This article was downloaded by: [University of Idaho], [Gopi Kafle]On: 06 June 2014, At: 08:38Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Environmental TechnologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tent20

Anaerobic digestion of Chinese cabbage waste silagewith swine manure for biogas production: batch andcontinuous studyGopi Krishna Kaflea, Sujala Bhattaraib, Sang Hun Kimc & Lide Chena

a Department of Biological and Agricultural Engineering, University of Idaho, ID, USAb Department of Biological Systems Engineering, Washington State University, Pullman, WA99164, USAc Department of Biosystems Engineering, Kangwon National University, Chuncheon,Kangwon-do, Republic of KoreaAccepted author version posted online: 12 May 2014.Published online: 04 Jun 2014.

To cite this article: Gopi Krishna Kafle, Sujala Bhattarai, Sang Hun Kim & Lide Chen (2014): Anaerobic digestion of Chinesecabbage waste silage with swine manure for biogas production: batch and continuous study, Environmental Technology, DOI:10.1080/09593330.2014.919033

To link to this article: http://dx.doi.org/10.1080/09593330.2014.919033

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Environmental Technology, 2014http://dx.doi.org/10.1080/09593330.2014.919033

Anaerobic digestion of Chinese cabbage waste silage with swine manure for biogas production:batch and continuous study

Gopi Krishna Kaflea∗, Sujala Bhattaraib, Sang Hun Kimc∗ and Lide Chena

aDepartment of Biological and Agricultural Engineering, University of Idaho, ID, USA; bDepartment of Biological SystemsEngineering, Washington State University, Pullman, WA 99164, USA; cDepartment of Biosystems Engineering, Kangwon National

University, Chuncheon, Kangwon-do, Republic of Korea

(Received 21 October 2013; final version received 31 March 2014 )

The aim of this study was to investigate the potential for anaerobic co-digestion of Chinese cabbage waste silage (CCWS)with swine manure (SM). Batch and continuous experiments were carried out under mesophilic anaerobic conditions (36–38◦C). The batch test evaluated the effect of CCWS co-digestion with SM (SM: CCWS = 100:0; 25:75; 33:67; 0:100, %volatile solids (VS) basis). The continuous test evaluated the performance of a single stage completely stirred tank reactorwith SM alone and with a mixture of SM and CCWS. Batch test results showed no significant difference in biogas yield up to25–33% of CCWS; however, biogas yield was significantly decreased when CCWS contents in feed increased to 67% and100%. When testing continuous digestion, the biogas yield at organic loading rate (OLR) of 2.0 g VSL−1 d−1 increased by17% with a mixture of SM and CCWS (SM:CCWS = 75:25) (423 mL g−1 VS) than with SM alone (361 mL g−1 VS). Thecontinuous anaerobic digestion process (biogas production, pH, total volatile fatty acids (TVFA) and TVFA/total alkalinityratios) was stable when co-digesting SM and CCWS (75:25) at OLR of 2.0 g VSL−1 d−1 and hydraulic retention time of 20days under mesophilic conditions.

Keywords: anaerobic digestion; co-digestion; Chinese cabbage waste silage; swine manure; batch and continuous study

Nomenclature% percentageC/N carbon to nitrogen ratioF/I feed to inoculum ratioG gramHRT hydraulic retention time (days)Kimchi fermented cabbageL litremL milli litreOLR organic loading rate (g VSL−1 d−1)

STP standard temperature (0◦C) and pressure (1 atm.)TA total alkalinity (mg L−1 CaCO3)

TAN total ammonia nitrogen (mg L−1)

TCOD total chemical oxygen demand (mg L−1)

TS total solids (%)TVFA total volatile fatty acids (mg L−1 acetate)VS volatile solids (%)w.b. weight basisw/w weight per weight

1. IntroductionThe total annual production of Chinese cabbage is approxi-mately three million tonnes in Korea, and up to 30% of thetotal production is discarded as waste.[1] Cabbage waste

∗Corresponding authors. Emails: gopikafle@yahoo.com, gopik@uidaho.edu; shkim@kangwon.ac.kr

is produced during harvest, transport and at the wholesalemarkets. A large amount of cabbage waste is also generatedfrom kimchi factories during the trimming process. Anaero-bic digestion could be a good approach for Chinese cabbagewaste (CCW) utilization because it can not only produce abiofuel but also a mineralized solid residue that can be usedas a bio-fertilizer with a high NPK concentration.[2] CCWis seasonal and it may accumulate in quantities larger thanneeded for immediate use. Ensiling has been envisionedfor storage before biogas production.[3] Herrmann et al.[3] reported that there will be some storage losses duringthe ensiling process but this loss can be compensated by theincrease in the methane yield per kg of organic dry matteradded. Pakarinen et al. [4] reported an increase in the biogasyield from ensiled substrates compared to fresh substrates.Similarly, Kafle and Kim [5] reported increases in biogasproduction by 2.6–15.2% after ensiling.

Although there are many successful anaerobic digesterscurrently treating swine manure (SM), the high ammoniacontent in this type of manure is a major limitation thathas plagued digesters for many years.[6–8] Similarly, cab-bage waste can be used as sole feed stock in anaerobicdigestion processes but its tendency for rapid acidifica-tion (VFA build-up) is the major limitation which can stallmethanogenesis.[9]

© 2014 Taylor & Francis

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

2 G.K. Kafle et al.

Table 1. Characteristics of materials used in batch and continuous tests.

Particulars Batch test Continuous test

SM CCWS Inoculum CCWS SM

TS (%) 5.3 13.5 3.3 13.0–15.0 9.2–7.4VS (%) 3.8 8.3 1.8 8.5–9.9 6.9–5.3VS/TS ratio 0.72 0.61 0.55 0.65–0.66 0.72–0.75TVFA (mg L−1) 14,407 14,789 1758 9000–12,100 21,372–29,041TA (mg L−1) 10,478 4537 18,719 5040–6200 11,540–10,070TVFA/TA ratio 1.38 3.26 0.09 1.79–1.95 1.85–2.88TAN (mg L−1) ND ND ND 2840–3000 5720–5900pH 7.72 5.58 8.22 5.5–5.8 7.3–7.4TCOD (mg L−1) ND ND ND 120,000–134,400 100,000–110,000

CCWS, Chinese cabbage waste silage; SM, swine manure; ND, not determined.

Anaerobic digestion tends to fail without the addition ofexternal nutrients and buffering agents [10]; therefore, co-digestion with substrates having a high buffering capacity(alkalinity) such as manure can be a good alternative foreffective treatment of highly biodegradable materials (likeCCW). The advantages of anaerobic co-digestion processoften lie in balancing the carbon/nitrogen (C/N) ratio inthe mixture, macro and micronutrients, pH, inhibitors/toxiccompounds as well as dry matter.[11] Thus, inputs of readilybiodegradable organic matter into animal manure digestercould significantly increase biogas production.[12,13] Soutilizing the cabbage waste with SM for biogas produc-tion can be one of the appropriate solutions for its effectivetreatment and energy recovery.

In co-digestion, the digester performance is influencedby the mixing ratio or influent composition. Kaparaju andRintala [7] suggested that 15–20% (on a volatile solid (VS)basis) potato waste can be included when treating SM forsuccessful digester operation. Similarly, Li et al. [13] rec-ommended 75% (VS basis) of kitchen waste and Jagadabhiet al. [14] recommended 30% (VS basis) of grass silageas optimal for co-digestion with cow manure. The abovestudies showed that, depending on the characteristics of thesubstrates used, the optimal mixing ratio will be differentfor different waste products being co-digested. In the co-digestion of CCWS and SM, specific study is needed forthe mixing ratio of CCWS to SM; however, this has notbeen investigated yet.

To examine the effects of co-digestion of CCWS withSM, batch tests were first conducted using different mixingratios of SM and CCWS. Based on the batch test results,the optimum mixing ratio for co-digestion of CCWS withSM was chosen and its performance under continuous feedconditions was evaluated using single stage completelystirred tank reactor (CSTR). Thus objectives of this studywere: (1) to define the optimum mixing ratio of SM andCCWS for anaerobic co-digestion; and (2) to evaluate theperformance of a single stage continuous digester with amixing ratio of SM and CCWS recommended from thebatch test.

2. Materials and methods2.1. Substrates and inoculumSM was obtained from swine farm and stored at 4◦C. Thescrew pressed CCW was obtained from kimchi factory. Thesilage was prepared by fermenting the CCW for 45 days inanaerobic bags at mesophilic conditions. The mesophilicanaerobic digested sludge (inoculum) was obtained fromlab scale anaerobic digester treating SM under mesophilicconditions. The characteristics of feed stock and inoculumused for batch and continuous tests are shown in Table 1.

2.2. Batch test set-up and designBatch anaerobic digestion tests were carried out in 2.3 Lglass bottles (liquid volume 1.5 L) (Figure 1). The batchtest was conducted at different mixing ratios of SM andCCWS keeping the constant F/I ratio of 1.6. The F/I ratiowas fixed at 1.6 based on preliminary test performed withSM only at F/I ratios of 0.8, 1.6 and 3.2. In the preliminarytest with SM only the biogas yield (mL g−1VS) decreasedwith increase in F/I ratio. Biogas yield decreased by 9.0%at F/I = 1.6 and by 26.0% at F/I = 3.2 compared to biogasat F/I = 0.8(unpublished data). Thus F/I = 1.6 was chosenfor co-digestion test in this study. Liu et al. [15] conductedbatch anaerobic digestion tests at F/I ratios of 1.6–5.0 andfound the highest methane yields for food waste, greenwaste and their mixture (1:1) at an F/I ratio of 1.6. TheF/I ratio was calculated based on the initial VS of substrateand inoculums.[16] The experimental designs are shown inTable 2.

After adding the required amounts of inoculum andsubstrate, each digester was filled with tap water to reachthe designated volume (1.5 L). The batch digesters werechecked for any leakage and flushed with 100% purenitrogen for approximately 2–3 min to ensure anaerobicconditions. The anaerobic digesters were maintained at38◦C in a temperature-controlled incubator. Each digesterwas connected to a 2 L Tedlar gas sampling bag throughhose and a 2-way valve. The daily biogas production was

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

Environmental Technology 3

Figure 1. Schematic diagram and picture of experimental set-up: (a) batch test set-up; (b) continuous test set-up.

determined by the volume of biogas produced. The gas vol-ume was measured using a syringe by displacement method.All the experiments were carried out in duplicate and resultswere expressed as mean. Each digester was mixed manu-ally for 30 s once a day. Assays with inoculums alone werealso used as controls. Biogas and methane produced fromthe inoculums (blanks) were subtracted from the sampleassays. The results thus represent only the gas productionfrom the substrate and not from the inoculum.[5,17–21]

2.3. Continuous test set-up and designA single-stage continuous process was performed in a 5.5 LCSTR with 4.0 L of working volume (Figure 1). The reac-tor was installed inside a temperature-controlled chamber(36–38◦C) and fed once a day using a peristaltic pump.An equivalent volume of digester content was dischargedprior to feeding. The biogas was collected in a gas collectorby water displacement method.[22] The reactor was stirred(7 min for each hour) by circulating the produced biogasusing the peristaltic pump.

The reactor was inoculated on day 0 with 4.0 L ofdigested sludge (inoculum), and anaerobic conditions werecreated by flushing the head space with nitrogen gas. A con-tinuous anaerobic digestion test was performed with SMalone and with a mixture of SM and CCWS at hydraulic

retention time (HRT) of 20–30 days. The experimentaldesign for the continuous test is shown in Table 2.

2.4. Analytical methodsThe total solids (TS) and VS in the well-mixed samples weredetermined in triplicate according to standard methods.[23]The closed reflux titration method was used for total chem-ical oxygen demand (TCOD) analysis, and the pH valuewas determined using a pH metre (YK-2001 PH, Taiwan).The total ammonia nitrogen (TAN) was measured using theNessler method and determined using a spectrophotome-ter (DR 2500, Hach, USA). The total volatile fatty acids(TVFA), total alkalinity (TA) and TVFA/TA ratio weredetermined using the Nordmann-titration method.[24]

Methane concentration (CH4, %) and carbon dioxideconcentration (CO2, %) in the biogas were analysed usinga gas analyser (BioGas Check – Geotechnical Instruments(UK), Ltd.).[19,25] The measured wet biogas and methanevolumes were adjusted to the volumes at standard tempera-ture (0◦C) and pressure (1 atm).[26] The weighted averagemethane contents over the digestion period were calculatedas described by Kafle and Kim [27] and Kafle and Kim.[16]

Biogas and methane yields were calculated by dividingthe cumulative biogas and methane yields by the amount ofVS initially added to each digester. The VS removal of feed

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

4 G.K. Kafle et al.

Table 2. Experimental design for batch and continuous tests.

Test OLR (g VS L−1 or Feed compositionTest duration g VS L−1d−1) (%, VS)

Batch test 40 days 16.0a SM:CCWS = 100:0, 75:25, 67:33, 33:67, 0:100Continuous test 98 days Period I (days 1–45) 1.0, 2.0b SM:CCWS = 100:0

Period II (days 46–98) 1.8, 2.0b SM:CCWS = 75:25

CCWS, Chinese cabbage waste silage; SM, swine manure.ag VS L−1.bg VS L−1 d−1.

was calculated using Equation (1).[19]

VS removal of feed(%) = (F + I ) × X − I × YF

, (1)

where F is the total VS feed added to reactor (g); I is the totalVS inoculum added to reactor (g); X is the calculated VSremoval of mixture of feed and inoculum based on totalinitial and final gram VS present in the testing reactors(%); Y is the calculated VS removal of inoculum in blankreactors (%).

2.5. Statistical analysisThe significance of differences in average biogas yields,methane yields, methane content and VS removal weredetermined by single factor ANOVA (Analysis of Vari-ance) in Excel software 2007. If the calculated F value washigher than the tabulated F value, Least Significant Differ-ence (LSD) was calculated to judge whether two or moreaverages were significantly different or not. LSD was cal-culated at α = 0.05 (LSD0.05) and at α = 0.01 (LSD0.01) asfollows.[16,28]

LSDα = tα

√2s2

r, (2)

where tα is the abulated value chosen for the degree of free-dom for error and level of significance (α) desired; s2 is themean square for error (MSE); r is the number of replicationson which the means to be separated are based.

3. Results and discussion3.1. Characterization of the SM and CCWSThere were significant differences in the composition ofdifferent biomass wastes used (Table 1). The CCWShad a lower TAN concentration in the range of 2840–3000 mg L−1, as compared to SM, which presented TANconcentration in the range of 5720–7160 mg L−1. TheCCWS was significantly acidic (pH = 5.50–5.80) com-pared to SM (pH = 7.34–7.44). There was noticed a greatchange in the chemical properties of CCW when convertedto CCWS. The reason for changes in properties of theCCW was due to the fermentation process during ensil-ing. The TVFA increased from the range of 1200–1400 to

9000–14,789 mg L−1 and TA increased from the range of500–620 to 2640–6200 mg L−1 when CCW was convertedto CCWS. Similarly, TAN concentration increased from therange of 720–1200 to 2840–3000 mg L−1. The increase inthe TAN concentrations was due to protein degradation dur-ing the ensiling process.[3] However, the pH dropped fromthe range of 6.5–7.2 to 5.5–5.8 (Table 1).

To determine the biogas and methane potential ofCCWS and SM, additional batch test was also performedat OLR of 5 g VS L−1 and F/I ratio of 1.0. The F/I ratiosuggested for the biogas potential test was 0.5–1.0.[14,26]The total biogas and methane potential of SM was found tobe 667 mL g−1 VS added and 450 mL g−1 VS added, respec-tively. Similarly, the biogas and methane potential of CCWSwas found to be 609 mL g−1 VS added and 301 mL g−1 VSadded, respectively. Shin et al. [29] reported methane poten-tial value for SM (492 mL CH4 g−1 VS added) closer toour results. In contrast, Zhang et al., [30], Li et al. [31]and Browne et al. [32] reported methane potential valuefor SM to be only 242, 207 and 99 mL CH4 g−1 VS added,respectively.

3.2. Batch digester performance at different mixingratios of SM and CCWS

The cumulative biogas yield (mL/g VS added) and methanecontent in the biogas produced from different mixingratios of SM and CCWS are shown in Figure 2. Bio-gas production started immediately on the first day ofdigestion in all of the digesters. More than 90% of bio-gas was produced within 28 days in all the digesters.The biogas yield was calculated to be 425, 408, 410,342 and 196 mL g−1 VS added to the SM:CCWS(100:0),SM:CCWS(75:25), SM:CCWS(67:33), SM:CCWS(33:67)and SM:CCWS(0:100), respectively (Table 3). Therewas found no significant difference in biogas yield(p < 0.05) (Table 4, LSD0.05 = 70 mL g−1 VS, LSD0.01 =100 mL g−1 VS) up to 33% of CCWS, but when the CCWScontents in the feed increased to 67% and 100%, biogasproduction significantly decreased.

The weighted average methane contents were cal-culated to be 80.4%, 74.5%,72.7%, 63.5% and 50.9%,respectively, for SM:CCWS(100:0), SM:CCWS(75:25),SM:CCWS(67:33), SM:CCWS(33:67) and SM:CCWS(0:

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

Environmental Technology 5

Figure 2. (a) Cumulative biogas yield; (b) methane content in biogas produced from different mixing ratios of CCWS and SM at OLRof 16 g VS/L and F/I ratio of 1.6.

Table 3. Calculated gas yield, methane content and VS removal for different mixing ratios of SM andCCWS (batch test).

Feed ratio (%, VS)Substrate Biogas yield Methane yield Methane VS removal

SM CCWS (g VS L−1) (mL g−1 VS) (mL g−1 VS) content (%) (%)

100 0 16 425 (35) 341 (10) 80.4 (4.4) 53.6 (0.0)75 25 16 408 (14) 304 (21) 74.5 (2.6) 67.5 (3.8)67 33 16 410 (24) 297 (17) 72.7 (8.4) 67.0 (3.1)33 67 16 342 (28) 217 (11) 63.5 (1.9) 76.3 (2.3)0 100 16 196 (44) 99 (16) 50.9 (3.4) 91.3 (1.6)

CCWS, Chinese cabbage waste silage; SM, swine manure. Values are expressed as mean (standarddeviation, n = 2).

100) (Table 3). No significant difference was observedin methane content up to 33% of CCWS, but methanecontent significantly decreased when CCWS contentsof the feed increased to 67% and 100% (Table 4,LSD0.05 = 10.7%, LSD0.01 = 15.4%). The methane yieldswere calculated to be 341, 304, 297, 217 and99 mL g−1 VS for SM:CCWS(100:0), SM:CCWS(75:25),SM:CCWS(67:33), SM:CCWS(33:67) and SM:CCWS(0:100), respectively. Similar to biogas yield, there was found

no significance difference in methane yield for CCWScontents up to 33%, but methane yields significantlydecreased (p < 0.01) when CCWS contents increasedto 67% and 100% (Table 4, LSD0.05 = 35 mL g−1 VS,LSD0.01 = 50 mL g−1 VS). The reasons for decrease inmethane yield with an increase in CCWS% may be dueto higher methane potential of SM than CCWS, due tohigher TVFA loading with SM for the same g VS added(Table 5) and another reason may be due to inhibition caused

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

6 G.K. Kafle et al.

Table 4. Data obtained from ANOVA analysis and LSD analysis at significance level (α) of 5%and 1%.

Parameter Units LSD F value MSE p-Value

α = 0.05 α = 0.01

Biogas yield mL g−1 VS 70 100 19.0 950 0.003151894Methane yield mL g−1 VS 35 50 77.3 241.2 0.000110584Methane content % 10.7 15.4 11.8 22.4 0.009241704VS removal % 5.7 8.1 61.7 6.2 0.000191874

by rapid acidification of CCWS.[9] Similar to our results,González-Fernández et al. [33] reported lower methaneyield for a mixture of algal biomass and SM compared to SMalone. The VS removal from feed significantly increased(P < 0.01) (Table 4, LSD0.05 = 5.7%, LSD0.01 = 8.1%)with mixture feed (SM and CCWS) compared to SM alone(Table 3). The reason for the increase in VS removal may bedue to the higher quantity of readily degradable compoundsin the CCWS compared to SM.

The initial and final characteristics of the digester con-tents during batch test are shown in Table 5. The TVFA,TVFA/TA ratio, TAN and pH decreased, while the TAincreased after anaerobic digestion. The final pH valuedecreased with increase in CCWS contents in the feed(Table 4). One reason for the decrease in the final pHcould be due to low pH, low TAN and TA concentration(Table 1) of CCWS compared to SM. Another reason couldbe rapid acidification of CCWS which decreased the pH byneutralizing the alkalinity capacity.[9,34]

In the digesters with a mixture ratio of 0:100(SM:CCWS), large amount of biogas was produced on the firstand second days of digestion, but it dropped rapidly ondays 4 and 5, and then gas production almost ceased fora week. For the initial two days, maybe the microbes ininoculum were sufficient to consume the produced acidbut from the third day the acid production rate rapidlyincreased compared to the consumption rate which causedaccumulation of a large quantity of acid.[9] The accumu-lation of acid decreased pH which reduced the activityas well as the growth rate of the methanogens.[35] Thus,results showed an F/I ratio of 1.6 (at OLR=16 g VS L−1) ishigher for sole anaerobic batch digestion of CCWS with-out the use of external additions. Li et al. [36] reportedthat at a loading rate of 10 g VS L−1 kitchen waste as a sin-gle substrate achieved the highest methane yield comparedto cow manure but when the loading rate was increasedto 20 g VS L−1 the digestion process failed due to highF/I ratio (1.90) which lowered the methanogenic activ-ity. Raposo et al. [37] observed inhibition phenomenonwhen F/I ratio increased from 0.3 to 2.0 by increasing sub-strate concentration from 5 to 30 g VS L−1. Thus, batchtest results suggest that at the higher loading rate (OLR≥ 16 g VS L−1 and F/I≥1.6) the CCWS content in feed notmore than 25–33% could be useful for co-digestion with SMkeeping stable digester operation and optimum recovery

of the biogas from the system. Also, results do not sug-gest for exceeding the CCWS contents in the feed greaterthan 67% for co-digestion with the SM without any externaladditions.

3.3. Continuous digester performance with SM andmixture of SM and CCWS

The operation and performance parameters of the contin-uously fed digester are illustrated in Figure 3. Table 6summarizes the steady-state data for reactor processes.After inoculating digester on day 0, SM alone was fed forthe entire test Period I (days 1–45). During the digester start-up period (days 1–20) SM was fed at OLR of 1.0 g VS L−1

d−1. The average biogas yield and methane content was502 mL g−1 VS and 72.5%, respectively. After the biogasproduction was stable the OLR was increased (from day21) to 2.0 g VS L−1 d−1. The biogas yield from SM droppedby around 36% (from 486 to 318 mL g−1 VS) when OLRincreased from 1.0 to 2.0 g VS L−1 d−1. The average pHTVFA, TVFA/TA ratio and TAN were measured to be8.12, 8233, 0.49 and 5400 mg L−1, respectively, at OLRof 2.0 g VS L−1 d−1.

After the end of test Period I, from day 46, the mixtureof SM and CCWS (75:25) was fed to the digester at a load-ing rate of 1.8 g VS L−1 d−1. The methane content, pH andTVFA/TA ratio were 74%, 7.77 and 0.45, respectively, anddigester process was stable. The loading rate was increasedto 2.0 g VS L−1 day−1 from day 82. The increase in OLRto 2.0 g VS L−1day−1 resulted in a drop in biogas yield by11% (from 476 to 423 mL g−1 VS). Also, TVFA/TA ratioincreased from 0.45 to 0.50 and pH, methane content hadslightly dropped. pH dropped down due to the increase inOLR, which results in high VFA concentration during theacidogenesis phase.[34,38] The TVFA/TA ratio is one ofthe most important monitoring parameters of the anaero-bic process. It is more important than the values of eitherparameter alone, because it indicates the buffer capacity ofsystem.[17,39] Callaghan et al. [40] reported that digestercan have potential for instability if the TVFA/alkalinityratio is in the range of 0.40–0.80. However, in our casethe biogas production and methane contents were stablethroughout test Period II even if the TVFA/TA>0.40. Thus,the test data shows that the CCWS up to 25% can be co-digested with SM without disturbing the digester process

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

Environm

entalTechnology7

Table 5. Initial and final characteristics of the digester contents (batch test).

Feed ratio VS (%) Initial Final

SM CCWS TVFAa TAa TVFA/TA TANa pH TVFAa TAa TVFA/TA TANa pH

100 0 7488 (57) 15,087 (18) 0.497 (0.004) 5210 (156) 8.15 (0.05) 3607 (80) 19,138 (31) 0.189 (0.004) 5200 (57) 8.10 (0.03)75 25 6587 (584) 14,458 (322) 0.456 (0.030) 4737 (52) 8.13 (0.07) 3019 (150) 18,148 (322) 0.167 (0.005) 4570 (240) 7.95 (0.06)67 33 6196 (117) 13,545 (291) 0.458 (0.018) 4350 (71) 8.10 (0.01) 3305 (150) 17,797 (165) 0.186 (0.010) 3770 (240) 7.91 (0.01)33 67 4563 (47) 12,571 (42) 0.363 (0.003) 4170 (99) 8.06 (0.02) 2741 (342) 15,392 (587) 0.178 (0.016) 3210 (156) 7.85 (0.04)0 100 3254 (217) 11,268 (42) 0.289 (0.021) 3470 (184) 8.03 (0.03) 2552 (272) 14,335 (290) 0.178 (0.023) 3650 (99) 7.73 (0.031)

CCWS, Chinese cabbage waste silage; SM, swine manure. Values are expressed as mean (standard deviation, n = 2).aUnits in mg L−1.

Table 6. Performance of continuous digester with SM and mixture of SM and CCWS.

Feed ratio (VS %)Test OLR HRT Biogas yield Methane TVFA TA TVFA/ TANperiod days SM CCWS g VS L−1 d−1 (days) (mL g−1 VS) content (%) pH (mg L−1) (mg L−1) TA (mg L−1)

Period I 1–20 100 0 1.0 30 502 (26) 72.3 (1.0) 8.08 (0.07) 7010 (415) 15,844 (240) 0.44 (0.03) ND21–45 100 0 2.0 30 361 (26) 71.6 (1.0) 8.12 (0.07) 8233 (801) 16,991 (465) 0.49 (0.05) 5400 (200)

Period II 46–81 75 25 1.8 20 476 (35) 74.3 (1.2) 7.77 (0.05) 6590 (365) 14,400 (400) 0.45 (0.03) 4320 (90)82–98 75 25 2.0 20 423 (18) 71.2 (0.5) 7.70 (0.04) 6720 (280) 13,560 (140) 0.50 (0.03) 4150 (120)

CCWS, Chinese cabbage waste silage; SM, Swine manure. Values in parentheses are standard deviation (SD). ND, not determined.

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

8 G.K. Kafle et al.

Figure 3. Process performance of continuous digester: (a) bio-gas yield; (b) methane content; (c) TVFA, alkalinity (TA)and TVFA/TA; (d) pH. SM only was fed during days 1–45(1.0 g VS L−1 d−1 during days 1–20, 2.0 g VS L−1 d−1 duringdays 21–45) and mixture of SM and CCWS (75:25) wasfed during days 46–98 (1.8 g VS L−1 d−1 during days 46–81,2.0 g VS L−1 d−1 during days 82–98).

for OLR up to 1.8–2.0 g VS L−1 d−1 and HRT of 20 daysin single stage CSTR under mesophilic conditions.

When comparison was made between gas productionfrom SM and a mixture of SM and CCWS at the same OLR(2.0 g VS L−1) the test results showed that co-digestion of

SM and CCWS (75:25) increased the biogas and methaneyields by 17% and 14%, respectively, compared to SMalone. However, co-digestion resulted in a drop in pH from8.12 to 7.70 and TA from 16991 to 13560 mg L−1 (Table 6)compared to SM alone fed at the same OLR. One rea-son for drop in pH and TA was due to rapid acidificationof CCWS which increased the acid generation and subse-quently neutralized the alkalinity capacity.[9,34] Anotherreason could be decrease in TAN concentration with mix-ture feed because TAN concentration contributes to thealkalinity and helps to maintain the pH value.[17]

Although the digester stability was decreased slightlywith the mixture feed, the gas production, pH, TVFA/TAratio and methane content were constant until the end of thetest. Additionally, the TAN concentration was decreasedfrom 5400 to 4150 mg L−1 (Table 6) with mixture feedwhich may have reduced the ammonia inhibition to methaneformers. Kaparaju and Rintala [7] reported that a mixtureof potato waste (20%) and SM (80%) increased methaneyield up to 30% compared to SM alone at OLR of 2.0 g VSL−1 d−1. Similarly, the specific methane yield was increasedby 19.2% when herbal extraction residues (25%) wereco-digested with SM (75%) compared to SM alone.[31]

4. ConclusionsCCWS was anaerobically treated with SM in both batch andcontinuous modes. The CCWS was found to be a potentialsubstrate for co-digestion with the SM for biogas produc-tion. The batch test results of this study demonstrated that25–33% (VS basis) of CCWS in the feed could be optimumfor anaerobic co-digestion with SM for biogas productionunder mesophilic conditions. The results of the continu-ous tests showed that CCWS can be co-digested with SM(SM:CCWS=75:25, % VS basis) without disturbing thedigestion process at OLR of 2.0 g VS L−1 and HRT of 20days. Additionally, the mixture of SM and CCWS undercontinuous mode showed positive synergetic effects on bothbiogas and methane productions. Thus, this study showedthe feasibility of anaerobic co-digestion of CCWS with SMfor biogas production under mesophilic conditions. Thecontinuous test performed in this study was preliminarywork; thus further long-term investigations are needed fordefining the optimum OLR and operational conditions foranaerobic co-digestion of CCWS with SM.

AcknowledgementsThe authors would like to extend sincere thanks to Rural Devel-opment Administration (RDA, Korea), Kangwon National Uni-versity and University of Idaho College of Agricultural and LifeSciences for their support.

References[1] Choi MH, Park YH. Production of yeast biomass using waste

Chinese cabbage. Biomass Bioenerg. 2003;25:221–226.

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

Environmental Technology 9

[2] Díaz JP, Reyes IP, Lundin M, Horvath IS. Co-digestionof different waste mixtures from agro-industrial activities:kinetic evaluation and synergetic effects. Bioresour Technol.2011;102:10834–10840.

[3] Herrmann C, Heiermann M, Idler C. Effects of ensiling,silage additives and storage period on methane formationof biogas crops. Bioresour Technol. 2011;102:5153–5161.

[4] Pakarinen A, Maijala P, Jaakkola S, Stoddard FL, Kymäläi-nen M, and Viikari L. Evaluation of preservation methodsfor improving biogas production and enzymatic conversionyields of annual crops. Biotechnol Biofuels. 2011;4(20):1–13.

[5] Kafle GK, Kim SH. Effects of chemical compositions andensiling on the biogasproductivity and degradation ratesof agricultural and food processing by-product. BioresourTechnol. 2013;142:553–561.

[6] Hansen KH, Angelidaki I, Ahring BK. Anaerobic diges-tion of swine manure: inhibition by ammoni. Water Res.1998;32(1):5–12.

[7] Kaparaju P, Rintala J. Anaerobic co-digestion of potatotuber and its industrial by-products with pig manure. ResourConserv Recy. 2005;43:175–188.

[8] Strik DPBTB, Domnanovich AM, Holubar P. A pH-basedcontrol of ammonia in biogas during anaerobic digestionof artificial pig manure and maize silage. Process Biochem.2006;41:1235–1238.

[9] Chanakya HN, Sharma I, Ramachandra TV. Micro-scale anaerobic digestion of point source components oforganic fraction of municipal solid waste. Waste Manage.2009;29:1306–1312.

[10] Demirel B, Scherer P. Production of methane from sugarbeetsilage without manure addition by a single-stage anaerobicdigestion process. Biomass Bioenerg. 2008;32:203–209.

[11] Hartmann H, Angelidaki I, Ahring BK. Co-digestion of theorganic fraction of municipal waste types. In: Mata-álvarez J,editor. Biomethanization of the organic fraction of municipalsolid wastes. London: IWA Publishing; 2003. p. 181–197.

[12] EI-Mashad HM, Zhang R. Anaerobic co-digestion offood waste and dairy manure. Trans ASABE. 2007;50:1815–1821.

[13] Li R, Chen S, Li X. Anaerobic co-digestion of kitchen wasteand cattle manure for methane production. Energ Source PartA. 2009;31:1848–1856.

[14] Jagadabhi PS, Lehtomäki A, Rintala J. Co-digestion ofgrass silage and cow manure in a CSTR by re-circulationof alkali treated solids of the digestate. Environ Technol.2008;29:1085–1093.

[15] Liu G, Zhang R, EI-Mashad HM, Dong R. Effect of feed toinoculum ratios on biogas yields of food and green wastes.Bioresour Technol. 2009;100:5103–5108.

[16] Kafle GK, Kim SH. Anaerobic treatment of apple waste withswine manure for biogas production: Batch and continuousoperation. Appl Energy. 2013;103:61–72.

[17] Allen E, Browne JD, Murphy JD. Evaluation of thebiomethane yield from anaerobic co-digestion of nitrogenoussubstrates. Environ Technol. 2013;34(13–14):2059–2068.

[18] Hansen TL, Schmidt JE, Angelidaki I, Marca E,Jansen JC, Mosbæk H, Christensen TH. Measurement ofmethane potentials of solid organic waste. Waste Manage.2004;24(4):393–400.

[19] Kafle GK, Kim SH, Sung KI. Ensiling of fish industry wastefor biogas production: a lab scale evaluation of biochemicalmethane potential (BMP) and kinetic. Bioresour Technol.2013;127:326–336.

[20] Rani RU, Kaliappan S, Kumar SA, Banu JR. Combined treat-ment of alkaline and disperser for improving solubilization

and anaerobic biodegradability of dairy waste activatedsludge. Bioresour Technol. 2012;126:107–116.

[21] Wall DM, O’ Kiely P, Murphy JD. The potential forbiomethane from grass and slurry to satisfy renewable energytargets. Bioresour Technol. 2013;149:425–431.

[22] Merrylin J, Kumar SA, Kaliappan S, Yeom IT, BanuJR. Biological pretreatment of non-flocculated sludge aug-ments the biogas production in the anaerobic digestion ofthe pretreated waste activated sludge. Environ Technol.2013;4(13–14):2113–2123.

[23] APHA. Standard methods for the examination of water andwastewater. 20th ed. Washington, DC: American PublicHealth Association; 1998.

[24] Kafle GK, Kim SH. Sludge exchange process on two serialCSTRs anaerobic digestions: process failure and recovery.Bioresour Technol. 2011;102:6815–6822.

[25] Kafle GK, Kim SH, Sung KI. Batch anaerobic co-digestion of Kimchi factory waste silage and swine manureunder mesophilic conditions. Bioresour Technol. 2012;124:489–494.

[26] VDI 4630. Fermentation of organic materials: character-ization of the substrate, sampling, collection of mate-rial data, fermentation tests. In: Verein DI, editor. Vol.10772. Berlin: VDI-Handbuch Energietechnik/Düsseldorf,Germany: Beuth Verlag GmbH; 2006.

[27] Kafle GK, Kim SH. Kinetic study of the anaerobic digestionof swine manure at mesophilic temperature: a lab scale batchoperation. J Biosystems Eng. 2012;37(4):233–244.

[28] Kafle GK, Bhattarai S, Kim SH, Chen L. Effect of feedto microbe ratios on anaerobic digestion of Chinese cab-bage waste under mesophilic and thermophilic conditions:biogas potential and kinetic study. J. Environ. Manag.2014;133:293–301.

[29] Shin JD, Han SS, Eom KC, Sung S, Park SW, Kim H.Predicting methane production potential of anaerobic co-digestion of swine manure and food waste. Environ Eng Res.2008;13(2):93–97.

[30] Zhang L, Lee YW, Jahng D. Anaerobic co-digestionof food waste and piggery water: focusing on the roleof trace elements. Bioresour Technol. 2011;102:5048–5059.

[31] Li Y, Yan XL, Fan JP, Zhu JH. Feasibility of bio-gas production from anaerobic co-digestion of herbal-extraction residues with swine manure. Bioresour Technol.2011;102(11):6458–6463.

[32] Browne JD, Allen E, Murphy JD. Evaluation of thebiomethane potential from multiple waste streams for a pro-posed community scale anaerobic digester. Environ Technol.2013;34(13–14):2027–2038.

[33] González-Fernández C, Molinuevo-Salces B, García-González MC. Evaluation of anaerobic codigestion ofmicroalgal biomass and swine manure via response surfacemethodology. Appl Energy. 2011;88:3448–3453.

[34] Rani RU, Kumar SA, Kaliappan S, Yeom IC, Banu JR.Impacts of microwave pretreatments on the semi-continuousanaerobic digestion of dairy waste activated sludge. WasteManag. 2013;33:1119–1127.

[35] Neves L, Oliveira R, Alves MM. Influence of inocu-lum activity on the bio-methanization of a kitchen wasteunder different waste/inoculum ratio. Process Biochem.2004;39:2019–2024.

[36] Li R, Chen S, Li X, Lar JS, He Y, Zhu B. Anaerobic co-digestion of kitchen waste with cattle manure for biogasproduction. Energ Fuel. 2009;23:2225–2228.

[37] Raposo F, Borja R, Martín MA, Martín A, De la RubiaMA, Rincón B. Influence of inoculum-substrate ratio on

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014

10 G.K. Kafle et al.

the anaerobic digestion of sunflower oil cake in batchmode: process stability and kinetic evaluation. Chem EngJ. 2009;149:70–77.

[38] Rani RU, Kumar SA, Kaliappan S, Yeom IT, Banu JR. Lowtemperature thermo-chemical pretreatment of dairy wasteactivated sludge for anaerobic digestion process. BioresourTechnol. 2012;103:415–424.

[39] Costa RD, Tavares CRG, Cossich ES. Stabilization ofswine wastes by anaerobic digestion. Environ Technol.2007;28:1145–1151.

[40] Callaghan FJ, Wase DAJ, Thayanithy K, Forster CF. Con-tinuous co-digestion of cattle slurry with fruit and vegetablewastes and chicken manure. Biomass Bioenerg. 2002;27:71–77.

Dow

nloa

ded

by [

Uni

vers

ity o

f Id

aho]

, [G

opi K

afle

] at

08:

38 0

6 Ju

ne 2

014