The dark side of crystal engineering: Creating glasses from small symmetric molecules that form multiple hydrogen bonds

Article abstract of DOI:10.1021/ja063353s

Glasses made from compounds of low molecular weight are useful materials with many attractive features, including well-defined compositions. At present, there are no reliable ways to identify molecules that will form long-lived glasses, and efforts to des

Full text of DOI:10.1021/ja063353s

Published on Web 07/22/2006
The Dark Side of Crystal Engineering: Creating Glasses from Small
Symmetric Molecules that Form Multiple Hydrogen Bonds
Olivier Lebel,1 Thierry Maris, Marie-EÅ ve Perron, Eric Demers, and James D. Wuest*
De´partement de Chimie, UniVersite´ de Montre´al, Montre´al, Que´bec H3C 3J7, Canada
Received May 14, 2006; E-mail: james.d.wuest@umontreal.ca
Amorphous molecular materials (molecular glasses) are a prime
source of isotropic thin films for use in optoelectronic devices.2,3
At present, molecular glasses are most commonly derived from
polymers, but compounds of low molecular weight can also exist
in amorphous forms with high glass transition temperatures (Tg >
25 °C).2,3 Because such glasses are derived from small molecules,
they offer the advantage of working with materials of precise mass,
well-defined composition, high purity, and good processability.
However, materials made from small molecules tend to reach
thermodynamic equilibrium quickly, thereby converting amorphous
forms into ordered crystalline phases. Few low-molecular-weight
glasses can forestall this process indefinitely, especially at temper-
atures above Tg.
Regrettably, there are few reliable guidelines for identifying new
small molecules that will form long-lived glasses. Current efforts
to design such glasses are based on relatively crude principles, such
as inhibiting crystallization by incorporating long alkyl chains and
bulky substituents, lowering symmetry, avoiding planarity, increas-
ing molecular size, selecting awkward globular shapes, reducing
intermolecular cohesion, and favoring multiple conformations.4
Recent research in crystal engineering has provided deeper under-
standing of how molecular crystals can be created by design, thereby
raising the possibility of using this understanding not to favor
crystallization but to thwart it. For example, crystal engineers have
established that hydrogen bonding provides strong directional
cohesion that can favor crystallization according to established
patterns; at the same time, however, hydrogen bonding creates an
opportunity to inhibit crystallization (1) by inducing molecules to
aggregate unproductively in ways that preclude effective packing
and disfavor crystallization, and (2) by slowing diffusion and
reorientation in the amorphous state. At present, high-temperature
long-lived glasses made from small molecules that engage in
hydrogen bonding are rare, and the role of hydrogen bonding in
these materials is not clear.5 In this paper, we report the results of
an initial exploration of the “dark side” of crystal engineering, in
which structural features and directional interactions that normally
favor crystallization are subverted to work against it.
cooled under similar conditions, so the mexyl group appears to play
a crucial role in inhibiting crystallization. Monomexyl derivatives
2b,c, prepared from mexyldicyandiamide by standard methods,6,7,11
yielded glasses with Tg ) 51 and 38 °C, respectively, establishing
that a single mexyl group suffices to block crystallization. Not
surprisingly, the lower symmetry of derivatives 2b,c leads to lower
values of Tg. Esters 2d-f, prepared from dimexyl compound 2a
by transesterification,11 also yielded glasses, and their values of Tg
(62, 58, and 54 °C, respectively) decreased as expected as the length
of the alkyl chain increased.
In addition, we examined the behavior of derivatives 4a,b, in
which the carbomethoxy group of bis(mexylamino)triazine 2a has
been replaced by other substituents of similar size. Compounds 4a,b,
which were prepared from cyanuric chloride by classic methods,11,12
both formed glasses (Tg ) 54 and 94 °C, respectively). In sharp
contrast, very close analogue 4c, which cannot self-associate by
hydrogen bonding, did not produce a glass under similar conditions.
Together, these observations suggest that the mexyl group and
hydrogen bonding are both essential and act in concert to forestall
crystallization, even though strong directional association in small,
symmetric, and relatively inflexible molecules normally promotes
crystallization.
To probe this unusual behavior, we crystallized bis(mexylamino)-
triazine 2a from CHCl3 and solved its structure by X-ray crystal-
lography. The crystals were found to belong to the monoclinic space
group C2/c and to correspond to an inclusion compound of
composition 2a‚1CHCl3‚xH2O.13 Approximately 39% of the volume
of the crystals is accessible to guests, as measured by standard
methods.14 We suggest that the failure of compound 2a to yield a
normal close-packed structure without included guests helps provide
the free volume typically associated with molecular glasses.2,3
Moreover, it is a sign that structural features in compound 2a may
make efficient crystallization inherently difficult. Like related
diarylbiguanides and bis(arylamino)triazines,6,7 compound 2a crys-
In exploring the supramolecular chemistry of biguanides and
aminotriazines,6,7 we noted that both 1,5-bis(3,5-dimethylphenyl)-
biguanide (1 ) 1,5-dimexylbiguanide)6,8 and methyl 4,6-bis-
(mexylamino)-1,3,5-triazine-2-carboxylate (2a)7 form glasses with
high Tg (37 and 75 °C, respectively), as measured by modulated
differential scanning calorimetry (mDSC).9 These glasses result even
when melts are cooled slowly, and they do not show a propensity
to crystallize, despite the small size of the molecules, their high
symmetry, their low degree of flexibility, and their conspicuous
ability to form hydrogen bonds.10 To identify the structural elements
that underlie this surprising behavior, we examined the properties
of aminotriazine 2a and related compounds in detail.
Neither simple bis(phenylamino)triazine 3a7 nor its methyl-
substituted derivatives 3b-d7 formed glasses when melted and
9
10372
J. AM. CHEM. SOC. 2006, 128, 10372-10373
10.1021/ja063353s CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
a crucial role in thwarting crystallization, presumably by causing
aminotriazine units to self-associate and to form aggregates with
reduced mobility and particular structural features that cannot
readily be accommodated in closely packed periodic structures.
Our observations are noteworthy because they demonstrate that
long-lived glasses can be made from families of small molecules
with properties that normally promote rapid crystallization, includ-
ing high symmetry, low flexibility, and strong intermolecular
cohesion. New molecular glasses can be produced in a rational way
by turning to the “dark side” of crystal engineering and making
small but carefully selected structural modifications specifically
designed to thwart established patterns of crystallization.
Figure 1. View of the structure of crystals of methyl 4,6-bis(mexylamino)-
1,3,5-triazine-2-carboxylate (2a) grown from CHCl3. The view shows how
molecules of compound 2a adopt an amphiphilic conformation and form
hydrogen bonds with two neighbors to define tapes that lie along the ac
diagonal. Hydrogen bonds are represented by dotted lines. Carbon atoms
are shown in gray, hydrogen atoms in white, nitrogen atoms in blue, and
oxygen atoms in red.
Acknowledgment. We are grateful to the Natural Sciences and
Engineering Research Council of Canada, the Ministe`re de l’EÄ du-
cation du Que´bec, the Canada Foundation for Innovation, and the
Canada Research Chairs Program for financial support.
Supporting Information Available: Experimental procedures for
making compounds 2b-f and 4a-c and characterization of these
compounds; representative mDSC traces, additional crystallographic
details, and FT-IR spectra for compound 2a. This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Fellow of the Ministe`re de l’EÄ ducation du Que´bec, 2002-2004.
(2) Fuhrman, T.; Salbeck, J. AdV. Photochem. 2002, 27, 83-166.
(3) (a) Shirota, Y. J. Mater. Chem. 2005, 15, 75-93. (b) Strohriegel, P.;
Grazulevicius, J. V. AdV. Mater. 2002, 14, 1439-1452.
(4) (a) Miao, Q.; Chi, X.; Xiao, S.; Zeis, R.; Lefenfeld, M.; Siegrist, T.;
Steigerwald, M. L.; Nuckolls, C. J. Am. Chem. Soc. 2006, 128, 1340-
1345. (b) Aujard, I.; Baltaze, J.-P.; Baudin, J.-B.; Cogne´, E.; Ferrage, F.;
Jullien, L.; Perez, EÄ .; Pre´vost, V.; Qian, L. M.; Ruel, O. J. Am. Chem.
Soc. 2001, 123, 8177-8188. (c) Wang, S.; Oldham, W. J., Jr.; Hudack,
R. A., Jr.; Bazan, G. C. J. Am. Chem. Soc. 2000, 122, 5695-5709. (d)
Shirota, Y. J. Mater. Chem. 2000, 10, 1-25.
Figure 2. View along the ac diagonal showing a truncated 2 × 4 × 1
array of unit cells in the structure of bis(mexylamino)triazine 2a. The view
shows how the characteristic hydrogen-bonded tapes that lie along the ac
diagonal (Figure 1) are packed along the b axis. For clarification, the cross
section of one tape is depicted in green, and guest molecules are omitted.
Carbon atoms are shown in gray, hydrogen atoms in white, nitrogen atoms
in blue, and oxygen atoms in red.
(5) (a) Baran, J.; Davydova, N. A.; Pietrasko, A. J. Mol. Struct. 2005, 744-
747, 301-305. (b) Dirama, T. E.; Carri, G. A.; Sokolov, A. P. J. Chem.
Phys. 2005, 122, 114505. (c) Boils, D.; Perron, M.-EÅ .; Monchamp, F.;
Duval, H.; Maris, T.; Wuest, J. D. Macromolecules 2004, 37, 7351-
7357. (d) Takahashi, A.; Mallia, V. A.; Tamaoki, N. J. Mater. Chem.
2003, 13, 1582-1587. (e) Boileau, S.; Bouteiller, L.; Foucat, E.; Lacoudre,
N. J. Mater. Chem. 2002, 12, 195-199. (f) Chelli, R.; Cardini, G.;
Procacci, P.; Righini, R.; Califano, S. J. Chem. Phys. 2002, 116, 6205-
6215. (g) Kim, S.-J.; Karis, T. E. J. Mater. Res. 1995, 10, 2128-2136.
(h) Naito, K. Chem. Mater. 1994, 6, 2343-2350.
tallizes in an amphiphilic conformation with a lipophilic aryl region
and a polar triazinecarboxylate headgroup that forms multiple
hydrogen bonds (Figure 1). Association of the headgroups creates
tapes with figure-8-shaped cross sections, which then pack with
interdigitation of the aryl groups (Figure 2). Related bilayered
structures are a common feature of diarylbiguanides and bis-
(arylamino)triazines.6,7 In the case of compound 2a, however, the
mexyl groups can be seen to interfere by preventing efficient
packing. In particular, no aromatic interactions are present, and the
closest distance between the centers of aromatic rings is 4.84(1)
Å.
FT-IR spectra of solutions of bis(mexylamino)triazine 2a (CH2-
Cl2) showed an N-H stretch at 3403 cm-1, which shifted to 3346
cm-1 in the glassy state and to 3359 cm-1 in crystals, thereby
confirming that compound 2a forms hydrogen-bonded aggregates
in the glass. This aggregation does not appear to involve the
carbonyl group, which gives rise to similar bands in solution (1747
cm-1) and in the glassy state (1751 cm-1), whereas stretching occurs
at 1738 cm-1 in crystals. We conclude that hydrogen bonding plays
(6) Lebel, O.; Maris, T.; Duval, H.; Wuest, J. D. Can. J. Chem. 2005, 83,
615-625.
(7) Lebel, O.; Perron, M.-EÅ .; Maris, T.; Zalzal, S. F.; Nanci, A.; Wuest, J. D.
Chem. Mater. 2006, in press.
(8) The 1,3-dimethyl-5-phenyl group can be abbreviated by mexyl: Ryu, D.
H.; Corey, E. J. J. Am. Chem. Soc. 2005, 127, 5384-5387.
(9) Reading, M.; Luget, A.; Wilson, R. Thermochim. Acta 1994, 238, 295-
307.
(10) Such glasses have been kept at 25 °C for more than one year without
crystallizing.
(11) See the Supporting Information for details.
(12) (a) Thurston, J. T.; Schaefer, F. C.; Dudley, J. R.; Holm-Hansen, D. J.
Am. Chem. Soc. 1951, 73, 2992-2996. (b) Dudley, J. R.; Thurston, J. T.;
Schaefer, F. C.; Holm-Hansen, D.; Hull, C. J.; Adams, P. J. Am. Chem.
Soc. 1951, 73, 2986-2990.
(13) The value of x is estimated to be 6-7.
(14) (a) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool; Utrecht
University: Utrecht, The Netherlands, 2001. (b) van der Sluis, P.; Spek,
A. L. Acta Crystallogr. 1990, A46, 194-201.
JA063353S
9
J. AM. CHEM. SOC. VOL. 128, NO. 32, 2006 10373