International Confrence on Sustainable Deve lopent for Water and Wmte Water Treatment Yagyakarta, INDONESIA, December 14-15,2009.

Membrane Bioreactor for Palm Oil Mi21 Effluent and Resource Recovery

2. Ahmad Lecturer, Faculty of Civil and Environmental Engineering, Universiti Tun Hussein Onn Malaysia

Email: ztil@,uthm.edu. ITZY

M. B. Ridzuan Lecturer. Faculty of Civil and Environmental Engineering, Universiti Tun Hussein Onn Malaysia

Email: mdbahar@,z~thm. edu. mv

Z. Daud Dr, Faculty of Civil and Emironmental Engineering, Universiti Tun Hussein Onn Malaysia

Email: zawawil@,uthm.edu.mv

Abstract

The per$omance of a membrane bioreactor (MBR) far treating palm oil mill efluent ( P O U ) has been studied The object* of this study was to observe and evaluate the pm$ormonce of MBR for POME treatment. A high strength POME with organic loadng rate uf 1.4 kg BOD per m3d war fed into an aerobic MBR The MBR was operated with constant flux of 15 LMH. The reactor was operated at a mixed liquor suspended solid (MLSS) concentration about 4.5 dl. The removal eflciency of COD, SS, and Turbid@ were achieved 88%, 99%, and 99% .respectively. The system were also capable to remove 61% of BOD, 75% of colmr, 74% of TN, and 49% of P. To observe the gect of biomass on membrane fouling, activated slugde for MBR was investigated ta determine concentration of carbohybate for soluble microbial product (SMP) and extracelZuZm polymeric sub~tames @Ps). The aerobic MBR was farnd to be able to degrade POME signij?cantly and high qualiq efluenf could be reused for various other applications.

1. INTRODUCTION

. . . . . A membrane bioreactor (MBR) process consists of suspended growth biological reaction combine with membrane module separating the biomass from the treated water. Recently MBR has been used increasingly for municipal and i n m .n"e advmcement of

'

membrane technology in relation to materials and process design. MBR systems feature many advantages over the conventional processes with their highly improved effluent quality, increased organic loading, reduced footprint md sludge production [I].

Membrane technology for POME treatment has been developed using side-stream membrane anaerobic system (MAS) to determine the performance of removal organic in the system [2]. Resource recovery related to POME treatment was investigated with the combination of ultrafiration and reverse osmosis membrane as the membrane separation treatment water recycling [3]. Membrane

Membrane Bioreactor for Palm Oil Mill Efluenf and Resource Recovery

technology coupled with coagulation and flocculation as pretreatment reclaimed drinking water from POME with water recovery [4]. The pilot-scale study deals with highly polluting agro-industry's wastewater, palm oil mill effluent (POME) using ceramic ultrafiltration membrane effecting transmembrane pressure and crossflow velocity on permeate flux, limiting flux and percentage rejection for suspended solids and dissolved organic matters were investigated [S].

Membrane bioreactor (MBR) for POME treatment was studied using flat sheet membrane modules to determhe the performance of the system on flux variable of 10 to 12 LMH with sludge concentration of 7.2 g/l MLSS and cleaning membrane processes [6]. Hybrid MBR to treat POME was reported that the systm was operated in the critical flux of 15 LMH with sludge concentration of 8 g/l and cakes resistance rn~~hardsrn was also investigated relating to membrane fouling 17, 8, 9). An active biomass (sludge) inside hybrid MBR was investigated to observe the membrane fouling effect due to extracellular polymeric substances (EPS), soiuble microbial products ( S M P ) and inert biomass. Protein was dominant component in the PMP and EPS that caused the fouling membrane mechanism [lo]. Most bacteria produce EPS of biological origin than participate in the formation of microbial aggregate whether the bacteria grow in suspended cultures or in biofilms. The biofdm or floc consists of W a cell enveloped by a matrix of large polymeric molecules where are located at or outside the cell surface [ll]. SMP are s01uble cellular components that are released during cell lysis, difkse through the cell membrane, are lost during synthesis, or are excreted for some purposed [12, 131.

The objective of this study is to evaluate the performance of the ME3R system and to observe biomass effect focusing on carbohydrate in SMP and EPS related to membrane fouling mechanism..

2. MATERIAL AND METHODS

2.1 Fxperiineqtal set-up and operating conditions

The experimental set-up is shown in Fig.1. The MBR consists of aerobic reador where a module of flat sheet membrane is i m m d in the reactor. The membrane modules are made h m chlorinated polyethylene (Kubota, Japan) with nominal pore size of 0.4 ,U m and effective area of ~ . l r n ~ / ~ c . The workiag volume is 20 litres and operatting condition of the MBR is mentioned in Tables 1. An airlift was installed underneath the membrane module in order to provide aeration to the membrane and oxygen to the biomass.

Fig. 1: Scbematk of MBR

IntemationaI Confeeme on SrrstainabIe DeveZopntentfor Water and Wizrte Wbter Treatment

Table 1: Operation parameters

The MBR was seeded with activated sludge obtained &om the SBR POME pilot plant at Pertubuhan Peladang Negeri Johor (PPNJ) Palm Oil Mill, Kahang, Johor, Malaysia. After 24 hours of acclimatisation, the membrane filtration was turned on progressively. The feeding characteristics are showed in Table 2.

No. 1. 2- 3. 4. 5. 6. 7. 8.

Table 2: Feeding chrracteristics

Operation Condition Organic loading Qi,

Suction time (onloff) Constant flux MLSS DO Temperature Air flow

2.2 Analytical method3

Laboratory experiments were carried out in the University of Tun Hussein Onn Malaysia (UTHM), Environmental Analysis Laboratory. Experimental analysis was conducted according to standard methods [14]. The activated sludge was regularly tested for MLSS and MLVSS concentrations. cnern1ca.I oxygen demand ( spectrophotometer (HACHDR 5000). Biochemical oxygen demand (BOD) was determined using BODtrak apparatus. Dissolved oxygen o) concentration and temperature were monitored using portable HACH kits rn-ement, turbidity was measured using HACH DR/2010 and pH was also monitored using pH meter. Soluble microbial products (SMP) and extracellular polymer substances (EPS) were evaluated in the biomass following the modified heating method [I$]. Carbohydrate and protein concentrations of the SMP and EPS were measured using the method originally introduced by Dubois [16]. The Carbohydrate in the EPS and SMP were measured using spectrophotometer (HACWDR 5000).

Unit kg B O D I ~ ~ . ~

I/d min

LMH mg/l mgA "C

LPM

1

Range 750 f 112 1305 k 106 665 & 130

27 f 2 49 + 5

1604 f 368 24 zk 2

A

Range 1.4 36

. 1012 15

Up to 450 6 to 8

25 to 27 10 to 15

Unit

m a mgll mg/l mgA mgn NTU Copt

No.

1. 2. 3. 4. 5. 6. 7.

Parameter BOD COD SS TN TP

Turbidity Colour

Membrane Bioreactor for Palm Oil Mill &'fluent and Resource Recovery

3. RESULTS AND DISCUSSION

3.1 PerJormance of the M%R system

Table 3 shows the concentration of feeding and permeate quality. The average SS and turbidity removal were about 99%. The COD and BOD removal were an average of 88% and 61% respectively. At rm average the total removal was about 75% for colour, 74% for TN and 49% for'^^. ~%wever, the preesetment processes played a significant role in reducing almost 99.9% of suspended solids, 863% of BOD, 85.0% of COD and 88.2% of turbidity in POME before entered the membrane treatment [5, 17, 181.

Table 3: Concentration of feeding and permeate quality

Table 4 shows the comparison of removal efficiency for three (3) cases of study using flat sheet Kubota membrane for POME treatment with different flux and MLSS. In this case of study, the MBR was operated with flux of 15LMH and MLSS of 4.5 gll. The second cases, an aerobic MBR [6] was operated with flux of 12 LMH and MLSS of 7.2 gA. And, the Hybrid MBR [7J which consists of sequencing processes of anaerobic, anoxic and aerobic MBR The reactor was operated with flux of 15 LMH and MLSS of 8 g/l. An average, COD removal achieves 88% for this study. It is very good LU

removal efficiency of SS for both reactors, in this study and hybrid MBR, are approximately 99%. Hence, membrane is suitable for SS removal in wastewater treatment. The achievement of removal of turbidity and colour are better in this study if comparing to aerobic MBR The concentration of sludge affected the removal potential of turbidity and colour. In this study, the concentration of MLSS was 4.5 g/l while in the aerobic MBR was 7.2 gll. On the contrary, BOD, TN and TP for this study are lower if comparing to the aerobic MBR and hybrid MBR Also concentration of sludge aflected the removal potential of the BOD, TN and TP. The MLSS in aembic MBR and hybrid MBR were quite high comparing to the MLSS in this study.

% Removal Efficiency

6 1

88

99

74

49

99

75

295 =t 32

153 ;t 42

8k5

7k2

25 k 5

16*2

6=t 1

Influent

750 =t 112

1305 z t 106

665 k 130

2 7 k 2

49h5

1604 & 368

24rt2

Unit . .

mg/l

mgfi

m@J1

mg/l

NTU

Copt

No.

1.

2.

3.

4.

5.

6.

7.

Parameter

BOD

COD

SS

TN

TP

Turbidity

Colour

International Conference on Swuinable Development for Water and Waste Water Treatment

Table 4: Comparison of rtmoval efficiency

3.2 Biomass influence in the MBR system

No.

1.

2.

3.

4.

5.

6.

7.

8.

Activated sludge inside the MBR was started-up with MLSS concentration of 700 mgA. The system was fed continuously with raw POME for 42 days and biomass kept growing until they achieved steady-state condition with MLSS maintained about 4500 mg/l. The avemge MLVSS/MLSS ratio during the period of the experimentation was 0.87.

4 SMP h, EPS

Parameter

BOD

COD

SS

TKN

TN

TP

Turbidity

Colour

MLSS (man)

Fig. 2: Concentration of carbohydrate and MLSS

% Removal Efficiency

MBR (This study)

61

88

99

- 74

49

99

75

Aerobic MBR [q

87

90

77

7 1

93

23

Hybrid MBR [7J

- 94

98

83

64

-

Membrane Bioreactor for Palm Oil Mill Efluent and Resource Recovery

Figure 2 is compared concentration of carbohydrate in SMP and EPS correlate to MLSS. The concentration of EPS shows very high at the early stage of 55 mg1gVSS for MLSS 700 mg/l and abruptly reduce to about 7 mg/gVSS at the steady-state condition. While the concentration of SMP shows slightly reducing from 9 to 6 rng/gVSS at the early stage to the steady-state condition. When the concentration MLSS is high (4500mgn), the concentrations of EPS and SMP are narrow to similar values of 7 mg/gVSS for EPS and of 6 mg/gVSS for SMP. The trend of narrow variation region was indicated by researchers [19].

The relationship concentration of protein and MLSS [lo} (Figure 31, SMP reveals slightly higher than concentration of EPS. The concentration of SMP and EPS as well as protein and carbohydrate, protein was found to be the dominant component in EPS and PMP for five MBR plants, with SMP values consistently lower than the EPS one [20].

2000 3000 4000 5000 6000 7000 8000

~ ~ L S S (moll)

Fig. 3: Concentration of Protein and MLSS [lo] .

4, CONCLUSIONS

Performance of MBR system is very good to treat and resource 'recovery for wastewater treatment

Ml3R depends on the operation and biomass condition in the reactor. For high MLSS of 7 to 8 g/l, operation flux of 12 to 15 LMH, the removal efficiency achives more.than 90% for BOD, 87% for BOD, 77% for TKN, 83% for TN and 64% for TP. For the sludge of 4.5 g/l MLSS, with constant flux of 15 LMH, the MBR system reveals good on SS, turbidity and colour removal. The romoval efficiency of these components are 99% for SS, and turbidity, and 75% for colour.

Biomass characteristics and relate to membrane fouling ,were investigated in this studies. SMP and EPS are two major components of active biomass that contribute to membrane fouling, For carbohydrate, EPS concentration was slightly higher than SMP concentration during the steady-state condition. However, for protein studied [lo], SMP was founded that it was the dominent component in the protein concentration.

Intemtional Conference on Sustainable Development for Water and Wmte Water Treatment

5. ACKNOWLEDGEMENT

The authors wouid like to thank to the Research Management and Innovation Centre, UHTM for sponsoring the present research under short grant vot no. 0567, and to the Faculty of Civil and Environmental Engineering, UTHM for giving full supports and commitment of this research.

6. REFERENCES

[I] Stephenson, T., Judd, S,, Jefferson, B., and Brindle, K., "Membrane bioreactor for wastewater treatmenty', 2002, WA Publishing, London, 2002.

[2] Fakhru'l-Razi, A., and Noor, M.J.M.M., 'Treatment of palm oil mill effluent (POME) with the membrane anaerobic system (MAS)", 1999, Water Science and Technology, 39(10), 159-163, 1999.

[3] Ahmad, A.L., Ismail, S. and Bhatia, S., "Water recycling from palm oil mill ef'fluent (POME) using membrane technologyn, 2003, Desalination, 157,87-95,2003.

[4] Ahrnad, A.L., Chong, M.F., Bhatia, S., Ismail, S., "Drinking water reclamation from palm oil , mill effluent (POME) using membrane techulogy", 2006, DesaIi1t~tion,l91,35-44,2006.

[S] Ahrnad, A.L., Ismail, S., Bhatia, S,, "UlUlb.afiltration behavior in the treatment of ago-industry effluent: Pilot scale studiesn, 2005, Chemical Engineering Science, 60,5385 - 5394,2005

[6] Lim, LP. and Ujang, Z., "Critical flux of membrane bioreactor (MBR) for digested palm oil mill effluent (POME) treatment", 2004, Proceedings of the Water Environment -Membrane Technology (wEMlZ004). June 7 - 10. Seoul, Korea: 793-799,2004.

[n Ahmad, Z., Ujang, Z., Yunus, S., Damaymti, A, and Olsson, G., "A novel approach for palm oil mill effluent (POME) treatment and recovery using hybrid membrane bioraictor (MBR)", 2008, Proceeding of Asiawater 2008, An Infernational Conference on Water & Wastewater, Kuala Lumpur, Malaysi, 1-2 April 2008.

f8] Ahrnad, Z., Ujang Z., and Olsson G., " Membrane fouling resistance in a hybrid membrane bioreador for palm oil mill effluent and resource recoveryn, 2008, Proceeding of the International Conference and Expo on Environmental Management and Technologies (ICEEAUT'08), PWTC, Kuala Lumpw, Malaysia, 10th - 12th December 2008.

[9] Ahmad, Z., Ujang, Z., Abdul Latie A.A., and Ahmat Nor, N.I., Tffects of membrane fouling and cake resistance in a hybrid membrane bioreactor for palm oil mill effluent treatment ", 2008, Proceeding of the international conference on environment 2008 (ICENV 2008), Penang, Malaysia, 15-17 December 2008.

[lo] Ahmad, Z., Ujang, Z., Abdul L a t e A.A., and Mustapha, S., "Biomass characteristics and effect on membrane fouling in a hybrid membrane bioreactor for palm oil mill effluent treatment", 2008, Proceeding of the International conferem 2008 on sustainable emironmental technology and sanitationfor tropical region, Surabaya, Indonesia, 18 - 19 November, 2008.

[ l 11 Chrysi, S., Laspidou, Rittmann, B.E., "A unified theory for extracellular polymeric substances, soluble microbial products, and actives and inert biomass", 2001 Water Research. 36, 2711- 2720,200 1.

[12] Noguera, D. R., Araki, N., and Rittmann, B. E., "Soluble microbial products -SMP in anaerobic chemostats" , 1994, Biotechnol. Bioeng., 44, 1040-1047,1994.

Membrane Bioreacfor for Palm Oil Mill Efluenf and Resource Recovery

[13] Laspidou, C.S. and Rittrnann, B.E., "A unified theory for extracellular polymeric substances, soluble microbial products, and adive and inert biomass", 2002, Water Resoures, 36, 271 1- 2720,2002.

r14 1 APHA, A W A , WEF, "Standard Methods", 1998, American Public Health: Association, American Water Works Association and Water Environment Federd'on, Washington D.C, USA.

[15] Zhang, X., Bishop, L.P., and W e , B.K., " Comparison of extraction methods for quantifying extracellular polymers in biofilms", 1999, Water Science and Technology, 39(7), 21 1-218, 1999.

[16] Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., and Smith F., "Colorimetric methods for determination of sugars and related substancesy', 1956, Analytical Chemistry. 28,350-356, 1956.

1171 Najafpour, G., Yieng, H. A., Younesi, H., and Zinatizadeh, A., "Effect of organic loading on performmce of rotating biological contactm using Palm Oil Mill effluents", 2005, Process Biochemistry, 40,2879-2884,2005.

1181 Wu, T.Y., Mohammada, A. W., Jahia, J. M., and Anuar N., "Palm oil mi11 effluent (POME) treatment and bioresources recovery using ultrafiltration membrane: Effect of pressure on membrane fouling" 2007, Biochemical Engineering Journal, 2007.

[19] Cho, J., Song, K.G., Yun, H,, Ahn, K.H., Kim, J.Y., and Chung, T.H., 'Quantitative analysis of biological effect on membrane fouling in submerged membrane bioreactor", 2005, Water Science & Technology, 51(6-7), 9-18,200$.

1201 Reid, E., Liu, X., Judd, S. J , "Effect of high salinity on activated sludge characteristics and membrane permeability in an immersed membrane bioreactor", 2006, Jout.nal of Membrane Science 283,164-171,2006.