Storage stability of biocrude oils from fast pyrolysis of poultry litter

Waste Management 32 (2012) 67–76

Contents lists available at SciVerse ScienceDirect

Waste Management journal homepage: www.elsevier.com/locate/wasman

Storage stability of biocrude oils from fast pyrolysis of poultry litter Ofei D. Mante ⇑, Foster A. Agblevor Biological Systems Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA

a r t i c l e

i n f o

Article history: Received 1 March 2011 Accepted 1 September 2011 Available online 2 October 2011 Keywords: Fast pyrolysis Poultry litter Aging of biocrude oil Wood shavings

a b s t r a c t The unstable nature of biocrude oils produced from conventional pyrolysis of biomass is one of the properties that limits its application. In the disposal of poultry litter via pyrolysis technology, the biocrude oil produced as a value-added product can be used for on farm applications. In this study, we investigated the influence of bedding material (wood shavings) on the storage stability of biocrude oils produced from the fast pyrolysis of poultry litter. The biocrude oils produced from manure, wood (pine and oak), and mixtures of manure and wood in proportions (75:25 50:50, and 25:75 w/w%) were stored under ambient conditions in sealed glass vials for a period of 6 months and their stability were monitored by measuring the changes in viscosity over time. The manure oil had the lowest rate of viscosity change and thus was relatively the most stable and the oils from the 50:50 w/w% litter mixtures were the least stable. The rate of viscosity change of the manure biocrude oil was 1.33 cP/day and that of the 50/50 litter mixture was 7.6 cP/day for pine and 4.17 cP/day for oak. The spectrometric analyses of the biocrude oils showed that the presence of highly reactive oxygenated functionalities in the oil were responsible for the instability characteristic of the litter biocrude oils. The poor stability of the biocrude oil from the 50:50 w/w% litter mixtures was attributed to reactions between nitrogenous compounds (amides) from protein degradation and oxygenated compounds from the decomposition of polysaccharides and lignin. The addition of 10% methanol and 10% ethanol to the oil from 50% manure and 50% pine reduced the initial viscosity of the oil and was also beneficial in slowing down the rate of viscosity change during storage. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Poultry litter is a valuable waste of the poultry industry if handled properly. It is traditionally used as a source of plant nutrients to substitute for inorganic fertilizers in agriculture and also as a protein source in animal feed. However, recent federal and state regulations on the use of poultry litter due to environmental issues related to over application of the litter on farm lands and as animal feed supplement (Bitzer and Sims, 1988; Herson et al., 2004; Kelleher et al., 2002; Virginia Department of Environmental Quality, 1998) limits the traditional utilization of poultry litter especially in areas with concentrated poultry production. Currently, alternative methods are being investigated for the safe and economical disposal of poultry litter. Pyrolysis technology is one of the thermochemical techniques that have been shown to be efficient for the disposal of the litter (Agblevor et al., 2010; Kim et al., 2009; Mante and Agblevor, 2010). Pyrolysis is considered as a dual tool for the production of valuable products from poultry litter and a solution to environmental waste disposal. Fast pyrolysis of poultry litter at

⇑ Corresponding author. Tel.: +1 5404491980. E-mail address: [email protected] (O.D. Mante). 0956-053X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2011.09.004

moderate temperatures (450–500 °C) converts the organic portion into biocrude oil and syngas whilst the inorganic fraction is trapped in the bio-char. The biocrude oil could potentially be used for on farm heating of poultry houses and the biochar which contains both macro and micro plant nutrients could be used as a slow-release organic fertilizer and soil amendment (Agblevor et al., 2010). Other studies have also shown poultry litter bio-char can be processed further into activated carbon (Koutcheiko et al., 2007; Lima and Marshall, 2005a–c; Lima et al., 2008; Qiu and Guo, 2010). The biocrude oil produced from poultry litter is generally higher in energy density (26–30 MJ/kg) and less acidic (pH values between 3.9–6.3) (Agblevor et al., 2010; Mante and Agblevor, 2010) relative to other biocrude oils produced from conventional pyrolysis of woody biomass. Nonetheless, poultry litter biocrude oils like any other biocrude oil from woody biomass are viscous and unstable compared to petroleum fuels (Agblevor et al., 2010; Kim et al., 2009). Typically, biocrude oils produced by conventional pyrolysis increases in viscosity when stored or undergoes solidification to form a thick glue-like material when exposed to air (Oasmaa and Peacocke, 2001). The presence of reactive oxygenates (Bridgwater and Bridge, 1991) and char particles in the biocrude oil (Agblevor et al., 1998) affect the stability characteristic. Most studies have

68

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76

shown that reactions such as polymerization and condensation occur during storage (Chaala et al., 2004; Diebold, 1999; Oasmaa and Kuoppala, 2003; Oasmaa et al., 2004). A detailed overview of the chemical and physical mechanisms of the storage stability of various biocrude oils by Diebold (Diebold, 1999) suggested that the main reactions responsible for biocrude oil aging were reactions of aldehydes leading to the formation of hydrates, esters, acetal, resin, heavy compounds and water. It is well understood that the properties of biocrude oils from biomass are functions of feedstock, pyrolysis operating conditions, reactor configuration, condensation system, and storage conditions (Czernik et al., 1994; Diebold, 1999). The effect of biocrude oil composition, storage conditions and the addition of solvents on the storage stability have been reviewed in literature (Agblevor and Besler, 1996; Ba et al., 2004a,b; Boucher et al., 2000a,b; Chaala et al., 2004; Czernik et al., 1994; Diebold, 1999; Diebold and Czernik, 1997; Garcia-Perez et al., 2006) but none have been reported on poultry litter biocrude oils. The presence of wood shavings (bedding material) in poultry litter influences both the chemical composition and the physical property of the pyrolysis derived biocrude oil. In our previous studies (Agblevor et al., 2010; Mante and Agblevor, 2010), we showed that the pyrolysis of different poultry litters produced biocrude oils with different properties. The biocrude oil produced from starter turkey litter (containing pine shavings) and chicken litters (containing hardwood shavings) were reported to have unusual variations in their viscosities. The variation in the properties were attributed to the manure and bedding material content in the poultry litter. Since fast pyrolysis involves only the partial decomposition of biomass, the property of the resulting biocrude oil is feedstock dependent and hence biocrude oils from poultry litter with different amount of manure and bedding material would potentially store differently. The fundamental understanding of how poultry litter compositions affect the stability characteristic is necessary in the production of biocrude oil that is more stable during shipping and storage. Thus, in this paper we report on the storage stability of various biocrude oils derived from poultry litter containing different amount and type of bedding material. 2. Materials and methods 2.1. Materials Most poultry operations raise birds with 2–6-in. layer of wood shavings or other bedding material as an absorptive base. The litter is removed once or more times a year. Consequently, the composition of the resulting litter is mainly dependent on the manure build up and the amount of bedding material removed with the manure. Due to variability in poultry litter, mixtures of known amount of manure and bedding materials were prepared. Layer manure, pine, and oak shavings were obtained from poultry growers in the Shenandoah Valley, VA. The wood shavings and manure were air dried at ambient conditions to equilibrium moisture content (moisture content ranged from 8% to 10%) and ground in a Wiley mill to pass a 1-mm mesh screen. Mixtures of manure and wood shavings (pine and oak) were prepared in a blender. In both type of mixtures, the manure and wood were mixed in the following ratios; 100:0, 75:25, 50:50, 25:75 and 0:100 w/w. The elemental analysis and the higher heating value (HHV) of the manure, pine and oak wood materials are shown in Table 1. 2.2. Fast pyrolysis of poultry litter Each feedstock; manure, wood (pine and oak), and mixtures of manure and wood in proportions (75:25 50:50, and 25:75 w/w%)

Table 1 Sample characterization (layer manure, pine and oak). Elemental composition (wt%)b

C H N O S Cl HHV (MJ/kg) Ash (wt%)a Bulk density (kg/L) a b

Feedstock Layer manure

Pine

Oak

29.15 4.13 6.42 36.56 0.36 0.62 14.79 23.53 0.41

50.03 5.64 <0.5 41.88 <0.05 180 ppm 19.58 1.95 0.24

49.52 6.06 <0.5 43.23 <0.1 0.032 19.50 1.05 0.23

ASTM E1755, Standard test method for ash in biomass. Moisture free basis.

were pyrolyzed at 450 °C with 18 L/min of N2 and a feed rate of 320 g/h in a bench-scale bubbling fluidized bed unit located at the Biological Systems Engineering Bioresource Lab, Virginia Tech, VA, USA (Fig 1). The unit comprised of a K-Tron volumetric feeder (K-Tron Process Group, Pitman, NJ), a 50 mm bubbling fluidized bed reactor equipped with a 100 lm porous metal gas distributor, hot gas filter, two chilled water condensers, an electrostatic precipitator and a packed column. The reactor was externally heated with a three-zone electric furnace (Thermcraft, Winston-Salem, NC). The reactor temperatures were measured and controlled by three K-thermocouples inserted into a thermal well in the reactor. The feedstocks were fed continuously with 6 L/min of nitrogen gas through a jacketed air-cooled feeder tube into the fluidized bed. During pyrolysis, the mixture of char, gases and vapors that exited the reactor were separated by a hot gas filter maintained at 380 °C. The separated gases and vapors were then passed through two condensers connected in series. The condensers were maintained at 8 °C with a 50/50 cooling mixture of ethylene glycol and water from an 18-liter refrigerated circulating bath (Haake, Karlsruhe, W. Germany). Any condensable gases and aerosols that escaped from the condenser were captured by an electrostatic precipitator (ESP) kept at 16–20 kV and a packed column of glass beads. The temperatures across the reactor, hot gas filter and the condensers were controlled and or monitored using an Omega multiscan (1200) acquisition system with TempView 2.1 program. Pressure drop across the hot gas filter was monitored by a pressure gauge. The non-condensable gases were sampled at intervals and were analyzed by a SRI Multiple Gas Analyzer#2 (Model 8610C, SRI Instruments) with two packed columns. The molecular Sieve 13 column separates H2, CH4 and CO. All the compounds in the C1– C6 range were separated by the Hayesep-D column and the gases were determined using a flame ionization detector (FID) equipped with a methanizer. The GC oven temperature was programmed to maintain 50 °C for 8 min after injection, followed by a 20 °C/min ramp to 200 °C and holding at 200 °C for 25 min. The gas chromatogram was processes using PeakSimple 3.67 program. The GC was calibrated with a standard gas mixture consisting of CO, CO2, CH4, C2H2, C2H4, and C2H6 in nitrogen balance (Supelco, Bellefonte, PA). Silica sand (100 g) was used as the pyrolysis medium and the reactor temperature, feed rate, and gas flow rate were kept constant for each run.

2.3. Storage stability test The stability test was performed on the oil fraction from the ESP since they contained mostly organics and less amount of water. The aging of the oils were conducted with about 30 g of sample in 50 mL glass vials tightly closed with a plastic cap. In the first test, all the raw biocrude oil samples were stored on a closed

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76

69

Fig. 1. Schematic diagram of the fluidized bed reactor unit (1 – fluidized bed, 2 – furnace, 3 – thermocouple, 4 – mass flow controller, 5 – jacketed air-cooled feeder tube, 6 – hopper, 7 – screw feeder, 8 – computer, 9 – heating tape, 10 – hot gas filter, 11 – reservoir, 12 – condenser, 13 – ESP, 14 – AC power supply, 15 – filter, 16 – wet gas meter, 17 – gas chromatograph).

cabinet in the lab at ambient conditions for 6 months. Another set of tests were conducted specifically on the bio oil from the mixture of 50% manure and 50% pine. 10 wt% each of methanol and ethanol was added to the bio oil and the samples were stored for 6 month at room temperature to assess its role in the stability of the oil. The viscosity measurements of all the samples were initially taken every week for the first month and then monthly for the rest of the storage period. 2.4. Biocrude oil characterization The physical properties and chemical composition of the biocrude oils were determined. The viscosity of the freshly produced biocrude oils and the stored oils were measured with a Brookfield DV-II + Pro viscometer (Brookfield Engineering laboratories Inc. MA, USA). The measurements were performed at 60 °C with Spindle SC4-18 and speed ranging from 1 to 30 rpm depending on the viscosity of the oil. The temperature was automatically maintained at the desired value with a Brookfield thermosel unit. Sample volume of 7 mL was used and the viscosity readings of the instrument were allowed to stabilize within 5–15 min before the value was recorded. The pH of the pyrolysis oils was measured using a Corning pH Meter 440 equipped with F-55500-10 Accumet pH probe (ColeParmer Instrument Company, Vernon Hills, Illinois, USA). The pH data were obtained after 10 min stabilization of the mechanically stirred oil. The densities of the oils were determined at 23 °C using a Mettler Toledo DA-110 M density meter (Greifensee, Switzerland) according to ASTM D4052. Calibrations were done prior to measurements with distilled water free of bubbles. The oil sample was introduced into an oscillating sample tube and the density was calculated from measured resonance frequency. The values were

reported to three decimal places in g/cm3. Volumetric Karl Fischer titration was used to measure the moisture content of the biocrude oils. A Metrohm 701KF Titrino (Brinkmann Instruments, Inc., NY, USA) and HydranalÒ Composite 5 reagent was used. About 1 g of oil sample was titrated in a solvent mixture of 40 mL of methanol and 20 mL of toluene. The 13C NMR spectra of the pyrolysis oils were recorded on a Varian Unity 400 MHz NMR spectrometer. The observing frequency for the 13C nucleus was 100.58 MHz. Approximately 2.0 g of oil was dissolved in 2.5 mL of deuterated dimethyl sulfoxide (d-DMSO) and 0.25 mL of tetramethylsilane (TMS) reference were placed in a 10 mm probe. The 1H NMR spectra were also recorded on a Varian Unity 400 MHz NMR spectrometer. The observing frequency for the 1H was 400.0 MHz. The oil was weighed to 400 mg and was dissolved in 0.5 mL of dimethyl sulfoxide (d-DMSO) and 0.05 mL of tetramethylsilane (TMS) was added as a reference and spectrum was collected in a 5-mm probe. Fourier-transform infrared (FTIR) analysis was performed on a Nicolet IR spectrometer (Nicolet Avatar, 370 DTGS). Approximately 50 mg of each sample was used on Zinc Selenide crystal material. The spectra were obtained over a range of 4000–650 cm 1 using 64 scans at a 4 cm 1 resolution and a background gain of 4.0 was used. Elemental composition (C, H, N, O, S, and Cl), ash and HHV of the biocrude oils (pine, manure and litter mixtures) collected from the ESP were determined by Galbraith Analytical Laboratory (Knoxville, TN, USA). Thermogravimetric analysis was conducted using a TA Instruments Q600 SDT. Sample (20 mg) was placed in an alumina crucible and was subjected to thermogravimetric analysis with 20 mL/min of N2 as a carrier gas. The heating rate was at 5 °C/ min from 25 °C to 700 °C. All the tests were repeated three times to ensure good reproducibility of the results.

70

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76

3. Results and discussion

3.2. Properties of biocrude oil

3.1. Pyrolysis products

The liquid fractions from the pyrolysis experiment were collected in the condenser and the ESP. The aqueous phase collected from the condenser was light yellow in color and contained about 35–75 wt% of water. The ESP fraction (biocrude oil) was heavy, dark brown and had moisture content of about 3–5 wt%. The summary of the physical properties of the various biocrude oils are shown in Table 2. The physical properties of the oils differed as a result of the litter compositions. The pH decreased from 5.89 to 3.72 and the density increased from 1.14 to 1.22 g/cm3 as oak wood content in the litter increased to 75 wt%. The higher heating value (HHV) of the oils generally decreased with increased wood content, however the HHV values for both the 75/25 and the 50/ 50 mixtures were about the same. The decrease in pH values, HHV and increase in density of the biocrude oil as oak content in the litter mixture increased was attributed to the increased amount of oxygenated functionalities from the decomposition of wood as will be shown later. The kinematic viscosity of the oils was influenced by the amount of wood, the viscosity increased from 130 to 225 cSt when 50 wt% of oak was added but decreased to 145.5 cSt when the oak content in the litter was 75%. A comparison between the properties of the oils obtained from the litter containing pine and those that had oak shavings suggested that both the amount and type of wood shavings in the litter may influence the physical properties. The graphs of HHV, pH, density and viscosity in Fig. 3 evidently show a difference in oak and pine for the biocrude oil especially produced from the litter mixture containing 75/25 wt% of manure/wood. The 75/25 wt% litter mixtures from oak had a pH of 5.31, a density of 1.16 g/cm3 and a viscosity of 144 cSt compared to a pH of 4.47, density of 1.19 g/ cm3, and viscosity of 224 cSt for the pine mixture. The biocrude oil from pine mixture had HHV value of 28.9 MJ/kg compared with 26.82 MJ/kg from the oak mixture.

The mass of biochar was determined gravimetrically by weighing the hot gas filter and the reactor before and after each pyrolysis experiment. The total mass of biocrude oil was also determined gravimetrically by weighing the condensers and electrostatic precipitator before and after each experiment. The yields were expressed in percentage on a moisture free basis. The total yield of the non-condensable gas was calculated by difference. The product distribution from the pyrolysis of manure and the litter mixtures containing oak wood as bedding material are shown in Fig. 2. The product yields and the properties of biocrude oils from the litter mixtures containing pine shavings were reported by Mante and Agblevor (2010) and will not be discussed here. From the pyrolysis of the litter mixtures containing oak shavings (Fig. 2), it appeared that more biocrude oil and less char were produced when compared with mixtures of manure and pine shavings (Mante and Agblevor, 2010). In the pyrolysis of oak wood, biocrude oil yield of 64.5 wt% and a char yield of 16.7 wt% were achieved. The biocrude oil yield from the pyrolysis of pine was 62.0 wt% with char yield of 22.9 wt%. The higher biocrude oil yield reported for oak litter mixtures relative to pine litter mixtures can be explained from the elemental composition analysis (see Table 1). The oak shaving used was slightly higher in oxygen (O) and hydrogen (H) than pine. Besides, the ash content of pine (1.95 wt%) was almost twice that of oak shavings (1.05 wt%). Thus, it can be inferred that the oak wood contained a little more organic fraction than pine which probably lead to higher biocrude oil yields and lower char yields. Generally, oak and pine wood shavings exhibited similar effect on the product distribution of the pyrolysis of the litter mixtures. An increase in oak wood in the litter increased the overall biocrude yield and decreased the char yield as was reported by Mante and Agblevor (2010) with the use of pine wood. The biocrude oil yield from pure manure was 43.3 wt% and increased to 55.05 wt% when the litter had 75 wt% of oak and the char yield decreased from 43.1 wt% to 23.78 wt%. The lower oil yield, higher char yield and lower gas yield from the pyrolysis of the litter containing higher amount of manure can be attributed to the higher inorganic content due to high ash content of the manure. The ash content of the manure was 23.5 wt% compared to 1.05 wt% in oak wood.

Fig. 2. Effect of oak wood content on the pyrolysis product distribution.

3.3. Spectrometric analysis The spectrometric analysis by 13C NMR, 1H NMR and FT-IR were used to identify changes in the functional groups of the biocrude oils as the poultry litter composition were varied. All the techniques were consistent and complimented the observation made by each technique. The biocrude from manure was found to be rich in aliphatic hydrocarbons, amides and N-heterocyclic compounds (see Tables 3 and 4). It suggests therefore that, the proteins in manure underwent fragmentation during pyrolysis giving rise to nitrogenous compounds and hydrocarbon chains. The wood biocrude oils on the other hand were rich in carbohydrate sugars (levoglucosan), phenolics, aromatic ethers, carboxylic acids, aldehydes and ketones due to the degradation of carbohydrate and lignin. The 13C NMR analysis of the biocrude oils from the litters containing varying amount of manure and oak wood showed that increase in oak content in the litter mixture consequently increased the intensity of methoxyl group (56 ppm carbon), levoglucosan (65, 72, 74, 77 and 108 ppm carbon), aromatic carbons (105– 160 ppm) and the overall carbonyl carbons (180–210 ppm). However, the amide carbonyl signal at 174.8 ppm decreased as the oak content in the litter increased. This resulted in a strong carboxylic signal at 173 ppm which was attributed to acetic acid. Furthermore, the carbons showing methylene in protein (30 ppm) and saturated n-alkyl groups (5–35 ppm) decreased as the oak content increased. The general increase in decomposition product signals for both carbohydrate and lignin and the gradual decrease in the intensity of signals from protein decomposition products as wood content in the litter increased suggested that the change in the chemical composition was largely as a result of dilution effect

71

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76 Table 2 Physical properties of manure and oak wood bio oils. Physical property

Analysis method

pH Density @ 23 °C (g/m3) Kinematic viscosity, at 60 °C (cSt) Heating value (MJ/kg)

pH meter ASTM D4052 Rotational viscometer ASTM D5865

Biocrude oils (ESP) 100% M

M:O (75/25)

M:O (50/50)

M:O (25/75)

100% O

5.93 1.14 130.0 29.7

5.31 1.16 143.9 26.82

4.31 1.21 226 26.88

3.72 1.22 145.5 25.09

3.31 1.26 99 24.00

M = Manure, O = Oak wood.

Fig. 3. Comparison of the physical properties between ESP biocrude oils produced from oak and pine litter mixtures.

Table 3 NMR results of the oils from manure, pine wood and their mixtures. Chemical shift, d (ppm)

Relative intensity 100% Manure 1

Hydrogen type Aliphatic protons in CH3, CH2, & CH (further from an aromatic ring) Aliphatic protons in CH3, CH2, & CH (Attached to an aromatic ring) Protons in methoxyl and hydroxyl Amide, N2H groups Aromatic protons, conjugated olefins

1.6–0.5 3.3–1.6 5.5–3.3 6.1–5.7 9.0–6.0

Carbon type Short aliphatic hydrocarbons Methylene carbons of saturated n-alkyl groups Total aliphatic carbons Methoxyl (–OCH3) in lignin Carbohydrate sugars, alcohols, ethers Aromatic C, C in N-heterocyclic & C in heteroaromatic C@O groups (Amides, carboxylic acids and derivatives) Aldehydes, ketones

28–5 38–28 5–38 55–57 105–60 160–105 180–160 180–210

M:P (75/25)

M:P (50/50)

H NMR of biocrude oil (percentage hydrogen 24.30 22.13 18.31 23.54 24.39 24.38 16.71 21.12 27.80 24.07 15.53 9.96 11.39 16.77 19.35

M:P (25/75)

100% Pine

total) 16.85 24.12 29.15 7.27 22.42

10.36 20.78 41.97 2.94 21.31

C NMR of biocrude oil (percentage carbon total) 29.53 19.01 15.82 15.92 21.07 17.74 12.87 11.14 50.6 36.75 28.69 27.04 2.11 7.41 9.00 9.08 2.91 20.34 26.26 28.98 39.54 30.86 32.11 30.97 3.93 4.18 3.09 2.96 0.21 0.44 0.85 0.97

12.13 8.26 20.41 9.45 34.03 32.31 2.60 1.20

13

M = Manure, P = Pine wood.

and to a lesser extent chemical reactions between the decomposition products of protein, carbohydrate and lignin. 3.4. Storage stability of biocrude oils Aging of biocrude oil is described as the increase in viscosity with time. The biocrude oils age when stored and their physical property change undesirably. The most significant physical property that changed during storage of the biocrude oils was viscosity. Thus, in this study we monitored the changes in viscosity of the biocrude oils to investigate how poultry litter composition influences oil storage stability. The changes in viscosities during the

storage period of the biocrude oils from the litter mixtures containing pine and those containing oak shavings are shown in Figs. 3a and 3b respectively. It appeared that all the biocrude oils aged differently and their rate of viscosity change was higher within the first month but did not show any significant reduction in the rate of viscosity change at the end of the storage time. This suggests that possible reactions responsible for the aging of the biocrude oils were still apparent during the 6 months of storage. The biocrude oil from wood free litter (manure) had the lowest rate of viscosity change and thus was relatively the most stable whilst the 50/50 litter mixture (both pine and oak) were the least stable (see Fig. 4a). The rate of viscosity change of the manure biocrude

72

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76

Table 4 NMR results of the oils from manure, oak wood and their mixtures. Chemical shift, d (ppm)

Relative intensity 100% Manure

M:O (75/ 25)

M:O (50/ 50)

M:O (25/ 75)

100% Oak

1

Hydrogen type Aliphatic protons in CH3, CH2, & CH (further from an aromatic ring) Aliphatic protons in CH3, CH2, & CH (Attached to an aromatic ring) Protons in methoxyl and hydroxyl Amide, N2H groups Aromatic protons, conjugated olefins

1.6–0.5 3.3–1.6 5.5–3.3 6.1–5.7 9.0–6.0

Carbon type Short aliphatic hydrocarbons Methylene carbons of saturated n-alkyl groups Total aliphatic carbons Methoxyl (–OCH3) in lignin Carbohydrate sugars, alcohols, ethers Aromatic C, C in N-heterocyclic & C in heteroaromatic C@O groups (Amides, carboxylic acids and derivatives) Aldehydes, ketones

28–5 38–28 5–38 55–57 105–60 160–105 180–160 180–210

H NMR of biocrude oil (percentage hydrogen total) 24.30 22.69 17.32 13.09 23.54 25.22 24.85 23.63 16.71 19.93 26.90 34.94 24.07 10.44 6.82 4.75 11.39 21.59 24.07 23.42

8.91 23.36 45.71 3.70 17.82

13

C NMR of biocrude oil (percentage carbon total) 29.53 20.32 14.54 12.61 21.07 14.89 11.31 9.64 50.6 35.21 25.85 22.24 2.11 4.16 5.98 7.16 2.91 10.66 17.18 21.21 39.54 45.62 46.93 44.64 3.93 3.54 3.42 3.20 0.21 0.64 0.81 1.55

10.23 8.50 18.73 8.0 27.90 40.19 3.36 1.81

M = Manure, O = Oak wood.

Fig. 3a. Aging of biocrude oils produced from litter mixtures containing pine wood.

Fig. 4a. Rate of viscosity change of biocrude oils.

Fig. 3b. Aging of biocrude oils produced from litter mixtures containing oak wood.

oil was 1.33 cP/day and that of the 50/50 litter mixture was 7.6 cP/ day for pine and 4.17 cP/day for oak. Furthermore, the biocrude oil from wood shavings also aged at a higher rate but it is worth mentioning that the oak oil aged at a relatively slower rate when compared to the pine biocrude oil. The pine biocrude oil aged at a rate of 5.67 cP/day and oak biocrude oil aged at 3.55 cP/day.

The relative stability of the biocrude from manure can be attributed to the presence of less oxygenated functionalities like aldehydes, ketones, carboxylic acids, anhydrosugars and phenolics which are major decomposition products of wood. These compounds are known to be reactive and participate in aging reactions (Bridgwater and Bridge, 1991; Diebold, 1999). Even more, levoglucosan produced from the pyrolysis of cellulose (Adam et al., 2005; Carlson et al., 2010; Shafizadeh et al., 1979; Shen and Gu, 2009; Wolfrom et al., 1959) could potentially contribute to the aging of biocrude oil. At room temperature, levoglucosan solidifies over time. This affects both viscosity and stability of the oil when stored. Nokkosmäki et al. (2000) reported that the removal of sugars from biocrude oils can improve the stability of the oil significantly. Nevertheless, the underlying mechanism of levoglucosan reactivity during aging is still under investigation. It appeared therefore that the presence of hydrocarbons and nitrogenous compounds coupled with the absence of reactive oxygenated contributed to the stability of the manure biocrude oil. Furthermore, unlike other carbonyl functionalities (aldehydes and ketones), amide carbonyl is more stable and less reactive due to the presence of a lone pair of electrons on the nitrogen which interacts with the carbonyl p-bond

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76

(Ege, 2003). This may also be a possible contributing factor to the stability of the manure oil. From the stability test of the biocrude oils, it appeared that the presence of wood in the litter caused the biocrude oil to be less stable. As previously mentioned, biocrude oils produced from the 50/ 50 wt% litter mixture of manure and wood had the highest rate of viscosity change during the six months storage period. The unusual high rate of viscosity change could be due to a number of reasons. Firstly, the pyrolysis of the litter mixtures containing substantial amount of both wood and manure produces reactive aldehydes from wood degradation and amide groups from protein degradation. The aldehydes are known to undergo condensation reactions with amides to give polymeric compounds which eventually increase the viscosity and the average molecular weight (Acharya and Manning, 1983; Noyes and Forman, 1933), thus it appeared that crosslinking of the protein fragments was highly favored in the litter mixture of 50/50 manure and wood. Secondly, reaction between lignin and protein decomposition products might have contributed to the unstable nature of the litter biocrude oil. Studies have shown that lignin degradation compounds interact with proteins by the formation of hydrogen bonds between lignin reactive hydroxyl groups and protein carbonyl groups to form protein–polyphenol cross-linked network (Kunanopparat et al., 2009; Pouteau et al., 2003; Zahedifar et al., 2002). Lastly, the nitrogen compounds in the biocrude could serve as free radicals to catalyze olefinic condensation reactions to form polyolefins (Diebold, 1999) and hence increase the viscosity of the oil. Since we are not sure of the underlying mechanism, the above explanations need to be confirmed by more systematic experiments. It is worth noting that the rates of viscosity change of the stored biocrude oils from oak mixtures were lower than those from pine for the same storage period. Nonetheless, the elemental composition of the biocrude oils obtained from the litter with 50% wood (oak and pine) did not show any difference in their C, H, N, O, and S contents (Table 5). Thus, the difference in the stability characteristic between pine and oak on the aging of the oils may be attributed to differences between hardwood and softwood pyrolytic lignin. The lignin in hardwood (oak) are more of syringols (two methoxy groups) and that of softwood (pine) are mainly guaiacols (one methoxy) (Hon and Shiraishi, 1990; Pandey, 1999). Also, the average molecular weight (Mw) of softwood pyrolytic lignin is larger than that of hardwood. The softwood pyrolytic lignins are known to have a larger tendency to polymerize, because of the existence of unsubstituted positions ortho to the phenol hydroxyl in softwood lignins (Pandey, 1999; Scholze et al., 2001; Scholze and Meier, 2001). Therefore, it appeared that the biocrude oils from pine and litter mixtures were more reactive because most of the pyrolytic lignin (guaiacol) had free C5 position for polymerization reactions. The presence of dimethoxy phenols (syringols) in oak wood litter mixtures is likely a contributing factor in the relatively

73

low rate of viscosity change compared to biocrude oils from pine. Hence, it can be inferred that poultry litters containing hardwood produces biocrude oils that store better than litters with softwood.

3.5. Effect of solvent additives The influence of solvent additives on the storage stability of the biocrude oil was examined. Ten percentage (w/w) each of methanol and ethanol were added to the least stable biocrude oil produced (50% manure and 50% of pine wood). The viscosity of the biocrude oil was reduced by a factor of 6.4 and 5.12 upon the addition of 10% methanol and 10% ethanol respectively. In comparison with the raw biocrude oil, both solvents effectively retarded the rate of change in viscosity of the oil for a storage period of 6 months. The raw oil aged at 7.6 cP/day compared to 0.53 cP/ day and 0.66 cP/day for the oils which had 10% ethanol and methanol respectively (Fig. 4b). The result is in agreement with other studies reported in literature on the stability of wood derived bio-oils. Oasmaa et al. investigated the effect of solvent on hardwood bio oil by adding 2 wt%, 5%, 10%, and 20% ethanol (Oasmaa et al., 1997). They found that the addition of 20% ethanol at 50 °C decreased the initial viscosity of the bio-oil from 50 to 10 cSt and also decreased the rate of viscosity change from 0.12 to 0.01 cSt/day during the aging studies. Diebold and Czernik also investigated stabilizing hardwood bio-oil by adding 10 wt% methanol, ethanol, acetone, ethyl acetate, a 1/1 mixture of acetone and methanol, or a 1/1 mixture of methanol and methyl iso-butyl ketone (Diebold and Czernik, 1997). They reported that the solvents had a dramatic impact on slowing the aging rate and viscosity increase of the bio oil. Another work by Boucher et al. examined the effect of methanol and pyrolytic aqueous phase on the stability and aging of bio oils at 40, 50, and 80 °C (Boucher et al., 2000b). They found that the addition of methanol to the biooil improved its properties and increased its stability. The increase in viscosity of the stored raw biocrude oil obviously suggests the occurrence of chemical reactions during storage. However, specific reactions were not easily identified from our study considering the complexity of the mixture of compounds present in the oil from the decomposition of wood and manure. Nonetheless, the 13C NMR spectra of the stored oils for both raw and solvent added oils in Fig. 5 showed clearly that the carbonyl signals between 200 ppm and 210 ppm decreased for the stored raw biocrude oil. Conversely, the biocrude stored with methanol and ethanol did not show such reduction in the carbonyl region

Table 5 Elemental composition of 50% manure & 50% wood biocrude oils (moisture free basis). Elemental (wt%)

C H N O S Cl H/C molar ratio O/C molar ratio N/C molar ratio Ash

50% Manure & 50% wood biocrude oil (ESP) Pine

Oak

63.71 6.93 3.02 26.16 0.16 155 ppm 1.31 0.31 0.04 <0.06

64.36 7.05 2.58 25.82 0.16 0.032 1.31 0.30 0.034 <0.1

Fig. 4b. Effect of 10% solvent addition on the rate of viscosity change of biocrude oil.

74

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76

A

B

C

D

200

150

100

50

0

PPM

13

Fig. 5. C NMR spectra of biocrude oils produced from litter mixture of 50% manure and 50% pine wood. (A) – biocrude oil stored with 10% ethanol; (B) – biocrude oil stored with 10% methanol; (C) – fresh raw biocrude oil; (D) – stored raw biocrude oil.

(200–210 ppm). This suggests therefore that the alcohols prevented reactions involving aldehydes and ketones which would have otherwise occurred. Similar observation was made by Oas-

maa (Oasmaa et al., 2004). They suggested that the alcohol additives protect aldehydes and ketones from further reactions likely due to acetal formation. Other studies have shown that pyrolysis

0.4 Stored raw biocrude oil Biocrude oil stored with 10% ethanol Biocrude oil stored with 10% methanol Fresh biocrude oil

Deriv. Weight (%/°C)

0.3

0.2

0.1

0.0 0

100

200

300

400

Temperature (°C)

500

600

700 Universal V4.7A TA Instruments

Fig. 6. DTA plots of fresh and stored biocrude oil from the pyrolysis of litter mixture containing 50% manure and 50% pine wood. Key – fresh biocrude oil: red, stored biocrude oil: black, biocrude oil stored with 10% ethanol: blue, biocrude oil with 10% methanol: green. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76

oil stored with methanol promotes esterification reaction as well as slows the polymerization of aromatics by dilution and making fewer reactive sites available for polymerization reactions at any one time. The role of solvent in slowing aging by preventing the formation of larger molecules during storage in reacting with active groups like acids and aldehydes is still unclear. The differential thermogravimetry (DTG) plots of the biocrude oils stored with and without solvent are shown in Fig. 6. It can be seen that the addition of 10% of solvent especially in the case of ethanol clearly reduced the decomposition rate of the well defined peaks at 200 and 350 °C. The effect was attributed to the fact that, methanol and ethanol are good bio oil solvents and their addition to the oil dissolved some of the structured high molecular weight component resulting in a diluted system with low viscosity and hence the reduction in the decomposition peaks at 220, 350 °C. However, the stored raw biocrude oil and those with 10% methanol showed some similarities. The stored raw oil had a reduction in the rate of decomposition of the peak at 350 °C to about the same rate for both oils stored with solvents. It is therefore suggestive that during aging, chemical components in the raw oil corresponding to peak at 350 °C underwent some chemical reactions that consequently increased the viscosity during storage. It can be inferred that the use of solvents as additives mainly slows down the formation of compounds that leads to viscosity increase. 4. Conclusions The biocrude oil from manure which was high in nitrogenous compounds and hydrocarbons was the most stable. The presence of oxygenated compounds from the decomposition of wood in the litters made the oils relatively unstable. The reactions of proteins with reactive carbonyls (aldehydes) and pyrolytic lignin seem to contribute to the unstable nature of the biocrude oils. The addition of solvent to decrease the rate of aging appears to involve molecular dilution to slow the chemical reactions and the formation of intermediate products. In conclusion, the elimination of specific oxygenated groups of bio-oils derived from the mixture of manure and wood is necessary for improving their stability. The aging of the oils from the litter mixtures were dependent on the amount and the type of wood (pine or oak). Generally, the oils produced from the litter mixture with oak stored better than those produced from the litter mixture with pine. Acknowledgements The National Fish and Wildlife Foundation, Farm Pilot Projects Coordination Inc, Virginia Poultry Federation, Shenandoah Resource and Conservation Council, Chesapeake Bay Foundation and Bluemoon Fund are all acknowledged for their financial support of this work. References Acharya, A.S., Manning, J.M., 1983. Reaction of glycolaldehyde with proteins: latent crosslinking potential of alpha-hydroxyaldehydes. Proceedings of the National Academy of Sciences of the United States of America 80, 3590–3594. Adam, J., Blazsó, M., Mészáros, E., Stöcker, M., Nilsen, M.H., Bouzga, A., Hustad, J.E., Grønli, M., Øye, G., 2005. Pyrolysis of biomass in the presence of Al-MCM-41 type catalysts. Fuel 84, 1494–1502. Agblevor, F.A., Beis, S., Kim, S.S., Tarrant, R., Mante, N.O., 2010. Biocrude oils from the fast pyrolysis of poultry litter and hardwood. Waste Management 30, 298– 307. Agblevor, F.A., Besler, S., 1996. Inorganic compounds in biomass feedstocks. 1. Effect on the quality of fast pyrolysis oils. Energy & Fuels 10, 293–298. Agblevor, F.A., Scahill, J., Johnson, D.K., 1998. Pyrolysis char catalyzed destabilization of biocrude oils. AIChE Symposium Series 319, 146–150. Ba, T., Chaala, A., Garcia-Perez, M., Rodrigue, D., Roy, C., 2004a. Colloidal properties of bio-oils obtained by vacuum pyrolysis of softwood bark. Characterization of water-soluble and water-insoluble fractions. Energy & Fuels 18, 704–712.

75

Ba, T., Chaala, A., Garcia-Perez, M., Roy, C., 2004b. Colloidal properties of bio-oils obtained by vacuum pyrolysis of softwood bark. Storage stability. Energy & Fuels 18, 188–201. Bitzer, C.C., Sims, J.T., 1988. Estimating the availability of nitrogen in poultry manure through laboratory and field studies. Journal of Environmental Quality 17, 47–54. Boucher, M.E., Chaala, A., Pakdel, H., Roy, C., 2000a. Bio-oils obtained by vacuum pyrolysis of softwood bark as a liquid fuel for gas turbines. Part II: Stability and ageing of bio-oil and its blends with methanol and a pyrolytic aqueous phase. Biomass and Bioenergy 19, 351–361. Boucher, M.E., Chaala, A., Roy, C., 2000b. Bio-oils obtained by vacuum pyrolysis of softwood bark as a liquid fuel for gas turbines. Part I: Properties of bio-oil and its blends with methanol and a pyrolytic aqueous phase. Biomass and Bioenergy 19, 337–350. Bridgwater, A.V., Bridge, S.A., 1991. Biomass Pyrolysis Liquids, Upgrading and Utilisation. Elsevier, London. Carlson, T.R., Jae, J., Lin, Y.-C., Tompsett, G.A., Huber, G.W., 2010. Catalytic fast pyrolysis of glucose with HZSM-5: the combined homogeneous and heterogeneous reactions. Journal of Catalysis 270, 110–124. Chaala, A., Ba, T., Garcia-Perez, M., Roy, C., 2004. Colloidal properties of bio-oils obtained by vacuum pyrolysis of softwood bark: aging and thermal stability. Energy & Fuels 18, 1535–1542. Czernik, S., Johnson, D.K., Black, S., 1994. Stability of wood fast pyrolysis oil. Biomass and Bioenergy 7, 187–192. Diebold, J.P., 1999. A review of the chemical and physical mechanisms of the storage stability of fast pyrolysis bio-oils, Other Information: PBD: 27 Jan 1999. Diebold, J.P., Czernik, S., 1997. Additives to lower and stabilize the viscosity of pyrolysis oils during storage. Energy & Fuels 11, 1081–1091. Ege, S.N., 2003. Organic Chemistry: Structure and Reactivity, fifth ed. Houghton Mifflin Company, Boston. Garcia-Perez, M., Chaala, A., Pakdel, H., Kretschmer, D., Rodrigue, D., Roy, C., 2006. Evaluation of the influence of stainless steel and copper on the aging process of bio-oil. Energy & Fuels 20, 786–795. Herson, M., Dubberly, D., Kelly, F., 2004. Poultry litter feeding ban: implications of the 2nd interim rule for animal feeding. Florida Cooperative Extension Service. Hon, D.N.-S., Shiraishi, N., 1990. Wood and Cellulosic Chemistry. Marcel Dekker Inc., New York, NY. Kelleher, B.P., Leahy, J.J., Henihan, A.M., O’Dwyer, T.F., Sutton, D., Leahy, M.J., 2002. Advances in poultry litter disposal technology – a review. Bioresource Technology 83, 27–36. Kim, S.-S., Agblevor, F.A., Lim, J., 2009. Fast pyrolysis of chicken litter and turkey litter in a fluidized bed reactor. Journal of Industrial and Engineering Chemistry 15, 247–252. Koutcheiko, S., Monreal, C.M., Kodama, H., McCracken, T., Kotlyar, L., 2007. Preparation and characterization of activated carbon derived from the thermo-chemical conversion of chicken manure. Bioresource Technology 98, 2459–2464. Kunanopparat, T., Menut, P., Morel, M.-H.l.n., Guilbert, S.p., 2009. Modification of the wheat gluten network by kraft lignin addition. Journal of Agricultural and Food Chemistry 57, 8526–8533. Lima, I., Marshall, W.E., 2005a. Utilization of turkey manure as granular activated carbon: Physical, chemical and adsorptive properties. Waste Management 25, 726–732. Lima, I.M., Marshall, W.E., 2005b. Adsorption of selected environmentally important metals by poultry manure-based granular activated carbons. Journal of Chemical Technology and Biotechnology 80, 1054–1061. Lima, I.M., Marshall, W.E., 2005c. Granular activated carbons from broiler manure: physical, chemical and adsorptive properties. Bioresource Technology 96, 699– 706. Lima, I.M., McAloon, A., Boateng, A.A., 2008. Activated carbon from broiler litter: process description and cost of production. Biomass and Bioenergy 32, 568– 572. Mante, O.D., Agblevor, F.A., 2010. Influence of pine wood shavings on the pyrolysis of poultry litter. Waste Management 30, 2537–2547. Nokkosmäki, M.I., Kuoppala, E.T., Leppämäki, E.A., Krause, A.O.I., 2000. Catalytic conversion of biomass pyrolysis vapours with zinc oxide. Journal of Analytical and Applied Pyrolysis 55, 119–131. Noyes, W.A., Forman, D.B., 1933. aldehyde–amide condensation. I. Reactions between aldehydes and acetamide. Journal of the American Chemical Society 55, 3493–3494. Oasmaa, A., Kuoppala, E., 2003. Fast pyrolysis of forestry residue. 3. Storage stability of liquid fuel. Energy & Fuels 17, 1075–1084. Oasmaa, A., Kuoppala, E., Selin, J.F., Gust, S., Solantausta, Y., 2004. Fast pyrolysis of forestry residue and pine. 4. Improvement of the product quality by solvent addition. Energy & Fuels 18, 1578–1583. Oasmaa, A., Leppamaki, E., Koponen, P., Levander, J., Tapola, E., 1997. Physical characterisation of biomass-based pyrolysis liquids. Application of standard fuel oil analyses. Technical Research Centre of Finland, Espoo. Oasmaa, A., Peacocke, C., 2001. A guide to physical property characterisation of biomass derived fast pyrolysis liquids. Technical Research Centre of Finland, Espoo. Pandey, K.K., 1999. A study of chemical structure of soft and hardwood and wood polymers by FTIR spectroscopy. Journal of Applied Polymer Science 71, 1969– 1975. Pouteau, C., Dole, P., Cathala, B., Averous, L., Boquillon, N., 2003. Antioxidant properties of lignin in polypropylene. Polymer Degradation and Stability 81, 9–18.

76

O.D. Mante, F.A. Agblevor / Waste Management 32 (2012) 67–76

Qiu, G., Guo, M., 2010. Quality of poultry litter-derived granular activated carbon. Bioresource Technology 101, 379–386. Scholze, B., Hanser, C., Meier, D., 2001. Characterization of the water-insoluble fraction from fast pyrolysis liquids (pyrolytic lignin): Part II. GPC, carbonyl goups, and 13C-NMR. Journal of Analytical and Applied Pyrolysis 58–59, 387– 400. Scholze, B., Meier, D., 2001. Characterization of the water-insoluble fraction from pyrolysis oil (pyrolytic lignin). Part I. PY-GC/MS, FTIR, and functional groups. Journal of Analytical and Applied Pyrolysis 60, 41–54. Shafizadeh, F., Furneaux, R.H., Cochran, T.G., Scholl, J.P., Sakai, Y., 1979. Production of levoglucosan and glucose from pyrolysis of cellulosic materials. Journal of Applied Polymer Science 23, 3525–3539.

Shen, D.K., Gu, S., 2009. The mechanism for thermal decomposition of cellulose and its main products. Bioresource Technology 100, 6496–6504. Virginia Department of Environmental Quality, 1998, 1998. Poultry Waste Management. Report for the Senate Committee on Agriculture, Conservation and Natural Resource. Wolfrom, M.L., Thompson, A., Ward, R.B., 1959. The composition of pyrodextrins. II. Thermal polymerization of levoglucosan1. Journal of the American Chemical Society 81, 4623–4625. Zahedifar, M., Castro, F.B., Ørskov, E.R., 2002. Effect of hydrolytic lignin on formation of protein-lignin complexes and protein degradation by rumen microbes. Animal Feed Science and Technology 95, 83–92.