RSC Advances

PAPER

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article OnlineView Journal | View Issue

Suspended grow

aSchool of Industrial Technology, Universit

Pinang, Malaysia. E-mail: ttteng@usm.my; FbSchool of Environmental Engineering, Un

Pengajian Jejawi 3, 02600 Arau, Perlis, Mal

Cite this: RSC Adv., 2014, 4, 64659

Received 11th August 2014Accepted 18th November 2014

DOI: 10.1039/c4ra08483g

www.rsc.org/advances

This journal is © The Royal Society of C

th kinetic analysis on biogasgeneration from newly isolated anaerobic bacterialcommunities for palm oil mill effluent at mesophilictemperature

Yee-Shian Wong,ab Tjoon Tow Teng,*a Soon-An Ong,b Norhashimah Morada

and Mohd Rafatullaha

The anaerobic degradation of palm oil mill effluent (POME) was carried out undermesophilic temperature in

an anaerobic suspended growth closed bioreactor (ASGCB). Monod model was applied to describe the

kinetic analysis of POME at different organic loading rates (OLR) in the range of 2.75–8.2 g TCOD per L

per day. The hydraulic retention time (HRT) ranged between 8 and 24 days. The TCOD removal

efficiency achieved was between 89.66% and 79.83%. The evaluated kinetic coefficients were: growth

yield, YG (0.357 g VSS per g TCOD), specific biomass decay rate, b (0.07 per day), maximum specific

biomass growth rate, mmax (0.27 per day), saturation constant for substrate, Ks (25.03 g TCOD per L),

critical retention time, QC (3.72 day) and methane yield, YCH4(0.34 L CH4 per TCODremoved). Besides, new

fermentative anaerobic bacteria isolated from POME were identified as Escherichia fergusonii,

Enterobacter asburiae, Enterobacter cloacae, Desulfovibrio aerotolerans, Desulfobulbus propionicus,

Fusobacterium nucleatum, Paenibacillus pabuli, Bacillus subtilis, Methanobacterium sp., Methanosaeta

concilii, Methanofollis tationis, Methanosarcina mazei and Methanosarcina acetivorans using 16S rDNA.

1. Introduction

The palm oil industry produces high strength organic waste-water known as Palm Oil Mill Effluent (POME). POME isa viscous, brownish liquid containing about 95–96% water,0.6–0.7% oil and 4–5% total solids (including 2–4% SS, mainlydebris from fruit). It is acidic (pH 4–5), hot (80–90 �C)with averageChemical Oxygen Demand (COD) and Biochemical OxygenDemand (BOD) values of 50 000 mg L�1 and 25 000 mg L�1,respectively.1

Anaerobic digestion has oen been used for high strengthorganic wastewater, since it produces less sludge compared toconventional aerobic treatment. An advantage of anaerobictreatment is operational cost effectiveness with simultaneousproduction of methane or hydrogen gas as energy resource.Anaerobic treatment is a complex process which involvesdecomposition of organic compounds and production ofmethane, hydrogen and carbon dioxide gases in the absence ofmolecular oxygen. The degradation process takes place bydifferent types of anaerobic bacteria. The degradation mecha-nisms involve hydrolysis, acidogenesis (including acetogenesis)

i Sains Malaysia, 11600 Gelugor, Pulau

ax: +60-4-657-3678; Tel: +60-4-653-2215

iversiti Malaysia Perlis, Kompleks Pusat

aysia

hemistry 2014

and methanogenesis.2 Hydrolysis is the conversion reaction inwhich complex molecules such as carbohydrate, lipids andprotein are converted into sugar, organic acids and etc. Acido-genic bacteria are responsible for breaking down the sugar, fattyacids and amino acids in acidogenesis process. Methano-genesis process occurs in which hydrogenotropic methanogensutilize the hydrogen and carbon dioxide gases produced byacetoclastic methanogens from short chain fatty acids toproduce benecial end product of methane gas.3

For better control of degradation process with high opera-tional efficiency, identication of microbial community ofanaerobic digestion process is an essential requirement.4

Currently only few anaerobic bacteria such as Clostridium ssp.,Streptococcus cohnii ssp., Lactobacillus and Thermoanaer-obacterium spp. have been reported with hydrogen producingability from palm oil wastewater.5–7 However, hitherto no workregarding anaerobic bacteria with methane producing abilityfrom palm oil wastewater has been reported. Kinetic study isrequired for a better understanding of the underlyingphenomena in the anaerobic digestion process. Among manyimportant applications, it can predict the compounds producedor consumed on their corresponding rates. The process kineticsfor the anaerobic digestion are composed of several aspects.Firstly, there are the mass balance models involving input andoutput streams of the anaerobic system. Secondly, the selectedmodel of ow or hydraulic must be formed for the anaerobic

RSC Adv., 2014, 4, 64659–64667 | 64659

RSC Advances Paper

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article Online

process. Finally, the selection of kinetic model for the reactionand stoichiometry can be conducted to determine the kineticcoefficients. Numerous kinetic models for anaerobic digestionsuch as Monod, rst-order, second-order, modied Stover–Kincannon, Contois, Chen and Hashimoto have beenproposed.8

The literature study is abound with results of research onattached growth anaerobic treatment such as immobilised cellbioreactor,9 two stage up-ow anaerobic sludge blanket(UASB),10 high rate up-ow anaerobic sludge xed lm,11 andcombined high-rate anaerobic reactors12 for the treatment ofPOME. Little research has been conducted on the application ofanaerobic suspended growth process and also identication ofanaerobe bacteria from POME for better control of the degra-dation process. Therefore, the aim of the present work wasfocused on the identication of anaerobic bacteria and kineticcoefficients for the anaerobic digestion of POME using sus-pended growth closed bioreactor.

2. Materials and methods2.1 Wastewater preparation

The raw POME was collected by onsite sampling fromMALPOMIndustries Bhd, in Nibong Tebal, Penang, Malaysia. The seedingrequired to acclimatize the Anaerobic Suspended GrowthClosed Bioreactor (ASGCB) was taken from the anaerobic pondof MALPOM Industries Bhd wastewater treatment plant. Thecomposition and properties of the collected POME used in thepresent study are summarized in Table 1. The POME was storedin the refrigerator at 4 �C until further experimental work wasconducted. This storage observed no change of composition.

Table 1 Composition and properties of POME used

ParameteraConcentration range(mg L�1) Mean

pH 4.3–4.75 4.5COD 62 500–65 000 63 750SCOD 30 500–34 500 32 500BOD 31 500–38 500 35 000MLSS 35 050–41 090 38 070MLVSS 30 230–34 700 32 465O & G 15 500–16 600 16 050TN 725–835 780NH3-N 65–70 67.5Palmitic acid 7420–7500 7460Acetic acid 1900–2110 2005Propanoic acid 53.2–56 54.6Butanoic acid 28.6–33.4 31iso-Butanoic acid 2.8–5.8 4.3Carbon (%) 35.8–9.5 47.7Hydrogen (%) 5.3–9.1 7.2Nitrogen (%) 0.9–2.7 1.8Sulfur (%) 0.5–1.3 0.9Oxygen (%) 21–55 38

a Unit for all parameter is mg L�1 except pH, carbon, hydrogen,nitrogen, sulfur and oxygen.

64660 | RSC Adv., 2014, 4, 64659–64667

2.2 ASGCB set-up

The schematic conguration of the ASGCB set-up is shown inFig. 1. The ASGCB is 0.25 m in diameter and 0.36 m in height.The reactor consists of a cylindrical-shape exi glass vessel withtotal and working volumes of 17.7 L and 14 L, respectively. Itcomprises an integrated on-line pH data recording systemconnected to a pH probe and an overhead stirrer. The operatingtemperature of ASGCB was maintained constant at 35 �C usingthermal jacket. The ASGCB has a gas sampling valve (or clamp)which allows gas samples to be collected without interferencewith head space composition and is connected to a 50 L gascollection bag.

2.3 Bacteria cultivation and isolation

Cultivation of bacteria was conducted according to isolationtechniques of microbiology.13 The culture medium contained(in g L�1): heart extract and peptone 20; yeast extract 5.0;sodium chloride 5.0; bacteriological agar 15; sodium thio-glycollate 2.0 and sodium formaldehyde sulfoxylate 1.0. The pHof the medium was adjusted to pH 7.0 � 0.1 with 1 N sodiumhydroxide (NaOH) or 0.1 N of hydrochloric acid (HCl) by usingpH meter (Metrohm, 826 pH Mobile). The medium was steril-ized at 121 �C for 15 minutes. Isolation of bacteria in the samplewas done with serial dilution where one mL of the sample wasdiluted with 9 mL of sterile saline water. The dilution wasrepeated until the colonies on the agar plate was countable.Pure isolates were then stored on culture medium at 4 �C untilfurther analysis. The cultivation and isolation were done in theanaerobic chamber (Fisher Scientic, Forma Anaerobic System)at 35 �C.

2.4 Molecular identication of isolates

2.4.1 DNA extraction and PCR amplication. The bacteriagenomic DNA was extracted by using I-genomic DNA extractionkit (QIAGEN) according to the manufacturer's instruction from

Fig. 1 Schematic configuration of the Anaerobic Suspended GrowthClosed Bioreactor (ASGCB).

This journal is © The Royal Society of Chemistry 2014

Paper RSC Advances

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article Online

the cultivated medium. Then, PCR amplication of 16S rDNAwas conducted using the amplication kit (PROMEGA)according to the manufacturer's operation procedure. Theamplication reaction was done in a 50 mL tube withmixtures of10 mL of 5X colorless GoTaq buffer, 1 mM of MgCl2, 0.1 mM ofdNTP, 0.5 mM forward primer, 0.5 mM reverse primer, 0.02 unitsof Taq DNA polymerase and nuclease-free water. Two types ofprimer pair were applied by using universal primer of 27F and1492R (27F: AGA GTT TGA TCM TGG CTC AG; 1492R: CGG TTACCT TGT TAC GAC TT) and archaea primer of ARC344F andARC915R (ARC344F: ACG GGG YGC AGC AGG GGC GA;ARC915R: GTG CTC CCC CGC CAA TTC CT). The PCR ampli-cation was 2 min at 95 �C followed by 34 cycles of denaturationat 95 �C for 50 s, annealing at 58 �C for 45 s, elongation at 72 �Cat 45 s and the nal elongation at 72 �C for 8 minutes. The PCRamplication procedures for archaea were set similar to thatof bacteria with only difference in the annealing temperature at54 �C for 1 min. The PCR products were then analyzed with 1%agarose gel electrophoresis (AGE).

2.4.2 Bacteria identication based on 16S rDNA sequences.The PCR products were sent to the Genomics BioScience andTechnology Co., Ltd. in Malaysia for 16S rDNA sequence anal-ysis. The clone libraries were nally built based on the obtainedsequences. The obtained sequences were analyzed through thegene bank database of National Center for BiotechnologyInformation (NCBI) for highest similarity identication ofbacteria. The maximum bacteria identity was determinedthrough the Basic Local Alignment Search Tool (BLAST)between the similarity identication of aligned sequencenucleotides and nucleotides stored in the gene bank database.

2.5 ASGCB experimental procedure

The seeding required for starting the ASGCB was taken from theanaerobic pond of MALPOM Industries Bhd Penang, Malaysiawastewater treatment plant. About 14 liters of the anaerobicdigested POME were used to acclimatize the laboratory ASGCB.The start-up of the ASGCB involved step increases of theinuent organic volumetric loading rates from 2.72 g COD per Lper day to 6.55 g COD per L per day. This acclimatization phasewas important to allow the microorganisms present in themixed liquor perfectly acclimatize to the new environment andreach a steady state condition before proceeding to the nextphase of research studies. The pH in the ASGCB was adjusted to7.50 � 0.05 using 1 N sodium hydroxide (NaOH).

Once the start-up phase has been completed, a series ofcontinuous experiments using feed ow-rates of 0.58, 0.7, 0.88,1.17 and 1.75 liters of raw POME per day, corresponding toHydraulic Retention Time (HRT) of 24, 20, 16, 12 and 8 days wasfollowed. The rate of inuent substrate concentration of TCODwas controlled in the range of 62 500–65 500 mg L�1. Biogasand samples from ASGCB were collected for analysis aer 24hours of input raw POME. For each batch of HRT, the ASGCBwas continuously operated until steady state condition wasachieved. The steady-state condition was reached once thevalues of TCOD removal efficiency, VFA, biogas production rateand biogas composition remained unchanged for ve

This journal is © The Royal Society of Chemistry 2014

consecutive days. The variation of the actual results from themean values was <3%.

2.6 Analytical method

The TCOD of a sample was measured using reactor digestionmethod (HACH, range: 20–1500mg L�1) for two hours at 150 �C.The MLSS, MLVSS, O & G, NH3-N, VFA and Alk were determinedaccording to the Standard Methods for the Examination ofWater and Wastewater (APHA, 2005). Daily biogas productionwas recorded using Binder COMBIMASS gas analyzer consistedof ve gas channels: CO2, CH4, O2, H2S and H2. The volatile fattyacids such as acetic, propanoic, butanoic and iso-butanoic wereextracted from the sample using 10% phosphoric acid (H3PO4)and methanol. The extracted samples were analyzed by Shi-madzu GCMS QP 2010 Plus equipped with SGE BP21 capillarycolumn (25 m length, 0.22 mm internal diameter and 0.25 umthickness). Helium was used as the carrier gas. The injectiontemperature was 220 �C and the column temperature wasinitially held at 120 �C for 6 minutes then was raised progres-sively at 15 �C minute�1 until 220 �C. The carbon, hydrogen,nitrogen, sulfur and oxygen contents of POME wastewater weremeasured by elemental analyzer (PerkinElmer 2400 Series IICHNS/O).

2.7 Kinetic model development

2.7.1 Substrate balance model. Determination of kineticcoefficients can be done on the specic parameters of thesystem performance. Different kinetic models were used fordifferent biological systems to evaluate kinetic coefficients. Forthis purpose, various kinetic models such as Monod, Stover–Kincannon, Michaelis–Menten, rst order and second orderhave been applied. In the present study, the mass balance,reaction rate equation and Monod models were used. Theequations describing the performance of the system are themass balance equations of both the substrate and biomass.

2.7.2 Mass balance equation. The substrate balance can beexpressed as (1):

dS

dt¼ QðS1 � S2Þ

V� rv (1)

where, S ¼ substrate concentration in the reactor, Q ¼ volu-metric ow rate, S1 ¼ inuent substrate concentration, S2 ¼effluent substrate concentration (S2 ¼ S), V ¼ working volume,rv ¼ substrate utilization rate per volume, and t ¼ time.

Once the reactor was completely mixed and continuously fed

without recycle,dSdt

¼ 0: Eqn (1) then becomes:

rv ¼ (S1 � S2)D (2)

where, D ¼ dilution rate ¼ QV¼ 1

HRT:

The specic substrate utilization rate, rX is dened as rX ¼ rvX,

where X ¼ biomass concentration in the reactor:

rX ¼ ðS1 � S2ÞDX

(3)

RSC Adv., 2014, 4, 64659–64667 | 64661

RSC Advances Paper

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article Online

On the other hand, the biomass balance gives:

dX

dt¼ �QX

Vþ mX (4)

where, m¼ specic biomass growth rate. At steady state,dXdt

¼ 0:Hence, eqn (4) yields

m ¼ D ¼ 1

HRT(5)

The growth of microorganism is proportional to thesubstrate utilization. Biomass decay or endogenousmetabolismshould also be taken into account to describe the specicgrowth rate:11

m ¼ YGrX � b (6)

where, YG ¼ growth yield coefficient and b ¼ specic biomassdecay coefficient.

Combining eqn (5) and (6) gives:

D ¼ YGrX � b (7)

or

rX ¼ D

YG

þ b

YG

(8)

2.7.3 Reaction rate equation. Several types of reaction rateequations have been proposed to describe the relationshipbetween the growth rate and the substrate utilization. Monodequation is the simplest and well known in biological treatmentsystem.10 The Monod equation is:

rX ¼ S

Ks þ SrX ; max (9)

where, rX,max ¼ maximum specic substrate utilization rate,Ks ¼ saturation constant for substrate, and S ¼ substrateconcentration in the reactor. rX is half of rX,max when thesubstrate concentration is Ks. Eqn (9) can be rearranged to thefollowing three types of linear equation to obtain rX,max and Ks

from experimental data of rX and S.

1

rX¼

�Ks

rX ; max

��1

S

�þ 1

rX ; max

Lineweaver� Burk plot (10)

rX ¼ rX ; max � Ks

�rXS

�Eadie plot (11)

S

rX¼ S

rX ; max

þ Ks

rX ; max

Hofstee plot (12)

The maximum specic biomass growth rate, mmax, isobtained by substituting rX,max for rX in eqn (6):

mmax ¼ YGrX,max � b (13)

Critical retention time QC is a safety factor for anaerobicdegradation process where substrate utilization does not

64662 | RSC Adv., 2014, 4, 64659–64667

happen but bacteria washout could happen if the reactor isloaded below the critical retention time.10 QC can be expressedas:

QC ¼ 1

mmax

(14)

2.7.4 Methane yield. Methane yield is a vital economicfactor for the anaerobic digestion process. In most cases, themethane yield coefficient, YCH4

is determined from experi-mental data. If the volume of methane gas produced, VCH4

isassumed to be proportional to the amount of substrateconsumed, then eqn (15) is formed:

VCH4¼ YCH4

Q(S1 � S2) (15)

A plot of VCH4versus Q(S1 � S2) should yield a straight line

passing through the origin with a slope equal to YCH4.

3. Results and discussion3.1 Bacteria identication

Bacteria cultivation and isolation processes were conductedfrom the ASGCB mix sludge samples at the end of each batchoperating HRT. Table 2 summarizes the microbial identica-tion of ASGCB mix sludge sample by using 16S rDNA. Eighttypes of bacteria and ve types of archaea were obtained. The16S rDNA data of ASGCB sample illustrates that the microor-ganisms consisted of Phylum proteobacteria (38.5%), Firmicutes(15.4%), Fusobacteria (7.6%) and Euryarchaeota (38.5%).

Escherichia fergusonii, Enterobacter asburiae and Enterobactercloacae which belong to the family of Enterobacteriaceae wereidentied in the ASGCB. These types of bacteria are known toproduce hydrogen gas from carbohydrates and fatty acids.14,15

Sulfur-reducing anaerobe bacteria such as Desulfovibrio aero-tolerans, Desulfobulbus propionicus, Fusobacterium nucleatum,Paenibacillus pabuli and Bacillus subtilis were also obtained.These types of bacteria survived with sulfur content in POMEwastewater in the ASGCB (Table 1). To our knowledge this is therst report that Fusobacterium nucleatum, Paenibacillus pabuliand Bacillus subtilis as sulfur-reducing bacteria are present inthe anaerobic process of POME wastewater. The sulfur-reducingbacteria consumed of sulfur as the electron acceptor to formhydrogen sulde gas (H2S) from the hydrogen gas (H2) as theelectron donor.16–19 Therefore, high concentration of H2S gaswas obtained from the anaerobic degradation process (Table 3).Low H2 gas production was detected as most of the hydrogenion had been used to form H2S gas (Table 3).

High contents of fatty acids were characterized from POMEwastewater (Table 1). Therefore, ve species of methaneproducing bacteria such as Methanofollis ationis, Meth-anobacterium sp., Methanosaeta concilii, Methanosarcina mazei,and Methanosarcina acetivorans have been obtained in ASGCB.The methanogenic archaea bacteria converted acetates intomethane gas (CH4) and carbon dioxide gas (CO2).20,21 Perhaps,the presence of the identied microbial community in the

This journal is © The Royal Society of Chemistry 2014

Table 2 Microbial community identification in ASGCB

Microbial groupSequencesimilarity (%) Accession no.

Phylogenetic affiliation

Phylum Class Family

Bacteria Escherichia fergusonii 99 NR 074902 Proteobacteria Gammaproteobacteria EnterobacteriaceaeEnterobacter asburiae 99 NR 024640 Proteobacteria Gammaproteobacteria EnterobacteriaceaeEnterobacter cloacae 99 NR 044978 Proteobacteria Gammaproteobacteria EnterobacteriaceaeDesulfovibrio aerotolerans 99 NR 043163 Proteobacteria Deltaproteobacteria DesulfobulbaceaeDesulfobulbus propionicus 99 NR 074930 Proteobacteria Deltaproteobacteria DesulfovibrionaceaePaenibacillus pabuli 99 NR 040853 Firmicutes Bacilli PaenibacillaceaeBacillus subtilis 99 NR 102783 Firmicutes Bacilli BacillaceaeFusobacterium nucleatum 99 NR 074412 Fusobacteria Fusobacteria Fusobacteriaceae

Archaea Methanobacterium sp. 99 NR 102889 Euryarchaeota Methanobacteria MethanobacteriaceaeMethanofollis tationis 99 NR 041717 Euryarchaeota Methanomicrobia MethanomicrobiaceaeMethanosarcina mazei 99 NR 118787 Euryarchaeota Methanomicrobia MethanosarcinaceaeMethanosaeta concilii 99 NR 104707 Euryarchaeota Methanomicrobia MethanosaetaceaeMethanosarcina acetivorans 99 NR 044724 Euryarchaeota Methanomicrobia Methanosarcinaceae

Table 3 The stable operation results obtained under various HRT in ASGCBa

Composition

HRT (day)

24 20 16 12 8

OLR (g TCOD per L day) 2.75 � 0.01 3.3 � 0.06 4.1 � 0.05 5.45 � 0.06 8.2 � 0.05Volumetric feed ow rate(L POME per day)

0.583 � 0.00 0.7 � 0.00 0.875 � 0.00 1.167 � 0.00 1.75 � 0.00

TCOD removal efficiency (%) 89.66 � 0.03 86.06 � 0.10 85.15 � 0.03 84.56 � 0.29 79.83 � 0.03Effluent TCOD (mg per L) 6810 � 4.65 9130 � 2.90 9720 � 2.90 10 130 � 10.00 13 200 � 5.85Inside TCOD (mg per L) 12 650 � 5.00 14 250 � 14.45 16 860 � 44.80 21 270 � 36.00 32 150 � 11.55Total alkalinity in ASGCB(mg CaCO3 per L)

11 400 � 13.30 10 400 � 28.90 9160 � 5.00 8400 � 5.75 7650 � 2.90

Total VFA in ASGCB(mg CH3COOH per L)

514.29 � 2.50 1017.14 � 0.50 2028.57 � 0.82 2497.14 � 9.80 3860.57 � 5.35

VFA : Alk 0.05 � 0.01 0.10 � 0.01 0.22 � 0.01 0.30 � 0.01 0.50 � 0.01MLVSS in ASGCB (mg per L) 7610 � 2.90 8220 � 28.85 9290 � 25.05 10 780 � 5.05 11 570 � 5.80Total volume biogas production(L per day)

17.79 � 0.05 19.85 � 0.03 23.82 � 0.01 32 � 0.06 46.76 � 0.11

Daily volume methane production, QCH4

(L CH4 per day)12.90 � 0.06 14.17 � 0.06 16.41 � 0.06 21.71 � 0.05 30.81 � 0.05

Methane gas (CH4) % 72.50 � 0.03 71.40 � 0.29 68.90 � 0.15 67.90 � 0.30 65.90 � 0.26Carbon dioxide gas (CO2) % 27.30 � 0.25 28.40 � 0.11 31 � 0.23 31.80 � 0.06 33.90 � 0.06Hydrogen sulde gas (H2S) (mg per L) 1825 � 3.00 3245 � 5.05 5820 � 28.80 6975 � 15.25 12 370 � 5.25Hydrogen gas (H2) (mg per L) 214 � 0.50 488 � 0.55 717 � 0.50 1998 � 0.55 3841 � 0.55

a � standard deviation.

Paper RSC Advances

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article Online

ASGCB played an important role in degrading POME in thecomplex anaerobic degradation processes.

3.2 Operational parameters and biodegradability of POMEwastewater in ASGCB

Table 3 summarizes the results obtained under various HRT. Ananalysis of variance (ANOVA) was conducted to test the signi-cance of experimental data at different HRT in Table 3. Table 4states the outcome of ANOVA test as F-value and p-value for eachoperational variables under various HRT. High F-value with lowp-value (<0.05) was obtained through ANOVA signicant test.

This journal is © The Royal Society of Chemistry 2014

Therefore, the operational results are statistically proven to besignicant in the present study.22,23

Throughout the study, the organic loading rates were variedfrom 2.75 g TCOD per L per day to 8.2 g TCOD per L per day. Ateach HRT, the operational parameters were operated untilsteady state was reached as indicated by a constant effluentTCOD, biogas production, total VFA, alkalinity, and MLVSSlevel. The effluent TCOD of the reactor increased as the HRTreduced. This is due to the increase of input organic loadingrates as the volumetric ow rate of POME was increased. On theother hand, the anaerobic degradation process of POME isoperated without acidication risk in the ASGCB with the level

RSC Adv., 2014, 4, 64659–64667 | 64663

Table 4 Analysis of variance (ANOVA) for the anaerobic degradation inASGCB

Factors

Statistics

F-value p-value Remark

TCOD removal efficiency 48.08 6.15 � 10�3 SignicantEffluent TCOD 49.22 5.95 � 10�3 SignicantReactor TCOD 87.76 2.58 � 10�3 SignicantTotal alkalinity in ASGCB 284.76 4.53 � 10�4 SignicantTotal VFA in ASGCB 97.22 2.22 � 10�3 SignicantVFA : Alk 230.09 6.22 � 10�4 SignicantMLVSS inside ASGCB 70.21 3.57 � 10�3 SignicantTotal biogas production 54.94 5.08 � 10�3 SignicantDaily volume methaneproduction

50.36 5.76 � 10�3 Signicant

Methane gas 96.43 2.25 � 10�3 SignicantCarbon dioxide gas 27.37 4.24 � 10�2 SignicantHydrogen sulde gas 27.37 1.36 � 10�2 SignicantHydrogen gas 251.13 5.46 � 10�4 Signicant

Fig. 2 Plot of specific substrate utilization rate, rX versus dilution rate,D (eqn (8)).

RSC Advances Paper

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article Online

of VFA/Alk between 0.05 and 0.50. The VFA/Alk levels fallbetween the recommended condition of 0.3–0.5 whereby thisprocess is considered to be operating favorably in anaerobiccondition.24,25 The level of alkalinity reduced from 11 400 mgCaCO3 per L to 7650 mg CaCO3 per L as the HRT reduced. Thehigh level of bicarbonate alkalinity was reached in the reactordue to the reaction of ammonia with carbon dioxide and waterin forming ammonium bicarbonate.26 The mixed liquor volatilesuspended solid (MLVSS) inside the ASGCB was analyzed as themicrobial growth or biomass concentration. The MLVSS variedfrom 7610 mg MLVSS per L to 11 570 mg MLVSS per L. Thisresult proves that the anaerobic degradation of POME could bewell performed as the increase of microbial growth withincreasing organic loading rate (OLR).

The daily biogas production increased from 17.79 L per dayto 46.76 L per day with increasing OLR. This is because thePOME was characterized with high concentration of palmiticfatty acids as stated in Table 1. Palmitic acids containing 16carbons long-chain fatty acid decomposed faster to shorterchain fatty acids such as acetic; propanoic; butanoic; iso-butanoic and etc. The identied anaerobic bacteria Escherichiafergusonii, Enterobacter asburiae, Enterobacter cloacae, Desulfo-vibrio aerotolerans, Desulfobulbus propionicus, Fusobacteriumnucleatum, Paenibacillus pabuli, Bacillus subtilis, Meth-anobacterium sp., Methanosaeta concilii, Methanofollis tationis,Methanosarcina mazei and Methanosarcina acetivorans furtherdecomposed the fatty acids to produce methane (CH4), carbondioxide (CO2), hydrogen (H2) and hydrogen sulde (H2S) gassesthrough acidogenesis, acetogenesis and methanogenesis ofanaerobic degradation pathway. The anaerobic degradationreactions are:

3C16H32O2 + 42H2O / 7CH3COOH + 4CH3CH2COOH

+ 3CH3CH2CH2COOH + 10CO2 + 52H2 (acidogenesis) (16)

2CH3CH2COOH + 2H2O /

3CH3COOH + 2H2 (acetogenesis) (17)

64664 | RSC Adv., 2014, 4, 64659–64667

CH3CH2CH2COOH + 2H2O /

2CH3COOH + H2 (acetogenesis) (18)

CH3COOH / CH4 + CO2 (methanogenesis) (19)

CH3COO� + SO42� / HS� + 2HCO3

� (sulphidogenesis) (20)

H+ + HS� 4 H2S (21)

The contents of the CH4, CO2, H2 and H2S produced duringthe steady-state period at different HRTs are shown in Table 3.The CH4 content decreased from 72.5% to 65.9% with decreaseof HRT. A decrease in CH4 content is caused by an increase ofthe VFA level and poor performance of effluent TCOD. This isbecause high OLR was characterized by high concentration ofVFA at the low HRT. These results are in agreement with thending by Hikmet27 who reported that the CH4 contentdecreased with TCOD removal efficiency and decrease in HRT.

3.3 Evaluation of kinetic coefficients

3.3.1 Kinetic coefficients YG and b. The experimental dataunder steady state condition (Table 3) were analyzed and bio-logical kinetic models were tested. The growth yield (YG) andspecic biomass decay (b) were obtained using eqn (8) byplotting rX versus D. As illustrated in Fig. 2, the points tted wellon a straight line with a regression coefficient of 0.99. Thevalues of YG and b were calculated to be 0.357 g VSS per gTCODremoved and 0.07 per day, respectively from the slope andintercept. The evaluated growth yield, YG was in the range of0.181 g VSS per g TCODremoved and 0.99 g VSS per g TCODremoved

as reported in the literature: 0.69 g VSS per g TCODremoved;28

0.174 g VSS per g TCODremoved;11 0.99 g VSS per g TCODremoved29

and 0.181 g VSS per g TCODremoved.30 Hitherto, this is due to thesimilar substrate of POME wastewater was treated in theanaerobic degradation process.

3.3.2 rX,max; Ks; mmax and QC biokinetic coefficients. Themaximum specic substrate utilization rate (rX,max) and satu-ration constant for substrate (Ks) were obtained through eqn(10) by Lineweaver–Burk plot. By plotting 1/rX and 1/S datapairs, a straight line with regression coefficient of 0.98 isobtained (Fig. 3). The values of rX,max and Ks are 0.957 g TCODper g VSS day and 25.03 g TCOD per L, respectively, from the

This journal is © The Royal Society of Chemistry 2014

Fig. 3 Determination of maximum specific substrate utilization, rX,max

and saturation constant, Ks through Lineweaver–Burk plot (eqn (10)).

Paper RSC Advances

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article Online

intercept and slope. Themaximum specic biomass growth rate(mmax) and the critical retention time (QC) are 0.27 per day and3.72 day, respectively, obtained from eqn (13) and (14). Thevalue of mmax for the substrate of TCOD is close to that of thereported work of 0.207 per day 11 and 0.304 per day.31 This is dueto similar activity for microbial growth in the treatment ofPOME. The critical retention time, QC at which the washout ofmicroorganisms was 3.72 days for the substrate of TCOD. Thus,the ASGCB was operated without risk to avoid low performanceof anaerobic degradation process as the kinetic analysis exper-iments were conducted between 24 days and 8 days of retentiontime.

The small values of rX,max, Ks and mmax were achieved as theanaerobe bacteria could easily degrade the organic matters inthe input of palm oil wastewater. Therefore, the ASGCB signif-icantly performed well between 79.83% and 89.66% reductionof TCOD with increase of OLR (Table 3). This signicant ndingis reected by the newly isolated anaerobe bacteria: Escherichiafergusonii, Enterobacter asburiae, Enterobacter cloacae, Desulfo-vibrio aerotolerans, Desulfobulbus propionicus, Fusobacteriumnucleatum, Paenibacillus pabuli, Bacillus subtilis, Meth-anobacterium sp., Methanosaeta concilii, Methanofollis tationis,Methanosarcina mazei and Methanosarcina acetivorans that areable to degrade the fatty acid contents of POME wastewater.

3.3.3 Kinetics of methane production. The experiment datalisted in Table 3 were used to determine the yield of methane,YCH4

. By plotting VCH4against (S1 � S2)Q in eqn (15), a value of

Fig. 4 Determination of methane yield, YCH4by eqn (15).

This journal is © The Royal Society of Chemistry 2014

0.34 L CH4 per g TCODremoved was obtained from the slope(Fig. 4). In Fig. 4, the straight line passing through the originwith a correlation coefficient of 0.99 suggests the validation ofeqn (15). The value of YCH4

of 0.34 L CH4 per g TCODremoved isreported so far to be the highest methane yield for palm oilwastewater compared to the previously reported values of0.325 L CH4 per g TCODremoved;9 0.325 L CH4 perg TCODremoved

11 and recently 0.32 L CH4 per g TCODremoved.30

Therefore, the presence of Methanobacterium sp., Methanosaetaconcilii, Methanofollis tationis, Methanosarcina mazei andMethanosarcina acetivorans anaerobes performs well insuspended growth anaerobic process.

4. Conclusion

New anaerobic bacteria such as Escherichia fergusonii, Enter-obacter asburiae, Enterobacter cloacae, Desulfovibrio aerotolerans,Desulfobulbus propionicus, Fusobacterium nucleatum, Paeniba-cillus pabuli, Bacillus subtilis, Methanobacterium sp., Meth-anosaeta concilii, Methanofollis tationis, Methanosarcina mazeiand Methanosarcina acetivorans were isolated from ASGCB.Their presence in the ASGCB produced the highest evermethane yield at 0.34 L CH4 per TCODremoved. Monod modelobtained for the suspended growth anaerobic process producesthe following biological kinetic coefficients: YG (0.357 g VSS perg TCOD), b (0.07 per day), mmax (0.27 per day), Ks (25.03 g TCODper L), QC (3.72 day) and YCH4

(0.34 L CH4 per TCODremoved),respectively.

Nomenclature

b

Specic biomass decay (1/day) rX Specic substrate utilization rate (g TCOD per g VSS

day)

rv Substrate utilization rate per volume (g TCOD per day) rX,max Maximum specic substrate utilization (g TCOD per g

VSS day)

t Time (day) ASGCB Anaerobic suspended growth closed bioreactor Alk Total alkalinity BOD Biochemical Oxygen Demand (mg L�1) CO2 Carbon dioxide gas CH4 Methane gas D Dilution rate, 1/HRT (1/day) DNA Deoxyribonucleic acid HRT Hydraulic Retention Time (day) H2 Hydrogen gas H2S Hydrogen sulde gas Ks Saturation constant for substrate (g TCOD per L) L Liter NH3-N Ammonia nitrogen (mg L�1) O & G Oil and Grease (mg L�1) OLR Organic loading rate (g TCOD per L day) PCR Polymerase chain reaction POME Palm oil mill effluent Q Volumetric ow rate (L per day) S Substrate concentration in the reactor (mg L�1)

RSC Adv., 2014, 4, 64659–64667 | 64665

RSC Advances Paper

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article Online

S1

64666 |

Inuent substrate concentration (mg L�1)

S2 Effluent substrate concentration (mg L�1) SCOD Soluble Chemical Oxygen Demand (mg L�1) SRT Solid retention time SS Suspended solid (mg L�1) TCOD Total Chemical Oxygen Demand (mg L�1) TN Total nitrogen (mg L�1) TS Total solid (mg L�1) TVS Total volatile solid (mg L�1) V Reactor volume (L) VCH4

Methane gas volume (L)

VFA Volatile fatty acid (mg L�1) VSS Volatile suspended solid (mg L�1) X Biomass concentration in the reactor (mg L�1) YG Growth yield (g VSS per g TCODremoved) YCH4

Methane yield (L CH4 per TCODremoved)

Greek letter

mmax

RSC

Maximum specic biomass growth rate (1/day)

QC Critical retention time (day) m Specic biomass growth rate (1/day)

Acknowledgements

The authors would like to express their sincere gratitude toUniversity Sains Malaysia Research University (RU) grant: 1001/PTEKIND/814160 for nancial support, School of Environ-mental Engineering of University Malaysia Perlis (UniMAP) andMALPOM Industries Sdn Bhd.

References

1 G. D. Najafpour, A. A. L. Zinatizadeh, A. R. Mohamed, H. Isaand H. Nasrollahzadeh, High-rate anaerobic digestion ofpalm oil mill effluent in an upow anaerobic sludge-xedlm bioreactor, Process Biochem., 2006, 41, 370–379.

2 Z. Dong, C. Yinguang, Z. Yuxiao and Z. Xiaoyu, New sludgepretreatment method to improve methane production inwaste activated sludge digestion, Environ. Sci. Technol.,2010, 44(12), 4802–4808.

3 S. Tariq, P. Tara, S. Kathleen and M. Foght Julia, AnaerobicBiodegradation of longer chain n-Akanes coupled tomethane production in oil sands tailings, Environ. Sci.Technol., 2011, 45(13), 5892–5899.

4 M. Tabatabaei, A. Rahim Raha, N. Abdullah, A. Wright,Y. Shirai, K. Sakai, A. Sulaiman and M. A. Hassan,Important of methanogenic archea populations inanaerobic wastewater treatmets, Process Biochem., 2010, 45,1214–1225.

5 M. Badiei, J. J. Md, N. Anuar, R. S. Abdullah Siti, L. Swee Suand M. A. Kamaruzzaman, Microbial community analysis ofmixed anaerobic microora in suspended sludge of ABSR

Adv., 2014, 4, 64659–64667

producing hydrogen from palm oil mill effluent, Int. J.Hydrogen Energy, 2012, 37, 3169–3176.

6 Y.-S. Wong, S.-A. Ong, T.-T. Teng, N. Aminah Lutpi andK. Kumaran, Production of Bioocculant by Staphylococcuscohnii ssp.from Palm Oil Mill Effluent (POME), Water, Air,Soil Pollut., 2012, 223, 3775–3781.

7 L. Singh, M. F. Siddiqui, A. Anwar, R. M. H. Ab, M. Sakinahand Z. A. Wahid, Application of polyethylene glycolimmobilized Clostridium sp. LS2 for continuous hydrogenproduction from palm oil effluent in upow anaerobicsludge blanket reactor, Biochem. Eng. J., 2013, 70, 158–165.

8 B. Nurdan and F. Ayse, Determination of kinetic constant ofan anaerobic hybrid reactor, Process Biochem., 2002, 38, 73–79.

9 R. Borja and C. J. Banks, Anaerobic digestion of palm oil milleffluent using an up ow anaerobic sludge blanket (UASB)reactor, Biomass Bioenergy, 1994, 6, 381–389.

10 R. Borja, C. J. Banks and E. Sanchez, Anaerobic treatment ofpalm oil mill effluent in a two-stage upow anaerobic sludgeblanket (UASB) system, J. Biotechnol., 1996, 45, 125–135.

11 A. A. L. Zinatizadeh, A. R. Mohamed and G. D. Najafpour,Kinetic evalution of palm oil mill effluent digestion in ahigh rate up-ow anaerobic sludge xed lm bioreactor,Process Biochem., 2006, 41, 1038–1046.

12 W.-H. Choi, C.-H. Shin, S.-M. Son, A. Ghorpade, J.-J. Kim andJ.-Y. Park, Anaerobic treatment of palm oil mill effluentusing combined high rate anaerobic reactors, Bioresour.Technol., 2013, 141, 138–144.

13 Bacteriological Analytical Manual Website: http://cfsan.fda.bam.gov.

14 L. Blairon, M. L. Maza, I. Wybo, D. Pierard, A. Dediste andO. Vandenberg, Vitek 2 ANC card vesus BBL CrystalAnaerobe and RapID ANA II for identication of clinicalanaerobic bacteria, Anaerobe, 2010, 16, 355–361.

15 M. I. Sriram, S. Gayathiri, U. Gnanaselvi, P. S. Jenifer,S. M. Raj and S. Gurunathan, Novel lipopeptidebiosurfactant produced by hydrocarbon degrading andheavy metal tolerant bacterium Escherichia fergusoniiKLU01 as a potential tool for bioremediation, Bioresour.Technol., 2001, 102, 9291–9295.

16 G. L. Mogensen, K. U. Kjeldsen and K. Ingvorsen,Desulfovibrio aerotolerans sp. nov., an oygen tolerantsulphate-reducing bacterium isolated from activatedsludge, Anaerobe, 2005, 11, 339–349.

17 S. S. Paul, S. M. Deb and D. Singh, Isolation andcharacterization of novel sulphate-reducing Fusobacteriumsp. and their effects on in vitro methane emission anddigestion of wheat straw by rumen uid from indian riverinebuffaloes, Anim. Feed Sci. Technol., 2011, 166–167, 132–140.

18 M. H. M. Zainudin, M. A. Hassan, M. Tokura and Y. Shirai,Indigenous cellulolytic and hemicellulolytic bacteriaenhanced rapid co-composting of lignocellulose oil palmempty fruit bunch with palm oil mill effluent anaerobicsludge, Bioresour. Technol., 2013, 147, 632–635.

19 A. A. El-Midany andM. A. Abdel-Khalek, Reducing sulfur andash from coal using Bacillus subtitis and paenibacilluspolymyxa, Fuel, 2014, 115, 589–595.

This journal is © The Royal Society of Chemistry 2014

Paper RSC Advances

Publ

ishe

d on

20

Nov

embe

r 20

14. D

ownl

oade

d by

Tok

yo U

nive

rsity

of

Scie

nce

on 2

4/12

/201

4 03

:16:

06.

View Article Online

20 P. Lins, C. Reitschuler and P. Iiimer, Methanosarcina spp.,the key to relieve the start-up of a thermophilic anaerobicdigestion suffering from high acetic acid loads, Bioresour.Technol., 2014, 152, 347–354.

21 I. Maus, D. Wibberg, R. Stantscheff, K. Cibis, F.-G. Eikmeyer,H. Konig, A. Puhler and A. Schluter, Complete genomesequence of hydrogenotrophic ArchaeonMethanobacterium sp. Mb1 isolated from a productionscale biogas plant, J. Biotechnol., 2013, 168, 734–736.

22 S. R. Chaganti, D. H. Kimb, J. A. Lalmanc and W. A. Shewac,Statistical optimization of factors affecting biohydrogenproduction from xylose fermentation using inhibitedmixed anaerobic cultures, Int. J. Hydrogen Energy, 2012, 37,11710–11718.

23 W. Y. Shian, T.-T. Teng, S.-A. Ong, M. Norhashimah,M. Rafatullah and J.-Y. Leong, Methane Gas Productionfrom Palm oil wastewater- An Anaerobic MethanogenicDegradation Process in Continuous Stirrer SuspendedClosed Anaerobic Reactor, J. Taiwan Inst. Chem. Eng., 2014,45, 896–900.

24 R. Borja, A. Martin, E. Sanchez, B. Rincon and F. Raposo,Kinetic modeling of the hydrolysis, acidogenic andmethanogenic steps in the anaerobic digestion of two-phase olive pomace (TOPO), Process Biochem., 2005, 40,1841–1847.

This journal is © The Royal Society of Chemistry 2014

25 R. Borja, E. Sanchez, B. Rincon, F. Raposo, M. A. Martin andA. Martin, Study and optimization of the anaerobicacidogenic fermentation of two phase olive pomace,Process Biochem., 2005, 40, 281–291.

26 H. W. Zhao and T. Viraraghavan, Analysis of theperformance of an anaerobic digestion system at theRegina wastewater treatment plant, Bioresour. Technol.,2004, 95, 301–307.

27 T. Hikmet, Temperature and organic loading dependency ofmethane and carbon dioxide emission rates of a full-scaleanaerobic waste stabilization pond, Pergamon, 1994, 0043–1354, 251–257.

28 B. G. Yeoh, A kinetic based design for thermophilicanaerobic treatment of a high strenght agro industrialwastewater, Environ. Technol., 1986, 7(1), 509–518.

29 Yee-Shian Wong, A. B. K. Mohd Omar and T.-T. Teng,Biological kinetics evaluation of anaerobic stabilizationpond treatment of palm oil mill effluent, Bioresour.Technol., 2009, 100, 4969–4975.

30 Y. J. Chan, M. F. Chong and C. L. Law, An integratedanaerobic-aerobic bioreactor (IAAB) for the treatment ofpalm oil mill effluent (POME): Start-up and steady stateperformance, Process Biochem., 2012, 47, 485–495.

31 M. Faisal and U. Hajime, Kinetic analysis of palm oil millwastewater treatment by a modied anaerobic baffledreactor, Biochem. Eng. J., 2011, 9, 25–31.

RSC Adv., 2014, 4, 64659–64667 | 64667