Synthesis of new polyelectrolytes via backbone quaternization of poly(aryloxy- and alkoxyphosphazenes) and their small molecule counterparts

Article abstract of DOI:10.1021/ma202619j

Novel polyelectrolytes were synthesized by quaternization of the backbone of poly(alkoxy- and aryloxyphosphazenes) with strong alkylation reagents. As models for the synthesis of these polymers, similar quaternization reactions were also carried out on small-molecule alkoxy and aryloxy cyclotriphosphazenes. The quaternized small molecules and high polymers were characterized by ¹H NMR, ³¹P NMR, DSC, TGA, and AC impedance studies. The quaternized poly(alkoxyphosphazenes) showed ionic conductivities of 2.58 × 10^-4 S?cm^-1 at 25 °C and 2.09 × 10 ^-3 S?cm^-1 at 80 °C, which are among the highest values for known solvent-free ionically conducting polymers.

Full text of DOI:10.1021/ma202619j

Article
pubs.acs.org/Macromolecules
Synthesis of New Polyelectrolytes via Backbone Quaternization of
Poly(aryloxy- and alkoxyphosphazenes) and their Small Molecule
Counterparts
Chen Chen, Andrew R. Hess, Adam R. Jones, Xiao Liu, Greg D. Barber, Thomas E. Mallouk,
and Harry R. Allcock*
Department of Chemistry, The Pennsylvania State University, University Park, Pennsylvania 16802, United States
ABSTRACT: Novel polyelectrolytes were synthesized by quaternization of the
backbone of poly(alkoxy- and aryloxyphosphazenes) with strong alkylation
reagents. As models for the synthesis of these polymers, similar quaternization
reactions were also carried out on small-molecule alkoxy and aryloxy
cyclotriphosphazenes. The quaternized small molecules and high polymers
1
were characterized by H NMR, 31P NMR, DSC, TGA, and AC impedance
studies. The quaternized poly(alkoxyphosphazenes) showed ionic conductivities
of 2.58 × 10−4 S·cm−1 at 25 °C and 2.09 × 10−3 S·cm−1 at 80 °C, which are among the highest values for known solvent-free
ionically conducting polymers.
considerably higher than in most other ionic conducting
polymers including poly(ethylene oxide), allows sufficient
segmental motion for the facile migration of ions through the
solid electrolyte medium. Polyphosphazenes with oligoethylene
oxy side chains are among the most ionically conductive
polymers.6−8 One of the disadvantages of polymer/salt
complexes as polymer electrolytes is that both cations and
anions are mobile, which leads to concentration polarization
which significantly reduces cell performance. In addition, the
mobility of both anions and cations also hampers a fundamental
understanding of the ion transport mechanisms in polymer
electrolytes. A solution to these problems is to covalently link
anions or cations to the polymer backbone, to yield
polyelectrolytes. Side groups with both cations and anions
have been tethered to the polyphosphazene backbones as side
groups to form polycations and polyanions.9−13
Functionalization of the backbone nitrogen atoms of
polyphosphazenes has received far less attention than the
variation of side groups at the phosphorus centers. Backbone
nitrogen atoms have been shown to play key role in
coordination chemistry of oligo- and polyphosphazenes.14−19
Organo- and organoamino-substituted cyclooligophosphazenes
have been quaternized at the backbone nitrogen atoms to form
phosphazenium cations,20−25 but the quaternization of organo-
aminophosphazene high polymer has not been studied
extensively.26
INTRODUCTION
■
Since the first report of a solid-state polymer electrolyte
composed of poly(ethylene oxide)/salt complexes by Wright et
al.1 the development of polymeric solid-state electrolytes has
drawn the attention of many researchers. Solid-state polymeric
electrolytes have many advantages over conventional liquid
electrolytes such as high dimensional stability, excellent
processability, and improved safety, which make them ideal
electrolytes for solid-state electrochemical devices such as high
energy density rechargeable batteries, fuel cells, electric double-
layer capacitors, and electrochromic displays.2
Dye-sensitized solar cells (DSSCs) have been under intensive
investigation as a low cost alternative to silicon solar cells.3,4
The most important factor affecting the practicality of DSSCs is
the electrolyte, which usually consists of an I−/I3− redox couple
in a volatile liquid organic solvent. The serious drawbacks of
organic solvent based liquid electrolytes are their volatility and
the possibility of leakage during extended operation, especially
at elevated temperatures, together with the necessary complex-
ity of sealing the unit. Solid-state polymeric electrolytes are a
logical solution to overcome both of the problems caused by
liquid organic solvent.
Polyphosphazenes are long chain macromolecules that
contain an inorganic backbone composed of alternating
phosphorus and nitrogen atoms joined by unusual quasi-
delocalized single−double bonds, and organic side groups
attached to the backbone phosphorus atoms. Many different
organic side groups can be linked to the polyphosphazene
backbone in variable ratios, a feature that provides this polymer
platform with versatile properties and applications.5 With
appropriate side groups attached to the backbone, polyphos-
phazenes are known to have low glass transition temperatures
(Tg) and zero crystallinity: in other words, they remain highly
flexible even at low temperatures due to the unusual torsional
mobility of the PN backbone. This flexibility, which is
We report the first synthesis of novel polyelectrolytes via
backbone quaternization of poly(aryloxy- and alkoxyphospha-
zenes), in which the charge carriers are located on skeletal
nitrogen atoms . These new polyelectrolytes show good ionic
conductivity without the presence of plasticizers or solvents.
Received: December 2, 2011
Revised: January 10, 2012
Published: January 30, 2012
© 2012 American Chemical Society
1182
dx.doi.org/10.1021/ma202619j | Macromolecules 2012, 45, 1182−1189

Macromolecules
Article
fonate (0.686 mL, 6.06 mmol) at room temperature dropwise, and the
mixture was stirred at room temperature for 24 h. All volatiles were
evaporated under vacuum and the resulting oil solidified slowly under
high vacuum to yield the trimer (6) as a reddish waxy solid (2.42 g,
Although dealkylation reactions, which cause loss of charge
carriers, were detected when these quaternized species were
exposed to anions with strong nucleophilicity, the promising
conductivity results reveal a new concept for the synthesis of
highly conductive polyelectrolytes via backbone functionaliza-
tion.
1
2.82 mmol), which is sufficiently pure by H and 31P NMR spectra.
Yield: 93%. 1H NMR (360 MHz, CDCl3): δ 3.77 (t, 3JPH = 10.8 Hz, 3
H), 6.59−6.65 (m, 4 H), 6.95−7.03 (m, 8 H), 7.15−7.25 (m, 6 H),
7.28−7.47 (m, 12 H). 31P NMR (145 MHz, CDCl3): δ −3.60 (1 P),
2
1.12 (2 P), AB2, JPP = 78.4 Hz. ESIMS: m/z 708.2 [M − OTf]+.
EXPERIMENTAL SECTION
■
A tt e m p t e d Q u a t e r n i z at i o n o f H e xa ( p h e n o x y ) -
cyclotriphosphazene (2) with Triethyloxonium Tetrafluoroborate.
To a solution of hexa(phenoxy)cyclotriphosphazene (2) (0.347 g,
0.500 mmol) in 5 mL of DCM was added triethyloxonium
tetrafluoroborate (1 M in DCM, 1.00 mL, 1.00 mmol) at room
temperature dropwise, and the mixture was stirred at room
temperature for 2 days. 31P NMR spectroscopy indicated that only
24% of trimer (2) had been quaternized. 31P NMR (145 MHz, crude
reaction mixture with D2O capillary as external reference: δ −4.08 (1
Reagents and Equipment. All synthesis reactions were carried
out under a dry argon atmosphere using standard Schlenk line
techniques. Tetrahydrofuran (THF), 1,4-dioxane and dichloromethane
(DCM) (EMD) were dried using a solvent purification system, in
which the solvents pass through columns of molecular sieves under a
dry argon atmosphere.27 Di(ethylene glycol) methyl ether (Aldrich),
1-butanol (EMD), and 2,2,2-trifluoroethanol (Aldrich) were distilled
over calcium hydride. Phenol was purified by sublimation under
vacuum. Acetone (EMD), methanol (EMD), hexanes (EMD), N,N-
dimethylformamide (EMD), sodium hydride (60% dispersion in
mineral oil, Aldrich), sodium iodide (Aldrich), 1,2-dichloroethane
(Aldrich), chorobenzene (Aldrich), sodium thiosulfate pentahydrate
(Aldrich), sodium bicarbonate (EMD), triethyloxonium tetrafluor-
oborate (1 M solution in dichloromethane, Aldrich) and methyl
trifluoromethanesulfonate (Aldrich) were used as received. Poly-
(dichlorophosphazene) was prepared via the thermal ring-opening
polymerization of recrystallized and sublimed hexachlorocyclotriphos-
phazene (Fushimi Pharmaceutical Co., Japan or Ningbo Chemical Co.,
China) in evacuated Pyrex tubes at 250 °C.28 Unpolymerized
hexachlorocyclotriphosphazene was removed by vacuum sublimation.
31P and 1H NMR spectra were obtained with a Bruker 360 WM
instrument operated at 145 and 360 MHz, respectively. 31P NMR
2
P), 1.28 (2 P), AB2, JPP = 77.1 Hz, 5.08 (broad, s, 76%). After
quenching with saturated sodium bicarbonate solution, the broad
singlet peak at 1.30 shifted back to 9.40, which could be assigned to
trimer 2.
Synthesis of N-Methyl Hexa(n-butoxy)cyclotriphosphazenium
Trifluoromethanesulfonate (7). To a solution of hexa(n-butoxy)-
cyclotriphosphazene (3) (1.41 g, 2.46 mmol) in 25 mL of DCM was
added a solution of methyl trifluoromethanesulfonate (0.334 mL, 2.95
mmol) in 10 mL of DCM dropwise at −78 °C. The mixture was
allowed to warm to room temperature gradually over 1 h. The
resulting mixture was diluted with 30 mL of DCM, washed with 10%
sodium bicarbonate solution, dried over anhydrous sodium sulfate, and
concentrated under vacuum. The residue was purified on a Sephadex
LH-20 column using THF/DCM (1/1, v/v) as eluent to yield trimer
1
spectra were proton decoupled. H shifts are reported in ppm relative
1
to tetramethylsilane at 0 ppm. 31P shifts are reported in ppm relative to
85% H3PO4 at 0 ppm. 31P NMR spectra were simulated with
SpinWorks (version 3.1.8.1). Glass transition temperatures were
measured with a TA Instruments Q10 differential scanning calorimetry
apparatus at a heating rate of 10 °C/min and a sample size of ca. 10
mg. Gel permeation chromatograms were obtained using a Hewlett-
Packard HP 1100 gel permeation chromatograph equipped with two
Phenomenex Phenogel linear 10 columns and a Hewlett-Packard
1047A refractive index detector. The samples were eluted at 1.0 mL/
min with a 10 mM solution of tetra-n-butylammonium nitrate in THF,
and the elution times were calibrated with polystyrene standards. Mass
spectrometric analysis data were collected using the turbospray
ionization technique with an Applied Biosystems API 150EX LC/
MS mass spectrometer. Ionic conductivities were measured with a HP
4192A impedance analyzer in custom-made cells. All the ionic
conductivity tests were carried out at a fixed AC level of 0.1 V, and the
frequency was scanned from 5 to 6,000,000 Hz. Thermal
decomposition traces were obtained using a Perkin−Elmer TGA 7
thermogravimetric analyzer. Heating occurred at a rate of 20 °C/min
from 50 to 800 °C under a nitrogen atmosphere and a flow rate of 50
mL/min.
Synthesis of Unquaternized Phosphazenes. Hexa(phenoxy)-
cyclotriphosphazene (2),29 hexa(n-butoxy)cyclotriphosphazene (3),30
hexakis(2-(2-methoxyethoxy)ethoxy)cyclotriphosphazene (4),31
hexakis(2,2,2-trifluoroethoxy)cyclotriphosphazene (5),32 poly-
(diphenoxyphosphazene) (13),28,33 poly[di(n-butoxy)phosphazene]
(14),34 poly[bis(2-(2-methoxyethoxy)ethoxy)phosphazene] (15)7
and poly[bis(2,2,2-trifluoroethoxy)phosphazene] (16)28,33 were syn-
thesized by allowing sodium phenoxide or sodium alkoxides to react
with hexachlorocyclotriphosphazene or poly(dichlorophosphazene) in
THF or 1,4-dioxane at room temperature (25 °C) or reflux
temperature. Cyclotriphosphazenes were purified by recrystallization
or flash column chromatography, and polyphosphazenes were purified
by reprecipitation or dialysis prior to being subjected to quaternization.
Synthesis of Cyclic Trimer Models. Synthesis of N-Methyl
Hexa(phenoxy)cyclotriphosphazenium Trifluoromethanesulfonate
(6). To a solution of hexa(phenoxy)cyclotriphosphazene (2) (2.10 g,
3.03 mmol) in 20 mL of DCM was added methyl trifluoromethanesul-
(7) as a colorless liquid (1.57 g, 2.13 mmol). Yield: 87%. H NMR
(360 MHz, CDCl3): δ 0.90−1.00 (m, 18 H), 1.35−1.50 (m, 12 H),
3
1.65−1.83 (m, 12 H), 2.96 (t, JPH = 9.3 Hz, 3 H), 3.96−4.05 (m, 4
H), 4.10−4.23 (m, 8 H). 31P NMR (145 MHz, CDCl3): δ 6.03 (1 P),
2
9.38 (2 P), AB2, JPP = 72.3 Hz. ESIMS: m/z 588.4 [M − OTf]+.
Synthesis of N-Methyl Hexakis(2-(2-methoxyethoxy)ethoxy)-
cyclotriphosphazenium Trifluoromethanesulfonate (8). Hexakis(2-
(2-methoxyethoxy)ethoxy)cyclotriphosphazene (4) (1.01 g, 1.18
mmol) was quaternized to yield trimer (8) as a colorless liquid
(1.02 g, 1.01 mmol) by following a similar procedure to the one
1
described for the synthesis of trimer (7). Yield: 86%. H NMR (360
3
MHz, CDCl3): δ 3.00 (t, JPH = 9.3 Hz, 3 H), 3.36 (s, 18 H), 3.50−
3.57 (m, 12 H), 3.61−3.68 (m, 12 H), 3.70−3.80 (m, 12 H), 4.18−
4.25 (m, 4 H), 4.29−4.37 (m, 8 H). 31P NMR (145 MHz, CDCl3): δ
2
6.55 (1 P), 9.79 (2 P), AB2, JPP = 75.5 Hz. ESIMS: m/z 864.4 [M −
OTf]+.
Attempted Quaternization of Hexakis(2,2,2-trifluoroethoxy)-
cyclotriphosphazene (5). To a solution of hexakis(2,2,2-
trifluoroethoxy)cyclotriphosphazene (5) (0.365 g, 0.501 mmol) in
10 mL of DCM was added methyl trifluoromethanesulfonate (0.113
mL, 1.00 mmol) at room temperature dropwise and the mixture. After
stirring for 48 h at room temperature, no quaternization could be
observed based on the 31P NMR spectra of the reaction mixture. The
reaction mixture was then refluxed for 48 h. 31P NMR spectroscopy
indicated that only 26% of trimer (5) had been quaternized, and the
presence of quaternized species (9) could be detected in ESIMS
spectrum as well. 31P NMR (145 MHz, crude reaction mixture with
D2O capillary as external reference): δ 5.24 (1 P), 9.02 (2 P), AB2, 2JPP
= 82.3 Hz, 17.24 (s, 74.5%). ESIMS: m/z 864.4 [M − OTf]+.
Treatment of N-Methyl Hexa(phenoxy)cyclotriphosphazenium
Trifluoromethanesulfonate (6) with Sodium Iodide. To a solution of
N-methyl hexa(phenoxy)cyclotriphosphazene trifluoromethanesulfo-
nate (6) (429 mg, 0.500 mmol) in 10 mL of DCM was added sodium
iodide (150 mg, 1.00 mmol) and the mixture was stirred at room
temperature for 48 h. Then the reaction mixture was diluted with 10
mL of DCM, washed with 5% aqueous sodium thiosulfate and brine
successively, dried over anhydrous sodium sulfate, and concentrated
under vacuum to yield trimer (2) as a yellowish solid (309 mg, 0.446
1183
dx.doi.org/10.1021/ma202619j | Macromolecules 2012, 45, 1182−1189

Macromolecules
Article
1
mmol). The yield was 89%. The H NMR and 31P NMR spectra
6000−8000) to yield polymer (18) as a clear colorless gum (1.90 g,
correspond well with unquaternized trimer (2).
8.42 mmol), which was soluble in acetone and THF and was slightly
1
Synthesis of 1,3,3,5,5-Pentakis(n-butoxy)-1-oxo-2-methylcyclotri-
phosphazadiene (10). To a solution of N-methyl hexa(n-butoxy)-
cyclotriphosphazene trifluoromethanesulfonate (7) (2.02 g, 2.74
mmol) in 25 mL of DCM was added sodium iodide (821 mg, 5.48
mmol) and the mixture was stirred at room temperature for 24 h. The
reaction mixture was then diluted with 25 mL of DCM, washed with
5% aqueous sodium thiosulfate and brine successively, dried over
anhydrous sodium sulfate, and concentrated under vacuum. The
residue was purified on a Sephadex LH-20 column using methanol/
DCM (1/1) as an eluent to yield compound (10) as a pale yellow
liquid (1.16 g, 2.18 mmol). Yield: 80%. 1H NMR (360 MHz, CDCl3):
δ 0.85−1.00 (m, 15 H), 1.33−1.47 (m, 10 H), 1.58−1.75 (m, 10 H),
soluble in DCM and chloroform. Yield: 80%. H NMR (360 MHz,
acetone-d6): δ 0.87−1.05 (m, 6 H), 1.48−1.55 (m, 4 H), 1.58−1.85
(m, 4 H), 3.13−3.31 (m, 0.63 H), 3.95−4.43 (m, 4 H). 31P NMR (145
MHz, acetone-d6): δ −5.8 to −3.0 (m, 60.4%), 3.0 to 5.0 (m, 39.6%).
Synthesis of {[NP(OBu-n)2]0.87[N+(CH3)(OTf)−P(OBu-n)2]0.13}n (18a).
Poly[di(n-butoxy)phosphazene] (14) (1.15 g, 6.01 mmol) was
quaternized with 0.12 equiv of methyl trifluoromethanesulfonate (82
μL, 0.725 mmol) to yield polymer 18a as a clear colorless gum (1.01 g,
4.75 mmol) by following a similar procedure to the one described for
the synthesis of polymer 18. Polymer 18a was soluble in acetone,
1
methanol, DCM, and chloroform. Yield: 79%. H NMR (360 MHz,
acetone-d6): δ 0.87−1.05 (m, 6 H), 1.48−1.55 (m, 4 H), 1.58−1.85
(m, 4 H), 3.13−3.25 (m, 0.40 H), 3.95−4.43 (m, 4 H). 31P NMR (145
MHz, acetone-d6): δ −5.8 to −3.0 (m, 72.7%), 3.0−5.0 (m, 27.3%).
Synthesis of {[NP(OCH2CH2OCH2CH2OCH3)2]0.81[N+(CH3)(OTf)−P-
(OCH2CH2OCH2CH2OCH3)2]0.19}n (19). Poly[bis(2-(2-methoxyethoxy)-
ethoxy)phosphazene] (15) (2.40 g, 8.47 mmol) was quaternized to
yield polymer 19 as a clear yellow gum (2.26 g, 7.19 mmol) by
following a similar procedure to the one described for the synthesis of
polymer 18. Polymer 19 was soluble in acetone, methanol, DCM, and
2.87 (dd, JPH = 9.9, 8.2 Hz, 3 H), 3.70−4.10 (m, 10 H). 31P NMR
3
2
(145 MHz, CDCl3): δ 1.63 (dd, JPP = 58.0, 46.4 Hz, 1 P), 7.64 (dd,
2JPP = 75.4, 58.0 Hz, 1 P), 15.02 (dd, JPP = 75.4, 46.4 Hz, 1 P).
2
ESIMS: m/z 532.2 [M + H]+.
Synthesis of 1,3,3,5,5-Pentakis(2-(2-methoxyethoxy)ethoxy)-l-
oxo-2-methylcyclotriphosphazadiene (11). N-Methyl hexakis(2-(2-
methoxyethoxy)ethoxy)cyclotriphosphazene trifluoromethanesulfo-
nate (8) (1.90 g, 1.88 mmol) was treated with sodium iodide to
form compound 11 as a colorless liquid (1.20 g, 1.58 mmol) by
following a similar procedure to the one described in the synthesis of
1
chloroform. Yield: 85%. H NMR (360 MHz, CDCl3): δ 3.00−3.16
(m, 0.58 H), 3.34 (s, 6 H), 3.48−3.55 (m, 4 H), 3.58−3.79 (m, 8 H),
4.00−4.42 (m 4 H); 31P NMR (145 MHz, CDCl3) δ −8.2 to −6.0 (m,
65.6%), 0−2.5 (m, 34.4%).
1
compound 10. Yield: 84%. H NMR (360 MHz, CDCl3): δ 2.91 (dd,
3JPH = 9.8, 8.5 Hz, 3 H), 3.33−3.42 (m, 15 H), 3.49−3.58 (m, 10 H),
3.62−3.68 (m, 10 H), 3.68−3.76 (m, 10H), 3.95−4.22 (m, 10 H). 31
P
Synthesis of {[NP(OCH2CH2OCH2CH2OCH3)2]0.88[N+(CH3)(OTf)−P-
NMR (145 MHz, CDCl3): δ 1.44 (dd, 2JPP = 60.9, 47.9 Hz, 1 P), 8.22
(OCH2 CH2 OCH2 CH2 OCH3 )2 ]0 . 1 2
}
(19a). Poly[bis(2-(2-
n
2
2
(dd, JPP = 75.4, 60.9 Hz, 1 P), 15.15 (dd, JPP = 75.4, 47.9 Hz, 1 P).
methoxyethoxy)ethoxy)phosphazene] (15) (1.28 g, 4.52 mmol) was
quaternized to yield polymer 19a as a clear yellow gum (1.10 g, 3.63
mmol) by following a procedure similar to the one described for the
synthesis of polymer 18a. Polymer 19a was soluble in acetone,
ESIMS: m/z 762.3 [M + H]+.
Treatment of Hexa(phenoxy)cyclotriphosphazene (2) with Iodo-
methane. To a solution of hexa(phenoxy)cyclotriphosphazene (2)
(0.347 g, 0.500 mmol) in 5 mL pf chlorobenzene was added
iodomethane (0.311 mL, 5.00 mmol), and the mixture was stirred at
100 °C for 2 days. No quaternization was detected according to 31P
NMR spectroscopy.
1
methanol, DCM and chloroform. Yield: 80%. H NMR (360 MHz,
CDCl3): δ 3.00−3.12 (m, 0.37 H), 3.35 (s, 6 H), 3.46−3.54 (m, 4 H),
3.56−3.73 (m, 8 H), 3.97−4.38 (m 4 H); 31P NMR (145 MHz,
CDCl3) δ −8.2 to −6.0 (m, 80.5%), 0−2.5 (m, 19.5%).
Treatment of Hexakis(2-(2-methoxyethoxy)ethoxy)-
cyclotriphosphazene (4) with Iodomethane. To a solution of
hexakis(2-(2-methoxyethoxy)ethoxy)cyclotriphosphazene (4) (0.425
g, 0.500 mmol) in 8 mL of 1,2-dichloroethane was added iodomethane
(0.156 mL, 2.50 mmol). No quaternization could be detected after
stirring at 40 °C for 24 h. The mixture was then stirred at 60 °C for 24
h, and 31P NMR indicated that 37% of trimer 5 had been converted to
compound 11, which was further confirmed by ESIMS. Quaternization
at 100 °C in chlorobenzene for 24 h resulted in a complex mixture
with 31P NMR signals ranging from −5 to +20 ppm. The yield of
compound 11 in the mixture was 54% based on the integration in the
31P NMR spectrum.
Treatment of {[NP(OPh)2]0.50[N+(CH3)(OTf)−P(OPh)2]0.50 n
(17)
}
with Sodium Iodide. To a solution of {[NP(OPh)2]0.50[N+(CH3)-
(OTf)−P(OPh)2]0.50}n (17) (332 mg, 1.06 mmol) in 25 mL of THF
was added sodium iodide (0.159 g, 1.05 mmol) at room temperature,
and the mixture was stirred for 2 days, and was concentrated and
precipitated from THF into water four times to yield polymer 13 as a
white fibrous solid (198 mg, 0.856 mmol). The yield was 81%. The 1H
NMR and 31P NMR spectra corresponded well to the unquaternized
polymer 13.
Synthesis of {[NP(OBu-n)2]0.81[N(CH3)P(O)(OBu-n)]0.19}n (20). To a
solution of {[NP(OBu-n)2]0.81[N+(CH3)(OTf)−P(OBu-n)2]0.19}n (18)
(332 mg, 1.06 mmol) in 25 mL of acetone was added sodium iodide
(0.159 g, 1.05 mmol) at room temperature and the mixture was stirred
for 2 days. The mixture was concentrated and precipitated from
acetone into water once and from THF into water twice to yield
polymer 20 as a gray elastomer (380 mg, 1.93 mmol). It was soluble in
THF and was slightly soluble in DCM, acetone and chloroform. Yield:
90%. 1H NMR (360 MHz, THF-d8/acetone-d6 1/1): δ 0.96 (br, s, 5.45
H), 1.38−1.55 (m, 3.61 H), 1.58−1.85 (m, 3.58 H), 2.94−3.08 (m,
0.57 H), 3.80−4.38 (m, 3.63 H). 31P NMR (145 MHz, THF-d8/
acetone-d6 1/1) δ −6.0 to −3.0 (m, 60.2%), −3.0 to 0.5 (m, 17.2%),
6.5 to 9.0 (m, 22.6%).
Synthesis of Polymers. Synthesis of {[NP(OPh)2]0.50[N+(CH3)-
(OTf)− P(OPh)2 ]0 . 5 0
}
(17). To a suspension of poly-
n
(diphenoxyphosphazene) (13) (462 mg, 2.00 mmol) in 20 mL of
DCM was added a solution of methyl trifluoromethanesulfonate
(0.226 mL, 2.00 mmol) in 10 mL of DCM dropwise at 0 °C. The
mixture was allowed to warm to room temperature gradually and was
stirred for 3 days until it turned into a clear solution. The reaction
mixture was then precipitated into hexanes. The precipitation was
repeated twice from THF into hexanes to yield polymer 17 as a white,
fibrous solid (469 mg, 1.50 mmol), which was soluble in THF and was
1
slightly soluble in DCM and chloroform. Yield: 75%. H NMR (360
Synthesis of {[NP(OCH2CH2OCH2CH2OCH3)2]0.81[N(CH3)P(O)-
MHz, CDCl3/THF-d8 1/1 v/v): δ 4.00 (broad, s, 1.5 H), 6.30−7.00
(m, 10 H). 31P NMR (145 MHz, CDCl3/THF-d8 1/1 v/v): δ −18.0 to
−16.5 (m, 11.7%), −15.55 (broad, s, 75.8%), −11.2 to −10.2 (m,
12.5%).
(OCH2CH2OCH2CH2OCH3)]0.19
}
(21). To a solution of
n
{[NP(OCH2CH2OCH2CH2OCH3)2]0.81[N+(CH3)(OTf)−P-
(OCH2CH2OCH2CH2OCH3)2]0.19}n (19) (640 mg, 2.04 mmol) in 25
mL of acetone was added sodium iodide (306 mg, 2.04 mmol) at
room temperature and the mixture was stirred for 4 days. The mixture
was dialyzed versus methanol for 3 days (MWCO 6000−8000) to
yield polymer 21 as a clear yellow gum (500 mg, 1.88 mmol), which is
Synthesis of {[NP(OBu-n)2]0.79[N+(CH3)(OTf)−P(OBu-n)2]0.21}n (18).
To a solution of poly[di(n-butoxy)phosphazene] (14) (2.00 g, 10.5
mmol) in 80 mL of DCM was added a solution of methyl
trifluoromethanesulfonate (1.19 mL, 10.5 mmol) in 10 mL of DCM
dropwise at −78 °C. The mixture was allowed to warm to room
temperature gradually and was stirred for 2 days. The mixture was
dialyzed versus acetone/methanol (1/1, v/v) for 3 days (MWCO
1
soluble in acetone and THF, DCM and chloroform. Yield: 92%. H
NMR (360 MHz, CDCl3): δ 2.86−2.95 (m, 0.58 H), 3.34 (s, 5.65 H),
3.45−3.53 (m, 3.94 H), 3.55−3.74 (m, 7.48), 3.95−4.35 (m, 3.6 H).
1184
dx.doi.org/10.1021/ma202619j | Macromolecules 2012, 45, 1182−1189

Macromolecules
Article
31P NMR (145 MHz, CDCl3) δ −7.0 to −5.0 (m, 66.9%), −4.0 to 2.5
noteworthy that the basicity of a compound will not necessarily
be in the same order as its nucleophilicity or electron density,
because the nucleophilicity is influenced by other factors such
as steric hindrance. However, the basicity of cyclotriphospha-
zenes with similar structures may serve as a preliminary
indicator of nucleophilicity. The basicities of cyclotriphospha-
zenes with various substituents are reported to decrease in the
order: N3P3(NHR)6 or N3P3(NR2)6 (pKa = 8.8−7.6) > N3P3R6
(pKa = 6.4−1.5) > N3P3(OR)6 (pKa = 1.4∼-5.8) > N3P3Cl6
(pKa < −6.0).37−39 Fairly basic organoamino-substituted
cyclooligophosphazenes are readily quaternized at the ring
nitrogen atoms to form phosphazenium cations using simple
alkyl halides or stronger alkylating reagents such as methyl
trifluoromethanesulfonate or trimethyloxonium fluorobo-
rate.20−22,26 Because of the relatively lower electron densities
at the ring nitrogen atoms, alkoxy or aryloxy-substituted
phosphazenes are inert to iodomethane at room temperature,26
or they undergo rearrangement of the alkoxyphosphazene to
the N-alkyloxophosphazane at higher temperatures.36 In this
study, methyl trifluoromethanesulfonate and triethyloxonium
tetrafluoroborate, two of the most aggressive alkylation
reagents, were used for methyl or ethyl quaternization
respectively. Dichloromethane (DCM) was chosen as the
solvent because of its ability to solubilize all the trimers and its
resistance to aggressive alkylation reagents. The reactions were
monitored using 31P NMR spectroscopy. Monoquaternization
of trimer 2 was achieved at room temperature with methyl
trifluoromethanesulfonate. Further quaternization did not occur
even after prolonged reaction times in the presence of excess
amounts of methyl trifluoromethanesulfonate. Triethyloxonium
tetrafluoroborate was less reactive to trimer 2 than was methyl
trifluoromethanesulfonate (see Experimental Section). Thus,
only methyl trifluoromethanesulfonate was utilized for the
remaining quaternization studies. For the quaternization of
trimers 3 and 4, methyl trifluoromethanesulfonate was added at
−78 °C. Meanwhile, the reaction temperatures, reaction times,
and the stoichiometries were carefully controlled to avoid the
formation of byproduct. Only 26% of trimer 5 had been
monoquaternized to form compound 9 after 48 h of refluxing.
Prolonged reaction times or elevated temperatures led to the
formation of a complex mixture of products. Quaternized
trimer 6 could be heated at 60 °C under vacuum for drying for
1 day or stored at room temperature unsealed for a few months
without detectable decomposition. Quaternized trimers 7 and 8
were found to be unstable at 60 °C while drying under vacuum.
The reactivity of the trimers with different substituents
corresponded well with the basicities of their skeletal nitrogen
(m, 16.4%), 4.5 to 6.2 (m, 16.7%).
Treatment of Poly[bis(2-(2-methoxyethoxy)ethoxy)-
phosphazene] (15) with Iodomethane. To a solution of poly[bis-
(2-(2-methoxyethoxy)ethoxy)phosphazene] (15) (567 g, 2.00 mmol)
in 8 mL of 1,2-dichloroethane was added iodomethane (0.623 mL,
10.0 mmol) at room temperature. The mixture was stirred at 40 °C for
17 h and at 60 °C for 27 h. Then the solution was concentrated and
dried under vacuum to yield a brown gum (608 mg), which is
insoluble in common solvents.
RESULTS AND DISCUSSION
■
Quaternization of Cyclic Trimer Models. To our
knowledge, the skeletal quaternization of organophosphazenes
that bear P−O linked side groups has not been reported. Thus,
phenoxy, n-butoxy, 2-(2-methoxyethoxy)ethoxy (MEEO) and
2,2,2-trifluoroethoxy (TFEO) were chosen as side groups to
cover the spectrum of P−OR linked species. Reactions carried
out on high polymeric phosphazenes are always more
challenging than their small-molecule counterparts, especially
for purification and characterization and, for this reason, small-
molecule cyclic phosphazenes have played a major role as
models for the development of polyphosphazene chemistry.35
In this field, cyclotriphosphazenes (usually referred to as
“trimers”) are the most readily available models, and are
favored for their ease of synthesis. Model reactions are shown
in Scheme 1.
Scheme 1. Quaternization of Trimer Models
The syntheses of the small-molecule unquaternized trimers
were carried out by the treatment of hexachlorocyclotriphos-
phazene 1 with an excess amount of the corresponding sodium
aryloxide or alkoxides. The yields of all four trimers were over
80%. The susceptibility of the backbone nitrogen atoms in
phosphazenes in the presence of alkylation reagents is
controlled by their electron densities, which are ultimately
determined by the substituents bonded to the neighboring
phosphorus atoms. Basicity studies with small molecule cyclic
phosphazenes show the ring nitrogen atoms to be the centers
of greatest electron density in these molecules.36 It is
Table 1. NMR Data for Small-Molecule Models
side
group
a
trimer
31P NMR
1H NMR
OPh
2
6
9.35 (s)
−
2
−3.60 (1 P), 1.12 (2 P), AB2, JPP = 78.4 Hz
9.34 (s)
3.77 (t, 3JPH = 10.8 Hz, 3 H)
b
2
−
OBn-n
3
18.90 (s)
−
2
7
6.03 (1 P), 9.38 (2 P), AB2, JPP = 72.3 Hz
1.63 (dd, 2JPP = 58.0, 46.4 Hz, 1 P), 7.64 (dd, 2JPP = 75.4, 58.0 Hz, 1 P), 15.02 (dd, 2JPP = 75.4, 46.4 Hz, 1 P)
2.96 (t, 3JPH = 10.8 Hz, 3 H)
10
4
2.87 (dd, 3JPH = 9.9, 8.2 Hz, 3 H)
OMEE
19.05 (s)
−
2
8
6.55 (1 P), 9.79 (2 P), AB2, JPP = 75.5 Hz
1.44 (dd, 2JPP = 60.9, 47.9 Hz, 1 P), 8.22 (dd, 2JPP = 75.4, 60.9 Hz, 1 P), 15.15 (dd, 2JPP = 75.4, 47.9 Hz, 1 P)
3.00 (t, 3JPH = 10.8 Hz, 3 H)
11
2.91 (dd, 3JPH = 9.8, 8.5 Hz, 3 H)
a
b
Only the methyl groups attached to backbones are shown here. Trimer 2 recovered from treatment of compound 17 with sodium iodide.
1185
dx.doi.org/10.1021/ma202619j | Macromolecules 2012, 45, 1182−1189

Macromolecules
Article
atoms. The pKa values of trimers 2, 3, 4, and 5 are −5.8, 0.1,
∼−1 (estimated), and −6.0 respectively. The higher stability of
trimer 6 than those of trimers 7 and 8 might result from a
shielding effect by bulky phenoxy groups to stabilize the
quaternization site, and from the resistance of the sp2 carbons
to migration and rearrangement reactions.
Scheme 2. Quaternization of Polymers
Because of the lack of hydrogen bonding sites and the
asymmetric geometries of the quaternized trimers, quaternized
trimer 6 is an amorphous solid at room temperature, whereas 7
and 8 are liquids. Thus, no X-ray crystallography data were
available. The NMR spectra of unquaternized and quaternized
trimers are shown in Table 1. Triplet signals at 3.77, 2.96, and
1
3.00 in H NMR spectra clearly indicate the methyl groups
attached to backbone nitrogen atoms in quaternized trimers 6,
7, and 8. The two adjacent phosphorus atoms split the methyl
groups with formally identical coupling constants (3JPH). The
the 31P NMR data due to the inherent structural complexity of
the polymers compared to those of their small-molecule models
as well as the random distribution of quaternization sites along
the backbone. The 31P NMR spectrum of polymer 18, as a
representative, is shown in Figure 1B. Broadened peaks with
1
ambiguous coupling in H NMR spectra were also detected,
which is also quite common for polyphosphazenes and some
other polymers. These phenomena further emphasized the key
role of model reactions in this chemistry. With the presence of
1 equiv of quaternization reagent per repeating unit, the
percentages of quaternized backbone nitrogen atoms in
polymers 17, 18, and 19 were 50%, 21%, and 19% respectively,
with 1−2% variation from batch to batch. The surprisingly high
degree of quaternization for polymer 17 can be attributed to
the shielding effect of bulky phenoxy substituents as observed
in the quaternization of trimer 2. Lower degrees of
quaternization were achieved by controlling the stoichiometry
of the quaternization reagent. Thermogravimetric analysis
(TGA) showed that the thermal stability of quaternized
polymers is lower than their unquaternized counterparts, but
they are considerably more stable than the small-molecule
models.
Molecular weight data for quaternized polymers 17−19
could not be obtained by gel permeation chromatography
(GPC) although they are completely soluble in THF, the
mobile phase used. This is a common problem for
polyelectrolytes in the absence of additional salting.40 Because
of the large quantity of alkali trifluoromethanesulfonate
required to saturate the mobile phase and the solubility and
possible instability of these polymers in aqueous media, we did
not attempt to measure the molecular weights with the use of
additional salting.
Figure 1. 31P NMR spectra of quaternized phosphazenes: (A) trimer 7
and (B) polymer 18.
31P NMR spectrum of trimer 7 is shown in Figure 1A. 31P
NMR signals of quaternized trimers are shifted upfield
compared to the singlet signals of the parent phosphazenes.
Software simulation indicated that the spin systems are AB2,
similar to the reported quaternized amino-substituted trimers.20
However, the extra splittings might be the result of the large
coupling constants. Mass spectra also provide evidence for the
formation of phosphazenium cations. (See Experimental
Section)
Quaternization of Polymers. Following similar synthesis
procedures to those employed for the small molecule models,
the quaternization of polymers 13−16 were carried out as
shown in Scheme 2. Unquaternized polymers were synthesized
by replacement of the chlorine atoms in poly-
(dichlorophosphazene) using sodium salts of the corresponding
aryloxide or alkoxide. Quaternization of the backbone nitrogen
atoms of polymers 13−16 was attempted using methyl
trifluoromethanesulfonate in dichloromethane (DCM) solu-
tions. In all cases except for polymer 16, the quaternization of
the polymers followed the same pattern as for the small-
molecule models. Quaternization of polymer 16 was
unsuccessful because of its poor solubility in DCM and in
other solvents which are inert to methyl trifluoromethanesul-
fonate and the low electron density on the backbone nitrogen
atoms. The physical data for polymers 17−19 are shown in
Table 2. Little structural information could be obtained from
Elimination Reactions of Quaternized Phosphazenes.
In order to expand the possible uses of these new materials,
further attempts were conducted with the quaternized
polyphosphazenes to exchange the anions from the stable
trifluoromethanesulfonate ion to other anions. A motivation for
this aspect was the need for an I−/I3 redox couple for use in
−
DSSC electrolytes.
Ion exchange was first studied by dialyzing polymer 19 versus
2% sodium iodide solution in methanol. The resultant polymer
showed a much lower ionic conductivity. This prompted a
study of the chemistry that occurs during the anion exchange.
All the quaternized phophazenes were subjected to treatment
with excess sodium iodide in DCM, THF, or acetone at room
temperature, as shown in Scheme 3. All were found to be
1186
dx.doi.org/10.1021/ma202619j | Macromolecules 2012, 45, 1182−1189

Macromolecules
Article
Table 2. Structural and Physical Properties for Polymers
Tg
(°C)
a
side group polymer
31P NMR
1H NMR
Mw (×103) Tonset (°C)
OPh
13
17
13
14
18
−18.75 (s)
−
1596.2
−
345.5
2242.3
−
426.6
351.6
372.8
260.0
157.1
192.7
176.1
268.7
173.9
211.1
242.5
−4.5
10.3
−18.0 to −16.5 (m, 11.7%), −15.55 (br, s, 75.8%), −11.2 to −10.2 (m, 12.5%)
−18.81 (s)
−7.32 (s)
−5.8 to −3.0 (m, 57.8%), 3.0 to 5.0 (m, 42.2%)
−6.0 to −3.5 (m, 72.7%), 2.2 to 4.5 (m, 27.3%)
−8.5 to −5.5 (m, 59.1%), −4.7 to −3.0 (m, 17.7%), 3.8 to 6.2 (m, 23.3%)
−7.74 (s)
−8.2 to −6.0 (m, 65.6%), 0 to 2.5 (m, 34.4%)
−7.2 to −6.8 (m, 80.5%), 0 to 2.5 (m, 19.5%)
4.00 (br, s)
b
−
−
−1.7
OBu-n
−108.7
−81.8
−93.2
−87.7
−79.2
−70.6
−74.2
−73.9
3.13−3.32 (m)
3.13−3.24 (m)
2.85−3.01
18a
20
−
659.2
778.0
−
−
211.0
OMEE
15
−
19
3.00−3.16 (m)
3.00−3.16 (m)
2.85−2.95 (m)
19a
21
−8.5 to −5.0 (m, 66.9%), −4.0 to −2.5 (m, 16.4%), 4.3 to 6.2 (m, 23.3%)
a
b
Only the methyl groups attached to backbones are shown here. Polymer 13 recovered from treatment of polymer 17 with sodium iodide.
Scheme 3. Treatment of Quaternized Phosphazenes with Sodium Iodide
susceptible to the strong nucleophilicity of iodide ion.
Depending on the structures of the side groups, dealkylation
of the quaternary nitrogen atoms or of the side-chain alkyl
groups was detected. The progress of the elimination reactions
was monitored with 31P NMR spectra. For the phenoxy
substituted quaternized trimer 6 or polymer 17 iodide ion
attacks the methyl groups attached to the skeletal nitrogen
atoms to regenerate unquaternized trimer 2 or polymer 13.
This is due to the inertness of sp2 carbon atoms to
nucleophiles. For alkoxy substituted quaternized species 7, 8,
18, and 19, iodide ion attacks the α-carbon atoms of the side
groups to form oxophosphazanes 10, 11, 20, and 21, by
following the mechanism proposed in Scheme 3. (see Table 1
and Table 2 for structural and physical properties and Figure 2
for 31P NMR spectra) The molecular weights of the resultant
polymers 13 (product by elimination), 20 and 21 were lower
than the corresponding unquaternized parent polymers 13, 14,
and 15. This suggests that scission of the polymer backbone
occurred during either the quaternization or the elimination
reactions. To further investigate the mechanism of the
rearrangements, unquaternized phosphazenes were treated
with iodomethane. Trimer 2 was stable in the presence of 10
equiv of iodomethane at 100 °C for 2 days. In the presence of
iodomethane, trimer 4 underwent rearrangement to form
oxophosphazane 11 in 37% yield after being stirred at 60 °C for
24 h. Polymer 15 became an insoluble elastomer after
treatment with iodomethane at 60 °C, probably due to cross-
linking induced by the elevated temperature.
Figure 2. 31P NMR spectra of dealkylization products: (A) trimer 10
and (B) polymer 20.
values that are lower than imidazolium based ionic liquids with
trifluoromethanesulfonate counterion, which usually show
conductivities above 10−3 S·cm−1.41 This can be explained by
the large sizes of the cations and the high viscosity. Polymer 17
has a low ionic conductivity of 2.53 × 10−9 S·cm−1, due to its
high Tg. No further research was carried out on these
compounds.
The ionic conductivities of solvent-free quaternized poly-
alkoxyphosphazenes and their salt complexes with lithium
trifluoromethanesulfonate and silver trifluoromethanesulfonate
are shown in Table 3. For the pure quaternized polymers, a
similar trend in conductivity with increasing degree of
quaternization was found for all the polymers with same side
Ionic Conductivity and Thermal Analysis. The ionic
conductivities of quaternized trimers 6, 7, and 8 are 3.36 × 10−5
S·cm−1, 9.46 × 10−4 S·cm−1 and 6.21 × 10−4 S·cm−1 at 25 °C,
1187
dx.doi.org/10.1021/ma202619j | Macromolecules 2012, 45, 1182−1189

Macromolecules
Article
ionic conductivities, compared to pure polymers 18 and 19
which have same ion concentrations, and even compared to
polymers 18a and 19a which have lower ion concentrations.
(entry 2−4 and entry 10−12).
Table 3. Conductivities and Tg’s of Quaternized
Polyphosphazenes
−
a
OTf /
conductivity
−1
entry polymer salt (mol %)
ru
(S·cm )
Tg (°C)
Tg’s for polyelectroytes and polyelectrolyte/salt complexes
were measured by differential scanning calorimetry (DSC). All
the salt/polyelectrolyte complexes underwent increases in Tg.
(See Table 3.) Metal ions coordinate to the polymers to form
ionic cross-links. These cross-links hinder the motion of the
polymer side chains, and this is reflected in the Tg’s. The
elevated Tg’s conform to the trend of lower conductivity of
polyelectrolyte/salt complexes.
The higher conductivity and lower Tg of backbone
quaternized polyelectrolytes compared to conventional poly-
mer electrolytes can be attributed to the following factors. First,
the shielding effect by side chains linked to the backbone
immobilizes cations and hinders ion pairing, whereas cations
and anions are more likely to pair with each other in
conventional polymer electrolytes. Second, Coulombic repul-
sion between side-chain shielded cations is weaker compared to
that in polymer electrolytes, which could help to maintain high
segmental flexibility. Furthermore, ionic cross-linking is avoided
or minimized in backbone quaternized polymers as there are no
metal ions in the backbone quaternized species.
1
18a
18
−
0.13
0.21
0.21
0.21
0.26
0.31
0.26
0.31
0.12
0.19
0.19
0.19
0.24
0.29
0.24
0.29
1.72 × 10−4
7.45 × 10−5
9.88 × 10−5
2.14 × 10−4
8.21 × 10−5
4.44 × 10−5
9.59 × 10−5
5.34 × 10−5
1.55 × 10−4
1.23 × 10−4
2.09 × 10−4
2.58 × 10−4
1.92 × 10−4
1.28 × 10−4
2.04 × 10−4
1.49 × 10−4
−93.2
−90.7
−75.8
−81.8
−77.4
−74.8
−71.0
−67.0
−74.2
−69.9
−70.7
−70.8
−67.6
−63.1
−66.9
−61.9
2
LiOTf (8%)
AgOTf (8%)
−
LiOTf (5%)
LiOTf (10%)
AgOTf (5%)
AgOTf (10%)
−
3
4
5
6
7
8
9
19a
19
10
11
12
13
14
15
16
LiOTf (7%)
AgOTf (7%)
-
LiOTf (5%)
LiOTf (10%)
AgOTf (5%)
AgOTf (10%)
a
Repeating unit.
group. Even if only the anions are mobile, the conductivities of
solvent-free quaternized polymers 18 (entry 4) and 19 (entry
12) are higher than that of a MEEP/silver trifluoromethanesul-
fonate complex (1.68 × 10−4 S·cm−1, 12 mol %), or a MEEP/
lithium trifluoromethanesulfonate complex (2.8 × 10−5 S·cm−1,
17 mol %) at similar temperature.42 The temperature
dependent conductivities of polymers 18 and 19 are illustrated
in Figure 3. The highest conductivities of polymer 19 of 2.58 ×
SUMMARY
■
The quaternization of poly(alkoxyphosphazenes) and poly-
[(bisphenoxy)phosphazene] has been accomplished. Because
of the lower electron donating ability of alkoxy and aryloxy side
groups than in the case of alkylamino side groups, efficient
quaternization could only be achieved with methyl trifluome-
thanesulfonate as the reagent. This is the first report of
backbone functionalization of phosphazenes that bear P−O
linked side groups at both the small molecule and high polymer
levels. However, the quaternized poly(alkoxyphosphazenes) are
sensitive to dealkylation in the presence of nucleophilic iodide
anions.
These new backbone quaternized poly(alkoxyphosphazenes)
have high ionic conductivity without plasticizers or additional
salts. This offers the potential for insight into the mechanism of
ion transport and for the applications of polymers in solid-state
electrochemical devices. The new concept of immobilizing ions
onto or near to a polymer backbone may have several
advantages over conventional polymer/salt complexes or side
chain functionalized polyelectrolytes. The corresponding
quaternization of poly(alkylaminophosphazenes) to improve
the stability of polyelectrolytes is currently ongoing.
AUTHOR INFORMATION
■
Corresponding Author
Figure 3. Arrhenius plots for the conductivity of solvent/salt-free
polymers 18 and 19.
*E-mail: hra@chem.psu.edu.
Notes
10−4 S·cm−1 at 25 °C and 2.09 × 10−3 S·cm−1 at 80 °C were
achieved, which are among the highest conductivities of
unplasticized solid-state polymer electrolytes or polyelectro-
lytes. To further compare the effect of backbone-quaternization
versus simple salt-complexing on conductivity, polyelectrolyte/
salt complexes were formulated by drying an acetone solution
of a polyelectrolyte and a salt under vacuum. Additional salt
applied to polymers 18 or 19 resulted in decreased ionic
conductivities (entry 5−8 and entry 13−16). Polyelectrolyte/
salt complexes formulated with polymer 18a and 19a had lower
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the U.S. Department of Energy for financial support
(Grant DE-FG36-08GO18011) and Dr. Kirk Marat (Depart-
ment of Chemistry, University of Manitoba), Dr. Wenbin Luo
and Dr. Alan Benesi (Department of Chemistry, The
Pennsylvania State University) for discussions on 31P NMR
spectra.
1188
dx.doi.org/10.1021/ma202619j | Macromolecules 2012, 45, 1182−1189

Macromolecules
Article
(37) Feakins, D.; Shaw, R. A.; Last, W. A. J. Chem. Soc. 1964, 2387−
2396.
(38) Feakins, D.; Shaw, R. A.; Last, W. A. J. Chem. Soc. 1964, 4464−
4471.
(39) Feakins, D.; Last, W. A.; Neemuchw, N.; Shaw, R. A. J. Chem.
Soc. 1965, 2804−2811.
(40) Dautzenberg, H.; Jaeger, W.; Kotz, J.; Philipp, B.; Seidel, C.;
Stscherbina, D., Polyelectrolytes: formation, characterization, and
application. Hanser: New York, 1994; pp 233−247.
(41) Holbrey, J. D.; Rogers, R. D., In Ionic liquids in synthesis, 2nd ed.;
Wasserscheid, P.; Welton, T., Eds. Wiley-VCH: Weinheim, Germany,
2008; Vol. 1, p 149.
REFERENCES
■
(1) Fenton, D. E.; Parker, J. M.; Wright, P. V. Polymer 1973, 14, 589.
(2) Gray, F., Solid Polymer Electrolytes. VCH Publishers: New York,
1991.
(3) Oregan, B.; Gratzel, M. Nature 1991, 353, 737−740.
(4) Hagfeldt, A.; Boschloo, G.; Sun, L. C.; Kloo, L.; Pettersson, H.
Chem. Rev. 2010, 110, 6595−6663.
(5) Allcock, H. R., Chemistry and applications of polyphosphazenes.
John Wiley & Sons, Inc: Hoboken, NJ, 2003.
(6) Blonsky, P. M.; Shriver, D. F.; Austin, P.; Allcock, H. R. J. Am.
Chem. Soc. 1984, 106, 6854−6855.
(7) Allcock, H. R.; Austin, P. E.; Neenan, T. X.; Sisko, J. T.; Blonsky,
P. M.; Shriver, D. F. Macromolecules 1986, 19, 1508−1512.
(8) Allcock, H. R.; Pucher, S. R.; Turner, M. L.; Fitzpatrick, R. J.
Macromolecules 1992, 25, 5573−5577.
(42) Allcock, H. R.; Napierala, M. E.; Olmeijer, D. L.; Best, S. A.;
Merz, K. M. Macromolecules 1999, 32, 732−741.
(9) Ganapathiappan, S.; Chen, K.; Shriver, D. F. Macromolecules
1988, 21, 2299−2301.
(10) Ganapathiappan, S.; Chen, K. M.; Shriver, D. F. J. Am. Chem.
Soc. 1989, 111, 4091−4095.
(11) Allcock, H. R.; Fitzpatrick, R. J.; Salvati, L. Chem. Mater. 1991, 3,
1120−1132.
(12) Allcock, H. R.; Welna, D. T.; Maher, A. E. Solid State Ionics
2006, 177, 741−747.
(13) Klein, R. J.; Welna, D. T.; Weikel, A. L.; Allcock, H. R.; Runt, J.
Macromolecules 2007, 40, 3990−3995.
(14) Wisian-Neilson, P.; Garciaalonso, F. J. Macromolecules 1993, 26,
7156−7160.
(15) Luther, T. A.; Stewart, F. F.; Budzien, J. L.; LaViolette, R. A.;
Bauer, W. F.; Harrup, M. K.; Allen, C. W.; Elayan, A. J. Phys. Chem. B
2003, 107, 3168−3176.
(16) Chandrasekhar, V.; Krishnan, V., In Phosphazenes: A worldwide
insight, Gleria, M.; De Jaeger, R., Eds. Nova Science Pub Inc: New
York, 2004; p 827.
(17) Heston, A. J.; Panzner, M. J.; Youngs, W. J.; Tessier, C. A. Inorg.
Chem. 2005, 44, 6518−6520.
(18) Zhang, Y.; Huynh, K.; Manners, I.; Reed, C. A. Chem. Commun.
2008, 494−496.
(19) Tun, Z. M.; Heston, A. J.; Panzner, M. J.; Medvetz, D. A.;
Wright, B. D.; Savant, D.; Dudipala, V. R.; Banerjee, D.; Rinaldi, P. L.;
Youngs, W. J.; Tessier, C. A. Inorg. Chem. 2011, 50, 8937−8945.
(20) Benson, M. A.; Zacchini, S.; Boomishankar, R.; Chan, Y.;
Steiner, A. Inorg. Chem. 2007, 46, 7097−7108.
(21) Benson, M. A.; Boomishankar, R.; Wright, D. S.; Steiner, A. J.
Organomet. Chem. 2007, 692, 2768−2772.
(22) Benson, M. A.; Steiner, A. Chem. Commun. 2005, 5026−5028.
(23) Rapko, J. N.; Feistel, G. Inorg. Chem. 1970, 9, 1401−1405.
(24) Searle, H. T.; Dyson, J.; Ranganathan, T. N.; Paddock, N. L. J.
Chem. Soc., Dalton Trans. 1975, 203−208.
(25) Calhoun, H. P.; Trotter, J. J. Chem. Soc., Dalton Trans. 1974,
377−381.
(26) Allcock, H. R.; Levin, M. L.; Austin, P. E. Inorg. Chem. 1986, 25,
2281−2288.
(27) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518−1520.
(28) Allcock, H. R.; Kugel, R. L. J. Am. Chem. Soc. 1965, 87, 4216−
4217.
(29) Allcock, H. R.; Walsh, E. J. J. Am. Chem. Soc. 1972, 94, 4538−
4545.
(30) Fitzsimmons, B. W.; Shaw, R. A. J. Chem. Soc. 1964, 1735−1741.
(31) Fei, S. T.; Allcock, H. R. J. Power Sources 2010, 195, 2082−2088.
(32) Allcock, H. R.; Walsh, E. J. J. Am. Chem. Soc. 1972, 94, 119−124.
(33) Allcock, H. R.; Kugel, R. L.; Valan, K. J. Inorg. Chem. 1966, 5,
1709−1715.
(34) Allcock, H. R.; Connolly, M. S.; Sisko, J. T.; Alshali, S.
Macromolecules 1988, 21, 323−334.
(35) Allcock, H. R. Acc. Chem. Res. 1979, 12, 351−358.
(36) Fitzsimmons, B. W.; Hewlett, C.; Shaw, R. A. J. Chem. Soc. 1964,
4459−4464.
1189
dx.doi.org/10.1021/ma202619j | Macromolecules 2012, 45, 1182−1189