performance comparison between mesophilic and thermophilic anaerobic reactors for treatment of palm...

Bioresource Technology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Bioresource Technology

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

Performance comparison between mesophilic and thermophilicanaerobic reactors for treatment of palm oil mill effluent

http://dx.doi.org/10.1016/j.biortech.2014.04.0070960-8524/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +82 2 2220 0411; fax: +82 2 2220 1945.E-mail address: [email protected] (J.-Y. Park).

Please cite this article in press as: Jeong, J.-Y., et al. Performance comparison between mesophilic and thermophilic anaerobic reactors for treatmpalm oil mill effluent. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.biortech.2014.04.007

Joo-Young Jeong, Sung-Min Son, Jun-Hyeon Pyon, Joo-Yang Park ⇑Department of Civil and Environmental Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 133-791, Republic of Korea

h i g h l i g h t s

� Anaerobic combined system was used to treat fresh POME in Sumatra, Indonesia.� Fresh POME was mechanically pretreated by using screw decanter.� Total COD removal was 90–95% in both conditions at the OLR of 15 kg [COD]/m3/d.� Thermophilic was better for COD removal and biogas production than mesophilic.� Maximum biogas production was 20.0 l/d at 15 kg [COD]/m3/d OLR in thermophilic.

a r t i c l e i n f o

Article history:Received 10 January 2014Received in revised form 3 April 2014Accepted 4 April 2014Available online xxxx

Keywords:Anaerobic digestionOrganic loading ratePalm oil mill effluentMesophilicThermophilic

a b s t r a c t

The anaerobic digestion of palm oil mill effluent (POME) was carried out under mesophilic (37 �C) andthermophilic (55 �C) conditions without long-time POME storage in order to compare the performanceof each condition in the field of Sumatra Island, Indonesia. The anaerobic treatment system was com-posed of anaerobic hybrid reactor and anaerobic baffled filter. Raw POME was pretreated by screw decan-ter to reduce suspended solids and residual oil. The total COD removal rate of 90–95% was achieved inboth conditions at the OLR of 15 kg [COD]/m3/d. The COD removal in thermophilic conditions was slightlybetter, however the biogas production was much higher than that in the mesophilic one at high OLR. Theorganic contents in pretreated POME were highly biodegradable in mesophilic under the lower OLRs. Thebiogas production was 13.5–20.0 l/d at the 15 kg [COD]/m3/d OLR, and the average content of carbondioxide was 5–35% in both conditions.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Palm oil is one of the most popular vegetable oils in the worldand its consumption has increased steadily (Lim and Teong,2010). Palm oil mill effluent (POME) is wastewater generated frompalm oil milling process. About 5–7 tons of water are required toproduce 1 ton of crude palm oil and half of water used in the mill-ing process is turned into POME. Raw POME contains 0.6–0.7%residual oil and 2–4% of suspended solids, which are mainly debrisof palm fruit mesocarp. POME is a colloidal suspension of brownishcolor which has very high concentrations of chemical oxygendemand (COD) due to low carbon numbers (8–20) of amino andfatty acids dissolved in it. Therefore, the discharge of untreatedPOME is a great threat to surrounding watershed and water body(Wu et al., 2009; Yeoh et al., 2011).

The aerobic process is not suitable for treatment of POME due tounbalanced nutrient content. On the other hand, anaerobic treat-ment is favorable for POME treatment since POME contains highconcentration of COD (Chin et al., 1996). The ponding system isthe most commonly used conventional anaerobic method for thetreatment of POME (Yacob et al., 2005). The ponding systemrequires low operational costs; however, it needs a long retentiontime and large treatment area. Furthermore, the generated biogassuch as methane and carbon dioxide is released directly into theatmosphere from the open pond that accelerates global warming(Gobi et al., 2011). To solve these problems, closed anaerobicdigesters such as the up-flow anaerobic sludge blanket (UASB)and anaerobic filter (AF) has been investigated for developing ofan efficient process (Poh and Chong, 2009; Fang et al., 2011). Theup-flow anaerobic sludge-fixed film was used for treating POME,and a COD removal efficiency of 89–97% was achieved (Najafpouret al., 2006). Zhang et al. (2008) obtained a COD removal efficiencyof 91% using expanded granular sludge bed. The sequentialanaerobic treatment system composed of UASB and down flow

ent of

http://dx.doi.org/10.1016/j.biortech.2014.04.007

mailto:[email protected]


http://www.sciencedirect.com/science/journal/09608524

http://www.elsevier.com/locate/biortech


2 J.-Y. Jeong et al. / Bioresource Technology xxx (2014) xxx–xxx

anaerobic filter was investigated to treat confectionery wastewaterunder mesophilic condition at 25–35 �C (Beal and Raman, 2000). Inthe last case, the total COD removal efficiency was 98% at theorganic loading rate (OLR) of 12.5 kg [COD]/m3/d, and this resultindicated that high efficiency can be achieved with the sequentialanaerobic treatment system in comparison with that obtainedfrom a single process.

The operating temperature is one of the key factors which affectthe COD removal efficiency of anaerobic treatment. In general, it isknown that thermophilic digester can be operated at high OLRsand can produce more biogas than that generated from mesophilicdigester (Dinsdale et al., 1997; Yu et al., 2002). Several researchprojects have conducted investigations to find the effect of operat-ing temperature on anaerobic digestion by using various wastewa-ters (Choorit and Wisarnwan, 2007; Ramakrishnan and Surampalli,2013). However, the accumulation of volatile fatty acid (VFA) fre-quently occurred on account of washout biomass at thermophiliccondition.

A higher substrate degradation rate and biogas production ratehave been commonly reported as advantages in the thermophilicanaerobic POME digestion. Raw POME discharged at high temper-atures around 80–90 �C was expected to be treated by thermo-philic anaerobic condition without difficulty. From this point ofview, several researchers have validated the feasibility of the ther-mophilic anaerobic process with various conditions (Khemkhaoet al., 2012, 2011; Chou et al., 2010; Chan et al., 2010). Neverthe-less, the mesophilic condition still can be preferable due to greaterprocess stability. In addition, usually raw POME was collected inpalm oil refinery then brought to the laboratory then stored untilit was used in all cases. During this time, the characteristics of col-lected POME can be changed.

In this study, the combined anaerobic digestion system wasexamined for comparison between mesophilic and thermophilicconditions. The combined anaerobic digestion system was com-posed of two hybrid reactors for stable operation. The primarydigester, anaerobic hybrid reactor (AHR), consists of UASB andAF. In AHR, the wastewater was treated in two steps; by the micro-organisms from the sludge bed at the lower part of the reactors andby a biofilm formed on the filter media in the upper part of thereactors. Additionally, the media prevents the washout of the gran-ule. The secondary digester, anaerobic baffled filter (ABF), wasmade of plug flow reactor with a vertically installed baffle whichis a combination of anaerobic baffle reactor and AF. In comparisonwith a previous study conducted by Beal and Raman (2000), ABFshowed better performance than that of anaerobic downflow filterin terms of accumulated sludge collection. Therefore, ABF couldfacilitate high solids removal and long biomass retention. Also,ABF may have resilience to both hydraulic and organic shock load-ings. Finally, the objectives of this study were to compare and toevaluate the performance of combined system of AHR and ABFunder mesophilic and thermophilic anaerobic condition for POMEtreatment.

Table 1The characteristics of the POME used in this study.

Parameter Fresh POME Pretreated POME

pH 3.7–4.4 3.7–4.3COD 71,800–98,000 58,480–59,520Total suspended solids 25,380–34,900 3,400–3,860Volatile suspended solids 22,820–27,910 2,040–2,536Total nitrogen 570–1,120 380–880Total phosphorus 628–2,370 182–1,200

Note: All parameters are in mg/l except pH.

2. Methods

2.1. Characteristics of the raw POME

The raw POME was sampled from Amagra palm oil plantationon Sumatra Island, Indonesia. To prevent chemical change of sam-ples such as acidogenesis, all experiments were performed in thenext building of palm oil refinery. High concentrations ofsuspended solids (SS) in the POME can lead to problems, such asproducing scum and clogging (Latif et al., 2011). The residual oilin the POME also causes a problem in anaerobic treatment bycoating the surface of anaerobic granule. In this study, in order to

Please cite this article in press as: Jeong, J.-Y., et al. Performance comparisonpalm oil mill effluent. Bioresour. Technol. (2014), http://dx.doi.org/10.1016/j.b

avoid these problems, the raw POME was pretreated by a screwdecanter to remove the SS and residual oil. Thus, considerableamount of SS (around 90%) was removed. The characteristics ofthe POME are presented in Table 1. The suitable ratio of COD:N:Pin anaerobic digestion was reported to be 300–500:5:1(Annachhatre, 1996). Since the pretreated POME in this studyhad an appropriate content of nitrogen and extra surplus contentof phosphorous, the addition of nutrients during treatment wasnot necessary.

2.2. Reactor design and configuration

The schematic diagram of experimental set-up is shown inFig. 1. Exactly the same set-up was used to operate the anaerobictreatment at two different temperatures: 37 �C and 55 �C. The feedflows were closely controlled to maintain same OLR using one peri-staltic pump with silicon tubes (I.D. 8 mm, Masterflex�). The pri-mary reactor, AHR, was made from acrylic resin so as to have avolume of 3 l (H: 60 cm � I.D. 8 cm). The upper zone (20% of reac-tor volume) of AHR was filled with media (Tri-pack, polyethylene,diameter: 1 in., surface area: 85 ft2/ft3, specific gravity: 0.95, vol-ume void space: 90%, Solmaro Trading & Engineering Company,South Korea) for tri-phase separation as well as prevention ofwashout. The surge tank (H: 20 cm � I.D. 5 cm) with a volume of0.4 l was located next to AHR in order to recycle the accumulatedVSS to the AHR again. The recycle flow rate was maintained consis-tently 30 times of influent flow rate for mixing of AHR. The second-ary reactor, ABF was made from acrylic resin with a volume 1.5 l(H: 30 cm � I.D. 8 cm). The vertical baffle (H: 20 cm) in ABF forcesthe POME to flow under and over to improve the degradation. 80%of ABF volume was filled with the same media of AHR. All reactorsexcept the surge tanks were equipped with water jacket to keepconstant the temperatures (37 �C or 55 �C).

2.3. Inoculum

The mesophilic seed slurry was acquired from a brewery inSouth Korea. The total suspended solids (TSS) and volatile sus-pended solids (VSS) of slurry containing granule were 53–59 g/land 33–35 g/l, respectively. The mesophilic seed slurry was useddirectly for the initial thermophilic start-up. After failure of thestart-up, the seed slurry had been re-circulated at 55 �C for 4 weekswithout any feed for its acclimation to thermophilic conditions.After filling the reactors with fresh water, anaerobic reactors,except surge tanks, were inoculated with the seed slurry: 0.9 l(30% of the reactor volume) for AHR and 0.225 l (15%) for ABF.

2.4. Experimental method

The OLR was increased incrementally from 2 to 15 kg[COD]/m3/d in consideration of the criteria for steady states sug-gested by Wu et al. (2000). The inflow rate was fixed to 0.73 and0.77 l/d the OLR of 2–4 and 6.5–15 kg [COD]/m3/d, respectively.The total hydraulic retention time of combined systems were 5.8

between mesophilic and thermophilic anaerobic reactors for treatment ofiortech.2014.04.007


Fig. 1. The schematic diagram of experimental set-up.

J.-Y. Jeong et al. / Bioresource Technology xxx (2014) xxx–xxx 3

and 6.2 days, respectively. The OLR was adjusted by diluting POMEwith freshwater, thus the COD of the influent was ranging from8220 to 53,920 mg/l. Diluted POME was stirred continuously inorder to prevent changes of properties due to settling. The totalexperiment period was 93 and 114 days at mesophilic and thermo-philic conditions, respectively.

2.5. Analytical method

All samples were analyzed in duplicate. The COD values weredetermined by dichromate reactor digestion method (HACH,method 8000, range: 20–1500 mg/l). The TSS and VSS were mea-sured as prescribed by the standard method (20th edition). Thebiogas generation was determined using water displacementmethod. To measure volumes of produced biogas, the vent pipesof the reactors are connected with aspirator bottle filled withdeionized water. As the produced biogas enters the bottle, itreplaces same volume of water. The carbon dioxide contents ofcollected biogas were determined by using a gas detection tube(Gastech, 2HH, range: 2.5–40%), then the methane contents weredetermined supposing that the remaining volume of the biogaswas methane.

3. Results and Discussion

3.1. Start-up

The slurry of mesophilic granules were seeded for bothmesophilic and thermophilic reactors directly. The start-up ofmesophilic reactor progressed fairly well showing more than 90%of COD removal rate at the OLR of 6.5 kg [COD]/m3/d. Howeverthe initial start-up of thermophilic reactor failed after 30 days.Even at the low OLR of 4 kg [COD]/m3/d, it showed less than 30%of COD removal rate and the pH of an effluent of 5.7, indicatingorganic overload to a limited amount of the microorganismacclimated to thermophilic temperature (Dos Santos et al., 2004).So, it has decided to stop the start-up of both reactors and thento start the acclimation operation for thermophilic reactors. Thesame amount of slurry seeded again to the failed reactor and hadbeen re-circulated without any feed of POME at 55 �C. After4 weeks, the feed of POME was restarted for both reactors. Thestart-up of thermophilic reactors was longer than that of themesophilic reactor. It took about 86 days to reach the OLR of


15 kg [COD]/m3/d in thermophilic reactors whereas 72 days inmesophilic reactors.

3.2. Variation of pH with different OLRs

The variations of the effluent pH with different OLR are pre-sented in Fig. 2A and B. The pH is one of the most important factorsin anaerobic digestion which can be operated optimally at neutralpH. The POME is an acidic solution with its pH in the range of 3.4–3.7, and thus it is deemed to be not suitable for anaerobic diges-tion. Therefore, the pH inside of reactor was adjusted around 6.4by using sodium bicarbonate during initial stage, and then theadded amount of sodium bicarbonate was gradually decreasedwith increasing the OLR. Thus, the influent pH was decreased from6.4 to 3.7.

The pH of all reactors was maintained over 6 during the initialstage in both mesophilic and thermophilic conditions becausefreshwater filled with granule during the seeding process. The pHof mesophilic (6.9–8.2) AHR was maintained slightly higher thanthat of thermophilic (6.4–7.9) at the OLR of 2–4 kg [COD]/m3/d.This seems to be due to more active acidogenic bacteria in the ther-mophilic AHR in this stage. From the point where the OLR reached6.5 kg [COD]/m3/d, the pH of the effluent was maintained nearneutral value despite the fact that the pH of the influent was verylow. This indicates that the amino and fatty acids dissolved inPOME had been fairly well degraded into intermediates, and subse-quently into methane and carbon dioxide, thereby raising pHinside of the reactors. This also may be due to generation of ammo-nia (Khanal, 2009), consumption of hydrogen ion during anaerobicbiotransformation (Chui et al., 1994), and high recycle ratio wascontributing not only to supply the alkalinity but also to eliminatethe organics (Najafpour et al., 2006). The pH of the both secondaryABF reactors was maintained at same or higher values than that ofthe primary AHR reactors. It seems that methanogenesis is moreactive in ABF than in AHR reactor.

3.3. Variation of COD of the effluent with different OLRs

At the OLR of 2–4 kg [COD]/m3/d, the COD of the effluent ofmesophilic AHR was decreased approximately from 3300 to1780 mg/l at Fig. 3A. On the other hand, at the same OLR condition,the COD of the effluent of thermophilic AHR was temporarilyincreased then steadily decreased repeatedly as it can be seen fromFig. 3B. At the OLR of 15 kg [COD]/m3/d, the COD of the effluent



Fig. 2. Variation of the effluent pH of mesophilic (A) and thermophilic (B)conditions with different OLR (kg [COD]/m3/d).

Fig. 3. Variation of the effluent COD of mesophilic (A) and thermophilic (B)conditions with different OLR ranging between 2 and 15 kg [COD]/m3/d. Thebehavior of the organic removal rate with an increase OLR in ABF (C).


values of both AHR were gradually increased, indicating aninsufficient accumulation of microorganism to cover the organicload. The results also indicate that thermophilic AHR was moreunstable with OLR changes.

The organic removal rate was increased from 1.4 to12.6 kg [COD]/m3/d with an increase of OLRs from 2 to15 kg [COD]/m3/d. A well-fitted linear relationship was observedwhen the organic removal rate was plotted as a function of theOLR (Choi et al., 2013). It shows high correlation with thecoefficient R2 over 0.99. The slope of regression lines of both of themindicates an average COD removal rate of 87.2% and 87.1% in themesophilic and thermophilic AHR, respectively. Therefore the tworeactors were performed almost identically in removing COD.

The effluent from the primary AHR was additionally treated bythe secondary ABF. As it can be seen from Fig. 3A and B, thebehavior of the secondary reactor was greatly affected by theprimary reactor. Fluctuation of AHR COD of the effluent wasmitigated by the following ABF. Fig. 3C shows the organic removalrate of the ABF along with increasing of OLR. Fluctuations of COD inAHR effluents produced very complicated plots on the performanceof the secondary ABF as it is shown in Fig. 3C. From the slopesof the regression lines, even though its correlations were very


low, the average organic removal rates of the mesophilic and ther-mophilic conditions in ABF were found to be 31.6% and 31.0%,respectively.




The ABF COD of the effluent in mesophilic and thermophilicconditions were increased from 1,050 to maximum 4600 mg/land 870 to maximum 3800 mg/l, respectively. Total COD removalrates of the combined system were 90–93% and 93–95%, respec-tively. The thermophilic reactor was performed slightly better thanthe mesophilic one in treating POME. Thermophilic anaerobicreactors have been reported to remove organics better at thehigher OLR than mesophilic reactors (Khemkhao et al., 2012;Ramakrishnan and Surampalli, 2013). Nevertheless, the thermo-philic reactors did not show much improved performance for thetreatment of the POME. It seems that organic content in POMEare highly biodegradable in mesophilic anaerobic condition. There-fore additional removal of organic is hardly expected by switchingPOME treatment into thermophilic anaerobic condition.

3.4. Variation of TSS and VSS in mesophilic and thermophilic operatedreactors with different OLRs

As it is shown in Fig. 4A and B, the TSS in the effluent of AHR andABF reached maximum 2300 mg/l and 790 mg/l, respectively inmesophilic reactors, whereas it reached maximum 2500 mg/l and680 mg/l, respectively in thermophilic reactors. There was notmuch difference between mesophilic and thermophilic reactors.

Fig. 4. Variation of the ABF effluent TSS of mesophilic (A) and thermophilic (B) conditionVSS/TSS ratios (C) and COD/ VSS ratios (D) with an increase OLR ranging between 2 and


These results indicate that the washout of granule was indepen-dent of the fermentation temperature in this combined system.

The TSS in the effluent from both AHR was fluctuated withincrease of OLRs while the TSS in the effluent of ABF was consis-tent. It seems that the tri-pack media in ABF played a key role inthis behavior. The ABF had been effective as a filter to preventgranule wash out. The VSS in the effluent showed a similar ten-dency with the TSS in the effluent regardless of operational tem-perature. It seems that the majority of TSS is VSS, and there arenot much non-biodegradable organics in the pretreated POME.The values of AHR and ABF VSS in the effluent were around2000 mg/l and 500 mg/l in the both mesophilic and thermophilicconditions. However VSS in the effluent of mesophilic reactorwas a little higher than thermophilic one. The removed VSS wouldbe proportional to residual VFA (Song et al., 2004). Thus, the VFAvalue of thermophilic reactor seems to be a little higher than mes-ophilic reactor. This can be supported by the lower values of pH inthe thermophilic reactor (Fig. 2A and B).

Fig. 4C shows the VSS/TSS ratios of APF effluents at the OLRranging between 2 and 15 kg [COD]/m3/d. The VSS/TSS ratios ofthe influent varied with the increase of OLRs. The ratios in meso-philic conditions were initially higher than those in thermophilicconditions up to an OLR of 4 kg [COD]/m3/d, but there became

s with different OLR ranging between 2 and 15 kg [COD]/m3/d. The behavior of the15 kg [COD]/m3/d (all values are ABF effluent except influent).



Fig. 6. A comparison of specific methane production with different OLR rangingbetween 4 and 15 kg [COD]/m3/d.


slightly lower between the OLR of 6.5 and 15 kg [COD]/m3/d. Thisindicates that the organics in POME were consumed slightly morein thermophilic conditions than in mesophilic conditions,especially at higher OLRs. The decrease in VSS/TSS ratios at higherOLRs seems to be due to substrate inhibition effect (Chan et al.,2012). Fig. 4D shows the variation of the COD/VSS ratios atdifferent OLRs. The ratios of AHR effluents from both mesophilicand thermophilic reactors showed similar behaviors in terms ofORL. With increases in OLRs, the ratios of thermophilic reactorsbecame higher than those of mesophilic reactors. This indicatesagain that the organics in POME became more biodegradable inthermophilic reactors.

3.5. Biogas production

Biogas production in mesophilic condition began at OLR of4 kg [COD]/m3/d, after 26 days from the start. Its rate of productiongradually increased from 1.3 l/d to maximum 13.5 l/d with theincrease in OLR as it is shown in Fig. 5A. In thermophilic conditions,the biogas began to produce after 22 days. The production rate ofbiogas in thermophilic condition was around 1–3 l/d at the OLRs

Fig. 5. Variation of the biogas production and carbon dioxide content of mesophilic(A) and thermophilic (B) conditions with different OLR ranging between 2 and15 kg [COD]/m3/d.


of 4–10.5 kg [COD]/m3/d and then reached 20.0 l/d which wasmuch higher than that in mesophilic condition at OLR of15 kg [COD]/m3/d as it is shown in Fig. 5B. The thermophilic reac-tor was much more effective in biogas production at higher OLR.Immediately after a step increase in OLR, the biogas productionwas temporarily ceased and then recovered soon within 1 week.It seems to be due to adaptation lag of methanogenic bacteriaagainst temporary shock load. The carbon dioxide content of theproduced biogas in mesophilic and thermophilic reactors was inthe similar range of 24–32% and 20–35%, respectively over theOLR of 10.5 kg [COD]/m3/d.

The specific methane production ratios (methane produced perCOD removed) at different OLRs of both conditions are compared inFig. 6. The ratios in thermophilic reactors were similar or a littlelower at the OLRs between 4 and 10.5 kg [COD]/m3/d. Then theybecame much higher at the OLR of 15 kg [COD]/m3/d than thosein mesophilic reactors. This indicates that thermophilic reactorswere more effective in biogas production at the higher OLR.Yilmaz et al. (2008) reported similar observations with a paper millwastewater. In this respect the thermophilic reactor produced lessmethane at the OLR lower than 2.4 kg [COD]/m3/d and more meth-ane at the OLR ranging between 5.3 and 11.3 kg [COD]/m3/d.

4. Conclusions

POME was anaerobically treated in mesophilic and thermo-philic conditions using a combination system of AHR and ABF infield of Sumatra Island, Indonesia. A thermophilic reactor per-formed better for COD removal rate and biogas production, whileSS removal was almost same in both reactors. Mesophilic reactorswould be more suitable considering the heating energy for lowOLRs less than 10 kg [COD]/m3/d, and that is because anaerobicreactors performed similarly in both conditions. In terms of biogasproduction, a thermophilic reactor would be more effective at theOLRs higher than 10 kg [COD]/m3/d.

Acknowledgements

This research was supported by the Energy TechnologyDevelopment Program of the Korea Institute of Energy TechnologyEvaluation and Planning (KETEP), granted financial resource from




the Ministry of Trade, Industry & Energy, Republic of Korea (No.20121620100050).

References

Annachhatre, A.P., 1996. Anaerobic treatment of industrial wastewaters. Resour.Conserv. Recycl. 16 (1–4), 161–166.

Beal, L.J., Raman, D.R., 2000. Sequential two-stage anaerobic treatment ofconfectionery waste water. J. Agric. Eng. Res. 76 (2), 211–217.

Chan, Y.J., Chong, M.F., Law, C.L., 2010. Effects of temperature on aerobic treatmentof anaerobically digested palm oil mill effluent (POME). Ind. Eng. Chem. Res. 49(15), 7093–7101.

Chan, Y.J., Chong, M.F., Law, C.L., 2012. Start-up, steady state performance andkinetic evaluation of a thermophilic integrated anaerobic-aerobic bioreactor(IAAB). Bioresour. Technol. 125, 145–157.

Chin, K.K., Lee, S.W., Mohammad, H.H., 1996. A study of palm oil mill effluenttreatment using a pond system. Water Sci. Technol. 34 (11), 119–123.

Choi, W.H., Shin, C.H., Son, S.M., Ghorpade, P.A., Kim, J.J., Park, J.Y., 2013. Anaerobictreatment of palm oil mill effluent using combined high�rate anaerobicreactors. Bioresour. Technol. 141, 138–144.

Choorit, W., Wisarnwan, P., 2007. Effect of temperature on the anaerobic digestionof palm oil mill effluent. Electron. J. Biotechnol. 10 (3), 376–385.

Chou, K.W., Norli, I., Anees, A., 2010. Evaluation of the effect of temperature, NaOHconcentration and time on solubilization of palm oil mill effluent (POME) usingresponse surface methodology (RSM). Bioresour. Technol. 101 (22), 8616–8622.

Chui, H.K., Fang, H.H.P., Li, Y.Y., 1994. Removal of formate from wastewater byanaerobic process. J. Environ. Eng. 120 (5), 1308–1320.

Dinsdale, R.M., Hawkes, F.R., Hawkes, D.L., 1997. Comparison of mesophilic andthermophilic upflow anaerobic sludge blanket reactors treating instant coffeeproduction wastewater. Water Res. 31 (1), 163–169.

Dos Santos, A.B., Bisschops, I.A.E., Cervantes, F.J., Van Lier, J.B., 2004. Effect ofdifferent redox mediators during thermophilicazo dye reduction by anaerobicgranular sludge and comparative study between mesophilic (30 �C) andthermophilic (55 �C) treatments for decolourisation of textile wastewaters.Chemosphere 55 (9), 1149–1157.

Fang, C., O-Thong, S., Boe, K., Angelidaki, I., 2011. Comparison of UASB and EGSBreactors performance, for treatment of raw and deoiled palm oil mill effluent(POME). J. Hazard. Mater. 189 (1–2), 229–234.

Gobi, K., Mashitah, M.D., Vadivelu, V.M., 2011. Development and utilization ofaerobic granules for the palm oil mill (POM) wastewater treatment. Chem. Eng.J. 174 (1), 213–220.


Khanal, S.K., 2009. Anaerobic Biotechnology for bioenergy Production. Principlesand Applications. A John Wiley & Sons Inc., Publication, USA.

Khemkhao, M., Nuntakumjorn, B., Techkarnjanaruk, S., Phalakornkule, C., 2011.Effect of chitosan on UASB treating POME during a transition from mesophilic tothermophilic conditions. Bioresour. Technol. 102 (7), 4674–4681.

Khemkhao, M., Nuntakumjorn, B., Techkarnjanaruk, S., Phalakornkule, C., 2012.Comparative mesophilic and thermophilic anaerobic digestion of palm oil milleffluent using upflow anaerobic sludge blanket. Water Environ. Res. 84 (7),577–587.

Latif, M.A., Ghufran, R., Wahid, Z.A., Ahmad, A., 2011. Integrated application ofupflow anaerobic sludge blanket reactor for the treatment of wastewaters.Water Res. 45 (16), 4683–4699.

Lim, S., Teong, L.K., 2010. Recent trends, opportunities and challenges of biodiesel inMalaysia: an overview. Renewable Sustainable Energy Rev. 14 (3), 938–954.

Najafpour, G.D., Zinatizadeh, A.A.L., Mohamed, A.R., Hasnain Isa, M., Nasrollahzadeh,H., 2006. High-rate anaerobic digestion of palm oil mill effluent in an upflowanaerobic sludge-fixed film bioreactor. Process Biochem. 41 (2), 370–379.

Poh, P.E., Chong, M.F., 2009. Development of anaerobic digestion methods for palmoil mill effluent (POME) treatment. Bioresour. Technol. 100 (1), 1–9.

Ramakrishnan, A., Surampalli, R.Y., 2013. Performance and energy economics ofmesophilic and thermophilic digestion in anaerobic hybrid reactor treating coalwastewater. Bioresour. Technol. 127, 9–17.

Song, Y.C., Kwon, S.J., Woo, J.H., 2004. Mesophilic and thermophilic temperature co-phase anaerobic digestion compared with single-stage mesophilic-andthermophilic digestion of sewage sludge. Water Res. 38 (7), 1653–1662.

Wu, M., Wilson, F., Tay, J.H., 2000. Influence of media-packing ratio on performanceof anaerobic hybrid reactors. Bioresour. Technol. 71 (2), 151–157.

Wu, T.Y., Mohammad, A.W., Jahim, J.Md., Anuar, N., 2009. A holistic approach tomanaging palm oil mill effluent (POME): biotechnological advances in thesustainable reuse of POME. Biotechnol. Adv. 27 (1), 40–52.

Yacob, S., Hassan, M.A., Shirai, Y., Wakisaka, M., Subash, S., 2005. Baseline study ofmethane emission from open digesting tanks of palm oil mill effluenttreatment. Chemosphere 59 (11), 1575–1581.

Yeoh, C.Y., Chin, N.L., Tan, C.S., 2011. Co-composting of palm oil mill wastes. J. FoodAgric. Environ. 9 (3–4), 880–885.

Yilmaz, T., Yuceer, A., Basibuyuk, M., 2008. A comparison of the performance ofmesophilic and thermophilic anaerobic filters treating paper mill wastewater.Bioresour. Technol. 99 (1), 156–163.

Yu, H.Q., Fang, H.H.P., Gu, G.W., 2002. Comparative performance of mesophilic andthermophilicacidogenic upflow reactors. Process Biochem. 38 (3), 447–454.

Zhang, Y., Yan, L., Chi, L., Long, X., Mei, Z., Zhang, Z., 2008. Startup and operation ofanaerobic EGSB reactor treating palm oil mill effluent. J. Environ. Sci. 20 (6),658–663.


http://refhub.elsevier.com/S0960-8524(14)00502-1/h0005



















































































performance comparison between mesophilic and thermophilic anaerobic reactors for treatment of palm...

Documents