Bioresource Technology 100 (2009) 1–9

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /b ior tech

Review

Development of anaerobic digestion methods for palm oilmill effluent (POME) treatment

P.E. Poh, M.F. Chong *

School of Chemical and Environmental Engineering, Faculty of Engineering, The University of Nottingham Malaysia Campus, Jalan Broga, 43500 Semenyih, Selangor, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history:Received 10 April 2008Received in revised form 11 June 2008Accepted 12 June 2008Available online 25 July 2008

Keywords:POMEAnaerobic treatmentUASBUASFF

0960-8524/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.biortech.2008.06.022

* Corresponding author. Tel.: +603 8924 8347; fax:E-mail addresses: MeiFong.Chong@nottingham

yahoo.com (M.F. Chong).

Palm oil mill effluent (POME) is a highly polluting wastewater that pollutes the environment if dis-charged directly due to its high chemical oxygen demand (COD) and biochemical oxygen demand(BOD) concentration. Anaerobic digestion has been widely used for POME treatment with large emphasisplaced on capturing the methane gas released as a product of this biodegradation treatment method. Theanaerobic digestion method is recognized as a clean development mechanism (CDM) under the Kyotoprotocol. Certified emission reduction (CER) can be obtained by using methane gas as a renewable energy.This review aims to discuss the various anaerobic treatments of POME and factors that influence theoperation of anaerobic treatment. The POME treatment at both mesophilic and thermophilic temperatureranges are also analyzed.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

In the process of palm oil milling, POME is generated throughsterilization of fresh oil palm fruit bunches, clarification of palmoil and effluent from hydrocyclone operations (Borja et al.,1996a). POME is a viscous brown liquid with fine suspended solidsat pH ranging between 4 and 5 (Najafpour et al., 2006). The char-acteristics of POME could be referred to the Data for Engineers:POME (2004). Direct discharge of POME into the environment isnot encouraged due to the high values of COD and BOD. Further-more, with the introduction of effluent discharge standards im-posed by the Department of Environment in Malaysia, POME hasto be treated before being released into the environment (FederalSubsidiary Legislation, 1974).

Anaerobic digestion has been employed by most palm oil millsas their primary treatment of POME (Tay, 1991). More than 85% ofpalm oil mills in Malaysia have adopted the ponding system forPOME treatment (Ma et al., 1993) while the rest opted for opendigesting tank (Yacob et al., 2005). These methods are regardedas conventional POME treatment method whereby long retentiontimes and large treatment areas are required. High-rate anaerobicbioreactors have also been applied in laboratory-scaled POMEtreatment such as up-flow anaerobic sludge blanket (UASB) reactor(Borja and Banks, 1994a); up-flow anaerobic filtration (Borja andBanks, 1994b); fluidized bed reactor (Borja and Banks, 1995a,b)

ll rights reserved.

+603 8924 8017..edu.my, chong_mei_fong@

and up-flow anaerobic sludge fixed-film (UASFF) reactor(Najafpour et al., 2006). Anaerobic contact digester (Ibrahimet al., 1984) and continuous stirred tank reactor (CSTR) have alsobeen studied for treatment of POME (Chin, 1981).

Other than anaerobic digestion, POME has also been treatedusing membrane technology (Ahmad et al., 2006, 2007), aerobicactivated sludge reactor (Vijayaraghavan et al., 2007), and evapora-tion method (Ma et al., 1997).

1.1. Clean development mechanism (CDM)

The utilization of methane gas as a renewable energy from theanaerobic digestion can be used to obtain certified emission reduc-tion (CER) credit by clean development mechanism (CDM) underthe Kyoto protocol (Tong and Jaafar, 2006). Besides helping to re-duce carbon emission to the environment, CDM has the advantageto offer developing countries such as Malaysia to attract foreigninvestments to sustain renewable energy projects (Menon, 2002).Thus, palm oil mills could earn carbon credits as revenue by theutilization of methane gas as renewable energy from anaerobicdigestion of palm oil mill effluent. More emphasis has been givento develop anaerobic treatment for POME since the implementa-tion of CDM. Currently, there are two CDM projects that have beenregistered to recover methane from palm oil mill effluent whichare hosted by Kim Loong Power Sdn. Bhd. (Project, 0867) and Uni-ted Plantations Bhd. (Project, 1153).

Subsequent to this, investigation by Yacob et al. (2006a) on themethane emission from anaerobic pond shows that 1043.1 kg/day/pond of methane gas is emitted. Based on the ponding system

2 P.E. Poh, M.F. Chong / Bioresource Technology 100 (2009) 1–9

investigated, there were 4 anaerobic ponds, which in total pro-duced 4172.4 kg/day of methane gas. It is estimated that approxi-mately RM 1,027,975 per year (€ 228,438.9) can be generated asrevenue if methane gas emitted from the anaerobic ponds are cap-tured as renewable energy. Calculations are based on 300 workingdays and carbon credit price of € 10 per ton of carbon quoted byMenon (2007).

Payback period for investment on anaerobic bioreactors can beshort if carbon credit prices remain high (Menon, 2007). Consider-ing the revenue and advantages achieved through capturing meth-ane gas, palm oil mills could switch to anaerobic bioreactor forPOME treatment.

2. Anaerobic digestion

Anaerobic digestion is the degradation of complex organic mat-ters under the absence of oxygen. This process is time consumingas bacterial consortia responsible for the degradation process re-quires time to adapt to the new environment before they start toconsume on organic matters to grow.

In the process of degrading POME into methane, carbon diox-ide and water, there is a sequence of reactions involved; hydroly-sis, acidogenesis (including acetogenesis) and methanogenesis(Gerardi, 2003). Hydrolysis is where complex molecules (i.e., car-bohydrates, lipids, proteins) are converted into sugar, amino acidand etc. In the step of acidogenesis, acidogenic bacteria will breakdown these sugar, fatty acids and amino acids into organic acidswhich mainly consist of acetic acid (from acetogenesis) togetherwith hydrogen and carbon dioxide. Hydrogen and carbon dioxidewill be utilized by hydrogenotropic methanogens while acetic acidand carbon dioxide will be utilized by acetoclastic methanogens togive methane as a final product.

Methanogenesis is the rate limiting step in anaerobic digestionof POME (Ibrahim et al., 1984). As such, conventional anaerobicdigesters require large reactors and long retention time to ensurecomplete digestion of treated influent. Nonetheless, high-rateanaerobic bioreactors have been proposed (Borja and Banks,1994a,b, 1995a,b; Najafpour et al., 2006; Ibrahim et al., 1984) to re-duce reactor volume, shorten retention time as well as capturemethane gas for utilization.

2.1. Anaerobic and alternative POME treatment methods

Aerobic treatment, membrane treatment system and evapora-tion method are the currently available alternative methods forPOME treatment. The advantages and disadvantages for anaerobicand alternative treatment methods are shown in Table 1. In termsof energy requirement for POME treatment operation, anaerobicdigestion has a stronger advantage over other alternative methodsas it does not require energy for aeration. Furthermore, anaerobic

Table 1Advantages and disadvantages between anaerobic and alternative treatment methods

Treatmenttypes

Advantages Disadv

Anaerobic Low energy requirements (no aeration), producing methanegas as a valuable end product, generated sludge from processcould be used for land applications

Long rarea r

Aerobic Shorter retention time, more effective in handling toxic wastes High einactivsludge

Membrane Produce consistent and good water quality after treatment,smaller space required for membrane treatment plants, candisinfect treated water

Shortconve

Evaporation Solid concentrate from process can be utilized as feed materialfor fertilizer manufacturing

High e

POME treatment produces methane gas which is a value-addedproduct of digestion that can be utilized in the mill to gain morerevenue in terms of CER.

Take for instance the open digesting tank for POME treatmentwithout land application, capital cost quoted by Gopal and Ma(1986) for a palm oil mill processing 30 tons FFB/h is RM750,000. Based on the Chemical Engineering Plant Cost Index in2006 (Ullrich and Vasudevan, 2004) the capital cost for this systemis estimated to be RM 1,147,842 in 2006. Comparing this to thecapital cost for a membrane system in POME treatment for a palmoil mill processing 36 tons FFB/h at RM 3,950,000 (Chong, 2007), itis obvious that the former anaerobic treatment has better advan-tage over other treatment methods in terms of capital cost. Theonly two significant drawbacks of anaerobic treatment are longretention times and long start-up period. However, the problemof long retention times can be rectified by using high-rate anaero-bic bioreactors while the long start-up period can be shortened byusing granulated seed sludge (McHugh et al., 2003), utilizing seedsludge from same process (Yacob et al., 2006b) or maintaining suit-able pH and temperature in the high-rate anaerobic bioreactor forgrowth of bacteria consortia (Liu et al., 2002).

Untreated wastewater with BOD/COD ratio of 0.5 and greatercan be treated easily by biological means (Metcalf and Eddy,2003). With reference to the published values of BOD and COD inData for Engineers: POME (2004), aerobic and anaerobic treatmentis suitable for POME treatment since the BOD/COD ratio is of 0.5. Incomparison of these two treatment methods, the anaerobic treat-ment can be regarded to be more suitable for POME treatmentdue to its lower energy consumption while producing methaneas a value-added product in the process.

2.2. Anaerobic treatment methods

2.2.1. Conventional treatment systemsPonding system is the most common treatment system that is

employed in palm oil mills to treat POME with more than 85% ofthe mills having adopted this method. Ponding system comprisesof de-oiling tank, acidification ponds, anaerobic ponds and faculta-tive or aerobic ponds (Chan and Chooi, 1984). Number of pondsvaries according to the capacity of the palm oil mill. Facultativeor aerobic ponds are necessary to further reduce BOD concentra-tion in order to produce effluent that complies with Federal Sub-sidiary Legislation, 1974 effluent discharge standards.

A typical size of an anaerobic pond in a palm oil mill which hasa processing capacity of 54 tons per hour is 60.0 � 29.6 � 5.8 m(length �width � depth) (Yacob et al., 2006a) which is approxi-mately equivalent to half the size of a soccer field. Size of pond de-pends on the capacity of the palm oil mill as well as the areaavailable for ponds. Anaerobic ponds have the longest retentiontime in ponding system which is around 20–200 days (Chan and

antages Reference

etention time, slow start-up (granulating reactors), largeequired for conventional digesters

Metcalf and Eddy(2003), Borja et al.(1996a)

nergy requirement (aeration), rate of pathogenation is lower in aerobic sludge compared to anaerobic, thus unsuitable for land applications

Leslie Grady et al.(1999), Doble andKumar (2005)

membrane life, membrane fouling, expensive compared tontional treatment

Ahmad et al. (2006),Metcalf and Eddy(2003)

nergy consumption Ma et al. (1997)

P.E. Poh, M.F. Chong / Bioresource Technology 100 (2009) 1–9 3

Chooi, 1984). Investigations by Yacob et al. (2006a) showed thatanaerobic pond had a higher emission of methane with an averagemethane composition of 54.4% compared to open digester tank. Inaddition to that, the methane composition from anaerobic pondswas also found to be more consistent in the gaseous mixture.Methane emission in anaerobic ponds is influenced by mill activi-ties and seasonal cropping of oil palm (Yacob et al., 2006a).

Open digesting tanks are used for POME treatment when lim-ited land area is available for ponding system. Yacob et al. (2005)investigated on the methane emission from open digesting tankswhere each tanks was half the capacity of anaerobic ponds(3600 m3) with retention time of 20 days. Emission of methanegas from open digesting tank was found to be less than anaerobicpond with an average methane composition of 36.0%. Lower meth-ane composition is due to the transfer of oxygen into the tankwhen feed is induced into the tank. Mixing in digesting tanks im-proves the digestion process as bacteria consortia are brought intomore contact with food (Leslie Grady et al., 1999). Nevertheless,mixing in open digesting tank only depends on slow bubblingand eruption of biogas which causes low conversion of methanegas.

2.2.2. Anaerobic filtrationAnaerobic filter has been applied to treat various types of

wastewater including soybean processing wastewater (Yu et al.,2002a), wine vinases (Nebot et al., 1995; Pérez et al., 1998), landfillleachate (Wang and Banks, 2007), municipal wastewater (Bodkhe,2008), brewery wastewater (Leal et al., 1998), slaughterhousewastewater (Ruiz et al., 1997), drug wastewater (Gangagni Raoet al., 2005), distillery wastewater (Acharya et al., 2008), beet sugarwater (Farhadian et al., 2007) and wastewater from ice-creammanufacture (Hawkes et al., 1995; Monroy et al., 1994). Borjaand Banks (1994b, 1995b) have also utilized anaerobic filter forPOME treatment. The packing allows biomass to attach on the sur-face when raw POME feed enters from the bottom of the bioreactorwhile treated effluent together with generated biogas will leavefrom the top of the bioreactor.

Anaerobic filter is selected for wastewater treatment because (i)it requires a smaller reactor volume which operates on a shorterhydraulic retention times (HRTs) (ii) high substrate removal effi-ciency (Borja and Banks, 1994b), (iii) the ability to maintain highconcentration of biomass in contact with the wastewater withoutaffecting treatment efficiency (Reyes et al., 1999; Wang and Banks,2007), and (iv) tolerance to shock loadings (Reyes et al., 1999; VanDer Merwe and Britz, 1993). Besides, construction and operation ofanaerobic filter is less expensive and small amount of suspendedsolids in the effluent eliminates the need for solid separation or re-cycle (Russo et al., 1985).

However, filter clogging is a major problem in the continuousoperation of anaerobic filters (Bodkhe, 2008; Jawed and Tare,2000; Parawira et al., 2006). So far, clogging of anaerobic filterhas only been reported in the treatment of POME at an organicloading rate (OLR) of 20 g COD/l/day (Borja and Banks, 1995b)

Table 2Operating OLR range; COD removal efficiency in various wastewater treatments using ana

Types of Wastewater Operating OLR range(kg COD/m3/day)

COD removal e(%)

Slaughterhouse wastewater 1.0–6.5 79.9 (91.5)POME 1.2–11.4 94.0 (94.0)Baker’s yeast factory effluent 1.8–10.0 69.0 (74.0)

Distillery wastewaters 0.42–3.4 91.0 (93.0)Landfill leachate 0.76–7.63 90.8 (90.8)

() – number in bracket denotes highest COD removal efficiency. N/A – data unavailable.

and also in the treatment of slaughterhouse wastewater at 6 gCOD/l/day. This is due to the fact that other studies were con-ducted at lower OLRs which had lower suspended solid contentcompared to POME.

In general, anaerobic filter is capable of treating wastewaters togive good effluent quality with at least 70% of COD removal effi-ciency with methane composition of more than 50%. Table 2 indi-cates the COD removal efficiency of some treated wastewater usinganaerobic filtration based on highest achievable percentage ofmethane in the generated biogas. In terms of POME treatment,the highest COD removal efficiency recorded was 94% with 63%of methane at an OLR of 4.5 kg COD/m3/day, while overall COD re-moval efficiency was up to 90% with an average methane gas com-position of 60% (Borja and Banks, 1994b).

Investigations have been done to improve the efficiency ofanaerobic filtration in wastewater treatment. For instance, Yuet al. (2002a) found that operating at an optimal recycle ratiowhich varies depending on OLR will enhance COD removal. How-ever, methane percentage will be compromised with increase inoptimal recycle ratio. Higher retention of biomass in the filter willalso lead to a better COD removal efficiency. In order to optimizethe retention of biomass on the filter media surface and trappedsuspended biomass within the interstitial void spaces, Show andTay (1999) suggested the use of support media with high porosityor open-pored surfaces. It was also suggested that continuously fedsystem gives better stability and greater degradation efficiency inanaerobic filters (Nebot et al., 1995).

2.2.3. Fluidized bed reactorFluidized bed reactor exhibits several advantages that make it

useful for treatment of high-strength wastewaters. It has very largesurface areas for biomass attachment (Borja et al., 2001; Toldráet al., 1987), enabling high OLR and short HRTs during operation(Garcia-Calderon et al., 1998; Sowmeyan and Swaminathan,2008). Furthermore, fluidized bed has minimal problems of chan-neling, plugging or gas hold-up (Borja et al., 2001; Toldrá et al.,1987). Higher up-flow velocity of raw POME is maintained for flu-idized bed reactor to enable expansion of the support material bed.Biomass will then attach and grow on the support material. In thisway, biomass can be retained in the reactor. Investigations havebeen done on the application of fluidized bed to treat cutting-oilwastewater (Perez et al., 2007); real textile wastewater (S�en andDemirer, 2003); wine and distillery wastewater (Garcia-Calderonet al., 1998; Sowmeyan and Swaminathan, 2008); brewery waste-water (Alvarado-Lassman et al., 2008); ice-cream wastewater(Borja and Banks, 1995a; Hawkes et al., 1995); slaughterhousewastewater (Toldrá et al., 1987); pharmaceutical effluent (Saravan-ane et al., 2001) and POME (Borja and Banks, 1995b).

OLR ranges and COD removal efficiencies of various wastewatertreatments using fluidized bed is tabulated in Table 3. Based onTable 3, it can be concluded that anaerobic fluidized bed can typi-cally remove at least 65% and up to more than 90% of COD. Inverseflow anaerobic fluidized bed is capable of tolerating higher OLRs

erobicfiltration based on highest % of methane production

fficiency Highest methane composition(%)

Reference

51.1 Ruiz et al. (1997)63.0 Borja and Banks (1994b)65.0 Van Der Merwe and Britz

(1993)63.0 Russo et al. (1985)

N/A Wang and Banks (2007)

Table 3Operating OLR range, COD removal efficiency of various wastewater treatments using fluidized bed reactor

Types of wastewater Operating OLR range(kg COD/m3/day)

COD removal efficiency(%)

Reactorconfiguration

Reference

POME 10.0–40.0 78.0–94.0 UF Borja and Banks (1995b)Protein production from extracted sunflower

flour effluent0.6–9.3 80.0–98.3 UF Borja et al. (2001)

Ice-cream wastewater 3.2–15.6 94.4 UF Borja and Banks (1995a)Cutting-oil wastewater 11.9–51.3 67.1–95.9 UF Perez et al. (2007)Distillery effluent 6.11–35.09 80.0–92.0 DF Sowmeyan and Swaminathan

(2008)Brewery wastewater 0.5–70.0 80.0–90.0 DF Alvarado-Lassman et al. (2008)Real textile wastewater 0.4–5.0 78.0–89.0 UF S�en and Demirer (2003)

UF – upward flow; DF – downward/inverse flow.

4 P.E. Poh, M.F. Chong / Bioresource Technology 100 (2009) 1–9

compared to up-flow configuration. Alvarado-Lassman et al. (2008)showed that inverse flow fluidized bed shows excellent stabilitywhen overload is applied. By making a comparison between theoperating OLR ranges for POME from Tables 2 and 4, it was foundthat in general, anaerobic fluidized bed is able to operate at higherOLRs, implying that less reactor volume will be required to operateat lower OLRs.

The type of support material in the fluidized bed plays animportant role to determine the efficiency of the entire treatmentsystem (Garcia-Calderon et al., 1998; Sowmeyan and Swamina-than, 2008) for both inverse flow and up-flow systems. Studiesusing fluidized bed to treat ice-cream wastewater showed differentCOD removal efficiencies when different support materials wereused. Hawkes et al. (1995) found that fluidized bed using granularactivated carbon (GAC) gave about 60% COD removal while Borjaand Banks (1995a) obtained 94.4% of COD removal using ovoidsaponite. Thus suitable support material needs to be selected toobtain high COD removal efficiency in the system.

In POME treatment, fluidized bed was found to be a better treat-ment method compared to anaerobic filter due to its ability to tol-erate higher OLRs and its better methane gas production. ShorterHRT (6 h) also proved to be an advantage of fluidized bed overanaerobic filter (1.5–4.5 days) in POME treatment.

2.2.4. Up-flow anaerobic sludge blanket (UASB) reactorUASB was developed by Lettinga et al. (1980) whereby this sys-

tem has been successful in treating a wide range of industrial efflu-ents including those with inhibitory compounds. The underlyingprinciple of the UASB operation is to have an anaerobic sludgewhich exhibits good settling properties (Lettinga, 1995). So far,UASB has been applied for the treatment of potato wastewater(Kalyuzhnyi et al., 1998; Lettinga et al., 1980; Parawira et al.,2006); domestic wastewater (Barbosa and Sant’Anna, 1989; Beh-ling et al., 1997); slaughterhouse wastewater (Sayed et al., 1984);ice-cream wastewater (Hawkes et al., 1995); POME (Borja and

Table 4Performance of UASB in various wastewater treatments

Types of wastewater Operating OLR range(kgCOD/m3/day)

Potato wastewater 1.8–13.91.5–6.125.0–45.0

POME single-stage two-stage(based on methanogenic reactor)

1.27–10.631.1–60.0

Ice-cream wastewater 0.5–5.0Sugar-beet 4.0–5.0Confectionary wastewater 1.25–2.25Pharmaceutical wastewater 0.27–2.00Domestic sewage 3.76

Slaughterhouse wastewater 7.0–11.0

N/A – data unavailable.

Banks, 1994c); pharmaceutical wastewater (Stronach et al.,1987); instant coffee wastewater (Dinsdale et al., 1997); sugar-beet wastewater (Lettinga et al., 1980) and etc. UASB has arelatively simple design where sludge from organic matter degra-dation and biomass settles in the reactor. Organic matter fromwastewater that comes in contact with sludge will be digestedby the biomass granules.

Table 4 indicates some performances of wastewater treatmentusing UASB system. For potato wastewater treatment, Kalyuzhnyiet al. (1998) and Parawira et al. (2006) both observed foamingand sludge floatation in the UASB reactor when operating at higherOLRs (>6.1 kg COD/m3 day). The ability of UASB to tolerate higherOLR for potato wastewater investigated by Lettinga et al. (1980)compared to Kalyuzhnyi et al. (1998) and Parawira et al., 2006 isdue to the fact that the latter two studies were conducted at labo-ratory scale. In general, UASB is successful in COD removal of morethan 60% for most wastewater types except for ice-cream waste-water. Hawkes et al. (1995) suggested that the lower COD removalpercentage from ice-cream wastewater was due to design faults inthe reactor’s three phase separator and high contents of milk fatthat were hard to degrade.

POME treatment has been successful with UASB reactor, achiev-ing COD removal efficiency up to 98.4% with the highest operatingOLR of 10.63 kg COD/m3day (Borja and Banks, 1994c). However,reactor operated under overload conditions with high volatile fattyacid content became unstable after 15 days. Due to high amount ofPOME discharge daily from milling process, it is necessary to oper-ate treatment system at higher OLR. Borja et al. (1996a) imple-mented a two-stage UASB system for POME treatment with theobjective of preventing inhibition of granule formation at higherOLRs without having to remove solids from POME prior to treat-ment. This method is desirable since suspended solids in POMEhave high potential for gas production while extra costs fromsludge disposal can be avoided. Results from this study showedthe feasibility of separating anaerobic digestion into two-stages

COD removal efficiency(%)

Methane composition(%)

Reference

63.0–81.0 54.0–67.0 Kalyuzhnyi et al. (1998)92.0–98.0 59.0–70.0 Parawira et al. (2006)93.0 N/A Lettinga et al. (1980)96.7–98.4 54.2–62.0 Borja and Banks (1994c)

>90.0 60.0–83.0 Beccari et al. (1996)50.0 69.6 Hawkes et al. (1995)95.0 N/A Lettinga et al. (1980)66.0 N/A Forster and Wase (1983)26.0–69.0 N/A Stronach et al. (1987)74.0 69.0 Barbosa and Sant’Anna

(1989)55.0–85.0 65.0–75.0 Sayed et al. (1984)

P.E. Poh, M.F. Chong / Bioresource Technology 100 (2009) 1–9 5

(acidogensis and methanogenesis) using a pair of UASB reactors.The methanogenic reactor was found to adapt quickly with thefeed from the acidogenic reactor and also tolerate higher OLRs. Itwas suggested that OLR of 30 kg COD/m3day could ensure an over-all of 90% COD reduction and efficient methane conversion.

UASB reactor is advantageous for its ability to treat wastewaterwith high suspended solid content (Fang and Chui, 1994; Kal-yuzhnyi et al., 1998) that may clog reactors with packing materialand also provide higher methane production (Kalyuzhnyi et al.,1996; Stronach et al., 1987). However, this reactor might face longstart-up periods if seeded sludge is not granulated. A study byGoodwin et al. (1992) has proved that reactors seeded with granu-lated sludge achieved high performance levels within a shorterstart-up period. It could also adapt quickly to gradual increase ofOLR (Kalyuzhnyi et al., 1996).

2.2.5. Up-flow anaerobic sludge fixed-film (UASFF) reactorUASB and anaerobic filter has been integrated to form a hybrid

bioreactor – UASFF. This hybrid reactor combines the advantagesof both reactors while eliminating their respective drawbacks. Assuch, UASFF is superior in terms of biomass retention, reactor sta-bility at shock loadings and operation at high OLRs while eliminat-ing the problems of clogging and biomass washout in anaerobicfilter and UASB. Ayati and Ganjidoust (2006) has proven thatUASFF is more efficient compared to UASB and anaerobic filter inthe treatment of wood fiber wastewater.

Other investigations of wastewater treatments using UASFF in-cludes sugar wastewater (Guiot and van den Berg, 1985); dairywastewater (Córdoba et al., 1995); slaughterhouse wastewater(Borja et al., 1995c, 1998; Lo et al., 1994); wash waters from puri-fication of virgin olive oil (Borja et al., 1996b); coffee wastewater(Bello-Mendoza and Castillo-Rivera, 1998); brewery wastewater(Yu and Gu, 1996) and POME (Najafpour et al., 2006). Performancesof UASFF for wastewater treatments are tabulated in Table 5.

This hybrid reactor is generally capable of tolerating OLRs high-er than UASB and anaerobic filter. Clogging is not reported in stud-ies on the performance of hybrid reactor. UASFF is also able toachieve COD removal efficiency of at least 70% and above exceptfor wood fiber wastewater as wood fiber is harder to degrade.Methane production for UASFF is also at a satisfactory level. Inthe treatment of POME, Najafpour et al. (2006) found that internalpacking and high ratio of effluent recycle are both vital to controlthe stability of the UASFF reactor. Internal packing effectively re-tained biomass in the column while effluent recycle producedinternal dilution to eliminate effects of high OLR.

2.2.6. Continuous stirred tank reactor (CSTR)CSTR is equivalent to a closed-tank digester with mixer. The

mechanical agitator provides more area of contact with thebiomass thus improving gas production. In POME treatment, CSTRhas been applied by a mill under Keck Seng (Malaysia) Berhad inMasai, Johor and it is apparently the only one which has been oper-ating continuously since early 1980s (Tong and Jaafar, 2006). Otherapplications of CSTR on wastewater treatment include dilute dairy

Table 5Performance of UASFF in various wastewater treatments

Types of wastewater OLR (kg COD/m3day) COD removal efficie

Sugar wastewater 5.0–51.0 63.0–96.0Wash waters from olive oil purification 2.6–17.8 75.7–90.8POME 1.75–23.15 89.5–97.5Slaughterhouse wastewater 2.49–20.82 90.2–93.4Dairy wastewater 1.8–8.4 90.1–92.0Wood fiber wastewater 1.0–15.0 52.0–72.5Coffee wastewater 1.06–6.0 22.4–88.6

N/A: data unavailable.

wastewater (Chen and Shyu, 1996); jam wastewater (Mohan andSunny, 2008) and coke wastewater (Vázquez et al., 2006) wherecoke wastewater was treated in aerobic conditions.

The CSTR in Keck Seng’s palm oil mill has COD removal effi-ciency of approximately 83% and CSTR treating dairy wastewaterhas COD removal efficiency of 60%. In terms of methane composi-tion in generated biogas, it was found to be 62.5% for POME treat-ment and 22.5–76.9% for dairy wastewater treatment. Anotherstudy on POME treatment using CSTR has been investigated byUgoji (1997) where results indicated that COD removal efficiencyis between 93.6–97.7%. The difference of COD removal efficiencybetween the two published results by Keck Seng and Ugoji is dueto the different operating conditions where the latter study wasdone in laboratory scale. In the plant scale POME treatment at KeckSeng’s palm oil mill, the treated wastewater could not be assumedto be well mixed due to the large volume of feed which might af-fect the overall efficiency of the COD removal.

Ramasamy and Abbasi (2000) attempted to upgrade the perfor-mance of CSTR by incorporating a biofilm support system (BSS)within the existing reactor. Low-density nylon mesh were rolledinto cylinders and inserted into the CSTR. This BSS functions as asupport media for growth of biomass. From this study, it was foundthat efficiency of CSTRs can be improved without biomass recy-cling. The implementation of BSS into CSTR can be useful to in-crease COD removal efficiency as well as biogas production inPOME treatment.

2.2.7. Anaerobic contact digestionContact process involves a digester and a sedimentation tank

where sludge from digester effluent is left to settle and the effluentis recycled back into the digester. This process has been imple-mented in POME (Ibrahim et al., 1984); ice-cream wastewater,alcohol distillery wastewater (Vlissidis and Zouboulis, 1993) andfermented olive mill wastewater treatment (Hamdi and Garcia,1991).

Concentrated wastewaters are suitable to be treated by anaer-obic contact digestion since relatively high quality effluent canbe achieved (Leslie Grady et al., 1999). In the study of fermentedolive mill wastewater treatment, anaerobic contact was capableof reaching steady state more quickly compared to anaerobic fil-ter; however, more oxygen transfer in the digester (due to mix-ing) causes this process to be less stable. While scum formationwas reported in POME treatment pilot plant (Ibrahim et al.,1984), instability was not reported in other treatment systems.Despite the problems that might be encountered in anaerobiccontact, this system has been able to remove COD efficiently,achieving up to 80% removal efficiency (Vlissidis and Zouboulis,1993)

2.3. Comparison of various anaerobic treatment methods in POMEtreatment

Table 6 lists the performance of various anaerobic treatmentmethods of POME. Although the fluidized bed reactor has the

ncy (%) Methane composition (%) Reference

N/A Guiot and van den Berg (1985)69.0–75.0 Borja et al. (1996b)62.0–84.0 Najafpour et al. (2006)56.0–74.0 Borja et al. (1998)65.3 Córdoba et al. (1995)N/A Ayati and Ganjidoust (2006)N/A Bello-Mendoza and Castillo-Rivera (1998)

Table 6Performance of various anaerobic treatment methods on POME treatment

OLR (kg COD/m3day) Hydraulic retention time(days)

Methane composition(%)

COD removal efficiency(%)

Reference

Anaerobic pond 1.4 40 54.4 97.8 Yacob et al. (2006a)Anaerobic digester 2.16 20 36 80.7 Yacob et al. (2005)Anaerobic filtration 4.5 15 63 94 Borja and Banks

(1994b)Fluidized bed 40.0 0.25 N/A 78 Borja and Banks

(1995b)UASB 10.63 4 54.2 98.4 Borja and Banks (1994c)UASFF 11.58 3 71.9 97 Najafpour et al. (2006)CSTR 3.33 18 62.5 80 Tong and Jaafar (2006)Anaerobic contact

processa3.44 4.7 63 93.3 Ibrahim et al. (1984)

N/A: data unavailable.a In terms of BOD.

6 P.E. Poh, M.F. Chong / Bioresource Technology 100 (2009) 1–9

ability to treat POME at high OLR with a short retention time, bio-gas capture is not emphasized using this process. Therefore it canbe concluded that UASFF currently gives the best performance inPOME treatment, achieving high COD removal efficiency and highOLR methane production at relatively short HRT compared to con-ventional and other available anaerobic treatment methods.

Table 7 shows the advantages and disadvantages for eachanaerobic treatment method. It can be clearly seen that conven-tional methods are lacking in terms of treatment time, area re-quired for treatment and facilities to capture biogas.Nevertheless, these methods are more economically viable and

Table 7Advantages and disadvantages of various anaerobic treatment methods

Advantages

Conventional anaerobicdigestion (pond anddigester)

Low capital costLow operating and maintenance costAble to tolerate big range of OLR (pond) thus can eascope POME discharge during high crop seasonRecovered sludge cake from pond can be sold as fertil

Anaerobic filtration Small reactor volumeProducing high quality effluentShort hydraulic retention timesAble to tolerate shock loadingsRetains high biomass concentration in the packing

Fluidized bed Most compact of all high-rate processes

Very well mixed conditions in the reactorLarge surface area for biomass attachment

No channeling, plugging or gas hold-upFaster start-up

UASB Useful for treatment of high suspended solid wastewa

Producing high quality effluent

No media required (less cost)

High concentration of biomass retained in the reactorHigh methane production

UASFF Higher OLR achievable compared to operating UASB oanaerobic filtration aloneProblems of clogging eliminatedHigher biomass retentionMore stable operationAbility to tolerate shock loadingsSuitable for diluted wastewater

CSTR Provides more contact of wastewater with biomassthrough mixingIncreased gas production compared to conventionalmethod

Anaerobic contact process Reaches steady state quickly

Short hydraulic retention timeProduces relatively high effluent quality

have the capacity to tolerate a wider range of OLR. High-rate biore-actors are more effective in biodegradation as shorter retentiontimes are needed, producing higher methane yield while compro-mising the OLR, capital and operating cost (i.e. power requirementfor bed fluidization, support media and investment on controlsystems).

Since the implementation and development of CDM, manyearlier registered CDM projects have shown good results and thishas become an encouragement to palm oil millers to switch tohigh-rate bioreactors in POME treatment. Payback period forinvestments on capital and operating costs for high-rate bioreactors

Drawbacks References

large volume for digestion Chan and Chooi (1984)long retention times

ily no facilities to capture biogas

izer lower methane emissionClogging at high OLRs Borja and Banks (1994b, 1995b)High media and support costUnsuitable for high suspended solidwastewater

High power requirements for bedfluidization

Leslie Grady et al. (1999)

High cost of carrier mediaNot suitable for high suspendedsolid wastewatersNormally does not capturegenerated biogas

ter Performance dependant on sludgesettleability

Lettinga (1995), Kalyuzhnyi et al.(1998), Goodwin et al. (1992)

Foaming and sludge floatation athigh OLRsLong start-up period if granulatedseed sludge is not usedGranulation inhibition at highvolatile fatty acid concentration

r Lower OLR when treatingsuspended solid wastewaters

Ayati and Ganjidoust (2006)

Less efficient gas production at hightreatment volumeLess biomass retention

Less stable due to oxygen transferin digesting tank

Hamdi and Garcia (1991)

Settleability of biomass is critical tosuccessful performance

P.E. Poh, M.F. Chong / Bioresource Technology 100 (2009) 1–9 7

could be shortened by registering methane recovering POME treat-ment systems as a CDM project.

There is a tradeoff between performance and cost factor inanaerobic POME treatment. Therefore, plant optimization study isessential in order to obtain the most suitable operating parameterfor the treatment process and at the same time prove to be eco-nomically viable, providing the fastest payback period for the palmoil mill investment.

3. Factors influencing anaerobic digester performance

Many factors govern the performance of anaerobic digesterswhere adequate control is required to prevent reactor failure. Thefew major factors that greatly influence digester performances inPOME treatment are pH, mixing, operating temperature, nutrientsfor bacteria and organic loading rates into the digester.

3.1. pH

The microbial community in anaerobic digesters are sensitive topH changes and methanogens are affected to a greater extend (Les-lie Grady et al., 1999). An investigation by Beccari et al. (1996) con-firmed that methanogenesis is strongly affected by pH. As such,methanogenic activity will decrease when pH in the digester devi-ates from the optimum value. Optimum pH for most microbialgrowth is between 6.8 and 7.2 while pH lower than 4 and higherthan 9.5 are not tolerable (Gerardi, 2006).

Several cases of reactor failure reported in studies of wastewa-ter treatment are due to accumulation of high volatile fatty acidconcentration, causing a drop in pH which inhibited methanogen-esis (Parawira et al., 2006; Patel and Madamwar, 2002). Thus, vol-atile fatty acid concentration is an important parameter to monitorto guarantee reactor performance (Buyukkamaci and Filibeli,2004). It was found that digester could tolerate acetic acid concen-trations up to 4000 mg/l without inhibition of gas production (Staf-ford, 1982). To control the level of volatile fatty acid in the system,alkalinity has to be maintained by recirculation of treated effluent(Najafpour et al., 2006; Borja et al., 1996a) to the digester or addi-tion of lime and bicarbonate salt (Gerardi, 2003).

3.2. Mixing

Mixing provides good contact between microbes and sub-strates, reduces resistance to mass transfer, minimizes buildup ofinhibitory intermediates and stabilizes environmental conditions(Leslie Grady et al., 1999). When mixing is inefficient, overall rateof process will be impaired by pockets of material at differentstages of digestion whereby every stage has a different pH andtemperature (Stafford, 1982). Mixing can be accomplished throughmechanical mixing, biogas recirculation or through slurry recircu-lation (Karim et al., 2005a).

Investigations have been done to observe the effects of mixingto the performance of anaerobic digesters. It was found that mixingimproved the performance of digesters treating waste with higherconcentration (Karim et al., 2005b) while slurry recirculationshowed better results compared to impeller and biogas recircula-tion mixing mode (Karim et al., 2005c). Mixing also improved gasproduction as compared to unmixed digesters (Karim et al.,2005b). Intermittent mixing is advantageous over vigorous mixing(Kaparaju et al., 2008; Stafford, 1982), where this has been adoptedwidely in large-scale municipal and farm waste digesters (Stafford,1982). Rapid mixing is not encouraged as methanogens can be lessefficient in this mode of operation (Gerardi, 2003).

However, Karim et al. (2005b) mentioned that mixing duringstart-up is not beneficial due to the fact that digester pH will be

lowered, resulting in performance instability as well as leading toa prolonged start-up period. Mixing in palm oil mills which dependon biogas produced (Ma and Ong, 1985) are less efficient comparedto mechanical mixing as digesters are not perfectly mixed. Furtherinvestigation on effects of mixing on POME should be undertaken toobtain a suitable mode of mixing for the best digester performance.

3.3. Operating temperature

POME is discharged at temperatures around 80–90 �C (Zinati-zadeh et al., 2006) which actually makes treatment at both meso-philic and thermophilic temperatures feasible especially in tropicalcountries like Malaysia. Yet, anaerobic POME treatments in Malay-sia are conducted only in the mesophilic temperature range. Vari-ous studies have been conducted to investigate the feasibility ofoperating wastewater treatment systems in the thermophilic tem-perature range such as sugar, high-strength wastewater (Wiegantet al., 1985; Wiegant and Lettinga, 1985) and POME (Cail and Bar-ford, 1985; Choorit and Wisarnwan, 2007). These studies have re-ported successful system operation in the thermophilictemperature range, with POME treatment having treatment ratesmore than four times faster than operation in the mesophilic tem-perature range (Cail and Barford, 1985). Similarly, high productionof methane was also observed from the treatment of sugar waste-water in this higher temperature range.

Effect of temperature on the performance of anaerobic digestionwas investigated. Yu et al. (2002b) found that substrate degrada-tion rate and biogas production rate at 55 �C was higher than oper-ation at 37 �C. Studies have reported that thermophilic digestersare able to tolerate higher OLRs and operate at shorter HRT whileproducing more biogas (Ahn and Forster, 2002; Kim et al., 2006;Yilmaz et al., 2008). However, failure to control temperature in-crease can result in biomass washout (Lau and Fang, 1997) withaccumulation of volatile fatty acid due to inhibition of methano-genesis. At high temperatures, production of volatile fatty acid ishigher compared to mesophilic temperature range (Yu et al.,2002b). Many operators prefer to have digesters operating in mes-ophilic temperature due to better process stability. Nevertheless,investigation on digester stability by Kim et al. (2002) proved thatdisadvantages of thermophilic digesters can be resolved by keep-ing microbial consortia in close proximity.

A cost benefit analysis done on anaerobic POME treatment sys-tem with biogas recovery for heat generation and digester effluentfor land application indicated that operation in the thermophilicrange provide the fastest payback to investment. The cost benefitanalysis for POME treatment system that utilizes biogas for elec-tricity generation and digester effluent for land application alsoshowed a faster payback (Yeoh, 2004). Yeoh (2004) also stated thatif all POME in Malaysia is to be treated at thermophilic tempera-ture where recovered biogas is fully utilized for electricity energygeneration, it would generate 2250 million kWh which contributesapproximately 4% of national electricity demand in 1999. Thisshows the potential of operating POME treatment systems in ther-mophilic temperature.

3.4. Organic loading rates

Various studies have proven that higher OLRs will reduce CODremoval efficiency in wastewater treatment systems (Torkianet al., 2003; Sánchez et al., 2005; Patel and Madamwar, 2002).However, gas production will increase with OLR until a stage whenmethanogens could not work quick enough to convert acetic acidto methane. OLR is related to substrate concentration and HRT,thus a good balance between these two parameters has to beobtained for good digester operation. Short HRT will reduce thetime of contact between substrate and biomass.

8 P.E. Poh, M.F. Chong / Bioresource Technology 100 (2009) 1–9

4. Future development and conclusion

There is a great potential to anaerobically treat POME in thethermophilic temperature range. Further research can be done todevelop a thermophilic anaerobic bioreactor with minimal controlto ease system operation. Moreover, intensity of mixing in thethermophilic range should be investigated to obtain an optimummixing rate that will keep microbial consortia in close proximityand at the same time improve the system efficiency.

Seed sludge used in various anaerobic digestion studies is nor-mally a mixture of mesophilic and thermophilic microbial consor-tia. Therefore, it would be beneficial to isolate bacteria culture forspecific operation (i.e. thermophilic bacteria culture for use in bio-reactors operating in the thermophilic range) and also to study theperformance of the isolated culture on the system efficiency.

In conclusion, anaerobic digestion is an advantageous methodfor POME treatment as it generates valuable end product thatcan be exchanged into revenue when registered as a CDM project.Furthermore, operation costs can be reduced through utilization ofbiogas for heat or electricity energy generation in the plant. Thismethod is also able to treat effluent to a satisfactory quality for dis-charge at lower costs. With operation of anaerobic digestion inthermophilic temperature, POME treatment can be acceleratedand still produce good effluent quality.

References

Acharya, B.K., Mohana, S., Madamwar, D., 2008. Anaerobic treatment of distilleryspent wash – a study on upflow anaerobic fixed film bioreactor. BioresourceTechnology 99, 4621–4626.

Ahmad, A.L., Chong, M.F., Bhatia, S., Ismail, S., 2006. Drinking water reclamationfrom palm oil mill effluent (POME) using membrane technology. Desalination191, 35–44.

Ahmad, A.L., Chong, M.F., Bhatia, S., 2007. Mathematical modeling of multiplesolutes system for reverse osmosis process in palm oil mill effluent (POME)treatment. Chemical Engineering Journal 132, 183–193.

Ahn, J.-H., Forster, C.F., 2002. A comparison of mesophilic and thermophilicanaerobic upflow filters treating paper-pulp-liquors. Process Biochemistry 38,257–262.

Alvarado-Lassman, A., Rustrián, E., García-Alvarado, M.A., Rodríguez-Jiménez, G.C.,Houbron, E., 2008. Brewery wastewater treatment using anaerobic inversefluidized bed reactors. Bioresource Technology 99, 3009–3015.

Ayati, B., Ganjidoust, H., 2006. Comparing the efficiency of UAFF and UASB withhybrid reactor in treating wood fiber wastewater. Iranian Journal ofEnvironmental Health Science Engineering 3, 39–44.

Barbosa, R.A., Sant’Anna Jr., G.L., 1989. Treatment of raw domestic sewage in anUASB reactor. Water Research 23, 1483–1490.

Beccari, M., Bonemazzi, F., Majone, M., Riccardi, C., 1996. Interaction betweenacidogenesis and methanogenesis in the anaerobic treatment of olive oil milleffluents. Water Research 30, 183–189.

Behling, E., Diaz, A., Colina, G., Herrera, M., Gutierrez, E., Chacin, E., Fernandez, N.,Forster, C.F., 1997. Domestic wastewater treatment using UASB reactor.Bioresource Technology 61, 239–245.

Bello-Mendoza, R., Castillo-Rivera, M.F., 1998. Start-up of an anaerobic hybrid(UASB/filter) reactor treating wastewater from a coffee processing plant.Anaerobe 4, 219–225.

Bodkhe, S., 2008. Development of an improved anaerobic filter for municipalwastewater treatment. Bioresources Technology 99, 222–226.

Borja, R., Banks, C.J., 1994a. Anaerobic digestion of palm oil mill effluent using anup-flow anaerobic sludge blanket reactor. Biomass and Bioenergy 6, 381–389.

Borja, R., Banks, C.J., 1994b. Treatment of palm oil mill effluent by upflow anaerobicfiltration. Journal of Chemical Technology and Biotechnology 61,103–109.

Borja, R., Banks, C.J., 1994c. Anaerobic digestion of palm oil mill effluent using anup-flow anaerobic sludge blanket reactor. Biomass and Bioenergy 6, 381–389.

Borja, R., Banks, C.J., 1995a. Response of an anaerobic fluidized bed reactor treatingice-cream wastewater to organic, hydraulic, temperature and pH shocks.Journal of Biotechnology 39, 251–259.

Borja, R., Banks, C.J., 1995b. Comparison of an anaerobic filter and an anaerobicfluidized bed reactor treating palm oil mill effluent. Process Biochemistry 30,511–521.

Borja, R., Banks, C.J., Wang, Z., 1995c. Performance of a hybrid anaerobic reactor,combining a sludge blanket and a filter, treating slaughterhouse wastewater.Applied Microbiology and Biotechnology 43, 351–357.

Borja, R., Banks, C.J., Sánchez, E., 1996a. Anaerobic treatment of palm oil milleffluent in a two-stage up-flow anaerobic sludge blanket (UASB) reactor.Journal of Biotechnology 45, 125–135.

Borja, R., Alba, J., Banks, C.J., 1996b. Anaerobic digestion of wash waters derivedfrom the purification of virgin olive oil using a hybrid reactor combining a filterand a sludge blanket. Process Biochemistry 31, 219–224.

Borja, R., Banks, C.J., Wang, Z., Mancha, A., 1998. Anaerobic digestion ofslaughterhouse wastewater using a combination sludge blanket and filterarrangement in a single reactor. Bioresource Technology 65, 125–133.

Borja, R., González, E., Raposo, F., Millán, F., Martín, A., 2001. Performance evaluationof a mesophilic anaerobic fluidized-bed reactor treating wastewater derivedfrom the production of proteins from extracted sunflower flour. BioresourceTechnology 76, 45–52.

Buyukkamaci, N., Filibeli, A., 2004. Volatile fatty acid formation in anaerobic hybridreactor. Process Biochemistry 39, 1491–1494.

Cail, R.G., Barford, J.P., 1985. Thermophilic semi-continuous anaerobic digestion ofpalm-oil mill effluent. Agricultural Wastes 13, 295–304.

CDM, 2007. Project 0867: Kim Loong Methane Recovery for Onsite UtilizationProject at Kota Tinggi, Johor, Malaysia. [Online] Available: <http://cdm.unfccc.int/Projects/DB/TUEV-SUED1169205863.92/view> (29/2/2008).

CDM, 2007. Project 1153: Methane recovery and utilization project at UnitedPlantations Berhad, Jendarata Palm Oil Mill Malaysia. [Online] Available: <http://cdm.unfccc.int/Projects/DB/DNV-CUK1181122330.1/view> (29/2/2008).

Chan, K.S., Chooi, C.F., 1984. Ponding System for palm oil mill effluent treatment. In:Proceedings of the Regional Workshop on Palm Oil Mill Technology & EffluentTreatment, pp. 185–192.

Chen, T.-H., Shyu, W.-H., 1996. Performance of four types of anaerobic reactors intreating very dilute dairy wastewater. Biomass and Bioenergy 11, 431–440.

Chin, K.K., 1981. Anaerobic treatment kinetics of palm oil sludge. Water Research15, 199–202.

Chong, M.F., 2007. Modeling, simulation and design of membrane based palm oilmill effluent (POME) treatment plant from pilot plant studies. Ph.D. Thesis.Universiti Sains Malaysia.

Choorit, W., Wisarnwan, P., 2007. Effect of temperature on the anaerobic digestionof palm oil mill effluent. Electronic Journal of Biotechnology 10, 376–385.

Córdoba, P.R., Francese, A.P., Sineriz, F., 1995. Improved performance of a hybriddesign over an anaerobic filter for the treatment of dairy industry wastewater atlaboratory scale. Journal of Fermentation and Bioengineering 79, 270–272.

Data for Engineers: POME, 2004. Palm Oil Engineering Bulletin 71, pp. 34–35.Dinsdale, R.M., Hawkes, F.R., Hawkes, D.L., 1997. Comparison of mesophilic and

thermophilic upflow anaerobic sludge blanket reactors treating instant coffeeproduction wastewater. Water Research 31, 163–169.

Doble, M., Kumar, A., 2005. Biotreatment of Industrial Effluents. ElsevierButterworth-Heinemann, Oxford, United Kingdom. pp.19-38.

Fang, H.H.P., Chui, H.K., 1994. Comparison of startup performance of four anaerobicreactors for the treatment of high-strength wastewater. Resources,Conservation and Recycling 11, 123–138.

Farhadian, M., Borghei, M., Umrania, V.V., 2007. Treatment of beet sugar water byUAFB bioprocess. Bioresource Technology 98, 3080–3083.

Federal Subsidiary Legislation – Environmental Quality Act 1974 [ACT 127],Environmental Quality (Sewage and Industrial Effluents) Regulation 1979.[Online] Available: <http://www.doe.gov.my> (29/2/2008).

Forster, C.F., Wase, D.A.J., 1983. Anaerobic treatment of dilute wastewaters usingan upflow sludge blanket reactor. Environmental Pollution (Series A) 31, 57–66.

Garcia-Calderon, D., Buffiere, P., Moeltta, R., Elmaleh, S., 1998. Anaerobic digestionof wine distillery wastewater in down-flow fluidized bed. Water Research 32,3593–3600.

Gangagni Rao, A., Venkata Naidu, G., Krishna Prasad, K., Chandrasekhar Rao, N.,Venkata Mohan, N., Jetty, A., Sarma, P.N., 2005. Anaerobic treatment ofwastewater with high suspended solids from a bulk drug industry using fixedfilm reactor (AFFR). Bioresource Technology 96, 87–93.

Gerardi, M.H., 2003. The Microbiology of Anaerobic Digesters. Wiley-Interscience,New Jersey. pp. 51-57.

Gerardi, M.H., 2006. Wastewater Bacteria. Wiley-Interscience, New Jersey. pp. 19–31.

Goodwin, J.A.S., Wase, D.A.J., Forster, C.F., 1992. Pre-granulated seeds for UASBreactors: how necessary are they? Bioresource Technology 41, 71–79.

Gopal, J., Ma, A.N., 1986. The comparative economics of palm oil mill effluenttreatment and resource recovery systems. National Workshop on RecentDevelopments in Palm Oil Milling Technology & Pollution Control.

Guiot, S.R., van den Berg, L., 1985. Performance of an upflow anaerobic reactorcombining a sludge blanket and a filter treating sugar waste. Biotechnology andBioengineering 27, 800–806.

Hamdi, M., Garcia, J.L., 1991. Comparison between anaerobic filter and anaerobiccontact process for fermented olive mill wastewaters. Bioresource Technology38, 23–29.

Hawkes, F.R., Donnelly, T., Anderson, G.K., 1995. Comparative performance ofanaerobic digesters operating on ice-cream wastewater. Water Research 29,525–533.

Ibrahim, A., Yeoh, B.G., Cheah, S.C., Ma, A.N., Ahmad, S., Chew, T.Y., Raj, R., Wahid,M.J.A., 1984. Thermophilic anaerobic contact digestion of palm oil mill effluent.Water Science and Technology 17, 155–165.

Jawed, M., Tare, V., 2000. Post-mortem examination and analysis of anaerobicfilters. Bioresources Technology 72, 75–84.

Kalyuzhnyi, S.V., Skylar, V.I., Davlyatshina, M.A., Parshina, S.N., Simankova, M.V.,Kostrikina, N.A., Nozhevnikova, A.N., 1996. Organic removal andmicrobiological features of UASB-reactor under various organic loading rates.Bioresource Technology 55, 47–54.

P.E. Poh, M.F. Chong / Bioresource Technology 100 (2009) 1–9 9

Kalyuzhnyi, S., de los Santos, L.E., Martinez, J.R., 1998. Anaerobic treatment of rawand preclarified potato-maize wastewater in a UASB reactor. BioresourceTechnology 66, 198–199.

Kaparaju, P., Buendia, I., Ellegaard, L., Angelidakia, I., 2008. Effects of mixingon methane production during thermophilic anaerobic digestion of manure:lab-scale and pilot-scale studies. Bioresource Technology 99, 4919–4928.

Karim, K., Klasson, K.T., Hoffmann, R., Drescher, S.R., DePaoli, D.W., Al-Dahhan, M.H.,2005a. Anaerobic digestion of animal waste: effect of mixing. BioresourceTechnology 96, 1607–1612.

Karim, K., Hoffmann, R., Klasson, K.T., Al-Dahhan, M.H., 2005b. Anaerobic digestionof animal waste: effect of mode of mixing. Water Research 39,3597–3606.

Karim, K., Hoffmann, R., Klasson, T., Al-Dahhan, M.H., 2005c. Anaerobic digestion ofanimal waste: waste strength versus impact of mixing. Bioresource Technology96, 1771–1781.

Kim, J.K., Oh, B.R., Chun, Y.N., Kim, S.W., 2006. Effects of temperature and hydraulicretention time on anaerobic digestion of food waste. Journal of Bioscience andBioengineering 102, 328–332.

Kim, M., Ahn, Y.-H., Speece, R.E., 2002. Comparative process stability and efficiencyof anaerobic digestion; mesophilic vs. thermophilic. Water Research 36, 4369–4385.

Lau, I.W.C., Fang, H.H.P., 1997. Effect of temperature shock to thermophilic granules.Water Research 31, 2626–2632.

Leal, K., Chachin, E., Gutierez, E., Fernandez, N., Forster, C.F., 1998. A mesophilicdigestion of brewery wastewater in an unheated anaerobic filter. BioresourceTechnology 65, 51–55.

Leslie Grady Jr., C.P., Daigger, G.T., Lim, H.C., 1999. Biological WastewaterTreatment, second ed. CRC Press. Revised & Expanded.

Lettinga, G., van Velson, A.F.M., Hobma, S.W., de Zeeuw, W., Klapwijk, A., 1980. Useof the upflow sludge blanket (USB) reactor concept for biological wastewatertreatment, especially for anaerobic treatment. Biotechnology andBioengineering 22, 699–734.

Lettinga, G., 1995. Anaerobic digestion and wastewater treatment systems. Antonievan Leeuwenhoek 67, 3–28.

Liu, W.-T., Chan, O.-C., Fang, H.H.P., 2002. Microbial community dynamics duringstart-up of acidogenic anaerobic reactors. Water Research 36,3203–3210.

Lo, K.V., Liao, P.H., Gao, Y.C., 1994. Anaerobic treatment of swine wastewater usinghybrid UASB reactor. Bioresource Technology 47, 153–157.

Ma, A.N., Ong, A.S.H., 1985. Anaerobic digestion of palm oil mill. PORIM Bulletin 4,35–45.

Ma, A.N., Cheah, S.C., Chow, M.C., 1993. Current status of palm oil processing wastesmanagement. In: Waste Management in Malaysia: Current Status and Prospectsfor Bioremediation, pp. 111–136.

Ma, A.N., Tajima, Y., Asahi, M., Hanif, J., 1997. Effluent treatment – evaporationmethod. PORIM Engineering News 44, 7–8.

McHugh, S., O’Reilly, C., Mahony, T., Colleran, E., O’Flaherty, V., 2003. Anaerobicgranular sludge bioreactor technology. Reviews in Environmental Science andBiotechnology 2, 225–245.

Menon, R., 2002. Carbon credits and clean development mechanism. Palm OilEngineering Bulletin 65, 11–14.

Menon, R.N., 2007. Dialogue session with the palm oil industry and stakeholders.Palm Oil Engineering Bulletin 83, 11–14.

Metcalf, Eddy, 2003. Wastewater Engineering Treatment and Reuse, fourth ed.McGraw Hill, pp. 96–97.

Mohan, S., Sunny, N., 2008. Study of biomethanization of waste water from jamindustries. Bioresource Technology 99, 210–213.

Monroy, O., Johnson, K.A., Wheatley, A.D., Hawkes, F., Caine, M., 1994. The anaerobicfiltration of dairy waste: results of a pilot trial. Bioresource Technology 50, 243–251.

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 Biochemistry 41,370–379.

Nebot, E., Romero, L.I., Quiroga, J.M., Sales, D., 1995. Effect of the feed frequency onthe performance of anaerobic filters. Anaerobe 1, 113–120.

Parawira, W., Murto, M., Zvauya, R., Mattiasson, B., 2006. Comparative performanceof a UASB reactor and an anaerobic packed-bed reactor when treating potatowaste leachate. Renewable Energy 31, 893–903.

Patel, H., Madamwar, D., 2002. Effects of temperature and organic loading rates onbiomethanation of acidic petrochemical wastewater using an anaerobic upflowfixed-film reactor. Bioresource Technology 82, 65–71.

Pérez, M., Romero, L.I., Sales, D., 1998. Comparative performance of high rateanaerobic thermophilic technologies treating industrial wastewater. WaterResearch 32, 559–564.

Perez, M., Rodriguez-Cano, R., Romero, L.I., Sales, D., 2007. Performance of anaerobicthermophilic fluidized bed in the treatment of cutting-oil wastewater.Bioresource Technology 98, 3456–3463.

Ramasamy, E.V., Abbasi, S.A., 2000. Energy recovery from dairy waste-waters:impacts of biofilm support systems on anaerobic CST reactors. Applied Energy65, 91–98.

Reyes, O., Sánchez, E., Rovirosa, N., Borja, R., Cruz, M., Colmenarejo, M.F., Escobedo,R., Ruiz, M., Rodríguez, X., Correa, O., 1999. Low-strength wastewater treatmentby a multistage anaerobic filter packed with waste tyre rubber. BioresourceTechnology 70, 55–60.

Ruiz, I., Veiga, M.C., de Santiago, P., Blázquez, R., 1997. Treatment of slaughterhousewastewater in a UASB reactor and an anaerobic filter. Bioresource Technology60, 251–258.

Russo, C., Sant’ Anna Jr., G.L., de Carvalho Pereira, S.E., 1985. An anaerobic filterapplied to the treatment of distillery wastewaters. Agricultural Wastes 14, 301–313.

Sánchez, E., Borja, R., Travieso, L., Martín, A., Colmenarejo, M.F., 2005. Effect oforganic loading rate on the stability, operational parameters and performanceof a secondary upflow anaerobic sludge bed reactor treating piggery waste.Bioresource Technology 96, 335–344.

Saravanane, R., Murthy, D.V.S., Krishnaiah, K., 2001. Treatment of anti-osmotic drugbased pharmaceutical effluent in an upflow anaerobic fluidized bed system.Waste Management 21, 563–568.

Sayed, S., de Zeeuw, W., Lettinga, G., 1984. Anaerobic treatment of slaughterhousewaste using a flocculant sludge UASB reactor. Agricultural Wastes 11, 197–226.

S�en, S., Demirer, G.N., 2003. Anaerobic treatment of real textile wastewater with afluidized bed reactor. Water Research 37, 1868–1878.

Show, K.-Y., Tay, J.-H., 1999. Influence of support media on biomass growth andretention in anaerobic filters. Water Research 33, 1471–1481.

Sowmeyan, R., Swaminathan, G., 2008. Evaluation of inverse anaerobic fluidized bedreactor for treating high strength organic wastewater. Bioresource Technology99, 3877–3880.

Stafford, D.A., 1982. The effects of mixing and volatile fatty acid concentrations onanaerobic digester performance. Biomass 2, 43–55.

Stronach, S.M., Rudd, T., Lester, J.N., 1987. Start-up of anaerobic bioreactors on highstrength industrial wastes. Biomass 13, 173–197.

Tay, J.H., 1991. Complete reclamation of oil palm wastes. Resources Conservationand Recycling 5, 383–392.

Toldrá, F., Flors, A., Lequerica, J.L., Vallés, S., 1987. Fluidized bed anaerobicbiodegradation of food industry wastewaters. Biological Wastes 21, 55–61.

Tong, S.L., Jaafar, A.B., 2006. POME Biogas capture, upgrading and utilization. PalmOil Engineering Bulletin 78, 11–17.

Torkian, A., Eqbali, A., Hashemian, S.J., 2003. The effect of organic loading rate on theperformance of UASB reactor treating slaughterhouse effluent. ResourcesConservation & Recycling 40, 1–11.

Ugoji, E.O., 1997. Anaerobic digestion of palm oil mill effluent and its utilizationas fertilizer for environmental protection. Renewable Energy 10, 291–294.

Ulrich, G.D., Vasudevan, P.T., 2004. Chemical Engineering Process Design andEconomics: A Practical Guide, second ed. Process Publishing Company.

Van Der Merwe, M., Britz, T.J., 1993. Anaerobic digestion of baker’s yeast factoryeffluent using an anaerobic filter and hybrid digester. Bioresource Technology43, 169–174.

Vázquez, I., Rodríguez, J., Marañón, E., Castrillón, L., Fernández, Y., 2006.Simultaneous removal of phenol, ammonium and thiocyanate from cokewastewater by aerobic degradation. Journal of Hazardous Materials B 137,1773–1780.

Vijayaraghavan, K., Ahmad, D., Abdul Aziz, M.E., 2007. Aerobic treatment of palm oilmill effluent. Journal of Environmental Management 82, 24–31.

Vlissidis, A., Zouboulis, A.I., 1993. Thermophilic anaerobic digestion of alcoholdistillery wastewaters. Bioresource Technology 43, 131–140.

Wang, Z., Banks, C.J., 2007. Treatment of a high-strength sulphate-rich alkalineleachate using an anaerobic filter. Waste Management 27, 359–366.

Wiegant, W.M., Lettinga, G., 1985. Thermophilic anaerobic digestion of sugars inupflow anaerobic sludge blanket reactors. Biotechnology and Bioengineering27, 1603–1607.

Wiegant, W.M., Classen, J.A., Lettinga, G., 1985. Thermophilic anaerobic digestion ofhigh strength wastewaters. Biotechnology and Bioengineering 27, 1374–1381.

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, 1575–1581.

Yacob, S., Hassan, M.A., Shirai, Y., Wakisaka, M., Subash, S., 2006a. Baseline study ofmethane emission from anaerobic ponds of palm oil mill effluent treatment.Science of the Total Environment 366, 187–196.

Yacob, S., Shirai, Y., Hassan, M.A., Wakisaka, M., Subash, S., 2006b. Start-upoperation of semi-commercial closed anaerobic digester for palm oil milleffluent treatment. Process Biochemistry 41, 962–964.

Yeoh, B.G., 2004. A technical and economic analysis of heat and power generationfrom biomethanation of palm oil mill effluent. Electric Supply Industry inTransition: Issues and Prospect for Asia, 20–60.

Yilmaz, T., Yuceer, A., Basibuyuk, M., 2008. A comparison of the performance ofmesophilic and thermophilic anaerobic filters treating papermill wastewater.Bioresource Technology 99, 156–163.

Yu, H., Gu, G., 1996. Biomethanation of brewery wastewater using an anaerobicupflow blanket filter. Journal of Cleaner Production 4, 219–223.

Yu, H.-Q., Hu, Z.-H., Hong, T.-Q., Gu, G.-W., 2002a. Performance of an anaerobic filtertreating soybean processing wastewater with and without effluent recycle.Process Biochemistry 38, 507–513.

Yu, H.-Q., Fang, H.H.P., Gu, G.-W., 2002b. Comparative performance of mesophilicand thermophilic acidogenic upflow reactors. Process Biochemistry 38, 447–454.

Zinatizadeh, A.A.L., Mohamed, A.R., Abdullah, A.Z., Mashitah, M.D., Hasnain Isa, M.,Najafpour, G.D., 2006. Process modeling and analysis of palm oil mill effluenttreatment in an up-flow anaerobic sludge fixed film bioreactor using responsesurface methodology (RSM). Water Research 40, 3193–3208.