Branched polyamide/poly (acrylonitrile) blend membranes: Synthesis,characterisation and pervaporation applications

Abstract

Herein, a novel branched polyamide (BPA-1and BPA-2) was synthesised by direct polycondensation using branched polyethyleneimine and two different dicarboxylic acids (adipic acid and Hexafluoroisopropylidene bis benzoic acid) as monomers. The newly developed BPA was blended with poly (acrylonitrile) (PAN) separately at ratio of 3:1 (BPA/PAN) to form a branched BPA/PAN membrane via the solution-casting method. The prepared membrane was characterised using FTIR, X-ray, SEM and TGA and so that used for the dehydration of alcohols by pervaporation process. The experimental results revealed that the permeation flux of the BPA/PAN membrane increased with increasing feed ethanol concentration and feed temperature, but an increase in the number of carbon atoms inalcohol results in an increase in the permeation rate. The best separation performance could be achieved for the BPA-2/PAN membrane at 60°C for 80% ethanol in the feed with a total flux of 925 g m⁻² h⁻¹ and a separation selectivity of 1696.

Keywords

Branched polyamides with polyethyleneimine units blend membranes aromatic fluorinated dicarboxylic acid pervaporation alcohol–water mixture

Introduction

Dehydration of organic solvents containing a small amount of water, which forms azeotrope is very important in the area of organic synthesis [1]. Because these solvents are used in a large scale in chemical industries such as paints, adhesives, cosmetics and coatings as well as in pharmaceutical laboratories [2, 3]. The separation of the above mixtures needs the use of processes such as azeotropic distillation, molecular sieve adsorption, extractive, solvent extraction and vacuum distillation [4]. These methods are high energy consuming processes. Membrane separation has gained attraction in many industries such as treatment of drinking water, water softening, textile, pervaporation (PV) and gas separation [5-7]. PV is an excellent membrane technology of liquid mixture separation that is more energy saving, environmentally safe and low operation cost process [8]. The separation mechanism in this technology is based on the difference in sorption and diffusion characteristics of the permeating components. The separation of mixtures by means of PV methods can be classified into three main fields, (i) separation of organic-organic solvent mixtures, (ii) dehydration of aqueous–organic mixtures and (iii) removal of trace volatile organic compounds from aqueous solution [9, 10]. According to the solution–diffusion model, the permselectivity of a membrane is related to solubility and diffusivity. Therefore, the efficiency of the PV technique depends basically on the inherent properties of the polymers utilised in the fabrication of membrane [11]. Since polyamides have attractive physical, chemical and mechanical properties, as well as commercial availability, they can be used in many membrane separation technologies such as gas separation and PV [12, 13]. As a consequence of this, the development of novel polyamide PV membranes with appropriate separation efficiency is important. Nevertheless, polyamides exhibit low permeation rate due to their low water solubility and low free volume [14]. Several researchers have focused their attention on improving the polyamide membrane separation performance, such as polymer blending, introduction of bulky pendent groups or flexible links into the polymer backbone, plasma grafting and chemical grafting [15]. Blending is an ideal and easy method to improve organo-solubility and film-forming ability of polyamide membranes [16]. On the other hand, poly (acrylonitrile) (PAN) is one of the most semi-crystalline versatile thermoplastic polymers which has been widely used for the preparation of reverse osmosis, microfiltration, nanofiltration and PV membranes because of its good thermal stability and commercial availability [17, 18]. Since PAN is relatively hydrophilic polymer and has a polar nitrile (CN) group attached to a repeat unit of main chain, it is understandable that PAN demonstrates a high permselectivity to water [19]. To our best knowledge, there is no report on the dehydration of alcohol–water mixtures using BPA/Polyacrylonitrile blend membranes. It was anticipated that blending polyacrylonitrile with branched polyamides increase the flux and separation factor. Since blend membranes containing the hydrophilic areas of polyacrylonitrile which increase the solubility selectivity of water in the blend membrane by providing more sorption sites of water. With this view in mind, in this study, two new branched polyamides were synthesised by phosphorylation polycondensation using polyethyleneimine (PEI) and two different dicarboxylic acids as monomers. Then, these polyamides blended with PAN separately and BPA/PAN blend membranes were prepared via solution-casting method and characterised. These membranes were used for the dehydration of alcohols by PV process. The effects of ethanol concentrations in the feed solution, feed solution temperature and various alcohols on PV performance were investigated.

Experimental

Materials

Branched poly (ethylenimine) (BPEI, Mw = 25,000 g mol⁻¹) and Hexafluoroisopropylidene bis benzoic acid, adipic acid (AA) and PAN (Mw = 52,000 g mol⁻¹) all were purchased from Sigma-Aldrich (USA) and were purified by sublimation under reduced pressure. N-methyl-2-pyrrolidone (NMP) and N, N-dimethylacetamide (DMAc) (Fluka) were purified by distillation under reduced pressure over calcium hydride before use. Triphenyl phosphite (TPP) received from Sigma-Aldrich, was purified by distillation under reduced pressure. Anhydrous calcium chloride (CaCl₂) (Fluka) was dried at 180°C for 8 h under reduced pressure. Pyridine (PY), methanol, ethanol, n-propanol and t-butanol were obtained from Merck (Germany). Deionised water was used throughout the research work. The chemical structures of monomers are shown in Figure 1.

Synthesis of branched polyamides

Synthesis of branched polyamides was performed by Yamazaki's phosphorylation method [20].

Synthesis of BPA-1 from branched PEI and AA

A mixture of branched PEI (1.98 g, 2.5 mmol), AA (3.82 g, 2.5 mmol), calcium chloride (CaCl₂, 1.4 g), triphenyl phosphite (TPP, 1.8 ml), pyridine (PY, 2.1 ml) and N-methyl-2-pyrrolidinone (NMP, 7 ml) was heated in nitrogen with stirring at 120°C for 4 h. As the polycondensation proceeded, the reaction mixture became viscous gradually, the resultant solution of the polymer was poured slowly into 300 ml of methanol while constantly stirring producing a fiber-like polymer precipitate. The precipitate was collected, washed thoroughly with methanol and hot water successively for three times. The precipitate was collected on a filter and dried at 110°C under vacuum oven overnight, to obtain the desired polyamide (BPA-1) as a yellow powder in excellent yield.

BPA-2 also was synthesised from branched PEI and hexafluoroisopropylidene bis benzoic acid with a similar procedure.

Characteristics and solubility behaviour of both fabricated branched polyamides are shown in Tables 1 and 2 respectively.

Table 1

Reaction yield, inherent viscosities and molecular weight data of branched polyamides.

Polymer	Diacid	Yield (%)	η_inh ^a (dl g⁻¹)	M_n ^b (Da)	PDI ^b	P_n ^c
BPA-1	AA	98	0.75	26.4	1.25	61
BPA-2	Hexafluoroisopropylidene bis benzoic acid	99	0.89	28.2	1.32	64

Table 2

Solubility behaviour of branched polyamides.

Polymer	NMP	DMAc	DMSO	DMF	THF	CHCl₃
BPA-1	+ +	+ +	+ +	+	−	−
BPA-2	+ +	+ +	+ +	+	−	−

Notes: (++) soluble at room temperature, (+) soluble on heating, (−) insoluble even on heating

Polymerisation was carried out with 2.5 mmol each of BPA and diacid a Number average molecular weight (Mn) and polydispersity index (PDI = Mw/Mn) were determined using a polymer/N, N-dimethyl formamide (DMF) solution at a flow rate of 1 ml min⁻¹ at 70°C and calibrated with polystyrene standards. B Degree of polymerisation (Pn) was determined from Mn. c Inherent viscosity was measured for 0.5% (w v⁻¹) solutions of weight branched polyamides in DMAc at 30°C.

Preparation of BPA-1/PAN and BPA-2/PAN blend membranes

Dense BPA/PAN membranes were prepared by solution-casting technique. Both of polymer blends were prepared by dissolving BPA (12 wt-%) and PAN (4 wt-%) in DMAC with 3/1 weight ratio. The mixture was stirred for 9 h and then kept in an ultrasonic bath for about 3 h. Solutions were then filtered to remove insoluble impurities and left overnight to obtain a homogeneous and bubble free solutions. The resulting solutions were poured on to a clean glass plate and dried at room temperature for about 24 h. After that, membranes continued to dry in a vacuum oven at 50°C for 2 days, to remove residual solvent, after complete drying, the BPA/PAN membranes were peeled off carefully from the glass plate. The obtained membranes were designated as BPA-1/PAN and BPA-2/PAN membranes. The thickness of the resulting membranes by a digital micrometre (Mitutoyo, model ID-C112E) were 35 and 40 μm for BPA-1/PAN and BPA-2/PAN membranes, respectively.

Characterisation

Fourier transform infrared (FTIR) spectroscopy measurement

The FTIR spectra were scanned in the range of 400–4000 cm⁻¹ using a Nicolet-740, Perkin-Elmer 283 B FTIR spectrophotometer by KBr pellet method.

X-ray diffraction (XRD)

A Siemens D 5000 powder X-ray diffractometer was utilised to estimate the solid-state morphology of BPA-1, BPA-2, PAN, BPA-1/PAN membrane and BPA-2/PAN membrane in powdered form. X-rays of 1.54 Å wavelength were generated by a Cuk source.

Scanning electron microscopy

SEM micrographs of the BPA-1, BPA-2, BPA-1/PAN membrane and BPA-2/PAN membrane were obtained under high resolution (Ma: 300X, 6kv) utilising a JOEL MODEL JSM 840 A, scanning electron microscopy (SEM).

Thermal gravimetric analysis (TGA)

Thermal stability of the BPA-1, BPA-2, PAN, BPA-1/PAN membrane and BPA-2/PAN membrane and were examined (Seiko 220TG/DTA analyzer) in the temperature range of 25–600°C at a heating rate of 10 °C min⁻¹ with continuous flushing under nitrogen gas at 250 mL min⁻¹.

GEL permeation chromatographic

GEL permeation chromatographic (GPC) analysis was performed on a Lab Alliance RI2000 instrument (one column, MIXED-D) connected with one refractive index detector from Schambeck SFD Gmbh. All GPC analyses were performed using a polymer/N, N-dimethyl formamide solution at a flow rate of 1 mL min⁻¹ at 70°C and calibrated with polystyrene standards.

Viscosity

The inherent viscosities of branched polyamides were determined on an Ubbelohde suspended level viscometer for 0.5% (w v⁻¹) solutions of branched polyamides in DMAc at 30°C.

Solubility

Solubility of branched polyamides was determined at 3 wt-% concentration in various solvents at room temperature or on heating. The polymer samples were tested for solubility initially at room temperature for 24 h and observations were made. The polymer samples that were not soluble at room temperature were heated at 70°C for 30 min and observations were made.

PV measurement

The PV system used in this study is exhibited in Figure 2 [21]. The PV cell was assembled from two cylindrical half cells made of stainless steel. The membranes were supported on a perforated stainless steel plate placed at the junction of two cells. In PV, the feed solution is in direct contact with the membrane. The effective membrane area for PV was 25.6 cm². For all measurements, the downstream pressure was maintained at 6 mmHg by using a vacuum pump. For each experiment, the operating time was 4 h to ensure that a steady state was obtained. The permeate vapour containing alcohol and water was condensed in a glass condenser suspended inside a cryogenic trap kept at −40°C. The permeation rate was characterised by measuring the weight of the permeate. The composition of the feed solution and the permeate were measured by gas chromatography (Varian 3300 gas chromatograph). The separation performances of these membranes were investigated for a range of feed composition from 5–80% water in feed. All the PV runs were repeated thrice for each of the feed compositions studied with 0.2–0.5% variations in flux data and the average results are reported.

The permeation flux was measured from the weight of permeate collected as follow: (1) where P, W, A and t indicate the permeation flux (g m⁻² h⁻¹), weight of collected permeate (g), effective area of the membrane (m²) and operation time (h), respectively. The separation factor is defined by: (2) where Y_A and Y_B are the weight fraction of the component in feed and X_A and X_B are the weight fraction of those in the permeate (A being the more permeative species), respectively. Component A stands for water and B stands for alcohol in this experiment.

Result and discussion

Fourier transform infrared (FTIR) studies

The FT-IR spectra of the BPA-1, BPA-2, neat PAN, BPA-1/PAN membrane and BPA-2/PAN membrane are shown in Figure 3. For the BPA-1, as can be observed from Figure 3a, the spectrum demonstrates the characteristic absorption band at around 3300–3500 cm⁻¹ correspond to stretching vibration of –NH, the peak at 1640 cm⁻¹ is due to C = O stretching and the aliphatic segment is identified in the spectra by –CH₂– stretching vibration at 2940 cm⁻¹. Similarly, the FT-IR spectrum of BPA-2 (Figure 3(b)) shows characteristic absorption band at 1651 cm⁻¹ due to C = O stretching, the absorption bands at 3300–3500 and 1245 cm⁻¹ is attributed to -NH stretching and C–F stretching, respectively. Apart from these, peaks appeared at around 3128 and 3045 cm⁻¹ are due to aromatic C–H stretching and at 2908 cm⁻¹, aliphatic C–H stretching is also seen. The spectra of neat PAN is exhibited in Figure 3(c). Typical characteristic absorption peaks at around 2240 and 1629 cm⁻¹, represents the presence of –C≡Nand C = C groups in polyacrylonitrile, respectively. In addition, the bands at 1452 and 1360 cm⁻¹ are deformation vibrations in the –CH₂ and C–H groups, respectively. Notably, the FT-IR spectra of BPA-1/PAN membrane (Figure 3(d)) indicates the characteristic absorption peaks at 3300–3500, 1628 and 2241 cm⁻¹ that correspond to the –N–H, -C = O and –C≡N groups, respectively. In this spectrum, some changes in the spectra of branched polyamides after addition of PAN have been appeared. The enhancement of the –C = O stretching peak at 1641 cm⁻¹ and the decrease in the –NH band intensity can be observed. Furthermore, there is a decrease of the –C≡N band intensity in comparison with pure PAN spectra, that indicates a molecular interaction between BPA-1 and PAN. Figure 3(e) shows the FT-IR spectra of BPA-2/PAN membrane. The characteristic peaks for the BPA-2/PAN membrane are observed at 3300–3500 cm⁻¹ (stretching band of –NH), 1590 cm⁻¹ (stretching band of –C = O) and 2242 cm⁻¹ (stretching band of –C≡N); which are consistent with the inherent characteristic of BPA-2 and PAN. In addition to these, the peak at 2850, 2790 cm⁻¹ are attributed to aromatic C–H stretching of BPA-2. All of these results indicate that the branched polyamides were successfully synthesised and blended with polyacrylonitrile.

XRD analysis

In order to get an idea of the morphological state of the synthesised membranes, X-ray diffraction analysis was carried out. The X-ray diffraction patterns of the BPA-1, BPA-2, neat PAN, BPA-1/PAN membrane and BPA-2/PAN membrane are shown in Figure 4. As can be observed from Figure 4(a,b), in general, the broad spectra of BPA-1 and BPA-2 confirming amorphous nature of these branched polyamides. This is due to the branched poly (ethylenimine) units in polymer backbone that reduce the extent of close packing and hence loss of crystallinity. For neat PAN (Figure 4(c)) the peak at 2θ=17° is correspond to the diffraction of the hexagonal lattice of PAN. From the spectra of BPA-1/PAN (Figure 4(d)) and BPA-2/PAN (Figure 4(e)) it can be noted that the XRD patterns of BPA/PAN membranes are weaker than those of PAN, indicating that BPA/PAN membranes have lower crystallinity than PAN and since branched polyamides are amorphous, thus the addition of the PAN to branched polyamides may increase the crystallinity and decrease the amorphous nature of the branched polyamides. This can improve the selective permeation of the membranes.

SEM analysis

The scanning electron micrographs of the external surface of BPA-1, BPA-2, BPA-1/PAN and BPA-2/PAN are demonstrated in Figure 5(a–d), respectively. Figure 5(a,b) shows the surface of BPA-1 and BPA-2, respectively. Different diacid monomers with different reactivity led to the different reaction behaviour, resulting in the formation of branched polyamides with different types of morphology. In general, the SEM images of the branched polyamides showed a rougher surface morphology. The SEM images of the BPA-1/PAN and BPA-2/PAN membranes are exhibited in Figure 5(c,d), respectively. As shown in the SEM images, the morphology of both BPA/PAN membranes was significantly different from the branched polyamides. The surface of the membranes was dense and flat and there were no cracks or pinholes, indicating the complete miscibilities of the PAN with BPAs matrix. Hence, it can be concluded that the prepared membranes are suitable for PV separation process since the process needs a dense structure membrane.

Thermal gravimetric analysis

Thermal stability of the fabricated membranes is an important factor in PV performance at high temperature. Furthermore, membranes with stability over a wide temperature range are preferred for PV application. The thermal stability of the BPA-1, BPA-2, PAN, BPA-1/PAN membrane and BPA-2/PAN membrane were analysed utilising thermo gravimetric analysis under the nitrogen atmosphere and the resulting thermograms are shown in Figure 6. Figure 6(a,b) shows that BPA-1 and BPA-2 begin to undergo a weight loss starting at 298 and 301°C respectively, followed by the final decomposition at 423°C for BPA-1 and 441°C for BPA-2. For the neat PAN, as can be observed from Figure 6(c), maximum weight loss occurs at the temperature range of 310–450°C. In addition, for the BPA-1/PAN membrane (Figure 6(d)) and BPA-2/PAN membrane (Figure 6(e)) the curves show that the weight loss starts at 306 and 462°C respectively, followed by a final decomposition at 482°C for BPA-1/PAN membrane and 449°C for BPA-2/PAN membrane. From the above results, it is evident that the blending of two branched polyamides with polyacrylonitrile had more thermal stability when compared with the branched polyamides alone. This TGA study reveals that synthesised membranes have satisfactory thermal stability and the BPA-1/PAN membrane and BPA-2/PAN membrane can be effectively used in PV experiments, even at high operating temperatures.

Effect of ethanol concentration in feed solution on the PV performances of BPA/PAN membranes

The PV performance of BPA-1/PAN and BPA-2/PAN membranes were investigated for varying ethanol concentrations from 5 to 80 wt-% at different temperatures and results are shown in Figure 7(a,b). It can be seen from Figure 7(a,b) that the permeation rates were found to increase for all feed ethanol concentrations from 5 to 80 wt-% at different temperatures. For example, as the feed ethanol concentration increases from 5 to 80 wt-% at 60°C for BPA-1/PAN membrane (see Figure 7(a)), the permeation rate increases gradually from 163 to 854 g m⁻² h⁻¹. Similar results were observed for the BPA-2/PAN membrane (see Figure 7(b)), as the feed ethanol concentration increases from 5 to 80 wt-% at 60 °C the permeation rate increases from 289 to 925 g m⁻² h⁻¹. Besides, as can be observed from Figure 8, the separation factor goes to a higher level from 958 to 1295 for BPA-1/PAN membrane and from 1104 to 1696 in the case of BPA-2/PAN membrane with an increase in the feed ethanol concentration from 5 to 80 wt-% at 60°C. These results might be due to the plasticising effect of ethanol. When the ethanol concentration in the feed is higher, the amorphous regions of the membranes is more swollen. Therefore, the polymer chains in the swollen regions become more flexible and the transport through the membrane becomes easier resulting in high permeation rate. Furthermore, these membranes show a high permselectricity to water rather than ethanol because both the BPA/PAN membranes containing the hydrophilic areas of PAN which increase the solubility selectivity of water in the blend membrane by providing more sorption sites of water. Hence, these results indicate that blending PAN with branched polyamides increase the flux and separation factor. In addition, as can be observed from Figures 7 and 8, the highest separation selectivity (1696) was observed for the BPA-2/PAN membrane with a flux of 925 g m⁻²h⁻¹ at 60 °C for 80% ethanol in the feed. The higher flux and separation factor is attributed to the increase of the molecular volume of the substituted group in the polymer backbone, these phenomena might be due to the fact that the BPA-2/PAN membrane containing a bulky aromatic segment (hexafluoroisopropylidene bis benzen) in the polymer backbone which this modification in polymer design decrease inter-molecular hydrogen bonds (CO … ..NH), weaken inter-chain interactions and consequently the free volume of the BPA-2/PAN membrane is higher than BPA-1/PAN membrane. Thus, the permeation rate and separation factor of BPA-2/PAN membrane is high.

Effect of feed solution temperature on the PV performances of BPA/PAN membranes

Feed temperature is an important factor impacting PV process. The effect of feed solution temperature on the PV performance for ethanol–water mixtures at 25, 40 and 60°C through BPA-1/PAN and BPA-2/PAN membranes at different feed concentrations from 5 to 80 wt-% are shown in Figure 7(a,b). It can be observed from Figure 7(a,b) that the permeation rates were found to increase from 25 to 60°C for all feed ethanol concentrations. For example, in the case of BPA-1/PAN membrane for 80 wt-% aqueous alcohol solution as the temperature increases from 25 to 60°C, the permeation rate increases from 725 to 854 g m⁻² h⁻¹. Similar results were observed for the BPA-2/PAN membrane with 80 wt-% ethanol in feed mixture, an increases of temperature from 25 to 60°C results in an increase in permeation rate from 769 to 925 g m⁻²h⁻¹. Higher feed temperature, caused by an increase in membrane chain mobility and swelling of the membrane matrix, resulting in the higher permeation rates. In other words, the plasticising effects of the permeant and the interaction between permeant and polymer can be used to further explain the above phenomena. Furthermore, the partial pressure of ethanol in the vapour phase increases with increasing the feed solution temperature, resulting in the membrane more swelling. This facilitates the transport of ethanol molecules along with water, thereby increasing the permeation rate. This is also evident from Figure 7(a,b) that the highest permeation rate (892 g m⁻²h⁻¹) was observed for the BPA-2/PAN membrane at 60°C for 80% ethanol in the feed. The higher flux indicates that the PV separation through the BPA-2/PAN membrane is more temperature-sensitive than that through the BPA-1/PAN membrane.

Effect of different feed types on PV performances of BPA/PAN membranes

The separation efficiency of BPA-1/PAN and BPA-2/PAN membranes in dehydration of four alcohol–water mixtures with 80 wt-% alcohol in feed mixture at 60°C are shown in Table 3. The data show that for BPA-1/PAN membrane with 80 wt-% alcohol in feed mixture an increase in the number of carbon atoms in alcohol results in an increase in the permeation rate from 742 g m⁻² h⁻¹ for methanol to 956 g m⁻²h⁻¹ for t-butanol but gives a decrease in the separation factor from 1369 for methanol to 1067 for t-butanol. Similar results were observed for the BPA-2/PAN membrane with 80 wt-% alcohol in feed mixture, an increase in the number of carbon atoms in alcohol results in an increase in the permeation rate from 812 g m⁻²h⁻¹ for methanol to 991 g m⁻²h⁻¹ for t-butanol but gives a decrease in the separation factor from 1745 for methanol to 1394 for t-butanol. This is also evident from Figure 7(a,b) that the highest permeation rate (991 g m⁻²h⁻¹) was observed for the BPA-2/PAN membrane at 60°C for 80% t-butanol in the feed. These results indicate that the PV dehydration of aqueous t-butanol mixtures proceeds more easily than that of aqueous methanol, ethanol and n-propanol mixtures because of the larger molecular size of t-butanol and relatively weak coupling phenomenon (less polar than methanol, ethanol and n-propanol) with water molecules and hydrophilic membranes. The higher permeation fluxes for t-butanol mixtures compared to other three mixtures can be attributed to less polarity and less coupling in the t-butanol mixtures. Again for the same reasons BPA/PAN membranes are more efficient in dehydration of n-propanol/water and ethanol/water mixtures compared to methanol/water mixture. A comparison of the pervaporation performance of polyamide membranes tested under comparable conditions and reported in the literature is given in Table 4.

Table 3

PV of BPA/PAN membranes for alcohol–water mixtures at 60°C.

Membrane	Alcohol solution (80 wt-%)	Degree of swelling (%)	Selectivity	Permeation rate (g m⁻²h⁻¹)
BPA-1/PAN	Methanol	15.3	1369	742
	Ethanol	21.5	1295	854
	n-propanol	29.4	1184	927
	t-butanol	35.9	1067	956
BPA-2/PAN	Methanol	27.5	1745	812
	Ethanol	32.2	1696	925
	n-propanol	40.1	1548	974
	t-butanol	45.7	1394	991

Table 4

Comparison of water permeability data of the present membranes with literature data for binary mixtures.

Membrane	Ethanol concentration in the feed (wt-%)	Feed temperature (°C)	Total flux (g m⁻²h⁻¹)	Water selectivity	Ref.
Nylon 4	90	25	350	4.5	[22]
Modified nylon 4	90	25	420	13.5	[23]
Poly(amide-sulfonamide)s	90	25	7–34	1984	[24]
Polyamide/clay hybrid nanocomposite	90	25	380	26	[25]
BPA-1/PAN	80	60	854	1295	Current study
BPA-2/PAN	80	60	925	1696	current study

Conclusion

In the present study, novel branched polyamides (BPA-1and BPA-2) were synthesised by direct polycondensation using branched PEI and two different dicarboxylic acids, then, the newly developed branched polyamides were blended with PAN separately at ratio of 3:1 (BPA/PAN) to form the BPA/PAN membranes via the solution-casting method. The experimental results revealed that the permeation flux of both BPA/PAN membranes increased with increasing feed ethanol concentration and feed temperature, but an increase in the number of carbon atoms in alcohol results in an increase in the permeation rate and a decrease in the separation factor. Furthermore, both BPA/PAN membranes showed a high permselectricity to water rather than ethanol because these membranes containing the hydrophilic areas of polyacrylonitrile which increase the solubility selectivity of water in the blend membrane by providing more sorption sites of water. Therefore, these results indicate that blending polyacrylonitrile with branched polyamides rise the flux and separation factor. The best separation performance could be achieved for the BPA-2/PAN membrane at 60°C for 80% ethanol in the feed with a total flux of 925 g m⁻² h and a separation selectivity of 1696. In addition, the results indicate that the PV dehydration of aqueous t-butanol mixtures proceeds more easily than that of aqueous methanol, ethanol and n-propanol mixtures due to the larger molecular size of t-butanol. Compared with the BPA-1/PAN membrane the BPA-2/PAN membrane containing aromatic fluorinated dicarboxylic acid showed a much higher permeation rate than that of BPA-1/PAN membrane (925 versus 854) at 60°C for 80 wt-% ethanol. The higher flux is attributed to the increase of the molecular volume of the bulky aromatic segment (hexafluoroisopropylidene bis benzen) in the polymer backbone. High dehydration performance of alcohol–water mixtures have exhibited the utility of these membranes for use in solvent dehydration application with higher efficiency.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Featherstone

, Cox

Separation of aqueous-organic mixtures by pervaporation. Br Chem Eng Process Technol. 1971;16:817–819.

Shao

, Huang

RYM.

Polymeric membrane pervaporation. J Memb Sci. 2007;287:162–179. doi: 10.1016/j.memsci.2006.10.043

Uragami

, Matsugi

, Miyata

Effect of membrane structures on separation characteristics for an aqueous ethanol solution through hydrophilic organic-inorganic hybrid membranes. J Network Polym Japan. 2001;22:157–162.

Yeom

, Lee

Preparative permeations of homologous series of alcohol aqueous mixtures through a hydrophilic membrane. J Appl Polym Sci. 2001;79:703–713. doi: 10.1002/1097-4628(20010124)79:4<703::AID-APP150>3.0.CO;2-O

Mansourpanah

, Madaeni

, Rahimpour

, The effect of non contact heating (Microwave Irradiation) and contact heating (Annealing process) on morphology and performance of polyethersulfone nanofiltration membranes. Appl Surf Sci. 2001;225:8395–8402.

Mansourpanah

, Madaeni

, Adeli

, Surface modification and preparation of nanofiltration membrane from polyethersulfone/polyimide blend-use of new material (PEG-triazine). J Appl Polym Sci. 2009;112:2888–2895. doi: 10.1002/app.29821

Salimi

, Mansourpanah

Construction of a full hands-free liquid droplet flowmeter for low-permeable gas separation membranes. J Memb Sci. 2017;537:202–208. doi: 10.1016/j.memsci.2017.05.021

Yamasaki

, Mizoguchi

Preparation of PVA membranes containing β-cyclodextrin oligomer (PVA/CD membrane) and their pervaporation characteristics for ethanol/water mixtures. J Appl Polym Sci. 1994;51:2057–2062. doi: 10.1002/app.1994.070511211

Burshe

, Sawant

, Joshi

, Sorption and permeation of binary water-alcohol systems through PVA membranes crosslinked with multifunctional crosslinking agents. Sep Purif Technol. 1997;12:145–156. doi: 10.1016/S1383-5866(97)00042-7

10.

Park

, Nam

, Lee

YM.

Pervaporation separation of an aqueous organic mixture through a poly (acrylonitrile-co-vinyl phosphonic acid) membrane. J Appl Polym Sci. 1999;74:83–89. doi: 10.1002/(SICI)1097-4628(19991003)74:1<83::AID-APP9>3.0.CO;2-6

11.

Gupta

, Pradhan

, Adhikari

Separation of phenol from aqueous solution by pervaporation using HTPB-based polyurethaneurea membrane. J Memb Sci. 2003;217:43–53. doi: 10.1016/S0376-7388(03)00069-3

12.

Albo

, Wang

, Tsuru

Application of interfacially polymerized polyamide composite membranes toisopropanol dehydration: effect of membrane pretreatment and temperature. J Memb Sci. 2014;453:384–393. doi: 10.1016/j.memsci.2013.11.030

13.

Kujawski

, Waczyński

, Lasota

Pervaporation properties of dense polyamide-6 membranes in separation of water-ethanol mixtures. Sep Sci Technol. 1996;31:953–963. doi: 10.1080/01496399608002497

14.

Lai

, Chu

, Huang

, Separation of water–alcohol mixtures by pervaporation through asymmetric nylon 4 membrane. J Appl Polym Sci. 1994;53:999–1009. doi: 10.1002/app.1994.070530803

15.

EL-Gendi

, Deratani

, Ahmed

, Development of polyamide-6/chitosan membranes for desalination. Egyptian J Petroleum. 2014;23:169–173. doi: 10.1016/j.ejpe.2014.05.003

16.

Albo

, Wang

, Tsuru

Application of interfacially polymerized polyamide composite membranes to isopropanol dehydration: effect of membrane pretreatment and temperature. J Memb Sci. 2014;453:384–393. doi: 10.1016/j.memsci.2013.11.030

17.

Aminabhavi

, Naik

HG.

Pervaporation separation of water/dimethylformamide mixtures using poly (vinyl alcohol) -g-polyacrylamide copolymeric membranes. J Appl Polym Sci. 2002;83:273–282. doi: 10.1002/app.2242

18.

Hoshi

, Saitoh

, Yoshioka

, Pervaporation separation of 1, 1, 2-trichloroethane–water mixture through crosslinked acrylate copolymer composite membranes. J Appl Polym Sci. 1999;74:983–994. doi: 10.1002/(SICI)1097-4628(19991024)74:4<983::AID-APP25>3.0.CO;2-V

19.

Ping

, Nguyen

, Chen

, Pervaporation of water–ethanol mixtures with polyacrylate-grafted polyethylene acid (PE-g-AA) membranes: physico-chemical analysis of the transport mechanis. J Memb Sci. 2002;195:23–34. doi: 10.1016/S0376-7388(01)00484-7

20.

Yamazaki

, Matsumoto

, Higashi

FJ.

Studies on reactions of the N-phosphonium salts of pyridines. XIV. Wholly aromatic polyamides by the direct polycondensation reaction by using phosphites in the presence of metal salts. J Polym Sci Part A Polym Chem. 1975;13:1373–1380. doi: 10.1002/pol.1975.170130609

21.

Lee

, Teng

, Lee

, Dehydration of ethanol/water mixtures by pervaporation with composite membranes of polyacrylic acid and plasma-treated polycarbonate. J Memb Sci. 2000;164:13–23. doi: 10.1016/S0376-7388(99)00097-6

22.

El-Gendi

, Abdallah

Pervaporation of methanol/methylacetate mixture using polyamide-6 membranes. Desalin Water Treat. 2013;51:7807–7814. doi: 10.1080/19443994.2012.749040

23.

Shieh

J-J

, Huang

RYM.

Chitosan/N-methylol nylon 6 blend membranes for the pervaporation separation of ethanol–water mixtures. J Memb Sci. 1998;148:243–255. doi: 10.1016/S0376-7388(98)00165-3

24.

Meier-Haack

, Lenk

, Berwald

, Influence of thermal treatment on the pervaporation separation properties of polyamide-6 membranes. Sep Purif Technol. 2000;19:199–207. doi: 10.1016/S1383-5866(00)00053-8

25.

Kopeć

, Meller

, Kujawski

, Polyamide-6 based pervaporation membranes for organic–organic separation. Sep Purif Technol. 2013;110:63–73. doi: 10.1016/j.seppur.2013.03.007