Mesophilic and thermophilic anaerobic co-digestion of rendering plant and slaughterhouse wastes

Download Mesophilic and thermophilic anaerobic co-digestion of rendering plant and slaughterhouse wastes

Post on 04-Sep-2016

213 views

Category:

Documents

1 download

TRANSCRIPT

  • o-

    kk5, FI

    Thermophilic

    ndnk rtion57wimee u

    2011 Elsevier Ltd. All rights reserved.

    to sustgradabconveane aroduceigestat

    tion, animal by-products and their process products and wastes are pressed into cake for processing into crude animal feed. Dependingon the raw material characteristics, rendering is carried out as wetor dry processes and operated either as a single batch or multiplecontinuous process.

    Slaughterhouse and rendering wastes are considered as idealsubstrates for biogas production, because they usually contain highconcentration of organic matter and are rich in proteins and lipids.However, anaerobic digestion of these materials is extremelyprone to failure due to production of inhibitory compounds suchas ammonia, VFAs and LCFAs (Cuetos et al., 2008; Hejnfelt andAngelidaki, 2009).

    Abbreviations: CSTR, continuously stirred tank reactor; FM, fresh matter; GC, gaschromatograph; HRT, hydraulic retention time; LCFA, long chain fatty acid; NH3,free ammonia nitrogen; NH4, ammonia nitrogen; OFMSW, organic fraction ofmunicipal solid waste; OLR, organic loading rate; SCOD, soluble chemical oxygendemand; TKN, total Kjehldahl nitrogen; TS, total solids; TVFA, total volatile fattyacids; uVFA, unionised volatile fatty acid; VFA, volatile fatty acid; VS, volatile solids. Corresponding author. Tel.: +358 45 6790087; fax: +358 14 2602321.

    E-mail address: suvi.bayr@jyu. (S. Bayr).1 Present address: Tampere University of Technology, Department of Chemistry

    Bioresource Technology 104 (2012) 2836

    Contents lists available at

    Bioresource T

    journal homepage: www.elsand Bioengineering, P.O. Box 541, FI-33101 Tampere, Finland.ple as a soil conditioner. In slaughterhouse, animals are slaugh-tered and processed into meat products for human consumption.Besides meat products, animal by-products not intended for hu-man consumption are produced. Treatment of these wastes is reg-ulated due to risk of disease spread, mainly Bovine SpongiformEncephalopathy (BSE). In European Union (EU), treatment ofslaughterhouse wastes is regulated by the Animal By-product Reg-ulation (ABPR 1069/2009/EC replacing the ABPR 1774/2002/EC,European Parliament and Council, 2009). According to ABP regula-

    process includes carcases, parts of carcases, heads, feet, offal, ex-cess fat, excess meat, hides, feathers, bones and blood (EuropeanCommission, 2005; Woodgate and van der Veen, 2004). In render-ing process, raw materials are ground to a uniform size (

  • CSTRs at 35 and 55 C. In addition, methane potentials of 10 differ-

    materials froma full-scale biogas plant (Stormossen, Vaasa, Finland)

    TechAmmonia is produced by the biological degradation of thenitrogenous matter, mostly proteins (Kayhanian, 1999). In anaero-bic digestion process, proteins are hydrolysed to amino acids,which are degraded to VFAs (Mata-Alvarez, 2003). In the anaerobicdigestion process, ammonia exists in two forms viz. ammoniumions (NH4 ) and ammonia gas (NH3, free ammonia), the relativeamount of each depends on pH (Hobson and Wheatley, 1993;Kayhanian, 1999). Free ammonia has been suggested to be moreinhibitory than total ammonia nitrogen (Angelidaki and Ahring,1994), because it can readily diffuse across the cell membrane(Kadam and Boone, 1996). Besides pH, temperature also affectsthe dissociation constant of ammonia nitrogen and concentrationof free ammonia in the process; the higher is the temperaturethe higher is the concentration of free ammonia (Kayhanian,1999). In the literature a wide range of inhibitory levels, 0.0451.1 g/l, have been reported for NH3 (e.g. Angelidaki and Ahring,1994; Hansen et al., 1998; Kayhanian, 1999). Similarly, for NH4N, concentrations of 3.86 and 5.60 g/l were found to decreasemethane production by 50% in treating synthetic substrate (petfood) to simulate organic fraction of municipal solid waste(OFMSW) at 35 and 55 C, respectively (Benabdallah El Hadjet al., 2009), whilst concentration of 6 g/l led to inhibited steadystate at 55 C, but was less inhibitory at 35 C in CSTRs treatingswine manure (Hansen et al., 1998).

    Similarly, under anaerobic conditions, lipids are rst hydrolysedto LCFAs, which are oxidised to acetate or propionate and hydrogenthrough the b-oxidation pathway (Cirne et al., 2007; Mata-Alvarez,2003). Inhibitory concentrations of individual LCFAs vary, as e.g.30 mg/l of oleic or linoleic acid inhibited acetoclastic methanogen-esis, whilst 100 mg/l of stearic acid was not inhibitory (Lalman andBagley, 2000, 2001). Furthermore, oleic acid was found less inhib-itory to acidogenesis than for methanogenesis (Beccari et al.,1996). LCFAs were speculated to be reason for process failures inCSTRs treating poultry slaughterhouse wastes at 35 C (Cuetoset al., 2010) and pig slaughterhouse waste with manure at 35and 55 C (Hejnfelt and Angelidaki, 2009). Moreover, temperaturewas reported to effect oleic acid inhibition as acetate-utilisingmethanogenesis was more susceptible to oleic acid toxicity at55 C than at 35 C (Hwu and Lettinga, 1997). Nevertheless LCFAinhibition has been shown to be temporary, but recovery timemight be long (Cirne et al., 2007). Studies on anaerobic digestionof rendering wastes are scarce whereas, several studies on theanaerobic digestion of slaughterhouse waste, especially, poultrywaste have been reported in the literature. For rendering wastes,methane potentials of 497 and 487 dm3 CH4/kg VS were reportedfor bone our and fat, respectively, in batch assays at 55 C(Hejnfelt and Angelidaki, 2009). In a similar study, methane poten-tial of 351381 dm3 CH4/kg VS was reported for meat and bonemeal at 35 C (Wu et al., 2009). For slaughterhouse wastes, meth-ane potentials of 225619 dm3 CH4/kg VS were reported for pigwaste at 55 C (Hejnfelt and Angelidaki, 2009) and 580 dm3 CH4/kg VS at 35 C (Rodrguez-Abalde et al., 2011). Similarly, methanepotentials of 460 dm3 CH4/kg VS were reported for poultry wasteat 35 C (Rodrguez-Abalde et al., 2011) and 210910 dm3 CH4/kg VS both at 35 and 55 C; highest potential was obtained for offal(Salminen et al., 2003). On the other hand, methane yields of 520700 dm3 CH4/kg VS were reported during semi-continuous diges-tion of poultry slaughterhouse wastes in CSTR experiments at35 C (Cuetos et al., 2008; Salminen and Rintala, 2002).

    Co-digestion of slaughterhouse wastes with other materialscontaining low nitrogen and/or lipid content is one option to im-prove the digestion process stability and increase methane yields.Methane yields of 270500 dm3 CH4/kg VS have been reported

    S. Bayr et al. / Bioresourceduring the co-digestion of slaughterhouse wastes with other sub-strates such as manure, sewage sludge and OFMSW (Alvarez andLiden, 2008; Cuetos et al., 2008; Hejnfelt and Angelidaki, 2009;treating putrescible organic fraction ofmunicipalwastewas used asinoculum. Characteristics of the inocula are presented in Table 1.

    2.2. Batch experiments

    Biochemical methane potential experiment was carried out in1 l glass bottles with a working volume of 750 ml. To each bottle,rendering plant wastes (seven materials) or slaughterhouse wastes(three materials) and inoculum (250 ml/bottle) were added at asubstrate to inoculum (Neninniemi) VS ratio of 0.25. Tap waterwas added in order to obtain the working volume of 750 ml. NaH-CO3 (3 g/l) was added as buffer. Thereafter, bottles were ushedwith N2 for 3 min in order to create anaerobic conditions andsealed with silicon stoppers. Assays containing only inoculument rendering plant and slaughterhouse waste fractions were stud-ied in batch assays at 35 C.

    2. Methods

    2.1. Substrates and inocula

    Seven different types of rendering plant wastes and three dif-ferent slaughterhouse by-products were used as substrates in theexperiments. Characteristics of the substrates are presented inTable 1. The rendering wastes viz. melt (sterilized (133 C,20 min, 3 bar) mass), biosludge (sludge from wastewater treat-ment), fat from fat separation well (fat separated with H2O2from wastewater of production equipments and rooms), separa-tor sludge (water, protein and fat extracted in nal puricationby centrifuge from sterilized and solids separated fat), decantersludge (solids, separated by centrifuge from fat separated bypressing from sterilized mass), fat (sterilized and puried fat)and boneour (solids separated by pressing from sterilized mass)were collected from a rendering plant (Honkajoki Ltd., Finland).The above mentioned waste fractions were chosen in the studyas there is need to develop waste treatment method. Slaughter-house wastes viz. the contents of stomach and intestines of bo-vine (without rumen and reticulum) and swine were from ameat producing factory (Saarioinen Ltd., Jyvskyl). Discardedpoultry (turkey) was delivered by Honkajoki Ltd., Finland. Atthe laboratory, the slaughterhouse by-products were macerated(5 mm) by using a meat mincer (Talsa W 22). The well homog-enised materials were stored at 20 C until further use. All thesubstrates, rendering plant as well as slaughterhouse wastes, areheterogeneous, which needs to be taken into account when con-sidering the results.

    Digested sludge from a municipal wastewater treatment plant(Neninniemi, Jyvskyl, Central Finland) was used as inoculum inmesophilic experiments. For thermophilic experiments, digestedLuste and Luostarinen, 2010). In many cases, large quantities ofslaughterhouse and rendering wastes are generated at the sameprocessing plant. Moreover, treatment and use of these organicwastes, rich in protein and lipids, through anaerobic digestion isconsidered as a sustainable solution for simultaneous recovery ofenergy and nutrients. As far as we know, there are no studies onsemi-continuous anaerobic digestion of rendering plant wastesalone or co-digestion of rendering plant wastes with slaughter-house wastes. The objective of the present study was to evaluatethe feasibility of anaerobic semi-continuous co-digestion of ren-dering plant wastes with slaughterhouse wastes in laboratory scale

    nology 104 (2012) 2836 29and water were used as controls. The methane production fromcontrol assays was subtracted from the sample assays. Theprepared assays were incubated statically at 35 1 C. The

  • Table1

    Chem

    ical

    characteristicsan

    dmetha

    nepo

    tentials(STP

    )of

    thestud

    iedslau

    ghterhou

    sean

    drend

    eringplan

    twastesan

    dinocula.Stan

    dard

    deviationin

    parenthe

    siswhe

    nap

    plicable.

    Material

    Batch

    TS (%)

    VS

    (%)

    VS/TS

    (%)

    TKN(g/

    kgFM

    )NH4(g/

    kgFM

    )Protein(g/

    kgFM

    )Protein(g/

    kgVS)

    Lipid(g/

    kgFM

    )Lipid(g/

    kgVS)

    Methan

    eprod

    .(dm

    3CH4/kgVS a

    dded)

    Methan

    eprod

    .(dm

    3CH4/kgFM

    added)

    Theoreticalmethan

    eprod

    .(dm

    3CH4/kgVS)

    **

    Partof

    feed

    (%of

    FM)*

    Ren

    dering

    wastes

    Fatfrom

    fat

    sepa

    ration

    124

    .122

    .292

    4.2

    0.80

    2611

    719

    889

    227

    5(52)

    61(12)

    962

    1.4

    278

    .376

    .097

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    Sepa

    rator

    sludg

    e1

    2.2

    2.0

    910.29

    0.01

    210

    016

    800

    572(187

    )12

    (4)

    861

    4.0

    222

    .421

    .194

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    Melt

    198

    .466

    .668

    690.28

    429

    644

    220

    330

    515(54)

    343(37)

    654

    1.1

    298

    .474

    .576

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    Decan

    ter

    sludg

    e1

    98.2

    62.0

    6361

    0.26

    384

    619

    221

    356

    476(164

    )29

    5(102

    )66

    91.3

    298

    .974

    .675

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    Biosludg

    e1

    1.0

    0.9

    901.2

    0.74

    884

    41

    111

    16(23)

    0.1(0.2)

    532

    2.2

    22.3

    2.0

    87N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    Fat

    199

    .399

    .210

    01.1

    0.02

    77

    935

    943

    406(39)

    403(39)

    959

    Bon

    eou

    r1

    98.7

    56.0

    5772

    0.34

    450

    804

    100

    179

    287(123

    )16

    1(69)

    580

    Slau

    ghterhou

    sewastes

    Poultry

    138

    .233

    .187

    26.3

    0.72

    164

    483

    151

    453

    266(74)

    90(25)

    699

    38.2

    232

    .629

    .590

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.

    262(93)

    76(27)

    Sw

    ine

    131

    .828

    .289

    15.8

    1.1

    9935

    114

    852

    542

    8(25)

    120(7)

    706

    25.2

    Bovine

    153

    .252

    .699

    5.6

    0.24

    3557

    461

    767

    572(89)

    296(46)

    806

    26.7

    Inocula

    Mesop

    hilic

    11.49

    0.77

    521.3

    0.78

    N.d.

    N.d.

    N.d.

    N.d.

    Thermop

    hilic

    12.67

    1.24

    46N.d.

    1.73

    N.d.

    N.d.

    N.d.

    N.d.

    N.d.=

    not

    determ

    ined

    .*Partof

    feed

    ,water

    dilution

    not

    takeninto

    account.

    **Calculatedbasedon

    proteinan

    dlipidcontents.

    30 S. Bayr et al. / Bioresource Technologproduced biogas was collected into aluminium gas bags. Bot-tles were mixed manually before each gas analysis. On day24, as the methane production slowed down, one of the repli-cate bottles digesting rendering plant wastes was moved from35 to 55 C in order to study the effect of change in tempera-ture on the residual methane potential.

    2.3. Reactor experiments

    Anaerobic co-digestion of rendering plant and slaughter-house wastes was studied in two semi-continuously fed stain-less steel CSTRs with a total volume of 13 l and liquid volumeof 10 l. Reactors, R1 and R2, were operated at 35 and 55 C,respectively. Temperature in the reactors was maintained bya heating coil wrapped around the insulated reactor and con-trolled thermostatically. Reactors were mixed mechanicallywith a timer (13 min on/16 min off). Reactors were fed manu-ally on ve days per week on days 1448 and daily on days 49178. Reactors were fed when stirring was off, at a feed rate of200 ml/d and HRT of 50 d (expect for days 112168 in R2). Ateach feeding, an equal amount of efuent was collected intothe digestate storage container (10 l capacity) via the gas trap.The produced biogas was collected in aluminium gas bags.

    Feed was prepared by mixing the ve rendering plant andthree slaughterhouse waste streams according to the actualamount of the by-products produced at the factories as indi-cated in Table 1 (wet weight basis). The prepared feed was fro-zen in plastic containers (500 ml) as weekly feed portions.Frozen feed was thawed and diluted with tap water to obtainthe desired HRT and OLR (Tables 2 and 3).

    During the start-up, 9.7 l of inoculum and 50 g of separatorsludge was added to each reactor. Reactors were ushed withN2 for 5 min to create anaerobic conditions. Separator sludgewas fed twice before actual feeding was initiated on day 14.The initial OLR used in both reactors was 1.5 kg VS/m3 d. Onday 21, OLR of mesophilic reactor (R1) was decreased from1.5 to 0.5 kg VS/m3 d. Thereafter, OLR was increased stepwiselyto 1 kg VS/m3 d on day 69, to 1.5 kg VS/m3 d on day 107 and to2.0 kg VS/m3 d on day 154. On the other hand, OLR of thermo-philic reactor (R2) was increased to 2.25 kg VS/m3 d on day107. Due to process inhibition, feeding in R2 was withheld be-tween days 112 and 168. However, R2 was fed 50200 g onindividual days (day 113,119120,133168) at an averageOLR of 0.4 kg VS/m3 d and HRT 109 d.

    2.4. Analyses

    Total solids (TS) and VSwere analysed according to standardmethods (APHA, 1998). pH was measured with Mettler ToledoSevenEasy pH-metre immediately after each sampling. Solublechemical oxygen demand (SCOD) was analysed according toFinnish StandardMethods (SFS 5504, Finnish Standard Associa-tion, 1988). Ammonianitrogen (NH4) and totalKjeldahlnitrogen(TKN) were determined according to Tecator application note(efuents, Perstorp Analytical/Tecator AB, 1995) or standardmethod (substrates, method 984.13, AOAC, 1990). VFAs weredetermined by using a gas chromatograph (PerkinElmer Auto-system XL GC, PE FFAP column 30 m 0.32 mm 25 lm). He-lium was used as a carrier gas and operation conditions were:oven 100160 C (25 C/min), detector 225 C and injector230 C. The samples for SCODandVFAanalysiswere centrifuged(Sanyo Harrier 18/80 Refrigerated) at 2500 rpm (15 min), forVFA also at 12,000 rpm (10 min) and ltered through glass

    y 104 (2012) 2836microlter (VWR Glass microbers lter 691, particle retention1.6 lm). Lipid content was analysed with ether extract afterhydrolysis with 3 M HCl (Anon, 1971).

  • anaeesis)

    69106 107153 154178

    1.0 1.5 2.057F58321N44

    anaeesis)

    TechTable 2Process parameters and characteristics of feeds and efuents during semi-continuousvalues, except for pH, are mean values during the run (standard deviation in parenth

    Days 1420 2168

    OLR (kg VS/ m3d) 1.5 0.5HRT (d) 50 50CH4 yield (dm3/kg VSfed) 35 () 965 (489)

    Feed Efuent Feed EfuentpH 6.06.1 6.87.4 5.86.3 6.67.4SCOD (g/l) 7.9 (2.4) 2.1 () 3.3 (0.3) 1.8 (1.4)TVFA (mg/l) 586 () 1143 (381) 131 (27) 417 (554)TKN (mg/l) 2682 (8) 1189 () 1546 (354) 1370 (184)NH4 (mg/l) 144 (4) 896 () 53 (4) 916 (70)NH3 (mg/l) N.d. 16 (13) N.d. 15 (8)TS (%) 8.2 () 1.4 (0.5) 2.3 (0.4) 1.3 (0.1)VS (%) 7.9 () 0.9 (0.4) 2.2 (0.4) 0.7 (0.04)VSremoval (%) 89 68

    N.d. = not determined.

    Table 3Process parameters and characteristics of feeds and efuents during semi-continuousvalues, except for pH, are mean values during the run (standard deviation in parenth

    Days 14106

    OLR (kg VS/m3 d) 1.5HRT (d) 50

    S. Bayr et al. / BioresourceGas composition (CH4, CO2 and H2) was measured by using a GC(PerkinElmer Arnel Clarus 500, Alumina column 30 m 0.53 mm)equipped with a thermal conductivity detector (TCD, reactor exper-iments) or ame-ionization detector (FID, batch experiments).Argon was used as a carrier gas and operation conditions were:oven 100 C, detector and injection port 225 C. The biogas volumewas measured by water displacement method. Biogas results wereconverted to standard temperature and pressure conditions (STP,T = 273 K, p = 1 bar).

    2.5. Calculations

    In CSTR-studies methane yields were calculated on weekly val-ues. Protein content was calculated by 6.25 TKN. Theoreticalmethane yields based on protein and lipid contents of the sub-strates (STP, T = 273 K, p = 1 bar) were calculated by X 496 dm3 CH4/kg VS + Y 1014 dm3 CH4/kg VS, were X = % ofproteins and Y = % of lipids (Angelidaki and Sanders, 2004). Theunionised fraction of the ammonia nitrogen was calculated bythe following equation (Watcharasukarn et al., 2009):

    FNH3 1 10pkwpkbpH 1

    1

    The unionised fraction of VFA (uVFA) was calculated by the fol-lowing equation (Watcharasukarn et al., 2009):

    FVFA 10pkapH

    1 10pkapH2

    CH4 yield (dm3/kg VSfed) 766 (173)Feed Efuent

    pH 5.66.1 7.78.1SCOD (g/l) 10.1 (1.7) 3.9 (0.8)TVFA (mg/l) 453 (66) 465 (461)TKN (mg/l) 3177 (507) 2785 (350)NH4 (mg/l) 168 (22) 2049 (255)NH3 (mg/l) N.d. 455 (137)TS (%) 7.4 (0.8) 2.2 (0.5)VS (%) 7.0 (0.9) 1.3 (0.2)VSremoval (%) 81

    N.d. = not determined.0 50 5025 (28) 717 (45) 639 (37)eed Efuent Feed Efuent Feed Efuent.76.1 7.17.5 5.45.9 7.57.9 5.55.7 7.67.8.6 (0.04) 0.7 (0.1) 12.3 (1.2) 1.2 (0.2) 13.4 (4.3) 3.0 (0.8)71 (48) 34 (9) 624 (101) 95 (98) N.d. 1110 (406)815 (216) 1897 (162) 4027 (71) 2720 (320) 5160 (88) 3275 (121)38 (5) 1029 (100) 214 (26) 1617 (283) 293 () 2230 (159).d. 21 (10) N.d. 68 (22) N.d. 94 (10).5 (0.4) 1.1 (0.04) 7.5 (0.1) 1.3 (0.1) 9.9 (0.3) 1.4 (0.04).2 (0.4) 0.7 (0.04) 7.1 (0.1) 0.9 (0.1) 9.4 (0.2) 1.0 (0.02)

    83 87 89

    robic co-digestion of rendering plant and slaughterhouse wastes in CSTR at 55 C. All.

    107111 112168

    2.25 0.450 109robic co-digestion of rendering plant and slaughterhouse wastes in CSTR at 35 C. All.

    nology 104 (2012) 2836 313. Results and discussion

    3.1. Material characterisation

    The characteristics of the 10 studied materials varied greatly(Table 1). Rendering materials, except for biosludge, separatorsludge and fat from fat separation, had TS more than 98%, withVS accounting for >85% of TS in all materials, except for melt, de-canter sludge and boneour. Fat from fat separation unit, separatorsludge and biosludge varied greatly between two batches appar-ently due to differences in rendering and wastewater treatmentprocesses. Based on chemical composition, melt, decanter sludge,fat and boneour of the rendering wastes and bovine from slaugh-terhouse wastes, were expected to produce high methane yieldsdue to high protein and lipid contents taken together (P495 g/kgfresh matter (FM)). All materials, except for biosludge, could beconsidered as promising substrates for methane production. Inthe reactor feed (without dilution) calculated proportion of lipidswas 22.7% of FM and calculated proportion of proteins was 10.5%of FM.

    3.2. Batch experiments

    Methane yields for the studied materials are presented in Ta-ble 1. Methane yields for rendering materials varied between 275and 572 dm3 CH4/kg VSadded (except for biosludge 16 dm3 CH4/kg VSadded). Similar methane yields of 262572 dm3 CH4/kg VSadded

    617 () 401 (128)Feed Efuent Feed Efuent5.9 7.9 5.45.9 7.57.917.8 () 6.9 (0.6) 5.0 (0.6) 12.9 (1.8)983 () 2181 (196) 258 (13) 4135 (438)5118 () 3392 () 1548 (43) 3368 (298)341 () 2647 (89) 86 (8) 2756 (221)N.d. 598 () N.d. 488 (123)11.6 () 1.7 () 2.9 (0.03) 2.4 (0.2)10.8 () 1.1 () 2.8 (0.04) 1.7 (0.2)

    90 39

  • obic digestion of poultry slaughterhouse waste was possible at HRTof 25 d, if the HRT was progressively decreased from 50 to 25 d.

    Techwere also obtained for slaughterhouse wastes. On VS basis, highestmethane yield of 572 dm3 CH4/kg VSadded, was obtained for separa-tor sludge and bovine. Both these materials had high VS/TS ratioand high concentration of proteins and lipids taken together. Onthe other hand, fat, with high lipid content (935 g/kg FMadded)and VS/TS ratio 100%, produced the highest methane yield perFM, (403 dm3 CH4/kg FMadded). No difference was noticed when as-says incubated at 35 C were moved to thermophilic conditions (onday 24), compared to bottles left to mesophilic conditions (data notshown).

    As far as we know this is the rst study to report methane yieldsfrom rendering plant wastes widely. The methane yields obtainedfor bone our (287 dm3 CH4/kg VSadded) and fat (406 dm3 CH4/kg VSadded) in the present study were slightly lower than the yieldsof 497 and 487 dm3 CH4/kg VS, reported for similar materials byHejnfelt and Angelidaki (2009). For meat and bone meal, methaneyields of 351381 dm3 CH4/kg VS were reported (Wu et al., 2009).The methane yields obtained for pig waste (428 dm3 CH4/kg VSadded) in the present study, were within the range of 225619 dm3 CH4/kg VS reported for pig wastes at 55 C (Hejnfelt andAngelidaki, 2009). On the other hand, the methane yield obtainedfor poultry slaughterhouse waste in the present study(266 dm3 CH4/kg VSadded) was similar or lower than the yieldsreported at 35 and 55 C. For instance, Salminen et al. (2003)reported methane yields of 210910 dm3 CH4/kg VS, with highestyield reported for poultry offal. For bovine slaughterhouse wastethere are no comparable results reported in the literature.

    3.3. Reactor experiments

    3.3.1. OLRs and methane productionCo-digestion of rendering plant and slaughterhouse wastes was

    studied in semi-continuously fed CSTR-reactors for 178 days at35 C and 168 days at 55 C. The reactor performances are shownin Figs. 13 and summarised in Tables 2 and 3.

    In both reactors, daily feeding was started on day 14 with anOLR of 1.5 kg VS/m3 d. Methane production started immediatelyin thermophilic reactor, even though some low yields were de-tected around day 35. On the other hand, no methane productionwas noticed in mesophilic reactor and therefore, OLR of the meso-philic reactor was decreased from 1.5 to 0.5 kg VS/m3 d on day 21after which methane production rapidly peaked indicating metha-nation of accumulated VFAs. Subsequently, the OLR in mesophilicreactor was then increased stepwisely according to increasedmethane production up to OLR 2.0 kg VS/m3 d on day 154. Simi-larly, OLR in thermophilic reactor was increased as well, to2.25 kg VS/m3 d on day 107. However, methane production of ther-mophilic reactor started to decrease and from day 112 onwards, forrest of the run, reactor was fed with 50200 g/d on individual days(d 113,119120,133168), average OLR being 0.4 kg VS/m3 d andaverage HRT 109 d.

    Overall, better process performance and higher methane yieldswere noticed in the mesophilic reactor than in the thermophilicreactor (Figs. 13, Tables 2 and 3). During the stable period, meth-ane content of the biogas was 6575% in the mesophilic reactor(days 25178) and 5572% in the thermophilic reactor (days 7168). For mesophilic reactor the highest average methane yieldof 965 dm3 CH4/kg VSfed, was obtained when the reactor was oper-ated with an OLR of 0.5 kg VS/m3 d (Table 2, Fig. 1). However, someof the methane was from accumulated substrate from the initialOLR of 1.5 kg VS/m3 d period. Average methane yields 725 and717 dm3 CH4/kg VSfed were obtained with OLRs of 1 and1.5 kg VS/m3 d, respectively. Thus, it seems that mesophilic reactor

    3

    32 S. Bayr et al. / Bioresourcecan be operated with OLR 1 kg VS/m d and apparently also withOLR 1.5 kg VS/m3 d as the experimental periods with these OLRslasted together for 2 HRTs. Good performance of the mesophilicThis improvement in process stability in the above study wasmainly attributed due to the adaptation of micro-organisms. Thusthe need for low OLRs and long HRTs at the beginning of the exper-iment are crucial for process success. Before OLRs can be increased,the microbes need to adapt well for feedstocks with high lipid andprotein content such as slaughterhouse wastes (Cuetos et al., 2008;Edstrm et al., 2003).

    As far as we know, there are no comparable studies on anaero-bic digestion of rendering plant wastes alone or co-digestion ofrendering with slaughterhouse wastes, especially semi-continuousreactor studies. Therefore, the methane yields obtained in the pres-ent study are compared with the yields obtained during digestionof slaughterhouse wastes alone or during co-digestion of slaugh-terhouse wastes and other wastes such as sewage sludge orOFMSW. The methane yields obtained in the present study weresimilar to the yields of 520700 dm3 CH4/kg VS reported duringthe anaerobic digestion of slaughterhouse wastes alone at 35 C(Cuetos et al., 2008; Salminen and Rintala, 2002) and were higherthan the yields of 380430 dm3 CH4/kg VS, reported during the co-digestion of animal by-products and sewage sludge (VS ratios of1:3 and 1:7) at 35 C (Luste and Luostarinen, 2010).

    3.3.2. VFAs and SCODVFA production proles are presented in Fig. 3. Total volatile

    fatty acids (TVFA) concentration in the thermophilic reactor in-creased gradually from 25 mg/l (day 17) to reach the highest con-centration of 4750 mg/l on day 143 and thereafter dropped to3300 mg/l at the end of the run (day 168). On the other hand, TVFAbuild up in the mesophilic reactor was noticed during the initialstart-up (days 139), slight increase after day 140, and faster in-crease after day 156 reaching 1655 mg/l in the end, suggesting thatOLR of 2 kg VS/m3 d was too high. However, the VFA levels in themesophilic reactor were lower than the concentrations of900 mg/l of propionic acid considered as inhibitory for methano-gens and 2400 mg/l of acetic or butyric acid considered non-inhib-itory for methanogens (Wang et al., 2009). Acetic acid was themain VFA, accounting for >80% of the TVFA, presented in the mes-ophilic reactor indicating that aceticlastic methanogenesis was therate-limiting step in this reactor. A similar observation was also re-ported during the mesophilic digestion of poultry slaughterhousewaste (Salminen and Rintala, 2002). Lowest concentration of SCODprocess was also supported with high VS reductions, between83% and 89% during the run. The lower VS reduction noticed duringdays 2168 (68%) was due to the conversion of VFA that had accu-mulated during the rst phase (day 1420). On the other hand, inthe thermophilic process the highest average methane yield of766 dm3 CH4/kg VSfed, was obtained when the reactor was oper-ated with an OLR of 1.5 kg VS/m3 d with VS reduction of 81%. How-ever, after 1.5 HRT, thermophilic process became unstable and VFAaccumulation was noticed indicating that OLR of 1.5 kg VS/m3 dwas too high. These results are in agreement with previous studies,where OLRs of 0.81.7 kg VS/m3 d were successfully applied fortreating slaughterhouse wastes alone (Cuetos et al., 2008;Salminen and Rintala, 2002) and 1.853.7 kg VS/m3 d for co-diges-tion of slaughterhouse waste with OFMSW (Cuetos et al., 2008). Inthe present study, both reactors were operated with 50 d HRT (ex-cept for reactor R2 during days 112168), which was the same orlower than HRTs (50100 d) reported during the digestion of poul-try slaughterhouse waste in CSTRs at 35 C (Salminen and Rintala,2002). Nevertheless, Cuetos et al. (2008) demonstrated that anaer-

    nology 104 (2012) 2836(ca 0.6 g/l) was measured in mesophilic process when TVFA wasthe lowest (

  • Tech2000

    3000

    yie

    ld (m

    l gV

    S)

    S. Bayr et al. / BioresourceConcentrations of calculated uVFA values uctuated more in themesophilic than in the thermophilic reactor, but remained mainlyat the level of 0.012 mg/l in the mesophilic reactor and 0.014 mg/l in the thermophilic reactor. Highest calculated uVFA con-centrations of 23.1 and 5.6 mg/l were noticed on day 22 and day168 in the mesophilic and the thermophilic reactor, respectively.The increase in calculated uVFA concentrations in both reactors

    0

    1000

    Met

    han

    e

    CH

    4/

    6

    7

    8

    9pH

    0

    2

    4

    6

    SCO

    D (g

    /l)

    0

    1

    2

    3

    NH

    3o

    r N

    H4-

    N (g

    /l)

    0

    1000

    2000

    3000

    0 50 1

    TVFA

    (m

    g/l)

    0

    2

    4

    6

    TKN

    (g/

    l)

    Fig. 1. Characteristics of the efuent during semi-continuous anaerobic co-digestion o() OLR, (j) pH, (N) SCOD, (e) NH4N, () NH3, () TKN and (s) TVFA.2

    3

    VS/

    m3 d

    )

    nology 104 (2012) 2836 33were apparently due to an increase in TVFA concentrations asthe concentration of uVFA depends on concentration of TVFA, pHand temperature (Kayhanian, 1999).

    3.3.3. Ammonia and pHDepending upon the applied OLR, feed TKN concentration ran-

    ged from 1.5 to 5.1 g/l with NH4N accounting for 37% of TKN.

    0

    1

    OLR

    (kg

    00 150 200

    f rendering plant and slaughterhouse wastes in CSTR at 35 C; (h) methane yield,

  • Tech800

    1200

    ld (m

    l S)

    34 S. Bayr et al. / BioresourceNH4N concentration in the efuent of mesophilic and thermo-philic reactor was 5475% and 7482% of TKN, respectively, indi-cating high ammonication of organic nitrogen in the feed.Salminen and Rintala (2002) also reported a similar ammonica-tion of 5267% ammonia of TKN in the efuent, compared to littleammonia present in the feed, during the anaerobic digestion of

    0

    400

    Met

    han

    e yi

    e

    CH

    4/gV

    6

    7

    8

    9pH

    0

    6

    12

    18

    SCO

    D (g/

    l)

    0

    1

    2

    3

    NH

    3o

    r N

    H4-

    N (g/

    l)

    0

    2000

    4000

    6000

    0 50 10

    TVFA

    (m

    g/l)

    0

    2

    4

    6

    TKN

    (g/

    l)

    Fig. 2. Characteristics of the efuent during semi-continuous anaerobic co-digestion o() OLR, (j) pH, (N) SCOD, (e) NH4N, () NH3, () TKN and (s) TVFA.2

    3

    m3 d

    )

    nology 104 (2012) 2836poultry slaughterhouse waste at mesophilic conditions. These re-sults suggest that high amounts of ammonia would be formed dur-ing anaerobic digestion of the studied substrates that might inhibitmethanogenesis.

    As expected, concentration of NH4N was higher in the thermo-philic than in the mesophilic reactor. In thermophilic reactor,

    0

    1

    OLR

    (kg

    VS/

    0 150 200

    f rendering plant and slaughterhouse wastes in CSTR at 55 C; (h) methane yield,

  • Tech0

    1000

    2000

    3000

    4000

    5000

    0 50 100 150 200

    VFA

    (m

    g/l)

    Time (d)

    0

    500

    1000

    1500

    2000

    2500

    3000

    0 50 100 150 200

    VFA

    (m

    g/l)

    Fig. 3. TVFA, acetic acid, propionic acid and iso-valeric acid concentrations in semi-continuous CSTR experiments of rendering plant and slaughterhouse wastes at

    S. Bayr et al. / BioresourceNH4N concentration increased throughout the experiment from1.6 to 3.0 g/l. On the other hand, NH4N concentration in the mes-ophilic reactor was 0.81.1 g/l during days 1101, but increased to2.4 g/l (on day 178), when OLR was increased to 1.5 kg VS/m3 d.However, the ammonia concentrations in the present study werebelow 3.86 and 5.60 g/l found to decrease methane productionby 50% in treating simulated OFMSW at 35 and 55 C, respectively(Benabdallah El Hadj et al., 2009).

    The calculated NH3 concentration, was 36 times higher inthermophilic than in mesophilic reactor and followed the sametrend as that of NH4N. The calculated NH3 concentration in thethermophilic reactor was between 290 and 635 mg/l during thewhole experiment (except for days 114). The corresponding valuefor the mesophilic reactor was

  • Anon, 1971. Determination of crude oils and fats. Off. J. Eur. Commun. Legis. 297,995997.

    AOAC, 1990. Ofcial Methods of Analysis. Association of Ofcial Analytical ChemistsInc., Arlington, VA, pp. 1298.

    APHA, 1998. Standard Methods for the Examination of Water and Wastewater,twentyth ed. American Public Health Association, Washington DC.

    Beccari, M., Bonemazzi, F., Majone, M., Riccardi, C., 1996. Interaction betweenacidogenesis and methanogenesis in the anaerobic treatment of olive oil millefuents. Water Res. 30, 183189.

    Benabdallah El Hajd, T., Astals, S., Gal, A., Mace, S., Mata-lvarez, J., 2009. Ammoniainuence in anaerobic digestion of OFMSW. Water Sci. Technol. 59,11531158.

    Bjrnsson, L., Murto, M., Mattiasson, B., 2000. Evaluation of parameters formonitoring an anaerobic co-digestion process. Appl. Microbiol. Biotechnol. 54,844849.

    Cirne, D.G., Paloumet, X., Bjrnsson, L., Alves, M.M., Mattiasson, B., 2007. Anaerobicdigestion of lipid-rich wasteeffects of lipid concentration. Renew. Eng. 32,965975.

    Cuetos, M.J., Gomez, X., Otero, M., Moran, A., 2008. Anaerobic digestion of solidslaughterhouse waste (SHW) at laboratory scale: inuence of co-digestion withthe organic graction of municipal solid waste (OFMSW). Biochem. Eng. J. 40,99106.

    Cuetos, M.J., Gomez, X., Otero, M., Moran, A., 2010. Anaerobic digestion and co-digestion of slaughterhouse waste (SHW): inuence of heat and pressure pre-treatment in biogas yield. Waste Manage. 30, 17801789.

    Edstrm, M., Nordberg, ., Thyselius, L.T., 2003. Anaerobic treatment of animalbyproducts from slaughterhouses at laboratory and pilot scale. Appl. Biochem.Biotechnol. 109, 127138.

    European Commission, 2005. Integrated pollution prevention and control.Reference document on best available techniques in the slaughterhouses andanimal by-products industries.

    European Parliament and Council, 2009. Regulation (EC) No 1069/2009 of theEuropean Parliament and of the Council of 21 October 2009 laying downhealth rules as regards animal by-products and derived products notintended for human consumption. Ofcial Journal of European Union L27314/11/2009.

    Finnish Standard Association, 1988. SFS 5504 Determination of chemical oxygendemand (CODCr) in water with closed tube method, oxidation with dichromate.Finnish Standard Association, Helsinki, Finland.

    Hobson, P.N., Wheatley, A.D., 1993. Anaerobic digestion, Modern theory andpractice. Elsevier Science Publishers Ltd., England.

    Hwu, C.-S., Lettinga, G., 1997. Acute toxicity of oleate to acetate-utilizingmethanogens in mesophilic and thermophilic anaerobic sludges. EnzymeMicrob. Technol. 21, 297301.

    Kadam, P.C., Boone, D.R., 1996. Inuence of pH on ammonia accumulation andtoxicity in halophilic, methyltrophic methanogens. Appl. Environ. Microbiol. 62,44864492.

    Kayhanian, 1999. Ammonia inhibition in high-solids biogasication: an overviewand practical solutions. Environ. Technol. 20, 355365.

    Lalman, J.A., Bagley, D.M., 2000. Anaerobic degradation and inhibitory effects oflinoleic acid. Water Res. 34, 42204228.

    Lalman, J.A., Bagley, D.M., 2001. Anaerobic degradation and methanogenicinhibitory effects of oleic and stearic acids. Water Res. 35, 29752983.

    Luste, S., Luostarinen, S., 2010. Anaerobic co-digestion of meat-processing by-products and sewage sludgeeffect of hygienization and organic loading rate.Bioresour. Technol. 101, 26572664.

    Mata-Alvarez, J., 2003. Fundamentals of the anaerobic digestion process. In: Mata-Alvarez, J. (Ed.), Biomethanation of the Organic Fraction of Municipal SolidWastes. IWA Publishing, UK, pp. 122.

    Perstorp Analytical/Tecator AB, 1995. Determination of nitrogen according tokjeldahl using block digestion and steam distillation. Tecator application note.

    Rodrguez-Abalde, A., Fernndez, B., Silvestre, G., Flotats, X., 2011. Effects of thermalpre-treatments on solid slaughterhouse waste methane potential. WasteManage. 31, 14881493.

    Salminen, E.A., Rintala, J.A., 2002. Semi-continuous anaerobic digestion of solidpoultry slaughterhouse waste: effect of hydraulic retention time and loading.Water Res. 36, 31753182.

    Salminen, E., Einola, J., Rintala, J., 2003. The methane production of poultryslaughtering residues and effects of pretreatments on the methane productionof poultry feather. Environ. Technol. 24, 10791086.

    Wang, Y., Zhang, Y., Wang, J., Meng, L., 2009. Effects of volatile fatty acidconcentrations on methane yield and methanogenic bacteria. Biomass andBioenergy 33, 848853.

    Watcharasukarn, M., Kaparaju, P., Steyer, J.-P., Krogfelt, K.A., Angelidaki, I., 2009.Screening Escherichia coli, Enterococcus faecalis, and Clostridium perfringens asindicator organisms in evaluating pathogen-reducing capacity in biogas plants.Microb. Ecol. 58, 221230.

    36 S. Bayr et al. / Bioresource Technology 104 (2012) 2836Hansen, K.H., Angelidaki, I., Ahring, B.K., 1998. Anaerobic digestion of swinemanure: inhibition by ammonia. Water Res. 32, 512.

    Hejnfelt, A., Angelidaki, I., 2009. Anaerobic digestion of slaughterhouse by-products.Biomass Bioenergy 33, 10461054.Woodgate, S., van der Veen, J., 2004. The role of fat processing and rendering in theEuropean Union animal production industry. Biotechnol. Agron. Soc. Environ. 8,283294.

    Wu, G., Healy, M.G., Zhan, X., 2009. Effect of the solid content on anaerobic digestionof meat and bone meal. Bioresour. Technol. 100, 43264331.

    Mesophilic and thermophilic anaerobic co-digestion of rendering plant and slaughterhouse wastes1 Introduction2 Methods2.1 Substrates and inocula2.2 Batch experiments2.3 Reactor experiments2.4 Analyses2.5 Calculations

    3 Results and discussion3.1 Material characterisation3.2 Batch experiments3.3 Reactor experiments3.3.1 OLRs and methane production3.3.2 VFAs and SCOD3.3.3 Ammonia and pH3.3.4 LCFAs and inhibition

    4 ConclusionsAcknowledgementsReferences

Recommended

View more >