201406031353105198

Journal of Agricultural Science and Applications (JASA)

JASA Volume 1, Issue 2 June 2012 PP. 37-44 DOI: 10.14511/jasa.2012.010201 © American V-King Scientific Publishing - 37 -

Pyrolysis of Poultry Litter Fractions for Bio-Char and Bio-Oil Production

Kaushlendra Singh1, L. Mark Risse2, K. C. Das2, John Worley2, Sidney Thompson2

Division of Forestry and Natural Resources, West Virginia University, Morgantown, WV USA Department of Biological and Agricultural Engineering, University of Georgia, Athens, GA USA

[email protected]

Abstract-Although production of activated carbon, catalytic gasification of poultry litter char, combustion, and physical treatment of poultry litter using screening have been thoroughly studied, yet eco-friendly and value added utilization of litter has not been adopted by the poultry industry. This paper examines whether or not screening and pyrolysis can be combined for value added utilization of poultry litter. Poultry litter was sampled from three commercial farms and each of the samples was divided into nine subsamples. These subsamples were randomly given nine treatments as follows: one control, screen #5 coarse fraction, screen #5 fine fraction, screen #10 coarse fraction, screen #10 fine fraction, screen #18 coarse fraction, screen #18 fine fraction, screen #20 coarse fraction, and screen #20 fine fraction. All coarse and fine fractions were pyrolyzed in a batch reactor at 500°C under a nitrogen flow rate of 2 lpm to produce char and condensate. The condensate was separated into three fractions based on density. Bomb calorimeter, proximate, and ultimate analysis were performed on un-pyrolyzed but screen-treated samples, char, and three phases of condensate. The results showed that the pyrolysis of the coarse fraction (screen#20) produced 44.5% char which retained 43.5% of total feedstock energy. Overall, the pyrolysis products captured 57.2% of total feedstock energy and 53.8% of total feedstock carbon. The light phase of the condensate (4.94% yields on dry biomass) had a calorific value of 34.83MJ/kg.

Keywords-Poultry Litter; Pyrolysis; Screening; Bioenergy; Biofuels

I. INTRODUCTION Poultry litter is a combination of bedding material

(typically pine shavings), spilled feed, feathers, manure, and dead birds. It is the product of total cleanout of a poultry house, typically after 3-5 flocks of birds grown on it. Utilization of poultry litter for energy production can improve water quality by reducing nutrient imbalance in areas where poultry litter production exceeds the needs for nutrients on local croplands and pastures. Reference [1 and 2] recommended a simple screening of poultry litter, using screen # 18 and screen # 20 with mesh opening of 1.0 mm and 0.85 mm respectively, to concentrate the nitrogen and minerals in a fine fraction which increases its value as fertilizer thus making it more transportable. Reference [2] suggested energy production from the remaining coarse fraction by combustion or gasification as it retained most of the woody material.

In combustion studies, [3 and 4] documented the thermal performance and poultry litter consumption rate in a two staged combustion furnace. Reference [4] concluded that inclusion of 20% poultry manure with coal didn’t change the emission pattern but approximately 50% additional ash was produced. The direct combustion of poultry litter exhibited problems like boiler corrosion and ammonia emission, which motivated studies in the gasification of poultry litter.

Catalytic steam gasification of poultry litter, its reaction kinetics, and ammonia emissions during gasification have been widely studied [5, 6, and 7]. Char derived from pyrolysis of poultry litter was the primary fuel source used for these studies. Reference [8 and 9] presented pyrolysis kinetics for poultry litter and reference [10] used poultry litter to produce granular activated carbon and optimized conditions for its production. The kinetics of biomass decomposition under pyrolysis conditions and proportion of end pyrolysis products (char, condensate, and gasses) depend upon chemical composition of the biomass in terms of cellulose, hemi-cellulose, lignin, extractives, and ash. Cellulose degrades in the temperature range of 240 to 350˚C producing anhydrocellulose and levoglucosan. The later is produced when a glucosan radical forms, and it does not get a chance to bridge with oxygen present in cellulose polymer [11].Hemi-cellulose contains hetero-polysaccharide and thermally degrades in the temperature range of 130-194˚C [11]. The differences in thermal decomposition chemistry of hemi-cellulose and cellulose are not well-known. The thermal decomposition of lignin occurs in the broad temperature range of 280 to 500̊C yielding phenol v ia cleavage of ether and carbon-carbon linkages. According to [11] lignin produces more residual char than cellulose or hemi-cellulose. Inorganic minerals do not degrade during pyrolysis and remain in the solid residue; however, some cationic metals like sodium and potassium can catalyze pyrolysis reactions of cellulose, hemi-cellulose, and lignin [12]. Extractives decompose at a higher rate and lower temperature than lignin but in a similar way [13]. In addition, the physical and chemical changes in the biomass during pyrolysis and in turn, the physiochemical properties of pyrolysis products also depend on the pyrolysis conditions (heating rate, peak temperature, sweep gas flow rate, pressure, and the presence of a catalyst.

In a study of the kinetics of thermal degradation of poultry litter, [14] devised a four independent parallel reaction models involving cellulose, hemi-cellulose, lignin, and protein as the main degradable components. Reference [14] identified four distinct stages of poultry litter decomposition under pyrolysis conditions as following: hemicelluloses decomposition (180°C to 365°C), cellulose decomposition (300°C to 460°C), protein decomposition (300°C to 600°C), and lignin decomposition (180°C to 800°C).

Most studies on alternative uses of poultry litter have either focused on physical treatment (screening) to separate nutrient rich fraction of poultry litter to reduce nutrient transport cost or thermal treatment (combustion, gasification, or pyrolysis) to produce heat, gas fuel, or char. Combining the screening and pyrolysis process should serve a dual purpose. It should reduce nutrient transport cost and hence reduce potential phosphorous pollution in water bodies. It should also produce some value added products like char and bio-oil. This paper presents the results of a study which combined



screening and pyrolysis for value added utilization of poultry litter. The specific objectives of this study were:

• To study the effect of screen size and pyrolysis on fuel properties of poultry litter.

• To characterize the pyrolysis condensate to determine its nutrient and energy content.

• To study the efficiency of the pyrolysis process.

The set of sieves chosen for this study was based on previous research work done in this area. However, those studies were focused on fractionation of mass and nutrients as a result of screening. This study was focused on fractionation of energy as a result of screening. Our goal was to find out which sieve gives a coarse fraction which can be used to produce char with high energy and low ash content. So, the selection was not only based on calorific value data but also ash content results. The study results will help the poultry industry by adding value to poultry litter and by reducing environmental impact due to its land application.

II. MATERIALS AND METHODS

A. Poultry Litter Sampling Method Container loads of (approximately 100 kg) poultry litter

were obtained from three commercial broiler farms in northeast Georgia. The poultry litter came from total cleanout of poultry houses that had grown at least three flocks of broiler chickens since the last cleanout and contained pine shavings as bedding material. Information about the farm size, size of chickens, litter management etc was not collected. The first lot of poultry litter was obtained by collecting samples from random locations in the building according to a standard procedure described in Appendix E of [16]. In this method, the poultry house is mentally divided into three zones of equal size and six cores are collected within each zone. The first core is collected within a foot of the feed line using a spade and a small trench the width of the spade to the depth of the litter is cleared and removed. This process is repeated, gathering six cores from each zone in a zigzag pattern along the length of the building, taking the last core within each zone within one foot of the water line. After collecting samples from all three zones, all of the litter samples are mixed and crumbled in a pile and transferred into a storage bucket.

The second lot was taken from a storage pile that had been delivered to a site for land application. The third lot was taken from a pile in the middle of a poultry house that was ready to be loaded onto a truck. Samples for both the second and third lots were taken from several locations in the piles. The three different lots of poultry litter represented the target population and variability among the poultry litter loads coming from various poultry farms. The sampled litter was stored in a refrigerated room below 4ºC to avoid microbial decomposition. Sub-samples of 6.0 to8.0 kg drawn from the containers of poultry litter were used for assigning treatments randomly.

B. Experiment Design Randomized Complete Block Design was used to analyze

treatment responses. There were three blocks (poultry litter collected from three poultry farms). Blocking effect takes care of variability across poultry litter lots due to factors other than treatments. These factors might include different feeding

material used to grow chickens, different varieties of pine wood shavings used for bedding material, or different poultry house operations. There were nine treatments. First was the control treatment. The other eight treatments were a combination of two factors: screen size and type of fraction. The nine treatments are given below:

1. Control (no screening),

2. Screen #5 (coarse or retained material),

3. Screen #5 (fine or screened material),






9. Screen #20 (fine or screened material).

One sub-sample was used to obtain a coarse fraction (material which did not pass through a screen) while another experimental unit was used to obtain a fine fraction (material which passed through the same screen). Amount of each sub-sample used for screening was not recorded because our target was to get 4 to 5 Kg of screen treated samples. That’s why the sub-samples fed to screening varied from treatment to treatment. A similar process was repeated for the rest of the treatments. This approach preserved the randomness and independence of the nine treatments. The treatment responses were calorific value, ash, volatile matter, and fixed carbon content. The data was analyzed according to a randomized complete block design with three poultry farms considered as blocks using statistical analysis software (SAS 9.1, SAS Institute Inc., Cary, N.C.) with N=27 (3 blocks, 9 treatments/block). Pair-wise least significant difference (LSD) test was performed to study the differences between two responses. The differences in calorific values were explained by proximate analysis data [9]. After the statistical analysis, the screening treatment for producing the highest calorific value and lowest ash content char was selected. The three phases of condensate corresponding to that screening treatment were characterized for moisture content, calorific value, carbon, hydrogen, nitrogen, and oxygen content.

C. Laboratory Procedure The poultry litter sub-samples were screened using screens

listed in the above section (USA Standard Testing Sieve, Fisher Scientific Co., Pittsburgh, PA) on an electrical sieve shaker (Model CL 5028, Soil Testing Inc, Evanston, IL.). The separation of mass due to the screening process was not recorded because it was well established by [17] and [18]. The coarse and fine fractions were initially dried at 45°C for at least two days in an oven (Isotempoven, Fisher Scientific Co., Pittsburgh, PA) to control the effect of moisture during pyrolysis. A known amount of the dried fractions was pyrolyzed in a batch reactor at 500ºC under a nitrogen atmosphere and the weights of condensate and char produced were recorded (described in the next section). The actual weight of the feedstock and the amount of char and condensate produced were used to estimate char and condensate yields. Actual mass input as well as actual mass output of solids and liquids were used to estimate gas yield.



The condensate was separated into three phases (heavy, medium, and light phase) using a separatory funnel (2 liter Wilmad Lab glass Separatory Funnel Suibb Glass Stopcock, Fisher Scientific, Pittsburgh, PA) by gravity-density separation (fig. 1). All the condensate produced during the pyrolysis process was poured in a separatory funnel and allowed to settle. After 7-8 hours, the condensate settled into three distinct layers: heavy, medium, and light phase. Each phase was carefully collected in a dark glass bottle (Fisher brand Amber Rounds with PTFE face PE-lined closures Cat# 02-911-740, Fisher Scientific, Pittsburgh, PA) and its amount was recorded to calculate phase yield. The heavy phase was semi-solid at room temperature and it would hardly flow through the nozzle of the separatory funnel. A fiber optic illuminator (Model: 41500-50, Cole-Parmer Instrument Company, Vernon Hills, IL) was used to heat the heavy phase and to provide enhanced visibility during separation of layers. Once, heated by the light source, the heavy phase flowed through the funnel nozzle. The dark glass bottles containing liquid phases were stored at room temperature prior to analysis.

Figure 1. Separation of condensate into three phases was done using a separatory funnel

( . .)

1100

CharChar m f

feedstock

YY

mc=

−

(1)

( . .)

*

1100

Condensate feedstock char Char

Condensate m ffeedstock

Y mc mc YY

mc− +

=

−

(2)

Where YChar = char yield with moisture; YCondensate = condensate yield with moisture; YChar (m.f.) = char yield without moisture; YCondensate (m.f.) = condensate yield without moisture; mcfeedstock = moisture content of feedstock; and mcChar = moisture content of char.

To perform the bomb calorimeter, proximate, and ultimate analyzes, the samples were prepared according to the following procedure. The dried materials were ground to 4mm (FRITSCH, Pulverisette, Industries Trane, Idar, Oberstein) followed by further reduction in particle size to less than 1 mm (Thomas Scientific, Swedesboro, NJ). Char samples were ground in a mixer mill (8000 M MIXER/MILL, Spex Sample Prep, NJ). Calorific value of the ground char and poultry litter fractions without pyrolysis were measured according to ASTM D5865 [19] using a Bomb Calorimeter (Model 1351, Parr Instrument Company (Parr), Moline, IL). According to the standard, a known amount of sample (less than 1g or less than 8000 calorie content) is combusted in excessive oxygen in an adiabatic bomb calorimeter and heat released during the combustion process is harvested by a known amount of water

(2 kg), thereby raising its temperature. The bomb calorimeter must detect at least 0.002°C temperature change in order to estimate calorific value of the substance. Later, calorific value of the sample is computed using the amount of water, the increase in temperature, specific heat, and amount of the sample. The instrument gives calorific value of the sample in calories/g within 0.2 to 0.3% of a relative standard deviation which is converted to MJ/kg (one calorie = 4.1868 joule). ASTM D5865 procedure requires that correction should be made for heat of formation of sulfuric acid due to the presence of sulfur in the sample and heat of formation of nitric acid due to the presence of nitrogen in the sample. During the correction process, both heats of formation are subtracted from the calorimeter output. These corrections were not taken into account in this study because both chemical reactions contribute to the heat produced during the combustion process and the authors wanted to report the total heat released due to combustion. Another reason for not taking both corrections into account is that the amount of sulfur and nitrogen is pretty much the same in all the samples so error due to sulfur and nitrogen would not cause problems in comparing the samples. Proximate analysis was performed according to ASTM D 5142-04 using a 2 g sample in the Thermo-gravimetric Analyzer (Model TGA701, LECO, St. Joseph, MI) [20]. The "proximate" analysis gives moisture content, volatile content (when heated to 950 C) within 0.02% of a relative standard deviation, the free carbon remaining at that point, and the ash (mineral) in the sample based on the complete combustion of the sample to carbon dioxide and liquid water.

The moisture contents of the three phases of condensate were measured using a Karl Fisher Titrator (Model Mettler Toledo DL 31, Star Systems, Columbus, OH). The fundamental principle behind the Karl Fisher method is based on the Bunsen Reaction between iodine and sulfur dioxide in an aqueous medium. During the reaction alcohol (methanol) is used as solvent and a base (pyridine) as a buffering agent. The alcohol reacts with sulfur dioxide (SO2) and a base to form an intermediate alkyl-sulfite salt, which is then oxidized by iodine to an alkyl-sulfate salt. Water and iodine are consumed in a 1:1 ratio in the above reaction. Once all of the water present is consumed, the presence of excess iodine is detected volumetrically by the titrator’s indicator electrode. The amount of water present in the sample is calculated based on the concentration of iodine in the Karl Fisher titrating reagent (i.e., titer) and the amount of Karl Fisher Reagent consumed in the titration.

Calorific values of heavy and the light phases of condensate were measured using a bomb calorimeter. The calorific value of the medium phase was estimated using the following equation [21] because its sample was not able to create a temperature rise above 0.002°C inside the bomb calorimeter (precision 0.26% and accuracy 0.23%).

ANOSHCHHV 0211.00151.01034.01005.01783.13491.0 −−−++= (3)

The equation can estimate the calorific values with an absolute error of 1.45% and holds for the following ranges of carbon (C), hydrogen (H), nitrogen (N), sulfur (S), oxygen (O), and ash (A): 0.0%≤C≤92.3%, 0.4%≤H≤25.2%, 0.0%≤O≤50.0%, 0.0%≤N≤5.6%,0%≤S≤94.1%, 0%≤A≤71.4% .

The ultimate analysis was performed on the coarse fraction, char, and three phases of condensate using a CHNS analyzer



(Model LECO CHNS-932, LECO Corporation, St. Joseph, MI). In this analysis, approximately 1mg of sample is placed in a pressed silver capsules and combusted in a furnace at 1000°C in the presence of excessive oxygen creating a mixture of water vapor, CO2, NO2, SO2, and O2. The mixture of gases is then passed through a copper column which absorbs oxygen. The remaining mixture of gases is passed through four infra-red sensors which detect hydrogen, carbon, nitrogen, and sulfur. The "ultimate" analysis" gives the composition of the biomass in wt% of carbon (±0.001%), hydrogen (±0.01%) and oxygen (the major components) as well as sulfur (±0.02%) and nitrogen (±0.01%). The data obtained from this analysis was used to calculate the efficiency of the pyrolysis process. The efficiency of the pyrolysis process may be expressed using energy conversion efficiency of the char, fixed carbon yield, and char carbon yield [22]. Similar to energy conversion efficiency [23] used the term process thermal efficiency. The energy conversion efficiency is an estimation of percentage of feedstock energy left in products and it is not an energy balance of the process; hence, it does not include external heat supplied to the process, heat lost during the condensation process, heat loss due to radiation from the reactor body, or heat produced during the pyrolysis process due to chemical reactions. A detailed energy as well as entropy balance for a continuous pyrolysis reactor has been presented by [24]. The energy conversion efficiency (ηEnergy) and fixed carbon yield (FCYield) were estimated using the following expressions [22]:

//

Char CondensateEnergy Char CondensateYield

Feedstock

HHVYHHV

η = (4)

Where ηEnergy= energy conversion efficiency; YChar/Condensate

Yield= char or condensate yield (%); HHVChar/Condensate= calorific value of char or condensate (MJ/kg); and HHVFeedstock= calorific value of the feedstock biomass [22].

100Yield CharyieldFCFC Y

Ash=

− (5)

Where FCYield = fixed carbon yield; Ash = percent ash content of the feedstock and FC = fixed carbon content of the char. To check char carbon yield (YC) equation 6 was used [22].

%%C Char

charCY YfeedC

= (6)

Where % charC is the percent carbon content of the dry char, and % feedC is the percent carbon content of the dry feed. A similar equation may be extended to estimate condensate carbon yield (YCondensateC).

1) Batch Pyrolysis Experiments Approximately, 2.0 to 3.0 kg of poultry litter biomass was

pyrolyzed inside a cubical pyrolysis reactor. Before placing the biomass inside the reactor, empty weights of the reactor and condensing units were recorded. The reactor was filled with biomass only up to 75% of its volumetric capacity to leave some head space. The reactor filled with biomass was weighed to calculate the actual amount of biomass fed. The cubical reactor had a fixed volume, and thus weight of the biomass fed to the reactor depended upon the density of the biomass. The density of the biomass (and thus the weight fed

to the reactor) varied from fraction to fraction. After placing the sample in the reactor and before heating began, nitrogen was purged into the reactor for 10 minutes to create inert conditions. The reactor was heated under inert atmosphere until the biomass temperature reached 500̊ C using a furnace (Model: 30400 thermolyne, BarnStead International, Dubuque, Iowa) (fig. 2). The temperature of biomass was recorded by thermocouple located at the center of the biomass. The furnace has heating elements located in the side walls only and not on top or bottom. The reactor pressure was not measured during the experiments. The nitrogen was purged at a flow rate of 2 liters per minute to maintain an inert atmosphere and to ensure the quick removal of volatiles. The operating conditions of the pyrolysis process were not changed during the experiments.

Figure 2. Batch Pyrolysis installations used and biomass temperature profile recorded on a computer (1) Computer connected to thermocouples, (2) Carrier

gas cylinder, (3) Mass flow controller (4) Oven, (5) Pyrolysis reactor, (6) Chiller, (7) Condensing traps (Source: Garcia-Perez et al., 2007)

The pyrolysis vapors were rapidly removed from the hot biomass and quenched in five ice cooled traps connected in series and maintained at 0°C temperature. The remaining uncondensed gases were released to the atmosphere (fig. 2). Once the biomass temperature reached 500°C, the furnace was shutdown and the reactor was allowed to cool under a nitrogen atmosphere. After the biomass cooled to room temperature for four-five hours, the reactor was disconnected from the condensing unit and each was weighed separately. These weights of reactor and condensers and their weight before pyrolysis were used to estimate char and condensate yields. The reactor was opened and the char was collected in a ziplock bag. The condensing unit was opened and condensate from each of the five stainless steel cylinders was collected in a transparent one liter glass bottle and the amount was recorded. This condensate was later transferred to a separatory funnel to get condensate fractions: heavy, medium, and light phases (fig. 1). After each pyrolysis run, all connecting tubing and condenser cylinders were thoroughly cleaned using acetone and air dried.

III. RESULTS AND DISCUSSION

A. Calorific value of Fractionated Poultry Litter without pyrolysis The calorific values obtained for raw poultry litter of

14.73MJ/kg was comparable to the calorific values reported in literature 14.45MJ/kg by [25], 14.26MJ/kg by [26], 14.98MJ/kg by [27], and 14.40MJ/kg by [9]. The pair-wise comparison showed that the screens #18 and #20 divided poultry litter into an energy-dense coarse and an energy-scarce fine fraction (table I).



TABLE I. CALORIFIC VALUE (MEAN ± SE ON MOISTURE FREE BASIS, n=3) OF POULTRY LITTER FRACTIONS

Sl. No.

Poultry litter fraction Calorific value, MJ/kg

1 Raw 14.73 ± 0.83b

2 Screen #5 Coarse (4 mm opening) 14.95 ± 0.76b

3 Screen #5 Fine (4 mm opening) 14.37 ± 1.00b

4 Screen #10 Coarse (2 mm opening) 15.16 ± 0.79b

5 Screen #10 Fine (2 mm opening) 14.18 ± 1.19b

6 Screen #18 Coarse (1 mm opening) 15.05 ± 0.85c

7 Screen #18 Fine (1 mm opening) 13.66 ± 1.10a

8 Screen #20 Coarse (0. 85 mm

opening) 15.19 ± 0.77c

9 Screen #20 Fine (0.85 mm

opening) 12.96 ± 1.70a

Any two numbers in a column followed by the same letter are not significantly different at 95% confidence level

The coarse fraction obtained from screen #20 had the calorific value (15.19MJ/kg) significantly higher than the fine fraction (12.96MJ/kg) obtained from the same screen (p-value= 0.0021). Similar observation was recorded screen #18 (p-value= 0.0360). The differences in calorific values can be explained by proximate analysis (table II) because the screening process did not divide cellulose, hemicelluloses, fat, and lignin content; however, it concentrated protein and ash into the fine [9]. Both screens, #20 and #18, not only concentrated volatile matter in the coarse fractions but also yielded ash-rich fine fractions (table II).

TABLE II. PROXIMATE ANALYSIS (MEAN ± SE ON MOISTURE FREE BASIS, n=3) OF POULTRY LITTER FRACTIONS OBTAINED FROM SCREENING

TREATMENT

Sl. No.

Treatment Moisture[a] Volatile Matter

Ash Fixed Carbon

1 Raw 15.22 ± 3.76b

65.15 ± 2.47c

24.55 ± 2.24b

10.30 ± 0.39a

2 Screen #5

Coarse 11.31 ±

0.31a 65.41 ±

2.34c 23.75

± 2.62a 10.85 ±

0.40a

3 Screen #5

Fine 14.77 ± 0.64b

63.37 ± 3.51b

26.31 ± 4.61b

10.32 ± 1.12a

4 Screen #10

Coarse 13.21 ±

0.74a 65.56 ±

2.73c 23.45

± 2.79a 10.99 ±

0.13a

5 Screen #10

Fine 12.35 ±

2.56a 63.63 ± 3.54b

25.85 ± 4.84b

10.53 ± 1.31a

6 Screen #18

Coarse 13.34 ±

1.50a 66.38 ±

2.07c 22.48

± 2.34a 11.13 ±

0.29a

7 Screen #18

Fine 12.04 ±

2.59a 62.24 ±

4.13a 27.98

± 5.03c 9.77 ± 0.97a

8 Screen #20

Coarse 13.07 ±

3.06a 65.97 ±

2.14c 23.08

± 2.65a 10.95 ±

0.29a

9 Screen #20

Fine 11.04 ±

2.47a 61.87 ±

4.43a 28.66

± 5.57c 9.47 ± 1.15a

[a] The moisture content shown is after initial 48 hours drying at 45°C. Any two numbers in a column followed by the same letters in a column are not

significantly different at 95% confidence level

B. Pyrolysis Products’ Yeild

1) Char, Condensate, and Gas Yeild The pyrolysis products’ yield could not be compared to

other published research since no published data were found for batch pyrolysis of poultry litter. According to [28], fast pyrolysis of poultry litter gave 27% char, 15% light bio-oil, 35% heavy bio-oil, and 10% gas yield. Screen #20 coarse fraction gave significantly lower char yield (44.47%) than its corresponding fine fraction (49.01%) (p-value= 0.01) (table

III). A similar influence for screen #18 was observed (p-value= 0.04). The higher char yield from the fine fractions than coarse fractions may be attributed to their significantly high ash content (p-value= 0.05). There was no difference in ash content between the two coarse fractions obtained from screen #18 and screen #20. Figure 3 shows screen#20 coarse fraction and pyrolysis products derived from it (char and heavy, medium, and light phases of condensate).

Figure 3. Screen # 20 coarse fraction and pyrolysis products derived from it. (a) Screen # 20 coarse fraction, (b) char, (c) heavy, medium, and light phases

of condensate

TABLE III. CHAR, CONDENSATE, AND GAS YIELD (MEAN ± SE ON MOISTURE FREE BASIS, N=3) RESULTING FROM PYROLYSIS OF POULTRY LITTER

AFTER SCREENING TREATMENT

Sl. No.

Treatment Char Condensate Gases*

1 Raw 46.22 ±

3.54b 30.33 ± 3.51a 23.45 ±

1.55a

2 Screen #5 Coarse

45.41 ±

4.01b 29.62 ±1.99a 24.97 ±

5.80a

3 Screen #5 Fine

44.70 ±

3.75b 30.71 ± 4.67a 24.59 ±

3.01a

4 Screen #10 Coarse

43.89 ±

3.22b 32.09 ± 3.50a 24.02 ±

0.67a

5 Screen #10 Fine

46.38 ±

2.75b 30.12 ± 1.24a 23.51 ±

1.56a

6 Screen #18 Coarse

43.15 ±

1.97a 31.88 ± 1.82a 24.98 ±

0.38a

7 Screen #18 Fine

46.74 ±

3.72b 30.35 ± 2.44a 22.91 ±

2.46a

8 Screen #20 Coarse

44.47 ±

2.41a 30.68 ± 2.80a 24.85 ±

0.88a

9 Screen #20 Fine

49.01 ±

3.89b 30.15 ± 4.41a 20.85 ±

1.88a

Any two numbers in a column followed by the same letters are not significantly different at 95% confidence level.*Gas yield was calculated by

difference

Statistically, screening treatments had no significant effect on either condensate or gas yield (table III). The temperature-time data obtained from the thermocouple was used to calculate the actual heating rate during the pyrolysis process. The calculated average heating rates were 2.57, 4.14, 2.06, 3.61, 1.81, 2.19, 2.22, 2.06, and 4.89°C/min respectively, for treatments one through nine.



2) Condensate Phase Yield

Screening treatment had no significant influence on the yields of heavy, medium, or light phases of condensate (table IV).

TABLE IV. PHASE YIELD AS PERCENTAGE OF TOTAL CONDENSATE (AS RECORDED) (MEAN ± SE ON MOISTURE FREE BASIS, n=3) RESULTING FROM

PYROLYSIS OF POULTRY LITTER AFTER SCREENING TREATMENT

Sl. No.

Treatment Heavy Phase

Medium Phase

Light Phase

1 Raw 5.45 ± 1.36a

83.58 ± 1.18a

10.97 ± 0.98a

2 Screen #5 Coarse

2.43 ± 1.77a

80.08 ± 2.23a

17.50 ± 2.74b

3 Screen #5 Fine

3.12 ± 0.35a

85.61 ± 1.59a

11.27 ± 1.40a

4 Screen #10 Coarse

2.95 ± 0.93a

84.11 ± 1.83a

12.94 ± 2.39ab

5 Screen #10 Fine

3.66 ± 1.01a

83.68 ± 3.13a

12.66 ± 2.13ab

6 Screen #18 Coarse

2.76 ± 0.25a

82.77 ± 3.14a

14.47 ± 3.35a

7 Screen #18 Fine

4.16 ± 0.49b

82.87 ± 4.65a

12.96 ± 2.92a

8 Screen #20 Coarse

3.64 ± 1.94a

80.78 ± 2.68a

15.58 ± 4.32b

9 Screen #20 Fine

4.65 ± 1.33a

81.06 ± 2.31a

14.29 ± 3.10b

Any two numbers in a column followed by the same letters are not significantly different at 95% confidence level

Moisture had a significant effect on the yield of medium phase (p-value= 0.0048) as well as on the yield of light phase (p-value= 0.0126). Other possible factors affecting the phase yield would be initial sample weight, and particle size which may affect heat and mass transfer.

C. Fuel Properties of Pyrolysis Products 1) Fuel Properties of Char made from Fractionated

Poultry Litter

Chars produced from coarse fractions of screen #20 and #18 had the highest calorific values but not significantly different from their un-pyrolyzed parentfeedstock (table V). It was interesting to observe that screen# 20 coarse fraction char had significantly higher calorific value than #20 fine fraction char (p-value= 0.033) but similar to the calorific value of #18 coarse fraction char.

The differences in calorific value of chars made from pyrolysis of screened poultry litter may be explained by proximate analysis results [9]. High calorific value is associated with high volatile matter and fixed carbon; however, high moisture and ash content tend to decrease the calorific value of biomass. Ash content was the most influential variable (p-value < 0.0001) (table VI).

There was no difference in volatile matter content of char produced as result of the nine treatments. Volatile matter in the raw litter was affected by screening; however, the volatile matter content in chars was independent of screening. The screening caused separation of wood rich material into the coarse fraction and mineral rich material in the fine fraction. The wood rich material had more volatile matter compared to mineral rich material which explains why volatile matter in raw poultry litter was affected by sieving.

TABLE V. CALORIFIC VALUE (MEAN ± SE ON MOISTURE FREE BASIS, n=3) OF CHAR MADE FROM FRACTIONATED POULTRY LITTER

Sl. No.

Poultry litter fraction Calorific value, MJ/kg

1 Raw 13.86 ± 1.74a

2 Screen #5 Coarse 14.12 ± 1.25a

3 Screen #5 Fine 14.42 ± 1.78a

4 Screen #10 Coarse 14.51 ± 1.03a

5 Screen #10 Fine 14.52 ± 1.98a

6 Screen #18 Coarse 15.04 ± 1.17b

7 Screen #18 Fine 13.81 ± 1.98a

8 Screen #20 Coarse 14.91 ± 1.06b

9 Screen #20 Fine 13.17 ± 1.75a

Any two numbers column followed by the same letters are not significantly different at 95% confidence level

TABLE VI. PROXIMATE ANALYSIS (MEAN ± SE ON MOISTURE FREE BASIS, n=3) OF POULTRY LITTER CHAR MADE AFTER SCREENING TREATMENT OF

POULTRY LITTER

Sl. No.

Treatment Moisture Volatile Ash Fixed Carbon

1 Raw 1.17 ± 0.56a

15.19 ± 0.52a

52.45 ± 4.33b

32.36 ± 3.83b

2 Screen #5

Coarse 1.03 ± 0.48a

15.40 ± 0.54a

52.11 ± 3.69b

32.50 ± 3.17b

3 Screen #5

Fine 0.99 ± 0.49a

14.50 ± 1.34a

52.36 ± 6.27b

33.14 ± 4.93b

4 Screen #10

Coarse 1.08 ± 0.52a

15.04 ± 0.87a

51.02 ± 3.83b

33.95 ± 3.00b

5 Screen #10

Fine 0.43 ± 0.36a

15.85 ± 2.07a

51.20 ± 6.97b

32.96 ± 4.90b

6 Screen #18

Coarse 0.98 ± 0.46a

14.84 ± 0.71a

50.05 ± 3.35a

35.11 ± 3.02c

7 Screen #18

Fine 1.60 ± 0.22a

15.00 ± 1.38a

54.06 ± 6.57b

30.94 ± 5.22a

8 Screen #20

Coarse 1.11 ± 0.55a

14.91 ± 0.98a

50.23 ± 3.26a

34.85 ± 3.03c

9 Screen #20

Fine 1.40 ± 0.13a

15.51 ± 1.41a

55.54 ± 5.91c

28.95 ± 5.29a

Any two numbers a column followed by the same letters are not significantly different at 95% confidence level

After pyrolysis, most of the volatile matter was lost from both coarse as well as fine fractions leaving material with almost no volatile matter (chars). That is why chars had similar volatile matter content. All the chars had roughly 50% ash content. Generally, high fixed carbon content gives high calorific value [29] and screen #20 and #18 coarse fraction chars had the highest fixed carbon contents which justified its energy richness. Both screens #18 and #20 are suitable for producing energy rich coarse fraction and one may choose either of these. We decided to select screen #20 to further analyze pyrolysis products produced from it.

2) Fuel Properties of Condensate The three fractions of the condensate were distinct in

appearance (fig. 1). The heavy fraction was black in appearance and looked semi-solid but was ductile at room temperature. The medium fraction was an orange-red water-like liquid. The light fraction was a dark grey easy flowing liquid. The heavy and the light fractions were energy rich; however, the medium phase was mostly water (p-value< 0.0001).



Though the nitrogen content (% by wt) of the medium phase was significantly lower than the other two phases (p-value< 0.0001), the total nitrogen content contained in the medium phase would be higher than in the other two phases (discussed in later section) due to the high yield of the medium

phase (80.78% of the total condensate) (table IV). The medium phase liquid had significantly higher hydrogen and oxygen content than the other two phases (p-value<0.0001), which may be explained by its high water content.

TABLE VII. PROPERTIES OF THE THREE PHASES OF THE CONDENSATE OBTAINED FROM PYROLYSIS OF THE COARSE FRACTION (SCREEN # 20, 0.85 MM) OF POULTRY LITTER. THE VALUES ARE MEAN ± SD ON MOISTURE FREE BASIS, n=3

Type Carbon Hydrogen Nitrogen Sulfur Oxygen Calorific value, MJ/kg Moisture Content Heavy Phase 67.36 ± 2.12a 8.79 ± 0.07a 7.33 ± 0.10b 0.29 ± 0.02b 16.22 ± 5.32a 27.85 ± 0.95b 4.58 ±0.91a

Medium Phase 6.90 ± 0.82b 10.98 ± 0.21b 4.30 ± 0.45a 0.08 ± 0.01a 77.75 ± 1.25b 7.23 ± 0.80[a]a 43.80 ± 0.80b

Light Phase 73.01 ± 1.53c 8.93 ± 0.22b 6.39 ± 0.32b 0.63 ± 0.06b 11.03 ± 4.38a 34.83 ± 0.59b 3.17 ± 0.50b

[a] This calorific value was estimated using an empirical [21]. Any two numbers a column followed by the same letters are not significantly different at 95% confidence level

The light phase contained significantly higher carbon than the medium phase (p-value<0.0001) and the heavy phase (p-value= 0.0251) (table VII). The differences in carbon content were consistent with the differences in calorific values of the three phases because high carbon content is associated with high calorific value. Most of the sulfur was retained in the light phase followed by heavy and medium phases (p-value <0.0001). Based on this analysis, it may be concluded that the light and heavy phases are good for energy production, and the medium phase may best be used for fertilizer or other applications.

D. Efficiency of the Pyrolysis Process

1) Energy Conversion Efficiency The char energy conversion efficiency and fixed carbon

yield for various treatments are listed in table VIII. The greatest energy conversion efficiency was recorded for the fine fraction from screen #20 at 49.85%. It was significantly greater than its corresponding coarse fraction (43.53%). Heavy and light phases of the condensate held 2.84% and 10.89% of the feedstock energy. Therefore, it can be concluded that pyrolysis products retained 57.26% of the total energy in the feedstock (coarse fraction of poultry litter from screen #20).

The condensate energy conversion efficiency was only calculated for the #20 coarse fraction condensate. The heavy, medium, and light phases captured 2.64%, 11.54%, and 10.89% of the total feedstock energy, respectively.

2) Fixed Carbon Yield A meaningful measure of carbonization efficiency is fixed

carbon yield. The fixed carbon yield represents the efficiency realized by the pyrolytic conversion of the ash-free organic matter in the feedstock into a pure ash-free carbon [22]. All the chars had the same fixed carbon yield as expected because fixed carbon is produced from decomposition of lignocellulosic biomass and screening did not significantly separate cellulose, lignin, and hemicelluloses [9]. The fixed carbon yields shown in table VIII are lower than those reported for chestnut (21.4%) and oak (29.5%) [22].

TABLE VIII. CHAR ENERGY CONVERSION EFFICIENCY AND FIXED CARBON YIELD RESULTING FROM PYROLYSIS OF POULTRY LITTER AFTER SCREENING

TREATMENT. THE VALUES ARE MEAN ± SE, n=3

Sl. No.

Treatment Energy Conversion

Efficiency (%)

Fixed Carbon Yield (%)

1 Raw 42.76 ± 1.64a 19.47 ± 0.50a

2 Screen #5 Coarse 42.47 ± 2.35a 19.07 ± 0.62a

3 Screen #5 Fine 44.15 ± 1.58a 19.61 ± 0.22a

4 Screen #10 Coarse 41.84 ± 2.39a 19.25 ± 0.51a 5 Screen #10 Fine 46.81 ± 0.23b 20.19 ± 0.57a

6 Screen #18 Coarse 42.92 ± 1.11a 19.38 ± 0.31a

7 Screen # 18 Fine 46.37 ± 1.28b 19.54 ± 0.74a

8 Screen #20 Coarse 43.53 ± 1.87a 19.98 ± 0.55a

9 Screen #20 Fine 49.85 ± 4.14b 19.27 ± 0.31a

Any two numbers a column followed by the same letters are not significantly different at 95% confidence level

E. Distribution of Carbon, Hydrogen, Nitrogen, and Sulfur among Pyrolysis Products Figure 4 gives a visual presentation of the distribution of

carbon, hydrogen, nitrogen, and sulfur among pyrolysis products: char, condensate (heavy, medium, and light phases) and gases presented as percentage of parent feedstock carbon, hydrogen, nitrogen, and sulfur (coarse fraction, screen # 20, 0.85 mm). Loss in gases was calculated by subtracting the other component (char, heavy, medium, and light phase) percentages from 100.

Char produced by pyrolysis of coarse fraction from screen#20 captured 47.71% of the feedstock carbon. Similarly, heavy, medium, and light phases captured 2.71%, 4.38%, and 9.07% of the feedstock carbon. Based on this analysis, it may be concluded that the pyrolysis products, over all, captured 53.86% of the feedstock carbon. The majority of sulfur was retained in the char (77.69%). Char held the maximum nitrogen (46.63%), followed by the medium phase of the condensate (33.16%).

If someone chooses land application of poultry litter char, then carbon in char will remain in the soil for longer period of time than carbon added through direct land application of poultry litter or its compost; hence, giving benefits of carbon sequestration [30].

Figure 4. Distribution of carbon, hydrogen, nitrogen, and sulfur among

pyrolysis products (char, condensate heavy, medium, and light phases, and gases) was presented as percentage of parent feedstock carbon, hydrogen,

nitrogen, and sulfur (coarse fraction, screen # 20, 0.85 mm)



IV. SUMMARY Based on this study, we recommend that either screen #18

or #20 can be used for screening of poultry litter. The pyrolysis of the #20 coarse fraction produced 44.47% char retaining 43.53% of total feedstock energy. Over all, the pyrolysis char and condensate produced from #20 coarse fraction captured 57.23% of total feedstock energy and 53.86% of the total feedstock carbon. The light phase condensate had calorific value 34.83MJ/kg and could be further refined as low grade fuel.

The conclusions drawn from this study may only apply to the poultry houses which grow broilers and use pine wood shavings as a bedding material. The results may significantly vary for industries using rice hulls or peanut hulls as a bedding material. Screening and pyrolysis could be a good option for value added utilization of poultry litter for energy production. The process produced energy products: light phase of bio-oil and char; and nutrient rich fine fraction and medium phase of condensates which could be transported to remote locations to remove nutrients from areas with nutrient imbalances. The medium phase of the condensate can be mixed with the nutrient rich fine fraction to produce fertilizer.

ACKNOWLEDGMENT The authors would like to thank the National Animal and

Poultry Waste Management Center, N.C. for funding this research work.

REFERENCES [1] P. M. Ndegwa, “Fractionation of poultry litter for enhanced utilization

and reduction of environmental pollution,”. MS thesis. Clarke, GA: The University of Georgia, Department of Biological and Agricultural Engineering, 1990.

[2] A. Coloma, “Treatment of poultry litter by screening,” Unpublished M. S. thesis, Clemson University, Clemson, SC, USA, 2005.

[3] M. Lopez, T. A. Costello, and L. A. Roe, “Combustion of poultry litter for space heating,” ASAE Paper No. 058019, St. Josheph, Mich.: ASAE, 2005.

[4] K. M.Keener, R. Shook, K. Anderson, and C. R. Parrish,“Characterization of poultry manure for potential co-combustion with coal in an electricity generation plant,” ASAE Paper No. 026058, St. Joseph, Mich.: ASAE, 2002.

[5] J.Jones, A,“From waste to Energy- Catalytic steam gasification of broiler litter,” Unpublished M. S. Thesis, The University of Tennessee, Knoxville, TN, USA, 1998.

[6] A. C.Sheth, and A. D. Turner, “Kinetics and economics of catalytic steam gasification of broiler litter,”Transactions of ASAE, vol. 45(4), pp 1111-1121, 2002.

[7] A. C.Sheth, and B. Bagchi, “Investigation of nitrogen-bearing species in catalytic steam gasification of poultry litter,”Journal of Air &Waste Management Association,vol. 55, pp. 619-628, 2005.

[8] S. Kim, F.A Agblevor, “Pyrolysis characteristics and kinetics of chicken litter,”Waste Management, vol. 27, pp. 135-140, 2007.

[9] K. Singh, M. Risse, J. Worley, K. C. Das, S. Thompson, “Effect of Fractionation on Fuel Properties of Poultry Litter,”Applied Engineering in Agriculture, vol.24(3), pp. 383-388, 2008.

[10] I. M.Lima, and W. E. Marshall, “Granular activated carbons from broiler manure: physical, chemical and adsorptive properties,”Bioresource Technology,vol. 96, pp. 699-706, 2005.

[11] D.Mohan, C. U. Pittman Jr., and P. H. Steele, “Pyrolysis of wood/biomass for bio-oil: A critical review,”Energy and Fuelsvol. 20, pp. 848-889, 2006.

[12] K.Raveendran, A. Ganesh, and K. C. Khilar, “Influence of mineral matter on biomass pyrolysis characteristics,”Fuel, vol. 74(12), pp. 1812-1822, 1995.

[13] K.Raveendran, A. Ganesh, and K. C. Khilar, “ Pyrolysis characteristics of biomass and biomass components,”Fuel, vol. 75(8), pp. 987-998, 1996.

[14] M. E. Bernhart and O. Fasina, “Physical properties and pyrolysis behavior of fractionated poultry litter,”Transactions of ASABE,vol. 52(2), pp. 531-538, 2009.

[15] N.Whitely, R. Ozao, R. Artiaga, Y. Cao, and P. Wei-Ping, “Multi-Utilization of chicken litter as Biomass source. Part 1. Pyrolysis,”Energy and Fuels,vol. 20(6), pp. 2666-2671, 2006.

[16] EPA-821-B-04-006, “Managing manure nutrients at concentrated animal feeding operations,” Washington, D. C.: USEPA Office of Water (4303T), 2004.

[17] P.M.Ndegwa, S. A. Thompson, and W. C. Merka, “Fractionation of Poultry Litter for Enhanced Utilization,”Transactions of ASABE,vol. 34(3), pp. 992-997, 1991.

[18] A.Coloma, J. P. Chastain, and K. P. Moore, “Fractionation of Solids, Plant Nutrients, and Carbon as a Result of Screening Broiler Litter,” ASABE Paper No. 044086. St. Joseph, Mich.: ASABE, 2004.

[19] ASTM Standards,“D 5865: Standard test method for gross calorific value of coal or coke,” W. Conshohocken, PA: ASTM, 2003.

[20] ASTM Standards, “D 5142: Standard test methods for proximate analysis of the analysis sample of coal and coke by Instrumental Procedures,” W. Conshohocken, PA: ASTM, 2004.

[21] S. A.Channiwala and P. P. Parikh, “A unified correlation for estimating HHV of solid, liquid and gaseous fuels,”Fuel,vol. 81, pp. 1051-1063, 2002.

[22] M. J.Antal and M. Gronli, “The art, science, and technology of charcoal production,”Industrial & Engineering Chemistry Research,vol. 42(8), pp. 1619-1640, 2003.

[23] M.Garcia-Perez, T. T. Adams, J. W. Goodrum, D. P. Geller, and K. C. Das, “Production and fuel properties of pine chip bio-oil/biodiesel blends,”Energy & Fuels,vol. 21 (4), pp. 2363–2372, 2007.

[24] K. Singh, E. W. Tollner, S. Mani, L. M. Risse, K. C. Das, and John Worley, “Transforming solid wastes into high quality bioenergy products: Entropy analysis,” ASME Paper No. 1924. Peer reviewed proceedings, North American Waste to Energy Conference (NAWTEC-16), Philadelphia, PA, 2008.

[25] J. Z.Dávaloa, M. V. Roux, and P. Jiménez, “Evaluation of poultry litter as a feasible fuel,”Thermochemia Acta vol. 394, pp. 261-266, 2002.

[26] S.Mukhtar, K. Annamalai, B. Thien and S. C. Porter, “Co-firing of coal and Broiler litter (BL) for power 261 generation: BL fuel quality and characteristics,” ASAE Paper No. 024189. St. Joseph, Mich.: ASAE, 2002.

[27] S.Priyadarsan, K. Annamalai, J. M. Sweeten, S. Mukhtar, and M. T. Holtzapple, “Fixed-bed gasification of feedlot manure and poultry litter biomass,”Transaction of ASAE, vol. 47(5), pp. 1689-1696, 2004.

[28] M. I.Schnitzer, C. M. Monreal, G. A. Facey, and P. B. Fransham, “The conversion of chicken manure to biooil by fast pyrolysis 1. Analysis of chicken manure, biooils, and char by 13C and 1H NMR and FTIR Spectrophotometry,”Journal of Environmental Science and Health, vol. 42, pp. 71-77, 2007.

[29] A. Demirbaş,“Relationships between lignin contents and heating values of the biomass,”Energy Conversion and Management,vol.42(2), pp. 183‐188, 2001.

[30] J. Gaskin,W., C. Steiner, K. Harris, K. C. Das, and B. Bibens, “Effect of low-temperature pyrolysis conditions on biochar for agricultural use,”Transactions of the ASABE,vol. 51(6), pp. 2061-2069, 2008.

201406031353105198

Documents

Transcript of 201406031353105198