Tetrazolation of side chains and anhydrous conductivity in a hydrophobic polymer

Article abstract of DOI:10.1021/ma501068j

1H-Tetrazoles possess the lowest pK_a within the azole family of nitrogen-containing heterocycles, making them attractive amphoteric moieties for anhydrous proton conduction. The synthesis of a styrenic, random coil polymer with pendent C-substituted 1H-tetrazoles is described in detail. This facile route results in a polymer containing free proton-bearing tetrazoles, limits side reactions, and proceeds using low polarity, halogenated solvents (e.g., 1,2-dichlorobenzene or chlorobenzene), which improve solubility during tetrazole cyclization. The resulting polymer, when probed by NMR spectroscopy and elemental analysis, had no measurable salt content, ensuring accurate proton conductivity measurements. Utilizing interdigitated electrodes (IDEs) in conjunction with electrochemical impedance spectroscopy (EIS), undoped anhydrous proton conductivities were measured to be as high as 10^-5 S cm ^-1 at 120 °C. This weakly acidic, 1H-tetrazole-bearing polymer is thermally stable to 210 °C, possesses two distinct glass transitions (T_g) at 49 and 74 °C, and exhibits surprisingly low water uptake, despite its acidic and amphoteric nature. Reduction of T_gs, achieved by synthesis of low molecular weight poly(4-vinylphenol) via acid polymerization, shows a minimal dependence of anhydrous proton conductivity on backbone motion.

Full text of DOI:10.1021/ma501068j

Article
pubs.acs.org/Macromolecules
Tetrazolation of Side Chains and Anhydrous Conductivity in a
Hydrophobic Polymer
,†
†
†
†
†
Holly L. Ricks-Laskoski,* Brian L. Chaloux, Stephen M. Deese, Matthew Laskoski, Joel B. Miller,
†
‡
§
†
Mary A. Buckley, Jeffrey W. Baldwin, Michael A. Hickner, Kaitlin M. Saunders,
and Caroline M. Christensen§
†
Chemistry Division, Branch 6120, Naval Research Laboratory, Washington, D.C. 20375-5320, United States
Acoustics Division, Branch 7130, Naval Research Laboratory, Washington, D.C. 20375-5320, United States
Department of Materials Science and Engineering, The Pennsylvania State University, 310 Steidle Building, University Park,
‡
§
Pennsylvania 16802, United States
ABSTRACT: 1H-Tetrazoles possess the lowest pK within the azole family of
a
nitrogen-containing heterocycles, making them attractive amphoteric moieties for
anhydrous proton conduction. The synthesis of a styrenic, random coil polymer
with pendent C-substituted 1H-tetrazoles is described in detail. This facile route
results in a polymer containing free proton-bearing tetrazoles, limits side
reactions, and proceeds using low polarity, halogenated solvents (e.g., 1,2-
dichlorobenzene or chlorobenzene), which improve solubility during tetrazole
cyclization. The resulting polymer, when probed by NMR spectroscopy and
elemental analysis, had no measurable salt content, ensuring accurate proton
conductivity measurements. Utilizing interdigitated electrodes (IDEs) in
conjunction with electrochemical impedance spectroscopy (EIS), undoped
−5
−1
anhydrous proton conductivities were measured to be as high as 10 S cm at
1
2
20 °C. This weakly acidic, 1H-tetrazole-bearing polymer is thermally stable to
10 °C, possesses two distinct glass transitions (T ) at 49 and 74 °C, and exhibits
g
surprisingly low water uptake, despite its acidic and amphoteric nature. Reduction of T s, achieved by synthesis of low molecular
g
weight poly(4-vinylphenol) via acid polymerization, shows a minimal dependence of anhydrous proton conductivity on backbone
motion.
INTRODUCTION
Additionally, these reactions typically require transition metal
catalysts, organic azides, and purification by column chroma-
tography. Removal of spent catalyst using this type of
purification method is often impractical postpolymerization
due to poor polymer solubility in common solvents,
necessitating the development of an alternative synthetic
scheme for polymeric materials. Polymer tetrazolation which
foregoes transition metal catalysts has, in the past, relied upon
strongly dipolar solvents (e.g., DMF or water) and the in situ
generation of hydrazoic acid, giving rise to safety con-
■
(
Amphoteric molecules, acting simultaneously as an acid
proton donor) and base (proton acceptor), similar to water,
have been shown to aid in the anhydrous transport of
1
,2
protons.
Many recent examples of anhydrous proton
conductors rely upon moieties from the azole family of
nitrogen-containing heterocycles for charge transport.
most acidic member of the azole family, having a pK of 4.9,
similar to that of acetic acid, is 1H-tetrazole, an aromatic five-
membered ring containing four nitrogens and one carbon atom.
For the purposes of proton conduction, it is the least studied
azole to date due to the arduous synthesis typically involved.
However, 1H-tetrazoles are particularly attractive for proton
conducting applications due to their high potential degree of
coordination and acidity. A variety of synthetic strategies
have been developed to access functionalized azoles with intact
N−H protons, which enable structural proton transport.
the absence of a free proton, these materials require additional
proton sources (e.g., −SO H, H PO ) to generate mobile
protons.
3
−5
The
a
9
,7,11,14,17,18
cerns.
Herein, an alternative synthesis, purification,
and characterization of a styrenic random-coil polymer
containing 1H-tetrazole side chains 1 (Figure 1) will be
discussed along with its proton conductivity within an
anhydrous environment.
6,7
1
0,18,15,7,19,14
1
H-Tetrazole-bearing polymers are known.
8
−16
However, few have been studied in their capacity as materials
In
20−26
capable of anhydrous proton transport.
The majority of
these materials are derived from poly(acrylonitrile), where the
tetrazole is directly bound to the backbone and is highly
3
3
4
While click chemistry of nitriles and azides is an attractive
route to tetrazole formation, many methodologies produce N-
substituted species, resulting in loss of the single azole proton.
Received: May 22, 2014
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/ma501068j | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Synthesis of In-House Poly(4-vinylphenol) 7. To a 500 mL
Schlenk flask under nitrogen and fitted with a stir bar was added 4-
acetoxystyrene 5 (15.90 g, 98.03 mmol) and tetrahydrofuran (0.98 M).
After deoxygenation via the freeze−pump−thaw method (3 times),
the flask was backfilled with nitrogen, and a 5 M solution of aqueous
sodium hydroxide (100 mL, sparged with nitrogen) was syringed into
the reaction mixture. The reaction was stirred at room temperature for
1 h or until the reaction was complete as indicated by TLC (SiO : R =
0.7, 30% ethyl acetate/hexanes). The crude reaction mixture was
chilled to 0 °C, at which time 4 N HCl (90 mL) was added slowly
until a pH of neutral was reached. The product was extracted into
ethyl acetate, washed with distilled water three times, dried over
Figure 1. A polystyrenic random-coil polymer with pendent C-
substituted 1H-tetrazole side chains 1.
2
f
dependent on backbone flexibility for increased conductivity or
where the reaction of 5-amino-1H-tetrazole, having modified
acid−base properties, is reacted with suitable electrophilic
precursor polymers. The nature and length of the side chain
connecting 1H-tetrazole to the polymer backbone (e.g.,
shortest in PAN-derived materials) are likely to have a great
impact on material properties through both its electronic effects
and effects on the local environment of the tethered tetrazole.
Because of the complexities involved in synthesis and increasing
the spatial distance between the polymer backbone and
tetrazole moiety, the postpolymerization synthesis of polymers
bearing pendent C-alkyl-1H-tetrazoles has, thus far, not been
reported. Inspiration for our synthetic method can be attributed
to the work by Aureggi and Sedelmeier and their attention to
the safe and thorough quenching of unreacted azide, which is
MgSO , filtered, and concentrated by rotary evaporation. The resultant
4
oily product was precipitated into cold hexanes, decanted, and dried
under vacuum to yield 4-vinylphenol 6 (9.99 g, 85%) as a white
powder which was immediately polymerized as described in the
2
7
literature. A Schlenk flask under nitrogen fitted with a stir bar was
charged with 4-vinylphenol 6 (5.02 g, 41.8 mmol) and ethanol (4.2 M)
while a second Schlenk flask, also under nitrogen, was charged with
ethanol (15 mL) and 12 N HCl (17.5 mL). Both flasks were
deoxygenated via the freeze−pump−thaw method (3 times) and
backfilled with nitrogen, after which the HCl solution was syringed
into the reaction mixture containing the 4-vinylphenol 6 and stirred at
room temperature for 72 h. Although this reaction was not optimized,
it was confirmed by TLC (SiO : R = 0.7 for 4-vinylphenol, 30% ethyl
2
f
acetate/hexanes) that consumption of the monomer was complete.
The resultant polymer was precipitated into distilled water, filtered,
and rinsed with distilled water until a pH of neural was reached. The
crude materials were then redissolved in acetone, dried over MgSO4,
filtered, and concentrated by rotary evaporation. Further drying under
vacuum yielded poly(4-vinylphenol) 7 (3.80 g, 78%) as a light tan
16
overlooked in most synthetic works. Although their strategy
involved small molecules, the reaction of dialkylaluminum azide
with nitriles in nonpolar, aprotic organic solvents to achieve the
desired 1,3-dipolar cycloaddition has opened the door for
further investigation and utilization of this species in the
postpolymerization formation of covalently linked 1H-tetrazole
side chains.
powder which was used without further purification.
1
4
-Vinylphenol 6. H NMR (300 MHz, DMSO-d ): δ 9.53 (s, 1H),
6
3
3
3
7
1
.28 (d, 2H, J = 8.7 Hz), 6.75 (d, 2H, J = 8.4 Hz), 6.61 (q, 1H, J =
7.7 Hz), 5.57 (d, 1H, J = 17.7 Hz), 5.03 (d, 1H, J = 11.1 Hz).
3
3
13
C
NMR (75 MHz, DMSO-d ): δ 157.40, 136.45, 128.31, 127.46, 115.38,
EXPERIMENTAL SECTION
6
■
1
10.66.
Reagents. All reagents were purchased from the Aldrich Chemical
Co. and used without further purification unless otherwise noted.
Azide handling: precautions should always be taken when handling
azides to forbid exposure to metals, mitigate possible shock
sensitivities, and to avoid the hydrolysis product, hydrozoic acid.
Caution should be exercised when using halogenated solvents with
azides. Care was exercised to form the non-nucleophilic dialkylalumi-
num azide prior to introduction of chlorobenzene or 1,2-
1
Poly(4-vinylphenol) 7. H NMR (300 MHz, DMSO-d ): δ 9.02 (br,
6
13
1
H), 6.52 (br, 4H), 2.3−0.9 (m, 3H). C NMR (75 MHz, DMSO-
d ): δ 155.37, 135.86, 128.59, 115.14, 43.25, 40.77. Commercial
6
poly(4-vinylphenol) (Aldrich Chemical, M ∼ 11 000 Da) displayed a
w
monomodal molecular weight distribution and exhibited M of 9500
w
Da and M /M = 1.37; the HCl-polymerized poly(4-vinylphenol) was
w
n
found to have a Mw of 4000 Da and M /M = 1.33, also in a
w
n
monomodal distribution.
dichlorobenzene.
Characterization. All solution 1H, 13C, 23Na, and 27Al NMR
Synthesis of Polystyrenic Alkoxynitrile 8. To a 500 mL
Schlenk flask under nitrogen fitted with a stir bar was added either in-
house or commercially sourced poly(4-vinylphenol) 7 (9.96 g, 82.9
mmol), 6-bromohexanenitrile (21.25 g, 91.9 mmol), and anhydrous
DMF (0.4 M). This mixture was deoxygenated via the freeze−pump−
thaw method (3 times), backfilled with nitrogen, and heated to 65 °C.
Potassium carbonate (58.98 g, 355 mmol) was added to the reaction
vessel and allowed to stir for 16 h. After cooling to room temperature,
the reaction mixture was neutralized with the addition of 12 N HCl
and distilled water. The crude product was extracted into ethyl acetate
and washed with 1 N HCl, brine, and excess distilled water until
spectra were obtained using a Bruker AC-300 spectrometer using
DMSO-d , CDCl , or acetone-d as solvent. All differential scanning
6
3
6
calorimetric (DSC) analyses were performed on a TA Instruments
DSC Q100 modulated thermal analyzer at a heating rate of 10 °C
−
1
3
−1
min and a helium purge of 25 cm min . Thermal gravimetric
analysis (TGA) was performed on a TA Instruments SDT2960 TGA-
−1
DTA at a heating rate of 10 °C min and air or nitrogen purge of 100
3
−1
1
cm min . Magic angle spinning solid-state H NMR spectra were
obtained on an Agilent spectrometer operating at 500 MHz, with
sample spinning at 22 kHz. Second moments of proton resonances
were calculated by fitting the NMR data to 12 Lorentzian peaks and
summing the Lorentzians associated with each band. The reported
temperatures of the solid-state NMR data were adjusted upward by 12
reaching a pH of neutral. The organic layer was dried over MgSO4,
filtered, and concentrated by rotary evaporation. To remove any
remaining 6-bromohexanenitrile, the product was stirred over hexanes
and decanted. To remove color, carbon black was added to a round-
bottom flask containing the product dissolved in ethyl acetate, which
was then filtered via gravity filtration. The product was again
concentrated by rotary evaporation and further dried under vacuum
°
C from the recorded temperature to account for frictional heating of
the sample due to spinning. All elemental analysis were conducted by
Galbraith Laboratories, Knoxville, TN. A Waters gel permeation
chromatography (GPC) system was used to obtain molecular weights
of the poly(4-vinylphenol)s only, due to limited solubility of the
tetrazolated polymers. The GPC system consisted of Waters Breeze
software, a Waters 1515 isocratic HPLC pump, Styragel columns, and
a Waters 2414 refractive index detector. Tetrahydrofuran was used as
eluent at 35 °C, and the system was calibrated to polystyrene
standards.
to yield polymer 8 (15.85 g, 89%) as an off-white solid and used
1
without further purification. H NMR (300 MHz, CDCl
): δ 6.65 (br,
3
4H), 3.87 (br, 2H), 2.35 (br, 2H), 1.69 (br, 7H), 1.25 (br, 2H) and
(300 MHz, DMSO-d ) δ 6.65 (br, 4H), 3.86 (br, 2H), 2.47 (br, 2H),
6
13
1.56 (br, 7H), 1.39 (br, 2H). C NMR (75 MHz, acetone-d ): δ
6
157.71, 138.00, 129.08, 120.49, 114.62, 67.88, 41.29, 41.28, 29.15,
B
dx.doi.org/10.1021/ma501068j | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
2
7
5.88, 25.79, 17.00. Elem. Anal. Calcd: C, 78.10; H, 7.96; N, 6.51; O,
.43. Found: C, 77.54; H, 8.03; N, 6.16; O, 7.59.
Synthesis of Polystyrenic Alkoxy-1H-Tetrazole 1. Using
calibration to 0.05, 0.1, 0.5, and 1.0 mM KCl solutions using a typical
commercial conductivity probe as a reference. The measured cell
constant was close to the value predicted by IDE geometry, 0.106
cm , and was valid for film thicknesses greater than 40 μm. Thick
films (∼100 μm, measured by stylus profilometry) of polystyrenic
alkoxy-1H-tetrazole 1 were drop-cast onto the IDE array from a
concentrated DMSO solution and annealed under vacuum at 70 °C
prior to measurement. Film resistance was calculated by fitting
impedance data to an empirical circuit model.
−1
glovebox techniques, a 1000 mL Schlenk flask fitted with a stir bar
was charged with diethylaluminum chloride (25 wt % in toluene, 34.0
mL, 62.5 mmol). The sealed Schlenk flask was taken out of the
glovebox, attached to a Schlenk line under nitrogen, and chilled to 0
°
C. Sodium azide (4.06 g, 62.43 mmol) was then added to the reaction
flask, warmed to room temperature, and stirred for 16 h, which
16
produced a cloudy-white mixture, as previously described. A separate
00 mL Schlenk flask under nitrogen also fitted with a stir bar was
5
RESULTS AND DISCUSSION
Traditionally, the conversion of nitrile to 1H-tetrazole has been
accomplished by reaction either with hydrazoic acid or an
■
charged with polymer 8 at a ratio of 1:3 (ratio of polymer 8 to
Et AlN ) (4.46 g, 20.72 mmol) and anhydrous toluene and/or
2
3
chlorobenzene or 1,2-dichlorobenzene (0.14 M) added (previously
deoxygenated) and then backfilled with nitrogen. The diethylalumi-
num azide mixture was heated to 80 °C, and the solution containing
polymer 8 was warmed to 80 °C and then slowly added to the azide
mixture via cannula. The reaction flask was sealed and stirred at 80 °C
for 18 h. The reaction flask containing the crude reaction mixture was
chilled to 0 °C and opened to a Schlenk line under nitrogen while a
equivalent species, such as NH N , in dimethylformamide
4
3
(
DMF) or sodium azide in water, the latter catalyzed by zinc
9,7,11,14,17,18
salts.
These reagents are less reactive toward alkyl
nitriles than aryl and vinyl derivatives, limiting their general
utility. Additionally, water is an unsuitable solvent for
hydrophobic materials, including the nitrile precursor to
polystyrenic alkoxy-1H-tetrazole 1. The direct generation of
hydrazoic acid is also a safety concern, given its toxicity and
shock sensitivity. Dialkylaluminum azide, however, is a suitable
alternative reagent for the conversion of unactivated nitriles to
NaOH/NaNO solution (3.7 M each) was added until a pH of 11 was
2
reached. To this mixture was then added 6 M HCl until a pH of 3 was
reached. The product was obtained by filtering via gravity filtration and
washed with excess water until a pH of neutral was reached. The crude
product was dissolved in a minimal amount of DMSO, precipitated
into 1 N HCl, filtered, and dried under vacuum, and polymer 1,
1
H-tetrazoles due to its reactivity toward a wide variety of
16
isolated as a light tan powder (4.75 g, 89%), was stored in a desiccator.
nitriles and its solubility in nonpolar solvents. These solvents
and the dialkyl functionality provide intimate mixing of polymer
and azide during cyclization and facilitate azide quenching on
completion of the reaction.
1
H NMR (300 MHz, DMSO-d ): δ 16.20 (bs, tetrazole-H, 1H), 6.57
6
(
br, 4H), 3.81 (br, 2H), 2.85 (br, 2H), 1.69 (br, 7H), 1.39 (br, 2H).
1
3
C NMR (75 MHz, DMSO-d ): δ 156.92, 155.86, 137.51, 128.57,
6
1
14.26, 67.48, 40.90, 40.85, 28.85, 27.25, 25.50, 23.09. Elem. Anal.
Initial attempts at functionalization of a polymer substrate by
Calcd: C, 65.09; H, 7.02; N, 21.69; O, 6.19. Found: C, 64.80; H, 6.99;
1
1
H-tetrazole-containing moieties proved unfruitful. Per Scheme
, compound 3 was chosen as a simple, representative structure
N, 21.31; O, 6.69.
Measuring Proton Conductivity. Proton conductivity was
measured by electrical impedance spectroscopy (EIS) utilizing a
Gamry Reference 600 potentiostat with a 100 mV applied RMS
Scheme 1. Synthesis of the Bicyclic Structure 1,5-
Pentamethylene Tetrazole 4
(
1
sinusoidal) voltage from 1 MHz to 1 Hz. Temperature control, from
20 to 30 °C, was provided by an ESPEC SH-240 humidity and
temperature chamber under a flow of dry nitrogen (0% RH) and
verified through the addition of an Omega humidity and temperature
data logger OM-CP-RHTEMP101A. The cell consisted of inter-
digitated electrodes (IDE, Figure 2) which were fabricated in-house,
to determine the side reaction that inhibits the desired process.
Upon attempting the Williamson ether reaction of phenol with
bromohexane-1H-tetrazole 2, the unexpected bicyclic structure
1
,5-pentamethylene tetrazole 4, also known as the drug
pentylenetetrazol, was the only product isolated. Use of a
stoiciometric amount of base with formation of phenolate anion
prior to the addition of 2 was insufficient to avoid formation of
4
. Although diethylaluminum azide is a convenient, non-
nucleophilic source of azide, tolerant of a variety of functional
groups, the low pK of the resulting 1H-tetrazole causes its
a
selective deprotonation, leading to an intramolecular SN2
reaction of tetrazolate with the ε-carbon of compound 2.
Protection of 1H-tetrazole likewise leads to undesired side
products. Thus, to achieve polymers with 1H-tetrazoles
containing intact N−H protons, formation of the acidic
tetrazole was required to proceed after reaction of the
polymeric phenolic group with halo-alkyl nitrile, necessitating
a modified tetrazolation methodology.
Figure 2. Schematic for gold-on-chromium interdigitated electrodes,
fabricated on 3 in. × 1 in. glass microscope slides.
composed of a 100 nm Au layer and a 10 nm Cr adhesion layer
between the gold and borosilicate glass substrate. The IDE consisted
of 50 teeth of 10 μm in spacing and width and 3 mm in length. A
comprehensive review of both the physics of IDEs and the theoretical
and experimental calculations of cell constants can be found in the
Polystyrenic alkoxy-1H-tetrazole 1 serves as a first example of
both the synthesis and properties of polymers containing alkyl-
tethered 1H-tetrazole moieties. To afford a compact, time-
2
8,29
literature.
For the IDE cells used to measure the conductivity of
−1
polymer 1, cell constants of κ = 0.115 ± 0.003 cm were calculated by
C
dx.doi.org/10.1021/ma501068j | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 2. Synthesis of Polystyrenic Alkoxy-1H-tetrazole 1 via [2 + 3] Cycloaddition of Polystyrenic Alkoxynitrile 8 and
Diethylaluminum Azide
1
Figure 3. Solution H NMR spectra of parent polystyrene alkoxynitrile 8 and polystyrenic alkoxy-1H-tetrazole 1 in deuterated dimethyl sulfoxide
(
DMSO-d6).
1
Figure 4. Solid state H NMR spectrum of polystyrenic alkoxy-1H-tetrazole 1 at 100 °C, showing raw data and 12 component Lorentzian fits.
1
6
efficient synthesis of compound 1, a two-step strategy was
employed, first forming the ether linkage between poly(4-
vinylphenol) (Mw ≈ 11 000, Sigma-Aldrich) 7, and 6-
bromohexane nitrile (Scheme 2). Polystyrenic alkoxynitrile 8
resulted when K CO was used as base and DMF as solvent. To
obtain lower molecular weights, 4-vinylphenol 6 was synthe-
sized from 4-acetoxystyrene 5 and subsequently polymerized in
the presence of HCl to yield poly(4-vinylphenol) 7 of variable
molecular weight and narrow molecular weight distribution. It
should be noted that NMR spectra of commercial and in-house
synthesized poly(4-vinylphenol)s 7 are identical.
in prior work. Tetrazolation commenced on addition of
polystyrenic alkoxynitrile 8, dissolved in chlorobenzene, 1,2-
dichlorobenzene (DCB), and/or toluene, to the diethylalumi-
num azide solution, followed by heating and stirring of the
reaction mixture overnight. After chilling the solution,
diethylaluminum was cleaved from the tetrazole nitrogen by
2
3
addition of aqueous NaOH and NaNO . Subsequent acid-
2
ification formed the 1H-tetrazole and quenched remaining
azide by in situ formation and immediate reaction of HNO2
with HN (to N , NO, and H O). The entire process was
3
2
2
conducted under an inert atmosphere. Conversion to
polystyrenic alkyoxy-1H-tetrazole 1 was confirmed by the
growth of a broad 1H-tetrazole N−H peak at 16 ppm and
Diethylaluminum azide was generated in situ by the addition
of NaN to a solution of diethylaluminum chloride in toluene.
3
The formation and general use of this reagent are documented
shifting of the terminal CH protons from 2.4 to 2.9 ppm in
2
D
dx.doi.org/10.1021/ma501068j | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
solution 1H NMR (Figure 3). Although the 1H-tetrazole
proton is broad in solution NMR and often exhibits reduced
intensity due to exchange with residual water, the full integrated
1
intensity of this peak was observed in solid-state H NMR of
the polymer, illustrated at 100 °C in Figure 4. Because of
changes in tetrazole environment and interactions in the solid
state, the N−H proton shifts to ∼12 ppm at high temperature.
During the early stages of tetrazolation using toluene as a
solvent, polymer 8 precipitated from solution as swollen solid
particulate, most likely due to the disparate solubility of the
backbone and tetrazole-containing pendent moieties. Never-
theless, conversions in excess of 90% were achieved in the
swollen solid, showcasing the facile nature of this reaction. To
attain quantitative conversion, DCB was used as cosolvent to
maintain solubility of polymer product and facilitate complete
removal of residual azide and aluminum salts upon quenching.
It was noted that at least 2 equiv of dialkylaluminum azide were
required for complete conversion of nitrile to tetrazole. Early
experiments with small molecule alkoxy- and alkylnitrile
compounds determined that an additional equivalent was
required for each alkoxy group present in the structure. It is
suspected that the electrophilic aluminum stoichiometrically
coordinates to the oxygen in the alkoxy linkage, requiring a
second equivalent to completely react with the nitrile. Although
using alkoxy substituents requires doubling of the azide, this
procedure forgoes potentially explosive and acutely toxic
intermediates (e.g., HN , NH N ) and allows safe, in situ
Figure 5. Thermal gravimetric analysis of polystyrenic alkoxy-1H-
tetrazole 1 under nitrogen and air. Weight loss of 1% is observed by
2
10 °C and 5% by 250 °C.
water, which has a pK of 15.7 and pK of −1.7, the sum of
a
b
these values gives rise to the autodissociation constant, pK .
d
This results in a proton concentration of only 10− M,
7
approximately the square root of K . In 1H-tetrazoles, where
d
pK ≈ 22 based on aqueous measures of pK = 4.9 and pK =
d
a
b
17, autodissociation alone should result in low charge
concentrations and conductivities even where charge mobility
6,32
is high.
To measure low conductivities by electrochemical
impedance spectroscopy (EIS), blocking interdigitated electro-
des (IDEs) were fabricated from gold. Compared to standard
parallel plate and plane parallel cell configurations, an
interdigitated design requires only small sample volumes (ca.
10 μL) and can exhibit improved signal-to-noise at low
3
4
3
quenching of excess azide. This methodology has been
demonstrated as a safe and useful reaction pathway to obtain
1H-tetrazoles with labile protons that are covalently tethered by
−1
an alkyl substituent to a polymeric backbone.
conductivity. The IDE cell constant, κ (cm ), is determined in
the thick film limit by the effective spacing between electrode
teeth and the total electrode area. Equation 1 relates
Cycloaddition of pendant nitriles with diethylaluminum azide
permitted the synthesis of polystyrenic alkoxy-1H-tetrazole 1.
Although conversions of nitrile to 1H-tetrazole between 90%
and 100% were observed depending upon reaction conditions,
use of 1,2-dichlorobenzene as cosolvent enabled consistent,
quantitative conversion. The various properties of polystyrenic
alkxoy-1H-tetrazole 1 reported herein were conducted only on
samples of 100% conversion. This class of polymers bearing
alkyl-tethered 1H-tetrazoles is of interest for its potential
anhydrous proton conducting capability. Proton conduction in
tethered systems can only be structural in nature (per the
Grotthuss mechanism) due to the lack of long-range transla-
tional diffusion in a solid polymer and should thus be strongly
−1
conductivity (σ, S cm ) to κ and solution/film resistance
(Rsoln, Ω).
σ = Rκ
soln
(1)
A comprehensive review of both the physics of IDEs and the
theoretical and experimental calculations of cell constants can
29,28
be found in the literature.
For the cells used to measure the
conductivity of polymer 1, cell constants of κ = 0.115 ± 0.003
−1
cm were calculated by calibration to 0.05, 0.1, 0.5, and 1.0
mM KCl solutions using a typical commercial conductivity
probe as a reference. The measured cell constant was close to
3
0,31
coupled to polymer motion.
convert nitriles to 1H-tetrazoles generate high levels of salts
e.g., aluminum and sodium salts), care must be taken to
As the synthetic methods to
−1
the value predicted by IDE geometry, 0.106 cm , and was valid
for thicknesses greater than 40 μm. Thick films (∼100 μm,
measured by stylus profilometry) of polystyrenic alkoxy-1H-
tetrazole 1 were drop-cast onto the IDE array from
concentrated DMSO solution and annealed under vacuum at
70 °C prior to EIS. Electrochemical impedance spectroscopy
was conducted on a Gamry Reference 600 potentiostat from 1
MHz to 1 Hz and 120 to 30 °C under flowing nitrogen, using a
100 mV applied RMS (sinusoidal) voltage. A representive
(
rigorously exclude them from the final product. To ensure that
measured conductivity could arise only from dissociated azoles,
sodium and aluminum salt concentrations were monitored by
2
3
27
Na and Al NMR in deuterated dimethyl sulfoxide (DMSO-
d6). Samples were reprecipitated from DMSO into 1 N HCl
and then repeatedly rinsed with distilled water until neutral to
reduce salt concentrations below detection limits. Despite the
explosive nature of many small molecule tetrazoles, polystyrenic
alkoxy-1H-tetrazole 1 was thermally stable under both nitrogen
and air to 210 °C. The 14% weight loss observed from 210 to
̂
Nyquist plot (−Z″ vs Z′, where total impedance Z = Z′ + iZ″)
and the empirical circuit model used to fit the raw data are
illustrated in Figure 6. The four-element, empirical circuit
model depicted in Figure 6 was originally proposed by Karan et
al. for modeling conductivity in thin, hydrated Nafion films with
2
80 °C by TGA (Figure 5) was controlled under both nitrogen
and air and presumably corresponds to off-gassing of N and
HN by retrocyclization of 1H-tetrazole.
2
7
33
IDE cell geometry. It includes constant phase elements
3
For pure, undoped 1H-azole materials, autodissociation of
azoles to form charged species is expected to be quite low based
on acid and base dissociation constants (pK and pK ). For
(CPEs) approximating geometric capacitance and electrode
polarization (double-layer capacitance) effects, which are
known to be good empirical descriptors, over a finite frequency
a
b
E
dx.doi.org/10.1021/ma501068j | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 6. Left: Nyquist plot of imaginary (−Z″) vs real (Z′) impedance of a film of polymer 1, used to calculate resistance by fitting to a circuit
model (red line). Full spectrum from 1 MHz to 1 Hz inset. Right: empirical circuit model used to fit impedance data to extract solution resistance
and film conductivity.
range, of the nonideal behavior encountered in polymer ionic
conductors.
The anhydrous conductivity of polystyrenic alkoxy-1H-
tetrazole 1 between 30 and 120 °C is illustrated in Figure 7,
Figure 8. Differential scanning calorimetry thermogram of polystyr-
enic 1H-tetrazole 1, illustrating two distinct T s at 49 and 74 °C.
g
mechanical and microscopic methods can be found else-
35
where.
Traditionally, VFT dependence of conductivity on temper-
ature suggests coupling of conduction to polymer motion.
Figure 7. Anhydrous conductivity of polystyrenic alkoxy-1H-tetrazole
between 30 and 120 °C, averaged over three separate films, with
standard deviations (log-weighted). Line is VFT fit to conductivity (T0
−33 °C).
However, T for polymer 1 was observed to be significantly
0
1
lower (82 °C) than the T of its softer phase (49 °C), raising
g
doubts that conductivity is coupled to polymer segmental
relaxation. Temperature-dependent, solid-state NMR spectros-
copy, illustrated in Figure 9, was used to examine 1H-tetrazole
proton motion, which was found to occur with a similar
activation energy, and in a similar temperature range, to side
chain motion. The onset of side chain mobility, observed as an
increase in the slope of log line width vs inverse temperature,
occurs at ∼47 °C. Noting that NMR transitions typically occur
at higher temperatures than DSC transitions due to differences
in the time scales of the processes being probed, side chain
motion onsets prior to segmental relaxation of the soft phase.
Segmental motion at a similar time scale to side chain motion
would additionally be visible as a change in the line width of the
aryl protons of 1, which remain relatively broad, and thus
immobile, over the range of temperatures investigated. The
coupling of proton motion to side chain dynamics, which
appear to be independent of backbone dynamics, may explain
the VFT behavior of conductivity observed above and below
both Tgs.
=
averaged over three batches. The large, non-Arrhenius
dependence of conductivity on temperature fits well to a
Vogel−Fulcher−Tammann (VFT) empirical relationship,
where T = −33 °C. At 120 °C, polymer 1 exhibits conductivity
0
−5
−1
on the order of 10 S cm , which is reasonable for an
undoped, azole-containing polymer well above its T . As frames
g
−
8
of reference, deionized water exhibits a conductivity of 6 × 10
−
1
−2
S cm at 25 °C and phosphoric acid varies between 5 × 10
S
−
1
−1
−1
34
cm at 25 °C and 4 × 10 S cm at 120 °C. Differential
scanning calorimetry of compound 1 (Figure 8) shows two
distinct T s at 49 and 74 °C, suggesting unexpected, but
g
unambiguous, two-phase behavior. This polymer was exten-
sively purified and verified by NMR and elemental analysis to
be free of residual solvents; thus, this behavior cannot be
explained by molecular impurities and must arise from the
polymer structure. The precise nature of this phase separation
is currently unknown. However, as the subject of ongoing
investigation, results confirming the existence of two phases by
If proton mobility is truly decoupled from segmental motion,
changes in polymer Tg should not significantly affect
F
dx.doi.org/10.1021/ma501068j | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
understood and thus the subject of ongoing work. However,
a preliminary study of the polymer physics of this material,
including mechanical and dielectric relaxation behavior, has
35
recently been published.
CONCLUSIONS
The Brønsted acidity of 1H-tetrazoles (pK = 4.9 for
■
a
unsubstituted 1H-tetrazole) causes them to be incompatible
with many reactions that might be leveraged to attach a
tetrazole-containing small molecule to a polymer backbone.
The Williamson ether reaction is one such polymer
functionalization technique, where poly(4-vinylphenol) is a
convenient, commercially available starting material. However,
as illustrated in Scheme 1, under alkaline conditions 1H-
tetrazole readily deprotonates, forming a nucleophile that
competes with phenolate to form 1,5-pentamethylenetetrazole
exclusively. Thus, it is synthetically prudent to perform the
tetrazole [2 + 3] cycloaddition as the final step in a synthetic
procedure. We have demonstrated the facile synthesis of
polymeric C-substituted 1H-tetrazoles through the [2 + 3]
cycloaddition of diethylaluminum azide, a non-nucleophilic
azide donor, and alkyl nitriles. This protocol offers advantages
over earlier postpolymerization procedures and employs
moderate temperatures, reasonable reaction times, and safe
reagents. Because of the facile nature of this cycloaddition
reaction, conversion of unactivated nitriles can even be
achieved on sparingly soluble or swollen, insoluble polymers.
With the correct choice of cosolvent, quantitative conversions
of nitrile to 1H-tetrazole are consistently achievable.
1
Figure 9. Average solid-state H NMR line widths of alkyl, aryl, and
azole protons in polystyrenic alkoxy-1H-tetrazole 1 vs inverse
temperature. Inset depicts representative, full polymer spectrum at
9
2 °C, from which peaks are fit.
conductivity. Aside from modifying the backbone structure, Tg
can be decreased by either plasticization or reduction of
polymer molecular weight. Obtaining a poly(4-vinylphenol)
with a lower molecular weight than commercially available was
synthetically achievable in only a few additional steps (see
Scheme 2). Acid polymerization of 4-vinylphenol with
concentrated aqueous HCl was attractive, producing a low
molecular weight with a narrow molecular weight distribution
and no visible side products. The HCl-polymerized poly(4-
vinylphenol) was found to have a M of 4000 Da and M /M
of 1.33, comparing favorably to the commercial poly(4-
vinylphenol) (Aldrich Chemical, M ≈ 11 000 Da) used for
prior syntheses, which was determined to have a M of 9500 Da
w
w
n
w
The model polystyrenic alkoxy-1H-tetrazole 1, synthesized
using the reported procedure, exhibits intrinsic anhydrous
w
with a M /M of 1.37. The lower molecular weight polystyrenic
alkoxy-1H-tetrazole derived from this polymer exhibits two
distinct T s, shifted to 26 and 44 °C, compared to 49 and 74 °C
−5
−1
w
n
proton conductivity of up to 10 S cm , despite its relatively
high T s of 74 and 49 °C. The duals T s of this material are
g
g
g
unlike other polymers containing pendent 1H-azoles reported
in the literature, which characteristically exhibit only one glass
transition. This behavior is especially unusual given this
material’s short side chains and homopolymer nature. Addi-
tionally, despite the apparently VFT behavior of conductivity,
common to most azole-functionalized polymers, side chain
motion and conductivity appear to be decoupled from
segmental motion, according to solid-state NMR spectroscopic
measurements. This is supported by observation of only a
factor of 2 increase in anhydrous proton conductivity on
in the higher molecular weight material. Although conductivity
of the low molecular weight polystyrenic alkoxy-1H-tetrazole
was found to increase uniformly by a factor of 2 compared to
the higher molecular weight polymer, illustrated in Figure 10,
this effect is much lower than might be expected for a decrease
in T of approximately 25 °C. Synthetically, the glass transition
can be easily varied in this material. The physics governing both
phase separation and proton motion are still not well
g
decreasing the T s of both phases by ∼25 °C. As this behavior
g
has not been reported in other 1H-azole-containing homopol-
ymers, it is currently under further investigation. As part of this
ongoing work, mechanical spectroscopy, dielectric relaxation
spectroscopy, and atomic force microscopy are published
elsewhere which provide preliminary insight into the physical
35
characteristics of this material.
AUTHOR INFORMATION
■
Corresponding Author
Tel 202-404-7352; e-mail holly.ricks-laskoski@nrl.navy.mil
H.L.R.-L.).
*
(
Notes
The authors declare no competing financial interest.
Figure 10. Anhydrous proton conductivity of low and high molecular
weight polystyrenic alkoxy-1H-tetrazole 1, derived from in-house
polymerized (low M ) and commercial (high M ) poly(4-vinyl-
ACKNOWLEDGMENTS
■
The authors thank the Office of Naval Research and the Naval
Research Laboratory for financial support of this project.
Additionally, B.L.C., S.M.D., and M.A.B. acknowledge NRL’s
w
w
phenol). Color-coded vertical lines illustrate T s of the respective
g
materials.
G
dx.doi.org/10.1021/ma501068j | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Student Temporary Employment and Pathways Programs
while K.M.S. acknowledges ONR’s Science & Engineering
Apprentice Program.
(27) Gupta, B. B. G. Process for the preparation of poly(vinylphenol)
from poly(acetoxystyrene) by acid catalyzed transesterification. US
4
,989,916, February 6, 1990.
(28) Sheppard, N. F. J.; Tucker, R. C.; Wu, C. Anal. Chem. 1993, 65,
1
199−1202.
ABBREVIATIONS
■
(29) Olthuis, W.; Streekstra, W.; Bergveld, P. Sens. Actuators 1995,
2
(
(
(
4−25, 252−256.
CPE, constant phase element; DCB, 1,2-dichlorobenzene;
DSC, differential scanning calorimetry; DMSO, dimethyl
sulfoxide; EIS, electrochemical impedance spectroscopy; IDE,
interdigitated electrode; NMR, nuclear magnetic resonance
30) Cukierman, S. Biochim. Biophys. Acta 2006, 1757 (8), 876−885.
31) de Grotthuss, C. J. T. Ann. Chim. 1806, 58, 54−78.
32) Hauptmann, S.; Eicher, T.; Speicher, A. In The Chemistry of
Heterocycles; Wiley-VCH: Weinheim, 2008; pp 212−217.
(
(
spectroscopy); PS-Tet, polystyrenic alkoxy-1H-tetrazole; Tg,
33) Paul, D. K.; Fraser, A.; Karan, K. Electrochem. Commun. 2011,
glass transition; TGA, thermal gravimetric analysis; TLC, thin
1
(
3, 774−777.
layer chromatography; VFT, Vogel−Fulcher−Tammann.
34) Chin, D.-T.; Chang, H. H. J. Appl. Electrochem. 1989, 19, 95−99.
35) Casalini, R.; Chaloux, B. L.; Roland, C. M.; Ricks-Laskoski, H. L.
(
J. Phys. Chem. C 2014, 118, 6661−6667.
REFERENCES
■
(
1) Kawada, A.; McGhie, A. R.; Labes, M. M. J. Chem. Phys. 1970, 52
6), 3121−3125.
(
(
2) Celik, S. U.; Bozkurt, A.; Hosseini, S. S. Prog. Polym. Sci. 2012, 37,
265−1291.
3) Granados-Focil, S.; Woudenberg, R. C.; Yavuzcetin, O.;
Tuominen, M. T.; Coughlin, E. B. Macromolecules 2007, 40, 8708−
713.
4) Scharfenberger, G.; Meyer, W. H.; Wegner, G.; Schuster, M.;
Kreuer, K.-D.; Maier, J. Fuel Cells 2006, 3−4, 237−250.
5) Martwiset, S.; Woudenberg, R. C.; Granados-Focil, S.;
Yavuzcetin, O.; Tuominen, M. T.; Coughlin, E. B. Solid State Ionics
007, 178, 1398−1403.
6) Trifonov, R. E.; Ostrovskii, V. A. Russ. J. Org. Chem. 2006, 42
11), 1585−1605.
7) Tsarevski, N. V.; Bernaerts, K. V.; Dufour, B.; Du Prez, F. E.;
Matyjaszewski, K. Macromolecules 2004, 37, 9308−9313.
8) Loren, J. C.; Krasinski, A.; Fokin, V. V.; Sharpless, K. B. Synlett
005, 18, 2847−2850.
9) Demko, Z. P.; Sharpless, K. B. J. Org. Chem. 2001, 66, 7945−
950.
10) Taden, A.; Tait, A. H.; Kraft, A. J. Polym. Sci., Part A 2002, 40,
333−4343.
11) Gaponik, P. N.; Ivashkevich, O. A.; Karavai, V. P.; Lesnikovich,
1
(
8
(
(
2
(
(
(
(
2
(
7
(
4
(
A. I.; Chernavina, N. I.; Sukhanov, G. T.; Gareev, G. A. Angew.
Makromol. Chem. 1994, 219, 77−88.
(
12) Celik, S. U.; Cosgun, S.; Akbey, U.; Bozkurt, A. Ionics 2012, 18,
1
(
2
(
2
(
1
(
01−107.
13) Jin, T.; Kitahara, F.; Kamijo, S.; Yamamoto, Y. Tetrahedron Lett.
008, 49, 2824−2827.
14) Huang, M.-R.; Li, X.-G.; Li, S.-X.; Zhang, W. React. Funct. Polym.
004, 59, 53−61.
15) Annenkov, V. V.; Kruglova, V. J. Polym. Sci., Part A 1993, 31,
903−1906.
16) Aureggi, V.; Sedelmeier, G. Angew. Chem., Int. Ed. 2007, 46,
440−8444.
17) Du, N.; Park, H. B.; Robertson, G. P.; Dal-Cin, M. M.; Visser,
T.; Scoles, L.; Guiver, M. D. Nat. Mater. 2011, 10, 372−375.
18) Du, N.; Robertson, G. P.; Dal-Cin, M. M.; Scoles, L.; Guiver, M.
D. Polymer 2012, 53, 4367−4372.
19) Kizhnyaev, V. N.; Vereshchagin, L. I. Rus. Chem. Rev. 2003, 72
2), 143−164.
20) Pu, H.; Wu, J.; Wan, D.; Chang, Z. J. Membr. Sci. 2008, 322,
8
(
(
(
(
(
3
(
(
92−399.
21) Pu, H.-T.; Ye, S. React. Funct. Polym. 2006, 66, 856−862.
22) Ye, S.; Pu, H.; Wan, D. Polym. Eng. Sci. 2012, 52 (7), 1450−
1
456.
(
3
(
2
23) Pu, H.; Luo, H.; Wan, D. J. Polym. Sci., Part A 2013, 51, 3486−
493.
24) Sinirlioglu, D.; Muftuoglu, A. E.; Bozkurt, A. J. Polym. Res. 2013,
0, 242.
(
25) Celik, S. U.; Bozkurt, A. Solid State Ionics 2011, 199−200, 1−5.
(
26) Ozden, S.; Celik, S. U.; Bozkurt, A. Electrochim. Acta 2010, 55,
8
498−8503.
H
dx.doi.org/10.1021/ma501068j | Macromolecules XXXX, XXX, XXX−XXX