1292 Ind. Eng. Chem. Res.

2009, 48, 1292–1297

Pyrolysis of Broiler Manure: Char and Product Gas Characterization

Isabel M. Lima,*,‡ Akwasi A. Boateng,§ and K. Thomas Klasson‡
USDA ARS Southern Regional Research Center, PO Box 19687, New Orleans, Louisiana 70179, and USDA
ARS Eastern Regional Research Center, 600 East Mermaid Lane, Wyndmoor, PennsylVania 19038

Value-added materials produced from broiler manure can be a renewable alternative to its problematic disposal.
Pyrolysis of broiler manure (litter and cake) produces char, bio-oil, and biogas. In this study broiler manure
chars were characterized for their ability to adsorb selected metal ions. Char yields for pelletized broiler
manure slowly pyrolyzed at 700 and 800 °C for 1 h ranged between 29.6% and 33.5%. Surface areas for the
broiler litter and cake decreased with pyrolysis temperature with 238 and 318 m2/g and 199 and 261 m2/g for
700 and 800 °C, respectively. Broiler manure chars removed significantly more metal ions from solution (up
to 0.91 mmol/g Cu2+ and 0.96 mmol/g Zn2+) when compared to chars produced from coal, wood, or coconut
shells (up to 0.0 mmol/g Cu2+ and 0.10 mmol/g Zn2+). Composition of the biogas generated during fast
pyrolysis between 700-900 °C, as a measurement of its energy potential, was also determined. The major
components of the noncondensable gas produced during fast pyrolysis of broiler manure, included CO, CO2,
low molecular weight hydrocarbon gases, and H2. Except for CO, noncondensable gases increased with pyrolysis
temperature.

Introduction waste generally has lower elemental carbon content and may
Pyrolytic products or chars are low porosity, low surface area produce lower char yields than plant material when both sources
materials that are intermediate products in the development of are pyrolyzed under the same time and temperature conditions.
activated carbons. They are not commonly used in industry for Cattle manure has been used1 to produce a pyrolysis product
purification and chemical recovery operations or environmental as a soil amender due to its high content in phosphate ion,
remediation, as are the more structurally developed activated organic nitrate, and potassium ion. The pyrolysis product,
carbons. Toxic metals contamination of various water sources produced by heating cattle manure to 380 °C in an atmosphere
is a significant problem in many parts of the United States. of limited air, had a very low surface area of 2.2 m2/g, an ash
Chars, which can be produced from a number of precursor content of 25.6%, and an elemental carbon content of 49.2%.
materials including coal, biomass, and municipal wastes, have The production of pyrolyzed cattle manure was also described
not been examined for remediation of this problem. There are over the temperature range of 250-800 °C in closed containers.2
few if any commercial outlets for chars because of their inferior Surface areas were also low (<20 m2/g) over this temperature
adsorption characteristics compared to activated carbon. While range. The pyrolyzed cattle manure had high ash content (up
carbon production is an expanding industry in the United States, to 60%), a high pH (>10), and a total carbon content of less
with a present production rate of over 400 million pounds a than 40%. With a subsequent chemical activation step using
year and a growth rate of over 3% annually, char production is zinc chloride, BET values surpassed 2000 m2/g for chemically
simply an intermediate step in carbon manufacture and has no activated chars from cattle-manure.3 Alternative to chemical
significant presence in the marketplace. Besides a char product, activation, steam activation yielded low surface area (60.5 m2/
pyrolysis also delivers biosyn-gas and bio-oil which can have g) for a carbon product produced from steam-activated poultry
an associated energy potential. droppings in a fluidized bed reactor.4 It was concluded that
The production of pyrolytic products, in the form of chars, poultry droppings will yield only a very low grade carbon
is an age-old art. Char, when produced in the absence of air by mainly because of its low surface area and high ash content.
nonoxidative carbonization or pyrolysis, is a relatively inactive More recently, Koutcheiko et al.5 pyrolyzed chicken manure to
material possessing a surface area limited to several square produce a char with BET surface area of 4 m2/g which
meters per gram. Chars are normally produced to reduce the augmented by subsequent separate consecutive activations, using
volume and mass of a particular feedstock and provide a soil NaOH and CO2. The resulting product was a demineralized
amender that improves the physical and nutritive properties activated carbon with BET of 877 m2/g and a mineral content
through its ash content of hard, compact soils with a high clay of 10% wt. Sewage sludge was used by Jindarom et al.6 to
content or highly porous soils with a high silica or sand content. produce a char under either N2 or CO2 atmospheres as a potential
Because chars have poorly developed internal structure, they adsorbent for dye removal. Maximum surface area of mainly
adsorb few if any metal ions or organic compounds and nonpolar mesopores was 60.7 m2/g, and dye adsorption was found to
materials from either liquid or gaseous media. increase with pyrolysis temperature.
Very few studies have evaluated animal waste as source of While various methodologies exist for the creation of
pyrolytic products or their respective activated carbons. Animal activated carbons from plant or plant-derived material that are
* To whom correspondence should be addressed. E-mail: effective in the adsorption of metal ions, there remains a need
[email protected]. Tel.: 504/286-4515. Fax: 504 /286-4419. to create lower cost pyrolytic products, such as chars, from
Address: Isabel M. Lima and Thomas K. Klasson Commodity Utiliza- alternate sources of carbonaceous material that are in great
tion Research, USDA-ARS Southern Regional Research Center, 1100
Robert E. Lee Boulevard, P.O. Box 19687, New Orleans, LA 70179.
abundance and that have enhanced adsorption properties toward
‡
USDA ARS Southern Regional Research Center. metal ions. Additionally, thermo-chemical conversion of bio-
§
USDA ARS Eastern Regional Research Center. mass is becoming increasingly popular as an alternative means
10.1021/ie800989s This article not subject to U.S. Copyright. Published 2009 by the American Chemical Society
Published on Web 12/19/2008
 Ind. Eng. Chem. Res., Vol. 48, No. 3, 2009 1293
for renewable energy production and some studies in the where qe ) amount of solute adsorbed per unit weight of
literature report the thermochemical conversion of animal adsorbent in mmol/g; Ce ) equilibrium concentration of solute
manures and sewage sludges in particular.7-10 The thermo- in mM; Q0 ) monolayer adsorption capacity of adsorbent for
chemical conversion of biomass for bio-oil and/or biogas solute in mmol/g, and b ) constant related to the free energy
production generates a char material that should be characterized of adsorption in L/mM. The nonlinear least-squares regression
for its potential use as adsorbent. With these conversion method of Marquardt was implemented using Sigma Plot v.
technologies for energy production from bio-oil and/or biogas, 7.101 for Windows 2000 (SPSS Inc. Chicago, IL). A correlation
chars are also generated for which their use as an adsorbent coefficient (r2) and a probability value (p-value) representing
needs to be investigated. Whether chars are produced as the the goodness of fit of the Langmuir model to the data were
primary choice material from pyrolysis or considered a byprod- obtained.
uct, thermochemical conversion as a whole benefits from the Metals Adsorption. Chars were analyzed for metal ion
exploration of all generated streams and their respective potential uptake using 20 mM individual metal solutions of cadmium
use. The objective of this study was to produce value-added chloride (CdCl2), copper chloride (CuCl2), nickel nitrate [Ni-
chars from poultry manure with excellent metal ion adsorptive (NO3)2], or zinc chloride (ZnCl2), which were made up in an
properties. Additionally, the biogas resulting from the pyrolysis 0.07 M sodium acetate-0.03 M acetic acid buffer (pH 4.8).
process was further characterized for its potential as a fuel Multiple metal solutions included all four metal ions at 5 mM
source. each. Adsorption assays and metal concentration were deter-
mined as in the method by Lima and Marshall.11
Materials and Methods Determination of Biogas Composition. Biogas produced
Raw Material and Sample Preparation. Broiler cake and from the same two broiler manure samples was determined by
litter samples were obtained from the USDA-ARS, Poultry fast pyrolysis. A CDS analytical (Oxford, PA) pyroprobe was
Research Unit, Starkville, MS. Broiler cake results from partial used for the fast pyrolysis. It consisted of a 1-cm quartz tube
cleaning of the top layer in a poultry house and litter results heated by a platinum filament of 2-3 mm diameter, which is
from a complete cleaning of the facility. Cake consists of bird capable of maintaining up to 1200 °C temperature at a nominal
droppings, feathers, and small amounts (<5%) of bedding heating rate of 20 °C/ms. Pulverized samples were sifted and
material while litter contains larger amounts of bedding material particle sizes with 90% passing a 500 µm screen were used.
(20-30%), usually made of wood shavings, typically soft wood, The average weight charged into the pyrolyzer (PY) was about
such as pine. Samples were milled in a Retsch cross-beater mill 1 mg (0.96-1.12) and occupied about 1-1.5 mm in height in
(Glen Mills, Clifton, NJ) to a particle size of less than US 20 the quartz tube holder over packed quartz wool. Helium, the
mesh (<1 mm). Ground samples were then pelletized in a carrier gas for the GC, was also used to purge air in the sample
PMCL5 Laboratory pellet mill (California Pellet Mill, Merri- prior to pyrolysis and the pyrolysis gas yield to the GC.
mack, NH) equipped with a 3/16 in. die plate. Sample moisture Although the nominal heating rate is about 20 °C ms-1, the
content was monitored with a Sartorius moisture analyzer model sample heating rate can be much lower and typically esti-
MA 51 (Sartorius, Brentwood, NJ). The pellets produced were mated12 at 300 °C s-1. The experimental sample preparation
cylinders of 3/16 in. diameter and 3/16 in. length. Pelletized procedure is reported in Boateng et al.13 and is consistent with
poultry cake and litter samples were placed in a ceramic other reports.12 Using samples weighing less than 2 mg does
evaporating dish and placed in a Lindberg bench furnace not significantly change gas yield.12 The pyrolyzer was inter-
equipped with a retort (Lindberg/Blue M, Waterton, WI). Pellets faced to a gas chromatograph (SRI, CA). The pyroprobe/gas
were pyrolyzed (under slow pyrolysis conditions) at either 700 chromatograph (PY-GC) system allowed for a variety of
or 800 °C for one hour under a flow of nitrogen gas set at a compounds formed during flash pyrolysis to be characterized.
flow rate of 0.1 m3/h. Chars were allowed to cool to room The noncondensable gas products from pyrolysis were separated
temperature overnight in the retort. Samples were washed with using a Shincarbon ST 80/100, 2 m × 2.0 mm packed column
0.1 M HCl and subsequently given three separate water (Restek, Bellefonte, PA). The GC was programmed to maintain
washings before being dried overnight at 80 °C. 45 °C for 3 min after injection, followed by a 10 C/min ramp
Physical and Chemical Properties. The charcoals resulting to 250 °C, followed by holding at 250 °C for 10 min, yielding
from slow pyrolysis were evaluated for select physical and a total time of 34.4 min. Hydrogen was detected using the TCD
chemical properties that include char yield, surface area, SRI Instruments Wheatstone bridge with four filaments. The
apparent (bulk) density, attrition (hardness), and pH. Surface yields of the major noncondensable gas products from primary
area measurements were obtained from nitrogen adsorption and secondary pyrolysis reactions were quantified by calibration
isotherms at 77 K using a Nova 2000 surface area analyzer with a standard gas mixture consisting of CO, CO2, H2, CH4,
(Quantachrome Corp., Boynton Beach, FL). Specific surface C2H4, C2H6, C3H8, and C4H10 in helium (custom-mixed by Scott
areas (SBET) were taken from adsorption isotherms using the Specialty Gases, Plumsteadville, PA). Gas yields were quantified
Brunauer, Emmett, and Teller (BET) equation. Bulk density and on the basis of a linear relationship between the mass and area
attrition were determined as in the methods described in Lima counts of the programs. The coefficient of determination, R2,
and Marshall.11 A Thermo Orion pH meter was used to measure for the linear fits ranged between 70% (CO) to 97% (CH4). Char
pH, where 0.5 g of sample was placed in 50 mL of deionized yield was determined gravimetrically. Tests were run in
water, covered with parafilm, and allowed to equilibrate by triplicate, and average values are reported, for three pyrolysis
stirring at 300 rpm for 72 h. temperatures: 700, 800, and 900 °C, for 20 s retention time. At
Adsorption Isotherms. Adsorption isotherms were run for this time there was no further change and therefore the
copper ion per the method described in Lima and Marshall.11 devolatilization reaction was complete.
Adsorption capacities (Q0) and affinity constants (b) were
calculated by fitting the isotherm data to the nonlinear Langmuir Results and Discussion
adsorption model:
1. Physical Properties of Char. Pyrolysis is a series of
qe ) Q0bCe ⁄ (1 + bCe) (1) cleavage reactions and polymerization reactions that make a
1294 Ind. Eng. Chem. Res., Vol. 48, No. 3, 2009

Table 1. Select Physical and Chemical Properties at Two Different Pyrolysis Temperatures, PT, 700 and 800 °Ca
PT sample Y (%) wt loss (%) BET (m2/g) PM % BD (g/cm3) A (%) pH
700 °C broiler cake 40.3 24.5 318 ( 15 88.4 ( 0.4 0.54 ( 0.00 15.1 ( 0.2 8.6
broiler litter 40.7 17.7 238 ( 13 90.4 ( 6.6 0.60 ( 0.00 14.4 ( 2.0 8.1
coal 78.3 22.3 3.8 ( 2 0.42 ( 0.00 34.1 ( 3.0 4.2
coconut shell 27.6 0.0 35 ( 12 34.6 ( 2.8 0.61 ( 0.00 20.5 ( 2.9 6.6
wood 25.3 4.8 301 ( 24 86.3 ( 10.0 0.38 ( 0.01 23.4 ( 0.3 5.1

800 °C broiler cake 39.3 24.5 261 ( 47 93.3 ( 4.9 0.53 ( 0.00 7.6 ( 2.6 9.4
broiler litter 39.4 17.5 199 ( 33 84.9 ( 3.2 0.62 ( 0.01 7.7 ( 0.3 9.1
a
Physical and chemical properties of poultry manure-based chars and chars made from traditional sources (coal, coconut shell, wood): yield, Y;
percent loss from acid washing, wt loss; BET surface area; percent micropores, PM; bulk density, BD; and attrition, A; pH.
fixed carbon structure involving both carbonization and devola-
tilization processes. Carbonization is the stripping off of
noncarbon atoms, like H, O, N, and S, and halides to create a
purer carbon form. During devolatilization, the physical vapor-
izing of free carbon structures or weakly bonded carbon chains
(volatile matter) occurs. Broiler litter and cake chars were
analyzed for their physical and chemical properties, and data
are summarized in Table 1. Char yield decreased slightly with
pyrolysis temperature and was slightly lower for broiler cake
char than for broiler litter chars. For 700 °C pyrolysis temper-
ature, char yields for the broiler manure chars were much larger
than those found for the plant-based chars. Nonetheless, weight
losses from acid washing were larger for the broiler manure
chars than their plant counterparts (wood and coconut shells)
(Table 1). Broiler manure contains much higher amounts of
inorganic compounds than either wood or coconut shells, which
are removed during the acid wash step. Coal char yields were
Figure 1. Copper ion isotherm for broiler cake char fitted to the Langmuir
highest with significant amounts of inorganic material also model.
removed by the acid wash as demonstrated by the percentage
losses (Table 1). In previous studies Lima and Marshall14,15 Table 2. Langmuir Parameters for Copper Ion Adsorption:
reported large ash values present in poultry manure carbons as Adsorption Capacity Q0 and Goodness of Fit Parameters r2 and p,
for Broiler Litter and Cake Chars Pyrolyzed at 700°C
a characteristic of the starting material. For this reason, both
chars and carbons made from these types of precursors should sample Q0 b R2 p
be acid washed to partially remove excess levels of inorganic broiler cake char 0.87 8.25 0.87 0.0008
material that can potentially interfere with its use as an adsorbent broiler litter char 0.78 6.60 0.86 0.0009
material. Additionally, the presence of this inorganic fraction
is responsible for the high pH values observed for broiler manure properties related to the composition of the starting material,
char pH. Values of pH were higher for broiler cake than broiler which produces a more cohesive pellet and yield a char more
litter due to larger amounts of inorganic material in the former. resistant to friction than the harder plant material char. It is
Surface area (BET) values for both broiler litter and cake chars important to add that attrition of wood- and coconut shell-chars
as well as for the wood char were unexpectedly high since chars in their native form (ground to 18 × 40 mesh or 0.47 to 1.119
are typically low porosity materials. BET surface areas for the mm) without pelletization instead of ground to a powder and
chars produced in this study were greater than 199 m2/g, even subsequently pelletized) should be significantly reduced. Overall,
as high as 318 m2/g, and a largely microporous structure was comparing properties of the broiler manure char with those
observed where 85 to 90% of pores were in the micropore size found by Lima and Marshall16 for activated carbons of the same
range (Table 1). With the exception of wood chars which also material, it is observed that in the char, yield and bulk density
displayed much higher than expected BET values, the plant are higher and surface area and attrition are lower. During
based chars contrasted with the broiler manure chars in that pyrolysis or charring, carbon structure rearranges and condenses
BET surfaces areas observed were significantly lower, ranging to a higher density, producing a more stable amorphous graphitic
from 3.8 m2/g to 35 m2/g. Surface area was significantly higher carbon structure. Further activation allows for pore development
for broiler cake char than for broiler litter and it decreased with rendering the carbons less dense, more porous, and less resistant
pyrolysis temperature (Table 1). BET values for broiler cake to abrasion. Previous work17 has estimated production costs for
chars at 318 m2/g compare very favorably with the highest BET broiler manure carbons at $0.65/lb. This value compared
values reported for broiler litter-based activated carbon14 of 548 favorably well with other published studies in the literature on
m2/g (activated for 60 min under a water flow of 3 mL/min). the cost of manufacturing carbons from pecan shells, almond
Attrition was lower for the broiler manure chars than the shells, and sugar cane bagasse.17 A lower cost of production
reference chars and contrary to what would be expected, attrition for the broiler manure char is estimated since the activation step
decreased with pyrolysis temperature. It is believed that attrition is eliminated and higher yields are obtained.
values are directly influenced by both the starting material and 2. Adsorption Properties of Char. Adsorption capacity
the pelletization process. Commonly, attrition is lower for harder (Table 2) for copper ion (Q0) was calculated from the adsorption
materials and denser pellets. Coconut shells, wood, and coal isotherm, by fitting data to the nonlinear Langmuir adsorption
are both harder materials than broiler manure. Therefore the model (Figures 1 and 2). It can be seen from the regression
lower attrition values of the latter are likely due to other parameters (r2 and p-values) that the nonlinear regression model
 Ind. Eng. Chem. Res., Vol. 48, No. 3, 2009 1295
Table 4. Broiler Litter and Cake Char Elemental Composition in
Phosphorous (P), Calcium (Ca), Magnesium (Mg), Sulfur (S), and
Potassium (K) in mg/g Char
P Ca Mg S K
litter 36.8 ( 2.7 53.0 ( 4.0 14.2 ( 1.2 17.3 ( 2.3 42.0 ( 5.3
cake 49.2 ( 10.6 68.5 ( 15.5 22.2 ( 6.1 20.0 ( 1.7 36.5 ( 5.5

Table 5. Percent Metal Ion Adsorption Efficiencies of Broiler

Manure Chars and Chars Made from Traditional Sources (Coal,
Coconut Shell, Wood) at Two Different Pyrolysis Temperatures, PT,
700 and 800 °C, for Solutions Containing Only a Single Metal Ion at
5 mM Concentration
PT sample Cu2+ Cd2+ Ni2+ Zn2+
700 °C broiler cake 95.4 83.2 6.6 89.8
broiler litter 95.0 82.3 5.1 90.9
coal 0.0 12.8 0.5 2.6
coconut shell 3.1 13.5 0.0 0.5
Figure 2. Copper ion isotherm for broiler litter char fitted to the Langmuir wood 6.3 13.3 0.0 1.8
model.
800 °C broiler cake 97.9 89.4 11.2 92.5
Table 3. Metal Ion Adsorption of Chars Made from Poultry broiler litter 86.9 59.9 6.0 75.7
Manure and Chars Made from Traditional Sources (Coal, Coconut
Shell, Wood) in mmol/g Char, at Two Different Pyrolysis Table 6. Percent Metal Ion Adsorption Efficiencies of Broiler
Temperatures, PT, 700 and 800 °C Manure Chars and Chars Made from Traditional Sources (Coal,
Coconut Shell, Wood) at Two Different Pyrolysis Temperatures, PT,
PT sample Cu2+ Cd2+ Ni2+ Zn2+ 700 and 800 °C, for Solutions Containing Four Metal Ions at 5 mM
Each
700 °C broiler cake 0.91 0.64 0.10 0.96
broiler litter 0.58 0.45 0.25 0.72 PT sample Cu2+ Cd2+ Ni2+ Zn2+
coal 0.00 0.05 0.13 0.03
coconut shell 0.00 0.04 0.13 0.10 700 °C broiler cake 71.1 18.8 3.8 23.7
wood 0.00 0.03 0.20 0.05 broiler litter 66.1 18.1 3.6 25.2
coal 0.6 0.3 0.7 1.0
800 °C broiler cake 1.05 0.80 0.13 0.80 coconut shell 0.2 0.9 0.7 3.8
broiler litter 0.68 0.43 0.03 0.43 wood 4.0 0.0 0.0 2.4
800 °C broiler cake 69.1 8.1 7.0 18.1
fits the copper ion adsorption data well for both broiler litter broiler litter 39.6 4.9 10.8 11.0
and cake chars. Adsorption capacity toward copper ions was
higher for the cake char than the litter char. This difference is broiler cake char (potassium excluded). From Table 4 it can be
linked to the presence of larger amounts of wood in litter than seen that the phosphorus content of broiler litter and cake chars
cake samples and, as will be discussed later, the reduced ability ranges between 3.7% and 5% by weight. In contrast, chars
for metal ion uptake by wood-based chars. Copper ion saturation produced from traditional sources, such as coal, coconut shell,
levels for the broiler manure chars can be extracted from the and wood, generally have less than 0.2% phosphorus. Chars
adsorption capacity values. For the broiler litter and cake chars, were also analyzed for metal ion uptake using less than
saturation occurred at a concentration of or around 0.8-0.9 saturating concentrations of metal ions with 5 mM solutions
mmoles Cu2+ per gram of char (Figures 1 and 2). Albeit high rather than 20 mM solutions. Tables 5 and 6 report adsorption
concentrations of this or other metals are not commonly found values individually and in competition, respectively, as percent
in waste waters, capacity values give an indication of how much metal ion adsorbed. In competition mode, the chars were placed
could be adsorbed in a worst case scenario. The chars were in solutions containing each of the four metal ions at concentra-
further tested for their ability to adsorb three other metal ions tions of 5 mM per metal ion with a total metal ion concentration
at 20 mM, and adsorption efficiencies toward Cu2+, Cd2+, Ni2+ of 20 mM. A comparison across all metal ions demonstrates
and Zn2+ are reported in Table 3. Broiler litter and cake chars strong adsorption of copper and zinc by the broiler litter and
generally showed good metal ion adsorption at pH 4.8 with the cake chars and less adsorption for cadmium and zinc ions.
exception of nickel. Broiler cake adsorbed more metal ion than Generally, affinity toward metal ions was highest for copper,
litter and both adsorbed significantly more metal ion than any followed by zinc, and then cadmium ions. Chars performed
of the reference chars (Table 3). With the exception of zinc better when exposed to individual solutions of each metal ion
ions, adsorption, in general, improved when pyrolysis temper- than when exposed to all at the same time. In competition mode
ature was increased from 700 to 800 °C. During pyrolysis, planar copper ions appeared to still be the choice for removal and the
sheets of graphite plates form and either stack upon each other presence of other metal ions had little effect on its binding. This
or form “v-shapes”. Interbonding occurs by aliphatic dislocations same trend was also observed in previous studies by Lima and
(imperfection) of plates held together by weak atomic forces Marshall.16 There is no apparent reason why nickel ions bind
called London dispersion forces. Inorganic constituents, such much less readily than the remaining three metal ions in this
as phosphorus, contained in the manure are either physically study, as it is an ion of similar size and weight. Similar results
“trapped” or covalently (chemically) bound within the lattice were observed by Brown and Lester18 with activated sludge,
structure. The presence of covalently bound phosphorus, Ho et al.19 with sphagnum moss, and Lima and Marshall16 with
primarily in the form of phosphate ion, can create centers of poultry manure carbons. Reference chars adsorbed negligent
negative charge on the carbon. These centers of negative charge amounts to none when exposed to the four metals at once.
can readily adsorb or ionically bind positively charged ions such 3. Yields for Noncondensable Gas, Char, and Tar. In
as copper. Table 4 reports the composition of both chars in addition to char characterization, two other products of pyrolysis,
certain elements, with larger amounts of each one found in the bio-oil and noncondensable gas were also quantified for yield,
1296 Ind. Eng. Chem. Res., Vol. 48, No. 3, 2009

Figure 4. Percent composition of quantified noncondensable gas yields as

a function of pyrolysis temperature and for broiler cake and litter samples.
Figure 3. Yields for noncondensable gas, char, and bio-oil as a function of
pyrolysis temperature for broiler cake and litter samples.
4. Gas Composition. Composition of the syn-gas will depend
and composition of the noncondensable gas was characterized. on the characteristics of the starting material. Syn-gas can be
During pyrolysis, carbon atoms fixed within the calcined used as primary fuel for direct combustion. The main combus-
structure via gas-solid chemical reactions are removed via tibles found in noncondensable gas are carbon dioxide, carbon
formation of product gases such as CO, CO2, CH4, C2H6, and
monoxide, hydrogen, and low molecular weight hydrocarbons
so on, leaving no free carbon as volatile matter in the charred
(HC). Compositional analysis of the noncondensable gas was
materials. Figure 3 presents the noncondensable gas yields for
the broiler litter and cake samples as a function of pyrolysis important to determine the quality of the gas. Figure 4 presents
temperature. Char and noncondensable gases yields were directly the major components of the noncondensable gas produced by
measured. All other gases, including the condensable gases such pyrolysis at three different temperatures, 700, 800, and 900 °C.
as reaction water and pyrolytic oil vapors, plus the noncon- These included CO, CO2, and the low molecular weights
densable gases that were not calibrated, were considered as hydrocarbon gases CH4, C2H6, C3H8, and C4H10 as well as
“condensable gas.” These included tar and hydrocarbon gases hydrogen. The largest fraction of gas was carbon dioxide
greater than C4H10 that were not quantified. The condensable followed by carbon monoxide and then hydrocarbons, with
gas yield that constitutes potential bio-oil or pyrolysis liquid hydrogen yields being the lowest for both samples. As Figure
was calculated as the difference between the biomass pyrolyzed 4 shows, the CO2 component of the noncondensable gas was
and the sum of the measured gases and the residual char. lowest at the lowest temperature (700 °C) for both samples and
Noncondensable gas yield increased for both samples with increased with increasing pyrolysis temperature. The sharpest
pyrolysis temperature and the increase was more significant for increase in carbon dioxide (approximately 50%) was observed
broiler litter than for broiler cake samples. Biogas yield was
for broiler litter samples from 800 to 900 °C. All the
lower for broiler litter samples pyrolyzed at 700 °C, but at 900
hydrocarbons increased with pyrolysis temperature as well, so
°C noncondensable gas yield for broiler litter surpassed that of
did hydrogen. Even though it was represented in the lowest
the broiler cake. Larger amounts of wood present in the broiler
litter than broiler cake will decompose more completely at higher amounts measured, the percentage of hydrogen tripled from 700
pyrolysis temperatures. Char yield was also a function of to 900 °C with both samples. The increase of hydrocarbon yield
pyrolysis temperature. Char yield decreased with pyrolysis with pyrolysis temperature is consistent with the depolymeri-
temperature with a much stronger decline observed for broiler zation of larger molecular weight hydrocarbons (i.e., tars greater
cake char than broiler litter char. When comparing the char than C4) which would otherwise condense to form part of the
yields from the slow pyrolysis (Table 1) with those from the bio-oil constituent.13 The only measured gas that did not increase
fast pyrolysis (Figure 3), it can be seen that there was a with pyrolysis temperature was carbon monoxide, where a slight
significant discrepancy. It is known that pyrolysis is a thermal (not significant) decrease was observed. The composition of the
process evolving in the absence of air which produces syn-gas, hydrocarbons consisted mainly of CH4 with gradually smaller
char, and tar in fractions that depend upon operating temperature, amounts of hydrocarbons of increasing number of carbon.
heating rate, and residence time. For high tar yields, a high Unless it can be converted to combustible gas products, CO2
heating rate and short residence time are requested (fast decreases the quality of syn-gas. For this reason, pyrolysis
pyrolysis), whereas to maximize char and syngas yields, a low
conditions that reduce the amount of CO2 produced are
heating rate and a long residence time should be applied (slow
favorable. Even though the amount of the primary combustible
pyrolysis).20 Pyrolysis temperature also affected the amount of
tar (condensable gases) produced (Figure 3). For broiler cake components of syn-gas (the lower molecular weight hydrocar-
samples, significantly more bio-oil was produced as the tem- bons, hydrogen, and carbon monoxide) increased in broiler litter
perature increased from 700 to 900 °C, whereas for broiler litter, and cake samples with increased temperature of pyrolysis, the
there was only a slight increase (Figure 3). The mass balance same was not true in percentage terms. For broiler litter samples,
on the broiler cake pyrolysis-products reveals that an increase the percentage of these components in the overall composition
in pyrolysis temperature translated predominantly into a mass of the syn-gas, decreased from 37% to 34% for broiler cake
conversion into condensable gases. and from 40% to 31% for broiler litter, when pyrolysis
 Ind. Eng. Chem. Res., Vol. 48, No. 3, 2009 1297
temperature increased from 700 to 900 °C. It is apparent that a Congress; Thome-Kozimiensky, K. J., Ed.; E. Freitag-Verlag fur Umwelt-
higher quality syn-gas is produced for lower pyrolysis temperature. technik: Berlin, 1979; Vol. 1, p 714.
(5) Koutcheiko, S.; Monreal, C. M.; Kodama, H.; McCracken, T.;
Kotlyar, L. Preparation and characterization of activated carbon derived
Conclusions from the thermo-chemical conversion of chicken manure. Bioresour.
Technol. J. 2007, 98, 2459.
In the present study, broiler manure is pyrolyzed to produce (6) Jindarom, C.; Meeyoo, V.; Kitiyanan, B.; Rirksomboon, T.; Rang-
chars that possess enhanced adsorption ability for metal ions. sunvigit, R. Surface characterization and dye adsorptive capacities of char
These chars, without being subjected to subsequent activation, obtained from pyrolysis/gasification of sewage sludge. Chem. Eng. J. 2007,
133, 239.
retained adsorption properties typically only associated with (7) Serio, M. A.; Bassilakis, R.; Kroo, E.; Wójtowicz, M. A. Pyrolysis
activated carbons. Their metal ion sequestering ability greatly Processing of Animal Manure to Produce Fuel Gases ; ACS Division of
exceeds that of chars made from traditional feedstocks such as Fuel Chemistry Preprints; 2002; Vol. 47, p 588.
coal, wood, and coconut shells, with Cu2+ adsorption between (8) Sánchez, M. E.; Martı́nez, O.; Gómez, X.; Morán, A. Pyrolysis of
0.33 and 0.91 mmol/g char. In contrast, chars produced from mixtures of sewage sludge and manure: A comparison of the results obtained
in the laboratory (semi-pilot) and in a pilot plant. J. Waste Manage. 2007,
coal, coconut, or wood exhibited no detectable adsorption of 27, 1328.
Cu2+. Slow pyrolysis of broiler manure entraps pre-existing (9) Kim, S. S.; Agblevor, F. A. Pyrolysis characteristics and kinetics of
phosphorus, primarily in the form of phosphate ions within the chicken litter. J. Waste Manage. 2007, 27, 135.
carbon matrix. It is postulated that the strategic exposure of (10) Domı́nguez, A.; Fernández, Y.; Fidalgo, B.; Pis, J. J.; Menéndez,
J. A. Bio-syngas production with low concentrations of CO2 and CH4 from
chemical surface groups, such as phosphate, contributes sig- microwave-induced pyrolysis of wet and dried sewage sludge. Chemosphere
nificantly to the adsorption phenomena between charged species. 2008, 70, 397.
A mass balance on the char, syn-gas, and tar produced during (11) Lima, I. M.; Marshall, W. E. Production of granular activated
pyrolysis revealed that yield for both condensable and noncon- carbons from pig manure for metal ions adsorption. J. Res. Sci. Technol.
densable gases increased with pyrolysis temperature to the 2007, 4, 9.
(12) Caballero, J. A.; Font, R.; Marcilla, A.; Garcia, A. N. Flash pyrolysis
expense of a decrease in char yield. The change was much more of Klason lignin in a Pyroprobe 1000. J. Anal. Appl. Pyrolysis 1993, 27,
significant for broiler cake than broiler litter. Pyrolysis temper- 221.
ature affected both char and syn-gas properties, with improved (13) Boateng, A. A.; Hicks, K. B.; Vogel, K. P. Pyrolysis of switchgrass
metal ion uptake contrasting with decreased syn-gas quality as (Panicum Virgatum) harvested at several stages of maturity. J. Anal. Appl.
Pyrolysis 2006, 75, 55.
pyrolysis temperature increased. Slow pyrolysis of broiler (14) Lima, I. M.; Marshall, W. E. Granular activated carbons from broiler
manure could possibly provide a char with enhanced ability to manure: physical, chemical and adsorptive properties. Bioresour. Technol.
adsorb metal ions in addition to a syn-gas that can be used as J. 2005a, 96, 699.
a fuel. (15) Lima, I. M.; Marshall, W. E. Utilization of Turkey manure as
granular activated carbon: physical, chemical and adsorptive properties. J.
Waste Manag. 2005b, 25, 726.
Acknowledgment (16) Lima, I. M.; Marshall, W. E. Adsorption of Select Environmentally
Important Metals by Poultry Manure-based Granular Activated Carbons.
The authors would like to thank Renee Bigner, Bonnie Dillon J. Chem. Technol. Biotechnol. 2005c, 80, 1054.
and Rick Cook for the various analyses and Dana Miles for (17) Lima, I. M.; McAloon, A.; Boateng, A. A. Activated carbon from
supplying the broiler litter samples. broiler litter: Process description and cost of production. J. Biomass
Bioenergy 2008, 32, 568.
(18) Brown, M. J.; Lester, J. N. Metal removal in activated sludge; the
Literature Cited role of bacterial extracellular polymers. Water Res. 1979, 13, 817.
(19) Ho, Y. S.; Wase, D. A. J.; Forster, C. F. Batch Nickel removal
(1) Shinogi, Y.; Yoshida, H.; Koizumi, T.; Yamaoka, M.; Saito, T. Basic from aqueous solution by sphagnum moss peat. Water Res. 1995, 29, 1327.
characteristics of low-temperature carbon products from waste sludge. AdV. (20) Baggio, P.; Baratieri, M.; Gasparella, A.; Longo, G. A. Energy and
EnViron. Res. 2003, 7, 661. environmental analysis of an innovative system based on municipal solid
(2) Shinogi, Y.; Karni, Y. Pyrolysis of plant, animal and human waste: waste (MSW) pyrolysis and combined cycle. Appl. Therm. Eng. 2007, 28,
physical and chemical characteristics of the pyrolytic products. Bioresour. 136.
Technol. J. 2003, 90, 241.
(3) Qian, Q.; Machida, M.; Tatsumoto, H. Preparation of activated ReceiVed for reView June 25, 2008
carbons from cattle-manure compost by zinc chloride activation. Bioresour. ReVised manuscript receiVed October 27, 2008
Technol. J. 2007, 98, 353. Accepted November 4, 2008
(4) Bilitewski, B. Production and Possible Applications of ActiVated
Carbon from Waste; Recycling Berlin, 1979 International Recycling IE800989S