Synthesis and Properties of Fluorocarbon–Hydrocarbon Hybrid Block Copolymers with Perfluorosulfonimide Acid

Abstract

Novel fluorocarbon–hydrocarbon hybrid block copolymer electrolytes were synthesized. The block copolymer electrolytes consist of poly(perfluoropropyl sulfonimide) (PC3SI) as a perfluorinated hydrophilic segment and poly(ether ether sulfone) as a hydrocarbon hydrophobic segment. The sulfonimide group of poly(perfluoropropyl sulfonimide) has superacidity, very low equivalent weight (EW = 293 g/equiv), and a proton conductivity of 1.2 × 10^–2 S/cm under dry conditions and 25 °C, although soluble in water. The proton conductivity of the block copolymer was 1.7 × 10^–3 S/cm at 20% relative humidity and 25 °C, which is three times as high as that of Nafion 112.

Introduction

Fuel cell (FC) vehicles that are equipped with polymer electrolyte fuel cells (PEFCs) and use hydrogen gas as the fuel have received considerable attention in recent years due to their lack of harmful emissions. The performance of PEFC systems is dependent on membrane-electrode assembly materials. The PEFC has been typically operated under humidified conditions using humidifying systems at 60–80 °C. Operation under low humidity could simplify the humidifying systems and, in the end, reduce the cost of FC systems. The proton conductivity of conventional polymer electrolyte membranes, however, is so low under low humidity. The main issue in the development of polymer electrolyte membranes is the improvement of proton conductivity under low humidity, and the electrolyte membranes should have high acid density (low EW) and strong acidity to form continuous proton conduction pathways. Perfluorosulfonic acid electrolytes meet these requirements because of the super acidity derived from the electronegativity of fluorine. However, perfluorosulfonic acid membranes have some issues with respect to high processing costs and a low glass transition temperature.¹

Many studies have been conducted on aromatic hydrocarbon polymer electrolytes as alternative membranes, which consist of hydrophilic hydrocarbon segments and hydrophobic hydrocarbon segments.²⁻²⁴ For hydrocarbon electrolytes to exhibit high proton conductivity, it has been reported that a block copolymer that consists of a regular arrangement of hydrophobic and hydrophilic segments maintains significantly higher proton conductivity under low humidified conditions than that which consists of a random arrangement of the segments.^{2−4,9,10,18} However, hydrocarbon-hydrophilic segments with sulfonic acids exhibit weaker acidity than perfluorosulfonic acids, and there have been a few reports on embedded perfluorosulfonic acids or perfluorosulfonimide acids as side chains of hydrophilic segments.²⁵⁻²⁸ Poly(perfluoroalkyl sulfonimides) (PCSIs) are promising candidates as the hydrophilic segment of block copolymers. Hu and DesMarteau reported the synthesis and properties of PCSIs, and the advantage of PCSIs in the control of equivalent weight (EW) by changing the alkyl chain length, and synthesized a polymer with a number of sulfonimide repetitions as intended.²⁹⁻³¹ There has been no report of the incorporation of segments with highly repetitive sulfonimide into main chains of hydrophilic segments. We have focused on poly(perfluoropropyl sulfonimide) (PC3SI), which should satisfy the requirements of a very low EW (293 g/equiv) together with super acidity similar to that of perfluorosulfonic acid. For hydrophobic segments, poly(ether ether sulfone) (PEES) was selected because of its ease of synthesis of hydrocarbon polymers.

For the block copolymer to achieve high proton conductivity, the control of block length in hydrophilic segments and hydrophobic segments is also important. MacGrath and co-worker have reported that proton conductivity of block copolymer membranes increased as the block length increased due to the sharply developed microphase-separated structure.⁴

In this paper, we report studies on novel fluorocarbon–hydrocarbon hybrid block copolymer electrolyte membranes including the synthesis of telechelic hydrophilic and hydrophobic segments by polycondensation (Scheme 1) and the properties of the obtained electrolyte membrane with different block lengths of j and k. The microphase-separated structure of the block copolymers may change not only by the block lengths of j and k but also by the ratio (j/k). In this paper, the effects of block lengths on the properties were investigated with j/k kept constant.

Scheme 1. Synthesis of Perfluorocarbon–Hydrocarbon Hybrid Block Copolymer Electrolyte.

Experimental Section

Materials

Perfluoropropyl-1,3-bissulfonylfluoride (C3F) was purchased from Daikin Industries, Ltd. (Osaka, Japan) and used as received. Lithium bromide was purchased from Sigma-Aldrich and used as received. 4-Fluorobenzenesulfonyl chloride, sodium hydroxide, potassium hydroxide, potassium carbonate, magnesium sulfate, diethyl ether (Et₂O), ethanol (EtOH), chloroform (CH₃CI), dimethyl sulfoxide (DMSO), DMSO-d₆, acetonitrile-d₃ (MeCN-d₃), anhydrous acetonitrile (MeCN), anhydrous N,N-dimethylacetamide (DMAc), N-methyl-2-pyrrolidone (NMP), p-cresol, 1,4-dioxane, hexane, and tetrahydrofuran (THF) were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan) and used as received. N,N-Diisopropylethylamine (DIPEA) was also purchased from Wako Pure Chemical Industries, Ltd., distilled with CaH₂, and stored under a nitrogen atmosphere. Anhydrous toluene was purchased from Kanto Chemical Co. (Tokyo, Japan) and used as received. 4,4′-Bisphenol, 4,4′-dichlorodiphenyl sulfone and 1 M lithium hexamethyldisilazide (LiHMDS) solution in THF were purchased from Tokyo Kasei Co. (Tokyo, Japan). 4,4′-Bisphenol and 4,4′-dichlorodiphenyl sulfone were recrystallized from 1,4-dioxane/hexane (5:2, v/v) and THF/hexane (1:1, v/v), respectively, before use.

Synthesis of Perfluoropropyl-1,3-bissulfonamide (C3A)

1 M LiHMDS solution in THF (66 mL, 66 mmol) was added to a 100 mL two-necked round-bottomed flask under a N₂ atmosphere, and then, 9.48 g (30 mmol) of C3F was slowly added at −80 °C. The mixture was stirred for 1 h at −80 °C and then for 12 h at room temperature. After removal of the volatiles, the remaining solid was dissolved in 1 N HCl solution and extracted with 300 mL of Et₂O. The organic phase was dried with anhydrous MgSO₄ and then stirred with activated carbon for 20 min until the color of the solution disappeared. The colorless solution was evaporated to give C3A as white powder (7.6 g, 82%). ¹⁹F nuclear magnetic resonance (NMR) (MeCN-d₃, 500 MHz): δ −118.28 (s, 2F), −112.84 (s, 4F).

Synthesis of PC3SI

C3F 2.50 g (7.92 mmol), C3A 2.48 g (8 mmol), DIPEA 6.13 mL (40 mmol), and anhydrous MeCN (8 mL) were added to a 30 mL Schlenk flask and stirred under a N₂ atmosphere at 80 °C for 2 days. The volatile contents were removed by N₂ flow, and the remaining oily product was dissolved in 50 mL of l N NaOH aqueous solution. The alkaline solution was bubbled with N₂ gas to remove DIPEA. The solution was then washed with Et₂O to remove impurities, and the Na salt was extracted with MeCN. The organic phase was evaporated, and the residue was dried under vacuum at 60 °C overnight to give the Na salt. The salt was acidified with 100 mL of 1 N HCl solution, and the acidic solution was washed again with Et₂O to remove impurities. This ether washing caused three-phase separation, the top layer was aqueous solution, the middle layer was Et₂O including impurities, and the bottom layer was Et₂O solution including PC3SI. The bottom layer was separated, and the residue was dried under vacuum at 80 °C to give PC3SI as pale yellow powder (5.0 g, 85%, i = 25, Scheme 2).

Scheme 2. Chemical Structure of PC3SI.

Synthesis of F-Ph-Terminated Telechelic Poly(perfluoropropyl sulfonimide) Potassium Salt (1 in Scheme 1)

F-Ph-terminated telechelic poly(perfluoropropyl sulfonimide) was synthesized with the comonomer compositions shown in Table 1. The polymerization is described as follows for run 1 in Table 1. C3F (13.8 g, 43.6 mmol), C3A (20.3 g, 65.4 mmol), and anhydrous MeCN (40 mL) were added to a Schlenk flask equipped with a PTFE stopcock, under a N₂ atmosphere. DIPEA (38 mL, 0.22 mol) was then added to the flask, and the solution was stirred at 80 °C for 2 days. After evaporation of the volatiles, 1 N NaOH (aqueous) was added to the remaining mixture, and the mixture was extracted with MeCN. The organic phase was evaporated to half the volume, and 1 N HCl (aqueous) was added for acidification, after which the solvents were evaporated. The residue was extracted with MeCN to remove NaCl by decantation. The extracted solution was then vacuum-dried, and the remaining solid was dissolved with a small amount of MeCN, followed by filtration of the solution with Celite. After the removal of MeCN, PC3SI was quantitatively obtained (37.0 g) after drying under vacuum.

Table 1. Reaction Conditions of F-Ph-Terminated PC3SI.

polycondensation						cap reaction
run	feed j	C3F (g)	C3A (g)	DIPEA (mL)	MeCN (mL)	PC3SI (g)	FPhSO2Cl (g)	DIPEA (mL)	MeCN (mL)
1	5	13.8	20.3	38	40	16.9	5.33	20	20
2	15	18.1	20.3	50	20	39.3	5.00	30	30

PC3SI (16.9 g 11.4 mmol), 4-fluorobenzenesulfonyl chloride (5.33 g, 27.4 mmol), and anhydrous MeCN (20 mL) were added to a Schlenk flask equipped with a PTFE stopcock. DIPEA (20 mL, 0.11 mol) was added to the flask under a N₂ atmosphere, and the mixture was stirred at 80 °C overnight. After removal of the volatiles, small amounts of MeCN and 1 N KOH (aqueous) were added to the remaining solution, and the aqueous phase was removed by decantation. This procedure was performed five times, and the organic solvent was then evaporated. The remaining solid was dissolved in MeCN, and the solution was filtered with Celite to remove salts. The F-Ph-terminated PC3SI potassium salt was obtained (20.4 g, 59%, j = 8.6) after drying under vacuum.

Synthesis of OH-Ph-Terminated Telechelic PEES (2 in Scheme 1)

OH-Ph-terminated telechelic PEES was synthesized with the comonomer compositions shown in Table 2. The polymerization is described as follows for run 2 in Table 2. 4,4′-Bisphenol (8.56 g, 45.9 mmol) and K₂CO₃ (25.4 g, 184 mmol) were added to a 300 mL three-necked round-bottomed flask equipped with a condenser, a Dean–Stark apparatus, and a 3-way stopcock, and the mixture was then dried under vacuum for 2 h at room temperature. Anhydrous DMAc (65 mL) and toluene (65 mL) were added to the flask under a N₂ atmosphere, and the solution was refluxed at 120 °C for 2 h. The reaction temperature was gradually increased to 150 °C, while azeotropic water was removed with the Dean–Stark apparatus. After cooling naturally to room temperature, 4,4′-chlorodiphenyl sulfone (12.0 g, 41.8 mmol) was added to the flask, and the solution was stirred at 165 °C for 24 h. The excess K₂CO₃ and KF byproducts were then filtered out from the reaction solution. The filtrate was reprecipitated in EtOH, and the precipitate was retrieved by centrifugation. The target compound was obtained after drying under vacuum overnight (15.4 g, 88%, M_n = 6.0×10³, determined from NMR spectra, k = 15).

Table 2. Reaction Conditions of OH-Ph-Terminated Telechelic PEES.

run	feed k	4,4′-dichlorodiphenyl sulfone (g)	4,4′-bisphenol (g)	K2CO3 (g)	DMAc (mL)	toluene (mL)
1	4	12.0	9.73	28.9	65	65
2	10	12.0	8.56	25.4	65	65

Synthesis of the Cl-Ph-Terminated Hydrophobic Segment (for Polymer Structural Analysis)

4,4′-Bisphenol (1.3 g, 7.0 mmol), 4,4′-dichlorodiphenyl sulfone (6.0 g, 21 mmol), and K₂CO₃ (0.97 g, 21 mmol) were added to a 100 mL three-necked round-bottomed flask equipped with a condenser, a Dean–Stark apparatus, and a 3-way stopcock, and the mixture was dried under vacuum at room temperature for 2 h. Anhydrous DMAc (28 mL) and toluene (28 mL) were added to the flask under a N₂ atmosphere, and the solution was refluxed at 120 °C for 2 h. The reaction temperature was gradually increased to 150 °C, while azeotropic water was removed with the Dean–Stark apparatus. The solution was kept at 165 °C for 24 h. The excess K₂CO₃ and KF byproducts were then filtered out from the reaction solution. The filtrate was reprecipitated in EtOH, and the precipitate was retrieved by centrifugation. The target compound was obtained after drying at 60 °C under vacuum overnight (4.7 g, 99%).

Synthesis of the Targeted Block Copolymer

F-Ph-terminated telechelic poly(perfluoropropylsulfonimide) potassium salt (run 1 in Table 1; 18.2 g, F-Ph 11 mmol), OH-Ph terminated PEES (run 1 in Table 2; 15.9 g, OH-Ph 11 mmol), and K₂CO₃ (3.06 g, 22 mmol) were added to a 1 L three-necked round-bottomed flask equipped with a condenser, a Dean–Stark apparatus, and a 3-way stopcock, and the mixture was dried at room temperature under vacuum for 2 h. Anhydrous DMAc (300 mL) and toluene (250 mL) were added to the flask under a N₂ atmosphere, and the solution was refluxed at 120 °C for 2 h. The reaction temperature was gradually increased to 150 °C, while azeotropic water was removed with the Dean–Stark apparatus. The solution was stirred and kept at 165 °C for 2 weeks. The excess K₂CO₃ and KF byproducts were then filtered from the reaction solution. The filtrate was reprecipitated in 1 N HCl aq, and the precipitate was retrieved by centrifugation. After the obtained precipitate was dried under vacuum overnight and dissolved in MeCN, the insoluble part was filtered, and the filtrate was evaporated to dryness (30 g, 89%). The dried copolymer (0.5 g) was dissolved in MeCN (5 g), and CHCl₃ (45 g) was added to the solution. The obtained precipitate was recovered by centrifugation, and the target compound was obtained after drying at 60 °C under vacuum overnight (0.28 g, 56%).

The F-Ph-terminated telechelic poly(perfluoropropylsulfonimide) potassium salt (run 2 in Table 1; 19.0 g, F-Ph 5.05 mmol), OH-Ph-terminated telechelic PEES (run 2 in Table 2; 15.2 g, OH-Ph 5.05 mmol), and K₂CO₃ (1.40 g, 10.1 mmol) were used to obtain another copolymer (34 g, 87%) according to the same above procedure. The obtained copolymer (0.5 g) was dissolved in MeCN (5 g), and 1,4-dioxane (25 g) was added to the solution. The resultant precipitate was recovered by centrifugation, and the target compound was obtained after drying at 60 °C under vacuum overnight (0.32 g, 63%).

Film Formation Method

The obtained block copolymer (0.24 g) was dissolved in a mixture of NMP and p-cresol (1 mL/1 mL) at room temperature, and the solution was cast onto a flat Petri dish (45 mm diameter). The cast film was dried at 80 °C and annealed at 100 °C for 1 h. The cast membrane was separated from the Petri dish and dried under vacuum overnight at 60 °C. The membrane was immersed in 1 N H₂SO₄ aqueous solution at 60 °C overnight and then washed with pure water until the pH of the water became neutral. The membrane was dried at 60 °C overnight.

Measurements

¹H NMR (500 MHz) and ¹⁹F NMR (500 MHz) measurements were conducted using samples dissolved in MeCN-d₃ or dimethyl sulfoxide (DMSO-d₆) on a JEOL JMTC-400/54/SS spectrometer.

Molecular weight and molecular weight distributions (M_w/M_n) were determined using gel permeation chromatography (GPC; Tosoh HPC-8120GPC) with a UV detector (Tosoh UV-820; 270 nm) and a TSK gel α-M column (Tosoh) with DMSO (50 mM LiBr) as the eluent at a flow rate of 0.5 mL/min at 40 °C. A molecular weight calibration curve was obtained using a polystyrene standard (Tosoh).

Transmission electron microscopy (TEM) observations were conducted using ultrathin sections of the membranes. The samples were embedded with epoxy resin and then cross-sectioned with an ultra-microtome (Leica ULTRACUT-S) at 20 °C to obtain 50 nm-thick sections. Bright-field images of the ultrathin sections were obtained using TEM (JEOL JEM-2000EX) at an acceleration voltage of 200 kV.

Evaluation of Membrane Properties

Water Uptake

The weight of the membrane after immersion in water at room temperature (W_H2O) and that of the membrane after drying overnight at 80 °C (W_dry) were measured. The water uptake was calculated using formula 1

1

Swelling Ratio

The size in plain and through plain of the membrane after immersion in water at room temperature (L_H2O) and that of the membrane after drying overnight at 80 °C (L_dry) were measured. The swelling ratio was calculated using formula 2

2

Proton Conductivity

The obtained membrane was set on a pair of platinum electrodes at a distance of 0.4 cm. The resistance of the membrane was measured using a chemical impedance meter (Hioki 3532-80) over the frequency range of 50 Hz to 1 MHz. The proton conductivity (σ) was calculated from the membrane resistance (R) and the distance between the electrodes (L = 0.4 cm) using the membrane thickness (T) and width (W) under the given measurement conditions using formula 3

3

Equivalent Weight

The obtained membrane was dried at 80 °C under vacuum overnight, and the weight of the dried membrane was measured. The dried membrane was immersed in 10 wt % NaCl aqueous solution (30 mL) at room temperature overnight to obtain an acidic aqueous solution for titration. After immersion, the membrane was removed from the solution and rinsed with water. The solution was diluted to a total volume of 80 mL with water as a titration sample. The titration was performed using an automatic titrator (GT-100, Mitsubishi Chemical Corporation) with 0.05 mol/L NaOH aqueous solution. The EW was calculated from the weight of the dried membrane (W_dry), the dripped volume of the basic aqueous solution (V_NaOH), and the concentration of the NaOH aqueous solution (C_NaOH) using formula 4

4

Results and Discussion

Synthesis of PC3SI and F-Ph-Terminated Telechelic Poly(perfluoropropyl sulfonimide) Potassium Salt (Hydrophilic Segment)

Table 3 shows the results for the synthesis of hydrophilic segments. The progress in these syntheses was confirmed by ¹⁹F NMR, as follows. Figure 1 shows the ¹⁹F NMR spectra for the obtained PC3SI and hydrophilic segments. In Figure 1a, the peaks that originate from perfluoropropyl groups [−112.4 ppm (−CF₂CF₂CF₂−), 118.4 ppm (−CF₂CF₂CF₂−)] and the peak that originate from the SO₂NH₂ end groups [−113.0 ppm (−CF₂CF₂CF₂–SO₂NH₂)] were observed, which indicates that the target PC3SI compound was obtained. In Figure 1b,c, the peak derived from SO₂NH₂ end groups [−113.0 ppm (−CF₂CF₂CF₂–SO₂NH₂)] was not observed, while the peak derived from the Ph-F end groups −109.5 ppm (−Ph–F) was observed, which indicates that the target hydrophilic segments were obtained. C3F was hydrolyzed by trace moisture in the reaction system when condensation between C3F and C3A proceeded, as evidenced by the peak for CF₂CF₂CF₂–SO₃H (−114.1 ppm). Implementation values of F-Ph end groups, which was calculated from the ratio of the integral values of F-Ph end groups and SO₃H end groups in the ¹⁹F NMR spectrum, were 98% (run 1) and 95% (run 2). The degree of polymerization, which was obtained from the ratio of the integral value of polymer terminals and repeat units in the ¹⁹F NMR spectrum, was higher than that of the feed. This is most likely because the lower molecular weight oligomers were removed, and the higher molecular weight polymers were recovered by purification.

Table 3. Results for Obtained F-Ph-Terminated Telechelic PC3SIs.

run	feed j	yield %	obtained ja	F-Ph (%)	Mna (×103)
1	5	59	8.6	98	2.9
2	15	51	21	95	6.3

^aCalculated from ¹⁹F NMR.

Figure 1.

¹⁹F NMR spectra for obtained PC3SI [(a); i = 25] and F-Ph-terminated hydrophilic segments [(b); j = 8.6, (c); j = 21] in DMSO-d₆, at RT.

Synthesis of OH-Ph-Terminated Telechelic PEES (Hydrophobic Segment)

Table 4 shows the results for the synthesis of the hydrophobic segments. The progress in these syntheses was confirmed by ¹H NMR, as follows. Figure 2 shows ¹H NMR spectra for the hydrophobic segments. In the spectra for the Cl-Ph-terminated model compound and 4,4′-bisphenol, Cl-Ph was not observed in the end groups of the hydrophobic segments, which indicates that implementation of the OH-Ph end groups was 100%. The degree of polymerization for the hydrophobic segments, which was calculated from the ratio of the integral value of polymer terminals (H_g) and repeat units (H_a, H_b, and H_c) in the ¹H NMR spectra, was 1.5 times that of the feed. This is also likely because the lower molecular weight oligomers were removed, and the higher molecular weight polymers were recovered by purification.

Table 4. Results for Obtained OH-Ph-Terminated Telechelic PEESs.

run	feed k	yield %	obtained ja	Mna (×103)	Mnb (×103)	Mwb (×103)	Mw/Mnb
1	4	81	6.7	2.9	6.7	12	1.8
2	10	88	15	6.0	16	33	1.9

^aCalculated from ¹H NMR.

^bCalculated from GPC (PSt standard).

Figure 2.

¹H NMR spectra of 4,4′-bisphenol (a), OH-Ph terminated PEES [run 1 (b) and run 2 (c) in Table 4] and C1-Ph-terminated model compound (d) in DMSO-d₆ at RT.

Synthesis of the Hydrophilic–Hydrophobic Block Copolymer

Table 5 shows the results for the block copolymerization. The charged j/k values were set so that the ratio of the hydrophilic segments to the hydrophobic segments in the block copolymers was almost equal. The progress in the reaction between the obtained hydrophilic and hydrophobic segments was confirmed by ¹⁹F and ¹H NMR, as follows. Figure 3 shows ¹⁹F and ¹H NMR spectra for the obtained block copolymers (run 2 in Table 5). The peaks derived from the F-Ph end groups of the hydrophilic segments (−109.5 ppm in ¹⁹F NMR spectrum) and the end groups of the hydrophobic segments [H_f (7.46 ppm) and H_g (6.85 ppm) in ¹H NMR spectrum] disappeared; these NMR results confirmed that the block copolymerization was successful.

Table 5. Results for Block Copolymerization.

		Feed				obtained
run	block copolymer	F-Ph PC3SI Ja	HO-Ph PEES kb	j/k	EWc (g equiv–1)	yield (%)	EWd (g equiv–1)	Mne (×103)	Mwe (×103)	Mw/Mne
1	1	8.6	6.7	1.28	592	56	510	2.6	6.4	2.5
2	2	21	15	1.40	574	63	558	3.3	6.0	2.3

^aCalculated from ¹⁹F NMR.

^bCalculated from ¹H NMR.

^cTheoretical.

^dCalculated from titration (NaCl method).

^eCalculated from GPC.

Figure 3.

¹⁹F NMR (upper) and ¹H NMR (lower) spectra for the obtained block copolymer (run 2 in Table 5) in DMSO-d₆ at RT. Asterisk peak is unknown.

Figure 4 shows the ratio of the remaining F-Ph groups in the copolymerization reaction to the feed and the molecular weight of each as a function of the reaction time (run 1 in Table 5). It took 2 weeks to consume all the F-Ph groups because the collision frequency between end groups of the hydrophilic segments and hydrophobic segments in this copolymerization was much lower than in typical reactions between low molecular weight compounds.

Figure 4.

(a) [F-Ph]/[F-Ph]₀ and (b) M_n as a function of reaction time.

Figure 5 shows size-exclusion chromatography (SEC) profiles for the purified block copolymer [(a) run 1 in Table 5] and the hydrophobic segments before reaction [(b) run 1 in Table 4]. The peak top of (b) was shifted to that of (a) for high molecular weight; therefore, it was also confirmed by SEC that the block copolymerization was successful. The profile for the block copolymer was monomodal, and the molecular weight distribution (M_w/M_n) of approximately 2.5 was larger than that of the theoretical value of 2 for the polycondensation polymer; therefore, multi block copolymers with various degrees of polymerization were obtained.

Figure 5.

SEC curves for the obtained copolymer [(a) run 1 in Table 5] and the obtained OH-Ph-terminated PEES [(b) run 1 in Table 4].

The EW values, 510 and 558, of the obtained block copolymers in runs 1 and 2 in Table 5, respectively, were smaller than the theoretical EW values of 592 and 574 of the feed. This is because of the purification process; in the purification step by reprecipitation of the block copolymers, the EW values in runs 1 and 2 for the block copolymer dissolved in the top clear layer were higher at 782 and 610 g/equiv, respectively. Therefore, the block copolymers recovered as a precipitate had lower EWs.

Evaluation of Physical Properties

The obtained block copolymer was dissolved in a mixture of NMP and p-cresol at room temperature, and the solution was cast onto a flat Petri dish, dried at 80 °C, and then annealed at 100 °C. After immersion in 1 N H₂SO₄ aqueous solution, the membrane was washed with pure water. Table 6 shows the physical properties of the obtained block copolymers. Phase separation of hydrophilic and hydrophobic segments in the membrane can affect properties such as the water uptake and proton conductivity. Figure 6 shows a TEM image of a section of the block copolymer (run 2 in Table 6) membrane. Dark parts and clear parts are hydrophilic segments and hydrophobic segments, respectively. Dark parts were partially continuous at the nanometer level and, therefore, the hydrophilic segments formed continuous proton conductive pathways to provide high proton conductivity. Although the phase separation occurred, the contrast between hydrophilic segments and hydrophobic segments was not clear, and the diameter of ion channels (a few nanometers) was smaller than that of the previously reported electrolyte membranes (tens of nanometers).^3,5,6 Watanabe et al. have reported the relationship between the phase separation of block copolymer electrolytes and block length, in which the contrast becomes clearer and the diameter becomes thicker as the block length increases.⁵ Using long block length, the similar effects can be obtained in this block copolymer system. The block copolymer, whose block length of j was greater than 21, was not obtained because the implementation of F-Ph end groups of hydrophilic segments was too low to copolymerize.

Table 6. Properties of the Obtained Block Copolymers.

						swelling ratio
run	electrolyte	EWa (g/equiv)	densityb (g/cm3)	NHc (mmol/cm3)	water uptake (%)	in-plane (%)	through-plane (%)
1	block copolymer 1	510	2.31	4.5	113	44	34
2	block copolymer 2	558	2.27	4.1	153	36	25
3	Nafion112	1075	2.03	1.9	22	14	13
4	S-PEEK	418	1.32	3.1	165	54	16

^aCalculated from titration (NaCl method).

^bCalculated from dry membrane weight and dimensions.

^cProton density (N_H = density/EW×1000).

Figure 6.

TEM image of the obtained block copolymer (run 2 in Table 6).

Despite of lower N_H (proton density) of block copolymer 2, the water uptake was higher than that of block copolymer 1. Calculating the ratio of the block length of the hydrophilic segments (j) to that of the hydrophobic segments (k) in Table 5, the ratio of block copolymer 2 (j/k = 1.40) is higher than that of block copolymer 1 (j/k = 1.28). The reason for high water uptake of the block copolymer 2 was that the hydrophobic segment had no enough length to sufficiently suppress the hydrophilic segment from absorbing water.

Although the obtained block copolymers (runs 1 and 2 in Table 6) were insoluble in hot water at 80 °C, the water uptake was extremely high (more than 100%) as with S-PEEK [sulfonated-poly(ether ether ketone)]. This suggests that the hydrophobic segments formed were not effectively continuous to resist swelling of the hydrophilic segments with water.

The swelling ratio of block copolymers 1 and 2 was anisotropic in comparison to Nafion 112. Both the swelling ratio in-plane and through-plane was decreased with increasing the block length. Long block length may be effective in controlling of the swelling ratio.

Figure 7 shows the dependency of the proton conductivity on the humidity for PC3SI (i = 25) and the block copolymer membrane (runs 1 and 2 in Table 6). The conductivity of PC3SI at 20% relative humidity (RH) was 1.2 × 10^–2 S/cm, which is approximately 10 times as high as that of Nafion 112. The conductivity of PC3SI at more than 50% RH was not obtained because the membrane was dissolved. The conductivity of the block copolymer decreased when humidity became low, but did not decrease suddenly as with the hydrocarbon S-PEEK, Figure 7e, which has randomly arranged sulfonic acid groups. The conductivity of the block copolymer at 20% RH was 1.7 × 10^–3 S/cm, which is approximately three times that of Nafion 112 and approximately 930 times that of S-PEEK. Although EW of the block copolymer 2 was higher than that of block copolymer 1, the conductivity of 2 was higher than that of 1. It is expected that the clearer phase-separated structure and more continuous proton conduction pathways of block copolymer 2 are formed because of long block lengths.

Figure 7.

Proton conductivity of PC3SI [(a) EW 293 g/equiv], the obtained block copolymer [(b) EW558 g/equiv run 2 in Table 6, (c) EW510 g/equiv run 1 in Table 6], Nafion112 [(d) EW1075 g/equiv], and S-PEEK [(e); EW 418 g/equiv].

According to the Nernst–Einstein equation, the proton conductivity is proportional to the proton diffusion coefficient and proton density:

5

where σ is the conductivity, Z is the charge of a proton, F is the Faraday constant, D_σ is the proton diffusion coefficient, N_H is the proton density, R is the gas constant, and T is the temperature.

N_H for the block copolymer was approximately twice that for Nafion 112 (Table 6). D_σ at 20% RH for the block copolymer was calculated from eq 5 and was approximately 1.5 times as high as that for Nafion 112, which indicates that continuous proton conductive pathways were formed in the block copolymer, even at low humidity. The high proton conductivity of the block copolymer is attributed to both the low EW and the continuous proton conductive pathways.

Conclusions

Novel fluorocarbon and hydrocarbon hybrid block copolymer electrolytes that consist of poly(perfluoropropyl sulfonimide) as hydrophilic segments and PEES as hydrophobic segments were prepared. The membranes of the obtained block copolymers had insufficient microphase-separated structures but had partially continuous proton conduction paths. The proton conductivity of the block copolymer did not suddenly decrease with a decrease in the humidity and was 1.7 × 10^–3 S/cm at 20% RH, which is approximately three times as high as that for Nafion 112. The proton diffusivity for the block copolymer was approximately 1.5 times of Nafion 112 because of the formation of continuous proton conductive pathways at low humidity.

The authors declare no competing financial interest.