This action might not be possible to undo. Are you sure you want to continue?
Characterisation of solid and liquid fractions of dairy manure with regard to their component distribution and methane production
´ , H. Garcıa a, C. Rico b, I. Tejero
a Department of Chemical Engineering and Inorganic Chemistry, University of Cantabria, Los Castros s/n, 39005 Santander, Spain Department of Sciences and Techniques of Water and Environment, University of Cantabria, Los Castros s/n, 39005 Santander, Spain
Received 18 October 2004; received in revised form 21 November 2005; accepted 26 April 2006 Available online 15 June 2006
Abstract Dairy manure with a total solids content of 77.2 g TS/l was separated by means of screening and coagulation–ﬂocculation treatments, using CaO as coagulant and a cationic polyacrylamide as ﬂocculant, obtaining liquid and solid fractions. The solid fraction separated contained 33.4% of the initial total mass of dairy manure plus chemical solutions, containing also 75.2% of the TS, 80.4% of the VS, 58.5% of the total Kjeldahl nitrogen (TKN) and 87.4% of the total phosphorus (PT) present in the initial dairy manure. 83.7% of the liquid fraction chemical oxygen demand (COD) was anaerobically biodegradable (CODBD). Methane production for the separated liquid fraction was 0.604 l CH4 NCTP/g VS added, being 0.307 and 0.371 l CH4 NCTP/g VS added for dairy manure and screened dairy manure, respectively. The characteristics of this liquid fraction would allow its treatment in high loading anaerobic reactors having shorter hydraulic retention times, smaller reactor size and a higher methane volumetric production rate than conventional anaerobic reactors treating either manure or screened manure. Ó 2006 Elsevier Ltd. All rights reserved.
Keywords: Dairy manure; Dewatering; Solids removal; Liquid fraction; Polyacrylamide; Anaerobic biodegradability; Methane production
1. Introduction The composition of cattle waste depends on the conditions under which animals are kept on the farm (dairy cows, breeding cows and calves). Cattle manure extracted from cowhouses contains residues of food and bedding (straw, sand, sawdust, etc.) in addition to solid and liquid animal wastes. Many of these solids are either not biodegradable or only slowly so. The waste can be collected by means of either mechanical or hydraulic scraping of the ﬂoor and dragging of the material into the manure pit. Separation of liquid and solid fractions of the waste is a desirable upstream operation in the treatment process: dewatering the solid fraction lowers the cost of shipping and increases the energetic yield in combustion processes. When the separated solid fractions have an adequate TS
Corresponding author. Tel.: +34 942201599; fax: +34 942201591. E-mail address: firstname.lastname@example.org (J.L. Rico).
content and VS/TS ratio, then they are suitable for production of compost (Lo et al., 1993) or energy (Møller et al., 2004). If the separated liquid fraction, which features the largest volume, presents low solids and ammonium contents, it can be spread on ﬁelds during most of the year without causing environmental problems. After separation, it will be possible to apply the concentrated dry matter and nitrogen-rich fraction during the growing crops season (Henriksen et al., 1998). When manure is collected by washing instead of scraping and dragging the ﬂoors, it is more dilute and dewaters more easily through a sieve. There seems to be a rinsing action on the sieve: the more plentiful ﬂuid favours detachment of ﬁne and colloidal solids from longer ﬁbres retained by the screen and these pass through the sieve. Thus, sieved liquid presents a higher volatile solids content in comparison with sieved manure collected from the animal house by scraping and dragging. Lo et al. (1983) proved that liquid–solid screening separation had a minimal eﬀect on the rate of methane
0960-8524/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2006.04.032
J.L. Rico et al. / Bioresource Technology 98 (2007) 971–979
production from mesophilic anaerobic digestion of dairy manure for equal volatile solids loading rates at 16 days hydraulic retention time (HRT). For shorter HRT a signiﬁcant increase was found in the biogas production rate per litre of digester or per gram volatile solids added for the screened manure over unscreened manure. Anaerobic digestion of screened dairy manure, with 3.78% TS, showed that optimum positive net energy yields of 0.213 l CH4/g VS added can be obtained at 30 °C and 20 days HRT, according to Hawkes et al. (1984). Liao and Lo (1985) studied the technical feasibility of adopting the ﬁxed-ﬁlm reactor concept for biogas production from screened dairy manure, with 4.4% TS, compared with the conventional reactor at 35 °C. Maximum methane volumetric production rate for the conventional reactor was 0.63 l CH4/l day at 6 days HRT. For the ﬁxed ﬁlm reactor the maximum production rate was 3.53 l CH4/l day when operated at 3 h HRT (loading rate 262 g VS/l day and methane yield 0.0135 l CH4/g VS added). Lo et al. (1985a) studied the performance of ﬁxed-ﬁlm reactors treating screened dairy manure at 22 °C on the basis of methane productivity and biodegradation eﬃciency. For 5.2% VS screened dairy manure feeds, maximum methane volumetric production rate was 1.29 l CH4/l day at 1 day HRT and maximum methane yields of 0.10 l CH4/g VS added occurred at 16 days HRT. Lo et al. (1985b) compared anaerobic digestion process results for screened dairy manure (3.4% and 5.2% VS) under mesophilic (35 °C) and thermophilic (55 °C) conditions. For the 3.4% VS at 35 °C, 0.63 l CH4/ l day were obtained as maximum methane volumetric production rate at 6 days HRT, and 0.152 l CH4/g VS added as maximum methane production at 12 days HRT. The same feed at 55 °C gave 1.03 l CH4/l day as maximum methane volumetric production rate at 2 days HRT, and 0.149 l CH4/g VS added as maximum methane production at 15 days HRT. By comparison of two reactors fed with 3.4% VS manure, for HRT’s of 6 days and greater, the gas production rates showed that there was no beneﬁt in operating at the thermophilic temperatures. Mesophilic digestion of screened dairy manure using an anaerobic rotating biological contact reactor, plus three sequencing batch reactors for the subsequent aerobic process, achieved a chemical oxygen demand reduction of 98% (Lo and Liao, 1986). In the anaerobic reactor, the maximum methane productivity was 1.89 l CH4/l day at 1 day HRT, with the highest methane yield of 0.093 l CH4/g VS added being obtained at 11 days HRT (Lo et al., 1986a). When screened dairy manure was treated using one-phase or two-phase anaerobic digestion systems, maximum production rates of 1.50 and 2.32 l CH4/l day were achieved, respectively, at 1 day HRT (Lo et al., 1986b). Liao and Lo (1987) reported that methane volumetric production rates were very similar for both the screened manure and supernatants from the sedimentation process. Kalyuzhnyi et al. (1999) reported data from the treatment of liquid cattle manure obtained by dilution and solid–liquid separation from 15–8% TS to 7–4% TS in a UASB reactor. Total COD
reduction of 41.5% and a methane volumetric production rate of 0.70 l CH4/l day at a maximum organic loading rate (OLR) of 6.63 g COD/l day at 1 day HRT were obtained. Excepting the data reported by Lo and Liao (1986), COD removal percentages were always below 50%, in accordance with the biodegradability of both unscreened and screened manure. Taking into account the diﬀerent kinetics of hydrolytic– acidogenic and methanogenic stages, Nozhevnikova et al. (1999) proposed a two-step anaerobic manure treatment in which both sanitation of manure and energy saving were present: (i) acidogenic fermentation at high temperature, (ii) solid and liquid fractions separation, (iii) treatment of liquid fractions under low temperature conditions. Previous experiments to separate liquid and solid fractions have been performed by Burcham et al. (1997), Gilbertson and Schulte (1987) and Mohri et al. (2000) with mechanical separators and settling tanks. If a better liquid–solid separation is desired, coagulation–ﬂocculation is a suitable way to attain this goal. In addition to the mechanical separation of liquid and solid fractions in animal wastes, salts of Fe3+ and Al3+ and CaO have been used together with polymers as methods to improve the eﬃciency of coagulating and settling ﬁne and colloidal solids. Barrow et al. (1997), Bolto et al. (1999), Vanotti and Hunt (1999), Shang-Kyu and Gregory (2001) and Sievers et al. (1994) have reported coagulation–ﬂocculation experiences with animal wastes. Zhang and Lei (1998) have reported the use of salts together with polymers to improve the removal of phosphorous considerably. McKenney (1998) tested aluminium sulphate as coagulant to improve the collection of solids, followed by sedimentation and dewatering of the sludge in a belt ﬁlter press. The main diﬃculty that appeared was that the high solids content in the sludge (10–13%) hindered the operation of the pumps. After various changes in the process, pumps and ﬁlters, solids removal eﬃciencies reached values between 71% and 78%. Removal of nutrients stayed around 70% for phosphorous and 40% for nitrogen. Cantabria, a region in Northern Spain, has about 350,000 dairy cows which generate around 4,500,000 tonne/year of semi-liquid manure (7–14% TS). Most of the intensive farming is operated on the coastal ﬂatlands, which are tourism areas. Manure storage capacity is limited and there is not enough land available for waste disposal by direct application during those periods when the soil beneﬁts from nutrient additions, which leads to environmental damage. Williamson et al. (1998) came to the conclusion that intensive-production farms should use specialised technologies if they wish to keep operating in sensitive zones, where the traditional method of using wastes as land fertilizer is either not applicable or exceeds the land acceptance capacity. In humid areas, particularly where the ground shows pronounced slopes, as happens in many zones of Cantabria, the liquid fraction of manure is responsible for part of the pollution found in superﬁcial and underground waters. The intention to develop solutions for this problem was the origin of
J.L. Rico et al. / Bioresource Technology 98 (2007) 971–979
the present work. The overall process undergone by the manure was considered: once the initial hydrolytic and acidogenic stages had taken place in a controlled manure pit, solid and liquid fractions of manure could be separated and then stabilised by means of biological treatments. Composting is an attractive alternative for the solid fraction. Once this material has been composted it can be used as a soil conditioner. If the liquid fraction presents most of its organic matter in soluble form and a higher percentage of biodegradable chemical oxygen demand (CODBD) than raw manure, then it could be treated in high-load anaerobic reactors. High load anaerobic reactors operate with low HRT. Microorganisms can only act on soluble matter. It is necessary that the major part of the total chemical oxygen demand (CODT) be in soluble form (CODﬁlt). The high-load reactors have lower volume requirements than continuous stirred tank reactors for the same organic load. However, treated eﬄuents from anaerobic processes cannot be discharged to public water bodies without further treatment, according to present environmental regulations. Additional decontamination processes have to be applied to the eﬄuent, alone or mixed with other wastewaters. This study was focused on the separation of solid (SF) and liquid (LF) fractions of dairy manure by means of screening and later coagulation–ﬂocculation. One of the objectives was to obtain a liquid fraction in which organic matter was mainly in soluble form, presenting a high anaerobic biodegradability. The other objective was that TS, total Kjeldahl nitrogen (TKN) and total phosphorus (P2O5) stayed retained in the solid fraction. Distribution of TS, VS, TKN and P2O5 in solid and liquid fractions was analysed. Chemical characteristics of LF and its anaerobic biodegradability were measured by their corresponding assays. Speciﬁc methane yields of the raw manure, screened and liquid fractions thus separated were determined as well. 2. Methods 2.1. Manure collection and experimental procedure For this work, manure with a mean TS concentration of 7–8% was collected from a farm in the Santander area, where 400 cows were present. Half of them were involved in dairy production. Once collected from the farm, manure was kept refrigerated at 4 °C. The raw manure was screened by means of a 1 mm mesh light sieve, and later the screened manure was subjected to coagulation–ﬂocculation. Operating conditions adopted were: (1) coagulation—addition of coagulant solution into the screened manure and mixing for 2 min at 175 rpm; (2) ﬂocculation—addition of polyacrylamide solution into the mixture and new mixing for 13 min at 50 rpm for ﬂocs formation; (3) solids settling—settling down, either for 2 h when clariﬁed supernatant was extracted, or for 5 min when a laboratory scale ﬁlter press was employed for dewatering the solid fraction.
2.2. Analytical methods Volatile fatty acids (VFA) were determined using a HP6890 gas chromatograph (GC) ﬁtted with a 2 m · 1/ 8 in glass column, liquid phase 10% AT 1000, packed with the solid support Chromosorb W-AW 80/100 mesh. Nitrogen was the carrier gas, ﬂow rate 14 ml/min, and a FID detector was installed. Gas composition was assayed on a 2 m Poropak T column in a HP 6890 GC system with helium as carrier gas, ﬂow rate 15 ml/min, and TCD detector. Cellulose, hemicellulose and lignin were determined according to Van Soest et al. (1991). All other analyses were performed following the Standard Methods for the Analysis of Waters and Wastewaters (APHA, 1995). Determination of biodegradability was performed according to the method described by Field et al. (1988). The calculated percentages of COD were the result of dividing the respective COD values by the initial value of the total chemical oxygen demand (COD0). CODVFA and CODCH4 are the percentages due to volatile fatty acid concentration and methane yield, respectively. %CODVFA ¼ CODVFA =COD0 ; %CODCH4 ¼ CODCH4 =COD0 : It is assumed that %CODacid ¼ %CODCH4 þ %CODVFA ; %CODBD ¼ %CODacid þ %CODcel ; where acid chemical oxygen demand is the addition of COD due to VFA present and CH4 produced. The percentage of biodegradable chemical oxygen demand is the sum of percentage CODacid and the COD used in biological reactions for biomass generation (CODcel). The CODcel is the sum of that employed in the growth of both the acidogenic and methanogenic biomass. %CODcel ¼ ð%CODcel Þacid þ ð%CODcel Þmeth ; ð%CODcel Þacid ¼ Y A =ð1 À Y A Þ Ã ð%CODacid À %CODVFA Þ; Y A ¼ 0:196 g CODcel =g CODused ; ð%CODcel Þmeth ¼ Y M =ð1 À Y M Þ Ã %CODCH4 ; Y M ¼ 0:028 g CODcel =g CODused ; where YA and YM are the estimated yield cellular coeﬃcients for acidogenic and methanogenic bacteria, respectively. To determine %CODBD, the value obtained on day 7 was considered. 2.3. Apparatus A laboratory scale stationary screen with a sieve with mesh size 1 mm, with a manual press piston was used. Jar-test coagulation–ﬂocculation equipment was used to perform the coagulation–ﬂocculation experiments.
J.L. Rico et al. / Bioresource Technology 98 (2007) 971–979 Table 2 Mean composition of elements in the screened dairy manure at the farm (mg/l) Sample Mean St. Dev. N–NHþ 4 716 45 TKN 2,198 117 P2O5 707 70 K2O 1,801 79 CaO 695 65 MgO 361 24
To measure the biodegradability and the methane production, cylindrical reactors, made of PVC, with an internal diameter of 22 cm (2.5 l total volume and 2.0 l useful volume), were used. Stirring, at 20 rpm, was supplied for 20 s every 15 min by means of blade stirrers controlled by timing devices employed during anaerobic biodegradability tests. Gas production was measured by means of a displacement system using an alkaline solution to absorb the CO2 produced. All the measurements are expressed at 20 °C and standard pressure of 760 mmHg (NCTP). 2.4. Chemical reagents Ferric chloride (FeCl3) and calcium oxide (CaO) were used as coagulants. The following Fennopol polyacrylamides supplied by Kemira Iberica S.A were employed: A medium anionic A305 (MAP); two medium cationic, K504 (MCP1) and K508 (MCP2); a strong cationic C-598 (SCP). Concentrations of the stock additive solutions before mixing with sludge were 100 g FeCl3/l, 50 g CaO/l and 1.00 g /l for all kinds of polyacrylamides. 3. Results and discussion 3.1. Screening The compositions of the main parameters of the manure screened through an industrial 1.5 mm mesh sieve are shown in Table 1. Since washing a cowhouse uses water, manure pumped to the sieve came from the slurry pit. When dairy manure is separated into liquid and solid fractions, the VS/ TS ratio diminishes in the liquid fraction, which indicates an increase in the percentage of inorganic compounds. The values in Table 2 show the concentrations of the main elements in this screened manure. In Table 3, the compositions of manure and screened manure in terms of fat, protein and carbohydrates as percentages of dry matter are shown. In Table 3 it can also be seen that cellulose and hemicellulose were the main compounds removed by screening, whereas lignin was only minimally removed. The screened manure increased its percentage compositions of fat, protein and inorganic compounds. The values for manure were in accordance with those reported by Zeeman (1991). 3.2. Coagulation–ﬂocculation The eﬃciencies of CaO and FeCl3 as coagulants in the removal of ﬁnely divided, suspended organic matter from
Table 3 Composition, as %, of the total solids of raw and screened dairy manure Manure Raw Screened Fat 5.7 7.3 Protein 14.8 18.5 Cellulose 24.2 6.0 Hemi cellulose 20.9 4.7 Lignin 13.2 18.7 Inorg. residue 20.7 28.6
the previously screened dairy manure were tested separately. Other iron salts such as FeSO4 and Fe2(SO4)3 were not used, in accordance with the results obtained by Barrow et al. (1997). The results obtained showed that additions of the chemicals mentioned above did not allow total solids removal eﬃciencies higher than 15% in supernatants of screened dairy manure with initial TS contents of 4%. The combined addition of FeCl3 and CaO did not improve the results. To test the eﬃciency of various ﬂocculants, screened dairy manure was treated with FeCl3 and diﬀerent polyacrylamides (PAM). In order to study the eﬀect of low PAM doses and decrease the reactive chemical cost, reactive ﬂocculant (3.00 €/kg) being 50 times more expensive than reactive coagulant (0.060 €/kg), the dosage ratio of FeCl3/PAM was maintained in all cases at 20:1 (w/w). Fig. 1 shows the results obtained for the removal of TS with diﬀerent ﬂocculants. All four combinations of FeCl3/polyacrylamides achieved TS removal eﬃciencies higher than 60%, but the medium cationic polyacrylamides MCP1 and MCP2 were the ones that gave the best results. The medium anionic polyacrylamide MAP gave the worst results. The strong cationic polyacrylamide SCP gave intermediate results. The MCP1 was the ﬂocculant that gave the best removal eﬃciencies in combination with FeCl3. Similar removal eﬃciencies were found for VS and COD. Vanotti and Hunt (1999) also reported that medium cationic polyacrylamides gave the best results. When CaO was used as coagulant together with the four polyacrylamides at the same dosage ratio of CaO/PAM 20:1 (w/w), MCP1 was again the polyacrylamide that gave the highest percentage removals for TS, VS and COD, data not shown. Percentage removals were similar to those obtained when FeCl3 was employed as coagulant.
Table 1 Mean characteristics of the screened dairy manure at the farm pH Mean Std. Dev. 7.0 0.07 TS 39,854 4870 VS 28,372 3548 CODT 48,026 5933 CODcent 18,370 3380 CODﬁlt 12,942 2577 CODVFA 5786 687 TOC 5589 564
All parameters except pH in mg/l.
J.L. Rico et al. / Bioresource Technology 98 (2007) 971–979
MAP MCP1 MCP2 SCP
72 70 68 66 64 62 60 2000 2500 3000 3500
Fig. 1. TS removal eﬃciencies adding FeCl3 and various ﬂocculants. Dosage ratio 20:1 (w/w) FeCl3/PAM.
In accordance with the results reported by Rico et al. (2006), MCP1 doses were increased whereas coagulant doses were diminished. The screened manure was treated with CaO and MCP1 in a 1/1 (w/w) ratio. MCP1 concentration was increased up to 3 g/l so as to prevent dilution. Doses are expressed as percentage TS concentration. As can be seen in Fig. 2, the percentages of TS and VS removed increased linearly up to 2% TS doses of CaO and MCP1. From this point, further rises in reactive doses only produced small increases in percentage removals. 3.3. Mass balance A dairy manure sample, taken from the cowhouse of the farm, was kept in a laboratory-scale manure pit for 30 days
% TS and VS removed
87 85 83 81 79 77 75 0 1 2 3 4
MCP1 and CaO (% TS)
Fig. 2. Percentage of TS (ﬁlled points) and VS (blank points) removed.
at laboratory temperature (20–23 °C). Previous experiments had shown that after 30 days of storage at ambient temperature, VFA concentration had increased while no methanization had occurred, so only the development of the hydrolytic and acidogenic stages had been allowed (Rico et al., in preparation). 10 kg of that dairy manure were screened through a 1 mm mesh sieve. The screened manure was subjected to a coagulation–ﬂocculation process. The selected doses were 10 mg CaO/g TS and 20 mg MCP1/g TS. The CaO/MCP1 ratio was decreased down to 1/2 (w/w) to ensure a lower pH increase in the separated liquid fraction, allowing its anaerobic digestion without pH adjustment. Concentration of the MCP1 solution in this test was 3.00 g/l again. A laboratory scale ﬁlter press was employed for dewatering the solid fraction after ﬂocculation. Analyses for determination of the chemical compositions of raw manure and also solid and liquid fractions separated by means of mechanical and chemical treatment were done in triplicate; the values of each parameter were accepted if none had a deviation higher than 5%. In such cases, analytical determinations were carried out again. The results of the separation process for 1.0 kg of dairy manure can be seen in Table 4. The screening separated 19.6% of the initial mass of manure as solids (solid 1 fraction) and 79.1% as screened manure. The separated solid 1 fraction contained 45.9% of the initial TS and 51.3% of the initial VS. It also contained 20.8% of the initial TKN and 45.1% of PT. The VS/TS ratio was 0.78 for dairy manure, being 0.87 and 0.72 for solid 1 fraction and the screened manure, respectively. Powers et al. (1995) studied the distribution of TS, TKN and PT for diﬀerent sizes of particle in dairy manure separated by screening, pointing out that 60% of the TS, 86% of the TKN and 94% of the PT passed through a 0.50 mm sieve. Møller et al. (2004) reported that particles >1 mm were low in phosphate, compared to small particles. The percentages of TS and TKN found in the liquid fraction passed through a 1 mm mesh sieve in the present study were a little lower than those reported by Powers et al. (1995). In the case of PT this diﬀerence was quite large, changing from 54.9% to 94%. A possible reason is that Powers et al. (1995) worked with dairy manure of 1.5% TS, whereas the present work was performed with dairy manure of 7.7% TS. As TS content increases, the solid fraction separated using a laboratory-scale sieve is associated with a greater mass of water. In addition, the water
Table 4 Dairy manure, solid and liquid fractions: composition (g/kg) and percentage distribution in relation to initial dairy manure mass Manure (g/kg) TS VS TKN PT Mass 77.163 ± 0.458 59.882 ± 0.407 3.078 ± 0.104 0.557 ± 0.0148 1000 g Solid 1 (g/kg) 180.302 ± 0.0860 156.317 ± 0.209 3.253 ± 0.0215 1.280 ± 0.0067 196.4 g Solid 1 (%) 45.9 51.3 20.8 45.1 19.6 Screened (g/kg) 49.940 ± 0.226 35.843 ± 0.133 3.069 ± 0.112 0.383 ± 0.0097 790.6 g Solid 2 (g/kg) 124.705 ± 0.0564 95.837 ± 0.122 6.223 ± 0.0396 1.257 ± 0.0062 186.8 g Solid 2 (%) 29.3 29.1 37.8 42.3 13.9 Liquid (g/kg) 19.985 ± 0.0197 12.040 ± 0.0343 1.401 ± 0.036 0.069 ± 0.0016 867.0 g Liquid (%) 21.8 17.0 39.5 10.8 64.1
% TS removed
J.L. Rico et al. / Bioresource Technology 98 (2007) 971–979
employed in the farm had high hardness characteristics, because it was collected from subsoil and the land was limy. Precipitation of calcium phosphate may be the reason for this high PT removal in the screening operation. Coagulation–ﬂocculation treatment was applied to the screened manure, 2.0% TS content dose of MCP1. The mass of the solid 2 fraction separated was 17.5% of the screened manure and the chemical mass. It contained 57.3% and 61.5% of the TS and VS of the screened manure, respectively. 47.9% of TKN and 77.9% of P2O5 of the screened manure also remained separated in this solid fraction. The VS/TS ratio was 0.77 for the solid 2 fraction and 0.60 for the liquid fraction. When both solid fractions, solid 1 and solid 2 were mixed, 33.4% of the initial mass of dairy manure and chemicals added were separated in the solid fraction, which contained 75.2% and 80.4% of the initial TS and VS respectively, as well as 55.8% of the initial TKN and 87.4% of the initial PT. The remainder of the initial components of the manure remained in the liquid fraction. TS, TKN and PT removed were in the same range as those reported by Powers et al. (1995), after coagulation–ﬂocculation and screening operations for 1.5% TS dairy manure. Removal percentages were higher than those obtained by centrifugation referred to by Møller et al. (2004). In Table 5, the characteristics of the screened manure and the liquid fraction can be seen. TS concentration was reduced from 49,940 mg/l in the screened manure to 19,985 mg/l in the liquid fraction. As can be seen, in the case of the screened manure the CODT and CODﬁltrate concentrations were 47,805 and 24,107 mg/l, respectively. After the process ﬂocculation the CODT concentration was 25,079 mg/l, while the CODﬁltrate value was 21,636 mg/l. The percentage of CODT due to CODﬁltrate changed from 50.4% to 86.3% after the ﬂocculation treatment. VFA are biodegradable compounds transformable to methane, whose elimination through the coagulation– ﬂocculation process is not desirable. Regarding COD due to VFA (CODVFA), the liquid fraction concentration was 90.7% of the screened manure value, as a result of the dilution produced by the addition of the ﬂocculant. CODﬁltrate is due to soluble compounds, VFA and others. There were three main VFA which contributed to CODVFA: acetic acid, 59.9%; propionic acid, 22.1% and butyric acid, 12.0%. These percentages are in accordance with those found by Hill (1970) and Patni and Jui (1985). On the other hand, as can be seen in Table 5, TKN and P2O5 concentrations in the liquid fraction decreased from 3069 to 1401 mg/l and from 880 to 159 mg/l, respectively. This decrease in nutrient concentration in the liquid fraction
is advantageous because during anaerobic digestion the removal of nitrogen and phosphorous nutrients is minimal, being due only to cellular synthesis, struvite (MgNH4PO4) precipitation and NH3 stripping, owing to a pH increase. The liquid fraction, according to data in Table 5, with about 86.3% of its CODT in soluble form (CODﬁltrate) and a high percentage of this CODﬁltrate due to VFA (CODVFA) 79.6%, forecast a high percentage of COD removal and its conversion to methane. The TS content of the ﬁnal solid fraction, shown in Table 4, was lower than that suitable (30–35%) for the development of a composting process, being necessary mixing it with other wastes with higher solids content such as straw and sawdust to make its treatment by the conventional aerobic composting process possible. Other options for this solid fraction are the semi-liquid anaerobic process in accumulative systems and/or its dewatering in electrical energy generation plants by means of cogeneration. For an industrial application, a belt ﬁlter press, screw press or decanter centrifuge could be necessary for dewatering the solid fraction and producing a liquid fraction free of suspended solids. To facilitate the separation process, development of a ﬂocculation process without previous screening, with higher TS content than screened manure, would allow an easier solid–liquid separation with the employment of just a ﬂocculant (O’Mellia, 1972). In this case the resultant SF would be more compact and only a small mesh sieve, 0.250 mm for example, would be necessary for solid–liquid separation. This operation implies greater reactive consumption due to higher TS content in the manure and as a result of that an increase in costs. Depending on the origin of manure, time and conditions of storage, some diﬀerences in the composition and distribution may appear after solid–liquid separation. With longer time and higher temperature, more organic matter will be in soluble form, organic nitrogen will be converted into ammonia and biodegradable organic matter will be transformed into VFA. If methanization took place, VFA in the liquid fraction would disappear as the result of CH4 formation. 3.4. Methane production The test of methanogenic production at 35 °C was carried out for the dairy manure, the screened manure and the liquid fraction obtained from the laboratory scale manure pit. The experiments were performed in duplicate. Six reactors of 2.5 l, as described above, were employed. Each reactor was loaded with 500 g of stabilized biomass
Table 5 Main characteristics of the screened manure and the liquid fraction Sample Screened Liquid f. pH 6.6 6.8 TS 49,940 19,985 VS 35,843 12,040 CODT 47,805 25,079 CODﬁlt 24,107 21,636 CODVFA 18,987 17,224 TKN 3069 1401 P2O5 880 159
All in mg/l except pH.
J.L. Rico et al. / Bioresource Technology 98 (2007) 971–979
obtained from previous assays and 1500 g of manure, screened manure or liquid fraction, respectively. Two 1.0 l bottles made of high density propylene, the tops of which were perforated in order to permit recovery of biogas, were seeded with 500 g of the same stabilized biomass and 250 ml of tap water and used as blanks. In all cases, air was removed from the reactors by passing biogas through them. The eight reactors were placed in two thermostatic vessels, four into each one, in order to get a temperature of 35 °C in the reactors. For 45 days the biogas produced in each reactor was collected in a gas meter as described above. The results shown are the mean values found in the two reactors with the same kind of waste. The deviations found were bigger for manure, but smaller than 2.4%. On day 45, the mean methane volumes accumulated, measured at NCTP (20 °C and 1 atm), were 18.371, 13.307 and 7.270 l CH4/kg for the dairy manure, the screened manure and the liquid fraction, respectively. In Fig. 3, the evolution of the speciﬁc methane production for each sample can be observed. These values were calculated from accumulated methane volumes and the values of VS in Table 4. The ﬁnal values were: 0.307, 0.371 and 0.604 l NCTP CH4/g VS added for the dairy manure, the screened and the liquid fractions, respectively. Saﬂey et al. (1992) found a yield of 0.24 m3 CH4/kg VS added for dairy manure. The value found in the present assay (0.307 m3 CH4/kg VS added) was higher than that. It must be taken into account that when VS content was measured in dairy manure, there was a high fraction of volatile organic components (VFA). The VS measurement includes drying (100 °C) and incineration (550 °C). According to Derikx et al. (1994), during drying operations losses of VFA can be expected for dairy manure depending on the sample pH value. It can be estimated that only 25% of the VFA present would have been included in the VS measurement, due to sample conditions. In addition, methane productions were minimised in the controlled manure pit. For these reasons, the speciﬁc methane productions for these samples of dairy manure, screened and liquid fractions were higher. Since hardly any biodegradable components were removed by means of physical–chemical treatment,
the speciﬁc methane production increased for the screened manure and mainly for the liquid fraction. The speciﬁc methane production reached a maximum value of 0.604 l CH4/g VS added for the liquid fraction and an intermediate value for the screened manure, albeit close to that of the dairy manure. One kilogram of dairy manure from the laboratory scale manure pit gave 18.371 l CH4 (NCTP). After screening 1 kg of dairy manure, 0.791 kg of screened manure was obtained, producing 10.525 l CH4 (NCTP). When this 0.791 kg of screened manure was subjected to the coagulation–ﬂocculation process, 0.867 kg of liquid fraction was obtained, which gave 6.303 l CH4 (NCTP). Separation of liquid and solid fractions by means of screening and coagulation–ﬂocculation decreases the total production of methane per unit mass for the liquid fraction separated, though the speciﬁc methane production increases. On the other hand, the respective solid fraction methane production per unit mass increases but speciﬁc methane production decreases. The important issue is the fact that the time required to methanize the biodegradable organic matter for the screened manure and the liquid fraction diminished as the percentage of poorly biodegradable organic matter diminished. In this assay, the amounts of CH4 produced after 8 days, expressed as percentages of the ﬁnal productions, were 38% for the manure, 46% for the screened manure and 90% for the liquid fraction. The dairy manure and screened manure needed 24 and 18 days respectively to produce 90% of the ﬁnal methane yield. The methane yield also depends on the origin of the manure, time and conditions of storage, pretreatments and the amount of seeding employed in methanogenic production assays. Increasing active biomass seeded increases the organic matter employed in biochemical methanogenic reactions, lessening that used in cell synthesis processes. In this way, methanogenic productivity will rise. For diﬀerent samples of dairy manure and changing storage times, the values for speciﬁc methanogenic productivity found at 35 °C were the following: 0.193–0.321 l NCTP CH4/g VS added for raw dairy manure, 0.287–0.378 l NCTP CH4/g VS added for screened manure and 0.462–0.635 cm3 NCTP CH4/g VS added for liquid fraction. 3.5. Anaerobic biodegradability of ﬂocculated dairy manure
600 500 400 300 200 100 0 0 10 20 30 40
Fig. 3. Speciﬁc methanogenic productions at 35 °C.
The same liquid fraction obtained by treatment of 10 kg of dairy manure from the laboratory scale manure pit was subjected to an anaerobic biodegradability test at 35 °C. In this way it was possible to know the percentage of anaerobically biodegradable COD in the liquid fraction. Two reactors were used to determine the anaerobic biodegradability of the liquid fraction. One reactor operated as a blank (500 g of biomass and 1.5 l of tap water) to measure the CH4 produced by biomass stabilization. The other was the treatment reactor (500 g of biomass and of 1.5 l of liquid fraction). Biomass from a UASB laboratory scale reactor treating leachates from a composting plant operating at
cm3 CH4 NCTP/g VS
J.L. Rico et al. / Bioresource Technology 98 (2007) 971–979
Table 6 Results of anaerobic biodegradability test for the liquid fraction Day CODT (g) CODF (g) CODVFA (g) CH4 (l) CODCH4 (g) CODacid (g) CODCH4 (%) CODVFA (%) CODacid (%) CODcel (%) CODBD (%) 0 37.619 32.454 25.836 0 0 25.836 0 68.7 68.7 0 68.7 4 6.970 5808 0.468 11.157 28.979 29.447 77.0 1.2 78.3 4.5 82.8 7 6.148 4.706 0 11.487 29.836 29.836 79.3 0.0 79.3 4.8 84.1
degradability test, it can be predicted that this liquid fraction could be treated in high load reactors, such as UASB, operating with HRT shorter than 2 days. 4. Conclusions Dairy manure with TS content of 7.72% was separated into two fractions, liquid and solid, by means of screening and coagulation–ﬂocculation. Manure screened through a 1 mm mesh sieve was subjected to a coagulation–ﬂocculation operation followed by gravitational settling. Reactives doses were 10 mg/g TS of CaO as the coagulant and 20 mg/g TS of a medium cationic polyacrylamide as the ﬂocculant. To separate liquid and solid fractions at full scale, a belt ﬁlter press or decanter centrifuge would be necessary after the coagulation–ﬂocculation process. The whole solid fraction, with 30.7% of the ﬁnal mass, contained 77.2% of the initial manure TS, 82.3% of the VS, 59.7% of the TKN and 89.0% of the PT. For the liquid fraction, 86.3% of CODT was due to soluble organic compounds. The liquid fraction CODVFA was 68.7% of CODT and its anaerobically biodegradable COD (CODBD) was 83.7%. The speciﬁc methane production for the liquid fraction was 0.604 l NCTP CH4/ g VS added, while for the dairy manure and the screened manure these yields were 0.307 and 0.371 l NCTP CH4/g VS, respectively. The liquid fraction properties could allow its treatment in high load anaerobic reactors. Consequently, the size of the plant and ancillaries would be reduced and the methane volumetric production rate would be increased. If eﬄuents from anaerobic digesters were going to be discharged, additional treatments would be necessary in order to reach legal discharge parameters. Acknowledgements The authors would like to thank the Ministry of Science and Technology, for ﬁnancially supporting this research under contract AGL2000-1281. References
APHA, 1995. Methods for the Examination of Water and Wastewater, 19th ed. American Public Health Association/American Water Works Association/Water Environment Federation, Washington, DC, USA. Barrow, J., Van Horn, H., Anderson, D., Nordstedt, R., 1997. Eﬀects of Fe and Ca additions to dairy wastewaters on solid and nutrient removal by sedimentation. Appl. Eng. Agric. 13 (2), 259–267. Bolto, B., Abbot-Braun, G., Dixon, D., Eldridge, R., Frimmel, F., Hesse, S., King, S.Y., Toiﬂ, M., 1999. Experimental evaluation of cationic polyelectrolytes for removing natural organic matter from water. Wat. Sci. Technol. 40 (9), 71–79. Burcham, T.N., Gill, S.K., Moore, R.B., 1997. Comparison of dairy manure separation technologies. In: Proceedings of the 1997 ASAE Annual International Meeting. Part 3. Derikx, P.J.L., Willer, H.C., Ten Have, P.J.W., 1994. Eﬀect of pH on the behaviour of volatile compounds in organic manures during drymatter determination. Bioresour. Technol. 49, 41–45.
35 °C was added. Those leachates had high concentrations of VFA and NHþ . The UASB biomass had a concentration 4 of 72.7 g VS/kg and a speciﬁc methanogenic activity at 35 °C of 0:613 g CODCH4 =g VS day. In this assay air was removed from the reactors. To measure the biogas produced, two gasometers of the type described above were used. On days 4 and 7, samples were withdrawn to measure CODT, CODﬁltrate and the CODVFA, calculating, as well, the COD equivalent of the generated CH4 (CODCH4 ). Corrected values were obtained from the data of the treatment reactor minus the data of the blank reactor. On day 8, 1.5 l of supernatant were withdrawn from the treatment reactor and another 1.5 l of liquid fraction were added to the reactor. The biodegradability test was carried out again. To determine %CODBD, the values obtained on day 7 were considered. %CODBD values were 83.3 and 84.1. Table 6 shows corrected data from the reactor corresponding to the assay that resulted in %CODBD of 84.1. These results indicate that the mean percentage of COD that was metabolised by anaerobic microorganisms in their biochemical reactions was 83.7%. The volumes of CH4 produced in both kinds of assays, methane production and anaerobic biodegradability, were consistent. In the biodegradability test for the liquid fraction, a mean value of 7.620 l CH4 NCTP/kg was produced, whereas in the methanogenic production assay the mean value was 7.270 l CH4 NCTP/kg. The CH4 production in the methanogenic production test was smaller than that obtained in the anaerobic biodegradability test, perhaps because the biomass in the two types of assay was diﬀerent. There was a fraction of biodegradable COD used for cellular synthesis that was not measured in the methanogenic production test. The diﬀerence was not very large because the major consumption by cellular synthesis corresponds to growth of acidogenic microorganisms, and as has been shown before, in this case 68.7% of the CODT was in the form of volatile fatty acids (VFA). On day 4, 97.7% of the initial VFA had been removed. On day 7 there were no VFA. This high rate of metabolization of biodegradable COD was due to the high methanogenic activity of the inoculum (22:3 g CODCH4 =day). Taking into account the relationship between the mass of the liquid fraction treated and biomass in the anaerobic bio-
J.L. Rico et al. / Bioresource Technology 98 (2007) 971–979 Field, J., Sierra, R., Lettinga, G., 1988. Ensayos anaerobios. 4° Seminario ´ sobre Depuracion Anaerobia de Aguas Residuales, Valladolid, pp 52– 81. Gilbertson, C.B., Schulte, D.D., 1987. Dewatering screen design for hydraulic settling of solids in swine manure. Trans. ASAE Am. Soc. Agric. Eng. 30 (1), 202–206. Hawkes, F., Rosser, B., Hawkes, D., Statham, M., 1984. Mesophilic anaerobic digestion of cattle slurry after passage through a mechanical separator: factors aﬀecting gas yield. Agric. Wastes 10, 241–256. Henriksen, K., Berthelsen, L., Matzen, R., 1998. Separation of liquid pig manure by ﬂocculation and ion exchange. Part 1: Laboratory experiments. J. Agric. Eng. Res. 69 (2), 115–125. Hill, K.J., 1970. Developmental and comparative aspects of digestion. In: Swenson, M.J. (Ed.), Duke’s Physiology of Domestic Animals, eighth ed. Comstock Publishing Assoc., London, UK, p. 413. Kalyuzhnyi, S., Sklyar, V., Fedorovich, V., Kovalev, A., Nozhevnikova, A., Klapwijk, A., 1999. The development of biological methods for utilisation and treatment of diluted manure streams. Wat. Sci. Tech. 40 (1), 223–229. Liao, P.H., Lo, K.V., 1985. Methane production using a whole and screened dairy manure in conventional and ﬁxed-ﬁlm reactors. Biotechnol. Bioeng. XXVII, 266–272. Liao, P.H., Lo, K.V., 1987. Methane production from dairy manure ﬁltrate and supernatants. Biomass 12, 1–6. Lo, K.V., Bulley, N.R., Liao, P.H., Whitehead, A.J., 1983. The eﬀect of solids-separation pre-treatment on biogas production from dairy manure. Agric. Wastes 8, 155–165. Lo, K.V., Liao, P.H., Chen, W.Y., 1985a. Eﬀect of feed strength on the performance of ﬁxed-ﬁlm reactors receiving screened dairy manure. Biomass 7, 59–72. Lo, K.V., Liao, P.H., March, A.C., 1985b. Thermophilic anaerobic digestion of screened dairy manure. Biomass 6, 301–315. Lo, K.V., Liao, P.H., 1986. Anaerobic-aerobic biological treatment of screened dairy manure. Biomass 10, 187–193. Lo, K.V., Chen, W.Y., Liao, P.H., 1986a. Mesophilic digestion of screened dairy manure using anaerobic rotating biological contact reactor. Biomass 9, 81–92. Lo, K.V., Liao, P.H., Bulley, N.R., 1986b. Two-phase mesophilic anaerobic digestion of screened dairy manure using conventional and ﬁxed-ﬁlm reactors. Agric. Wastes 17, 279–291. Lo, K.V., Lau, A., Liao, P.H., 1993. Composting of separated solid swine wastes. J. Agric. Eng. Res. 54 (4), 307–317. McKenney, L.A., 1998. Dewatering dairy manure using polymer and belt press technology. In: Proc. Manure Management in Harmony with the
Environment Conf., 354–358, Soil and Water Conservation Society, Ames, Iowa, USA, 10–12 Feb. Ames, Iowa: SWCS. Mohri, T., Fukagawa, R., Tateyama, K., Mori, K., Ambassah, N.O., 2000. Slurry dewatering system with planetary rotation chambers. TM Special Technical Publication, 1374, 279–292. Møller, H.B., Sommer, S.G., Ahring, B.K., 2004. Methane productivity of manure, straw and solid fractions of manure. Biomass Bioenergy 26, 485–495. Nozhevnikova, A.N., Kotsyurbenko, O.R., Parshina, S.N., 1999. Anaerobic manure treatment under extreme temperature conditions. Wat. Sci. Tech. 40, 215–221. O’Mellia, C.R., 1972. Coagulation and ﬂocculationPhysicochemical Processes for Water Quality Control. Wiley Interscience, New York, USA. Patni, N.K., Jui, P.Y., 1985. Volatile fatty acids in stored dairy-cattle slurry. Agric. Wastes 13, 159–178. Powers, W.J., Montoya, R.E., Van Horn, H.H., Nordstedt, R.A., Bucklin, R.A., 1995. Separation of manure solids from simulated ﬂushed manures by screening or sedimentation. Appl. Eng. Agric. 11 (3), 431– 436. ´ ´ Rico, J.L., Garcıa, H., Rico, C., Perez, N., 2006. Two factor design for coagulant/ﬂocculant doses eﬀect for solid–liquid separation of dairy manure. Environ. Technol. 27, 143–151. Saﬂey, L.M., Casada, E., Woodbury, J.W., Roos, K.F., 1992. Global methane emissions from livestock and poultry manure. EPA, February 1992. Shang-Kyu, K., Gregory, J., 2001. The interaction of humic substances with cationic polyelectrolytes. Wat. Res. 35 (15), 3557–3566. Sievers, D., Jenner, M., Hanna, M., 1994. Treatment of dilute manure wastewaters by chemical coagulation. Trans. ASAE 37 (2), 597– 601. Van Soest, P.J., Robertson, J.B., Lewis, B.A., 1991. Methods for dietary, neutral detergent ﬁber and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74, 3583–3597. Vanotti, M.B., Hunt, P.G., 1999. Solids and nutrient removal from ﬂushed swine manure using polyacrylamides. Trans. ASAE 42 (6), 1833–1840. Williamson, J.C., Taylor, M.D., Torrens, R.S., Vojvodic-Vukovic, M., 1998. Reducing nitrogen leaching from dairy farm eﬄuent-irrigated pasture using dicyandiamide: a lysimeter study. Agric. Ecosyst. Environ. 69 (1), 81–88. Zeeman, G., 1991. Mesophilic and Psychrophilic Digestion of Liquid Manure. Ph.D. Thesis, Department of Environmental Technology, Agricultural University, Wageningen, The Netherlands. Zhang, R.H., Lei, F., 1998. Chemical treatment of animal manure for solid–liquid separation. Trans. ASAE 41 (4), 1103–1108.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.