- Email: [email protected]
Methane productivity of manure, straw and solid fractions of manure
Available online at www.sciencedirect.com
Biomass and Bioenergy 26 (2004) 485 – 495
Methane productivity of manure, straw and solid fractions of manure H.B. M$llera;∗ , S.G. Sommera , B.K. Ahringb a Department
of Agricultural Engineering, Danish Institute of Agricultural Sciences, Research Center Bygholm, P.O. Box 536, DK-8700 Horsens, Denmark b Biocentrum-DTU, The Technical University of Denmark, DK-2800 Lyngby, Denmark
Received 24 February 2003; received in revised form 9 July 2003; accepted 25 August 2003
Abstract The methane productivity of manure in terms of volatile solids (VS), volume and livestock production was determined. The theoretical methane productivity is higher in pig (516 l kg−1 VS) and sow (530 l kg−1 VS) manure than in dairy cattle manure (469 l kg−1 VS), while the ultimate methane yield in terms of VS is considerably higher in pig (356 l kg−1 VS) and sow manure (275 l kg−1 VS) than in dairy cattle manure (148 l kg−1 VS). Methane productivity based on livestock units (LU) shows the lowest methane productivity for sows (165 m3 CH4 LU−1 ), while the other animal categories are in the same range (282–301 m3 CH4 LU−1 ). Pre-treatment of manure by separation is a way of making fractions of the manure that have a higher gas potential per volume. Theoretical methane potential and biodegradability of three types of fractions deriving from manure separation were tested. The volumetric methane yield of straw was found to be higher than the yield from total manure and the solid fractions of manure, due to the higher VS content, and hence the use of straw as bedding material will increase the volumetric as well as the livestock-based methane productivity. ? 2003 Elsevier Ltd. All rights reserved. Keywords: Methane yield; Anaerobic digestion; Manure; Straw; Manure separation
1. Introduction Anaerobic degradation of animal manure results in the production of biogas, which is composed mainly of methane and carbon dioxide [1]. Natural degradation of manure, leading to emissions of CH4 during storage, is very undesirable because of the global warming e>ects resulting from the release of greenhouse gases [2,3]. However, by using controlled anaerobic
∗ Corresponding author. Tel.: +45-7629-6043; fax: +45-76296100. E-mail address: [email protected] (H.B. M$ller).
0961-9534/$ - see front matter ? 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2003.08.008
digestion of animal manure, where the biogas is captured, the CH4 emissions during storage of manure can be reduced and the energy from manure can be used as a substitute for fossil fuels, thereby serving as a CO2 -neutral energy source [4]. The methane potential of manure comes from the digestion of the organic components in the faeces and in the straw used as bedding material, which is mainly: carbohydrates, proteins and lipids. Methane yield can be expressed in several ways [5]. The term ‘methane productivity’ is used to indicate the yield of methane per unit of a variable. Methane productivity can be measured in terms of volatile solids (VS) destroyed, VS loaded, volume, or animal production. Methane productivity measured
486
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
in terms of VS destroyed (l CH4 kg−1 VSdes ) corresponds to the theoretical methane yield (Bu ) if there is complete degradation of all organic components of the manure. The theoretical methane potential can be calculated from Bushwell’s formula [6]. Methane productivity in terms of VS loaded (l CH4 kg−1 VSload ) as residence time approaches inHnity is referred to as the ultimate methane yield (Bo ). The ultimate methane yield will always be lower than the theoretical yield because a fraction of the substrate is used to synthesize bacterial mass, a fraction of the organic material will be lost in the eJuent, and lignin-containing compounds will only be degraded to a limited degree [7]. Inhibition of the biological process by inhibitors such as ammonia and volatile fatty acids (VFA) is another factor contributing to the actual methane yield being lower than the potential yield which would be obtained if inhibition was not present [8]. It has been observed that both the ultimate methane yield (Bo ) and the volumetric methane production (l CH4 m−3 manure) of manure from di>erent origins can be very variable. The ultimate methane yield (l CH4 kg−1 VS) is affected by various factors, including: 1. 2. 3. 4.
species, breed and growth stage of the animals [1], feed [1], amount and type of bedding material [1], degradation processes during pre-storage [9].
The volumetric methane production (l CH4 m−3 manure) is a>ected by the same factors as the ultimate methane yield, but is also inLuenced by the amount of wasted drinking water and water added to the manure collection system. Both the ultimate methane yield and the volumetric methane production are of great importance for the economy of biogas plants. The amount of biogas that can be produced is proportional to the ultimate methane yield, while the volumetric methane production a>ects the necessary digester volume and hence the overall economics of a digestion facility. However, both the ultimate methane yield and the volumetric methane production are only concerned with the concentration and ‘quality’ of VS in the manure, rather than the methane yield that can be produced from a given animal production system. Methane productivity in terms of animal production can be determined by starting with the ultimate methane yield and applying an animal waste production factor. In
several countries, including Denmark, the term livestock unit (LU) is applied to describe the animal production factor and it is this productivity which is most useful in determining economic feasibility [5], since units represent the volume of methane per animal (m3 CH4 LU−1 ). The ultimate methane yield that can be produced from a given animal is determined by both the amount and the quality of the VS excreted. Often the VS excretion value from a given animal production system is not available and since feed intake levels, feedstu> quality and other factors di>er, the best estimate of VS excretion will be production data on feed intake and digestibility of the feed [10]. The volumetric methane production of manure can be increased by solid–liquid separation [11], thereby producing a fraction with a high VS concentration. However, a part of the VS will always be present in the liquid fraction, and if only the solid fraction is used as a substrate for anaerobic digestion, it is extremely important that the separation unit is eOcient in transferring VS to the solid fraction. M$ller et al. [12] found that chemical precipitation combined with screening and centrifugation was an eOcient method for transferring VS to the solid fraction. 2. Materials and methods 2.1. Gas production The ultimate methane yield (Bo ) was determined in a batch experiment. The experiments were performed in 1100-ml infusion bottles. The bottles were closed with butyl rubber stoppers, sealed with aluminium crimps and incubated at 35 ± 0:5◦ C, after Lushing with N2 . The method used is described in the international standard ISO 11734 [13]. The test medium was faeces taken directly after excretion from the pigs and cattle, wheat straw, and solid fractions of manure produced by centrifugation or by chemical precipitation (see Table 1 for details). The test medium was frozen until used in the batch experiment. Inoculum from a farm-scale biogas plant, which had been kept for 2 weeks before the test at 35◦ C to remove most of the remaining methane production, was used. The inoculum:substrate ratio is shown in Table 2. Each test medium was digested in triplicate. The gas produced by controls was subtracted from the actual gas
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
487
Table 1 Characteristics of the pig, cattle, solid fraction from manure and straw used in this study Manure No.
Manure/biomass type
Feed stu>
1 2 3 4 5 6 7
Cattle, Cattle, Cattle, Cattle, Cattle, Cattle, Cattle,
8
Pig fatteners, faeces
9 10 11
Pig fatteners, faeces Pig fatteners, faeces Pig fatteners, faeces
12 13
Pig fatteners, faeces Pig fatteners, slurry
14
Pig fatteners, faeces
15
Sows, faeces
16
Sows, faeces
17
Sows, faeces
18 19 20 21 22
Solid fraction (pig)—centrifugation Solid fraction (pig)—centrifugation Solid fraction (pig)—chemical treatment Evaporation concentrate Wheat straw
Hay from clover and minerals Concentrates and maize silage Concentrates and hay from alfalfa 19.6% concentrates, 80% roughage, 0.4% minerals 22.6% concentrates, 77% roughage, 0.4% minerals 21.6% concentrates, 55.2% roughage, 22.8% barley, and 0.4% minerals 52% roughage (maize and clovergrass silage), 47% concentrates and 0.4% minerals 24% soybean meal, 49.8% barley, 20% wheat, 3% molasses and fat, 3.2% minerals 7.5% soybean meal, 16.5% barley/wheat, 75% whey permeate, 1% fat 77% soybean meal, 20.2% barley/wheat, 3% concentrate 24% soybean meal, 49.8% barley, 20% wheat, 3% molasses and fat, 3.2% minerals No data available 24% soybean meal, 49.8% barley, 20% wheat, 3% molasses and fat, 3.2% minerals 24% soybean meal, 49.8% barley, 20% wheat, 3% molasses and fat, 3.2% minerals Protein concentrates 2%, barley/oat 22,7%, whey permeate 74.5%, animal fat 1% 13% soybean meal, 39% barley, 38% wheat, 3% fat, 6% protein concentrates, 3.2% minerals Dry feed: 8.5% soybean meal, 88.5% wheat/barley, 3% concentrates. Wet fed: 100% whey No data available No data available No data available No data available
dairy dairy dairy dairy dairy dairy dairy
faeces faeces faeces slurry slurry slurry faeces
produced through digestion of the medium. The volume of gas produced was calculated either by measuring pressure in the headspace [13], or by connecting the infusion vessel to a gas collection bag and using a large syringe to measure the volume of gas, as described by Steed and Hashimoto [3]. The two methods were compared and the di>erence was found to be negligible (¡ 1%), indicating that both methods were applicable. The gas samples were analysed for CO2 and CH4 with gas chromatography. CH4 was measured on a Hewlett Packard (5890, series II) gas chromatograph with an electron capture detector and a Lame ionization detector (FID). Methane was isolated on a 1:83 m × 3:1 mm column with Poropak N 80/100, the carrier gas was He at 30 ml min−1 and the temperatures of injection port, oven and detector were 110◦ C, 40◦ C and 270◦ C, respectively. CO2
was measured on a Varian 3350 gas chromatograph (1 m × 3:1 mm column with a 2 m × 3:1 mm Haysep R 80/100 molecular sieve) equipped with a thermal conductivity detector. The carrier gas was He at a Low rate of 30 ml min−1 , the temperatures of oven and detector were 40◦ C and 150◦ C, respectively. 2.2. Manure analysis The total nitrogen content (TN) was analysed by the Kjeldahl method and a Kjellfoss 16200 (Copenhagen, Denmark). The dry matter (DM) content was determined after a 24-h drying period at 105◦ C. The ammonium (NH+ 4 ) in the manure was analysed by means of a QuikChem 4200 Low injection analyzer (Lachat Instruments, Milwaukee, WI, USA). The chemical oxygen demand (COD) was analysed colorimetrically
488
Table 2 Composition, theoretical and ultimate methane yields Cattle
No. 2
No. 3
No. 4a
No. 5a
No. 6a
No. 7
Avg. Cattle
143.75 902.70 0.91 11.90 11.19 76.09 70.90 555.57 155.42 130.83 258.40 0.7 100 464 0.21
138.34 899.50 1.70 24.50 30.82 158.42 53.70 489.02 193.33 74.71 241.10 0.7 124 460 0.27
156.44 859.90 1.40 24.00 18.85 164.26 68.96 318.80 252.24 176.88 369.00 0.7 — 470 —
104.04 796.80 25.70 51.40 82.53 201.59 80.82 372.13 161.90 101.03 209.50 0.7 — 480 —
98.10 938.60 — — — — — — — — — 0.7 150 469 0.32
105.10 930.00 — — — — — — — — — 0.7 207(±5:9) 469 0.44
111.93 939.10 — — — — — — — — — 0.7 161(±2:9) 469 0.34
122.53 895.23 7.43 27.95 35.85 150.09 68.59 433.88 190.72 120.86 269.50 0.7 148(±41) 469 0.32
Sows
Pig, fatteners
DM incl. VFA (g/l) VS (g/kg DM) NH+ 4 -N (g/kg TS) Total N (g/kg TS) VFA (g/kg VS) Proteina (g/kg VS) Lipids (g/kg VS) Degradable carbohydrates (g/kg VS) Non degradable carbohydrates (g/kg VS) Lignin Crude Hbre (g/kg VS) Inoculum/substrate (I:S) ratio Bo (L CH4 /g VS) Bu (l CH4 /g VS) Biodegradability (Bo =Bu )
No. 8
No. 9
No. 10
No. 11
No. 12
No. 13a
No. 14
Avg. pig
No. 15
No. 16
No. 17
Avg. sows
206.69 821.00 5.00 39.30 78.32 261.11 152.01 352.40
233.40 849.00 5.90 42.20 64.57 267.23 130.98 357.48
238.00 803.50 5.50 44.00 67.98 299.47 110.64 285.44
292.19 868.30 8.00 34.70 62.62 192.19 137.97 349.82
237.99 850.20 5.40 29.30 78.52 175.69 170.67 344.47
95.50 836.50 46.10 70.10 80.00 179.32 120.50 393.04
260.00 912.10 — — — — — —
223.39 848.66 12.65 43.27 72.00 229.17 137.13 347.11
318.93 819.30 2.40 31.20 31.12 219.70 159.77 401.20
296.78 810.10 1.40 23.80 27.76 172.82 178.37 389.10
317.30 745.50 0.00 25.50 29.59 213.78 151.58 377.01
311.00 791.63 1.27 26.83 29.49 202.10 163.24 389.10
112.06
136.40
191.47
206.96
169.61
176.93
—
165.57
129.50
156.65
158.03
148.06
44.09 128.20 1.1 359
43.35 152.60 1.1 403
45.00 190.00 1.1 329
50.44 223.50 1.1 —
61.04 196.10 1.1 340
50.21 190.00 1.1 —
— —
58.71 154.20 4.2 250
75.30 187.90 4.2 260
70.00 170.00 4.2 317
528
516
507
514
532
502
49.02 180.07 1.1 356 (±28) 516 (±11) 0.69
529
536
524
68.00 170.70 4.2 275 (±36) 530 (±6) 0.52
0.67
0.78
0.65
—
0.62
—
1.1 358 (±8) — 0.69
0.47
0.48
0.60
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
DM incl. VFA (g/l) VS (g/kg DM) NH+ 4 -N (g/kg TS) Total N (g/kg TS) VFA (g/kg VS) Proteina (g/kg VS) Lipids (g/kg VS) Degradable carbohydrates (g/kg VS) Non degradable carbohydrates (g/kg VS) Lignin Crude Hbre (g/kg VS) Inoculum/substrate (I:S) ratio Bo (l CH4 /g VS) Bu (l CH4 /g VS) Biodegradability (Bo =Bu )
No. 1
13.32 0.00 7.3 506(±15) 542 0.93
905.40 959.00 0.00 5.90 0.00 38.45 23.57 501.49 358.00 78.50 418.60 0.7 195(±5:9) 432 0.45 285.00 812.00 14.00 44.50 0.00 234.76 130.67 440.73 103.08 90.76 157.40 0.7 247(±25) 514 0.48
2.3. Separation technologies The following types of separation technologies were used to produce solid fractions with a high VS concentration from the manure: • Decanting centrifuge produced by Alfa Laval, NX 309B-31 (R$dovre, Denmark) • Chemical treatment system, AnsagerSepTec (Ansager, Denmark) • Evaporation of water from the liquid fraction pre-treated with a decanting centrifuge (Funki Manura, S$nderborg, Denmark).
SD in brackets. a Slurry.
DM incl. VFA (g/l) VS (g/kg DM) NH+ 4 -N (g/kg TS) Total N (g/kg TS) VFA (g/kg VS) Proteina (g/kg VS) Lipids (g/kg VS) Degradable carbohydrates (g/kg VS) Non degradable carbohydrates (g/kg VS) Lignin Crude Hbre (g/kg VS) Inoculum:substrate (I:S) ratio Bo (l CH4 /g VS) Bu (l CH4 /g VS) Biodegradability (Bo =Bu )
489
(Spektroquant Nova 60; Merck, Germany) in accordance with a method described by the American Public Health Association [14]. VFA C2 –C5 concentration was determined by means of a gas chromatograph (Hewlett Packard 6850A) with a FID. The column was an HP-INNOWax, 30 m × 0:25 mm × 0:25 m. The carrier gas was He. The temperature of the column was gradually increased from 110◦ C to 220◦ C at a rate of 10◦ C min−1 . The lipid content was analyzed by measuring the amount of material, that can be extracted with diethyl ether in a soxleth extraction instrument after hydrolysis with 3 N hydrochloric acid. The lignin content was analyzed by determination of the suspended VS after boiling with a detergent in 1 mol=l sulfuric acid followed by suspending in 72% sulfuric acid. The protein content was determined by multiplying the di>erence between TN and NH4 -N with the factor 6.25. The crude Hbers were analyzed by determination of the suspended VS after boiling with 0:26 mol=l sulfuric acid and 0:23 mol=l potassium hydroxide dissolution.
290.60 703.20 15.60 42.80 14.68 241.75 119.60 400.14 119.03 104.81 157.40 4.8 261(±10) 508 0.51
309.00 807.40 26.80 47.10 11.69 180.43 150.88 433.16 119.03 104.81 157.40 0.7 159(±39) 521 0.30
198.00 637.91 71.40 102.30 158.35 302.75 170.87 354.71
No. 22 No. 20 No. 19 No. 18
Solid fraction from pig manure
No. 21
Straw
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
The decanter centrifuge consisted of a drum adjusted to rotate at a speed of 5600 rpm, giving a pressure of 4100g at the drum circumference; the rotational speed of the screw was 60 –80 rpm slower than the speed of the drum. For details see M$ller et al. [11]. The AnsagerSepTec separator consisted of three Locculation chambers with mixing and addition of chemicals to form new chemical compounds and to alter the physical state of dissolved and suspended solids and to facilitate their removal by a belt press. For details see M$ller et al. [12].
490
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
The Funki Manura technology uses only mechanical–thermal techniques. The separation process is divided into three main steps: mechanical pre-treatment (decanting centrifuge), evaporation and distillate cleaning. The pre-treated liquid is pumped to the evaporation unit, where it is degassed after an initial heating to about 100◦ C. During degassing, ammonia and carbon dioxide are removed. The degassed manure is pumped to the evaporation–heat exchanger. In the evaporation–heat exchanger the slurry is heated to more than 100◦ C. To avoid fouling the heat exchanger, the slurry is circulated with a high Low on the heated surface. By this process most of the water and the remaining ammonia evaporates. The remaining concentrate contains a high concentration of VS and nutrients. Evaporated water and ammonia are removed from the evaporation unit and are condensed. 2.4. Calculations The theoretical CH4 production (Bu ) was calculated from Bushwell’s formula [6]: a b H2 O C n Ha Ob + n − − 4 2 n a b n a b − + CO2 + + − CH4 ; (1) → 2 8 4 2 8 4 (n=2 + a=8 − b=4)22:4 (2) l CH4 kg−1 VS: 12n + a + 16b The average chemical composition of the VS in the manure was calculated with the assumption of an average composition of VSlipid (C57 H104 O6 ), VSprotein (C5 H7 O2 N), VScarbohydrate (C6 H10 O5 ), VSlignin (C6 H10 O5 ) and VSVFA (C2 H4 O2 ). The carbohydrates were divided into an easily degradable part (VSE; carbohydrate ) and a slowly degradable part(VSS; carbohydrate ). The degradable part of the carbohydrates was calculated by Eq. (4) and the slowly degradable part was calculated by Eq. (3): Bu =
VSS; carbohydrate = VScrudeHbre − VSlignin
(3)
VSE; carbohydrate = VS − VSprotein − VSlipid − VSVFA − VSS; carbohydrate
(4)
The theoretical CH4 production (Bu ) from VFA (acetic acid), carbohydrates, protein and lipid is
370, 415, 496 and 1014 l CH4 kg−1 VS, respectively (Eq. (2)). The ultimate methane yield from a given animal production unit can be calculated by using standard values for the excretion of VS from the given animal category and multiplying by Bo and number of animals, but it is more accurate to use data on feed intake and digestibility of the feed, to calculate the excretion of VS (Eq. (5)). VSexcretion i
= Intake(FU)i × DMi =FUi ×(1 − DEi %=100) × (1 − ASHi %=100);
(5)
where VSexcretion is the amount of VS excreted on an animal or farm level, Intake(FU)i the feed intake in feed units (FU) of feedstu> (i) on an animal or farm level, DMi =FUi the amount of dry matter (DM) per feed unit (FU) of feedstu> (i), DEi % the digestibility of feed DM of feedstu> (i), and ASHi % the ash content of the manure. 3. Results and discussion 3.1. Methane productivity of manure The chemical characteristics of the manures are given in Table 2. The VS were divided into four main groups, VSVFA ; VSlipid , VSprotein , VSlignin and VScarbohydrate . The carbohydrates were further divided into an easily degradable part (VSE; carbohydrate ) and a slowly degradable part (VSS; carbohydrate ) (Fig. 1). The average proportion of slowly degradable carbohydrates and lignin is larger in cattle than in pig manure (Fig. 1). This is as expected, since cattle are fed roughage which contains greater amounts of lignin complexed with cellulose than pig feed does. Furthermore the amounts of proteins and lipids are considerably higher in pig than cattle manure. Due to the higher proportion of lipid in pig manure, the theoretical CH4 production in the present study is higher in both pig (516 ± 11 l kg−1 VS) and sow (530 ± 6 l kg−1 VS) manure than in cattle manure (468 ± 6 l kg−1 VS). Hill [5] found that methane productivity per unit VS destroyed was relatively constant for all waste types (500 l kg−1 VS). Ianotti et al. [15] found almost the
90% 80% Lignin
60%
Carbohydrates (S) Carbohydrates (E)
50%
Lipid
40%
Protein VFA
30% 20% 10% 0% pig, fattening
sows
SFC
SFD
E-conc.
500 400 300 200 100 0 Cattle
Pig
Sow
SFD
SFC
straw
Fig. 1. The average composition of VS in freshly excreted cattle manure, fattening pig manure, sow manure, solid fraction from centrifugation of pig manure (SFD), solid fraction from chemical precipitation of manure (SFC), liquid fraction pretreated with a decanting centrifuge (E-conc.) and wheat straw.
same percentages of lipids (14.8%) and lignin (4.4%) but a slightly higher protein content (19.4%) in terms of VS in pig manure than we found in the present study. There is only a small di>erence in the VS composition of the manure from sows and fattening pigs (Table 2). The lignin and lipid contents are higher in sow manure, and thus the theoretical CH4 production (Bu ) is slightly but not signiHcantly (p ¿ 0:05, t-test) higher from sows than from fattening pigs (Fig. 2). The ultimate methane yield of dairy cattle manure (148 ± 41 l kg−1 VS) is signiHcantly (p ¡ 0:01, t-test) lower than the Bo of pig manure (356 ± 28 l kg−1 VS) and sow manure (275±36 l kg−1 VS). The variation between manure from individual cows is much higher than the variation between individual pigs, which may be explained by the fact that there are more di>erences in feeding practice and productivity between dairy farms than between pig farms; hence cows fed on only roughage give lower yields than cows fed on both roughage and concentrates. Furthermore, the plateau phase during batch digestion is reached much faster with pig manure than with cattle manure, indicating that the retention time seems to be longer during digestion of cattle manure. The Bo of manure from cows and pigs has been determined in several other studies [1,5,16,17]. IPCC [10] estimated the Bo of dairy cattle and pig manure in developed countries to be 240 l kg−1 VS and 450 l kg−1 VS,
E- Straw conc.
200 (l CH4 kg-1 product)
cattle
Volumetric methane yield
Ratio of VS
70%
491
600
(l CH4 kgVS-1)
100%
Theoretical and ultimate methane yield
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
167
150 100 50
12
19
10
Cattle
Pig
Sow
50
55
SFD
SFC
66
0 E- Straw conc.
Fig. 2. The theoretical Bu ( ) and ultimate methane yield Bo () in terms of VS and volume from cattle, pig manure, physical fractions of manure and straw. The DM content in manure has been assumed to be 10.3%, 6.6% and 4.5% in, respectively, cattle, pig and sow manure [26]. The VS:DM ratio in manure has been estimated as 0.8. The DM content in the fractions from manure has been assumed to be the same as in Table 2.
respectively, which is at the upper end of the Bo values determined in this study. Hill [5] estimated the Bo of dairy cattle and pig manure to be 131 l kg−1 VS and 300 l kg−1 VS, respectively (VS:DM ratio of 0.8), which is comparable to results presented here. The data in Table 3 show parameters of Eq. (5) and the methane productivity per LU (m3 CH4 LU−1 ) is calculated given certain assumptions about feed intake levels, digestibility in feed DM and the VS:DM ratio. The actual calculations were based on average diets for pigs and cattle in Denmark with average digestibility of feed DM (Table 3). If the actual feed intake and digestibility di>er from these assumptions, then methane productivity should be recalculated with the actual data using Eq. (5). Sows show the lowest methane productivity of 165 m3 CH4 LU−1 , while the other animal categories are in the same range
492
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
Table 3 Methane productivity in terms of livestock units Animal species
Dairy cattle (large breed) Dairy cattle (small) Fattening pigs (7.5 –30 kg) Fattening pigs (30 –100 kg) Sows (23 piglets to 7:5 kg)
Number to 1.LUa
Feed intake (FU animal−1 )
Digestibility in feed DM (%)
VSexcretion (kg animal−1 )
0.85 1 175 36 4.3
6065 5460 46.3 201.6 1340
71 71 84b 85b 83b
1759 1583 4.6 22.8 142
Methane productivity l CH4 kg−1 VS
m3 CH4 LU−1
190 190 350 350 270
284 301 282 287 165
Volatile solid (VS) excretion calculated (Eq. (5)) with assumptions of feed intake and feed digestibility and VS:DM ratio in manure. The VS:DM ratio in manure has been estimated to 0.8. The amount of DM per feed unit (FU) has been estimated to 0:83 kg DM FU−1 in pig feed and 1:25 kg DM FU−1 in cattle feed. a Livestock unit (LU) is deHned by the Ministry of Food, Agriculture and Fisheries [26]. b Poulsen et al. [25].
(282–301 m3 CH4 LU−1 ). From the calculation of methane productivity per LU (Eq. (5)), it is obvious that even small changes in feed intake and feed digestibility can make an important di>erence to LU-speciHc yields. In practice, Bo values, both in terms of VS and LU, are lower than those calculated in this study, since part of the easily degradable organic matter will be degraded and lost as CH4 and CO2 during pre-storage of manure inside buildings before digestion. In Western Europe, manure is stored in buildings for a period which varies between 15 days and 2 months at temperatures of 15 –20◦ C [4]. M$ller et al. [18] calculated Bo losses to be 4.3– 6.6% after 15 days storage and 7.7–11.9% after 30 days storage. 3.2. Methane productivity from fractions of manure deriving from separation Separation is a way to produce manure fractions with higher gas potentials based on volume, since the water can be drained from the solids, giving these fractions a higher VS concentration (Fig. 3). Theoretical methane potential and biodegradability of three types of fractions deriving from manure separation were tested. The fractions deriving from centrifugation (SFD), chemical precipitation (SFC) and evaporation (E-concentrate) have almost the same theoretical methane yield as pig manure (Fig. 2), but the ultimate methane yield in terms of VS is
lower for SFD (average 194 l kg−1 VS) and SFC (247 ± 25 l kg−1 VS) than for pig manure, while it is considerably higher in the E-concentrate (506 ± 15 l kg−1 VS). However, the volumetric methane yield of both the solid fractions and the evaporation concentrate has a considerably higher methane yield than pig manure, due to the higher VS content (Fig. 2); hence the volumetric methane productivity is 2.6 –3.5 higher in the physical fractions of manure compared with undiluted manure from fattening pigs, or 5 – 6.6 higher compared with undiluted sow manure. The methane productivity based on LU will be reduced compared with untreated manure by separation unless all the fractions are digested. On average 60% and 40% of the VS present in the manure will be transferred to SFD and E-concentrate, respectively, when a centrifuge is used as pre-treatment [9] and 85% of the VS will be transferred to SFC when chemical treatment is used [9]. Other studies concerning the biodegradability of physical fractions from manure have been carried out with the solid fractions derived after Hltering [17,19]. Andara and Esteban [19] found a Bo value of 165 l kg−1 VS in the fraction ¿ 1 mm from pig manure, which is less than that found in the solid fraction in this study. The solid fraction from a decanting centrifuge contains particles ¿ 0:025 mm [11] and, since the average particle size is smaller in the present study, a higher yield is expected than in the earlier
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495 Manure to separation
Decanting centrifuge
493
Solid fraction (SFD)
Liquid fraction
Water NH 3 -N
Evaporation unit
Evaporation concentrate (E-Conc .)
Manure to separation
Flocculation /dewatering
Solid fraction (SFC)
Liquid fraction
Fig. 3. Principles of production of concentrated fractions of manure by separation. Bo and Bu are determined in the solid fraction from a decanting centrifuge (SFD), solid fraction from chemical precipitation (SFC) and evaporation concentrate (E-concentrate).
investigation. Several studies reveal that the biogas potential of organic particulate matter increases with a decreasing particle size and an increase in speciHc surface area [20]. 3.3. Methane productivity of straw The chemical characteristics of wheat straw are given in Table 2. The amounts of lipid and protein, in particular, are lower in straw than in manure and thus the theoretical methane yield is signiHcantly lower in straw than in manure (Fig. 2). The ultimate methane yield (Bo ) is also lower in straw than in pig manure and at the same level as cattle manure. However, the volumetric methane yield of straw is higher than the yield from both total manure and the solid fractions from manure, due to the higher VS content, and thus the use of straw as bedding material will increase the volumetric methane yield of the manure. Each 10 g of straw added to 1 kg of manure will increase the methane production by about 1:8 l CH4 kg−1 manure or approximately a 10% increase in the methane yield
of the manure. This means that the methane yield in liquid manure will increase considerably due to the use of straw as bedding material. In the present study, di>erent particle size reductions of straw were tried and, after 60 days batch digestion, the methane productivity (Bo ) from straw cut to a maximum length of 1 mm (161 ± 10 l kg−1 VS) was signiHcantly (p ¡ 0:05) higher than the Bo value of straw cut to a maximum length of 30 mm (145 ± 3 l kg−1 VS) (Fig. 4). However, by extending the batch digestion to 110 days there was only a small and non-signiHcant e>ect of cutting straw to maximum length 1 mm (195 l kg−1 VS) (Fig. 4). The Bo of straw has been determined in several other studies [21–23]. In some other studies Bo has been shown to be inLuenced by particle size reductions [23] and the inoculum:substrate ratio [24]. Sharma et al. [23] found a signiHcant increase in Bo of wheat straw by size reduction from 30 mm (162 l kg−1 VS) to 1 mm (241 l kg−1 VS) but only a small e>ect of further size reduction to 0:1 mm. Torres-Castillo et al. [22] found ultimate methane yields of barley straw
494
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
-1
(l CH4 kgVS )
Methane production
Specific methane production from cattle manure 300 200 100 0 0
20
40
60
80
100
days after start Cattle (No. 7)
Cattle (No. 6)
Cattle (No. 2)
Cattle (No. 5)
Cattle (No. 1)
500 -1
(l CH4 kgVS )
Methane production
Specific methane production from pig manure and different fractions of pig manure
400 300 200 100 0 0
20
40
60
80
100
days after start Pig (No. 14)
Pig (No. 8)
Pig (No. 9)
SFC (No. 20)
SFD (No. 18)
SFD (No. 19)
E-conc. (No. 21)
Specific methane production from wheat straw
Methane production -1 (l CH4 kgVS )
300
200
100
0 0
20
40
60
80
100
days after start No cutting
max 30 mm
max. 1mm
Fig. 4. Methane production plotted against time from cattle, pig manure, physical fractions of manure and straw.
in the range from 240 l kg−1 VS to 370 l kg−1 VS. The present study and other studies clearly indicate a considerable variation in Bo depending on pre-treatment of the straw and digestion conditions. 4. Conclusions The chemical composition and the methane productivity in manure in terms of VS, volume and livestock production were determined. Furthermore,
the methane productivity in terms of VS and volume in straw and physical fractions of manure were also determined. The theoretical methane productivity is higher in pig (516 l kg−1 VS) and sow (530 l kg−1 VS) manure than in dairy cattle manure (468 l kg−1 VS) but is still relatively constant, while the ultimate methane yield in terms of VS is considerably higher in pig (356 l kg−1 VS) and sow manure (275 l kg−1 VS) than in dairy cattle manure (148 l kg−1 VS). Methane productivity based on livestock units (LU) shows the lowest methane productivity for sows (165 m3 CH4 LU−1 ), while the other animal categories are in the same range (282– 301 m3 CH4 LU−1 ). Variations in feed intake and feed digestibility can play an essential role in the LU-speciHc yields, and the actual production data are needed in order to make more precise evaluations of the energy production and economics of a biogas installation. Pre-treatment of manure by separation is a method of making fractions of the manure with higher gas potential in terms of volume, since the water can be drained from the solids, thus creating fractions with a higher VS concentration; hence the volumetric methane productivity is 2.6 –3.5 higher in the physical fractions of manure compared with undiluted manure from fattening pigs or 5 – 6.6 higher than undiluted sow manure. Theoretical methane potential and biodegradability of three types of fractions deriving from manure separation were tested. The fractions deriving from centrifugation (SFD), chemical precipitation (SFC) and evaporation (E-concentrate) have almost the same theoretical methane yield as pig manure in terms of VS, but the ultimate methane yield in terms of VS is lower for SFD (210 l kg−1 VS) and SFC (247 l kg−1 VS) than for pig manure, while it is considerably higher in the E-concentrate (506 l kg−1 VS). The methane productivity by separation of manure, in terms of LU, will depend on the separation eOciency of VS and, compared with untreated manure, it will be reduced unless all the fractions are digested. The volumetric methane yield of straw is higher than the yield from both manure and solid fractions from manure, due to the higher VS content, and therefore the use of straw as a bedding material will increase the volumetric as well as the livestock-based methane
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
productivity; thus every kilogram of straw added per 100 kg manure will increase methane productivity by 10%. Acknowledgements Thanks to Merete Maahn for her thorough and careful help during performance of the test. This paper is supported by grants from The Danish Energy Council. References [1] Hashimoto AG, Chen YR, Varel VH. Theoretical aspects of anaerobic fermentation: state of the art. In: Livestock waste: a renewable resource. Proceedings of the Fourth International Symposium on Livestock Wastes. ASAE, St Joseph, Michigan, 1981. p. 86 –91. [2] Husted S. Seasonal variation in methane emission from stored slurry and solid manures. Journal of Environmental Quality 1994;23:585–92. [3] Steed J, Hashimoto G. Methane emissions from typical manure management systems. Bioresource Technology 1994;50:123–30. [4] Sommer SG, M$ller HB, Petersen SO. Reduction in methane and nitrous oxide emission from animal slurry through anaerobic digestion. In: Proceedings of the Third International Symposium, 21–23 January, Maastricht, The Netherlands, 2002. [5] Hill DT. Methane productivity of the major animal waste types. Transactions of the ASAE 1984;27(2):530–4. [6] Symons GE, Bushwell AM. The methane fermentation of carbohydrate. Journal of the American Chemical Society 1933;55:2028–39. [7] Angelidaki I, Ahring BK, Ahring BK. Methods for increasing the biogas potential from the recalcitrant organic matter contained in manure. Water Science and Technology 2000;41(3):189–94. [8] Hansen KH, Angelidaki I, Ahring BK. Anaerobic digestion of swine manure: inhibition by ammonia. Water Resources 1998;32(1):5–12. [9] M$ller HB. Separation af slagtesvinegylle med Ansager SepTec gylleseparator. (Separation of pig manure with Ansager SepTec separator.) Internal Report No. 159. Danish Institute of Agricultural Sciences, 2002. [10] IPCC. Guidelines for National Greenhouse Gas Inventories: Reference Manual. Revised 1996. IPCC Guidelines, 1997. [11] M$ller HB, Sommer SG, Ahring BK. Separation eOciency and particle size distribution in relation to manure type and storage conditions. Bioresource Technology 2002;85(2): 189–96.
495
[12] M$ller HB, Hansen MN, Sommer SG. Heavy metal load from fractions of manure after separation. Bioresource Technology 2003; submitted for publication. [13] ISO. Water quality: evaluation of the ‘ultimate’ anaerobic biodegradability of organic compounds in digested sludge— method by measurements of the biogas production. International Standard, ISO/DIS 11734, 1995. [14] APHA. Standard methods for the examination of water and wastewater, 19th ed. Washington, DC: American Public Health Association; 1995. [15] Ianotti EL, Porter JH, Fisher JR, Sievers DM. Changes in swine manure during anaerobic digestion. Developments in Industrial Microbiology 1979;20(49):519–29. [16] Converse JC, Zeikus RE, Graves RE, Evans GW. Anaerobic degradation of dairy manure under mesophilic and thermophilic temperatures. Transactions of the ASAE 1977;20:336. [17] Haugen VJ, Lindley JA. Biogas production after solid-liquid separation of dairy manure. Transactions of the ASAE 1988;31(6):1782–6. [18] M$ller HB, Sommer SG, Ahring BK. Biological degradation and greenhouse gas emissions during pre-storage of liquid manure. Journal of Environmental Quality 2004;(33): Jan–Feb. [19] Andara AR, Esteban JML. Kinetic study of the anaerobic digestion of the solid fraction of piggery slurries. Biomass and Bioenergy 1999;17:435–43. [20] Hartmann H, Angelidaki I, Ahring BK. Increase of anaerobic digestion of particulate organic matter in full-scale biogas plants by mechanical separation. Water Science and Technology 2000;41(3):145–53. [21] Gunaseelan VN. Anaerobic digestion of biomass for methane production: a review. Biomass and Bioenergy 1997;13: 83–114. [22] Torres-Castillo R, Llabres-Luengo P, Mata-Alvaraz J. Temperature e>ect on anaerobic digestion of bedding straw in a one-phase system at di>erent inoculum concentration. Agriculture Ecosystems Environment 1995;54: 55–66. [23] Sharma SK, Mishra IM, Sharma MP, Saini JS. 1988. E>ect of particle size on biogas generation from biomass residues. Biomass 1988;17:251–63. [24] Hashimoto AG. E>ect of inoculum/substrate ratio on methane yield and production rate from straw. Biological Wastes 1987;28:247–55. [25] Poulsen HD, B$rsting CF, Rom HB, Sommer SG. KvXlstof, fosfor og kalium i husdyrg$dning. (Nitrogen, phosphorus and potassium contained in livestock manure.) DIAS Report No. 36, 2001. [26] Ministry of Food, Agriculture and Fisheries. Vejledning og skemaer, mark og g$dningsplan, g$dningsregnskaplantedXkke og harmoniregler. (Guidance concerning fertilizer, nutrient account and harmony rules.), 2002.
Biomass and Bioenergy 26 (2004) 485 – 495
Methane productivity of manure, straw and solid fractions of manure H.B. M$llera;∗ , S.G. Sommera , B.K. Ahringb a Department
of Agricultural Engineering, Danish Institute of Agricultural Sciences, Research Center Bygholm, P.O. Box 536, DK-8700 Horsens, Denmark b Biocentrum-DTU, The Technical University of Denmark, DK-2800 Lyngby, Denmark
Received 24 February 2003; received in revised form 9 July 2003; accepted 25 August 2003
Abstract The methane productivity of manure in terms of volatile solids (VS), volume and livestock production was determined. The theoretical methane productivity is higher in pig (516 l kg−1 VS) and sow (530 l kg−1 VS) manure than in dairy cattle manure (469 l kg−1 VS), while the ultimate methane yield in terms of VS is considerably higher in pig (356 l kg−1 VS) and sow manure (275 l kg−1 VS) than in dairy cattle manure (148 l kg−1 VS). Methane productivity based on livestock units (LU) shows the lowest methane productivity for sows (165 m3 CH4 LU−1 ), while the other animal categories are in the same range (282–301 m3 CH4 LU−1 ). Pre-treatment of manure by separation is a way of making fractions of the manure that have a higher gas potential per volume. Theoretical methane potential and biodegradability of three types of fractions deriving from manure separation were tested. The volumetric methane yield of straw was found to be higher than the yield from total manure and the solid fractions of manure, due to the higher VS content, and hence the use of straw as bedding material will increase the volumetric as well as the livestock-based methane productivity. ? 2003 Elsevier Ltd. All rights reserved. Keywords: Methane yield; Anaerobic digestion; Manure; Straw; Manure separation
1. Introduction Anaerobic degradation of animal manure results in the production of biogas, which is composed mainly of methane and carbon dioxide [1]. Natural degradation of manure, leading to emissions of CH4 during storage, is very undesirable because of the global warming e>ects resulting from the release of greenhouse gases [2,3]. However, by using controlled anaerobic
∗ Corresponding author. Tel.: +45-7629-6043; fax: +45-76296100. E-mail address: [email protected] (H.B. M$ller).
0961-9534/$ - see front matter ? 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2003.08.008
digestion of animal manure, where the biogas is captured, the CH4 emissions during storage of manure can be reduced and the energy from manure can be used as a substitute for fossil fuels, thereby serving as a CO2 -neutral energy source [4]. The methane potential of manure comes from the digestion of the organic components in the faeces and in the straw used as bedding material, which is mainly: carbohydrates, proteins and lipids. Methane yield can be expressed in several ways [5]. The term ‘methane productivity’ is used to indicate the yield of methane per unit of a variable. Methane productivity can be measured in terms of volatile solids (VS) destroyed, VS loaded, volume, or animal production. Methane productivity measured
486
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
in terms of VS destroyed (l CH4 kg−1 VSdes ) corresponds to the theoretical methane yield (Bu ) if there is complete degradation of all organic components of the manure. The theoretical methane potential can be calculated from Bushwell’s formula [6]. Methane productivity in terms of VS loaded (l CH4 kg−1 VSload ) as residence time approaches inHnity is referred to as the ultimate methane yield (Bo ). The ultimate methane yield will always be lower than the theoretical yield because a fraction of the substrate is used to synthesize bacterial mass, a fraction of the organic material will be lost in the eJuent, and lignin-containing compounds will only be degraded to a limited degree [7]. Inhibition of the biological process by inhibitors such as ammonia and volatile fatty acids (VFA) is another factor contributing to the actual methane yield being lower than the potential yield which would be obtained if inhibition was not present [8]. It has been observed that both the ultimate methane yield (Bo ) and the volumetric methane production (l CH4 m−3 manure) of manure from di>erent origins can be very variable. The ultimate methane yield (l CH4 kg−1 VS) is affected by various factors, including: 1. 2. 3. 4.
species, breed and growth stage of the animals [1], feed [1], amount and type of bedding material [1], degradation processes during pre-storage [9].
The volumetric methane production (l CH4 m−3 manure) is a>ected by the same factors as the ultimate methane yield, but is also inLuenced by the amount of wasted drinking water and water added to the manure collection system. Both the ultimate methane yield and the volumetric methane production are of great importance for the economy of biogas plants. The amount of biogas that can be produced is proportional to the ultimate methane yield, while the volumetric methane production a>ects the necessary digester volume and hence the overall economics of a digestion facility. However, both the ultimate methane yield and the volumetric methane production are only concerned with the concentration and ‘quality’ of VS in the manure, rather than the methane yield that can be produced from a given animal production system. Methane productivity in terms of animal production can be determined by starting with the ultimate methane yield and applying an animal waste production factor. In
several countries, including Denmark, the term livestock unit (LU) is applied to describe the animal production factor and it is this productivity which is most useful in determining economic feasibility [5], since units represent the volume of methane per animal (m3 CH4 LU−1 ). The ultimate methane yield that can be produced from a given animal is determined by both the amount and the quality of the VS excreted. Often the VS excretion value from a given animal production system is not available and since feed intake levels, feedstu> quality and other factors di>er, the best estimate of VS excretion will be production data on feed intake and digestibility of the feed [10]. The volumetric methane production of manure can be increased by solid–liquid separation [11], thereby producing a fraction with a high VS concentration. However, a part of the VS will always be present in the liquid fraction, and if only the solid fraction is used as a substrate for anaerobic digestion, it is extremely important that the separation unit is eOcient in transferring VS to the solid fraction. M$ller et al. [12] found that chemical precipitation combined with screening and centrifugation was an eOcient method for transferring VS to the solid fraction. 2. Materials and methods 2.1. Gas production The ultimate methane yield (Bo ) was determined in a batch experiment. The experiments were performed in 1100-ml infusion bottles. The bottles were closed with butyl rubber stoppers, sealed with aluminium crimps and incubated at 35 ± 0:5◦ C, after Lushing with N2 . The method used is described in the international standard ISO 11734 [13]. The test medium was faeces taken directly after excretion from the pigs and cattle, wheat straw, and solid fractions of manure produced by centrifugation or by chemical precipitation (see Table 1 for details). The test medium was frozen until used in the batch experiment. Inoculum from a farm-scale biogas plant, which had been kept for 2 weeks before the test at 35◦ C to remove most of the remaining methane production, was used. The inoculum:substrate ratio is shown in Table 2. Each test medium was digested in triplicate. The gas produced by controls was subtracted from the actual gas
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
487
Table 1 Characteristics of the pig, cattle, solid fraction from manure and straw used in this study Manure No.
Manure/biomass type
Feed stu>
1 2 3 4 5 6 7
Cattle, Cattle, Cattle, Cattle, Cattle, Cattle, Cattle,
8
Pig fatteners, faeces
9 10 11
Pig fatteners, faeces Pig fatteners, faeces Pig fatteners, faeces
12 13
Pig fatteners, faeces Pig fatteners, slurry
14
Pig fatteners, faeces
15
Sows, faeces
16
Sows, faeces
17
Sows, faeces
18 19 20 21 22
Solid fraction (pig)—centrifugation Solid fraction (pig)—centrifugation Solid fraction (pig)—chemical treatment Evaporation concentrate Wheat straw
Hay from clover and minerals Concentrates and maize silage Concentrates and hay from alfalfa 19.6% concentrates, 80% roughage, 0.4% minerals 22.6% concentrates, 77% roughage, 0.4% minerals 21.6% concentrates, 55.2% roughage, 22.8% barley, and 0.4% minerals 52% roughage (maize and clovergrass silage), 47% concentrates and 0.4% minerals 24% soybean meal, 49.8% barley, 20% wheat, 3% molasses and fat, 3.2% minerals 7.5% soybean meal, 16.5% barley/wheat, 75% whey permeate, 1% fat 77% soybean meal, 20.2% barley/wheat, 3% concentrate 24% soybean meal, 49.8% barley, 20% wheat, 3% molasses and fat, 3.2% minerals No data available 24% soybean meal, 49.8% barley, 20% wheat, 3% molasses and fat, 3.2% minerals 24% soybean meal, 49.8% barley, 20% wheat, 3% molasses and fat, 3.2% minerals Protein concentrates 2%, barley/oat 22,7%, whey permeate 74.5%, animal fat 1% 13% soybean meal, 39% barley, 38% wheat, 3% fat, 6% protein concentrates, 3.2% minerals Dry feed: 8.5% soybean meal, 88.5% wheat/barley, 3% concentrates. Wet fed: 100% whey No data available No data available No data available No data available
dairy dairy dairy dairy dairy dairy dairy
faeces faeces faeces slurry slurry slurry faeces
produced through digestion of the medium. The volume of gas produced was calculated either by measuring pressure in the headspace [13], or by connecting the infusion vessel to a gas collection bag and using a large syringe to measure the volume of gas, as described by Steed and Hashimoto [3]. The two methods were compared and the di>erence was found to be negligible (¡ 1%), indicating that both methods were applicable. The gas samples were analysed for CO2 and CH4 with gas chromatography. CH4 was measured on a Hewlett Packard (5890, series II) gas chromatograph with an electron capture detector and a Lame ionization detector (FID). Methane was isolated on a 1:83 m × 3:1 mm column with Poropak N 80/100, the carrier gas was He at 30 ml min−1 and the temperatures of injection port, oven and detector were 110◦ C, 40◦ C and 270◦ C, respectively. CO2
was measured on a Varian 3350 gas chromatograph (1 m × 3:1 mm column with a 2 m × 3:1 mm Haysep R 80/100 molecular sieve) equipped with a thermal conductivity detector. The carrier gas was He at a Low rate of 30 ml min−1 , the temperatures of oven and detector were 40◦ C and 150◦ C, respectively. 2.2. Manure analysis The total nitrogen content (TN) was analysed by the Kjeldahl method and a Kjellfoss 16200 (Copenhagen, Denmark). The dry matter (DM) content was determined after a 24-h drying period at 105◦ C. The ammonium (NH+ 4 ) in the manure was analysed by means of a QuikChem 4200 Low injection analyzer (Lachat Instruments, Milwaukee, WI, USA). The chemical oxygen demand (COD) was analysed colorimetrically
488
Table 2 Composition, theoretical and ultimate methane yields Cattle
No. 2
No. 3
No. 4a
No. 5a
No. 6a
No. 7
Avg. Cattle
143.75 902.70 0.91 11.90 11.19 76.09 70.90 555.57 155.42 130.83 258.40 0.7 100 464 0.21
138.34 899.50 1.70 24.50 30.82 158.42 53.70 489.02 193.33 74.71 241.10 0.7 124 460 0.27
156.44 859.90 1.40 24.00 18.85 164.26 68.96 318.80 252.24 176.88 369.00 0.7 — 470 —
104.04 796.80 25.70 51.40 82.53 201.59 80.82 372.13 161.90 101.03 209.50 0.7 — 480 —
98.10 938.60 — — — — — — — — — 0.7 150 469 0.32
105.10 930.00 — — — — — — — — — 0.7 207(±5:9) 469 0.44
111.93 939.10 — — — — — — — — — 0.7 161(±2:9) 469 0.34
122.53 895.23 7.43 27.95 35.85 150.09 68.59 433.88 190.72 120.86 269.50 0.7 148(±41) 469 0.32
Sows
Pig, fatteners
DM incl. VFA (g/l) VS (g/kg DM) NH+ 4 -N (g/kg TS) Total N (g/kg TS) VFA (g/kg VS) Proteina (g/kg VS) Lipids (g/kg VS) Degradable carbohydrates (g/kg VS) Non degradable carbohydrates (g/kg VS) Lignin Crude Hbre (g/kg VS) Inoculum/substrate (I:S) ratio Bo (L CH4 /g VS) Bu (l CH4 /g VS) Biodegradability (Bo =Bu )
No. 8
No. 9
No. 10
No. 11
No. 12
No. 13a
No. 14
Avg. pig
No. 15
No. 16
No. 17
Avg. sows
206.69 821.00 5.00 39.30 78.32 261.11 152.01 352.40
233.40 849.00 5.90 42.20 64.57 267.23 130.98 357.48
238.00 803.50 5.50 44.00 67.98 299.47 110.64 285.44
292.19 868.30 8.00 34.70 62.62 192.19 137.97 349.82
237.99 850.20 5.40 29.30 78.52 175.69 170.67 344.47
95.50 836.50 46.10 70.10 80.00 179.32 120.50 393.04
260.00 912.10 — — — — — —
223.39 848.66 12.65 43.27 72.00 229.17 137.13 347.11
318.93 819.30 2.40 31.20 31.12 219.70 159.77 401.20
296.78 810.10 1.40 23.80 27.76 172.82 178.37 389.10
317.30 745.50 0.00 25.50 29.59 213.78 151.58 377.01
311.00 791.63 1.27 26.83 29.49 202.10 163.24 389.10
112.06
136.40
191.47
206.96
169.61
176.93
—
165.57
129.50
156.65
158.03
148.06
44.09 128.20 1.1 359
43.35 152.60 1.1 403
45.00 190.00 1.1 329
50.44 223.50 1.1 —
61.04 196.10 1.1 340
50.21 190.00 1.1 —
— —
58.71 154.20 4.2 250
75.30 187.90 4.2 260
70.00 170.00 4.2 317
528
516
507
514
532
502
49.02 180.07 1.1 356 (±28) 516 (±11) 0.69
529
536
524
68.00 170.70 4.2 275 (±36) 530 (±6) 0.52
0.67
0.78
0.65
—
0.62
—
1.1 358 (±8) — 0.69
0.47
0.48
0.60
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
DM incl. VFA (g/l) VS (g/kg DM) NH+ 4 -N (g/kg TS) Total N (g/kg TS) VFA (g/kg VS) Proteina (g/kg VS) Lipids (g/kg VS) Degradable carbohydrates (g/kg VS) Non degradable carbohydrates (g/kg VS) Lignin Crude Hbre (g/kg VS) Inoculum/substrate (I:S) ratio Bo (l CH4 /g VS) Bu (l CH4 /g VS) Biodegradability (Bo =Bu )
No. 1
13.32 0.00 7.3 506(±15) 542 0.93
905.40 959.00 0.00 5.90 0.00 38.45 23.57 501.49 358.00 78.50 418.60 0.7 195(±5:9) 432 0.45 285.00 812.00 14.00 44.50 0.00 234.76 130.67 440.73 103.08 90.76 157.40 0.7 247(±25) 514 0.48
2.3. Separation technologies The following types of separation technologies were used to produce solid fractions with a high VS concentration from the manure: • Decanting centrifuge produced by Alfa Laval, NX 309B-31 (R$dovre, Denmark) • Chemical treatment system, AnsagerSepTec (Ansager, Denmark) • Evaporation of water from the liquid fraction pre-treated with a decanting centrifuge (Funki Manura, S$nderborg, Denmark).
SD in brackets. a Slurry.
DM incl. VFA (g/l) VS (g/kg DM) NH+ 4 -N (g/kg TS) Total N (g/kg TS) VFA (g/kg VS) Proteina (g/kg VS) Lipids (g/kg VS) Degradable carbohydrates (g/kg VS) Non degradable carbohydrates (g/kg VS) Lignin Crude Hbre (g/kg VS) Inoculum:substrate (I:S) ratio Bo (l CH4 /g VS) Bu (l CH4 /g VS) Biodegradability (Bo =Bu )
489
(Spektroquant Nova 60; Merck, Germany) in accordance with a method described by the American Public Health Association [14]. VFA C2 –C5 concentration was determined by means of a gas chromatograph (Hewlett Packard 6850A) with a FID. The column was an HP-INNOWax, 30 m × 0:25 mm × 0:25 m. The carrier gas was He. The temperature of the column was gradually increased from 110◦ C to 220◦ C at a rate of 10◦ C min−1 . The lipid content was analyzed by measuring the amount of material, that can be extracted with diethyl ether in a soxleth extraction instrument after hydrolysis with 3 N hydrochloric acid. The lignin content was analyzed by determination of the suspended VS after boiling with a detergent in 1 mol=l sulfuric acid followed by suspending in 72% sulfuric acid. The protein content was determined by multiplying the di>erence between TN and NH4 -N with the factor 6.25. The crude Hbers were analyzed by determination of the suspended VS after boiling with 0:26 mol=l sulfuric acid and 0:23 mol=l potassium hydroxide dissolution.
290.60 703.20 15.60 42.80 14.68 241.75 119.60 400.14 119.03 104.81 157.40 4.8 261(±10) 508 0.51
309.00 807.40 26.80 47.10 11.69 180.43 150.88 433.16 119.03 104.81 157.40 0.7 159(±39) 521 0.30
198.00 637.91 71.40 102.30 158.35 302.75 170.87 354.71
No. 22 No. 20 No. 19 No. 18
Solid fraction from pig manure
No. 21
Straw
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
The decanter centrifuge consisted of a drum adjusted to rotate at a speed of 5600 rpm, giving a pressure of 4100g at the drum circumference; the rotational speed of the screw was 60 –80 rpm slower than the speed of the drum. For details see M$ller et al. [11]. The AnsagerSepTec separator consisted of three Locculation chambers with mixing and addition of chemicals to form new chemical compounds and to alter the physical state of dissolved and suspended solids and to facilitate their removal by a belt press. For details see M$ller et al. [12].
490
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
The Funki Manura technology uses only mechanical–thermal techniques. The separation process is divided into three main steps: mechanical pre-treatment (decanting centrifuge), evaporation and distillate cleaning. The pre-treated liquid is pumped to the evaporation unit, where it is degassed after an initial heating to about 100◦ C. During degassing, ammonia and carbon dioxide are removed. The degassed manure is pumped to the evaporation–heat exchanger. In the evaporation–heat exchanger the slurry is heated to more than 100◦ C. To avoid fouling the heat exchanger, the slurry is circulated with a high Low on the heated surface. By this process most of the water and the remaining ammonia evaporates. The remaining concentrate contains a high concentration of VS and nutrients. Evaporated water and ammonia are removed from the evaporation unit and are condensed. 2.4. Calculations The theoretical CH4 production (Bu ) was calculated from Bushwell’s formula [6]: a b H2 O C n Ha Ob + n − − 4 2 n a b n a b − + CO2 + + − CH4 ; (1) → 2 8 4 2 8 4 (n=2 + a=8 − b=4)22:4 (2) l CH4 kg−1 VS: 12n + a + 16b The average chemical composition of the VS in the manure was calculated with the assumption of an average composition of VSlipid (C57 H104 O6 ), VSprotein (C5 H7 O2 N), VScarbohydrate (C6 H10 O5 ), VSlignin (C6 H10 O5 ) and VSVFA (C2 H4 O2 ). The carbohydrates were divided into an easily degradable part (VSE; carbohydrate ) and a slowly degradable part(VSS; carbohydrate ). The degradable part of the carbohydrates was calculated by Eq. (4) and the slowly degradable part was calculated by Eq. (3): Bu =
VSS; carbohydrate = VScrudeHbre − VSlignin
(3)
VSE; carbohydrate = VS − VSprotein − VSlipid − VSVFA − VSS; carbohydrate
(4)
The theoretical CH4 production (Bu ) from VFA (acetic acid), carbohydrates, protein and lipid is
370, 415, 496 and 1014 l CH4 kg−1 VS, respectively (Eq. (2)). The ultimate methane yield from a given animal production unit can be calculated by using standard values for the excretion of VS from the given animal category and multiplying by Bo and number of animals, but it is more accurate to use data on feed intake and digestibility of the feed, to calculate the excretion of VS (Eq. (5)). VSexcretion i
= Intake(FU)i × DMi =FUi ×(1 − DEi %=100) × (1 − ASHi %=100);
(5)
where VSexcretion is the amount of VS excreted on an animal or farm level, Intake(FU)i the feed intake in feed units (FU) of feedstu> (i) on an animal or farm level, DMi =FUi the amount of dry matter (DM) per feed unit (FU) of feedstu> (i), DEi % the digestibility of feed DM of feedstu> (i), and ASHi % the ash content of the manure. 3. Results and discussion 3.1. Methane productivity of manure The chemical characteristics of the manures are given in Table 2. The VS were divided into four main groups, VSVFA ; VSlipid , VSprotein , VSlignin and VScarbohydrate . The carbohydrates were further divided into an easily degradable part (VSE; carbohydrate ) and a slowly degradable part (VSS; carbohydrate ) (Fig. 1). The average proportion of slowly degradable carbohydrates and lignin is larger in cattle than in pig manure (Fig. 1). This is as expected, since cattle are fed roughage which contains greater amounts of lignin complexed with cellulose than pig feed does. Furthermore the amounts of proteins and lipids are considerably higher in pig than cattle manure. Due to the higher proportion of lipid in pig manure, the theoretical CH4 production in the present study is higher in both pig (516 ± 11 l kg−1 VS) and sow (530 ± 6 l kg−1 VS) manure than in cattle manure (468 ± 6 l kg−1 VS). Hill [5] found that methane productivity per unit VS destroyed was relatively constant for all waste types (500 l kg−1 VS). Ianotti et al. [15] found almost the
90% 80% Lignin
60%
Carbohydrates (S) Carbohydrates (E)
50%
Lipid
40%
Protein VFA
30% 20% 10% 0% pig, fattening
sows
SFC
SFD
E-conc.
500 400 300 200 100 0 Cattle
Pig
Sow
SFD
SFC
straw
Fig. 1. The average composition of VS in freshly excreted cattle manure, fattening pig manure, sow manure, solid fraction from centrifugation of pig manure (SFD), solid fraction from chemical precipitation of manure (SFC), liquid fraction pretreated with a decanting centrifuge (E-conc.) and wheat straw.
same percentages of lipids (14.8%) and lignin (4.4%) but a slightly higher protein content (19.4%) in terms of VS in pig manure than we found in the present study. There is only a small di>erence in the VS composition of the manure from sows and fattening pigs (Table 2). The lignin and lipid contents are higher in sow manure, and thus the theoretical CH4 production (Bu ) is slightly but not signiHcantly (p ¿ 0:05, t-test) higher from sows than from fattening pigs (Fig. 2). The ultimate methane yield of dairy cattle manure (148 ± 41 l kg−1 VS) is signiHcantly (p ¡ 0:01, t-test) lower than the Bo of pig manure (356 ± 28 l kg−1 VS) and sow manure (275±36 l kg−1 VS). The variation between manure from individual cows is much higher than the variation between individual pigs, which may be explained by the fact that there are more di>erences in feeding practice and productivity between dairy farms than between pig farms; hence cows fed on only roughage give lower yields than cows fed on both roughage and concentrates. Furthermore, the plateau phase during batch digestion is reached much faster with pig manure than with cattle manure, indicating that the retention time seems to be longer during digestion of cattle manure. The Bo of manure from cows and pigs has been determined in several other studies [1,5,16,17]. IPCC [10] estimated the Bo of dairy cattle and pig manure in developed countries to be 240 l kg−1 VS and 450 l kg−1 VS,
E- Straw conc.
200 (l CH4 kg-1 product)
cattle
Volumetric methane yield
Ratio of VS
70%
491
600
(l CH4 kgVS-1)
100%
Theoretical and ultimate methane yield
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
167
150 100 50
12
19
10
Cattle
Pig
Sow
50
55
SFD
SFC
66
0 E- Straw conc.
Fig. 2. The theoretical Bu ( ) and ultimate methane yield Bo () in terms of VS and volume from cattle, pig manure, physical fractions of manure and straw. The DM content in manure has been assumed to be 10.3%, 6.6% and 4.5% in, respectively, cattle, pig and sow manure [26]. The VS:DM ratio in manure has been estimated as 0.8. The DM content in the fractions from manure has been assumed to be the same as in Table 2.
respectively, which is at the upper end of the Bo values determined in this study. Hill [5] estimated the Bo of dairy cattle and pig manure to be 131 l kg−1 VS and 300 l kg−1 VS, respectively (VS:DM ratio of 0.8), which is comparable to results presented here. The data in Table 3 show parameters of Eq. (5) and the methane productivity per LU (m3 CH4 LU−1 ) is calculated given certain assumptions about feed intake levels, digestibility in feed DM and the VS:DM ratio. The actual calculations were based on average diets for pigs and cattle in Denmark with average digestibility of feed DM (Table 3). If the actual feed intake and digestibility di>er from these assumptions, then methane productivity should be recalculated with the actual data using Eq. (5). Sows show the lowest methane productivity of 165 m3 CH4 LU−1 , while the other animal categories are in the same range
492
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
Table 3 Methane productivity in terms of livestock units Animal species
Dairy cattle (large breed) Dairy cattle (small) Fattening pigs (7.5 –30 kg) Fattening pigs (30 –100 kg) Sows (23 piglets to 7:5 kg)
Number to 1.LUa
Feed intake (FU animal−1 )
Digestibility in feed DM (%)
VSexcretion (kg animal−1 )
0.85 1 175 36 4.3
6065 5460 46.3 201.6 1340
71 71 84b 85b 83b
1759 1583 4.6 22.8 142
Methane productivity l CH4 kg−1 VS
m3 CH4 LU−1
190 190 350 350 270
284 301 282 287 165
Volatile solid (VS) excretion calculated (Eq. (5)) with assumptions of feed intake and feed digestibility and VS:DM ratio in manure. The VS:DM ratio in manure has been estimated to 0.8. The amount of DM per feed unit (FU) has been estimated to 0:83 kg DM FU−1 in pig feed and 1:25 kg DM FU−1 in cattle feed. a Livestock unit (LU) is deHned by the Ministry of Food, Agriculture and Fisheries [26]. b Poulsen et al. [25].
(282–301 m3 CH4 LU−1 ). From the calculation of methane productivity per LU (Eq. (5)), it is obvious that even small changes in feed intake and feed digestibility can make an important di>erence to LU-speciHc yields. In practice, Bo values, both in terms of VS and LU, are lower than those calculated in this study, since part of the easily degradable organic matter will be degraded and lost as CH4 and CO2 during pre-storage of manure inside buildings before digestion. In Western Europe, manure is stored in buildings for a period which varies between 15 days and 2 months at temperatures of 15 –20◦ C [4]. M$ller et al. [18] calculated Bo losses to be 4.3– 6.6% after 15 days storage and 7.7–11.9% after 30 days storage. 3.2. Methane productivity from fractions of manure deriving from separation Separation is a way to produce manure fractions with higher gas potentials based on volume, since the water can be drained from the solids, giving these fractions a higher VS concentration (Fig. 3). Theoretical methane potential and biodegradability of three types of fractions deriving from manure separation were tested. The fractions deriving from centrifugation (SFD), chemical precipitation (SFC) and evaporation (E-concentrate) have almost the same theoretical methane yield as pig manure (Fig. 2), but the ultimate methane yield in terms of VS is
lower for SFD (average 194 l kg−1 VS) and SFC (247 ± 25 l kg−1 VS) than for pig manure, while it is considerably higher in the E-concentrate (506 ± 15 l kg−1 VS). However, the volumetric methane yield of both the solid fractions and the evaporation concentrate has a considerably higher methane yield than pig manure, due to the higher VS content (Fig. 2); hence the volumetric methane productivity is 2.6 –3.5 higher in the physical fractions of manure compared with undiluted manure from fattening pigs, or 5 – 6.6 higher compared with undiluted sow manure. The methane productivity based on LU will be reduced compared with untreated manure by separation unless all the fractions are digested. On average 60% and 40% of the VS present in the manure will be transferred to SFD and E-concentrate, respectively, when a centrifuge is used as pre-treatment [9] and 85% of the VS will be transferred to SFC when chemical treatment is used [9]. Other studies concerning the biodegradability of physical fractions from manure have been carried out with the solid fractions derived after Hltering [17,19]. Andara and Esteban [19] found a Bo value of 165 l kg−1 VS in the fraction ¿ 1 mm from pig manure, which is less than that found in the solid fraction in this study. The solid fraction from a decanting centrifuge contains particles ¿ 0:025 mm [11] and, since the average particle size is smaller in the present study, a higher yield is expected than in the earlier
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495 Manure to separation
Decanting centrifuge
493
Solid fraction (SFD)
Liquid fraction
Water NH 3 -N
Evaporation unit
Evaporation concentrate (E-Conc .)
Manure to separation
Flocculation /dewatering
Solid fraction (SFC)
Liquid fraction
Fig. 3. Principles of production of concentrated fractions of manure by separation. Bo and Bu are determined in the solid fraction from a decanting centrifuge (SFD), solid fraction from chemical precipitation (SFC) and evaporation concentrate (E-concentrate).
investigation. Several studies reveal that the biogas potential of organic particulate matter increases with a decreasing particle size and an increase in speciHc surface area [20]. 3.3. Methane productivity of straw The chemical characteristics of wheat straw are given in Table 2. The amounts of lipid and protein, in particular, are lower in straw than in manure and thus the theoretical methane yield is signiHcantly lower in straw than in manure (Fig. 2). The ultimate methane yield (Bo ) is also lower in straw than in pig manure and at the same level as cattle manure. However, the volumetric methane yield of straw is higher than the yield from both total manure and the solid fractions from manure, due to the higher VS content, and thus the use of straw as bedding material will increase the volumetric methane yield of the manure. Each 10 g of straw added to 1 kg of manure will increase the methane production by about 1:8 l CH4 kg−1 manure or approximately a 10% increase in the methane yield
of the manure. This means that the methane yield in liquid manure will increase considerably due to the use of straw as bedding material. In the present study, di>erent particle size reductions of straw were tried and, after 60 days batch digestion, the methane productivity (Bo ) from straw cut to a maximum length of 1 mm (161 ± 10 l kg−1 VS) was signiHcantly (p ¡ 0:05) higher than the Bo value of straw cut to a maximum length of 30 mm (145 ± 3 l kg−1 VS) (Fig. 4). However, by extending the batch digestion to 110 days there was only a small and non-signiHcant e>ect of cutting straw to maximum length 1 mm (195 l kg−1 VS) (Fig. 4). The Bo of straw has been determined in several other studies [21–23]. In some other studies Bo has been shown to be inLuenced by particle size reductions [23] and the inoculum:substrate ratio [24]. Sharma et al. [23] found a signiHcant increase in Bo of wheat straw by size reduction from 30 mm (162 l kg−1 VS) to 1 mm (241 l kg−1 VS) but only a small e>ect of further size reduction to 0:1 mm. Torres-Castillo et al. [22] found ultimate methane yields of barley straw
494
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
-1
(l CH4 kgVS )
Methane production
Specific methane production from cattle manure 300 200 100 0 0
20
40
60
80
100
days after start Cattle (No. 7)
Cattle (No. 6)
Cattle (No. 2)
Cattle (No. 5)
Cattle (No. 1)
500 -1
(l CH4 kgVS )
Methane production
Specific methane production from pig manure and different fractions of pig manure
400 300 200 100 0 0
20
40
60
80
100
days after start Pig (No. 14)
Pig (No. 8)
Pig (No. 9)
SFC (No. 20)
SFD (No. 18)
SFD (No. 19)
E-conc. (No. 21)
Specific methane production from wheat straw
Methane production -1 (l CH4 kgVS )
300
200
100
0 0
20
40
60
80
100
days after start No cutting
max 30 mm
max. 1mm
Fig. 4. Methane production plotted against time from cattle, pig manure, physical fractions of manure and straw.
in the range from 240 l kg−1 VS to 370 l kg−1 VS. The present study and other studies clearly indicate a considerable variation in Bo depending on pre-treatment of the straw and digestion conditions. 4. Conclusions The chemical composition and the methane productivity in manure in terms of VS, volume and livestock production were determined. Furthermore,
the methane productivity in terms of VS and volume in straw and physical fractions of manure were also determined. The theoretical methane productivity is higher in pig (516 l kg−1 VS) and sow (530 l kg−1 VS) manure than in dairy cattle manure (468 l kg−1 VS) but is still relatively constant, while the ultimate methane yield in terms of VS is considerably higher in pig (356 l kg−1 VS) and sow manure (275 l kg−1 VS) than in dairy cattle manure (148 l kg−1 VS). Methane productivity based on livestock units (LU) shows the lowest methane productivity for sows (165 m3 CH4 LU−1 ), while the other animal categories are in the same range (282– 301 m3 CH4 LU−1 ). Variations in feed intake and feed digestibility can play an essential role in the LU-speciHc yields, and the actual production data are needed in order to make more precise evaluations of the energy production and economics of a biogas installation. Pre-treatment of manure by separation is a method of making fractions of the manure with higher gas potential in terms of volume, since the water can be drained from the solids, thus creating fractions with a higher VS concentration; hence the volumetric methane productivity is 2.6 –3.5 higher in the physical fractions of manure compared with undiluted manure from fattening pigs or 5 – 6.6 higher than undiluted sow manure. Theoretical methane potential and biodegradability of three types of fractions deriving from manure separation were tested. The fractions deriving from centrifugation (SFD), chemical precipitation (SFC) and evaporation (E-concentrate) have almost the same theoretical methane yield as pig manure in terms of VS, but the ultimate methane yield in terms of VS is lower for SFD (210 l kg−1 VS) and SFC (247 l kg−1 VS) than for pig manure, while it is considerably higher in the E-concentrate (506 l kg−1 VS). The methane productivity by separation of manure, in terms of LU, will depend on the separation eOciency of VS and, compared with untreated manure, it will be reduced unless all the fractions are digested. The volumetric methane yield of straw is higher than the yield from both manure and solid fractions from manure, due to the higher VS content, and therefore the use of straw as a bedding material will increase the volumetric as well as the livestock-based methane
H.B. M3ller et al. / Biomass and Bioenergy 26 (2004) 485 – 495
productivity; thus every kilogram of straw added per 100 kg manure will increase methane productivity by 10%. Acknowledgements Thanks to Merete Maahn for her thorough and careful help during performance of the test. This paper is supported by grants from The Danish Energy Council. References [1] Hashimoto AG, Chen YR, Varel VH. Theoretical aspects of anaerobic fermentation: state of the art. In: Livestock waste: a renewable resource. Proceedings of the Fourth International Symposium on Livestock Wastes. ASAE, St Joseph, Michigan, 1981. p. 86 –91. [2] Husted S. Seasonal variation in methane emission from stored slurry and solid manures. Journal of Environmental Quality 1994;23:585–92. [3] Steed J, Hashimoto G. Methane emissions from typical manure management systems. Bioresource Technology 1994;50:123–30. [4] Sommer SG, M$ller HB, Petersen SO. Reduction in methane and nitrous oxide emission from animal slurry through anaerobic digestion. In: Proceedings of the Third International Symposium, 21–23 January, Maastricht, The Netherlands, 2002. [5] Hill DT. Methane productivity of the major animal waste types. Transactions of the ASAE 1984;27(2):530–4. [6] Symons GE, Bushwell AM. The methane fermentation of carbohydrate. Journal of the American Chemical Society 1933;55:2028–39. [7] Angelidaki I, Ahring BK, Ahring BK. Methods for increasing the biogas potential from the recalcitrant organic matter contained in manure. Water Science and Technology 2000;41(3):189–94. [8] Hansen KH, Angelidaki I, Ahring BK. Anaerobic digestion of swine manure: inhibition by ammonia. Water Resources 1998;32(1):5–12. [9] M$ller HB. Separation af slagtesvinegylle med Ansager SepTec gylleseparator. (Separation of pig manure with Ansager SepTec separator.) Internal Report No. 159. Danish Institute of Agricultural Sciences, 2002. [10] IPCC. Guidelines for National Greenhouse Gas Inventories: Reference Manual. Revised 1996. IPCC Guidelines, 1997. [11] M$ller HB, Sommer SG, Ahring BK. Separation eOciency and particle size distribution in relation to manure type and storage conditions. Bioresource Technology 2002;85(2): 189–96.
495
[12] M$ller HB, Hansen MN, Sommer SG. Heavy metal load from fractions of manure after separation. Bioresource Technology 2003; submitted for publication. [13] ISO. Water quality: evaluation of the ‘ultimate’ anaerobic biodegradability of organic compounds in digested sludge— method by measurements of the biogas production. International Standard, ISO/DIS 11734, 1995. [14] APHA. Standard methods for the examination of water and wastewater, 19th ed. Washington, DC: American Public Health Association; 1995. [15] Ianotti EL, Porter JH, Fisher JR, Sievers DM. Changes in swine manure during anaerobic digestion. Developments in Industrial Microbiology 1979;20(49):519–29. [16] Converse JC, Zeikus RE, Graves RE, Evans GW. Anaerobic degradation of dairy manure under mesophilic and thermophilic temperatures. Transactions of the ASAE 1977;20:336. [17] Haugen VJ, Lindley JA. Biogas production after solid-liquid separation of dairy manure. Transactions of the ASAE 1988;31(6):1782–6. [18] M$ller HB, Sommer SG, Ahring BK. Biological degradation and greenhouse gas emissions during pre-storage of liquid manure. Journal of Environmental Quality 2004;(33): Jan–Feb. [19] Andara AR, Esteban JML. Kinetic study of the anaerobic digestion of the solid fraction of piggery slurries. Biomass and Bioenergy 1999;17:435–43. [20] Hartmann H, Angelidaki I, Ahring BK. Increase of anaerobic digestion of particulate organic matter in full-scale biogas plants by mechanical separation. Water Science and Technology 2000;41(3):145–53. [21] Gunaseelan VN. Anaerobic digestion of biomass for methane production: a review. Biomass and Bioenergy 1997;13: 83–114. [22] Torres-Castillo R, Llabres-Luengo P, Mata-Alvaraz J. Temperature e>ect on anaerobic digestion of bedding straw in a one-phase system at di>erent inoculum concentration. Agriculture Ecosystems Environment 1995;54: 55–66. [23] Sharma SK, Mishra IM, Sharma MP, Saini JS. 1988. E>ect of particle size on biogas generation from biomass residues. Biomass 1988;17:251–63. [24] Hashimoto AG. E>ect of inoculum/substrate ratio on methane yield and production rate from straw. Biological Wastes 1987;28:247–55. [25] Poulsen HD, B$rsting CF, Rom HB, Sommer SG. KvXlstof, fosfor og kalium i husdyrg$dning. (Nitrogen, phosphorus and potassium contained in livestock manure.) DIAS Report No. 36, 2001. [26] Ministry of Food, Agriculture and Fisheries. Vejledning og skemaer, mark og g$dningsplan, g$dningsregnskaplantedXkke og harmoniregler. (Guidance concerning fertilizer, nutrient account and harmony rules.), 2002.