Ion Conducting ROMP Monomers Based on (Oxa)norbornenes with Pendant Imidazolium Salts Connected via Oligo(oxyethylene) Units and with Oligo(ethyleneoxy) Terminal Moieties

Article abstract of DOI:10.1021/acs.macromol.8b02295

A matrix of 22 two-armed norbornene-based imidazolium TFSI monomers (8) was synthesized to determine the optimal structure in terms of single ion conductivity. For the chain tethering the imidazolium ring to the norbornene ring three or four oxyethylene units are optimal. A terminal group of two ethyleneoxy units was optimal. NMR studies indicated that both the tether oxyethylene units and the terminal ethyleneoxy units interact with the imidazolium cation via hydrogen bonding. 8r (X = 4, Y = 2) exhibited a conductivity of 9.57 × 10^-5 S/cm at 25 °C and a T_g of -46 °C. Low T_g values do not correlate with higher conductivity as a result of the H-bonding interactions. Stability toward autopolymerization and reasonable conductivities provide an acceptable platform for ion conducting ROMP polymers. Four one-armed norbornene-based imidazolium TFSI monomers (15) were prepared with tetra(ethyleneoxy) linkers/spacers and variable terminal groups. All of these exhibited low T_gs (<-55 °C) and room temperature conductivities >10^-4 S/cm, the highest being 4.39 × 10^-4 S/cm for 15c (T_g = -69 °C), the analogue of 8r, providing hope for outstanding polymers. Three oxanorbornene-based two-armed imidazolium TFSI monomers (18) were prepared with varied linkers and terminal groups. 18b possesses a room temperature conductivity of 1.2 × 10^-4 S/cm, again augering well for polymers derived therefrom by ROMP.

Full text of DOI:10.1021/acs.macromol.8b02295

Article
pubs.acs.org/Macromolecules
Cite This: Macromolecules XXXX, XXX, XXX−XXX
Ion Conducting ROMP Monomers Based on (Oxa)norbornenes with
Pendant Imidazolium Salts Connected via Oligo(oxyethylene) Units
and with Oligo(ethyleneoxy) Terminal Moieties
†
‡,§
†
†
§
Terry L. Price, Jr., U Hyeok Choi,
Daniel V. Schoonover, Murugan Arunachalam, Renxuan Xie,
†
§
,†
Steven Lyle, Ralph H. Colby, and Harry W. Gibson*
†
Department of Chemistry and Macromolecules Innovations Institute, Virginia Tech, Blacksburg, Virginia 24061, United States
‡
Department of Polymer Engineering, Pukyong National University, Busan 48513, Korea
§
Department of Materials Science and Engineering, Penn State University, University Park, Pennsylvania 16802, United States
*
S Supporting Information
ABSTRACT: A matrix of 22 two-armed norbornene-based imidazolium TFSI monomers (8) was synthesized to determine the
optimal structure in terms of single ion conductivity. For the chain tethering the imidazolium ring to the norbornene ring three
or four oxyethylene units are optimal. A terminal group of two ethyleneoxy units was optimal. NMR studies indicated that both
the tether oxyethylene units and the terminal ethyleneoxy units interact with the imidazolium cation via hydrogen bonding. 8r
−
5
(
X = 4, Y = 2) exhibited a conductivity of 9.57 × 10 S/cm at 25 °C and a T of −46 °C. Low T values do not correlate with
g
g
higher conductivity as a result of the H-bonding interactions. Stability toward autopolymerization and reasonable conductivities
provide an acceptable platform for ion conducting ROMP polymers. Four one-armed norbornene-based imidazolium TFSI
monomers (15) were prepared with tetra(ethyleneoxy) linkers/spacers and variable terminal groups. All of these exhibited low
−
4
−4
T s (<−55 °C) and room temperature conductivities >10 S/cm, the highest being 4.39 × 10 S/cm for 15c (T = −69 °C),
g
g
the analogue of 8r, providing hope for outstanding polymers. Three oxanorbornene-based two-armed imidazolium TFSI
monomers (18) were prepared with varied linkers and terminal groups. 18b possesses a room temperature conductivity of 1.2 ×
−
4
1
0
S/cm, again augering well for polymers derived therefrom by ROMP.
INTRODUCTION
electrodes placed on both sides of the film. Under a direct
current (dc) potential the “free” ions move toward the
respective counter electrode. This electrical transport of mass
leads to strain and thus deforms the polymer in the direction
away from the ion movement. The actuator application thus
depends on the induced strain, and as a result large, bulky
mobile ions are more effective than smaller ones, even though
the mobility or conductivity of the latter may be higher. For
controllable movement in one direction, only one of the ions
should be mobile, i.e., a single ion conductor. If both cations
and anions are mobile, motion occurs in both directions
sequentially.
■
Recently there has been considerable interest in ionically
1
−4
conductive polymers.
In contrast to electronically con-
ductive materials, ion conducting materials move charge by the
transport of ions instead of electrons. These systems may be
thought of as analogous to solutions of electrolytes. The focus
5
of the present work was the synthesis of monomers and
polymers that conduct by the movement of anionic counter-
ions of polymer-bound imidazolium cations, as other workers
6
−12
have reported.
Imidazolium cationic species were selected
for this work due to their outstanding physical and ion
transport properties. Many imidazolium salts fall into the
category known as ionic liquids. Ionic liquids generally have
melting temperatures below 100 °C; a subset, room temper-
High ionic conductivity is most easily achieved in a low-
4
viscosity liquid phase that allows free or minimally hindered
ature ionic liquids, have melting points below 25 °C. They
17
movement of ions in an electric field. Consequently, using a
liquid salt that would show high ionic mobility in its liquid
phase was attractive. In addition to being liquids at low
possess large, delocalized ions that reduce Coulombic
attractive forces, leading to lower melting points and less ion
pairing than normal salts such as NaCl.
The ultimate purpose of this work was to generate materials
1
3−16
that can be used as electromechanical actuators.
The ideal
Received: October 26, 2018
Revised: January 14, 2019
material is a thin film of an ionic polymer whose mobile ions
are made to move using an electrical potential applied through
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
temperatures, imidazolium salts are less toxic than some other
organic cations such as paraquat.
these systems is that the monomers very easily autopolymerize.
18,19
Ideally for actuators the low-T , ion conducting component
g
In previous work, we found that there is a general
correspondence of the glass transition temperature of a
monomer or polymer with its ionic conductivity;
samples that have lower glass transition temperatures (Tg)
tend to have higher room temperature ionic conductivities
would be part of a block copolymer whose other component
would provide the mechanical properties necessary to do real
work; the modulus of the material is important. In this case,
the ideal morphology would be a bicontinuous microphase-
separated block copolymerone phase for mechanical
integrity and the other for ionic conduction. The autopoly-
merization issue basically prevents the preparation of such
block copolymers from the (meth)acrylates. As an alternative
ring-opening metathesis polymerization (ROMP) appeared to
be a viable alternative for the production of well-defined ion-
3
,20−30
(
σ25°C). Ethyleneoxy polymers often have very low glass
transition temperatures and thus are liquids at room temper-
ature.
2
0,26,30−32
Ethyleneoxy moieties are believed to stabilize
the imidazolium cation and encourage the separation and
transport of the counterion away from it. Indeed, there are
many examples of complexes of imidazolium cations with
polyethers, both cyclic (crown ethers) and linear, e.g.,
8
,11,42−46
containing polymers,
and sequential addition of
33−41
monomers can lead to the synthesis of the requisite block
poly(ethylene glycol)s.
The ethyleneoxy moieties provide
46−48
an electron-rich environment via the multiple lone pairs
associated with the oxygen atoms, which stabilizes the cationic
species and allows for ion pair dissociation, yielding more
copolymers.
The earlier work employing the imidazolium cation with a
−
PF6 anion gave acceptable conductivity values, but switching
−
“
free” or less tightly bound anions necessary for charge
to the bis(trifluoromethylsulfonyl)imide anion {(CF SO ) N
= TFSI } improved conductivities dramatically;
3
2 2
2
9,30
23,25
−
transport.
the
20,25,26,29
Some of our previous work
focused on the synthesis
thermal/chemical stability was also improved, and the larger,
asymmetric TFSI anion is expected to induce more strain than
of (meth)acrylate esters that carried an imidazolium salt with
differing tether chain lengths (Figure 1). A critical issue with
the smaller, more symmetric PF anion. In addition, this
6
original work provided a rubric for monomer design.
Oxyethylene (OE) spacer units (X and Y, Figure 1) were
employed to lower the T of the monomer and thus increase
g
2
0
the conductivity. And Tg values were decreased and
conductivities were increased by placing ethyleneoxy (EO)
20,26
units on the terminus of the pendant moiety (Z, Figure 1).
Our strategy was to use the well-developed chemistry of
norbornene and oxanorbornene. By use of these basic
monomer units, there are two different tactics: “one-armed”
and “two-armed” derivatives. Our primary effort was devoted
to “two armed” derivatives to maximize the ionic content, with
the idea that this might afford the most conductive systems,
but we also examined “one-armed” and oxanorbornene
analogues in a selective manner. Here we report the results
of these monomer studies, beginning with “two-armed”
monomers. Our figure of merit here is the room temperature
(
RT) conductivity, σ25°C; elsewhere, we will discuss detailed
Figure 1. General structure of prior (meth)acrylate imidazolium TFSI
polymers.
analyses of the dielectric relaxation spectra to elucidate more
basic parameters such as free carrier densities, mobilities,
dielectric constants, and so forth.
Scheme 1. Synthesis of Norbornene-Based Two-Armed Derivatives as Monomer Precursors
B
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
RESULTS AND DISCUSSION
Scheme 3. Norbornene Two-Armed Imidazolium TFSI
a
■
Monomer Syntheses
Synthesis of Two-Armed Norbornene-Based Imida-
zolium TFSI Monomers. Taking into consideration the
previous observations, synthetic schemes were developed to
include oligo(ethylene glycol) units in the norbornene
monomers (Schemes 1, 2, and 3). A uniform synthetic path
Scheme 2. Synthesis of Substituted Imidazole Precursors
to ditosylates 3 was not chosen in Scheme 1 due to the
differing availabilities and costs associated with oligo(ethylene
glycol) precursors. Scheme 2 shows the syntheses of imidazole
precursors. The exo-norbornene isomer was specifically
selected over the endo isomer for its ability to more readily
4
7−51
undergo ROMP;
stereo purity of anhydride 1 was
achieved through repeated recrystallizations and maintained
throughout the synthetic pathways. Although previous results
26
indicated longer ethyleneoxy chains would be advantageous,
with the addition of every ethyleneoxy unit, the overall charge
density of the molecule is diluted, so obviously there is a limit
and the optimal length was not easily predicted.
In an attempt to find the optimal lengths of the linkers and
terminal groups a matrix of monomers 8 was synthesized
(Scheme 3). X was varied from one to six oxyethylene (OE)
units, while Y of the terminal group was varied from one to
four ethyleneoxy (EO) units, with a butyl chain for
comparison. Three routes were established for the syntheses
of these norbornene monomers, and each has its own unique
advantage for a specific subset of monomers.
The first synthetic method (used for monomers 8a−8e)
involved reaction of the norbornene precursor 1 with 2-
bromoethanol in an esterification reaction to provide
dibromide 4 (Scheme 1). The imidazoles 6a−6e (Scheme 2)
were then quaternized with 4 to provide monomers 7a−7e
−
−
(
A = Br , Scheme 3).
The second synthetic method (for monomers 8f−8t) used
the oligo(ethylene glycol) of the desired length X in an
esterification reaction with anhydride 1 to produce the diols
a
Abbreviations: NB = norbornene, OE = oxyethylene = OCH CH ,
2
2
2
a−2c (Scheme 1). In this synthetic step the glycol was used
Im = imidazolium, Bu = n-butyl, EO = ethyleneoxy = CH CH O, Me
=
2
2
both as reagent and solvent for the reaction; product isolation
was performed via aqueous washings. These compounds were
then tosylated to give compounds 3a−3c.
CH , TFSI = N(SO CF ) .
3 2 3 2
−
−
Monomers 7f−7t (A = Tos ) were produced by
quaternization of the imidazoles 6a−6e with 3a−3c (Scheme
products, so that the ditosylate could be reconverted to the
glycol and the reaction repeated. The monotosylates were then
reacted with anhydride 1 via EDCI coupling to give the
ditosylates 3d and 3e (purification by column chromatography,
Scheme 1). The imidazole 6c was then quaternized with 3d
and 3e to give monomers 7u and 7v (A = Tos ), respectively
(Scheme 3).
Quaternizations to produce monomers 7a−7v were
achieved by solventless reactions of either dibromide 4 or
ditosylates 3a−3e with the desired imidazole 6 at <70 °C for 3
3
). This method is straightforward and used purification steps
that minimize the use of column chromatography; this method
was advantageous for ethylene glycols that were inexpensive.
The third synthetic method for monomers 8u and 8v was
developed due to the costs associated with penta(ethylene
glycol) and hexa(ethylene glycol). Their monotosylates were
prepared (Scheme 1) by reaction with tosyl chloride at a
stoichiometry of 1:1. Chromatography was used to isolate both
−
−
C
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
+
−
Figure 2. (A) Observed endo vs exo NOESY correlations for norbornenes. (B) Partial NOESY spectrum of NB[C(O)OE Im EO Me] (TFSI )
4
3
2
2
(
8s) taken in CDCl at room temperature.
3
a
Table 1. Glass Transition Temperatures (°C) of Two-Armed Norbornene TFSI Monomers 8
X
Y
1
2
3
4
5
6
butyl
(8a) −44
(8b) −39
(8c) −45
(8d) −50
(8e) −42
(8f) −47
(8g) −39
(8h) −49
(8i) −54
(8j) −57
(8k) −47
(8l) −43
(8m) −47
(8n) −46
(8o) −45
(8p) −45
(8q) −43
(8r) −46
(8s) −45
(8t) −47
1
2
3
4
(8u) −46
(8v) −54
a
Determined by DSC at 10 °C/min during the second heating scan. The midpoint of the change in heat capacity was defined as the glass transition
temperature.
days (Scheme 3). The use of solvent and/or higher
temperatures resulted in the formation of undesirable side
products by both retro-Diels−Alder and exo to endo inversion
processes.
Isolations of products 8a−8v were performed using liquid−
liquid extraction; the bulk material (bromide or tosylate salt 7)
was taken up in water and extracted with DCM for 2 days
before counterion exchange to produce the TFSI salt 8.
Because high purity was a concern for these monomers, the
isolation method focused on the purity of the resulting product
and not on maximizing the yield. Chromatography for these
products was avoided.
Syntheses of these monomers began with the stereo pure exo
compound 1. Care was taken to avoid inversions of the exo
1
isomer to the endo isomer by monitoring via H NMR. For
endo products, a NOESY correlation is observable between H1
and H and H couples with H , but these are not present in
2
2
3
exo products (Figure 2A). Figure 2B shows a partial NOESY
spectrum of 8s; peak assignments were carefully made using
COSY (not shown). No correlation exists between between H1
D
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 2. Room Temperature Conductivities (S/cm) of Two-Armed Norbornene TFSI Monomers 8
X
Y
1
2
3
4
5
6
5
(8f) 7.75 × 10−5
−5
−5
(8p) 6.82 × 10−5
(8q) 6.51 × 10
butyl
(8a) 3.07 × 10−
(8k) 7.66 × 10
(8l) 6.54 × 10
−
5
−5
−5
1
2
3
4
(8b) 2.42 × 10
(8g) 4.80 × 10
−
5
−5
−5
−5
(8u) 7.22 × 10−5
(8v) 9.27 × 10−5
(8c) 2.95 × 10
(8h) 7.72 × 10
(8m) 8.95 × 10
(8n) 8.93 × 10
(8o) 7.94 × 10
(8r) 9.57 × 10
−
5
5
−5
−5
−5
(8d) 7.10 × 10
(8e) 3.88 × 10
(8i) 5.39 × 10
(8s) 7.90 × 10
−
−5
−5
−5
(8j) 6.26 × 10
(8t) 6.99 × 10
and H and H and H are not coupled in 8s, thus confirming
the stereo purity of the exo product.
Characterization of Two-Armed Norbornene-Based
Imidazolium TFSI Monomers. Monomers 8a−8v comprise
a systematic matrix in which the numbers of oxyethylene units
T (8j) has a conductivity much (∼35%) lower than the
2
2
3
g
monomer with the highest conductivity (8r). This result
indicates that conductivity is influenced by parameters other
52
than T , as others have demonstrated. Figure 3 clearly shows
g
the lack of correlation of σ25°C with T across the whole series.
g
(
OE) in the spacer (X) and ethyleneoxy units (EO) in the
terminal groups (Y) were varied in an attempt to search out
the optimal monomer. Additionally, butyl groups were
positioned on the terminus for comparison. In addition to
the numbering system within this text, the naming system
+
−
NB[C(O)OE Im EO Me] (TFSI ) has been used for mono-
x
y
2
2
mers 8 (Scheme 3). Considering the linkage between the
norbornene and imidazolium (X ≤ 6 possible units) and on
the terminus of the imidazolium group (Y ≤ 6 possible units),
the potential existed for a matrix of 36 monomers besides the
butyl-terminated systems, for a total of 42. In Scheme 3 some
monomers were omitted; the reason for this will be discussed
shortly. The syntheses of 8a−8v offer a large array of
monomers such that the “best” monomer in terms of RT
conductivity could be found.
Glass transition temperatures (T ) (Table 1) were obtained
g
by DSC. Ionic conductivities (Table 2) of the monomers were
measured over a range of temperatures and frequencies by
dielectric relaxation spectroscopy (DRS). Here only the room
temperature conductivities are reported; a complete analysis of
the DRS results will be reported elsewhere.
Figure 3. Room temperature conductivity vs glass transition
temperature for monomers 8a through 8v. Error bars in T ± 1 °C
g
and in σ25°C ± 5% relative.
T values range from a high of −39 °C for compounds 8b
g
These observations likely arise from the fact that the
oxyethylene linkers and ethyleneoxy tails are both able to wrap
around the cation and free the anion as we establish below;
that is, the linker and the tail compete for interactions with the
imidazolium moieties. This indicates that for increased
conductivities the imidazolium unit should be well away
from the norbornene group, i.e., X ≥ 2, and contain a tail of
ethyleneoxy units. We propose that ethyleneoxy tail units that
strongly interact with the imidazolium cation not only increase
conductivity but also become somewhat ordered and thus do
and 8g to a low of −57 °C for 8j. As more oxyethylene and
ethyleneoxy units were added it was expected that more PEG-
like behavior would ensue. This, however, was not the case.
RT conductivity values in Table 2 range from a high of 9.57
−
5
−5
×
10 S/cm for 8r down to a low of 2.42 × 10 S/cm for 8b,
a relatively small 4-fold change. The conductivity is influenced
by several parameters such as T , free (mobile) ion content,
g
25,29
and dielectric constant.
the spacer between the norbornene and imidazolium group
X), the better the conductivity, up to a point. In addition to
The data suggest that the longer
(
not reduce T as much as might be expected. Detailed analyses
g
this, the data also show that Y (tail length) plays a large role.
For the X = 1 column, a tail of three ethyleneoxy units (Y = 3)
proves to be optimal; however, for X = 2−4, generally a tail of
two ethyleneoxy units (Y = 2) was found to be superior. This is
why only Y = 2 monomers were synthesized for X = 5 and 6
of the relationship between RT conductivity and T for series
with constant X or Y values (Supporting Information, Figures
S43 and S44) are consistent with this picture.
Intramolecular Interactions in Two-Armed Norbor-
nene-Based Imidazolium TFSI Monomers. If increasing
the number of ethyleneoxy tail units would cause the resulting
g
(8u and 8v, respectively). And monomer 8u continues the
trend of decreasing from the maximum value observed for 8r in
the series with Y = 2. Again, the “extra” oxyethylene unit in 8u
is viewed as diluting the ionic content and hence reducing the
conductivity relative to 8r. Comparison of ethyleneoxy tail
units with the butyl tail shows an RT conductivity trend
monomer to be more “PEG-like”, it should drive T downward.
Considering that the expected trend is not observed, the
g
unexpected T trends must be attributed to the ethyleneoxy tail
g
units acting in a different manner, e.g., interacting with the
1
imidazolium cation. As support for this concept, H NMR
slightly different from that of T . The butyl tail for X = 1−4 is
studies were performed in DMSO-d , a dipolar solvent that
g
6
better than the Y = 1 tail.
greatly inhibits any other hydrogen bonding, and compared to
results in other solvents that allow such interactions. Figure 4
What is unusual, however, is that RT conductivity values do
not parallel T values; that is, the lowest T s do not correlate
1
shows partial H NMR spectra of 8c taken in six different
g
g
with the highest conductivities. The monomer with the lowest
solvents.
E
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
+
−
Figure 4. Partial H NMR (500 MHz) spectra of NB[C(O)OEIm EO Me] (TFSI ) (8c) at 8.8 mM in (a) CDCl , (b) CD Cl , (c) acetone-d ,
2
2
2
3
2
2
6
(
d) acetonitrile-d , (e) methanol-d , and (f) DMSO-d .
3 4 6
8
c was specifically selected due to its linkages: one
oxyethylene unit between the norbornene and imidazolium
moieties and two ethyleneoxy tail units (X = 1, Y = 2). This
combination provides a situation in which only the tail moiety
has the possibility of interacting with the imidazolium group.
In Figure 4 it can be seen that the protons of the imidazolium
group in 8c are clearly in different chemical environments in
different solvents. As the solvent becomes more polar, the two
signals observed for both H and H come closer together, and
1
2
the integration difference between the two peaks shrinks. Two
peaks are observed for H in all solvents except DMSO-d . It is
1
6
Figure 5. 1H NMR peak ratio of
H
of NB[C(O)-
reasoned that the acidic hydrogens of the imidazolium units
interact with the lone pairs of electrons of the ethyleneoxy
chain, yielding the major set of signals, and also with the
solvents via hydrogen bonding and other intermolecular
solvation forces, leading to a separate minor set of signals.
Interestingly, plotting the peak ratios (equilibrium constants)
1
+
−
OEIm EO Me] (TFSI ) (8c) (Figure 4) vs solvent dielectric
2
2
2
constant.
downfield peaks are due to solvent interactions, while the
major signals arise from intramolecular hydrogen bonding of
for H vs the dielectric constant of the solvent yields a linear
1
H with the ether oxygens of the tail. The peak ratio is only
1
53−55
relationship (Figure 5). No correlation exists with E (30).
weakly dependent on concentration, the fraction of downfield
T
Proton H was chosen for analysis because the two signals of
signal changing from 80.6% to 75.9% over the range 0.10−10
1
this proton were better resolved than those for H . The higher
mM in CD CN, indicating that the observations are based on
2
3
the dielectric constant, the more competitive the solvent would
be for complexation of the imidazolium moiety compared to
the ethyleneoxy tail; in DMSO only the solvent interaction is
observed. Figure 5 supports the theory that the smaller
intramolecular interactions in the imidzolium salt. As an
additional note, studies have shown that the main component
responsible for solvation of imidazolium compounds is the
21,56
solvent’s ability to hydrogen bond with H1.
One study
F
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
+
−
Figure 6. Partial H NMR spectrum of NB[C(O)OE Im EO Me] (TFSI ) (8t), 600 MHz, CDCl : (A) 25 °C and (B) −26 °C.
4
4
2
2
3
Figure 7. Partial 13C NMR of NB[C(O)OEIm EO Me] (TFSI ) (8c) at 126 MHz: (A) solvent CD Cl and (B) solvent DMSO-d .
+
−
2
2
2
2
2
6
using a pillararene-based rotaxane with an imidazolium cation
showed that the interaction was limited when hydrogen-
bonding solvents were used and maximized with solvents
57
lacking a hydrogen-bond acceptor. These and other prior
G
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
+
−
Figure 8. Partial 2D NOESY NMR spectrum of NB[C(O)OEIm EO Me] (TFSI ) (8d) in acetone-d .
3
2
2
6
+
−
Figure 9. Left: summary of 2-D NOESY correlations for NB[C(O)OEIm EO Me] (TFSI ) (8d) in acetone-d . Center: possible conformation of
3
2
2
6
the tail group of 8d for H-bonding with 2-H of the imidazolium moiety. Right: CPK model showing a similar interaction in
+
−
NB[C(O)OEIm EO Me] (TFSI ) (8c).
2
2
2
3
1−41
studies
along with the present data support the hypothesis
brium between the two states. Thus, in 8t and analogues, the
longer ethyleneoxy tails allow for fast exchanging interactions
with the imidazolium units. Note that the chemical shift of H1
in 8t lies between the two peaks observed in the same solvent
that as the solvent becomes more nonpolar or contains less/
weaker hydrogen-bond acceptors, the interaction of the
imidazolium cation with intramolecular ethyleneoxy units
increases.
for H of 8c (Figure 4a), as expected for this situation. An
1
However, it is important to note that some monomers 8 in
attempt was made to test this hypothesis by using low-
temperature NMR; Figure 6B shows the spectrum of 8t at −26
°C (the instrumental limit); no noticeable change was
observed. This indicates the temperature was not low enough
to slow the exchange sufficiently to observe the individual
signals.
relatively nonpolar solvents such as CDCl display only a single
3
1
set of signals in their H NMR spectra; no extra peaks are
observed for the imidazolium protons. This is typically
observed with longer ethyleneoxy tail units (Y = 3 or more).
1
As an example, Figure 6A shows the H NMR spectrum of 8t
in CDCl . We consider the two different possibilities for
The observation of peaks in CDCl
is not limited to H NMR; this effect is also observed in the
3
that disappear in DMSO
3
1
complexation: fast and slow exchange. Under slow exchange
two peaks are present for each hydrogen, representing the self-
complex and simply solvated states, whereas under fast
exchange a single peak represents the time-averaged equili-
1
3
13
C NMR spectra. Figure 7 contains partial C NMR spectra
of 8c in CD Cl and DMSO-d ; in DMSO-d a single set of
2
2
6
6
1
3
C signals was obtained, but in CD Cl two distinct sets of
2
2
H
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
+
−
Figure 10. H NMR spectrum (500 MHz) of NB[C(O)OE Im EO Me] (TFSI ) (8r) in DMSO-d at room temperature.
4
2
2
2
6
Figure 11. 13C NMR spectrum (126 MHz) of NB[C(O)OE Im EO Me] (TFSI ) (8r) in DMSO-d at room temperature.
+
−
4
2
2
2
6
I
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
+
−
Figure 12. 2D NOESY (500 mHz) of NB[C(O)OE Im EO Me] (TFSI ) (8r) in CDCl at room temperature; degassed using argon.
4
2
2
2
3
signals were observed. So clearly two different environments
are involved in the nonpolar medium, consistent with a
hydrogen-bonding interaction in competition with another
solvated form.
because of its superior RT conductivity compared to 8b and
8e (Table 2); all are monomers that contain a single
oxyethylene unit linking the norbornene to the imidazolium
unit (X = 1). 8d exhibited a RT conductivity more than double
those of 8b, 8c, or 8e, perhaps attributable to interaction of its
ethyleneoxy tail and imidazolium units, thus liberating a higher
fraction of TFSI counterions. Figure 8 is a partial 2D NOESY
spectrum of 8d. Peak assignments were made by careful
A model system, N,N′-dimethylimidazolium TFSI was
studied in five ethereal solvents; as shown in Figures S45−
S47, the chemical shift of H depends on the number of ether
1
oxygen atoms in the solvent.
1
To further explore the proposition that the monomers
experience self-complexation, 2D NOESY was employed; this
experiment probes through-space interactions among the
protons. For this experiment monomer 8d was selected
analysis of H NMR and COSY data (not shown).
In Figure 8, hydrogens B, F, and G of 8d all show
correlations with the 2-proton (C) of the imidazolium ring, as
expected due to their close proximity. The following key points
J
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
+
−
Figure 13. Partial H NMR spectra (500 MHz, CDCl ) of NB[C(O)OE Im EO Me] (TFSI ) (8r): (i) normal spectrum, (ii) 1D NOESY
3
4
2
2
2
irradiation of N (signal not shown), (iii) 1D NOESY irradiation of O (signal not shown), and (iv) 1D NOESY irradiation of P (signal not shown).
from the NOESY spectrum give insight into the geometry of
the molecule: (1) A is not coupled to C; (2) G is coupled to C,
D, and E; and (3) I and/or J are/is coupled to C, D, and E,
while H and K are not. These correlations are shown in Figure
restricted rotation, which could result from intramolecular
1
3
hydrogen bonding. To probe this, C NMR and NOESY
1
3
spectra were obtained. In the C NMR spectrum (Figure 11)
no carbon signals are duplicated; 22 signals were expected, and
21 signals were found; the peak located at 68.62 ppm is
suspected to be two overlapping peaks. This provides evidence
that only protons F in each arm are nonequivalent and not the
attached carbons in the two arms.
9
with double arrows indicating hydrogens that are close to
one another. Figure 9 also contains a suggested structure for
the imidazolium tail group of 8d. One possible explanation for
hydrogens I and J being a considerable distance from the
imidazolium ring but showing coupling, while hydrogens H
and K do not is that the ethyleneoxy tail group folds back on
top of the imidazolium ring, as shown in Figure 9. This could
account for the coupling observed in the NOESY and allow
one of the oxygen atoms in the ethyleneoxy unit to be within
the needed proximity for hydrogen bonding to the
imidazolium cation.
In the 2D NOESY spectrum in CDCl (Figure 12) one of
3
the F proton signals of 8r is coupled to either or both of the L
or R protons while the other is not. However, 1D NOESY
proved that protons F were within close proximity of the
imidazolium cation, since irradiation of protons N, O, or P
quenched both F signals (Figure 13, ii, iii, and iv). Figure 14
provides a summary of notable interactions in both the 2D and
1D NOESY. The methyl group protons U couple weakly to the
imidazolium ring and the norbornene protons B and C couple
to the imidazolium protons N, O, and P. One possibility is that
the ethyleneoxy units wrap around the imidazolium ring; this
would explain why proton P is through-space coupled to
proton C. Given this result and considering the electron
deficiency of the imidazolium cation, it is likely that the
tetra(oxyethylene) tether also coordinates with the imidazo-
lium cation. Thus, the tetra(oxyethylene) tether in 8r plays two
roles: (1) it allows the imidazolium moiety to be far enough
from the norbornene unit to allow optimal binding with the
di(ethyleneoxy) tail, and (2) it also directly interacts with the
cation.
To understand the role of the oxyethylene spacer in 8r (the
1
most conducting monomer), H NMR data were assessed in
DMSO (Figure 10), which diminishes or eliminates intra-
molecular hydrogen bonding. Assignments were made via
careful analysis of the 2D COSY spectrum (not shown). The
first distinguishing feature observed in all solvents is the
splitting of protons F into two separate signals. Further
inspection reveals that these two peaks are overlapping triplets,
which would derive from a classic ABX system through large
2
A−X and B−X couplings (J ∼ 12 Hz) and small A−B coupling
(
J ∼ 5 Hz). This indicates that the two arms are equivalent, but
within each arm the two hydrogens F are in different
environments, as a result of conformational effects, i.e.,
K
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
product was the endo isomer. Consequently, this product was
isomerized and hydrolyzed in a single process, according to the
58
work by Kanao et al., to provide an exo isomer rich
norbornene carboxylic acid. The exo isomer is desirable for
polymerizations as it has less steric hindrance and polymerizes
47−51,58
more readily.
The exo isomer 13 was isolated by
selectively lactonizing the endo isomer in the presence of
iodine. The exo acid 13 was converted into the acid chloride, in
the presence of magnesium to preserve the alkene, and then
esterified with tetra(ethylene glycol) monotosylate. The
resulting tosylate monomer precursor 14 was then used to
quaternize the appropriately functionalized imidazole 6,
yielding the desired one-armed norbornene-based imidazolium
TFSI monomers 15. These are analogues of two-armed
analogues 8p−8t.
Figure 14. 1D and 2D NOESY summary for monomer NB[C(O)-
+
−
OE Im EO Me] (TFSI ) (8r).
4
2
2
2
The glass transition temperatures and room temperature
conductivities for the four monomers are shown in Table 3. All
These NMR studies support the hypothesis that the
oxyethylene tethers and the terminal ethyleneoxy segments
both interact with the imidazolium cations. This has two
consequences: (1) lowering of the binding energy of the TFSI
anions, thus leading to an increase in the fraction of charge
carriers, the free TFSI anions, and (2) an increase in the glass
transition temperature relative to a situation without the H-
bonding interactions (monomers with Y = 1 and those with a
butyl tail, Table 1).
20,25
the T values are low compared to similar acrylate
and two-
g
armed norbornene monomers (Table 1). This was a good
indication that the conductivities of these monomers would be
higher, as indeed they are.
The acrylate TFSI monomers displayed room temperature
−
4
20
conductivities up to 2.2 × 10 S/cm. All the present one-
armed ROMP monomers had conductivities higher than 10−
S/cm at room temperature, 3−4-fold higher than their two-
armed analogues 8p−8t (Table 2). 15c had the highest room
temperature conductivity, corresponding well with the result
for two-armed analogue 8r (Table 2).
4
As a final note, it should be mentioned that the two-armed
norbornene imidazolium TFSI monomers developed here were
1
shelf stable with no noticeable change in their H NMR spectra
over the course of several years; i.e., they did not undergo
autopolymerization.
Synthesis and Properties of Two-Armed Norbornene-
All of these monomers have X = 4, and the maximium
conductivity was observed with Y = 2. Likewise, Y = 1 has the
minimum conductivity in this series as in series 8p−8t (Table
2). In other words, the performance of series 15 nicely parallels
that of series 8.
Based Imidazolium Monomers with Other Counter-
anions. The advantages of the TFSI counterion over PF are
6
due to it being a much larger, but flexible, ion with well-
Synthesis and Properties of Two-Armed Oxanorbor-
nene-Based Imidazolium TFSI Monomers 18. The
syntheses of the two-armed oxanorbornenes 18 mirrored
those of the two-armed norbornenes 8 in terms of method-
ology, except that the starting material was the exo diacid 16
instead of the anhydride (Scheme 6); esterification was
achieved via EDCI coupling. These monomers undergo
retro-Diels−Alder reactions at modest temperatures (∼>50
delocalized charge. This lowers the T , reduces ion pairing, and
g
23,25
in turn results in higher conductivities.
For this reason, we
chose to evaluate the anions shown in Scheme 4: tetrakis[3,5-
bis(trifluoromethyl)phenyl]borate (BARF in 9r, X = 4, Y = 2),
tetrakis(perfluorophenyl)borate (TPFB in 10r, X = 4, Y = 2),
1
,4-bis-2-ethylhexylsulfosuccinate (AOT in 11a, X = 1, Y = n-
butyl), and dodecyl sulfate (DDS in 12a, X = 1, Y = n-butyl).
59−62
As detailed in Table S1, these resulted in RT conductivities
°C).
2
5
1
0 −10 lower than the TFSI analogues!
Table 4 lists the glass transition temperatures of these
Synthesis and Properties of One-Armed Norbornene-
oxanorbornene derivatives. The T values of 18a, 18b, and 18c
g
Based Imidazolium TFSI Monomers 15. One-armed
norbornene-based imidazolium TFSI monomers were synthe-
sized through a multistep process as depicted in Scheme 5.
Methyl norbornenate was synthesized from methyl acrylate
and freshly cracked cyclopentadiene. The majority of this
are each within 2 °C of their norbornene analogues 8l, 8q, and
8r (Table 1) but >21 °C higher than the one-armed
norbornenes 15b and 15c (Table 3).
Table 4 also lists the RT conductivities of oxanorbornenes
18. The oxonorbornenes 18b and 18c display RT conductivity
Scheme 4. Bulky Anions Used in This Study
L
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 5. Synthesis of One-Armed Norbornene-Based Imidazolium TFSI Monomers 15
Table 3. Glass Transition Temperatures and Room
Temperature Conductivities of One-Armed Norbornene
Imidazolium TFSI Monomers 15
Table 4. Glass Transitions and Room Temperature
Conductivities of Two-Armed Oxanorbornene Imidazolium
TFSI Monomers 18
b
b
T
αDC at 25 °C
(S/cm)
T
αDC at 25 °C
(S/cm)
g
g
a
a
monomer
Y
NAME
(°C)
compd
X
Y
name
(°C)
+
−
−4
1
1
5a
5b
0
1
NBC(O)OE Im Me TFSI
−55
−65
2.80 × 10
18a
3
1
ONB[C(O)
−47
−
4
+
−
+
−4
OE Im EOMe] (TFSI )
NBC(O)OE Im EOMe
2.07 × 10
3
2
2
4
−
1.2 × 10−4
7.9 × 10−5
TFSI
18b
18c
4
4
1
2
ONB[C(O)
−42
−44
+
−
+
4.39 × 10−4
2.79 × 10−4
OE Im EOMe] (TFSI )
1
1
5c
2
3
NBC(O)OE Im
−69
4
2
2
4
−
EO MeTFSI
ONB[C(O)
2
+
−
+
OE Im EO Me] (TFSI )
5d
NBC(O)OE Im EO Me
−56
4
2
2
2
4
3
−
a
TFSI
Abbreviations for names: same as Scheme 3 with ONB =
a
b
b
Abbreviations for names: same as Scheme 3. Determined by DSC at
0 °C/min during the second heating scan. The midpoint of the
change in heat capacity was defined as the glass transition
oxanorbornene. Determined by DSC at 10 °C/min during the
second heating scan. The midpoint of the change in heat capacity was
defined as the glass transition temperature.
1
temperature.
values comparable to their two-armed orbornene analogues 8q
and 8r (Table 2) but lower than the one-armed norbornenes
Scheme 6. Synthesis of Two-Armed Oxanorbornene-Based Imidazolium TFSI Monomers 18
M
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
8
=
5 (Table 3). Although there are only two points, unlike series
and 15, here the maximum conductivity is not observed for Y
2, but rather for Y = 1.
100 °C for 24 h. Each sample was then annealed in the Novocontrol
at 120 °C in a heated stream of nitrogen for 1 h prior to
measurements. However, since the oxanorbornenes are thermally
labile and undergo retro-Diels−Alder reactions, these films were
simply dried in vacuo at room temperature. The conductivity was
CONCLUSIONS
−
1
■
measured using a sinusoidal voltage with amplitude 0.1 V and 10 −
7
A matrix of two-armed norbornene-based imidazolium TFSI
monomers (8) was synthesized to determine the optimal
structure in terms of single ion conductivity. Thirty-seven new
compounds were synthesized, 22 of which were two-armed
norbornene imidazolium TFSI monomers. For the chain
tethering the imidazolium ring to the norbornene ring three or
four oxyethylene units (X = 3 or 4) are optimal. A terminal
group of two ethyleneoxy units (Y = 2) was optimal. NMR
studies indicated that the terminal ethyleneoxy units interact
with the imidazolium cation via hydrogen bonding; the longer
tethers also interact with the imidazolium moieties. The
1
0 Hz frequency range for all experiments. Data were collected in
isothermal frequency sweeps every 5 K from 120 to −100 °C.
Materials. Reagents were purchased and used as received.
63
64,65
64
20
Compounds 1, 5a,
5b, and 6c were prepared in accordance
with literature procedures.
General Procedure 1: Norbornene Esters 2. exo,exo-Bis(2-(2′-
hydroxyethoxy)ethyl) Bicyclo[2.2.1]hept-5-ene-2,3-dicarboxylate
{
1
NB[C(O)OE OH] , 2a}. exo-Norbornene anhydride (1) (19.63 g,
19.6 mmol) was added to a round-bottom flask containing
2 2
di(ethylene glycol) (500 mL, 3.74 mol), and the mixture was stirred
briefly before adding sulfuric acid (0.5 mL, 9 mmol). The reaction
mixture was heated at 70 °C with stirring for 20 h. After cooling to
room temperature, the reaction mixture was poured into water (1 L)
and extracted with DCM (200 mL) 3×. The DCM extracts were
combined and washed with 1 M HCl (100 mL × 3), saturated
−
5
optimal system 8r exhibited a conductivity of 9.57 × 10 S/
cm at 25 °C and a T of −46 °C. A decoupling of lower T
g
g
values with higher conductivity values was observed as a result
of the competition between the tether and terminal ethers for
H-bonding with the imidazolium cations. The observed shelf
stability and reasonably high conductivities of monomers 8a−
NaHCO (100 mL × 2), and water (100 mL × 3) and dried over
3
sodium sulfate. Filtration and removal of solvent provided the desired
1
product as a yellow tinted oil (32.98 g, 77%). H NMR (500 MHz,
CDCl ): δ 6.22 (s, 2H), 4.24 (m, 4H), 3.77−3.65 (m, 8H), 3.59 (m,
3
8
v render them promising precursors for effective polymers for
4
H), 3.18 (s, 2H), 3.13−3.08 (m, 2H), 2.68 (s, 2H), 2.12 (d, J = 9
1
3
use in ion conducting devices.
Hz, 1H), 1.50 (d, J = 9 Hz, 1H). C NMR (126 MHz, CDCl ): δ
3
Four new imidazolium/anion combinations (9r, 10r, 11a,
and 12a) were tested for RT conductivity with BARF, TPFB,
AOT, and DDS anions, respectively. When compared to their
TFSI counterpart, in terms of conductivity, the resulting
materials were vastly inferior.
Four new one-armed norbornene-based imidazolium TFSI
monomers (15) were prepared with tetra(oxyethylene) linkers
173.60, 137.94, 72.58, 68.82, 63.99, 61.57, 47.44, 45.55, 45.44 (9
signals expected and 9 signals found). HR MS: calcd for C H O [M
+ H] : m/z 359.1700; found: m/z 359.1705 (error 1 ppm).
1
7
27
8
+
General Procedure 2: Norbornene Ditosylates 3. exo,exo-Bis(2-
(
2′-tosyloxyethoxy)ethyl) Bicyclo[2.2.1]hept-5-ene-2,3-dicarboxy-
late {NB[C(O)OE OTs] , 3a}. exo,exo-Norbornene diol 2 (5.05 g,
2
2
1
4.1 mmol), TEA (33 mL, 0.24 mol), and DCM (20 mL) were
combined in a round-bottom flask and cooled to 0 °C. Tosyl chloride
41.97 g, 220.1 mmol) was dissolved in DCM (80 mL) and added
(
X = 4) and variable ethyleneoxy terminal groups (Y). All of
(
−4
these exhibited room temperature conductivities >10 S/cm,
the highest being 4.39 × 10− S/cm for 15c (Y = 2), the
analogue of 8r. The results parallel those for the two-armed
analogues and provide hope for outstanding polymers.
dropwise via an addition funnel to the norbornene diol solution; once
addition was complete, the reaction mixture was allowed to warm to
room temperature and stirred for 2 days. The reaction mixture was
poured into water (100 mL), and the organic layer was washed with
water (15 mL × 4) and saturated NaCl (15 mL × 1) and dried over
sodium sulfate. After filtration and removal of solvent the oil was
washed with hexanes (100 mL × 2) and purified by flash column
4
Three new oxanorbornene-based two-armed imidazolium
TFSI monomers (18) were prepared with varied linkers and
terminal groups. 18b (X = 4, Y = 1) possesses a room
temperature conductivity of 1.2 × 10− S/cm. However, retro-
Diels−Alder reactions are problematic with these monomers.
chromatography (silica eluting with DCM to acetonitrile) to yield a
4
1
clear oil (3.97 g, 42%). H NMR (500 MHz, CDCl ): δ 7.81 (d, J = 8
3
Hz, 4H), 7.36 (d, J = 8 Hz, 4H), 6.22 (m, 2H), 4.23−4.15 (m, 6H),
4
2
.11−4.04 (m, 2H), 3.72−3.67 (m, 4H), 3.62 (m, 4H), 3.10 (s, 2H),
EXPERIMENTAL SECTION
.65 (d, J = 2 Hz, 2H), 2.46 (s, 6H), 2.07 (d, J = 9 Hz, 1H), 1.49 (d, J
■
Measurements. 1H NMR spectra were obtained on JEOL
ECLIPSE-500, BRUKER 500, and AGILENT NMR vnmrs400
spectrometers; 13C NMR spectra were collected on these instruments
13
= 9 Hz, 1H). C NMR (126 MHz, CDCl
): δ 173.43, 144.89, 137.95,
3
132.98, 129.86, 127.95, 69.15, 69.14, 68.58, 63.60, 47.23, 45.74, 45.33,
21.64 (14 peaks expected and 14 peaks found). HR MS: calcd for
1
+
at 126, 126, and 101 MHz, respectively. H NMR splitting
C
H
31
39
O
S
12
2
[M + H] : m/z 667.1877; found: m/z 667.1842 (error
+
abbreviations: s (singlet), d (doublet), t (triplet), q (quartet), quin
−5.3 ppm); calcd for C31
found: m/z 684.2124 (error 2.8 ppm); calcd for C31H O S
39c 12 2
+ Na] : m/z 689.1697; found: m/z 689.1681 (error 2.3 ppm).
H
42
O
12
S
2
N [M + NH
4
] : m/z 684.2143;
(
quintet), sex (sextet), h (heptet); coupling constants, J, are in Hertz.
Na [M
+
HR-MS were obtained using an Agilent LC-ESI-TOF system. DSC
data were obtained with a TA Instruments Q2000 differential
General Procedure 3: Monotosylation of Glycols. Penta(ethylene
glycol) Monotosylate, HOE OTs. Penta(ethylene glycol) (19.31 g,
scanning calorimeter under N at heating and cooling rates of 10 °C/
5
2
min; T values were recorded on second heating scans as the midpoint
81.04 mmol) and NaOH (3.21 g, 80.2 mmol) were dissolved in water
(75 mL) and cooled to 0 °C with magnetic stirring. Tosyl chloride
(15.51 g, 81.35 mmol) in THF (50 mL) was added dropwise to the
reaction mixture via an addition funnel. After the tosyl chloride
solution was completely added, the reaction mixture was stirred at
room temperature for 24 h. THF was removed by rotary evaporation,
and the aqueous layer was extracted with DCM (3 × 50 mL). The
organic mixture was washed with 10% HCl (4 × 25 mL), water (3 ×
25 mL), and saturated NaCl (1 × 25 mL). DCM was removed by
rotary evaporation, and the resulting material was passed over silica;
100% DCM eluted unreacted TsCl; 100% EA eluted a ditosylated
g
of the change in heat capacity. The ionic conductivity measurements
were performed by dielectric relaxation spectroscopy. Samples were
prepared by allowing them to flow to cover a 30 mm diameter freshly
polished brass electrode at 100 °C in vacuo. To control the sample
thickness at 100 μm, silica spacers were placed on top of the sample
after it flowed to cover the electrode. Then a 15 mm diameter freshly
polished brass electrode was placed on top to make a parallel plate
capacitor cell which was squeezed to a gap of 100 μm in the
instrument (with precise thickness checked after dielectric measure-
ments were complete). The samples sandwiched between two
electrodes were positioned in a Novocontrol GmbH Concept 40
broadband dielectric spectrometer, after being in a vacuum oven at
product (13.28 g), and last 100% acetonitrile eluted the desired
1
product as a clear oil, 8.74 g (27%). H NMR (500 MHz, CDCl ): δ
3
N
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
7
3
1
7
.80 (d, J = 8 Hz, 2H), 7.34 (d, J = 8 Hz, 2H), 4.21−4.11 (m, 2H),
(t, J = 7 Hz, 6H). 13C NMR (126 MHz, DMSO-d ): δ 172.45, 137.71,
136.48, 122.76, 122.50, 119.44 (q, J = 323 Hz), 62.25, 48.62, 48.00,
46.46, 45.07, 44.70, 31.31, 18.68, 13.23 (14 signals expected and 14
6
1
3
.78−3.52 (m, 18H), 2.45 (s, 3H). C NMR (126 MHz, CDCl ): δ
3
44.78, 133.01, 129.81, 127.98, 72.45, 70.75, 70.61, 70.60, 70.55,
+
0.53, 70.35, 69.24, 68.68, 61.75, 21.65 (15 signals expected and 15
signals found). HR MS: calcd for C H O N S F [M − TFSI] : m/z
2
9
40
8
5 2 6
+
signals found). HR MS: calcd for C H O SN [M + NH ] : m/z
764.2217; found: m/z 764.2215 (error −0.2 ppm); calcd for
1
7
32
8
4
2+
4
10.1843; found: m/z 410.1836 (error 2 ppm); calcd for C H O S
C H O N [M − 2 TFSI] : m/z 242.1525; found: m/z 242.1524
1
7
29
8
27 40
4
4
+
[
M + H] : m/z 393.1583; found: m/z 393.1563 (error 5.1 ppm);
(error −0.4 ppm). T (DSC) = −44 °C. Conductivity at 25 °C = 3.07
g
+
−5
calcd for C H O SK [M + K] : m/z 431.1142; found: m/z
× 10 S/cm.
17
28
8
4
31.1126 (error 3.7 ppm).
exo,exo-Bis(2-bromoethyl) Bicyclo[2.2.1]hept-5-ene-2,3-dicar-
boxylate {NB[C(O)OEBr] , 4}. Norbornene anhydride 1 (3.58 g,
General Procedure 4. Esterification of 5-Norbornene-exo-2,3-
2
dicarboxylic Anhydride. exo,exo-Bis(2-(2′-(2″-(2’’’-(2’’’’-
tosyloxyethoxy)ethoxy)ethoxy)ethoxy)ethyl) Bicyclo[2.2.1]hept-5-
ene-2,3-dicarboxylate {NB[C(O)OE OTs] , 3d}. 5-Norbornene-exo-
21.8 mmol) was added to a round-bottom flask containing 2-
bromoethanol (37.8 g, 302 mmol), and the mixture was stirred briefly
before adding sulfuric acid (0.1 mL, 2 mmol). The reaction mixture
was heated at 70 °C with stirring for 20 h. After cooling to room
temperature, the reaction mixture was poured into water and
extracted with DCM three times. The DCM extracts were combined
5
2
2
,3-dicarboxylic anhydride (1) (0.88 g, 5.4 mmol) was combined in a
round-bottom flask with penta(ethylene glycol) monotosylate (4.06 g,
0.3 mmol), EDCI (1.61 g, 10.4 mmol), DMAP (0.13 g, 1.1 mmol),
1
and DCM (20 mL). The reaction mixture was stirred at room
temperature for 3 days, after which 30 mL of DCM was added, and
the mixture was poured into water. The organic layer was collected
and washed with water (3 × 10 mL). Solvent was then removed by
rotary evaporation. Column chromatography (silica) using 100%
DCM provided the desired product, 4.16 g (83%), as a viscous oil
and washed with NaHCO (×4) and water (×3) and dried over
3
sodium sulfate. Filtration and removal of solvent provided a material
that was passed through a silica plug, eluted with DCM to provide the
1
desired product as an oil (6.87 g, 80%). H NMR (500 MHz,
CDCl ): δ 6.24 (m, 2H), 4.38 (m, 4H), 3.54−3.48 (m, 4H), 3.17−
3
13
3
.13 (m, 2H), 2.68 (m, 2H), 2.07 (m, 1H), 1.52 (m, 1H). C NMR
containing slight amounts of EDCI and DMAP. The product was
(
126 MHz, CDCl ): δ 173.07, 137.94, 64.10, 47.13, 45.93, 45.27,
1
3
used without further purification. H NMR (400 MHz, CDCl ): δ
3
2
8.81 (7 signals expected and 7 signals found). HR MS: calcd for
7
4
2
.71−7.69 (m, 4H), 7.17−7.15 (m, 4H), 6.17 (s, 2H), 4.29−4.27 (m,
H), 3.62−3.55 (m, 48H; overlaps with EDCI protons), 3.06 (s,
H), 2.61 (s, 2H), 2.33 (s, 6H), 2.04 (d, J = 9 Hz, 1H), 1.45 (s, 1H).
+
C H O fragment [M − C H Br + H] : m/z 209.0814; found: m/z
1
1
13
4
2
4
2
209.0795 (error −9.1 ppm).
+
−
exo,exo-NB[C(O)OE Im EO Me] ( OTs) (7r). Norbornene ditosy-
1
3
4
2
2
2
C NMR (101 MHz, CDCl ): δ 173.47, 144.79, 137.90, 132.88,
3
late 3c (1.89 g, 2.24 mmol) and imidazole 6c (0.708 g, 5.61 mmol)
were combined in a round-bottom flask under nitrogen and heated at
1
4
29.02, 126.08, 70.65, 70.50, 70.48, 70.36, 70.21, 70.13, 68.80, 67.03,
7.18, 45.70, 45.28, 21.36 (20 signals expected and 18 signals found, 2
70 °C with stirring for 3 days. After cooling to room temperature, the
signals in the ethyleneoxy region are suspected to be overlapping).
EDCI and DMAP NMR peaks were omitted for clarity. HR MS: calcd
for C H O S N [M + NH ] : m/z 948.3716; found: m/z 948.3721
reaction mixture was taken up in water (50 mL) and extracted with
DCM (200 mL) in a liquid−liquid apparatus for 2 days. The aqueous
layer was collected, and water was removed by rotary evaporation.
The resulting material was placed under a stream of nitrogen for 1 day
+
43
66 18
2
4
+
(
error 0.5 ppm); calcd for C H O S Na [M + Na] : m/z 953.3275;
43 62 18 2
found: m/z 953.3205 (error 7.3 ppm).
1
and vacuum for 2 days (1.75 g, 80%, a viscous liquid). H NMR (500
General Procedure 5: N-Alkylimidazoles 6. 1-n-Butylimidazole
ImBu, 6a). Imidazole (16.95 g, 249 mmol) and sodium hydroxide
MHz, D O): δ 7.58 (d, J = 8 Hz, 4H), 7.48−7.43 (m, 4H), 7.26 (d, J
2
(
=
8 Hz, 4H), 6.17 (t, J = 2 Hz, 2H), 4.33−4.28 (m, 8H), 4.25−4.17
(
(
9.40 g, 235 mmol) were added to a round-bottom flask with water
10 mL) and stirred until the imidazole had completely dissolved. 1-
(
m, 2H), 4.06−3.99 (m, 2H), 3.80 (q, J = 5 Hz, 8H), 3.64 (t, J = 5
Hz, 4H), 3.60−3.52 (m, 20H), 3.49−3.45 (m, 4H), 3.24 (s, 6H),
Chlorobutane (21 mL, 0.26 mol) was diluted with THF (40 mL) and
added to the reaction flask. Reflux was achieved and maintained for 12
h. THF was removed by rotary evaporation, and the crude material
was poured into water (50 mL). After extraction with DCM (3 × 100
mL), the organic layers were combined and washed with 10% sodium
hydroxide (20 mL × 2) and water (20 mL × 3) and dried over
sodium sulfate. Filtration and removal of the solvent provided the
3
9
1
6
4
.03−2.98 (m, 2H), 2.66 (d, J = 2 Hz, 2H), 2.29 (s, 6H), 1.80 (d, J =
1
3
Hz, 1H), 1.37 (d, J = 9 Hz, 1H). C NMR (126 MHz, D O): δ
2
75.87, 142.41, 139.42, 137.94, 129.40, 125.32, 122.62, 122.54, 70.88,
9.63, 69.51, 69.50, 69.47, 68.38, 68.31, 64.16, 58.01, 49.13, 49.11,
7.28, 45.47, 44.62, 20.44 (23 signals expected and 23 signals found).
2
+
HR MS: calcd for C H N O [M − 2 OTs] : m/z 420.2361;
found: m/z 420.2350 (error 2.6 ppm).
1R,2S,4R)-Bicyclo[2.2.1]hept-5-ene-2-carboxylic Acid (13). Cy-
41 82
4
20
1
desired product as a clear oil (21.27 g, 86%). H NMR (500 MHz,
(
CDCl ): δ 7.45 (s, 1H), 7.05 (s, 1H), 6.90 (s, 1H), 3.93 (t, J = 7 Hz,
3
clopentadiene (66 g, 1.0 mol) was cracked from the dimer by heating
2
H), 1.80−1.71 (m, 2H), 1.38−1.28 (m, 2H), 0.94 (t, J = 7 Hz, 3H).
1
3
under reflux and distilling the product below 45 °C. Methyl acrylate
C NMR (126 MHz, CDCl ): δ 137.09, 129.40, 118.78, 46.73, 33.10,
3
(86 g, 1.0 mol) was added to the cyclopentadiene. DCM (200 mL)
1
9.74, 13.51 (7 signals expected and 7 signals found). HR MS: calcd
+
was added. The resulting solution was heated under reflux for 8 h.
DCM and unreacted starting materials were removed under vacuum.
The product (145 g, 95% yield) was a light yellow oil. The mixture of
endo and exo isomers of the methyl ester (50.01 g, 328.6 mmol) was
heated in a solution of 250 mL of methanol and 20.12 g (372.6
mmol) of sodium methoxide under reflux to increase the exo isomer
content by isomerization. The ester was then slowly hydrolyzed at low
temperature by the slow addition of water (25 mL) over 8 h. The slow
hydrolysis coupled with the isomerization inverted the isomer mixture
from ∼80:20 endo:exo to ∼20:80 endo:exo. At this point the two
isomers were separated by iodo-lactonization. The acid isomers
(45.34 g, 328.6 mmol) were dissolved in 250 mL of an aqueous
solution of sodium carbonate (36.26 g, 342.1 mmol) and titrated with
80 mL of an aqueous solution of iodine (20.0 g, 78.8 mmol) and
potassium iodide (12.5 g, 75.3 mmol). The lactonized endo isomer
was extracted with diethyl ether. The aqueous solution was acidified
and the exo isomer was extracted with diethyl ether, 35.41 g (71%),
for C H N [M + H] : m/z 125.1073; found: m/z 125.1073 (error 0
7
13
2
ppm).
General Procedure 6: Norbornene Imidazolium TFSI Salts.
+
−
exo,exo-NB[C(O)OEIm Bu] (TFSI ) (8a). exo,exo-Norbornene dibro-
2
2
mide 4 (3.32 g, 8.39 mmol) and imidazole 6a (4.35 g, 35.0 mmol)
were combined in a round-bottom flask under nitrogen and heated at
7
0 °C with stirring for 3 days. After cooling to room temperature, the
reaction mixture was taken up in water (50 mL) and extracted with
DCM (200 mL) in a liquid−liquid apparatus for 2 days. LiTFSI (5.85
g, 20.4 mmol) was dissolved in water (10 mL) and added to the
aqueous solution. The aqueous mixture was stirred for 4 h and
allowed to sit overnight for the salt to “oil out”. The aqueous layer was
decanted and extracted twice with DCM (50 mL). The DCM extracts
were combined with the remaining oil and washed with water (15 mL
×
4). Solvent was removed by rotary evaporation, and the product
was dried under high vacuum for 12 h (3.29 g, 37%). H NMR (500
1
MHz, DMSO-d ): δ 9.19 (s, 2H), 7.82 (s, 2H), 7.79 (s, 2H), 6.23 (s,
6
6
6
67
1
2
H), 4.43 (m, 6H), 4.19 (m, 6H), 2.96−2.91 (m, 2H), 2.61 (m, 2H),
mp 36.6−39.0 °C, lit. mp 32.5−35.0, 43−44 °C. H NMR (500
1
.82−1.72 (m, 4H), 1.69 (d, J = 9 Hz, 1H), 1.34−1.18 (m, 5H), 0.90
MHz, CDCl ): δ 6.13 (dd, J = 6, 3 Hz, 1H), 6.09 (dd, J = 6, 3 Hz,
3
O
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
1
H), 3.14 (s, 1H), 2.97 (s, 1H), 2.30 (ddd, J = 9, 4, 3 Hz, 1H), 2.01−
.96 (m, 1H), 1.56 (d, J = 8 Hz, 1H), 1.42 (m, 2H).
OE OTs] , 17b}. A solution of tetra(ethylene glycol) monotosylate
4
2
(8.71 g, 25 mmol), oxanorbornene dicarboxylic acid (16, 2.21 g, 12
mmol), a catalytic amount of DMAP (305 mg, 2.5 mmol), and EDCI
(5.75 g, 30 mmol) in 100 mL of DCM was vigorously stirred for 3
days at room temperature. Then the reaction mixture was washed
with water. The organic layer was concentrated and purified through
flash column chromatography on a neutral alumina column using
exo-(1R,2S,4R)-2-(2′-(2″-(2”’-(Tosyloxy)ethoxy)ethoxy)ethoxy)-
ethyl Bicyclo[2.2.1]hept-5-ene-2-carboxylate (NBC(O)OE OTs, 14).
4
exo-Norbornene acid 13 (10.01 g, 72.45 mmol) was dissolved in 50
mL of thionyl chloride with 2 drops of dimethylformamide.
Magnesium turnings (2.04 g, 83.9 mmol) were added to the solution.
The mixture was stirred for 4 h at room temperature. Excess thionyl
chloride was removed under vacuum, and the resulting material was
extracted with dry DCM and cooled in an ice bath. Tetra(ethylene
glycol) monotosylate (12.08 g, 34.67 mmol) and dry pyridine (5.0
mL 62 mmol) were dissolved in dry DCM and added over 30 min to
the acid chloride solution. The resulting mixture was stirred at room
temperature for 8 h before the product was isolated by aqueous work-
CHCl as eluent. The product was a mixture of retro-DA adduct and
3
the title compound. The mixture was purified again by flash column
chromatography with CHCl −MeOH (95:5), which yielded 8.98 g
3
68
1
(82%) of 17b as colorless viscous oil. H NMR (CDCl , 500 MHz): δ
3
2.44 (s, 6H), 2.83 (s, 2H), 3.62−3.69 (m, 28H), 4.15 (t, 4H), 5.25 (s,
2H), 6.44 (s, 2H), 7.34 (d, J = 10 Hz, 4H), 7.79 (d, J = 10 Hz, 4H).
1
3
C NMR (CDCl , 125 MHz): δ 21.6, 46.8, 64.2, 68.7, 68.9, 69.3,
3
1
up as a light brown liquid (13.97g, 86%). H NMR (400 MHz,
CDCl ): δ 7.8 (m, 2H), 7.2 (m, 2H), 6.13 (dd, J = 6, 3 Hz, 1H), 6.09
70.4, 70.5, 70.6, 70.7, 80.6, 127.9, 129.8, 132.9, 136.6, 144.8, 171.4
(17 peaks observed, theory: 17).
3
+
−
(
(
2
dd, J = 6, 3 Hz, 1H), 4.26−4.21 (m, 4H), 4.17−4.12 (m, 1H), 3.68
General Procedure 9. exo,exo-ONB[C(O)OE Im EOMe] (TFSI )
3 2 2
m, 4H), 3.62 (m, 2H), 3.57 (d, J = 2 Hz, 2H), 3.15 (s, 1H), 2.96 (m,
(18a). The ditosylate 17a (1.89 g, 2.50 mmol) and 1-(2′-
methoxyethyl)imidazole (6b, 1 g, 6 mmol) were mixed and heated
at 40−45 °C for 19 h. The resultant viscous liquid was dissolved in 25
mL of water and washed several times with DCM to remove the
excess imidazole. To the aqueous solution was added a saturated
LiTFSI solution; after stirring 6 h, the separated viscous liquid was
H), 2.44 (s, 3H), 2.32 (dd, J = 10, 4 Hz, 1H), 2.24 (dd, J = 10, 4 Hz,
H), 1.99 (dt, J = 12, 4 Hz, 1H), 1.94−1.87 (m, 1H), 1.51 (d, J = 9
1
Hz, 1H), 1.42 (d, J = 10 Hz, 1H), 1.34 (d, J = 8 Hz, 1H). HR MS: m/
+
z 469.1841, (M + H) , calcd for C H O S m/z 469.1896 (error 12
ppm); m/z 486.2134, (M + NH ) , calcd for C H O NS m/z
2
3
33
8
+
4
23 36
8
+
4
86.2156 (error 4.5 ppm); m/z 491.1693 Da = (M + Na) , calcd for
decanted and washed freely with water and vacuum-dried to get the
1
product as pale yellow oil (0.23 g, 7%). H NMR (CDCl , 500 MHz):
C H NaO S m/z 491.1716 (error 4.7 ppm).
3
23
32
8
δ 2.90 (s, 2H), 3.37 (s, 6H), 3.63−3.70 (m, 12H), 3.84 (t, J = 5 Hz,
General Procedure 8: One-Armed Norbornene Imidazolium TFSI
+
−
Salts, X = 4. exo-NBC(O)OE Im MeTFSI (15a). Tosylate 14 (2.34 g,
5
4H), 3.94 (m, 4H), 4.12−4.38 (m, 4H), 4.56 (t, J = 5 Hz, 8H), 5.24
4
1
3
mmol) was added to a round-bottom flask equipped with a stir bar.
(s, 2H), 6.53 (s, 2H), 7.75 (s, 2H), 7.79 (s, 2H), 9.04 (s, 2H).
C
Imidazole 6f (1.36 g, 20 mmol) was added to the flask, and the
contents were dissolved in 50 mL of nitromethane. The resulting
solution was heated under reflux for 2 days. Nitromethane was
removed under vacuum, and the residue was suspended in
acetonitrile. The tosylate salt was isolated using silica gel column
chromatography, eluting first with acetonitrile and then with a mixture
of methanol, nitromethane, and 2 M aqueous ammonium chloride
NMR (acetone-d , 125 MHz): δ 47.6, 50.5, 50.6, 59.0, 64.6, 69.4,
6
69.6, 70.9, 71.0, 71.1, 81.5, 121.0 (q, J = 320), 123.7, 123.8, 137.5,
137.7, 172.4 (17 peaks observed, theory: 17). HR MS: m/z 333.1734
2
+
2+
(M − 2 TFSI) , calcd for (C32
H
N
O
)
m/z 333.1733 (error 0.3
+ m/
= −46.5 °C. Conductivity at 25 °C
50
4
11
+
ppm); m/z 946.2639 (M − TFSI) , calcd for C34
z 946.2644 (error 0.5 ppm). T
H F N O S
50 6 5 15 2
g
not measured.
(
7:2:1 v:v:v). The resulting solution was distilled under vacuum to
remove as much solvent as possible, and the residue was dissolved in a
minimum of water. Lithium TFSI (2.87 g, 10 mmol) was dissolved in
water and added to the aqueous solution. The monomer salt was then
extracted with DCM. The organic solution was washed three times
with 10% HCl and dried over sodium sulfate, and DCM was removed
ASSOCIATED CONTENT
Supporting Information
■
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.macro-
mol.8b02295.
under vacuum. The resulting product was a brown oil (3.07 g, 93%).
1
H NMR (400 MHz, DMSO-d ): δ 9.04 (s, 1H), 7.72 (m, 1H), 7.68
6
Syntheses of various intermediates and monomers, high-
resolution mass spectra for intermediates and mono-
mers; representative differential scanning calorimetric
determinations of glass transition temperatures (PDF)
(
3
(
m, 1H), 6.15 (m, 2H), 4.34 (t, J = 4, 2H), 4.15 (m, 2H), 3.86 (s,
H), 3.76 (t, J = 4, 2H), 3.62 (t, J = 6, 2H), 3.45−3.57 (m, 8H), 2.96
bs, 1H), 2.89 (bs, 1H), 2.17 (m, 1H), 1.81 (m, 1H) 1.40 (d, J = 8,
1
3
1
1
7
4
H), 1.28 (m, 2H). C NMR (101 MHz, DMSO-d ): δ 175.63,
6
38.31, 137.20, 136.02, 123.76, 123.10, 119.89 (q, J = 323), 70.13,
0.08, 69.98, 69.93, 68.75, 68.55, 63.63, 49.17, 46.48, 46.29, 42.86,
1.46, 36.12 (21 signals expected and 21 signals found). HR MS:
AUTHOR INFORMATION
■
+
calcd for C H N O (M − TFSI) m/z 379.2227; found m/z
3
20
31
2
5
Corresponding Author
*E-mail: hwgibson@vt.edu.
79.2227, (error 0 ppm). DSC (5.0 °C/min): T = −55 °C.
g
−4
Conductivity at 25 °C = 2.80 × 10 S/cm.
ORCID
U Hyeok Choi: 0000-0002-0048-9550
Harry W. Gibson: 0000-0001-9178-6691
Present Addresses
3
2
T.L.P.: Zestron Corp., 11285 Assett Loop, Manassas, VA
2
0109.
D.V.S: Momentive Performance Materials, Frendly, WV
26146.
δ 2.45 (s, 6H), 2.84 (s, 2H), 3.59 (s, 8H), 3.65−3.70 (m, 8H), 4.15−
M.A.: The Gandhigram Rural Institute-Deemed University,
Gandhigram, India.
S.L.: Department of Chemistry, University of California,
Berkeley, CA 94720.
4
.19 (m, 6H), 4.26−4.31 (m, 2H), 5.25 (s, 2H), 6.45 (s, 2H), 7.34 (d,
1
3
J = 10 Hz, 4H), 7.80 (d, J = 10 Hz, 4H). C NMR (CDCl , 125
3
MHz): δ 21.6, 46.9, 68.7, 69.0, 69.2, 70.4, 70.7, 80.7, 100.0, 127.96,
1
29.83, 133.0, 136.7, 144.8, 171.4 (15 peaks observed, theory: 15).
exo,exo-Bis(2-(2′-(2″-(2”’-(tosyloxy)ethoxy)ethoxy)ethoxy)ethyl)
-Oxabicyclo[2.2.1]hept-5-ene-2,3-dicarboxylate {ONB[C(O)-
Notes
7
The authors declare no competing financial interest.
P
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
liquids: biological effects in fish channel catfish ovary (CCO) cell line.
ACKNOWLEDGMENTS
■
Ecotoxicol. Environ. Saf. 2013, 92, 112−118.
We are very appreciative of financial support from the U.S.
Army Research Office under Grant W911NF-07-1-0452 Ionic
Liquids in Electroactive Devices (ILEAD) MURI. We are also
grateful to Prof. James McGrath (VT, deceased) and Tim
Long (VT) for use of their thermal analysis equipment and
Prof. James Runt (PS) for use of his dielectric spectrometer.
(19) Pereiro, A. B.; Araujo, J. M. M.; Martinho, S.; Alves, F.; Nunes,
S.; Matias, A.; Duarte, C. M. M.; Rebelo, L. P. N.; Marrucho, I. M.
Fluorinated ionic liquids: properties and applications. ACS Sustainable
Chem. Eng. 2013, 1, 427−439.
(20) Lee, M.; Choi, U H.; Colby, R. H.; Gibson, H. W. Structure-
property relationships: ion conduction in polymerizable ionic liquid
acrylates and their polymers. Chem. Mater. 2010, 22, 5814−5822.
(21) Lee, M.; Niu, Z.; Slebodnick, C.; Gibson, H. W. Structure and
REFERENCES
■
Properties of N,N-Alkylene Bis(N’-alkylimidazolium) Salts. J. Phys.
Chem. B 2010, 114, 7312−7319.
(
1) Ohno, H.; Yoshizawa-Fujita, M. Electrochemical Aspects of Ionic
Liquids, 2nd ed.; Wiley: New York, 2011.
2) Armand, M. B.; Bruce, P. G.; Forsyth, M.; Scrosati, B.;
Wieczorek, W. Energy Materials; Wiley: New York, 2011; pp 1−31.
3) Green, M. D.; Long, T. E. Designing imidazole-based ionic
liquids and ionic liquid monomers for emerging technologies. Polym.
Rev. 2009, 49, 291−314.
4) Singh, V. V.; Nigam, A. K.; Batra, A.; Boopathi, M.; Singh, B.;
Vijayaraghavan, R. Applications of ionic liquids in electrochemical
sensors and biosensors. Int. J. Electrochem. 2012, 2012, 1.
5) Meziane, R.; Bonnet, J.-P.; Courty, M.; Djellab, K.; Armand, M.
Single-ion polymer electrolytes based on a delocalized polyanion for
lithium batteries. Electrochim. Acta 2011, 57, 14−19.
6) Hirao, M.; Ito-Akita, K.; Ohno, H. Polymerization of molten salt
monomers having a phenylimidazolium group. Polym. Adv. Technol.
000, 11, 534−538.
7) Ohno, H. Molten salt type polymer electrolytes. Electrochim. Acta
001, 46, 1407−1411.
8) Zheng, L.; Chen, F.; Xie, M.; Han, H.; Dai, Q.; Zhang, Y.; Song,
(22) Aitken, B. S.; Lee, M.; Hunley, M. T.; Gibson, H. W.; Wagener,
(
K. B. Synthesis of precision ionic polyolefins derived from ionic
liquids. Macromolecules 2010, 43, 1699−1701.
(
(23) Lee, M.; Choi, U H.; Salas-de la Cruz, D.; Mittal, A.; Winey, K.
I.; Colby, R. H.; Gibson, H. W. Imidazolium polyesters. Adv. Funct.
Mater. 2011, 21, 708−717.
(
(24) Aitken, B. S.; Buitrago, C. F.; Heffley, J. D.; Lee, M.; Gibson, H.
W.; Winey, K. I.; Wagener, K. B. Precision ionomers: synthesis and
thermal/mechanical characterization. Macromolecules 2012, 45, 681−
(
6
(
87.
25) Choi, U H.; Lee, M.; Wang, S.; Liu, W.; Winey, K. I.; Gibson,
H. W.; Colby, R. H. Ionic conduction and dielectric response of
poly(imidazolium acrylate) ionomers. Macromolecules 2012, 45,
(
3
(
974−3985.
2
(
2
(
26) Choi, U H.; Mittal, A.; Price, T. L., Jr.; Gibson, H. W.; Runt, J.;
Colby, R. H. Polymerized ionic liquids with enhanced static dielectric
constant. Macromolecules 2013, 46, 1175−1186.
(27) Kidd, B. E.; Lingwood, M. D.; Lee, M.; Gibson, H. W.; Madsen,
C. New polyelectrolytes by ring-opening metathesis polymerization of
norbornene derivatives with imidazolium functionalized oligo-
alkylether pendant groups. React. Funct. Polym. 2007, 67, 19−24.
L. A. Cation and anion transport in a dicationic imidazolium-based
plastic crystal ion conductor. J. Phys. Chem. B 2014, 118, 2176−2185.
(28) Lee, M.; Choi, U H.; Colby, R. H.; Gibson, H. W. Ion
(
9) Shaplov, A. S.; Vlasov, P. S.; Armand, M.; Lozinskaya, E. I.;
conduction in a semi-crystalline polyviologen and its polyether
Ponkratov, D. O.; Malyshkina, I. A.; Vidal, F.; Okatova, O. V.; Pavlov,
G. M.; Wandrey, C.; Godovikov, I. A.; Vygodskii, Y. S. Design and
synthesis of new anionic “polymeric ionic liquids” with high charge
delocalization. Polym. Chem. 2011, 2, 2609−2618.
10) Smith, T. W.; Zhao, M.; Yang, F.; Smith, D.; Cebe, P. Imidazole
polymers derived from ionic liquid 4-vinylimidazolium monomers:
their synthesis and thermal and dielectric properties. Macromolecules
013, 46, 1133−1143.
11) Ozturk, B. O.; Yakut, E.; Ak, E.; Sehitoglu, S. K. Imidazole-
substituted ROMP polymers: Post-modification of poly(norbornene
diester) derivatives with aminolysis reactions. React. Funct. Polym.
017, 111, 22−29.
12) Cheng, Y.; Yang, J.; Hung, J.-H.; Patra, T. K.; Simmons, D. S.
mixtures. Macromol. Chem. Phys. 2015, 216, 344−349.
(29) Choi, U H.; Mittal, A.; Price, T. L., Jr.; Lee, M.; Gibson, H. W.;
Runt, J.; Colby, R. H. Molecular volume effects on the dynamics of
polymerized ionic liquids and their monomers. Electrochim. Acta
(
2
(
015, 175, 55−61.
30) Choi, U H.; Mittal, A.; Price, T.; Colby, R. H.; Gibson, H. W.
Imidazolium-based Ionic liquids as initiators in ring opening
polymerization: ionic conduction and dielectric response of end-
functional polycaprolactones and their block copolymers. Macromol.
Chem. Phys. 2016, 217, 1270−1281.
2
(
(31) Jourdain, A.; Serghei, A.; Drockenmuller, E. Enhanced ionic
2
(
conductivity of a 1,2,3-triazolium-based poly(siloxane ionic liquid)
homopolymer. ACS Macro Lett. 2016, 5, 1283−1286.
Design rules for highly conductive polymeric ionic liquids from
(32) Elmore, C. T.; Seidler, M. E.; Ford, H. O.; Merrill, L. C.;
molecular dynamics simulations. Macromolecules 2018, 51, 6630−
Upadhyay, S. P.; Schneider, W. E.; Schaefer, J. L. Ion transport in
solvent-free, crosslinked, single-ion conducting polymer electrolytes
for post-lithium ion batteries. Batteries 2018, 4, 28.
6
(
644.
13) Green, M. D.; Wang, D.; Hemp, S. T.; Choi, J.-H.; Winey, K. I.;
Heflin, J. R.; Long, T. E. Synthesis of imidazolium ABA triblock
copolymers for electromechanical transducers. Polymer 2012, 53,
(33) Gjikaj, M.; Brockner, W.; Namyslo, J.; Adam, A. Crown-ether
enclosure generated by ionic liquid componentssynthesis, crystal
structure and Raman spectra of compounds of imidazolium based
salts and 18-crown-6. CrystEngComm 2008, 10, 103−110.
(34) Noujeim, N.; Leclercq, L.; Schmitzer, A. R. N,N′-Disubstituted
methylenediimidazolium salts: a versatile guest for various macro-
cycles. J. Org. Chem. 2008, 73, 3784−3790.
3
(
677−3686.
14) Jangu, C.; Wang, J.-H. H.; Wang, D.; Sharick, S.; Heflin, J. R.;
Winey, K. I.; Colby, R. H.; Long, T. E. Well-defined imidazolium ABA
triblock copolymers as ionic-liquid-containing electroactive mem-
branes. Macromol. Chem. Phys. 2014, 215, 1319−1331.
(
15) Jangu, C.; Wang, J.-H. H.; Wang, D.; Fahs, G.; Heflin, J. R.;
Moore, R. B.; Colby, R. H.; Long, T. E. Imidazole-containing triblock
copolymers with a synergy of ether and imidazolium sites. J. Mater.
Chem. C 2015, 3, 3891−3901.
(35) Ganesan, K.; Alias, Y. Synthesis and characterization of novel
dimeric ionic liquids by conventional approaches. Int. J. Mol. Sci.
2008, 9, 1207−1213.
(
16) Margaretta, E.; Fahs, G. B.; Inglefield, D. L.; Jangu, C.; Wang,
(36) Luo, S.; Zhang, S.; Wang, Y.; Xia, A.; Zhang, G.; Du, X.; Xu, D.
Complexes of ionic liquids with poly(ethylene glycol)s. J. Org. Chem.
2010, 75, 1888−1891.
(37) Lee, M.; Niu, Z.; Schoonover, D.; Slebodnick, C.; Gibson, H.
W. 1,2-Bis[N-(N’-alkylimidazolium)]ethane salts as new guests for
crown ethers and cryptands. Tetrahedron 2010, 66, 7077−7082.
(38) Suresh, M.; Mandal, A. K.; Kesharwani, M. K.; Adarsh, N. N.;
Ganguly, B.; Kanaparthi, R. K.; Samanta, A.; Das, A. Folding and
D.; Heflin, J. R.; Moore, R. B.; Long, T. E. Imidazolium-containing
ABA triblock copolymers as electroactive devices. ACS Appl. Mater.
Interfaces 2016, 8, 1280−1288.
(
17) Hanabusa, K. Electrochemical Aspects of Ionic Liquids, 2nd ed.;
2
(
011; pp 395−401.
18) Radosevic, K.; Cvjetko, M.; Kopjar, N.; Novak, R.; Dumic, J.;
Srcek, V. G. In vitro cytotoxicity assessment of imidazolium ionic
Q
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
unfolding movements in a [2]pseudorotaxane. J. Org. Chem. 2011, 76,
(56) Headley, A. D.; Kotti, S. R. S. S.; Ni, B. Solvation effects on
imidazolium salts that contain alkyl side chains. Heterocycles 2007, 71,
589−596.
1
(
38−144.
39) Xiao, Z.; Larson, R. G.; Chen, Y.; Zhou, C.; Niu, Y.; Li, G.
Unusual phase separation and rheological behavior of poly(ethylene
oxide)/ionic liquid mixtures with specific interactions. Soft Matter
(57) Liu, Y.; Chipot, C.; Shao, X.; Cai, W. How does the solvent
modulate shuttling in a pillararene/imidazolium [2]rotaxane? Insights
from free energy calculations. J. Phys. Chem. C 2016, 120, 6287−6293.
(58) Kanao, M.; Otake, A.; Tsuchiya, K.; Ogino, K. Stereo-selective
synthesis of 5-norbornene-2-exo-carboxylic acidrapid isomerization
and kinetically selective hydrolysis. Int. J. Org. Chem. 2012, 2, 26−30.
(59) Ouellette, R. J.; Rosenblum, A.; Booth, G. Conformational
analysis. XI. 2-Carbomethoxy-7-oxabicyclo[2.2.1]heptane and 2-
carbomethoxy-7-oxabicyclo[2.2.1]hept-5-ene. J. Org. Chem. 1968,
33, 4302−4303.
2
(
016, 12, 7613−7623.
40) Roy, A.; Saha, S.; Datta, B.; Roy, M. N. Insertion behavior of
imidazolium and pyrrolidinium based ionic liquids into α and β-
cyclodextrins: mechanism and factors leading to host−guest inclusion
complexes. RSC Adv. 2016, 6, 100016−100027.
(
41) Roy, M. N.; Saha, S.; Choudhury, S.; Roy, A. Study of
complexation of an ionic liquid inside crown ethers in aqueous
environments by physicochemical techniques. Phys. Chem. Liq. 2017,
5
(
5, 659−668.
(60) Lee, M. W.; Herndon, W. C. J. Stereochemistry of the furan-
maleic anhydride cycloaddition. J. Org. Chem. 1978, 43, 518.
(61) Hillmyer, M. A.; Lepetit, C.; McGrath, D. V.; Novak, B. M.;
Grubbs, R. H. Aqueous ring-opening metathesis polymerization of
carboximide-functionalized 7-oxanorbornenes. Macromolecules 1992,
42) Boyd, T. J.; Schrock, R. R. Sulfonation and epoxidation of
substituted polynorbornenes and construction of light-emitting
devices. Macromolecules 1999, 32, 6608−6618.
(
43) Santiago, A. A.; Vargas, J.; Cruz-Gomez, J.; Tlenkopatchev, M.
2
(
5, 3345−3350.
62) Rickborn, B. Retro−Diels−Alder reaction. Part I. C−C
dienophiles. Org. Reactions 1998, 52.
63) Matson, J. B.; Grubbs, R. H. Synthesis of fluorine-18
A.; Gavino, R.; Lopez-Gonzalez, M.; Riande, E. Synthesis and ionic
transport of sulfonated ring-opened polynorbornene-based copoly-
mers. Polymer 2011, 52, 4208−4220.
(
(
44) Santiago, A. A.; Vargas, J.; Cruz-Morales, J. A.; Tlenkopatchev,
functionalized nanoparticles for use as in vivo molecular imaging
M. A.; Gavino, R.; Malkanduev, Y. A.; Sivov, N. A. Synthesis of new
polymer ionomers via ring-opening metathesis polymerization. Open J.
Org. Polym. Mater. 2014, 4, 84−91.
agents. J. Am. Chem. Soc. 2008, 130, 6731−6733.
(64) Ouchi, M.; Inoue, Y.; Wada, K.; Iketani, S.; Hakushi, T.; Weber,
E. Molecular design of crown ethers. 4. Syntheses and selective cation
binding of 16-crown-5 and 19-crown-6 lariats. J. Org. Chem. 1987, 52,
(
45) Suga, T.; Sakata, M.; Aoki, K.; Nishide, H. Synthesis of pendant
radical- and ion-containing block copolymers via ring-opening
metathesis polymerization for organic resistive memory. ACS Macro
Lett. 2014, 3, 703−707.
2
(
420−2427.
65) Ouchi, M.; Inoue, Y.; Liu, Y.; Nagamune, S.; Nakamura, S.;
Wada, K.; Hakushi, T. Convenient and efficient tosylation of
oligoethylene glycols and the related alcohols in tetrahydrofuran−
water in the presence of sodium hydroxide. Bull. Chem. Soc. Jpn. 1990,
(
46) Wiesenauer, E. F.; Nguyen, P. T.; Newell, B. S.; Bailey, T. S.;
Noble, R. D.; Gin, D. L. Imidazolium-containing, hydrophobic−
ionic−hydrophilic ABC triblock copolymers: synthesis, ordered
phase-separation, and supported membrane fabrication. Soft Matter
2
(
norbornene esters possessing Pd SCS pincer complexes or
diaminopyridines. Macromolecules 2003, 36, 2230−2234.
(
Comparative reactivity of exo- and endo-isomers in the Ru-initiated
ring-opening metathesis polymerization of doubly functionalized
norbornenes with both cyano and ester groups. Macromolecules
2
(
6
(
3, 1260−1262.
66) Roberts, J. D.; Trumbull, E. R., Jr.; Bennett, W.; Armstrong, R.
013, 9, 7923−7927.
47) Pollino, J. M.; Stubbs, L. P.; Weck, M. Living ROMP of exo-
The reaction of norbornylene with N-bromosuccinimide. nortricy-
II
clene and its derivatives. J. Am. Chem. Soc. 1950, 72, 3116−3124.
(67) Krapcho, A. P.; Dundulis, E. A. Stereochemistry of alkylation of
carboxylic acid salt and ester alpha. anions derived from cyclic
48) Nishihara, Y.; inoue, Y.; Nakayama, Y.; Shiono, T.; Takagi, K.
systems. J. Org. Chem. 1980, 45, 3236−45.
(
68) Zhang, L.; Sun, L.; Cui, Z.; Gottlieb, R. L.; Zhang, B. 5′-
Sulfhydryl-modified RNA: initiator synthesis, in vitro transcription,
and enzymatic incorporation. Bioconjugate Chem. 2001, 12, 939−948.
006, 39, 7458−7460.
49) Lapinte, V.; Brosse, J.-C.; Fontaine, L. Synthesis and ring-
opening metathesis polymerization (ROMP) reactivity of endo-and
exo-norbornenylazlactone using ruthenium catalysts. Macromol. Chem.
Phys. 2004, 205, 824−833.
(
50) Moatsou, D.; Hansell, C. F.; O’Reilly, R. K. Precision polymers:
a kinetic approach for functional poly(norbornenes). Chem. Sci. 2014,
, 2246−2250.
51) Ashirov, R. V.; Zemlyakov, D.; Lyapkov, A. A.; Kiselev, S.;
5
(
Vervacke, D. The relative reactivity of 2,3-dicarbomethoxy-5-
norbornenes in metathesis polymerization using the original N-
chelating ruthenium carbene complex. J. Appl. Polym. Sci. 2014, 131,
4
(
0130.
52) Wang, Y.; Fan, F.; Agapov, A. L.; Saito, T.; Yang, J.; Yu, X.;
Hong, K.; Mays, J.; Sokolov, A. P. Examination of the fundamental
relation between ionic transport and segmental relaxation in polymer
electrolytes. Polymer 2014, 55, 4067−4076.
̈
53) Reichardt, C.; Harbusch-Goernert, E. Uber pyridinium-n-
(
phenolat-betaine und ihre verwendung zur charakterisierung der
polaritat von losungsmitteln, x. erweiterung, korrektur und neudefi-
nition der e -losungsmittelpolaritatsskala mit hilfe eines lipophilen
̈
̈
̈
̈
t
penta-tert-butyl-substituierten pyridinium-n-phenolat-betainfarb-
stoffes. Liebigs Ann. Chem. 1983, 1983, 721−743.
(
54) Harrod, W. B.; Pienta, N. J. Solvent polarity scales. 1.
determination of E and Π* values for phosphonium and ammonium
T
melts. J. Phys. Org. Chem. 1990, 3, 534−544.
(
55) Machado, V. G.; Stock, R. I.; Reichardt, C. Pyridinium N-
phenolate betaine dyes. Chem. Rev. 2014, 114, 10429−10475.
R
DOI: 10.1021/acs.macromol.8b02295
Macromolecules XXXX, XXX, XXX−XXX