J of Applied Polymer Sci - 2022 - Kandiyil - Efficient Selective Methylene Blue Adsorption by Polyurethane

Received: 31 May 2022 Revised: 11 November 2022 Accepted: 14 November 2022
DOI: 10.1002/app.53464
RESEARCH ARTICLE
Efficient selective methylene blue adsorption

by polyurethane/montmorillonite-based antifouling
electrospun composite membranes
Juraij Kandiyil 1 | Suni Vasudevan 2 | Sujith Athiyanathil 1
1
Department of Chemistry, Materials
Research Laboratory, National Institute of Abstract
Technology Calicut, Kozhikode, India A robust antifouling membrane with pollutant-removing capacity is in high
2
Department of Chemistry, Inorganic and demand for industrial wastewater treatment. Here, we report a new multifunc-
Bio-inorganic Laboratory, National
tional polyurethane/montmorillonite-oxidized multi-walled carbon nanotube
Institute of Technology Calicut,
Kozhikode, India (PU/MMT) nanofibrous composite membrane, beneficial for both water filtra-
tion and selective methylene blue adsorption. It was observed that oxidized
Correspondence
Sujith Athiyanathil, Department of
multi-walled carbon nanotube (o-MWCNT) imparted high flexibility with a
Chemistry, Materials Research high percentage elongation (318%–405%) to the composites, whereas montmo-
Laboratory, National Institute of rillonite (Mt) improved the hydrophilicity, antifouling property, and dye
Technology Calicut, Kozhikode, 673601,
India. adsorption capacity of the fibrous PU membrane. The sample with 20 wt% Mt
Email: [email protected] (PU/20-MMT) showed a minimum water contact angle of 57 . It showed a
high water flux, oil rejection, and flux recovery ratio compared to the pristine
electrospun PU membrane. The Hermia fouling mechanism suggested that
electrospun PU obeyed an intermediate blocking model, whereas PU/20-MMT
was best fitted with a cake formation model. During adsorption, the cationic
dye methylene blue (MB) was selectively adsorbed on the sample from a
methyl orange/methylene blue (MO/MB) mixture. It showed a maximum
adsorption capacity (qm) of 112.35 mg/g. The anionic dye methyl orange, on
the other hand, remained unadsorbed. The adsorption kinetics and the adsorp-
tion isotherm were pseudo-second-order kinetics and the Langmuir adsorption
isotherm, respectively. The developed, robust electrospun composite mem-
brane opens a new avenue for the selective adsorption of MB from dye mix-
tures and contaminated water.
KEYWORDS
antifouling, dye adsorption, electrospun membrane, montmorillonite clay, nanocomposite,
polyurethane
1 | INTRODUCTION Among them, methylene blue (MB) dye is hazardous to

the environment and human health. The most severe risk
For decades, industries related to cosmetics, textiles, paper, of industrial dyes is their resistance to degradation in natu-
food, plastics, and rubber have produced a tremendous ral environments and their adverse impact on the aquatic
amount of wastewater containing indispensable dyes.1 ecosystem.2 Because of the adverse effect of dyes on water,
J Appl Polym Sci. 2023;140:e53464. wileyonlinelibrary.com/journal/app © 2022 Wiley Periodicals LLC. 1 of 18

https://doi.org/10.1002/app.53464
10974628, 2023, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app.53464 by The Director National Institute Of Technology Calicut, Wiley Online Library on [06/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 of 18 KANDIYIL ET AL.
their removal and purification are a global challenge.3–7 composite PU membrane with a reduced graphene oxide-
Hence, the development of effective methods and mate- titaninum dioxide (r-GO-TiO2) hybrid filler. Significant oil
rials for dye removal from wastewater is highly needed. rejection and water flux were observed compared to the
The removal of dyes from wastewater has been pristine electrospun PU membrane.
achieved using different techniques such as precipitation, Nanomaterials can contribute a variety of functional
chemical oxidation, coagulation, photocatalysis, and properties, apart from antifouling properties, to virgin
adsorption.8,9 Adsorption has been explored in particular polymeric membranes.21 Many inorganic, organic, and
because of its benefits of low cost, effectiveness, ability to hybrid fillers have been used to fabricate polymer nano-
reuse and regenerate adsorbents, and ease of surface func- composites.22 Multi-walled carbon nanotube (MWCNT)
tionalization of the adsorbent.10 Among the various adsor- is an excellent reinforcing nanomaterial for producing
bent materials, polymer membranes have gained much high-performance polymer composite membranes by
attention in dye adsorption due to their high sorption effi- virtue of its solitary properties, such as exceptional
ciency, cost-effectiveness, and lack of secondary pollu- mechanical strength and tunable chemical and electrical
tion.11 Among the membranes, electrospun membranes properties.23 Oxidation of MWCNT generates many
are highly versatile because of their high porosity, ease of functional groups such as C O, OH, and COOH on
surface functionalization, tunable fiber diameter, high sur- its surface. It converts the hydrophobic nature of
face area-to-volume ratio, fabrication flexibility, and MWCNT into a hydrophilic nature; hence, it is suitable
enhanced filtration ability.12 for water filtration applications.24 Montmorillonite
However, fouling on polymer electrospun membranes (Mt) clay [(Na, Ca)0.33(Al, Mg)2(Si4O10)(OH)2nH2O] is a
is a serious problem in water filtration applications as well valuable smectite group clay mineral that can be found
as dye removal from wastewater.13 It reduces water perme- in nature.25 Many features, such as cation exchange,
ability and, hence, purification efficiency. Hydrophobic electrical conductivity, heat resistance, water sorption
moieties such as oil and protein often act as foulants. In capacity, and antibacterial capabilities, make them a
this scenario, the antifouling ability of the membrane is better material for fouling tolerance with increased per-
considered an essential property for water treatment. Many formance.26 Mt is employed in multiple disciplines,
strategies, such as composite preparation, blend formation, including food additives, catalysis, antimicrobial, and
and surface functionalization were adopted for the develop- sorbents. Exclusively, it is an excellent material for
ment of antifouling membranes.14 Recently, Ouyang et al.15 increasing the hydrophilicity and selective adsorption
developed an antifouling polypropylene fibrous membrane capacity of cations such as cationic dyes and heavy
(PPM) by grafting poly(N-vinylpyrrolidone) onto PPM. The metals.27
designed fabrics showed adsorption toward organic micro- Synthetic polymers such as polyacrylonitrile (PAN),28,29
pollutants. Gandhimathi et al.16 converted hydrophobic PES,30 cellulose acetate (CA),31 polyesters,32 PVDF,33 and
poly(vinylidene fluoride) (PVDF) into a hydrophilic electro- PU34 are widely used in water filtration applications.
spun membrane by blending it with a polyethylene glycol Among these polymers, thermoplastic polyurethane
copolymer. It showed significant water flux due to the (TPU) has some advantages. It has high hydrolytic,
superhydrophilicity and large pore size. Similarly, Meimei mechanical, and thermal stability.35 But pristine PU
et al.17 prepared a highly hydrophilic PVDF-based electro- polymer has a fouling problem due to its hydrophobic
spun membrane by blending it with poly(vinyl alcohol) backbone.36 Fouling-tolerant PU electrospun membrane
(PVA). The optimized PVDF/PVA blended electrospun is considered a good candidate in water filtration appli-
membrane exhibited a high water flux of 45,900 Lm1 h1 cations. Apart from antifouling properties, electrospun
and significant antifouling performance. Issa et al.18 used membranes with dye removal capacity have gained
a hydrophilic polyvinylpyrrolidone (PVP) polymer to much attention in wastewater treatment. There are few
improve the hydrophilicity of polyethersulfone (PES). The reports on dye removal from wastewater using PU-based
developed PES/PVP blended electrospun membrane materials based on photocatalytic dye degradation rather
showed a high water flux compared to the PES membrane. than dye adsorption. Rajeswari et al.37 developed a photo-
A few attempts were made to prepare polyurethane-based catalytic material based on cellulose acetate-polyurethane
hydrophilic electrospun membranes. Suja et al.19 utilized a (CA-PU) electrospun membrane by employing zinc oxide
blending strategy for developing antifouling polyurethane nanoparticles as a functional filler. Significant degradation
(PU)-based electrospun membranes using hydrophilic PVP. of dyes such as reactive orange 84 and reactive red 11 was
The PU/PVP blend system showed an appreciable reduc- observed in water. Stefano et al.34 fabricated a TiO2
tion in fouling and increased water flux compared to the nanoparticle-modified PU electrospun composite mem-
pristine electrospun PU membrane. On the other hand, brane for the photocatalytic degradation of dyes present in
Suja et al.20 prepared a fouling-tolerant electrospun wastewater.
KANDIYIL ET AL. 3 of 18
Although a few articles have investigated the develop- TABLE 1 Formulations of PU/MMT electrospun composite
ment of antifouling PU nanofibrous membranes, none of membranes
them have explored the fabrication and performance of Sample PU (%) o-MWCNT (wt%) Mt (%)
Mt clay-modified PU electrospun composite membranes
PU 100 0 0
for synergistic water filtration and selective MB adsorp-
tion. Herein, for the first time, we present an antifouling, PU/5-MMT 95 0.02 5
robust montmorillonite/oxidized multi-walled carbon PU/10-MMT 90 0.02 10
nanotube (Mt/o-MWCNT) hybrid filler-based PU electro- PU/15-MMT 85 0.02 15
spun composite membrane for both water filtration and PU/20-MMT 80 0.02 20
selective dye adsorption. The developed membrane showed
versatile properties such as high mechanical strength,
hydrophilicity, and dye adsorption capacity compared to
the pristine electrospun PU. Thus, the major purpose of
this article is to develop an antifouling, robust PU-based
electrospun composite membrane and investigate the selec-
tive adsorption of MB from the MB/MO dye mixture.
2 | MATERIALS AND METHODS
2.1 | Materials
Lubrizol Advanced Materials, Massachusetts, USA, pro-

vided TECOFLEX™ EG-85A grade thermoplastic poly-
urethane. Sigma Aldrich, Germany, supplied MWCNT
powder with inner and outer diameters of 3–5 nm and
10–14 nm, respectively. As per the production details, the
length of the MWCNT is 0.1–10 μm. Dyes such as methy- F I G U R E 1 Schematic representation of electrospun composite
lene blue and methyl orange were brought from Sisco membrane fabrication [Color figure can be viewed at
Research Laboratories Pvt. Ltd., India. Montmorillonite wileyonlinelibrary.com]
clay with a particle size of 25 micron was obtained from
Sigma-Aldrich Chemical Pvt. Ltd., India. Avra Chemi-
cals, India, provided dimethyl formamide (DMF) and tet- Initially, a 0.02 wt% dispersion of o-MWCNT was pre-
rahydrofuran (THF) solvents. pared in a 3 ml DMF and 7 ml THF solvent mixture. It
was further ultrasonicated for 30 min. To the o-MWCNT
dispersion, a specific amount of Mt (5–20 wt%) was
2.2 | Oxidation of MWCNT added. Subsequently, it was sonicated for 30 minutes. A
suitable amount of PU pellets (0.8–1 g) was gently added
The MWCNT was oxidized with 10 ml of 30% H2O2. to the Mt/o-MWCNT dispersion and stirred for 6 h to
MWCNT (0.5 g) was kept under reflux in H2O2 for six ensure complete dissolution. The formulations of the
days at 120 C. A sufficient concentration of H2O2 was samples are given in Table 1.
maintained by supplying 5 ml of H2O2 daily. To remove The PU/MMT spinning solution was filled into a
contaminants, the oxidized MWCNT was thoroughly syringe (10 ml) with a metallic spinneret with an inner
washed with deionized water. It was filtered via a milli- diameter of 0.514 mm (21G 100 ). The syringe was con-
pore PVDF membrane with a 0.45 μm pore size. The fil- nected to a high-voltage supply electrode, which was
trate was vacuum-dried overnight at 110 C. placed over a syringe pump (0–30 kV, HO-NFES-041
Holmarc, India). The other electrode was banded
together with a rotating mandrel at 500 rpm. The electro-
2.3 | Preparation of PU/MMT spinning process was prolonged for another 10 h to
nanofibrous composite membrane achieve the requisite membrane thickness. The nanofi-
brous membranes were subjected to vacuum drying for
The electrospinning solutions were primarily prepared by 24 h to remove the surplus solvent. Electrospun mem-
the dissolution of 10 wt% of PU in a 7THF: 3DMF system. brane fabrication is depicted in Figure 1.
2.4 | Characterization where d represents the average fiber diameter and ε

denotes the porosity.
The bonding characteristics were studied using attenuated
total reflectance-Fourier transform infrared spectroscopy
(ATR-FTIR) between 4000 and 400 cm1 (FT-IR) (Jasco 2.6 | Water flux, fouling model, flux
FT/IR-4700). The morphology of samples was analyzed by recovery ratio, and rejection ratio
a field emission scanning electron microscope (FE-SEM)
(Hitachi 6600) at an accelerating voltage of 5 kV. Image J A stirred ultrafiltration cell with a cross-sectional area of
software was employed to determine the average fiber 4.1 cm2 was used for water flux analysis. The reservoir
diameter. Fifty fibers were chosen at random from each was filled with 1000 ml of distilled water for flux mea-
membrane for this purpose. The o-MWCNT functionaliza- surements. N2 gas was employed to keep an appropriate
tion was confirmed by Raman spectroscopy (Horiba Lab- pressure during the experiment. At various pressures
Ram Confocal Raman Spectrometer, 532 nm laser). The (5, 10, 15, 20, 25, 30, and 35 psi), water flux was mea-
thermal characteristics of membranes were investigated sured. Equation (4) was used to determine the pure water
using TGA (TGA Q-50, TA instruments). The experiment flux (J).40
was performed between 0 and 600 C, with a scan rate of
5 C/min. A Universal Testing Machine (UTM, Schimadzu V
J¼ , ð4Þ
Autograph, AG-Xplus series) with a 10 N applied stress A Δt
was used to analyze the mechanical characteristics of the
membranes. At room temperature, the crystalline property where V represents the volume of permeating liquid in
was investigated by x-ray diffraction (XRD) (Miniflex liter, A denotes the membrane area (m2), and t is the flux
600 Rigaku X-ray generator, Cu Kα radiation with time (h).
λ = 1.5406 Å). Equation (1) was used to calculate the per- The membrane was tested for oil rejection using a
centage crystallinity of electrospun membranes.38 1 g/L oil–water emulsion. The ultrasonication strategy
was used to prepare an oil–water emulsion. UV–visible
Crystalline peak area spectroscopy was employed to examine the concentration
Percentage crystallinity ¼ 100: ð1Þ
Total peak area of permeate. Standard emulsions (0–150 ppm) were pre-
pared, and absorbance was monitored to evaluate the
A GBX Digidrop goniometer was used to determine unknown concentrations of permeate solution. The
the membrane's wettability. The adsorption properties of membrane was thoroughly washed using distilled water
the PU/MMT electrospun samples were investigated by a after emulsion filtration, and the pure water flux was
UV–Visible spectrophotometer (UV-2600 Shimadzu). tested again to evaluate the membrane's efficiency.
Equation (5) was used to calculate the oil rejection.41
2.5 | Pore size and porosity Cf Cp

Rejection ð%Þ ¼ 100, ð5Þ
measurement Cf
The porosity (ϵ%) of nanofibrous membranes was evalu- where Cp and Cf denote the permeate and feed emulsion
ated using Equation (2).39 concentrations, respectively.
The flux recovery ratio (FRR) was measured by
ðW w W d Þ=ρw Equation (6).41
ε¼ 100, ð2Þ
ðW w W d Þ=ρw þ ðW d =ρPU Þ
J W2
FRR ð%Þ ¼ 100, ð6Þ
where Wd and Ww denote the mass of the dry and wet J W1
nanofibrous membranes, respectively. The ρpu and ρw
represent the densities of PU (ρpu = 1.05 g/cm3) and deio- where JW1 is the water flux before oil rejection study and
nized water, respectively. JW2 is water flux after oil rejection study.
The pore size (r) of the nanofibrous membranes was eval- To obtain the fouling model, the flux of the oil–water
uated by employing the equation given below (Equation 3).38 emulsion was analyzed at 5 psi for 2 h. Various fouling
models responsible for flux reduction at constant pres-
pffiffiffi !
π π sure were proposed by Hermia. The following mathemat-
r¼ 1 d, ð3Þ ical equation is used to present flux decline under
4 2 log 1ε
constant pressure.42
T A B L E 2 Various fouling models

Fouling model m Expression of flux Linear form
proposed by Hermia
Cake formation 0 J ¼ pffiffiffiffiffiffiffiffiffiffi
J0
1þJ kt
1
J2
¼ J12 þ kt
0
0
¼ ð1þJJ 00 ktÞ J ¼ J 0 þ kt
Intermediate blocking 1 1 1
J
Standard blocking 1.5 J¼ J0

pffiffiffiffi 2 p1ffiffi ¼ p1ffiffiffiffi þ kt
ð1þ J 0 kt Þ J J0
1
Complete blocking 2 J ¼ J 0 ekt ln ¼ ln 1
þ kt
J J0
m
d2 t dt where C0 and Ct denote the initial and concentration at
¼ k , ð7Þ
dV 2 dV time t, respectively (mg/L), V represents the volume in
liter, and M denotes the weight of the adsorbent (g).
where V denotes the volume, k represents resistance coef- The dynamic filtration of MB from MB solution was
ficient, t is the filtration time, and m is the fouling model investigated using dead-end ultrafiltration apparatus.46
constant. 10 ppm MB solution was filled into the reservoir, and N2
The best fouling mechanism of the PU and PU/MMT gas was employed to apply 5 psi pressure during the
electrospun membranes was identified using the linear experiment. MB filtration efficiency was calculated using
form of the Hermia model (Table 2). The cake formation Equation (9). Filtration was continued for five filtration
model suggests that the diameter of foulants is higher cycles to evaluate the reusability of the membrane. The
than that of pores, which results in the formation of a adsorbed MB on the electrospun membrane was des-
foulant layer on the membrane surface. The chance of a orbed for the subsequent filtration test. Further, the reus-
pore getting blocked by a foulant that is approaching the ability of the PU/MMT electrospun membrane was
surface is described by the intermediate blocking model. expressed by percentage removal efficiency (R%) and was
The standard blocking model suggests that particle evaluated by adsorption–desorption experiments in dupli-
adsorption on the pore walls of the membrane causes a cate for both batch and dynamic adsorption. The desorp-
reduction in pore width. According to the complete tion of MB from the composite membrane was performed
blocking model, each foulant that reaches the surface by treating it with a 1 N HCl solution. For 2 h, the MB
blocks a fraction of the pores.43 adsorbed PU/MMT fibrous membrane was stirred with
HCl solution at 300 rpm. It was continued for 10 h by
adding the membrane to a newly prepared HCl solution
2.7 | Batch methylene blue adsorption at a 2-h regular interval. Finally, the desorbed membrane
and dynamic filtration was thoroughly washed with distilled water, dried and
again subjected to adsorption–desorption experiments.
Batch adsorption studies were carried out to validate the The percentage removal efficiency (R%) was measured by
effects of pH, the initial dye concentration, temperature, Equation (9).7
and contact time.44 Dried PU/MMT electrospun com-
posite membranes were added to 10 ml of an equimolar C0 Cf
R ð %Þ ¼ 100, ð9Þ
mixture (10 ppm) of methyl orange (MO) and MB solu- C0
tion and stirred at 300 rpm in a batch condition. Adsorp-
tion kinetics was evaluated by measuring the amount of where C0 and Cf stand for the initial and final MB con-
MB adsorbed on the composite membrane with respect centrations, respectively.
to time. The adsorption equilibrium isotherm was evalu-
ated by analyzing the amount of MB adsorbed at various
MB concentrations (10–50 ppm). At 665 nm, the 3 | RESULTS A ND DISCUSSION
adsorbed amount of MB was calculated using a UV–VIS
spectrophotometer. The adsorbed amount of MB (qt; 3.1 | Morphology evaluation
mg/g) on the PU/MMT membrane was calculated using
Equation (8).45 A SEM micrograph of the pristine electrospun PU mem-
brane is shown in Figure 2a. Figure 2b shows a SEM image
C0 Ct of the o-MWCNT-incorporated PU electrospun composite
qt ¼ V, ð8Þ
M membrane (PU/o-MWCNT). SEM micrographs of
F I G U R E 2 SEM
micrographs of (a) PU
electrospun membrane,
(b) PU/o-MWCNT electrospun
membrane, and PU/MMT
electrospun membrane at
various Mt loadings, (c) 5 wt%,
(d) 10 wt%, (e) 15 wt%, (f)
20 wt%
F I G U R E 3 (a) Graph
showing the variation in average
fiber diameter of different
samples and (b) variation of
porosity and pore size of various
electrospun samples [Color
figure can be viewed at
wileyonlinelibrary.com]
polyurethane/montmorillonite-oxidized multi-walled car- fibers. Significant growth in fiber diameter was noted for
bon nanotube (PU/MMT) nanofibrous composite mem- 20 wt% Mt-loaded electrospun fiber. It was due to the over-
branes at four different Mt loadings (5, 10, 15, and 20%) are charged density of the polymer solution, which resulted in
given in Figure 2c–f. TEM micrographs of neat o-MWCNT the fast jet of the polymer solution.38 The fast polymer jet
and PU/MMT electrospun composite fiber are given in had a shorter flight time during the whipping motion to
Data S1. It showed axially aligned o-MWCNTs along the stretch and elongate the polymer jet properly. As a result,
PU/MMT electrospun fiber. Mt loading had a significant the production of fibers with a large fiber diameter was
impact on morphology as well as the fiber diameter. As Mt observed.
loading was increased, interconnected, narrow fibers were Pore size and porosity are the important factors affect-
formed. Pristine electrospun PU showed the largest fiber ing the filtration performance of membranes (Figure 3b).
diameter of 615 nm, whereas PU/o-MWCNT electrospun The pristine PU membrane showed a minimum porosity of
membranes showed a fiber diameter of 463 nm (Figure 3a). 48.39%. As the Mt loading increased, porosity also
This reduction in the fiber diameter may be due to the increased. PU/20-MMT showed the highest porosity of
enhanced electrical conductivity of the o-MWCNT- 61.43%. This enhanced porosity caused better filtration per-
modified PU solution. Further fiber diameter reduction was formance in composite samples. Since the pore size
observed for Mt/o-MWCNT hybrid filler-modified compos- depends on the porosity and fiber diameter, there was no
ite nanofibrous membranes. A progressive decrease in the regular trend in pore size variation. However, electrospun
fiber diameter was observed from 344 to 254 nm as Mt PU showed the largest pore size of 1130.50 nm, and
loading was increased from 5 to 15 wt%, respectively. The PU/20-MMT showed a relatively large pore size among all
improved electrical conductivity of the polymer solution PU/MMT composite membranes. According to the Hagen–
due to the presence of exchangeable free ions in the Mt was Poiseuille equation, a membrane with large porosity and
responsible for the decrease in fiber diameter.47 Owing to a pore size can show better water permeability.48 So, the
more extensive distribution over the polymer solution, the PU/20-MMT electrospun composite membrane is consid-
polymer jet elongates during flight, resulting in narrow ered a promising material for water filtration applications.
F I G U R E 4 ATR-FTIR
spectra of (a) electrospun PU,
Mt, and composite membranes,
(b) enlarged portion of FTIR
showing intensity variation,
(c) Raman spectra of o-MWCNT
and PU/MMT electrospun
membranes. XRD patterns of
(d) Mt, o-MWCNT, and
electrospun PU (e) PU/MMT
composites at various Mt
loadings [Color figure can be
viewed at
3.2 | Bonding and crystalline properties TABLE 3 Percentage crystallinity and peak position of samples
in XRD
The ATR-FTIR spectra of various samples are shown in Sample Peak position (degree) Crystallinity (%)
Figure 4a. The nanofibrous PU showed an absorption peak PU 18.64 58.86
at 3325 cm1, responsible for the N H group stretching.
PU/5-MMT 19.88 69.93
The stretching frequencies at 1688 and 1716 cm1 were
PU/10-MMT 19.80 68.25
due to the intermolecular hydrogen-bonded C O stretch-
ing and bare carbonyl stretching (C O), respectively.38 Mt PU/15-MMT 20.04 65.98
showed a broad absorption band at 3401 cm1 related to PU/20-MMT 19.83 64.02
the OH stretching vibration (υOH) of interlayer water
molecules. The peak at 3621 cm1 was due to the υOH of
the OH groups coordinated to octahedral cations in the samples confirmed that o-MWCNT had properly modified
crystal lattice. The Si O out-of-plane and in-plane stretch- the PU. The FTIR and Raman analysis of o-MWCNT are
ing were represented by 1110 and 987 cm1, respectively. provided under Data S2.
Al Al OH, Al O Si, and Si O Si bending were shown Figure 4d shows the XRD patterns of various samples.
at 917, 512, and 440 cm1, respectively. All the characteris- Table 3 shows the 2θ value and percentage crystallinity of
tic peaks of PU and Mt were visible in the PU/MMT elec- fibrous membranes. The percentage crystallinity was
trospun composite membrane. In the composite samples, increased gradually from the electrospun PU membrane to
the peaks of Mt at 987, 512, and 440 cm1 were slightly the PU/10-MMT electrospun composite membrane. The
shifted to 1030, 517, and 460 cm1, respectively. Moreover, improved crystallinity was mostly due to the enhanced
the intensity of these peaks was increased in composites as molecular order within the nanofibers and the proper dis-
Mt loading was increased (Figure 4b). It may be due to the tribution of smooth fibers.50 The crystallinity of the
proper distribution of Mt in the PU matrix and strain- PU/20-MMT composite membrane, on the other hand,
induced crystal phase change due to the polymer infu- deteriorated slightly due to the increased fiber diameter
sion.49 Figure 4c shows the Raman spectra of o-MWCNT and random fiber distribution. With a percentage crystal-
and PU/MMT nanofibrous membranes. The o-MWCNT linity of 58.86%, electrospun PU showed a pronounced
showed an absorption peak at 1340 cm1, corresponding peak at 18.64 ; thus, nanofibrous PU is semi-crystalline. It
to the D band, and another peak at 1573 cm1, represent- was mainly due to the proper orientation of nanofibers and
ing the G band. These prominent peaks in PU/MMT the intermolecular hydrogen bonding in hard segments.19
F I G U R E 5 Mechanical
properties of samples (a) stress–
strain curve, (b) Young's
modulus and tensile strength,
(c) percentage elongation,
(d) TGA curves of samples
[Color figure can be viewed at
The o-MWCNT showed major XRD patterns at 3.3 | Mechanical and thermal properties
78.58 , 44.86 , and 25.95 , which correspond to planes
(100), (110), and (002), respectively.51 Mt showed Flexible and high-strength electrospun membranes have
its crystalline nature with a series of diffraction patterns received massive interest in the filtration sector because
at 5.8 , 19.17 , 29.16 , 35.01 , and 61.86 corresponding of their long-term use. The stress–strain curve of all sam-
to the planes (001), (100), (005), (110), and (300), ples is shown in Figure 5a. Due to the sliding tendency of
respectively (JCPDS card number: 46–1045).52 The x-ray randomly distributed interconnected nanofibers over others,
diffractograms of PU/MMT nanofibrous composite all samples had a nonlinear initial elastic region.57 Electro-
membranes at different Mt compositions are shown spun PU showed the least tensile strength (3.28 MPa) and
in Figure 4e. The peak corresponding to o-MWCNT the least elongation at break (209%) (Figure 5b,c). 20 wt%
was merged with the PU peak at 2θ = 18.64 in the Mt alone modified PU electrospun membrane (PU/Mt)
PU/MMT composite membrane, and the peak of PU showed better tensile strength (6.16 MPa) and percentage
was shifted to a higher 2θ value compared to the pristine elongation (318%) than electrospun PU. Our previous work
PU, which may be due to the interaction between the showed the existence of an outstanding percentage elonga-
PU and fillers. The prominent peak of Mt at 5.8 showed tion (502%) of MWCNT alone modified superoleophilic
that the basal spacing (d001) was 1.52 nm, which con- electrospun PU.38 In this work, Mt was employed for the
firmed the existence of Ca-montmorillonite.53 In the development of hydrophilic membranes. It was observed to
composite samples, a slight shift in the peak correspond- have an intermediately significant tensile strength and per-
ing to the basal spacing occurred (6.06 –6.22 ), which centage elongation for Mt/o-MWCNT hybrid filler-modified
was attributed to the change in lattice parameters dur- samples (PU/MMT).
ing the formation of the composite due to the polymer The PU/15-MMT composite showed the highest tensile
infusion.54,55 As Mt loading increased, the intensity of strength of 8.58 MPa due to the exceptional mechanical
the basal peak also increased. At lower Mt loading, the properties of o-MWCNT, and the uniform distribution of
intensity of the basal peak was found to be very low Mt/o-MWCNT hybrid filler in the PU matrix.62 The
compared to the higher loading, probably due to the PU/20-MMT composite membrane exhibited a significant
proper distribution of Mt at lower concentrations and elongation at break of 405%. It confirmed the improved
the exfoliation of a minor fraction of Mt due to the inter- flexibility of electrospun composite membranes compared
action of the PU backbone.56 to the virgin fibrous PU membrane (inset of Figure 5a).
T A B L E 4 Comparison data of the flexibility, adsorption capacities, and water flux of previously reported adsorbents with PU/MMT
electrospun membrane
Elongation at Adsorption capacity Water flux

Adsorbents break (%) (qm) (mg/g) (lm2 h1) References
PU/GO electrospun membrane 109.48 109.88 17,706 58
PDA modified cellulose acetate Not reported 88.2 Not reported 59
nanofiber
β cyclodextrin-based electrospun fiber 9.19 826.45 Not reported 60
p-PEN/a-CNTs@TA/CC electrospun Not reported 633 Not reported 61
membrane
PU/MMT electrospun membrane 405 112.35 3785 This work
TABLE 5 TGA data of samples
Sample Tonset ( C) Tmax1 ( C) Tmax2 ( C) Residue (%) Filler content (%)

Mt 26.18 56.63 - 82.44 99.98
PU 256.45 305.33 392.86 1.33 0
PU/5-MMT 291.79 312.58 396.48 9.46 4.48
PU/10-MMT 286.45 313.33 390.74 16.27 8.75
PU/15-MMT 287.45 320.09 401.18 19.06 12.46
PU/20-MMT 287.82 310 395.18 21.93 18.07
The obtained percentage elongation is higher than the slightly less than the theoretical Mt loading (input load-
reported electrospun membranes for dual functionalities ing) due to the mass loss during electrospinning.
such as water filtration and methylene blue adsorption
from contaminated water. The comparative data is shown
in Table 4, indicating that the present work could produce 3.4 | Hydrophilicity, water flux,
a highly flexible electrospun membrane with antifouling and antifouling properties
and methylene blue adsorption properties.
TGA curves (Figure 5d) and derivative thermogravi- Hydrophilicity is a crucial parameter of the membrane in
metry (DTG) (Data S3) revealed that both PU and water filtration applications. Figure 6a shows the contact
PU/MMT nanofibrous membranes have two-step degra- angle of electrospun PU and all composite samples. Pris-
dation. The putrefaction of the CO NH bond in the tine fibrous PU showed the highest water contact angle
hard segment caused the first weight loss, while the of 117 ± 2 , mainly due to the hydrophobic backbone of
breakdown of the C O C bond in the soft segment PU. As the loading of Mt was increased, a progressive
caused the second weight loss.20 Due to the intercalated decline in the contact angle was observed.
water molecules, Mt showed a weight loss in the range of The PU/20-MMT nanofibrous composite membrane
31–105 C. The onset and Tmax temperatures of all showed a minimum contact angle of 57 , which was attrib-
PU/MMT nanofibrous composite membranes were ele- uted to the higher loading of hydrophilic Mt. Hydration is
vated considerably compared to the virgin fibrous PU an inherent property of Mt by virtue of hydrophilic func-
membrane, showing the improved thermal stability of tional groups like OH and exchangeable ions such as
the composites (Table 5). It may be due to the uniform Na+, K+, and Ca2+.65 Increased hydrophilicity prevents the
distribution of thermally stable Mt/o-MWCNT hybrid adhesion of foulants during water filtration.
filler in the polymer matrix. As Mt filler loading Water flux is a vital quantity to consider while evalu-
increased, the percentage of residue also increased. The ating the filtration performance of the membrane. The
calculated percentage filler content is given in Table 5, pure water flux of PU/MMT electrospun membranes was
which was evaluated using TGA.63,64 A detailed explana- measured using a stirred ultrafiltration cell. Figure 6b
tion is given in Data S3. It was found that the percentage shows the pure water flux of various electrospun compos-
filler content in the electrospun composite samples was ite membranes at a range of pressures (5–35 psi). In all
F I G U R E 6 (a) Water
contact angle of electrospun PU
and PU/MMT composites,
(b) water flux of different
samples at different pressure,
(c) water flux of different
samples at different spinning
duration, (d) oil rejection and
FRR of PU and PU/MMT
electrospun membranes [Color
PU/MMT membranes, the water flux was increased up observed for virgin PU fibrous membrane due to the
to 25 psi (1.76 kg/cm2). However, flux declined above presence of the lipophilic backbone of electrospun
25 psi by reason of membrane flattening, which caused PU. The oil rejection percentage for PU/20-MMT was
changes in fiber orientation, porosity, and pore size. At found to be 95.45%. The hydration layer developed on the
up to 25 psi, all composite membranes had adequate nanofibrous membrane was primarily responsible for
water flux, indicating that the membrane morphology the antifouling properties due to the improved hydrophi-
remained intact even at high pressure. Figure 6c shows licity of the composite membrane. Moreover, the small
the water flux at a fixed feeding pressure of 25 psi with pore size (699.42 nm) compared to the size of oil droplets
different spinning durations, such as 10 and 20 h. Elec- (12236.13 nm) prevented the permeation of oil droplets
trospun membranes spun for 10 h had a thickness of through the nanofibers. To support this, an optical image
about 0.054 mm, whereas those spun for 20 h had a of the oil–water feed solution before the rejection study
thickness of around 0.175 mm. The flux of the 10 h and permeate after the rejection study is given in Data S4.
electrospun membrane was larger than that of the 20 h The feed solution showed many oil droplets, with an aver-
electrospun membrane. It may be due to the higher age diameter of 12236.13 nm, whereas the permeate was
porosity and larger pore size. A 10 h spun pristine PU free from oil droplets. The FRR was also used to assess the
fibrous membrane showed the lowest water flux of antifouling behavior of PU/MMT membranes. FRR
935 Lm2 h1, and it was due to its hydrophobic nature increased as Mt concentration in the membrane increased.
and reduced porosity. In terms of water flux, Mt-loaded The highly hydrophilic PU/20-MMT electrospun compos-
electrospun membranes surpassed pure PU mem- ite membrane showed the highest FRR of 98.45% for the
branes. The improved hydrophilicity, large pore size, first oil–water filtration cycle. Moreover, the FRR of
and porosity of the PU/20-MMT electrospun composite PU/20-MMT was tested for six oil–water filtration cycles
membrane significantly increased water flux to 4029 to investigate dynamic fouling (Data S4, Figure S5c). Elec-
Lm2 h1. So, PU/20-MMT nanofibrous membrane has trospun PU showed a significant decline in FRR due to the
been chosen for further antifouling and adsorption severe fouling caused by the adhered oil foulant on the
studies. hydrophobic backbone. The PU/20-MMT composite mem-
The antifouling behavior of electrospun membranes brane showed only a slight decrease in the FRR as the
was studied by percentage rejection (R%) and flux recov- number of cycles was increased, probably due to the com-
ery ratio (FRR). Figure 6d shows the oil rejection percent- paction of fibers during dynamic filtration. Hence, the
age and FRR of PU/MMT membranes. The oil rejection electrospun composite membrane is more suitable for
of Mt-modified membranes increased as Mt content water filtration applications with promising antifouling
arose. A minimum oil rejection ratio of 45.5% was properties than the pristine electrospun PU.
F I G U R E 7 (a) Flux decline of oil emulsion through PU and PU/20-MMT electrospun membranes. Linear fitting of the PU nanofibrous
membranes for different Hermia models, (b) m = 0, (c) m = 1, (d) m = 1.5, (e) m = 2. Linear fitting of the PU/20-MMT electrospun
composite membranes for different Hermia models (f) m = 0, (g) m = 1, (h) m = 1.5, (i) m = 2 [Color figure can be viewed at
3.5 | Theoretical study of fouling foulant formed an oily layer on the surface of the electro-
mechanism spun membrane. The enhanced hydrophilicity of the Mt-
modified membrane reduced the synergy between the oil
Figure 7a shows the flux decline during oil–water emul- foulant and the membrane surface. Hence, the oily cake
sion filtration. The pristine PU electrospun membrane layer on the PU/20-MMT membrane surface can be read-
showed a significant flux decline due to severe fouling ily wiped away. So, the PU/MMT nanofibrous composite
problems. The hydrophobic surface was responsible for membrane is considered a promising material against
this increased fouling. The PU/20-MMT electrospun com- fouling during water filtration.
posite membrane showed a reduced flux decline com-
pared to the PU, mostly due to the hydrophilic surface of
the sample. 3.6 | Selective MB adsorption
Figure 7b–e shows the fitting curves of nanofibrous from MO/MB mixture
PU, and Figure 7f–i depicts the fitting curves of optimized
PU/20-MMT electrospun composite membrane for vari- The selective dye adsorption property of an adsorbent is
ous Hermia fouling models (m = 0, 1, 1.5, and 2). Table 6 an attractive characteristic in dye removal from contami-
shows the R2 values obtained from the Hermia model for nated water. An adsorption study was performed on an
PU and composite membranes. The intermediate block- equimolar solution of a methylene blue/methyl orange
ing model appeared to be the best fit for the PU electro- mixture (MB/MO). The UV–Visible spectra of the selec-
spun membrane. On the other hand, the PU/20-MMT tive adsorption of MB from MB/MO mixture are given in
membrane provided the best fit for the cake filtration Figure 8a. The color of the mixture gradually changed
model. According to the cake formation model, the oil from dark green to bright yellow with the addition of the
TABLE 6 R2 and k values for different Hermia fouling models
Samples
PU PU/20-MMT
Model k R2
k R2
m=0 1.40825 106 0.94825 4.43269 107 0.98535
4 5
m=1 1.18821 10 0.98512 5.69587 10 0.91867
4 4
m = 1.5 9.46123 10 0.94442 4.8759 10 0.82607
m=2 3.8150 102 0.7972 1.7340 102 0.68667
F I G U R E 8 (a) UV–visible
spectra of adsorption of MB
from an MB/MO mixture using
the PU/20-MMT electrospun
membrane, (b) MB adsorption
isotherm of different PU/MMT
electrospun composite
membranes [Color figure can be
viewed at
PU/20-MMT electrospun composite membrane (inset of 3.7 | Adsorption kinetics and adsorption
Figure 8a). After 15 min, the intensity of the peak at isotherm
665 nm corresponding to MB was noticeably reduced, but
there was no noticeable decline at 464 nm related to Since no adsorption of MO had been observed, detailed
MO. It showed that the electrospun composite membrane adsorption studies on MB were performed. Adsorption
had good selectivity for the adsorption of MB from an kinetics were investigated using pseudo-second-order and
MB/MO solution. The negatively charged Mt surface is pseudo-first-order kinetic models. Equation (10) repre-
accountable for the adsorption of the cationic dyes. Iso- sents the pseudo-first-order model.67
morphically substituting Fe3+ or Al3+ in place of Si4+ in
the tetrahedral sheet is responsible for the negatively ln ðqe qt Þ ¼ ln ðqe Þ k 1 t, ð10Þ
charged Mt surface.66 The adsorption phenomena was
caused by electrostatic interaction between the negatively where qt (g/g) represents the adsorption capacity at t, qe
charged Mt on the PU/MMT electrospun composite (g/g) denotes the equilibrium adsorption capacity, and k1
membrane and the positively charged MB.52 Hence, the stands for the pseudo-first-order rate constant (min1).
negatively charged anionic dye MO remained unad- Equation (11) describes the pseudo-second-order
sorbed due to the electrostatic repulsion. model.67
Figure 8b shows the amount of MB adsorbed (qt) on
the various PU/MMT electrospun composite membranes t 1 t
¼ þ , ð11Þ
at different time intervals during batch experiments. qt k 2 q2e qe
There are three major stages that can be observed. Ini-
tially, an abrupt increase in the adsorption capacity was where k2 represents the pseudo-second-order rate con-
observed. A slow rise in dye adsorption was observed stant (gg1 min1), qe (g/g) represents the equilibrium
between 30 and 70 min, and within 100 min, the equilib- adsorption capacity, and qt (g/g) is the adsorption capac-
rium condition was achieved. After adsorptive equilib- ity at time t.
rium, the white membrane turned a dark blue color Figure 9a,b shows the pseudo-first-order and pseudo-
(inset of Figure 8b). For detailed kinetic and adsorption second-order kinetic models for the adsorption of MB on
isotherm experiments, the PU/20-MMT electrospun com- the electrospun composite sample, respectively. Table 7
posite membrane, which had the highest equilibrium shows the kinetic parameters and the correlation coeffi-
adsorption capacity of 33.19 mg/g was used. cient (R2). For the pseudo-second-order kinetic model,
F I G U R E 9 (a) Pseudo-first-
order kinetic model, (b) pseudo-
second-order kinetic model,
(c) Freundlich adsorption
isotherm, (d) Langmuir
adsorption isotherm [Color
TABLE 7 Regression coefficient and kinetic parameters for the adsorption of MB on the PU/20-MMT electrospun membrane
Model qe, experimental (mg/g) qe, calculated (mg/g) k R2

Pseudo-first-order 33.193 27.11 0.0469 0.98877
Pseudo-second-order 33.193 36.33 0.00276 0.99915
the R2 value of dye adsorption is closer to one (R2 ≈ 1), where C is the concentration, n and KF (L/mg) are the
and the calculated adsorption capacity at equilibrium Freundlich constant and empirical parameter, respectively.
(qe, cal) is close to the experimental data. So, PU/20-MMT Figure 9c,d shows the Freundlich and Langmuir iso-
electrospun composite membranes obeyed the pseudo- therms of MB adsorption on PU/20-MMT, respectively.
second-order kinetic mechanism during the MB adsorption. Table 8 shows the R2 value and isotherm parameters.
The adsorption isotherm of MB adsorption on The R2 value of the Langmuir adsorption isotherm
PU/20-MMT was studied using the Freundlich and Lang- was found to be closer to unity. As a result, the Langmuir
muir adsorption isotherms. adsorption isotherm was satisfied during MB adsorption
The Langmuir isotherm is shown below.68 on the PU/20-MMT electrospun composite membrane. It
is mainly owing to the monolayer adsorption of MB on the
Ce Ce 1 electrospun composite membrane. The maximum MB
¼ þ , ð12Þ
qe qm K L qm adsorption capacity was found to be 112.35 mg/g. The
“RL” value of adsorption was identified as 0.095, which
where qe (mg/g) represents the equilibrium adsorption indicated the reaction was favorable (0 < RL <1).
capacity, Ce (mg/L) represents the adsorbate concentra-
tion at equilibrium, qm (mg/g) denotes the maximum
adsorption capacity, and KL (L/mg) stands for the Lang- 3.8 | Thermodynamics parameters and
muir adsorption equilibrium constant. effect of pH on adsorption
The Freundlich isotherm is shown below.69
1 The MB solution's pH significantly influences the ads-
log qe ¼ log Ce þ log K F , ð13Þ orption capacity of MB on PU/20-MMT electrospun
n
composite membranes, mainly due to the shift in surface attract positively charged MB to the PU/MMT mem-
charges with the pH changes. Figure 10 shows the sche- brane surface.70
matic representation of the pH dependency on MB The dye adsorption behavior with respect to tempera-
adsorption. When the pH was increased, the MB adsorp- ture variation was investigated using thermodynamic
tion increased (Figure 11a). At lower pH, the hydroxyl parameters (Table 9). The spontaneity was analyzed
( OH) and carboxyl ( COOH) groups in Mt and o- using thermodynamic parameters such as entropy change
MWCNT are in protonated forms ( OH2+ and (ΔS), Gibbs free energy change (ΔG), and enthalpy
CO2H2+). This positively charged surface of the mem- change (ΔH). Thermodynamic parameters were calcu-
brane prevents cationic dyes from adhering to it. At a lated using Equations (14) and (15).71
basic pH, the carboxyl and hydroxyl groups form car-
boxylate ions ( COO) and oxide ions ( O), which ΔS ΔH
ln ðqe =Ce Þ ¼ , ð14Þ
R RT
TABLE 8 Adsorption isotherm parameters and R2 values ΔG ¼ ΔH TΔS, ð15Þ

Langmuir Freundlich
qm = 112.35 mg/g n = 2.46 where T temperature in Kelvin, R represents the gas con-
stant (8.314 JK1 mol1), and qe/Ce is adsorption affinity.
KL = 0.1598 L/mg KF = 19.35
ΔG, ΔS, and ΔH, are Gibb's free energy change, entropy
R2 = 0.9996 R2 = 0.9299
change, and enthalpy change, respectively.
F I G U R E 1 0 A schematic
representation of the pH
dependency on MB adsorption
on PU/MMT fibrous membrane.
[Color figure can be viewed at
F I G U R E 1 1 (a) pH
dependency on the adsorption,
(b) temperature effect on the
adsorption, (c) determination of
thermodynamics parameters
from the plot ln(qe/Ce) versus
1/T [Color figure can be viewed
at wileyonlinelibrary.com]
Figure 11b shows the dye adsorption behavior of the to the fouling, which originated from the MB foulants.
PU/20-MMT electrospun composite membrane at various The fouling mechanism during MB filtration through the
temperatures (298.15 to 338.15 K). As the temperature PU/20-MMT electrospun membrane was investigated,
increased, the adsorption capacity also increased. So, the and the regression coefficient (R2) obtained for various
adsorption process was endothermic, and Gibb's free fouling models is tabulated in Table 10. It showed that
energy indicated that the process was thermodynamically the fouling mechanism was best fitted with the standard
feasible. blocking model (R2 = 0.9999), in which MB molecules
were adsorbed on the inner walls of the electrospun
fiber.72 It further confirmed the adsorption of MB mole-
3.9 | Dynamic MB removal and cules on the electrospun composite membrane during
reusability dynamic filtration.
The reusability of the PU/20-MMT membrane was
The dynamic removal of MB from MB-contaminated studied by adsorption–desorption experiments in dynamic
water was investigated using a stirred ultrafiltration cell filtration and batch conditions. Figure 12c shows the
(Figure 12a). During filtration, clean water was obtained percentage removal efficiency (R%) of MB from the MB
from the blue-colored MB solution. A video showing the solution by the PU/20-MMT electrospun composite mem-
dynamic MB removal from MB solution using a stirred brane in six cycles in both dynamic filtration and batch
cell at 5 psi pressure is provided under supplementary conditions. The electrospun membrane showed an
video S1. A gradual decline in the flux of permeate from enhanced removal efficiency of MB in the dynamic
the MB solution was observed (Figure 12b), mainly due removal experiment compared to the batch experiment.
The membrane showed removal efficiency greater than
90% during dynamic MB removal for all cycles. In the
T A B L E 9 Thermodynamics parameters of MB adsorption on batch experiment, the removal efficiency dropped slightly
electrospun composite membrane after the third cycle to 81.05%. This minor decline in the R
% could be attributed to the fractional coverage of methy-
ΔS ΔH
(kJ/mol) (kJ/mol) ΔG (kJ/mol) lene blue on the membrane surface even after desorption.
Figure 12d shows the SEM micrograph of the PU/20-MMT
308.15 K 318.15 K 328.15 K
nanofibrous membrane after six cycles of dynamic MB
0.2423 66.8696 7.8114 10.2349 12.6584
removal from MB-contaminated water. It preserved an
FIGURE 12
(a) Photographs of the dynamic
removal of MB from MB
solution using stirred
ultrafiltration cell, (b) a graph
showing the variation of flux
during the dynamic removal of
MB, (c) removal efficiencies of
MB at dynamic and batch
experiments, (d) FE-SEM image
of PU/20-MMT after six cycles of
dynamic removal of MB [Color
T A B L E 1 0 R2 value obtained from the Hermia fouling model AUTHOR CONTRIBUTIONS

for MB flux decline Juraij Kandiyil: Conceptualization (lead); investigation
Fouling models Regression coefficient (R2)
(lead); methodology (lead); writing – original draft (lead).
Suni Vasudevan: Supervision (equal); writing – review
Cake formation 0.9692
and editing (supporting). Sujith Athiyanathil: Concep-
Intermediate blocking 0.9962 tualization (equal); supervision (lead); writing – review
Standard blocking 0.9999 and editing (equal).
Complete blocking 0.9964
ACKNOWLEDGMENT
The authors gratefully acknowledge CMC-CRS, NIT
interconnected, web-like morphology and porosity. How- Calicut for Raman analysis.
ever, a slight flattening of the fibers on the surface was
noticed, along with an increase in fiber diameter to F U N D I N G IN F O R M A T I O N
494 nm, probably due to the swelling of fibers during MB MHRD Govt. of India.
adsorption and dynamic flux. So, the electrospun compos-
ite membrane is considered a better material for dynamic CONFLICT OF INTEREST
MB removal and water filtration applications due to its The authors declare no potential conflict of interest.
high percentage removal efficiency and retention in the
porosity. DA TA AVAI LA BI LI TY S T ATE ME NT
The data that support the findings of this study are avail-
able from the corresponding author upon reasonable
4 | C ON C L U S I ON S request.
In summary, a new, highly flexible, fouling-tolerant ORCID

montmorillonite/o-MWCNT-modified polyurethane elec- Juraij Kandiyil https://orcid.org/0000-0002-3196-5759
trospun composite membrane (PU/MMT) has been Sujith Athiyanathil https://orcid.org/0000-0003-3278-
developed for the selective MB adsorption from the dye 5343
mixture. The PU/MMT nanofibrous membranes outper-
formed electrospun PU membranes in terms of mechani- RE FER EN CES
cal strength and thermal stability. The optimized [1] S. R. Yousefi, H. A. Alshamsi, O. Amiri, M. Salavati-Niasari,
PU/20-MMT electrospun composite membrane showed J. Mol. Liq. 2021, 337, 116405.
enhanced hydrophilicity and water flux. The maximum [2] S. R. Yousefi, M. Ghanbari, O. Amiri, Z. Marzhoseyni, P.
flux of electrospun PU and optimized composite samples Mehdizadeh, M. Hajizadeh-Oghaz, M. Salavati-Niasari, J. Am.
Ceram. Soc. 2021, 104, 2952.
were 935 ± 101 and 4029 ± 123 Lm2 h1, respectively.
[3] N. Mohammad, Y. Atassi, Sci. Rep., 2020, 10, 102020.
The optimized PU/20-MMT electrospun composite mem- [4] S. Swaminathan, N. M. Imayathamizhan, A. Muthumanickkam,
brane showed outstanding antifouling properties and J. Elastomers Plast. 2020, 53, 48.
selective MB adsorption. Pristine electrospun PU fol- [5] S. Haider, F. F. Binagag, A. Haider, A. Mahmood, N. Shah,
lowed an intermediate blocking fouling model, whereas W. A. Al-Masry, S. U.-D. Khan, S. M. Ramay, Desalin. Water
the optimized composite membrane corresponded to the Treat. 2014, 55, 1609.
cake formation fouling model. The optimized membrane [6] S. Radoor, J. Karayil, A. Jayakumar, J. Parameswaranpillai, S.
exhibited a high rejection of 95.45% and an FRR of Siengchin, J. Polym. Environ. 2021, 29, 3185.
[7] A. Aluigi, F. Rombaldoni, C. Tonetti, L. Jannoke, J. Hazard.
98.45%. It showed a maximum adsorption capacity (qm)
Mater. 2014, 268, 156.
of 112.35 mg/g. The pseudo-second-order kinetic model [8] A. Ahmad, S. H. Mohd-Setapar, C. S. Chuong, A. Khatoon,
and Langmuir adsorption isotherms can describe the W. A. Wani, R. Kumar, M. Rafatullah, RSC Adv. 2015, 5,
adsorption phenomena of the fibrous membrane. Positive 30801.
ΔH (66.86 kJ/mol) indicated the adsorption was endo- [9] S. R. Yousefi, A. Sobhani, H. A. Alshamsi, M. Salavati-Niasari,
thermic. Negative ΔG values (7.81 to 12.81 kJ/mol) RSC Adv. 2021, 11, 11500.
confirmed that the adsorption process was feasible. [10] L. N. Ndlovu, K. I. Malatjie, M. B. Chabalala, A. K. Mishra,
S. B. Mishra, E. N. Nxumalo, J. Appl. Polym. Sci. 2022, 139,
PU/20-MMT showed a significant removal efficiency of
52302.
MB in batch adsorption conditions and dynamic MB fil-
[11] N. Xu, J. Han, Y. Feng, C. Xiao, J. Membr. Sci. 2022, 641,
tration. Thus, the PU/MMT nanofibrous membrane is a 119935.
promising fouling-tolerant material for removing methy- [12] P. Bansal, R. Batra, R. Yadav, R. Purwar, J. Appl. Polym. Sci.
lene blue from the dye mixture and contaminated water. 2022, 139, 51582.
[13] A. Iranpoury, M. R. Mehrnia, S. H. Jafari, M. Najmi, J. Appl. [39] R. S. Bangari, A. Yadav, J. Bharadwaj, N. Sinha, J. Environ.
Polym. Sci. 2022, 139, e52733. Chem. Eng. 2022, 10, 107052.
[14] T. A. Sherazi, S. Azam, S. H. Shah, S. Hussain, S. A. R. Naqvi, [40] I. I. C. Bates, I. B. S. Carrillo, H. Germain, E. Loranger, B.
S. Li, J. Membr. Sci. 2022, 643, 120060. Chabot, J. Environ. Chem. Eng. 2021, 9, 106204.
[15] Q. Ouyang, Q. Gui, C. Liu, J. Zhang, X. Chen, Chem. Eng. J. [41] H. Yang, B. Zhu, L. Zhu, Z. Zeng, G. Wang, Z. Xiong, ACS
2023, 451, 138596. Appl. Mater. Interfaces 2021, 13, 43648.
[16] C. Gandhimathi, S. Sundarrajan, T. Matsuura, D. K. [42] M. A. Indok Nurul Hasyimah, A. W. Mohammad, Ind. Eng.
Srinivasan, H. Wei, D. Xuecheng, S. Ramakrishna, J. Appl. Chem. Res. 2014, 53, 15213.
Polym. Sci. 2021, 138, 50296. [43] M. Enfrin, J. Lee, P. Le-Clech, L. F. Dumée, J. Membr. Sci.
[17] M. Li, J. Li, M. Zhou, Y. Xian, Y. Shui, M. Wu, Y. Yao, J. Appl. 2020, 601, 117890.
Polym. Sci. 2020, 137, 48416. [44] I. P. Meneses, S. D. Novaes, R. S. Dezotti, P. V. Oliveira,
[18] I. S. Al-Husaini, W.-J. Lau, A. R. M. Yusoff, M. Z. Al-Abri, D. F. S. Petri, J. Hazard. Mater. 2022, 421, 126804.
B. A. A. Farsi, J. Environ. Chem. Eng. 2021, 9, 104728. [45] Q. Gui, Q. Ouyang, J. Zhang, S. Shi, X. Chen, ACS Appl. Mater.
[19] S. P. Sundaran, C. Reshmi, A. Sujith, New J. Chem. 2018, 42, Interfaces 2021, 13, 20796.
1958. [46] P. Shi, X. Hu, M. Duan, J. Cleaner Prod. 2021, 290, 125794.
[20] S. P. Sundaran, C. R. Reshmi, P. Sagitha, A. Sujith, J. Mater. [47] J. Xue, T. Wu, Y. Dai, Y. Xia, Chem. Rev. 2019, 119,
Sci. 2020, 55, 5892. 5298.
[21] M. A. Mahdi, S. R. Yousefi, L. S. Jasim, M. Salavati-Niasari, [48] J.-W. Lee, J. Jung, Y. H. Cho, S. K. Yadav, K. Y. Baek, H. B.
Int. J. Hydrogen Energy 2022, 47, 14319. Park, S. M. Hong, C. M. Koo, ACS Appl. Mater. Interfaces
[22] F. E. Ahmed, B. S. Lalia, R. Hashaikeh, Desalination 2015, 2014, 6, 14600.
356, 15. [49] M. Forhad Mina, S. Seema, R. Matin, M. Jellur Rahaman, R.
[23] F. Jahanmard, M. Baghban Eslaminejad, M. Amani-Tehran, F. Bijoy Sarker, M. Abdul Gafur, M. Abu Hashan Bhuiyan,
Zarei, N. Rezaei, M. Croes, S. Amin Yavari, Mater. Sci. Eng. C Polym. Degrad. Stab. 2009, 94, 183.
2020, 106, 110163. [50] S. Ramakrishna, An Introduction to Electrospinning and Nano-
[24] Z. Yin, C. Cui, H. Chen, X. Yu, W. Qian, Small 2020, 16, fibers, World Scientific, Singapore 2005.
1902301. [51] B. Kamaliya, P. N. Dave, P. M. Macwan, J. Appl. Polym. Sci.
[25] A. Rehab, N. Salahuddin, Mater. Sci. Eng., A 2005, 399, 368. 2022, 139, e52888.
[26] L. Brochard, J. Phys. Chem. C 2021, 125, 15527. [52] B. Zhang, T. Zhang, Z. Zhang, M. Xie, J. Mater. Sci. 2019, 54,
[27] K. Buruga, H. Song, J. Shang, N. Bolan, T. K. Jagannathan, 11037.
K.-H. Kim, J. Hazard. Mater. 2019, 379, 120584. [53] D. S. Tong, C. W. Wu, M. O. Adebajo, G. C. Jin, W. H. Yu,
[28] M. Makaremi, R. T. de Silva, P. Pasbakhsh, J. Phys. Chem. C S. F. Ji, C. H. Zhou, Appl. Clay Sci. 2018, 161, 256.
2015, 119, 7949. [54] S. Tunç, O. Duman, Appl. Clay Sci. 2010, 48, 414.
[29] N. Li, W. Wang, L. Zhu, W. Cui, X. Chen, B. Zhang, Z. Zhang, [55] C. Reshmi, S. P. Sundaran, T. Subija, S. Athiyanathil, Appl.
Chem. Eng. J. 2021, 421, 127857. Clay Sci. 2018, 166, 262.
[30] S. Chen, Y. Guo, D. Yuan, C. He, J. Bao, S. Ai, F. Li, [56] M. S. Islam, M. S. Rahaman, J. H. Yeum, Carbohydr. Polym.
W. Zhao, Y. Xu, C. Zhao, Compos. Sci. Technol. 2021, 214, 2015, 115, 69.
108993. [57] C. Reshmi, S. P. Sundaran, A. Juraij, S. Athiyanathil, RSC Adv.
[31] M. R. De Guzman, C. K. A. Andra, M. B. M. Y. Ang, G. V. C. 2017, 7, 2092.
Dizon, A. R. Caparanga, S.-H. Huang, K.-R. Lee, J. Membr. Sci. [58] S. P. Sundaran, C. Reshmi, P. Sagitha, O. Manaf, A. Sujith,
2021, 620, 118881. J. Environ. Manag. 2019, 240, 494.
[32] K. Selatile, S. S. Ray, V. Ojijo, R. E. Sadiku, ACS Omega 2021, [59] J. Cheng, C. Zhan, J. Wu, Z. Cui, J. Si, Q. Wang, X. Peng, L.-S.
6, 21005. Turng, ACS Omega 2020, 5, 5389.
[33] P. Peer, M. Janalikova, J. Sedlarikova, P. Pleva, P. Filip, J. [60] R. Zhao, Y. Wang, X. Li, B. Sun, C. Wang, ACS Appl. Mater.
Zelenkova, A. O. Siskova, ACS Appl. Mater. Interfaces 2021, Interfaces 2015, 7, 26649.
13, 41021. [61] S. Zhao, Y. Zhan, Q. Feng, W. Yang, H. Dong, A. Sun,
[34] S. Alberti, A. Dodero, E. Sartori, S. Vicini, M. Ferretti, M. X. Wen, Y.-H. Chiao, S. Zhang, Colloids Surf. A 2021, 626,
Castellano, ACS Appl. Polym. Mater. 2021, 3, 6157. 127069.
[35] K.-S. Kang, A. Phan, C. Olikagu, T. Lee, D. A. Loy, M. Kwon, [62] N. D. Wanasekara, L. E. Matolyak, L. T. J. Korley, ACS Appl.
H.-j. Paik, S. J. Hong, J. Bang, W. O. Parker Jr., M. Sciarra, Mater. Interfaces 2015, 7, 22970.
A. R. de Angelis, J. Pyun, Angew. Chem. Int. Ed. 2021, 60, [63] M. A. Fuad, M. Zaini, M. Jamaludin, R. Ridzuan, Polym. Test.
22900. 1994, 13, 15.
[36] J. Chen, Z.-X. Low, S. Feng, Z. Zhong, W. Xing, H. Wang, ACS [64] M. A. Fuad, M. Zaini, M. Jamaludin, Z. M. Ishak, A. M. Omar,
Appl. Mater. Interfaces 2021, 13, 60763. J. Appl. Polym. Sci. 1994, 51, 1875.
[37] A. Rajeswari, S. Vismaiya, A. Pius, Chem. Eng. J. 2017, [65] J. Lindholm, J.-F. Boily, M. Holmboe, ACS Earth Space Chem.
313, 928. 2019, 3, 2490.
[38] K. Juraij, C. Chingakham, O. Manaf, P. Sagitha, V. Suni, V. [66] K. Mukherjee, A. Kedia, K. Jagajjanani Rao, S. Dhir, S. Paria,
Sajith, A. Sujith, J. Appl. Polym. Sci. 2022, 139, 52117. RSC Adv. 2015, 5, 30654.
[67] E. Ajenifuja, S. O. Alayande, O. A. Aromolaran, J. A. Ajao, SU PP O R TI N G I N F O RMA TI O N

E. O. Dare, T. A. M. Msagati, E. O. B. Ajayi, J. Water Process Additional supporting information can be found online
Eng. 2017, 19, 253. in the Supporting Information section at the end of this
[68] V. Vaid, R. Jindal, J. Appl. Polym. Sci. 2022, 139, 52218.
article.
[69] Y. Q. Liu, C. G. Song, G. Ding, J. Yang, J. R. Wu, G. Wu, M. Z.
Zhang, C. Song, L. P. Guo, J. C. Qin, Adv. Mater. Interfaces
2022, 9, 2102480. How to cite this article: J. Kandiyil,
[70] W. Peng, H. Li, Y. Liu, S. Song, J. Mol. Liq. 2016, 221, 82.
S. Vasudevan, S. Athiyanathil, J. Appl. Polym. Sci.
[71] Z. Mahmoodi, A. R. Abhari, R. S. Lalehloo, Z. H. Bakr, G. A.
2023, 140(10), e53464. https://doi.org/10.1002/app.
Ali, ChemistrySelect 2022, 7, e202103619.
[72] S. Kim, M. Yu, Y. Yoon, ACS Appl. Mater. Interfaces 2020, 12, 53464
16557.

J of Applied Polymer Sci - 2022 - Kandiyil - Efficient Selective Methylene Blue Adsorption by Polyurethane

Uploaded by

Copyright:

Available Formats

J of Applied Polymer Sci - 2022 - Kandiyil - Efficient Selective Methylene Blue Adsorption by Polyurethane

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

J of Applied Polymer Sci - 2022 - Kandiyil - Efficient Selective Methylene Blue Adsorption by Polyurethane

Uploaded by

Copyright:

Available Formats

Received: 31 May 2022 Revised: 11 November 2022 Accepted: 14 November 2022

Efficient selective methylene blue adsorption

Juraij Kandiyil 1 | Suni Vasudevan 2 | Sujith Athiyanathil 1

1 | INTRODUCTION Among them, methylene blue (MB) dye is hazardous to

J Appl Polym Sci. 2023;140:e53464. wileyonlinelibrary.com/journal/app © 2022 Wiley Periodicals LLC. 1 of 18

2 | MATERIALS AND METHODS

Lubrizol Advanced Materials, Massachusetts, USA, pro-

2.4 | Characterization where d represents the average fiber diameter and ε

2.5 | Pore size and porosity Cf Cp

T A B L E 2 Various fouling models

Standard blocking 1.5 J¼ J0

Elongation at Adsorption capacity Water flux

TABLE 5 TGA data of samples

Sample Tonset ( C) Tmax1 ( C) Tmax2 ( C) Residue (%) Filler content (%)

TABLE 6 R2 and k values for different Hermia fouling models

Model qe, experimental (mg/g) qe, calculated (mg/g) k R2

TABLE 8 Adsorption isotherm parameters and R2 values ΔG ¼ ΔH TΔS, ð15Þ

T A B L E 1 0 R2 value obtained from the Hermia fouling model AUTHOR CONTRIBUTIONS

In summary, a new, highly flexible, fouling-tolerant ORCID

[67] E. Ajenifuja, S. O. Alayande, O. A. Aromolaran, J. A. Ajao, SU PP O R TI N G I N F O RMA TI O N

You might also like