Environmental Technology

ISSN: 0959-3330 (Print) 1479-487X (Online) Journal homepage: https://www.tandfonline.com/loi/tent20

Synthesis, characterization of cellulose acetate

membrane and application for the treatment of
oily wastewater

B Shoba, J Jeyanthi & S Vairam

To cite this article: B Shoba, J Jeyanthi & S Vairam (2020) Synthesis, characterization of cellulose
acetate membrane and application for the treatment of oily wastewater, Environmental Technology,
41:12, 1590-1605, DOI: 10.1080/09593330.2018.1543353

To link to this article: https://doi.org/10.1080/09593330.2018.1543353

Published online: 15 Nov 2018.

Submit your article to this journal

Article views: 201

View related articles

View Crossmark data

Citing articles: 4 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tent20
ENVIRONMENTAL TECHNOLOGY
2020, VOL. 41, NO. 12, 1590–1605
https://doi.org/10.1080/09593330.2018.1543353

Synthesis, characterization of cellulose acetate membrane and application for the

treatment of oily wastewater
Shoba Ba, Jeyanthi Ja and Vairam Sb
a
Department of Civil Engineering, Government College of Technology, Coimbatore, India; bDepartment of Chemistry, Government College of
Technology, Coimbatore, India

ABSTRACT ARTICLE HISTORY

This study reports the synthesis of cellulose acetate (CA) membranes of different weight Received 6 April 2018
percentages of the polymer ranging from 5 to 17.5 wt% with 2.5% increment and their Accepted 28 October 2018
behaviour towards oil removal from water. The membrane showed decreased water content and
KEYWORDS
porosity and also increased hydraulic resistance upto 3.38 kPa/(l m−2 h−1) as the concentration of Cellulose acetate membrane;
polymer increases. The Atomic Force Microscopic studies reveal that the membrane shows water flux; oily wastewater;
decreased arithmetic mean roughness from 36.291 to 5.935 nm as the concentration of the oil rejection; TG-DTA
polymer increases from 5% to 17.5%. Field Emission Scanning Electron Microscopy shows the
surface morphology of the CA membrane. X-ray diffraction studies indicate that in the
membrane having above 15% polymer concentration, the polymer chains orient to regularity
showing higher crystallinity. The thermal studies indicate their stability upto 280°C, and
exothermic decompositions from 338°C to 380°C implying the compactness of chains as polymer
concentration increases. From Total Organic Carbon studies, the oil rejection was found to be
maximum (96.4%) for the highest concentration of polymer in the membrane.

1. Introduction
aquatic environment. The used engine oil derived from
In recent years, water resources are becoming limited automobile service station contains highly pollutant
and due to this problem, reuse of water becomes una- materials. This wastewater requires efficient treatment
voidable. Therefore, the treatment of reused water in before letting into the ground, water bodies or sewer
an effective manner may reduce the impact on human [3]. According to a recent survey, it is understood that
health and the environment. One of the major environ- 45% of used engine oil are collected worldwide for
mental concerns is oily water and the direct discharge reuse and recycle purpose and the remaining 55% of
of oily wastewater into the environment [1]. Nowadays, used engine oil are being thrown into the environment
petroleum, automobile and oil refineries generate oil- [4]. Thereby, the maximum discharge value of oil–water
in-water emulsion in the range of 50–1000 mg/l [2]. emulsion is restricted to 5–10 mg/l for disposal into
Direct discharge of this pollutant into the aquatic water and land and 20 mg/l into public sewer by Pol-
environment affects both the human health and lution Control Boards and other Governmental agencies

CONTACT Jeyanthi J [email protected]; [email protected] Department of Civil Engineering, Government College of Technology, Coim-
batore 641013, India
© 2018 Informa UK Limited, trading as Taylor & Francis Group
 ENVIRONMENTAL TECHNOLOGY 1591

[5,6]. Hence, it is essential to treat the oily wastewater porous and nonporous or symmetric and asymmetric
before discharging into water bodies. membranes [22]. It is known that the membranes are
The conventional treatment techniques such as skim- also classified as filters, microfilters and ultrafilters when
ming, floatation, dissolved air flotation, gravity separ- they have pore size greater than 5000, 100–5000 and 2–
ation, emulsion breaking, coagulation and flocculation 100 nm, respectively [23]. The pore size of the membrane
have been used for the treatment of oily wastewater is an important parameter for the assessment of mem-
for the last two decades [7,8]. However, these methods brane performance and the development of pores in
have some drawbacks, which include high operational the membrane is mainly dependent on the selection of
cost and complex operational procedure. In addition, suitable polymer [24]. Though many works have been per-
these methods are ineffective to remove the micron or formed with CA membrane with regard to its character-
submicron-sized oil droplets, especially when the size istics, no work has so far been attempted to assess the
of the oil droplet is less than 20 µm [9]. Therefore, mem- characteristics of CA membrane with change in the con-
brane technology is considered as one of the alternative centration of polymer. In this work, the important charac-
techniques to overcome the issues mentioned. It is quite teristics of CA with different polymer concentration (wt%)
popular due to the reasons such as ease of fabrication, ranging from 5 to 17.5 have been studied and the
less energy consumption and minimum use of chemical efficiency of oil rejection has been presented.
constituents for removing organic pollutants present in
wastewater [10].
Recently, membrane technology is being used for oil 2. Experimental
water separation, especially water-soluble oil emulsion
2.1. Materials and membrane preparation
[11]. The mechanism involved in membrane technology
methods
is macromolecular solute separation, which allows the
passage of solvents through the membrane retaining the Cellulose acetate powder with an average molecular
larger molecules [12]. Literature survey [13] indicates that weight of 30,000 g/mol powder, sodium dodecyl sul-
polymeric membranes, ceramic membranes and carbon phate (SDS) and N, N-dimethyl formamide (DMF) with
membranes can be used effectively for the treatment of 99% of purity were purchased from Merck. Ultrapure
oil-in-water emulsion. The use of ceramic membranes water (Direct-Q 8UV system – 18.2 MΩ cm @ 25°C) was
always includes high cost and complex fabrication
methods. Carbon membranes possess advantages such
as excellent chemical stability, thermal and mechanical Table 1. Various concentrations of solute and solvent used for
resistance and use of cheap precursors. Even though, the preparation of cellulose acetate membrane.
carbon membranes are quite restricted because their Membrane composition
(wt%)
pore dimensions are suitable only for gas permeation. Membrane ID
Membrane CA Solvent
The polymeric membranes are widely used and have
CA1 CA 5% 5 95
proven for the treatment of industrial wastewater effec- CA2 CA 7.5% 7.5 92.5
tively. Membranes are synthesized in various forms CA3 CA 10% 10 90
CA4 CA 12.5% 12.5 87.5
such as Micro Filtration [MF], Ultra Filtration [UF], Nano CA5 CA 15% 15 85
Filtration [NF], Reverse Osmosis [RO] and gas separation CA6 CA 17.5% 17.5 82.5
units [14–18]. Generally, CA is used as a base polymer for
the membrane. CA is also an environmental friendly
cheap polymer and can be derived from sustainable
resources. Further, it can offer moderate water flux,
high biocompatibility, good chlorine tolerance and
desalting capability [19]. In addition, CA membranes
are also hydrophilic in nature, due to which they repel
the droplets of oil from adhering to the membrane
surface contributing to minimize the membrane fouling.
Cellulose acetate membranes have been synthesized
by either using wet or dry phase inversion technique
[20]. During the synthesis of membrane, macro voids are
formed in the membrane in the casting process, which
is strongly dependent on cast solution thickness [21].
Based on the structure, membranes are classified as Figure 1. Cellulose membrane.
1592 SHOBA B ET AL.

Figure 2. FTIR spectrum of CA membrane.

used for this study. Waste engine oil (10W-30) was col- 2.2. Characterization of cellulose acetate
lected from nearby two wheeler service stations and membrane
used.
2.2.1. FTIR spectrum analysis
Different quantities of CA powder (5% to 17.5 wt%
The IR spectrum of the membrane was recorded in the
with 2.5% increment, beyond which due to high vis-
range between 500 and 4000 cm−1 at 4 cm−1 resolution
cosity, membrane was not formed) were dissolved in
in the mid-infrared region with 16 scans per sample to
the DMF as shown in Table 1 and heated using a mag-
investigate functional groups present in the membrane
netic stirrer maintained at 60°C for 3 h [25]. The sol-
and shown in Figure 2.
utions were kept in an ultrasonic probe sonicator
(Pro-500, Make:Labman) for 1 h to obtain a uniform dis-
persion and kept undisturbed for 24 h to get rid of air
bubbles. Then, each solution was poured into a glass 2.2.2. Water content, porosity and pore size testing
plate and then drawn to the speed of 20 mm/s using Water content testing was performed by the procedure
(UF-membrane casting unit) doctor blade to get a described by Velu et al. [26] and the percentage of
uniform thickness of 250 µm of the film. After casting, water content was calculated by using Equation (1).
the glass plate was immediately immersed into the The porosity and mean pore size of the membrane
coagulation bath containing distilled water at 28 ± 2°C were determined using the method reported by
immediately with care in order to avoid solvent evapor- Haolong Bai et al. [27] and Jianjun Zhou et al. [28]
ation in a surrounding relative humidity as 55 ± 2%. The (Equations (2) and (3)).
membranes formed were washed several times with
w1 − w 2
distilled water to remove residual solvent. Figure 1 Water content (%) = × 100, (1)
w1
shows the CA membrane. All the membranes were
stored in 0.1% formalin solution to avoid microbial where w1 and w2 are the wet and dry weight of mem-
growth. branes respectively. The values were checked for
 ENVIRONMENTAL TECHNOLOGY 1593

Table 2. Water content, porosity, pore size and hydraulic trials were run and the mean value was used for calcu-
resistance of the cellulose acetate membranes. lation.
Water Hydraulic
Membrane content Porosity Mean pore resistance (kPa)/ V
ID (%) (%) diameter (nm) (l/m2 h) Pure water flux (Jw) = , (4)
AT
CA1 91.03 89.47 128.25 0.05
CA2 89.41 86.85 88.45 0.10 where Jw = pure water flux in l/(m2 h), A = membrane
CA3 82.40 78.38 60.67 0.26 area in m2, T = permeation time in hour, V = Quantity of
CA4 79.05 72.39 48.74 0.47
CA5 76.42 74.55 39.66 0.68 permeate in l.
CA6 68.16 61.25 21.08 3.38

2.2.4. Membrane hydraulic resistance (Rm)

concordance. It is a measure of tolerance of a membrane towards

 hydraulic pressure. The Rm value of the membrane was
Ww − W d found out by substituting the water flux (Jw) of the CA
Porosity, Pr (%) = ∗100, (2)
(rAh) membrane at transmembrane pressure (ΔP) in the
where ww and wd is the weight of wet and dry mem- Equation (5).
brane in g, ρ = density of water in g/cc, A = area of the Dp
Rm = . (5)
membrane in wet state in cm2, h = thickness of the mem- Jw
brane in cm.

[(2.9 − 1.75Pr )∗(8hlJw )] 2.2.5. Field emission scanning electron microscopy
rm = , (3) (FESEM)
3600Pr Dp
The images of the surfaces of membranes and the
where Pr = porosity %, η = viscosity of water in Pa.s, l = cross-sectional areas were found by using FESEM-
thickness of membrane in m, Jw = pure water flux in l/ ΣIGMA HV-ZEISS with Bruker quantax 200. In this
(m2 h), ΔP = operating pressure (Pa), rm = mean pore method, the membranes were cut into smaller pieces
size in m. The pore size was calculated for three times and sputter coated with the thin film of gold and the
and the mean values are given in Table 2. images of the surface were recorded.
In order to find the images of cross sections of mem-
2.2.3. Pure water flux branes, the membranes were first immersed into liquid
Permeation experiments of CA membranes were carried nitrogen for 20–30 s and then broken into smaller
out using vacuum filtration setup with a processing pieces. The membranes were mounted in such a way
volume of 300 ml. The effective permeation membrane that the cross-sectional area is visible on a brass plate.
area is 15.90 cm2. The pure water flux of the membrane All the images are shown in Figure 4.
was measured under the steady state flow using ultra-
pure water. The pure water flux was found by using 2.2.6. Atomic force microscopy (AFM)
the Equation (4) (Figure 3). For each membrane, three The surface morphology and roughness of the cellulose
acetate membranes were found by AFM (XE-7, Park
Systems KANC 4F, South Korea) under non-contact
mode. A silicon cantilever with nominal spring constant
of 40 N/m and a resonance frequency of 300 KHz were
used to conduct this experiment [29]. A small square
(less than 1 cm2) of the membrane was used for this
analysis. The membranes were cut and mounted with
the magnetic steel using double sided tape. The mem-
branes were scanned (shown in Figure 5) at the surface
area of the 5 µm × 5 µm.

2.2.7. Thermo gravimetric analysis (TGA–DTA)

Thermo gravimetric analyser was used (Shimadzu DTG-
60, Japan) to found the thermal behaviour of the cellu-
Figure 3. Effect of pure water flux on the cellulose acetate mem- lose acetate membrane. TGA and DTA curves were
branes with different time intervals. plotted simultaneously for the dry membrane samples
1594 SHOBA B ET AL.

Figure 4 . a FESEM images of (a) CA1 (b) CA2 (c) CA3 (d) CA4 (e) CA5 and (f) CA6 and Cross-sectional images of (a1) CA1 (b1) CA2 (c1) CA3
(d1) CA4 (e1) CA5 and (f1) CA6. (b) FESEM cross-sectional image of skin layer and porous structure of membrane (CA6).

from ambient temperature to 1000°C at a heating rate of 6100) run with Cu Ka radiation at a wavelength of
10°C/min and the curves recorded are given in Figure 6. 1.5406 Å operated at 40 kV and 30 mA. The scan
range varies from 5° to 80° in a continuous mode
2.2.8. XRD analysis with the scan speed of 10° per min. The patterns are
X-ray diffraction pattern was obtained for CA mem- given in Figure 7. The crystallite size of particle was
brane using X-ray diffractometer (SHIMADZU-XRD found out using the Scherrer’s equation as given in
 ENVIRONMENTAL TECHNOLOGY 1595

Figure 4 Continued
1596 SHOBA B ET AL.

Figure 5. AFM-3d images (a) CA1 (b) CA2 (c) CA3 (d) CA4 (e) CA5 and (f) CA6 and 2d images (a1) CA1 (b1) CA2 (c1) CA3 (d1) CA4 (e1) CA5
and (f1) CA6.

Equation (6). 2.2.9. Rejection studies

Oil-in-water emulsion was prepared as follows. The
Kl
Dp = , (6) engine oil (0.9 g) was mixed with 1litre of water and
b cos u
0.01 g of sodium dodecyl sulfate and sonicated using
where Dp is the crystallite size in nm, λ is the diffraction probe sonication (Pro-500 Labman) for 15 min at an
wavelength, β is the full width at half maximum amplitude of 75% to obtain a homogenous solution.
(FWHM), θ is the diffraction angle and K is the Scherrer’s The particle size of the oily wastewater was found in
constant and the value is close to unity. the Zetasizer Nano ZS90 (Figure 8). Optical microscope
 ENVIRONMENTAL TECHNOLOGY 1597

Figure 5 Continued

was used to assess the dispersive nature of oil-in-water where TOCfeed and TOCpermeate are the concentration of
emulsion (Figure 9). The oil removal efficiency was deter- TOC in the feed and permeate, respectively.
mined (Figure 10) in terms of TOC removal (by using
Equation (7)) by a total carbon analyzer (TOC-L, Shi-
madzu TOC analyzer) [30,31]. 2.2.10. Flux recovery ratio (FRR)
The flux behaviour of the membrane was analysed
based on the FRR results (shown in Figure 12).
TOCfeed − TOCpermeate Initially, the pure water flux (Jw1) of the membrane
TOC removal (%) = X 100, (7)
TOCfeed was obtained at 69 KPa. After that, oily wastewater is
1598 SHOBA B ET AL.

Figure 6. TGA and DTA curves of (a) CA1 (b) CA2 (c) CA3 (d) CA4 (e) CA5 and (f) CA6 membranes.

allowed to pass through the membrane for 1 h and the the cleaned membrane was measured at 69 KPa. In
membranes were washed thoroughly with the distilled order to evaluate the antifouling ability of the membrane,
water for 20 min and then the pure water flux (Jw2) of FRR was calculated using the following Equation (8).

JW2
FRR (%) = × 100. (8)
JW1

Figure 7. XRD patterns of CA membranes (a) CA1 (b) CA2 (c) CA3
(d) CA4 (e) CA5 and (f) CA6. Figure 8. Oil droplet size distribution curve.
 ENVIRONMENTAL TECHNOLOGY 1599

Moreover, C–H stretching of cellulose are also revealed

by the peaks around 1385 cm−1. All the observed
peaks substantiate that the membrane is made of cellu-
lose acetate alone without any modification. Similar
observations were also reported by Kanagaraj et al. [32].

3.2. Water content, porosity and pore size

From the data given in Table 2, it is observed that the
water content is found to be decreasing as the compo-
sition of the polymer increases in the membrane. It is
generally expected that water content can increase
with increasing polymer composition due to more
number of hydrophilic groups in the polymer, influen-
cing water content with the involvement of hydrogen
bonding [33]. However, in our case, it is found to be con-
tradictory as it is evident from the FESEM images, which
show a decrease in porosity with increasing composition
of the polymer. Moreover, the pores should be an open
Figure 9. Optical micrograph of oil-in-water emulsion. paving way for the flow of water through the membrane.
This is implied from the values of hydraulic resistance,
3. Results and discussion which is lower for the membrane having a low quantity
of the polymer (CA1). However, CA6 with 61.25%
3.1. FTIR spectrum
porosity, has water content 68.16%, and the hydraulic
The important vibrations of the CA membrane are shown resistance is abnormally high. This may be because of
in the IR spectrum (Figure 2). The inset picture indicates the closure of pores at the top and bottom layer of the
the expanded version of the range from 2750 to membrane during casting preventing water flow [34]
500 cm−1 of the spectrum. The spectrum shows the (Figure 4). These results are comparable with the report
νOH at 3408 cm−1 due to hydrogen bonding. Further, in in the case of polyether sulphone membrane [35]. The
the expanded spectrum, sharp peaks found at 1669 pore diameter is decreasing with increasing the concen-
and 1733 cm−1 are due to carbonyl stretching of CA tration of polymer ranging from 128.25 to 21.08 nm. This
(C=O group). The peaks observed at 1229 and clearly indicates that the formation of membranes are
1092 cm−1 are due to the presence of C–O–C bond. from microfilters to ultrafilters.

Figure 10. TOC rejection percentage of various proportions of cellulose acetate membranes.
1600 SHOBA B ET AL.

30 min and at 180 min the value of pure water flux

decreases to 1096.34 Lm−2 h−1. This flux reduction in
the membrane is altered to the compaction of the mem-
brane under pressure. Further, the reorganization of the
polymeric chain occurs leading to the change in the mem-
brane structure. This observation is similar to that of
Arthanareeswaran et al. [36]. Further, it indicates that
the pure water flux of the membrane has a direct
influence on the pore size and porosity [37].

3.4. Membrane hydraulic resistance (Rm)

Hydraulic resistance of the membrane is the intrinsic resist-
Figure 11. Flux variations of different concentrations of the CA ance of the cellulose acetate membrane. Membrane
membranes.
hydraulic resistance was used to estimate the tolerance
level of membrane towards hydraulic pressure. The hydrau-
3.3. Pure water flux lic resistances were found out using ultra-pure water as a
The pure water flux of the membrane is used to evaluate feed sample [38]. The hydraulic resistance offered by the
the permeation performance of the cellulose acetate membrane is more at the higher concentration of
membrane. When the polymer concentration is increased polymer due to its low porosity as discussed earlier [39].
from lower weight percentage (CA1) to higher weight per-
centage (CA6) then the average pure water flux (for 3 h)
3.5. Field emission scanning electron microscopy
reduces from 1304 to 20.41 Lm−2 h−1, respectively
(FESEM)
(Figure 3). The reason for the lower flux of the membrane
is the reduction in pore size of the membrane. The The surface and cross sections of the CA membranes were
decrease in pure water flux with increasing concentration also investigated by recording SEM micrographs. The pore
of polymer may be due to the formation of tight polymer structure of the membrane can be identified through
matrix in higher concentration. All the membranes exhibit FESEM images. More number of pores are seen at the
uniform, pure water flux except CA1. In this observation, surface of the membrane for CA1, CA2, CA3 and CA4.
CA1 has the initial water flux of 1583.21 Lm−2 h−1 at When the film casting glass plate is immersed in the

Figure 12. Flux recovery ratio of cellulose acetate membranes.

 ENVIRONMENTAL TECHNOLOGY 1601

coagulation bath, phase inversion starts immediately due results are in good agreement with FRR results. When
to the miscibility between DMF and water due to this for- the polymer concentration increases, the flux recovery
mation of pores occur. The increase in viscosity of the ratio of the membrane tends to increase.
casting solution affects the miscibility between DMF and The surface roughness of the membrane mainly
water during phase inversion. This leads to the formation depends on the composition of the casting solution.
of dense skin structure at the surfaces of the membrane From the experimental results, it is revealed that the
as shown in Figure 4(a). In this image, no pores are visibly surface roughness value gets lower at the polymer con-
found on the surface of CA5 and CA6 membrane. centration increases. In addition to that, the decrease
The images obtained from FESEM confirm that the in polymer concentration causes an increase in pore
membranes become less porous due to the increase in size of the membrane. This is due to the formation of
the concentration of polymers. After phase inversion, the gaps between the polymer matrix. These gaps are seen
membranes show a decrease in thickness, though the in the FESEM images at the lower concentration of the
initial casting thickness is 250 µm for all membranes. The polymer. Further, all the CA membrane surfaces
decrease in thickness is due to the escaping of DMF contain peaks and valleys. The membrane with lower
during the solidification process of polymer in water. concentration of the polymer contains more valleys. Nor-
However, the increase in the concentration of the mally, these valleys exhibit the lower resistance against
polymer leads to relative increase in the thickness of the the pressure. Therefore, the permeate water easily
membrane. This is because of the slow outward diffusion diffuses through the valley. When the polymer concen-
of DMF and intrusion of water into the polymer solution. tration is higher, the surface of the membrane contains
These results are comparable with the results reported by more peaks. Due to this, the hydraulic resistance of the
Darvishmanesh et al. [40]. From the cross-sectional membranes gets improved. In the higher polymeric con-
images, it is understood that the structure of the membrane centration, the valleys almost disappear due to the lower
consists of a dense skin layer supported with a spongy water flux. Therefore, the polymer concentration is an
macroporous structure as shown in Figure 4(b). [41]. effective parameter on the roughness of the membrane.
Akbari et al. [44] have also reported that increasing the
concentration of the polymer PVDF (polyvinylidene
3.6. Atomic force microscopy (AFM)
fluoride) from 14% to 20% causes lower surface rough-
AFM technique is widely used to study the membrane ness of the membrane. Thus, our report conforms to
surface at the atomic resolution level. The AFM tech- the literature report.
nique does not require any special sample preparation Similarly, in the 3d images, have the bright peaks rep-
methods. The analysis was carried out from the CA5% resent high points or nodules, whereas pores of the
to CA17.5% with the interval of 2.5% increment using membrane may be considered as a dark depression
the non-contact mode. Three important parameters region or valley [45]. The three-dimensional roughness
such as the arithmetic mean roughness (Ra), square values are usually abbreviated as S. AFM is also used to
average roughness (Rq) and a mean depth (Rz) were cal- analyse the roughness average (Sa), root mean square
culated to estimate the surface roughness of the mem- (Sq), surface skewness (Ssk) and surface kurtosis (Sku) of
brane with an area of 5 µm × 5 µm as given by Han the membrane and the results are collected in Table 3.
et al. [42]. The given parameter is used to measure the Ssk and Sku are usually used to measure the deviation
2d roughness and these are abbreviated as R. The rough- from the height distribution values. The negative value
ness values are obtained from AFM images using XEI of Ssk indicates the structure is a porous one and these
software (Version 1.7.6). These roughness parameters porous structures are found only in CA1 and CA2 mem-
are tabulated in Table 3. Hobbs et al. [43] reported that brane (Figure 5(a, b); Table 3). These negative values
increasing the roughness of polymeric membrane almost vanish at the higher concentration of CA mem-
causes the more fouling on the membrane. These brane from CA3 to CA6.

Table 3. Surface parameters of the cellulose acetate membranes at various compositions.

Membrane ID Ra (nm) Rq (nm) Rz (nm) Sa (nm) Sq (nm) Ssk (nm) Sku (nm)
CA1 36.291 46.481 297.864 0.0363 0.0466 –0.0794 3.1931
CA2 21.919 29.445 228.816 0.0219 0.0294 –0.1246 3.908
CA3 13.793 17.948 192.845 0.0138 0.0179 0.7225 5.0487
CA4 11.795 15.509 126.422 0.0118 0.0155 0.6856 4.2234
CA5 9.362 12.026 93.059 0.0094 0.012 0.3929 3.6688
CA6 5.935 8.447 93.180 0.0059 0.0084 1.7305 10.595
1602 SHOBA B ET AL.

3.7. Thermo gravimetric analysis (TGA–DTA) 20 µm). Further, these oil droplet sizes are larger than
the pore size of all the membranes except CA1. Due to
Simultaneous TG-DTA was performed for all the mem-
these reasons, the oil particles are difficult to diffuse
branes and shown in Figure 6. It shows three step degra-
through these membranes. Further, CA1 membrane exhi-
dation and the first step (DTA peaks at 60°C) is the
bits a pore size as 128.25 nm. This value lies in the size
evaporation of volatile particles held in the pores of
range of the oil droplets. Due to this reason, some of the
the membrane, which is an endothermic process,
oil droplets pass through the membrane. The permeate
showing weight loss of almost 3–4%. The second step
water quality was checked by the optical microscope.
is the decomposition of the membrane starting from
The optical micrograph (Figure 9) shows that the few oil
250°C shown by small exothermic peak centred at 338°
droplets are present in the water after the CA1 membrane
C, 345°C, 353°C, 349°C, 373°C and 380°C in the DTA for
treatment. The passage of these oil droplets through the
CA1 to CA6 membrane respectively. This step is followed
membrane is due to the pressure applied in the mem-
by a complete decomposition in third step showing
brane as well as large pores present in the CA1 membrane.
sharp exothermic peak at 454°C, 473°C, 491°C, 472°C,
Oil removal efficiency was estimated in terms of TOC
489°C and 504°C for CA1–CA6 membrane, respectively.
rejection as shown in Figure 10. The results indicate that
It was observed that the decomposition temperature
the removal efficiency was low for CA membranes with
was slightly increased for the membrane with a higher
lower concentrations of the polymer. The TOC rejection
concentration of the polymer. All the degradations are
values obtained for CA membranes with a different con-
similar to those reported by Rajesh et al. [45].
centration range from 90 (CA1) to 96.4% (CA6), which go
along with the results observed by Karakulski et al. [48]
3.8. XRD analysis and Wu et al. [49] who treated oily wastewater using
microfiltration carbon membrane.
Analysis of crystallites size deduced from FWHM values
from XRD peaks using Scherrer’s formula. XRD patterns of
samples show the humps centred approximately as 2θ 3.10. Effect of CA concentration on the permeate
values as 8.4° for all the samples (Figure 7). The crystallite flux
size of the membranes with various polymer concentration From the observed results, it is understood that the
range from 1.90 to 4.24 nm. From these observations, the permeate flux decreases as the concentration of CA
crystallite size is found to be increasing as the concen- increases. The permeate flux of oily wastewater for the
tration increases from CA1 to CA6 membranes. The diffrac- CA1 membrane was maximum (1217 L m−2 h−1) and
togram indicating that all membranes have been made by for CA6, the value was minimum (13.58 L m−2 h−1). The
the amorphous CA because of which porous nature is permeate flux of all the membrane decreases with time
developed. Among the patterns of membranes of [50]. CA1 membrane exhibits the permeate flux of
different concentrations of CA, CA5 and CA6 have a rela- 1275 L m−2 h−1 at 15 and at 60 min the flux decreases
tively good ordered state. This may be attributed to more to 1180 L m−2 h−1. The reasons for the less permeate
number of polymer molecules tend to align with one flux is the deposition of oil droplets on the surface of
another, which it is not possible in the case of other the membrane. Further, the oil deposition forms the
samples of lower concentration. Similar, trend has been cake layer on the surface of the membrane and this
observed by Vetrivel et al. [46] and Rajesh et al. [47]. leads to the lower permeate flux due to the concen-
tration polarization over time [51,52]. It is observed that
the removal of oily wastewater is dependent on the mor-
3.9. Rejection studies
phological structures of the membrane. In addition, the
The used engine oil and water was mixed thoroughly with permeate flux of the membrane is always lower than
the help of stirrer (REMI-5MLH Plus) followed by sonication. the pure water flux of the membrane (Figure 11).
The filtration experiment was conducted for the prepared
feed solution (900 mg of oil in 1 l of water) to estimate the
3.11. Flux recovery ratio
removal efficiency of an oil-in-water emulsion.
The oil droplet size was measured in the particle size Membrane fouling is one of the major reasons that affect
analyzer using dynamic light scattering technique and the permeate efficiency and restricting the wide usage of
the graph shows that the oil particle is in the range the membrane. To overcome this drawback, membrane
between 100 and 1000 nm and mean size of oil droplet cleaning is often required to recover the flux. The anti-
is 300 nm as shown in Figure 8. The obtained results fouling property of the membrane was estimated
confirm that the oil is in the emulsified state (less than based on the FRR value (Figure 12). The higher FRR
 ENVIRONMENTAL TECHNOLOGY 1603

value indicates better antifouling properties. From exper- [5] Abadi SRH, Sebzari MR, Hemati M, et al. Ceramic mem-
imental results, it is observed that FRR value slightly brane performance in microfiltration of oily wastewater.
Desalination. 2011;265:222–228. doi:10.1016/j.desal.2010.
increases with the higher concentration of polymeric
07.055.
membrane. Similar results were found in the previous [6] Suresh K, Pugazhenthi G, Uppaluri R. Fly ash based
studies [53]. The pure water flux of the fouled mem- ceramic microfiltration membranes for oil-water emulsion
branes (Jw2) restored nearly 90% of the initial flux (Jw1) treatment: parametric optimization using response
by simple washing of the membranes [54]. The results surface methodology. J. Water Process Eng. 2016;13:27–
show that the CA6 membrane has the capability of pro- 43. doi:10.1016/j.jwpe.2016.07.008.
[7] Al-Shamrania AA, Jamesa A, Xiao H. Destabilisation of oil–
viding better fouling resistance and provide better per-
water emulsions and separation by dissolved air flotation.
formance [55]. Water Resour. 2002;36:1503–1512.
[8] Adebajo MO, Frost RL, Kloprogge JT, et al. Porous
materials for oil spill cleanup: a review of synthesis and
4. Conclusion absorbing properties. J. Porous Mater. 2003;10:159–170.
[9] Ahmad AL, Majid MA, Ooi BS. Functionalized PSf/SiO2
This study led us to conclude that as the concentration of nanocomposite membrane for oil-in-water emulsion sep-
CA increases, the pore size of membrane decreases from aration. Desalination. 2011;268:266–269. doi:10.1016/j.
128.25 to 21.08 nm due to the alignment of polymer mol- desal.2010.10.017.
ecules one over the other leading to increased crystalli- [10] Cheryan M. Ultrafiltration and microfiltration handbook.
Florida: CRC press; 1988.
nity. Further, the change in polymer concentration [11] Zhong J, Sun XJ, Wang CL. Treatment of oily wastewater
affects not only the pore size but also roughness, oil rejec- produced from refinery processes using flocculation and
tion, antifouling property, porosity, pure water flux and ceramic membrane filtration. Sep. Purif. Technol.
permeate flux of the membrane. Further, the optimum 2003;32:93–98.
concentration of CA formation is 17.5 wt%, above which [12] Sairam M, Sereewatthanawut E, Li K, et al. Method for the
preparation of cellulose acetate flat sheet composite
the viscosity of polymer does not permit membrane for-
membranes for forward osmosis – desalination using
mation. The TOC removal of 96.4% is achieved for the MgSO4 draw solution. Desalination. 2011;273:299–307.
initial oil concentration of 900 mg/l in the CA6 membrane. doi:10.1016/j.desal.2011.01.050.
Due to the lower roughness of CA6, the antifouling prop- [13] Zhang X, Zhang B, Wu Y, et al. Preparation and character-
erty of the membrane gets improved. The water flux of the ization of a diatomite hybrid microfiltration carbon mem-
fouled membranes is also restored as 90% of the initial brane for oily wastewater treatment. J. Taiwan Inst. Chem.
Eng. 2018;89:39–48. doi:10.1016/j.jtice.2018.04.035.
flux of the membrane after 1 h of treatment of oily waste- [14] Baker RW. Membrane technology and application.
water. The pure water flux and permeate flux of the mem- London: Mc-Graw-Hill; 2000.
brane increases initially and decline after the addition of [15] Thakur VK, Voicu SI. Recent advances in cellulose and chit-
various concentrations of CA. osan based membranes for water purification: a concise
review. Carbohydr. Polym. 2016;146:148–165.
[16] Miculescu F, Maidaniuc A, Voicu SI, et al. Progress in
hydroxyapatite-starch based sustainable biomaterials for
Disclosure statement
biomedical bone substitution applications. ACS Sustain.
No potential conflict of interest was reported by the authors. Chem. Eng. 2017;5:8491–8512.
[17] Wróblewska-Krepsztul J, Rydzkowski T, Borowski G, et al.
Recent progress in biodegradable polymers and nano-
composite-based packaging materials for sustainable
References
environment. Int. J. Polym. Anal. Charact. 2018;23:383–
[1] Asatekin A, Mayes AM. Oil industry wastewater treatment 395. doi:10.1080/1023666x.2018.1455382.
with fouling resistant membranes containing amphiphilic [18] Thakur S, Govender PP, Mamo MA, et al. Progress in lignin
comb copolymers. Environ. Sci. Technol. 2009;43:4487– hydrogels and nanocomposites for water purification:
4492. doi:10.1021/es803677k. future perspectives. Vacuum. 2017;146:342–355.
[2] Nandi BK, Uppaluri R, Purkait MK. Treatment of oily waste [19] Ferjani E, Lajimi RH, Deratani A, et al. Bulk and
water using low-cost ceramic membrane: flux decline surface modification of cellulose diacetate based RO/
mechanism and economic feasibility. Sep. Sci. Technol. NF membranes by polymethyl hydrosiloxane
2009;44:2840–2869. preparation and characterization. Desalination. 2002;146:
[3] Hamawand TY, Rafat S. Recycling of waste engine oils 325–330.
using a new washing agent. Energies. 2013;6:1023–1049. [20] Rambabu K K, Velu S. Modified cellulose acetate ultrafiltra-
doi:10.3390/en6021023. tion composite membranes for enhanced dye removal.
[4] Zitte LF, Awi-Waadu GDB, Okorodike CG. Used-oil gener- Int. J. Chem. Sci. 2016;14:195–205.
ation and its disposal along east-west road, port [21] Vogrin N, Stropnik C, Musil V, et al. The wet phase separ-
Harcourt Nigeria. Int. J. Waste Resour. 2016;6:1–5. doi:10. ation: the effect of cast solution thickness on appearance
4172/2252-5211.1000195. of macrovoids in the membrane forming ternary cellulose
1604 SHOBA B ET AL.

acetate/acetone/water system. J. Membr. Sci. 2002;207: ultrafiltration membranes: part I. Sep. Purif. Technol.
139–141. 2008;64:38–47.
[22] Miculescu M, Thakur VK, Miculescu F, et al. Graphene- [37] Feng C, Shi B, Li G, et al. Preparation and properties of
based polymer nanocomposite membranes: a review. microporous membrane from poly (vinylidene fluoride–
Polym. Adv. Technol. 2016;27:844–859. doi:10.1002/pat. co-tetrtafluroethylene) (F2.4) for membrane distillation.
3751. J. Membr. Sci. 2004;237:15–24.
[23] Pandey P, Chauhan RS. Membranes for gas separation. [38] Arthanareeswaran G, Thanikaivelan P, Srinivasn K, et al.
Prog. Polym. Sci. 2001;26:853–893. Synthesis, characterization and thermal studies on cellu-
[24] Sivakumar M, Mohan DR, Rangarajan R. Studies on cellu- lose acetate membranes with additive. Eur. Polym. J.
lose acetate-polysulfone ultrafiltration membranes II. 2004;40:2153–2159. doi:10.1016/j.eurpolymj.2004.04.024.
Effect of additive concentration. J. Membr. Sci. [39] Vidya S, Vijayalakshmi A, Nagendran A, et al. Effect of addi-
2006;268:208–219. doi:10.1016/j.memsci.2005.06.017. tive concentration on cellulose acetate blend mem-
[25] Qin JJ, Oo MH, Cao YM, et al. Development of a LCST branes-preparation, characterization and application
membrane forming system for cellulose acetate ultrafiltra- studies. Sep. Sci. Technol. 2008;43:1933–1954. doi:10.
tion hollow fiber. Sep. Purif. Technol. 2005;42:291–295. 1080/01496390802063846.
doi:10.1016/j.seppur.2004.07.015. [40] Darvishmanesh S, Jansen JC, Tasselli F, et al. Novel poly-
[26] Velu S, Rambabu K, Muruganandam L. Development, phenylsulfone membrane for potential use in solvent
characterization and application studies of cellulose nanofiltration. J. Membr. Sci. 2011;379:60–68.
acetate – activated carbon blend ultra filtration mem- [41] Nagendran A, Mohan D. Cellulose acetate and polyether-
branes. Int. J. Chem Tech Res. 2014;6(ISSN: 0974-4290): imide blend ultrafiltration membranes: II. Effect of addi-
565–577. tive. Polym. Adv. Technol. 2008;19:24–35.
[27] Bai H, Zhou Y, Wang X, et al. The permeability and mech- [42] Han B, Zhang D, Shao Z, et al. Preparation and character-
anical properties of cellulose acetate membranes blended ization of cellulose acetate/carboxymethyl cellulose
with polyethylene glycol 600 for treatment of municipal acetate blend ultrafiltration membranes. Desalination.
sewage. Proc. Environ. Sci. 2012;16:346–351. doi:10. 2013;311:80–89. doi:10.1016/j.desal.2012.11.002.
1016/j.proenv.2012.10.049. [43] Hobbs C, Hong S, Taylor J. Effect of surface roughness on
[28] Zhou J, Chen J, He M, et al. Cellulose acetate ultrafiltration fouling of RO and NF membranes during filtration of a
membranes reinforced by cellulose nanocrystals: prep- high organic surficial groundwater. J. Water Supply Res.
aration and characterization. J. Appl. Polym. Sci. Technol. 2006;55:559–570. doi:10.2166/aqua.2006.038.
2016;133:1–7. doi:10.1002/app.43946. [44] Akbari A, Hamadanian M, Jabbari V, et al. Influence of
[29] Zhang L, Chen G, Tang H, et al. Preparation and character- PVDF concentration on the morphology, surface rough-
ization of composite membranes of polysulfone and ness, crystalline structure, and filtration separation proper-
microcrystalline cellulose. J. Appl. Polym. Sci. 2009;112: ties of semicrystalline phase inversion polymeric
550–556. doi:10.1002/app.29434. membranes. Desalin. Water. Treat. 2012;46:96–106.
[30] Yin J, Fan H, Zhou J. Cellulose acetate/poly(vinyl alcohol) doi:10.1080/19443994.2012.677524.
and cellulose acetate/crosslinked poly (vinyl alcohol) [45] Rajesh S, Maheswari P, Senthilkumar S, et al. Preparation
blend membranes: preparation, characterization, and and characterization of poly (amide-imide) incorporated
antifouling properties. Desalination Water Treat. cellulose acetate membranes for polymer enhanced
2015;57:10572–10584. doi:10.1080/19443994.2015. ultrafiltration of metal ions. Chem. Eng. J. 2011;171:33–
1040846. 44. doi:10.1016/j.cej.2011.03.033.
[31] Hua F, Wang YJ, Tsang F, et al. Study of microfiltration [46] Vetrivel S, Sri Abirami Sarasvathi M, Rana D, et al.
behavior of oily wastewater. J. Environ. Sci. Health. Part Fabrication of cellulose acetate nanocomposite mem-
A. 2007;42:489–496. doi:10.1080/10934520601188342. branes using 2D layered nanomaterials for macromolecu-
[32] Kanagaraj P, Nagendran A, Rana D, et al. Separation of lar separation. Int. J. Biol. Macromol. 2018;107:1607–1612.
macromolecular proteins and removal of humic acid by [47] Rajesh S, Shobana KH, Anitharaj S, et al. Preparation, mor-
cellulose acetate modified UF membranes. Int. J. Biol. phology, performance and hydrophilicity studies of poly
Macromol. 2016;89:81–88. (amide-imide) incorporated cellulose acetate ultrafiltra-
[33] Abedini R, Mousavi SM, Aminzadeh R. A novel cellulose tion membranes. Ind. Eng. Chem. Res. 2011;50:5550–5564.
acetate (CA) membrane using TiO2 nanoparticles: prep- [48] Karakulski K, Kozlowski A, Morawski AW. Purification of
aration, characterization and permeation study. oily wastewater by ultrafiltration. Sep. Technol. 1995;5:
Desalination. 2011;277:40–45. doi:10.1016/j.desal.2011. 197–205.
03.089. [49] Wu Y, Zhang X, Liu S, et al. Preparation and applications of
[34] Altinkaya SA, Yenal H, Ozbas B. Membrane formation by microfiltration carbon membranes for the purification of
dry-cast process: model validation through morphological oily wastewater. Sep. Sci. Technol. 2016;51:1872–1880.
studies. J. Membr. Sci. 2005;249:163–172. [50] Mittal P, Jana S, Mohanty K. Synthesis of low-cost hydro-
[35] Velu S, Muruganandam L, Arthanareeswaran G. Effect of philic ceramic-polymeric composite membrane for treat-
solvents on performance of polyethersulfone ultrafiltraion ment of oily wastewater. Desalination. 2011;282:54–62.
membranes for separation of metal ions. Int. J. Chem. [51] Kutowy O, Thayer WL, Tlgner J, et al. Tubular cellulose
Analyt. Sci. 2011;2:82–86. acetate reverse osmosis membranes for treatment of
[36] Arthanareeswaran G, Sriyamuna Devi TK, Raajenthiren M. oily wastewaters. Ind. Eng. Chem. prod. Res. Dev.
Effect of silica particles on cellulose acetate blend 1981;20:354–361.
 ENVIRONMENTAL TECHNOLOGY 1605

[52] Duong PHH, Chung TS, Wei S, et al. Highly permeable [54] Han G, de Wit JS, Chung TS. Water reclamation from emul-
double-skinned forward osmosis membranes for anti- sified oily wastewater via effective forward osmosis
fouling in the emulsified oil-water separation process. hollow fiber membranes under the PRO mode. Water.
Environ. Sci. Technol. 2014;48:4537–4545. Res. 2015;81:54–63.
[53] Mafirad S, Mehrnia MR, Sarrafzadeh MH. Effect of mem- [55] Duong PHH, Nunes SP, Chung TS. Dual-skinned polya-
brane characteristics on the performance of membrane mide/poly(vinylidene fluoride)/cellulose acetate mem-
bioreactors for oily wastewater treatment. Water Sci. branes with embedded woven. J. Membr. Sci. 2016;520:
Technol. 2011;64:1154–1160. 840–849.