sustainable approach in palm oil industry – green …kilang minyak kelapa sawit untuk rawatan air...

Jurnal Kejuruteraan SI 1(7) 2018: 11-20https://doi.org/10.17576/jkukm-2018-si1(7)-02

Sustainable Approach in Palm Oil Industry – Green Synthesis of Palm Oil Mill Effluent Based Graphene Sand Composite (P-GSC) for Aerobic Palm Oil Mill

Effluent Treatment

(Pendekatan Mampan dalam Industri Minyak Kelapa Sawit – Sintesis Komposit Pasir Grafin (P-GSC) Berasaskan Efluen Kilang Minyak Kelapa Sawit untuk Rawatan Air Kumbahan Aerobik Kilang Minyak Kelapa Sawit)

Wan Nur Athirah Wan Mohammad Hamdana, Teow Yeit Haana,b*, Abdul Wahab Mohammada,b

aResearch Centre for Sustainable Process Technology (CESPRO),bChemical Engineering Programme,

Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia

ABSTRACT

Graphene sand composite (GSC), sand media coated with graphene, possess higher adsorption capacity for wastewater treatment compared to conventional activated carbon (AC). However, large-scale synthesis of GSC due to costly carbonaceous source appears as a great challenge towards commercial application. Palm oil mill effluent (POME), which abundantly discharged from the palm oil industry, appears to be a potentially low cost carbonaceous source for the synthesis of GSC. This study presented a green chemistry approach by utilizing the POME as carbonaceous source coated on the different particle size of river sand in synthesizing POME-based GSC (P-GSC). Field Emission Scanning Electron Microscope (FESEM), Energy Dispersive X-ray (EDX), and X-ray Diffraction (XRD) analysis indicated the successful graphinization of POME coated onto the river sand surface. Adsorption studies – batch column study and batch equilibrium study showed that small size P-GSC (0.30 – 0.60 mm) had the best performance in removing chemical oxygen demand (COD), colour, turbidity, and total dissolved solids (TDS) from diluted aerobic POME up to 94.7%, 92.3%, 83.3%, and 51.5%, respectively. This could be attributed to its larger surface area that provides more active sites for adsorption. Equilibrium sorption data of small size P-GSC was well-fitted to the Freundlich model and its associated adsorption kinetic could be described by Pseudo First Order model. The concept of sustainable waste-to-treat-waste has been proven throughout this study, where POME can be utilized to produce high performance P-GSC that can be used to treat aerobic POME.

Keywords: POME; POME-Dased Graphene Sand Composite (P-GSC); Graphinization; Sustainability Development; Wastewater Treatment

ABSTRAK

Komposit pasir grafin (GSC) merupakan media pasir yang disalut dengan grafin, mempunyai kapasiti penjerapan yang lebih tinggi dalam merawat air kumbahan berbanding karbon yang diaktifkan (AC). Namun begitu, sintesis GSC berskala besar seacra komersial menggunakan sumber karbohidrat yang berkos mahal, muncul sebagai suatu cabaran yang besar. Efluen minyak kelapa sawit (POME), yang dihasilkan secara banyak dari industri minyak kelapa sawit berpotensi digunakan sebagai sumber karbon yang berkos rendah dalam proses mensintesis GSC. Kajian ini bertujuan untuk menyampaikan pendekatan dalam menggunakan bahan semulajadi iaitu POME sebagai sumber karbon untuk disalut pada pasir sungai yang berbeza saiz zarah dalam mensintesis GSC yang berasaskan efluen kelapa sawit (P-GSC). Analisa pengimbasan mikroskop electron (FESEM), spektroskopi tenaga serakan sinar-x (EDX), dan pembelauan sinar-x (XRD) menunjukkan hasil kejayaan grafin yang telah disalut ke atas permukaan pasir sungai. Kajian penyerapan - kajian kelompok turus dan kajian kelompok keseimbangan, menunjukkan bahawa saiz kecil P-GSC (0.30 – 0.60 mm) mempunyai prestasi yang terbaik dalam mengurangkan peratus keperluan oksigen kimia (COD), warna, kekeruhan dan jumlah pepejal terlarut (TDS) daripada POME aerobik cair sehingga mencapai 94.7%, 92.3%, 83.3% dan 51.5%, masing-masing. Ini boleh dikaitkan dengan kawasan permukaan yang lebih luas, menyediakan lebih banyak tapak yang lebih aktif untuk proses penjerapan. Data penyerapan keseimbangan saiz kecil P-GSC mengikuti model Freundlich dan kinetik penjerapan yang berkaitan dapat diterangkan melalui model Pseudo-tertib pertama. Konsep sisa merawat sisa ini telah terbukti di sepanjang kajian ini, di mana POME boleh digunakan dalam menghasilkan P-GSC yang lebih berprestasi tinggi dalam merawat POME aerobik.

Kata kunci: POME; Komposit Grafin Pasir Berasaskan Efluen Kelapa Sawit (P-GSC);Penggrafinan; Pembangunan Mampan; Rawatan Air Kumbahan

JK 30 SI1(7) Bab 2new.indd 11 3/18/2019 2:20:40 PM

12

INTRODUCTION

Conventional water and wastewater treatment processes such as ponding system (Ahmad & Krimly 2014; Madaki & Seng 2013), biological treatment (Ahmad et al. 2003; Azmi et al. 2014), membrane technology (Azmi et al. 2014; Rupani & Singh 2010), biofilm (Takriff 2014), and chemical coagulation (Igwe & Abia 2007; Wu et al. 2010) resulted in high expenses is the major constraint for developing countries. Therefore, a more economical feasible treatment process – adsorption appears to be an attractive option for water and wastewater treatment applications (Namasivayam & Kavitha 2002). Adsorption, being a simple and reliable technology, has been commonly employed in water and wastewater treatment processes to remove a wide range of undesirable compounds (Ahmad et al. 2003; Ahmad & Krimly 2014; Azmi et al. 2014). Historically and presently, activated carbon (AC) has been extensively used as the adsorbent for water purification processes (Rupani & Singh 2010; Takriff 2014). The wide acceptance of AC in water and wastewater treatment processes could be attributed to its high specific surface area, high porosity, and high thermal stability (Gupta et al. 2012; Igwe & Onyegbado 2007; Kumar et al. 2014; Namasivayam & Kavitha 2002; Wu et al. 2010).

Recently, a significant evolution has been made by some researchers in greatly expanded the avenue of adsorption technology. Graphene, the latest member of carbon family is believed to be one of the most interesting materials of this century. Graphene is a single atomic layer of sp2-hybridized carbon arranged in a honeycomb structure. It has been reported that graphene (as an adsorbent) possesses excellent adsorption capacity for the removal of heavy metal ions (Pb2+, Cd2+, Cr6+, etc.) (Deng et al. 2010; Zhu et al. 2012) and dyes (methylene orange, methylene blue, rhodamine B, etc.) (Gupta & Suhas 2009). The great adsorption capacity displayed by graphene can thus convert it into adsorbent and apply in water and wastewater treatment. Gao et al. (2011) and Sreeprasad et al. (2011) had proved that chemically synthesized graphene, as well as graphene oxide, can be anchored onto the surface of river sand with suitable binder such as chitosan, to produce highly effective adsorbents. This new product has been named as graphene sand composite (GSC) (Dubey et al. 2015; Gupta et al. 2012). Gao et al. (2011) reported that GSC retained at least 5-fold higher amount of unwanted substances compared to AC with equal mass of carbon. However, the use of graphenic materials for large-scale and down to earth applications like water purification and wastewater treatment is quite challenging. This is mainly due to the difficulty in large-scale synthesis of GSC, as the carbonaceous source such as sugar is costly (Gupta et al. 2012). In general, an adsorbent can be termed as a low-cost adsorbent if it requires little processing, abundant carbonaceous source in nature, or the carbonaceous source is a by-product or waste material from industry. Hence, cheap and easily available carbonaceous source with simple synthesis method to produce GSC is utmost concern.

Palm oil industry is one of the major agro-industries in Malaysia. Malaysia produces 19.96 million tons of crude oil palms (CPO) in year 2015 after Indonesia (Antone & Spencer 2015; Loh et al. 2014; Ramli 2011). The palm oil extraction process consumes huge amount of water for steam sterilizing the fresh oil palm fruit bunches (PFB) and clarifying the extracted oil which eventually end up as palm oil mill effluent (POME), the single largest source of industrial wastewater in Malaysia (Rupani & Singh 2010). POME contains large amount of organic constituents, residual oil, and suspended particles (Igberaharha 1998). Direct disposal of POME into the waterway is disastrous to the surrounding ecosystem due to the depletion of oxygen in water body and thus suffocate the aquatic life. Due to these reasons, the concept of using POME as carbonaceous source to produce POME-based GSC (P-GSC) appears to be an attractive practice that could be economical and eco-friendly beneficial. The synthesized P-GSC can be used as the adsorbent in adsorption treatment process to remove the undesirable constituents in POME. This concept echoes with the increasingly important of sustainable development, where the waste (POME) is converted into value-added product (P-GSC) to treat the waste (POME). To the best of our knowledge, there is no study on the synthesis of GSC using POME as carbonaceous source. Hence, the novelty of this study was to investigate the feasibility of using POME as low cost and easily available carbonaceous source in synthesizing P-GSC. The efficiency of P-GSC was evaluated by analysing the removal of unwanted constituents from aerobic POME.

METHODOLOGY

MATERIALS

Raw POME and aerobic POME were collected from East Mill, Sime Darby Plantation, Carey Island, Selangor, Malaysia. Raw POME was used as the carbonaceous source for the synthesis of P-GSC while diluted aerobic POME with chemical oxygen demand (COD) at 270 mg/L was used as the feed solution in adsorption study. River sand was purchased from local hardware shop, whereas commercial AC with 8 × 30 mesh size (2.38 – 2.97 mm) was purchased from Eastnova Frontier Sdn. Bhd. Sulphuric acid (H2SO4) (purity: 95 – 98%) supplied by R&M Chemicals, United Kingdom was used as the activating agent to activate the synthesized P-GSC.

SYNTHESIS OF P-GSC

The river sand was sieved and separated into three different range of particle size: large size (1.18 – 2.36 mm), medium size (0.60 – 1.18 mm), and small size (0.30 – 0.60 mm). Prior the synthesis process, river sand was first soaked into diluted H2SO4 solution for a few minutes to remove the impurities attached onto the river sand surface. The river sand was then washed thoroughly with ultra-pure (UP) water and dried in an oven. Next, 60 mL of H2SO4 was slowly poured into 1 L of raw POME. The POME mixture was stirred for 6 hours


13

with constant heating at 300°C until the POME mixture was almost dried. 300 g of river sand was then added into the concentrated POME mixture and mixed thoroughly to initiate the coating. The POME-coated river sand was transferred into a crucible and heated in a muffle furnace at temperature programmed as below:

1. From room temperature to 100°C in 30 minutes2. From 100°C to 200°C in 30 minutes3. From 200°C to 300°C in 60 minutes4. From 300°C to 400°C in 60 minutes5. From 400°C to 500°C in 60 minutes6. Held at 500°C for 180 minutes (to ensure complete

graphinization of POME)7. From 500°C to 400°C in 60 minutes8. From 400°C to 300°C in 60 minutes

After the heating cycle, the muffle furnace was switched off and the black material inside the crucible was cooled into room temperature. The obtained black material was named as P-GSC. For activation of P-GSC, the P-GSC was soaked with concentrated H2SO4 and kept undisturbed at room temperature for 20 minutes. The activated P-GSC was then washed thoroughly with UP water. Following, the P-GSC was dried in an oven at 100°C prior characterization and adsorption study.

CHARACTERIZATION OF P-GSC

P-GSC was characterized by Field Emission Scanning Electron Microscope (FESEM), Energy Dispersive X-Ray (EDX), and X-Ray Diffraction (XRD). The surface morphology of the synthesized P-GSC was examined by FESEM, SUPRA 55VP-ZEISS (Merlin-Zeiss, Germany). Prior FESEM analysis, the outer surface of P-GSC was coated with a thin layer of platinum using the K550 sputter coater. The P-GSC was then examined under electron microscope at potential of 15 kV. Using same sample, EDX (Merlin-Zeiss, Germany) was applied to analyse the element distributed on P-GSC surface. In order to ascertain the crystallinity of P-GSC, XRD analysis was performed at the scanning range of 0 – 80° using Bruker D8 Advance AXS (Alpha Instruments Supplies & Services, Germany).

ADSORPTION STUDY

BATCH EqUILIBRIUM STUDY

Batch equilibrium study was performed using Erlenmeyer flasks. 250 mL diluted aerobic POME (adsorbate solution) was poured into Erlenmeyer flasks and added with different weight of P-GSC: 5 g, 10 g, 15 g, and 20 g respectively. These Erlenmeyer flasks were then shaken on an isothermal shaker at constant speed of 230 rpm for 7 hours to attain equilibrium stage at room temperature. Sample was taken every 1 hour to analyze the COD value of the adsorbate solution. The equilibrium adsorption capacity per gram dry weight of the adsorbent, qe (mg/g) was calculated by the following equation:

qC C

WV

F P

R

C

q Q b

C

Q

x

m

eo e

e

e o

e

o

=−

×

=−

×

= +

Rejection (%) %

log

100

1

== = +

− = −

log log log

log ( ) log.

q kC

n

q q qk

t

t

e fe

e t e1

2 303

qq k q q tt e e

= +1 1

2

(1)

Where Co and Ce are the initial and equilibrium concentration (mg/L), respectively, V is the volume of adsorbate solution (L), and W is the dry weight of the adsorbent (g).

Similar procedures were conducted to study the effect of P-GSC particle size. 250 mL of diluted aerobic POME was poured into Erlenmeyer flasks and added with pre-weight P-GSC (5 g, 10 g, 15 g, and 20 g) at different size. These Erlenmeyer flasks were then shaken on an isothermal shaker at constant speed of 230 rpm for 7 hours to attain equilibrium stage at room temperature. Sample was taken every 1 hour to analyze the COD value of the adsorbate solution. The qe (mg/g) was calculated using Eq. (1).

BATCH COLUMN STUDY

Batch column study was carried out in a chromatography column of 100 mL capacity at room temperature. The working volume of diluted aerobic POME was 40 mL while 10 cm height of adsorbent (P-GSC and AC) was packed into the column. The experimental setup for batch column study was shown in Figure 1, operated at down flow mode. The treated water was then analysed to calculate the percentage of rejection for each parameter, including COD, colour, turbidity, and TDS. The percentage of rejection (%) was calculated using Eq. (2)

qC C

WV

F P

R

C

q Q b

C

Q

x

m

eo e

e

e o

e

o

=−

×

=−

×

= +

Rejection (%) %

log

100

1

== = +

− = −

log log log

log ( ) log.

q kC

n

q q qk

t

t

e fe

e t e1

2 303

qq k q q tt e e

= +1 1

2

(2)

Where F is the initial concentration of the interested parameter in feed solution and P is the final concentration of the interested parameter in treated water at the end of the adsorption process.

FIGURE 1. Experiment setup for batch column study


14

ANALYTICAL METHODS

The treated water sample was assessing by several parameters such as COD, total dissolved solid (TDS), colour, turbidity, and pH. COD analysis was measured by preheating the sample at 150°C in Hach digital reactor RBC200 (Hach Company, Colorado, USA) for 2 hours and analysed using DR3900 benchtop spectrophotometer (Hach Company, Colorado, USA) at low range COD (3 – 150 mg/L). Colour of the sample was measured based on platinum-cobalt standard method using DR3900 benchtop spectrophotometer. Whereas, TDS and pH were measured using HI2550 benchtop meter (Hanna Instrument, Colorado, USA) by immersed the probe into the sample while turbidity of the sample was measured using 2100AN turbidity meter (Hach Company, Colorado, USA).

ADSORPTION ISOTHERMS

EqUILIBRIUM ISOTHERMS

Two equilibrium isotherms – Langmuir model and Freundlich model were used to describe the interactive behaviour between solutes and adsorbent at equilibrium stage.

LANGMUIR MODEL

Langmuir model quantitatively describes the formation of monolayer adsorbate on the outer surface of the adsorbent in which no further adsorption takes place after the first sorption on an active site. This model assumes uniform adsorption energy throughout the adsorbate surface and no transmigration of adsorbate in the plane of the surface (State et al. 2012). Based upon these assumptions, Langmuir model is expressed as in Eq. (3):

qC C

WV

F P

R

C

q Q b

C

Q

x

m

eo e

e

e o

e

o

=−

×

=−

×

= +

Rejection (%) %

log

100

1

== = +

− = −

log log log

log ( ) log.

q kC

n

q q qk

t

t

e fe

e t e1

2 303

qq k q q tt e e

= +1 1

2

(3)

Where x is the amount of adsorbate (mg), m is the weight of the adsorbent used (g), qe or x/m is the amount of adsorbate at equilibrium (mg/g), Ce is the adsorbate concentration at equilibrium (mg/L), Kf is adsorption capacity, and n is the intensity.

FREUNDLICH MODEL

Freundlich model assumes that the uptake of adsorbate occurs on heterogeneous surface by multilayer adsorption. The amount of adsorbate adsorbed onto adsorbate surface is varied due to the variation of adsorption energy at different active site. The linearized Freundlich model is presented in Eq. (4):

qC C

WV

F P

R

C

q Q b

C

Q

x

m

eo e

e

e o

e

o

=−

×

=−

×

= +

Rejection (%) %

log

100

1

== = +

− = −

log log log

log ( ) log.

q kC

n

q q qk

t

t

e fe

e t e1

2 303

qq k q q tt e e

= +1 1

2

(4)

Where x is the amount of adsorbate (mg), m is the weight of the adsorbent used (g), qe or x/m is the amount of adsorbate

at equilibrium (mg/g), Ce is the adsorbate concentration at equilibrium (mg/L), Kf is adsorption capacity, and n is the intensity.

KINETIC ISOTHERM

Two kinetic isotherms – Pseudo First Order model and Pseudo Second Order model were used to describe the rate of adsorbate uptake by P-GSC and to understand the dynamics of adsorption mechanism.

PSEUDO FIRST ORDER

Pseudo First Order of Lagergren (1898) revealed that the adsorption was preceded by diffusion through a boundary, in which the adsorbates were attached on the adsorbent surface through physical attachment (Bousba & Meniai 2013; Mahmoudi et al. 2014). The linearized form of Pseudo First Order is expressed by Eq (5):

qC C

WV

F P

R

C

q Q b

C

Q

x

m

eo e

e

e o

e

o

=−

×

=−

×

= +

Rejection (%) %

log

100

1

== = +

− = −

log log log

log ( ) log.

q kC

n

q q qk

t

t

e fe

e t e1

2 303

qq k q q tt e e

= +1 1

2

(5)

Where qe and qt are the sorption capacity at equilibrium and at time t (min), respectively (mg/g). k1 is the rate constant of Pseudo First Order model (min-1). Values of qe and k1 were calculated from ln (qe − qt) versus t graph plotted.

PSEUDO SECOND ORDER

Pseudo Second Order assumed that the adsorption process is controlled by chemical process (Can 2015) or chemisorption that involved the valence forces through sharing or exchange of electrons between adsorbate ions and adsorbent, as the limiting step in determining the adsorption rate in an adsorption process (Marczewski et al. 2013; Matouq et al. 2015; Rout et al. 2015). Pseudo Second Order was represented by Eq. (6):

qC C

WV

F P

R

C

q Q b

C

Q

x

m

eo e

e

e o

e

o

=−

×

=−

×

= +

Rejection (%) %

log

100

1

== = +

− = −

log log log

log ( ) log.

q kC

n

q q qk

t

t

e fe

e t e1

2 303

qq k q q tt e e

= +1 1

2

(6)

Where, qe, qt, and t have the same definition as explained in Eq. (5). k2 is the overall rate constant of Pseudo Second Order model (g/mg/min).

RESULTS AND DISCUSSION

CHARACTERIZATION OF P-GSC

Figure 2 shows the FESEM micrographs of small size P-GSC at the magnification of 100 × and 2.50 K×. It was observed that the surface of P-GSC was rough, uneven, and irregular with the presence of pores. This observation was in good agreement with Dubey et al. (2015) finding where similar morphology has been presented by GSC synthesized using sugar as carbonaceous source. The rough surface and the presence of pores on P-GSC provided more active sites for


15

adsorption and possibly enhanced the removal efficiency of P-GSC. Gupta et al. (2012) had mentioned that the thin sheet layers anchored on the river sand surface was actually were graphene layers. Graphene has specific surface area

(SSA) larger than typical carbon black or carbon nanotubes (Bonaccorso et al. 2015). Hence, it was possibly contributed to high adsorption capability for the synthesized P-GSC.

FIGURE 2. FESEM micrographs of small size P-GSC at the magnification of (a) 100 × and (b) 2.50 K×

Figure 3 presented the elemental analysis and EDX mapping of small size P-GSC. The associated elements – carbon (C), oxygen (O), and silica (Si) were detected as the major elements presented on P-GSC adsorbents. The weight percentage of C, O, and Si elements were found to be 47.2 wt%, 34.4 wt%, and 18.3 wt%, respectively. High weight percentage (wt%) of C element and its uniform distribution on the synthesized P-GSC surface (shown in Figure 3b) was attributed by the carbonaceous source – POME, confirming the success graphinization of fresh POME into graphene.

FIGURE 3. Elemental analysis and EDX mapping of small size P-GSC (a) Si element distribution and (b) C element distribution

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0

Cou

nts

10 20 30 40 50 60 702Theta (Coupled Two Theta/Theta) WL=1.54060

XRD was used to analyse the crystallinity of the synthesized small size P-GSC. As depicted in Figure 4, XRD spectra of P-GSC showed the highest peak at 26.42°. This confirmed the success conversion of POME into graphene where graphene was known to exhibit a strong peak at the range of 20 – 27° (Dubey et al. 2015).

BATCH EqUILIBRIUM STUDY

COD is one of the crucial parameters in wastewater treatment. Thus, batch equilibrium study was carried out in accessing the maximum number of undesirable constituents contributed to COD could adsorb onto P-GSC and AC at equilibrium stage. Figure 5 shows the adsorption profile of P-GSC and AC at different weight for COD removal within 7 hours of batch equilibrium study. It is clearly seen in Figure 5 that the adsorption underwent two-stage processes – rapid initial adsorption followed by slow adsorption until equilibrium was achieved, which is similar to the study conducted by Domga et al. (2015); Rocha et al. 2015). Rapid adsorption at the initial stage of adsorption process was possibly due to the abundant number of active sites on P-GSC surface for the adsorption to take place. After a period of time, adsorption process was slow down as equilibrium point was reached.

The adsorption capacity decreases as the weight of P-GSC and AC increases from 5 g to 20 g, irrespective to the particle size of the adsorbent. Theoretically, the increase of adsorbent’s weight should increase the adsorption capacity

FIGURE 4. XRD diffraction pattern of small size P-GSC


16

FIGURE 5. Adsorption profile of adsorbent with different weight for COD removal (a) large size P-GSC, (b) medium size P-GSC, (c) small size P-GSC, and (d) AC

due to the presence of more active sites for adsorption (Tan and Hameed 2010). However, in this study, a contradict phenomenon was observed. This could be ascribed to some of adsorption sites at high adsorbent weight were remained unsaturated (underused) during the batch equilibrium study (Domga et al. 2015). Attribute to the result in batch equilibrium study, 5 g of adsorbent is sufficient to adsorb undesirable constituents contributed to COD in diluted aerobic POME.

Not only that, trends of the adsorption profile graph showed a rapid adsorption at the initial of the process and followed by slow adsorption until the equilibrium points were achieved. But at the end of the process, the trend was decreasing due to leaching of fine black dust (graphene layers) from the P-GSC adsorbents and seeped into the feed solution. Thus, effecting the quality of the treated water.

Small size P-GSC had the highest adsorption capacity compared to medium size P-GSC and large size P-GSC. Small size P-GSC with largest surface area was expected to have more active sites for the adsorption process. Meanwhile,

the adsorption capacity for commercial AC was the lowest among the studied adsorbent. Although commercial AC is known to be more porous than river sand (Hussaro 2014), coverage of river sand with graphene layers had enhancing the P-GSC adsorption capability (Gupta et al. 2012). Thus, P-GSC synthesized from river sand coated with POME had contributed to higher adsorption capacity than commercial AC.

ADSORPTION ISOTHERMS

EqUILIBRIUM ISOTHERMS

The experimental data from batch equilibrium study was fitted into Langmuir model and Freundlich model to describe the adsorption mechanism at equilibrium stage. Constant of each model was determined using linear regression analysis and the square of the correlation coefficient (R²) was calculated. A list of equilibrium isotherm parameters together with R² values were tabulated in Table 2. Between two equilibrium


17

isotherms, Freundlich model was the most well-fitted equilibrium isotherm with R² = 0.1538, R² = 0.2366, R² = 0.9334, and R² = 0.8897 for large size P-GSC, medium size P-GSC, small size P-GSC, and AC, respectively. Freundlich model assuming the adsorption of solutes from adsorbate solution (POME effluent) onto the adsorbent’s surface (P-GSC and AC) attributed by different adsorption energy (Dubey et al. 2015; Gao et al. 2011; Ramli 2011). Solutes will form a layer when they adsorbed onto the surface of the adsorbent. Later, other solutes will have adsorbed on top of the first layer and thus heterogeneous adsorption occurs.

Kf and n are the constants for an adsorbent-adsorbate system, where Kf is Freundlich model onstant which act as an approximate indicator of adsorption capacity, while 1/n is the function of the adsorption strength (adsorption intensity) in an adsorption process (Voudrias 2002). Constant n ranges between 1 to 10 shows a beneficial adsorption process. If n is close to 1, the surface heterogeneity of adsorbent could be assumed to be less significant, whereas, if n approaches 10, the impact of surface heterogeneity becomes more significant (Khan et al. 2005; Uddin et al. 2007). As depicted in Table 1, AC was having greater Kf value as compared to P-GSC. This could be due to large surface area of AC which has microporous character and adequate porous (Chen et al. 2010). In general, as the Kf value increases, the adsorption capacity of the adsorbent, for the given adsorbate, also increases (Noroozi et al. 2007). Though, experimental data from batch equilibrium study indicates that small size P-GSC was having higher adsorption capacity than AC. This is possibly due to higher n value of Freundlich model constant for small size P-GSC. High n value of small size P-GSC signify wide distribution of bonded ions on small size P-GSC surface which was then devoted to high adsorption capacity (Igwe and Abia 2007).

cases, the correlation coefficients (R2) closed to unity. Pseudo First Order model suggested that the adsorption process was attributed by physical adsorption with the existence of Van der Waals forces between the P-GSC surface and the adsorbate solution (Sarkar et al. 2015). As Van der Waals forces is greatly affected by the friction, P-GSC with rough and uneven surface is therefore a great adsorbent for grabbing the adsorbates tightly to the active sites.

TABLE 1. Equilibrium isotherms parameters for P-GSC and AC

Isotherms P-GSC AC Large Medium Small

Langmuir Model Qo (mg/g) 0.0053 0.0040 0.0051 0.0064 b (L/mg) 0.4124 1.5088 1.8420 0.2340 R² 0.0391 0.0476 0.8646 0.7399Freundlich Model Kf (mg/g) 0.0000 0.0003 0.0040 4.4031 n 0.0840 0.2594 0.3462 0.1154 R² 0.1538 0.2366 0.9334 0.8897

KINETIC ISOTHERMS

Pseudo First Order and Pseudo Second Order models were used to govern the solute uptake rate in adsorbent-adsorbate system for this study. The kinetic isotherm’s parameters together with R² values were summarized in Table 2, Table 3, Table 4, and Table 5. Judging on the correlation coefficient, R2 values obtained from Pseudo First Order model and Pseudo Second Order model, Pseudo First Order model was the well-fitted kinetic isotherm because for most of the

TABLE 2. Kinetic isotherms parameters for 5 g of P-GSC and AC


Pseudo First Order model qe (mg/g) 1.0392 0.6994 1.6362 1.4493 k1 (mg-1) 0.3392 0.1789 0.5893 0.1350 R² 0.6343 0.0755 0.6049 0.4715Pseudo Second Order model qe (mg/g) 0.0019 0.0008 -0.0040 0.0018 k2 (g/mg/min) 9251.2700 10705.0600 14812.9300 1683.8700 R² 0.0298 0.1727 0.0001 0.3684



Pseudo First Order model qe (mg/g) 0.9825 0.5780 0.8962 0.7182 k1 (mg-1) 0.8535 0.0140 0.2195 0.3551 R² 0.7913 0.0002 0.1877 0.4157Pseudo Second Order model qe (mg/g) 0.0005 0.0006 0.0009 0.0008 k2 (g/mg/min) 8657.7300 10900.2500 4957.4800 10222.5400 R² 0.3927 0.1462 0.2731 0.334



Pseudo First Order model qe (mg/g) 0.9162 0.8173 0.5131 0.4652 k1 (mg-1) 1.1918 0.8841 0.1490 0.1414 R² 0.8490 0.3062 0.0716 0.0033 Pseudo Second Order model qe (mg/g) 0.0008 0.0007 0.0009 0.0005 k2 (g/mg/min) 3821.8700 11224.2900 7295.9400 10920.7400 R² 0.1385 0.0536 0.3225 0.1006



Pseudo First Order model qe (mg/g) 0.9768 1.1147 0.8938 0.8555 k1 (mg-1) 3.0811 0.3001 0.5771 2.7530 R² 0.8623 0.7500 0.4700 0.8396 Pseudo Second Order model qe (mg/g) -0.0134 -0.0011 0.0075 0.0040 k2 (g/mg/min) 38.1200 8573.3000 531.5800 251.6800 R² 0.0486 0.2447 0.0073 0.0278


18

BATCH COLUMN STUDY

Batch column study was conducted to access performance of the synthesized P-GSC in treating diluted aerobic POME. The performance of the synthesized P-GSC was benchmarked by commercial AC. Several parameters such as COD, colour, turbidity, and TDS were employed to evaluate the efficiency of the adsorbent in batch column study. The parameters of feed solution (diluted aerobic POME) and treated water after the batch column study were tabulated in Table 6 whereas Figure 6 shows the percentage of rejection of each parameter after the experiments.

Repeating coating was then significantly hindered the active sites on large size P-GSC and medium size P-GSC which in turn reduced its efficiency in adsorption process. Whereas, for small size P-GSC, the coated process was completed within one cycled.

On the other hand, commercial AC only managed to remove 75.4% of COD, 59.3% of colour, and 17.4% of turbidity, which were lower than small size P-GSC. However, pH and TDS of the treated water had increased, which could be attributed by the residual acid on AC that seeped into the treated water. The presence of graphene layers on river sand surface has granted a better adsorption capability for the small size P-GSC adsorbent.

CONCLUSION

POME has been successfully employed as low cost and easily available carbonaceous source in synthesizing P-GSC. Small size P-GSC has shown the greatest performance in treating diluted aerobic POME effluent with 94.7%, 92.3%, 83.3%, and 51.5% of removal for COD, colour, turbidity, and TDS, respectively for batch column study. This could be attributed to its larger surface area and the presence of adsorptive graphene, which provided more adsorption sites and higher adsorption capacity towards undesirable constituents. According to batch equilibrium study, the adsorption mechanism and rate of adsorption of P-GSC can be described by Freundlich model and Pseudo First Order model, respectively. This has illustrating the uptake of adsorbate solutions by P-GSC was a multilayer adsorption through physical attachment where the rate of occupation of sorption sites was proportional to the number of unoccupied sites. As conclusion for this study, POME can be used as a cheap carbonaceous source to produce P-GSC that has better adsorption capacity. The concept of sustainable waste-to-treat-waste has been proven in this study. The findings obtained in this study will promote the sustainable practice in oil palm industry by converting the waste into value-added product.

ACKNOWLEDGEMENT

The authors wish to gratefully acknowledge the financial support for this work by Skim Geran Penyelidikan Fundamental (FRGS/1/2015/TK02/UKM/01/1) and Dana Penyelidikan Strategik (KRA-2017-016).

REFERENCES

Ahmad, A. & Mohd, Z. K. 2014. Palm oil mill effluent treatment process evaluation and fate of priority components in an open and closed digestion system. Current World Environment 9(2): 321-330.

Ahmad, A. L., Suzylawati, I. & Subhash, B. 2003. Water recycling from palm oil mill effluent (POME) using membrane technology. Desalination 157(1): 87-95.

TABLE 6. Treated water parameters after adsorption test

P-GSC Parameters Feed AC Large Medium Small

COD (mg/L) 270.0 153.0 157.0 14.0 66.0Colour (PtCo) 1674.0 1037.0 937.3 129.7 681.3Turbidity (NTU) 34.8 29.6 29.5 5.8 28.7TDS (mg/L) 873.3 739.0 735.7 424.0 897.0pH 8.1 8.1 7.9 6.8 9.6

FIGURE 6. Percentage of rejection of each parameter for different particle sizes of P-GSC and commercial AC

Figure 6 revealed that small size P-GSC was the best in removing the parameters of concern, with 94.7%, 92.3%, 83.3%, and 51.5% of removal for COD, colour, turbidity, and TDS, respectively. While, the least performance was attributed by large size P-GSC, with only 43.2%, 38.0%, 15.0%, and 15.4% of removal for COD, colour, turbidity, and TDS, respectively. Although all P-GSC with different particle size was synthesized using the same procedures, but the greater surface area of small size P-GSC has provided more activation sites for solutes to adsorbed onto it (Gupta et al. 2012). Consequently, small size P-GSC had a better removal capability compared to medium size and large size P-GSC.

Additionally, Figure 6 illustrated that the medium size and large size P-GSC were having the similar performance in batch column study. This is mainly due to the difficulty in coating and anchoring the POME onto the large size and medium size river sand. Several coatings should be done to fully coat the large size and medium size river sand surface.


19

Antone, D. & Spencer, S. R. 2015. Organising for sustainable palm oil consumption: a market-based approach. Consumption Markets and Culture, 18(1): 55-71.

Azmi, N. S., Khairul, F. & Md. Y. 2014. Wastewater treatment of palm oil mill effluent (POME) by ultrafiltration membrane separation technique coupled with adsorption treatment as pre-treatment. Agriculture and Agricultural Science Procedia 2 257-264.

Bonaccorso, F., Colombo, L., Yu, G., Stoller, M., Tozzini, V., Ferrari, A. C., Ruoff, R. S. & Pellegrini, V. 2015. Graphene, related two-dimensional crystals & hybrid systems for energy conversion and storage. Science 347(6217): 1246501.

Bousba, S. & Abdeslam, H. M. 2013. Adsorption of 2-chlorophenol onto sewage sludge based adsorbent: equilibrium and kinetic study. Chemical Engineering Transactions 35: 859-864.

Can, M. 2015. Studies of the kinetics for rhodium adsorption onto gallic acid derived polymer: the application of nonlinear regression analysis. Acta Physica Polonica A 127(4): 1308-1310.

Chen, S., Jian, Z., Chenglu, Z., qinyan, Y., Yan, L. & Chao, L. 2010. Equilibrium and kinetic studies of methyl orange and methyl violet adsorption on activated carbon derived from phragmites australis. Desalination 252(1-3): 149-156.

Deng, X. J., Lu, L. L., Li, H. W. & Luo, F. 2010. The adsorption properties of Pb(II) and Cd(II) on functionalized graphene prepared by electrolysis method. Journal Hazard Materials 183(1-3): 923-30.

Domga, R., Massai, H., Constant, T., Jean, B. T., Arbam, T., Nathalie, K. D. & Dangwang, D. 2015. Batch equilibrium, kinetic, and thermodynamic studies on adsorption of methylene blue in aqueous solution onto activated carbon prepared from bos indicus gudali bones. Chemistry Journal 1(6): 172-181.

Dubey, R., Bajpai, J. & Bajpai. A. K. 2015. Green synthesis of graphene sand composite (GSC) as novel adsorbent for efficient removal of Cr(VI) ions from aqueous solution. Journal of Water Process Engineering 5: 83-94.

Gao, W., Majumder, M., Alemany, L., Narayanan, B., Ibarra, T. N., Pradhan, M. A. & Ajayan, B. K. 2011. Engineered graphite oxide materials for application in water purification. ACS Applied Materials and Interfaces 3(6): 1821-1826.

Gupta, S. S., Theruvakkattil, S. S., Shihabudheen, Mundampra, M., Sarit, K. & Thalappil, P. 2012. Graphene from sugar and its application in water purification. ACS Applied Materials and Interfaces 4(8): 4156-63.

Gupta, V. & Suhas, K. 2009. Application of low-cost adsorbents for dye removal - a review. Journal of Environmental Management 90(8): 2313-42.

Hussaro, K. 2014. Preparation of activated carbon from palm oil shell by chemical activation with Na2CO3 and ZnCl2 as imprenated agents for H2S adsorption. American Journal of Environmental Science 10(4): 336-346.

Igberaharha, L. O. 1998. Zero discharge layout of palm oil mill effluent (POME). Research Project, Department of Chemical Engineering University of Port Harcourt 1-5.

Igwe, J. C. & Onyegbado, C. C. 2007. A review of palm oil mill effluent (POME) water treatment. Global Journal of Environmental Resources 1: 54-62.

Igwe, J. C. & Abia, A. A. 2007. Adsorption isotherm studies of Cd(II), Pb(II), and Zn(II) ions bioremediation from aqueous solution using unmodified and EDTA-modified maize cob. Ecletica Quimica 32(1): 33-42.

Keno, D. K., Erastus, G., Godfrey, O. M. & Eliakim, M, M. 2016. Kinetic, sorption isotherms, pseudo-first-order model and pseudo-second-order model studies of Cu(II) and Pb(II) using defatted moringa oleifera seed powder. The Journal of Phytopharmacology 5(2): 71-78.

Khan, A. R., Tahir, H., Uddin, F. & Uzma, H. 2005. Adsorption of methylene blue from aqueous solution on the surface of wool fiber and cotton fiber. Journal of Applied Science and Environmental Management 9(2): 29-35.

Kumar, B. A., Renu, D. & Jaya, B. 2015. Graphene sand composite (GSC) as novel adsorbent for efficient removal of Cr(VI) ions from aqueous solutions. Journal of Water Process Engineering 5: 83-94.

Loh, S. K., Muzzammil, N., Weng, S. L. & Yuen, M. C. 2013. Zero discharge treatment technology of palm oil mill effluent. Journal of Oil Palm Research 25(Dec): 273-281.

Madaki, Y. S. & Lau, S. 2013. Palm oil mill effluent (POME) from malaysia palm oil mills: waste or resource. International Journal of Science Environment and Technology 2(6): 1138-1155.

Mahmoudi, K., Noureddine, H. & Ezzeddine, S. 2014. Preparation and characterization of activated carbon from date pits by chemical activation with zinc chloride for methyl orange adsorption. Environmental Science 5(6): 1758-1769.

Marczewski, A. W., Deryło-Marczewska, A. & Agata, S. 2013. Adsorption and desorption kinetics of benzene derivatives on mesoporous carbons. Adsorption 19(2-4): 391-406.

Matouq, M., Nina, J., Mohammed, q., Muna, H. & Maha, q. A. 2015. The adsorption kinetics and modelling for heavy metals removal from wastewater by moringa pods. Journal of Environmental Chemical Engineering 3(2): 775-784.

Namasivayam, C. & Kavitha, D. 2002. Removal of congo red from water by adsorption onto activated carbon prepared from coir pith, an agricutural solid waste. Dyes and Pigments 54(1): 47-58.

Noroozi, B., Sorial, G. A., Bahrami, H. & Arami, M. 2007. Equilibrium and kinetic adsorption study of a cationic dye by a natural adsorbent-silkworm pupa. Journal of Hazardous Materials 139(1): 167-174.

Ramli, A. 2011. World palm oil supply, demand, price and prospects - focus on malaysian and indonesian palm


20

oil industries. Oil Palm Industry Economic Journal 11: 13-25.

Rocha, D., Pablo, A., Franca, S. & Leandro, S. 2015. Batch and column studies of phenol adsorption by an activated carbon based on acid treatment of corn cobs. International Journal of Engineering and Technology 7(6): 459-464.

Rout, S., Ajay, K., Pazhayath, M. R., Raj M. T., Pazhayath, M. R. & Raj, M. T. 2015. Pseudo second order kinetic model for the sorption of U(VI) onto soil: a comparison of linear and non-linear methods. International Journal of Environmental Sciences 6(1): 145-154.

Rupani, P. & Singh, R. 2010. Review of current palm oil mill effluent (POME) treatment methods: vermicomposting as a sustainable practice. World Applied Sciences Journal 10(1): 1190-1201.

Sarkar, S., Chiranjib, B. & Stefano, C. 2015. Studies on adsorption, reaction mechanisms, and kinetics for photocatalytic degradation of CHD, a pharmaceutical waste. Ecotoxicology and Environmental Safety 121: 154-163.

Sreeprasad, T. S., Maliyekkal, S. M., Lisha, K. P. & Pradeep, T. 2011. Reduced graphene oxide–metal/metal oxide composites: facile synthesis and application in water purification. Journal of Hazardous Materials 186(1): 921-31.

State, K., Ekiti, S. & Kwara, S. 2012. Langmuir, freundlich, temkin and dubinin - radushkevich isotherms studies of equilibrium sorption of Zn2+ unto phosphoric acid modified rice husk. Journal of Applied Chemistry 3(1): 38-45.

Takriff, M. 2014. A review of biofilm treatment systems in treating downstream palm oil mill effluent (POME). Journal of Applied Sciences 14(12): 1334-1338.

Tan, I., Abdul, W. & Hameed, B. H. 2010. Adsorption isotherms, kinetics, thermodynamics and desorption studies of basic dye on activated carbon derived from oil palm empty fruit bunch. Journal of Applied Sciences 10(21): 2565-2571.

Uddin, M. T., Islam, M. S. & Abedin, M. S. 2007. Adsorption of phenol from aqueous solution by water hyacinth ash. Engineering and Applied Sciences 2(2): 121-128.

Voudrias, E., Fytianos, F. & Bozani, E. 2002. Sorption description isotherms of dyes from aqueous solutions and wastewaters with different sorbent materials. Global Nest, The International Journal 4(1): 75-83.

Wu, T. Y., Abdul, W. M., Jamaliah, M. J. & Nurina, A. 2010. Pollution control technologies for the treatment of palm oil mill effluent (POME) through end-of-pipe processes. Journal of Environmental Management 91(7): 1467-1490.

Zhu, J., Wei, S., Gu, H., Rapole, S. B., Wang, q., Luo, Z., Haldolaarachchige, N., Young, D. P. & Guo, Z. 2012. One-pot synthesis of magnetic graphene nanocomposites decorated with core at double–shell nanoparticles for fast chromium removal, environment. Science Technology, 46(2): 977-85.

Wan Nur Athirah Wan Mohammad Hamdan, *Teow Yeit Haan, Abdul Wahab MohammadChemical Engineering Programme, Research Centre for Sustainable Process Technology (CESPRO), Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia.

*Corresponding author; email: [email protected]

Received date: 13th April 2018Accepted date: 17st July 2018Online First date: 1st October 2018Published date: 30th November 2018


sustainable approach in palm oil industry – green …kilang minyak kelapa sawit untuk rawatan air...

Documents