Industrial Crops and Products 56 (2014) 1726

Contents lists available at ScienceDirect

Industrial Crops and Products

journal homepage: www.elsevier.com/locate/indcrop

Optimization of agro-industrial wastewater treatment using

unmodied rice starch as a natural coagulant
Chee Yang Teh a , Ta Yeong Wu a, , Joon Ching Juan b
a
Chemical Engineering Discipline, School of Engineering, Monash University, Jalan Lagoon Selatan, Bandar Sunway, 46150, Selangor Darul Ehsan, Malaysia
b
Nanotechnology & Catalysis Research Centre (NANOCAT), University of Malaya, 50603, Kuala Lumpur, Malaysia

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

Article history: There have been increasing interests on the use of organic starch derivative coagulants such as cationic
Received 9 September 2013 starch in coagulation-occulation process due to environment and health concerns regarding the use of
Received in revised form 20 January 2014 inorganic coagulants. However, studies on the potential utilization of unmodied starch in wastewater
Accepted 16 February 2014
treatment remain limited. In this study, treatment performance of unmodied rice starch and alum
Available online 16 March 2014
was tested on agro-industrial wastewater produced from crude palm oil extraction, namely palm oil
mill efuent (POME). Response surface methodology (RSM) showed that an addition of unmodied rice
Keywords:
starch not only enhanced total suspended solids (TSS) and chemical oxygen demand (COD) removals;
Alum
Coagulation-occulation
but it also signicantly improved the process by reducing both the dosage of alum (47.95%) and settling
Palm oil mill efuent (POME) time (58.66%), while shifting the operating pH closer to the natural pH of POME. The treatment enabled
Response surface methodology (RSM) TSS and COD removals up to 86.65 and 49.23%, respectively under the optimum conditions of 0.38 g/L
Rice starch alum, 0.28 g/L unmodied rice starch, pH 4.45 and settling time of 5.54 min. An addition of unmodied
Sludge rice starch to alum also resulted in 33.8% reduction of chemical cost as compared to the treatment using
alum alone.
2014 Elsevier B.V. All rights reserved.

1. Introduction
Over the years, agro-industries have been one of the major con-
tributors in the world economy especially for developing nations
which often rely on subsistence farming and income generation
from formal and/or informal agro-industries. At the same time,
this industry is also one of the main contributors to worldwide
industrial pollution problem. With such a drastic rate of population
growth, increasing in per-capita energy consumption and the over-
exploitation of natural resources, a rise of highly complex wastes
generated from agro-industries is often inevitable. Although the
composition and quantity of agro-industrial waste depend on the
source of raw materials, nature of the products, operations and
processing steps, it generally contains large amount of organics
materials with high values of chemical oxygen demand (COD) and
total suspended solids (TSS) which could potentially cause severe
pollution problems (Prasertsan et al., 2007). If the untreated waste
is released into the environments, it is certain to cause signicant
environmental problems due to its accumulation in soil and water
environments (Wu et al., 2013a).
Corresponding author. Tel.: +60 3 55146258; fax: +60 3 55146207.
E-mail addresses: wu.ta.yeong@monash.edu, tayeong@hotmail.com (T.Y. Wu).
Wet process of palm oil milling has been the most common
method to extract crude palm oil for large scale productions. How-
ever, some of the processes such as sterilization of fresh oil palm
fruit bunches, palm oil clarication and hydrocyclone operations
generate a thick colloidal mixture of agro-industrial wastewater
known as palm oil mill efuent (POME) (Wu et al., 2010). Crude
palm oil extraction produces approximately 50 million tonnes of
POME annually in Malaysia (Wu et al., 2013b) and this gure is
expected to rise in the future. Improper discharge of this highly
polluting wastewater is causing environmental problems due to
its high polluting characteristics with considerable amount of COD
(53,630 mg/L) and TSS (19,020 mg/L) (Wu et al., 2009a). Pond-
ing system is the most typical method used in POME treatment
but it has many disadvantages such as long retention time, large
treatment area, accumulations of large amounts of sludge and
uncontrollable release of methane gas into the atmosphere (Wu
et al., 2010; Lim et al., 2014). In many countries, waste manage-
ment systems are undergoing changes due to the threat of global
climate change and other environmental issues (Nouri et al., 2012).
Hence, continuous studies on other treatment processes will pro-
vide invaluable insights towards improving the current treatment
of POME (Wu et al., 2010).
As one of the most important processes to treat both urban
and industrial wastewater, coagulation-occulation could be a

http://dx.doi.org/10.1016/j.indcrop.2014.02.018
0926-6690/ 2014 Elsevier B.V. All rights reserved.
18 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726

potential alternative for POME treatment. However, there are con- Table 1
Experimental range and levels of independent variables for treatment using alum
cerns regarding the use of inorganic metal salts (e.g. alum) in
or unmodied rice starch alone.
coagulation-occulation process of wastewater. Besides producing
large volumes of sludge, overdosing of these chemicals increases Variables Symbols Coded levels
metal ions content in coagulated water and can potentially cause Actual Coded 1.682 1 0 1 1.682
health hazards (Renault et al., 2009). Increasing awareness of this
Alum alone
issue shifted interest towards using natural plant-based materials Alum dosage (g/L) X1 x1 0.16 0.30 0.50 0.70 0.84
such as chitosan (Ahmad et al., 2006), Moringa oleifera seeds (Bhatia Initial pH X2 x2 1.14 2.50 4.50 6.50 7.86
et al., 2007), guar gum (Pritchard et al., 2009), tannin (Beltrn- Settling time (min) X3 x3 1.59 5.00 10.00 15.00 18.41
Heredia et al., 2010), Jatropha curcas seeds (Abidin et al., 2013), Unmodied rice starch alone
mustard seeds (Bodlund et al., 2013) and other common vegeta- Unmodied rice Y1 y1 0.00 0.20 0.50 0.80 1.00
bles and legumes (Choy et al., 2013) in coagulation-occulation starch dosage
(g/L)
treatment of raw water and wastewater.
Initial pH Y2 y2 1.48 2.50 4.00 5.50 6.52
Starch, one of the most abundant natural polymers in the world, Settling time (min) Y3 y3 0.31 1.50 3.25 5.00 6.19
is an interesting material to be used as a coagulant. According to
Kraak (1992), starch products have a special role in purication
process for drinking water through occulation. In its crude form,
it consists of a mixture of two polymers of anhydroglucose units, Coagulation-occulation experiments were carried out using a
amylose and amylopectin (Wei et al., 2008). Due to its renewability, occulator (VELP Scientica Flocculator JLT4) in 500 mL beakers.
biodegradability and low cost, starch is often modied chemically pH of the POME (300 mL) was rstly adjusted using 1 mol/L HCl or
or biochemically to produce coagulant. Currently, cationic starch NaOH. The beaker was then placed in the occulator and different
is one of the most commonly investigated starch derivative coagu- dosage of coagulant (alum or unmodied rice starch) was added.
lants to be used to effectively treat organic and inorganic matters in The sample was then stirred rapidly at 150 rpm for 5 min, followed
wastewater carrying negative charge (Wei et al., 2008). However, by slow stirring at 40 rpm for 15 min. For combination use of coag-
potentially hazardous chemicals such as formaldehyde, caustic ulant and coagulant aid, specic dosages of alum and unmodied
soda and various solvents are often used for starch modications. rice starch were added at the beginning of rapid and slow stirring,
Until now, study on the optimized use of unmodied starch respectively. The ocs were then allowed to settle for a specic
as a natural coagulant still remains scarce. Selective occulation duration of time. Analysis samples were taken from approximately
of dark-colored minerals (DCM) using starch by Dogu and Arol 2 cm below the water level for TSS and COD analysis.
(2004) was among the few studies done in the previous years. TSS was measured using photometric method (HACH Method
Due to the elimination of chemical modication steps, the use of 8006), while COD was determined by using United States Envi-
unmodied starch to treat real industrial wastewater could be a ronmental Protection Agency (USEPA) reactor digestion method
better option as compared to its modied counterparts, either as (HACH Method 8000). The removal efciency from the treatment
a primary coagulant or coagulant aid to reduce the risk caused process was calculated by using equation (1) as stated below:
by the use of metal coagulants. Thus, the main aim of this work
was to improve and optimize the overall efciency and process
 Cnal

Removal efciency(%) = 1 100% (1)
conditions in coagulation-occulation treatment of POME using Cinitial
(1) alum, (2) unmodied rice starch and (3) combination between
alum and unmodied rice starch. Accurate evaluations on the treat- where Cinitial and Cnal are the concentration of TSS or COD in POME
ment performances were done by optimizing coagulant/coagulant before and after the treatment, respectively.
aid dosage, initial pH and settling time for each treatment through
response surface methodology (RSM). This study would provide a
better understanding on the feasibility of using unmodied rice 2.2. Experimental design
starch in agro-industrial wastewater treatment as well as the
potential to combine and reduce the amount of alum during the Traditionally, one-factor-at-a-time technique was used to opti-
treatment of POME. mize the treatment process, in which one parameter would be
varied while the others were kept at constant level. However, this
technique not only fails to identify possible interactions between
2. Materials and methods variables, but is also time-consuming and expensive because large
number of experiments must be carried out (Wu et al., 2009b).
2.1. Chemicals, collection of POME and procedures To overcome these drawbacks, RSM was used in this study for
analyzing and modeling the effects of multiple variables and their
Aluminum sulfate octadecahydrate (Al2 (SO4 )3 18H2 O) and rice interactions using a limited number of experiments. Recently, RSM
starch were purchased from SigmaAldrich Corporation. Both is used widely in waste treatment and management (Jadhav et al.,
analytical-reagent-grade chemicals were used without further 2013; Snchez-Martn et al., 2013; Usharani and Muthukumar,
purication. Rice starch solution of 30 g/L was prepared by dissolv- 2013). Central Composite Design (CCD) was selected in this study
ing the starch in distilled water. The mixture was then autoclaved to optimize the variables which were affecting the treatment ef-
using a pressure steam sterilizer (Electric Model No.50X, All Amer- ciency. Independent factors chosen for the optimization process
ican) at 121 C and 17 psi for 20 min. The solution was then kept were alum dosage or/and unmodied rice starch dosage, initial pH
at constant temperature of 80 C with agitation speed of 400 rpm and settling time. The range and levels of the chosen variables were
using a hot plate stirrer. The rice starch solution was prepared daily shown in Tables 1 and 2. Both TSS and COD removals were the
to eliminate inconsistency of results due to biodegradation. Raw dependent variables (responses) investigated in this study. Using
POME was obtained from Seri Ulu Langat Palm Oil Mill Sdn. Bhd., Design Expert (Version 6.0.10, Stat-Ease Inc., Minneapolis, USA)
Malaysia and was preserved at 4 C to prevent biodegradation. The software, all responses were tted using a predictive polynomial
initial TSS and COD of raw POME were 25,750 and 65,667 mg/L, quadratic equation to obtain a correlation of both the response and
respectively. independent variables. General form of second-order polynomial
 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726 19

Table 2 alum alone. This was due to the combined action between the
Experimental range and levels of independent variables for treatment using alum
coagulant and coagulant aid which will be discussed in subse-
plus unmodied rice starch.
quent sections. On the other hand, the treatment using unmodied
Variables Symbols Coded levels rice starch alone successfully yielded about 6 and 10% higher
Actual Coded 2 1 0 1 2 TSS removal as compared to the treatment using unmodied
rice starch plus alum and alum alone, respectively. These results
Alum dosage (g/L) Z1 z1 0.00 0.10 0.20 0.30 0.40
Unmodied rice starch Z2 z2 0.05 0.20 0.35 0.50 0.65 clearly showed that destabilization mechanism by unmodied
dosage (g/L) rice starch was effective in removing TSS in POME. However,
Initial pH Z3 z3 1.50 2.50 3.50 4.50 5.50 COD removal using unmodied rice starch alone was about 20%
Settling time (min) Z4 z4 0.00 2.00 4.00 6.00 8.00 lower than using alum alone or alum plus unmodied rice starch.
As unmodied rice starch is a natural coagulant derived from
plants, an addition of unmodied rice starch will contribute to the
quadratic model used to describe the effects of different factors on
organic matters in the wastewater and thus, COD removal was
the response was given as (Hay et al., 2012):
reduced at higher dosages of unmodied rice starch (Song et al.,

k 
k  2004).
Y = 0 + i Xi + ii Xi2 + ii Xi Xj (2)
i=1 i=1 i<j 3.2. Parameters affecting coagulation-occulation process and
improvements of treatment by using unmodied rice starch as a
where Y = response (TSS or COD removal), 0 = constant, i = linear primary coagulant or coagulant aid
coefcient, ii = quadratic coefcient, ij = interactive coefcient,
both Xi and Xj = actual factor level. Interactive effects of the inde- 3.2.1. Effect of coagulant/coagulant aid dosage
pendent variables on the dependant variables were plotted as Generally, organic polymer is used as a coagulant aid to bridge
three- and two-dimensional contour plots. Optimum treatment the coagulated particles when aluminum or iron salt has been
conditions were also determined by solving this regression equa- used as a primary coagulant in water and wastewater treatment
tion and analyzing the contour plots. Conrmation experiments (Bolto and Gregory, 2007). In this study, an introduction of 0.28 g/L
were then conducted in triplicate to verify the predicted optimum unmodied rice starch reduced 47.95% of the optimum dosage of
removals. alum needed while achieving higher TSS (86.65%) and COD (49.23%)
removals as compared to the treatment using alum alone. This
3. Results and discussion observation was concurrent with Hall (1997) who reported that
an addition of anionic polyacrylamide of low or medium charge
3.1. Improvements on TSS and COD removals by unmodied rice density and high molecular weight into slightly positive surface
starch as a coagulant aid using RSM charge ocs formed by inorganic salts might result in a substantial
lowering of alum dosage to about 4060%.
In order to evaluate the extent of both TSS and COD removals Fig. 1(ac) and Fig. 2(ac) show that an increase of alum dosage
by using unmodied rice starch as a coagulant aid, optimum enhanced both TSS and COD removals of POME. The destabiliza-
conditions for each treatment were estimated by applying CCD tion of POME colloidal particles using alum was most likely caused
experimental design based on RSM. Regression coefcient esti- by the charge neutralization and sweep occulation mechanism.
mates and the corresponding p values are presented in Table 3. Charge neutralization involves adsorption of hydrolysis product
The statistical signicance of response function using F-test showed and/or hydroxide precipitates onto the negatively charged particles
that all models for TSS and COD removals were signicant at 95% for coagulation to occur while sweep occulation involves enmesh-
condence level with p values 0.05. Analysis of variance (ANOVA) ment of particles in growing hydroxide precipitates for its removal
results for nal reduced quadratic model for different treatments (Duan and Gregory, 2003). Higher dosages gave positive effect on
were tabulated in Table 4. High R2 indicated that total variations both TSS and COD removals due to the higher amount of ocs for-
for TSS and COD removals could be represented by most of the mation. However, unlike COD removal, there exists a point where
established model with good quadratic t. Only models for COD further addition of alum would not result in a signicant increase
removal for treatment using alum alone and alum plus unmodied in TSS removal. Fig. 1(a) shows a slight decrease in TSS removal
rice starch gave R2 values lower than 0.9. The model adequate pre- when the dosage exceeded the optimum points, indicating resta-
cision ratios were greater than 4, indicating adequate signal for all bilization of particles due to charge reversal. Similar observations
response variables and the models could be used to navigate the were also reported by Jangkorn et al. (2011) where restabilization
design space. Final reduced models in terms of actual factors are of pollutants in industrial wastewater occurred when the coagulant
expressed in Table 5. dosage was greater than the optimum value due to the change of
Figs. 1 and 2 illustrate the response surface plots for TSS and pH and types of aluminum species (anionic or cationic) present in
COD removals, respectively based on the nal reduced model using the wastewater.
alum plus unmodied rice starch. Plots were obtained by holding This study also showed that TSS removal increased with an
two variable constants at each corresponding center points while increase of unmodied rice starch dosage from 0.2 to 0.5 g/L
varying the two within their range of factorial points. In most of (Fig. 1(ce)) due to higher availability of polymers present in the
the surface responses, the convex shapes obtained indicated that solution for aggregate formation with the organic matters. Desta-
there was a well-dened optimum point for removal efciencies bilization of POME particles by unmodied rice starch was mainly
within the investigated parameter ranges (Wu et al., 2009b; Tan through bridging occulation mechanism as the starch has high
et al., 2013). From the respective nal reduced quadratic equations, molecular weight and is non-ionic/weakly ionic in nature. This
optimum treatment conditions to achieve the highest TSS and COD mechanism requires the polymer chains (amylose or amylopectin)
removals were obtained (Table 6). to be adsorbed on the particle surfaces, with only a few point of
As shown in Table 6, experimental values indicated that alum attachment while the bulk of the chains projecting into the sur-
plus unmodied rice starch achieved 4% higher TSS removal by rounding solution for contact and adherence to other pollutant
using about half of the optimum dosage of alum and similar particles (Nasser and James, 2006). Attachment will continue until
COD removal was obtained as compared to the treatment using a large, rapid settling oc is formed. However, as the dosage of
20 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726

Fig. 1. Contour and response surface plots for TSS removal of POME treatment using alum plus unmodied rice starch as a function of (a) alum dosage and initial pH; (b)
alum dosage and settling time; (c) alum dosage and unmodied rice starch dosage; (d) unmodied rice starch dosage and initial pH; (e) unmodied rice starch dosage and
settling time; (f) initial pH and settling time. The other two variables were held constant at their corresponding center points.
 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726 21

Fig. 2. Contour and response surface plots for COD removal of POME treatment using alum plus unmodied rice starch as a function of (a) alum dosage and initial pH; (b)
alum dosage and settling time; (c) alum dosage and unmodied rice starch dosage; (d) unmodied rice starch dosage and initial pH; (e) unmodied rice starch dosage and
settling time; (f) initial pH and settling time. The other two variables were held constant at their corresponding center points.
22 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726

Table 3
Regression coefcient estimates and its signicance test for full model for treatment using alum alone, unmodied rice starch alone and alum plus unmodied rice starch.

TSS removal COD removal

Source Sum of squares df Mean square F value p value Source Sum of squares df Mean square F value p value

Alum alone
Model 7469.54 9 829.95 26.55 <0.0001* Model 583.10 9 64.79 5.90 0.0052*
x1 1773.51 1 1773.51 56.74 <0.0001* x1 132.11 1 132.11 12.02 0.0060*
x2 65.63 1 65.63 2.10 0.1779 x2 129.90 1 129.90 11.82 0.0063*
x3 2770.82 1 2770.82 88.65 <0.0001* x3 114.71 1 114.71 10.44 0.0090*
2
x1 1289.93 1 1289.93 41.27 <0.0001* x21 5.91 1 5.91 0.54 0.4803
x22 981.28 1 981.28 31.40 0.0002* x22 74.15 1 74.15 6.75 0.0266*
x23 978.76 1 978.76 31.32 0.0002* x23 117.05 1 117.05 10.65 0.0085*
x1 x2 12.03 1 12.03 0.38 0.5489 x1 x2 28.80 1 28.80 2.62 0.1365
x1 x3 0.85 1 0.85 0.027 0.8722 x1 x3 1.39 1 1.39 0.13 0.7290
x2 x3 131.30 1 131.30 4.20 0.0675 x2 x3 1.31 1 1.31 0.12 0.7368

Unmodied rice starch alone

Model 10028.61 9 1114.29 57.72 <0.0001* Model 2140.70 9 237.86 43.84 <0.0001*
y1 3478.48 1 3478.48 180.20 <0.0001* y1 67.88 1 67.88 12.51 0.0054*
y2 3144.51 1 3144.51 162.90 <0.0001* y2 1337.24 1 1337.24 246.49 <0.0001*
y3 1493.45 1 1493.45 77.37 <0.0001* y3 150.96 1 150.96 27.83 0.0004*
2
y1 722.47 1 722.47 37.43 0.0001* y21 91.45 1 91.45 16.86 0.0021*
2
y2 180.74 1 180.74 9.36 0.0120* y22 348.55 1 348.55 64.25 <0.0001*
y23 779.47 1 779.47 40.38 <0.0001* y23 78.71 1 78.71 14.51 0.0034*
y1 y2 17.14 1 17.14 0.89 0.3682 y1 y2 93.16 1 93.16 17.17 0.0020*
y1 y3 266.69 1 266.69 13.82 0.0040* y1 y3 14.96 1 14.96 2.76 0.1278
y2 y3 188.47 1 188.47 9.76 0.0108* y2 y3 27.23 1 27.23 5.02 0.0490*

Alum plus unmodied rice starch

Model 10134.76 14 723.91 10.54 <0.0001* Model 977.51 14 69.82 3.47 0.0113*
z1 3824.87 1 3824.87 55.69 <0.0001* z1 161.82 1 161.82 8.05 0.0125*
z2 442.56 1 442.56 6.44 0.0227* z2 97.85 1 97.85 4.87 0.0434*
z3 153.32 1 153.32 2.23 0.1559 z3 1.20 1 1.20 0.060 0.8106
z4 1215.24 1 1215.24 17.70 0.0008* z4 222.53 1 222.53 11.07 0.0046*
2
z1 1339.76 1 1339.76 19.51 0.0005* z21 8.87 1 8.87 0.44 0.5168
2
z2 370.48 1 370.48 5.39 0.0347* z22 6.56 1 6.56 0.33 0.5764
z23 975.60 1 975.60 14.21 0.0019* z23 15.89 1 15.89 0.79 0.3881
z24 1835.41 1 1835.41 26.73 0.0001* z24 291.69 1 291.69 14.50 0.0017*
z1 z2 9.73 1 9.73 0.14 0.7118 z1 z2 19.54 1 19.54 0.97 0.3399
z1 z3 1152.94 1 1152.94 16.79 0.0010* z1 z3 96.14 1 96.14 4.78 0.0450*
z1 z4 2.92 1 2.92 0.043 0.8393 z1 z4 2.98 1 2.98 0.15 0.7059
z2 z3 0.0081 1 0.0081 0.000118 0.9915 z2 z3 0.92 1 0.92 0.046 0.8334
z2 z4 3.22 1 3.22 0.047 0.8314 z2 z4 0.92 1 0.92 0.046 0.8334
z3 z4 24.60 1 24.60 0.36 0.5584 z3 z4 50.62 1 50.62 2.52 0.1335
*
Signicant variable.

unmodied rice starch approached 0.5 g/L, any further increase in of pollutants (Bolto and Gregory, 2007). As shown in Fig. 2(ce),
dosage led to restabilization of the particles. When particle surfaces an addition of unmodied rice starch slightly decreased the COD
were highly covered with the polymers, insufcient adsorption removal due to the added organic matters from unmodied rice
sites resulted in little or no further improvement on removals starch into POME.

Table 4
ANOVA results for nal reduced quadratic model for the treatments.

TSS removal COD removal

Source Sum of squares df Mean square F value p value Source Sum of squares df Mean square F value p value

Alum alone
Model 7325.36 6 1220.89 34.75 <0.0001 Model 545.68 5 109.14 10.37 0.0003
Lack of t 381.56 8 47.70 3.17 0.1095 Lack of t 100.20 9 11.13 1.18 0.4496
Pure error 75.16 5 15.03 Pure error 47.08 5 9.42
Total 7782.09 19 Total 692.96 19
R2 = 0.9413, adjusted R2 = 0.9142, adequate precision = 15.874 R2 = 0.7875, adjusted R2 = 0.7116, adequate precision = 10.872

Unmodied rice starch alone

Model 10011.47 8 1251.43 65.50 <0.0001 Model 2125.74 8 265.72 42.23 <0.0001
Lack of t 175.67 6 29.28 4.24 0.0671 Lack of t 56.85 6 9.47 3.83 0.0810
Pure error 34.51 5 6.90 Pure error 12.37 5 2.47
Total 10221.65 19 Total 2194.95 19
R2 = 0.9794, adjusted R2 = 0.9645, adequate precision = 28.368 R2 = 0.9685, adjusted R2 = 0.9455, adequate precision = 24.654

Alum plus unmodied rice starch

Model 10094.27 9 1121.59 20.95 <0.0001 Model 868.51 6 144.75 8.11 <0.0001
Lack of t 988.04 15 65.87 3.99 0.0671 Lack of t 362.42 18 20.13 2.09 0.2128
Pure error 82.59 5 16.52 Pure error 48.22 5 9.64
Total 11164.90 29 Total 1279.16 29
R2 = 0.9041, adjusted R2 = 0.8610, adequate precision = 15.398 R2 = 0.6790, adjusted R2 = 0.5952, adequate precision = 12.484
 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726 23

Table 5
Final reduced quadratic model for POME treatment.

Treatment Responses Final reduced quadratic model

Alum only TSS removal 113.29 + 293.50X1 + 19.66X2 + 9.44X3 236.52X12 2.06X22 0.33X32
COD removal 2.66 + 15.55X1 + 6.50X2 + 2.81X3 0.55X22 0.11X32

Unmodied rice starch only TSS removal 45.18 + 167.61Y1 + 8.48Y2 + 34.48Y3 78.67Y12 1.574Y22 2.40Y32 11.00Y1 Y3 1.85Y2 Y3
COD removal 31.15 + 65.75Y1 + 16.96Y2 + 9.67Y3 27.99Y12 2.19Y22 0.76Y32 7.58Y1 Y2 0.70Y2 Y3

Alum plus unmodied rice starch TSS removal 54.80 + 108.69Z1 + 142.97Z2 + 22.24Z3 + 19.92Z3 698.90Z12 163.34Z22 5.96Z32 2.05Z42 + 84.89Z1 Z3
COD removal +46.67 59.83Z1 13.46Z2 5.13Z3 + 7.86Z4 0.79Z42 + 24.51Z1 Z3

During coagulation-occulation treatment of POME using alum
plus unmodied rice starch, lesser alum (47.95%) and unmodi-
ed rice starch (62.16%) were used in comparison with treatment
using alum or unmodied rice starch alone (Table 6). The lower
dosage required in the treatment using alum plus unmodied rice
starch could be due to the reason that at an optimized dosage, both
coagulant and coagulant aid exhibited the capability of removing
a certain amount of pollutants in POME using similar mechanism
(charge neutralization and sweep occulation for alum; bridging
occulation for unmodied rice starch) when it was present in
POME alone. In addition, a reduction of zeta potential via charge
neutralization by alum may ease the bridging occulation process
played by the unmodied rice starch due to the lower repulsion
forces between pollutant particles and polymers. Lower dosage of
unmodied rice starch during the coagulation-occulation treat-
ment of POME could also be due to the reduction of surface area
coverage required for polymer bridging because large initial ocs
were formed when pollutant particles were covered with alu-
minum hydroxo complexes during the initial addition of alum
(Narkis et al., 1991).
3.2.2. Effect of initial pH
It is desired to have the treatment process operated closer to the
natural pH of POME (pH 4.5) (Wu et al., 2010) without needing
pH adjustments. In this study, ANOVA showed that generally, initial
pH signicantly affected both TSS and COD removals of POME for
the treatment processes as indicated by the low p value (<0.05)
(Table 3).
Table 6 shows that POME treatment using alum plus unmodi-
ed rice starch required initial pH at pH 4.45 as compared to the
treatment using alum alone (pH 5.31) or unmodied rice starch
alone (pH 2.58). Initial pH of 4.45 was desirable in POME treat-
ment because it was closer to the natural pH of this wastewater.
Typically, for a non-ionic polymer, hydrogen bonding is usually
the main mechanism in which the polymer chain adsorbs onto
the pollutant particles. For adsorption to occur, both particle and
polymer chain should be sufciently close enough due to the short
distance between the atoms linked by the hydrogen (Ibrahim and
Abdel-Khalek, 1992). Hence, the success of bridging occulation
would depend on the zeta potential of unmodied rice starch.
Wongsagonsup et al. (2005) conducted a zeta potential analysis as
a mean to determine the residual protein content of rice our dur-
ing rice starch processing. It was found that zeta potential increased
with decreasing pH from 12 to 2 regardless of protein content. Gen-
erally, in acidic conditions, lone pair of nitrogen of the amino groups
in protein is protonated and becomes positively charged (NH3 + ),
while carboxyl group of protein is neutral (COOH) (Wongsagonsup
et al., 2005). With negative zeta potential value of POME (approx.
20 mV), lower pH could accelerate the adsorption process of par-
ticles into rice starch polymer chain, resulting in higher TSS and
COD removals. On the other hand, carboxyl groups of the proteins
become negatively charged (COO ) while some amino groups
become neutral (NH3 ) at high pH (Wongsagonsup et al., 2005),
causing repulsion between particles and polymer chain, which
could discourage the bridging occulation process. Hence, lower pH
was favorable in the present study when unmodied rice starch was
used as a primary coagulant or coagulant aid in POME treatment.
At optimum pH, when alum was used at any dosage with
unmodied rice starch, hydroxoaluminum cations (monomeric and
polymeric) coexisted with aluminum hydroxide precipitates. The
monomeric metal cations could neutralize the negative charge
of pollutants through surface adsorption while polymeric metal
cations promoted binding between pollutant particles. POME pol-
lutants could also be enmeshed into the growing aluminum
hydroxide precipitates via adsorption to be effectively removed
(Canizares et al., 2009). At pH lower than the optimum, charge
neutralization was the main removal mechanism due to the pre-
dominant existence of monomeric species such as Al3+ , AlOH2+ and
Al(OH)2+ . As sweep occulation mechanism generally gives bet-
ter particle removal as compared to destabilization by just charge
neutralization (Duan and Gregory, 2003), the relatively low con-
centration of aluminum hydroxide precipitates at pH lower than
optimum resulted in lower TSS removal (Fig. 1(a)). A decrease of
removal was also observed at pH higher than optimum due to the
predominant existence of negatively charged monomeric hydrox-
oaluminum anions such as Al(OH)4 (Canizares et al., 2007).
3.2.3. Effect of settling time
The present study showed that the settling time had a profound
effect on POME treatment by either using alum or rice starch. If the
initial pH and rice dosage were xed, increasing settling time from
2 to 6 min enhanced TSS and COD removals (4460% and 3945%,
respectively) until it reached a point where increasing settling time
did not show any signicant effect on the removals (Fig. 1(b) and
Fig. 2(b)). Similar trend was also observed while varying the dosage
of unmodied rice starch from 0.2 to 0.5 g/L (Fig. 1(e) and Fig. 2(e)).
Sufcient settling time allowed the newly formed ocs to settle,
resulting in low suspended solids concentration in the supernatant.
However, prolonged settling time after the optimum will have little
effect on pollutant removal as most of the ocs were already settled.

Table 6
Comparison between experimental and predicted optimum TSS and COD values.

Treatment Conditions TSS removal (%) COD removal (%)

Alum dosage (g/L) Unmodied rice Initial pH Settling time (min) Predicted Experimental Predicted Experimental
starch dosage (g/L)

Alum alone 0.73 5.31 13.4 88.29 82.24 0.18 45.36 49.12 4.64
Unmodied rice starch alone 0.74 2.58 4.66 95.53 92.50 0.30 35.57 30.91 1.24
Alum plus unmodied rice starch 0.38 0.28 4.45 5.54 84.79 86.65 2.32 58.12 49.23 4.68
24 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726
During the treatment using unmodied rice starch alone, a sig-
nicantly shorter settling time (4.66 min) was needed to achieve
the highest TSS and COD removals as compared to the treatment
using alum alone (13.4 min). It was evident that in this study, bridg-
ing occulation produced ocs which were larger and stronger than
those produced through destabilization by reducing the repulsion
between them (by using metal salts) (Zhu et al., 2009). The high
settling velocity obtained may be due to the high molecular weight
of rice starch (Weight-average molar mass: 0.822.5 108 g/mol)
because longer molecular chain of rice starch could interact with
pollutants at longer distances, leading to a formation of ocs with
larger size (Park et al., 2007; Zhu et al., 2009).
Moreover, the use of unmodied rice starch as a coagulant aid
enabled most of the ocs to settle in just 5.54 min with TSS and COD
removals of 86.65 and 49.23%, respectively as compared 82.24 and
49.12% achieved by alum alone in 13.4 min. Optimum settling time
using alum plus unmodied rice starch was slightly longer than
the time needed in treatment using unmodied rice starch alone
(5.54 min) because lower dosage of unmodied rice starch was used
in the combination process (Table 6). It was found that the ocs
formed by alum plus unmodied rice starch were larger because the
particles, which have been undergone charge neutralization (due to
the presence of alum), were bridged more effectively by the starch
polymers. Ahmad et al. (2008) also reported that the ocs exhibited
higher settling rate when alum were used together with polyacry-
lamides in the treatment of pulp and paper mill wastewater.
3.3. Other advantages of using unmodied rice starch in
coagulation-occulation treatment of POME
Reduction of alum dosage in wastewater treatment presents
a positive benet to human health and environment. Neurotox-
icity of high levels of aluminum and epidemiological correlation
between ingested aluminum and the incidence of Alzheimers dis-
ease are some of the most accepted facts related to the continuous
use of aluminum (Bondy, 2010). Study conducted by Rondeau et al.
(2000) showed that aluminum levels above 0.1 mg/L in drink-
ing posed an elevated risk of dementia and Alzheimers disease
to elderly population. Hence, the reduction of alum dosage by
using unmodied rice starch during POME treatment (Table 6) may
reduce these risks when the treated efuent is discharged into
the watercourses. Furthermore, improper disposal of sludge con-
taining high levels of aluminum to land has an adverse effect on
plants because aluminum inhibits root growth at the organ, tissue,
and cellular level even at micromolecular concentrations (Panda
and Matsumoto, 2007). Phytotoxicity of aluminum in plants could
be minimized through reduction or elimination of metal salts in
the sludge using unmodied rice starch in coagulation-occulation
process of wastewater.
For economic evaluation between the POME treatments, cost of
coagulants and sludge managements are the two main costs in any
wastewater treatments involving coagulation-occulation process.
Generally, sludge managements consist of both sludge handling
and disposal, in which they account for a signicant part of the over-
all operating cost in wastewater treatment plant (Haydar and Aziz,
2009). In order to determine the amount of sludge produced in this
study, volume of sludge generated was observed. Both treatments
using alum alone and the combination use of alum plus unmodi-
ed rice starch produced comparable sludge volume (20 mL/L) and
thus, similar cost in terms of sludge handling and disposal was
concluded.
In a cost evaluation of coagulant, unmodied rice starch used for
POME treatment could be those produced from broken rice. Dur-
ing rice processing, 2030% of milled rice ends up as broken rice,
resulting a loss of production (Raina et al., 2005). Normally, broken
rice is used as a low-cost animal feed (Vasupen et al., 2008). Hence,
the reutilization of this by-product is highly favorable (Ugwu et al.,
2012). The costs of alum and rice starch produced from broken rice
per kg are stated in Table 7 (Lumdubwong and Seib, 2000; Rama
Rao et al., 2000; Haydar and Aziz, 2009).
An addition of unmodied rice starch as a coagulant aid reduced
the cost of coagulant by 33.8% as compared to the treatment
using alum alone to achieve similar pollutant removals in the
POME. Further investigations also showed that the treatment using
unmodied rice starch alone could result in 62.1% savings in coagu-
lant cost as compared to the treatment using alum alone. However,
higher sludge handling cost would be incurred because sludge pro-
duced in the treatment using unmodied rice starch alone was 50%
higher than the other treatments. Furthermore, COD removal of
POME using unmodied rice starch alone was lower than other
treatments (Table 6).
3.4. Possible ways for managing settled sludge
With increasing world demand of crude palm oil, rapid expan-
sion of palm oil industry with concomitant growth in wastewater
treatment plants has led to an increase of sludge production. Gener-
ally, three basic disposal routes for sludge include land application,
landll and incineration (Campbell, 2000). Of late, numerous stud-
ies have examined the zero waste concept. Waste avoidance and
reduction technology is considered to be the prime challenge rather
than the development of new waste treatment technology (Zaman,
2013).
In this study, the sludge which was produced from coagulation-
occulation treatment of POME has the potential to be transformed
into value added products. When unmodied rice starch alone was
used in the treatment, the produced sludge could be reused as a
substrate in agro-industries, either as a fertilizer or animal feed
because both the unmodied rice starch and POME contain high
organic contents and are non-toxic in nature. Proper use of the
settled sludge in land environment not only improves soil fer-
tility, but it also represents a cheaper alternative to commonly
used chemical fertilizers such as phosphorus. However, an appli-
cation of this organic residue should be carried out with caution
and its composition should be rstly determined to avoid imbal-
ances in the nutrient compositions in soil which will affect plant
growth (Embrandiri et al., 2012). Sludge produced by unmodied
rice starch from the POME coagulation is also suitable to be reused
as dietary substitute for pigs, poultry and small ruminants. For
example, Hutagalung et al. (1977) reported that POME-based feed
consisting of palm oil sludge, cassava root meal, palm kernel cake
and grass meal could economically replace 50% of maize as feed for
growing-nishing pigs.
On the other hand, the use of alum in POME treatment ren-
ders the settled sludge to be unsuitable for reuse purposes in land
or agricultural applications. However, many studies have shown
the potential of recovering and reusing the alum embedded in the
sludge produced from coagulation-occulation process. The reuse
of alum which was recovered from the sludge not only signicantly
reduced the cost of coagulants used in wastewater treatment pro-
cess, but it also help reduce the sludge volume and its disposal cost.
For example, Xu et al. (2009) showed that the alum was effectively
recovered (84.5%) from the sludge of both water and wastewa-
ter treatment via acidication technique, in which the recovered
alum could to be reused in chemically enhanced primary treat-
ment. Besides successfully reducing 35.5% of the sludge produced,
the use of this recovered alum also achieved similar turbidity, UV254
and COD as compared to fresh alum. In addition, the use of sludge
produced by coagulation-occulation using alum for biogas gener-
ation has been investigated by Krishnan et al. (2006). They showed
that the anaerobic biomass, when subjected to varying alum dosage
in the coagulated palm oil sludge, did not exhibit inhibition as the
 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726
Table 7
Cost of coagulants.
Treatment Coagulant (s)
Alum alone Alum
Unmodied rice starch alone Unmodied rice starch*
Alum plus unmodied rice starch Alum
Unmodied rice starch*
*
Cost of unmodied rice starch was based on the cost of broken rice.
digester performance was in conformity with the regular treatment
process.
4. Conclusion
This work demonstrated the effectiveness of unmodied rice
starch to be used as a primary coagulant or coagulant aid in
coagulation-occulation treatment of POME for enhancing both
TSS and COD removals. RSM was successfully applied and convex
shapes were obtained for most of the surface responses, indicating
the existence of a well-dened optimum point for removal efcien-
cies within the investigated parameter ranges. The use of respective
nal reduced quadratic model to obtain optimum treatment con-
ditions enabled the accurate evaluation between the performance
of POME treatment using alum alone, unmodied rice starch alone
and alum plus unmodied rice starch. The treatment using alum
plus unmodied rice starch achieved TSS and COD removals up to
86.65 and 49.23%, respectively utilizing the optimum conditions of
0.38 g/L alum, 0.28 g/L unmodied rice starch, pH 4.45 and settling
time of 5.54 min. As compared to the treatment using alum alone, an
addition of unmodied rice starch resulted in 47.95% reduction of
alum dosage, 2.4 times faster settling time and a shift of optimum
pH closer to natural pH of POME to achieve higher TSS and COD
removals. Also, introducing unmodied rice starch together with
alum reduced the cost of chemicals by 33.8% as compared to the
POME treatment using alum alone. Completion of this study shows
the potential of using unmodied rice starch in POME treatment
either as a primary natural coagulant or coagulant aid.
Acknowledgments
The authors would like to thank Monash University Malaysia for
providing C.Y. Teh with a PhD scholarship.
References
Abidin, Z.Z., Mohd Shamsudin, N.S., Madehi, N., Sobri, S., 2013. Optimisation of a
method to extract the active coagulant agent from Jatropha curcas seeds for use
in turbidity removal. Ind. Crop. Prod. 41, 319323.
Ahmad, A.L., Sumathi, S., Hameed, B.H., 2006. Coagulation of residue oil and sus-
pended solids in palm oil mill efuent by chitosan, alum and PAC. Chem. Eng. J.
118, 99105.
Ahmad, A.L., Wong, S.S., Teng, T.T., Zuhairi, A., 2008. Improvement of alum and PACI
coagulation by polyacrylamides (PAMs) for the treatment of pulp and paper mill
wastewater. Chem. Eng. J. 137, 510517.
Beltrn-Heredia, J., Snchez-Martn, J., Gmez-Munoz, M.C., 2010. New coagulant
agents from tannin extracts: preliminary optimisation studies. Chem. Eng. J.
162, 10191025.
Bhatia, S., Othman, Z., Ahmad, A.L., 2007. Pretreatment of palm oil mill efuent
(POME) using Moringa oleifera seeds as natural coagulant. J. Hazard. Mater. 145,
120126.
Bodlund, I., Pavankumar, A.R., Chelliah, R., Kasi, S., Sankaran, K., Rajarao, G.K., 2013.
Coagulant proteins identied in Mustard: a potential water treatment agent. Int.
J. Environ. Sci. Technol., http://dx.doi.org/10.1007/s13762-013-0282-4.
Bolto, B., Gregory, J., 2007. Organic polyelectrolytes in water treatment. Water Res.
41, 23012324.
Bondy, S.C., 2010. The neurotoxicity of environmental aluminum is still an issue.
Neurotoxicology 31, 575581.
Campbell, H.W., 2000. Sludge management future issues and trends. Water Sci.
Technol. 41, 18.
Canizares, P., Martnez, F., Jimnez, C., Lobato, J., Rodrigo, M.A., 2007. Coagulation
and electrocoagulation of wastes polluted with colloids. Sep. Sci. Technol. 42,
21572175.
25
Unit price ($/kg) Optimum dosage (g/L) Coagulant cost, 104 ($/L)
0.43 0.73 3.14
0.16 0.74 1.18
0.43 0.38 2.08
0.16 0.28
Canizares, P., Jimnez, C., Martnez, F., Rodrigo, M.A., Sez, C., 2009. The pH as a key
parameter in the choice between coagulation and electrocoagulation for the
treatment of wastewaters. J. Hazard. Mater. 163, 158164.
Choy, S.Y., Prasad, K.M.N., Wu, T.Y., Ramanan, R.N., 2013. A review on
common vegetables and legumes as promising plant-based natu-
ral coagulants in water clarication. Int. J. Environ. Sci. Technol.,
http://dx.doi.org/10.1007/s13762-013-0446-2.
Dogu, I., Arol, A.I., 2004. Separation of dark-colored minerals from feldspar by selec-
tive occulation using starch. Powder Technol. 139, 258263.
Duan, J., Gregory, J., 2003. Coagulation by hydrolysing metal salts. Adv. Colloid.
Interface Sci. 100102, 475502.
Embrandiri, A., Singh, R.P., Ibrahim, H.M., Ramli, A.A., 2012. Land application of
biomass residue generated from palm oil processing: its potential benets and
threats. Environmentalist 32, 111117.
Hall, T., 1997. Water Treatment Processes and Practices, second ed. WRc Publica-
tions, Blagrove.
Hay, J.X.W., Wu, T.Y., Teh, C.Y., Jahim, J.M., 2012. Optimized growth of Rhodobacter
sphaeroides O.U.001 using response surface methodology (RSM). J. Sci. Ind. Res.
71, 149154.
Haydar, S., Aziz, J.A., 2009. Coagulation-occulation studies of tannery wastewater
using combination of alum with cationic and anionic polymers. J. Hazard. Mater.
168, 10351040.
Hutagalung, R.I., Chang, C.C., Toh, K.M., Chan, H.C., 1977. Potential of palm oil mill
efuent as feed for growing-nishing pigs. Planter 53, 29.
Ibrahim, S.S., Abdel-Khalek, N.A., 1992. The action of different types of corn starch
on the occulation of phosphate slimes. Miner. Eng. 5, 907916.
Jadhav, S.B., Surwase, S.N., Phugare, S.S., Jadhav, J.P., 2013. Response surface method-
ology mediated optimization of Remazol Orange decolorization in plain distilled
water by Pseudomonas aeruginosa BCH. Int. J. Environ. Sci. Technol. 10, 181
190.
Jangkorn, S., Kuhakaew, S., Theantanoo, S., Klinla-or, H., Sriwiriyarat, T., 2011. Eval-
uation of reusing alum sludge for the coagulation of industrial wastewater
containing mixed anionic surfactants. J. Environ Sci. 23, 587594.
Kraak, A., 1992. Industrial applications of potato starch products. Ind. Crop. Prod. 1,
107112.
Krishnan, V., Ahmad, D., Endut, E.M., 2006. Effect of coagulation on palm oil mill
efuent and subsequent treatment of coagulated sludge by anaerobic digestion.
J. Chem. Technol. Biotechnol. 81, 16521660.
Lim, S.L., Wu, T.Y., Clarke, C., 2014. Treatment and biotransformation of highly pol-
luted agro-industrial wastewater from a palm oil mill into vermicompost using
earthworms. J. Agric. Food Chem. 62, 691698.
Lumdubwong, N., Seib, P.A., 2000. Rice starch isolation by alkaline protease digestion
of wet-milled rice our. J. Cereal Sci. 31, 6374.
Narkis, N., Ghattas, B., Rebhun, M., Rubin, A.J., 1991. The mechanism of occulation
with aluminium salts in combination with polymeric occulants as occulant
aids. Water Supply 9, 3744.
Nasser, M.S., James, A.E., 2006. The effect of polyacrylamide charge density and
molecular weight on the occulation and sedimentation behaviour of kaolinite
suspensions. Sep. Purif. Technol. 52, 241252.
Nouri, J., Nouri, N., Moeeni, M., 2012. Development of industrial waste disposal
scenarios using life-cycle assessment approach. Int. J. Environ. Sci. Technol. 9,
417424.
Panda, S.K., Matsumoto, H., 2007. Molecular physiology of aluminum toxicity and
tolerance in plants. Bot. Rev. 73, 326347.
Park, I.M., Ibnez, A.M., Shoemaker, C.F., 2007. Rice starch molecular size and its
relationship with amylose content. Starch 59, 6977.
Prasertsan, P., Prasertsan, S., H-Kittikun, A., 2007. Recycling of agro-industrial wastes
through cleaner technology. In: Doelle, H.W., Prasertsan, P. (Eds.), Environmen-
tal BiotechnologySocio-economic Strategies for Sustainability, Encyclopedia of
Life Support Systems (EOLSS). Developed under the Auspices of the UNESCO.
Eolss Publishers, Oxford.
Pritchard, M., Mkandawire, T., Edmondson, A., ONeill, J.G., Kululanga, G., 2009.
Potential of using plant extracts for purication of shallow well water in Malawi.
Phys. Chem. Earth 34, 799805.
Raina, C.S., Singh, S., Bawa, A.S., Saxena, D.C., 2005. Effect of vital gluten and gum
Arabic on the textural properties of pasta made from pre-gelatinised broken rice
our. Food Sci. Technol. Int. 11, 433442.
Rama Rao, S.V.R., Reddy, M.R., Prarharaj, N.K., Shyam Sunder, G., 2000. Laying per-
formance of broiler breeder chickens fed various millets or broken rice as source
of energy at a constant nutrient intake. Trop. Anim. Health Prod. 32, 329
338.
Renault, F., Sancey, B., Badot, P.M., Crini, G., 2009. Chitosan for coagula-
tion/occulation processes an eco-friendly approach. Eur. Polym. J. 45,
13371348.
26 C.Y. Teh et al. / Industrial Crops and Products 56 (2014) 1726
Rondeau, V., Commenges, D., Jacmin-Gadda, H., Dartigues, J.F., 2000. Relation
between aluminum concentrations in drinking water and Alzheimers disease:
an 8-year follow-up study. Am. J. Epidemiol. 152, 5966.
Snchez-Martn, J., Beltrn-Heredia, J., Delgado-Regana, A., Rodrguez-Gonzlez,
M.A., Rubio-Alonso, F., 2013. Optimization of tannin rigid foam as adsorbents
for wastewater treatment. Ind. Crop. Prod. 49, 507514.
Song, Z., Williams, C.J., Edyvean, R.G.J., 2004. Treatment of tannery wastewater by
chemical coagulation. Desalination 164, 249259.
Tan, H.T., Dykes, G.A., Wu, T.Y., Siow, L.F., 2013. Enhanced xylose recovery from oil
palm empty fruit bunch by efcient acid hydrolysis. Appl. Biochem. Biotechnol.
170, 16021613.
Ugwu, C.U., Tokiwa, Y., Aoyagi, H., 2012. Utilization of broken rice for the production
of poly(3-hydroxybutyrate). J. Polym. Environ. 20, 254257.
Usharani, K., Muthukumar, M., 2013. Optimization of aqueous methylparathion
biodegradation by Fusarium sp in batch scale process using response surface
methodology. Int. J. Environ. Sci. Technol. 10, 591606.
Vasupen, K., Yuangklang, C., Wongsuthavas, S., Srenanul, P., Mitchaothai, J., Beynen,
A.C., 2008. Macronutrient digestibility in Kadon pigs fed diets with isonitroge-
nous amounts of various carbohydrate sources. Trop. Anim. Health Prod. 40,
249253.
Wei, Y., Cheng, F., Zheng, H., 2008. Synthesis and occulating properties of cationic
starch derivatives. Carbohyd. Polym. 74, 673679.
Wongsagonsup, R., Shobsngob, S., Oonkhanond, B., Varavinit, S., 2005. Zeta potential
() analysis for the determination of protein content in rice our. Starch 57,
2531.
Wu, T.Y., Mohammad, A.W., Jahim, J.M., Anuar, N., 2009a. A holistic approach to man-
aging palm oil efuent (POME): biotechnological advances in the sustainable
reuse of POME. Biotechnol. Adv. 27, 4052.
Wu, T.Y., Mohammad, A.W., Jahim, J.M., Anuar, N., 2009b. Optimized reuse and bio-
conversion from retentate of pre-ltered palm oil mill efuent (POME) into
microbial protease by Aspergillus terreus using response surface methodology. J.
Chem. Technol. Biotechnol. 84, 13901396.
Wu, T.Y., Mohammad, A.W., Jahim, J.M., Anuar, N., 2010. Pollution control technolo-
gies for the treatment of palm oil mill efuent (POME) through end-of-pipe
processes. J. Environ. Manage. 91, 14671490.
Wu, T.Y., Guo, N., Teh, C.Y., Hay, J.X.W., 2013a. Advances in Ultrasound
Technology for Environmental Remediation. Springer, Netherlands,
http://dx.doi.org/10.1007/978-94-007-5533-8.
Wu, T.Y., Mohammad, A.W., Lim, S.L., Lim, P.N., Hay, J.X.W., 2013b. Recent advances
in the reuse of wastewaters for promoting sustainable development. In:
Sharma, S.K., Sanghi, R. (Eds.), Wastewater Reuse and Management. Springer,
Netherlands, pp. 47103, http://dx.doi.org/10.1007/978-94-007-4942-9 3.
Xu, G.R., Yan, Z.C., Wang, Y.C., Wang, N., 2009. Recycle of alum from water treat-
ment sludge in chemically enhanced primary treatment. J. Hazard. Mater. 161,
663669.
Zaman, A.U., 2013. Identication of waste management development drivers and
potential emerging waste treatment technologies. Int. J. Environ. Sci. Technol.
10, 455464.
Zhu, Z., Li, T., Lu, J., Wang, D., Yao, C., 2009. Characterization of kaolin ocs formed
by polyacrylamide as occulation aids. Int. J. Miner. Process. 91, 9499.