Membrane Formation and Modification - American Chemical Society

Chapter 11

Polyphosphazene-Based Cation-Exchange Membranes: Polymer Manipulation and Membrane Fabrication 1

2

Qunhui Guo , Hao Tang , Peter N. Pintauro

2,3

and Sally O'Connor

1

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on June 4, 2015 | http://pubs.acs.org Publication Date: December 29, 1999 | doi: 10.1021/bk-2000-0744.ch011

1

Department of Chemistry, Xavier University, New Orleans, L A 70125 Department of Chemical Engineering, Tulane University, New Orleans, L A 70118 2

Poly[bis(3-methylphenoxy)phosphazene] was sulfonated in solution with S 0 and solution-cast into ion-exchange membranes from Ν,Ν-dimethylacetamide. Water insoluble membranes were prepared with an ion-exchange capacity (IEC) as high as 2.1 mmol/g. For water insoluble polymers with an IEC < 1.92 mmol/g, there was no evidence of polymer degradation during sulfonation. The glass transition temperature of the sulfonated polymer increased from -28°C (for the base polymer) to -10°C for an IEC of 2.1 mmol/g. Equilibrium water swelling of a phosphazene membrane with an IEC of 0.95 mmol/g was 24% greater than that of a DuPont Nafion 117 cation-exchange membrane. The proton conductivity of a water-equilibrated 0.95 mmol/g IEC phosphazene membrane in the H form ranged from 0.012 S/cm at 25°C to 0.058 S/cm at 60°C. The water diffusion coefficient in a 0.95 mmol/g IEC membrane, at saturated vapor conditions, ranged from 8.0 x 10 cm /s at 25°C to 4.1 x 10 cm /s at 60°C. 3

+

-8

2

-7

2

Polyphosphazenes are a potentially useful class of base-polymers for ion-exchange membranes because of their reported thermal and chemical stability and the ease of chemically altering the polymer by adding various sidechains onto the -P=Nbackbone. Sulfonated polyphosphazene cation-exchange membranes, for example, may be an attractive alternative to perfluorosulfonic acid and polystyrene sulfonate membranes. The difficulty associated with producing such membranes lies in preparing the sulfonated polyphosphazene and, more importantly, in balancing the resulting hydrophilicity of the polymer to prevent dissolution in aqueous solutions. In practice, there are three synthesis routes leading to water insoluble sulfonic acid membranes from polyphosphazene polymers. The first method involves crosslinking polyphosphazene membranes followed by heterogeneous sulfonation Corresponding author.

162

© 2000 American Chemical Society

In Membrane Formation and Modification; Pinnau, I., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.


163 (/). In a second technique, homogeneous or heterogeneous polymer sulfonation is carried out, followed by film casting and crosslinking. These two methods require that the polyphosphazene contains sidegroups that can be used for crosslinking or in the second case, sulfonic groups can serve the purpose (2). The third route to a sulfonated polyphosphazene ion-exchange membrane is through appropriate balancing the hydrophilicity/hydrophobicity and, i f possible, crystallinity of the polymer. In this case, a hydrophobic and/or semicrystalline polyphosphazene polymer is sulfonated to such an extent that it only swells and does not dissolve in aqueous media. This imposes some limitations on the magnitude of the ion-exchange capacity of the final membrane but, at the same time, offers significant processing advantages. There are several reports in the literature on the sulfonation of phosphazene polymers although none of these studies was directed at fabricating an ion-exchange membrane nor was the structure of the sulfonated polyphosphazenes examined in detail. The sulfonation of aryloxy- and (arylamino)phosphazenes via reaction with sulfuric acid was studied by Allcock et al. (3) while the reaction of (aryloxy)polyphosphazenes with sulfur trioxide was studied by Montoneri and co workers (4,5). During the initial stages of S 0 addition, Montoneri et al. found no C-sulfonation, but rather the formation of a sN—>SOs complex. When the S0 /repeating monomer molar ratio was > 1, C-sulfonation was observed. A comprehensive review of the literature on sulfonated phosphazene polymers can be found elsewhere (6). In a previous study by Wycisk and Pintauro (d), two semi-crystalline phosphazene polymers, poly[(3-methylphenoxy)(phenoxy)phosphazene] and poly[(4-methylphenoxy)(phenoxy) phosphazene], with an alkylphenoxy/phenoxy molar ratio of 1.0, were sulfonated successfully to varying degrees with SO3. Water-insoluble cation-exchange membranes with good mechanical properties were fabricated with an ion-exchange capacity as high as 2.0 mmol/g. Polyphosphazenes with an ion-exchange capacity in the range of 2-3 mmol/g were also synthesized, but these polymers were found to be water soluble. Preliminary C N M R and U N M R analyses indicated that the methylphenoxy side groups were sulfonated preferentially. When poly[(4-ethylphenoxy)(phenoxy) phosphazene] was sulfonated with SO3, the sulfonation rate was very slow (due to both steric interference by the ethyl substituents on the ethylphenoxy side groups and the poor reactivity of the phenoxy groups) and significant polymer degradation was observed. In a more recent paper, Graves and Pintauro (7) examined the U V photocrosslinking of alkylphenoxy/phenoxy-substituted polyphosphazenes. Solution-cast films contained poly[(methylphenoxy) (phenoxy)phosphazene], poly[(ethylphenoxy)(phenoxy)phosphazene], or poly[(isopropylphenoxy) (phenoxy)phosphazene] (where the alkyl substituent was in either the meta or para position) and benzophenone photo-initiator (at a concentration of 1-25 mol%). Crosslinking was carried out at either 25°C or 70°C. The methylphenoxy/phenoxy phosphazenes were the best materials for crosslinking, indicating that steric effects of the alkyl group were playing a role during crosslink formation. For U V light3

3

13

l


164 exposed poly[(3-methylphenoxy)(phenoxy)phosphazene] films, the glass transition temperature increased by approximately 25°C (from -15°C to 10°C) and the film swelling (in DMAc) decreased from infinity (complete solubilization) to 25% as the benzophenone concentration was increased from 0 to 25 mol%. It is obvious from previous work that methyl substituents on the phenoxy side chains of the phosphazene polymer were activating the ring for electrophilic attack by S 0 and were the optimum alkyl group for U V photo-crosslinking (6,7). In the present paper, we report on the sulfonation of poly[bis(3-methyl phenoxy) phosphazene], a semi-crystalline polymer with only methylphenoxy side groups. The chemical structure of the repeating polymer unit is shown in Figure 1. The method of sulfonation and the physical and ion-exchange properties of the resulting non-crosslinked sulfonated polymer membranes are described below.


3

Experimental Poly[bis(3-methylphenoxy)phosphazene], purchased from Technically Inc., Andover, M A , was used as the base polymer without further purification. The molecular weight of this polyphosphazene, as determined by gel permeation chromatography (Waters Styragel H T 6E column in THF with polystyrene calibration standards), was 2.0 χ 10 g/mol. To sulfonate the polyphosphazene, a known weight of polymer (1.0 g) was first dissolved in 40 ml of 1,2-dichloroethane (DCE) and stirred for 24 h at 50°C. A given amount of S 0 in 10 ml of D C E was then added dropwise to the polymer solution in a dry nitrogen atmosphere. The resulting precipitate was stirred for 3 h at 0°C followed by the addition of 50 ml of a NaOH solution (water/methanol solvent) to terminate the reaction. After evaporation of solvent at 70°C for 24 h, the polymer was pre-conditioned by soaking sequentially in distilled water, 0.1 M NaOH, distilled water, 0.1 M HC1, and distilled water (each soaking step was carried out for 48 hours). The polymer product was then dried thoroughly and dissolved in N , N-dimethylacetamide (DMAc) at a concentration of 5 wt%. Membranes were cast from this solution on a polypropylene plate and then dried at 70°C for 3 days (until there was no change in the film weight). The thickness of the dry membranes was between 200 and 600 μηι. The ion-exchange capacity (IEC, with unit of mmol/g of dry polymer) of sulfonated polyphosphazene membranes was determined by measuring the concentration of H * that exchanged with N a when membrane samples were equilibrated with a NaCl solution. A known weight of dry polymer (0.2-0.4 g) in the acid form (after pre-conditioning) was placed into 100 ml of a 2.0 M NaCl solution and shaken occasionally for 48 h. Three 25 ml samples were then removed and the amount of H released by the polymer was determined by titration with 0.01 M NaOH. Equilibrium water swelling of sulfonated membrane samples at room temperature was measured in terms of the % increase in dry membrane weight, according to the following formula, 6

3

+

+


165 Swelling = ( m

wet

- m )/m χ 100 (%) dry

(1)

dry

where m and are the weights of the water swollen and dry membrane (in its reform), respectively. The diffusion coefficient of water in sulfonated polyphosphazene films at saturated vapor conditions was detennined by performing vapor-phase sorption/desorption experiments with a McBain sorption apparatus (8,9). A small membrane specimen of known dry weight was suspended on a quartz spring (enclosed in a thermostated glass chamber) and allowed to equilibrate with a water vapor atmosphere of activity 0.98 (an activity < 1.0 was needed to avoid water condensation). After the membrane was fully equilibrated and absorbed no more water, the vapor pressure of water was lowered by evacuating the chamber and the decrease in membrane weight was monitored as a function of time. A typical water desorption curve for the polyphosphazene membranes is shown in Figure 2, where the weight loss is plotted as a function of the square-root of time. In this figure M is the equilibrium, zero time, weight of water in the membrane sample and M is the water weight remaining in the membrane at a given time after desorption. The water diffusion coefficient was computed from the initial slope of the desorption curve and the following equation (10),


w e t

0

t

2

2

D = n6 (slope) /16

(2)

where δ is the thickness of the dry membrane. This method has been used previously to determine the diffusion coefficient of penetrants in polymers (11,12) and is accurate when the membrane is sufficiently thick so that errors due to polymer shrinkage at the membrane surface (where the water swelling is small) are minimized. Also, the membrane thickness must be much smaller than the sample width and length to ensure one-dimensional diffusion. In the present study the sulfonated polyphosphazene membranes were 600 μηι in thickness, with a width and length of 0.5 cm. DSC measurements were carried out using a T A Instruments DSC 2970 apparatus at a heating rate of 10°C/min, under a dry nitrogen atmosphere. Calibration was performed using indium. C solid-state N M R spectra were recorded on a Bruker A S X 400 M H z spectrometer. A sample spinning rate of 10 K H z and proton decoupling at a level of 45 K H z were employed under magic angle spinning (MAS) conditions using dipolar decoupling with cross-polarization. Tetramethylsilane was used as a reference for C . The electrical conductivity of protons in water equilibrated membranes in the H form was determined using an A C impedance method. Membrane samples were first soaked in deionized and distilled water for 24 hours. The conductivity was measured using a pair of pressure-attached, high surface area platinum electrodes, as described elsewhere (13). The mounted sample was immersed in deionized and distilled water at a given temperature and measurements were made from 1 Hz to 10 Hz using a Paar Model 5210 amplifier and a Paar Model 273 1 3

13

+

5



166

Figure 1. Chemical reaction sequence during the sulfonation of poly[bis(3methyl phenoxy)phosphazene] with SO3.

0

0

0

é

1

'

1

1

0

40

80

120

160

1 2

1

200

1/2

Time ' (sec )

Figure 2. Membrane water weight loss vs square-root of time during a McBain balance vapor desorption experiment using a sulfonated polyphosphazene membrane.


167 potentiostat/galvanostat. Both real and imaginary components of the impedance were measured and the real Z-axis intercept was closely approximated. The cell constant was calculated from the spacing of the electrodes, the thickness of the membrane, and the area of the platinum electrodes.


Results and Discussion Membrane Ion-Exchange Capacity (IEC) and Water Swelling. The measured ion-exchange capacity of sulfonated poly[bis(3-methylphenoxy)phosphazene] polymer samples and the equilibrium swelling of the samples in water at 25°C are listed in Table I for different SO3/POP molar ratios (where POP denotes a polyphosphazene monomer unit). When the SO3/POP ratio was < 0.64, the IEC of the membranes was detectable, but very low (0.01-0.06 mmol/g). A gradual increase in IEC was observed as the SO3/POP ratio was increased from 0.64 to 1.92. The overall reaction of S 0 with poly[bis(3-methylphenoxy)phosphazene] is shown in Figure 1. During polymer preconditioning with NaOH and water after sulfonation, most of the =N—>S0 complex was hydrolyzed and the resulting =N H HS0 " product was flushed from the sample. The difference between the total sulfur content and the amounts of S0 " (equal to the measured ion-exchange capacity) and resulting s N H H S 0 " gave the concentration of Ξ=Ν—»S0 complex. The s=N-»S0 and s N l H H S 0 " species make the polymer more hydrophilic which affects polymer swelling in water, but their presence had essentially no bearing on the measurement of the polymer's ion-exchange capacity (the concentration of = N H H S 0 " was so low that its decomposition and the elution of H S 0 affected the IEC by no more than 7%). 3

3

+

4

3

+

4

3

3

4

+

4

2

4

Table I. Ion-exchange capacity and swelling of sulfonated poly[bis(3-methylphenoxy)phosphazene]. SO3/POP molar ratio 0.32 0.64 0.80 0.96 1.28 1.60 1.92

Ion-exchange capacity Membrane swelling in water (wt.% dry membrane) (mmol/g) 1,92 (an IEC > 2.1 mmol/g) were found to be water soluble. For comparison, a sulfonated poly[bis(3-methylphenoxy)phosphazene] membrane with an ion-exchange capacity of 0.95 mmol/g swelled 42% at room temperature, whereas a Nafion 117 perfluorosulfonic acid membrane (with an IEC of 0.909 mmol/g) swelled 34% (14). 3

3

3


3

4

IR and N M R Spectra. IR spectra of sulfonated bis(3-methylphenoxy) phosphazene polymer samples were collected for wave numbers in the 600-1,650 cm" range. There was no obvious change in the P-O-methylphenoxy stretching band (at 1,140 cm" ) and the P-N band (at 1,243 cm* ) for polymers sulfonated with a SO3/POP ratio < 1.28. A new peak corresponding to S=0 (at 1,085 cm' ) appeared and grew with increasing SO3/POP. At a SO3/POP ratio of 1.92, the P=N band decreased slightly, which was attributed to some degradation of the polyphosphazene main chain. To determine the location of S 0 attack on the polyphosphazene's methylphenoxy side groups, C solid-state N M R spectra of the base and sulfonated polymers were collected. Six resonance peaks in the 100-180 ppm range were assigned to the six aromatic carbon nuclei on the methylphenoxy side-groups (aromatic carbon signals from sulfonated and non-sulfonated side-chains were indistinguishable). For a poly[bis(3-methylphenoxy)phosphazene] sample sulfonated at a SO3/POP ratio of 0.76, the C4 (para position) signal (at δ =125.9 ppm) decreased significantly and shifted downfield by 14-15 ppm. This observation, along with subsequent calculations, is consistent with para-position sulfonation of the polyphosphazene's methylphenoxy side-groups. When poly[bis(3-methylphenoxy)phosphazene] was sulfonated at SO3/POP molar ratios > 1.0, the intensities of the four resonance signals for aromatic carbons at the C5, C4, C2, and C6 positions decreased significantly, indicating that all phenoxy carbons except C I and C3 were undergoing S 0 attack. 1

1

1

1

3

1 3

3

DSC Measurements. The glass transition temperatures (Tg) of the base-polymer and sulfonated poly[bis(3-methylphenoxy)phosphazene] samples, as determined from DSC curves, are plotted in Figure 3 as a function of polymer ion-exchange capacity. No change in Tg was observed when the polymer was exposed to a low concentration of S 0 (i.e., when there was essentially no aromatic C-sulfonation). 3


169


Ο

0

1

2

3

Ion-exchange capacity (mmol/g) Figure 3. Variation in glass transition temperature of sulfonated poly[bis(3methyl phenoxy)phosphazene] as a function of polymer ion-exchange capacity.


170


The polymer's Tg increased with increasing ion-exchange capacity when the SO3/POP molar ratio was > 0.64. The observed increase in Tg was associated with electrostatic interactions among fixed-charge groups in the ionomeric portions of the polymer and the general incompatibility of charged domains and nonpolar polymer backbone, which restricted polymer chain mobility and produced a rise in Tg (15,16). Proton Conductivity. The proton conductivity of sulfonated polyphosphazene membranes in the FT form and equilibrated in water were determined at various temperatures ranging from 25°C - 60°C. The results for polyphosphazene membranes with an IEC of 0.8 and 1.0 mmol/g are plotted vs. the reciprocal temperature in Figure 4. The proton conductivity of a water-equilibrated Nafion 117 cation-exchange membrane with an IEC of 0.909 mmol/g (17) is also shown in this figure. As expected, the polyphosphazene membrane conductivity increased with increasing IEC and with increasing temperature. The proton conductivity of a 1.0 mmol/g IEC sulfonated phosphazene membrane was high, at approximately 60% of that for Nafion 117 (the lower conductivity was associated with the greater water swelling of the polyphosphazene membrane, coupled with a less well-defined "cluster network" micro-structure). The membrane conductivity data for Nafion 117 and the two polyphosphazene membranes in Figure 4 were fitted to the following equation, K = K

Membrane Formation and Modification - American Chemical Society

Recommend Documents