Alkaline Chemical Stability of Polymerized Ionic Liquids with Various Cations

Article abstract of DOI:10.1021/acs.macromol.5b01223

The success of long-lasting low-cost (nonplatinum) alkaline fuel cells is dependent on the development of anion exchange membranes (electrolyte separator) with high alkaline chemical stability. In this study, a series of methacrylate-based polymerized ionic liquids (PILs) were synthesized with various covalently attached cations: butylimidazolium, butylmethylimidazolium, trimethylammonium, pentamethylguanidinium, butylpyrrolidinium, and trimethylphosphonium. The alkaline chemical stability of these PILs was examined in tandem with their analogous ionic salts: 1-butyl-3-methylimidizolium chloride, 1-butyl-2,3-dimethylimidazolium chloride, tetramethylammonium chloride, benzyltrimethylammonium chloride, hexamethylguanidinium chloride, 1,1-butylmethylpyrrolidinium chloride, and tetramethylphosphonium chloride. The degradation mechanisms and extent of degradation were quantified using ¹H NMR spectroscopy at various pHs (in D₂O), and temperature. The PILs with imidazolium and pyrrolidinium cations showed enhanced chemical stability relative to the PILs with ammonium and phosphonium cations. Interestingly, direct correlations were not observed between the PILs and their analogous small molecule ionic salts; significant degradation was observed in imidazolium ionic salts, most notably at high temperature/high pH conditions, while the pyrrolidinium-, ammonium-, and phosphonium-based ionic salts showed no degradation under any of the conditions examined. Additionally, results on the imidazolium ionic salts showed that methyl substitution in the C2 position limited the ring-opening degradation reaction, whereas the PIL with the unsubstituted imidazolium actually showed higher chemical stability relative to its substituted PIL counterpart. Overall, the alkaline chemical stability of the PILs in this study showed no correlation to that of their analogous small molecule ionic salts, suggesting that alkaline chemical stability studies on small molecules may not provide a solid basis for evaluating alkaline stability in polymers, counter to the hypothesis in many previous studies.

Full text of DOI:10.1021/acs.macromol.5b01223

Article
pubs.acs.org/Macromolecules
Alkaline Chemical Stability of Polymerized Ionic Liquids with Various
Cations
Kelly M. Meek and Yossef A. Elabd*
Department of Chemical Engineering, Texas A&M University, College Station, Texas 77843, United States
*
S Supporting Information
ABSTRACT: The success of long-lasting low-cost (non-
platinum) alkaline fuel cells is dependent on the development
of anion exchange membranes (electrolyte separator) with
high alkaline chemical stability. In this study, a series of
methacrylate-based polymerized ionic liquids (PILs) were
synthesized with various covalently attached cations: butyl-
imidazolium, butylmethylimidazolium, trimethylammonium,
pentamethylguanidinium, butylpyrrolidinium, and trimethyl-
phosphonium. The alkaline chemical stability of these PILs
was examined in tandem with their analogous ionic salts: 1-butyl-3-methylimidizolium chloride, 1-butyl-2,3-dimethylimidazolium
chloride, tetramethylammonium chloride, benzyltrimethylammonium chloride, hexamethylguanidinium chloride, 1,1-
butylmethylpyrrolidinium chloride, and tetramethylphosphonium chloride. The degradation mechanisms and extent of
1
degradation were quantified using H NMR spectroscopy at various pHs (in D O), and temperature. The PILs with imidazolium
2
and pyrrolidinium cations showed enhanced chemical stability relative to the PILs with ammonium and phosphonium cations.
Interestingly, direct correlations were not observed between the PILs and their analogous small molecule ionic salts; significant
degradation was observed in imidazolium ionic salts, most notably at high temperature/high pH conditions, while the
pyrrolidinium-, ammonium-, and phosphonium-based ionic salts showed no degradation under any of the conditions examined.
Additionally, results on the imidazolium ionic salts showed that methyl substitution in the C2 position limited the ring-opening
degradation reaction, whereas the PIL with the unsubstituted imidazolium actually showed higher chemical stability relative to its
substituted PIL counterpart. Overall, the alkaline chemical stability of the PILs in this study showed no correlation to that of their
analogous small molecule ionic salts, suggesting that alkaline chemical stability studies on small molecules may not provide a solid
basis for evaluating alkaline stability in polymers, counter to the hypothesis in many previous studies.
13−19
INTRODUCTION
conditions.
However, BTMA is known to degrade under
■
alkaline conditions over time by nucleophilic substitution, thus
leading to the consideration of other cations that may show
improved alkaline chemical stability. Alternative AEMs with
Alkaline fuel cells (AFCs) employing solid-state anion exchange
membranes (AEMs) as electrolytes are of great interest, as they
produce high power densities at low operating temperatures
2
−12
23−25
other cations, such as imidazolium,
phosphonium,
(<200 °C) and enable the use of non-platinum electrodes (e.g.,
26−29
30
guanidinium,
and pyrrolidinium, have recently been
nickel), significantly reducing cost relative to proton exchange
membrane fuel cells. Recently, a number of AEMs have been
developed for the AFC.
limiting the wide scale use of solid-state AFCs is the alkaline
chemical stability of the AEM. Degradation of the covalently
1
investigated. Specifically, many recent stability studies on AEMs
for AFCs have focused on monitoring ionic conductivity over a
designated period of time after an AEM is immersed in a
concentrated alkaline solution, as a means of measuring
chemical stability of the AEM under specific test condi-
2
−30
One of the critical challenges
tethered cationic groups, as well as the polymer backbone, ma−y
5
−7,20,24
20
tions.
Hibbs compared various benzylic cations on the
be triggered by the high nucleophilicity and basicity of OH
polyphenylene backbone: BTMA, pentamethylguanidinum, and
N-methylimidazolium. The resonance stabilized cations,
benzylic pentamethylguanidinium and N-methylimidazolium,
showed >25% decrease in ionic conductivity and ion exchange
capacity (IEC) after 1 day in 4 M KOH at 90 °C, while the
BTMA showed ∼33% decrease after 14 days. No change in IEC
and only ∼5% decrease in conductivity were observed under
ions, which are produced in an AFC and transported through
the AEM. Although degradation of the polymer backbone
should be considered, it is generally accepted as more
chemically stable than the cation. Further investigations are
necessary to determine the most promising cations for long-
term stability in the presence of hydroxide in order to achieve
long-lasting AFC performance.
The benzyltrimethylammonium (BTMA) cation is the most
frequently employed cation in AEMs due to ease of
functionalization, high conductivity and thermal stability, and
adequate alkaline stability on short time scales in hydrated
Received: June 5, 2015
Revised: September 9, 2015
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 1. Chemical Structures of PIL Homopolymers with Various Pendant Cations
Scheme 2. Chemical Structures of Ionic Salts
the same conditions for 14 days when a hexamethylene spacer
was included between the polymer backbone and BTMA,
demonstrating the importance of not only cation type but also
how the cation is attached to the polymer backbone. Coates
styrene-based polymer; samples were degraded in hydroxide
concentrations measured in molar equivalents, a useful
technique for comparing studies, and a practice that we have
adopted for our current study. Arges and Ramani used two-
multidimensional NMR correlation spectroscopy to report on
the chemical stability of five different cations tethered to a PSF
backbones: BTMA, 1,4-dimethylpiperazinium, trimethyl-
phosphonium, 1-methylimidazolium, tris(2,4,6-trimethoxy-
phenyl)phosphonium. BTMA was more chemically stable
than all but 1,4-dimethylpiperazinium, which could potentially
21
2
4
and co-workers showed the potential benefits of a sterically
crowded cation by exploring the stability of tetrakis(dialkyl-
amino)phosphonium-functionalized polyethylene relative to
BTMA-functionalized polyethylene. The bulky phosphonium-
based cation outperformed the BTMA cation, maintaining its
conductivity over a 20 week period in 15 M KOH at 22 °C and
experiencing only a small decrease in conductivity (from 22 to
be attributed to its lower pK value or steric hindrance effects,
A
−
1
1
8 mS cm ) over 22 days in 1 M KOH at 80 °C. Prior to this
while trimethylphoshonium and 1-methylimidazolium degraded
rapidly in alkaline media. This technique of comparing several
cations on the same backbone within one study is useful, as
polymer backbone type may contribute to degradation; thus,
changes in backbone from study to study may lead to
inconsistencies when comparing cation stability.
25
study, Yan and co-workers found that an AEM with a bulky
phosphonium-based cation, tris(2,4,6-trimethoxyphenyl)-
polysulfone−methylene quaternary phosphonium hydroxide,
maintained its high ionic conductivity for 48 h in 10 M KOH at
room temperature and in 1 M KOH at 60 °C.
This approach of monitoring ionic conductivity and IEC
under alkaline conditions has provided valuable insights
regarding AEM chemical stability. Further insights into
chemical stability of the covalently attached cation have also
recently been achieved in the literature through the use of
In order to more accurately assess cation stability, it is
necessary to compare several cation types on the same polymer
backbone under the same degradation conditions. While a few
20,21,30
additional studies have used this approach,
to date, no
studies to our knowledge have quantified the alkaline chemical
degradation of AEMs with various covalently linked cation
types in tandem with their analogous small molecule ionic salts
to definitively determine if there is a correlation. One recent
1
proton nuclear magnetic resonance ( H NMR) spectroscopy to
quantify degradation and identify degradation mechanisms of
small molecule ionic salt analogues exposed to alkaline
9
,10,17,30,31
30
1
solutions.
Analogous salts are investigated instead of
study by Yan and co-workers was successful in using H
NMR to quantify the degradation of several pyrrolidinium salts
alongside two differently substituted pyrrolidinium-based
AEMs and showed the more stable small molecule to also be
the more stable polymeric cation. Additional studies attempted
a similar approach to examine polymer stability but were unable
to draw definitive conclusions for polymeric cation degradation
due to commonly faced issues of polymer gelation and H/D
1
AEMs because the polymer backbone typically weakens the H
NMR signal, making it difficult to determine the mechanism of
cation degradation. Additionally, the degraded polymer may no
longer be soluble in the same solvent as the undegraded
1
polymer, therefore making degradation analysis via H NMR
inaccurate. A few studies have successfully quantified AEM
1
4,13,21,22,30
degradation with the use of H NMR spectroscopy;
9
,10
conditions for degradation experiments (e.g., temperature,
solvent, hydroxide concentration) varied by study such that
relative cation stability between studies is difficult to assess.
exchange.
The usefulness of past and future small molecule
ionic salt degradation studies as tools for predicting relative
cation stability in AEMs is dependent upon proof of a
correlation between free cation and polymeric cation chemical
stability.
In this study, we synthesized polymerized ionic liquids
(PILs) containing various covalently attached cations (Scheme
1): butylimidazolium, butylmethylimidazolium, trimethyl-
ammonium, pentamethylguanidinium, butylpyrrolidinium, and
trimethylphosphonium. The alkaline chemical stabilities of
32
Holdcroft and co-workers found poly[2,2′-(m-mesitylene)-
5
,5′-bis(N,N′-dimethylbenzimidazolium)], an AEM that con-
tains its benzimidazolium cation in the polymer backbone, to
show no degradation over 10 days in 2 M KOH at 60 °C when
1
monitoring the chemical structure via H NMR. Nun
̃
ez and
1
3
1
Hickner used H NMR to examine the half-life of BTMA on
various backbones, finding it to be particularly stable on a
B
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
a
these PILs were investigated and compared to their analogous
ionic salts (Scheme 2): 1-butyl-3-methylimidizolium chloride,
Scheme 3. Synthesis of PIL Homopolymer, Poly(MEX-Br);
X = Various Cations (a−f)
1
-butyl-2,3-dimethylimidazolium chloride, tetramethyl-
ammonium chloride, benzyltrimethylammonium chloride,
hexamethylguanidinium chloride, 1,1-butylmethylpyrrolidinium
chloride, and tetramethylphosphonium chloride. In order to
overcome the aforementioned difficulties in quantifying cation
1
degradation in polymers via H NMR (e.g., polymer
precipitation, weakening of NMR signal by polymer backbone,
hydrogen/deuterium exchange), polymers were degraded in
D O as an alternative to the harsher standard degradation
2
solvent methanol, the PIL chemistry provided high water
solubility, such that a significant amount of polymer could be
1
dissolved in D O for H NMR analysis to enable integration,
2
and small amounts of H O were used to suppress H/D
2
a
(
1) AIBN, DMF, 60 °C, 2 h; (2a) 1-butylimidazole, DMF, 80 °C, 48
exchange where necessary. The degradation pathways and
extent of chemical stability of the PILs and ionic salts were
h; (2b) 1-butyl-2-methylimidazole, DMF, 80 °C, 48 h; (2c) aqueous
trimethylamine, DMF, room temperature, 48 h; (2d) 1,1,2,3,3-
pentamethylguanidine (PMG), DMF, room temperature, 24 h; (2e)
1
determined using H NMR spectroscopy after exposure to 2−
20 mol equiv KOH per cation at 30 or 80 °C for 168 h. This
1
-butylpyrrolidine, DMF, 80 °C, 24 h; (2f) trimethylphosphine/THF
tandem study of AEMs alongside analogous small molecule
cations establishes whether small molecule studies provide
insight into alkaline chemical stability behavior of polymeric
cations. Additionally, these results provide a consistent measure
of relative alkaline chemical stability of several major cation
types.
solution, DMF, room temperature, 24 h.
EXPERIMENTAL SECTION
■
≥
Materials. N,N-Dimethylformamide (DMF, ACS Reagent,
99.8%), 1-butylimidazole (98%), trimethylamine solution (45 wt %
in H O), 1-butylpyrrolidine (98%), trimethylphosphine solution (1 M
2
in THF), tetramethylammonium chloride (≥98%), benzyltrimethyl-
ammonium chloride (97%), tetramethylphosphonium chloride (98%),
oxalyl chloride (reagent grade, 98%), ethanol, tetramethylurea, 1,2-
dichloroethane (anhydrous, 99.8%), methylamine solution (33 wt % in
absolute ethanol), magnesium sulfate (anhydrous, ReagentPlus, 99%),
diethyl ether (anhydrous, ≥99.7%, contains 1 ppm BHT inhibitor),
hexane (ACS Reagent, ≥98.5%), tetrahydrofuran (THF, ≥99.9%),
methanol (ACS Reagent, ≥99.8%,), potassium hydroxide (KOH,
≥
90%, reagent grade), deuterium oxide (D O, 99.98 atom % D), and
2
dimethyl-d sulfoxide (DMSO-d , 99.9 atom % D, contains 0.03% v/v
6
6
TMS) were used as received from Sigma-Aldrich. 1-Butyl-2-
methylimidazole (>98%), 1-butyl-3-methylimidazolium chloride
(
>99%), 1-butyl-2,3-dimethylimidazolium chloride (>99%), and 1-
butyl-1-methylpyrrolidinium chloride (99%) were used as received
from Iolitec. Hexamethylguanidinium chloride (98%) was used as
received from Fisher Scientific. Azobis(isobutyronitrile) (AIBN, 98%,
Sigma-Aldrich) was purified by recrystallization twice from methanol.
Synthesis of Poly(BrEMA). Synthesis of the nonionic precursor
monomer 2-bromoethyl methacrylate (BrEMA) was performed
according to a procedure in the literature. The synthesis of the
nonionic precursor homopolymer, poly(BrEMA), was performed
using conventional free-radical polymerization as shown in Scheme
33
1
Figure 1. H NMR spectra for (I) poly(BrEMA), (II) poly(MEBIm-
Br), (III) poly(MEBMIm-Br), (IV) poly(METMA-Br), (V) poly-
(MEBP-Br), and (VI) poly(METMP-Br).
3
(1). A typical example is given as follows. 5.0 g (25.9 mmol) of
BrEMA monomer in DMF (BrEMA/DMF 1/1 w/w) and 22.3 mg
0.129 mmol) of AIBN (BrEMA/AIBN 200/1 mol/mol) were mixed
Synthesis of Poly(MEBIm-Br). Synthesis of the PIL homopol-
ymer, poly(MEBIm-Br) [MEBIm-Br = 1-[(2-methacryloyloxy)ethyl]-
3-butylimidazolium bromide], is shown in Scheme 3(2a), i.e.,
functionalization of nonionic precursor homopolymer, 2-bromoethyl
methacrylate, to form an ionic homopolymer. A typical example is
given as follows. 0.200 g (1.03 mmol) of poly(BrEMA) was first
dissolved in ∼2 mL of DMF in a 50 mL vial. 0.642 g (5.17 mmol) of 1-
butylimidazole (poly(BrEMA)/1-butylimidazole, 1/5 mol/mol) was
then mixed into the vial. The solution was stirred at 80 °C for 48 h.
The resulting polymer was precipitated twice into hexane and dried
(
in a 200 mL round-bottom flask and reacted under N for 2 h at 60 °C.
2
The resulting polymer was precipitated into methanol twice and dried
under vacuum in an oven at room temperature for 24 h. Yield: 0.85 g
1
of solid particles (17.0%). H NMR (500 MHz, CDCl , 23 °C) δ
3
(
ppm): 4.47−4.19 (s, 2H, O−CH −CH −Br), 3.66−3.47 (s, 2H,O−
2
2
CH −CH −Br), 2.12−1.84 (d, 3H, CH −C(CH )), 1.12 (s, 1H,
2
2
2
3
HCH−C(CH )), 0.98 (s,1H, HCH−C(CH )) (NMR, Figure 1). SEC
3
3
−1
(
THF, 40 °C): M = 59.15 kg mol , M /M = 1.85 (against PS
n w n
standards). Three batches of poly(BrEMA) were synthesized (0.85,
.83, and 1.13 g with yields of 17.0, 16.6, and 18.8%, respectively) to
produce the PILs in this study with similar M and PDIs.
under vacuum in an oven at room temperature for 24 h. Yield: 0.267 g
1
0
(0.841 mmol) of solid particles (81.2%). H NMR (500 MHz, D O,
2
23 °C) δ (ppm): 9.38−8.76 (s,1H, N−CHN), 7.98−7.17 (s, 2H,
n
C
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
N−CHCH−N), 4.54−3.91 (m, 6H, N−CH −CH −O, N−CH −
3(2d). A typical example is given as follows. 0.400 g (2.07 mmol) of
poly(BrEMA) was first dissolved in ∼2 mL of DMF in a 50 mL vial.
0.486 g (3.77 mmol) of 1,1,2,3,3-pentamethylguanidine (PMG)
(poly(BrEMA)/PMG, 1/2 mol/mol) was then mixed into the vial.
The solution was stirred at room temperature for 24 h. The resulting
polymer was precipitated twice into diethyl ether and dried under
2
2
2
CH −CH −CH ), 1.89 (s, 5H, CH −C(CH ), N−CH −CH −CH −
2
2
3
2
3
2
2
2
CH ), 1.34 (s, 3H, N−CH −CH −CH −CH , HCH−C(CH )), 0.93
3
2
2
2
3
3
(
1
s, 4H, N−CH −CH −CH −CH , HCH−C(CH )) (NMR, Figure
2
2
2
3
3
).
Synthesis of Poly(MEBMIm-Br). Synthesis of the PIL homopol-
ymer, poly(MEBMIm-Br) [MEBMIm-Br = 1-[(2-methacryloyloxy)-
ethyl]-3-butylmethylimidazolium bromide], is shown in Scheme
vacuum in an oven at room temperature for 24 h. Yield: 0.53 g (1.708
1
mmol) of solid particles (82.5%). H NMR (500 MHz, DMSO-d , 23
6
3
(2b). A typical example is given as follows. 0.200 g (1.03 mmol) of
poly(BrEMA) was first dissolved in ∼2 mL of DMF in a 50 mL vial.
.715 g (5.18 mmol) of 1-butyl-3-methylimidazole (poly(BrEMA)/1-
°C) δ (ppm): 3.55−3.19 (s, 12H, C−(N−(CH ) ) ) (NMR, Figure
3
2 2
S1).
0
Synthesis of Poly(MEBP-Br). Synthesis of the PIL homopolymer,
poly(MEBP-Br) [MEBP-Br = 1-[(2-methacryloyloxy)ethyl]-1-butyl-
pyrrolidinium bromide], is shown in Scheme 3(2e). A typical example
is given as follows. 0.500 g (2.59 mmol) of poly(BrEMA) was first
dissolved in ∼2 mL of DMF in a 50 mL vial. 0.989 g (7.77 mmol) of 1-
butylpyrrolidine (poly(BrEMA)/1-butylpyrrolidine, 1/3 mol/mol)
was then mixed into the vial. The solution was stirred at 80 °C for
24 h. The resulting polymer was precipitated twice into hexane and
dried under vacuum in an oven at room temperature for 24 h. Yield:
butyl-3-methylimidazole, 1/5 mol/mol) was then mixed into the vial.
The solution was stirred at 80 °C for 48 h. The resulting polymer was
precipitated twice into hexane and dried under vacuum in an oven at
room temperature for 24 h. Yield: 0.261 g (0.788 mmol) of solid
1
particles (76.5%). H NMR (500 MHz, D O, 23 °C) δ (ppm): 7.75−
2
7
.37 (s, 2H, N−CHCH−N), 4.68−4.01 (m, 6H, N−CH −CH −O,
2
2
N−CH −CH −CH ), 3.28−3.58 (s, 3H, N−C−CH N), 1.91 (s,
2
2
2
3
4
H, CH −C(CH ), N−CH −CH −CH −CH ), 1.35 (s, 3H, N−
2
3
2
2
2
3
1
CH −CH −CH −CH , HCH−C(CH )), 0.91 (s, 4H, N−CH −
CH −CH −CH , HCH−C(CH )) (NMR, Figure 1).
0.819 g (2.442 mmol) of solid particles (94.3%). H NMR (500 MHz,
2
2
2
3
3
2
D O, 23 °C) δ (ppm): 4.64−4.18 (s, 2H, N−CH −CH −O)), 3.90−
2
2
3
3
2
2
2
Synthesis of Poly(METMA-Br). Synthesis of the PIL homopol-
ymer, poly(METMA-Br) [METMA-Br = 1-[(2-methacryloyloxy)-
ethyl]-trimethylammonium bromide], is shown in Scheme 3(2c). A
typical example is given as follows. 0.730 g (3.78 mmol) of
poly(BrEMA) was first dissolved in ∼2 mL of DMF in a 50 mL
vial. 1.118 g (18.9 mmol) of trimethylamine in aqueous solution
3.45 (s, 6H, N−CH −CH −CH −CH −N, N−CH −CH −O),
2
2
2
2
2
2
2.35−2.02 (s, 4H, N−CH −CH −CH −CH −N), 1.51−1.27 (s,
2
2
2
2
7H, N−CH −CH −CH −CH , CH −C(CH )), 1.08−0.80 (s, 5H,
2
2
2
3
2
3
N−CH −CH −CH −CH , CH −C(CH ) (NMR, Figure 1).
2
2
2
3
2
3
Synthesis of Poly(METMP-Br). Synthesis of the PIL homopol-
ymer, poly(METMP-Br) [METMP-Br = 1-[(2-methacryloyloxy)-
ethyl]-trimethylphosponium bromide], is shown in Scheme 3(2f). A
typical example is given as follows. 0.500 g (2.59 mmol) of
poly(BrEMA) was first dissolved in ∼2 mL of DMF in a 50 mL
vial. 0.591 g (7.77 mmol) of trimethylphosphine in THF solution
(poly(BrEMA)/trimethylphosphine, 1/3 mol/mol) was then mixed
into the vial. The solution was stirred at room temperature for 24 h.
The resulting polymer was precipitated twice into diethyl ether and
dried under vacuum in an oven at room temperature for 24 h. Yield:
(
poly(BrEMA)/trimethylamine, 1/5 mol/mol) was then mixed into
the vial. The solution was stirred at room temperature for 48 h. The
resulting polymer was precipitated twice into diethyl ether and dried
under vacuum in an oven at room temperature for 24 h. Yield: 0.85 g
1
(
3.37 mmol) of solid particles (89.2%). H NMR (500 MHz, D O, 23
2
°C) δ (ppm): 4.64−4.25 (s, 2H, N−CH −CH −O), 3.98−3.62 (s,
2
2
2
3
1
H, N−CH −CH −O), 3.64−2.97 (s, 9H, N−(CH ) ), 2.17−1.86 (s,
2
2
3 3
H, CH −C(CH )), 1.28−0.81 (s, 2H, CH −C(CH )) (NMR, Figure
2
3
2
3
1
).
Synthesis of 1,1,2,3,3-Pentamethylguanidine (PMG). To
0.467 g (1.735 mmol) of solid particles (67.0%). H NMR (500 MHz,
D O, 23 °C) δ (ppm): 4.55−4.14 (s, 2H, P−CH −CH −O), 3.93−
2
2
2
synthesize a PIL homopolymer with pentamethylguanidinium as the
cation, 1,1,2,3,3-pentamethylguanidine was first synthesized (Scheme
3.50 (s, 2H, P−CH −CH −O), 2.50−1.62 (s, 9H, P−(CH ) ), 1.44−
2
2
3 3
0.61 (s, 5H, CH −C(CH )) (NMR, Figure 1).
2
3
4). A 250 mL two-necked flask was charged with 50 mL of dry 1,2-
Characterization. All chemical structures of the PIL homopol-
1
ymers were characterized by H NMR spectroscopy using a Varian 500
MHz spectrometer at 23 °C with D O as the solvent. The chemical
Scheme 4. Synthesis of 1,1,2,3,3-Pentamethylguanidine
2
a
shifts were referenced to water at 4.75 ppm. Chemical structures of
(PMG)
1
poly(BrEMA) and PMG were also characterized by H NMR
spectroscopy and referenced to CDCl at 7.27 ppm. The molecular
3
weights and molecular weight distributions of precursor PIL
homopolymer were determined by size exclusion chromatography
(
SEC) using a Waters GPC system equipped with a THF Styragel
column (Styragel@HR 5E, effective separation of molecular weight
a
−1
(
1) Oxalyl chloride, dichloroethane, 60 °C, 2 h; (2) methylamine/
range: 2−4000 kg mol ) and a 2414 reflective index (RI) detector. All
measurements were performed at 40 °C. THF was used as the mobile
phase at a flow rate of 1.0 mL/min. PS standards (Shodex, Japan) with
molecular weights ranging from 2.97 to 983 kg mol were used for
calibration.
ethanol solution, dropwise, 0 °C, 1 h.
−1
dichloroethane and 4.65 g (40 mmol) of tetramethylurea. Oxalyl
chloride (5.65 mL, 64.8 mmol) was added at room temperature, and
the solution was heated for 2 h at 60 °C. The solvent was removed
under vacuum, the residual yellow solid was dissolved in 20 mL of dry
ethanol, and 33.0 g of 33 wt % methylamine in dry ethanol (11.0 g or
Alkaline Chemical Stability Analysis. The chemical stability of
ionic salts and PIL homopolymers was examined using H NMR
1
spectroscopy with D O as the solvent. The stability study was
2
performed under alkaline conditions in NMR tubes. Ionic salts (25.0
mg) were exposed to different KOH concentrations (2, 20, and 50 mol
3
54.8 mmol of methylamine) was added dropwise at 0 °C. The
1
equiv) in 1 mL of D O at 30 or 80 °C for 168 h, followed by H NMR
reaction mixture was allowed to warm slowly to room temperature,
stirred overnight, and then refluxed for 4 h. The solvent was
evaporated under vacuum, and the residue was treated with 30%
aqueous NaOH. The organic layer was extracted with ether and dried
with anhydrous magnesium sulfate, and the solvent was evaporated.
2
experiments to quantify chemical degradation. PIL homopolymers (8.0
mg) dissolved in 1 mL of D O were exposed to 2 and 20 equiv KOH
2
1
at 30 or 80 °C for 168 h, followed by H NMR experiments. Note that
all KOH concentrations are in excess of 1 equiv to ensure full ion
exchange of ionic salts and PILs.
Distillation of the residue under reduced pressure at 100 °C yielded
1
3
(
.22 g (62.3%) of PMG. H NMR (500 MHz, CDCl , 23 °C) δ
3
ppm): 2.90 (s, 3H), 2.80 (s, 6H) and 2.66 (s, 6H) (NMR, Figure S1).
RESULTS AND DISCUSSION
■
Synthesis of Poly(MEPMG-Br). Synthesis of the PIL homopol-
ymer, poly(MEPMG-Br) [MEPMG-Br = 1-[(2-methacryloyloxy)-
ethyl]-pentamethylguanidinium bromide], is shown in Scheme
Ionic Salt Alkaline Chemical Stability. In order to
investigate the alkaline stability of PILs with covalently tethered
D
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
E
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
cations, the analogous small molecule model compounds were
20 equiv and 80 °C, degradation is further increased to 98.7%
and 7.5% (compared with 20.0% and 2.0% at 30 °C) for
first studied: 1-butyl-3-methylimidizolium chloride
+
−
+
−
+
−
+
−
(
(
BMIm Cl ), 1-butyl-2,3-dimethylimidazolium chloride
BMIm Cl and BDMIm Cl , respectively. BDMIm Cl
+
−
+
−
BDMIm Cl ), tetramethylammonium chloride (TMA Cl ),
benzyltrimethylammonium chloride (BTMA Cl ), hexamethyl-
guanidinium chloride (HMG Cl ), 1,1-butylmethyl-
pyrrolidinium chloride (BMP Cl ), and tetramethyl-
appears to be the more chemically stable imidazolium salt at
high temperature as well as low temperature.
+
−
+
−
1
Figures 2C,D and 3C,D show H NMR spectra for
+
−
+
−
+
−
ammonium-based ionic salts, TMA Cl and BTMA Cl , after
+
−
phosphonium chloride (TMP Cl ) (Scheme 2). The purity
168 h exposure to KOH/D O at 30 and 80 °C, respectively.
2
and chemical structure of these ionic salts were confirmed by
Though cation peaks do shift as expected after addition of
hydroxide, no other changes were observed as the original
cation peaks are still present and no new peaks appear under
any measured conditions, indicating high alkaline stability in up
1
H NMR prior to alkaline chemical stability experiments
(
Figure 2).
1
The degradation of the ionic salts was quantified by H NMR
spectroscopy, where the effects of alkaline concentration and
to at least 50 equiv KOH/D O at 30 °C and 20 equiv KOH/
2
2
reaction temperature on the alkaline chemical stability of the
D O at 80 °C. It is possible to show chemical degradation of
1
cations were investigated. Figure 2 shows H NMR spectra of
ammonium salts via nucleophilic attack at higher temperatures
2
4
each ionic salt after 1 week (168 h) exposure to 2, 20, and 50
or with other solvents (e.g., methanol); however, for
consistency within this study, the ionic salts were degraded
under the same conditions as the PILs, which were degraded in
the conditions most applicable to a fuel cell (e.g., water, 80 °C),
which would also not cause precipitation of the degraded
1
equiv KOH/D O solutions at 30 °C. In Figure 2A, the H
2
+
−
NMR spectra of BMIm Cl shows no degradation upon
exposure to 2 equiv, while exposure to 20 and 50 equiv shows a
new peak at ∼1.8 ppm (labeled 1′), where this peak can be
attributed to a ring-opening reaction as described in the
+
−
+
−
polymer. The phosphonium analogue to TMA Cl , TMP Cl
4
literature. Scheme 5 shows the ring-opening degradation
(Figures 2G and 3G), likewise shows no chemical degradation
after 168 h under similar measured conditions: 2, 20, and 50
equiv KOH/D O at 30 °C; 2 and 20 equiv KOH/D O at 80
Scheme 5. Ring-Opening Reaction of 3-Butyl-1-
methylimidazolium
2
2
°C.
1
Figures 2E and 3E show H NMR spectra for guanidinium
+
−
ionic salt, HMG Cl , after 168 h exposure to KOH/D O at 30
2
and 80 °C, respectively. As observed for the imidazolium salts,
degradation increased with increasing hydroxide concentration
as well as increasing temperature. In Figure 2E, new peaks
observed at ∼1.8 ppm (labeled 1′) were attributed to a
nucleophilic substitution reaction as described in the
28
+
−
literature. Scheme 6 shows the S 2 degradation of HMG Cl
N
in alkaline media, where the new peak was attributed to the
protons at position 1′. The degree of degradation was
1
determined by the relative integrations of H resonances of
the degraded and nondegraded cation (i.e., 2(1′)/(3(1) +
2
(1′)), Scheme 6) and was found to be 0.7, 10.2, and 66.4% in
+
−
mechanism for BMIm Cl in alkaline media, where the new
peak is attributed to the proton at position 1′. The degree of
degradation by this ring-opening reaction was calculated by the
2, 20, and 50 equiv KOH/D O solutions, respectively, at 30 °C
2
for 168 h. Figure 3E shows that as temperature increased to 80
°C, degradation increased to 60.9% and 100% for 2 and 20
equiv, respectively.
1
relative integrations of the indicated H resonances (i.e., 1′/(1
+
1′), Scheme 5) and was calculated to be 20.0% and 88.1% for
Similar to results for ammonium and phosphonium salts, the
+
−
+
−
BMIm Cl in 20 and 50 equiv KOH/D O solutions,
pyrrolidinium salt BMP Cl in Figures 2F and 3F shows no
chemical degradation in alkaline media. This is an interesting
result, as other ring structures (imidazolium salts) show
quantifiable degradation by ring-opening reaction under the
same conditions. One potential explanation for improved
chemical stability is increased resistance to bond breakage due
to lack of polarizability in the resonance-free ring structure of a
pyrrolidinium versus the conjugated bonds of the imidazolium
ring; steric hindrance as a result of the butyl and methyl
substituents in the N1 position may also be providing
protection from hydroxide attack.
2
respectively, at 30 °C for 168 h. Similarly, a ring-opening
reaction was observed in the C2-methyl-substituted imidazo-
lium salt (Figure 2B), BDMIm Cl ; the degree of degradation
+
−
at 30 °C for 168 h was found to be 0.0, 2.0, and 13.8% in 2, 20,
and 50 equiv KOH/D O solutions, respectively, indicating that
2
methyl substitution in the C2 position of imidazolium salts
hinders hydroxide attack compared to the C2-unsubstituted
ionic salt (BMIm Cl ). These results corroborate with recent
findings in the literature. It is also notable that in both salts the
proton peaks associated with the C2, C4, and C5 positions of
+
−
9
imidazole rings disappear in the presence of D O due to the
Overall, high alkaline chemical stability was observed in the
ammonium, phosphonium, and pyrrolidinium ionic salts,
2
hydrogen/deuterium (H/D) exchange reaction.
1
+
−
+
−
+
−
+
−
Figure 3 shows H NMR spectra of each ionic salt in this
TMA Cl , BTMA Cl , TMP Cl , and BMP Cl , as no
quantifiable degradation was evidenced inasmuch as 50 equiv
KOH/D O at 30 °C and 20 equiv KOH/D O at 80 °C. Of the
study after 168 h exposure to 2 and 20 equiv KOH/D O
2
solutions at a higher temperature (80 °C). Degradation is
accelerated by increased temperature, as seen in Figure 3A,B,
where at 80 °C both imidazolium salts show earlier signs of
ring-opening reactions with 0.5% degradation for 2 equiv
compared with 0.0% at 30 °C at this KOH concentration. For
2
2
remaining salts, the C2-methyl substituted imidazolium,
+
−
BDMIm Cl , was observed to be the most chemically stable,
with 7.5% degradation at 20 equiv KOH/D O at 80 °C,
2
+
−
compared with 98.7% and 100% degradation for BMIm Cl
F
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
G
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 6. Nucleophilic Substitution Reaction of
Hexamethylguanidinium
reaction temperature on the alkaline chemical stability of the
covalently tethered cations were investigated.
Figures 4 and 5 show H NMR spectra of each PIL after 168
1
h exposure to 2 and 20 equiv KOH/D O solutions at 30 and 80
2
1
°
C, respectively. In Figure 4A, the H NMR spectra of
poly(MEBIm-Br) show two new peaks at ∼7.5 and ∼8.43 ppm
(
attributed to ring-opening degradation as seen in the analogous
small molecule salt and described in the literature. Likewise, in
labeled f′ and e′, respectively), where these peaks are
4
+
−
and HMG Cl , respectively, under the same conditions. A
summary of the alkaline chemical degradation results for the
small molecule ionic salts is listed in Table 1.
Figure 4B, two new peaks appear at ∼7.48 and ∼8.42 ppm
1
(
labeled h′ and g′, respectively) in the H NMR spectra of
poly(MEBMIm-Br), also indicating a ring-opening mechanism.
Alkaline degradation by this ring-opening reaction was
PIL Alkaline Chemical Stability. We synthesized and
investigated the chemical stability of methacrylate-based PILs
consisting of various covalently attached cations: butylimida-
zolium, butylmethylimidazolium, trimethylammonium,
pentamethylguanidinium, butylpyrrolidinium, and trimethyl-
phosphonium. These cations are analogous to the ionic salts in
this study. The purity, chemical structure, and degree of
1
calculated by the relative integrations of the indicated H
resonances (i.e., f′/(f + f′) for poly(MEBIm-Br) and g′/(g + g′)
for poly(MEBMIm-Br); degradation mechanism shown in
Scheme 7) and was calculated as 12.1 and 22.3% for
poly(MEBIm-Br) and 15.5 and 23.9% for poly(MEBMIm-Br)
at 2 and 20 equiv, respectively, at 30 °C for 168 h. Figure 5A,B
shows the same new peaks arising under high temperature
conditions; the extent of degradation was calculated as 19.4 and
33.5% for poly(MEBIm-Br) and 31.3 and 38.2% for poly-
1
functionalization of these PILs were confirmed by H NMR
prior to alkaline chemical stability experiments (see Figure 1).
1
PIL cation degradation was also quantified by H NMR
spectroscopy, where the effects of alkaline concentration and
Table 1. Alkaline Chemical Degradation Results for Ionic Salts
a
ionic salt
temp (°C)
KOH conc (mol equiv)
KOH conc (M)
NMR spectrum
degradation (%)
degradation pathway
Scheme 5
+
−
BMIm Cl
30
2
20
50
2
0.29
2.86
7.16
0.29
2.86
0.27
2.65
6.63
0.27
2.65
0.46
4.56
11.41
0.46
4.56
0.27
2.69
6.74
0.27
2.69
0.29
2.86
7.16
0.29
2.86
0.28
2.81
7.03
0.28
2.81
0.39
3.89
9.72
0.39
3.89
Figure 2A
Figure 2A
Figure 2A
Figure 3A
Figure 3A
Figure 2B
Figure 2B
Figure 2B
Figure 3B
Figure 3B
Figure 2C
Figure 2C
Figure 2C
Figure 3C
Figure 3C
Figure 2D
Figure 2D
Figure 2D
Figure 3D
Figure 3D
Figure 2E
Figure 2E
Figure 2E
Figure 3E
Figure 3E
Figure 2F
Figure 2F
Figure 2F
Figure 3F
Figure 3F
Figure 2G
Figure 2G
Figure 2G
Figure 3G
Figure 3G
0.0
20.0
88.1
0.4
98.7
0.0
2.0
13.8
0.8
7.5
0
30
30
80
80
20
2
+
−
b
BDMIm Cl
30
Scheme 5
30
30
80
80
20
50
2
20
2
+
−
TMA Cl
30
30
30
80
80
20
50
2
0
0
0
20
2
0
+
−
BTMA Cl
30
0
30
30
80
80
20
50
2
0
0
0
20
2
0
+
−
HMG Cl
30
0.7
10.2
66.4
60.9
100
0
Scheme 6
30
30
80
80
20
50
2
20
2
+
−
BMP Cl
30
30
30
80
80
20
50
2
0
0
0
20
2
0
+
−
TMP Cl
30
0
30
30
80
80
20
50
2
0
0
0
20
0
a
b
All samples were degraded for 168 h. Analogous to degradation pathway of similar ionic salt.
H
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 4. 1H NMR spectra for (A) poly(MEBIm-Br), (B) poly(MEBMIm-Br), (C) poly(METMA-Br), (D) poly(MEBP-Br), and (E)
poly(METMP-Br) in 2 and 20 equiv KOH/D O at 30 °C for 168 h.
2
(
MEBMIm-Br) at 2 and 20 equiv, respectively, at 80 °C for 168
associated with the C2, C4, and C5 positions of imidazole rings
diminished more rapidly than could be attributed solely to ring-
opening degradation due to an accompanying hydrogen/
deuterium (H/D) exchange reaction. It has been shown in
previous studies that it is possible to suppress this H/D
h. In contrast to the ionic salt chemical stability results, the C2-
methyl-substituted cation showed slightly higher degradation at
each condition compared to the unsubstituted cation, indicating
that the resistance to ring-opening mechanism provided by
steric hindrance of additional moieties in the C2 position that
has been seen in the small molecule studies may not translate to
covalently tethered cations. It should be noted that for both
exchange in order to accurately calculate cation degradation by
4
including 10 wt % H O (relative to mass of polymer).
2
Therefore, for the imidazolium PILs only, 80 mg of D O was
2
imidazolium polymers in KOH/D O solution the proton peaks
replaced with 80 mg of H O, to maintain a total of 1 mL water
2
2
I
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
Figure 5. H NMR spectra for (A) poly(MEBIm-Br), (B) poly(MEBMIm-Br), (C) poly(METMA-Br), (D) poly(MEBP-Br) and and (E)
poly(METMP-Br) in 2 and 20 equiv KOH/D O at 80 °C for 168 h.
2
(
deuterated and nondeuterated), while also preventing H/D
include H O in the solution. It was also confirmed (Figure S6)
2
1
exchange. Additional H NMR spectra (Supporting Informa-
that including H O would not change the results for these PILs,
2
tion Figures S2 and S3) confirm that the exchange is fully
and thus the imidazolium results in KOH/D O/H O solutions
2
2
suppressed for the imidazolium PILs when H O is included.
For the remaining three PILs, H NMR spectra (Figures S4 and
S5) confirm that as anticipated, there is no H/D exchange in
are analogous to results for the other three PILs in KOH/D O
2
2
1
solutions. Similar results were obtained for the imidazolium
salts, where no difference in degradation was observed between
D O or H O solutions (Figure S7).
D O at 80 °C for 168 h, such that any reduction in the cation
2
2
2
1
peaks in the presence of KOH can be attributed solely to
Figures 4C and 5C show H NMR spectra for the quaternary
alkaline degradation reaction; it is therefore not necessary to
ammonium-based PIL, poly(METMA-Br), after 168 h exposure
J
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 7. Ring-Opening Reaction of Poly(MEBMIm-Br)
pathway that resulted in a pendant hydroxyl and an amine
byproduct, was the primary degradation mechanism at both low
and high temperatures.
The guanidinium-based PIL, poly(MEPMG-Br), was not
water-soluble and was therefore excluded from the chemical
stability study. For the pyrrolidinium-based PIL, poly(MEBP-
Br), in Figures 4D and 5D, limited degradation was observed in
1
the H NMR spectra after 168 h exposure to 2 or 20 equiv
KOH/D O at 30 and 80 °C, respectively, as a new peak arises
2
at ∼5.60 ppm (labeled j′). Similar degradation has been
described in the literature; the proposed Hofmann elimination
mechanism is shown in Scheme 9. The degree of degradation
to KOH/D O at 30 and 80 °C, respectively. Two major
degradation reactions are generally accepted for pendant
quaternary ammonium cations: (1) nucleophilic substitution
2
30
Scheme 9. Hofmann Elimination Reaction of Poly(MEBP-
Br)
(
SN2 reaction) via hydroxide attack on the α carbons resulting
in amine and alcohol byproducts and (2) Hofmann degradation
E2 elimination) from the removal of the β hydrogen by
hydroxide, resulting in amine and alkene byproducts (Scheme
(
3
4
8
). The S 2 reaction and E2 elimination occur simulta-
N
Scheme 8. Degradation Reactions of Poly(METMA-Br)
of the cation can be calculated by the relative integrated
1
intensities of corresponding H NMR peaks (i.e., j′/(k + j′))
and was 9.3 and 10.3% for 2 and 20 equiv, respectively, at 80
°
C, indicating a high alkaline chemical stability at these
conditions. The Hofmann elimination reaction was further
confirmed by the minimal degradation at 30 °C (1.5 and 3.1%
for 2 and 20 equiv, respectively), as this mechanism is known to
favor higher temperatures. 1
Figures 4E and 5E show H NMR spectra for the quaternary
phosponium-based PIL, poly(METMP-Br), at 30 and 80 °C,
respectively, after 168 h exposure to 2 and 20 equiv KOH/D O
2
solutions. The phosphonium cation is known to degrade by the
Cahours−Hofmann reaction, where hydroxide attacks the
phosphonium tetrahedron, resulting in phosphine oxide and
35,36
an alkyl group as byproducts (Scheme 10).
In both Figures
Scheme 10. Cahours−Hofmann Reaction of Poly(METMP-
Br)
neously, leading to a mixture of byproducts; one mechanism
may be preferential depending upon the chemical structure and
degradation conditions. In Figure 4C, nucleophilic substitution
reactions were observed for the trimethylammonium cation as
new peaks appeared at ∼3.92 ppm (labeled i′) and ∼2.09 ppm
(
labeled i″) in solutions of 2 and 20 equiv KOH/D O at 30 °C;
2
Hofmann elimination reaction was not observed at 30 °C.
Extent of degradation of the cation, defined as the relative
1
integrated intensities of corresponding H NMR peaks (i.e., (9i′
+
3i″)/(2i + 9i′ + 3i″)), was 46.4 and 75.5% for 2 or 20 equiv,
respectively, at 30 °C. At 80 °C, in addition to the degradation
peaks seen in Figure 4C, Figure 5C shows two additional peaks
located at ∼5.45 and ∼5.67 ppm (labeled i‴). These new peaks
indicate degradation by Hofmann elimination occurred at this
higher temperature condition; subsequently, relative integration
ratios include the new peaks (i.e., (9i′ + 3i″ + 9i‴)/(2i + 9i′ +
4E and 5E, the two new peaks appearing at ∼3.84 and ∼3.77
ppm (labeled l′) are associated with the −CH − of the alkyl
2
chain remaining on the polymer after the phosponium group
has been cleaved, while the new peak at ∼1.12 ppm (labeled l″)
is associated with the small molecule byproduct, trimethyl-
phosphonium oxide. Integration of the degradation peaks
relative to the remaining covalently tethered polymer cation
3
9
i″ + 9i‴)). Cation degradation was quantified as 72.1 and
4.8% upon exposure to 2 and 20 equiv KOH/D O solutions,
2
respectively, at 80 °C. While degradation increased with
increasing temperature, the S 2 reaction, specifically the
N
K
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 2. Alkaline Chemical Degradation Results for PILs with Various Cations
a
b
PIL
temp (°C)
KOH conc (mol equiv)
KOH conc (M)
NMR spectrum
cation degradation (%)
degradation pathway
c
poly(MEBIm-Br)
30
2
20
2
0.05
0.50
0.05
0.50
0.05
0.48
0.05
0.48
0.06
0.63
0.06
0.63
0.05
0.50
0.05
0.50
0.06
0.59
0.06
0.59
Figure 4A
Figure 4A
Figure 5A
Figure 5A
Figure 4B
Figure 4B
Figure 5B
Figure 5B
Figure 4C
Figure 4C
Figure 5C
Figure 5C
Figure 4D
Figure 4D
Figure 5D
Figure 5D
Figure 4E
Figure 4E
Figure 5E
Figure 5E
12.1
22.3
19.4
33.5
15.5
23.9
31.3
38.2
46.4
75.5
72.1
94.8
1.5
Scheme 7
30
80
80
20
2
poly(MEBMIm-Br)
poly(METMA-Br)
poly(MEBP-Br)
30
Scheme 7
Scheme 8
Scheme 9
30
80
80
20
2
20
2
30
30
80
80
20
2
20
2
30
30
80
80
20
2
3.1
9.3
20
2
10.3
63.4
93.1
d
poly(METMP-Br)
30
Scheme 10
30
80
80
20
2
20
d
a
b
In samples of poly(MEBIm-Br) and poly(MEBMIm-Br), 80 mg of D O was replaced with 80 mg of H O to suppress H/D exchange. All samples
2
2
were degraded for 168 h; amounts refer to both cleaved and uncleaved cation; see Supporting Information for backbone degradation due to
c
d
hydrolysis. Degradation pathway of poly(MEBIm-Br) is analogous to degradation pathway of poly(MEBMIm-Br). Polymer precipitated under
these conditions.
peak (i.e., 9l′/(l + 9l′)) yields degradation degrees of 63.4 and
3.1% for 2 or 20 equiv, respectively, at 30 °C. At 80 °C,
degradation could not be quantified because the polymer
precipitated from solution and the precipitate was black in
color. A summary of the alkaline chemical degradation results
for the PILs with various cations are listed in Table 2.
cation peaks, as well as the peaks that arise due to the E2
elimination reaction; overall, 65.5% of the backbone undergoes
hydrolysis (see Figures S9 and S10, Scheme S3). This could
provide a potential explanation for the extraordinarily high
chemical stability of poly(MEBP-Br), as the small molecule
byproduct of hydrolysis may have higher resistance to
hydroxide attack, keeping the pyrrolidinium ring intact.
However, of the cations that remained tethered to the polymer
9
Note that in addition to the chemical stability of cations,
another potential concern is hydrolysis degradation of the
carboxylate ester linkage of the methacrylate-based PIL
backbone under alkaline conditions. Although the hydrolysis
of a polymeric ester has been reported in the literature, it
typically occurs at a much slower rate compared to small
backbone in 20 equiv KOH/D O at 80 °C for 168 h, only
2
2
4.1% underwent E2 elimination; butylpyrrolidinium remains
the most chemically stable polymeric cation of those
considered. Ester hydrolysis was not identified as a primary
mechanism of degradation in the ammonium-based PIL (see
Figure S11).
3
7
molecule esters; degradation rate is significantly dependent
on the structure of the polymer backbone, with the hydrolysis
of a methacrylate-based polymer observed to occur at a
38
significantly slower rate compared to the acrylate analogue.
CONCLUSIONS
■
For poly(methyl methacrylate) (PMMA), studies have
The alkaline chemical stability of PILs with various cations was
confirmed that ∼9% of the monomer units are susceptible to
1
examined with H NMR spectroscopy in comparison with their
3
8
hydrolysis. Previous research has suggested ester hydrolysis of
the hydroxide-exchanged methacrylate-based PIL, poly-
analogous ionic salts. Results show enhanced chemical stability
of imidazolium- and pyrrolidinium-based PILs relative to
quaternary ammonium- and phosphonium-based PILs. Results
for the ionic salts are in stark contrast to the PILs, as the
imidazolium salts showed significantly more degradation than
the quaternary ammonium and phosphonium salts at all
conditions examined, while the pyrrolidinium-based salt
showed no degradation equivalent to that of the ammonium-
and phosphonium-based salts. Another inconsistency between
salt and polymer was observed when comparing the two
imidazolium cations; ionic salt results showed that methyl
substitution in the C2 position significantly lessened degrada-
tion, while the PIL with unsubstituted imidazolium actually
showed higher chemical stability under all conditions measured
compared to its substituted PIL counterpart. Overall, the
alkaline chemical stability of the free cations in solution
investigated in this study showed no correlation to that of the
(
MEBIm-OH), occurs after the ring-opening mechanism of
the butylimidazolium cation, as ring-opened imidazolium
byproducts were present in both covalently attached (non-
hydrolyzed) and cleaved (hydrolyzed) forms; ∼30% of the ring-
opened byproducts (∼11% of the total monomer units) were
cleaved from the polymer via ester hydrolysis after 168 h at 80
4
°
C and 1 M KOH. In this study, for the imidazolium-based
PILs, poly(MEBIm-Br) and poly(MEBMIm-Br), ∼18% and
∼15% of the total monomer units (∼55% and ∼40% of the
ring-opened byproducts), respectively, had cleaved from the
polymer after 168 h at 80 °C and 20 equiv KOH (see Figure
S8, Schemes S1 and S2); these results are in relatively good
agreement with results for PMMA and poly(MEBIm-OH). For
poly(MEBP-Br), after 168 h in 20 equiv KOH/D O at 80 °C,
2
evidence of significant hydrolysis is noted in the undegraded
L
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
tethered cations on the PILs, suggesting that small molecule
studies may not provide a solid basis for comparison of relative
cation stability as a preliminary screening tool for determining
the most promising cations.
(14) Vandiver, M. A.; Caire, B. R.; Poskin, Z.; Li, Y.; Seifert, S.;
Knauss, D. M.; Herring, A. M.; Liberatore, M. W. Durability and
Performance of Polystyrene-b-poly(vinylbenzyl trimethylammonium)
Diblock Copolymer and Equivalent Blend Anion Exchange Mem-
branes. J. Appl. Polym. Sci. 2015, 132, 41596.
(
15) Yan, J.; Hickner, M. A. Anion Exchange Membranes by
ASSOCIATED CONTENT
Supporting Information
■
Bromination of Benzylmethyl-Containing Poly(sulfone)s. Macro-
molecules 2010, 43, 2349−2356.
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.macro-
mol.5b01223.
(16) Tsai, T.-H.; Maes, A. M.; Vandiver, M. A.; Versek, C.; Seifert, S.;
Tuominen, M.; Liberatore, M. W.; Herring, A. M.; Coughlin, E. B.
Synthesis and Structure−conductivity Relationship of Polystyrene-
block-poly(vinyl benzyl trimethylammonium) for Alkaline Anion
Exchange Membrane Fuel Cells. J. Polym. Sci., Part B: Polym. Phys.
1
Additional H NMR data (PDF)
2
013, 51, 1751−1760.
AUTHOR INFORMATION
Corresponding Author
E-mail: elabd@tamu.edu (Y.A.E.).
Notes
■
(17) Sturgeon, M. R.; Macomber, C. S.; Engtrakul, C.; Long, H.;
Pivovar, B. S. Hydroxide based Benzyltrimethylammonium Degrada-
tion: Quantification of Rates and Degradation Technique Develop-
ment. J. Electrochem. Soc. 2015, 162, F366−F372.
*
(
18) Choe, Y.-K.; Fujimoto, C.; Lee, K.-S.; Dalton, L. T.; Ayers, K.;
The authors declare no competing financial interest.
Henson, N. J.; Kim, Y. S. Alkaline Stability of Benzyl Trimethyl
Ammonium Functionalized Polyaromatics: A Computational and
Experimental Study. Chem. Mater. 2014, 26, 5675−5682.
ACKNOWLEDGMENTS
This work is supported in part by the U.S. Army Research
Office under Grant W911NF-14-0310.
■
(
19) Chen, D.; Hickner, M. A. Ion Clustering in Quaternary
Ammonium Functionalized Benzylmethyl Containing Poly(arylene
ether ketone)s. Macromolecules 2013, 46, 9270−9278.
(20) Hibbs, M. R. Alkaline Stability of Poly(phenylene)-based Anion
Exchange Membranes with Various Cations. J. Polym. Sci., Part B:
Polym. Phys. 2013, 51, 1736−1742.
(21) Arges, C. G.; Ramani, V. Investigation of Cation Degradation in
Anion Exchange Membranes Using Multi-Dimensional NMR Spec-
troscopy. J. Electrochem. Soc. 2013, 160, F1006−F1021.
(22) Arges, C. G.; Ramani, V. Two-dimensional NMR Spectroscopy
Reveals Cation-triggered Backbone Degradation in Polysulfone-based
Anion Exchange Membranes. Proc. Natl. Acad. Sci. U. S. A. 2013, 110,
REFERENCES
■
(
1) Varcoe, J. R.; Slade, R. C. T. Prospects for Alkaline Anion-
Exchange Membranes in Low Temperature Fuel Cells. Fuel Cells 2005,
5
(
, 187−200.
2) Ye, Y.; Sharick, S.; Davis, E. M.; Winey, K. I.; Elabd, Y. A. High
Hydroxide Conductivity in Polymerized Ionic Liquid Block Copoly-
mers. ACS Macro Lett. 2013, 2, 575−580.
3) Nykaza, J. R.; Ye, Y.; Elabd, Y. A. Polymerized Ionic Liquid
Diblock Copolymers with Long Alkyl Side-chain Length. Polymer
014, 55, 3360−3369.
4) Ye, Y.; Elabd, Y. A. Relative Chemical Stability of Imidazolium-
Based Alkaline Anion Exchange Polymerized Ionic Liquids. Macro-
molecules 2011, 44, 8494−8503.
5) Chen, D.; Hickner, M. A. Degradation of Imidazolium- and
Quaternary Ammonium-Functionalized Poly(fluorenyl ether ketone
sulfone) Anion Exchange Membranes. ACS Appl. Mater. Interfaces
012, 4, 5775−5781.
6) Qiu, B.; Lin, B.; Si, Z.; Qiu, L.; Chu, F.; Zhao, J.; Yan, F. Bis-
imidazolium-based Anion-Exchange Membranes for Alkaline Fuel
Cells. J. Power Sources 2012, 217, 329−335.
7) Ran, J.; Wu, L.; Varcoe, J. R.; Ong, A. L.; Poynton, S. D.; Xu, T.
Development of Imidazolium-type Alkaline Anion Exchange Mem-
branes for Fuel Cell Application. J. Membr. Sci. 2012, 415, 242−249.
8) Li, W.; Fang, J.; Lv, M.; Chen, C.; Chi, X.; Yang, Y.; Zhang, Y.
Novel Anion Exchange Membranes Based on Polymerizable
Imidazolium Salt for Alkaline Fuel Cell Applications. J. Mater. Chem.
011, 21, 11340−11346.
9) Lin, B.; Dong, H.; Li, Y.; Si, Z.; Gu, F.; Yan, F. Alkaline Stable
C2-Substituted Imidazolium-Based Anion-Exchange Membranes.
Chem. Mater. 2013, 25, 1858−1867.
10) Gu, F.; Dong, H.; Li, Y.; Si, Z.; Yan, F. Highly Stable N3-
(
2
(
2
(
490−2495.
23) Arges, C. G.; Kulkarni, S.; Baranek, A.; Pan, K.-J.; Jung, M.-S.;
Patton, D.; Mauritz, K. A.; Ramani, V. Quarternary Ammonium and
Phosphonium Based Anion Exchange Membranes for Alkaline Fuel
Cells. ECS Trans. 2010, 33, 1903−1913.
(
(
24) Noonan, K. J. T.; Hugar, K. M.; Kostalik, H. A.; Lobkovsky, E.
B.; Abrun
̃
a, H. D.; Coates, G. W. Phosphonium-Functionalized
2
(
Polyethylene: A New Class of Base-Stable Alkaline Anion Exchange
Membranes. J. Am. Chem. Soc. 2012, 134, 18161−18164.
(
Y. A Soluble and Highly Conductive Ionomer for High-Performance
Hydroxide Exchange Membrane Fuel Cells. Angew. Chem., Int. Ed.
2
(
25) Gu, S.; Cai, R.; Luo, T.; Chen, Z.; Sun, M.; Liu, Y.; He, G.; Yan,
(
009, 48, 6499−6502.
26) Wang, J.; Li, S.; Zhang, S. Novel Hydroxide-Conducting
(
Polyelectrolyte Composed of an Poly(arylene ether sulfone)
Containing Pendant Quaternary Guanidinium Groups for Alkaline
Fuel Cell Applications. Macromolecules 2010, 43, 3890−3896.
2
(
(
27) Liu, L.; Li, Q.; Dai, J.; Wang, H.; Jin, B.; Bai, R. A Facile Strategy
for the Synthesis of Guanidinium-functionalized Polymer as Alkaline
Anion Exchange Membrane with Improved Alkaline Stability. J.
Membr. Sci. 2014, 453, 52−60.
(
(
28) Kim, D. S.; Labouriau, A.; Guiver, M. D.; Kim, Y. S.
Substituted Imidazolium-Based Alkaline Anion Exchange Membranes:
Experimental Studies and Theoretical Calculations. Macromolecules
014, 47, 208−216.
11) Qiu, B.; Lin, B.; Qiu, L.; Yan, F. Alkaline Imidazolium- and
Quaternary Ammonium-Functionalized Anion Exchange Membranes
for Alkaline Fuel Cell Applications. J. Mater. Chem. 2012, 22, 1040−
045.
Guanidinium-Functionalized Anion Exchange Polymer Electrolytes
via Activated Fluorophenyl-Amine Reaction. Chem. Mater. 2011, 23,
2
(
3
(
795−3797.
29) Sajjad, S. D.; Hong, Y.; Liu, F. Synthesis of Guanidinium-based
Anion Exchange Membranes and their Stability Assessment. Polym.
Adv. Technol. 2014, 25, 108−116.
1
(
12) Zhang, F.; Zhang, H.; Qu, C. Imidazolium Functionalized
Polysulfone Anion Exchange Membrane for Fuel Cell Application. J.
(30) Gu, F.; Dong, H.; Li, Y.; Sun, Z.; Yan, F. Base Stable
Pyrrolidinium Cations for Alkaline Anion Exchange Membrane
Applications. Macromolecules 2014, 47, 6740−6747.
Mater. Chem. 2011, 21, 12744−12752.
1
(
13) Nun
̃
ez, S. A.; Hickner, M. A. Quantitative H NMR Analysis of
(31) Price, S. C.; Williams, K. S.; Beyer, F. L. Relationships between
Structure and Alkaline Stability of Imidazolium Cations for Fuel Cell
Membrane Applications. ACS Macro Lett. 2014, 3, 160−165.
Chemical Stabilities in Anion-Exchange Membranes. ACS Macro Lett.
2
013, 2, 49−52.
M
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(
32) Thomas, O. D.; Soo, K. J. W. Y.; Peckham, T. J.; Kulkarni, M. P.;
Holdcroft, S. A Stable Hydroxide-Conducting Polymer. J. Am. Chem.
Soc. 2012, 134, 10753−10756.
(
33) Ye, Y.; Elabd, Y. A. Anion Exchanged Polymerized Ionic
Liquids: High Free Volume Single Ion Conductors. Polymer 2011, 52,
309−1317.
34) Yuesheng, Y.; Elabd, Y. A. Chemical Stability of Anion Exchange
1
(
Membranes for Alkaline Fuel Cells. In Polymers for Energy Storage and
Delivery: Polyelectrolytes for Batteries and Fuel Cells; ACS Symposium
Series 1096; American Chemical Society: Washington, DC, 2012; Vol.
1
(
1
(
096, pp 233−251.
35) Hofmann, A. W.; Cahours, A. Philos. Trans. R. Soc. London 1857,
47, 575.
36) Zanger, M.; Vander Werf, C. A.; McEwen, W. E. Kinetic Study
of the Decomposition of Quaternary Phosphonium Hydroxides. J. Am.
Chem. Soc. 1959, 81, 3806−3807.
(
37) Gaetjens, E.; Morawetz, H. Intramolecular Carboxylate Attack
on Ester Groups. II. The Effect of Diastereoisomerism in Polymers
and their Low Molecular Weight Models. J. Am. Chem. Soc. 1961, 83,
1
(
738−1742.
38) Baines, F. C.; Bevington, J. C. A Tracer Study of the Hydrolysis
of Methyl Methacrylate and Methyl Acrylate Units in Homopolymers
and Copolymers. J. Polym. Sci., Part A-1: Polym. Chem. 1968, 6, 2433−
2440.
N
DOI: 10.1021/acs.macromol.5b01223
Macromolecules XXXX, XXX, XXX−XXX