Research article

1. Journal of the Taiwan Institute of Chemical Engineers 53 (2015) 15–21
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers
journal homepage: www.elsevier.com/locate/jtice
Optimization of pulp fibre removal by flotation using colloidal gas
aphrons generated from a natural surfactant
Sumona Mukherjeea
, Soumyadeep Mukhopadhyayb
, Agamuthu Pariatambya
,
Mohd Ali Hashimb,∗
, Ghufran Redzwana
, Bhaskar Sen Guptac
a
Institute of Biological Sciences, University of Malaya, 50603, Kuala Lumpur, Malaysia
b
Department of Chemical Engineering, University of Malaya, 50603, Kuala Lumpur, Malaysia
c
School of the Built Environment, Heriot-Watt University, Edinburgh Campus, Currie EH14 4AS, UK
a r t i c l e i n f o
Article history:
Received 7 January 2015
Accepted 26 February 2015
Available online 14 March 2015
Keywords:
Colloidal gas aphrons (CGAs)
Flotation
Dispersion
Sapindus mukorossi
Saponin
Paper fibre recovery
a b s t r a c t
Colloidal gas aphrons (CGAs) are a system of highly stable micro bubbles in colloidal state. In this study, the
CGAs prepared from a natural surfactant saponin, extracted from the fruit pericarp of Sapindus mukorossi or
soapnut plant, was utilized for the recovery of pulp fibres from paper machine backwater in a flotation column.
The performance of soapnut CGAs was compared with that of CGAs generated from cationic, anionic and non-
ionic surfactants. Performance optimization of soapnut CGAs was undertaken using central composite design
(CCD). CGAs characterization showed that soapnut surfactant produced the most stable CGAs. Under various
CGAs sparging rate, pH and flow rate of wastewater, soapnut CGAs performed best by removing up to 60%
total suspended solids (TSS) from paper machine effluent as compared to 50%, 37% and 30% TSS removal by
cationic, anionic and non-ionic surfactants respectively. Optimized TSS removal of 76% was attained through
CCD at soapnut CGAs sparging rate of 0.013 L/min, wastewater flow rate of 16 L/min and pH of 7.5. Treatment
of effluent using natural surfactant CGAs is a cost effective and green process which can be replicated in
industries.
© 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
1. Introduction
Colloidal gas aphrons (CGAs) are a system of microbubbles mostly
above 25 μm diameter and classified as kugelschaums or “ball foams”,
first described by Sebba [1]. CGAs can be generated by high speed stir-
ring of the surfactant solution (ß6000 rpm), whereby air is entrapped
and microbubbles are formed. The CGAs are suitable for process ap-
plications due to their ability to adsorb particles at microbubble inter-
faces, their large interfacial area and their stability during transport
for enhanced mass transfer [1]. Earlier, CGAs had been applied for
the separation of fine particles through flotation process in a column
[2–6]. Froth flotation process using foams has several advantages over
other processes particularly in the removal of fine particles, which do
not have practical settling rates under gravity, and in the separation
of light particles which tend to float. Foam, however is hard to be
pumped as it loses its characteristics due to its rheology while CGAs
can be easily pumped.
Paper production is a highly water intensive process and conse-
quently generates large quantity of waste comprising fine pulp fibres
which escape through the fine wire mesh on which paper is formed
[7,8]. Recently, chitosan has been used in dissolved air flotation (DAF)
∗
Corresponding author. Tel.: +603 7967 5296; fax: +603 7967 5319.
E-mail address: alihashim@um.edu.my (M.A. Hashim).
process to recover pulp fibres [9]. However, flotation of paper fibres
by CGAs generated from saponin has never been undertaken and this
is completely different from DAF process. The nature and character-
istics of the CGAs are influenced by the type and concentration of the
surfactants, and the ionic nature of the surfactant has been shown to
be very important for the functioning of the CGAs.
This work aims to explore the efficiency and optimize the perfor-
mance of CGAs generated from natural surfactant saponin extracted
from soapnut fruit pericarp for TSS removal from paper mill effluent.
The CGAs generated by soapnut was compared with other common
synthetic surfactants, based on stability and liquid drainage time. The
generated CGAs were applied for the recovery of pulp fibres from
paper mill effluent and the operating parameters for saponin were
optimized using central composite design.
2. Materials and methods
2.1. Surfactants
Four surfactants were used in this study, of which one is of
plant origin saponin and three were synthetic. Saponin is a natu-
ral surfactant traditionally used as an environmental friendly de-
tergent [10] and is non-ionic at pH 3.5 and displays slightly an-
ionic character with increasing pH [11]. It was extracted from the
http://dx.doi.org/10.1016/j.jtice.2015.02.037
1876-1070/© 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

2. 16 S. Mukherjee et al. / Journal of the Taiwan Institute of Chemical Engineers 53 (2015) 15–21
soapnut fruit pericarp by water [12] and the extract contained about
65% saponin as determined by UV–vis spectrophotometer [13]. The
synthetic surfactants used were sodium dodecyl sulphate (SDS),
Triton-X100 and cetyltrimethylammonium bromide (CTAB). Soapnut
solution was used at a concentration of 0.5% (w/v). The synthetic
surfactants were used at concentrations of 7 mM for SDS, 1 mM for
Triton-X100 and 1 mM for CTAB. The CGAs generated by the surfac-
tants were characterized by liquid drainage, air hold-up and half-life
(t1/2) as proposed by Zhang et al. [14].
2.2. Paper mill effluent preparation
Synthetic paper machine backwater effluent stock solution was
prepared in the laboratory by mixing 2 g of ordinary tissue paper
in 1 L distilled water to prepare the stock solution in order to main-
tain uniformity throughout the extensive batch experiments [15]. The
stock solution was diluted 10 times to mimic paper machine back-
water fibre concentration. The resultant wastewater had 200 mg/L of
paper fibre concentration. The pH of the wastewater is near neutral
(ࣅ6.5) and the turbidity of the effluent is 80.6 NTU. No chemicals were
added to the diluted slurry and it was prepared fresh for each set of
experiments to prevent bacterial degradation.
2.3. Generation and characterization of CGAs
Colloidal gas aphrons were generated from surfactant using a ho-
mogenizer (IKA T 25 basic ULTRA-TURRAX R
). The surfactant solutions
were stirred at high speed (6500 rpm), starting with 500 mL of sur-
factant solution, until a constant volume of white creamy CGAs were
produced in 6 min. These CGAs once produced, were kept dispersed
under low stirring conditions at around 1000 rpm by a magnetic stir-
rer and were pumped into the flotation column using a peristaltic
pump (Sastec BT 100-2J) at different sparging rates of 0.007, 0.010,
0.013, 0.016 and 0.018 L/min.
2.4. Flotation of fibres using flotation columns
In order to remove the pulp fibres by flotation, the effluent and
CGAs were passed in counter-current direction. The CGA bubbles rise
up slowly due to their small sizes. The fine paper fibres coming down
with the wastewater from top of the column come in contact with
the bubbles rising upwards and are carried upwards by the bubbles
and are removed with the fomate. The flotation column is made of
Perspex glass, 0.05 m in diameter and 1 m in height. The CGAs inlet
was at 0.06 m from the base of the column and an outlet at the base
of the column for the tailings. A conical diffuser was positioned at
the base, just above the CGAs inlet to achieve a uniform distribution
of aphrons. The height of liquid in the column was maintained by
constantly pumping the wastewater from the top of the column at a
constant flow rate. The inlet for the wastewater was at 0.665 m above
the base of the column. The fomate and the entrapped particulate
matters were collected from the top of the column. The wastewater
was initially poured into the column until it reached just above the
feed inlet. Then the CGAs were pumped from bottom of the column.
The experimental scheme is shown in Fig. 1. Each set of experiments
was run for 80 min and samples were collected every 10 min. The
system required 30 min to stabilize and the data after the stabilization
phase is presented here.
2.5. Optimization experiments
Central composite design (CCD) having five levels effective for the
estimation of parameters in a second order model was developed by
Box–Hunter [16]. A second-degree polynomial equation is used to
Fig. 1. The scheme of the experiment.
explain the behaviour of the system, as shown in Eq. (1):
y = β0 +
k
i=1
βixi +
k
i=1
βiix2
i +
k
i⊇j
k
i=1
βijxixj (1)
where, y = predicted response, β0 = offset term, βi = linear effect,
βii = squared effect, βij = interaction effect.
Several factors that can influence the removal of TSS by CGAs
flotation were taken as variables and their coded and actual values
are listed in Table 1. All the experimental designs and optimization
were performed using Design Expert 7 software.
Table 1
Actual values of variables for the coded values.
Variables Actual values for the coded values
–α -1 0 +1 –α
CGA sparging rate (L/min)
(A)
0.00725 0.010 0.013 0.016 0.01805
Wastewater flow rate
(L/min) (B)
0.00725 0.010 0.013 0.016 0.01805
pH (C) 5.15 6 7.25 8.5 9.35

3. S. Mukherjee et al. / Journal of the Taiwan Institute of Chemical Engineers 53 (2015) 15–21 17
Fig. 2. Rise of CGA-liquid front with time for SDS, soapnut (SN), Triton-X100 and CTAB.
3. Results and discussion
3.1. Stability of the surfactants
Stability of CGAs is the most important characteristic which en-
ables them to be generated outside the point of application. It is de-
termined by liquid drainage from the foam and half-life (t1/2) [17]. Air
hold-up is another important parameter contributing significantly
to its applicability. Increased air hold-up indicates large number of
small size bubbles with increased interfacial surface area, which facil-
itates attachment of large number of particles to CGAs during flotation
experiments [18].
The rate of separation of CGAs from the liquid phase with time
is presented in Fig. 2. Triton-X100 and CTAB CGAs separate out
from the liquid phase faster than the soapnut or SDS CGAs. Soap-
nut CGAs took a slightly longer time to separate out and it is con-
cluded that it remained homogenized for a longer period of time than
other surfactant CGAs. Homogenization of CGAs is important so that
continuous flotation tests can be conducted over longer period of
time.
The half-lives of soapnut, CTAB, SDS and Triton-X CGAs increased
with concentrations. The half-life of soapnut is 130 s at 0.25% con-
centration, which increases to 180 s at 1% soapnut concentration
and increases insignificantly beyond 1%. Thus soapnut was used
at 0.5% concentration for all the flotation experiments. In case of
SDS, the half-life remains same for 3.5 mM and 7 mM concentra-
tion and then shows very small increase at higher concentration.
CGAs prepared from 0.5 mM, 1 mM and 2 mM of Triton-X100
solutions have half-lives of 130, 135 and 145 s respectively, but
the CGAs prepared from 0.5 mM, 1 mM and 2 mM of CTAB show
shorter half-lives of 70, 90, and 125 s respectively. Soapnut CGAs
are the most stable among all the surfactants having the highest
half-life.
As the air hold up in CGAs increases, less surfactant solution is
required for flotation. All the surfactants showed increased air hold-
up in CGAs with increasing concentrations. However, air hold-up of
soapnut is highest and ranges between 31 and 35%. Air hold-ups for
SDS, CTAB and Triton-X vary in the ranges of 29–31%, 25–29% and
14–21%, respectively. In case of soapnut CGAs, the term “100 mL of
CGAs” would imply that it contained up to 35% of air by volume and
the rest of it was surfactant solution.
3.2. Flotation of paper fibres by CGAs micro-bubbles
The removal of suspended pulp fibres from the effluent by CGAs
depends upon probability of collision of the bubbles with the fine
particles and their captures as well as retention, prior to being floated
upwards and removed. The attachment and capture is influenced by
the surface charge of the bubbles and the suspended particles [19].
In order to aid the particle-bubble collision, a counter current flow of
the CGAs and wastewater was maintained.
The sparging rate of the CGAs is a significant factor controlling the
removal of suspended particulate matter as exhibited by TSS concen-
tration in the fomate and the tailings (Fig. 3). TSS in fomate increased
with the increase in CGAs sparging rate from 0.013 to 0.018 L/min
for all surfactants. Particle removal by CGAs can be attributed to
two mechanisms, (i) buoyant action of small bubbles and (ii) bubble-
particle ionic interaction between.
Bubble-entrained particle-flotation is the principal mechanism,
where larger particles are floated up by the buoyant action of innu-
merable small bubbles [20]. According to Sebba [1], the small bubble
size and the high stability of CGAs aid the flotation process. Soapnut
CGAs were able to remove about 256 mg/L TSS at a sparging rate
of 0.018 L/min and a wastewater flow rate of 0.016 L/min, which
is the highest among all the surfactants. Better particle removal at
high sparging rate is due to the high stability of the soapnut CGAs
as observed by their longer half-life. CTAB CGAs were least stable,
but they were able to remove 244 mg/L TSS at a sparging rate of
0.016 L/min and an effluent flow rate of 0.016 L/min, highest removal
among the three synthetic surfactants studied. Out of the four sur-
factants used in the study, soapnut is mildly anionic at its natural pH
of 4.5, Triton-X100 is non-ionic, CTAB is cationic and SDS is anionic.
The ionic charge of the surfactant imparts a charge on the surface of
the CGA bubbles and hence it can be inferred that the CTAB aphrons
are positively charged and thus removed the suspended particulate
matters by ion flotation. According to the zeta potential values, the
pulp fibres in the effluent are negatively charged (−24.1 mV). An ion-
surfactant complex is formed by the negatively charged particles with
the positively charged surfactant molecules on the CGA microbubbles
and the complex are floated up to the surface [20,21]. On the other
hand, SDS is a strongly anionic surfactant and there is repulsive force
acting between the CGA bubbles and the suspended particles in the
effluent.

4. 18 S. Mukherjee et al. / Journal of the Taiwan Institute of Chemical Engineers 53 (2015) 15–21
Fig. 3. Variation of TSS concentration in the fomate and tailings due to the variation in sparging rate of the CGAs (SN: soapnut).
Fig. 4. Particle size distribution of raw paper mill effluent.
In this study it was found that amongst two factors responsible
for removal of suspended particulate matter by flotation, the bubble-
entrained floc flotation is a dominant factor in comparison with the
flotation due to ionic charges on the CGA bubbles. This is in sharp
contrast to the findings by other researchers, who deduced that the
charge of the surfactant was an important attribute for TSS removal
[20,22]. Soapnut exhibited higher removal due to highly stable bub-
ble structure as compared to positively charged CGAs produced by
cationic CTAB.
The concentration of TSS in the tailings decreases with increase in
TSS concentration in the fomate. With increase in sparging rate, the
recovery of pulp fibres through fomate becomes more efficient and
the turbidity of the effluent decreases.
3.3. Physiochemical characteristics of the removed suspended particles
Particle size distribution of the machine back water showed that
the pulp fibres mostly ranged between 500 and 2000 μm (Fig. 4).CGAs
have a size variation of 10–100 μm, which is much smaller than the
suspended particles. This is preferred as small bubble size increases
inter particular surface area which improves separation by flotation
[20,22]. Also it was observed that since bubble entrained flotation was
a dominant factor in case of fibre flotation by CGAs, large number of
smaller bubbles would be more effective. The SEM micrographs of the
pulp fibres recovered by soapnut CGAs are presented in Fig. 5 which
reveals the good condition of fibres during recovery and thus it can
be reused in the paper making process.
3.4. Optimization of TSS removal by soapnut CGAs
According to the results in Section 3.2, soapnut was found to be
the most effective surfactant, both in terms of CGA stability and in
terms of fibre removal. Hence, an optimization study was carried out
using soapnut CGAs at 0.5% (w/v) concentration in order to determine
the effect of CGAs sparging rate, effluent flow rate and pH of the
effluent. A three factor, five levels CCD statistical experimental design
was applied to optimize the important operating parameters for the
maximum removal of suspended fibres in the fomate. The results of
the ANOVA for response surface reduced cubic model are presented
in Table 2.
ANOVA is functional in graphical analysis of the data to assess
the nature of interaction between process variables and responses
[23]. In Table 2, the ANOVA of regression model shows that the re-
duced cubic model is highly significant for TSS removal in fomate, as
is evident from the Fisher’s F-test (Fmodel = 45.77), with a low prob-
ability value (P model > F = 0.0001), as suggested by Liu et al. [24].
There is only 0.01% possibility that this model value could occur due
to noise. The predicted R2 from ANOVA is a measure of accuracy of
the model. For the model to be sufficient, a difference of no more
than 0.20 between predicted and adjusted R2 values is allowable. For
TSS in fomate, the predicted R2 value is 0.9248, which is in sufficient
agreement with the adjusted R2 value of 0.9656. Adequate precision
is indicated by a signal to noise ratio of 4 or more, which determines
range of predicted response relative to the associated error. The de-
sired value is normally 4 or more [25,26]. The ratio of 28.999, for TSS
in fomate is indicative of adequate signal. The error expressed as a
percentage of the mean provides the coefficient of variation for this
model.
The data points of the graph containing the predicted versus actual
values are evenly distributed along a 45° line (Fig. A1), signifying
a good fit of data in the following reduced third order polynomial

5. S. Mukherjee et al. / Journal of the Taiwan Institute of Chemical Engineers 53 (2015) 15–21 19
Fig. 5. SEM micrographs of the pulp fibres after removal by soapnut CGAs.
Table 2
Statistical models obtained from the ANOVA for TSS removal.
Source Sum of squares df Mean square F value P-value probability > F
Model 17131.59 8 2141.448 45.77412 < 0.0001 Significant
A-CGA sparging rate 11131.92 1 11131.92 237.9483 < 0.0001
B-Wastewater flow rate 31.87246 1 31.87246 0.681284 0.4267
C-pH of wastewater 2154.781 1 2154.781 46.0591 < 0.0001
AB 15.125 1 15.125 0.323302 0.5811
AC 153.125 1 153.125 3.273094 0.0978
BC 78.125 1 78.125 1.669946 0.2228
A2
2238.481 1 2238.481 47.84822 < 0.0001
C2
1645.797 1 1645.797 35.17942 < 0.0001
Residual 514.6125 11 46.78295
Lack of fit 229.2792 6 38.21319 0.669624 0.6827 not significant
Pure error 285.3333 5 57.06667
Cor total 17646.2 19
Std. dev. 6.839807 R-squared 0.970837
Mean 115.3 Adj. R-squared 0.949628
C.V. % 5.932183 Pred. R-squared 0.889967
PRESS 1941.662 Adeq. precision 23.05126
equation:
TSS fomate = 638.648 − 32.29 × CGA sparging rate − 0.975
× wastewater flow rate− 83.928×pH of wastewater
+ 0.43 × CGA sparging rate × wastewater flow rate
− 0.7 × wastewater flow rate × pH of wastewater
+ 1.385 × CGA sparging rate2
+ 6.96
× pHofwastewater2
(1a)
The contour plot in Fig. 6(a) implies that as the sparging rate of
CGAs increases, the TSS in the fomate increases and reaches maximum
at 0.013 L/min, beyond which the concentration of fibres in the fo-
mate decreases. Since soapnut is a non-ionic surfactant, the removal is
mainly governed by bubble entrained flotation [20,21]. If the sparg-
ing rate of the CGA is very high, the viscous drag produced would
be dominant and cause the fibres to get detached from the bubble
surface and reduce flotation efficiency [27]. The removal of fibres is
marginally affected by the effluent flow. However, it can be observed
that at lower flow rates, the removal of fibres from the effluent is
higher and decreases at a flow rate of 0.013 L/min. A slight increase
in removal is observed at a flow rate of 0.016 L/min. However, the
batch experiments show that a very high flow rate of 0.018 L/min as
compared to 0.016 L/min sparging rate, the removal of fibres de-
creases drastically.
Fig. 6(b) shows the effect of pH change on the recovery of fibres in
the fomate. The concentration of pulp fibres in the fomate increases
with increase in pH and low CGA sparging rate. However, the ad-
justment of the wastewater pH to a higher alkaline pH requires the
addition of lime or other alkali salts and the treated water would also
be rendered highly alkaline, requiring further treatment. Thus recov-
ery of fibres at higher pH is not a feasible option from an economic
and environmental point of view.
3.5. Optimization and validation of model
Optimization of fibre removal in fomate was performed by a mul-
tiple response method called desirability function in Design Expert 7
software. In order to achieve maximum desirability of TSS removal,
the sparging rate of fomate and flow rate of effluent were kept in
range i.e., between 0.008 L/min and 0.018 L/min while the pH was set
near neutral, as shown in Table 3. The optimum values of the factors
were verified by confirmatory experiments. From the observed re-
sults, it can be concluded that the generated model was an adequate
prediction of turbidity removal with relatively small error of 3.09%.
After optimization 76% recovery of pulp fibres could be achieved.

6. 20 S. Mukherjee et al. / Journal of the Taiwan Institute of Chemical Engineers 53 (2015) 15–21
Fig. 6. Response surface plot of TSS in fomate (a) variation between wastewater flow rate and CGA sparging rate; (b) variation between pH of wastewater and CGA sparging rate.
Table 3
Optimum conditions and their desirability.
CGA sparging rate
(L/min−1
)
Wastewater flow
rate (L/min)
pH of the
wastewater
Optimization Validity
TSS fomate (mg/L) Desirability TSS fomate (mg/L) Error
12.99 16 7.25 147.44a
0.79211 152b
3.09%
a
TSS fomate of 147.44 mg/Lࣅ73.7% TSS removal from paper.
b
TSS fomate of 152 mg/L ࣅ 76% TSS removal.
4. Conclusion
This work investigates and optimizes the performance of soapnut
CGAs for recovering dispersed pulp fibres in the machine end of the
conventional paper making process. Characterization study showed
that the soapnut CGAs were the most stable having the longest half-
life of 180 s and air hold-up of 33.33% by volume. Batch experiments
using all the four surfactants demonstrated that soapnut removed
60% TSS from the effluent as compared to 50%, 37% and 30% removal
by CTAB, SDS and Triton-X100. Flotation of suspended fibres through
buoyancy of attached soapnut CGAs was more dominant mechanism
than flotation due to ionic interaction between the soapnut CGA bub-
bles and the particles. CCD was used to exhibit the influence of sig-
nificant operating parameters on TSS removal from the industrial
effluent. After process optimisation, about 76% fibre could be recov-
ered. Saponin being a plant origin surfactant, is biodegradable and

7. S. Mukherjee et al. / Journal of the Taiwan Institute of Chemical Engineers 53 (2015) 15–21 21
Fig. A1. Predicted TSS in fomate versus actual experimental values.
possesses anti-bacterial properties [28]. The actual surfactant con-
tent is only 0.09 g in 100 mL of surfactant solution. Therefore, use of
low amount of saponin as CGAs reduce the secondary pollution and
recover higher amount of pulp fibre in comparison to commonly used
synthetic surfactants.
Acknowledgements
The authors thank University of Malaya, Malaysia (Project no.:
UMC/HIR/MOHE/ENG/13) for providing the financial support.
Appendix A
Fig. A1
References
[1] Sebba F. Foams and biliquid foams – aphrons. Chichester: John Wiley and Sons;
1987.
[2] Hashim MA, Mukhopadhyay S, Gupta BS, Sahu JN. Application of colloidal gas
aphrons for pollution remediation. J Chem Technol Biotechnol 2012;87(3):305–
24.
[3] Hashim MA, Kumar SV, Sen Gupta B. Particle-bubble attachment in yeast flotation
by colloidal gas aphrons. Bioprocess Eng 2000;22(4):333–6.
[4] Spigno G, Dermiki M, Pastori C, Casanova F, Jauregi P. Recovery of gallic acid
with colloidal gas aphrons generated from a cationic surfactant. Sep Purif Technol
2010;71(1):56–62.
[5] Jauregi P, Dermiki M. 10 – Separation of value-added bioproducts by colloidal
gas aphrons (CGA) flotation and applications in the recovery of value-added food
products. In: Rizvi SSH, editor. Separation, extraction and concentration processes
in the food, beverage and nutraceutical industries. Cambridge: Woodhead Pub-
lishing; 2013. p. 284–314.
[6] Boonamnuayvitaya V, Jutaporn P, Sae-ung S, Jarudilokkul S. Removal of pyrene
by colloidal gas aphrons of a biodegradable surfactant. Sep Purif Technol
2009;68(3):411–16.
[7] Ahmad AL, Wong SS, Teng TT, Zuhairi A. Improvement of alum and PACl co-
agulation by polyacrylamides (PAMs) for the treatment of pulp and paper mill
wastewater. Chem Eng J 2008;137(3):510–17.
[8] Razali MAA, Ahmad Z, Ahmad MSB, Ariffin A. Treatment of pulp and paper mill
wastewater with various molecular weight of polyDADMAC induced flocculation.
Chem Eng J 2011;166(2):529–35.
[9] Miranda R, Nicu R, Latour I, Lupei M, Bobu E, Blanco A. Efficiency of chitosans for
the treatment of papermaking process water by dissolved air flotation. Chem Eng
J 2013;231(0):304–13.
[10] Suhagia BN, Rathod IS, Sindhu S. Sapindus Mukorossi (Areetha): An Overview. Int
J Pharm Sci Res 2011;2(8):1905–13 .
[11] Yang Y, Leser ME, Sher AA, McClements DJ. Formation and stability of emulsions
using a natural small molecule surfactant: Quillaja saponin (Q-Naturale). Food
Hydrocolloids 2013;30:589–96.
[12] Mukhopadhyay S, Mukherjee S, Hashim M, Gupta BS. Application of colloidal
gas aphron suspensions produced from Sapindus mukorossi for arsenic re-
moval from contaminated soil. Chemosphere 2014 http://dx.doi.org/10.1016/
j.chemosphere.2014.06.087 .
[13] Roy D, Kommalapati RR, Mandava S, Valsaraj KT, Constant WD. Soil washing
potential of a natural surfactant. Environ Sci Technol 1997;31(3):670–5.
[14] Zhang W, Tsang DCW, Lo IMC. Removal of Pb and MDF from contaminated soils
by EDTA- and SDS-enhanced washing. Chemosphere 2007;66(11):2025–34.
[15] Hashim MA, Gupta BS. The application of colloidal gas aphrons in the re-
covery of fine cellulose fibres from paper mill wastewater. Bioresour Technol
1998;64(3):199–204.
[16] Box GEP, Hunter JS. Multifactor experimental designs for exploring responses
surfaces. Ann Math Stat 1957;28:195–242.
[17] Yan Y-l, Qu C-t, Zhang N-s, Yang Z-g, Liu L. A study on the kinetics of liquid
drainage from colloidal gas aphrons (CGAs). Colloids Surf A: Physicochem Eng
Aspects 2005;259(1–3):167–72.
[18] Jianhong L, Jun L, Ping H, Meiying H. Kinetic rate constant of liquid drainage from
colloidal gas aphrons. Sep Sci Eng 2009;17(6):955–9.
[19] Reay D, Ratcliff GA. Removal of fine particles from water by dispersed air flotation:
effects of bubble size and particle size on collection efficiency. Can J Chem Eng
1973;51(2):178–85.
[20] Mansur EHA, Wang Y, Dai Y. Removal of suspensions of fine particles from water
by colloidal gas aphrons (cgas). Sep Purif Technol 2006;48(1):71–7.
[21] Fuda E, Jauregi P. An insight into the mechanism of protein separation by
colloidal gas aphrons (CGA) generated from ionic surfactants. J Chromatogr B
2006;843(2):317–26.
[22] Basu S, Malpani PR. Removal of methyl orange and methylene blue from wa-
ter using colloidal gas aphron-effect of processes parameters. Sep Sci Technol
2001;36(13):2997–3013.
[23] Sen R, Swaminathan T. Response surface modeling and optimization to elucidate
and analyze the effects of inoculum age and size on surfactin production. Biochem
Eng J 2004;21(2):141–8.
[24] Liu H-L, Lan Y-W, Cheng Y-C. Optimal production of sulphuric acid by
Thiobacillus thiooxidans using response surface methodology. Process Biochem
2004;39(12):1953–61.
[25] Aghamohammadi N, Aziz H, Isa MH, Zinatizadeh AA. Powdered activated
carbon augmented activated sludge process for treatment of semi-aerobic
landfill leachate using response surface methodology. Bioresour Technol
2007;98(18):3570–8.
[26] Mason RL, Gunst RF, Hess JL. Statistical design and analysis of experiments with
applications to engineering and science. 2nd ed. New York: Wiley; 2003.
[27] Hashim MA, Gupta BS, Kumar SV, Lim R, Lim SE, Tan CC. Effect of air to solid ratio
in the clarification of yeast by colloidal gas aphrons. J Chem Technol Biotechnol
1998;71(4):335–9.
[28] Upadhyay A, Singh DK. Molluscicidal activity of Sapindus mukorossi and Ter-
minalia chebula against the freshwater snail Lymnaea acuminata. Chemosphere
2011;83(4):468–74.