Bioresource Technology 154 (2014) 240–247

Contents lists available at ScienceDirect

Bioresource Technology

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

Solid-state anaerobic co-digestion of hay and soybean processing wastefor biogas production

0960-8524/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2013.12.045

⇑ Corresponding author. Tel.: +1 330 263 3855; fax: +1 330 263 3670.E-mail address: li.851@osu.edu (Y. Li).

1 Present address: Department of Environmental Engineering and Earth Sciences,Clemson University, Rich Lab, 342 Computer Court, Anderson, SC 29625, USA.

Jiying Zhu a,b, Yi Zheng a,1, Fuqing Xu a, Yebo Li a,⇑a Department of Food, Agricultural, and Biological Engineering, The Ohio State University/Ohio Agricultural Research and Development Center, 1680 Madison Ave., Wooster,OH 44691-4096, USAb School of Agricultural and Food Engineering, Shandong University of Technology, Zibo, Shandong 255049, China

h i g h l i g h t s

� Solid state anaerobic digestion (AD) of soybean processing waste (SPW) and hay.� Co-digestion of SPW and hay enhanced methane production.� Partial mixing of inoculum with feedstock did not impact cumulative methane yield.� Leachate recirculation accelerated start-up of partially premixed AD reactors.

a r t i c l e i n f o

Article history:Received 11 October 2013Received in revised form 6 December 2013Accepted 11 December 2013Available online 18 December 2013

Keywords:Solid state anaerobic digestionCo-digestionLeachate recirculationMethaneBiogas

a b s t r a c t

Co-digestion of soybean processing waste (SPW) and hay in solid-state anaerobic digestion (SS-AD) forbiogas production was investigated. Effects of the SPW to hay ratio, feedstock to effluent (inoculum) ratio,premixing of effluent with feedstock, and leachate recirculation on biogas production via SS-AD werestudied. The highest methane yield of 258 L/kg VS was obtained with a SPW/hay ratio of 75:25 and feed-stock/effluent (F/E) ratio of 3, which was 148% and 50% higher than that of 100% SPW and 100% hay,respectively. Increasing the F/E ratio from 1 to 5 decreased methane yield, however the highest volumet-ric methane yield (16.2 L/Lreactor) was obtained at an F/E of 3. There was no significant difference in meth-ane yields between premixing 50% and 100% of the effluent. Leachate recirculation significantlyaccelerated the SS-AD start-up process when effluent was not completely premixed.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Anaerobic digestion (AD) has been extensively used to convertorganic waste streams from various sources, such as agricultural,industrial, and municipal solid waste, to biogas. The AD processcan operate in both liquid and solid states in terms of total solid(TS) content. In general, the TS content of liquid AD (L-AD) systemsranges from 0.5 to 15%, while solid-state AD (SS-AD) systems usu-ally operate at TS contents of higher than 15% (Rapport et al.,2008). Comparisons between L-AD and SS-AD indicate that L-ADhas higher reaction rates and shorter retention times, while theadvantages of SS-AD are smaller reactor volume requirements, lessenergy input for heating, no processing energy needed for stirring,minimal material handling, and less total parasitic energy loss

(Guendouz et al., 2008). Problems related to the floating andstratification of fibrous material in L-AD does not occur in SS-AD(Chanakya et al., 1999; Kaparaju and Angelidaki, 2008). In compar-ison to the effluent of L-AD, the effluent of SS-AD (digestate) ismuch easier to handle because of its lower water content, and itcan be used as fertilizer or pelletized for fuels (Li et al., 2011a).Brown et al. (2012) compared SS-AD and L-AD of lignocellulosicfeedstocks for biogas production and found insignificant differ-ences in methane yield. However, they revealed that the volumet-ric methane productivity of SS-AD was 2- to 7-fold greater thanthat of L-AD.

Co-digestion is the simultaneous digestion of a mixture of twoor more substrates and offers many advantages, including ecolog-ical, technological, and economic benefits, compared to digestionof a single substrate (Rughoonundun et al., 2012). The purpose ofco-digestion is usually to balance nutrients (C/N ratio and macro-and micronutrients) (Girault et al., 2012; Liu et al., 2012) and diluteinhibitors/toxic compounds, thus enhancing methane production(Xu and Li, 2012; Luste et al., 2012; Xia et al., 2012). However,

Table 1Characteristics of feedstocks and inoculum.*

J. Zhu et al. / Bioresource Technology 154 (2014) 240–247 241

combining two or more feedstocks requires careful selection to im-prove efficiency of the AD process (Álvarez et al., 2010).

The start-up phase of an AD system is the most critical step inthe process (Pandey et al., 2011), especially for large-scale batchdigesters. The feedstock to inoculum (F/I) ratio, on a volatile solid(VS) basis, is one of the most important factors for the start of abalanced microbial community. Feedstock/effluent (F/E) ratio isused for studies in which the effluent from L-AD was used as boththe inoculum and nitrogen source. However, the effect of F/E ratioon biogas may vary due to differences between AD systems, suchas operating temperature and type of feedstock. In a study on theSS-AD of corn stover, the highest biogas yield was obtained at anF/E ratio of 2.43 under mesophlic condition and 4.58 under ther-mophilic conditions, respectively (Li et al., 2011b). In a differentstudy with an F/E ratio of 2, the co-digestion of 90% yard wasteand 10% food waste performed well but the reactor failed with80% yard waste and 20% food waste (Brown et al., 2012). In general,high F/E ratios can cause overproduction of volatile fatty acids(VFAs) resulting in low pH. The methanogens are subsequentlyinhibited by the low pH, and the AD reactor can fail. On the otherhand, as effluent provides a nitrogen source for SS-AD, low F/E ra-tios might lead to ammonia inhibition, especially when high-ammonia inoculum is used (Li et al., 2011b). As a result, an optimalF/E ratio should be determined for different AD systems.

Soybean processing usually generates two streams of by-prod-ucts: (1) high quality protein, fiber-rich stream; and (2) a low qualitywaste stream. In this research, the second stream, defined as soy-bean processing waste (SPW), was used as a feedstock. SPW usuallyconsists of soybean straw, beans, soybean oil residue, and diatoma-ceous earth used in the oil bleaching process. Although this mixtureof materials has relatively limited applications in industry, it couldbe a good feedstock for biogas production in the SS-AD processdue to its high organic matter content. However, the high proteincontent of SPW results in a low C/N ratio, thus when SPW is digestedalone, ammonia may accumulate creating toxic conditions for theAD microbes, which could lead to low biogas yield (Wang et al.,2013). Hay is a mixture of grass, legumes or other herbaceous plantsthat has been cut, dried, and stored for use as animal fodder and ani-mal bedding. The spent hay is usually spoiled and considered awaste stream. As spent hay has a high C/N ratio and TS content, itcould be co-digested with SPW to improve biogas yield. However,the optimum SPW/hay ratio needs to be determined.

Due to the poor water absorption capability of SPW, it becomesa slurry after being mixed with L-AD effluent at 20% total solid (TS)content, making it difficult to load into full-scale, garage-type solidstate digesters. Increasing the TS content to a stackable level willcause a higher F/E ratio, which might result in acidification andinhibition of methanogens (Park and Li, 2012; Xu and Li, 2012).Due to the high TS content for SS-AD, no agitation is applied duringthe process, so that contact between bacteria and feedstock is poor,resulting in low biogas yield. Therefore, mixing strategies, such aspremixing feedstock with partial inoculum and leachate recycling,may be effective for improving biogas yield.

In order to address the issues discussed above, the objectives ofthis research were to: (1) study the effects of SPW/hay and F/E ra-tios on the performance of SS-AD, and (2) investigate the effects ofpremixing and leachate recirculation on the performance of the SS-AD process.

Parameters Soybean processing waste Hay Effluent

Total solids (%) 49.2 ± 0.6 87.8 ± 0.3 7.6 ± 0.2Volatile solids (%) 35.9 ± 0.5 84.3 ± 0.6 3.8 ± 0.2Total carbon (%) 24.0 ± 2.3 45.9 ± 0.3 2.6 ± 0.0Total nitrogen (%) 1.9 ± 0.3 0.6 ± 0.0 0.4 ± 0.0Carbon to nitrogen ratio 12.5 ± 0.8 76.0 ± 3.5 6.5 ± 0.0pH 8.8 ± 0.1 6.8 ± 0.0 7.6 ± 0.0

* All numbers are wet basis.

2. Methods

2.1. Feedstocks and effluent

SPW was provided by the facility of quasar energy group inZanesville, OH, USA. It mainly consisted of soybean straw, soy-

beans, soybean oil extraction residues, and diatomaceous earthused in oil bleaching process. The SPW was kept in a cooler(4 �C) before use. Hay was obtained from a local farm in Wooster,OH, USA, and ground through a 10-mm sieve with a grinder(Mighty Mac, Mackissic Inc., Parker Ford, PA, USA). The groundhay was stored in air tight containers until use. Effluent from amesophilic liquid anaerobic digester (run by quasar energy groupin Zanesville, OH, USA) that was fed with biosolids was used asinoculum. Characteristics of the feedstocks and effluent are pre-sented in Table 1.

2.2. Co-digestion of SPW and hay for SS-AD

Batch AD with complete premixing was carried out at five SPW/hay ratios (100:0, 75:25, 50:50, 25:75, and 0:100) at an F/E ratio of3 and at five F/E ratios (F/E = 1, 2, 3, 4, and 5) at a SPW/hay ratio of75:25. For evaluation of both SPW/hay and F/E ratios, feedstockand effluent were well mixed and loaded into a 2-L reactor thatwas then sealed with a rubber stopper having a gas outlet con-nected to 5-L gas bags (CEL Scientific Tedlar gas bag, Santa FeSprings, CA, USA) (Fig. 1a). Reactors were incubated in a walk-inthermostat chamber at 37 �C for 42 days. All tests were run withduplicate reactors.

2.3. Pre-mixing of feedstocks and effluent

The effects of three methods for pre-mixing feedstocks andeffluent on SS-AD performance were investigated in 5-L batch reac-tors without leachate recirculation (Fig. 1a). An overall F/E ratio of3 and an SPW/hay ratio of 75:25 were used. Prior to placing mate-rials into 5-L reactors, 100%, 50% or 0% of the required effluent wasmixed thoroughly with all of the feedstock. The TS of the materialwith 50% and 0% effluent premixing was 24% and 54%, respectively(Table 2), which resulted in stackable materials that could beloaded by a tractor with a loader and toe pip bucket into garage-type digesters. After the mixture was loaded into the reactor, theremaining effluent (0%, 50%, or 100%, respectively) was immedi-ately added from the top of the reactor to obtain an overall F/E ratioof 3 and TS of 17.6, which was more favorable for SS-AD (Table 2).Biogas was collected in a 10-L Tedlar gas bag attached to the outletof the reactor. The operating procedure and mixing parameters areshown in Fig. 1b and Table 2, respectively.

2.4. Leachate recirculation

Pre-mixing and loading methods of feedstocks and effluentwere the same as those in Section 2.3. Batch reactors with a 5-Lcapacity and leachate recycling system were used (Fig. 1c). Theleachate recirculation system included a leachate sump at the bot-tom of the reactor, a peristaltic pump (Thermo Fisher Scientific™,Master flex 77200-62, Waltham, MA, USA) and plastic tubing con-necting the sump to the leachate inlet on the top of the reactor. Inorder to prevent the feedstock particles from dropping into thesump, a metal grid combined with a 2-mm hardware cloth was

(a)

(b)

(c)

(1) Loading SPW + Hay

Effluent(100%,50% or 0%)

Reactor

(2) Add the remainingeffluent from the top

Premix

Gas bag

Leachate

Pump

Fig. 1. Experimental set-up and pre-mixing scheme of SPW and hay: batch reactorwithout leachate recirculation (a), partial premixing and loading process diagram(b), and batch reactor with leachate recirculation (c).

242 J. Zhu et al. / Bioresource Technology 154 (2014) 240–247

placed between the reactor and the sump. During the SS-AD pro-cess, leachate dripped to the sump and was recirculated into thereactor from the top every 2 days using the peristaltic pump.Leachate in the reactor sump was pumped into a flask and sampledevery 6 days for analysis, and the remaining leachate in the flaskwas pumped into the reactor from the top. No buffer or alkaliwas added to the leachate to adjust pH.

2.5. Analytical methods

TS and VS contents were analyzed according to the APHA Stan-dard Methods for the Examination of Water and Wastewater(APHA, 2005). Total carbon and nitrogen contents were determinedby an elemental analyzer (Vario Max CNS, Elementar Americas, Mt.Laurel, NJ, USA) and were used to calculate the C/N ratio. Totalammonia nitrogen (TAN) was determined using a colorimetricmethod with an ammonia nitrogen kit (HT832, Hach Company,

Table 2Characteristics of pre-mixed material and overall material.

Premixing method F/E ratio TS (%)

Premixed material Overall material Premixed

Completely premixed 3 3 17.650% effluent premixed 6 3 24.0No premixing � 3 54.0

Düsseldorf, Germany) and a spectrophotometer (Model 3900, HachCompany, Düsseldorf, Germany).

VFAs in the AD leachate were measured using a gas chromatog-raphy (GC) (Shimadzu, GC-2010, Columbia, MD, USA) equippedwith a Stabilwax-DA column (30 m � 0.32 mm � 0.5 lm) and aflame ionization detector (FID). The temperatures for the columnand FID were 150 �C and 250 �C, respectively. Helium gas was usedas the mobile phase at a flow rate of 16.9 mL/min. The sampleswere prepared by diluting 3 g leachate with 3 g deionized waterand the solution was acidified to a pH of around 2.5 using 2 Nhydrochloric acid. The acidified solution was filtered through a0.2 lm nylon syringe filter for GC analysis.

The volume of biogas collected in the Tedlar bags was measuredby a drum-type gas meter (Ritter, TG 5, Bochum, Germany) and thecomposition of biogas (CO2, CH4, N2, and O2) was analyzed by a GC(HP 6890, Agilent Technologies, Wilmington, DE, USA) equippedwith a 30 m � 0.53 mm � 10 mm alumina/KCl deactivation col-umn and a thermal conductivity detector (TCD). Helium gas wasused as a carrier gas at a flow rate of 5.2 mL/min. The temperaturesof the column and detector were set at 40 �C and 200 �C,respectively.

2.6. Data analysis

The modified Gompertz equation (Eq. (1)) was used to calculatethe maximum methane production (Dechrugsa et al., 2013):

Y ðtÞ ¼ Y � exp �expRm � e

Yðk� tÞ þ 1

� �� �ð1Þ

where Y(t) is cumulative methane yield (L/kg VSfeed) at time t(day); e is exp(1) = 2.71828; Rm is the maximum specific methaneproduction rate (L/kgVSfeed/day); Y is methane production poten-tial (L/kgVSfeed), and k is lag phase time (days). The parameters inthis equation were estimated by least square method using SolverFunction in Microsoft� Office Excel 2010.

Statistical significance was determined by using analysis of var-iance (ANOVA) and least significant difference (LSD) (a = 0.05 andpcritical = 0.05) methods. JMP 10.0 software (version 10.0; SAS Insti-tute, Raleigh, NC, USA) was used to perform statistical analysis. Alltreatments were performed in two replicates in this study.

3. Results and discussion

3.1. Effect of SPW/hay ratio on biogas production of SS-AD

As shown in Fig. 2a and c, the SPW/hay ratio had a significanteffect on daily and cumulative methane yields (p < 0.05). The dailymethane yield of the reactors with 100% SPW stayed at a low levelduring the 42-day digestion and the cumulative methane yield wasthe lowest among all the tested SPW/hay ratios (Fig. 2a and c). Forthe cumulative methane yields in the figures, the symbols were theexperimental data and the lines represented the theoretical valuescalculated with the modified Gompertz equation (Eq. (1)). Themethane contents of the co-digested reactors were higher thanthat of the reactors with 100% SPW and 100% hay during the start-up stage, although all of them kept stable at around 65% after start-

C/N ratio

material Overall material Premixed material Overall material

17.6 10.8 10.817.6 12.3 10.817.6 15.1 10.8

J. Zhu et al. / Bioresource Technology 154 (2014) 240–247 243

up (Fig. 2b). Daily methane yield of the reactors with an SPW/hayratio of 75:25 increased to 3.6 L/kg VSfeed/d on the second day andthe methane content reached 68% on day 10 (Fig. 2a and b). Thecumulative methane yield of the reactors was 258 L/kg VS, whichwas 148% and 50% higher than that of 100% SPW and 100% hay,respectively. The cumulative methane yield of the reactors withSPW/hay ratio of 50:50 and 25:75 were similar (p > 0.05), and bothof them were higher than that of the reactors with 100% SPW and100% hay. The daily methane yields of the co-digestion reactorsmaintained above 6 L/kg VSfeed/d till day 36 and multiple peakswere also observed, while the daily methane yields of reactors with100% of SPW or hay dropped to below 3/L kg VSfeed/d after the ma-jor peak on day 18. These results indicated that co-digestion ofSPW and hay can improve the performance of SS-AD.

The TAN and TVFA concentrations in the digestate of the reactorsare shown in Fig. 2d. The TAN of the digestate with 100% and 75%SPW was 4730 mg TAN/kg and 3720 mg TAN/kg, respectively, bothof which were significantly higher (p < 0.05) than the digestate with100% hay. These high TN concentrations are most likely caused bythe high protein content in SPW. Feedstocks with high protein con-tent have been shown to cause accumulation of fatty acids or ammo-nia, which are produced by the degradation of proteins during AD(Cuetos et al., 2010; Palatsi et al., 2011; Xu and Li, 2012). Althoughammonia can provide nitrogen for bacterial growth (Strik et al.,2006) and buffer capacity (Procházka et al., 2012) in the AD process,

0

2

4

6

8

10

0 4 8 12 16 20 24 28 32 36 40

Dai

ly m

etha

ne y

ield

(L/k

gVS fe

ed/d

)

Time (day)

100:075:2550:5025:750:100

SPW:Hay

0

50

100

150

200

250

300

0 10 20 30 40

feed

Cum

lati

ve m

etha

ne y

ield

(L

/kg

VS

)

Time (d)

SPW:hay

100:075:2550:5025:750:100

(a)

(c)

Fig. 2. Effects of SPW/hay ratio on daily methane yield (a); methane content (b); cumul(TVFA), and pH of the digestate (d) (F/E ratio = 3).

high concentrations of ammonia may cause severe inhibition ofmethanogenesis. Rajagopal et al. (2013) reported that TAN concen-trations from 1500 to 7000 mg/L inhibited methanogenesis depend-ing on the nature of the substrates, inoculum, temperature, pH, andacclimation periods. One study on SS-AD reported that at TS of 18%,inhibition to methane production was observed at TAN higher than4300 mg/kg (Wang et al., 2013).

In this study, the reactors with 4730 mg TAN/kg in the digestate(100% SPW) resulted in low methane yield, while the reactors with3720 mg TAN/kg in the digestate (75% SPW) performed well. Theseresults indicated that the excessive ammonia nitrogen caused bydegradation of protein in SPW might have inhibited the methano-genesis in the reactor with 100% SPW and caused low methaneyield. Co-digestion of SPW and hay diluted the toxicity of ammoniaand improved the digestion performance. Although the TVFA con-centration of the digestate in the 100% SPW reactors was signifi-cantly higher than that in the reactors with 75% SPW, the pHvalues of these reactors were almost equal (Fig. 2d), which wasprobably due to a high buffering capacity caused by the highammonia concentration in the 100% SPW reactors.

3.2. Effect of F/E ratio on reactor performance

The reactor with an F/E ratio of 1 had a significantly higher peakvalue (p < 0.05) of daily methane yield and cumulative methane

0

10

20

30

40

50

60

70

80

0 4 8 12 16 20 24 28 32 36 40

Met

hane

con

tent

(%

)

Time (day)

100:075:2550:5025:750:100

SPW:Hay

0

2

4

6

8

10

12

14

16

TAN (g/kg) TVFA (g/kg) pH

TA

N a

nd T

VF

A (

g/kg

)

100:075:250:100

SPW:Hay

(b)

(d)

ative methane yield (c); and total ammonia nitrogen (TAN), total volatile fatty acid

(a) (b)

(c) (d)

0

5

10

15

20

25

30

0 4 8 12 16 20 24 28 32 36 40

Dai

ly m

etha

ne y

ield

(L/k

gVS f

eed/d

)

Time (day)

F/E=1F/E=2F/E=3F/E=4F/E=5

0

10

20

30

40

50

60

70

80

90

0 4 8 12 16 20 24 28 32 36 40

Met

hane

cont

ent (

%)

Time (day)

F/E=1F/E=2F/E=3F/E=4F/E=5

50

100

150

200

250

300

350

400

450

1 2 3 4 5

Cum

ulat

ive

met

hane

yie

ld(L

/kg

VS

feed

)

F/E ratio

0

2

4

6

8

10

12

14

16

18

1 2 3 4 5

Vol

umet

ric

met

hane

pro

duct

ivity

(L/L

wor

k)

F/E ratio

Fig. 3. Effects of F/E ratio on daily methane yield (a), methane content (b), cumulative methane yield (c), and volumetric methane productivity (d) (SPW/hay ratio = 75:25).

244 J. Zhu et al. / Bioresource Technology 154 (2014) 240–247

yield (400 L/kgVS) compared to the other reactors (Fig. 3a and c).The methane contents of the reactors with F/E of 1, 2 and 3 wereclose to each other with a steady methane content at around 70%after startup (Fig. 3b). Xu and Li (2012) investigated the effect ofF/E ratio on solid state co-digestion of expired dog food and cornstover and also observed that lower F/E ratios achieved higher dailymethane yield peak values. However, low F/E ratios resulted in alow organic loading rate, low TS content of the digestion mediumand increased working volume. In this study, when the F/E ratio in-creased from 1 to 5, the initial TS of the digestion medium in-creased from 11.5% to 22.1%. The volumetric methaneproductivity of the reactor with an F/E ratio of 3 was 16.2 L/Lreactor,which was 65% and 28% higher than that of F/E of 1 and F/E of 2,respectively (Fig. 3d). Higher volumetric productivity of SS-AD isa major advantage over L-AD due to the smaller reactor volume re-quired for the same solids loading (Brown et al., 2012). However,excessive organic loading rate at F/E ratios of 4 and 5 resulted inlower methane contents and yields (Fig. 3), causing failure of thedigestion process. Overloading of organic material can cause accu-mulation of VFAs which might lead to inhibition of methanogensand failure of the digester (Abbassi-Guendouz et al., 2012; Parkand Li, 2012).

3.3. Effect of premixing and leachate recirculation on SS-ADperformance

Without leachate recirculation, start-up of the reactors with 0%and 50% effluent pre-mixed with the feedstocks was significantlyslower compared to the one with 100% effluent pre-mixed. The dai-ly methane yield of the reactors with 100% effluent pre-mixedreached 5.0 L/kg VSfeed/d on day 2 while reactors with 50% and0% effluent pre-mixed achieved this methane yield on day 14 andday 20 (Fig. 4a), respectively. Moreover, the methane content ofthe 100% effluent pre-mixed reactors increased to the stable levelof �60% on day 8, while that of the partial premixed reactorswas still kept at around 40% (Fig. 4c). This result can be explainedby the higher regional TS content and F/E ratio in the partially pre-mixed reactors than that of the 100% premixed reactors. The over-all TS and F/E ratio for all the reactors was 18% and 3, respectively,but as the effluent at the top layer of the reactors dripped veryslowly, the actual initial TS (except the top layer) and F/E ratio inthe 50% effluent premixed reactor was 24% and 6, respectively,while all the inoculum was above the substrate in reactors with0% effluent premixing. Solid content has an important effect onthe performance of SS-AD. Increasing TS from 20% to 30% has been

(a) (b)

(c) (d)

(e)

0

2

4

6

8

10

12

14

16

0 4 8 12 16 20 24 28 32 36 40

Dai

ly m

etha

ne y

ield

(L/k

gVS f

eed/d

)

Time (day)

100% premixed50% premixed0% premixed

No leachate recirculation0

2

4

6

8

10

12

14

16

0 4 8 12 16 20 24 28 32 36 40

Dai

ly m

etha

ne y

ield

(L/k

gVS f

eed/d

)

Time (day)

100% premixed50% premixed0% premixed

With leachate recirculation

0

10

20

30

40

50

60

70

80

0 4 8 12 16 20 24 28 32 36 40

Met

hane

cont

ent (

%)

Time (day)

100% premixed50% premixed0% premixed

No leachate recirculation0

10

20

30

40

50

60

70

80

0 4 8 12 16 20 24 28 32 36 40

Met

hane

cont

ent (

%)

Time (day)

100%premixed50% premixed0% premixed

With leachate recirculation

0

50

100

150

200

250

300

350

400

No leachate recirculation With leachate recirculation

Cum

ulat

ive

met

hane

yie

ld (L

/kg

VS fe

ed)

100% premixed50% premixed0% premixed

Fig. 4. Effects of premixing on daily methane yields with (a) and without (b) leachate recirculation, on methane content with (c) and without (d) leachate recirculation, andon accumulative methane yield (e) (SPW/hay ratio = 75:25, F/E ratio = 3).

J. Zhu et al. / Bioresource Technology 154 (2014) 240–247 245

(a) (b)

4

5

6

7

8

9

6 12 24 30 36 42

pH

Time (d)

100% premixed

50% premixed

0% premixed

0

20

40

60

80

100

120

6 12 24 30 36 42

TV

FA (g

/kg)

Time (d)

100% premixed

50% premixed

0% premixed

Fig. 5. Variation of pH (a) and total VFAs (b) of the leachate during SS-AD. (SPW/hay ratio = 75:25, F/E ratio = 3). �There was no leachate generated before day 10 for reactorswithout premixing.

246 J. Zhu et al. / Bioresource Technology 154 (2014) 240–247

shown to result in slower start-up of the SS-AD process, lower sub-strate degradation, and lower methane yield (Fernández et al.,2008; Forster-Carneiro et al., 2008). Le Hyaric et al. (2011) foundthat the specific methanogenic activity on propionate increasedlinearly by a factor of 3.5 when the solid content decreased from35% to 18%. Abbassi-Guendouz et al. (2012) demonstrated thathigh TS content (30% and 35%) resulted in high VFA accumulationdue to the mass transfer limitation and caused serious inhibition ofthe AD process. Lower concentrations of methanogens in the 50%and 0% effluent premixed materials could be another reason forthe slow start-up of the digestion. The low concentrations of meth-anogens caused accumulation of VFAs, which consequently inhib-ited the growth and activity of methanogens and slowed downthe start-up process. El-Mashad et al. (2006) indicated that feed-stocks mixed well with inoculum significantly improved AD sys-tem performance. In this study, the cumulative methane yield ofthe reactors with 50% effluent premixing was not significantly dif-ferent from that of the reactors with 100% effluent premixing(p > 0.05), while the reactors without premixing had significantlylower (p < 0.05) cumulative methane yield compared to the others(Fig. 4e). The effluent added from the top of the reactor permeatedthrough the feedstock gradually, diluted the VFAs, and increasedthe buffering capacity of the premixed material. Meanwhile, themicrobes (especially methanogens) in the effluent were also trans-ferred to the premixed material. Therefore, the reactors with 50%and 0% effluent premixing started slowly, but performed well afterstartup (Fig. 4a).

Fig. 4b shows that leachate recirculation significantly improvedthe startup performance of the reactors with 50% and 0% effluentpremixing. Daily methane yield peaks for the reactors with 50%and 0% effluent premixing shifted from day 24 and day 30, respec-tively, without leachate recirculation, to day 12 and day 16 (Fig. 4aand b), respectively, with leachate recirculation. Compared to thereactors without leachate recirculation, methane content peaks ofthe 100% effluent premixing reactors was delayed, however thatof the reactors with 50% effluent premixing increased more quicklyduring the startup stage (Fig. 4c and d). This result is probably be-cause leachate recirculation can promote mass transfer of VFAsfrom acidogenic to methanogenic pockets (Veeken and Hamelers,2000) and improve the mixing of the feedstock and inoculum(El-Mashad et al., 2006). However, leachate recirculation of thereactor with 100% effluent premixing resulted in a lower dailymethane yield in the first 10 days and a significantly higher dailymethane yield peak of 12.3 L/kg/VSfeed/d compared to the reactorwithout leachate recirculation. When the methanogenic activity

is low, leachate recirculation during startup of a batch AD processcould enhance acidogenesis resulting in over production of VFAsand inhibition of methanogenesis (Kusch et al., 2012; Veeken andHamelers, 2000). On day 6, the pH of the leachate collected fromthe reactors premixed with 100% effluent was 6.8 (Fig. 5a) whichis in the optimal pH range for methanogenesis. The increase ofthe daily methane yield of these reactors after day 6 might bedue to the abundance of VFAs in the leachate and the optimumpH condition for methanogenesis.

Leachate recirculation for the reactors without premixingstarted on day 10 as there was no leachate generated for thesereactors in the first 10-days due to the high TS content of the feed-stocks in the bottom half of the reactor and the slow effluent drip-ping. During the startup stage, the pH value of the leachate fromthe reactors with 0% effluent premixing was substantially lowerand the TVFA concentration was higher compared to the otherreactors with 100% and 50% effluent premixing as only limitedamount of inoculum existed in the lower part of the reactor (highlocal F/E ratio) at the start up stage (Fig. 5a and b). With leachaterecirculation, VFAs generated in the lower part of the reactor wererecirculated to the top of the reactors where there was abundantinoculum. This probably caused the sharp increase of daily meth-ane yield of these reactors after leachate recirculation (Fig. 4band c).

4. Conclusion

Co-digestion of SPW and hay significantly improved the meth-ane yield of SS-AD. The highest volumetric methane productivitywas achieved at an F/E ratio of 3 although reactors with an F/E ratioof 1 had the highest methane yield. Partially premixing 50% of theeffluent with the feedstocks and adding the remaining effluentfrom the top of the reactor resulted in a methane yield almostequal to that of 100% premixing. Leachate recirculation acceleratedthe start-up of SS-AD reactors with partial or no premixing, but didnot improve the 40-day cumulative methane production of thesereactors.

Acknowledgements

This project was funded by USDA NIFA Biomass Research andDevelopment Initiative Program (Award No. 2012-10008-20302).The authors wish to thank Mrs. Mary Wicks (Department of Food,Agricultural and Biological Engineering, OSU) for critical review.

J. Zhu et al. / Bioresource Technology 154 (2014) 240–247 247

References

Abbassi-Guendouz, A., Brockmann, D., Trably, E., Dumas, C., Delgenès, J.P., Steyer,J.P., Escudié, R., 2012. Total solids content drives high solid anaerobic digestionvia mass transfer limitation. Bioresour. Technol. 111, 55–61.

Álvarez, J.A., Otero, L., Lema, J.M., 2010. A methodology for optimizing feedcomposition for anaerobic co-digestion of agro-industrial wastes. Bioresour.Technol. 101, 1153–1158.

Brown, D., Shi, J., Li, Y., 2012. Comparison of solid-state to liquid anaerobic digestionof lignocellulosic feedstocks for biogas production. Bioresour. Technol. 124,379–386.

Chanakya, H.N., Srikumar, K.G., Anand, V., Modak, J., Jagadish, K.S., 1999.Fermentation properties of agro-residues, leaf biomass and urban marketgarbage in a solid phase biogas fermenter. Biomass Bioenerg. 16, 417–429.

Cuetos, M.J., Gómez, X., Otero, M., Morán, A., 2010. Anaerobic digestion and co-digestion of slaughterhouse waste (SHW): influence of heat and pressure pre-treatment in biogas yield. Waste Manage. 30, 1780–1789.

Dechrugsa, S., Kantachote, D., Chaiprapat, S., 2013. Effects of inoculum to substrateratio, substrate mix ratio and inoculum source on batch co-digestion of grassand pig manure. Bioresour. Technol. 146, 101–108.

El-Mashad, H.M., van Loon, W.K.P., Zeeman, G., Bot, G.P.A., Lettinga, G., 2006. Effectof inoculum addition modes and leachate recirculation on anaerobic digestionof solid cattle manure in an accumulation system. Biosyst. Eng. 95, 245–254.

Fernández, J., Pérez, M., Romero, L.I., 2008. Effect of substrate concentration on drymesophilic anaerobic digestion of organic fraction of municipal solid waste(OFMSW). Bioresour. Technol. 99, 6075–6080.

Forster-Carneiro, T., Perez, M., Romero, L., 2008. Influence of total solid andinoculum contents on performance of anaerobic reactors treating food waste.Bioresour. Technol. 99, 6994–7002.

Girault, R., Bridoux, G., Nauleau, F., Poullain, C., Buffet, J., Peu, P., Sadowski, A.G.,Beline, F., 2012. Anaerobic co-digestion of waste activated sludge and greasysludge from flotation process: batch versus CSTR experiments to investigateoptimal design. Bioresour. Technol. 105, 1–8.

Guendouz, J., Buffière, P., Cacho, J., Carrère, M., Delgenes, J.P., 2008. High-solidsanaerobic digestion: comparison of three pilot scales. Water Sci. Technol. 58,1757–1763.

Kaparaju, P., Angelidaki, I., 2008. Effect of temperature and active biogas process onpassive separation of digested cow manure. Bioresour. Technol. 99, 1345–1352.

Kusch, S., Oechsner, H., Jungbluth, T., 2012. Effect of various leachate recirculationstrategies on batch anaerobic digestion of solid substrates. Int. J. Environ. WasteManage. 9, 69–88.

Le Hyaric, R., Chardin, C., Benbelkacem, H., Bollon, J., Bayard, R., Escudié, R., Buffière,P., 2011. Influence of substrate concentration and moisture content on thespecific methanogenic activity of dry mesophilic municipal solid wastedigestate spiked with propionate. Bioresour. Technol. 102, 822–827.

Li, Y., Park, S.Y., Zhu, J., 2011a. Solid-state anaerobic digestion for methaneproduction from organic waste. Renew. Sustain. Energ. Rev. 15, 821–826.

Li, Y., Zhu, J., Wan, C., Park, S.Y., 2011b. Solid-state anaerobic digestion of corn stoverfor biogas production. ASABE Trans. 54, 1415–1421.

Liu, X., Gao, X., Wang, W., Zheng, L., Zhou, Y., Sun, Y., 2012. Pilot-scale anaerobiccodigestion of municipal biomass waste: focusing on biogas production andGHG reduction. Renew. Energ. 44, 463–468.

Luste, S., Heinonen-Tanski, H., Luostarinen, S., 2012. Co-digestion of dairy cattleslurry and industrial meat-processing by-products – effect of ultrasound andhygienization pre-treatments. Bioresour. Technol. 104, 195–201.

Palatsi, J., Viñas, M., Guivernau, M., Fernandez, B., Flotats, X., 2011. Anaerobicdigestion of slaughterhouse waste: main process limitations and microbialcommunity interactions. Bioresour. Technol. 102, 2219–2227.

Pandey, P.K., Ndegwa, P.M., Soupir, M.L., Alldredge, J.R., Pitts, M.J., 2011. Efficacies ofinocula on the startup of anaerobic reactors treating dairy manure under stirredand unstirred conditions. Biomass Bioenerg. 35, 2705–2720.

Park, S., Li, Y., 2012. Evaluation of methane production and macronutrientdegradation in the anaerobic co-digestion of algae biomass residue and lipidwaste. Bioresour. Technol. 111, 42–48.

Procházka, J., Dolejš, P., Máca, J., Dohányos, M., 2012. Stability and inhibition ofanaerobic processes caused by insufficiency or excess of ammonia nitrogen.Appl. Microbiol. Biotechnol. 93, 439–447.

Rajagopal, R., Massé, D.I., Singh, G., 2013. A critical review on inhibition of anaerobicdigestion process by excess ammonia. Bioresour. Technol. 143, 632–641.

Rapport, J., Zhang, R., Jenkins, B.M., and Williams, R.B., 2008. Current anaerobicdigestion technologies used for treatment of municipal organic solid waste.Sacramento, Cal.: California Environmental Protection Agency. Available at:<www.calrecycle.ca.gov/Publications/Organics/2008011.pdf>.

Rughoonundun, H., Mohee, R., Holtzapple, M., 2012. Influence of carbon-to-nitrogenratio on the mixed-acid fermentation of wastewater sludge and pretreatedbagasse. Bioresour. Technol. 112, 91–97.

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

Veeken, H.M., Hamelers, B.V.M., 2000. Effect of subtract-seed mixing and leachaterecirculation on solid state digestion of biowaste. Water Sci. Technol. 41, 255–262.

Wang, Z., Xu, F., Li, Y., 2013. Effects of total ammonia nitrogen concentration onsolid-state anaerobic digestion of corn stover. Bioresour. Technol. 144, 281–287.

Xia, Y., Daniel, I., McAllister, T.A., Kong, Y., Seviour, R., Beaulieu, C., 2012. Identityand diversity of archaeal communities during anaerobic co-digestion of chickenfeathers and other animal wastes. Bioresour. Technol. 110, 111–119.

Xu, F., Li, Y., 2012. Solid-state co-digestion of expired dog food and corn stover formethane production. Bioresour. Technol. 118, 219–226.