Anaerobic co-digestion of food waste with MSW incineration plant fresh leachate: process performance and synergistic effects

Download Anaerobic co-digestion of food waste with MSW incineration plant fresh leachate: process performance and synergistic effects

Post on 14-Mar-2017




2 download


  • Chemical Engineering Journal 259 (2015) 795805Contents lists available at ScienceDirect

    Chemical Engineering Journal

    journal homepage: www.elsevier .com/locate /cejAnaerobic co-digestion of food waste with MSW incineration plant freshleachate: process performance and synergistic effects 2014 Elsevier B.V. All rights reserved.

    Corresponding authors. Tel.: +86 411 84707448; fax: +86 411 84706679.E-mail addresses: (W. Zhang), (L. Zhang), (A. Li).Wanli Zhang, Lei Zhang , Aimin Li Key Laboratory of Industrial Ecology and Environmental Engineering (Ministry of Education), School of Environmental Science and Technology, Dalian University ofTechnology, Linggong Road 2, Dalian 116024, PR Chinah i g h l i g h t s

    Anaerobic mono-digestion of foodwaste exhibited poor performanceand stability.

    The deficiency of essential metalelements of food waste was thelimiting factor.

    Propionate inhibition was eliminatedby supplementing metal elements.

    Anaerobic digestion of food wastewas improved by co-digesting withfresh leachate.

    A feasible approach to co-treat foodwaste and fresh leachate wasprovided.g r a p h i c a l a b s t r a c ta r t i c l e i n f o

    Article history:Received 7 June 2014Received in revised form 5 August 2014Accepted 13 August 2014Available online 20 August 2014

    Keywords:Anaerobic co-digestionFood wasteFresh leachateTrace metal elementVFA accumulationa b s t r a c t

    The objectives of this study were to examine the feasibility of improving biogas production and processstability of anaerobic mono-digestion of food waste by co-digesting with MSW incineration plant freshleachate, and to identify the key factors governing performance and stability of anaerobic digestion.For this purpose, a series of semi-continuous experiments were carried out. During a long-term operationperiod, contrary to the failure of mono-digestion of food waste, anaerobic co-digestion with fresh leach-ate exhibited a much better performance and stability in terms of high CH4 yields (375.9506.3 mL/gVSadded), high VS removals (66.981.7%), no VFA inhibition, and stable pH (7.27.8). The unstablemono-digestion of food waste was recovered from process imbalance by supplementing trace metal ele-ments (Fe, Co, Mo, Ni), as indicated by the increased CH4 yields (from 384.1 to 456.5 mL/g VSadded), thedecreased propionate concentration (from 899.0 to 10.0 mg/L), and the increased pH (from 6.9 to 7.4).These results were in line with our analytical results that the food waste was deficient in trace metal ele-ments, and fresh leachate was rich in them. Co-digestion strategy provided abundant trace elements foranaerobic process. Our results clearly demonstrated that the deficiency of metal elements was the reasoncausing the unstable performance of anaerobic mono-digestion of food waste, which was corrected by co-digesting with fresh leachate. This research provides a more technically and economically feasibleapproach to co-treating and co-utilizing food waste and fresh leachate from MSW incineration plant.

    2014 Elsevier B.V. All rights reserved.

  • 796 W. Zhang et al. / Chemical Engineering Journal 259 (2015) 7958051. Introduction

    In 2012, around 171 million tons of municipal solid waste(MSW) were generated in China [1], 5060% (by weight) of whichwas food waste [2]. With the explosive growth of food waste pro-duction, the management strategies for food waste have raised alot of environmental concerns. Due to the features of high moisturecontent, low calorific value, and easy decay of food waste, currentmanagement practices of food waste such as landfill, incineration,composting, and animal feed are less satisfactory in terms ofunsustainable resource recycle, great environmental impact, andcostly investment [3]. It is a challenge to develop an environmen-tally-friendly, efficient, and cost-effective strategy for food wastetreatment.

    Anaerobic digestion is considered to be a desirable disposalmethod for organic wastes owing to valuable products, low emis-sion of secondary pollutants, and high economic feasibility [4-6].Recent researches indicated that food waste was an easily biode-gradable substrate for anaerobic digestion with high carbohydrate,lipid, protein, and moderate moisture content. Browne and Mur-phy [7] reported that the biochemical methane potential (BMP)of food waste (467529 mL CH4/g VSadded) was high. Nonetheless,the poor stability has been often reported as the main problemfor anaerobic digestion of food waste. First, high-calorie food wastewas easily acidified into volatile fatty acid (VFA) by fermentativebacteria, and VFA accumulation led to the decreased pH and theinhibition of methanogenic system [3]. Furthermore, high proteincontent of food waste typically gave a high nitrogen content byhydrolysis and led to the elevated ammonia concentration indigester [8]. In addition, the deficiency of essential trace elementsof food waste limited CH4 production and led to the failure ofanaerobic digestion process [9]. Thus, it is an urgent task to solvethe problem of instability of anaerobic digestion of food waste.

    Fresh leachate from MSW incineration plant (MSWIP) isanother major organic waste stream. Recently, MSW incinerationhas been developing rapidly in China due to the excellent efficiencyof waste reduction. However, the high moisture content and lowcalorific value of MSW due to the large proportion of food waste(5060% by weight) both suggested that it could not be incineratedeffectively without fuel addition or meticulous separation [2]. Inpractice, MSW was stored in storage bunkers for 37 days beforeincineration. By this way, the moisture content was reduced, andthe calorific value was increased [10]. During the storage period,a large amount of fresh leachate (1020% of MSW (by weight))was generated. This is different from some developed countries,where less fresh leachate was produced from source-sorted MSW[11]. Thus, fresh leachate from storage bunkers of MSWIP may bea specific waste stream for developing countries. The fresh leachatewas rich in organic matters, VFA, ammonia-N, and metal elements[12]. In conventional completely stirring tank reactor (CSTR), ahigh organic load rate (OLR) could not be achieved by reducingthe hydraulic retention time (HRT) due to the relative low organicstrength. In some cases, the up-flow anaerobic sludge blanket(UASB) reactor and expanded granular sludge bed (EGSB) wereemployed to treat the fresh leachate of MSWIP [10,13]. However,due to the features of fresh leachate rich in calcium ion and ammo-nia-N, UASB and EGSB may encounter the problems of scaling,blocking, poor performance of granule sludge, and sludge loss.Therefore, it is still a challenging task to treat MSWIP fresh leachatein a more effective manner.

    Recently, anaerobic co-digestion attracted more attentions dueto the sharing facility and complementary features of substrates[14,15]. Anaerobic co-digestion of food waste with other compo-nents have been studied, including sewage sludge, cattle manure,piggery wastewater, and other components [16]. With regard tothe fresh leachate, its high alkalinity could improve the bufferingability of anaerobic digestion system. Moreover, fresh leachate richin trace elements due to the diverse sources could increase theenzymatic activity of anaerobes. Besides, the abundant VFA of freshleachate could be used by acetogens and methanogens withouthydrolysis and acidogenesis, which might accelerate the start-upof anaerobic digestion process. Considering complementary fea-tures of food waste and fresh leachate, it seemed to be feasibleto co-digest food waste with fresh leachate. So far, there is noreport about anaerobic co-digestion of food waste and MSWIPfresh leachate. This was the reason to initiate the present study,where the influence of MSWIP fresh leachate addition on perfor-mance and stability of anaerobic digestion of food waste wasexamined. The co-digestion process was evaluated under varyingOLR and HRT conditions, VFA profiles, CH4 yields, and organic rem-ovals in semi-continuous mode. Moreover, the food waste, freshleachate, and digestate were characterized, and special focus wasput on metal elements. In addition, the effect of trace elements(Fe, Co, Mo, Ni) supplementation on the recovery of unstableanaerobic mono-digestion of food waste was tested to explorethe role of metal elements of fresh leachate in maintaining goodperformance and stability of co-digestion process.2. Materials and methods

    2.1. Materials and sample preparation procedures

    The food waste (FW) containing mainly rice, vegetable, andmeat was obtained from a campus restaurant of Dalian Universityof Technology, China. The impurities, such as bones, napkins, plas-tic bags, etc., were picked out by hand. The food waste was homog-enized using an electrical kitchen blender, and then screenedthrough a 14-meshes screen. The food waste sample was frozenat 18 C and thawed for 12 h at 5 C before use.

    The fresh leachate (FL) was obtained from a waste storage bun-ker of a MSW incineration plant, Dalian, China. The incinerationplant had a daily handling capacity of 1500 t MSW, and about175 t/day fresh leachate was generated. The preparation and stor-age methods of fresh leachate sample were the same as those forfood waste.

    The seed sludge was obtained from the anaerobic digester ofone municipal sewage sludge treatment plant, Dalian, China. Thevolatile suspended solid (VSS) of seed sludge was 29.3 0.6 g/L.2.2. Experimental procedures

    In order to evaluate the biodegradability of the food waste andfresh leachate, BMP tests were carried out in triplicate in 500-mLSchott Duran bottles with silica gel stoppers. The working volumeof each bottle was 300 mL, which was filled with 200 mL of seedsludge and 100 mL of substrate. The OLR of each reactor was 10 gVS/L. The control was filled with 200 mL of seed sludge and100 mL of distilled water for background CH4 production. Nitrogengas was injected into the bottles for 5 min to remove the air. Then,the reactors were incubated in a shaking incubator at 150 rpm,37 C. The batch experiments lasted for 30 days.

    The semi-continuous anaerobic digestion experiments werealso carried out in 500-mL Schott Duran bottles. The method ofnitrogen gas injection and the operating conditions of the shakingincubator were the same as BMP tests. The initial working volumeof each bottle was 300 mL and was fed and discharged regularlyonce a day. The proportions of fresh leachate (by VS) in thefeedstock of R1, R2, R3, and R4 were 0%, 5.8%, 11.6%, and 22.7%,respectively. The VS contents were kept at 8.09.3%.

    The experimental period was divided into three phases. InPhase 1, R1R4 were operated at a relative low OLRs (44.1 g VS/

  • W. Zhang et al. / Chemical Engineering Journal 259 (2015) 795805 797L day) with a long HRT (20 days) for 70 days (Day 170). In Phase 2,R1R4 were operated at a moderate OLRs (66.2 g VS/L day) with ashorter HRT (15 days) for 70 days (Day 71140).

    In Phase 3, on Day 141, half of the digestate of R2 was separatedand transferred into a new reactor named as R20. R20 was fed withfood waste alone at a moderate OLR (6 g VS/L day) and HRT(15 days) for 78 days (Day 141218), which was the same as theoperating conditions of R1 in Phase 2. The other half of the dige-state was continued as R2, and R2 was operated at a high OLR(8.1 g VS/L day) and a short HRT (10 days) for 78 days (Day 141218). On Day 196, half of the digestate of R20 was separated andtransferred into another new reactor named as R200, which wasfed with metal elements (Fe, Co, Mo, Ni) and food waste underthe same operating condition as R20 for 23 days (Day 196218).The other half of the digestate in R20 was continued as R20 untilDay 218. The metal elements were added as metal element solu-tion, which was prepared by dissolving FeCl24H2O, CoCl26H2O,Na2MoO42H2O, and NiCl26H2O in distilled water. 0.08 mL of tracemetal elements solution was supplemented to R200 once a day. Thedesignated concentrations of Fe, Co, Mo, and Ni in R200 were 100,2.0, 5.0, and 10.0 mg/L, respectively. R3 and R4 were operated ata high OLR (8.18.3 g VS/L day) and a short HRT (10 days) for78 days (Day 141218). The detailed experimental proceduresand operating conditions of semi-continuous digestion were sum-marized in Fig. 1.

    2.3. Analytical methods

    The pH was determined with a pH meter (PB-10, Sartorius, Ger-many). The total solid (TS) and volatile solid (VS) were measuredaccording to the standard method [17]. The measuring errors ofTS and VS of fresh leachate caused by VFA volatilization duringdrying at 105 C were compensated by VS/TS calibration tests[18]. The total COD (TCOD) and soluble COD (SCOD) were mea-sured with a microwave digestion method [19]. Lipid was analyzedwith a Soxhlet extractor (SXT-06, Shanghai, China). Ammonia-NFig. 1. Summary of experimental procedures and opewas determined by a spectrophotometry (DR-5000, Hach, USA).Total Kjeldahl nitrogen (TKN) was analyzed with a Kjeldahl appa-ratus (KDY9820, Beijing, China). Protein content was estimatedby multiplying the organic nitrogen value (TKN subtracted byammonia nitrogen) by 6.25 [20].

    The composition of biogas was measured using a gas chromato-graph (GC-7800, Beijing) with a thermal conductivity detector(TCD) and a packed column (TDX-01). Biogas volume wascalculated according to the change of molar ratio of CH4 or CO2to N2 [21]. The concentrations of acetate, propionate, iso-butyrate,butyrate, iso-valerate, and valerate were measured using a gaschromatograph (GC-7900, TECHCOMP, Shanghai) with a flame ion-ization detector (FID) and a capillary column (DB-FFAP,30 m 0.25 mm 0.25 lm). The operating temperatures of theinjection port and detector were 250 C and 260 C, respectively.The initial temperature of oven column was 80 C for 1 min, thenincreased to 200 C at the rate of 5 C /min and maintained for1 min. The digestate was centrifuged at 14,000 rpm (13,210g)for 10 min. Then, the supernatant was acidified to below pH 3 by20 w% H3PO4, and 2-ethyl butyric acid was added as an internalstandard before injection [22]. Metal elements were quantifiedby an ion coupled plasma-atomic emission spectrometer (ICP-AES, Optima 2000DV, PerkinElmer, USA).

    3. Results and discussion

    3.1. Characteristics of food waste and fresh leachate

    The characteristics of food waste and fresh leachate are summa-rized in Tables 1 and 2. As shown in Table 1, the food waste wasrich in organic matters. The TS and VS content of food wastereached 232.0 g/L and 217.0 g/L, respectively. Carbohydrate(137.7 g/L), lipid (64.9 g/L), and protein (29.4 g/L) constituted theVS of food waste. The high VS/TS ratio (93.5%) suggested that mostsolid components of food waste were organics. The food waste alsocontained high TCOD (304.4 g/L) and SCOD (155.3 g/L). Theserating conditions of semi-continuous digestion.

  • Table 1Characteristics of food waste used during experiments as compared to the literature reports.

    Parameter Unit Zhang et al. [23] Banks et al. [9] Vrieze et al. [24] This study

    pH 4.71 0.01 a 4.4 0.1SCOD g/L 106 5.3 155.3 1.2TCOD g/L 238.5 3.8 260 47 304.4 2.8TS g/L 181 6 237.4 0.8 255 4 232.0 1.8VS g/L 171 6 217.1 0.9 240 6 217.0 1.5VS/TS ratio % 94 91.44 94.12 93.5Lipid g/L 23.3 0.45 64.9 1.3Proteinb g/L 32.9 1.4 29.4 2.3Carbohydratec g/L 111.7 6.2 137.7 1.4TKN g/L 5.42 0.26 8.12 0.01 11.93 1.1 4.9 0.2Ammonia-N g/L 0.16 0.04 0.384 0.004 0.23 0.01BMP mL CH4/g VSadded 479.5 21.3 480.5 20.2Na mg/L 3458.91 1875K mg/L 1236.909 3390 190 680.538Mg mg/L 141.406 218 190.506Fe mg/L 7.172 54 9.52 38.908Zn mg/L 18.711 7.8 2.6 4.83 13.033Ni mg/L 0.430 1.7 0.7 0.252 3.551Mn mg/L 2.172 20 3 4.69 3.171Cu mg/L 6.923 1.7 0.2 2.06 1.858Mo mg/L 0.057 0.11 0.01 0.1 0.360Co mg/L

  • W. Zhang et al. / Chemical Engineering Journal 259 (2015) 795805 799total COD (53.8 and 62.4 g/L, respectively) and TVFA concentrationof fresh leachate were much higher than aged leachate of landfill[25]. In addition, according to the standard method [17], freshleachate was dried at 105 C and ignited at 550 C to determinethe TS and VS content. However, VFA volatilization may causemeasuring errors during drying at 105 C. According to the resultsof VS/TS calibration tests, the VFA volatilization loss reached5866.6 mg/L. The corrected TS and VS values of fresh leachate were57.7 and 31.7 g/L, respectively. As shown in Tables 1 and 2, freshleachate contained much lower VS and TS content than food waste.The low VS/TS ratio (54.9%) suggested that only half of solid com-ponents of fresh leachate were organics. This feature is associatedwith the complex compositions of MSW. The high TKN concentra-tion (4.9 g/L) and low ammonia-N concentration (0.23 g/L) of foodwaste indicated that most of nitrogen of food waste existed in theform of organic nitrogen like protein. The high concentration ofammonia-N (2.29 g/L) of fresh leachate mainly resulted from thehydrolysis of the organic nitrogen of MSW.

    The BMP of food waste and fresh leachate reached 480.5 and667.8 mL CH4/g VSadded, respectively. The results showed the muchgreater methane production potential of food waste and freshleachate than other organic wastes, such as waste activated sludge(157.1 mL CH4/g VSadded) [26], wheat straw (233 mL CH4/g VSadded)[27], and dairy manure (242.7 mL CH4/g VSadded) [28]. The highorganic content (VS and COD) and BMP value indicated that thefood waste was an excellent substrate of anaerobic digestion. Com-pared to food waste, fresh leachate had a higher BMP and was alsoconsidered to be a highly desirable substrate of anaerobicdigestion.

    The concentrations of metal elements of food waste and freshleachate are shown in Tables 1 and 2. Both food waste and freshleachate were rich in light metals (Na, K, Mg) ranged from 190 to1875 mg/L. The similar results were reported in other literatures[13,23]. The content of light metal elements might be high enoughfor metabolic activities of microorganisms [29]. By contrast, thefood waste contained less trace metal elements than fresh leach-ate. Especially, the concentrations of Fe, Co, and Mo of food wastewere only 18.9%, 9.8%, and 15.9% of their respective concentrationsof fresh leachate. Considering the significant role of trace metals inanaerobic digestion [3,30-32], food waste seemed to be deficient intrace metal elements. Therefore, co-digesting with fresh leachatemay be a good method to adjust the essential trace metal elementsof food waste.

    3.2. Preliminary anaerobic co-digestion of food waste and freshleachate

    In order to verify the feasibility of improving anaerobic diges-tion of food waste by co-digesting with fresh leachate, the perfor-mance and stability of anaerobic co-digestion of food waste andfresh leachate was examined in semi-continuous mode for a longtime. During the experimental period, the OLRs of R1R4 weregradually increased from 44.1 to 8.18.3 g VS/Lday and the HRTswere reduced from 20 to 10 days. Fig. 2 shows the operational con-ditions and performances of R1R4 for 218 days (140 days for R1),including OLR, CH4 productivity, CH4 content, CH4 yield, pH value,TVFA concentration, and VS removal. The specific parameters aresummarized in Table 3.

    As shown in Fig. 2, R1 was stable under a relative low OLR (4 gVS/L day) and a long HRT (20 days) conditions in Phase 1, as indi-cated by high CH4 yield (425.2 ml/g VSadded), high CH4 content(61.3%), high VS removal (79.7%), suitable pH (7.5), and low VFAconcentration. However, after the OLR was increased to 6 g VS/L day and the HRT was reduced to 15 days in Phase 2, R1 exhibitedan unsustainable anaerobic digestion of food waste and failed, i.e.methane production declined to zero (from 371.5 mL/g VSadded to0), when pH reached 3.4 (from 7.2 to 3.4) due to accumulation ofTVFA (from 229.0 to 7831.9 mg/L). These results confirmed theinstability of anaerobic mono-digestion of food wastes observedby others [3,9,23,32]. The VFA accumulation in R1 could beexplained by the imbalance between the rate of hydrolysis andacidogenesis and that of the generation of acetic acid and itsconsumption by methanogens for CH4 production [33]. The VFAaccumulation resulted in the decreased pH and low activity ofmethanogens, and finally led to the failure of R1. Thus, VFA accu-mulation was the result of process imbalance, and the fluctuationof VFA level could be considered to be a process indicator for insta-bility. In addition, it was found that the TVFA concentration of R1still increased after the failure of CH4 production (on Day 127)and reached the peak on Day 129, then decreased slightly from8712.7 mg/L to 6169.6 mg/L during Day 130140. The results couldbe explained that the methanogens were more sensitive to thefluctuation of pH than acidogens and were inactivated under lowpH condition. The decomposition rate of VFA declined with thedecreased activity of methanogens. However, the pH-insensitiveacidogens were still active and VFA was continually generatedunder the suitable pH (5.9 on Day 127) for acidogenesis, accumu-lating the VFA to higher concentration. As pH further decreased,the activity of acidogens also declined, and the VFA concentrationof R1 decreased slightly with the regular feedstock supplementa-tion and digestate discharge once a day.

    In contrast to the failure of R1, anaerobic co-digestion in R2R4exhibited the better performance and stability during the wholeoperating period, as indicated by high CH4 yield, high CH4 content,suitable pH, and high VS removal. The CH4 yields of R2R4 were375.9450.2 mL/g VSadded, 419.1466.4 mL/g VSadded, and 452.2506.3 mL/g VSadded, respectively. It should be noted that theimproved CH4 yields in R2R4 were accompanied by the increasedproportion of fresh leachate addition, and R4 (22.7% fresh leachate)exhibited the best performance. The CH4 contents of R2R4 were58.862.6%, 61.963.7%, and 64.265.5%, respectively, which weremuch higher than those of anaerobic digestion of organic wastesreported in literatures [3,23]. The VS removals of R2-R4 were above66% during Phases 13. The result indicated that most of theorganic matters were converted into biogas by anaerobic microor-ganisms even under the conditions of high OLRs and short HRTs. Inaddition, the pH values of R2R4 were between 7.23 and 7.79, andrare fluctuation was observed. The results suggested that the addi-tion of fresh leachate effectively increased the pH value and thebuffering ability of anaerobic digestion system and played animportant role in maintaining stability of co-digestion process,especially in the case when basic substrate had low pH value (typ-ical for food waste). Thus, the good performance and stability ofR2R4 showed the positive effect of fresh leachate addition onsemi-continuous anaerobic digestion of food waste.

    Fig. 3 shows the VFA profiles of R1R4. During the late stage ofPhase 2, the rapidly increased concentration of propionate (from 0to 6470.6 mg/L) and the decreased pH values (from 7.2 to 3.4) in R1were accompanied by the decreased CH4 yields (from 371.5 mL/gVSadded to 0). Conversely, the concentrations of other acids in R1only showed a small fluctuation during the same period. Theseresults confirmed previous observations that propionate playedan important role in VFA inhibition of anaerobic mono-digestionof food waste [9]. Wang et al. [34] suggested that the decomposi-tion rate of propionate was much lower than other acids. Thus,once propionate has accumulated to a certain level, it is difficultto recover from propionate inhibition spontaneously. In the pres-ent experiment, the imbalance between the rates of propionategeneration and degradation in R1 resulted in propionate accumu-lation, which might exert a strong inhibition of CH4 fermentationand eventually led to the failure of R1. Thus, propionateaccumulation was considered to be a sensitive indicator for process

  • Fig. 2. The performance of semi-continuous anaerobic co-digestion of food waste and fresh leachate under varying OLR and HRT conditions: CH4 productivity (A); CH4content (B); CH4 yield (C); pH value (D); total VFA concentration (E); VS removal (F).

    800 W. Zhang et al. / Chemical Engineering Journal 259 (2015) 795805instability. Interestingly, during the initial stage of Phase 1 (Day 120), the propionate concentrations of four reactors were extremelyhigh and even higher than propionate inhibition concentration ofR1 in Phase 2, but no propionate inhibition was observed. It mightbe attributed to the different speciation of propionate under vary-ing state of the process. Wang and Wang [35] showed that VFAinhibition level was closely related to pH, and the inhibition ofundissociated fraction of VFA was much greater than that of thedissociated fraction. According to HendersonHasselbalch equa-tion, the concentration of undissociated acid is a function of pHand pKa (the negative logarithm of acid dissociation constant),and increases with decreasing pH [36]. The concentrations of theundissociated propionate of R1R4 during start-up stage (Day 120) were much lower at a high pH (7.07.6) than that of R1 duringthe late stage of Phase 2 at a low pH (3.45.9). Thus, no propionateinhibition was observed in R1R4 during the start-up period.

    As shown in Fig. 3(D), when the OLR was increased to 8.3 g VS/L day and the HRT was reduced to 10 days in Phase 3, some fluctu-ations of acetate and propionate concentrations in R4 appeared,but few fluctuation was observed in R2 and R3. It might be attrib-uted to the highest proportion of fresh leachate addition in thefeedstock of R4. As shown in Table 2, the fresh leachate was richin VFA (25476.4 mg/L), especially acetate (7117.5 mg/L), propio-nate (3196.9 mg/L), and butyrate (9085.2 mg/L), which increasedthe VFA concentration of the feedstock of R4. Meanwhile, theseorganic acids (except for acetate) could not be metabolized bymethanogens directly and have to be oxidized to acetate, hydro-gen, and carbon dioxide, thus accumulating acetate to a high level.Especially, the conversion rate of propionate was much lower thanother acids [34], and that was why there were some fluctuations ofpropionate concentration in R4 with the largest proportion of freshleachate addition. Nevertheless, it should be noted that even ifthere were some fluctuations of VFA concentration in Phase 3, noVFA inhibition occurred in R4, as indicated by the extremely highCH4 yield, high CH4 content, stable pH value, and high VS removal.The reason was that the concentrations of the undissociated propi-onate and acetate in R4 at a high pH (7.5) were so low that theycould not affect process performance and stability. These resultsindicated that the addition of fresh leachate effectively increasedthe buffering ability of anaerobic co-digestion system, and the fluc-tuations of acetate and propionate in R4 were self-controllable andcould recover spontaneously. Thus, R4 still maintained satisfactoryperformance and stability even if under the conditions of a highOLR (8.3 g VS/L day) and a short HRT (10 days).

    In summary, these results showed that the unstable perfor-mance of anaerobic digestion of food waste was greatly improvedby co-digesting with fresh leachate. With increasing the proportionof fresh leachate addition, anaerobic co-digestion showed thebetter performance, and specially, the best performance wasobtained in R4 (22.7% fresh leachate) in terms of high CH4 yields

  • Table3











































































































































































    L/gVS a















































































    W. Zhang et al. / Chemical Engineering Journal 259 (2015) 795805 801(452.2506.3 mL/g VSadded), high CH4 contents (64.265.5%),suitable pH values (7.57.8), and high VS removals (71.678.0%).On the contrary, anaerobic mono-digestion of food waste in R1 suf-fered from process imbalance and propionate accumulation, whichinhibited CH4 production and led to the process failure.

    3.3. Verification of the positive effect of fresh leachate addition onperformance and stability of anaerobic digestion of food waste

    In order to further confirm the positive effect of fresh leachateaddition on the performance and stability of anaerobic digestionof food waste, the experiment of R20 was carried out. On Day141, R2 was divided into two parts, one still was named as R2,the other as R20. R20 was fed with food waste alone at a moderateOLR (6 g VS/L day) and HRT (15 days) for 78 days (Day 141218),which were the same as the operating conditions of R1 in Phase2. For comparison, R2 was fed with mixture of food waste and freshleachate as before. Fig. 4 shows the profiles of CH4 yield, CH4 con-tent, pH value, VFA concentration, and VS removal for R20 as com-pared with R1 and R2.

    As mentioned above, R2 (94.2% food waste + 5.8% fresh leach-ate) showed a good performance and stability during Phases 1and 2. In contrast, after changing the mixed feedstock to 100% foodwaste in Phase 3, an unstable anaerobic digestion of food wastewas observed in R20, as indicated by the low CH4 yield and theaccumulation of VFA after 60 days. For instance, during the latestage of Phase 3 (Day 200218), the CH4 yield and CH4 content ofR20 gradually decreased to 0 with the decline of pH (from 6.9 to4.7) and the accumulation of TVFA (from 1689.4 to 8751.8 mg/L).Meanwhile, the concentration of propionate in R20 sharplyincreased from 1197.9 to 6721.7 mg/L, which might exert an addi-tional inhibitory effect on process performance of R20. These resultsagreed with the profiles of R1 during the late stage of Phase 2.

    In summary, the failure of R20 again confirmed that anaerobicmono-digestion of food waste was unstable. On the contrary tothe failure of R1 and R20, co-digestion of food waste and freshleachate (R2R4) exhibited a satisfactory performance and stabil-ity. The results suggested that some factors were limiting in thefood waste, which could be corrected by co-digestion strategy.

    3.4. Confirmation of the key role of trace metal elements (Fe, Co, Mo,Ni) of fresh leachate in maintaining stability of co-digestion

    Although Figs. 24 clearly showed that the poor stability ofanaerobic digestion of food waste was improved by supplementingfresh leachate, it was still unclear what factors of fresh leachatewere responsible for this improvement. By comparing the featuresof food waste and fresh leachate in Tables 1 and 2, it was noticedthat fresh leachate contained muchmore metal elements than foodwaste, especially trace metal elements. Several specific metal ele-ments are indispensable to the enzyme cofactors of the biochemis-try of CH4 formation. Schattauer et al. [29] summarized the generalfunctions of the essential metal elements in various enzymes andmethanogenesis involved in anaerobic reaction and transforma-tion. In practice, the requirements and positive effects of tracemetal elements like Fe, Co, Mo, and Ni in anaerobic digestionsystem of various substrates such as food waste [3,9,23,24], maizesilage [30], and agricultural waste [31] were reported. The charac-teristics of food waste were strongly dependent on the regionalculture and the sources of raw food material. Taking those reasonstogether, it is necessary to re-examine the role of trace elements inco-digestion of food waste and fresh leachate.

    Fig. 5 presents the concentrations of Fe, Co, Mo, Ni in thedigestates of R1R4 on Day 140 (the end of Phase 2). It was foundthat the concentrations of Fe, Co, Mo, Ni in the digestate were sig-nificantly increased by fresh leachate addition. For example, the

  • Fig. 3. The VFA concentrations of semi-continuous anaerobic co-digestion of food waste and fresh leachate under varying OLR and HRT conditions.

    802 W. Zhang et al. / Chemical Engineering Journal 259 (2015) 795805concentrations of Fe, Co, Mo, Ni in R4 were 123.066, 0.306, 0.334,and 2.668 mg/L, respectively, which was 5.2, 6.7, 4.6 and 1.5-foldsof those in R1. It should be noticed that the improved performancewas coincided with the increased proportion of fresh leachateaddition in feedstock and the increased concentrations of Fe, Co,Mo, Ni in reactors. For instance, R1 with least metal elements,showed the worst performance and stability. In contrast, R2R4containing more metal elements showed the better performanceand stability than R1, and specially, R4 containing highest amountof metal elements exhibited the best performance. These resultshighly suggested that the trace metal elements from fresh leachatecontributed to the improved performance and stability.

    In order to examine above speculation and further confirm thekey role of metal elements of fresh leachate in maintaining processstability, metal elements (Fe, Co, Mo, Ni) were supplied to unstableanaerobic digestion of food waste. On Day 196, half of the digestateof R20 was transferred into another new reactor named as R200,which was fed with metal elements (Fe, Co, Mo, Ni) and food wasteunder the same conditions as R20 for another 23 days (Day 196218). The other half digestate of R20 was continuously operatedas before. As mentioned above, during the late stage of Phase 3,R20 exhibited instability and finally failed. It should be noted thatR20 was already unstable on Day 196, as indicated by the decreasedpH (from 7.1 to 6.9) and propionate accumulation (from 0 to899.0 mg/L). In other words, R200, derived from R20, was also unsta-ble at the beginning of the experiment (on Day 196). As shown inFig. 6, compared to the failure of R20, the improved performance ofR200 was observed after supplementing metal elements from Day196. During the operating period (Day 196218), the average CH4yield and CH4 content of R200 reached 456.5 mL/g VSadded and59.7%, respectively. The pH value gradually increased from 6.9 to7.4, and the TVFA concentration decreased from 1258.9 to73.4 mg/L. The results indicated that R200 was recovered fromimbalance state by trace elements addition. In contrast, R20 with-out metal elements supplementation, suffered from VFA inhibitionand finally failed. Compared to the working volume of R200

    (100 mL), the volume of metal elements supplementary solution(0.08 mL/day) was negligible, which would not affect the OLRand HRT. In addition, the trace elements solution did not containadditional buffering substance, so the impact of the alkalinity ofsupplementary solution on process could be ruled out. Taking allthings together, the recovery of the unstable anaerobic mono-digestion of food waste in R200 was attributed to the trace elementssupplementation rather than the alkalinity of metal supplementarysolution. These comparative results clearly indicated the essentialrole of trace elements in the recovery of process imbalance.

    Fig. 6 also shows the changes of the VFA concentrations in R20

    and R200. It was found that propionate appeared as the maincomponent of VFA under the unstable state. As mentioned before,propionate is a sensitive process indicator for instability. After sup-plementing metal elements from Day 196, the improved perfor-mance of R200 was accompanied by the decreased propionateconcentration from 899.0 to about 10.0 mg/L. On the contrary,the propionate concentration of R20 grew fast and reached the peakof 6721.7 mg/L on Day 218 without metal elements supplementa-tion. The results indicated that the supplementation of Fe, Co, Mo,and Ni was effective to the rebalance of anaerobic digestion pro-cess by increasing propionate turnover rate and abating propionateinhibition. These results confirmed the positive effect of Fe, Co, Mo,and Ni elements on propionate degradation observed by others[9,37,38]. These four trace elements are essential in the synthesisof enzymes or coenzymes involved in biochemical reaction ofanaerobic digestion, such as CO-dehydrogenase, acetyl-CoA syn-thase, formate dehydrogenase, hydrogenase, B12-enzymes, methyl-transferase, methyl-CoM reductase (F430), etc. [29,39,40]. Some ofthese enzymes play important roles in propionate degradationand subsequent methanogenesis. Thus, the supplementation ofFe, Co, Mo, and Ni overcame the deficiency of trace elements offood waste and accelerated the conversion of propionate.

    In summary, the deficiency of the essential metal elements lim-ited the stability of anaerobic mono-digestion of food waste, asindicated by the failure of R1 and R20. In contrast, R200 was recov-ered from imbalance by trace metal elements (Fe, Co, Mo, Ni), sincetrace metal elements increased propionate degradation rate andabated propionate inhibition. Fresh leachate provided abundantmetal elements for co-digestion process. These results confirmed

  • Fig. 4. The performance of R1, R2, and R20 under varying OLR, HRT, and feedstock conditions: CH4 yield (A); CH4 content (B); pH value (C); total VFA concentration (D); VFAconcentration (E); VS removal (F).

    Fig. 5. Metal element concentrations of Fe, Co, Mo, Ni in R1R4 at the end of Phase 2 (Day 140).

    W. Zhang et al. / Chemical Engineering Journal 259 (2015) 795805 803

  • Fig. 6. The influence of metal elements (Fe, Co, Mo, Ni) supplementation on recovery of unstable anaerobic digestion of food waste: CH4 yield (A); CH4 content (B); pH value(C); total VFA concentration (D); VFA accumulation of R20 without metal elements supplementation (E); R200 recovered from VFA inhibition with metal elementssupplementation (F).

    804 W. Zhang et al. / Chemical Engineering Journal 259 (2015) 795805the key role of metal elements of fresh leachate in maintaining pro-cess stability of co-digestion.4. Conclusions

    This study showed the poor performance and stability of anaer-obic mono-digestion of food waste was significantly improved byco-digesting with MSW incineration plant fresh leachate. Anaero-bic mono-digestion of food waste was unstable, as indicated bythe VFA accumulation and the failed biogas production. In contrast,anaerobic co-digestion of food waste and fresh leachatemaintained satisfactory performance and stability even under theconditions of high OLRs (8.18.3 g VS/L day) and the short HRT(10 days). Specially, 22.7% fresh leachate addition (on VS basis)exhibited the best performance in terms of high CH4 yields(452.2506.3 mL/g VSadded), high CH4 contents (64.265.5%), suit-able pH (7.57.8), no VFA inhibition, and high VS removals(71.678.0%). In addition, the unstable anaerobic mono-digestionof food waste was recovered by metal elements (Fe, Co, Mo, Ni)resulting from stimulating propionate degradation and abatingpropionate inhibition. These results were in line with our analyti-cal results that the food waste was deficient in trace metal ele-ments, and fresh leachate was rich in them. Co-digestion strategyprovided abundant trace elements for anaerobic process. Ourresults clearly showed that the deficiency of essential metal ele-ments were the key factors limiting the performance and stabilityof anaerobic mono-digestion of food waste, which was reversed byco-digesting with MSW incineration plant fresh leachate.Acknowledgments

    The authors acknowledge the financial support from the Natu-ral Science Foundation of China (NSFC) (No. 51208075), NationalKey Technology R&D Program (No. 2012BAC05B04), and LiaoningProvince Education Administration (L2012023).

  • W. Zhang et al. / Chemical Engineering Journal 259 (2015) 795805 805References

    [1] China National Bureau of Statistics, China Statistical Yearbook, China StatisticsPress, Beijing, 2013.

    [2] Z.S. Li, L. Yang, X.Y. Qu, Y.M. Sui, Municipal solid waste management in BeijingCity, Waste Manage. 29 (2009) 25962599.

    [3] L. Zhang, D. Jahng, Long-term anaerobic digestion of food waste stabilized bytrace elements, Waste Manage. 32 (2012) 15091515.

    [4] F. Nez Fernndez, L.A. Fdez-Gelfo, M. Prez Garca, J.L. Garca-Morales, Newapproach for integral treatment of OFMSW: comparative analysis of itsmethane performance versus a conventional continuously stirred tankreactor, Chem. Eng. J. 233 (2013) 282291.

    [5] A. Vintiloiu, M. Boxriker, A. Lemmer, H. Oechsner, T. Jungbluth, E. Mathies, D.Ramhold, Effect of ethylenediaminetetraacetic acid (EDTA) on thebioavailability of trace elements during anaerobic digestion, Chem. Eng. J.223 (2013) 436441.

    [6] S. Shen, I.A. Nges, J. Yun, J. Liu, Pre-treatments for enhanced biochemicalmethane potential of bamboo waste, Chem. Eng. J. 240 (2014) 253259.

    [7] J.D. Browne, J.D. Murphy, Assessment of the resource associated withbiomethane from food waste, Appl. Energy 104 (2013) 170177.

    [8] C.J. Banks, M. Chesshire, S. Heaven, R. Arnold, Anaerobic digestion of source-segregated domestic food waste: performance assessment by mass and energybalance, Bioresour. Technol. 102 (2011) 612620.

    [9] C.J. Banks, Y. Zhang, Y. Jiang, S. Heaven, Trace element requirements for stablefood waste digestion at elevated ammonia concentrations, Bioresour. Technol.104 (2012) 127135.

    [10] Y. Dang, J. Ye, Y. Mu, B. Qiu, D. Sun, Effective anaerobic treatment of freshleachate from MSW incineration plant and dynamic characteristics ofmicrobial community in granular sludge, Appl. Microbiol. Biotechnol. 97(2013) 1056310574.

    [11] D. Chen, T.H. Christensen, Life-cycle assessment (EASEWASTE) of twomunicipal solid waste incineration technologies in China, Waste Manage.Res. 28 (2010) 508519.

    [12] F. Fang, A.A. Abbas, Y.-P. Chen, Z.-P. Liu, X. Gao, J.-S. Guo, Anaerobic/aerobic/coagulation treatment of leachate from a municipal solid wastes incinerationplant, Environ. Technol. 33 (2012) 927935.

    [13] J. Ye, Y. Mu, X. Cheng, D. Sun, Treatment of fresh leachate with high-strengthorganics and calcium from municipal solid waste incineration plant usingUASB reactor, Bioresour. Technol. 102 (2011) 54985503.

    [14] J. Pags-Daz, I. Pereda-Reyes, M.J. Taherzadeh, I. Srvri-Horvth, M. Lundin,Anaerobic co-digestion of solid slaughterhouse wastes with agro-residues:synergistic and antagonistic interactions determined in batch digestion assays,Chem. Eng. J. 245 (2014) 8998.

    [15] X. Shi, X. Yuan, Y. Wang, S. Zeng, Y. Qiu, R. Guo, L. Wang, Modeling of themethane production and pH value during the anaerobic co-digestion of dairymanure and spent mushroom substrate, Chem. Eng. J. 244 (2014) 258263.

    [16] J. Mata-Alvarez, S. Mace, P. Llabres, Anaerobic digestion of organic solidwastes. An overview of research achievements and perspectives, Bioresour.Technol. 74 (2000) 316.

    [17] APHA, WEF, Standard Methods for the Examination of Water & Wastewater,American Public Health Association, Washington, DC, 2005.

    [18] L. Masse, D. Masse, V. Beaudette, M. Muir, Size distribution and composition ofparticles in raw and anaerobically digested swine manure, Trans. ASAE 48(2005) 19431949.

    [19] W.F. Jardim, J.J.R. Rohwedder, Chemical oxygen demand (COD) usingmicrowave digestion, Water Res. 23 (1989) 10691071.

    [20] J.-H. Ahn, T.H. Do, S.D. Kim, S. Hwang, The effect of calcium on the anaerobicdigestion treating swine wastewater, Biochem. Eng. J. 30 (2006) 3338.[21] I. Angelidaki, W. Sanders, Assessment of the anaerobic biodegradability ofmacropollutants, Rev. Environ. Sci. Biotechnol. 3 (2004) 117129.

    [22] L. Zhang, D. Jahng, Enhanced anaerobic digestion of piggery wastewater byammonia stripping: effects of alkali types, J. Hazard. Mater. 182 (2010) 536543.

    [23] L. Zhang, Y.W. Lee, D. Jahng, Anaerobic co-digestion of food waste and piggerywastewater: focusing on the role of trace elements, Bioresour. Technol. 102(2011) 50485059.

    [24] J. De Vrieze, L. De Lathouwer, W. Verstraete, N. Boon, High-rate iron-richactivated sludge as stabilizing agent for the anaerobic digestion of kitchenwaste, Water Res. 47 (2013) 37323741.

    [25] S. Renou, J.G. Givaudan, S. Poulain, F. Dirassouyan, P. Moulin, Landfill leachatetreatment: review and opportunity, J. Hazard. Mater. 150 (2008) 468493.

    [26] X. Liu, W. Wang, X. Gao, Y. Zhou, R. Shen, Effect of thermal pretreatment on thephysical and chemical properties of municipal biomass waste, Waste Manage.32 (2012) 249255.

    [27] L.C. Ferreira, P.J. Nilsen, F. Fdz-Polanco, S.I. Prez-Elvira, Biomethane potentialof wheat straw: influence of particle size, water impregnation and thermalhydrolysis, Chem. Eng. J. 242 (2014) 254259.

    [28] R.A. Labatut, L.T. Angenent, N.R. Scott, Biochemical methane potential andbiodegradability of complex organic substrates, Bioresour. Technol. 102 (2011)22552264.

    [29] A. Schattauer, E. Abdoun, P. Weiland, M. Plchl, M. Heiermann, Abundance oftrace elements in demonstration biogas plants, Biosyst. Eng. 108 (2011) 5765.

    [30] H. Pobeheim, B. Munk, H. Lindorfer, G.M. Guebitz, Impact of nickel and cobalton biogas production and process stability during semi-continuous anaerobicfermentation of a model substrate for maize silage, Water Res. 45 (2011) 781787.

    [31] B. Demirel, P. Scherer, Trace element requirements of agricultural biogasdigesters during biological conversion of renewable biomass to methane,Biomass Bioenergy 35 (2011) 992998.

    [32] H. Qiang, Q. Niu, Y. Chi, Y. Li, Trace metals requirements for continuousthermophilic methane fermentation of high-solid food waste, Chem. Eng. J.222 (2013) 330336.

    [33] A. Montusiewicz, M. Lebiocka, Co-digestion of intermediate landfill leachateand sewage sludge as a method of leachate utilization, Bioresour. Technol. 102(2011) 25632571.

    [34] Q. Wang, M. Kuninobu, H.I. Ogawa, Y. Kato, Degradation of volatile fattyacids in highly efficient anaerobic digestion, Biomass Bioenergy 16 (1999)407416.

    [35] G. Wang, D.I. Wang, Elucidation of growth inhibition and acetic acidproduction by Clostridium thermoaceticum, Appl. Environ. Microbiol. 47(1984) 294298.

    [36] E. Palmqvist, B. Hahn-Hgerdal, Fermentation of lignocellulosic hydrolysates.II: inhibitors and mechanisms of inhibition, Bioresour. Technol. 74 (2000) 2533.

    [37] A. Espinosa, L. Rosas, K. Ilangovan, A. Noyola, Effect of trace metals on theanaerobic degradation of volatile fatty acids in molasses stillage, Water Sci.Technol. 32 (1995) 121129.

    [38] M.B. Osuna, M.H. Zandvoort, J.M. Iza, G. Lettinga, P.N. Lens, Effects of traceelement addition on volatile fatty acid conversions in anaerobic granularsludge reactors, Environ. Technol. 24 (2003) 573587.

    [39] F.G. Fermoso, J. Bartacek, S. Jansen, P.N. Lens, Metal supplementation to UASBbioreactors: from cell-metal interactions to full-scale application, Sci. TotalEnviron. 407 (2009) 36523667.

    [40] M.H. Zandvoort, E.D. van Hullebusch, F.G. Fermoso, P.N.L. Lens, Trace metals inanaerobic granular sludge reactors: bioavailability and dosing strategies, Eng.Life Sci. 6 (2006) 293301.

    Anaerobic co-digestion of food waste with MSW incineration plant fresh leachate: process performance and synergistic effects1 Introduction2 Materials and methods2.1 Materials and sample preparation procedures2.2 Experimental procedures2.3 Analytical methods

    3 Results and discussion3.1 Characteristics of food waste and fresh leachate3.2 Preliminary anaerobic co-digestion of food waste and fresh leachate3.3 Verification of the positive effect of fresh leachate addition on performance and stability of anaerobic digestion of food waste3.4 Confirmation of the key role of trace metal elements (Fe, Co, Mo, Ni) of fresh leachate in maintaining stability of co-digestion

    4 ConclusionsAcknowledgmentsReferences


View more >