The states of sulfate groups affect the mechanical and separation properties of carboxymethyl cellulose/chitosan complex membranes

Abstract

Solution-processable polyelectrolyte complexes (PECs) were fabricated from sulfonated carboxymethyl cellulose (SCMC) and chitosan (CS) by two methods. Two states of sulfate groups (SO₃⁻) in PECs, i.e., complexed and free sulfate groups, were distinguished and shown to play important roles in determining mechanical properties and separation performances of PEC membranes (PECMs). PECMs containing different states of SO₃⁻ were subjected to pervaporation dehydration, and their hydrophilic properties and swelling behaviors were thoroughly discussed. It was found that the tensile strength of PECMs could be effectively enhanced by increasing the ratio of complexed SO₃⁻ groups, while the hydration and flexibility were modulated by free SO₃⁻ groups. In the dehydration of water–ethanol mixtures, both the flux and selectivity of PECMs were simultaneously improved through the incorporation of SO₃⁻ groups, whereby the improved pervaporation performance was mainly attributed to the free SO₃⁻ groups.

---

Introduction

Chitosan (CS) is β-(1-4)-2-amino-2-deoxy-D-glucose, which is produced from chitin as the second most abundant polysaccharide in nature.¹ Currently, CS has been explored extensively because of its biocompatibility and non-toxicity.^2–5 Carboxymethyl cellulose (CMC), which has been commercially available since the 1920s, is one of the biggest industrial cellulose derivatives in terms of volume production.⁶ Bearing several hydroxyl and carboxylate groups in its repeating units, CMC is a anionic polyelectrolyte which is readily modular for biomedical applications,⁷ flocculation,⁸ papermaking processes,⁹ and membranes separations.¹⁰

Polyelectrolyte complexes (PECs) are usually prepared by mixing aqueous solutions of oppositely charged polyelectrolytes, which are produced driven by electrostatic interaction.^11,12 PECs have been proven as promising materials for sensing,¹³ flocculation,¹⁴ actuation,¹⁵ and molecular separation.¹⁶ PEC membranes (PECMs) containing polysaccharide such as CMC or CS show high pervaporation dehydration performance,^17–19 owing to their ionic crosslinking structures and hydrophilic properties. Pervaporation could realize the molecular separation between water and organics, and has been shown energy-saving, easy manipulation and low cost compared with the conventional distillation or adsorption as well.^20,21 In this regard, membranes with high flux, good selectivity and sufficient mechanical properties are sought after.²²

However PEC solids are generally insoluble in common solvents and infusible at high temperature, which impede the processing and applications of bulk PEC materials.²³ In previous work, we proposed a facile strategy, named “acid-protection”,¹⁷ to resolve this issue. In detail, CMC containing carboxylate (COONa) groups and polysaccharides such as CS and cationic cellulose were used as polyanion and polycations, respectively. PECs were formed by mixing these oppositely charged polyelectrolytes in acidic conditions, where a part of carboxyl groups in CMC could be protected. As such, the prepared PECs, containing protonated carboxylic acid groups, render themselves dispersible in aqueous alkaline solutions. These homogenous PECMs feature improved permeability that is crucial for dehydration of alcohols. However, membranes based on weak carboxylic acid groups lack sufficient electrostatic complexation force and hydrophilicity for stable performance, particularly when the feed solution is acidic. Hence we prepared PECMs based on electrostatic interactions between sulfonate (SO₃⁻) and ammonium groups. Compared with COONa groups bearing weak charge, the incorporation of sulfonate (SO₃⁻) groups manifests improved pervaporation performance in the dehydration of alcohols. Despite the beneficial properties fostered by SO₃⁻ groups; however, the detailed effects of SO₃⁻ states on the properties of homogenous PECMs were poorly understood so far. For example, a further understanding of the impact of SO₃⁻ groups on the mechanical properties and pervaporation performance of homogenous PECMs will shield more light on the rational design of high performance PEC membranes.

In this work, two strategies were exploited to prepare solution-processable PECs containing different states of SO₃⁻ groups from CMC and CS. The hydration ability of different states of SO₃⁻ groups was explored by thermal gravity analysis (TGA) and differential scanning calorimeter (DSC). The effects of SO₃⁻ states on the mechanical properties and pervaporation dehydration of homogenous PECMs were discussed on the basis of their structural features and water binding capacity. It was found that the complexed SO₃⁻ groups contributed to the PECMs strength, while the free ones enhanced its hydration ability and flexibility. An improved flux and selectivity of PECMs were achieved by free SO₃⁻ groups in comparison to complexed ones.

Experimental

Materials

Chitosan (CS) (M_n = 200 000 g mol⁻¹, deacetylation degree = 90%) was obtained from Yuhuan Chemical Company (Zhejiang, China). Carboxymethyl cellulose (CMC) was obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China), and its intrinsic viscosity was 625.1 mL g⁻¹ in 0.1 M sodium hydroxide (NaOH) aqueous solutions at 30 °C. Sulfur trioxide pyridine complex (SO₃/pyridine) and sulfur trioxide trimethylamine complex ((CH₃)₃N·SO₃) were received from Aldrich, and were applied without further purification. All organic solvents (analytical grade) such as p-toluenesulphonic acid (p-TsOH), ethanol, acetone, N,N-dimethylacetamide (DMAc), hydrochloric acid (HCl), sodium hydroxide (NaOH), and sodium carbonate (Na₂CO₃) were purchased from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China, and were used as received. It should be mentioned that p-TsOH was dried at 120 °C under vacuum for 3 h to complete dry before usage. Polysulfone ultra-filtration membranes were kindly provided by Development Centre of Water Treatment Technology, Hangzhou, China, which were used as membrane supports. De-ionized water, with a resistivity of 18 MΩ cm, was used in all experiments.

Preparation of PECs containing different states of SO₃⁻ groups and their membranes

As shown in Fig. 1, solution-processable PECs containing different states of SO₃⁻ groups were fabricated with two methods from CMC and CS. PECs containing complexed SO₃⁻ groups (SPECs) were prepared by using sulfated carboxymethyl cellulose (SCMC) and CS with acid-protection method,²⁴ whereas PECs containing both complexed and free SO₃⁻ groups (PECSs) were prepared with complexation-sulfation method, i.e., the sulfation of the formed CMC/CS complexes.²⁵

Fig. 1 Schematic diagram for fabricating solution-processable SPEC (a) and PECS (a′), uniformly dispersed PEC (b), PECM (c).

In detail, the solution-processable SPECs were prepared from SCMC and CS in the acid-protection method which was previously reported by us (Fig. 1a).²⁴ CMC was sulfated in DMAc solvent using SO₃/pyridine complex as sulfating agent. By tuning the feed ratio of SO₃/pyridine complex to CMC, SCMCs with different compositions were prepared, and SO₃⁻ molar fractions in SCMC were marked as X. The SO₃⁻ molar fractions (Y) in SPECs were controlled by varying the contents of SO₃⁻ groups in SCMC. Another solution-processable PECSs shown in Fig. 1a′ were fabricated by using (CH₃)₃N·SO₃ to sulfate the NH₂ groups of the formed CS/CMC complexes.²⁵ SO₃⁻ molar fractions (Z) in PESCs were controlled by adjusting the ratio of (CH₃)₃N·SO₃ to CMC/CS complexes.

The obtained SPECs and PECSs were dispersed in NaOH aqueous solution (0.10 mol L⁻¹) to form a homogeneous dispersion (2 wt%) with pH ∼ 7.0, because the protonation of COOH group under base condition yielded COO⁻ groups which were more hydrophilic (Fig. 1b). The free-standing PECM with large size (12 × 10 cm) was easily prepared (Fig. 1c), which was highly transparent. PECMs for pervaporation dehydration were prepared by casting homogeneous PEC dispersions on polysulfone ultra-filtration membranes, followed by drying at 40 °C for 8 h and then 60 °C for 2 h to remove residual solvent. The molar fractions of SO₃⁻ groups with respect to SPEC were varied as 0.10, 0.25, 0.35, and the resulting membranes were designed as SPECM-0.10, SPECM-0.25, SPECM-0.35. The molar fractions of SO₃⁻ groups relative to PECS were varied as 0.07, 0.13, 0.20, and the resulting membranes were designed as PECSM-0.07, PECSM-0.13, PECSM-0.20, respectively.

Characterization

Fourier transform infrared spectroscopy (FT-IR) for PECs was obtained using a FT-IR spectrometer (Bruker Vector 22, Germany) by dispersing PEC solids in KBr, and made as pellets. The ratio of S atom to O atom of SCMC and the ratio of S atom to N atom of SPECs and PECSs were analyzed by X-ray photoelectron spectra (XPS, PerkinElmer PHI 5300 ESCA), with Mg/Al Dual Anode Hel/Hell ultraviolet source (400 W, 15 kV, 1253.6 eV).

X, the molar fractions of SO₃⁻ in SCMC, was determined by the following equation:

where [S] and [O] were denoted the contents of S and O atoms in the SCMC.

Y, the molar fractions of SO₃⁻ groups in SPECs was calculated according to the following equation:

where [S] and [N] were the contents of S and N atoms in SPECs.

Z, the molar fractions of SO₃⁻ groups in PECSs was determined by the following equations:

where [CMC]/[CS] in PECS-0 represented the molar ratio of the monomer unit of CMC to that of CS.

Surface morphologies of PEC membranes were observed by a field emission scanning electron microscope (FESEM, Hitachi S4800, Japan). Samples were coated with Pt nanoparticles for 60 s before FESEM examination. The distribution S element in PEC membranes was determined through an energy dispersive X-ray spectrometer (EDX, SIRION-100, America). Water contact angles on the PECMs surface (at 25 °C) were measured using a contact angle meter (OCA 20, Data physics Instruments GmbH, Germany) through the sessile drop method, and the contact angle of PECMs at 80 s was used to characterize their surface hydrophilicity. The states of water absorbed in PECMs were determined by a differential scanning calorimetry (DSC, Perkin-Elemer Pyris 1, Waltham, MA) under a nitrogen atmosphere. Samples were cooled from 25 °C to −100 °C and then heated to 25 °C at a scanning rate of 10 °C min⁻¹. Thermal gravity analysis (TGA) of a membrane was obtained with a Perkin-Elemer Pyris 1 TGA at a heating rate of 20 °C min⁻¹. The membranes measured by TGA were dried at 50 °C for 12 h and then equilibrated at 30% relative humidity (RH) for ca. 6 h.

Mechanical and conductivity properties of PECMs

A 2 wt% homogenous PEC dispersion was casted onto a clean glass slide, and dried at 50 °C for 12 h. Free-standing PECMs (10 × 60 mm) with tailored thickness (30 ± 2 μm) were used for tensile tests using a universal test machine (SANS CMT4204, Shenzhen, China) at a stretching rate of 1 mm min⁻¹. Tensile strength of the PECMs was averaged by five repeats at 30% RH and 25 °C. Membrane strips with different sizes (8 cm × 8 cm) were prepared for testing their surface conductivity by using a high resistance meter (ZC 36, Shanghai REX Instrument Factory, China) at 30% RH and 25 °C. At least three samples were tested to determine the average values of surface conductivity.

Pervaporation performance of PECMs

PECMs were subjected to the pervaporation dehydration on the same apparatus as previously reported.²⁶ The feed temperature was tested by an electric control thermometer with an accuracy of 0.3 °C, and the downstream pressure was kept at ca. 180 Pa measured by a piezometer. The feed composition was stabilized at 10 wt% water–ethanol by circulation. The permeate condensed by liquid nitrogen was tested by a gas chromatograph (GC1690A, Hangzhou Ke Xiao Chemical Instrument Co., Ltd., China). Permeation flux (J) and water contents in permeate (wt%) or separation factor (α), and the pervaporation separation index (PSI) were used to evaluate the pervaporation performance of PECMs, they were defined as follows:where, Δg was the permeate weight collected in liquid nitrogen traps during the operation time Δt; A was the membrane area (18.09 cm²). P_w, P_E, F_w, and F_E represented the mass fractions of water (W) and ethanol (E) at the feed (F) and permeate (P) sides.

Equilibrium swelling degree (ESD) of PECMs

For the swelling tests, a thoroughly dried free-standing PECM was immersed into a 10 wt% water–ethanol mixture, and equilibrated at 50 °C for 24 h. Then the weight of this PECM was measured after removing their surface water with a tissue paper quickly. ESD was calculated in following equation:

ESD = (M_∞ − M₀)/M₀ (5)

where M₀ and M_∞ represented the weight of this membrane before and after being swollen, respectively.

Sorption experiments

A PECM was weighed after dried in a vacuum oven (50 °C, 12 h), and was then immersed in a 10 wt% water–ethanol solution at 50 °C for 24 h to equilibrate. Then the solution on the surface of this PECM was quickly wiped off with a tissue paper. The absorbed liquid in a PECM was desorbed by a vacuum pump and condensed in a liquid nitrogen trap, and its composition was analyzed by gas chromatography. All the data were averaged over three repeat experiments. The sorption selectivity (α_s) was calculated, and the diffusion selectivity (α_d) was also obtained according to the solution-diffusion principle:where α and α_s were the separation factor and the sorption selectivity at 50 °C, respectively. M and F were the weight fractions of water (W) and ethanol (E) in a membrane and feed solution. The contents of water and ethanol in the swollen PECMs, which were equilibrated in a 10 wt% water–ethanol mixtures, were determined by multiplying ESD with the M_W and M_E, respectively.

Results and discussion

Characterization of SCMCs and SPECs and PECSs

The chemical composition of SCMC, SPECs and PECSs were determined by XPS (Tables 1 and 2). SCMCs with different compositions were prepared. The molar fraction of SO₃⁻ groups in SCMC, i.e., X, increases with increasing the ratio of SO₃/pyridine complex to CMC in the feed. The molar fraction of SO₃⁻ groups (Y) in SPECs increases with increasing the contents of SO₃⁻ groups in SCMC. Hence, a tuneable amount of complexed SO₃⁻ groups have been successfully incorporated into SPECs. Z, denoting as the molar fraction of SO₃⁻ groups in PECSs, was determined by eqn (3) and shown in Table 2. PECSs containing modulated amount of SO₃⁻ groups were synthesized. To further characterize the states of SO₃⁻ groups in SPECs and PECSs, FTIR was carried out (Fig. 2). The peak at 810 cm⁻¹ for the C–O–S symmetrical stretching vibration appears in SCMC-0.58 and SPEC-0.35 but is absent in PEC (CS/CMC), indicating that SO₃⁻ groups were indeed introduced into SPECs.²⁷ The peak at 626 cm⁻¹ (attributed to free SO₃⁻ groups) appears only in SCMC-0.58 but not in SPEC-0.35 indicating that the SO₃⁻ groups in SPECs were all in complexed state.²⁸ In addition, the absorption peaks of SO₃⁻ groups appear at 754 cm⁻¹, 701 cm⁻¹ and 626 cm⁻¹ in PECS-0.20 while being absent in PEC(CS/CMC). The peak at 626 cm⁻¹ is associated with free SO₃⁻ groups, as shown in the sulfated CS, whereas the absorption peaks at 754 and 701 cm⁻¹ are attributed to the complexed SO₃⁻ groups.²⁹ The appearance of complexed SO₃⁻ groups in PECS-0.20 was due to the transfer of ion-pairs upon the incorporation of SO₃⁻ groups, which is ascribed to the stronger binding of SO₃⁻ with NH₃⁺ than that of COO⁻ groups.³⁰ The partial transfer of ion-pairs for SO₃⁻ groups in PECSs is irrevisible.^30,31 The ratio of free and complexed SO₃⁻ groups in PECSs is calculated via dividing the peak areas at 626 cm⁻¹ (free SO₃⁻ groups) by those of 701 cm⁻¹ (complexed SO₃⁻ groups). Combined with the XPS results (Table 2), the contents of free and complexed SO₃⁻ groups are obtained, as shown in Table 3. Both the contents of the free and the complexed SO₃⁻ groups exhibit an increase with increasing the ratio of (CH₃)₃N·SO₃ to CS/CMC in the feed. This result also indicates that free SO₃⁻ groups have been incorporated into PECSMs and their contents were tunable.

Table 1 Composition of SCMC and SPECs determined by XPS

Sample	S (at%)	N (at%)	O (at%)	S : N	S : O	X	Y
a The ratio of SO3/pyridine complex to CMC in the feed were 1 : 1, 1 : 2 and 1 : 3, respectively.b SPEC-0.10, SPEC-0.25, SPEC-0.35 are made from SCMC-0.12, SCMC-0.37, SCMC-0.58 respectively.
SCMC-0.12^a	0.96	—	56.14	—	0.0171	0.12	—
SCMC-0.37^a	2.38	—	57.48	—	0.0414	0.37	—
SCMC-0.58^a	3.92	—	58.77	—	0.0667	0.58	—
SPEC-0.10^b	0.74	1.72	—	0.4302	—	—	0.10
SPEC-0.25^b	1.51	1.95	—	0.7743	—	—	0.25
SPEC-0.35^b	2.39	2.74	—	0.8722	—	—	0.35

---

Table 2 Composition of PECSs determined by XPS

Sample	S (at%)	N (at%)	O (at%)	Z
a PECS-0 refers to CS/CMC complex which containing no SO3^− groups.b The ratio of (CH3)3N·SO3 to CS/CMC in the feed were 1 : 1, 1 : 2 and 1 : 3, respectively.
PECS-0^a	—	3.55	38.05	0
PECS-0.07^b	0.46	3.53	38.13	0.07
PECS-0.13^b	0.79	2.57	39.84	0.13
PECS-0.20^b	0.97	1.98	40.37	0.20

---

Fig. 2 FT-IR spectra for SCMC-0.58, SPEC-0.35, PEC (CS/CMC), Sulfated CS and PECS-0.20.

Table 3 Amounts of free and complexed SO₃⁻ groups in PECSs

Sample	Free SO3^− groups (mol%)	Complexed SO3^− groups (mol%)
PECS-0	0	0
PECS-0.07	4	3
PECS-0.13	8	5
PECS-0.20	14	6

---

The surface morphologies and sulfur distribution of SPECM-0.35 and PECSM-0.20 were studied by SEM and EDX (Fig. 3). Both membranes exhibit no noticeable pores indicating that the two membranes were dense and had no defects. The sulfur element was seen uniformly dispersed in SPECM-0.35 and PECSM-0.20, which indicates that SO₃⁻ groups were successfully introduced into SPECMs and PECSMs, and the SO₃⁻ groups were dispersed evenly.

Fig. 3 FESEM (×10.0k) surface morphologies of SPECM-0.35 (a), PECSM-0.20 (b). EDX elemental mapping images of sulfur in SPECM-0.35 (c) and PECSM-0.20 (d).

Composition of water in the SPECMs and PECSMs used for tensile testing

The water contents in SPECMs and PECSMs were determined by TGA (Fig. 4). A major weight loss (about 2.4–12.2 wt%) from 30 °C to 250 °C before decomposition was observed, which can be attributed to the evaporation of absorbed water in these membranes. The contents of absorbed water in both SPECMs and PECSMs increased with increasing the molar fraction of SO₃⁻ groups in PECMs. The specific contents of water in SPECMs and PECSMs based on the TGA results are given in Table 4. The water contents of all PECMs under the condition for tensile tests (30% RH, 25 °C) are less than 12.5 wt%.

Fig. 4 TGA curves for (a) SPECMs and (b) PECSMs. Samples were dried under vacuum at 50 °C for 24 h, then equilibrated at 30% RH and 25 °C for ca. 6 h.

Table 4 Water contents of SPECMs and PECSMs determined by TGA

Sample	Bound water (wt%)	Sample	Bound water (wt%)
SPECM-0	2.39%	PECSM-0	2.41%
SPECM-0.10	5.54%	PECSM-0.07	6.42%
SPECM-0.25	7.78%	PECSM-0.13	9.23%
SPECM-0.35	9.99%	PECSM-0.20	12.22%

---

To further determine the state of water in SPECMs and PECSMs, DSC for these membranes was implemented (Fig. 5). The peak of ice-melting at ca. 0 °C is absent on DSC curves of the water absorbed in SPECM-0.35 and PECSM-0.20, which indicates that no free water molecules existed in these membranes.³²

Fig. 5 DSC curves of water absorbed in PECSM-0.20 and SPECM-0.35 at 30% RH and 25 °C.

Mechanical properties of PECMs

Fig. 6 provides the effects of the contents and states of SO₃⁻ groups on the mechanical properties of PECMs. The tensile strengths and elongations at break of PECMs simultaneously improved with increasing the contents of complexed SO₃⁻ groups. For instance, the tensile strength and elongation at break for SPECM-0.35 are 87.5 MPa and 3.5%, which are 2.91 and 2.69 times as high as those for the pristine SPECM-0, respectively. It is noteworthy that the tensile strength of SPECM-0.35 outperforms that of the hybrid PECM containing carbon nanotubes (65.4 MPa).³³ As such, the mechanical properties of the PECMs are effectively improved by introducing the complexed SO₃⁻ groups, which is likely ascribed to stronger interaction of the SO₃⁻ groups with ammonium groups.³⁴ It has been documented that the difference in free energy of association of an ammonium group with a SO₃⁻ group and a COO⁻ group is as large as 14.9 kJ mol⁻¹ for the case of PECMs.³⁰ Moreover, the tensile strengths for SPECMs containing only complexed SO₃⁻ groups was increased to a higher percentage compared to those of PECSMs containing complexed and free SO₃⁻ groups. And the tensile strengths for SPECMs are higher than those for PECSMs as maintaining the same contents of SO₃⁻ groups (Fig. 6a). The bound water molecules existed in SPECMs and PECSMs are supposed to serve as plasticizers that enhance lubricity effects among the polyelectrolyte chains and augments the free volume,^35,36 and thus the elongations at break for SPECMs and PECSMs are improved. The higher water contents for PECSMs than those for SPECMs are observed and shown in Table 4. The higher hydration ability for PECSMs in comparison with SPECMs is due to the more affinity with water molecules for free SO₃⁻ groups than those for complexed ones (Fig. 7).³⁷ Herein, the tensile strengths of PECMs were effectively enhanced by complexed SO₃⁻ groups, while the hydration ability and elongations at break of PECMs were more effectively promoted by free SO₃⁻ groups compared to complexed SO₃⁻ groups.

Fig. 6 Effects of amounts of SO₃⁻ groups on tensile strengths (a), and elongations at break (b) of PECSMs and SPECMs.

Fig. 7 Schematic diagram of hydration ability of CMC/CS, SPECM, PECSM.

Surface conductivity of PECMs

Fig. 8 shows the effects of the states and contents of SO₃⁻ groups on surface conductivities of PECSMs and SPECMs. The surface conductivities of PECSMs and SPECMs simultaneously increases with increasing the contents of SO₃⁻ groups. For example, the surface conductivity of PECSM-0.20 (12.83 × 10⁻⁸ S m⁻¹) containing 20 mol% SO₃⁻ groups is 2.39 times high as that of PECSM-0 (5.36 × 10⁻⁸ S m⁻¹), while that of the SPECM-0.35 (7.08 × 10⁻⁸ S m⁻¹) containing 35 mol% SO₃⁻ groups is 1.32 times higher. In terms of surface conductivities, PECSMs possessing free SO₃⁻ groups exhibit a more visible increase compared to SPECMs.³⁸

Fig. 8 Effects of amounts of SO₃⁻ groups on surface conductivity of PECSMs and SPECMs.

Separation performance of PECMs

Fig. 9 presents the pervaporation performance of SPECMs and PECSMs containing complexed and free SO₃⁻ groups in dehydration of a 10 wt% water–ethanol mixture at 50 °C. Both of flux and separation factor for PECSMs and SPECMs increase with increasing SO₃⁻ groups contents. The flux of PECSM-0.20 incorporated 20 mol% SO₃⁻ groups is 880 g m⁻² h⁻¹ (Fig. 9a), which is 2.12 times as high as that of PECSM-0 (415 g m⁻² h⁻¹). Meanwhile, the flux of SPECM-0.35 incorporated 35 mol% SO₃⁻ groups is 645 g m⁻² h⁻¹, 1.55 times as high as that of PECSM-0. Moreover, the perm-selectively of PECSM-0.20 showed in Fig. 9b maintains at 1721 (1.58 times of PECSM-0), while that of SPECM-0.35 is 1491 (1.37 times of SPECM-0). This anti-trade-off phenomenon for PECMs performance is caused by the incorporated SO₃⁻ groups which improve both the charge density and hydrophilic property of PECMs.²⁶ Furthermore, the pervaporation dehydration performance for PECSMs containing free SO₃⁻ groups is more effectively improved compared to that of SPECMs incorporated only complexed SO₃⁻ groups. This result suggests that it is an effective strategy to improve PECMs pervaporation performance by introducing free SO₃⁻ groups to PECMs.

Fig. 9 Effects of amounts of SO₃⁻ groups on flux (a) and separation factor (b) for SPECMs and PECSMs in dehydration of 10 wt% water–ethanol mixtures at 50 °C.

Now that the role played by complexed and free SO₃⁻ groups is established, we set out to explore the correlation between the states of SO₃⁻ groups and the hydrophilic properties and swelling behavior of PECMs.³⁹ Water contact angles of SPECMs and PECSMs (Fig. 10) are smaller than those for PECM-0 and decrease with increasing the amount of SO₃⁻ groups. The promotion of hydrophilicity of PECSMs is more considerable than that of SPECMs, which is attributed to the higher hydration ability for free SO₃⁻ groups in PECSMs than that for complexed ones in SPECMs (Table 4). So the improved hydrophilicity of PECSMs with free SO₃⁻ groups results in a high flux. Besides, the sorption selectivity of both SPECMs and PECSMs (Fig. 11a) is much larger than the diffusion selectivity, which indicates that the sorption of feed component governs the PECMs separation process. Moreover, PECSMs incorporated with free SO₃⁻ groups show higher sorption selectivity than that for SPECMs incorporated only complexed SO₃⁻ groups (Fig. 11a), which renders a larger separation factor for PECSMs as shown in Fig. 9b. The swollen water experiences a rapid increase while the swollen ethanol maintains almost at the same level with increasing the contents of SO₃⁻ groups (Fig. 11b). This observation confirms that the incorporated SO₃⁻ groups in PECMs are conducive to absorb water but repel ethanol during pervaporation dehydration process.

Fig. 10 Effects of amounts of SO₃⁻ groups on hydrophilic properties of SPECMs and PECSMs.

Fig. 11 Effects of amounts of SO₃⁻ groups on the sorption selectivity (α_s) and diffusion selectivity (α_d) (a), and the absorption amounts of water and ethanol (b) in SPECMs and PECSMs in 10 wt% water–ethanol mixtures at 50 °C.

Table 5 gives the comparison of SPECM-0.35 and PECSM-0.20 in this study with other chitosan-based membranes reported in literature for the pervaporation dehydration for water–ethanol mixtures recently. The PECMs incorporated SO₃⁻ groups show higher performance compared with conventional poly (vinyl alcohol)⁴⁰ and chitosan membranes,⁴¹ as well as the nano-filled chitosan membranes.^42,43 This is attributed to the high hydrophilicity and ionic cross-linking structures of PECMs.⁴⁴ Furthermore, both the flux and selectivity for PECMs in this study are higher than those for other PECMs prepared with blending or layer-by-layer methods,^19,45 which is probably because of the improved hydrophilicity arisen from SO₃⁻ groups. The pervaporation separation index (PSI) for PECSM is 15.15 × 10⁵, which is much higher than the reported membranes in literature and that for SPECMs (9.62 × 10⁵) containing complexed SO₃⁻ groups. The exceptional separation performance of PECSMs is attributed to the higher hydrophilic property of free SO₃⁻ groups.

Table 5 Comparison of pervaporation dehydration performance of chitosan-based membranes for water–ethanol mixtures (10 wt%) in this study and the previous literatures

Membrane	Feed temperature (°C)	Flux (g m^−2 h^−1)	Water in permeate (wt%)	Separation factor (α)	PSI (×10^5)	Reference
a N,N′-methylene bisacrylamide crosslinked.b Chitosan cross-linked with glutaraldehyde (GA).c Chitosan–carbon nanotubes (CNT) functionalized with poly(styrene sulfonic acid) complex membranes.d Chitosan (CS) incorporated heteropolyacid, H14[NaP5W30O110] (HPA).e Chitosan (CS)–poly(acrylic acid) (PAA) blend membranes.f Self-assembled PEC membranes constructed by chitosan (CS) and poly(acrylic acid) (PAA) with layer-by-layer method.
PVA^a	40	410	98.2	491	2.01	40
GA/CS^b	50	450	94.5	155	0.69	41
CNT/CS^c	70	360	99.61	2270	8.17	42
HPA/CS^d	50	250	99.65	2562	6.41	43
PAA/CS^e	80	105	99.64	2491	2.62	45
PAA/CS^f	22	1050	91.05	92	0.97	19
SPECM-0.35	50	645	99.40	1491	9.62	This study
PECSM-0.20	50	880	99.48	1721	15.15	This study

---

Conclusions

Solution-processable PECs containing CMC and CS were prepared, and their homogenous membranes (PECMs) were successfully incorporated with different states of SO₃⁻ (i.e., complexed and free SO₃⁻ groups) by two methods. Tensile strength of PECMs was improved by complexed SO₃⁻, whereas the hydration and flexibility were largely determined by the free ones. The surface conductivity of PECMs was effectively increased owing to free SO₃⁻ groups. Both the flux and selectivity of PECMs were improved by free SO₃⁻ groups, which were attributed to the enhanced hydrophilicity and sorption selectivity of free SO₃⁻ groups. In dehydrating 10 wt% water–ethanol mixtures at 50 °C, the flux and selectivity of PECSM-0.20 are 880 g m⁻² h⁻¹ and 1721, respectively, which is much higher compared with conventional polyelectrolyte membranes reported recently in literature recently. This work provides an effective strategy to optimize the pervaporation performances and mechanical properties of polysaccharide-based membranes by engineering the different states of SO₃⁻ groups.

Acknowledgements

This research was financially supported by the program Key Laboratory of Novel Adsorption and Separation Materials and Application Technology of Zhejiang Province (No. XFFL2015).

Notes and references

1. R. A. A. Muzzarelli, Polym. Sci., 2012, 10, 153.
2. A. Franconetti, L. Contreras-Bernal, R. Prado-Gotor and F. Cabrera-Escribano, RSC Adv., 2015, 5, 74274.
3. X. Zhang, Y. Zhang, G. Ma, D. Yang and J. Nie, RSC Adv., 2015, 5, 79679.
4. P. Pereira, S. S. Pedrosa, A. Correia, C. F. Lima, M. P. Olmedo, Á. González-Fernández, M. Vilanova and F. M. Gama, Toxicol. in Vitro, 2015, 29, 638.
5. L. Chen, J. Hu, J. Ran, X. Shen and H. Tong, RSC Adv., 2015, 5, 56410.
6. T. Heinze and K. Pfeiffer, Macromol. Mater. Eng., 1999, 49, 37.
7. M. Boruah, P. Gogoi, A. K. Manhar, M. Khannam, M. Mandal and S. K. Dolui, RSC Adv., 2014, 4, 43865.
8. A. Hebeish, A. Higazy, A. El-Shafei and S. Sharaf, Carbohydr. Polym., 2010, 79, 60.
9. E. J. Siqueira, M. B. Salon, M. N. Belgacem and E. Mauret, J. Appl. Polym. Sci., 2015, 132, 42144.
10. S. B. Kuila and S. K. Ray, Sep. Purif. Technol., 2014, 123, 45.
11. Q. Zhao, P. F. Zhang, M. Antonietti and J. Y. Yuan, J. Am. Chem. Soc., 2012, 134, 11852.
12. Q. Zhao, Q. F. An, Z. W. Sun, J. W. Qian, K. R. Lee, C. J. Gao and J. Y. Lai, J. Phys. Chem. B, 2010, 114, 8100.
13. Q. Zhao, M. J. Yin, A. P. Zhang, S. Prescher, M. Antonietti and J. Yuan, J. Am. Chem. Soc., 2013, 135, 5549.
14. T. Swift, L. Swanson, A. Bretherick and S. Rimmer, Environ. Sci.: Water Res. Technol., 2015, 1, 332.
15. Q. Zhao, J. W. C. Dunlop, X. L. Qiu, F. H. Huang, Z. B. Zhang, J. Heyda, J. Dzubiella, M. Antonietti and J. Y. Yuan, Nat. Commun., 2014, 5, 4293.
16. Q. Zhao, Y. L. Ji, J. K. Wu, L. L. Shao, Q. F. An and C. J. Gao, RSC Adv., 2014, 4, 52808.
17. Q. Zhao, J. W. Qian, Q. F. An, C. J. Gao, Z. L. Gui and H. T. Jin, J. Membr. Sci., 2009, 333, 68.
18. T. Liu, Q. F. An, Q. Zhao, K. R. Lee, B. K. Zhu, J. W. Qian and C. J. Gao, J. Membr. Sci., 2013, 429, 181.
19. Y. Zhang, J. W. Rhim and X. S. Feng, J. Membr. Sci., 2013, 444, 22.
20. M. L. Gimenes, L. Liu and X. S. Feng, J. Membr. Sci., 2007, 295, 71.
21. P. Shao and R. Y. M. Huang, J. Membr. Sci., 2007, 287, 162.
22. B. Bolto, M. Hoang and Z. L. Xie, Chem. Eng. Process., 2011, 50, 227.
23. Z. J. Sui, J. A. Jaber and J. B. Schlenoff, Macromolecules, 2006, 39, 8145.
24. X. S. Wang, Q. F. An, F. Y. Zhao, Q. Zhao, K. R. Lee, J. W. Qian and C. J. Gao, Cellulose, 2014, 21, 3597.
25. X. S. Wang, Q. F. An, T. Liu, Q. Zhao, W. S. Hung, K. R. Lee and C. J. Gao, J. Membr. Sci., 2014, 452, 73.
26. X. S. Wang, Q. F. An, Q. Zhao, K. R. Lee, J. W. Qian and C. J. Gao, J. Membr. Sci., 2013, 435, 71.
27. H. Baumann and V. Faust, Carbohydr. Res., 2001, 331, 43.
28. G. Vikhoreva, G. Bannikova, P. Stolbushkina, A. Panov, N. Drozd, V. Makarov, V. Varlamov and L. Galbraikh, Carbohydr. Polym., 2005, 62, 327.
29. Y. Jiang and A. J. Epstein, J. Am. Chem. Soc., 1990, 112, 2800.
30. S. T. Dubas and J. B. Schlenoff, Langmuir, 2001, 17, 7725.
31. E. Tjipto, J. F. Quinn and F. Caruso, Langmuir, 2005, 21, 8785.
32. M. A. Hickner, C. H. Fujimoto and C. J. Cornelius, Polymer, 2006, 47, 4238.
33. Q. Zhao, J. W. Qian, M. H. Zhu and Q. F. An, J. Mater. Chem., 2009, 19, 8732.
34. S. A. Sukhishvili, E. Kharlampieva and V. Izumrudov, Macromolecules, 2006, 39, 8873.
35. H. H. Hariri, A. M. Lehaf and J. B. Schlenoff, Macromolecules, 2012, 45, 9364.
36. T. Liu, Q. F. An, Q. Zhao, J. K. Wu, Y. H. Song, B. K. Zhu and C. J. Gao, Sci. Rep., 2015, 5, 7782.
37. J. B. Schlenoff, A. H. Rmaile and C. B. Bucur, J. Am. Chem. Soc., 2008, 130, 13589.
38. P. C. Rieke and N. E. Vanderborgh, J. Membr. Sci., 1987, 32, 313.
39. D. P. Suhas, A. V. Raghu, H. M. Jeong and T. M. Aminabhavi, RSC Adv., 2013, 3, 17120.
40. J. J. Li, L. Zhang, J. Gu, Y. P. Sun and X. S. Ji, RSC Adv., 2015, 5, 19859.
41. J. Ma, M. H. Zhang, H. Wu, X. Yin, J. Chen and Z. Y. Jiang, J. Membr. Sci., 2010, 348, 150.
42. Y. L. Liu, W. H. Chen and Y. H. Chang, Carbohydr. Polym., 2009, 76, 232.
43. V. T. Magalad, G. S. Gokavi, M. N. Nadagouda and T. M. Aminabhavi, J. Phys. Chem. C, 2011, 115, 14731.
44. A. F. Thünemann, M. Müller, H. Dautzenberg, J. F. Joanny and H. Löwen, Adv. Polym. Sci., 2004, 166, 113.
45. S. Nam and Y. M. Lee, J. Membr. Sci., 1997, 135, 161.

---