Bioresource Technology 102 (2011) 729–735

Contents lists available at ScienceDirect

Bioresource Technology

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

Multi stage high rate biomethanation of poultry litter with self mixedanaerobic digester

A. Gangagni Rao a,⇑, S. Surya Prakash a, Johny Joseph a, A. Rajashekhara Reddy b, P.N. Sarma a

a Bioengineering and Environmental Centre, Indian Institute of Chemical Technology, Tarnaka, Hyderabad 500007, Andhra Pradesh, Indiab Department of Poultry Science, Sri Venkateswara Veterinary University, Rajendranagar, Hyderabad 500030, Andhra Pradesh, India

a r t i c l e i n f o

Article history:Received 16 April 2010Received in revised form 19 August 2010Accepted 20 August 2010Available online 26 August 2010

Keywords:Poultry litterBiomethanationUASBBiogasSMAD

0960-8524/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.biortech.2010.08.069

⇑ Corresponding author. Tel.: +91 40 27191663; faxE-mail address: gangagnirao@yahoo.com (A. Gang

a b s t r a c t

A multi stage high rate biomethanation process with novel self mixed anaerobic digester (SMAD) wasdeveloped in the present study to reduce the hydraulic residence time (HRT), increase the volatile solids(VS) loading rate, improve the VS destruction efficiency and enhance the methane yield. Specific designfeatures of SMAD were useful in mixing the digester contents without consuming power and de-alien-ated the problem of scum formation. In the first phase, poultry litter having 10% total solids (TS) was sub-jected to high rate biomethanation in multi stage configuration (SMAD-I and II in series with UASBreactor). It was observed that gross VS reduction of 58%, gross methane yield of 0.16 m3 kg�1 (VS reduced)and VS loading rate of 3.5 kg VS m�3 day�1 at HRT of 13 days was obtained. In the second phase SMAD-IIwas bypassed from the process scheme keeping the other parameters same as in the first phase. Theresults obtained were not as encouraging as in the first phase. The study showed that multi stage config-uration with SMAD design improved the anaerobic digestion process efficiency of poultry litter.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Poultry litter containing biodegradable organics and inorganicmatter with 75–80% moisture is generated in large quantities dur-ing intensive poultry farming. It contains nitrogen (1.22–1.63%),phosphorus (0.89–1.04%), potassium (1.34–1.7%) and also manymicronutrients such as zinc, copper, iron and selenium (Martinezet al., 2009; Vanotti et al., 2009; Ciborowski, 2001; Szogi and Vano-tti, 2009; Monteny et al., 2006). Poultry litter is usually treated byanaerobic methods for degradation of the organic matter. It isfound to be economically very attractive option for livestock andpoultry waste without any pretreatment (Kelleher et al., 2002;Ciborowski, 2001; Salminen and Rintala, 2002a; Sakar et al.,2009; Magbanua et al., 2001; Gangagni Rao et al., 2008a; Cantrellet al., 2008; Mata-Alvarez et al., 2000; Dubrovskis et al., 2008).Poultry litter generates more biogas compared to piggery and cat-tle waste (Itodo and Awulu, 1999; Callaghan et al., 1999). Thusthere is a great potential for generating biogas from poultry litter.There are certain limitations in using poultry litter for biogas gen-eration (Kelleher et al., 2002; Ciborowski, 2001; Salminen andRintala, 2002a; Sakar et al., 2009; Magbanua et al., 2001; GangagniRao et al., 2008a; Cantrell et al., 2008; Mata-Alvarez et al., 2000;Dubrovskis et al., 2008) as it is viscous in nature with high calciumcontent and sand/grit. Studies were reported on batch type

ll rights reserved.

: +91 40 27193159.agni Rao).

digesters for production of biogas from poultry litter where in allphases of anaerobic digestion (i.e., hydrolysis, acidification andmethanogenesis) take place in one vessel (Bujoczek et al., 2000;Callaghan et al., 1999; Magbanua et al., 2001; Abouelenien et al.,2009; Fantozzi and Buratti, 2009; Hill and Bolte, 2000). Hence tomaintain a feasible environment for optimal methanogenesis un-der such conditions various biochemical pathways need to be bal-anced. This can be achieved by providing high retention timesleading to increased volume of digester (Mata-Alvarez et al.,2000; Ciborowski, 2001; Salminen and Rintala, 2002a; Sakaret al., 2009; Dubrovskis et al., 2008). Moreover, such batch typeconventional plants without mixing are not suitable for the treat-ment of large quantities of poultry litter. Failure of such plantswas reported within two to three years of operation due to scumformation at the top and choking at the bottom (Dubrovskiset al., 2008; Sakar et al., 2009; Yadvika et al., 2004; Gangagni Raoet al., 2008b). The performance of batch digesters could be im-proved by installation of mechanical mixers (Kelleher et al.,2002; Sakar et al., 2009). It is reported that the performance ofthese mechanical mixers in pilot and full scale anaerobic digestersare not only unsatisfactory but also consume 20–30% of the energygenerated during the digestion process (Abouelenien et al., 2009;Kaparaju et al., 2008; Yadvika et al., 2004). However, high rate bio-methanation invariably requires complete mixing to enhance theperformance of the digester and accordingly digesters with novelmixing arrangements (Karim et al., 2005a,b; Yadvika et al., 2004)and two-phase biomethanation systems were developed to

730 A. Gangagni Rao et al. / Bioresource Technology 102 (2011) 729–735

improve the efficiency of the process in terms of hydraulic resi-dence time (HRT), volatile solids (VS) loading and destruction rateand biogas yield (Banks and Wang, 1999; Harikishan and Sung,2003; Nielsen et al., 2004; Nuri et al., 2001; Raynal et al., 1998;Salminen and Rintala, 2002b; Wang and Charles, 2003). Most ofthe earlier studies showed that mixing, staging and phasing werevery essential to enhance the efficiency of the high rate biometha-nation process. However, there are no literature reports combiningall these features for the anaerobic treatment of poultry litter. Inaddition to this, in India most of the poultry farms generate poultrylitter in the range of 2–20 tons per day (Mehta et al., 2002; Sivaku-mar et al., 2008). Conventional anaerobic batch digesters cannot beused for this type of farms. Therefore, the present studies areaimed at high rate biomethanation of poultry litter with multistage configuration and digester with novel mixing arrangementand the results obtained are presented and discussed in thiscommunication.

2. Methods

2.1. Poultry litter

Poultry litter was collected from the Live Stock Research stationof Sri Venkateswara Veterinary University (SVVU), Rajendranagar,Hyderabad, Andhra Pradesh, India in sun-dried conditions. The lit-ter was brought to work place as per requirement and stored in drycondition. Poultry litter slurry was prepared from the same storageand fed to the digester everyday as per requirement.

2.2. Experimental set up

The experimental set up consisted of 50 l feed slurry prepara-tion tank, two (I & II) Self Mixing Anaerobic Digesters (SMAD) inseries with Vibro screen-I & II, leachate tank, Upflow anaerobicsludge blanket (UASB) reactor, UASB recirculation tank and floatingdome type biogas holder. All the units were made of High DensityPoly Ethylene (HDPE). Schematic flow diagram of the process isshown in Fig. 1. The feed slurry preparation tank was circular inshape with conical bottom and a damper to remove grits. It wasalso fitted with aeration and overflow arrangement to dischargethe homogenized slurry to SMAD-I. SMAD-I & II was having twocompartments namely bottom and upper chamber. Both the cham-bers of SMAD were hydraulically connected with central draft pipe(Gangagni Rao et al., 2008b). Pressure developed in the bottomchamber due to the production of biogas was utilized for mixing

SMAD-I

GAS HOLDER

RecycleWater

FEED

Vibroscreen-I

Makeup Water

Recycle

Fig. 1. Schematic flow diagram of multi step high rate anaero

the slurry. Fresh slurry was fed to the bottom chamber of thedigester and the slurry travels up and down in the both the cham-bers of digester through draft tube due to the differential pressurein both the chambers. The movement of the slurry across twochambers occurred according to the automatic opening of the valveas per the set pressure. During this movement, whenever slurryfalls into the bottom chamber created vibrant mixing in the bottomchamber accordingly slurry in the bottom chamber becomeshomogeneous. SMAD-I and II having capacities of 364 and 187 l,respectively were connected in series and fitted with heating coil(on/off control) to maintain the temperature at 37 ± 1 �C. Provisionwas made for measuring slurry temperature at the digester outlet.UASB reactor (liquid retaining capacity 35 l and total volume 50 l)was of 120 cm height and 23 cm internal diameter and made ofHDPE. The reactor was insulated to maintain the temperature at37 ± 1 �C and the feed was pumped by means of a centrifugalpump. Provision was made for measuring liquid temperature atthe reactor outlet. This was found to be always in the range of36–38 �C. The leachate was introduced at the bottom of the UASBreactor via an inlet distribution network and the gas, solid and li-quid phases were separated at the top by means of a three-phaseseparator. The gas produced was measured using a wet gas flowmeter. Vibro screen consisting of mechanical sieve made of stain-less steel (SS) wire mesh fitted with electrically operated magneticvibrator was arranged to collect filtrate and cake without manualhandling. Vibro screens I & II were arranged successively forSMAD-I & II, respectively. The volume of the gasholder was around600 l approximately.

2.3. Inoculum

The seeding sludge was obtained from an anaerobic digester ofthe sewage treatment plant at Hyderabad and acclimatized for30 days with poultry litter slurry (10% TS) by mixing in 1:1 ratio.This was used as inoculum for all anaerobic reactors in the study(SMAD-I & II and UASB). All the chemicals used for the analysisduring the experiments were of AR grade.

2.4. Analytical methods

Initially, fresh poultry droppings were characterized for pH, to-tal solids (TS), volatile solids (VS), fixed solids (FS), total Kjeldahlnitrogen (TKN), ammoniacal nitrogen (NH4-N) and organic nitro-gen. During the course of experiments characteristics of poultry lit-ter slurry (Inlet for SMAD-I & II, outlet from SMAD-II) for pH, TS &VS and leachate (filtrate collected from Vibro screen-I & II and

UAS B

SMAD-II

OUTLET WATER

Vibroscree n-II

Drying yard

bic digestion process for the treatment of poultry litter.

A. Gangagni Rao et al. / Bioresource Technology 102 (2011) 729–735 731

treated leachate from UASB) for pH, COD, BOD, volatile fatty acids(VFA), alkalinity and sulfate were determined daily or weekly asper requirement. The characteristics of the poultry litter viz. pH,TS, VS, FS, TKN, NH4-N and organic nitrogen were determined asper standard methods (APHA, 1998). COD, BOD, VFA, alkalinity,pH and sulfate for leachate were also determined as per standardmethods (APHA, 1998). The volume of biogas produced was mea-sured and composition was estimated by using Orsat apparatus(BS Method, 1971). Analysis of each parameter was carried outusing the triplicates and variation was within ±1.0% and the varia-tion was added in the values.

2.5. Experimental procedure

2.5.1. Phase-IPoultry litter slurry (10% of TS) was soaked in the feed tank for

one day, thoroughly mixed with air for few minutes after soakingand allowed to settle. The homogenized slurry was charged tothe SMAD-I and allowed to digest under anaerobic conditions.The settled grit, sand and shells from the feed tank were separatedand sun dried. Shells were separated and reused. The dischargefrom SMAD-I was separated using Vibro screen-I. The separatedsolids from the Vibro screen-I were fed to SMAD-II by maintainingthe TS concentration in the range of 10–15% with fresh water. Theliquid (leachate) from the Vibro screen-I was pumped to the leach-ate tank. In SMAD-II also the slurry was allowed to digest underanaerobic conditions and the digested slurry from SMAD-II wasseparated using Vibro screen-II. Solids from Vibro screen-II weresun dried in trays and the liquid was pumped to the same leachatetank. The leachate was fed to the UASB reactor. The treated leach-ate from UASB was used along with fresh make up water to pre-pare the feed slurry. The biogas generated in the SMAD-I & II andUASB reactor was stored in biogas holder, measured and analyzedfor methane using Orsat apparatus.

2.5.2. Phase-IIIn this phase of experimentation, SMAD-II was bypassed from

the original scheme as shown in Fig.1 in order to asses the effectof single stage SMAD instead of two stage SMAD at same HRTand other conditions. A 16 l UASB reactor was used in place of50 l UASB reactor with same material of construction. Other oper-ational procedure remained the same as in the Phase-I.

3. Results and discussion

3.1. Poultry litter characteristics

The poultry litter used for studies in the present work was cagepoultry litter, which included the bedding material used during thepoultry production cycle. Materials used as bedding include straw,sawdust, wood shavings, shredded paper and peanut or rice hulls.During the production cycle bedding material was mixed with lit-ter. At the end of the cycle both were removed together. The mois-ture content (MC) of fresh droppings of poultry litter is 71 ± 1.0%and other characteristics of fresh droppings of poultry litter slurryon dry basis are as follows; TS: 29 ± 0.5%; VS: 81 ± 1.0% of TS; FS:19 ± 0.5% of TS; TKN: 1.4 ± 0.1% of TS; NH4-N: 0.2 ± 0.1% of TS; or-ganic nitrogen: 1.2 ± 0.1% of TS. It could be noted from the charac-teristics that composition of the litter was predominantly carbon(VS) with major amounts of nitrogen. The characteristics showedthat poultry litter is a very good material for biomethanation as re-ported earlier (Kelleher et al., 2002; Ciborowski, 2001; Salminenand Rintala, 2002a,b; Sakar et al., 2009; Magbanua et al., 2001;Gangagni Rao et al., 2008a,b; Cantrell et al., 2008; Mata-Alvarezet al., 2000; Dubrovskis et al., 2008). Earlier studies on biometha-

nation of poultry litter revealed that the efficiency of conversionto methane was found to decrease with increasing total solids(TS%) concentration of the feed slurry. Highest solids concentrationat which digestion was successful was approximately 10% total sol-ids (Bujoczek et al., 2000). Accordingly, throughout the study poul-try litter slurry of 10–12% TS concentration was used.

3.2. Start up and stabilization of process

Initially SMAD-I & II were seeded with 50 l of acclimatizedsludge (which was prepared as explained previously). Poultry litterwas soaked with water in the feed tank for 24 h and homogenizedslurry having approximately 10% TS was prepared and fed toSMAD-I at initial VS loading rate of 0.8 kg VS/m3/day and HRT of45 days. Subsequently, everyday SMAD-I was fed with the sameamount of slurry to maintain the aforesaid VS loading rate andHRT. From the second day onwards, slurry from the SMAD-I wassieved in Vibro screen-I to separate solids and leachate. Partiallydigested solids from SMAD-I were mixed with fresh water (approx-imately 12% TS) and fed to SMAD-II at VS loading rate of 0.8 kg VS/m3/day and HRT of 45 days. The following day digested slurry fromSMAD-II was sieved in Vibro screen-II and solids and leachate wasseparated. SMAD-II digested solids were dried under sun light intrays. The leachate from both the stages was collected in a tankand fed to the UASB reactor at organic loading rate (OLR) of3.5 kg COD/m3/day and HRT of 84 h. The process was continuedfor 20 days and VS reduction across SMAD-I & II and COD reductionof leachate in the UASB was determined regularly. Other parame-ters like pH, VFA and alkalinity were also determined. During thecourse of first 20 days operation, it was observed that VS reductionin SMAD-I and II improved from 10–45% to 10–35%, respectivelyand stabilized at 40–45% and 30–35%, respectively. Overall VSreduction of 55–60% was obtained cumulatively in both the digest-ers. Total methane gas production from the SMAD-I & II was ob-served to be in the range of 1.2–1.9m3 CH4/(kg VS reduced). CODreduction of 90–94% and methane yield of 0.31–0.32 m3 CH4/(kgCOD reduced) was obtained in the UASB reactor during this period.It was concluded from this observation that SMAD-I & II were sta-ble at VS loading rate of 0.8 kg VS/m3/day and HRT of 45 days. Itwas also observed that UASB reactor was stable at OLR of 3.5 kgCOD/m3/day and HRT of 84 h.

3.3. Optimization and performance evaluation of SMAD

It was essential that optimum design parameters of SMAD viz.VS loading rate, hydraulic residence time (HRT) need to be deter-mined at which VS reduction and methane yield was high. It wasalso essential to observe stable performance of SMAD at optimizedparameters. In order to establish this, the process was continuouslyoperated at different VS loading rates and the performance ofSMAD-I & II were evaluated for various important parameters.

3.3.1. SMAD-IAfter stabilization of performance at VS loading rate of

0.8 kg VS/m3/day and HRT of 45 days, the VS loading rate ofSMAD-I was increased in a stepwise manner up to 7.7 kg VS/m3/day by decreasing the HRT to 4.5 days in order to establish theoptimum loading rate. The reactor was operated until a steadystate condition was reached at each loading rate; steady state con-dition was characterized by a constant gas production rate (±5%)and effluent VS level (±8%). The performance of the SMAD-I at stea-dy state in terms of VS removal, HRT and methane yield at variousVS loading rates is depicted in Fig. 2. The figure shows that duringthe increase of VS loading rate from 0.8 to 5.9 kg/m3/day and cor-responding decrease of HRT from 45 to 6 days, the VS reductionand methane yield was in the range of 47.1–41.3% and 0.197–

Fig. 2. Optimization of SMAD-I (standard error: 1%).

732 A. Gangagni Rao et al. / Bioresource Technology 102 (2011) 729–735

0.192 m3/(kg VS reduced), respectively. When the digester wasoperated at VS loading rate in the range of 6.3–7.7 kg VS/m3/dayand HRT of 5.5–4.5 days, the VS reduction and methane yielddropped and were observed to be in the in the range of 28.7–22.3% and 0.16–0.128 m3/(kg VS reduced), respectively. Thereforeit could be established from the above results that, digester perfor-mance was optimum at VS loading rate of 5.9 kg VS/m3/day andHRT of 6 days yielding an average VS reduction and methane yieldof 40% and 1.8 m3/(kg Vs reduced) respectively. Accordingly,SMAD-I was operated at optimum values of 5.9 kg VS/m3/day (VSloading rate) and 6 days (HRT) for 15 weeks subsequently. The sta-ble performance of SMAD-I (weekly) in terms of VS loading rate, VSreduction and methane yield was tabulated and shown in Table 1.Table 1 shows that when the digester was operated at VS loadingrate in the range of 5.5–6.1 kg VS/m3/day, consistent VS reductionin the range of 38–43% and methane yield in the range of 0.18–0.19 m3/(kg VS reduced) was obtained.

3.3.2. SMAD-IIAfter stabilization of performance at VS loading rate of

0.8 kg VS/m3/day and HRT of 45 days, VS loading rate of SMAD-IIwas increased in a stepwise manner up to 11.64 kg VS/m3/day bydecreasing the HRT to 3 days in order to establish the optimumloading rate. The digester was operated until a steady state condi-tion was reached at each loading rate. Steady state condition wascharacterized by a constant gas production rate (±5%) and effluentVS level (±8%). The performance of the digester at steady state in

Table 1Long term performance of SMAD-I, SMAD-II and UASB (standard error: 1%).

S. No(Week)

SMAD-I SMAD-II

VS loading rate(kg VS/m3/day)

VSreduction(%)

Methane yield(m3/kg VSreduced)

VS loading rate(kg VS/m3/day)

Vre(%

1 5.8 39 0.19 6.8 32 5.9 41 0.19 6.7 23 5.7 41 0.19 6.5 24 5.9 40 0.19 7.0 35 5.8 39 0.18 6.8 26 6.0 43 0.18 6.7 37 5.8 40 0.19 6.8 28 5.7 41 0.19 6.6 29 5.9 38 0.19 7.2 2

10 5.6 39 0.19 6.7 311 5.8 38 0.19 7.0 312 5.5 40 0.18 6.4 313 6.1 42 0.19 7.0 214 5.9 40 0.18 6.9 215 5.7 38 0.18 6.9 3

terms of VS removal, HRT and methane yield at various VS loadingrates was plotted and presented in Fig. 3. Fig. 3 shows that duringthe increase of VS loading rate from 0.8 to 6.9 kg/m3/day and cor-responding decrease of HRT from 45 to 5 days, the VS reductionand methane yield was in the range of 33–28.3% and 0.12–0.13 m3/(kg VS reduced) respectively. When the digester was oper-ated at VS loading rate in the range of 8.8–11.64 kg VS/m3/day andHRT of 4–3 days, the VS reduction and methane yield dropped tothe range of 25–20% and 0.06–0.02 m3/(kg VS reduced) respec-tively. It could be concluded from the above results that, the diges-ter performance was optimum at VS loading rate of 6.9 kg VS/m3/day and HRT of 5 days yielding an average VS reduction and meth-ane yield of 28% and 0.13 m3/(kg Vs reduced), respectively. Accord-ingly, SMAD-II was operated at optimum values of 6.9 kg VS/m3/day (VS loading rate) and 5 days (HRT) for 15 weeks subsequently.The stable performance of SMAD-II (weekly) in terms of VS loadingrate, VS reduction and methane yield was tabulated and presentedin Table 1. Table 1 shows that when the digester was operated atVS loading rate in the range of 6.4–7.2 kg VS/m3/day, consistentVS reduction in the range of 27–31% and methane yield in therange of 0.11–0.14 m3/(kg VS reduced) was obtained.

3.4. Characteristics of leachate collected from SMAD

The leachate collected from SMAD-I & II was having pH in be-tween 6.8 and 7.4, COD in the range of 9000 ± 80–11,000 ± 100 mg/l, BOD5 in the range of 6000 ± 40–8000 ± 60 mg/l, VFA in the range of 4000 ± 20–6000 ± 40 mg/l, sulfate in between975±6 and 1425±8 mg/l, and alkalinity in the range of 2000 ± 15–4000 ± 25 mg/l. BOD/COD ratio of the leachate was in between0.65 and 0.75 which showed that leachate collected from theSMAD-I & II was highly biodegradable (Gangagni Rao et al., 2008a).

3.5. Optimization and performance evaluation of UASB

Optimum OLR of the UASB at which maximum COD reductionand methane yield at low HRT could be obtained was evaluated.After start up and stabilization, OLR of UASB was increased in astepwise manner from 4.1 to 30.6 kg COD/m3/day by decreasingthe HRT from 72 to 8 h. At each loading rate, the reactor was oper-ated until a steady state condition was established. Steady statecondition was characterized by a constant gas production rate(±5%) and effluent COD level (±8%). Stable performance of the reac-tor at each loading rate was plotted and is shown in Fig. 4. Fig 4shows that during the increase of OLR from 4.1 to 16.2 kg COD/

UASB

Sduction)

Methane yield(m3/kg VSreduced)

OLR(kg COD/m3/day)

CODReduction(%)

Methane yield (m3/kg COD reduced)

0 0.13 16.6 91 0.307 0.12 15.5 91 0.317 0.12 15.8 91 0.300 0.11 15.9 91 0.318 0.13 16.5 92 0.300 0.13 15.9 92 0.338 0.13 15.7 91 0.317 0.13 15.6 91 0.318 0.13 15.6 92 0.320 0.11 15.6 90 0.320 0.12 15.9 91 0.320 0.14 16.1 91 0.328 0.13 15.6 90 0.319 0.11 16.5 91 0.301 0.12 16.0 92 0.31

Fig. 3. Optimization of SMAD-II (standard error: 1%).

Fig. 4. Optimization of UASB (standard error: 1%).

Fig. 5. Long term overall performance of the process (standard error: 1%).

A. Gangagni Rao et al. / Bioresource Technology 102 (2011) 729–735 733

m3/day, COD reduction and methane yield were in the range of93–90% and 2.8–3.1 m3 CH4/(kg COD reduced), respectively. How-ever, when the OLR increased from 16.2 to 30.6 kg COD/m3/day,COD reduction and methane yield fell to the range of 66–55%and 2.0–1.38 m3 CH4/kg COD reduced, respectively. The results ob-tained revealed that UASB reactor performance was better in termsof COD reduction and methane yield at an OLR of 16.2 kg COD/m3/day compared to the higher OLR values. In the present study trea-ted leachate from the UASB reactor was recycled back to the feedpreparation tank for making slurry. Therefore, COD of the treatedleachate should be as low as possible so that extraction of solubleorganics from the litter was efficient. Hence, UASB operations wereoptimized at an OLR of 16.2 kg COD/m3/day and HRT of 18 h. Theleachate collected from SMAD-I & II was diluted if required while

feeding to the UASB reactor so that optimum OLR in the range of15–17 kg COD/m3/day and HRT of 18 h could be maintained. Per-formance of the UASB reactor (weekly) in terms of COD reduction,OLR and methane yield was tabulated and presented in Table 1.Table 1 shows that during the 15 weeks of operation when theOLR was maintained in the range of 15.5–16.6 kg COD/m3/day,COD reduction and methane yield was in the range of 90–92%and 0.3–0.33 m3/(kg COD reduced), respectively. The performanceof the UASB reactor in the present study was comparable to theUASB reactor used for treatment of poultry litter leachate (Gang-agni Rao et al., 2008a), liquid fraction of hen manure (Sergeyet al., 1998) and poultry slaughter wastewater (Chavez et al.,2005) in terms of COD reduction, OLR and methane yield.

3.6. Overall performance of the process

The overall performance of the process for all the 15 weeks ofoperation (weekly average values) in terms of VS reduction, VSloading rate and methane yield was plotted and shown Fig. 5. In or-der to calculate the gross VS loading rate, volume of the UASB reac-tor was also taken into consideration along with the volumes ofSMAD-I and II as leachate collected in SMAD-I and II was liquidfraction of the organic matter in the poultry litter. Similarly, biogasgenerated in the UASB reactor was also taken into considerationwhile calculating methane yield of the overall process. The resultsindicated that (Fig. 5) VS reduction and methane yield in the rangeof 55–60% and 0.31–0.34 m3/(kg VS reduced) respectively could beobtained at gross VS loading of 3.5 kg VS/m3/day. Average methanecontent in the biogas was found to be in the range of 60–65%. Thetotal HRT of the process was 12.75 days, which included 1 day infeed tank, 6 days in SMAD-I, 5 days in SMAD-II and 18 h in UASBreactor.

The anaerobic treatment of poultry litter involves two distinctstages (Kelleher et al., 2002; Ciborowski, 2001; Salminen and Rint-ala, 2002a; Sakar et al., 2009). In the first stage, complex compo-nents, including fats, proteins and polysaccharides, werehydrolysed and broken down to their component subunits. Thisfirst stage is commonly referred to as acid fermentation. The sec-ond stage involves the conversion of the hydrolysis products togases (mainly methane and CO2) by several different species ofstrictly anaerobic bacteria and is referred to as methane fermenta-tion. Generally, in two-phase systems, solid waste as such atappropriate TS concentration would be subjected to acidogenesisand methanogenesis in two digesters in series by maintainingthe appropriate pH in each digester. Due to this phase separation,the efficiency of the process may increase (Banks and Wang, 1999;Nielsen et al., 2004; Raynal et al., 1998; Wang and Charles, 2003)compared to the single stage process. But problems of solid diges-ter (Mata-Alvarez et al., 2000) remain the same as stated earlier.Further acid needs to be added to the acidogenic digester in orderto maintain the pH in the desired range. In the present study,leachate was extracted efficiently at its original pH of 7.2–7.4 inSMAD-I & II and treated separately in high rate UASB reactor with-out acid addition. Extraction of organics into the liquid phase notonly reduced the total HRT of the process, but also increased themethane yield and VS loading rate. A comparative study carriedout by Nielsen et al. (2004) with cattle manure with two-stageand a conventional single-stage reactor systems showed that atan organic loading of 3 g VS per liter per day, the two-stage setuphad 6–8% higher specific methane yield and 9% more effective VSremoval than the conventional single-stage reactor. Yet anotherstudy conducted by Karim et al. (2005a,b) showed that the effectof mixing and the mode of mixing became prominent in the caseof the digesters fed with thicker slurry (10%). In this study differentmodes of mixing like slurry recirculation, impeller and biogasrecirculation produced approximately 29%, 22% and 15% more bio-

Table 2Overall performance data for processes based on single stage and two stage SMAD (standard error: 1%).

Feed tank SMAD-I SMAD-II UASB Cumulative

Phase-IHRT (days) 1.00 6.00 5.00 0.75 12.75Volume (L) 50.00 364.00 187.00 35.00 636.00Methane gas produced (m3/day) 0.16 0.05 0.13 0.34Methane yield 0.18 0.13 0.30 0.29Gross methane yield (m3/kg VS added) 0.16VS or COD reduction (%) 40.00 28.00 90.00 57.00VLR or OLR* 5.90 6.90 16.20 3.50

Phase-IIHRT (days) 1.00 11.00 0.75 12.75Volume (L) 50.00 364.00 16.00 430.00Methane gas produced (m3/day) 0.09 0.06 0.15Methane yield 0.18 0.30 0.29Gross methane yield (m3/kg VS added) 0.12VS or COD reduction (%) 43.00 93.00 43.00VLR or OLR* 3.20 19.60 3.10

* OLR and COD are related to UASB and VLR and VS is related to SMAD.

734 A. Gangagni Rao et al. / Bioresource Technology 102 (2011) 729–735

gas than unmixed digester respectively. Most of the previous stud-ies showed that mixing, staging and phasing were very essential toenhance the efficiency of the high rate biomethanation process interms of HRT, VS reduction, VS loading and methane yield (Kelle-her et al., 2002; Ciborowski, 2001; Salminen and Rintala, 2002a;Sakar et al., 2009). However, there are no literature reports com-bining all these features for the anaerobic treatment of poultry lit-ter. Therefore, it could be stated that the present study wassuccessful in combining all the aforesaid configurations to getthe best effect of all the features of anaerobic treatment. In thepresent study, methane yield in the range of 0.186–0.2 m3/kg VSadded (0.31–0.34 m3/kg VS reduced) and VS reduction in the rangeof 55–60% at VS loading rate of 3.5 kg VS/m3/day and HRT of12.75 days was obtained for the overall process which could becomparable to the previous results (Kelleher et al., 2002; Ciborow-ski, 2001; Salminen and Rintala, 2002a; Sakar et al., 2009). In theearlier studies carried out with CSTR (Fantozzi and Buratti, 2009;Karim et al., 2005a,b; Kaparaju et al., 2008; Yadvika et al., 2004;Kelleher et al., 2002; Ciborowski, 2001; Salminen and Rintala,2002a; Sakar et al., 2009) approximately 20–25% of the energy gen-erated in the systems was reused for mixing the contents of the di-gester. This not only increases the capital cost (cost of mixer, motoretc.), but also reduces the net energy generated. Further, at higherscales CSTR design is not practicable. In the present investigations,pressure exerted by the biogas was utilized for mixing the digestercontents and thereby net energy generated was more compared tothe CSTR type digester, as there was no utilization of energy formixing. Hence, the results obtained in our study were better thanthe results obtained earlier for CSTR type designs.

3.7. Effect of staging of SMAD in the process

During the course of operation of the process at optimized con-ditions as specified above, in order to study the effect of staging,SMAD-II was bypassed in Phase-II and HRT of 11 days (total HRTof SMAD-I & II in Phase-I) was maintained in SMAD-I alone keepingthe total HRT of the process the same as in Phase-I (12.75 days).Process based on single stage SMAD was stabilized and operatedfor 30 days. At stable conditions overall performance in terms ofVS loading rate of 3.1 kg VS/m3/day and VS reduction of 43% wasobtained at the same HRT. The comparative overall performancedata for processes based on single stage and two stages SMADwas tabulated and given in Table 2. Table 2 reveals that in the pro-cess based on two stage SMAD, VS loading rate of 3.5 kg VS/m3/day,VS reduction of 57% VS reduction and methane yield of0.29 m3 CH4/(kg VS reduced) was achieved. In the process based

on single stage, VS loading rate of 3.1 kg VS /m3/day, VS reductionof 43% VS reduction and methane yield of 0.29 m3 CH4/(kg VS re-duced) was obtained at the same HRT of 12.75 days. The extractionof organics as VFA into the water from leachate is mainly depen-dent on the VFA concentration gradient (difference in VFA concen-tration of saturated leachate and unsaturated leachate at thatpressure and temperature) of leachate. The VFA concentration gra-dient was more in two-stage because the leachate was replacedwell before saturation with fresh water in the second stage. In sin-gle stage VFA concentration gradient was less because leachatewas not replaced with fresh water even though SMAD-I HRT was11 days (same as the HRT of SMAD-I and II in the first phase). Thisrevealed the fact that due to staging the extraction of organics intothe liquid phase from the solid organic matter was better in twostage system compared to the single stage. Accordingly highervalues in terms of VS loading rate and reduction was obtained intwo-stage process. The gross yield of methane was also more intwo-stage process compared to the single stage.

4. Recommendation

A pilot scale plant based on the process developed at laboratoryscale was installed at Live Stock Research Station of SVVU, Rajendr-anagar, Hyderabad, Andhra Pradesh for the treatment of approxi-mately 500 l of poultry litter slurry (TS of 10%) per day. The planthas been working for the past three years and plant performancehas been stable in terms of design parameters. Therefore, the de-sign in the present study could be scaled up for higher quantitiesof poultry litter and the process developed is remunerative.

5. Conclusions

SMAD is an improved digester for the high rate biomethanationof poultry litter and scum formation problem was delineated com-pletely due to its unique features of self mixing mechanism. Multistage configuration with two stage SMAD resulted in enhancementof overall efficiency of the process. The entire process was carriedout at the original pH of the poultry litter without the addition ofbuffers. Hence significant economic benefits could be derived forfull-scale installation.

Acknowledgements

The authors are grateful to Department of biotechnology (DBT),New Delhi Government of India for funding the project and theDirector, IICT, for his encouragement in carrying out this work.

A. Gangagni Rao et al. / Bioresource Technology 102 (2011) 729–735 735

References

Abouelenien, Fatma, Nakashimada, Yutaka, Nishio, Naomichi, 2009. Dry mesophilicfermentation of chicken manure for production of methane by repeated batchculture. Journal of Bioscience and Bioengineering 107 (3), 293–295.

APHA-AWWA-WPCF, 1998. Standard Methods for the Examination of Water andWastewater. American Public Health Association, Washington, D.C.

Banks, C.J., Wang, Z., 1999. Development of a two phase anaerobic digester forthe treatment of mixed abattoir wastes. Water Science and Technology 40 (1),69–76.

British, S. Institution, 1971. Methods for the Sampling and Analysis of Flue Gases.Analysis by the Orsat Apparatus, UK, p. S1756.

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

Callaghan, F.J., Wase, D.A.G., Thayanithy, K., Forster, C.F., 1999. Co-digestion ofwaste organic solids: batch studies. Bioresource Technology 67, 117–122.

Cantrell, Keri B., Ducey, Thomas F., Ro, Kyoung S., Hunt, Patrick G., 2008. Livestockwaste-to-bioenergy generation opportunities. Bioresource Technology 99 (17),7941–7953.

Chavez, C.P.R., Dendooven, Castillo L.L., Escamilla-Silva, E.M., 2005. Poultryslaughter wastewater treatment with an up-flow anaerobic sludge blanket(UASB) reactor. Bioresource Technology 96, 1730–1736.

Ciborowski, Peter, 2001. Anaerobic Digestion of Livestock Manure for PollutionControl and Energy Production: A Feasibility Assessment. Minnesota PollutionControl Agency. <http://www.pca.state.mn.us/publications/p-gen4-02.pdf>.

Dubrovskis, Vilis, Plume, Imants, Straume, Indulis, 2008. Anaerobic Digestion ofCow and Broiler Manure. <http://tf.llu.lv/conference/proceedings2008/Papers/10_DubrovskisPlume.pdf>.

Fantozzi, Francesco, Buratti, Cinzia, 2009. Biogas production from differentsubstrates in an experimental continuously stirred tank reactor anaerobicdigester. Bioresource Technology 100 (23), 5783–5789.

Gangagni Rao, A., Sasi Kanth Reddy, T., Surya Prakash, S., Vanajakshi, J., Joseph,Johny, Jetty, Annapurna, Rajashekhara Reddy, A., Sarma, P.N., 2008a.Biomethanation of poultry litter leachate in UASB reactor coupled withammonia stripper for enhancement of overall performance. BioresourceTechnology 99 (18), 8679–8684.

Gangagni Rao, A., Surya Prakash, S., Jetty, Annapurna, Rajashekhara Reddy, A.,Sarma, P.N., 2008b. Self Mixed Anaerobic Digester. WIPO Patent ApplicationWO/2008/026221. <http://www.sumobrain.com/patents/wipo/Self-mixing-anaerobic-digester-useful/WO2008026221.html>.

Harikishan, S., Sung, Shihwu, 2003. Cattle waste treatment and Class A biosolidproduction using temperature-phased anaerobic digester. Advances inEnvironmental Research 7, 701–706.

Hill, D.T., Bolte, J.P., 2000. Methane production from low solid concentration liquidswine waste using conventional anaerobic fermentation. BioresourceTechnology 74, 241–247.

Itodo, I.N., Awulu, J.O., 1999. Effects of total solids concentrations of poultry, cattle,and piggery waste slurries on biogas yield. Transactions of the ASAE 42, 1853–1855.

Kaparaju, Prasad, Buendia, Inmaculada, Ellegaard, Lars, Angelidakia, Irini, 2008.Effects of mixing on methane production during thermophilic anaerobicdigestion of manure: lab-scale and pilot-scale studies. BioresourceTechnology 99 (11), 4919–4928.

Karim, Khursheed, Rebecca, Hoffmann, Thomas Klasson, K., Al-Dahhan, M.H., 2005a.Anaerobic digestion of animal waste: effect of mode of mixing. Water Research39 (15), 3597–3606.

Karim, Khursheed, Rebecca, Hoffmann, Thomas Klasson, K., Al-Dahhan, M.H., 2005b.Anaerobic digestion of animal waste: waste strength versus impact of mixing.Bioresource Technology 96, 1771–1781.

Kelleher, B.P., Leahy, J.J., Henihan, A.M., O’Dwyer, T.F., Sutton, D., Leahy, M.J., 2002.Advances in poultry litter disposal technology – a review. BioresourceTechnology 83, 27–36.

Magbanua, Benjamin S., Adams, Thomas T., Johnston, Phillip, 2001. Anaerobiccodigestion of hog and poultry waste. Bioresource Technology 76, 165–168.

Martinez, Jose, Dabert, Patrick, Barrington, Suzelle, Burton, Colin, 2009. Livestockwaste treatment systems for environmental quality, food safety, andsustainability. Bioresource Technology 100 (22), 5527–5536.

Mata-Alvarez, J., Mace, S., Llabres, P., 2000. Anaerobic digestion of organic solidwastes: an overview of research achievements and perspectives. BioresourceTechnology 74, 3–16.

Mehta, R., Nambiar, R.G. et al., 2002. Livestock Industrialization, Trade and SocialHealth Environment Impact in Developing Countries: A Case Study of IndianPoultry Sector – (Mimeo) Phase I project report submitted to IFPRI. <http://www.fao.org/WAIRDOCS/LEAD/x6170e/x6170e2k.htm>.

Monteny, Gert-Jan, Bannink, Andre, Chadwick, David, 2006. Greenhouse gasabatement strategies for animal husbandry. Agriculture, Ecosystems andEnvironment 112 (2–3), 163–170.

Nielsen, H.B., Mladenovska, Z., Westermann, P., Ahring, B.K., 2004. Comparison oftwo-stage thermophilic (68 �C/55 �C) anaerobic digestion with one-stagethermophilic (55 �C) digestion of cattle manure. Biotechnology andBioengineering 86 (3), 291–300.

Nuri, A., Pepi, U., Richard, E.S., 2001. Effect of process configuration and substratecomplexity on the performance of anaerobic processes. Water Research 35,817–829.

Raynal, J.J., Delgenks, P., Moletta, R., 1998. Two-phase anaerobic digestion of solidwastes by a multiple liquefaction reactors process. Bioresource Technology 65,97–103.

Sakar, Suleyman, Yetilmezsoy, Kaan, Kocak, Emel, 2009. Anaerobic digestiontechnology in poultry and livestock waste treatment – a literature review.Waste Management and Research 27 (1), 3–18.

Salminen, E., Rintala, J., 2002a. Anaerobic digestion of organic solid poultryslaughter waste – a review. Bioresource Technology 83, 13–26.

Salminen, E.A., Rintala, Jukka A., 2002b. Semi-continuous anaerobic digestion ofsolid poultry slaughterhouse waste: effect of hydraulic retention time andloading. Water Research 36, 3175–3182.

Sergey, K., Vyacheslav, F., Alla, N., 1998. Anaerobic treatment of liquid fraction ofhen manure in UASB reactors. Bioresource Technology 65, 221–225.

Sivakumar, K., Ramesh Saravana Kumar, V., Richard Jagatheesan, P.N., Viswanathan,K., Chandrasekaran, D., 2008. Seasonal variations in composting process of deadpoultry birds. Bioresource Technology 99 (9), 3708–3713.

Szogi, A.A., Vanotti, M.B., 2009. Prospects for phosphorus recovery from poultrylitter. Bioresource Technology 100 (22), 5461–5465.

Vanotti, Matias, Szogi, Ariel, Pilar Bernal, M., Martinez, José, 2009. Livestock wastetreatment systems of the future: a challenge to environmental quality, foodsafety, and sustainability. Bioresource Technology 100 (22), 5371–5373.

Wang, Zhengjian, Charles, J.B., 2003. Evaluation of a two stage anaerobic digester forthe treatment of mixed abattoir wastes. Process Biochemistry 30 (9), 1267–1273.

Yadvika, Santosh, Sreekrishnan, T.R., Kohli, Sangeeta, Rana, Vineet, 2004.Enhancement of biogas production from solid substrates using differenttechniques – a review. Bioresource Technology 95 (1), 1–10.