Poly(2,6-dimethyl-1,4-phenylene oxide)-b-poly ... - ACS Publications

Jun 22, 2015 - ... Tsung-Han Tsai , E. Bryan Coughlin , Andrew M. Herring , and Vito Di .... Ryan C. Hayward , Adam Z. Weber , Andrew M. Herring , E. ...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/Macromolecules

Poly(2,6-dimethyl-1,4-phenylene oxide)‑b‑poly(vinylbenzyltrimethylammonium) Diblock Copolymers for Highly Conductive Anion Exchange Membranes Yating Yang and Daniel M. Knauss* Department of Chemistry and Geochemistry, Colorado School of Mines, Golden, Colorado 80401, United States ABSTRACT: Diblock copolymer membranes of poly(2,6-dimethyl-1,4phenylene oxide)-b-poly(vinylbenzyltrimethylammonium) (PPO-bPVBTMA) were prepared as anion exchange membranes. The membranes were produced by growing poly(vinylbenzyl chloride) (PVBC) blocks from a PPO macroinitiator using nitroxide-mediated polymerization to make PPO-b-PVBC copolymers with varying PVBC content, then melt pressing into films, and subsequently reacting with trimethylamine. Phase separation was indicated for the PPO-b-PVBC polymers based on the evidence of two Tgs from differential scanning calorimetry measurements. The meltprocessing conditions to form films produced lightly cross-linked materials. The conversion to PPO-b-PVBTMA membranes by reaction with trimethylamine led to materials with varying ion exchange capacities based on the amount of PVBC in the precursor. The water uptake, the dimensional changes with hydration, and the ionic conductivities in the hydroxide and bicarbonate forms were determined. The water uptake was suppressed by the light cross-linking in the hydrophilic PVBTMA component, allowing the films to remain insoluble in aqueous solutions at high IEC values. Membranes with IECs up to 2.9 mequiv/g were flexible under dry and hydrated conditions and showed hydroxide conductivity of 132 ± 3.8 mS/cm at 60 °C.

■

INTRODUCTION Solid anion exchange membranes (AEMs) are critical components for alkaline fuel cells, functioning as an electrolyte to transport anions between the electrodes while serving as a barrier to fuel and electrons. AEMs should be designed to have high ionic conductivity so that useful current can be achieved.1 In addition, AEMs should possess good mechanical properties and have high thermal and chemical stabilities to withstand the alkaline, humid, and elevated temperature operating conditions in a fuel cell.2,3 The challenge to making an ideal anion exchange membrane is to balance a high ion exchange capacity and water content with sufficient mechanical properties. Mechanical properties tend to be compromised by excessive water uptake that accompanies increasing the number of cationic pendent groups in the pursuit of higher ionic conductivity.4 Research has investigated anion exchange membranes with different polymer backbone structures including poly(arylene ether)s,5−10 poly(olefin)s,11−14 polystyrene,15−19 polybenzimidazolium,20,21 and polyphenylene.22,23 The engineering polymer, poly(2,6-dimethyl-1,4-phenylene oxide) (PPO), displays good mechanical properties and outstanding chemical and thermal stability24−26 and anion exchange membranes based on PPO have been recently studied.27−31 The interest in using PPO to prepare AEMs is in part because the benzylic methyl groups of PPO can be readily functionalized to cationic groups and a number of different cations can be anchored to PPO including benzyltrimethylammonium,32 methylimidazolium,33 dimethylimidazolium,29 benzimidazolium,34 and guanidinium.28 Recently, AEMs have been developed using quaternized PPO © XXXX American Chemical Society

with long alkyl side chains pendent to the nitrogen-centered cation.8 The resulting comb-shaped AEMs exhibited organized ionic domains and good hydroxide conductivity (43 mS/cm at room temperature).8 Other modifications to PPO have been done to prepare AEMs. A PPO-silica hybrid anion exchange membrane was investigated and exhibited good swelling resistance and thermal stability, and also showed reasonable hydroxide conductivity (11 mS/cm at room temperature).35 A cross-linked membrane was prepared through blending chloroacetylated PPO and brominated PPO followed by heat treatment and quaternization, and the resulting AEM displayed good conductivity (32 mS/cm at 25 °C) while the cross-linking was reported to improve tensile strength and increase the initial degradation temperature.36 The above studies all took advantage of the benzylic methyl groups on the PPO backbone to attach cationic groups by postpolymerization bromination and subsequent quaternization. This synthetic method is convenient, but can only provide randomly distributed cationic groups on the PPO backbone without the capability to control the bromination position along the polymer chain. Multiple research groups have demonstrated that microphase separation through the formation of diblock copolymers effectively improves the membrane performance with increased ionic conductivity and better mechanical properties.37−41 For example, our group has developed both random and block copolymers containing a Received: March 3, 2015 Revised: June 8, 2015

A

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

polymerization was terminated at specific time intervals by cooling the flask in a dry ice/acetone bath. After the solution was warmed to room temperature, chloroform was added to dilute the reaction mixture. The diluted solution was precipitated twice from methanol. The precipitated polymer was collected by filtration and washed with fresh methanol several times before drying under vacuum at 50 °C overnight. Membrane Preparation and Quaternization. The PPO-bPVBC copolymer powder was melt pressed for 20 min between two Teflon sheets in a Carver 3912 press with homemade heated/water cooled platens at 240 °C and 20000 pounds. Films were obtained after melt pressing with film thickness ranging from 50 to 90 μm. The films were immersed in 45% aqueous trimethylamine solution to convert benzyl chloride groups to quaternized benzyltrimethylammonium chloride groups. Conversion from chloride to the bicarbonate form was done by soaking the chloride form film in 1 M potassium bicarbonate for 48 h, followed by removing any excess salt by rinsing and subsequently soaking the film in fresh DI water for 12 h and changing the water every 2 h. Similarly, the film was converted to the hydroxide form by immersing the benzyltrimethylammonium chloride film in 1 M potassium hydroxide for 48 h. The film was then rinsed with nitrogen purged DI water and then soaked in DI water for 12 h with constant nitrogen purge and changing the water every 2 h. Characterization and Measurements. Proton nuclear magnetic resonance (1H NMR) spectra were recorded using a JEOL ECA-500 FT-NMR at room temperature with CDCl3 as solvent. Chemical shift values (δ) are reported downfield from tetramethylsilane at 0 ppm. The thermal stability of membranes was investigated by thermogravimetric analysis (TGA) using a Seiko TG/DTA 320 Thermal Analyzer with a RT Instruments software upgrade. Polymer samples were dried under vacuum at 60 °C overnight before the TGA measurement and were run under a nitrogen atmosphere with a heating rate of 10 °C per minute. Differential scanning calorimetry (DSC) was performed with a TA Instruments Q20 DSC. Samples were measured in aluminum pans over a temperature range of 20−250 °C with a heating rate of 10 °C/min and a cooling rate of 40 °C/min. The second and third cycles were used to determine the glass transition temperature (Tg), which was measured by the midpoint of the slope change in the heat flow plot. Molecular weights were determined relative to polystyrene standards by gel permeation chromatography (GPC) using a Waters 600-MS pump coupled with a Wyatt Technology Mini Dawn and Optilab DSP interferometric refractometer as detectors. GPC was performed using two PLgel 5 μm mixed D columns with THF as eluent at a flow rate of 1 mL/min. Ultraviolet−visible spectroscopy was done in quartz cuvettes from 200 to 600 nm with a Thermo Electron Corporation Evolution 300 PC spectrophotometer. Samples were prepared to 3 mL with concentrations of 500 ppm in chloroform, which was followed by adding 10 μL of 0.1 M TBAH. The analysis was conducted by running chloroform as background, followed by recording the sample. The presence of phenoxide groups were detected by the absorption between 310 and 380 nm.44 Attenuated total reflectance Fourier transform infrared spectroscopy was performed with a Thermo Electron Nicolet 4700 spectrometer in a range of frequencies between 400 and 4000 cm−1. Samples were measured in their powder forms for PPO, end-capped PPO, and PPOb-PVBC. PPO-b-PVBTMA was measured as a film. The percentage of water uptake of the PPO-b-PVBTMA copolymers was measured in the chloride form and calculated based on the following equation:

quaternized PPO moiety and a hydrophobic poly(2,6-diphenyl1,4-phenylene oxide) moiety.42 The diblock copolymer showed higher conductivities than random copolymers with similar ion exchange capacities. We designed a diblock copolymer of PPO and poly(vinylbenzyltrimethylammonium hydroxide) (PVBTMAOH) to thoroughly exploit the beneficial properties of PPO and direct the phase separation between hydrophilic and hydrophobic phases. We took advantage of the hydroxyl terminal group of PPO to attach a TEMPO functional group for nitroxide mediated polymerization and grow a poly(vinylbenzyl chloride) (PVBC) block from the chain end of PPO. The PVBC block was then conveniently converted to a PVBTMA block by reacting with trimethylamine through the Menshutkin reaction. With this design, we can tune the IEC of the polymer electrolyte by changing the length of the PVBTMA block while maintaining the PPO block intact, therefore benefiting from the mechanical properties and the chemical and thermal stability of the unaltered PPO chain. The quaternized PPO-b-PVBTMA membranes were evaluated for their application as anion exchange membranes by examining thermal stability, dimensional swelling, water uptake, morphology, and ionic conductivity.

■

EXPERIMENTAL SECTION

Materials. 4-Vinylbenzyl chloride (Sigma-Aldrich, 90%) was distilled from calcium hydride (Sigma-Aldrich, 95%) under reduced pressure prior to the block copolymer synthesis. For the alkoxyamine synthesis, 4-vinylbenzyl chloride was passed through a silica gel column before use. 1-[1-(4-Chloromethylphenyl)ethoxy]-2,2,6,6-tetramethylpiperidine (Cl-BzEt-TEMPO) was synthesized according to the reported method43 from manganese acetate (Matheson Co. Inc.), N,N′-disalicylidene-1,2-ethanediamine (Sigma-Aldrich), and 2,2,6,6tetramethylpiperidine (TEMPO) (Sigma-Aldrich). Poly(2,6-dimethyl1,4-phenylene oxide) (PPO) (Sigma-Aldrich) was dried under vacuum at room temperature overnight before use. Tetrabutylammonium hydroxide (TBAH) (1.0 M in methanol) (Sigma-Aldrich) was diluted to 0.1 M with chloroform before use. Trimethylamine (45% in water), anhydrous 1,2-dichlorobenzene, and all solvents were used as received from Sigma-Aldrich. Homopolymer of PVBC (20 kg/mol) was synthesized through nitroxide mediated polymerization. A PPO/ PVBC blend sample was prepared by precipitating the blend solution in chloroform into methanol and drying the precipitate under vacuum for 24 h. Synthesis of End-Capped Poly(2,6-dimethyl-1,4-phenylene oxide) (PPO-BzEt-TEMPO). PPO (5.0 g, 0.25 mmol) was dissolved in 120 mL of 1,2-dichlorobenzene in a 250 mL single-neck flask equipped with a nitrogen inlet and a magnetic stir bar. Sodium hydride (0.18 g, 7.5 mmol) was then added and the mixture was heated to 50 °C and stirred for 1 h. Cl-BzEt-TEMPO (0.93 g, 3.0 mmol) was introduced and the mixture allowed to react under magnetic stirring for 48 h at 50 °C. The mixture was cooled to room temperature and then poured into 1200 mL of methanol. The precipitate was collected by filtration through a glass fritted funnel and washed several times with fresh methanol. The polymer was redissolved in chloroform and precipitated into methanol two more times before drying under vacuum at 50 °C overnight. General Procedure To Synthesize PPO-b-PVBC Diblock Copolymers. The PPO-b-PVBC diblock copolymers were synthesized by employing PPO-BzEt-TEMPO as a macroinitiator for nitroxide mediated polymerization. In a typical synthesis of PPO-bPVBC diblock copolymers, PPO-BzEt-TEMPO macroinitiator (0.50 g, 0.025 mmol) was added to 4-vinylbenzyl chloride (10 mL, 71.0 mmol) in a 15 mL single neck, round-bottom flask. The flask was sealed with a rubber septum and purged for 30 min with nitrogen. The flask was then immersed in a 125 °C oil bath to initiate the polymerization. The size of the PVBC block was controlled by the reaction time. The

water uptake (%) =

Wwet − Wdry Wdry

× 100

The dry membrane weight (Wdry) was determined by drying the membrane at 60 °C under vacuum for 24 h. The wet membrane weight (Wwet) was measured after the membranes were fully hydrated by soaking the dry membranes in DI water until a constant weight was obtained. B

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules Scheme 1. Synthetic Route To Form PPO-b-PVBTMA Copolymer Membranes

The dimensional swelling of the PPO-b-PVBTMA copolymers was determined through the dimension difference between the dry membrane and the wet membrane in their chloride forms. The dimension swelling was calculated from the following equation:

dimension swelling (%) =

Dwet − Ddry Ddry

σ=

where σ is the ionic conductivity of the membrane (S cm−1), d is the distance between reference electrodes (mm), Ls and Ws are the thickness (mm) and width (mm) of the membrane, and R is the ohmic resistance of the membrane (Ω). The hydroxide conductivity after immersion in alkaline solution at elevated temperature was determined by soaking a sample in 1 M KOH at 60 °C for a period of time, washing with ultra pure deionized water for 3 h, changing to fresh water every 20 min, and then measuring the conductivity as described above. The sample was then reimmersed in the KOH solution for the next period of time for a total of 13 days. Tensile properties of PPO-b-PVBTMA membrane were measured with an Instru-Met frame tensile tester (MTS) A30-33 using a 250pound load cell with a stretching rate of 12.7 mm/min (0.5 in./min). The PPO-b-PVBTMA membrane was dried at 60 °C for 24 h and cut into rectangular samples with the dimensions precisely recorded (roughly 8 mm width, 50 mm length, and 75 μm thickness) before measurement. Young’s modulus, stress at break, and elongation at break values were determined. Transmission electron microscopy (TEM) experiments were imaged with an FEI Tecnai F20 electron microscope (FEI Corp, Hillsboro, OR) operating at 200 kV. TEM samples were prepared by embedding the quaternized polymer sample in epoxy and sectioning into ∼100 nm thick sections at room temerature using a Leica Ultracut UCT ultramicrotome (Leica Microsystems, Buffalo Grove, IL). The sections were collected onto 200 mesh copper EM grids, followed by staining for 3 h with osmium tetroxide vapor from a 4% aqueous solution. ImageJ analysis was conducted to develop histograms of bright and dark areas of the TEM micrographs. The dark/bright area percentages in the TEM micrographs were calculated to determine which phase was darkened after exposing to OsO4.

× 100

where Ddry and Dwet represent the dimension of the dry membrane and the dimension of the fully hydrated membrane, respectively. The ion exchange capacity values (IEC) of the membranes were determined by standard back-titration methods.11,17,45 The hydroxide form membranes were each immersed in 9.00 mL of standardized HCl solution (0.01 M) for 24 h to fully neutralize the hydroxide ions in the membrane. The residual HCl was titrated by a standardized NaOH solution (0.01 M) using an automatic titrator (Mettler Toledo, Model G20). The titrated samples were then dried under vacuum at 60 °C overnight before measuring the mass of the dry sample in the chloride form. The IEC value of each membrane was calculated from the following equation:

IEC =

VHCl × cHCl − VNaOH × cNaOH Wdry

where VHCl and cHCl represent the volume and concentration of the standardized HCl solution; VNaOH and cNaOH represent the volume and concentration of the standardized NaOH solution; Wdry represents the weight of the dry membrane. The hydration number (λ) was calculated from the following equation:

λ=

d LsWR s

WU × 1000 M H2O × IEC

where WU represents the percentage of water uptake of the membrane, MH2O represents the molecular weight of H2O; IEC represents the ion exchange capacity of the membrane. In-plane conductivity was measured by electrochemical impedance spectroscopy (EIS) using a 4-probe Teflon cell with Pt electrodes connected to a BioLogic VMP3 Potentiostat in a frequency range from 1 Hz to 100 kHz. The chloride conductivity at 95% relative humidity was determined in a TestEquity (Solatron 1007H Model) environmental chamber to control the temperature and percentage of relative humidity for measurement. The hydroxide and bicarbonate conductivity under fully hydrated conditions was performed with the same Teflon cell immersed in deionized ultra pure water (18 MΩ resistance). A control cell with a Teflon film was running as background while measuring the polymer samples. The ionic conductivity was obtained from the following equation:

■

RESULTS AND DISCUSSION The objective of this research was to develop highly conductive AEMs while maintaining robust membranes with suppressed water uptake and low dimensional swelling. We designed diblock copolymers with PPO blocks to provide mechanical properties and PVBTMA blocks to provide a hydrophilic medium for ionic conductivity. PPO was chosen because of its outstanding chemical and thermal stability and its expected ability to form robust films. PVBTMA was chosen as the hydrophilic domain due to the reasonable alkaline stability of benzyltrimethylammonium ion46 and the ease of polymerC

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

ization and quaternization resulting from the bifunctionality of 4-vinylbenzyl chloride. Similarly designed block copolymers of PPO and poly(vinylbenzyl phosphonic acid) have previously been prepared as proton exchange membranes.47 Atom transfer radical polymerization (ATRP) was employed to attach the poly(vinylbenzyl phosphonic acid) block to the PPO block. The resulting proton exchange membranes exhibited promising proton conductivity, good thermal stability and mechanical properties. More recent research has investigated phaseseparated copolymers incorporating PPO for AEM applications. Our group developed a block copolymer system with quaternized PPO as the hydrophilic block and poly(2,6diphenyl-1,4-phenylene oxide) as the hydrophobic block.42 The hydroxide conductivity for the best sample reached 84 mS/cm under 95% relative humidity at 80 °C. Bai and co-workers recently prepared a series of triblock copolymers with quaternized PPO as the hydrophilic end block and polysulfone as the hydrophobic middle block that showed promising hydroxide conductivity,48 while Xu and co-workers synthesized graft copolymers with PPO as the backbone and PVBTMA as grafted side chains.49 Low water uptake and encouraging hydroxide conductivity were observed in the graft copolymers. In the work described here, diblock copolymers of PPO-bPVBTMA were designed to combine the mechanically supportive PPO block with the ion conductive PVBTMA block. The first step was to end-cap the phenol end group of PPO with the benzylmethyl TEMPO moiety (Scheme 1). The PPO phenol end groups were deprotonated with sodium hydride at 50 °C before adding Cl-BzEt-TEMPO. The deprotonated phenoxide substituted for chloride on the benzylic carbon of Cl-BzEt-TEMPO to synthesize the PPOBzEt-TEMPO. End group analysis of the phenol in the phenoxide form was conducted using ultraviolet/visible (UV/ vis) spectroscopy, showing the phenoxide group at an absorbance between 310 nm to 380 nm.44 UV−vis spectra of the PPO are shown in Figure 1, before and after the introduction of the BzEt-TEMPO moiety. The lower spectrum shows the end-capped PPO with no absorbance in the region of 310 to 380 nm, indicating the successful nucleophilic substitution between PPO and Cl-BzEt-TEMPO.

Figure 1. UV−vis spectra of phenoxide-terminal PPO (upper) and end-capped PPO with BzEt-TEMPO moiety after nucleophilic substitution with Cl-BzEt-TEMPO (lower).

Figure 2. GPC curves of original PPO, PPO macroinitiator and two representative PPO-b-PVBC block copolymers.

Figure 3. 1H NMR spectrum of PPO-b-PVBC6 in CDCl3. D

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules Table 1. Molecular Weight Characterization of Each PPO-b-PVBC Block Copolymer Mn (NMR)a sample

Mn of PVBC block (kg/mol)

Mn of PPO-b-PVBC (kg/mol)

weight percentage of PVBCa (%)

Mn of PPO-b-PVBC (GPC)b (kg/mol)

Mw/Mnb

PPO-b-PVBC1 PPO-b-PVBC2 PPO-b-PVBC3 PPO-b-PVBC4 PPO-b-PVBC5 PPO-b-PVBC6 PPO-b-PVBC7 PPO-b-PVBC8 PPO-b-PVBC9 PPO-b-PVBC10 PPO-b-PVBC11 PPO-b-PVBC12

3.7 6.2 7.6 9.3 12 13 18 20 23 28 33 57

23.7 26.2 27.6 29 32 33 38 40 43 48 53 77

16 24 28 32 37 39 47 50 53 58 62 74

27 28 28 31 34 36 39 42 46 52 59 85

1.74 1.64 1.83 1.81 1.61 1.59 1.69 1.65 1.97 1.78 1.66 1.74

a Determined by proton NMR spectroscopy from the integrations of methyl groups of PPO and benzylic methylene groups of PVBC. bDetermined by GPC relative to polystyrene standards.

Figure 4. TGA and DTG thermograms of PPO-b-PVBC1, PPO-bPVBC10, and PPO-b-PVBC12 under flowing nitrogen at a heating rate of 10 °C/min. The two numbers in each pair of parentheses after the sample name represent the molecular weight of the PPO segment and PVBC segment, respectively.

Figure 5. DSC thermograms of blended PPO and PVBC with a weight percentage of PVBC of 50%, homopolymer PVBC (20 kg/mol) and homopolymer PPO (20 kg/mol), and diblock copolymers PPO-bPVBC1, PPO-b-PVBC3, PPO-b-PVBC8, PPO-b-PVBC10, and PPO-bPVBC12 with weight percentages of PVBC of 16%, 28%, 50%, 58%, and 74%. Vertical dashed lines are added for clarification.

PPO-BzEt-TEMPO was employed as an initiator for nitroxide mediated polymerization of vinylbenzyl chloride to synthesize the PPO-b-PVBC block copolymers. The polymerization was carried out in bulk vinylbenzyl chloride at 125 °C for different reaction times, and a series of block copolymers with different PVBC block lengths was synthesized in this research. The PPO-b-PVBC block copolymers were characterized by various techniques. The GPC traces are displayed in Figure 2 including the original PPO, the end-capped PPO, and two representative PPO-b-PVBC copolymers with different PVBC block lengths. The PPO Mn was determined to be 20 kg/mol relative to polystyrene stndards. The GPC traces of the original PPO and the end-capped PPO almost exactly overlapped due

to the negligible effect of the small end group on the molecular weight of the end-capped PPO. The clear shifts of the chromatograms to lower elution volume from the PPO macroinitiator to the block copolymers demonstrate the formation of the higher molecular weight copolymer, suggesting the successful synthesis of PPO-b-PVBC copolymers. The lower elution volume of PPO-b-PVBC10 than PPOb-PVBC6 indicated the higher molecular weight of PPO-bPVBC10 and its longer PVBC blocks. Also the unimodal appearance of the chromatograms of the copolymers indicated the absence of unreacted PPO or self-initiated polymerization of VBC. The PPO-b-PVBC copolymers were characterized by 1H NMR spectroscopy, from which the fraction of PVBC can be E

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 6. FTIR spectra of end-capped PPO, copolymer PPO-b-PVBC8, and quaternized copolymer PPO-b-PVBTMA8.

chosen as representative cases because they possess the lowest, middle, and the highest weight percentage of the PVBC segment, respectively. The copolymers exhibited stabilities up to 250 °C, beyond which there were two stages of degradation. On the basis of the TGA trace of the PVBC homopolymer, the first weight loss, occurring between 255 and 400 °C, was associated with the decomposition of the benzyl chloride groups. The TGA and DTG curves showed that a longer PVBC segment block resulted in a greater weight loss in the first degradation stage. The second weight loss above 400 °C revealed the degradation of the copolymer backbone, which was not affected by the PVBC segment length but resulted in similar weight percentage of the degradation residue. The three samples exhibited the same onset decomposition temperatures at approximately 255 °C. The results indicate that melt processing of the films could be accomplished below 255 °C. The block copolymers were synthesized in order to have phase-separated blocks with the PPO providing good film properties and the PVBTMA to provide for the ionic conductivity in a hydrophilic medium. While it is expected that the PPO and PVBTMA would be well phase-separated because of the distinct difference between the PPO and a cationic block, our processing of the films as the PPO-b-PVBC copolymers requires that the materials are phase separated prior to conversion to PVBTMA. The miscibility of many blends of PPO and styrenic polymers and copolymers have been studied,51,52 however studies of the miscibility of PPO and PVBC have not been reported. DSC analysis was employed to characterize the miscibility of the block copolymers. Figure 5 depicts DSC results for PPO homopolymer, PVBC homopolymer, a 50/50 blend of the two homopolymers, and representative PPO-b-PVBC copolymers. The PVBC and PPO homopolymers are found to have Tgs at 109 and 213 °C, respectively. A 50/50 blend of the two homopolymers shows two Tgs at close to the same relative temperatures of the pure homopolymers, indicating immiscibility of the blend. The diblock copolymers also show two Tgs, indicating phase separation, at least for samples where the amount of the minor component is sufficient to allow

calculated based on the distinguishable peaks from each block. The proton NMR spectrum of PPO-b-PVBC6 is illustrated in Figure 3 as a representative example. The NMR spectra exhibit characteristic peaks at δ = 4.5 ppm (a) from protons on the benzylic carbon of PVBC, as well as peaks at δ = 2.1 ppm (b) representing protons from the methyl groups of PPO. The quantitative ratio of PVBC block to PPO block was calculated from the integration of the “a” and “b” peaks. The repeat unit ratio of PVBC block to PPO block was calculated, and the total molecular weight of each block copolymer was then calculated based on the repeat unit ratio of the two blocks and the number-average molecular weight of PPO (20 kg/mol). The number-average molecular weights of the PPO-b-PVBC copolymers are displayed in Table 1 along with the weight percentage of PVBC. The weight percentage of PVBC in the block copolymers varied from 16% to 74% with the concomitant increase in number-average molecular weight of the PVBC. The copolymers were ultimately to be converted from PVBC to an ammonium salt form PVBTMA to form block polyelectrolyte membranes. Because of the solubility differences of PPO and the PVBTMA, processing from a common solvent is a difficult endeavor. It is desirable to develop a thermal process for preparing membranes, however melt processing of the polyelectrolyte block copolymers is also precluded. We have had good success with solvent casting of block copolymer films in the benzyl halide form and then subsequently converting to the ammonium salt form by reacting the films with aqueous trimethylamine.42,50 In this case, we expected to be able to melt process the block copolymers into films while in the benzyl chloride form and then convert into ionic block copolymers. Melt pressing of the films required the PPO-b-PVBC copolymers to be stable at the melt pressing temperature. Thermal gravimetric analysis (TGA) was used to determine the highest temperature that PPO-b-PVBC remained stable before degradation. The TGA thermograms and derivative thermogravimetry (DTG) curves are shown in Figure 4 for PPO-bPVBC1, PPO-b-PVBC10, and PPO-b-PVBC12 recorded between 200 and 600 °C. These three copolymers were F

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Article

Calculated from 1H NMR spectra of PPO-b-PVBC samples assuming complete conversion from PVBC to PVBTMA. bTitrated values. cNumber of water molecules per ammonium group. dIn-plane swelling in water. eThrough-plane swelling in water.

1.3 3.5 3.7 4.2 10.6 12.8 14.1 15.6 16.5 19.3 24.5 53.2 0.1 0.1 0.4 0.5 0.7 0.3 0.1 0.6 0.3 0.4 0.7 0.6 1.0 2.2 5.7 5.5 10.9 10.1 11.7 13.4 13.7 16.8 16.9 31.6 4.0 4.4 5.2 4.5 10.4 10.8 11.6 14.2 15.0 17.2 18.2 30.2 5 11 13 13 32 33 50 64 70 87 95 174 0.7 1.4 1.4 1.6 1.7 1.7 2.4 2.5 2.6 2.8 2.9 3.2 1.0 1.4 1.6 1.8 2.1 2.2 2.6 2.7 2.9 3.1 3.3 3.8 20 30 35 39 45 48 55 58 62 65 69 80 PPO-b-PVBTMA1 PPO-b-PVBTMA2 PPO-b-PVBTMA3 PPO-b-PVBTMA4 PPO-b-PVBTMA5 PPO-b-PVBTMA6 PPO-b-PVBTMA7 PPO-b-PVBTMA8 PPO-b-PVBTMA9 PPO-b-PVBTMA10 PPO-b-PVBTMA11 PPO-b-PVBTMA12

observation. The upper Tg correlates well with the Tg of PPO polymer for each sample. A small temperature increase is observed in the lower Tg for the samples that have enough PVBC block for observation. The temperature increase in the lower Tg can be attributed to a number of possible factors: The attachment of one chain end to the PPO in the formation of the block copolymer can increase the Tg; some phase mixing between the PPO and PVBC to form a PVBC-rich phase rather than a pure PVBC phase; and constraint of segmental motion by the glassy PPO in the dispersed system could increase the Tg of the PVBC phase. In addition, cross-linking can occur from the benzyl chloride groups53 that could increase the Tg. The remaining thermally labile TEMPO chain end was also expected to participate in any radical cross-linking at elevated temperatures. We do find some cross-linking to occur during the thermal treatment as the samples removed from the DSC pans were found to be insoluble. PPO-b-PVBC films were prepared by pressing the copolymer powders at 240 °C for 20 min. The processing temperature was determined from TGA and DSC measurements to exceed the Tgs of both the PVBC and the PPO phases yet avoid thermal degradation. The materials were all found to flow sufficiently prior to any cross-linking such that good consolidated films were obtained. Under the film formation temperature conditions some cross-linking of the PVBC does occur as evidenced by the fact that the PPO-b-PVBC copolymer films were found to be insoluble (but swelled) in the solvents that were good solvents prior to melt pressing the films. The crosslinking is expected to be advantageous for the water uptake and dimensional stability of the materials. The melt-pressed films were immersed in 45% aqueous trimethylamine solution to obtain quaternized membranes. The generation of benzyltrimethylammonium groups is demonstrated by FTIR spectroscopy. IR spectra of the PPO macroinitiator, PPO-b-PVBC8, and PPO-b-PVBTMA8 are displayed in Figure 6 as representative samples. In the IR spectrum of copolymer PPO-b-PVBC8, the presence of the absorption peak around 650 cm−1 is characteristic of C−Cl stretching, indicating the successful attachment of the PVBC block to the PPO block.54 After the quaternization, the disappearance of the C−Cl stretch absorption implies the effective substitution of chloride groups to form an ammonium group. A new absorption peak appears, centered around 3400 cm−1, in the IR spectrum of the quaternized copolymer as a result of moisture absorbed by the hygroscopic quaternary ammonium groups. By converting the PPO-b-PVBC films into PPO-b-PVBTMA membranes, the weight percentage of PVBTMA increases over that of the PVBC due to the extra mass of the trimethylammonium group. The samples are presented in Table 2 with characterization of ion exchange capacity (IEC), water uptake, ionic conductivity under different conditions, and dimensional swelling. Each of these parameters is important for determining the suitability of the materials for anion exchange membranes. The increase in the fraction of PVBTMA in the copolymer leads to a higher IEC and a concomitant increase in the water uptake. The IEC values of the melt-pressed PPO-b-PVBTMA films as determined by titration were mostly found to be slightly lower than the IEC values expected from the amount of benzyl chloride groups determined from the NMR spectra of the PPO-b-PVBC prior to melt processing. The IEC difference between that expected based on NMR analysis and the titrated

a

0.2 0.1 0.1 0.1 0.5 0.3 0.3 0.8 0.5 1.1 1.5 2.9 ± ± ± ± ± ± ± ± ± ± ± ±

60 °C

2 3 3 4 13 12 16 23 27 31 36 46 0.1 0.1 0.2 0.1 0.2 0.1 0.3 0.2 0.5 0.8 1.2 0.9 ± ± ± ± ± ± ± ± ± ± ± ±

20 °C

0.7 1.2 1 2 8 4 6 8 9 13 16 18 0.5 0.3 0.8 0.4 2.9 2.0 1.5 0.8 3.2 2.3 3.8 4.5 ± ± ± ± ± ± ± ± ± ± ± ±

60 °C

5 9 12 13 43 37 45 58 60 115 132 166 0.1 0.3 0.3 0.2 1.0 0.8 1.8 1.6 1.3 2.7 1.5 3.2 ± ± ± ± ± ± ± ± ± ± ± ±

20 °C

2 4 5 6 25 20 28 36 45 61 73 87 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.2 0.3 0.3 0.4 0.5 0.6 0.7 0.5 0.8 1.1 1.4 4.7

swellinge at 20 °C (%) swellingd at 20 °C (%) IECb (mequiv/g) IECa (mequiv/g) wt % of PVBTMA[Cl]a sample

Table 2. Characterization of PPO-b-PVBTMA Membranes

WU (%)

λc

hydroxide conductivity in water (mS/cm)

bicarbonate conductivity in water (mS/cm)

Macromolecules

G

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 7. Ionic conductivity of the PPO-b-PVBTMA series membranes under fully hydrated condition in their hydroxide forms (solid shape) and bicarbonate forms (hollow shape) at 20 and 60 °C as a function of (a) water uptake and (b) ion exchange capacity and (c) weight percentage of PVBTMA.

PVBTMA12 remained flexible as wet films. PPO-b-PVBTMA12 became noticeably more fragile under fully hydrated conditions and samples with higher fractions of PVBTMA were not studied because of the expectation of poor properties. The number of water molecules per cationic group under fully hydrated conditions, hydration number (λ), is an important parameter for AEMs. λ was calculated for the different membranes and was determined to remain relatively constant at a value of about 4−5 when the PVBTMA weight percentage was below 40 wt %. For samples above 40 wt % of PVBTMA, the hydration number was found to increase with an increase in the weight percentage of PVBTMA, increasing to a value of 30 for the sample with PVBTMA of 80 wt %. Changes in dimension of the membranes in the plane and through the plane of the films were also determined by measuring dry samples compared to fully hydrated samples. Small dimensional changes are important if the membranes are to be used in a fuel cell or other applications where the membranes are held in place. The samples show relatively small changes, in agreement with the relatively low water uptake for a given IEC, compared to other reported materials.17,48 The ionic conductivity was determined at 20 and 60 °C in both the hydroxide and bicarbonate counterion forms (Table 2). The conductivity is dependent on both IEC and the amount of water that is absorbed into the membrane, and an increase in ionic groups expectedly leads to higher water uptake because of the higher content of the hydrophilic component. Conductivity at 60 °C versus water uptake (Figure 7a) demonstrates higher ionic conductivity along with the increased water uptake that is a consequence of the higher IECs. The plot of conductivity against water uptake demonstrates an almost linear conductivity increase with increased water uptake for both counterions. The relationship between conductivity and IEC in Figure 7b

Figure 8. Hydroxide conductivity with time for PPO-b-PVBTMA9 immersion in 1 M KOH solution at 60 °C.

IEC was attributed in part to some cross-linking of the benzyl chloride groups that occurred during melt-pressing of the films. The cross-linking reduced the total number of benzyl chloride groups available for conversion to ammonium chloride groups. However, the cross-linking of chains in what ultimately becomes the ionic portion of the membranes ensures that the material cannot dissolve in aqueous solutions and inhibits excessive water uptake and dimensional changes during hydration. For example, PPO-b-PVBTMA11 with an IEC of 2.4 mequiv/g displayed a water uptake of only 50%. In comparison, some recently published AEM studies7,17,37,48 report much higher water uptake at equivalent or lower IECs. The PPO-b-PVBTMA membranes are able to support more charge carrier groups (higher IEC values) at lower water uptake, which can be attributed to the light cross-linking of the ionic block. The membranes in this research remained robust and flexible under dry conditions, and all but PPO-b-

Figure 9. Cross-sectional TEM micrographs of dry membranes of PPO-b-PVBTMA1 (a), PPO-b-PVBTMA10 (b), and PPO-b-PVBTMA12 (c). H

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

■

CONCLUSION This research has developed a method for the formation of PPO-b-PVBTMA membranes that demonstrate useful properties for application as anion exchange membranes. The work describes how PPO, as an engineering polymer with good mechanical properties and high chemical stability, can be converted into a macroinitiator for the nitroxide-mediated polymerization of a PVBC block and the subsequent melt processing and conversion into an ionic PVBTMA block. The melt processing has the benefit of producing a cross-linked membrane that inhibits dissolution in water for membranes with high IECs. The materials show distinct phase separation of the two blocks, which promotes the benefits of each block component. At high fractions of PVBTMA, the data, including TEM, water uptake, and conductivity studies indicate that the PVBTMA becomes well-connected leading to high conductivity. The PPO-b-PVBTMA materials remain as flexible membranes under both dry and wet conditions and samples with high hydroxide conductivity and relatively low water uptake are produced by the synthesis procedure. Further study is ongoing to examine the effect of cross-linking in the hydrophilic phase on the mechanical properties of these membranes and in examining the potential for these materials as alkaline fuel cell membranes.

illustrates that the change in conductivity with increasing IEC is small for samples with low IEC, and becomes a much larger change at higher IECs. This transition is proposed to be related to the change in morphology with increasing PVBTMA content and the formation of a continuous ion-conducting network facilitating ion transportation at higher IEC values. This is more clearly seen by plotting conductivity versus weight percent PVBTMA (Figure 7c) and the transition occurs above 40 wt % PVBTMA. Alkaline stability of PPO-b-PVBTMA9 membrane was investigated in 1 M KOH at 60 °C by measuring the change in hydroxide conductivity over 13 days (Figure 8). PPO-bPVBTMA9 showed no observable change in conductivity over the duration of the experiment. The appearance of the membranes after 13 days of exposure to the alkaline environment also showed little difference from that of the initially prepared membrane. The membrane remained flexible and no increase in brittleness was observed. The cross-linking within the membranes may improve the alkaline stability of quaternized ammonium ions.27 Sufficient mechanical properties are necessary for the application as anion exchange membranes. PPO-b-PVBTMA9 was chosen for examination of tensile properties due to its overall balance of high ionic conductivity from a relatively high IEC and acceptable water uptake and dimensional swelling. The tensile properties of PPO-b-PVBTMA9 showed a Young’s modulus of 3.5 ± 0.3 GPa, a stress at break of 10.2 ± 0.7 MPa, and an elongation at break of 4.8 ± 0.4%. The mechanical properties are comparable to those observed for other AEMs with PPO as the backbone.29,33 The properties are expected to change with the hydration level of the membrane and further work is being done to examine the hydrated properties of the different membranes. The PPO-b-PVBTMA membranes were investigated by TEM to observe the morphology with different weight percentages of PVBTMA. Typical cross-sectional TEM micrographs are presented in Figure 9 for the dry membranes of PPO-b-PVBTMA1, PPO-b-PVBTMA10, and PPO-bPVBTMA12. Distinct phase separation was observed in each case. The samples were stained with OsO4 to obtain contrast and analysis determined that the percentage of the darkened domains corresponded reasonably with the weight percentage of PVBTMA, indicating staining of the PVBTMA phase after exposure to OsO4. The membranes exhibited phase separation without long-range order in all TEM micrographs. PPO-bPVBTMA1 with the lowest fraction of PVBTMA shows the PVBTMA phase dispersed as spherical shapes within the PPO matrix (Figure 9a). An increase in the weight percentage of PVBTMA to 65 wt % for PPO-b-PVBTMA10 shows an increase in the amount of darker PVBTMA domains and an elongation of the domains (Figure 9b). When the weight percentage of PVBTMA increases to 80 wt % for PPO-bPVBTMA12, the PPO becomes the dispersed phase in a PVBTMA matrix (Figure 9c). The TEM micrographs for the three membranes reveal a more connected PVBTMA phase is achieved in the samples with higher weight percentage and longer length of PVBTMA blocks. No thermal or solvent annealing studies were conducted to improve the ordering of the membranes. Such studies could lead to improved order and higher conductivities although the cross-linking in the PVBTMA phase from the melt-processing could also inhibit any reorganization.

■

AUTHOR INFORMATION

Corresponding Author

*(D.M.K.) E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank the Army Research Office for support of this research under the MURI, No. W911NF-10-1-0520. NMR spectroscopy was made possible by a grant from the National Science Foundation (Grant No. CHE-0923537).

■

REFERENCES

(1) Merle, G.; Wessling, M.; Nijmeijer, K. J. Membr. Sci. 2011, 377, 1−35. (2) Brandon, N. P.; Skinner, S.; Steele, B. C. H. Annu. Rev. Mater. Res. 2003, 33, 183−213. (3) Yu, Y.; Li, H.; Wang, H.; Yuan, X.; Wang, G.; Pan, M. J. Power Sources 2012, 205, 10−23. (4) Kusoglu, A.; Karlsson, A. M.; Santare, M. H.; Cleghorn, S.; Johnson, W. B. J. Power Sources 2007, 170, 345−358. (5) Tanaka, M.; Fukasawa, K.; Nishino, E.; Yamaguchi, S.; Yamada, K.; Tanaka, H.; Bae, B.; Miyatake, K.; Watanabe, M. J. Am. Chem. Soc. 2011, 133, 10646−10654. (6) Li, N.; Zhang, Q.; Wang, C.; Lee, Y. M.; Guiver, M. D. Macromolecules 2012, 45, 2411−2419. (7) Mohanty, A. D.; Lee, Y. B.; Zhu, L.; Hickner, M. A.; Bae, C. Macromolecules 2014, 47, 1973−1980. (8) Li, N.; Leng, Y.; Hickner, M. A.; Wang, C. Y. J. Am. Chem. Soc. 2013, 135, 10124−10133. (9) Yan, J.; Hickner, M. A. Macromolecules 2010, 43, 2349−2356. (10) Li, L.; Wang, Y. J. Membr. Sci. 2005, 262, 1−4. (11) Noonan, K. J. T.; Hugar, K. M.; Kostalik, H. A.; Lobkovsky, E. B.; Abruña, H. D.; Coates, G. W. J. Am. Chem. Soc. 2012, 134, 18161− 4. (12) Kostalik, H. A.; Clark, T. J.; Robertson, N. J.; Mutolo, P. F.; Longo, J. M.; Abruña, H. D.; Coates, G. W. Macromolecules 2010, 43, 7147−7150. I

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules (13) Sherazi, T. A.; Yong Sohn, J.; Moo Lee, Y.; Guiver, M. D. J. Membr. Sci. 2013, 441, 148−157. (14) Robertson, N. J.; Kostalik, H. A.; Clark, T. J.; Mutolo, P. F.; Abruña, H. D.; Coates, G. W. J. Am. Chem. Soc. 2010, 132, 3400−4. (15) Zeng, Q. H.; Liu, Q. L.; Broadwell, I.; Zhu, A. M.; Xiong, Y.; Tu, X. P. J. Membr. Sci. 2010, 349, 237−243. (16) Lin, B.; Qiu, L.; Lu, J.; Yan, F. Chem. Mater. 2010, 22, 6718− 6725. (17) Price, S. C.; Ren, X.; Jackson, A. C.; Ye, Y.; Elabd, Y. A.; Beyer, F. L. Macromolecules 2013, 46, 7332−7340. (18) Tomoi, M.; Yamaguchi, K.; Ando, R.; Kantake, Y.; Aosaki, Y.; Kubota, H. J. Appl. Polym. Sci. 1997, 64, 1161−1167. (19) Cao, Y. C.; Wang, X.; Mamlouk, M.; Scott, K. J. Mater. Chem. 2011, 21, 12910−12916. (20) Thomas, O. D.; Soo, K. J. W. Y.; Peckham, T. J.; Kulkarni, M. P.; Holdcroft, S. J. Am. Chem. Soc. 2012, 134, 10753−6. (21) Wright, A. G.; Holdcroft, S. ACS Macro Lett. 2014, 3, 444−447. (22) Hibbs, M. R.; Fujimoto, C. H.; Cornelius, C. J. Macromolecules 2009, 42, 8316−8321. (23) Hibbs, M. R. J. Polym. Sci., Part B: Polym. Phys. 2013, 51, 1736− 1742. (24) Hay, A. S. In High performance polymers: their origin and development In Proceedings of the Symposium on the History of High Performance Polymers at the American Chemical Society Meeting; Elsevier: New York, 1986; pp 209−214. (25) Hay, A. S. J. Polym. Sci., Part A: Polym. Chem. 1998, 36, 505− 517. (26) Hay, A. S.; Blanchard, H. S.; Endres, G. F.; Eustance, J. W. J. Am. Chem. Soc. 1959, 81, 6335−6336. (27) Li, N.; Wang, L.; Hickner, M. Chem. Commun. 2014, 50, 4092− 4095. (28) Lin, X.; Wu, L.; Liu, Y.; Ong, A. L.; Poynton, S. D.; Varcoe, J. R.; Xu, T. J. Power Sources 2012, 217, 373−380. (29) Lin, X.; Varcoe, J. R.; Poynton, S. D.; Liang, X.; Ong, A. L.; Ran, J.; Li, Y.; Xu, T. J. Mater. Chem. A 2013, 1, 7262−7269. (30) Li, Q.; Liu, L.; Miao, Q.; Jin, B.; Bai, R. Chem. Commun. 2014, 50, 2791−2793. (31) Li, N.; Yan, T.; Li, Z.; Thurn-Albrecht, T.; Binder, W. H. Energy Environ. Sci. 2012, 5, 7888−7892. (32) Deavin, O. I.; Murphy, S.; Ong, A. L.; Poynton, S. D.; Zeng, R.; Herman, H.; Varcoe, J. R. Energy Environ. Sci. 2012, 5, 8584−8597. (33) Ran, J.; Wu, L.; Varcoe, J. R.; Ong, A. L.; Poynton, S. D.; Xu, T. J. Membr. Sci. 2012, 415−416, 242−249. (34) Lin, X.; Liang, X.; Poynton, S. D.; Varcoe, J. R.; Ong, A. L.; Ran, J.; Li, Y.; Li, Q.; Xu, T. J. Membr. Sci. 2013, 443, 193−200. (35) Wu, Y.; Wu, C.; Varcoe, J. R.; Poynton, S. D.; Xu, T.; Fu, Y. J. Power Sources 2010, 195, 3069−3076. (36) Wu, L.; Xu, T. J. Membr. Sci. 2008, 322, 286−292. (37) Tanaka, M.; Fukasawa, K.; Nishino, E.; Yamaguchi, S.; Yamada, K.; Tanaka, H.; Bae, B.; Miyatake, K.; Watanabe, M. J. Am. Chem. Soc. 2011, 133, 10646−54. (38) Ye, Y.; Sharick, S.; Davis, E. M.; Winey, K. I.; Elabd, Y. A. ACS Macro Lett. 2013, 2, 575−580. (39) Elabd, Y. A.; Hickner, M. A. Macromolecules 2011, 44, 1−11. (40) Weber, R. L.; Ye, Y.; Schmitt, A. L.; Banik, S. M.; Elabd, Y. A.; Mahanthappa, M. K. Macromolecules 2011, 44, 5727−5735. (41) Beers, K. M.; Balsara, N. P. ACS Macro Lett. 2012, 1, 1155− 1160. (42) Rebeck, N. T.; Li, Y.; Knauss, D. M. J. Polym. Sci., Part B: Polym. Phys. 2013, 51, 1770−1778. (43) Bothe, M.; Schmidt-Naake, G. Macromol. Rapid Commun. 2003, 24, 609−613. (44) Riffle, J. S.; Shchori, E.; Banthia, A. K.; Freelin, R. G.; Ward, T. C.; Mcgrath, J. E. J. Polym. Sci.: Polym. Chem. Ed. 1982, 20, 12. (45) Park, A. M.; Turley, F. E.; Wycisk, R. J.; Pintauro, P. N. Macromolecules 2013, 47, 227−235. (46) Einsla, B. R.; Chempath, S.; Pratt, L.; Boncella, J.; Rau, J.; Macomber, C.; Pivovar, B. ECS Trans. 2007, 11, 1173−1180.

(47) Kumar, A.; Pisula, W.; Markova, D.; Klapper, M.; Müllen, K. Macromol. Chem. Phys. 2012, 213, 489−499. (48) Li, Q.; Liu, L.; Miao, Q.; Jin, B.; Bai, R. Polym. Chem. 2014, 5, 2208−2213. (49) Ran, J.; Wu, L.; Wei, B.; Chen, Y.; Xu, T. Sci. Rep. 2014, 4. (50) Li, Y.; Jackson, A. C.; Beyer, F. L.; Knauss, D. M. Macromolecules 2014, 47, 6757−6767. (51) Messé, L.; Prud’homme, R. E. J. Polym. Sci., Part B: Polym. Phys. 2000, 38, 1405−1415. (52) Sharma, K. R. Polymer Thermodynamics: Blends, Copolymers and Reversible Polymerization. CRC Press: Boca Raton, FL, 2011. (53) Song, J.-M.; Lee, S.-Y.; Woo, H.-S.; Sohn, J.-Y.; Shin, J. J. Polym. Sci., Part B: Polym. Phys. 2014, 52, 517−525. (54) Xu, H.; Fang, J.; Guo, M.; Lu, X.; Wei, X.; Tu, S. J. Membr. Sci. 2010, 354, 206−211.

J

DOI: 10.1021/acs.macromol.5b00459 Macromolecules XXXX, XXX, XXX−XXX