Anhydrous proton conductivities of squaric acid derivatives

Article abstract of DOI:10.1039/c2cc31283b

In this communication, we introduce squaric acid derivatives as anhydrous proton conductors. We report the synthesis, characterization and proton conductivities of four squaric acid derivatives. The anhydrous proton conductivity of one of the derivatives

Full text of DOI:10.1039/c2cc31283b

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 5922–5924
www.rsc.org/chemcomm
COMMUNICATION
Anhydrous proton conductivities of squaric acid derivativesw
Dipankar Basak,a Craig Versek,b Daniel T. Toscano,a Scott Christensen,c
Mark T. Tuominenb and D. Venkataraman*a
Received 20th February 2012, Accepted 17th April 2012
DOI: 10.1039/c2cc31283b
In this communication, we introduce squaric acid derivatives as
anhydrous proton conductors. We report the synthesis, characteriza-
tion and proton conductivities of four squaric acid derivatives.
The anhydrous proton conductivity of one of the derivatives was
2.3 Â 10À3 S cmÀ1 at 110 8C, comparable to the conductivity of
molten 1H-1,2,3-triazole or 1H-imidazole.
There is a growing interest in developing new materials that can
be used as proton exchange membranes (PEM) in fuel cells.1,2
The target proton conductivity for PEMs is B1 Â 10À1 S cmÀ1 at
120 1C and 50% humidity.3 At the present time, membranes
derived from perfluorosulfonic acid-based Nafions or poly-
benzimidazole(PBI)/H3PO4 have high proton conductivities and
are used in commercial fuel cells.4,5 In these systems, the proton
transport is believed to occur through the diffusive mechanism
wherein the proton is transported through the carrier molecules
(water or H3PO4). Since the presence of such carriers is essential
for high proton conductivity, the fuel cell operation conditions
such as temperature are constrained by the physical properties of
the carriers.6 Therefore, there is widespread interest in developing
proton-transporting materials that have high intrinsic proton
conductivity wherein the protons are transported through
non-diffusive pathways (Grotthuss-type mechanism).
Fig. 1 1a. Crystal structure of squaric acid showing 2-D hydrogen
bonded network (CSD code: KECYBUO3); 1b. Illustration of the
re-organization of p bonds in squaric acid and the intermolecular
proton transfer between the tautomers.
Although squaric acid has been studied for its antiferroelectric
properties, its anhydrous proton transporting properties remain
unexplored.12–14 In this communication, we report the synthesis,
characterization, and proton conductivity under anhydrous
conditions for squaric acid derivatives.
Four derivatives (compounds 4a–d, Scheme 1) of squaric
acid were synthesized using synthetic protocols reported in the
literature.15–17 Addition of appropriate organolithium reagent to
commercially available 3,4-diisopropoxycyclobut-3-ene-1,2-dione
(1) furnished 2, which was then converted to cyclobutenedione 3
in the presence of a catalytic amount of acid. Deprotection of
isopropoxy group with 1 : 1 Conc.HCl/acetone (v/v) produced
compounds 4a–d in excellent yields.
Any molecule that exists in multiple tautomeric forms,
shows an extended hydrogen bonded network and has acidic
protons is a potential candidate for Grotthuss-type proton
transport. Thus far, the focus has been on nitrogen-based
heterocycles such as 1H-1,2,3-triazole and 1H-imidazole.7–10
Squaric acid (3,4-dihydroxycyclobut-3-ene-1,2-dione) exhibits
an extended hydrogen bonded network similar to 1H-imidazole or
1H-1,2,3-triazole in its crystalline state (Fig. 1a) and exists in
multiple tautomeric forms (Fig. 1b).11 Furthermore, like water, the
network consists of proton donor and acceptor sites. The
amphoteric nature of squaric acid network is an attractive and
important feature for translocation of protons over a long distance.
Anhydrous proton conductivities of squaric acid and its
derivatives (4a–d) were obtained by impedance spectroscopy.
Vacuum dried samples were placed into a cavity (length:
0.1 cm, cross-sectional area: 0.079 cmÀ2) between two custom
brass electrodes, forming liquid seals at the ends of a PTFE
tube. This set up was placed in a vacuum oven fitted with
a variable temperature controller. Samples were heated to
120 1C, the highest temperature investigated for this study,
a Department of Chemistry, University of Massachusetts Amherst,
710 N. Pleasant Street, Amherst, MA 01003, USA.
E-mail: dv@chem.umass.edu; Fax: +14135454490;
Tel: +14135452028
b Department of Physics, University of Massachusetts Amherst,
MA 01003, USA
c Department of Polymer Science and Engineering,
University of Massachusetts Amherst, MA 01003, USA
w Electronic supplementary information (ESI) available: general
information, TGA, proton conductivity vs. time and cyclic voltametry
data. See DOI: 10.1039/c2cc31283b
Scheme 1 Synthesis of of squaric acid derivatives.
c
5922 Chem. Commun., 2012, 48, 5922–5924
This journal is The Royal Society of Chemistry 2012

View Online
held for 12 h and subsequently cooled back to 40 1C
(see ESIw). Conductivity vs. time was plotted to ensure
integrity of the samples under applied experimental condition
(Fig. S3, ESIw). The proton conductivity of squaric acid was
found to be low (o10À8 S cmÀ1 at 120 1C) in our measurement,
comparable to the conductivities of 1H-imidazole and
4,4-1H-1H-bi-1,2,3-triazole (bitriazole) in their crystalline state
(melting point of squaric acid >300 1C).18,19 This poor
performance is attributed to a hindered molecular reorientation
process, which is considered to be the rate-limiting step for
anhydrous proton conduction.2 Similarly, the phenyl derivative
4d (melting point 207–208 1C) did not show any measurable
conductivity while compound 4a (melting point 160 1C) exhibited
low conductivity (2.0 Â 10À8 S cmÀ1 at 110 1C). Conversely,
the conductivity of 4b and 4c were measured as 2.2 Â 10À5 and
2.3 Â 10À3 S cmÀ1 at 110 1C, respectively (see Fig. 2). It is
noteworthy that proton conductivity of 4c is comparable to the
conductivity of molten 1H-imidazole at 90 1C (B10À3 S cmÀ1) and
higher than that of liquid 1H-1,2,3-triazole (1.3 Â 10À4 S cmÀ1) at
room temperature.8,18 From the Arrhenius plot, the activation
energy barrier for proton conduction was estimated as 40 kJ molÀ1
for 4c and 62 kJ molÀ1 for 4b by using the limiting slope at
high temperature.
Thermogravimetric (TGA) analysis of 4a–d under a nitrogen
atmosphere showed minimal weight loss for 4a until 310 1C, 4b
and 4d until 200 1C, and 4c until 150 1C (Fig. S1, ESIw); samples
were kept at 60 1C in vacuo for 12 h before performing TGA to
ensure removal of any adventitious water or volatile impurities.
As a result of the measured proton conductivities, we
decided to focus our attention on molecules 4b and 4c and
characterize them using differential scanning calorimetry
(DSC), X-ray scattering and cyclic voltammetry. In DSC,
upon first heating, 4b shows a broad endotherm corresponding
Fig. 3 Wide-angle X-ray scattering of compounds 4b (A) and 4c (B);
the samples were first measured at 25 1C, then heated to 130 1C and
cooled to 50 1C. I(q) values have been shifted vertically for clarity. The
X-ray scattering indicates the presence of a dimer for 4b (inset A) and a
tetramer for 4c (inset B).
performed at several different temperatures. Prior to any
thermal treatment, pristine 4c at 25 1C exhibits multiple peaks,
the first of which falls at q = 3.86 nmÀ1, corresponding to a
length scale (d) of 16.3 A (Fig. 3B; d is related to q by the
equation d = 2p/q). Although the structure cannot be determined
unambiguously from this data, Cerius 2 molecular modelling
suggests that 4c exists as a chain of hydrogen-bonded tetramers
with hydrogen bonding patterns similar to squaric acid (Fig. 3B
inset). The centre-to-centre distance between two tetrameric units
is B16.3 A, correlating very well with the observed distance in
WAXS. Furthermore, the presence of numerous sharp peaks
indicates that the material is indeed crystalline in nature,
consistent with the large endotherm observed in DSC for the
first heating cycle. Upon heating beyond the phase transition
temperature to 130 1C, the majority of crystalline peaks
disappear except for four distinct peaks. The position of the
primary peak varies only slightly from the crystal phase (q =
3.99 nmÀ1 vs. q = 3.86 nmÀ1), however it is significantly
broadened, indicating a decreased average feature size of
4.3 nm (full-width at half maximum, Dq = 1.46 nmÀ1). A broad
peak at q = 12.5 nmÀ1 corresponds well with the numerous
crystalline peaks evident at 25 1C, however the lack of sharp
reflections again indicates decreased order in this phase when
compared to the crystal state. Two additional small peaks at
q = 6.62 nmÀ1 and q = 7.17 nmÀ1 also emerge, and lie reasonably
close to peaks in the original crystal phase (q = 6.14 nmÀ1 and
q = 6.83 nmÀ1). The combination of these reflections indicates
that, upon heating, 4c forms a poorly ordered phase with a
similar unit cell to the initial crystal structure. Upon cooling to
to a phase transition at 101.7 1C (DHtransition = 24.07 J gÀ1
)
(Fig. S2 A, ESIw). A similar transition was observed for 4c at
92.4 1C (DHtransition = 86.32 J gÀ1) (Fig. S2 B, ESIw).
Exotherms were not observed upon cooling, indicating that
the initial crystalline phases are not recovered. Further, the
second heating cycle reveals two weak endotherms at 35.98 1C
and 115.87 1C for 4b, but no such peak was observed for 4c.
In order to investigate the morphology of these materials,
wide angle X-ray scattering (WAXS) of 4b and 4c was
Fig. 2 Comparison of temperature dependent proton conductivities
of 4a, 4b and 4c. Molecule 4c shows higher conductivity compared to
4a and 4b. Molecule 4d did not show any measurable conductivity and
therefore is not shown in the figure.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 5922–5924 5923

View Online
50 1C, the scattering pattern remains unchanged, indicating an
irreversible phase transition at 92 1C, correlating well with
DSC observation.
oxidation of 1H-imidazole under identical experimental
conditions. We believe that squaric acid derivatives are very
promising candidates as new anhydrous proton conducting
materials. Further studies regarding blends of squaric acid
derivatives into polymer matrix and immobilization of squaric
acid onto a polymer backbone by flexible linkers are currently
under investigation.
WAXS of 4b prior to any thermal treatment (25 1C) exhibits
fewer scattering peaks than 4c in its pristine state, indicating a
lower initial crystallinity, in agreement with DSC (Fig. 3A).
The primary peak falls at q = 4.00 nmÀ1, corresponding to a
length scale of 15.7 A. We note that the end-to-end distance
from the terminus of n-butyl chain to the CQO of squaric acid
is B8 A (obtained from the energy minimized structure of 4b
using Chem3D MM2 level molecular modelling), suggesting a
dimeric form (Fig. 3A inset). Heating the sample to 130 1C
causes a slight shift and broadening of the primary peak to
q = 3.53 nmÀ1 with an average feature size of 2.57 nm
(full-width at half maximum, Dq = 2.44 nmÀ1). Furthermore,
a broad halo region at q = 15.0 nmÀ1 was observed similar to
4c at elevated temperature, again indicating the presence of a
new phase with a similar unit cell as the initial crystal.
Based on the combination of DSC and WAXS data we
conclude that upon heating, 4b and 4c both undergo an
irreversible phase transition from a crystalline state to a poorly
ordered phase with a similar unit cell to the original crystal.
The presence of additional peaks with smaller full-width at
half maximum in 4c indicates that this phase is slightly more
ordered than that in 4b. The increased ordering of the new
phase in 4c likely indicates a higher concentration of squaric
acid connected through a hydrogen-bonded network, which in
turn facilitates anhydrous proton hopping. This observation
is in agreement with a previous report wherein continuous
nano-scale domains of proton transferring groups via self-
organization yield high conductivity.20
This work was initiated through a support by Fueling the
Future Center for Chemical Innovation (CHE 0739227),
sponsored by National Science Foundation. This material is
also based upon work supported in part by the U. S. Army
Research Laboratory and the U. S. Army Research Office
under grant number 54635CH, and by the National Science
Foundation Materials Research and Science Center on
Polymers (DMR 0820506). We sincerely thank Prof. Ryan C.
Hayward for his valuable input in interpreting the X-ray data.
Notes and references
1 K. D. Kreuer, Chem. Mater., 1996, 8, 610–641.
2 K. D. Kreuer, S. J. Paddison, E. Spohr and M. Schuster,
Chem. Rev., 2004, 104, 4637–4678.
3 M. A. Hickner, H. Ghassemi, Y. S. Kim, B. R. Einsla and
J. E. McGrath, Chem. Rev., 2004, 104, 4587–4611.
4 K. A. Mauritz and R. B. Moore, Chem. Rev., 2004, 104,
4535–4585.
5 M. E. Schuster and W. H. Meyer, Annu. Rev. Mater. Sci., 2003, 33,
233–261.
6 L. Carrette, K. A. Friedrich and U. Stimming, Fuel Cells, 2001, 1,
5–39.
7 M. F. H. Schuster, W. H. Meyer, M. Schuster and K. D. Kreuer,
Chem. Mater., 2004, 16, 329–337.
8 Z. Zhou, S. W. Li, Y. L. Zhang, M. L. Liu and W. Li, J. Am.
Chem. Soc., 2005, 127, 10824–10825.
9 S. Martwiset, R. C. Woudenberg, S. Granados-Focil,
O. Yavuzcetin, M. T. Tuominen and E. B. Coughlin, Solid State
Ionics, 2007, 178, 1398–1403.
10 D. Basak, S. Christensen, S. K. Surampudi, C. Versek,
D. T. Toscano, M. T. Tuominen, R. C. Hayward and
D. Venkataraman, Chem. Commun., 2011, 47, 5566–5568.
11 D. Semmingsen, F. J. Hollander and T. F. Koetzle, J. Chem. Phys.,
1977, 66, 4405–4412.
12 S. Horiuchi, Y. Tokunaga, G. Giovannetti, S. Picozzi, H. Itoh,
R. Shimano, R. Kumai and Y. Tokura, Nature, 2010, 463,
789–U797.
13 K. Yoshimura and L. S. Liebeskind, Macromol. Rapid Commun.,
2010, 31, 1725–1730.
14 H. Terao, T. Sugawara, Y. Kita, N. Sato, E. Kaho and S. Takeda,
J. Am. Chem. Soc., 2001, 123, 10468–10474.
15 M. W. Reed, D. J. Pollart, S. T. Perri, L. D. Foland and
H. W. Moore, J. Org. Chem., 1988, 53, 2477–2482.
16 L. S. Liebeskind, R. W. Fengl, K. R. Wirtz and T. T. Shawe,
J. Org. Chem., 1988, 53, 2482–2488.
17 D. Bellus, J. Am. Chem. Soc., 1978, 100, 8026–8028.
18 A. Kawada, A. R. McGhie and M. M. Labes, J. Chem. Phys.,
1970, 52, 3121–3125.
19 C. A. Alabi, Z. Chen, Y. S. Yan and M. E. Davis, Chem. Mater.,
2009, 21, 4645–4652.
20 Y. Chen, M. Thorn, S. Christensen, C. Versek, A. Poe,
R. C. Hayward, M. T. Tuominen and S. Thayumanavan, Nat.
Chem., 2010, 2, 503–508.
21 C. Yang, P. Costamagna, S. Srinivasan, J. Benziger and
A. B. Bocarsly, J. Power Sources, 2001, 103, 1–9.
In order to probe the electrochemical stability of 4b and 4c,
cyclic voltammograms (CVs) for these molecules were
recorded in acetonitrile solution at 1 Â 10À3 M under argon.
The CV traces for both samples were similar, exhibiting a
reversible oxidation peak at +0.6 V (vs. Ag/Ag+). CV traces
remain nearly identical even after ten consecutive cycles,
indicating little or no affinity of squaric acid derivatives
towards the platinum electrode (Fig. S4, ESIw). In contrast,
under identical electrochemical conditions, 1H-Imidazole
exhibits irreversible oxidation and significant adsorption onto
the platinum surface.8,21 This observation indicates that squa-
ric acid derivatives do not poison platinum electrodes and
show high electrochemical reversibility.
In conclusion, four different derivatives of squaric acid were
synthesized and investigated for anhydrous proton conduction.
The tert-butyl derivative of squaric acid exhibited proton
conductivity as high as 2.3 Â 10À3 S cmÀ1 at 110 1C while
its n-butyl analogue exhibited relatively poor conductivity.
This variation is attributed to an increased order of the new
phase present in tert-butyl derivative and a lower activation
energy barrier for proton movement. Squaric acid derivatives
also exhibit electrochemical stability compared to irreversible
c
5924 Chem. Commun., 2012, 48, 5922–5924
This journal is The Royal Society of Chemistry 2012