Bioresource Technology 100 (2009) 4117–4120

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Optimising the biogas production from leather fleshing wasteby co-digestion with MSW

P. Shanmugam *, N.J. HoranEnvironmental Engineering, School of Civil Engineering, University of Leeds, Leeds LS2 9JT, UK

a r t i c l e i n f o a b s t r a c t

Article history:Received 7 January 2009Received in revised form 16 March 2009Accepted 17 March 2009Available online 22 April 2009

Keywords:C:N ratiopHAmmoniaBiomass activityBiogas yield

0960-8524/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.biortech.2009.03.052

* Corresponding author. Address: Scientist – EI,Department, Central Leather Research Institute (CSI(off). Tel.: +91 44 24911 386.

E-mail address: pashanmugam@yahoo.com (P. Sha

Waste from the leather industry, known as limed leather fleshing (LF), has a low C:N (3.2) and an alkalinepH of 11.4. This is a major disadvantage for anaerobic digestion due to ammonia toxicity for methanogen-esis. This study describes co-digestion of LF with biodegradable fraction of municipal solids waste opti-mised over a range of C:N and pH to minimise ammonia and to maximise biogas yield. The optimumconditions were found with a blend that provided C:N of 15 and pH of 6.5 and the cumulative biogas yieldincreased from 560 mL using LF fraction alone, to 6518 mL with optimum blend. At higher pH of 8.5,unionised ammonia was high (2473 mg L�1) coincided with poor biogas yield (47 mL d�1) that confirmsammonia toxicity. By contrast at a pH of 4.5 the ammonia was minimum (510 mg L�1), but high VFA(26,803 mg L�1) inhibited the methanogens. Biomass activity measured using ATP correlated well withbiogas yield as reported previously.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction liquid (Shanmugam and Horan, 2008). This is conducive to the

Anaerobic digestion (AD) technology is increasingly importantfor waste management as it generates renewable energy fromindustrial and municipal solid wastes in an environmentally benignway. A number of regulatory drivers such as landfill allowances andtax escalation scheme (LATS), renewable obligation certificate(ROCs) in UK (Defra, 2007) and certified emission ratings (CERs),carbon off-settings and clean developing mechanisms (CDM) indeveloping countries, have added an economic incentive to thistechnology. These incentives together with the revenue from bothbioenergy and digestate sold as biofertilizer have contributed to areduction in the capital payback period, (Tomor, 1994; Murphyet al., 2004; Taleghani and Kia, 2005). Full-scale anaerobic digestionplants have been particularly successful for treating carbohydratewaste, but for the proteinaceous solid wastes such as leather,slaughterhouse, dairy, cow and chicken manure were reported toyield lower biogas than the carbohydrate rich solid wastes. This islargely a result of ammonia toxicity from the large nitrogen fractionin these wastes. Digestion of proteinaceous solid wastes results inthe hydrolysis and solubilisation of protein to amino acids throughproteolytic bacteria. Amino acids are further hydrolysed to releaseammonia, H2, CO2 and volatile fatty acids (VFA) by hydrolytic/hydrogenic bacteria. Release of ammonia from the hydrolysis ofamino acids increases both the alkalinity and pH of the digester

ll rights reserved.

Environmental EngineeringR), Adyar, Chennai 20, India

nmugam).

conversion of acetate to ammonium acetate or ammoniumbicarbonate, depletes the available acetate which is a food sourcefor the methanogens (Hoshimoto, 1986; Angelidaki et al., 1998;Sterling et al., 2001; Sung and Liu, 2003). Thus, inhibits the metha-nogenic activity and lowers the headspace CH4 content. The forma-tion of unionised free ammonia (NH3) was confirmed to be moretoxic than the ammonium ion (NH4

+) and such NH4+ or NH3 forma-

tion in an anaerobic digester is pH dependent (Sterling et al., 2001;Nielsen and Anglidaki, 2008; Chen et al., 2008).

Sterling et al. (2001) have reported that when ammonia isadded to a digester, the pH increases until a chemical equilibriumis reached: ammonia inhibits methanogen metabolism and as a re-sult VFA’s accumulate and this results in a reduction in pH, so thatammonia is converted into ionised ammonium. Although methaneproducing bacteria were able to acclimatise to ammonia concen-trations as high as 5000 mg L�1 (Strik et al., 2006), Koster (2007)reported that methane production ceased at an NH3 concentrationof 1900–2000 mg L�1 at pH values above 7.6. Hence a dynamicequilibrium between the hydrolytic and the methanogenicbacterial population is vital to achieve optimum biogas yield.Leather manufacturing is one of the world’s oldest industries andis important both for developing as well as for developed countries.Global production of leather from animal hides is around 1.8 mil-lion square feet per annum (Ramasami et al., 1999). The hide orskin consists of three layers, a top epidermal layer, middle coriumlayer (used to make actual leather) and bottom adipose tissue. Thetop epidermal layer and the bottom adipose tissue are the principalsolid wastes generated from leather making and are termed flesh-ing (LF) and this waste is around 15% by weight of raw hide/skin

4118 P. Shanmugam, N.J. Horan / Bioresource Technology 100 (2009) 4117–4120

processed to leather. Rosenwinkel and Meyer (1999) reported thatsuccessful treatment of a slaughterhouse waste, hog and cowstomach content with sewage sludge yielded maximum methaneof 0.23 m3 kg�1 of TS added. This low yield for such proteinaceouswaste is due to the low C:N and high alkaline pH of the wasteattributed to ammonia toxicity during anaerobic digestion(Hoshimoto, 1986; Hansen et al., 1998; Salerno et al., 2006).

There are many techniques for the control of ammonia in anaer-obic digesters such as struvite precipitation, Annamox, magnesiumammonium phosphate (Maqueda et al., 2003; Demirer et al., 2005),but which are expensive to implement in a large scale AD. By con-trast optimising the feedstock C:N ratio and implementing digesterpH control is a more feasible option to reduce the ammonia toxic-ity (Sievers and Brune, 1978; Hills, 1979; Ogunwande et al., 2008).Many researchers have reported varying optimum C:N ratios formaximum biogas yield, for example Sievers and Brune (1978)found that a C:N ratio of 15.5–19 was optimum for maximummethane production for swine manure. But it appears that no workhas been reported on the combined effect of C:N ratio and pH con-trol on ammonia reduction, in particular for leather fleshing waste.Therefore this present study was undertaken to evaluate the effectof admixture with MSW to vary the C:N ratio, together with pHcontrol, in order to minimise ammonia toxicity and thus maximisebiogas yield during anaerobic digestion.

2. Methods

Leather fleshing (LF) waste (before chrome tanning leathermanufacturing process) was collected following the liming process,from the Holmes Hall Tannery in Kingston upon Hull (UK). Thelimed LF was minced and homogenized with a commercial blenderto 6 mm diameter before feeding to the digester. The source sortedpre-autoclaved (130 �C) organic Municipal solid waste (MSW) wascollected in bulk and used in this experiment. The characterizationof total solids (TS) and volatile solids (VS) was carried out usingStandard Methods (APHA, 1998). The procedure for elementalcomposition of carbon, hydrogen, nitrogen and sulphur, ATP anal-ysis and the biochemical methane potential (BMP) batch experi-ment, and the total biogas measurement has been reportedpreviously (Shanmugam and Horan, 2009). The C:N ratio wascalculated based on the chemical composition (Shanmugam andHoran, 2009) and the TS content of LF and MSW and blends inthe ratio of 5, 10, 15, 20 and 30 were prepared. The supernatantof C:N adjusted reactors was collected periodically for the analysisof TS, VS, VFA, alkalinity, NH3–N. The pH value was measuredimmediately upon collection of samples.

After determination of the optimum C:N ratio, the experimentswere repeated at this ratio but with controlled pH values of 4.5, 5.5,6.5 7.5 and 8.5. The pH was checked daily and adjusted by additionof either 6 N NaOH or 1 N HCl. The head space biogas was sampledfrom each batch digesters using gas tight syringe for biogas compo-sition analysis. Biogas composition was analyzed with gas liquidchromatography (HP5890) using a molecular sieve column lengthby diameter of 30 � 0.320 (Agilent catalogue number 19091P).Carrier gas flow rate was 6 mL min�1 and the column, detectorand injector temperatures were 60, 200 and 200 �C, respectively.The components of the biogas were detected with a TCD detector.

3. Results and discussion

3.1. Methane yields from unblended wastes

The chemical characteristics of the raw LF and MSW used in thisstudy have been described previously (Shanmugam and Horan,2009). The LF waste is characterized by its very alkaline pH value

of 11.4 and also a very low C:N ratio of 3.2. As a result, duringthe digestion process it releases only small quantities of methanewith a methane yield of 0.08 Nm L CH4/gm VS removed. It wasthought that this low biogas yield is a result of ammonia releaseduring digestion which is toxic to methanogenesis. By contrastthe MSW has a neutral pH and a C:N of 21.6 and during digestionit has a methane yield three times higher at 0.24 Nm L CH4/gm VSremoved (Table 1). The average head space CH4 content in LF onlyreactor was 33%, whereas the same was noticed as 56% in the MSWonly anaerobic reactor.

3.2. NH3–N reduction through optimisation of C: N ratio

Among many NH3–N removal strategies, the optimisation ofC:N ratio of the feedstock by blending waste streams for co-diges-tion is potentially the least expensive and easiest to implement.The wastes chosen permitted blends with C:N ratio’s between 3.2and 30, and a total of seven blends were prepared. Wastes withC:N values of 15 and 20 gave the highest cumulative biogas pro-duction and there was little difference between these two blendsbut they both produced considerably more gas than the otherblends evaluated (Fig. 1). The corresponding maximum specificbiogas yield was 0.145 and 0.15 Nm L CH4/g VSr, respectively.Sievers and Brune (1978) have also reported an optimum C:N ratioof 19.9. Wastes with a low C:N (LF alone, 5 and 10) were associatedwith the release of large quantities of ammonia up to 4289 mg L�1

which resulted in the pH rising to as high as 11.4 and the alkalinityto 34,020 mg L�1. By contrast the wastes with the highest biogasyield had low concentrations of both ammonia and alkalinity at1736 and 8970 mg L�1, respectively. Similar observations weremade by Koster and Lettinga (1988) who investigated anaerobicdigestion at high ammonia concentrations. In view of the impor-tance of pH in enhancing methane production, the effects of arange of operating digester pH values were evaluated for the opti-mum C;N ratio of 15. The headspace CH4 content has increasedwith an increased C:N ratio of LF and MSW blend (Table 1).

3.3. NH3–N reduction through optimisation of pH with C:N ratio of 15

Researchers have reported different concentrations for thethreshold level of unionised NH3 that is toxic to methanogenesis.These ranges from 250 mg L�1 for chicken manure (Bujoczeket al., 2000), 95–297 mg L�1 for biowaste (Gallert et al., 2003),700–1000 mg L�1 for livestock waste (Angelidaki and Ahring,1994; Hansen et al., 1998) and as high as 7000 mg L�1 for poultrywastes (Pechan et al., 1987) and 8000–13,000 mg L�1 for syntheticwaste water (Sung and Liu, 2003). As the C:N ratio is important indetermining the likely free ammonia concentration this study wasundertaken to evaluate the combined effect of C:N with pH control.The BMP evaluation was carried out at C:N of 15 with controlledpH values of 4.5, 5.5, 6.5, 7.5 and 8.5 to study its impact on NH3–N toxicity on biogas yield. Over this pH range the cumulative bio-gas production ranged from 1258 mL at pH 8.5 to 6518 at pH 6.5.The observed minima at pH 8.5 equated to a specific biogas yieldof 0.17 Nm L CH4 g�1 VS removed and coincided with the highest(3473 mg L�1) ammonia production. At pH 6.5 the cumulative bio-gas yield volume was 6515 Nm mL and this coincided with a lowammonia production at 817 mg L�1. This was in line with workof Koster (2007) and Kayhanian (2004) who observed that theammonia toxicity only occurred at pH values greater than 7.6.However, the methane yield is not solely a function of the ammo-nia concentration as the lowest concentration of 515 mg L�1 wasobserved at a pH of 4.5 (Fig. 2). At this pH the VFA toxicity is thelimiting factor and Sterling et al. (2001), concluded that whenammonia is added to the digester, the pH was increased untila chemical equilibrium was established. As ammonia inhibits

Table 1Anaerobic digester performance indicators at different C:N ratio.

Parameters C:N 5 C:N 10 C:N 15 C:N 20 C:N 30 Only LF Only MSW

Daily average biogas yield (Nm mL d�1) 7.9 47.7 87.87 79.97 84.17 2.83 50.03VSr (%) 24.36 26.33 44.31 39.10 40.31 35 50Specific Biogas Yield (Nm L CH4/gm VS removed) 0.02 0.10 0.15 0.11 0.13 0.08 0.24Avg NH3–N (mg L) 3600.8 3215.3 2828 1611.8 1736.5 4289.5 614Avg. Alk (mg L) 23770.0 22731.0 15742.0 11957.0 2954.00 31,312 22,000Avg. VFA(mg L) 12450.0 25943.0 9879.0 8370.0 4135.0 13,240 12,600Avg. pH 8.70 7.50 7.00 6.80 7.30 11.4 7.2Residual TS (%) 6.4 7.1 7.2 7.0 6.2 7.6 6.8Residual VS (%) 54 53 48 50 48 60 42Ammonia reduction (%) 17.3 25.4 52.0 62.4 59.52 12.2 64ATP level (mg L) 3 3 5.2 6.2 3 1.5 5Biogas composition CH4 36 46 52 47.4 54.5 33 56

CO2 55 43.1 37 36 35 57 40

0

500

1000

1500

2000

2500

3000

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31Days Elapsed

Bio

gas

in m

L

C:N 5C:N 10C:N 15C:N 20C:N 30Only MSWOnly LF

Fig. 1. Cumulative biogas yield at different C:N ratio of LF admixed with MSW.

0

1000

2000

3000

4000

5000

6000

7000

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35

Days Elapsed

Cum

ulat

ive

biog

as (m

L)

pH 4.5pH 5.5pH 6.5pH 7.5pH 8.5

Fig. 2. Cumulative biogas yield at C:N ratio of 15 at different pH controlled reactorsof LF admixed with MSW.

0

1000

2000

3000

4000

5000

1 3 5 7 10 15 20 30Days elapsed

NH 3-N

(mg.

L-1)

C:N 5C:N 10C:N 15C:N 20C:N 30Only LF MSW

Fig. 3. NH3–N levels at different C:N ratio of admixed LF and MSW.

0

1000

2000

3000

4000

5000

0 5 10 15 20 25 30 35Days Elapsed

NH 3

-N (m

g.L-1

)

pH 4.5pH 5.5pH 6.5pH 7.5pH 8.5

Fig. 4. NH3–N levels at different controlled pH and C:N of 15.

P. Shanmugam, N.J. Horan / Bioresource Technology 100 (2009) 4117–4120 4119

methanogenesis then VFA accumulation occurs and this results in aconsequent reduction in pH. Higher alkalinity was observed atlower C/N. Thus the low biogas yield of 0.08 Nm L/g VS removedat pH 4.5 can be attributed to VFA toxicity whereas the low biogasyield at a pH value of 8.5 is attributed to NH3 toxicity.

A large reduction in NH3 concentration was observed with theC:N experiment (Fig. 3), pH controlled reactors (Fig. 4), and theNH3 level was correlated well between both the total alkanity atdifferent C:N ratio, and the operating pH and ammonia released.This confirms the validity of waste co-blending and pH control inthe digester to reduce ammonia release and thus optimise meth-ane yield. It has also been reported that in practice pH control of

the digester in the acidic range can be achieved by maintaining ahigh OLR (Kato et al., 1999). Higher headspace CH4 content (60%)was noticed at 6.5 pH than the 8.5 pH (Table 2).

3.4. Biomass activity and VS removal as AD performance indicators

The measurement of volatile suspended solids (VSS) is widelypracticed to assess the efficiency of the anaerobic digestion

Table 2Anaerobic digester performance evaluation at different controlled pH conditions and constant C:N of 15.

Parameters 4.5 pH 5.5 pH 6.5 pH 7.5 pH 8.5 pH

Daily average biogas yield (Nm mL d�1) 1.5 65 170 106 47VSr (%) 41.00 49.16 68.63 59.12 58.47Specific Biogas yield (Nm L CH4/gmVSr) 0.08 0.25 0.48 0.37 0.27Average NH3–N (mg L�1) 510.50 713.94 817.17 1245.39 2473.06Average Alkalinity (mg L�1) 2240.77 5910.23 22060.00 15870.95 11440.77Average VFA (mg L�1) 26803.00 21083.0 10350.68 10648.00 10848.00Residual TS (%) 5.52 4.24 2.46 2.23 3.79Residual VS (%) 48.00 47.00 33.00 39.00 35.00Ammonia reduction (%) 88.10 83.3 80.85 56.98 49.4ATP level (mg L�1) 2 4.2 7.2 4.2 3.1Biogas composition CO2 56 54 40 52.5 43

CH4 40 42 60 46 40

4120 P. Shanmugam, N.J. Horan / Bioresource Technology 100 (2009) 4117–4120

process. A simpler and more rapid alternative that measures theATP content of a digester has been proposed as an alternative(Shanmugam and Horan, 2009). This early study was extended tothe present work which has confirmed the validity of this tech-nique. A high concentration of ATP (6.2 mg L�1) at the C:N ratioof 20 coincided with higher level of VS removal efficiency (62%),whereas a lower level of ATP (1.5 mg L�1) was coincident withthe lower level of VS removal observed for the LF waste alone.The ATP concentration during the pH controlled experiments wasmaximum at a pH of 6.5 and minimum at pH values of 4.5 and8.5 (Table 2). This coincided with the highest VS removal efficiencyof 68.6% at 6.5 pH and 41.0% at 4.5 pH (Table 2).

4. Conclusions

� The higher ammonia observed at lower C:N ratio and highlyalkaline pH (8.5) of the blended LF with MSW coincided withlower head space CH4 content.

� Blending LF waste to a C:N of 15 and pH of 6.5 reduced the con-centration of NH3 released during digestion by 80%, compared tounblended LF wastes with a pH of 11.4. At pH values of 8.5 andabove, NH3 toxicity was observed leading to a reduced biogasyield.

� By contrast at pH 4.5, VFA toxicity was prevalent, particularly forhighly proteinaceous solid waste such as leather fleshing and asa result the biogas yield was also reduced.

� The highest observed specific biogas yield was at near neutralpH of 6.5 and this pH value is achievable naturally by maintain-ing a high organic loading rate (OLR).

� Blending of co-digested LF and MSW waste helps to minimiseNH3 toxicity and thus maximise the renewable energy recoverythrough the anaerobic co-digestion process.

� The ATP measured for biomass activity was correlated well withVS removal and biogas yield confirms well with our previouslyreported results.

References

Angelidaki, I., Ahring, B.K., 1994. Anaerobic thermophillic digestion of manure atdifferent ammonia loads: effect of temperature. Water Research 28 (3), 727–731.

Angelidaki, I., Ellegaard, L., Ahring, B.K., 1998. A mathematical model for dynamicsimulation of anaerobic digestion of complex substrates. Focusing on ammoniainhibition. Biotechnology and Bioengineering 42, 159–160.

APHA, 1998. Standard Methods for the Examination of Water and Waste Water,20th ed. American Public Health association, American Water WorksAssociation Water Pollution control Federation, Washington, DC.

Bujoczek, G., Oleskiewicz, J., Sparling, R., Cenkoswki, S., 2000. High solid anaerobicdigestion of chicken manure. Journal of Agricultural Engineering 76,51–60.

Chen, Y., Cheng, J., Creamer, K.S., 2008. Inhibition of anaerobic digestion process: areview. Bioresource Technology 99, 4044–4064.

Defra, 2007. Waste Strategy. Department of Environment Food Rural affair andAgriculture, UK.

Demirer, S.U., Demirer, G.N., Chen, I.S., 2005. Ammonia removal from anaerobicallydigested dairy manure by struvite precipitation. Process Biochemistry 40,3667–3674.

Gallert, C., Henning, A., Winter, J., 2003. Scale-up of anaerobic digestion of thebiowaste fraction from domestic wastes. Water Research 37, 1441–1443.

Hansen, K.H., Anglidaki, I., Ahring, B.K., 1998. Anaerobic digestion of swine manure:inhibition by ammonia. Water Research 32 (1), 5–12.

Hills, J.D., 1979. Effect of carbon to nitrogen ratio on the anaerobic digestion of dairymanure. Agricultural Wastes 1 (4), 267–278.

Hoshimoto, A.G., 1986. Ammonia inhibition of methanogenesis from cattle wastes.Agricultural Wastes 17, 241–261.

Kato, A., Kankyogken, M., Noike, T., 1999. Ammonia toxicity control by pHregulation and organic loading rate in anaerobic digestion of agricultural andlivestock industrial wastes. Japanese Journal of Industrial Water 490, 18–32.

Kayhanian, M., 2004. Performance of high solids anaerobic digestion process undervarious ammonia concentrations. Journal of Chemical Technology andBiotechnology 59 (4), 349–352.

Koster, I.W., 2007. Characteristics of pH influenced adaptations of methanogenicsludge to ammonium toxicity. Journal of Chemical Technology Biotechnology36 (10), 445–455.

Koster, I.W., Lettinga, G., 1988. Anaerobic digestion at extreme ammoniaconcentrations. Biological Wastes 25, 51–59.

Maqueda, C., Perez, R.J.L., Lebrato, J., 2003. Study of struvite precipitation inanaerobic digesters. Water Research 28 (2), 411–416.

Murphy, J.D., McKeogh, E., Kiely, E., 2004. Technical/economic/environmentalanalysis of biogas utilization. Applied Energy 77, 407–427.

Nielsen, H.B., Angelidaki, I., 2008. Strategies for optimizing recovery of the biogasprocess following ammonia inhibition. Bioresource Technology 99, 7995–8001.

Ogunwande, G.A., Osunade, J.A., Adekalu, K.O., Ogunjimi, K.A.O., 2008. Nitrogen lossin chicken litter compost as affected by carbon to nitrogen ratio and turningfrequency. Bioresource Technology 99, 7495–7503.

Pechan, Z., Knappovfi, O., Petrovi, B., Adamec, O., 1987. Anaerobic digestion ofpoultry manure at high ammonium nitrogen concentrations. Biological Wastes20, 117–131.

Ramasami, T., Sreeram, K.J., Gayathri, R., 1999. Sources of pollution in tanneriesmitigation and case studies in United Nations Industrial DevelopmentOrganisation (UNIDO) manual on design. Operation and Maintenance ofTannery Effluent Treatment Plants 21, 32.

Rosenwinkel, K.H., Meyer, H., 1999. Anaerobic treatment of slaughterhouse residuesin municipal digesters. Water Science Technology 40 (1), 101–111.

Salerno, M.B., Park, W., Zuo, Y., Logan, B.E., 2006. Inhibition of biohydrogenproduction by ammonia. Water Research 40, 1167–1172.

Shanmugam, P., Horan, N.J., 2009. Simple and rapid methods to evaluate themethane potential and biomass yield for range of mixed solid wastes.Bioresource Technology 100, 471–474.

Shanmugam, P., Horan, N.J., 2008. Effect of C:N ratio, and OLR on fermentative H2

production from leather and municipal solid wastes using batch and CSTRreactors. In: 8th World Hydrogen Energy Conference, Brisbane, Australia.Organised by International Hydrogen Energy Association, Proceedings Pages,June 2008.

Sievers, D.M., Brune, D.E., 1978. Carbon/nitrogen ratio and anaerobic digestion ofswine waste. Transactions of the ASAE 21 (3), 537–541.

Sterling, M.C., Lacey, R.E., Engler, C.R., Ricke, S.C., 2001. Effects of ammonia nitrogenon H2 and CH4 production during anaerobic digestion of dairy cattle manure.Bioresource Technology 77, 9–18.

Strik, D.P.B., Domnanovich, A.M., Holubar, D.P., 2006. A pH based control ofammonia in biogas during anaerobic digestion of artificial pig manure andmaize silage. Process Biochemistry 41, 1235–1238.

Sung, S., Liu, T., 2003. Ammonia inhibition on thermophillic anaerobic digestion.Chemosphere 53, 43–52.

Taleghani, G., Kia, A.S., 2005. Technical–economical analysis of the Saveh biogaspower plant. Renewable Energy 30, 441–446.

Tomor, S.S., 1994. The status of biogas plants in India. Renewable Energy l5, 829–831.