food Sci., 43, 1074-1080.
Kimball, D. A. 1982. Unpublished.
Kobatake, M., Kurata, H., and Komagata, K. 1978. Microbiological studies of fruit juices I & II,
J. Food Hygenic Sci. of Japan (Shokuin Eiseigaku Zasshi) , 19,449-461.
Parish, M. E. 1988. Microbiological aspects of fresh squeezed citrus juice. In Ready (0 Serve Citrus
Juices and Juice Added Beverages, R. F. Matthews, ed. Univ. of Florida, Gainsville, Fla., 79-
96.
Parish, M. E. and Higgins, D. P. 1989. Yeasts and molds isolated from spoilage citrus products
and by-products, J. of Food Protection, 52(4), 261-263.
York, G. K. 1988. Private communication with microbiologist, University of California, Davis,
Calif.
Chapter 16
Citrus processing plants generally are located in areas with mild climates, which
offer ideal living and growing conditions for many insects, rodents, and birds.
However, citrus juices may be processed and packaged in nearly every climate,
each with its own problems in regard to animal infestation. Regardless of the
processing conditions, the same basic principles of pest control are generally
applicable and important in producing a good-quality citrus product.
INSECTS
Even though the insects are a specific taxonomical class, the term "insect" is
generally applied to any creature that one would colloquially refer to as a' 'bug."
Insects are known for having six legs and three separate body sections. Such
creatures as spiders, mites, ticks, chiggers, millipedes, centipedes, snails, slugs,
and so on, are not insects, but for the purposes of food sanitation they can be
considered as such because of the methods used for their containment and con-
trol. Insects are the most abundant animal form on earth, with three to four
times as many species as the rest of the animal kingdom has put together. As
many as 700,000 species are known, with about two new species being added
each year. There is much that we do not know about them, but one thing we
do know is that insects cost the agricultural industry billions of dollars each
year. With respect to agricultural growth, insects are a concern because of their
effects on the quality and yield of certain crops. In the citrus processing indus-
try, however, the concern is with insects being embedded in fruit or being at-
tracted to the processing areas and entering the product during processing or
storage. Insect fragments in citrus products do not pose a health danger because
the acids of the juices kill any pathogenic microbes, but visible insect fragments
constitute a serious quality infraction, rendering the juice unpalatable. Also,
insect fragments and eggs that are visible only under a microscope can result
in lower defects scores upon USDA inspection; and citrus products contami-
nation by insects can lead to FDA action, up to costly fines or worse.
244
Hunger and thirst are the greatest motivators on earth, and insects are constantly
seeking food and water. Processing plants have a tendency not only to exist by
large bodies of war, but to have a large amount of water scattered around the
premises. Water, clean or dirty, has a tendency to draw insects, expecially
crawlers, some of which can detect water from long distances; and crickets and
cockroaches are especially attracted to wet areas. The field cricket (Gryllus
assimilis) (Fabricius)) is found in many parts of the world and across the United
States. The most common types of cockroaches are the Croton bug or the Blat-
tella germanica (German cockroach) (Fig. 16-1) and the Blatta orienta lis (ori-
ental cockroach); these cockroaches can hatch up to 50 offspring at one time.
These insects congregate around sinks, ditches, and wherever water collects.
They are most active at night or in the dark, but they often can be seen during
daylight hours. Crickets are generally easy to locate because of the noise they
make with their legs. Puddles of water not only attract insects, but they look
messy, and if they are in walkways or roadways, they become unpleasant ob-
stacles for people to wade through. Water should be drained or removed by
squeegee in such areas, and long ditches or waterways should be covered or
shortened to minimize insect attraction.
Insects feed on a variety of materials, especially those that humans consider
to be trash, such as paper. Trash receptacles should be well placed and used,
and should be emptied regularly. Floors in processing and eating areas should
be swept andlor mopped often. Fruit residue common around fruit conveyers
should be routinely removed, and juice spillage should not be left very long
before cleanup. Fruit waste that must be stored should be stored away from the
plant, and, if an insect infestation is noticed, the waste should be sprayed.
Weeds should not be allowed to grow near the plant, and landscaping should
be well cared for and sprayed if necessary. Unnecessary piles of parts, equip-
ment, or pallets provide excellent insect breeding and hiding areas, and should
be eliminated or removed from the plant. If the food and water are taken away,
insects will go somewhere else, not into your plant or your products.
The natural oils found in the outer peel of citrus fruit act as a natural barrier
246 CITRUS JUICE SANITATION
to most insects. Also, the acidity of the juice inside the fruit is unpalatable to
insects, as a rule. However, there are some insects that thrive in citrus fruit,
namely, the Mediterranean fruit fly, the Mexican fruit fly, the Oriental fruit fly ,
and a few others that are less common. These pests have been essentially erad-
icated from the continental United States but flourish in citrus processing areas
in other parts of the world. These insects enter the fruit and lay their eggs within
the fruit itself. This not only damages the fruit but provides unsightly contam-
ination when it is processed. Grade-out of infested fruit is not always 100%
efficient, and centrifuging of contaminated citrus juices is commonly used to
remove insect fragments. Imported juices often will have lower pulp levels be-
cause centrifuging to remove insect fragments also removes pulp. Other insects
grow only on the surface of citrus fruit and generally do not contaminate juice
INSECTS. RODENTS. AND BIRDS 247
extracted from the fruit. Lemons are not considered a host for the Mediterranean
fruit fly.
Drosophila
Even though the natrual aromas and essential oils of citrus repel most insects,
the common fruit fly or Drosophila (Insecta, diptera-also known as vinegar
flies or pomace flies) seems attracted by them. The genus Drosophila has up to
50 different species, with the predominant species being Drosophila melano-
gaster, which thrives in mild weather during the late summer and early fall (see
Fig. 16-1). Drosophila simulans also becomes abundant in areas such as central
California but does not thrive in hot weather. Drosophila pseudobscura pre-
dominates in cooler weather.
Drosophila melanogaster are 1.5 to 2.5 mm long with red eyes and a black
abdomen, the first three segments having a yellow band. They convey spoilage
organisms from fruit to fruit and are strong fliers, flying over 6 miles in one
day, although they become immobilized in wind over 5 miles per hour and a
light intensity greater than 150 foot-candles. They are day fliers with very rapid
reproductive cycles. Although the flies spend only 24 hours in the egg, the egg
may remain in the female to within an hour of hatching. The larva and pupa
stages take three days each, giving a reproductive cycle of about seven days.
Females may lay up to 2000 eggs at one time but average only about 1000.
Adults live from 40 to 70 days, depending on the temperature. Thus, if two
Drosophila mate on June 1, assuming an average life span and two days be-
tween reproductive cycles, by mid-July, when the original parents die, there
will have been over 30 quadrillion (3.125 x 10 16 ) fruit flies produced. If this
many fruit flies were placed end to end (assuming an average 2 mm length),
they would extend out into space nearly 2! light days or the distance that light
traveling at 186,000 miles per second would go in 2~ days. Within a few days
after this, the average rate of production from the time of original mating would
have exceeded the speed of light in placing the flies end to end as they were
born. If the parents were to remate during the larva and pupa stages of the last
generation, the reproduction rate would be two to three times faster. These
staggering statistics suggest the importance of insect control and the need for
action early in the season before the numbers get out of control. Vigorous re-
production also produces sex odors that drift in the wind and attract even more
insects-another reason why breeding areas should be removed and cleaned.
Temperature
Insects are cold-blooded animals; that is, they require an outside source for
body heat. The ideal temperature for insects is generally 28.0°C (82.4°F),
248 CITRUS JUICE SANITATION
which is very close to humans' ideal temperature. Therefore, insects will always
try to invade the human environment. The comfort range for insects is also very
narrow, generally 21 to 32°C (70-90°F); temperature outside this range seri-
ously diminish insect activity. Solar heat stored in cement and metals provides
a heat source for insects after sunset. The abundance of cement surfaces, brick
or stone buildings, metal buildings, vehicles, equipment, or railroad rails com-
monly found at processing plants provide a comfortable habitat for insects. Once
activated by temperature, they begin their search. for food and water, which are
generally plentiful around processing plants. When the climate is cooler or dur-
ing the winter, insects like to move indoors, where they can vacation for the
winter. Some like it so much that they never bother to go home, especially if
temperature controls are left on all the time. Turning off the temperature con-
trols over the weekend, especially during cold weather, inhibits insect activity
and growth. Insects also like to inhabit electrical control boxes and equipment
because of the warmth of the electronics. As no processing plant can operate
without these devices, the sealing of cracks and crevices and the use of chem-
icals (see below) are needed to control infestations in these areas. Insects often
cause damage to such equipment, especially sensitive laboratory equipment.
Light
All insects are attracted to light to some extent, a characteristic that is of great
importance to night operations where nocturnal (night-active) insects are at-
tracted to processing lighting. Some insects, such as moths, can see for hundreds
of yards, whereas others, such as Drosophila, can see only a few yards. The
compound eye structure that is common to insects prevents them from seeing
linearly, so they usually approach lighting in a wandering fashion.
Insect are particularly sensitive to the ultraviolet light range, just outside hu-
mans' visible region. Mercury vapor lights attract over 100 times as many in-
sects as softer yellow sodium vapor lights, which emit a smaller amount of
ultraviolet light. For this reason, the use of mercury vapor lights should always
be discouraged at food processing plants, especially in rural areas. When out-
side lighting is necessary, it should be placed away from entries into the plant
or locations where product is exposed to the air, such as tanker loading areas.
Tankers always need air access to the tanker during filling or emptying, which
usually means that insects have access as well. If these areas are well lit at
night, insects usually will congregate there. Lights within the plant are best
placed where they cannot be viewed directly from outside the plant. Blinds and
curtains should be closed at night to avoid attracting insects, and screens should
be used to cover windows that must remain open. In rural areas where a pro-
cessing plant is the only light source for miles around, insects especially can be
a problem, but in a city where there are many lights, the problem may be less
INSECTS, RODENTS, AND BIRDS 249
severe. Using lights with a lower wattage will also diminish insect attraction.
Again, once they are inside the plant, they may decide to stay there and repro-
duce.
The attraction insects have for light can also be used to the advantage of the
quality control department. Ultraviolet light traps, consisting of an ultraviolet
light surrounded by an electrified screen, attracts insects. When they land on
the screen, they are electrocuted and fall into a collecting tray at the bottom of
the trap. These trays, which should be changed often, will give a good indi-
cation of what kind of insects are infesting an area. These lights should be
placed in areas that draw the insects away from processing areas, and where
they cannot be seen from the outside-you would not want a light trap to lure
insects in from the outside. Most flying insects fly only about 5 feet from the
floor or less, so the light traps are best used at these heights. However, higher
processing areas that have attractive debris will lure insects higher off the
ground. Ultraviolet bulbs lose their ability to emit ultraviolet light with time
and should be changed periodically (at least annually), even if no visible change
is observed in their light.
Insecticides
Due to their obvious toxicity, most insecticides are not permitted for use in food
processing plants. The federal code (Title 21 11O.37(b)) states: "The use of
insecticides or rodenticides is permitted only under such precautions and restric-
tions as will prevent the contamination of food or packing materials with illegal
residues." The pesticide tolerances that define "illegal residues" are in a con-
stant state of change, so that only by subscribing to the Federal Register can
one be assured of staying on top of changes in the law, but this generally is not
required of citrus processors. The use of pyrethrin insecticides is permitted by
the Food and Drug Administration and has been shown to be effective against
Drosophila and many crawling insects. Pyrethrins have been used orally for the
treatment of intestinal worms in humans and thus pose a smaller toxic threat
than other chemicals used to control pests. Extracted from pyrethrin plants in
Africa and South America, pyrethrins consist basically of four compounds,
known as pyrethrin I and II and cinerin I (allethrin) and II; and they are effective
against a wide variety of insects including Drosophila, aphids, bettles, cabbage
worm, housefly, leaf hooper, louse, mealybug, mosquito, sod webworm, thrip,
and mare. These insecticides also work quickly, with a quick kill rate. Sunlight
quickly breaks pyrethrins down, making them a contact insecticide that must
be sprayed directly on the insect rather than a residual insecticide that will kill
the insect at a later time. This breakdown of pyrethrins prevents the buildup of
insecticides in the plant, but it also necessitates cyclic application. Pyrethrins
should never be used with alkaline materials, as they would thereby lose their
250 CITRUS JUICE SANITATION
Physical Barriers
An open door or product tank is an invitation to dinner for an insect that is best
canceled. All doors to processing areas should be closed when not in use and
equipped with automatic closing arms if necessary. Few insects can fly through
a closed door. Those doors that must remain open for long periods of time
should be equipped with properly functioning air screens. As shown in Fig.
16-2, air screens should not blow insects into the plant. Air flow should be
uniform across the entire door area without quiet "holes" where insects can
INSECTS. RODENTS. AND BIRDS 251
PROCESSING AREA
PROCESSING AREA
BAD!
Fig. 16-2. Proper and improper air screens over doorways that must be open for long periods of
time, showing good air flow and bad air flow. Care should be taken not to blow the insects into
the processing area.
slip through. Also, ventilation systems used to remove peel oil volatiles that
bum the eyes in extraction areas can disrupt air curtain flow, a situation that
must be avoided. Because fruit must be conveyed into the plant and the wastes
removed, there usually are openings in plant walls that can allow the entry of
insects. These should be as small as possible and covered with plastic strips or
other suitable barriers. Windows that open should always be screened. Also,
windows should be caulked, thresholds sealed, and other holes and cracks re-
paired. Even a solid wall a few feet high around the plant can be a significant
barrier to a crawling insect infestation.
Product tanks should always be closed except when it is absolutely necessary
that they be open. Some tanks require air vents during filling and emptying,
which should be screened to avoid the entrance of insects. Drum, can, or bottle
filling stations should be entirely free of insects. Filling is usually the most
vulnerable time in the process for product contamination by insects.
252 CITRUS JUICE SANITATION
Education
Often, in order for the previous methods to work, plant personnel need to be
educated on what the plant is doing for the control of insect contamination.
Purchased or home-made slide or video presentations can be used as a part of
new-employee orientations. Unlike microorganisms, insect problems are some-
thing that everyone can see and do something about, not just the quality control
department.
The important thing to remember is that all of these methds need to be used.
No one or two methods are sufficient. In fact, using all of these methods does
not guarantee that you will not have insect problems. As with many quality
control parameters, however, the consumer demands a sanitary product regard-
less of the problems involved.
DETECTION OF CONTAMINATION
QUESTIONS
REFERENCES
Considine, D. M. 1982. Foods and Food Production Encyclopedia. Van Nostrand Reinhold Com-
pany, New York, 1626.
Chapter 17
The sanitary production of foods as outlined in the federal code for good man-
ufacturing practice includes restrictions that prevent contamination by physical
or nonliving objects or chemicals. The physical objects range from dust parti-
cles to hammers, and the chemicals range from toxic lubricants to juices of
another variety. Most of these objects and chemicals are necessary for the op-
eration of a citrus processing plant. However, regardless of their value, they all
have the potential for misuse and can damage the quality of the plant's products.
Such misuse can range from accidental or careless contamination to intentional
adulteration, and, unlike the contamination discussed in previous chapters, can
be lethal to product consumers. For this reason, prevention of such contami-
nation should be a vital part of any sanitation or quality control system in citrus
processing plants.
PHYSICAL CONTAMINATION
254
wall as a sanitary hazard. Even visitors should wear hair protection when they
tour proudct areas. Also, good personal hygiene should be practiced by food
handlers, including clean clothes and hands.
Insulation near product containers, commonly used to insulate product and
refrigeration lines, may become worn and break off, causing unsightly or even
dangerous contamination. Dust and dirt inside the plant should be removed as
much as possible, and cleaned equipment should not be placed on the ground
or in any location where it may pick up dust or dirt. Dusty roads or travel ways
should not be allowed near or inside the plant, especially in tanker and tank
loading and unloading areas. After a tanker has been cleaned, it should never
travel any distance without all openings closed tightly. Condensate water that
commonly accumulates on cold surfaces, such as refrigerated lines and tanks,
should not fall in or near product containers, as this water picks up dust and
dirt from the air and the exterior surfaces of processing equipment.
Citrus products should never come into contact with unsanitary surfaces such
as wood or rusty steel. Drums and drum lids used to package citrus concentrates
should contain an inner baked enamel finish that is free from rust, dirt, or chip-
ping paint. Plastic containers should be free from plastic dust and residue. Pa-
perboard containers should be checked for dust and dirt accumulation prior to
use. During barreling, two plastic liners should be used, and the innermost liner
should not be opened until just before filling and should be closed just after
filling. Before barrel dumpers are used, all extraneous material, such as labels,
liner fasteners, and so on, should be removed so that it will not fall into the
product as it empties.
If glass objects are dropped into a tank full of product, there is no sanitary
way to get them out, and the potential for breakage could result in the total
rejection of the entire lot of juice. Plastic or metal containers should be used to
take samples, and bimetal thermometers should be used to take temperature
readings. Similarly, the Food and Drug Administration requires all glass light-
ing over product vessels or containers to have plastic covers or sleeves in order
to avoid contamination if the glass should break. Because debris from eating,
drinking, smoking, or chewing gum or tabacco often finds its way into product
containers, the FDA prohibits these activities in processing areas. Food ma-
chines should be placed in lunchrooms and not in processing areas, and em-
ployees should be provided a time and place for meals and breaks.
CHEMICAL CONTAMINATION
The processing of citrus juices always requires the use of chemicals in one form
or another. Water itself is a chemical, and is used not only for cleaning but to
adjust the concentration of concentrates and juices from concentrate. It should
go without saying that only potable water should be used in citrus products and
256 CITRUS JUICE SANITATION
for cleaning surfaces that come into contact with citrus products. Recycled or
reuse water should never be used for such purposes. Even though the water so
used in potable, that does not guarantee that the minerals, salts, or other chem-
icals (such as chlorine) in the water will not affect the quality of the citrus
product. Dilution water should be checked for off flavors and odors prior to use
or whenever contamination is suspected. Water used for cleanups also should
be potable and should not contain any extraneous debris. Many beverage man-
ufacturers go to great lengths to treat the water they use in the making of their
products.
Only food-grade caustics and sanitizers should be used for cleanups, and
FDA inspectors generally require information on the cleaners used in process-
ing. After cleanups, equipment should be thoroughly rinsed and drained in order
to avoid contamination the next time that it is used. Some mixing of the
FDA-approved cleaning agents may not pose much of a health hazard for con-
sumers; however, off flavors or odors may arise with such mixing, the acid
nature of the juice may change, and/or the federal standards of identity will
most likely be violated.
Citrus processing sometimes also includes the use of food additives, pesti-
cides, lubricants, and other chemicals. Nonfood and food chemicals should be
stored separately so that toxic materials will not be added accidentally to food
products. Care should be taken not to add the wrong food additive because so
doing not only could affect the quality of the product, but it could also violate
standards of identity. If polychlorinated biphenyls (PCBs) are used in any trans-
formers, in capacitors, as heat transfer or hydraulic fluids, or in lubricants,
coatings, inks, and so on, the federal code should be consulted (Title 21 110.40).
These compounds are considered toxic, and their use is regulated in food plants.
The many moving parts of citrus machinery require lubricants, and care should
be taken that these chemicals not drip into product containers or anyplace where
they could result in product contamination. Also, refrigerants used to chill citrus
juices and concentrates have the potential of leaking into the product, which
would diminish their chilling capacity in addition to contaminating the product.
Containers, such as drums, that have been previously used for nonfood pur-
poses, such as petroleum, should never be used for food products.
A good rule of thumb regarding physical or chemical contamination is some-
times referred to as Murphy's law: If anything has the possibility of getting into
a food product, most likely it will. Eliminate the possibility, and you will elim-
inate the contamination.
QUESTIONS
Processing Contamination
HESPERIDIN
HESPERIDIN
OH
OH ~OCH
~o-~w)g ,
OH HO OH ~
I
OH 0
f3-Rutinosyl
258
of the juice from the fruit, hesperidin comes into contact with the acidic juice
and will begin to form crystals. Under the microscope, these crystals are long
and needlelike. On processing equipment, they appear as a white scale or film
that can build up until it breaks off as white flakes. Hesperidin also contributes
to about 10 to 20% of the juice cloud. Its crystals can clog finishing screens,
thereby decreasing juice yield. Also, during evaporation concentration of the
juice accelerates hesperidin crystallization, producing white flakes that appear
in the concentrate or reconstituted juice.
Hesperidin levels within the fruit generally decrease with maturity because
as the fruit matures, it accumulates moisture, which dilutes the hesperidin con-
centration. In spite of this, the appearance of hesperidin flakes in citrus juices
increases with fruit maturity and may become an acute problem in the late sea-
son, especially in Valencia juice. This is probably due to lower acid levels,
which reduce the solubility of the hesperidin. Also, greater breakdown of the
peel in late-season fruit may cause increased amounts of hesperidin to get into
the juice. The USDA has set an arbitrary standard for the number of hesperidin
flakes permitted for various defects scores, as shown in Fig. 18-1. Hesperidin
flakes can be observed (USDA method) using 710 ml of 11.8 °Brix reconstituted
juice that has sat undisturbed for 5 minutes in a 1000 ml beaker with a 4-inch-
diameter bottom. White hesperidin flakes are apparent on the bottom of the
beaker, as observed from below with the aid of a flashlight.
Even though hesperidin formation cannot be prevented, routine hot caustic
cleanups can keep finisher screens and processing equipment free from hesper-
idin buildup. Again, late-season processing requires more frequent hot caustic
cleanups.
SCORING GUIDE
FOR HESPERIDIN
FROZEN CONCENTRATED ORANGE JUICE
AND CONCENTRATED ORANGE FOR MFG.
Fig. 18-1. Arbitrary defects scoring guide for hesperidin flakes used by the USDA.
260 CITRUS JUICE SANITATION
BLACK FLAKES
JUICE OXIDATION
cause no further changes in the juice. The organic acids present in citrus juices
also are catalysts and can assist in breaking up carbohydrate chains through acid
hydrolysis. However, the amine-assisted breakdown of carbohydrates has been
shown to be the major cause of the development of cooked off flavors (Hand-
werk and Coleman 1988). The mechanism for the amine catalysis is similar to
that of acid catalysis, but has been shown to occur under milder conditions of
heat and acidity than acid catalysis. Also, the oxidation of ascorbic acid or
vitamin C has been shown to be a significant factor in the development of off
flavors in citrus juices.
Citrus juices contain greater amounts of amino acids than other juices such
as apple or grape, which make citrus juices more vulnerable than other juices
to oxidation involving amino acids and the plentiful carbohydrates. These sugar-
amino acid reactions were first studied by Louis-Camille Maillard (1878-1935)
and have since taken his name as nonenzymatic Maillard reactions. The Mail-
lard reactions involve the reaction of the aldehyde group of the sugar with the
amino group of the amino acid, according to the following condensation reac-
tion:
HC =N-R
R'-CHO + R-NH z ...... I
R' (18-1 )
sugar amino Schiff base
acid
Further reactions can occur from the Schiff base, including various condensa-
tion reactions and Strecker degradations. At least 20 products of oxidation of
citrus juices have been isolated, and many more exist. However, only six have
been shown to contribute significantly to the quality of heat-abused citrus juices
(Handwerk and Coleman 1988).
1-Ethyl-2-formylpyrrole
CHO
orange juice, which generally contains larger amounts of oxidative products due
to the additional heat required in processing. No data are available for other
citrus juices, but the levels would be expected to be much less. Excess amounts
of this compound impart a piney stale odor and flavor to heat-abused juices. Of
the six heat-abuse products that affect citrus juice quality, this is the only one
that actually incorporates the nitrogen from the amino acid into its structure,
illustrative of the role of amino acids in the oxidative process. Its boiling point
is 48 to 53°C.
2-Hydroxy-3-methyl-2-cyclopenten-1-one
HO
5-Methyl-2-furaldehyde
It can be detected at 10 ppm, compared to less than O.S ppm found in dehy-
drated orange juice. Adding a hydroxy group to the methyl group to give
5-hydroxymethyl-2-furaldehyde results in a compound with a taste threshold in
canned orange juice of over 200 ppm with a typical level of 14 ppm. Grapefruit
juice held at SO°C for 12 weeks can exceed this taste threshold. This hydroxy
compound is virtually absent in fresh juices and may serve as a general measure
of product oxidation. It has a boiling point of 114 to 116°C, a melting point of
32 to 3SoC, and an index of refraction of I.S627 20 .
2, 5-Dimethyl-4-hydroxy-3( 2H )-furanone
o
o
This compound is also known as furaneol, and is also produced from hexoses
and amino acids. It has a taste threshold of O.OS ppm in orange juice and imparts
a pineapple-like flavor when it occurs in excessive amounts.
2-Methoxy-4-etheneylphenol
OH
et al. 1988). It has a melting point of 138 to 139°C and is slightly soluble in
water and soluble in ethanol. It has a taste threshold of only 0.05 ppm, with a
typical canned orange juice level of 1 ppm (which is over the taste threshold).
Excessive amounts impart an old fruit or rotten flavor and odor.
(1- Terpineol
General Observations
All these compounds have very low taste thresholds, and some of them have
been shown to impart off flavors below their thresholds when combined with
other heat-abuse products. For example, 5-hydroxymethyl-2-furaldehyde, hy-
droxymethylcyclopentenone, and N-ethylpyrrole-2-carboxaldehyde will impart
an off flavor even though each exists at only half of its taste threshold. Also,
2,5-dimethyl-4-hydroxy-3(2H)-furanone, a-terpineol, and 4-vinyl guaiacol im-
part off flavors even though they occur at only half of their individual taste
thresholds.
Additional amounts of citric acid also have been shown to enhance Maillard
oxidation, an effect attributed to the buffering action of the citrate ion because
the rates of these oxidative reactions are pH-sensitive. Also, the methyl groups
of citric acid assist in catalyzing oxidation to some extent, and phosphate ad-
dition accelerates oxidation regardless of buffering effects.
Restricting the amount of available oxygen obviously reduces the rate of ox-
PROCESSING CONTAMINATION 265
idation of citrus products. Steam injection or vacuum filling minimizes off flavor
development. Deaeration of citrus juices has not been shown to significantly
reduce oxidation because there is always enough air in the juice to facilitate
oxidation. Increasing the dissolved oxygen increased the rate of browning in
lemon juice but did not affect the rate of ascorbic acid degradation (Robertson
and Samaniego 1986). In another study, raising the initial levels of dissolved
oxygen increased browning and ascorbic acid degradation in orange juice, but
reducing the initial dissolved oxygen levels did not extend the taste-panel-de-
termined shelf life of pasteurized orange juice (Trammell, Dalsis, and Malone
1986).
Even though large amounts of heat are used during most citrus processing,
the majority of heat abuse occurs during storage. Shelf-life studies of 66°Brix
orange concentrate have shown that a reduction of storage temperature to 5.5 ° C
(4.4°C to -1.1 °C) can increase the shelf life by 4 months (5 months to 9
months) (Marcy et a1. 1984). With storage temperatures of -6.6°C (20°F) or
below, orange concentrate can be stored for at least a year. These results are
comparable to what is found industrially. Juices being evaporated are generally
held at 90 to 95°C (195-205°F) for about 10 to 20 seconds, with a total reten-
tion time in the evaporator of about 10 minutes, depending on the evaporator.
Most of this time the temperature of the juice is around 35 to 70°C (100-
150°F). Processed concentrate can withstand maximum evaporator tempera-
tures for about 30 minutes. Aseptic juices and concentrates are known for their
burnt-off flavors, most of which occur during unrefrigerated or prolonged stor-
age.
Analytical Techniques
Citrus juice oxidation products, like most chemicals, can be measured by using
sophisticated or time-consuming analytical techniques, including mass spec-
trometry, gas chromatography, and so forth. These methods are generally too
expensive and involved for routine quality control, and are generally not found
in quality control laboratories. For quality control purposes, the general degree
of oxidation and the shelf-life determination are the main factors of interest.
The simplest method for measuring the general heat abuse of citrus juices is the
formol or formaldehyde test, which measures the level of primary amino acids.
Most of the amino acids in citrus juices are primary amino acids, the exception
being proline, which is a secondary amino acid. Proline occurs in significant
amounts in citrus juices and even constitutes the major amino acid in many
varieties. Because amino acids are consumed during Maillard oxidation, the
loss of amino acids can be used as an indicator of heat abuse. The formol test
involves a Strecker degradation of the amino acids, generating acidity according
266 CITRUS JUICE SANITATION
Formol Test
Procedure
1. Proceed to the endpoint (8.2 pH) as described in the section on acid ti-
tration (see Chapter 3), and continue titrating to a pH of 8.4.
2. Add 10 ml of the neutralized formaldehyde solution. The pH will drop.
Rezero the buret containing the NaOH solution.
3. Again titrate to a pH of 8.4, and read the ml of NaOH needed to do so.
The total amino acids (except proline) can be calculated by using:
where N is the normality of the base (0.1562N) and Ss and Sj are the Ib
solids/gal of the sample and 1l.8°Brix juice. Ss and Sj are the same if
the sample has been reconstituted or adjusted to 11.8°Brix. Sj is equal to
1.029 lb solids / gal. The amino acid level is usually expressed as milli-
equivalents amino acids per 100 ml of juice. Equation 18-3 can be con-
densed to:
Another method suggested for the monitoring of heat abuse in citrus juice
products is the measurement of the furfural level. Furfural is a product of as-
corbic acid degradation, but it does not contribute directly to off flavor devel-
opment in the juice. Even though some oxidation reactions occur at different
rates, the rate of furfural development is believed to be an adequate indicator
of heat abuse for quality control purposes. The advantages of this method in-
clude the measurement of a specific oxidative product, and the fact that because
furfural is essentially nonexistent in fresh juice, only one test is needed to es-
timate the heat abuse of the juice. Also, furfural is easily distilled from the juice
by using an apparatus common in citrus quality control laboratories and is con-
centrated sixfold in so doing. The main disadvantage is that it involves the use
of aniline, a toxic chemical that can cause intoxication when inhaled, ingested,
or absorbed through the skin. Consumption of 0.25 ml of aniline can cause
serious poisoning. Acute symptoms include cyanosis, methemoglobinemia,
vertigo, headache, and mental confusion. Chronic symptoms include anemia,
anorexia, weight loss, skin lesions, and bladder tumors. The maximum safe
level is only 5 ppm. Analysis using aniline should be done under expert super-
vision.
The furfural procedure involves the following reaction:
o CHO
01 + +
FURFURAL ANILINE
This is a form of the Strecker degradation that results in a highly colored con-
jugated product that can be detected with a spectrophotometer. The procedure
uses SnCl2 to stabilize the color formation as well as to decrease interference
from codistilled compounds.
268 CITRUS JUICE SANITATION
• Spectrophotometer.
• Oil distillation apparatus used in oil determination plus boiling chips or
anti foaming agent.
• 10 ml ground glass stoppered graduated cylinder.
• 2 ml pipette.
• One 13 x 150 mm test tube.
• 1 ml pipette.
• Furfural standards (0.5, 1.0, 2.0 p.g/ml in distilled water; make a 1: 100
dilution of a 1 mg/ml stock solution, and use 5, 10, and 20 ml of this
solution, diluting each to 100 ml).
• 95 % ethanol.
• Aniline / acetic acid solution. (Mix 2 ml of purified aniline with 1 ml 20 %
SnCl z • 2H zO in concentrated HCI. Mix and add glacial acetic acid to 20
ml.) Observe necessary safety precautions in working with aniline.
Procedure
For grapefruit juices, multiply the furfural level by 3. Off flavor detection
correlates with furfural levels of about 50 to 70 J1.g/liter in canned and glass-
PROCESSING CONTAMINATION 269
packed orange juice and 150 to 175 ILg/liter in canned or glass-packed grape-
fruit juice. Grapefruit juice has a greater tendency to mask the off flavors formed
from product oxidation and thus has a higher taste threshold.
If the equipment is available, HPLC can be used to measure the amount of
furfural, which would eliminate the need to handle the toxic aniline.
Procedure
• Rotary evaporator.
• 100 g NaC!.
270 CITRUS JUICE SANITATION
• 1500 ml beaker.
• Two I-liter separatory funnels.
• 1 liter methylene chloride.
• Na2S04 desiccant.
• 10 ml volumetric flask or graduated cylinder.
• GLC with a 9' x 0.25" Carbowax 20M pack stainless steel column (20%
on 60-80 mesh Gas Chrom P) with a helium flow rate of 200 ml/min.
The temperature is programmed at 80°C for 6 min., and increased to 130°C
for 6 min., 135°C for 14 min., 140°C for 24 min., 155°C for 30 min.,
180°C for 46 min., 190°C for 56 min., 200°C for 64 min., and 220°C
for 76 min.
• 100 ILl syringe for injection into GLC system.
Procedure
Standard
Compound Source GLC Ret. Time (min.)
a-Terpineol c 21.0
2,5-Dimethy l-4-hydroxy-3(2H)-furanone x 43.0
4-Vinyl guaiacol a 54.5
5-Hydroxymethyl furfural c 75.0
5-Methyl-2-furaldehyde c *
l-Ethyl-2-formylpyrrole b *
2-hydroxy-3-methyl-2-cyclopenten-1-one c *
Furfural c 12.5
a-Synthesized as described in references (Tatum et al. 1975).
b-Synthesized as described in references (Tatum et al. 1967).
c-Commercially available.
x-Not commercially available and no reported synthesis.
'-Not reported using this method.
PROCESSING CONTAMINATION 271
Brown Color
As mentioned earlier, during the oxidation of citrus juices browning follows off
flavor development. Browning with heat is common in nearly all foods. The
development of brown compounds in crude plantation white sugars has been
attributed to the development of humic acids, caramel, 5-hydroxymethyl fur-
fural, and melanoidins (Cheng, Lin, and Wang 1983). 5-Hydroxymethyl fur-
fural has already been mentioned as a remote participator in off flavor devel-
opment in oxidized citrus juices. The dark pigment in caramel is probably related
to the melanoidins, as is the pigments found in humic acids (Benzing-Purdie,
Ripmeester, and Preston 1983). Melanoidins have long been associated with
Maillard reactions (Spark 1969) and are the result of a slow polymerization of
reducing sugars along with inorganic salts. Melanoidins also are what consti-
tutes the brown pigment in skin. Even though the chemical structures of mel-
anoidins have not been determined, the related compound melanin has been
shown to be a complex quinoidal compound consisting of the following units:
When high Brix (58-65°Brix), low Brix/ acid ratio citrus concentrates are stored
in commercial freezers, crystallization of some of their organic salts may occur.
These crystals generally have a negative effect on the citrus juice, resulting in
lower USDA defects scores as well as clogging plate heat exchangers and other
processing equipment that have restricted flow areas. These crystals also do not
reconstitute as easily as citrus concentrates. Larger crystals usually are laced
with insoluble hesperidin, which further hampers reconstitution attempts.
The crystals are hard and have a characteristic acid taste. They range in size
from fine sand up to 3 mm in diameter, depending on how long the crystalli-
zation has been allowed to proceed. The composition of the crystals from Cal-
ifornia and Brazilian concentrates is illustrated in Table 18-1. As can be seen,
the Brazil concentrate and crystals have higher levels of potassium than the
California counterparts. The higher potassium levels resulted in higher acid
levels in the Brazilian crystals as well. This predominance of potassium and
citric acid in the crystals agrees with other studies (Hils 1973; Koch 1980;
Bielig et al. 1983). The stoichiometric relationship between potassium and citric
acid suggests that monopotassium citrate (MPC) is the primary species of salt
in the crystals. The citrate ion has so many sites for internal rotations that it
takes on a variety of shapes, leading to an amorphous or irregular crystal lattice
that gives rise to amorphous crystals. Also, other juice components, especially
cloud material, pulp, and hesperidin, become entrapped in or adhere to the
crystals, adding to their amorphous nature.
There are three main factors that affect the appearance of these crystals, one
being the thermodynamic favorability of the crystallization. Solubility tests have
shown that the Ksp of MPC in sucrose solutions follows this relationship:
where (CA) and (K+) are the concentrations of the citric acid and potassium
ions in moles/liter, and T is the absolute temperature between 253 OK ( - 20°C
or -4°F) and 293°K (+20°C or 68°F) (Kimball 1985). If the product of the
molar concentrations of citric acid and potassium ion exceeds the Ksp value,
crystallization is thermodynamically favored. It should be remembered that cit-
rus juices, like most other fruit juices, contain excess amounts of citrate beyond
that measured by acid titrations-up to 20% more citrate (Shaw, Buslig, and
Wilson 1983). Measurements of the citric acid and potassium levels for navel
and Valencia concentrate throughout a season in California are shown in Fig.
18-2 (Kimball 1985). The dashed line at the bottom is the Ksp value calculated
from Equation 18-8 at O°C. As can be seen, throughout the entire season for
both varieties of orange concentrate the Ksp value was exceeded, a situation
that would strongly favor MPC crystallization. This suggests that the high-Brix
--
'"> . 0.10
.....
("(S
0
-E •
0
-50 o 50 100 150 200 250 300
Day Processed
Fig. 18-2. The product of the citric acid and potassium concentrations for California navel and
Valencia concentrates during a processing season. The dashed line represents the Ksp value cal-
culated from Equation \8-8 for O°C (Kimball 1985). (Reprint from Food Technology 1985, 39(9),
79-8\, copyright © by the Institute of Food Technologists.)
274 CITRUS JUICE SANITATION
70
65 ° Brix Concentrate
60 Supersaturated
in MPC
50
40
x
.;:: Unsaturated
in MPC
~ 30
Frozen
Juice \
\
20 \
\
\
\ Single Strength Juice
10 \
\
\
\
OL-----~--~~--~~--~~--~~----~--~~--~----~
·20 ·15 ·10 ·5 o 5 10 15 20
Temperature (0C)
Fig. 18-3. Saturation curves for California navel and Valencia orange juices for MPC crystalli-
zation as a function ofBrix and temperature (Kimball 1985). (Reprint from Food Technology 1985,
39(9),79-81, copyright © by the Institute of Food Technologists.)
+lV>+
PROCESSING CONTAMINATION 275
SO'in~i~~ty
Uti7 ~~~~id
o
&-
\utions
ffi
crystalization in
concentrated solutions
and/or acid solutions Fig. 18-4. The effect of acid on MPC
o
crystallization. The solubilizing water mol-
ecules favor the hydrogen ion. leaving the
potassium to crystallize with citrate (Kim-
ball 1985). (Reprint from Food Technology
1985,39(9),79-81, copyright © by the In-
Citrate stitute of Food Technologists.)
where D is the diffusivity factor, V is the ionic volume, C is the ionic concen-
tration, S is the ionic solubility, t is the time, and q is the charge of the ionic
species. The diffusivity factor is generally constant in dilute solutions, but here
this factor is significantly affected by the Brix, temperature, and acid content
of the concentrate. The acid levels can be reduced by ion exchange or blending.
It is desirable to keep the Brix as high as possible in order to minimize storage
and shipping costs. The temperature can be controlled somewhat. Bulk storage
tanks cannot be kept much below -lOoC (about 15°F) because if the temper-
ature is too low, the concentrate becomes too viscous to be pumped in and out
of the storage tank, but drum freezers are kept much colder (-20°C or O°F).
The higher heat needed by bulk storage systems increases the ion mobility in
276 CITRUS JUICE SANITATION
QUESTIONS
1. What quality problem does hesperidin produce in citrus juices, and how
can this problem be eliminated?
2. During what part of the season is hesperidin contamination the biggest
problem, and why?
3. What causes black flakes to form, and how can this contamination be
eliminated?
4. How many black flakes are generally acceptable in citrus concentrates?
5. What compounds in citrus juices are mostly responsible for off flavor
development during excessive heating?
6. Which of the six heat-abuse products mentioned in the chapter imparts
a pineapple-like flavor?
7. Which two of the six heat-abuse products mentioned in the chapter that
affect flavor are not produced by Maillard reactions between sugars and
amino acids or from ascorbic acid degradation?
PROCESSING CONTAMINATION 277
PROBLEMS
1. In the formol test, using 9.83 ml of 60.2°Brix concentrate for both tests, would the
concentrate be expected to have developed off flavors if the initial test titrated 9.82
ml of the base and the test after heat abuse titrated 9.51 ml of the base?
2. In the furfural test, suppose that the absorbance matched the 1.02 Ilg/ml standard.
Would this orange juice be expected to have off flavors from heat abuse?
3. What would be the Ksp value for MPC at O°C, the freezing point of water? (O°C =
273°K)
4. Would a concentrate containing 0.15M citric acid and O.IOM potassium ion be fa-
vored to form MPC crystals at the same temperature mentioned in problem 3?
5. About how long would it take to form 2 mm crystals in 60 Brix concentrate with a
0
REFERENCES
Benzing-Purdie, L., Ripmeester, J. A., and Preston, C. M. 1983. Elucidation of the nitrogen forms
in melanoidins and humic acid by nitrogen-IS cross polarization-magic angle spinning nuclear
magnetic resonance spectroscopy, 1. Agric. Food Chern., 31, 913-915.
Bielig, H. 1., Hofsommer, H. J., Fischer-Ayloff-Cook, K. P., and Bakke, K. J. 1983. Crystalline
precipitation in frozen orange juice concentrates, Flussiges Obst, 50(3), 105.
Cheng, H. T., Lin, W. F., and Wang, C. P. 1983. Studies on the color development in stored
plantation white sugars. In The Maillard Reaction in Foods and Nutrition, G. R., Waller and
M. S. Feather, eds. ACS Symposium Series 215, American Chemical Society, Washington,
DC., 91-102.
Handwerk, R. L. and Coleman, R. L. 1988. Approaches to the citrus browning problem, J. Agric.
Food Chern., 36, 231-236.
Hils, A. 1973. Identification of a crystalline substance (K-citrate) in orange concentrate, Flussiges
Obst, 40(12), 496.
Kimball, D. A. 1985. Crystallization of potassium citrate salts in citrus concentrates, Food Tech.,
39(9),76-81,97.
278 CITRUS JUICE SANITATION
Koch, J. 1980. Formation of monopotassium citrate (MPC) in frozen concentrated orange juice
(FCOJ) at 65°Brix. Presented at the 31 st Annual Citrus Processors' Meeting, Lake Alfred, Fla.,
September 10, 1980.
Marcy, J. E. and Rouseff, R. L. 1984. HPLC determination of furfural in orange juice, J. Agric.
Food Chern., 32, 979-981.
Marcy, 1. E., Graumlich, T. R., Crandall, P. G., and Marshall, M. R. 1984. Factors affecting
storage of orange concentrate, J. Food Sci., 49, 1628-1629.
Nairn, M., Striem, B. 1., Kanner, J., and Peleg, H. 1988. Potential offerulic acid as a precurser
to off-ftavors in stored orange juice, J. Food Sci., 53(2), 500-504.
Robertson, G. L. and Samaniego, C. M. L. 1986. Effect of initial dissolved oxygen levels on
degradation of ascorbic acid and the browning of lemon juice during storage, J. Food Sci., 51 (I),
184-187, 192.
Shaw, P. E., Buslig, B. S., and Wilson, C. W. III. 1983. Total citrate content of orange and
grapefruit juices, J. Agric. Food Chern., 31(1),182-184.
Spark, A. A. 1969. Role of amino acids in non-enzymatic browning, J. Sci. Food Agric., 20, 308-
316.
Tatum, J. H., Nagy, S., and Berry, R. E. 1975. Degradation products formed in canned single
strength orange juice during storage, J. Food Sci., 40, 707-709.
Tatum, J. H., Shaw, P. E., and Berry, R. E. 1967. Some compounds formed during nonenzymatic
browning of orange powder. J. Agric. Food Chern., 15, 773.
Ting, S. V. and Rouseff, R. L. 1986. Citrus Fruits and Their Products. Marcel Dekker, Inc., New
York, 177-178.
Trammell, D. J., Dalsis, D. E., and Malone, C. T. 1986. Effect of oxygen of the taste, ascorbic
acid loss, and browning for high-temperature-short-time pasteurized single strength orange juice,
J. Food Sci., 51(4), 1021-1023.
Chapter 19
Juice Adulteration
Fraudulent representation of foods and medicines has a long history. In the days
of the wild west, snake oils and medicinal tonics were sold to gullible passersby
at a price that was hard to tum down. These concoctions were consumed in the
hope of there being at least a remote possibility that some of their claims would
be valid but many of these "cure-aIls" caused more damage than they reme-
died.
In 1938, the United States Congress enacted the Federal Food, Drug, and
Cosmetic Act. Even though there was an earlier law, and the Act has been
modified somewhat since, it remains essentially intact today. This law gave
definition to the term "adulteration," which is described in section 402 of the
Act. The Food and Drug Administration (FDA) has the responsibility for en-
forcing this law, which requires different emphases and methods depending on
which type of adulteration is involved. The four basic types of adulteration are:
(1) filthy, putrid, decomposed, or harmful; (2) unsafe or unsanitary; (3) statu-
torily unsafe; and (4) economic (Stutsman 1988).
The first kind of adulteration is a concern for the manufacturers of any prod-
uct. Chapter 17 covers a portion of this type of contamination. There is no
product made that cannot be dangerous in the wrong situation. The production
process for any consumable food, such as citrus juices, is not totally invulner-
able to the possible addition of poisons or harmful materials. In recent times a
great deal of attention has been paid to packaging that prevents the addition of
dangerous substances by persons other than the manufacturer. Citrus juices often
are packaged in cans, plastic containers, or cartons that are difficult to tamper
with in retail stores. Industrial processing and storage areas should maintain a
security system to prevent outsiders from roaming around the plant. Processing
records should be detailed enough that any problems can be traced back to
specific employees. This way employees can be held accountable for their own
actions, in addition to serving as watch dogs for the product during processing.
The second type of adulteration involves processing in such a manner that
the food may become adulterated. Because the FDA cannot continually inspect
279
D. Kimball, Citrus Processing
© Van Nostrand Reinhold, New York, NY 1991
280 CITRUS JUICE SANITATION
STANDARDS OF IDENTITY
The standards of identity for various citrus products are contained in the Code
of Federal Regulations and the Federal Register. These standards and their
interpretation may change from time to time, and the current status of such
standards should be monitored by quality control management with the aid of
legal counsel if necessary. With this charge in mind, the following constitutes
a brief description of the most current standards of identity for citrus products
available:
Orange Juice (CFR Title 21 146.135). This product contains juice from un-
JUICE ADULTERATION 281
fermented fruit of sweet orange varieties (c. sinensis) with seeds and excess
pulp removed. The product may be chilled but not frozen.
Frozen Orange Juice (CFR Title 21 146.137). This product is the same as
orange juice except that it is frozen.
Pasteurized Orange Juice (CFR Title 21 146.140). This product is the same
as orange juice except that pasteurized orange juice can contain up to 10% of
tangerine juice (c. reticulata) or its hybrids. Also, oil and pulp (not washed or
spent pulp) can be added. Concentrated orange juice, sugar, invert sugar, dex-
trose dried com sirup, or dried glucose sirup can be added to raise the Brix to
a normal range for orange juice, but such addition must be declared on the label.
The juice can be heat-treated to reduce enzyme activity and the number of viable
microorganisms. This juice can be frozen and must have a Brix of at least 10.5,
not including added sweeteners. The B / A ratio must be at least 10.
Canned Orange Juice (CPR Title 21 146.141). This product is the same as
pasteurized orange juice with certain additional characteristics: the condensate
water from deoiling may be added, the minimum Brix is 10, and the minimum
B / A ratio is 9. Also, if the juice is not refrigerated, the word "canned" can be
omitted from the label.
Orange Juice from Concentrate (CPR Title 21 146.145). Orange juice from
concentrate contains reconstituted frozen concentrated orange juice or orange
concentrate for manufacturing and may contain orange juice, pasteurized orange
juice, orange juice for manufacturing that has been preserved by freezing and
not canning, orange oil, unwashed pulp, and water. It may contain the same
declared sweeteners as pasteurized orange juice, and may be heated or pasteur-
ized. It must contain at least 11.8°Brix exclusive of any added sweeteners.
Frozen Concentrated Orange Juice (CPR Title 21 146.146). This product
may contain 10% tangerine juice or its hybrids (c. reticulata) and/or 5 % sour
orange juice (c. aurantium). It may contain water, orange oil, orange essence,
unwashed pulp, and the same declared sweeteners as above. The product can
be heat-treated. In order to reconstitute this product to not less than 11. 8 °Brix,
the dilution ratio should be not less than 3 parts water to 1 part concentrate.
Reduced Acid Orange Juice (CPR Title 21 146.148). This product is the
same as frozen concentrate orange juice except that it must employ the use of
anionic ion exchange as permitted in 173.25 of Title 21 in reducing the acid
levels, and the B / A ratio must be between 21 and 26.
Canned Concentrated Orange Juice (CPR Title 21 146.150). This product
is the same as frozen concentrated orange juice except that it is not frozen, and
it is canned and heat-treated so as to prevent spoilage.
Orange Juice for Manufacturing (CFR Title 21 146.151). This product is the
same as orange juice except that it can have a lower Brix and B / A ratio than
required for orange juice and can contain up to 10% tangerine juice or its hy-
brids (c. reticulata). Unwashed pulp and orange oil also may be added, and it
may be heat-treated, chilled, and/or frozen.
282 CITRUS JUICE SANITATION
are defined as food additives, they can be used, provided that they comply with
section 409 of Title 21.
Economic Fraud
artificial colors to mask poor juice color would be a violation of the third charge.
The addition of cheaper carbohydrates would be a clear violation of the fourth
charge.
There are many types of economic adulteration. However, nearly every type of
adulteration that has been known to occur in the industry falls into one of four
different categories: water addition, carbohydrate addition, cover-up, and
blending of unauthorized juices or juice products.
Water Addition
Some of the standards of identity exclude the addition of water. These standards
involve single-strength juices that are neither concentrated nor made from con-
centrate. These juices include lemon juice, grapefruit juice, orange juice, frozen
orange juice, pasteurized orange juice, canned orange juice, orange juice for
manufacturing, orange juice with preservative, and canned tangerine juice. The
last is defined under USDA grade standards rather than as a separate standard
of identity. On the other hand, the citrus juice products whose standard of iden-
tity allows for the addition of water include orange juice from concentrate,
frozen concentrated orange juice, reduced acid frozen concentrated orange juice,
concentrated orange juice for manufacturing, and concentrated orange juice with
a preservative.
Water comprises over 80% of the composition of natural juices, so detection
of its illicit addition can be difficult; but it can be done in one of two ways.
One method is to monitor the mineral content of the juice. Natural juices have
a certain range of mineral levels as follows:
Orange Grapefruit
Sodium <50 ppm <50 ppm
Potassium > 1400 ppm 300-500 ppm
Calcium 65-120 ppm 100-150 ppm
Magnesium 95-170 ppm 90-140 ppm
Phosphorous 120-310 ppm
Potable water from the city, which is used in many plants, may contain much
higher levels of these minerals, and high mineral levels would indicate illicit
water addition. Because pulp washing utilizes plant water, high mineral levels
also may be indicative of pulp wash addition although legal water addition can
occur in citrus concentrates without the addition of pulp wash. Minerals can be
determined by using a variety of methods, which usually involve instrumenta-
JUICE ADULTERATION 285
tion beyond the resources of most quality control laboratories. Professional lab-
oratories provide an economic means of determining mineral levels.
Treated water naturally low in minerals could not be detected through mineral
analyses. However, a method has been developed that can distinguish between
water that is natural to the juice and water added from other sources (Winters
et al. 1988; Brause et al. 1984). This method involves the measurement of the
oxygen isotope ratios. When water is absorbed into the plant through the root
system, it has a tendency to lose the lighter IH and 160 isotopes through eva-
potranspiration and to keep the heavier deuterium eH) and 18 0 isotopes. The
oxygen isotope ratios are determined by using a complex procedure, also rec-
ommended for professional laboratories, and the results are reported as 0 180 by
use of the following equation:
In this equation the R values are the 180 / 6 0 isotope ratios. For concentrated
orange juice above 60 o Brix, the 0 180 value should be above + 10. For authentic
single-strength orange juice the 0 18 0 value should be positive. Anything less
than these values is indicative of water addition.
Carbohydrate Addition
Because citrus juices are comprised primarily of carbohydrates and water, the
addition of cheaper carbohydrates to authentic juices has been a lucrative way
to falsify citrus products. Citrus juices are sold by the weight of the soluble
solid equivalent of carbohydrates. When commercial sugars are sold as low as
5C lIb, and the equivalent orange juice price is up to $2.00 lIb soluble solids,
one can see how much the profit margin can be increased with substitution of
commercial sugars or similar products.
Several methods can be used to detect carbohydrate adulteration. One of the
simplest is based on the sugar composition of the juice itself. Citrus juices
contain sucrose, glucose, and fructose, in a ratio of about 2: 1 : 1. Glucose and
fructose are known as reducing sugars because they can be oxidized by using
mild oxidizing agents. Sucrose, however, cannot be oxidized by mild oxidizing
agents because its anomeric carbon forms the bond between its glucose and
fructose monomers. Thus a measure of the reducing sugars, a relatively easy
determination, should result in the determination of about half of the carbohy-
drates, which in tum constitutes about 80 to 90% of the total soluble solids of
the Brix value. The use of high-fructose com syrup or cane sugars (predomi-
nately sucrose) can be detected quickly by the following method. However,
invert beet sugars or special combinations of reducing and nonreducing sugars
286 CITRUS JUICE SANITATION
cannot be detected from a reducing sugar test; so this method should constitute
a preliminary screening and not serve as a verification of authenticity.
The following reducing sugar test is based on the reduction of Cu 2 + to Cu I +
by the reducing sugar. The Cu 2 + ions usually form a light blue color in aqueous
solution, when they are reduced to Cu 1+, an orange-colored oxide precipitate
is formed, according to the following equation:
Procedure
4. Titrate the copper solution while stirring with the sample serum until a
dark orange color is observed. Add 3 drops of the indicator, and continue
titrating until the bright orange of the copper oxide is fully restored.
5. The % reducing sugars (RS) can be calculated from the following:
For example:
23.93/4.77 ml titrated = 5.02 % reducing sugars
(19-5)
It has been found that there is a difference in this ratio for different plants in
nature. Simpler C4 plants, such as com and sugar cane, have O\3C values around
-10, whereas higher C 3 plants, such as flowers and trees (citrus), have a O\3C
value below -22. Therefore, if sugar from cane or com is used in citrus juices,
a higher D\3C value will be observed. The analytical methods required for such
a determination are, again, expensive and complex and best left to professional
laboratories.
Another procedure has been developed (Low 1989) to detect invert beet sugar
down as low as 1 %, and perhaps even as low as 0.1 %. This method employs
the use of HPLC with a pulse amperometric detector that is several orders of
magnitude more sensitive than refractive index detectors. The method detects
the adulterating carbohydrates by determining the oligosaccharide pattern. With
this new technique, any illicit carbohydrate adulteration can be detected.
Cover-Up
When juice has been adulterated, it often becomes necessary to add something
else in order to cover up the adulteration. For example, one of the simplest tests
288 CITRUS JUICE SANITATION
used in detecting the illicit addition of pulp wash has been the formol or for-
maldehyde test, which measures the amount of primary amines. Pulp wash has
a lower concentration of amino acids than authentic juice has, and its illicit
addition may be detected by this very fast and simple technique. However, it
is possible to get around the test with the addition of very cheap amino salts-
the addition of the amino salts being a second adulteration used to cover up the
first adulteration. Because it is so easily circumvented, the formol test has be-
come a weak test for juice adulteration. However, it is useful in other parts of
citrus quality control, as mentioned in the preceding chapter, and as a prelim-
inary adulteration screening.
Another common cover-up technique is the addition of colors. When a juice
is adulterated with something other than juice, it must become more dilute in
authentic components including colors. The addition of turmeric, annatto, or
other dyes (FD&C dyes), which makes the juice a darker orange, has been used
in adulteration cover-ups. However, these dyes can be easily tested for by any
quality control laboratory using the following techniques.
Turmeric is prepared from the dried roots of the tuberous rhizomes of Cur-
cuma tonga L., a member of the ginger family. It is native to southern and
southeastern Asia and is commercially cultivated in Malaysia. The color com-
ponent of turmeric is curcumin, also known as turmeric yellow, with the fol-
lowing chemical structure:
<OCH 3
rCH=CH-<O/OH
CH 2
~OCH=CH-<0 (-OH
OCH 3
Annatto dyes are natural carotenoid colorants derived from the seed of the
tropical annatto tree or Bixa orellana. The primary color component is bixin, a
carotenoid with the following chemical structure:
CH 3 CH 3 H H CH 3
I I I I I
H COOC-CH C CH C C~ Ct-i C=C C CH 3COOH
3 ~/~/ ~/~/ ,/" \//\//
CH CH CH CH CH C CH CH
I
CH 3
JUICE ADULTERATION 289
The water-soluble fonn of bixin is norbixin, which is fonned from the sa-
ponication of the methyl ester group. A single test can be used to detect either
annatto or tunneric. Even though you can continue the test to detennine which
dye has been added, the presence of either dye indicates adulteration.
Procedure
This test is a pass/fail test. Any discernible deepening of the color in the last
hexane layer indicates the presence of these dyes.
The Food and Drug Act of 1906 certified 7 dyes that could be used in foods.
By 1938, when the Food, Drug, and Cosmetic Act was passed, 11 more dyes
were included, and thereafter these dyes were referred to as FD&C dyes by
number. The azo dyes, FD&C Yellow No.5 and FD&C Yellow No.6, have
been used to restore diluted orange juice color. FD&C No. 5 is also called
tartrazine or hydrazine yellow. It has the following chemical structure:
In 1856 dyes from coal tar were discovered, and for the next 80 years these
dyes were used extensively in food products. More modem synthetic techniques
generally have replaced the extraction of these dyes from coal tar. The presence
of coal tar dyes and of FD&C dyes can be detected simultaneously using the
following simple test. Again this is a pass/fail test. Any detection of these dyes
would constitute adulteration in 100% orange juice products.
• 1 g of calcium carbonate.
• Acetone and petroleum ether in the ratio of 1 : 3.
• 500 ml separatory funnel.
JUICE ADULTERATION 291
Procedure
Most orange juice products are allowed to contain up to 10% tangerine or tan-
gerine hybrids while still being called 100% orange juice. The high color of
tangerine juice is used to enhance the appearance of orange juices of weaker
color, primarily in Florida. However, other fruit juices usually are less expen-
sive than tangerine juice and have been illegally added to orange juices in direct
violation of the standards of identity. Apple, pear, white grape, orange pulp
wash, and grapefruit juices usually are sold at lower prices than that of orange
juice, and when they are blended into orange juice products, their economical
returns are augmented greatly. Apple, pear, and white grape can be detected by
using isocitric tests or other more complex testing schemes based on composi-
tional differences. Grapefruit juice contains the characteristic compound narin-
gin, so naringin determinations can be used to detect grapefruit adulteration in
orange juice (see Chapter 10 for analytical methods).
Federal law allows pulp wash to be added to citrus juices on-line in an amount
that constitutes about 5 to 10% of the juice, but if the pulp wash is separated
and packaged separately, it can no longer be legally added to other citrus prod-
ucts. Florida law prohibits any addition of pulp wash to any 100% citrus juices,
and in order to detect illicit pulp wash addition, Florida requires the addition
of sodium benzoate to all pulp wash juices to act as a tracer. If sodium benzoate
is found in 100% orange juice products, it is assumed that Florida pulp wash
292 CITRUS JUICE SANITATION
Several adulteration screening procedures are used around the world, ranging
from special computerized packages available from a single manufacturer to
specialized techniques developed by individual laboratories to government-reg-
ulated procedures and criteria. These procedures are constantly in a state of
development and change, and new methods and technologies are constantly
appearing (Nagy, Attaway, and Rhodes 1988; Fruit Juice Adulteration Work-
shop 1989). However, most of these techniques are beyond the scope and re-
sources of most industrial quality control laboratories; so professional labora-
tories usually are consulted regarding the analytical work. However, a
knowledge of the standard methods can be helpful to any citrus quality control
facility.
Europe
Northern Europe has seen the establishment of some of the most sophisticated
adulteration procedures in the world, and presently three standards are being
used in Europe: the German RSK (Hofsommer 1989; Association of the Ger-
man Fruit Juice Industry 1987), the French AFNOR (Hofsommer 1989; Union
Nationale des Producteurs de Jus de Fruits 1989), and a criterion used by the
Netherlands and Belgium, referred to here as the Dutch standard (Hofsommer
1989; Warenwet P.B.O.-voorschriften (E-17a) 1977). The European Commu-
nity is working toward a single EC standard but has encountered some diffi-
culty. One problem, for example, is that many different products are of concern
to these countries. The following listing shows the juice products that each
standard covers:
RSK
AFNOR (French) (German) Dutch
Apple Apple Apple
Grape Apricot Grape
Grapefruit Black currant Grapefruit
Lemon Grape Orange
Orange Grapefruit
Pineapple Lemon
Tomato Orange
Passion fruit
Pear
Raspberry
Sour Cherry
JUICE ADUL TERA TION 293
United States
As yet, the United States does not have an official standard for authentic juices
that compares with the European standards. The standards of identity simply
declare that the juice must be authentic and describe the natural components
permitted in it, such as pulp, citrus oils, and so on. When citrus juices are
imported into the United States, or when a commercial product is suspected of
adulteration, any and all means can be used to prove such adulterations, in-
cluding employee depositions, chemical tests, and/or inspection of company
records. Imported juices usually go through a screening process by customs and
FDA officials that is somewhat easy to monitor, even though it is impossible to
thoroughly analyze every imported lot. However, the inspection of domestic
juice is not required, and the monitoring of adulteration is much more difficult
domestically. The most likely person to detect domestic adulteration is a com-
petitor or a customer of the juice adulterator. Adulteration may be suspected if
a product is selling for too Iowa price, or an in-house screening procedure may
unearth suspect material. A competitor may wish to notify the FDA and pursue
criminal prosecution of the adulterator, while a customer may wish just to cease
doing business with a suspected supplier. The latter does not require as much
rigorous proof.
Even a rumor that a company is adulterating its juices can cause it difficulty
in marketing its products. Even if a company purchases blend components from
a suspected company, the purchaser's reputation may be on the line. Not only
are reputations of companies at stake, but also the reputations of management
personnel, especially quality control management. It would be difficult to con-
vince a judge that quality control personnel were entirely ignorant of adultera-
tion that was going on within the company because it is their specific job to
monitor the composition of the company's products.
The adulteration tests are quite numerous and complex, including sophisti-
cated computerized systems that are now on the market. However, the resources
of most quality control facilities are usually quite limited and exclude the bulk
of these techniques. Professional laboratories have tried to find an economic
and thorough method of adulteration detection that can be of use to any citrus
quality control laboratory. One such method is the matrix method of detecting
juice adulteration (Brause et al. 1984), which consists of the measurement of
294 CITRUS JUICE SANITATION
In-house Methods
The purpose of this chapter is to acquaint citrus quality control personnel with
the problem of citrus juice adulteration and some of the things that have been
and are being done to correct the problem. Every quality control laboratory
should have some sort of screening system for all inbound products used in the
formulation of its products. Simple chemical tests such as those mentioned here
can constitute preliminary screenings. If these tests or information gathered from
other sources suggest the possibility of adulteration, more elaborate tests should
be performed. Periodic testing by professional laboratories is highly advisable,
and such spot checks are best done randomly.
Some observers have suggested the establishment of in-house or company
standards for the products the company purchases. The establishment of spec-
ifications is a common practice, but it rarely if ever includes standards specifi-
298 CITRUS JUICE SANITATION
QUESTIONS
1. When was the Food, Drug, and Cosmetic Act enacted, and in what sec-
tion of the Act is adulteration specifically mentioned?
JUICE ADULTERATION 299
2. What are the four types of adulteration according to the FDA? Give an
example of each.
3. What are the four charges or descriptions of adulteration according to
Section 402(b) of the Act? Give an example of each.
4. What are the four main types of economic adulteration that occurs in the
citrus industry?
5. What citrus products can one legally add water to?
6. What is the surest way to detect water addition to citrus juices? Then
why are other methods used?
7. Why are sugar or sugar products added to citrus juices?
8. What limitations does the carbon isotope ratio determination have in de-
tecting illicit carbohydrate addition to citrus juices?
9. What is the surest method to detect adulteration cover-up, and why is it
so definitive?
10. How much grapefruit juice can one blend with concentrated orange juice
for manufacturing? How much tangerine juice?
11. How much pulp wash (%) can be contained in orange juice according
to Florida standards? U.S. standards?
12. What are the 3 standards used in Europe, and to which countries do they
apply?
13. What is the procedure used by the U.S. government (FDA)?
14. Who does most of the policing of adulteration in the United States?
15. What is the adulteration screening system used by your company? What
feasible screening process would you recommend?
PROBLEMS
1. What could you say for an orange juice sample that had the following mineral com-
position?
Na: 55 ppm
Ca: 126 ppm
K: 1565 ppm
Mg: 165 ppm
P: 306 ppm
2. Suppose that in three reducing sugar tests you titrated the following volumes (ml).
Give the % reducing sugars and interpret the results. What would be the appropriate
action in each case?
3. What would 0 180 value of -0.15 indicate? What would a ol3e value of -20.5 in-
dicate?
4. A typical analysis of single-strength orange juice in the United States might be 0.65 %
w /w acid as citric acid, with a Brix/acid ratio of 18.2. Would this product pass the
RSK screening? Explain why or why not, and give reasons why the RSK may be
faulty in some circumstances.
5. Would the product in problem 4 meet the Dutch standards? How about the AFNOR
standards?
REFERENCES
AOAC 1980. Official Methods of Analysis, 13th Edition, 14.154. Association of Official Analytical
Chemists, Washington, D.C.
AOAC. 1984. Official Methods of Analysis, 14th Edition, 31.034-31.036. Association of Official
Analytical Chemists, Washington, D.C.
Association of the German Fruit Juice Industry. 1987. RSK Values, The Complete Manual. Fliis-
siges Obst GmbH, Bonn, Federal Republic of Germany, 1987.
Brause, A. R., Ratennan, J. M., Petrus, D. R., and Doner, L. W. 1984. Verification of authen-
ticity of orange juice, J. of the A. O.A. C., 67, 535.
Bricout, J. and Koziet, 1. 1987. Control of the authenticity of orange juice by isotopic analysis, J.
Agric. Food Chern., 35,758-760.
Fruit Juice Adulteration Workshop, August 9-10, 1989, Herndon, Va. Sponsored by General Phys-
ics Corporation, Columbia, Md.
Hofsommer, H. 1989. Analytical methodologies of detection as practiced in Europe. Presented at
the Fruit Juice Adulteration Workshop, August 9-10, 1989, Herndon, Va. Sponsored by General
Physics Corporation, Columbia, Md.
Low, N. 1989. Detection of beet medium invert sugar adulteration in high carbohydrate foods.
Presented at the Fruit Juice Adulteration Workshop, August 9-10, 1989, Herndon, Va. Spon-
sored by General Physics Corporation, Columbia, Md.
Nagy, S., Attaway, J. A., and Rhodes, M. E. eds. 1988. Adulteration of Fruit Juice Beverages.
Marcel Dekker, Inc., New York.
Page, S. W. 1986. Pattern recognition methods for the determination of food composition, Food
Tech., 40(1/) , 140-109.
Stutsman, M. J. 1988. Fruit juice adulteration, an overview of compliance and regulatory issues.
Presented at the Fruit Juice Adulteration Workshop, July 21, 1988, Herndon, Va. Sponsored by
General Physics Corporation, Columbia, Md.
Ting, S. V. and Rouseff, R. L. 1986. Citrus Fruits and Their Products. Marcel Dekker, Inc., New
York, 188.
Union Nationa1e des Producteurs de Jus de Fruits. AFNOR-UNPJF Fruit Juices Specifications.
Association Francaise de Normalisation, 1989.
Warenwet P. B. O.-voorschriften (E-17a), Verordening van 9 December 1976, Vb.Be.aft. 9, d.d.
7 -3-1977 van bei Produkischap van Groenten en Fruit.
Winters, K., Scalan, R. S., and Parker, P. L. Detection of orange juice adulteration by stable
isotope analyses. Presented at the Fruit Juice Adulteration Workshop, July 21, 1988, Herndon,
Va. Sponsored by General Physics Corporation, Columbia, Md.
Chapter 20
Besides citrus juices, other food-grade products are manufactured from citrus
fruits. The most common among them are the citrus oils and essences, discussed
in Chapter 6. Pulp products also are very common, and were discussed in Chap-
ter 7. These flavoring and texturizing materials are used extensively in juice
products themselves, as well as in other foods. Other food-grade by-products
will be discussed in this chapter, including pectin, jellies, jams, fruit sections,
juice drinks, and purees. Peel extracts are commonly used as beverage bases in
many parts of the world, mainly as a result of improved debittering techniques.
Citrus flavors, including encapsulated flavors, are popular, but their production
involves technology beyond the scope of this book. Much of this product in-
formation is proprietary, as confidentiality is necessary in order to maintain a
healthy competitive climate. However, basic principles and products common
to the industry are discussed here.
The manufacturing of jellies and jams using pectins has been in existence since
the French chemist Braconnat used pectins to make jellies as early as 1820.
However, it was not until the 1900s that large-scale pectin production emerged,
with the first pectin plant being built in Corona, California by the California
Fruit Growers Exchange.
Pectin Production
Pectin has been called "nature's glue." It consists of long and complex chains
primarily in the form of polygalacturonic acid units (a linear galacturonglycan
of a-(1-4 )-linked D-galactopyranosyl-uronic acid) with molecular weights of
100,000 to 200,000 (see Chapter 8). Pectin has a solubility in hot water of about
2 to 3 %, yielding a pH of 2.0 to 3.5. It belongs to a group of related compounds
303
D. Kimball, Citrus Processing
© Van Nostrand Reinhold, New York, NY 1991
304 CITRUS JUICE BY-PRODUCTS
are in the neighborhood of 3 % of the peel weight. The high cost of the solvents
used in pectin extraction justify the recovery and reuse of the wash solutions.
The Food Chemicals Codex (FCC) has established specifications for com-
mercial pectin along with procedures for measuring the required parameters
(Food Chemicals Codex 1931), and these specifications are illustrated in Table
20-1. As mentioned earlier, one of the quality problems of using technical grades
of acids in processing is the presence of heavy metal contaminants. The FCC
specifications for acid-insoluble ash, total ash, heavy metals, and lead are de-
signed to account for these contaminants and keep them at safe levels. Arsenic
contamination also must be accounted for in processing acids. The solvent con-
tent and the degree of esterification are parameters of the pectin itself, and the
sodium methyl sulfate specification provides for measuring the degree of de-
methoxylation that has occurred in the pectin as a result of pectinase enzymes
or acid hydrolysis. The FCC procedures for the determination of arsenic, lead,
and sodium methyl sulfate are not given here because they involve the use of
highly toxic chemicals or complex procedures not suited to routine quality con-
trol use by laboratory technicians. It is recommended that atomic absorption or
ion chromatography be used instead of FCC procedures. Sending spot samples
to professional laboratories also is a viable alternative if adequate in-house fa-
cilities do not exist. The following procedures are given by the FCC (Food
Chemicals Codex 1931).
Total Ash
• Crucible .
• Crucible tongs.
306 CITRUS JUICE BY-PRODUCTS
Procedure
Wash ( 100%)
Ash = ----"""--'-----'-- (20-1)
t Wsample
For example:
Acid-Soluble Ash
Procedure
1. Add the ash remaining from the total ash detennination to a boiling flask,
and add 25 ml of dilute HCl. Boil the mixture for 5 minutes.
2. Collect the insoluble material on a tared ashless filter paper or in a Gooch
crucible, and wash it with hot water.
3. Ignite the solid material, and after complete combustion cool it in a des-
iccator.
4. Weigh the remaining ash, and calculate the acid-soluble ash in the same
way as done with the total ash in the previous procedure.
Degree of Esterification
• 250 ml beaker.
• Magnetic stirrer.
• 250 ml Erlenmeyer flask.
• 100 ml graduated cylinder.
• 500 ml distillation flask with a Kjeldahl trap and a water-cooled condenser.
• Acid titration setup shown in Fig. 3-8.
• Balance.
• Drying oven (105°C or 221 OF).
• 30 to 60 ml fritted disc filter funnel of coarse porosity and vacuum source.
• About 50: 50 HCl, ethanol washing solution.
• 60% isopropyl alcohol (lPA) solution in water (about 100 ml).
• Anhydrous isopropyl alcohol (20 ml).
• Ethanol (2 ml).
• Phenolphthalein indicator solution (l % in IPA).
• Methyl red indicator solution (1 % in IPA).
• O.IN NaOH solution.
• 0.5N NaOH solution.
• 2.5N NaOH solution.
• O.IN HCl solution.
• 0.5N HCl solution.
• Concentrated HCl.
Procedure
V2 ( 100%)
EO = ~-'-----'- (20-2 )
Vr
(20-3 )
Heavy Metals
Procedure
3. Add 2 ml of nitric acid and 5 drops of sulfuric acid, and heat the crucible
until dense white sulfuric acid fumes evolve.
4. Ignite the sample, preferably in a muffle furnace (500-600 o e or 932-
1112°F), until complete combustion ofthe organic material occurs, and
cool it.
5. Add 4 ml of dilute hydrochloric acid, cover the sample, and heat it on a
steam bath for 10 to 15 minutes.
6. Uncover the sample, and evaporate it to dryness on the steam bath.
7. Moisten the residue with 1 drop of hydrochloric acid, add 10 ml of hot
water, and heat it for 2 minutes.
8. Add ammonium hydroxide dropwise until the sample is just alkaline to
litmus, dilute it to 25 ml with water, and adjust the pH to 3.0 to 4.0,
using short-range pH paper and dilute acetic acid.
9. Filter the sample if a solid or suspended material exists, rinse the cru-
cible and filter with 10 ml of water, and transfer the washings to a 50
ml color comparison tube, diluting the solution to 40 m!. Mix the solu-
tion to a uniform consistency. This is solution B.
10. Add 10 ml of freshly prepared hydrogen sulfide to each tube (solutions
A and B), and allow the tubes to stand for 5 minutes.
11. View the sample tube (B) and the lead standard tube (A), looking down-
ward from the top over a white surface. If the sample solution is darker
than the lead standard, then the pectin exceeds the 10 ppm lead or heavy
metal specification, and vice versa.
Solvent Content
Procedure
1. Heat the empty dry weighing bottle and stopper in a drying chamber for
30 minutes, and weigh the bottle and stopper.
2. Place 1 to 2 g of pectin, accurately weighed, in the weighing bottle, and
evenly distribute the pectin to a depth of about 10 mm.
3. Place the opened bottle and stopper in the same drying chamber ( 105°e
or 221°F) for 2 hours.
4. After drying, stopper the bottle immediately, cool it in a desiccator, and
weigh it.
FOOD-GRADE NONJUICE PRODUCTS 311
wt after drying )
% solvent = ( 1 - wt b" d.
elore rymg
100% (20-5)
For example:
The most common quality control parameter of pectins, however, is the jelly
grade or the pounds of sugar that one pound of pectin will support. The maxi-
mum jelly grade of pectin from lemon peel is about 300 to 350; from grapefruit
peel, 250 to 300; and from orange peel about 150 to 250. Most pectins used
for the commercial production of jams and jellies have a jelly grade of about
150. The Institute of Food Technologists has established a standard procedure
to measure the jelly grade, which has been officially accepted by the industry
(1FT Committee on Pectin Standardization 1959).
Procedure
..
Hazel-Atlas No. 35
glass tumblers removal of jelly
from tumbler
micrometer
R..Ii.---'----
Ridgelimeter
lass plate
9. Using the micrometer, lower the screw to just above the surface of the
jelly. After 2 minutes adjust the screw to just contact the top of the jelly
with the micrometer tip. The lowest line on the vertical scale is the per-
cent sag, with the micrometer reading giving tenths of a percent sag.
Repeat this step for each of the two remaining triplicate samples. If the
percent sag differs by more than 0.6 %, the jelly should be remade and
the test repeated.
10. Remove a portion of the jelly from the center, place it on the prism of a
refractometer, and read the Brix after about a minute. The temperature-
corrected Brix (see Chapter 2) must be 65.0 ± 1.0oBrix. A 1.0oBrix
error can mean an error of 3 to 4 jelly grade points.
11. The jelly grade is found by calculating a correction factor to the assumed
or estimated jelly grade. The factor (Fj) is found from:
Fj = 2 - (% sag)/23.5 (20-6 )
12. This factor is used to get the true jelly grade as follows:
For example:
Fj = 2 - (24.1)/23.5 = 0.97
Gelation has long been used as a means of preserving various fruits; the resul-
tant increased viscosity or decrease in water activity inhibits microbial growth
and spoilage. Jams and jellies usually can be stored at room temperature until
they are opened. Thereafter, the surface becomes subject to airborne molds,
which literally grow on the surface of almost any food product, especially one
as rich in carbohydrates and nutrients as fruit jams and jellies are.
Jellies can be defined as a clear jelled fruit product that retains its shape when
cut by a knife. The federal standards of identity describe jellies as a mixture of
fruit ingredients containing not less than 45 parts of fruit juice by weight to each
55 parts of saccharine ingredient (CFR Title 21 150.140). (The word "sac-
charine" is defined as a sweetening material.) For example, if you used 2.5
pounds of 65.0 o Brix orange concentrate, the amount of saccharine ingredient
that you could use while still calling the product fruit jelly would be:
acids are used to control the negatively charged pectin chains. By combining
the right proportions of pectin, sugar, water, and acid, gelation will occur, even
at temperatures as high as the boiling point of water (Joseph 1955). Also, the
mineral content will affect the balance and thus the gelation. Divalent cations
may act as linking agents between pectin chains, resulting in gelation. In fact,
low-methoxy pectins require polyvalent cations in order to gel. For the reason,
the mineral content of the water used in processing should be considered. Also,
the natural pectins, sugars, acids, buffering salts, and other soluble material
inherent in the juices or fruit used should be taken into consideration.
Slow-set pectins require the pH to be less than 3.30, whereas rapid-set pectins
require a pH ofless than 3.55. The slow-set pectins, however, are more tolerant
of excess acid levels than the rapid-set pectins. Jellies generally utilize slow-
set pectins so that gelation does not occur until the product is in the container.
Also, jellies generally use more pectin than jams to obtain a firmer gel. For
jams it is better to use a rapid-set pectin in order to avoid floating pieces of fruit
that separate if gelation is too slow. Most jellies and jams are made from a
standard 150 jelly grade pectin.
The general process of making jams or jellies involves premixing the pectin
with about three to five times as much sugar as pectin to help disperse the
pectin, and then dissolving this mixture in the fruit juice or pulp. When this
mixture is dissolved in fruit juice concentrates, the amount of sugar to be pre-
mixed with the pectin should not exceed about 20 to 25 % of the total solution.
This mixture is cooked or evaporated until it reaches an acid- and temperature-
corrected Brix of 65 to 68. The acid is added just before filling in order to
postpone the gelation as long as possible. Two examples of fruit jellies and
jams are presented in Tables 20-2 and 20-3.
The USDA divides the grade standards for fruitjeUy into two areas. Standard
fruit jelly meets the FDA standards of identity (CFR Title 21 150.140.), whereas
unstandardized jelly includes fruit ingredients not allowed in the same standards
Procedure
I. Mix pectin with 5 pounds of sugar, and add the mixture to warm water with stirring and bring
to a boil. Boil for 30 seconds.
2. Add the rest of the sugar, dissolve it, and add orange concentrate without further heating.
3. Add acid solution and orange oil and fill the container. Makes about 165 pounds of jelly.
316 CITRUS JUICE BY-PRODUCTS
Procedure
1. Mix pectin with about 5 pounds of sugar, and add the mixture to warm water with stirring
and bring to a boil. Boil for 30 seconds.
2. Add the rest of the sugar and peel strips, boil to dissolve, and then turn off the heat source.
3. Add the concentrated orange juice, oil, and acid solution. Yields about 152 pounds of mar-
malade.
from any caramelized or other objectionable flavor, whereas Grade B may have
only a slight caramel-like flavor.
FRUIT SECTIONS
The canning and the bottling of fruit sections have become a popular means of
preserving citrus fruit. Grapefruit sections have found acceptance in the United
States, and mandarin sections are well accepted in the Far East. Canned and
bottled citrus fruit sections have the advantage of long shelf lives and retention
of most of the whole-fruit nutrition, and can be conveyed to noncitrus or off-
season geographical areas that otherwise might not be able to partake of the
benefits of citrus.
Fruit Preparation
Prior to processing, citrus fruit should be held in storage bins for 4 to 6 days to
allow it to soften. A prescaling of the fruit may shorten this holding time to 2
to 3 days. The fruit then is scalded in hot water (90-100°C or 196-2100F) for
5 to 8 minutes for grapefruit or 4 minutes for oranges, which further softens the
fruit. Then the fruit is sprayed with low pressure steam that penetrates the peel,
which is followed by additional bin storage until the fruit is ready for peeling.
The softened fruit requires gentle handling, but the additional holding time can
reduce scalding requirements by 30 %. It also results in less fruit spoilage and
less cooked off flavors.
Peeling
Sectionizing
Manual sectionizing is done by placing the fruit against a spindle, and the sep-
arated sections are tapped or brushed against each other to remove the seeds.
318 CITRUS JUICE BY-PRODUCTS
Filling
After the sections are placed in the container, a liquid sweetener is added, which
can be a mixture of water, juices, and sugar at vaious concentrations. Allowable
filling solutions include water, grapefruit juice and water, grapefruit juice,
slightly sweetened sirup or slightly sweetened water (12.0-15.9°Brix), light
sirup (16.0-17.9°Brix), heavy sirup (~18.0oBrix), slightly sweetened
grapefruit juice and water (12.0-15.9°Brix), lightly sweetened grapefruit juice
and water (16.0-17.9°Brix), heavily sweetened grapefruit juice and water
(~18.0oBrix), slightly sweetened grapefruit juice (12.0-15.9°Brix), lightly
sweetened grapefruit juice (16.0-17.9°Brix), and heavily sweetened grapefruit
juice ( ~ 18.0 oBrix). Usually sections from grapefruit that pack 70 to 80 to a
box in fresh fruit packaging are packed in 8-ounce and 303 cans, and larger
grapefruit that pack 54 or 64 to a box are packaged in 26-ounce glass containers.
There are two basic methods of filling citrus sections into containers. One is
cold fill, where the container is chilled after sealing in a 35 to 40 0 P (2-4°C)
water bath. The container then is dried by an air stream, and dry ice sometimes
is sprinkled on it to maintain its temperature on the way to cold storage at 36
± 2°P. About 0.04 to 0.05% sodium benzoate sometimes is added as a pre-
servative, along with less than 0.035 % calcium chloride or calcium lactate to
enhance the firmness of the sections.
Hot-fill packaging includes heating the sealed container to 149 to 190 0 P (63-
88°C) for 20 to 45 minutes, depending on its size. Larger containers require
more heat for a longer time. The heat-treated container then is passed through
a 60 to 75°P (16-24 °C) water bath, which cools the contents to about 90 to
lOOoP (32-38°C). The residual heat dries the can.
Analytical Procedures
The Brix and acid levels of the sirup used in canning or bottling citrus fruit
sections are measured as explained in Chapters 2 and 3. The integrity of the
fruit sections themselves generally is determined by direct inspection. If the
section maintains its form when suspended above the liquid, it can be consid-
ered firm. The main quality parameter is the drain weight, which is determined
by using the following procedure.
FOOD-GRADE NONJUICE PRODUCTS 319
Drain Weight
Procedure
1. Pour the contents of the container through the sieve, collecting the liquid
in the pan or tray under the sieve.
2. Drain for 2 minutes, and tap the sieve to remove adhering liquid.
3. Weigh the liquid and the screened fruit sections separately.
4. Calculate the percent drain weight by using:
For example:
The federal standards of identity require that at least 50 % of the drain weight
be in the form of whole sections in order for the product to be called fruit
sections or segments. If less than 50% of the drain weight is whole sections,
the product must be called broken sections or broken segments. Also, they must
have less than 20 square centimeters (3.1 square inches) of tough membrane
or albedo on the segments per 500 grams ( 17.6 ounces) of the product. Also,
there can be no more than four developed seeds (9.0 mm or 0.35 inch or larger),
and no more than 15 % of the drain weight can contain injured, discolored, or
otherwise blemished fruit.
Grade A canned grapefruit sections must have at least 53 % of the water capacity
of the container in the form of fruit, with at least 65 % of the fruit being in the
form of whole sections. Grade B also requires that 53% of the water capacity
of the container be in the form of fruit, with only 50% or more of the fruit being
in the form of whole sections. Substandard canned grapefruit fails the above
requirements. The fill of the container or the composition of the liquid medium
320 CITRUS JUICE BY-PRODUCTS
is not considered in USDA grades. Table 20-5 illustrates the USDA scoring
system for canned grapefruit.
Citrus juice drinks can be defined as beverages that contain less than 100 % fruit
juices. They usually contain varying amounts of sweeteners and organic acids,
and perhaps also carbonation, flavors, colors, emulsifiers, texturizers, or other
components designed to enhance the beverage. A drink base can be defined as
the concentrated components of a drink minus the major portion of water,
sweetener, and perhaps even the acid, although many drink bases already con-
tain the needed acid. In the United Kingdom, drink bases are called squashes;
in Australia they are known as cordials and usually contain about 20 to 50%
juice. The main consumer appeal of juice drinks is their low cost. Drinks and
drink bases can be used for nonbeverage products as well, such as popsic1es
and purees. Purees are a type of fruit juice base used as a flavoring material for
sherbets, candy, baking products, and many other products. Purees once were
made from macerated whole fruit but now are made from juice concentrate in
a manner similar to that used for drink bases. Also included in puree bases are
juice sacs, sugar, color, flavors, citrus oils, essences, and citric acid to adjust
the pH to about 3.0. Sherbet bases generally are manufactured as 50-fold con-
centrates and are required by law to contain at least 2 % juice. Sugar and color
usually are added by the food processor using the puree base. Oil levels of puree
bases usually occur in the range of 0.030 to 0.035 %, as measured by the pro-
cedure outlined in Chapter 6. The pulp level ranges from 6 to 12.5 %, as deter-
mined by procedures mentioned in Chapter 7.
The most popular beverages in the United States are carbonated soft drinks,
and the increased concern in the late 1900s for healthful beverages has made
juice-added soft drinks especially appealing. Also, juice drinks with high levels
of juice have been developed that emphasize added nutrients such as calcium
and low calorie sweeteners. The flavor and the color quality of the juices used
in these drinks playa lesser role in the overall quality of the drink than they do
in citrus juices; so a lower quality can be used in these drink bases, such as
pulp wash juices or bitter juices as well as high-acid juices such as lemon and
lime juices. Components that make up citrus drinks can be broken down into
11 groups: juice, sweeteners, acids, colorants, flavors, texturizers, clouding
agents, nutrients, foaming agents, de foaming agents, and preservatives; but
these components may not all be contained in an individual juice drink or base.
Juices
Juices are used for several reasons, the main one being to enhance the health
appeal of the product. Depending on the amount of juice used, color and flavor
enhancement are the next most important contributions the juices make to the
drink. Cost often determines which juices are used in a drink, with cheaper
juices such as apple, pear, or white grape juices frequently used. Some highly
colored juices, such as pomegranate or cranberry juice, can add a desirable
deepening of the citrus color at around the 1 % level, but greater amounts impart
an unsightly brownish tinge to the product. Lemon and lime juices have been
used as citric acid sources for drinks, enhancing their citrusy flavor at the same
time. Pulp wash juices and peel extracts are ideal juice sources for citrus drinks,
which provide a use for these products that would be difficult to market other-
wise.
With the advent of debittering techniques, the use of bitter juices in drinks is
expected to decline. The most bitter navel juice can contain limonin levels as
high as 40 ppm, well over the 6 to 7 ppm generally accepted taste threshold. A
tenfold dilution in making a 10% juice drink would dilute the limonin level to
4 ppm maximum, below the taste threshold. Care must be taken in using bitter
juices with nonbitter citrus juices because even non bitter citrus juices contain
some limonin. Because pulp wash juice (or the juice produced from washing
the core material recovered from FMC-like juice extractors), and peel extracts
322 CITRUS JUICE BY-PRODUCTS
are highly bitter, they may require debittering before significant quantities can
be used in juice drinks.
Sweeteners
Sugar (sucrose), or its equivalent in high fructose com syrup (HFCS) or low
calorie sweeteners, is a major component in most juice drink formulations.
Commercial sugars generally cost only 5 to 15% as much as the sugars or sol-
uble solids found naturally in citrus juices, a difference that makes juice drinks
a more economical product for consumers. The only known nutritional disad-
vantage of citrus juices is their high calorie content, which can be largely over-
come by using low calorie sweeteners in juice-based drinks. The Federal Reg-
ister allows sugar (sucrose), invert sugar, fructose, com syrup, glucose syrup
sorbitol, or any combination of these ingredients to be used as sweeteners in
carbonated soft drinks (FR June 26, 1969,9867). Because sugar is readily avail-
able and comprises the major portion of most juice drinks, it is not added until
just prior to packaging of the final drink, along with the necessary water. This
procedure saves a considerable amount of money in shipping, storage, and con-
veying.
Acids
Acids are another major component of citrus juices and juice drinks, and give
the beverage a citrusy tangy taste. Because the hydronium ion is what imparts
this taste, theoretically most acids can be used; but not all acids are approved
for food use. Citric acid, which is the most common acid found in fruits and
juices, is considered to be one of the best acids to use in foods. One reason for
this is that it is easily buffered, being a triprotic weak acid. Also, citrus pro-
cessors who have quality control programs based on citric acid (citric acid cor-
rections to the Brix and lemon and lime inventories based on the weight of citric
acid) are better equipped to handle citric acid than other acids. However, the
Federal Register permits the use of acetic acid, adipic acid, fumaric acid, glu-
conic acid, lactic acid, malic acid, phosphoric acid, or tartaric acid in addition
to citric acid in carbonated soft drinks (FR June 26, 1969, 9867). Citric acid
can be buffered by adding tricalcium phosphate, sodium citrate, and citric acid
in the ratio of 1 : 5: 30 by weight. Again, the Federal Register allows acetate,
bicarbonate, carbonate, chloride, citrate, gluconate, lactate, orthophosphate, or
sulfate salts of calcium, magnesium, potassium, or sodium to be used as buff-
ering agents in carbonated soft drinks (FR June 26, 1969, 9867).
Colorants
Colorants play a minor role in regard to the Brix, but they have a significant
influence on the drink's overall appeal and therefore can be formulated inde-
FOOD-GRADE NONJUICE PRODUCTS 323
pendently of the sugar and the acid. Highly pigmented juices from other vari-
eties of fruit offer a natural and nutritional way to produce desirable colors in
fruit juice drinks. As mentioned above, the addition of 1 % of pomegranate or
cranberry juice to citrus juice drinks can deepen the color. Also, tangerine juices
are known for their deep orange color, as is juice produced in dry Mediterranean
climates such as that in California, Australia, and South Africa and around the
Mediterranean basin. Natural pigments can be used such as turmeric and anatto
(see Chapter 19).
Other common natural pigments used in citrus drinks include the carotenoids.
One of the most common carotenoids is iJ-carotene, which is extracted from
citrus peels or other natural sources. iJ-Carotene imparts a light yellow to orange
color to citrus juice drinks. It is not very soluble in water but is more soluble
in fats and oils. Its molecular structure is as follows:
FD&C dyes, described in the Chapter 19, also can be used to enhance the
color of citrus drinks. FD&C No.5 is used in lemon drinks in the range of 10
to 15 ppm, in lime drinks at 20 to 30 ppm, and in grapefruit drinks at 5 to 10
ppm. In orange drinks FD&C No.5 and FD&C No.6 can be used together,
each at about 0.008 to 0.2 %, depending on the amount of juice used in the
drink. Packaging also has an effect on the need for color. Generally, the bright
colors of citrus juices are a good selling point. However, if opaque cans or
containers are used, color enhancement becomes less important.
Flavors
Citrus flavors come in various forms. Natural citrus oils and essences extracted
from the peel of the fruit and recovered during juice evaporation or pasteur-
ization are the most common source of citrus flavors. Citrus oils can be con-
centrated or folded in order to strengthen the citrus flavor by concentrating the
oxygenated compounds in the oils. Again, the addition of these flavors has little
effect on the Brix measurements and can be formulated independently of the
324 CITRUS JUICE BY-PRODUCTS
sugars and acids. The most prevalent component of citrus oils is d-limonene,
which is easily measured by using the Scott method, described in Chapter 6.
Ideal oil ranges are 0.015 to 0.020% by volume. Other compounds that con-
tribute to the citrus-like taste of juice drinks include ethyl butyrate and various
aldehydes, such as decanal, citral, acetaldehyde, and octanal.
Texturizers
Clouding Agents
Clouding agents give citrus drinks the opaque appearance of juice. This is es-
pecially important in lemonades and limeades, where the natural cloud levels
in the juices must be extensively diluted in order to bring the acid levels in the
juice down to the desired levels. About 30 % of the natural cloud in juice is
made up of proteins, which come primarily from the cytoplasm and organelles
of the juice cells during extraction. About 10 to 15 % of the cloud is made up
of hemic ell uloses from the membrane material of the fruit, and about 5 % of the
juice cloud is composed of pectin (Bennett 1985). Natural flavonoids, such as
hesperidin and naringin, also contribute to the cloud material. Protein/pectin
combinations have been shown to produce cloud similar to that found in citrus
juices. Casein, a common protein, has been found to produce cloud with highly
methoxylated pectin in a ratio of about 500 to 1000 /kg/mg pectin (Bennett
1988). Soy flour or other suitable protein sources can be used with pectins as a
starting point in most drink cloud formulations.
FOOD-GRADE NONJUICE PRODUCTS 325
Nutrients
where p is the percent of juice soluble solids in the final drink compared to the
total soluble solids in the drink, and W is the total weight of the drink. The
weight of the drink can be found from the gallons by multiplying by the density
found from the Brix, using Equation 2-9 or 2-11. For example, if we had 1000
gallons of an 11.8 °Brix juice drink that contained 10 % orange juice soluble
solids, the weight of drink after converting the 1000 gallons of drink to pounds
would be 8717 pounds drink (d = 8.717 lb/gal). Equation 20-9 then would
give:
Calcium addition to juice drinks containing high levels of citrus juices has
recently become a popular way to increase the nutritional value of citrus prod-
ucts. The USRDA for calcium is 1 gram. Assuming a serving size of 6 ounces
( 177 ml), a maximum of 24 grams per gallon of drink or 53 pounds per 1000
gallons of drink should constitute the maximum level of calcium needed to
enhance the drink's nutrition. Because most people consume other foods that
are high in calcium (such as milk), an amount that is less than this maximum
level should be adequate. It must be remembered, as explained in Chapter 11,
that whenever nutrients are added, nutritional labeling is mandated by law.
Preservatives
Maillard oxidative reactions, which can cause off flavors and colors to develop.
Ascorbic acid or vitamin C is sometimes used as an antioxidant as well. How-
ever, the use of sulfites is becoming more restricted, and they have been banned
in many areas offood processing. Some ofthe most common preservatives used
in citrus juices and beverages are benzoates and sorbates. These preservatives
are more effective against bacteria in single-strength juices and are less effective
against the slower-growing yeasts, with the effectiveness of sodium benzoate
against yeast depending on the pH of the juice. For pH values of 2.5 to 3.5, a
sodium benzoate level of 0.05 % is effective against yeasts; at pH values of up
to 4.5, 0.10% sodium benzoate is required. Common sodium benzoate levels
of 0.15 % are found in many juice drinks, as well as levels of 0.05 % potassium
sorbate.
The Federal Register permits the following preservatives to be used in car-
bonated soft drinks: ascorbic acid, benzoic acid, BHA, BHT, calcium disodium
EDT A, erythorbic acid, glucose-oxidase-catalase enzyme, methylparaben or
propylparaben, propylgallate, potassium or sodium benzoate, potassium or so-
dium bisulfite, potassium or sodium metabisulfite, potassium or sodium sorbate,
sorbic acid, sulfur dioxide, tocopherols, and, in the case of canned soft drinks,
stannous chloride at less than 11 ppm as tin (FR June 26, 1969,9867).
Some drinks have a greater appeal when associated with foam, such as some
soft drinks, but others do not. Foaming and defoaming agents can be used to
control this characteristic. The Federal Register permits the use of glycyrrhizin,
gum ghatti, licorice or glycyrrhiza, yucca (Joshua tree or Mojave), and guillaia
(soap bark) as foaming agents in carbonated soft drinks (FR June 26, 1969,
9867). Dimethylpolysiloxane can be used as a defoaming agent in carbonated
soft drinks at levels equal to or below 10 ppm.
as well. The blending of these aqueous and nonaqueous components with juice
concentrates constitutes a drink base, which can be used by a bottler to make a
finished drink by the addition of sugar and water. However, the amounts of
sugar, acid, juice, and water significantly affect each other's concentrations; so
determining the right amount of these macrocomponents in relation to one an-
other can become confusing.
In order to determine how the four main macrocomponents of juice drinks
affect each other, their relationship must be defined. The sugar or soluble solid
balance in juice-like solutions can be defined as follows:
(20-10 )
where Vc' Dc, and Be are the volume, density, and Brix of the concentrate used
in making the base or drink, Ws and Wa are the weights of sugar and acid to be
added to the drink, and B, V, and D are the Brix, volume, and density of the
final juice drink. The 1.19 factor is used to obtain the sugar equivalent of the
added acid correction to the Brix as calculated by Equation 2-2. The first and
last terms in Equation 2-2 dropped because they are negligible in acid correc-
tions to juice drinks. To understand the validity of this assumption, consider
making 1000 gallons of drink at 11.8 °Brix and 1 % acid using 10% soluble
solids as orange juice concentrate with a Brix of 65.0 and an acid level of 3.0.
The error in calculating the amount of sugar needed would be about a third of
a pound of sugar out of 828 pounds needed. This suggests that the first and last
terms in Equation 2-2 are negligible. The percent of acid plus the correction is
I + O. 19, so that 1. 19 times the weight of the citric acid is equal to the sugar
contribution to the Brix or soluble solids.
The acid balance is:
(20-11 )
with Vo DC' and Ac being the volume, density, and acidity of the juice concen-
trate used in the drink and V, D, and A the volume, density, and acidity desired
in the final drink, with Wa again being the weight of the acid needed. Unlike
the Brix, which includes acid levels in its value, the percent acid does not in-
clude any sugar concentrations or sugar corrections. Therefore, the weight of
the sugar is not included in Equation 20-11.
The percent juice in the drink is an important parameter, which is defined as:
(20-12 )
that enable the calculation of the amount of sugar and acid needed in a particular
drink:
To find the volume of water needed, we can utilize a simple weight balance
equation as follows:
(20-15)
where Vw and Dw are the volume and density of water. Using Equations 20-12
through 20-15, we can obtain:
Equations 20-13, 20-14, 20-16, and 20-17 are all independent equations based
on the same drink parameters, which are usually known prior to formulation.
Care must be taken to use consistent units for densities, volumes, and weights.
We use mainly pounds and gallons here.
Let us consider an example to illustrate how these equations can be used.
Suppose that we want to make 1000 gallons of a 10% juice drink with a final
Brix of 11.8 and a final acid level of 1.0% from a 65.0 o Brix concentrate with
an acid level of 3.0%. How much concentrate, water, sugar, and acid are needed
to make the drink? Because Equations 20-13,20-14,20-16, and 20-17 are in-
dependent equations, we can calculate any of the four parameters in any order.
The densities can be obtained from Equation 2-9 or 2-11 from the Brix. Using
Equation 20-17 to get the volume of concentrate needed, we obtain:
= 827.7 lb sugar
(20-18 )
For example:
compared to the 827.7 pounds of sugar calculated by Equation 20-13. The vol-
ume of water needed can be calculated by using a form of Equation 20-15:
(20-19 )
(20-20)
where V' is the volume of water used to dissolve the acid. This value can be
calculated by using:
(20-21)
where X is the fraction of acid used in predissolving the acid. In the above
example, if we want to dissolve 82.4 pounds of citric acid to make a 50% (X
= 0.50) acid solution, Equation 20-21 becomes:
To calculate the fraction of citric acid in a predissolving solution, one can use
the following equation:
(20-22)
~~(714-;-.4~)~(~10~.9_7_7~)(~6_5.~0)~+~1_19~(_82~.4~)~
,. = 59.3°Brix for drink base
(14.4) (10.977) + 1.19(82.4) + 9.9(8.322)
FOOD-GRADE NONJUICE PRODUCTS 331
More water may be needed to dissolve the citric acid. The volume of the base
also needs to be known, and can be calculated by using the following equation:
(20-23 )
where the density of the base, Db' can be found from Equation 2-9 or 2-11
according to the Brix, B b . In the above example, Equation 20-23 becomes:
The user of the base then would need to use 31.7 gallons of base with 919.1
9.9 or 909.2 gallons of water and 827.7 pounds of sugar. This formulation can
be converted to the use of one 52-gallon drum of base by means of simple
proportions, to give 1 drum of drink base + 1491 gallons of water + 1358
pounds of sugar to make 1640 gallons of finished drink.
As you may have surmised, these calculations can become complex and te-
dious if performed often. Here again, computer programming can be used to
automate the calculations. Figure 20-2 shows a flow chart that can be used to
write a program to automate this procedure.
One problem that emerges from time to time is that the Brix and/or % acid
of the final drink does not come out right. Suppose, in our example, that for
some reason the Brix came out to 1l.5°Brix instead of the needed 1l.8°Brix
and with 0.9 % acid instead of the needed 1.0% acid. If citric acid were added,
the Brix would be affected; so addition of sugar and water would be required
to achieve the proper balance of components. To find the amounts of acid and
sugar needed to correct the situation, the following equations could be used:
(20-24 )
(20-25 )
where the subscript i refers to the parameters for the initial erroneous condi-
tions, and absence of a subscript indicates the desired final conditions. In the
example above we do not want to add water to the final drink; so the last term
332 CITRUS JUICE BY-PRODUCTS
EQ 20-21 V'
x=% acid desired in
predissolved
solution
EQUATIONS
20-13,20-14,20-16
20-17
Ws, Wa, Vw, Ve
yes
Ws, Wa, Vw, Ve
Fig. 20-2. Flow chart that can be used to program computers to formulate drink bases.
in the numerator of the two equations above goes to zero, and we get:
Vi=Vi+Vw
Vw = Vi(Ai/A-1)
(add Vw to drink)
EO 20-24
Wa
EO 2-14
V'w
EO 20-26 EO 20-25
V Ws
Vi=Vi+V'w
Fig. 20-3. Flow chart that can be used to program computers to calculate the sugar and acid
needed to adjust errors in any type of drink.
334 CITRUS JUICE BY-PRODUCTS
Again, computer programming can simplify the use of these equations. Figure
20-3 shows a flow chart that can be used to program computers to automate the
adjustment of sugar and acid in citrus or any type of drink.
Because a great amount of water is used in making juice drinks, it is impor-
tant to monitor it so that contaminants do not enter drinks through the water.
The alkalinity of the water is one of the major problems, as it can neutralize
some of the added acids as well as affect the drink flavor. Chemicals such as
Drink base
Orange pulp wash concentrate** 274.0 gal
Tangerine concentrate*** 7.4 gal
Lime concentrate 7.4 gal
Acid solution 736.0 gal
Oil emulsion 296.0 gal
Potassium sorbate 74.11b
Orange oil flavor 2.8 gal
*This value is the amount calculated in Equation 20-14 minus the citric
acid added in the oil emulsion and minus the acid in the lime concentrate,
assuming 400 GPL lime concentrate.
**60.0 o Brix and 25 B I A ratio or 2.40% acid.
"*60.0 o Brix and 10.0 B I A ratio or 6.00% acid.
Note: 1000 gal of base makes 17,38\ gal of drink, requiring 15,I691b of
sugar and 15,282 gal of water in addition to the 650 gal of water used to
make the acid solution. The final drink should have a Brix of 11.8 and
0.56% acid.
FOOD-GRADE NONJUICE PRODUCTS 335
chlorine also may influence the final product. In severe cases, water treatment
may be required in citrus drink production.
Table 20-6 shows a complete sample formulation.
QUESTIONS
PROBLEMS
1. In making 1000 gallons of a 50% orange juice drink, how much vitamin C do you
need to add in order to have about the same level of vitamin C as found in 100 %
orange juice?
2. Suppose you wanted to make a 60% juice drink base that was 23.0 oBrix, and that
would make 10,000 gallons of final drink at 11.8°Brix and 0.53 acid, using 62.0oBrix
orange concentrate with an acid level of 2. 12 %. How much citric acid, water, and
orange juice concentrate would be needed to make the base, and how much water,
base, and sugar would be needed to make the final drink? (Hint: You can use Equa-
tion 20-21 to find the volume of water V' needed to get the desired 23.0 oBrix.)
3. Suppose that you wanted to make a 30% juice drink, using pasteurized single-strength
juice (12.4°Brix and 0.95% acid) to make 20,000 gallons of 12.0oBrix, 0.67% acid
drink. How much citric acid, water, sugar, and juice would you need? (Hint: Treat
the pasteurized juice in the same way as concentrate.)
336 CITRUS JUICE BY -PRODUCTS
4. In making a 25,DOO-gallon juice drink, it was found that the Brix came out to 11.1
instead of the desired 11. 9 °Brix, but the acid level came out correct at 0.72%. How
much sugar and acid would be needed to correct the Brix, and what would the final
volume of the drink be?
S. In making a 1O,ODO-gallon drink, it was found that the Brix came out to 12.6 instead
of the desired 11.8°Brix with an acid level of 1.60% instead of the desired 1.00%.
How much acid, water, and sugar will be needed to correct the situation? (Hint:
Consider setting the Wa or Ws to zero and use Equation 20-24 and/or Equation 20-25
to solve for Vw ' )
REFERENCES
Bennett, R. D. 1985. From presentation to the Citrus Products Technical Committee at the USDA
Fruit and Vegetable Laboratory, Pasadena, Calif., December 16, 1985.
Bennett, R. D. 1988. From presentation to the Citrus Products Technical Committee at the USDA
Fruit and Vegetable Laboratory, Pasadena, Calif., June 9, 1988.
Food Chemicals Codex, 1931. National Academy Press, Washington, D.C., 215-217.
1FT Committee on Pectin Standardization. 1959. Pectin standardization, final report of the 1FT
committee, Food Tech., 13,496-500.
Joseph, G. H. 1955. Pectic substances, Adv. Chern. Ser., 12,49-56.
Swisher, H. E. and Swisher, L. H. 1977. Specialty citrus products. Citrus Science and Technology
Vol JI, S. Nagy, P. E. Shaw, and M. K. Veldhuis, eds. The AVI Publishing Company, Inc.,
Westport, Conn., 321-322.
Chapter 21
The waste products of the agricultural and food industry have long been used
as fertilizers and animal feeds, and to make special nonfood products such as
fuels, in the form of methane gas and alcohols. Methane gas production usually
is generated from waste water and will be discussed in Chapter 22. Fuel-grade
alcohols usually are manufactured from peel products and are discussed in this
chapter. Citrus peel and peel products have been and continue to be used as a
source of high-quality animal feeds, which require careful quality control.
Citrus peel often is viewed as a waste product even though some citrus proces-
sors are able to make substantial profits from it. The most popular peel products
are those used in animal feeds. Citrus plants frequently are located in rural areas
where livestock are raised that can use the citrus wastes as feed or a feed sup-
plement. Fertilizers from livestock or pOUltry wastes, in tum, are often used in
citrus groves, creating a compatible symbiotic relationship. Dried peel products
can be conveyed to distant markets as well, which provide the juice processor
with numerous opportunities for disposing of waste products economically.
Many factors affect the quality control of feeds, such as the type and breed
of animal, the desired end results or animal products, the effect of these prod-
ucts on humans, supplemental feeds, the complexities of animal nutrition, the
economics involved, and the storage capabilities of the feed. Many of these
parameters are too complex for routine quality control. Also, the processor has
control over only a few of them. Most of these characteristics are naturally
inherent in the feed, and once the general characteristics of the feed have been
established, it is up to the quality control department to guard against excessive
contamination or alteration of the basic attributes of the feed.
Most citrus processors guarantee their feeds to have at least a minimum
amount of protein, crude fiber, fat, and ash. Another important parameter is the
337
D. Kimball, Citrus Processing
© Van Nostrand Reinhold, New York, NY 1991
338 CITRUS JUICE BY -PRODUCTS
Unprocessed Peel
Pressed Peel
Peel that has been shredded, limed, and pressed to remove about 10% of the
moisture has the advantages of lower transportation costs and a more appealing
ANIMAL FEED AND FUEL BY-PRODUCTS 339
and unifonn texture and composition. Such treatment provides for the produc-
tion of press liquor and/or citric molasses, and allows the feed to be transported
longer distances because of a reduction in trailer leaks. Typical nutrient levels
of pressed peel are shown in Table 21-1. Little to no routine quality control is
perfonned on this lightly processed peel product.
Dried citrus peel, one of the most common feeds manufactured from citrus peel,
is second only to com in nutritive value for dairy and beef cattle and sheep
(Kirk and Davis 1954). The drying of citrus peel is somewhat involved, re-
quiring more quality control monitoring than does production of unprocessed
or pressed peel. Dried citrus peel is made by passing pressed peel through a
rotary dryer, with the consistency and speed of the peel through the dryer de-
tennining the drying efficiency. Studies have been made of the rate of drying
at various temperatures (Braddock and Miller 1978).
Citric molasses is often added to the pressed peel to aid in the drying process
and to help prevent burning of the peel. Too much molasses will cause the peel
to stick to the edges of the dryer, which will also cause burning. Molasses
addition also darkens the color of the feed. Some investigators have even used
the feed color to detennine the amount of molasses that has been added (Bissett
1950). When molasses is used in the drying process, the dried pulp is referred
to as sweetened dried citrus pulp. In wanner drier weather, wet or pressed peel
can be spread on the ground and solar-dried. This is a common and economical
practice in areas with a dry Mediterranean climate such as California. Even
though most peel dryers are equipped with dust collectors, dried citrus peel
usually is associated with fine dried particles. FMC juice extractors break up
more juice cells than other juice extractors, causing an increase in this fine dust
material in the dried feed.
The most important parameter to monitor in dried peel manufacturing is the
moisture content. Moisture levels must be below 10%, or micorbiological spoil-
age can set in along with a buildup of heat. This heat can reach a point where
the dried peel will spontaneously combust, not only causing a fire hazard but
destroying the product. Smoldering peel that has not been dried sufficiently can
create a serious disposal problem. Unless the peel is spread out and air- or solar-
dried immediately, it will continue to bum until it becomes ash. Some bulk
dried peel warehouses have thermocouple monitors with air vents in the floor
to adjust the temperature and correct moisture problems. The relative humidity
in storage areas should be kept at 52 % or less in order to avoid reabsorption of
moisture by the peel (Braddock and Miller 1978). Oil recovery systems some-
times add moisture to the peel through peel washings, which should be ac-
counted for in drying operations. Grapefruit peel generally contains more mois-
ture than orange peel and thus would require more drying. There are several
ways to perform moisture tests, and one of the simplest follows:
Procedure
1. If the peel has not been shredded or chopped fine, this needs to be done
with a shredder. Peel particles should be about 20 to 30 mesh or about
1 /81/ in diameter or smaller.
2. Place the empty drying pan or dish on the triple beam balance, and record
the weight ( Wpan ).
3. Place the shredded peel in the drying pan so as to just cover the bottom
of the pan, and record the weight (Wb ).
4. Leaving the drying pan on the balance, radiate the pan with a heat lamp
until no further change in weight is noticed. Record the weight ( Wa ). Care
should be taken not to overheat or scorch the sample.
5. The % moisture can be calculated from:
(21-1 )
For example:
higher in fiber, nitrogen-free extract, and ash than wet peel, but it is lower in
fat (Kirk and Davis 1954).
Pellets
Because of the dusty nature of dried citrus pulp and its low bulk density, many
processors pelletize the dried pulp. This compacting of the dried pulp volume
cuts shipping and storage costs about in half and facilitates a cleaner and more
efficiently conveyed product. The higher-density pellets are more easily con-
sumed by ruminants as well. Pellets made from dried pulp fines are generally
about! inch in diameter and about 1 to I! inches long. Whole dried citrus pulp
is made into pellets about ~ inch in diameter and about ~ to ~ inch long in order
to retain the pulp fiber and seeds (Kesterson and Braddock 1976).
Several factors affect the mechanical durability of the pellets, including the
thickness of the die, holding time, pellet size, energy used in pelletizing, and
bonding agents used. Citrus molasses has been found to be an excellent bonding
agent when used in the proportions of about 5 to 15 % of the total weight.
However, molasses usually is added before the peel is dried instead of during
pelletizing in order to aid in the drying process, as mentioned previously. The
extrusion rate and the bulk density of the pellets vary inversely with the die
length, and the energy needed during the extrusion is inversely proportional to
the die thickness.
Some additional advantages of pelletizing dried citrus peel include the re-
duction of microbial spoilage. It has been found that pellets with twice the
moisture content of dried pulp still are resistant to microbial deterioration (Dean
1966). However, pellets wetter than this lose their mechanical stability. Be-
cause pelletizing involves friction that can heat the product, and thus dry the
pellets, dried citrus pulp destined to become pellets must be dried to only 10 to
12 % moisture levels. After pelletizing, the moisture levels will drop to below
the needed 10% level. Like dried citrus pulp, pellets should be kept in areas
with 52 % or less humidity to avoid reabsorption of moisture into the peel. A
relative humidity of 90% for 30 days has been shown to induce mold growth
(Braddock and Miller 1978).
342 CITRUS JUICE BY-PRODUCTS
Insect Infestation
Another problem in the stomge of dried citrus pulp or pellets is the occurrence
of insect infestation. The almond moth, Cadra cautella (Walker) (Hagstrum
and Sharp 1975) and the saw-toothed gmin beetle, Oryzaephilus surinameusis
(L.), have been associated with dried citrus pulp, with the latter found in coarse
pulp (Laudens and Davis 1956). Pellets and fine meals have been found to
contain the cigarette beetle, Lasioderina serricorne (F.) (Laudens and Davis
1956). Other insects have also been found in dried citrus material and justify
thorough cleaning of warehouses as often as possible. Residual insecticides,
traps, and barriers should be used, as well as professional services if needed.
(See Chapter 16.)
Press liquor
Press liquor is the solution expressed from shredded and limed peel. Unlimed
peel is very slimy and retains moisture. To remove moisture, about 0.15 to
0.25 % lime is added to shredded peel, and is allowed to react for at least 15
minutes. The exact mechanism by which the lime helps to release water and
remove the slimy texture of the peel is not known, but probably the basicity of
the lime demethoxylates the pectins in the peel in a manner similar to the action
of pectinase enzymes:
o o
II ow II (21-2 )
R- COCH 3 - R- CO- + CH3 0H
which is similar to Equation 8-1. The divalent cations, Ca + 2 , from the lime
probably combine with the demethoxylated pectin in a manner similar to that
illustrated in Fig. 8-2. The end result is syneresis, or the natuml expression of
fluid, which is not unexpected considering that 80 to 85 % of the peel is water.
The slimy nature of the unlimed peel is probably due to hydrogen bonding of
the ;:!ster groups of the pectin with water. Lime demethoxylation increases the
ionic strength of these bonds, producing a less slimy texture. About 10% of the
peel moisture can be removed in primary presses. Secondary presses generally
can remove only about another 2 % of the moisture from the peel.
Press liquor usually contains about 9 to 15 % soluble solids expressed as °Brix,
with about 60 to 70% of these soluble solids being in the form of sugars. Peel
oil levels mnge from 0.2% to 0.5% in Florida press liquor and up to 0.8% in
California press liquor made from orange peel. The drier climate in California
produces peel with a higher oil content. Typical press liquor constituents are
shown in Table 21-3. Press liquor is occasionally sold as is for animal feed
production, or fermented to produce fuel-gmde alcohols, or concentrated to
citric molasses, the last being its most common use.
ANIMAL FEED AND FUEL BY-PRODUCTS 343
Citric Molasses
The manufacture of citric molasses offers several advantages over the manufac-
ture of press liquor alone. Storage and shipping costs are greatly reduced be-
cause molasses takes up less space; concentrated sugar solutions are more re-
sistant to microbial spoilage; and citric molasses can be mixed with pressed
peel to aid in peel drying, as mentioned previously, in addition to acting as a
bonding agent when pellets are made from dried pulp. The use of evaporators
that utilize waste heat from peel dryers can greatly reduce the cost of molasses
production. An example of the composition of citric molasses is given in Table
21-4.
Citric molasses generally has about the same characteristics as Blackstrap
molasses except that it contains half as much ash and twice as much protein.
Florida requires that at least 45 % of the total sugars be invert sugars (glucose
and fructose) with a Brix of at least 35.5 after mixing with an equal weight of
water (Florida State Dept. of Agriculture 1950).
Two common problems in the manufacture of citric molasses are foaming
and scale formation. Commercial antifoaming agents can be used to minimize
foaming problems; and in addition to foaming agents, removal of the suspended
solids through centrifuging or other suitable means can reduce foam formation.
Foaming can be due to microbial spoilage, alkaline processing, or the mixing
of old and new molasses (Hendrickson and Kesterson 1971). Scale formation
is due to calcium from the lime used to treat pressed peel, which often precip-
itates on evaporator surfaces as calcium citrate. Using preheaters will provide
a site for the precipitation of this scale in an area that is easier to clean, and the
use of two such preheaters allows the use of one while the other is being cleaned.
The scale can be removed by using a 5 to 15 % hot caustic solution containing
2 to 3 % chelating agents such as alkaline ethylene diaminetetraacetic acid (Kes-
terson and Braddock 1976).
Most citric molasses has been sold as a feed supplement for animal feeds.
Storage of the molasses usually will decrease the pH by about 0.4 unit per
season (Hendrickson and Kesterson 1950), and viscosity will increase with stor-
age. Vigorous agitation generally will restore fluidity, but prolonged storage
can result in solidification of the molasses. Molasses that is increasing its vis-
cosity at a rate of less than 500 cps per day can be stored for at least 3 months
(Hendrickson and Kesterson 1952).
The microbial flora of citric molasses would be expected to resemble that of
juice concentrates. Various species of Lactobacillus and Leuconostoc have been
isolated from citric molasses including L. Jermentum, Le. mesenteroides subsp.
mesenteroides, and Le. paramesenteroides (Parish and Higgins 1988).
A final footnote on citric molasses concerns the suspended matter. Citric
molasses manufactured in California generally contains a large amount of set-
tleable sludge material, which has been linked to the commercial waxes applied
in the fresh fruit packing houses. In Florida, where most citrus fruit is processed
directly from the field, this waxy sludge is less common. Also, the waxes used
in Florida and California are not the same. In California, juice fruit usually is
heavily coated with these waxes, which find their way into the press liquor and
then into the molasses. Commercial centrifuging to remove this sediment has
proved ill advised because sand impurities from the lime end up in the press
liquor and produce excessive wear in centrifuges. This sediment adds to the
viscosity of the molasses and may cause a conveyance problem if allowed to
settle out in storage tanks.
FUEL-GRADE ALCOHOL
With the increase in price of fossil fuels over the past several decades, alter-
native sources for fuels have been sought; and some of the more promising
ANIMAL FEED AND FUEL BY-PRODUCTS 345
sources are found in renewable agricultural products and waste materials. The
food industry has always tried to achieve economical recovery of waste mate-
rials. Now anhydrous ethanol is being used in premium unleaded gasolines, as
ethanol is an efficient fuel that can be used to enhance octane ratings.
Citrus press liquor and reconstituted citric molasses have been used as a feed-
stock for commercial alcoholic fermentations. A major problem encountered in
such fermentations is the presence of citrus peel oil, which is toxic to most
microorganisms. As mentioned previously, press liquor contains at least 0.2 %
oil (based on 1I.8°Brix). Citric molasses usually contains much less oil, as
most of it has been stripped out during evaporation. Oil levels must be 0.08 %
or less for a successful alcoholic fermentation; so the oil in citrus press liquor
must be removed before fermentation can take place. A common way to do this
is to use direct steam injection into press liquor (see Chapter 6 and section on
stripper oil). The viscosity of citric molasses also can inhibit fermentation, in-
cluding the contribution of waxy sludge material found in California citric mo-
lasses, as mentioned. Citric molasses usually is diluted to about 200Brix or less
prior to fermentation.
Like most microorganisms, yeasts undergo four main stages of growth, as
shown in Fig. 21-1. In the initial phase, they begin adjusting to conditions and
start to grow. Growth is rapid during the log phase, where reproduction occurs
exponentially. As the feedstock is depleted and the yeast waste material (alcohol
here) is concentrated, growth slows until fermentation reaches a steady state. It
is in the lag phase that fermentation is the most efficient. If fresh feedstock is
r
z
0 lag phase
~
::> death
a.. phase
0
a.. log phase
.....
en
L5
>-
TIME •
Fig. 21-1. The four stages of microbial growth. Alcoholic fermentation is most effective when
the yeasts are in the lag phase.
346 CITRUS JUICE BY-PRODUCTS
not introduced, the growth will eventually begin to decrease, resulting in the
death phase.
There are basically two types of fermentation, continuous and batch. Contin-
uous fermentation requires that yeast be kept in the lag phase by constantly
replenishing the feedstock while bleeding off the alcoholic solution without
changing the yeast population or activity. A popular and efficient continuous
fermentor is the Wick fermentor, shown in Fig. 21-2 (Wick 1980). The gas
generated in the fermentation causes a circular flow in the solution; and this
circular current not only mixes the solution to make it uniform, but it keeps the
yeast within it, providing a "dead zone" that contains little yeast in the comers
where the alcoholic solution can be bled off without significantly changing the
yeast population. Commercial use of the Wick fermentor has proved successful
(Eastman 1981). Continuous fermentation also is easier to monitor as far as
quality control is concerned because its conditions do not change as much as
those of batch operations. Batch fermentation requires greater tank space to
produce alcohol as rapidly as the continuous methods do. Also, the constant
starting and stopping of batches is very time-consuming and requires close at-
tention to ensure a successful fermentation.
Other factors also affect the rate of fermentation, including the type of yeast
used. Distiller's yeast has been found to grow faster in reconstituted citric mo-
lasses or steam-treated press liquor in California than do the natural yeasts or
Montrachet yeast. Excessive bacterial growth can destroy a fermentation, pro-
Carbon
Dioxide
Vent
Outbound
Alcoholic
Solution
,,--........
Currents Caused ! f -;; '\ )
By Gas Bubbles ~ Iff '/ J
From I
Fermentation tIL I;
t I...~ /
,-,.,
Fig. 21-2. Wick continuous fennentor used commercially to produce fuel-grade alcohol (Wick
\980).
ANIMAL FEED AND FUEL BY-PRODUCTS 347
ducing acetic acid rather than alcohol. Because bacteria grow much faster than
yeast, bacterial growth can quickly take over a fermentation. However, bacteria
are more sensitive than yeast to the osmotic effects in concentrated molasses,
as well as to benzoates and penicillin. Steam treatment of press liquor and pas-
teurization of molasses during evaporation are generally sufficient to prevent
bacterial contamination. Small amounts of benzoate or veterinary grade peni-
cillin can be used to inhibit bacterial growth as well. A good cleanup and san-
itation program is the best way to prevent bacterial spoilage.
In order to start up a fermentation process, using either a batch or a contin-
uous method, the feedstock must be diluted to 12 to 30 oBrix. Also, fermenta-
tion is more rapid with warm solutions. The peel oil level should be checked
to ensure that it is below the 0.08% v Iv limit (based on 1l.8°Brix). As the
container is filling with feedstock, one should add about 120 ml of veterinary
grade penicillin, 3 gallons of FMC defoamer or the equivalent, and 1 gallon of
pectinase if needed to reduce the viscosity of the solution, assuming a
1O,OOO-gallon fermentation volume. One should mix about 20 pounds of dis-
tiller's yeast in a 5-gallon bucket of lukewarm water to paint consistency, and
when gas evolves, add the mixture to a fermentation vessel containing about
half of the feedstock (5000 gallons here). It is important to allow a vent for
evolving carbon dioxide gas. The 20.0oBrix feedstock should decrease in Brix
down to about 13.0 oBrix after fermentation in batch processing. Fermentation
of 10,000 gallons takes up to 24 hours or so to really get under way and about
36 to 48 hours for completion, depending on feedstock, yeasts, and tempera-
ture.
To determine when fermentation is complete, a reducing sugar test can be
used. The Brix alone is insufficient because alcohol and other components of
the solution will contribute to it. Although the reducing sugars account for only
about half of the sugars present, their disappearance can be used as an indication
that all the sugars have been consumed in the fermentation. Reducing sugar
levels of 1 to 2 % usually indicate that the fermentation is complete. A method
that can be used to measure the reducing sugars is given in Chapter 19.
After fermentation, the beer is distilled in commercial stills. Aqueous alcohol
forms an azeotropic mixture upon distillation that prevents more than 95.6%
ethanol from being distilled at atmospheric pressure. Further heating cannot
result in any enrichment of the ethanol solution. Commercial plate column stills
can produce 90% ethanol solutions but generally achieve around 80% ethanol.
An 80% ethanol solution is sufficiently enriched to be used directly as a fuel in
automobiles with enlarged air intake jets, but 100% or anhydrous ethanol is
required for blending with gasoline. Alcohol bums hotter than gasoline and thus
requires a higher air mixture in the carburetor. Pure ethanol can be obtained by
further drying of the 80 % solution using absorbents, such as cornstarch or dried
citrus peel, or by the use of a vacuum (95 torr, 0.125 atm, or 26 inches of Hg)
during distillation that will shift the azeotropic composition up to as high as
348 CITRUS JUICE BY-PRODUCTS
Thermometer
Vapors
Beer --+
oo
:s
a.
CD
:s
...
~
Reflux
~-- Tank
Product
Tank
Distillage
Fig. 21-3. Commercial still used in production of alcohol from citric molasses or press liquor.
ANIMAL FEED AND FUEL BY-PRODUCTS 349
not affect its perfonnance as a fuel. Adequate signs and labeling must be used,
warning that the alcohol is toxic and is not intended for human consumption.
In order to measure the alcohol level in the distill age and the final product,
several methods can be used. One method involves measurement of the chem-
ical oxygen demand (COD) and is lengthy compared to other methods (AOAC
1984); another involves measurement of the specific gravity (AOAC 1984); a
third method involves the direct fractional distillation of the alcohol from the
sample. The latter two methods have proved sufficient for routine quality con-
trol. The specific gravity method is best for measuring the ethanol in the distil-
late or final product, and the distillation method works best for measuring the
ethanol in the distillage.
• Balance.
• 25 ml pipette.
• 25 ml beaker or flask.
• 25 ml of distilled water.
• Thennometer.
• AOAC table 52.003 (optional).
Procedure
Ethanol by Distillation
Procedure
1. Add 100 ml of the sample to the boiling flask in Fig. 21-4, and apply
heat. The temperature at the top of the column should reach 78.2°C
( 173 ° F) when the fractionating ethanol reaches the thermometer. This
temperature should remain steady until toward the end of the distillation.
2. When no more ethanol can be seen coming over into the condenser, or
when water beads are noticed in the column or condenser, record the
milliliters distilled as the % v I v of alcohol. For further accuracy, multi-
ply the milliliters of distillate by 0.95 to get the approximate true % al-
cohol in the sample, or use the specific gravity method on the distillate.
It should be readily apparent that the distillate cannot contain more than the
95.6% ethanol azeotropic limit. This is the reason for using the 0.95 factor in
step 2 above and for the suggestion that the specific gravity method be used on
the distillate. However, the % v Iv of ethanol measured by using this method
is adequate for most routine quality control.
QUESTIONS
1. What are the main nutritional parameters of citrus peel animal feeds?
2. What is the most common feed produced from citrus peel?
3. Why is citric molasses added back to pressed peel before peel drying?
4. What is the most important quality control parameter in the manufacture
of dried citrus peel, and why?
5. What are the advantages of pelletizing dried citrus pulp?
6. Why is lime added to press peel?
7. What is the major use of citric molasses?
ANIMAL FEED AND FUEL BY-PRODUCTS 351
Thermometer --.Il
Packing Material
Condenser
- - - - (No Water)
Condenser-' Water
Water-.
PROBLEMS
1. What is the TDN for wet peel that just meets a typical nutrient guarantee as men-
tioned in the chapter, assuming a nitrogen-free extract of 1O.3%?
2. What would be the TDN of pressed California orange peel, sweetened dried citrus
pulp, and citric molasses, according to the data in the chapter?
3. Is dried citrus pulp dry enough if a 6.832 g sample in a 165.132 g pan weighed
171.178 g (pan and sample) after drying under a heat lamp?
4. What would be the % alcohol in a distillate with a specific gravity of 0.832?
5. What would be the exact % alcohol in the distillage if 6.7 ml were distilled off in
the alcohol determination described in the chapter, yielding a distillate with a specific
gravity of 0.812? Would this be considered an acceptable distillation efficiency?
REFERENCES
AOAC. 1984. Official Methods of Analysis, 13th Edition, 9.012, 9.014, 10.024, 11.005, 11.007,
11.008. Association of Official Analytical Chemists, Washington, D.C.
Bennett, R. D. 1981. Private communication.
Bissett, O. W. 1950. A method for estimating soluble solids in dried citrus pulp, Proc. Fla. State
Hort. Soc., 63, 174-179.
Braddock, R. 1. and Miller, W. M. 1978. Some moisture properties of dried citrus peel, Proc.
Fla. State Hort. Soc., 91,106-109.
Dean, W. W. JT. Pelleting Wet Citrus Pulp with the Aid ofa Bonding Agent. M.S. Thesis. Uni-
versity of Florida, August 1965, University of Florida Agriculture and Engineering Mimeo Re-
port EG66-1, February 1966.
Eastman, R. 1981. Private communication.
Florida State Dept. of Agriculture. Lett. of December 8, 1950, Supplementary Feed Bull., 97.
Hagstrum, D. W. and Sharp, J. E. 1975. Population studies of Cadra cautella in citrus pulp ware-
house with particular reference to Diapause, 1. Econ. Entomol., 68(1), 11-14.
Hendrickson, R. and Kesterson, J. W. 1950. Storage changes in citrus molasses, Proc. Fla. State
Hort. Soc., 63, 154-162.
Hendrickson, R. and Kesterson, J. W. 1952. Viscosity of citrus molasses, Proc. Fla. State Hort.
Soc., 65, 226-228.
Hendrickson, R. and Kesterson, J. W. 1971. Citrus molasses, Fla. Agr. Exp. Sta. Bull., 677,
3-27.
Kesterson, J. W. and Braddock, R. 1. 1976. Byproducts and Specialty Products of Florida Citrus.
University of Florida, Gainsville, Fla., 10,22,31.
Kirk, W. G. and Davis, G. K. 1954. Citrus products for beef cattle, Fla. Agr. Exp. Sta. Bull.:
538,5-16.
Laudens, H. and Davis, D. F. 1956. Dried citrus pulp insect problem and its possible solution with
insecticides-coated paper bags, Proc. Fla. State Hort. Soc., 69, 191-195.
Nolte, A. J., von Loesedke, H. W. and Pulley, G. N. 1942. Feed yeast and industrial alcohol from
citrus wastes press juice, Ind. Eng. Chern., 34, 670-673.
Parish, M. and Higgins, D. 1988. Isolation and identification or lactic acid bacteria from samples
of citrus molasses and unpasteurized orange juice, 1. Food Sci., 53(2), 645-646.
Wick, E. 1980. New vessel design for rapid, continuous, fermentation. USDA Science and Edu-
cation Administration, Advances in Agricultural Technology, AAT-W-10/February.
Chapter 22
Citrus processing plants, like most other plants, generate materials of little or
no commercial value that generally need to be removed from the processing
site. It usually is the responsibility of the quality control department to monitor
and manage waste accumulation and treatment, especially in small processing
plants. These materials, which can be classified as solid, liquid, gaseous, or
hazardous wastes, can be conveyed from the plant through waste streams, by
air flow into the atmosphere, or by vehicle. Because these materials affect per-
sons in the vicinity of the plant, and because federal, state, and local regulations
apply to the disposal of these wastes, the quality control of wastes is an impor-
tant function.
The degree of pollution regulation depends largely upon the type and amount
of pollution as well as the surrounding environment. Citrus processing plants
and bottlers exist in a variety of locations, from rural to metropolitan, with a
wide range of existing pollution. Many plants are located next to large bodies
of water, which often act as disposal sites for wastes. Rural sites include large
land areas, which also may serve as disposal sites. Metropolitan plants often
take advantage of municpal disposal facilities, whereas plants located in sub-
urban areas must treat their own wastes or convey them long distances. The
federal government has passed several laws to control the pollution of the en-
vironment, including the Clean Air Act of 1977, the Clean Water Act, the Safe
Drinking Water Act, the Resource Conservation and Recovery Act, and the
Toxic Substance Control Act. State and local governments are likely to have
even stricter requirements and to act as watchdogs for the standards of the fed-
eral Environmental Protection Agency (EPA). It is estimated that except for
preconstruction environmental impact reports, a properly running citrus pro-
cessing plant that processes under 500,000 tons of fruit a year probably will not
need to deal with EPA on a continual basis. With good waste management,
even larger plants can escape close regulatory scrutiny.
353
D. Kimball, Citrus Processing
© Van Nostrand Reinhold, New York, NY 1991
354 CITRUS JUICE BY-PRODUCTS
AIR POLLUTION
Citrus processors and bottlers are not considered nationally to be major polluters
of the atmosphere. However, as with all manufacturers, their pollutant emis-
sions are significant enough to require some form of monitoring. Some of the
main sources of air pollution include boilers, evaporators, and furnaces, as well
as minor sources such as storage tanks, cooling equipment, and dried peel con-
veyance equipment. Rotting refuse and waste water are another source of off
odors that can contribute to atmospheric pollution.
The EPA has issued standards for ambient air quality, as summarized in Ta-
ble 22-1. Primary emissions are those that immediately escape the processing
equipment. Secondary emissions are those formed from primary emissions in
the air, which are of less importance to citrus processors than primary emis-
sions. Pollution problems often arise from errors in the feedmill where peel is
burned or volatile organic compounds (VOCs) are not sufficiently scrubbed. A
bluish gray smoke is indicative of d-limonene emission. Many local regulations
limit the VOCs that can be emitted, including the VOC content of paints and
thinners. For example, in Tulare County, California, only 420 grams of
VOC /liter of paint (excluding water) is allowed during drying below 194 OF
(90 ° C). The stringency of local regulations can be seen by comparing this
value to the federal maximum of 1470 grams VOC /liter of paint used in paint-
ing automobiles (45 FR 85414, December 24, 1980).
Sulfur compounds are considered one of the major pollutants in the combus-
tion of fossil fuels. In citrus processing, natural gas and oils, which contain
some sulfur, are the most common sources of energy. In Tulare County, Cali-
fornia, a rural area, the sulfur content must be below 0.7% in fuels. In Met-
ropolitan Los Angeles, the sulfur content must be 0.5 % or less in fuels. These
maximum values change from time to time, depending on existing conditions.
Local authorities generally require notification of changes from one fuel to an-
other or any other changes that may affect the pollution of the area. Good com-
munication and consideration for one's neighbors and the environment gener-
ally will prevent pollution regulation from becoming a serious issue.
SOLID WASTES
L1aUID WASTES
The primary waste from citrus processing plants and bottlers is in the form of
liquid or aqueous wastes. Aqueous wastes assume a variety of forms, from
relatively clean refrigeration water to highly organic discharges from oil cen-
trifuges. Regardless of the source, citrus waste streams basically contain four
types of contaminants: suspended and settleable solids (such as juice sac ma-
terial, pulp, and waxes), soluble organics (primarily sugars and acids), soluble
inorganics (caustic sodas), and volatile organics (d-limonene from peel oils,
etc.). Various methods and method combinations can be used to treat this type
of waste, depending on the need and the resources available.
Filtration
All waste treatment methods list filtration as the initial step. Self-cleaning grav-
ity-fed screens or shaker screens remove particulate matter, which can clog
equipment or lines as well as prove difficult to treat. Solid material thus re-
moved can be added to citrus pulp and processed in the feedmill, or collected
and transported to a remote disposal site.
356 CITRUS JUICE BY -PRODUCTS
Reuse Water
Waste effluents can be greatly reduced by reusing some of the waste streams.
For example, the relatively clean refrigeration water can easily be recycled
within the refrigeration system itself. Condensate water from the first effect in
evaporators can be reused in boilers as makeup water. Other condensate streams
from cooling towers can be used in belt sprayers to facilitate fruit conveyance,
as well as in some clean up hoses. This water may not be suitable for other
drinking or food processing purposes and should be so labeled, but potable
water that has been completely treated can be used anywhere in the plant. Be-
cause juice evaporators produces water, the waste water effluents will always
exceed water usage.
The introduction of fine air bubbles into static waste water is used extensively
in the food industry to separate grease, fats, and waxy materials. Under such
conditions, these materials have a tendency to float so that they can be skimmed
off and removed. In citrus processing plants OAF is used to help separate sus-
pended pulp and waxes from screened waste water and to aid in sludge settling
after aerobic treatment. As mentioned in Chapter 21, waxes washed from fresh
market culls produce a waxy sediment in waste streams, which can be separated
by OAF. The skimmed residue then can be added to citrus pulp entering the
feedmill. It is estimated that 10 to 20% of the effluent waste loading can be
removed by using OAF.
Flocculation
Irrigation
Municipal Disposal
Processing plants that exist in less-rural areas often use municipal sewage treat-
ment plants as a means of waste water disposal. This usually requires some
quality control monitoring, as disposal fees usually are based on the loading of
the effluent and the volume. Outside laboratory analyses often are used in order
to ensure the impartiality of the analysis results. In larger communities, citrus
effluents make minor contributions to overall waste streams; but in smaller com-
munities, citrus effluents run the risk of becoming political footballs, whether
or not that is warranted. Careful records should be kept regardless of actual
requirements.
358 CITRUS JUICE BY-PRODUCTS
Sample collection for such monitoring can be done in several ways. The
easiest and least accurate method is to take periodic grab samples. Waste water
composition varies dramatically throughout processing, and such grab samples
are not likely to be representative of the total stream. Periodic autosamplers are
costly and, again, may not accurately represent the waste streams. The most
accurate sampling system is a bleed line attached to the main waste stream,
which empties into a composite tank. This continuous sampling gives the clos-
est representation of the total stream. When sampling is to be done, the tank
contents are thoroughly mixed, a sample is taken, and the tank is drained so
that it can collect another composition of the waste stream.
Daily sampling usually is sufficient. Daily samples can be blended into weekly
composites, which in tum can be blended into monthly composites according
to the volumes of the waste water per time period. For example, if the daily
volumes were 50,142 gallons, 64,321 gallons, 36,478 gallons, 84,168 gallons,
and 47,171 gallons, you could make a weekly composite by adding 50.1 mlof
sample one, 64.3 ml of sample two, 36.5 ml of sample three, 84.2 ml of sample
four, and 47.2 ml of sample five, to get 282.3 ml of total weekly composite.
Samples should be well mixed before blending in order to contribute a repre-
sentative amount of settleable solids to the composite. Also, all waste water
samples should be refrigerated to avoid changes in composition before analysis.
The common tests performed on such samples can be found in the following
sections.
Aerobic Treatment
There are many types of microorganisms, each growing under a certain range
of conditions. In waste water treatment, aerobic microbes-those that require
oxygen to grow-are known to consume most organic wastes without the pro-
duction of off odors. Those organic wastes that can be thus consumed are called
biodegradable. Anaerobic microbes-those that do not require oxygen to grow
or require that oxygen be absent-produce waste products associated with rotten
odors, which are environmentally undesirable. For this reason, aerated lagoons
are used to induce aerobic growth in treating most organic wastes.
Microbial growth is very sensitive to certain conditions, one of them being
the acidity or pH. Aerobes grow best at neutral pH levels, but citrus effluents
generally have pH values of around 3 to 4, which means that pH control systems
may be required prior to aerobic treatment. Most municipal effluents are basic
and are partially neutralized by citrus waste streams. Shock loadings of low pH
or high pH values from caustic cleanups can seriously upset aerobic activity.
Also, waste streams that are left standing without aeration for long periods of
time, especially during warm weather, may undergo anaerobic growth that can
significantly reduce the pH. After the pH has been adjusted, the waste stream
WASTES FROM CITRUS PLANTS 359
or:
02(ppm) = (0.9916 - 3.049 x 1O- 5A)/(0.066 + 0.00216T) (22-2)
or:
1.136 x 10 14e-(A+1764110)2j9.616XIO 'O
02(ppm) - -----------=---- (22-4 )
6.025 x 1O- 6 (T + 156.8)2 - 0.07955
These equations are good from around O°C to 45°C, 502 mm Hg to 775 mm
Hg, and from - 540 feet to 11,273 feet above sea level. Once the oxygen meter
is calibrated, it can be used for several tests before recalibration is needed,
depending on the care of the electrode. The manufacturer's instructions should
be followed in the operation and care of the oxygen meter and the electrode.
When the DO levels in the aeration lagoon are too low, the aeration equipment
should be adjusted accordingly. The higher the loading, the higher the rate of
oxygen consumption will be. If insufficient aeration is performed, the water
will generate anaerobic growth, off odors, and low pH levels, and aerobic ac-
tivity will cease.
It also is important to monitor the d-limonene level in aerobic treatment. This
substance, obtained primarily from oil processing wastes, is toxic to most mi-
360 CITRUS JUICE BY -PRODUCTS
crobes and may affect aerobic activity (Murdock and Allen 1960; McNary,
Wolford, and Patton 1951). Water entering the aeration lagoon should have an
oil level of 0.005 % v Iv or less as measured by the Scott method (see Chapter
6). Oil strippers can be used on high-oil streams, or these oil-laden streams can
be isolated and treated separately. If the water content of high-oil streams is not
too high, they can be added to the press liquor in the feedmill and processed
into molasses, and then added back to the dried peel.
Determination of the biological oxygen demand (BOD) is the most common
way to find the loading or the necessary aerobic treatment for waste streams.
This is done by growing certain microbes in waste water samples and measuring
the rate of decrease of the DO, which usually takes 5 days. The seed bacteria
used in the test can be obtained from the aeration lagoon itself. The seed water
should be stored at 20°C for 24 to 36 hours, and 3.0 ml of the clear undisturbed
top layer should be used to seed the samples. The water used in the dilution of
the sample should be distilled from alkaline permanganate solution in order to
eliminate all organic residues. The use of water deionized by ion exchange is
not recommended because interfering organic residues may be contained therein.
The dilution water needs to be saturated with oxygen. This can be done by
filling one-gallon jugs three-quarters full with the distilled water and incubating
them at 20°C for at least 24 hours while sealed. The air in the head space above
the water will saturate the water. BOD nutrient buffer pillows can be purchased,
which can be added to a certain volume of dilution water to ensure proper pH
values and nutrients for microbial growth. The following procedure can be used.
Five-Day BOD
Procedure
1. Using the O2 meter, measure the DO of the sample, and record it.
2. Using the pipette bulb, pipette 1,2,3,4, and 5 ml of stirred sample each
into corresponding 300 ml BOD bottles.
WASTES FROM CITRUS PLANTS 361
3. Add 3 ml of seed water to each bottle, and fill the bottles to just below
the lip with the dilution water. When pouring water, decant it along the
sides of the bottle to avoid the formation of air bubbles. Stopper each
bottle, being careful not to entrap air bubbles, and invert the bottles sev-
eral times to mix the contents.
4. Remove the stoppers, and fill the bottles to the top so that no head space
will remain when they are capped. Incubate the samples (20 ± 1 DC) for
5 days in the dark.
5. After 5 days, remove the cap of each bottle, and insert the oxygen elec-
trode and measure the DO. of each. The oxygen meter should have been
previously calibrated, as described above in the text. When DO. readings
are made, some stirring may be required, but excess stirring will cause
atmospheric oxygen to dissolve in the solution and distort the results.
6. Plot the DO. readings versus milliliters of sample taken. At least three
points should be on a straight line. Discard any extraneous points. Deter-
mine the DO. consumed per milliliter of sample (the slope of the plot) and
the dissolved oxygen level at "0" sample, or where the line crosses the
DO. axis. This should be between 8 and 9 ppm DO., depending on the
elevation of the laboratory. The Bo.D can be calculated from:
where DO.; is the DO. where the line intersects the DO. axis, and DO., is
the DO. measured in the pure sample. The 300 refers to the Bo.D bottle
volume. The oxygen consumed is found by selecting two points used in
the plot in Fig. 22-1 and subtracting the DO. of the larger DO. (Do.larger)
from the DO. of the smaller DO. (Do.smaller) and dividing by the volume
of the smaller DO. (Vsmaller) minus the volume of the larger DO. (Vlarger)'
For example, if we obtained DO. readings as shown in Fig. 22-1 and mea-
sured a DO. value of 1 ppm in the original sample, we would obtain the
following by using Equation 22-5:
-
6
E
Q. 5
.s
0 4
C
1
0
0 2 3 4 5
mls of sample
Fig_ 22-1. Plot of DO (ppm) of incubated BOD wastewater samples versus milliliters of sample
used in the sample BOD problem.
or:
(22-7)
For example:
Even though BOD values give the amount of oxygen needed to aerobically
treat waste water, the analysis is time-consuming, requiring 5 days for comple-
tion. A more rapid method that can be used to estimate the BOD is the chemical
oxygen demand (COD). The COD can be measured in about 3 hours and pro-
vides for a good estimate of the BOD. Generally, BOD values range from 60
to 65 % of the COD values. In the COD determination, the organic material is
oxidized by the acid-catalyzed oxidizing agent potassium dichromate. After in-
cubation, the unreacted dichromate is titrated with a solution of ferrous am-
monium sulfate, which produces a greenish color. The endpoint is determined
by using a ferro in indicator, which reacts with excess ferrous ion to form a red
complex. The appearance of this red complex marks the end of the titration.
WASTES FROM CITRUS PLANTS 363
COD kits are commercially available and are recommended. These kits in-
volve less chemical handling and are faster and safer to use than the method
below. In the absence of such kits, the following procedure can be used.
COD Test
Procedure
For example:
9. Using the COD to estimate the BOD, you can multiply by 65% or 0.65
to get 4139 ppm BOD in the example.
WASTES FROM CITRUS PLANTS 365
Anaerobic Treatment
to adapt to low-pH and high-peel-oil conditions that can reduce the need for
pretreatment.
Anaerobic treatment also is used as a secondary treatment in aerobic methods
to remove those contaminates that aerobic treatment could not remove, espe-
cially volatile organics. Carbon granules sometimes are added to help absorb
volatile organics as well as to provide a site for anaerobic growth. Sludge pro-
duced in this secondary treatment generally needs to be disposed of in a remote
site in order to prevent off odors from lowering the quality of feed coming from
the feedmill. However, as mentioned above, less sludge is produced with an-
aerobic treatment than with aerobic treatment. After secondary anaerobic treat-
ment, a tertiary chemical oxidation may be employed, using chlorine gas to
remove remaining organics and nitrogenous compounds. A final carbon filter,
used to remove residual chlorine gas and volatile organics, generally will pro-
duce potable water that usually is of a higher quality than that used by the plant
from other sources.
Reverse Osmosis
The process of high pressure filtration (about 800 psi) through semipermeable
membranes has been investigated in the treatment of citrus effluents (Kimball
1982). This method, using both cellulose acetate membranes and non-cellulose
acetate membranes, can reduce COD levels by about 75 %. The filtered per-
meate can undergo further processing or may be recycled as belt sprays or in
other suitable applications. About 10% of the original volume of the untreated
waste water will emerge as a concentrate with the Brix increased about fourfold
(-0.5°Brix untreated to -2.0oBrix treated), which can be added to press li-
quor and can be concentrated to molasses and added back to the feed in the
feedmill. About half of the peel oil can be removed from the waste stream as
well, by using reverse osmosis. This method requires frequent cleaning and
durable membranes.
HAZARDOUS WASTES
Hazardous wastes may be defined as substances that can ignite, corrode, react
in a dangerous manner, or be toxic. Citrus juice plants utilize materials such as
cleaning agents, lubricants, fuels, refrigerants, boiler treatment chemicals, pro-
cessing chemicals, compressed gases, and laboratory chemicals that fall into
this category. Whenever these chemicals are spilled, contaminated, or decom-
pose, they become waste material that must be properly disposed of by using
licensed disposal services. Empty containers that contained hazardous material
also must be disposed of in proper fashion, according to local laws. Empty
WASTES FROM CITRUS PLANTS 367
caustic containers can be triple-rinsed to render them safe in most areas. Empty
closed-head oil or solvent drums that contained flammable material should never
be opened with a torch. Most chemical suppliers will pick up used drums or
partially used drums that contained hazardous materials for recycle. Empty con-
tainers (generally over 5 gallons) should never by simply discarded, with no
regard for proper and legal disposal requirements. Care should be taken in the
disposal of hazardous chemicals from laboratories. Direct disposal of such
chemicals down the drain, so that they end up in municipal treatment plants,
may cause a hazardous condition. Used compressed gas cylinders should be
promptly returned to the vendor and not allowed to accumulate. When food
containers are used for nonfood purposes, they should be labeled as such and
not used to contain food material again.
QUESTIONS
1. What are some of the sources of air pollution in citrus processing plants?
2. What is the "total sulfur oxides" maximum level permitted by EPA in
ambient air during a 24-hour period?
3. Why should inert solid wastes be removed from plant sites?
4. How can DAF aid in the treatment of aqueous citrus plant effluents?
5. How many pounds of BOD/acre day can Florida's sandy soils receive,
according to text?
6. What is the advantage of aerobic treatment?
7. What are the quality control parameters of aerobic treatment?
8. What DO levels must be maintained in aerobic treatment?
9. What maximum peel oil content is permissible for successful aerobic
treatment?
10. What is the advantage of COD measurements over BOD measurements?
PROBLEMS
1. Suppose that daily samples of waste water were collected for 5 days, as listed below,
corresponding to the given daily meter readings for gallons of effluent. How would
you make 200 ml of a weekly composite that would represent the week's effluent?
Monday: 61,452 gallons
Tuesday: 85,967 gallons
Wednesday: 31,002 gallons
Thursday: 46,972 gallons
Friday: 55,012 gallons
2. In air calibration of an oxygen meter, what would you set the meter to if you were
at an altitude of about 1600 feet at a temperature of 28°C?
368 CITRUS JUICE BY-PRODUCTS
3. Suppose you obtained the following data in a BOD analysis. What would be the BOD
(ppm)?
DO
Initial sample 1. 6
After incubation using:
I ml of sample 6.5
2 ml of sample 4.0
3 ml of sample 1.6
4 ml of sample 0.0
5 ml of sample 0.0
4. Suppose we obtained the following data in COD a analysis. What would be the COD
(ppm) and the estimated BOD (ppm)?
29.27 ml FAS titrated in 2nd blank
25.60 ml FAS titrated in 1st blank
24.78 ml FAS titrated in sample
5. How many estimated pounds of BOD are there in 156,472 gallons of waste water
with a COD value of 4367 ppm?
REFERENCES
Kimball, D. A. 1981. Unpublished data. California Citrus Producers, Inc., Lindsay, Calif.
Kimball, D. A. 1982. Unpublished data. California Citrus Producers, Inc., Lindsay, Calif.
McNary, R. R., Wolford, R. W., and Patton, V. D. 1951. Experimental treatment of citrus waste
water, Food Tech., 8, 319-323.
Murdock, D. I. and Allen, W. E. 1960. Germicidal effect of orange peel oil and d-limonene in
water and orange juice, Food Tech., 14, 441-445.
Perry, R. H. and Green, D. 1984. Perry's Chemical Engineer's Handbook, 6th edition. McGraw-
Hill Book Company, New York, 26.11.
Ratcliff, M. W. 1974. Unpublished research. Citrus World, Lake Wales, Fla.
Wheeler, R. W. 1977. Unpublished communication. FMC, Riverside, Calif.
Wood, C. 1973. Recycling citrus waste water, Sunshine State Agr., September-October, 6-7.
Chapter 23
Quality control of any product inherently involves the use of statistical methods.
Statistical mathematics is used to evaluate parameters that cannot be determined
absolutely. For example, it would be impossible to determine the precise di-
acetyl level in chilled juice being processed and packaged continuously from
fresh extracted fruit, as the juice composition is constantly undergoing change.
However, statistical or spot checking could give a good indication of what the
quality would be. Although the field of statistics is broad, only a few techniques
that may be of use to quality control personnel in citrus processing plants are
discussed here.
SIGNIFICANT FIGURES
Errors in regard to significant figures often occur in the citrus industry. Often
labels are encountered with Brix readings to two decimal places, or acid levels
are reported to one decimal place. The subject of significant figures is one of
the first studied in science classes, as the number of significant figures repre-
sents a way to show the accuracy of a measurement. For example, if we report
the milliliters titrated as 23, do we mean exactly 23 ml or 23 ± 0.5 ml? Ac-
cording to the rules of significant figures, the latter is true. If exactly 23 ml is
intended, we should write 23.0 or 23.00, depending on the accuracy of the
measurement. The reading of a standard buret generally is ±0.02 ml, which
means that all buret readings should be expressed to the hundredths decimal
place. Using the proper number of significant figures should not replace the
need to express error ranges. If we were to report the milliliters titrated as
23.234, we would be implying that we could make the measurement accurate
to the thousandths decimal place, which would be false. On the other hand, if
we were to report the milliliters titrated as 23.4, we would not be reporting the
accuracy of the measurement as close by as we could. A 23.4 ml reading really
means 23.4 ± 0.05 ml, whereas 23.40 means exactly 23.40 ml or 23.40 ±
0.005.
371
1. All nonzero figures are significant. For example, 123.45 = five significant
figures.
2. All zeros between nonzero figures are significant. For example, 3,076,002
= eight significant figures.
3. All zeros to the left of the first nonzero figure are not significant. For
example, 0.013 = two significant figures.
4. All zeros to the right of the last nonzero figure and to the right of the
decimal point are significant. For example, 3.2300 = five significant fig-
ures.
5. All zeros to the right of the last nonzero figure and to the left of the
decimal point mayor may not be significant. If the decimal point appears
(expressed decimal), then the zeros are significant. For example, 3700 =
two, three, or four significant figures, but 3700. = four significant fig-
ures.
6. Use scientific notation, when in doubt, to clarify the proper number of
significant figures in rule 5. For example, 3700 can be expressed as 3.700
X 103 , 3.70 X 103 , or 3.7 X 103 to indicate the exact number of intended
significant figures.
7. Some numbers have unlimited significant figures. For example, 5 people
means exactly 5 people because it generally is impossible to have a partial
person. In other words 5 people is the same as 5.0000000000000000000
people. Also, 60 seconds in one hour means the same as 60.00000000
seconds in one hour because exactly 60 seconds are in one hour. Absolute
or exact numbers have an unlimited number of significant figures.
8. When one is adding or subtracting numbers, the answer can be expressed
only to the number of significant figures that are completely known. To
begin with, one can insert any necessary zeros needed in the calculation
and then must round off to the nearest fully known figure. For example,
if we add or subtract the following:
25.3? 3455.346
+3.45 -321.1 ??
28.75 3134.246
or 28.8? 3134.2??
QUALITY CONTROL STATISTICS 373
The last figure(s) in the answer are not known for certain. Rounding off
to the nearest known figures would give 28.8 and 3134.2 for the answers
expressed to the correct number of significant figures. The least accurate
data determine the number of significant figures in the answer.
9. When one is multiplying or dividing, the total number of significant fig-
ures in the answer can be no more than the least number of significant
figures in the numbers multiplied or divided. For example, if we multi-
plied 34.56789 by 0.023, we would get an answer of 0.80. This is so
because even though the first number has seven significant figures, the
number with the least significant figures has only two. Thus, the answer
can have only two significant figures. In dividing 6.234 by 3.45633567,
we would get 1.804 because the least number of significant figures in the
data is four.
ERROR ANALYSIS
The expression of error is an inherent part of any precise science. All measure-
ments contain error, regardless of the effort or care exercised to avoid or min-
imize such. Error can be reported in many ways, including the following four
ways in which the error range can be expressed.
Actual Range
The simplest expression is the actual error range (AR). For example, if we have
three acid determinations of 3.41 %, 3.48 %, and 3.49 %, we can say the range
is between 3.41 and 3.49. This type of error analysis does not take into consid-
eration the fact that two of the three data points are close together with the third
somewhat distant from the others. It is useful for bracketing data within groups
of absolute occurrence, however. Mathematically it can be expressed as:
AR = Xmax - X min
Average Deviation
The average deviation, (AD) is calculated by averaging the deviation from either
the average value or the median value. (The median value is the value exactly
midway between the two extremes.) In the acid example above, the median
value would be 3.45 % and the average 3.46 %. In calculating the average de-
viation, the average difference from the median is the preferred method, rather
than the difference from the average value. In our example this would be:
(23-1 )
The square root of the average of the squared deviations from the mean is called
the root mean square deviation (RMSD). It can be calculated by using the fol-
QUALITY CONTROL STATISTICS 375
lowing expression:
(23-2)
This is similar to the average deviation except that one does not need to deal
with absolute values. However, the RMSD has an inherent bias and so is not
commonly used.
Standard Deviation
The standard deviation (SD) is the most commonly used means of determining
the error range of data. It is similar to the RMSD except that it eliminates the
inherent bias by multiplying the RMSD by the factor .In / (n - l). The SD can
be expressed as:
(23-3 )
The above equation requires calculation of the data mean (Xm). A more con-
venient form of the equation, which does not require the calculation of the
mean, is:
You will notice that the above answer is given to only one significant figure.
This is done because the % acid data is reported to the hundredths place; the
error range is reported likewise. Otherwise, if the error range were reported to
the proper number of significant figures (four here), it would appear to be more
accurate than the original data, which is not the case. Also, more accurate re-
sults are obtained if one does not round off the summations or any intermediate
answer until the final error range. If this practice had been followed, we would
have obtained a standard deviation of 0.04 instead of 0.05, using 35.9186 for
the first summation instead of 35.92. The former (0.04) is the more accurate
376 CITRUS JUICE MANAGEMENT
standard deviation, as seen in the use of Equation 23-3 with the sample data.
The calculation of standard deviations is a built-in feature of many calculators
and computers.
CONTROL CHARTS
Control charts have been used to monitor on-line quality throughout the food
industry. A control chart consists of upper and lower allowable limits, depicted
on graph paper as shown in Fig. 23-1. The actual levels of the target parameter
are plotted in relation to these limits to determine not only if the product is
within specifications, but if undesirable trends are developing.
The median value and limits of control charts are determined mathematically
by statisticians. Often, however, in the food industry, the upper and/or lower
limits are determined by specification or quality thresholds and limits. For ex-
ample, the maximum level of diacetyl acceptable in citrus juices is determined
by the taste threshold and not by the probability of finding a diacetyllevel more
than 95 % away from the mean value. This absolute nature of quality in food
processing requires changes in the interpretation of quality control charts.
Runs
A run on a control chart is a consecutive series of plots that are on the same
side of the median value. Runs found above the median value are called' 'runs
up" and those below the median value "runs down." Long runs that consist
of many consecutive points are indicative of a trend, and such trends may result
in the product drifting out of specifications if not corrected. The use of control
charts enables quality control personnel to detect such trends early enough to
Upper Limit
Lower Limit
TIME---...
Fig. 23-1. A typical control chart, including the median and upper and lower control limits.
QUALITY CONTROL STATISTICS 377
Applications
Quality control charts have been used in the monitoring of the diacetyl levels
in single-strength juice products and pulp levels during juice extraction, and
some attempt has been made to apply the use of control charts to fruit grading.
In the production of single-strength juices, if the diacetyl values constitute a
run up, spoilage is setting in, and the temperatures and/or retention times in the
pasteurizers need to be increased. If the diacetyl values form a run down, too
much heat may have been applied, and perhaps the pasteurizer temperature and/
or retention time should be reduced. The example of % pulp monitoring given
above is another illustration of the application of control charts to citrus pro-
cessing. One can monitor fruit graders by taking counts of broken and rotten
fruit leaving the grading table and plotting the results on a control chart. The
upper limits of fruit grading have been established by the USDA-1O% for
broken fruit and 2 % for rotten fruit leaving the grading table. The establishment
of a median value will permit the detection of runs up, which would require the
use of more graders or grading activity, or runs down, which would allow the
378 CITRUS JUICE MANAGEMENT
use of fewer graders or less grading activity. Control charts can be used for
packaging equipment, sample monitoring, and many other applications in con-
tinuous processing.
ANALYSIS OF VARIANCE
Suppose that three technicians measure the Brix on three samples of concentrate
as shown in the table:
1. Sum each column, square the sum, add the squares, and divide by the
number of items in each column. For example:
2 2 2
(1.2 + 1.3 + 0.9) + (1.4 + 1.0 + 1.2) + (0.8 + 0.9 + 0.9)
3
= 10.43
3. Sum all the items, square the sum, and divide by the total number of
items. For example:
4. Subtract the results in (3) from the results in (1). For example:
The variance ratio can be compared to Table 23-1, where VI equals the de-
grees of freedom for the columns, and V2 equals the degrees of freedom of the
residual. At the 0.05 probability level, the variance ratio must exceed 5.14 in
order for a significant difference to occur between the lab technicians. Because
3.33 is less than 5.14, we fail to reject the null hypothesis, which means there
is no significant difference between the lab technicians' from a statistical point
of view. That is not to say that the lab technicians are performing the Brix
measurements with sufficient accuracy; it only means that statistically they are
all performing equally at the 5 % level.
We can expand the analysis of variance to include the determination of
whether or not there is a significant difference between the concentrate samples
as well as between the lab technicians. This is called a two-way ANOVA. We
use a similar example to illustrate how this is done. Suppose that four lab tech-
nicians each analyze three samples of orange concentrate for Brix, as shown
below:
First, subtract 60.0 from each of the measurements in order to simplify the
calculations:
1. Sum each column, square the sums, add the results, and divide by the
number of cases in each column. For example:
2. Sum each row, square the sums, add the results, and divide by the number
of cases in each row. For example:
4. Sum all the cases, square the sum, and divide by the number of cases.
For example:
8. Subtract the results in (5) and (6) from (7). For example:
The variance ratios can be compared to Table 23-1, where VI equals the
degrees of freedom for the columns or rows, and V2 equals the degree of free-
dom of the residual. At the 0.05 probability level the variance ratio must exceed
4.76 for the columns and 5.14 for the rows in order for a significant difference
QUALITY CONTROL STATISTICS 383
REGRESSION
The best a and b values are those obtained when this sum of the squares is at a
minimum. In order to find the minimum point on this equation, we can try to
find the point where the slope is zero. This we can do by partially differentiating
384 CITRUS JUICE MANAGEMENT
Equation 23-5 with respect to each coefficient, and setting the result equal to
zero:
an+bL;x=L;y (23-8 )
a L; x + b L; x 2 = L;xy (23-9)
This gives two equations that can be used to find the two unknowns a and b,
where n is the number of (x, y) data pairs used in the analysis. Using Equations
23-8 and 23-9 to solve for a and b, we obtain the following:
L; x 2 L; Y - L; x L; xy
(23-1O)
Y
a=
n L; x 2 - (L; x
Thus, using these two equations, we can calculate the best values for the con-
stants a and b by performing the above summations, using the sample data
points x and y and the number of data pairs n.
For example, suppose we were interested in finding a linear equation that
would best describe the change of pulp level in freshly extracted orange juice
before and after evapomtion to 65°Brix. We first obtain the following data:
U sing Equations 23-10 and 23-11, we obtain the following parameters and the
resulting best a and b values:
:6 x = 85.3 a = -6.21
:6 y = 53.0
:6 x 2 = 1459.4 b = 0.985
:6 xy = 908.3
We can make a statistical prediction of pulp levels before and after evaporation
using the resulting equation:
Using this equation we can calculate predicted values and compare them to the
actual values:
This is a simple way to check the accuracy of the linear least squares best fit
of the data; however, the "goodness" of fit can be calculated mathematically
with more objective results by using the correlation coefficient, R2. The closer
the correlation coefficient is to 1.00, the better the data fit the regression equa-
tion. If the R2 equals exactly 1.00, the curve fits the data exactly, and exact
predictions can be made. For a linear equation, the R2 value can be calculated
from:
(23-14 )
In the above example, the R2 value becomes 0.892 or 89.2% of the y values
( % pulp after evaporation here) that can be accounted for using this least squares
equation.
It is important to keep in mind that "noisy" variations will give a lower
correlation coefficient than smooth or stretched deviations from the regression
386 CITRUS JUICE MANAGEMENT
line. Actual and calculated values should be checked to verify the correlation.
Also, the regression equation usually is valid only within the range of the data.
The linear least squares equation determined above may not fit the data as
closely as one would like. Because the above results represent the best linear
fit to the data, the only way to obtain a better fit is to use another equation-a
nonlinear equation. By using the principles illustrated above and substitution,
the method of least squares can be applied to a wide variety of equations. Also,
the least squares analysis can be programmed into computers for easy use. The
reader is referred to Kolb (1984), who provided not only the equations needed
to fit data to up to 19 different equations but also computer programs in a variety
of program languages, which can be used to determine regression relationships
automatically.
QUESTIONS
1. Give five areas where statistical methods can be applied to citrus quality
control in routine processing.
2. What is the purpose of using significant figures?
3. What is the difference between significant figures and the determination
of error ranges?
4. How are the control limits (upper and lower) usually determined in the
food industry?
5. How many points can constitute a run down in 32 measurements before
the process can be considered statistically influenced by some factor?
6. In what situation can an analysis of variance be useful in citrus quality
control?
7. Explain the null hypothesis.
8. Explain regression analysis.
9. What is the difference between regression analysis and least squares anal-
ysis?
10. How can the use of statistical analyses be greatly simplified in routine
use in citrus quality control?
PROBLEMS
1. How many significant figures does each of the following numbers have?
2,345.34
3000.1
2500
0.00230
6 dogs
QUALITY CONTROL STATISTICS 387
2. Express the answer to the following mathematical problems to the correct number of
significant figures:
3. What is the variance ratio (VR) for a one-way ANDV A analysis between lab tech-
nicians using the data below, and does it prove the null hypothesis?
4. What is the variance ratio for both rows (refractometers) and columns (lab techni-
cians) for the following data, and is there a significant variance between lab techni-
cians and/or refractometers?
5. It may be useful to have an equation that predicts the acid correction to the Brix base
on the % acid. Determine the linear least squares constants and correlation coefficient
using the following data. Compare the calculated Brix corrections with the actual
Brix calculations.
REFERENCES
Duncan, A. J. 1974. Quality Control and Industrial Statistics, 4th edition. Richard D. Irwin, Inc.,
Homewood, Ill., 659-662.
Kolb, W. M. 1984. Curve Fitting for Programmable Calculators. 3rd edition, Syntec Inc., Bowie,
Md.
Chapter 24
Quality control involves more than just laboratories and test tubes; a good qual-
ity control department requires administration and management. Managerial
policies and procedures vary widely from plant to plant, but the success of any
system is dependent on basic principles. Like any other employee, quality con-
trol personnel must understand the basic policies and objectives of the company.
Also, quality control departments are run by people who report to non-quality
control adminstrators.
Like all employees, quality control personnel should be shown courtesy and
respect. They should be given concrete goals and responsibilities as well as the
resources and latitude needed to achieve those goals. Quality control personnel
must be leaders, with the courage to speak up when something is going wrong.
They must be competent, and their competency must be recognized. They need
sufficient resources to perform their jobs, such as adequate laboratory space and
equipment.
The quality control laboratory is a major focus of outside visitors, who often
judge a company by the professional appearance of the quality control labora-
tory. The plant's "IQ" is often evalulated by the laboratory's veneer.
Good employees should have a vision of where they are going and why they
are there. Just as they are required to support management in decisions regard-
ing the operation of the plant, management should stand behind quality control
personnel in their quest to protect the quality of the company's products. More-
over, quality control procedures should be part of some type of plan that ulti-
mately will benefit the company. Such a plan must be managed; it should be
efficient, personally rewarding, and profitable.
OTHER DEPARTMENTS
As complex and intricate as a quality control department may seem, it com-
prises only a portion of the overall manufacture of citrus products; and books
could be written about the other areas as well. The quality control department
388
must interact with all of these other areas and act as a focal point for many of
the plant's decisions. It also acts as a helm that steers the course of many plant
activities.
One of the first things that should be done in setting up a quality control
department is to establish a realistic set of specifications, for both inbound and
outbound products. Quality control personnel can and should help to define
these specifications. The specifications should at least contain legal require-
ments as defined by the federal standards of identity, good manufacturing prac-
tice, and, if USDA grades are used, the requirements of the desired USDA
grade. Quality control then should be given the authority to reject products that
do not meet these specifications or sanitary conditions. A specification that is
not enforced is a specification that really does not exist.
It is important to keep the quality control department separate from the pro-
duction department because production often will have a tendency to cut comers
with respect to quality in order to get the product made and out on time. Quality
control, on the other hand, should be considerate of production's schedules,
and should not unduly delay or hamper their activities any longer than is nec-
essary to ensure quality products and sanitation.
Although company organizations may vary, the relationships between their
departments may have similarities. A general example of the communication
flow between departments is shown in Fig. 24-1. One should remember that
communication is a two-way street, and much follow-up communication is not
shown in the figure. The figure does show that quality control acts as a traffic
light that determines whether a product may proceed, or whether it must stop
and be rerouted.
In addition to routing quality control activities, the quality control department
may be called upon to become involved in other, non-quality control activities,
involving other departments. One such activity is the analysis of fruit samples
for maturity and juice yields, as mentioned in Chapter 5. The office generally
receives this information so that the owners of the fruit can be credited propor-
tionally. The sales department may ask quality control personnel to assist in
preparing and sending product samples to customers or potential customers.
Similarly, purchasing may ask quality control to analyze inbound product sam-
ples received from suppliers or potential suppliers. Quality control may be called
upon to work with research and development regarding new products, specifi-
cations, or complex problems. Quality control may and should be asked to assist
customers and suppliers in regard to quality problems and/or specifications.
PERSONNEL
The hiring of quality control personnel is generally left to the discretion of the
quality control manager. Job requirements and criteria usually are a function of
390 CITRUS JUICE MANAGEMENT
OUTBOUND INBOUND
mechanical?
yes
Fig. 24-1. Example of communication flow and interdepartmental relationships involving the
quality control department.
supply and demand in the local job market, but job descriptions for every po-
sition in the quality control department should be written and communicated to
personnel already on board as well as to potential employees. A new employee
should receive a complete orientation, including a written and an oral expla-
nation of the company and departmental policies, rules, and procedures. The
new employee should be given a comprehensive plant tour and be assigned an
on-the-job trainer. Every quality control department should have a manual out-
lining procedures and other information related to the quality control activities
QUALITY CONTROL MANAGEMENT 391
that are specific to the particular plant, and a copy of the manual should be
made available to both the new employee and older employees for reference.
The orientation should include instruction in the use of safety gear and safety
procedures (see below in this chapter). Also it should include a description of
the chain of command and the relationship that should exist between other su-
pervisors and members of management. As with any position, the one-boss rule
should be invoked so that communication up and down the chain of command
can go through one person in order to avoid confusion and conflicting direc-
tives. If more than one person is allowed to give quality control personnel di-
rectives, conflicts and confusion may arise.
Although training could begin right away, and even though the employee
may have existing experience and skills for the job, serious training should not
begin for several days in order to give the employee time to adapt to the new
environment. A new employee's success is determined to a significant degree
by how well he or she is allowed to adapt in the first week of employment.
Once the employee can perform the tasks contained in the job description at a
level that is satisfactory to the quality control manager, the initial training can
be considered complete. In addition to this initial training, all laboratory tech-
nicians should undergo continuous training, which may coincide with perfor-
mance evaluations. Cross training among employees of different job descrip-
tions protects quality control activities during times of sickness, layoff, leaves
of absence, vacations, and job changes.
Each employee should be evaluated by his or her supervisor at least once a
year, and should be rated on job performance as well as on how well he or she
has performed generally. A good employee may have poor technical ability,
and a good technician may find it difficult to arrive at work on time. Annual
evaluations provide a good opportunity to solicit employee input about the gen-
eral activities of the department and the company. This can be done during the
evaluation interview orally or by means of a written questionnaire. Many prob-
lems will not emerge without some degree of probing, and unless they do
emerge, they cannot be solved. Written exams on laboratory procedures and
tests as well as departmental and company policies stimulate thought and pro-
vide an excellent forum for instruction and review. Performance tests, where
all technicians analyze the same sample, can demonstrate the consistency of
analyses between lab technicians and ensure standardization in laboratory re-
sults. If possible, a program of incentives should be instigated to encourage
employees to improve and to excel, including such rewards as days off, longer
vacations, or cash awards. Morale is always high and work well done when an
employee feels that he or she is growing professionally and working toward a
goal. Evaluations also provide an important opportunity for the employee to
account for his or her performance. It is important for every employee to receive
appreciation and recognition for work well done.
392 CITRUS JUICE MANAGEMENT
such problems. A frank termination may be the last chance for such a person
to be jolted into getting the needed help. Employers must be sure that such
terminations are for the violation of company policies and not just the result of
someone's personal opinion about what ought to be done. It is advisable for the
employer and the employee to get professional advice in questionable situa-
tions.
EQUIPMENT MANAGEMENT
SAFETY
Safety training and accident avoidance should be integral parts of any industrial
operation. Mechanical, electrical, and fire safety should be emphasized in areas
where quality control personnel work. Shelf chemicals should be restrained in
case of an earthquake. Guards for belts and chains on equipment used by quality
control personnel in performing fruit maturity tests and yield tests should be in
place, and machinery should be locked out during cleaning. Employees should
not place their hands in machinery unless absolutely necessary, and even then
should do so only when the machine is locked out. Screw conveyers should
always be covered, to guard against the accidental insertion of a leg or an arm.
Employees should wear goggles and gloves when cleaning with caustic so-
lutions, and bump hats, hair nets, beard guards, and so on, should be used, not
only to avoid product contamination but to prevent hair from getting caught in
machinery. Likewise, loose clothing should not be worn around moving ma-
chinery. Proper hand rails and climbing equipment must be installed for sani-
tation inspections, the taking of samples, or the performance of other necessary
tasks. Quality control personnel should not climb onto tankers, rail cars, or
other moving vehicles until the vehicle has come to a complete stop, and the
driver is out of the cab. Finally, quality control personnel should avoid using
tools or equipment not included in their job description. Even if the lab tech-
nician is proficient in the use of such items, it could lead to unsafe and even
unsanitary conditions.
The written hazard communication program must include a list of the haz-
ardous materials in the plant and their location with an MSDS for each material.
Training should include methods to detect the release of the hazardous sub-
stance, information on physical and health hazards, and ways that employees
can protect themselves, especially in an emergency. It is a good idea to include
an outline of the training session in the written hazard communication program,
along with the date held and a list of employees who attended. For more details
on the Hazardous Communication Regulation, the reader is refered to the law
itself or to an appropriate consultant.
QUESTIONS
1. What is one of the first things that should be done in setting up a quality
control department?
2. Should production oversee quality control, or should quality control over-
see production?
3. What are some of the non-quality control activities that quality control
personnel may become involved with regarding other departments within
the company?
4. What is one of the biggest problems in any company?
5. Is it always true that an employee's personal problems are not the business
of the company? Why or Why not?
REFERENCES
Weast, R. C. 1968. Handbook of Chemistry and Physics, The Chemical Rubber Company, Cleve-
land, Ohio, F4.
Chapter 25
Inventory Management
SAMPLING
The sampling of bulk lots (drums, tankers, etc.) has been discussed in previous
chapters and is fairly straightforward. However, in continuous operations in the
manufacture of retail or consumer products, various sampling techniques must
be used because the product usually is not standardized or mixed to complete
uniformity. Questions arise about how many samples to take and analyze in
order to ensure representative sampling and how many defective samples can
be allowed. Numerous mathematical and statistical methods can be used to jus-
tify a wide variety of sampling plans. However, the simplest and easiest meth-
ods to consider can be copied from those sampling plans used by the Food and
Drug Administration (FDA) (CFR Title 21 145.3) and the U.S. Department of
Agriculture (USDA) (CFR Title 7 52.1-52.83), which are illustrated in Tables
25-1 and 25-2. These tables give the number of samples needed per total num-
ber of containers per lot and the number of permissible rejects. The USDA
sampling plan is based on container type or volume, and the FDA sampling
plan is based on the net weight. Because the FDA is looking for contamination
that would violate the standards of identity and/or health violations, its sampling
plan is more extensive than the USDA plan, which is looking for general grade
or quality characteristics. The continuous sampling performed by the USDA is
an involved procedure in which fewer samples are taken than the number listed
in Tables 25-1 and 25-2. Another sampling plan in the federal code is the one
397
FILL WEIGHT
Fill weights are important for some citrus products, and general methods for
their determination are regulated by the federal government, The "general
method for fill of containers" can be found in the Code of Federal Regulations
Title 21 130, 12(b), which describes how the percent fill of a container is to be
determined. The lid is cut out and the distance from the double seam to the
level of food (or juice) is measured. The food is removed, and the container is
cleaned, dried, and weighed. The container then is filled with water to 1'6 inch
from the double seam and weighed, and the net weight ofthe water is recorded.
Then the water, at the same temperature, is removed down to the previously
measured level of the food (or juice), and the net weight of remaining water is
400 CITRUS JUICE MANAGEMENT
determined. The latter water weight is divided by the former water weight and
multiplied by 100% to give the percent fill of the container. For grapefruit juice
and lemon juice, the standards of identity require a 90% fill of the container.
Even though there are no other" fill of container" requirements in the standards
of identity for other citrus juice products, generally a 90% fill is a standard
procedure in the food industry.
LABELS
Labels come in a wide variety of shapes, sizes, and colors. The nutritional
aspect of labeling has already been discussed, in Chapter 11. Retail or consumer
commodity labels generally are fixed or quite inflexible from lot to lot and must
comply with the Fair Packaging and Labeling Act (Public Law 89-755 89th
Congress, S. 985 November 3, 1966) as well as the regulations for the enforce-
ment of the Federal Food, Drug, and Cosmetic Act and the Fair Packaging and
Labeling Act (CFR Title 21 10 1.1-10 1.105). A new addition to the law is the
country of origin regulation, which states that 75 % ofthe imported juice (citrus
and/or noncitrus) must be declared by country of origin on all labels for both
retail and nonretail products. The standards of identity also have labeling re-
quirements, as do the USDA grades for retail products. In addition to govern-
mental requirements, industrial organizations have established guidelines for
labeling, such as those of the National Food Processors Association (1977).
Lawyers and graphic artists are asked to design labels, which then are used to
manufacture large inventories of labels and/or labeled containers. The quality
control department should monitor the general quality of labels and containers
as a part of sample analysis. Lot or day codes imprinted on retail containers
also should be inspected, as well as the fill levels and the container seal quality.
Nonretail or nonconsumer commodities (generally bulk) sold to other com-
mercial entities still fall under the labeling requirements of the standards of
identity. Each lot of bulk juice (drums, pails, tankers, etc.) should be clearly
marked as the product described in the standards of identity. Often a product
will meet the criteria for several standards of identity, and the seller can choose
the one he or she prefers to use to identify the product. (See Chapter 19 for
brief descriptions of the standards of identity.) Included with the product name
should be the name and address of the manufacturer, the lot number, the date
of manufacture, the net weight, and the Brix. Perhaps the Brix/acid ratio, tare,
and gross weights would be appropriate as well. If any other ingredients have
been added, such as preservatives or sugar, or any material that may affect the
standards of identity, those ingredients also should be declared on the label.
Defective product should be marked with a highly visible tag, such as a red tag,
indicating the nature of the defect and the lot number. This facilitates proper
handling of the defective product. Tanker or barge lots generally have the above-
listed information on the bill of lading.
INVENTORY MANAGEMENT 401
bologies. Labels can be customized on-site to fit the needs of any manufacturer.
Portable scanners can be used to scan the bar codes of bulk inventories and can
be programmed to process the data into any form or format. The portable scan-
ner then can be attached to a printer or a computer, and the finished manifest,
blend sheet, or inventory listing can be printed in the desired format with no
human error in counting or writing. This information also can go directly into
a central processing unit to automatically adjust product inventories, again with-
out human involvement. Examples of UPC/EAN symbology as well as an au-
tomatic identification drum label are shown in Fig. 25-1. Figure 25-2 shows a
UPC Symbol
1380 10121
CALIFORNIA CITRUS
PRODUCERS, INC.
P.O. 80x C
Lindsay , CA 93247 USA
UALENCIA
Concentrated Orange Juice
SAMPLE 1211
For Manufactur i ng
8PIX RAT ~T WT
162.eIL7.sl563 I
Dilute with 5 . 3
Pa.-ts Water
Fig. 25-1. Example of UPC lEAN bar code symbology and an industrial bar code label used
commercially for 52-gallon drums.
INVENTORY MANAGEMENT 403
CALIFORMIA CIYRU8
IPRODUCERS. INC.
P.O. Box C
525 E. Lindmore Ave.
Lindsay. CA 93247
(209) 562-5169
LOT NUMBER NUMBER OF DRUMS BRIX B/A RATIO DRUM NET WEIGHT
VB1758 4 59.8 12.5 560
VA1758 18 59.9 12.6 559
VB1698 1 59.5 12.6 562
VB1728 20 59.9 12.9 559
VC1728 20 60.0 12.9 558
VB1738 21 59.8 12.5 560
TOTAL DRUMS= 84
Fig. 25-2. Example of shipping manifest printed from data transmitted directly from a bar code
label scanner. Counting, sorting. and manifest formatting can be programmed into the portable bar
code scanning unit.
drum manifest printed from a scanning of drum bar code labels that was pro-
cessed and transmitted directly to a small inexpensive printer.
TANK MEASUREMENT
Another important area that quality control personnel often get involved with is
the determination of the volume of product in bulk tanks. Fluid levels are mea-
sured from a fixed standard point, or they are measured using various types of
level meters. Level meters designed for water measurements according to den-
sity usually are not adequate for juices or concentrates because their densities
are quite different from that of water. Once the level is determined , equations
can be used to calculate the volume of fluid in the tank. Net weights then can
be determined too , if needed. Most processing tanks have the shape illustrated
in Fig. 25-3, with a dome top, cylindrical center, and slant bottom . As the
product fills from the bottom, each of the sections is filled , with a varying
relationship existing between the product level and the volume of the tank .
The slant bottom portion is divided into two sections, the first half and the
second half, as shown in Fig. 25-4, where the shaded area represents the prod-
uct or fluid in the tank. As the product fills the first half of the slant bottom,
the top surface area of the product may be represented as a segment of a circle,
404 CITRUS JUICE MANAGEMENT
DOME TOP
CYLINDRICAL
CENTER
Fig. 25-3. Typical shape of bulk tanks used in citrus processing, with three main sections.
-------p
~a ~
hO
SIDE VIEW
Fig. 25-4. Isometric views of the slant bottom portion of bulk tanks and dimensions used to derive
volume equations.
INVENTORY MANAGEMENT 405
(25-1)
with the variables as depicted in Fig. 25-4. Looking at the side view and using
a trigonometric relationship we get:
(r - y) / h = tan () = 2r / ho (25-2)
With inches used as the unit of measurement, the volume of the slant bottom
during filling of the first half becomes:
(25-4 )
Substituting Equation 25-1 and 25-3 into Equation 25-4 gives the volume of the
first half of the slant bottom as a function of only one variable, the measured
vertical height h, as follows:
Vb = -hr
2
(7r- + 2 (2h
- - 1 ) -J(l - h/ho)h/ho - sin- I ( 1 - -2h)) (25-5)
~ 2 ~ ~
(25-6)
which is similar to Equation 25-1 except for a negative sign. Using the same
treatment as before, we obtain the following relationship for the volume of
product in the slant bottom as the product fills the second half of the slant
bottom:
which is similar to Equation 25-5 except for the signs of the last terms.
Once the total bottom section is filled, the total volume can be calculated
406 CITRUS JUICE MANAGEMENT
(25-8)
(25-9)
By adding the filled volume of the slant bottom section to Vc ' the relationship
between the product level and the volume while the level is within the cylin-
drical section becomes:
(25-10)
When the cylindrical and slant bottom portions are full, the dome top begins
to fill, as shown in Fig. 25-5. The volume of a segment of a sphere can be
expressed as:
using the dimensions in Fig. 25-5. The radius of the sphere must be determined
by measuring the curve of the dome, transposing it to paper, and graphically
measuring the radius. Equation 25-11 represents the total volume of the dome
portion of the tank. The product volume in the dome is determined by subtract-
.......1 - - - - - - 2r -------1.~1
~
v
+
Fig. 25-5. Dimensions useful in detennining the volume of the dome top in citrus bulk tanks.
The radius R must be detennined graphically or by measurement.
INVENTORY MANAGEMENT 407
ing the volume of the spherical segment of space remaining above the product
from Equation 25-11. After simplification we obtain the following:
With the cylindrical portion filled, the term h in Equation 25-10 can be ex-
pressed as the fixed constant d for the length of the cylindrical portion of the
tank, and the total volume of the tank with product in the dome top can be
expressed as follows:
(25-13 )
The total height from the bottom of the tank to the product level in the dome
top is:
H,=ho+d+v (25-14 )
To illustrate the use of these equations, we will take an imaginary tank and
calculate the volume of fluid at different levels. Suppose the tank has a diameter
of 12 feet (r = 72 inches), a slant bottom of 12 inches vertically, a cylindrical
length of 15 feet (180 inches), a dome top with height of 3 feet, and a spherical
radius of 64 feet (78 inches) . We measure the total height H from the very
bottom of the tank and determine the volume of product in the tank by using
the various equations shown above. Table 25-3 summarizes the data and the
results. With the hatch set at 6 inches above the cylindrical portion of the tank,
the maximum v value is 6 inches. Paddle or propeller agitators in product tanks
displace a portion of the product, and their volume can be subtracted from the
product volume. However, this displaced volume generally is not significant.
If a tank has a spherical bottom, the volume with the juice level in this spher-
ical segment again can be described as in Equation 25-11, but with Vo now
representing the actual fluid level. This form of Equation 25-11 then can be
used in place of the equations used for the slant bottom portion in the above
procedures.
Using the above-derived equations every time a volume determination is
needed is awkward and time-consuming. The best way to use such equations is
to generate a table consisting of a level measurement and the corresponding
volume. Such charts or tables can be used quickly and effectively, not only by
quality control but by production personnel as well. Programmable calculators
or computers are ideally suited for such table generation. A flow chart that can
be used to program computers to generate a volume table is illustrated in Fig.
25-6.
A problem commonly found with some computer languages, such as RPG,
is the inability to perform the sin -I function. In such cases, a Taylor series
expansion (shown below) out to six terms gives sufficient accuracy in most
cases:
Tankers
The Internal Revenue Service and general company accounting procedures gen-
erally require an annual inventory of products and supplies. Even if no inven-
tory is required, a knowledge of equipment and supplies on hand will help the
quality control manager to utilize resources efficiently and will facilitate the
timely restocking of supplies. Inventories can be classified according to such
INVENTORY MANAGEMENT 409
VOLUMES IN SLANT-
BOTTOM TANKS
Tk Btm to Juice (Ht)
Ht of Slant Btm (ho)
Tank Radius (r)
Ht Cyl Portion (d)
Radius Dome Top (R)
Max Ht in Dome Top (Vm) EO 25-5
Vb or Vt
EO 25-7
Vb' or Vt
EO 25-10
Vc+Vb or Vt
yes EO 25-13
Vb or Vt
TANK OVERFILL!!!!
Fig. 25-6. Flow chart that can be used to generate a volume table for a bulk tank using the
equations in the chapter.
410 CITRUS JUICE MANAGEMENT
QUESTIONS
PROBLEMS
1. If the FDA sampling plan was used to determine the number of samples needed for
50,000 half-gallon containers of orange juice from concentrate, how many samples
should be taken, and how many rejects would be allowed? If the half-gallon con-
tainers were produced at the rate of 100/ min., at what time intervals should samples
be taken using this plan?
2. Consider a bulk tank with a radius of 150 inches, a vertical slant bottom of 36 inches,
a dome top height of 24 inches, a cylindrical section 250 inches high, and a dome
spherical radius of 300 inches. The hatch is 8 inches above the top of the cylindrical
section. What is the maximum capacity of the tank?
INVENTORY MANAGEMENT 411
3. If the tank in problem 2 were filled to a height of 200 inches from the vety bottom
of the slant section, what would be the product volume?
4. If the tank in problem 2 were filled to a height of 30 inches from the bottom of the
slant section, what would be the product volume?
5. Suppose a tank had a spherical bottom with a depth of 18 inches and a spherical
radius of 120 inches and a product level of 13 inches. What volume of juice is in the
tank?
REFERENCES
Knill, B. 1988. New directions in automatic identification, Industry Week, 237(4), AI-A36.
National Food Processors Association Labeling Manual, August 1977.
UPC Council, Inc. 7051 Corporate Way, Suite 201, Dayton, Ohio 45459.
Appendix A
412
ACID CORRECTIONS TO THE BRIX 413
C***************************************************************************************
C* RPG CALCULATIONS SUBROUTINE THAT CAN BE USED TO CALCULATE
C* THE LBS SOLIDS/GALLON FROM THE BRIX USING EQUATION 2-10
C***************************************************************************************
CSR SUBI BEGSR
CSR Z-ADDI COUNT 20 INITIALIZE
CSR Z-ADDl.OOOOOOO TAYLOR 157 VARIABLES
CSR BRIX ADD 330.872 FO 154
CSR Z-ADDl.OOOO Fl 158
CSR Z-ADD 1.0000 F2 154
CSR Z-ADDl.OOOO F3 154
CSR Fl MULTF Fl H CALCULATE
CSR F2 MULT 170435 F2 FACTORS
CSR Fl DIV F2 Fl H FOR
CSR Fl DIV F3 Fl H TAYLOR
CSR Fl MULT FO Fl H SERIES
CSR Z-ADDFI F4 158
CSR RET TAG
CSR SETOF 10 TAYLOR
CSR TAYLOR ADDF4 TAYLOR SERIES
CSR F4 MULT F3 F4 H SUMMATION
CSR F4 MULT Fl F4 H
CSR COUNT ADD 1 COUNT H
CSR F3 MULT COUNT F3 H
CSR F4 DIV F3 F4 H
CSR COUNT COMP20 10
CSR N10 GOTO RET
CSR SETOF 10
CSR TAYLOR MULT .0437691 TAYLOR H
CSR TAYLOR MULT BRIX SPG 43H FINAL SPG
CSR ENDSR VALUE
418
GWBASIC AND RPG PROGRAMS 419
10 'THIS PROGRAM CALCULATES THE UNCORRECTED BRIX NEEDED FROM THE EVAPORA-
TOR IN ORDER TO ACHIEVE A DESIRED GPL IN PROCESSING LEMON CONCENTRATE
20 INPUT "Uncorrected Brix of inbound lemon single strength juice"; UBJ
30 'IF REFRACT CORRECTS FOR TEMP, DISREGARD STEPS 40-60
40 INPUT "Temperature for Brix correction to lemon SSJ";T
50 BB=UBJ
60 GOSUB 340
70 INPUT "mls of NaOH titrated with SSJ";MLS
80 'IF NaOH NORMALITY IS ALWAYS 0.3125N THEN DISREGARD STEPS 90-110
90 INPUT "Normality of NaOH";N
100 IF N=O THEN N=.3125
110 MLS=MLS*N/.3125
422 APPENDIX B
10 'THIS PROGRAM CALCULATES THE BRIX, ACID, AND RATIO FROM LABORATORY DATA
20 INPUT "Uncorrected Brix";UB
30 'IF AN AUTO TEMP CORRECTING REFRACT IS USED, DISREGARD STEPS 40-80
AND ANY REFERENCES TO "TC"
40 INPUT "Temperature for Brix correction";T
50 A=UB'2*( .0001425-8.605E-06*T+7 .138E-08*T'2)
60 AA=UB*( -.02oo9+.001378*T-1.857E-05*T'2)
70 AAA=-.7788+.017*T+.00II*T'2)
80 TC=A+AA+AAA
90 INPUT "mls of NaOH titrated";MLS
100 'IF THE NORMALITY OF THE NaOH IS ALWAYS 0.1562, DISREGARD STEPS 110-130
110 INPUT "Normality of NaOH";N
120 IF N =0 THEN N = .1562
130 MLS=MLS*N/.1562
140 INPUT "Weight of juice sample (1O.5g = 10 mls SSJ)";W
150 ACID=MLS/W
160 AC=.014+.192*ACID-.00035*ACID'2
169 'THE FOLLOWING FIXES THE CORRECT DECIMAL PLACES
170 BRIX=CINT((UB+AC+TC)*lO)
180 BRIX=BRIX/lO
190 ACID=CINT(ACID*100)
200 ACID=ACID/loo
210 RATIO = CINT((BRIX/ACID)* 10)
220 RATIO=RATIO/I0
230 PRINT BRIX; "Brix"
240 PRINT ACID;"% acid"
250 PRINT RATIO;"B/ A ratio"
260 INPUT "Desired Brix";DB
270 INPUT "Volume of high Brix juice"; V
280 SPGI= .0437691 *BRIX*EXP((BRIX +330.872) '2/ 170435!)
290 SPGF= .0437691 *DB*EXP( (DB+330.872)'2/ 170435!)
300 WATER=CINT(V*(SPGI-SPGF)/SPGF)
GWBASIC AND RPG PROGRAMS 425
10 'THIS PROGRAM DETN'S THE AVG OIL AND OIL ADJUSTMENT OF A BLEND
20 DIM B(80)
25 DIM 0(80)
30 DIM V(80)
35 DIM SPG(80)
50 N=N+l
60 PRINT "Brix of component ";N;
70 INPUT B(N)
75 IF B(N) =0 THEN GOTO 130
76 PRINT "% oil of component ";N;
77 INPUT O(N)
78 O(N)=CINT( looo*O(N»/I000
80 PRINT "Volume of component ";N;
90 INPUT YeN)
95 SPG(N)=.0437691 *B(N)*EXP( (B(N)+330.872 )"2/ 170435!)
100 Z=Z+V(N)*SPG(N)
105 ZZ=ZZ+ V(N) *SPG(N) *O(N)
110 V=V+V(N)
120 GOTO 50
130 N=N-l
135 PRINT" #"," BRIX"," OIL", "VOL"
140 FOR 1=1 TO N
150 PRINT I,B(I),O(I),V(I)
160 NEXT
170 S=Z/V
1750=CINT(l000*ZZ/Z)/1OOO
180 BF=S*1O
190 B=BF
200 BF=S!( .0437691 *EXP( (B+330.872) "2/170435!»
210 IF ABS(BF-B)< .0001 THEN GOTO 230
220 GOTO 190
428 APPENDIX B
230 BF=CINT(BF*IO)/IO
250 PRINT "AVG BRIX = ";BF;SPC(3);"AVG OIL = ";0;SPC(3);"TOT VOL=";V
255 INPUT "Do you want to add oil (Y /N)";Q$
256 IF Q$ = "N" THEN END
260 INPUT "Desired final oillevel";OF
261 INPUT "Is volume in drums (52-gallon)(Y /N)";Q$
265 IF Q$="Y" THEN DD=52 ELSE DD=I
2700A=CINT«OF-0)*36.19*S*V*DD)
280 PRINT "Oil needed = ";OA
290 INPUT "mls oil enhancer needed per specification";OE
295 0I=OF-OE/(V*S*36.19*DD)
300 IF OA> =OE THEN PRINT "mls oil enhancer needed = ";OA: END
310 INPUT "Can the total volume of the blend change (Y /N)";Q$
330 INPUT "Number of blend component you wish to exchange";NN
340 FOR J = I TO N
350 ZZZ = ZZZ + SPG(J)*V(J)*(OI -0(1))
360 NEXT
370 ZZZ=ZZZ-SPG(NN)*V(NN)*(OI-O(NN»
375 01 =CINT«OI + ZZZ/(SPG(NN)*V(NN»)*lOoo) / 1000
380 IF Q$="N" THEN PRINT "% oil needed in component";NN;" =";01
390 V I = CINT(ZZZ / (O(NN)*SPG(NN)*( 1 - 01 /O(NN))))
400 IF Q$="Y" THEN PRINT "Volume needed in component";NN;" = ";VI
410 ZZZ=O
420 GOTO 310
10 'THIS PROGRAM DETN'S THE AVG PULP AND PULP ADJUSTMENT OF A BLEND
20 DIM B(80)
25 DIM P(80)
30 DIM V(80)
35 DIM SPG(80)
50 N=N+l
60 PRINT "Brix of component ";N;
70 INPUT B(N)
75 IF B(N)=O THEN GOTO 130
76 PRINT" % pulp of component ";N;
77 INPUT peN)
7 8 peN) = CINT( 1000*0(N» / 1000
80 PRINT "Volume of component ";N;
90 INPUT YeN)
95 SPG(N)= .0437691 *B(N)*EXP«B(N) +330.872 (2/ 170435!)
100 Z2 = Z2 + V (N)*SPG(N)
105 ZI =Zl + V(N)*SPG(N)*P(N)
120 GOTO 50
130 N=N-I
135 PRINT " #"," BRIX"," PULP", "VOL"
140 FOR 1=1 TO N
150 PRINT I,B(I),P(I),V(I)
160 NEXT
170 PAVG=ZI/Z2
250 PRINT "AVG PULP=";PAVG;SPC(3);"TOT VOL=";V
310 INPUT "Can the total volume of the blend change (Y /N)";Q$
GWBASIC AND RPG PROGRAMS 429
HP-41 C Programs
The following HP-41C or HP-41CV programs can best be used by assigning the pro-
grams to convenient keys using the ASN function. Then the program will activate when
the calculator is in the USER mode. The programs are designed for the calculator to be
alloted 53 storage registers using the SIZE function (SIZE 053). Included with each
program is a brief explanation of how to use the program. It is assumed that the user is
familer with the HP-41C program language. Most programs are self-explanatory. You
enter the data when prompted. Consult the text and relevant flow charts for program
logic and the objectives of the programs.
This program will calculate the parameters needed to adjust a full tank of juice to the
proper Brix by adding either water or higher Brix concentrate. If no entry is made for
the needed Brix, the program defaults to 11.8 °Brix. If the prompt does not apply, enter
nothing.
430
HP-41 C PROGRAM 431
This program will calculate the Brix, % acid, and B / A ratio from laboratory results. No
entry when you are prompted for the temperature causes the program to default to 20°C
or to apply no temperature correction to the Brix. If nothing is entered for the normality,
the default is 0.1562N. This can be changed in the program to 0.3125N if desired. If no
entry is made when you are prompted for the sample weight, a weight of 10.5 g (10 ml
of SSJ) will be assumed.
Each program below should be assigned a USER key. The BLD G?D program sets the
volume mode of the blend. Gallons mode assumes all volumes are in gallons. Gal/drums
mode assumes that all volumes less than 52 represent the number of 52-gallon drums,
and all volumes 52 or over are gallons. For example, an entry of 45 would mean 45
drums of concentrate, whereas 58 would be taken as 58 gallons of concentrate. This
allows the blending from bulk tanks and drums simultaneously without conversion of
one quantity to the volume units of the other. Each component is numbered and appears
in the prompts and displays. In the view program, if you enter nothing when prompted
for the number of the blend component desired, the default is the first component. The
calculator then will automatically step through the entered blend components for com-
parison to a blend manifest. You can add, delete, correct, or view blend components at
any time by using the separate assigned USER keys. The blend calculation takes a few
minutes. The calculation program also will prompt for the desired Brix and will calculate
the water needed in the blend, as well as the resulting final volume of the blend.
This program can be used to calculate fruit sample data for oranges, tangerines, grape-
fruit, or other citrus fruit except lemons and limes. The default temperature for the
temperature correction to the Brix is, again, 20 C or no temperature correction. The
D
default normality is 0.1562, and the default sample weight is 10.5 g (10 ml SSJ).
This program will calculate the average % oil or % pulp in a blend, as the two calcu-
lations are similar. The difference between oil and pulp is determined by the magnitude
of their values. The Brix of each component must be entered, as well as the % oil or
pulp. The volume prompt assumes that all units are the same (all drums or all gallons).
When all the components have been entered, enter nothing, press R/S, and the weighted
average % oil or pulp will appear.
This program calculates the fruit sample information for lemons and limes. The default
for temperature is, again, 20°C; the default for the NaOH normality is 0.1562; and the
default for the sample weight is 10.5 g.
This program will calculate the uncorrected Brix needed by evaporator operators in order
to concentrate lemon or lime juices to a desired GPL level. The temperature, normality,
and sample weight defaults are as before-20°C, 0.1562N, and 10.5 g, respectively.
The default for the desired GPL is 400,GPL. If you want to subtract an acid correction
from the Brix observed by the evaporator operator, then enter the temperature of the
concentrate observed by the refractometer.
This program calculates the GPL from Brix and acid measurements. The temperature,
normality, sample weight, and desired GPL defaults are 20°C, 0.3125N, 10.5 g, and
400 GPL, respectively. Any volumes can be used, but you must use the same units
throughout the calculation. The program also calculates the water needed to adjust high
GPL concentrates or juices.
This program is convenient when dilution to a desired Brix is all that is wanted. The
default desired Brix is 11.8°Brix. Any units of volume can be used as long as they are
consistent.
This program will generate data to construct a volume table for a particular bulk tank.
After entering the needed tank dimensions (see text for greater details), you enter the
distances from the hatch to the juice level, and the equivalent volume of the tank will
appear. The hatch depth is the distance from the lower lip of the hatch to the edge of
the tank.
Chapter 2
Questions
3. No.
5. Temperature, acid level, lab technician or techniques, refractometer calibration,
wavelength or color of light source, sample concentration.
Problems
Chapter 3
Questions
Problems
4. 4.82% acid
7. 60.6°Brix
10. 245 gallons of the 398 GPL concentrate, 3146 total gallons.
444
ANSWERS TO SELECT QUESTIONS AND PROBLEMS 445
Chapter 4
Questions
Problems
Chapter 5
Questions
4. Because the smaller and irregular size of the lemon fruit causes more fruit to be
lost during conveyance, and the nonspherical shape results in less efficient juice
extraction.
Problems
Chapter 6
Questions
Problems
Chapter 7
Questions
2. Increased.
3. The yield would be low, also.
9. Pulp wash may be used in drinks or beverages, or in other food products.
Problems
4. 4.7% pulp.
5. 1.8% pulp.
Chapter 8
Questions
Problems
Chapter 9
Questions
2. California orange juice is usually deeper in color than Florida orange juice be-
cause of California's drier climate.
Chapter 10
Questions
4. USDA: no standards. Florida: Grade A, 600 ppm naringin and 5.0 ppm limonin;
Grade B, 750 ppm naringin and 7.0 ppm limonin.
7. Test the juice for limonin, heat, and retest until the heating no longer increases
the limonin level.
ANSWERS TO SELECT QUESTIONS AND PROBLEMS 447
Problems
Chapter 11
Questions
Problems
Chapter 12
4. Extractors.
5. 58 sec-I.
448 APPENDIX D
Chapter 13
Questions
Chapter 14
Questions
Chapter 15
Questions
9. This minimum plate count is unknown, perhaps in the tens of thousands or higher.
12. By a pure random correlation; or, it does not correlate.
14. Several months.
Problems
1. Too many to count (TMTC) or too numerous to count (TNTC) , as over 300
colonies appeared on the plate.
4. 3.2 ppm.
Chapter 16
Questions
Chapter 17
Questions
Chapter 18
Questions
Problems
1. 2.35 - 2.27 meq/IOO ml = 0.08, less than the 0.10 that produces off flavors.
Therefore, the shelf life has not expired, but it is about to do so shortly.
3. Ksp (O°C) = 0.014 M2
Chapter 19
Questions
Problems
4. No. Early-season juice is authentic and may contain this much acid.
450 APPENDIX 0
Chapter 20
Questions
Problems
3. 688 pounds of citric acid, 13,134 gallons of water, 13,838 pounds of sugar, and
5797 gallons of juice.
5. 0 pounds of citric acid, 5,645 gallons of water, and 5,492 pounds of sugar.
Chapter 21
Questions
Problems
1. 25.0.
5. 6.3% alcohol; no.
Chapter 22
Questions
Problems
Chapter 23
Questions
3. Significant figures determine the number of decimal places that legitimately can
be used but do not indicate the actual error range. Error analysis does not indicate
the precision of measurement, only the range of error.
4. By specification.
9. Least squares analysis is a form of regression analysis that uses calculus to math-
ematically determine the best fit of a particular curve to particular data.
Problems
Chapter 24
Questions
Chapter 25
Questions
Problems
453
454 INDEX
Deaeration of juice for Brix measurement, 9, concentrated juice addition to, 328
10, 12 core wash use in, 115
Debittering (see also Limonin) 2, 156-158, flavors in, 323, 324
280 flow chart to formulate a drink base,
adsorption resins use in, 157, 158 333
auxin sprays use in, 157 foaming agents in, 326
immobilized bacteria use in, 157 juices used in, 321, 322
insect antifeedant use of wastes from, 2 low pulp juices in, 107
neodiosmin use in, 157 nutrients in, 325
of core wash, 115 preservatives in, 325, 326
supercritical carbon dioxide use in, 157 sample base formulation, 334
Decolorization of grapefruit, 134 sugar addition to, 327-329, 333
Defoaming agents in juice drinks, 326 sugar correction in, 331, 332, 334
Density (see also Specific gravity), 21,94 sulfites in, 325
of juices, 7-9,13,14,31,40,47,48,50, sweeteners used in, 322
52, 68, 70, 148, 150, 154, 166, 324- texturizers in, 324
328,331,383,393,408 vitamin C addition to, 325
of pellets, 341 vitamin C as an antioxidant in, 326
of oils, 87, 94, 95 water addition to, 328-330
of water, 94, 95, 403 water quality used to make, 334, 335
optical, 94, 143 Drosophila fruit fly, 246, 247
radioactive, 178 light effects on, 248
tables of, 8 Dutch standards, 292, 295
temperature corrections to juice, 31
a-terpineol, 264
Deoiling of single strength juice, 83 Employee discipline, 392, 393
Diacetyl Employee evaluations, 391
determination of, 232-234 Employee training, 391
non-random effects of, levels, 377 Enzymes
statistical analysis of, 371 aconitase, in citrus acid cycle, 35
Dried juice sacs, 110-112 deactivation of pectinase, 83, 119, 120,
color of, 112 177,239,281
composition of, (table), 112 deactivation of pectinase in juice sac
Dried peel, 339-341 production, 110
dust from, 339 effect of. on pressed peel due to liming,
heat production from, 339, 340 342
moisture addition to, by oil recovery immobilized, used in debittering, 157
systems, 340 immunoassay limonin determination, 142-
moisture content of, 339, 340 144
moisture test, 340 lemon juice clarification using, 124
nutrients in, (table), 341 melanin formation via, 271
Drinks, 303, 320-335 oxidation via, 260
acid addition to, 327-331, 333 PEV measurement of, activity, 121-123
acid correction, 331, 332, 334 protein as, 164
acids used in, 322 Steven's cloud test, 120
anti foaming agents in, 326 thiamine as coenzyme, 171
base formulation, 320, 326-331, 333 use of, as drink preservative, 326
calcium addition to, 325 in first stage of pulp wash, 124
~-carotene, 323 in Florida not allowed, 124
clouding agents in, 324 in oil production, 85
colors in, 322, 323 in pectin extraction, 304, 305
computer applications, 333, 334 in production of cold pressed oil, 87
458 INDEX
potassium citrate in, juice, 273, 274 deoiling of single strength juice, 83, 88, 89
use of, juice in drinks, 321 effect on alcoholic fermentation of
Neodiosmin, 157 molasses, 345
Nomilin, precurser to limonin, 140 essence oils, 88
Nootkatone evaporative residue, 97, 98
from valencene in excessively heated FCC specifications for, 89-91
essence oils, 88 flavor enhancer use of, 73
unique component in grapefruit oil, 74 flavor source for juice drinks, 323
Nutrients, 162-172 folded oils, 88
calcium, 172,321,322,325 freshly extracted juice levels, 83
calories, 164 grapefruit oils, 86
carbohydrates, 165, 166 in citric molasses fermentation, 347
drink base components, 325, 326, 334 in sample drink base formulation, 334
fat, 166 insect barrier of, to fruit, 245
flavonoids, 172 juice oils, 88
labeling, 162-167 levels in press liquor, 342
labeling example, 163 moisture addition to peel from, recovery,
macronutrients, 164 340
micronutrients, 166, 167 non-food product use, 73
minerals, 166, 167, 172 optical rotation of, 89, 92, 93
pectin, 172 production of, 83-99
potassium, 166 refractive index of, 93, 94
protein, 164, 165 removal during debittering, 157
sample plan for determination of nutrional Scott method of determination of, 74-79
label compliance, 398 significant figures of, levels, 373
sodium, 166 specific gravity of, 94, 95
thiamine, 171, 172,334 standards of identity allowance, 281
vitamin C, I, 166-172, 188,202,325 steam distilled, 87
vitamins, 166, 167,334 storage of, 85
taste threshold of, 79
Occupational Safety and Health Administra- USDA grades of, in juices, 80, 83
tion (OSHA), 394, 395 USP specifications for, 89-91
Oils, 2, 73-99, 303 use of, in juice drink flavor emulsions, 326
a-pinene in, 74 in juice drinks, 323, 324
a-terpineol oxidative by-product, 264 uv absorption of, 98, 99
addition of, to juice products, 80-82 winterizing of, 85
air pollution from d-limonene burning in yield change of, with maturity of fruit, 86
feedmill, 354 Oriental fruit fly, 246
aldehyde determination in, 95-97 Oxidation, 97, 260-272
blending of, levels in juices, 79-83, 427, acid contributions to, 264
428, 436, 437 amino acid contribution to, 261
centrifuge use in production of, 86 cyclotene formation due to, 262
Clevenger method determination of, 74, 75 detection of, 265
cold pressed, production, 85-87 2,5-dimethyl-4-hydroxy-3(2H)-furanone
composition of, 74 formation due to, 264, 270
computer applications, 83, 84, 427, 428, N-ethylpyrrole-2-carboxaldehyde formation
436,437 due to, 264
d-limonene effects in aerobic wastewater l-ethyl-2-formylpyrrole formation due to,
treatment, 359, 365 261,270
d-limonene effects on aerobic wastewater formol test, 267
treatment, 360, 366 furaneol formation due to, 263
d-limonene in, 74 furfural, 267, 270
464 INDEX
Citrus limetta Risso (Lemon-like), 201 tangerine juice addition and, 195, 281
Citrus limettoides Tan. (Sweet Lime), 203, varietal blending and, 291
204 vitamin C addition and, 188
Citrus limon (L.) Burm. f. (Lemon), 181, Statistics, 371-386
201 actual range, 374
Citrus maxima (Burm.) (Pummelo), 181, analysis of variance, 378-383
183 average deviation, 374
Citrus nobilis Loureiro (King Tangerine), control charts, 376, 377
181,200 error analysis, 374-376
Citrus paradisi Macfadyen (Grapefruit), F-distribution, 380
181, 192 least squares analysis, 383-386
Citrus reticulata Blanco (Common pulp level control applications, 104
Tangerine), 181, 195, 196 regression, 383-386
Citrus sinensis (L.) Osbeck (Sweet Orange), root mean square deviation. 374, 375
181-185, 187, 189, 190 runs, 376, 377
Citrus sudachi Hort. Shirai (Tangerine significant figures, 371-373
Hybrid), 201 standard deviation, 375
Citrus unshiu Marc. (Satsuma Tangerine), Steven's cloud test, 120
181 Sulfites
Specifications, 3, 234 as antioxidants, 272, 325, 326
black flake, 260 as preservative, 325, 326
Brix,28 in juice drinks. 325, 326
color, 130 Sucrose (see also Brix)
FCC, for oils, 89-91 adulteration detection of, 285
for pectin, 305 as Brix, 8
inbound and outbound, 389 as soluble solids. 7
microbial, 235 in nutritional labeling, 164-166
oil, 79 potassium citrate solubility in, solutions,
pulp, 107, 109 273
use of, to prevent adulteration, 297, 298 refractometer calibration using, solutions.
with control charts, 376 15
USP, for oils, 89-91 regression analysis of, tables, 21. 383
SPG (pounds soluble solids per gallon, see slime molds metabolized from. 228
Brix) structure of. 8
Standard deviation, 375, 376 viscosity contributions of, 177
Standards of identity, 280-284, 293. 389, 397 use of, density tables, 21. 47.154,383
antioxidants and. 272 in juice drinks. 322
calcium addition and, 172 solutions to increase limonin solubility,
cleaning agents and, 256 141
dried juice sacs and, 110 yeast fermentation of, 230, 231
food additives and, 256 Supercritical extraction, 2, 157
fruit sections, 319 Sweet orange (Citrus sinensis L. Osbeck).
grapefruit juices, 282 181, 182
imported juice, 220
jellies, 314, 315
labeling requirements, 400 Tangeretin, unique component in tangerine
lemon juices, 47, 282, 283, 400 oils, 74
oils and, 73, 281 Tangerines or mandarins, 181, 194-197
orange juices, 52, 53, 280-282 acid determination of, juice, 46
pulp and, 281 blending of, juice in orange juices, 126,
pulpwash and, 105, 114 281,291
468 INDEX