See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/27474567

Some Aspects of Calcium Phosphate Chemistry in Sugarcane Clarification

Article  in  International Sugar Journal · July 2009

Source: OAI

CITATIONS READS

11 908

2 authors:

William Doherty Darryn Rackemann

Queensland University of Technology Queensland University of Technology
185 PUBLICATIONS   4,786 CITATIONS    51 PUBLICATIONS   1,022 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Specifications for renewable crude oils View project

High performance pyrolysis reactor for resource recovery from comingled municipal waste View project

All content following this page was uploaded by Darryn Rackemann on 10 June 2014.

The user has requested enhancement of the downloaded file.

QUT Digital Repository:
http://eprints.qut.edu.au/

This is the published version of this journal article. Published as:

Doherty, William O. S. and Rackemann, Darryn W. (2009) Some

aspects of calcium phosphate chemistry in sugarcane clarification.
International Sugar Journal, 111(1327). pp. 448-455. (In Press)

© Copyright 2008 William O.S. Doherty and Darryn W. Rackemann

Some aspects of calcium phosphate chemistry
in sugarcane clarification †

By W.O.S. Doherty * and D.W. Rackemann

Sugar Research and Innovation, Queensland University of Technology, Brisbane, Qld, Australia.
*
Contact author: Email: [email protected]

abstract

This paper reviews some aspects of calcium phosphate chemistry since phosphate in juice is an important
parameter in all sugar juice clarification systems. It uses basic concepts to try and explain the observed differences
in clarification performance obtained with various liming techniques. The paper also examines the current
colorimetric method used for the determination of phosphate in sugar juice. In this method, a phosphomolybdate
blue complex formed due to the addition of a dye is measured at 660 nm. Unfortunately, at this wavelength there is
interference of the colour arising from within the juice and results in the underestimation of the amount of soluble
inorganic phosphate content of juice. It is suggested that phosphate analysis be conducted at the higher
wavelength of 875 nm where the interference of the juice colour is minimised.

Keywords: absorbance, calcium phosphate, clarification chemistry, colorimetric method, liming

Introduction Rackemann (2008) that suggests that the current colorimetric
method used for the determination of soluble inorganic phosphate in
Whilst the basic chemistry of sugar juice clarification is understood, sugarcane juice is inadequate.
details of some of the interactions between various impurities
remain unknown. The main reason stems from the complexity of Calcium phosphate precipitation
the interactions between constituents of the juice and inconsistency
in juice composition. Juice composition depends on a variety of Sugarcane juice contains soluble inorganic phosphates, which, after
conditions such as the cane variety, age of the cane, geographical the addition of lime/lime saccharate to increase the pH to
location and growing conditions, as well as on the harvesting and prevent sucrose inversion, form insoluble calcium phosphate
milling processes. The work by Naidoo and Lionnet (2000) on South phases. Calcium phosphate particles are involved in the formation of
African sugarcane clearly highlights the effect of cane variety and flocs, and hence are pivotal in the clarification process.
agricultural factors (e.g. cane age, ratoon number, rainfed and In aqueous solution, the main reactions that occur to produce
irrigation) on colour, phosphate, silica, calcium, potassium and calcium phosphates are:
sulphate contents of juice. Cane variety and geographical location
were found to be statistically significant (5% level) on colour, Ca2 + + 2H2PO4- Ý Ca(H2PO4)2 [1]
gums and inorganic phosphate. Geographical location was also (monocalcium phosphate, MCP)
statistically significant (5% level) on calcium and potassium, while Ca2 + + HPO42 - Ý CaHPO4 [2]
cane variety was weakly statistically significant (10% level) with (dicalcium phosphate, DCP)
these elements.
The phosphate level in sugar juice is a key parameter in DCP can be present as dicalcium phosphate dihydrate (DCPD)
the success of the clarification process by defecation. Juices with or dicalcium phosphate anhydrate, (DCPA).
insufficient phosphate (<~300 mg/L as P2O5) generally clarify
poorly and produce hazy clarified juices (Deben, 1976; Steindl, 3Ca2 + + 2PO43- Ý Ca3(PO4)2 [3]
1998). Factory experience has shown that stale cane normally (tricalcium phosphate, TCP)

clarifies better when phosphate is added to juice prior to clar- 2CaHPO4 + 2Ca3(PO4)2 Ý Ca8H2(PO4)6 [4]
ification (Dymond, 1950). However, excessive phosphate in the (octacalcium phosphate, OCP)

Ý Ca5(PO4)3OH + 2H+
clarified juice impacts negatively on sugar filterability (Bennett
Ca3(PO4) 2 + 2Ca2 + + HPO42 - + H2O [5]
and Regnauth, 1960) and increases scaling of the evaporators. (hydroxyapatite, HAP)
Conversely, if too much lime is added or the mixed juice contains
high levels of residual calcium, the clarified juice becomes dark, In addition, tetracalcium phosphate [Ca4(PO4)2OH2] is formed.
increased scaling occurs in the evaporators and pans, ash in
sugar increases, and molasses exhaustion is less efficient. Because some of these reactions are sequential, i.e. some
In many situations where the calcium level in mixed juice is reaction products participate as reactants in further reactions, the
adequate and the pH is appropriate for maximum calcium distribution of calcium phosphate products depend on the kinetics
phosphate precipitation, the variable that requires adjustment is of each reaction. Thermodynamically, HAP is the most stable form
the phosphate content of juice.
Steindl (1998) found from a Figure 1. Calcium concentrations and pH values of solutions saturated with various
phosphate (as P2O5) survey calcium phosphate phases (Nancollas et al., 1979)
conducted in a factory over a
24-hour period that the level in
mixed juice varied between 64
mg/L and 414 mg/L P2O5. The
weighted average was 197 mg/L
P2O5, implying that extra
Total calcium (moles/L)

phosphate should be added to

mixed juice for optimum clarifi-
cation performance. However,
it is known that certain juice
types may contain far in excess
of the normal requirement of
inorganic phosphate and yet
clarify poorly. The work reported
here provides information to
the process chemist on
some aspects of phosphate
chemistry in juice clarification.
This paper also provides some
preliminary work by Doherty and
of the calcium phosphate phases (Kibabezyc et al., 1988), however, chemical parameters, including the molecular weight of the polymer,
it has been found experimentally that the calcium to phosphate ratio the charge density of the polymer, the concentration of the
is 1.45, significantly lower than the expected ratio of 1.67 for HAP polymer, the size and charge of the juice particles, the juice pH, and
(Nancollas et al., 1979). This is because of differences in the juice temperature. In addition, the speed of flocculation is directly
stoichiometric ratios between calcium and phophorus lattice ions in proportional to the velocity gradient in the solution and it is known
the HAP structure. that the type, speed and duration of agitation have a marked effect
Figure 1 shows the calcium ion concentration versus pH of an on floc structure (Peng and Williams, 1993).
aqueous solution saturated with calcium phosphate. At pH ~7.8
(the normal pH of limed cane juice in Australian sugar mills), HAP is Effect of liming technique on calcium phosphate flocs
the most stable followed by TCP. However in limed factory juices
amorphous TCP is the predominant phase formed in the clarifier. As reported by Doherty et al. (2002), and illustrated in table 1, lower
The effectiveness of the juice clarification process would be consid- turbidity is achieved with lime saccharate rather than with milk of
erably increased if a larger proportion of HAP was formed instead of lime. The mean value of turbidity (per 100 brix) for liming with
TCP since HAP would remove a significant proportion of proteins saccaharate is 27.9 and 62.3 for liming with milk of lime, and from
and consequently reduce the formation of melanoidins. The the (two tailed) t-test on paired samples the difference is highly
challenge is for researchers to develop conditions during clarification significant although greater variability was observed for the milk of
conducive for the formation of HAP. lime results. The mean value for phosphate (as P2O5 mg/L on brix)
with saccharate is 16.1 and 35.6 with liming with milk of lime, and
Flocculation from the (two tailed) t-test on paired samples the difference is
highly significant. This implies that lime saccharate reduces the
Particles suspended in solution are subject to a number of different amount of suspended solids in clarified juice and also reduces the
forces. These include: phosphate content. Why?
The process chemist is aware that the ratio of syrup to lime in
• Gravity, the force which separates particles according to density, lime saccharate has a strong bearing on clarified juice turbidity - as
with the densest suspended solids moving towards the bottom of turbidity decreases with increase in syrup to lime ratio. The question
the vessel in which they are suspended. is, how can we account for these occurrences? Why does syrup
• Van der Waals forces are short range attractive forces between added to lime to form saccharate influence the nature of calcium
suspended particles. These forces tend to draw suspended particles phosphate, the material primarily responsible for the removal of
towards each other. suspended solids?
• Brownian forces, dependent on temperature, impart motion to These observed differences can be explained by considering the
colloidal particles through collision of particles and molecules of nature of the calcium phosphate formed and the physico-chemical
the suspending medium. Electrostatic repulsion between charged interactions that occur during clarification. When lime/milk of lime is
particles tends to keep the particles apart, though shearing has an mixed with syrup, the calcium in the lime interacts with sucrose to
opposite effect. form mono-, di- and tri-calcium saccharate. The predominant phase
is dependent on the reaction conditions. Of the three, mono-calcium
Coagulation of the suspended particles occurs when the forces saccharate is the most soluble form hence providing a higher
of attraction exceed the forces of repulsion. However, the major proportion of calcium ions readily available for interaction with
physical reaction that occurs during clarification is the enhanced inorganic phosphate ions. In order to have the mono-calcium
flocculation of micro-particles by the addition of an anionic copoly- saccharate as the predominant phase, a sucrose to calcium molar
mer flocculant. There are three mechanistic pathways possible for ratio of 7:1 is recommended. So, any ratio less than this would have
the binding of such a polymer to a particle. These are (a) depletion a mixture of saccharates with varying degrees of available calcium
flocculation; (b) charge annihilation/charge patch flocculation; and ions. As lime saccharate provides more available calcium ions than
(c) bridging flocculation. In bridging flocculation, the high molecular milk of lime due to enhanced solubility, more calcium phosphate
weight polymer acts as long threads connecting particles together. would be precipitated, resulting in reduced amounts of phosphate in
The bridges may form through electrostatic attraction, hydrogen clarified juice.
bonding or van der Waals forces. In sugarcane juice, anionic The suspended particles in juice generally carry a negative
acrylamide-sodium acrylate copolymer is used to achieve floccula- charge. For coagulation of the particles to take place the overall
tion (Fellows and Doherty, 2006). While it should be expected that charge on the particles has to be neutralised to near zero charge.
electrostatic repulsion would prevent flocculation due to the Calcium phosphate forms when milk of lime/lime saccharate is
close proximity of negatively charged juice particles, it has been added to the juice. The particles are positively charged and form
established that cationic bridging through calcium ions bonding with flocs which entrap suspended solids. When lime saccaharate (with
the ionic acrylate group occurs (Crees et al., 1991). Large floc high syrup to lime ratios) is added to juice, spontaneous precipita-
structures are formed by this process. tion of fine calcium phosphate particles occurs because of the
Muhle and Domacsch (1990) noted that the strength of flocs readily available calcium ions from the highly soluble mono-calcium
formed by bridging flocculation depends on the number of bridges saccharate. With milk of lime or with lime saccharate mixtures with
formed between mud particles, the free segment/surface adsorption low syrup to lime ratios, the proportion of readily available calcium
energy and the fraction of polymer segments in the adsorbed state. ions is relatively small, resulting in the formation of bigger calcium
Therefore, the efficiency of bridging is a function of several physico- phosphate particles. The calcium phosphate particles formed with
Table 1. Results for clarified juices (Doherty et al., 2002)
Test No. Type of lime solution Turbidity @900 nm Turbidity per 100 brix Phosphate (as P2O5),
mg/L on brix
1 Intermediate* liming with saccharate 6.3 33.0 31.9
1 Intermediate liming with milk of lime 20.8 112.1 69.0
1 Hot** liming with saccharate 5.5 28.4 7.2
1 Hot liming with milk of lime 5.1 27.1 4.3
2 Intermediate liming saccharate 7.3 39.6 10.9
2 Intermediate liming with milk of lime 11.4 60.4 28.6
2 Hot liming with saccharate 4.4 22.0 2.5
2 Hot liming with milk of lime 4.0 20.0 4.0
3 Intermediate liming with milk of lime 11.2 59.8 50.7
3 Intermediate liming with saccharate 4.9 26.1 17.1
3 Hot liming with milk of lime 14.3 71.3 24.9
3 Hot liming with saccharate 5.5 26.6 9.2
4 Hot liming with milk of lime 15.6 85.6 67.5
4 Hot liming with saccharate 3.5 19.3 34.2
* Primary juice (at 76°C), lime saccharate/milk of lime to pH = 7.8, heat to 100°C flocculate and clarify in settling tube.
**Primary juice (at 76°C), heat to 100°C, add lime saccharate/milk of lime to pH = 7.8, flocculate and clarify in settling tube.

Figure 2. Phosphate and turbidity trends in clarified juice effects is that the calcium
phosphate particles within the
floc aggregate formed with lime
ESJ turbidity (%), ESJ phosphate level (mg/L P2O5)

ESJ turbidity Phosphoric acid saccharate are relatively small in

Phosphate level pump turned on
size and are more likely to form
Phosphoric acid
Pump operation a loose floc structure due to
electrostatic repulsion forces.

Turbidity versus clarified

juice phosphate

As has been stated and is

generally accepted, clarified
juice with low turbidity is
achieved when sufficient phos-
phate is present in the mixed
juice. Some sugar mills link
their turbidity output to a phos-
phate pump so that phosphoric
acid is added to mixed juice
when the turbidity outputs
reach unacceptably high levels.
lime saccharate would therefore have higher charge to mass ratio Recently, studies were conducted in a mill, where an on-line
than the particles obtained with milk of lime (or lime saccharate mix- phosphate device was used to monitor the level of phosphate in
tures with low syrup to lime ratios). As such, they are more effective clarified juice, so that comparisons could be made with turbidity
in neutralising the negatively charged suspended particles in juice. outputs. Figure 2 presents the turbidity and phosphate results. There
This would invariably increase the rate of collision between the small is no correlation between turbidity and the phosphate content of
sized juice particles, thereby forming larger floc aggregates that can the clarified juice. For this set of results, the addition of phosphoric
more readily trap more particles. This will then lead to reduced juice acid did not reduce the turbidity of the clarified juice. In fact, its
turbidity. addition increased the level of phosphate in clarified juice which
Studies on calcium phosphate flocs indicated that liming with impacts on filterability and scale formation in the evaporators. In
saccharate gave larger and looser flocs than liming with milk of lime this instance, it is more than likely that the mixed juice contained
(Doherty et al., 2002), and therefore is expected to be better for sufficient inorganic phosphate to achieve satisfactory clarification.
removing impurities through physical entrapment and adsorption. In So, it would be beneficial for the process chemist to have at his
addition, the reduced settling rate obtained with saccharate aids the disposal an on-line device to monitor phosphate level (as well as
‘sweeping’ of suspended solids. A probable explanation for these calcium) in mixed juices.
In situations where there is excessive phosphate in juice but poor The Australian sugar industry uses the method for the
clarification, the option open to the process chemist is either to determination of total (i.e. acid-soluble inorganic phosphate) and
increase the liming pH or to add, if available, soluble calcium salts soluble (i.e. water-soluble inorganic phosphate) phosphate (BSES,
i.e. calcium chloride. 2001). If the amount of soluble inorganic phosphate is being
determined, no pH adjustment of the juice prior to analysis is
Phosphomolybdate method for the determination of required. However, if the total amount of phosphate is being
inorganic phosphate in juice determined, the pH of the juice is adjusted to 4.0 by adding
sulphuric acid. In the current method, the juice is then filtered
Taylor and Miller (1914) introduced a method for the colorimetric under vacuum with acid-washed Supercel. Known amounts of
determination of inorganic phosphate. This method has been acid, molybdate, amidol (2, 4-diaminophenol dihydrochloride) and
modified over the years and is the primary method used for the sodium metabisulphite are added to the filtered solution and the
determination of phosphate in aqueous solutions. absorbance is measured at 660 nm after about 10 min to allow full
The principle of the method involves the condensation of ortho- development of the blue-coloured complex.
phosphate and molybdic acid to form phosphomolydic acid, which Berenblum and Chain (1938) reported on the class of sub-
on reduction produces a blue colour complex. The intensity of the stances which interfere with the colorimetric determination of
blue colour is proportional to the amount of phosphate in a sample. phosphates. These include:

Figure 3.The absorbance of the phosphomolybdate colour complex in • Compounds which alter the acidity of
clarified juice the solution. These are acids, alkalis and
buffering salts.
• Compounds which form molybdenum
complexes which are difficult to reduce and
therefore reduce the amount of molybdic acid
available for reduction. These compounds
include oxalates, fluorides, and citrates.
Absorbance, mAU

• Compounds which reduce the concentra-

tion of the reducing agent e.g. nitrites.

It is reported in the modified method used

by the Australian sugar industry that the
presence of low concentrations of organo-
phosphorus compounds and cations and
anions found in cane juice do not interfere to
a significant extent with the phosphate deter-
mination (BSES, 2001).
As the determination of phosphate is a
colorimetric method, it would suggest that
the colour of the cane juice could add to
the error of the absolute phosphate value
Figure 4. Absorption curve of untreated clarified sugar juice
obtained. As the colour of the cane juice
varies with cane variety and location, it
was necessary to re-examine whether the
measurement of the blue complex at 660 nm
was the most suitable wavelength for the
determination of phosphate in cane juice,
particularly with clarified juices.
Absorption, mAU

Absorbance of colour in juice

A UV scan of a clarified juice solution

treated for phosphate determination (i.e.
colour reagents added to juice) was
carried out over the range 400 to 900 nm. The
absorbance of the molybdate blue complex is
shown in figure 3. The absorption reaches a
maximum at around 725 nm (not at 660 nm)
and thereafter plateaus. Similar results were
obtained with a standard phosphate solution.
Table 2. Variation of soluble phosphate concentrations Acknowledgements
(as P2O5, mg/L) of cane juices
The financial support of the Sugar Research Development
Mixed juice Clarified juice
Corporation and a syndicate of Australian mills are acknowledged.
Factory 660 nm 875 nm 660 nm 875 nm The authors wish to thank James Whyte of MEP instruments for
A 233 264 4 5 evaluating both sensors.
B 129 152 26 37
†
This paper was presented at the 2008 Australian Society of Sugar
C 538 593 n.d. n.d.
Cane Technologists annual conference and is published here with
D 204 222 21 27
the agreement of the Society.
F 135 156 22 29
G 189 207 8 11 References
H 430 485 28 34
I 170 191 16 19 Bennett, M.C. and Regnauth, J.M. (1960) The effects of calcium and
J 491 528 21 26 phosphate in cane juice clarification. International Sugar Journal 62: 13-16.

K 141 150 13 15 Berenblum, I. and Chain, E. (1938) Studies on the colorimetric deter-
mination of phosphate. Biochemical Journal 32: 286-294.
The absorption curve of untreated clarified juice (figure 4) shows Bureau of Sugar Experimental Stations (BSES) (2001) Laboratory man-
that absorption decreases with wavelength. Comparing figures 3 ual for Australian sugar mills, vol. 2 - Analytical Methods and Tables.
and 4 shows that interference of colour is more significant at 660 nm Crees, O.L., Senogles, E. and Whayman, E. (1991) The flocculation of
than at 800-900 nm. Based on this, it was decided to conduct tests cane sugar muds with polyacrylamide-sodium acrylate copolymers.
using sensors with 660 nm and 875 nm wavelengths. Journal of Applied Polymer Science 42: 837-844.
Deben, J.G. (1976) The chemistry of calcium phosphate precipitation
Analysis of phosphate in mixed and clarified juices in cane juice clarification. International Sugar Journal 78: 35-39, 73-75.
Doherty, W.O.S., Greenwood, J., Pilaski, D. and Wright, P.G. (2002) The
Samples of mixed juices and clarified juices were collected from effect of liming conditions in juice clarification. Proceedings of the
a number of Australian sugar mills and examined with the two Australian Society of Sugar Cane Technologists 24: 443-451.
sensors. The results are shown in table 2 and indicate differences in Doherty, W.O.S. and Rackemann, D.W. (2008) Stability of sugar cane
the phosphate level between the two sensors. The differences in the juice - A preliminary assessment of the colorimetric method used for phos-
measured phosphate levels varied between 8% and 15% for mixed phate analysis. Zuckerindustrie (accepted).
juice samples and between 10% and 30% for clarified juice Dymond, G.C. (1950) A progress report on some clarification problems
samples. While the tests were not exhaustive, these results suggest of some Natal cane juices. South African Sugar Journal 34: 367-381.
that the current BSES method for phosphate determination using a Fellows, C.M. and Doherty, W.O.S. (2006) Insights into bridging floccu-
660 nm wavelength underestimates the phosphate level in cane lation. Macromolecules Symposium 23: 1-10.
juice. Using the current BSES method for phosphate determination, Kibabezyc, C.W., Zielenkeiwicz, A. and Zielenkeiwicz, W. (1988)
results would become increasingly important if measurements are Colorimetric investigation of calcium phosphate precipitation in relation to
used for factory processing control, which is not the case at present solution composition and temperature. Thermochimica Acta 131: 47-55.
in most Australian factories. Muhle, K. and Domasch, K. (1990) Floc Strength in Bridging
The results of table 2 also show a wide variation in phosphate Flocculation. In: ‘Chemical Water and Wastewater Treatment’, Proceedings
levels in cane juice over the various milling districts. A number of of the 4th Gothenburg Symposium, (Ed. H.H. Klute), Springer Verlag,
sugar factories clearly had phosphate levels far below the Berlin: 105-115.
recommended P2O5 value of 300 mg/L. Naidoo. L. and Lionnet. G.R.E. (2000) The effect of cane variety and
other agricultural factors in juice composition. Proceedings of the South
Conclusion African Sugar Technologists Association 74: 19-24.
Nancollas, G.H., Amjad, Z. and Koutsoukos, P. (1979) Calcium phos-
The success of a clarification process depends on a fundamental phates-speciation, solubility, and kinetic considerations, chemical model-
understanding of the influential role played by soluble inorganic ing in aqueous solutions. American Chemical Society, Advanced
phosphate. Although the role of soluble inorganic phosphate is very Chemicals Symposium 23: 475-497.
important in the understanding of the mechanisms of clarification, Peng, S.J. and Williams, R.A. (1993) Control and optimization of min-
the disregard of the role of the physico-chemical nature of juice, the eral flocculation and transport processes using on-line particle size analy-
chemical and physical reactions during clarification (including the sis. Minerals Engineering 6: 133-153.
reactions of organic salts and proteins, degradation of sucrose and Steindl, R.J. (1998) Dirt - its implications for the clarifier and filter sta-
reducing and polymerisation reactions), and the role of calcium tions. Proceedings of the Australian Society of Sugar Cane Technologists
would result in the process chemist being unable to maintain and 20: 484-490.
improve productivity and produce sugar with increased compliance Taylor, A.E. and Miller, C. (1914) On the estimation of phosphorus in
with premium grade criteria. The use of turbidity alone to monitor biological materials. Journal of Biological Chemistry 18: 215-220.
juice quality completely disregards the role of soluble impurities in
sugar manufacture.

View publication stats