In situ Observation of Molecular Swapping in a Crystal by X-ray Analysis

Article abstract of DOI:10.1246/cl.2000.1372

X-ray diffraction analysis elucidated that a pair of hexaarylbiimidazoles was transformed to a complex composed of two lophyl radicals and a piezodimer on irradiation at low temperatures. The piezodimer, the existence of which has been assumed spectroscopically as an unstable isomer, was produced via molecular swapping of two photo-induced lophyl radicals with drastic conformational change. The black crystal reverted to the original structure on warming to 25 deg C.

Full text of DOI:10.1246/cl.2000.1372

1
372
Chemistry Letters 2000
In situ Observation of Molecular Swapping in a Crystal by X-ray Analysis
Masaki Kawano,* Tomokatsu Sano, Jiro Abe,†††,# and Yuji Ohashi*†,††
†
†††
†
CREST, Japan Science and Technology Corporation
†
†
Department of Chemistry and Materials Science, Tokyo Institute of Technology, Tokyo 152-8551
Department of Photo-Optical Engineering, Tokyo Institute of Polytechnics, Kanagawa 243-0297
†††
(Received September 13, 2000; CL-000850)
X-ray diffraction analysis elucidated that a pair of hexa-
(1.482(2) Å) connecting two imidazolyl groups coincided with
1
3
arylbiimidazoles was transformed to a complex composed of
two lophyl radicals and a piezodimer on irradiation at low tem-
peratures. The piezodimer, the existence of which has been
assumed spectroscopically as an unstable isomer, was produced
via molecular swapping of two photo-induced lophyl radicals
with drastic conformational change. The black crystal reverted
to the original structure on warming to 25 °C.
that of o-Cl-HABI. The shortest intermolecular distance (4.32
Å) among non-hydrogen atoms was more than the sum of van
der Waals radii.
Since Schmidt and Cohen started systematic analysis of the
photodimerization of cinnamic acids in crystals almost four
1
decades ago, the solid-state organic reactions have been exten-
sively investigated from various aspects.2 Spectroscopic meth-
ods have been playing an important role in analyzing the tran-
3
sient unstable species. We have extensively studied a variety
of reactions in organic and organometallic crystals with retain-
ing single crystal form, which are called crystalline-state reac-
tions.4 The structural change in the process of crystalline-
state reaction can be observed directly by X-ray analysis.
Although such in situ observations are very informative
to make clear the reaction mechanism, there are many experi-
mental difficulties when we intend to observe unstable or
metastable species produced in organic crystals. In the previ-
ous paper we reported the observation of the metastable pho-
5
tochromic species using two-photon excitation. If the molecu-
lar motion was somewhat large during the reaction, the cry-
otrapping method, which has been utilized in a variety of
physicochemical measurements, was very effective to keep the
crystallinity.6 Even if the product content was less than 10%
in a crystal, the product structure was successfully distinguished
A pale yellow crystal of HABI was irradiated at low tem-
peratures under 108 K with a high-pressure mercury lamp for
30 min. The disordered molecular structures of 1,2'-dimers
,7
8
from the original one.
(open line) and photoproducts (solid line) are depicted in Figure
–
1
4
The hexaarylbiimidazole (HABI) derivatives, which are
well-known as photo/thermo/piezo-chromic materials, have been
used as polymerization initiators in industry and therefore have
1. The space group (P1) was retained after irradiation. The
structure of the photoproduct clearly indicated the progress of
reaction (Scheme 1). Two C–N bonds connecting two imida-
9,10
been investigated spectroscopically since Maeda and Hayashi
prepared them.11 It has been proposed that the photolysis of the
HABI derivative in benzene instantaneously produces a pair of
lophyl radicals. This lophyl radical has been proposed to be an
initiator for the radical polymerization. In the previous paper the
photo-produced radical pair of the o-Cl-HABI derivative was
successfully observed by X-rays using the cryotrapping
6
method. When HABI was used as the mother compound, unex-
pected intermediates were observed. Here we report the in
situ observation of photo-induced unstable intermediates derived
from HABI in a crystal using the cryotrapping method.
zolyl groups in two inversion-center-related 1,2'-dimers were
cleaved on irradiation to produce two planar lophyl radicals and
a 2,2'-dimer in 13.9(2)% yields in a unit cell. Two imidazolyl
rings with a 6π-electron system in two 1,2'-dimers moved
inward (by max 1.69 Å) to form a bridging C–C bond of 2,2'-
Before irradiation the crystal structure of HABI (1,2'-
1
2
dimer) was determined at 108 K by X-ray diffraction. The
bond distances and angles of HABI were identical to those of o-
Cl-HABI within experimental errors. The C–N bond distance
Copyright © 2000 The Chemical Society of Japan

Chemistry Letters 2000
1373
dimer. This was accompanied with drastic conformational
changes of one of phenyl rings (the central one in Figure 1 was
rotated by 60°). The intermolecular distance between the bridg-
ing carbon atoms before irradiation was 4.87 Å. Two other imi-
dazolyl rings with a 4π-electron system moved outward (by max
References and Notes
#
Present address: Department of Applied Chemistry, Graduate School
of Engineering, Tokyo Metropolitan University, Tokyo 192-0397
G. M. J. Schmidt and M. D. Cohen, J. Chem. Soc., 1964, 1996.
M. D. Cohen, Angew. Chem., Int. Ed. Engl., 14, 386 (1975); J. M.
McBride, B. E. Segmuller, M. D. Hollingsworth, D. E. Mills, and B.
A. Weber, Science, 234, 830 (1986); J. R. Scheffer, in “Organic
Solid State Chemistry,” ed. by G. R. Desiraju, Elsevier, Amsterdam
(1987), Chap. 1; J. N. Gamlin, J. N. Gamlin, R. Jones, M. Leibovitch,
B. Patrick, J. R. Scheffer, and J. Trotter, Acc. Chem. Res., 29, 203
1
2
1
.86 Å) to form two planar lophyl radicals sandwiching the 2,2'-
dimer. On warming to 25 °C, the black crystal returned to the
initial pale-yellow HABI crystal. The black crystal was consider-
ably stable at room temperature in contrast with the irradiated
crystal of o-Cl-HABI containing lophyl radical pairs only.
(1996).
3
4
N. J. Turro, “Modern Molecular Photochemistry,” University
Science Books, Sausalito, CA (1991).
ESR spectra of a single crystal of HABI irradiated at 77 K
for 30 min showed a broad doublet signal corresponding to free
lophyl radicals at g = 2.003. Reflective visible spectra of an
irradiated KBr pellet of HABI also indicated the presence of
lophyl radicals (λmax = 566 nm). In addition, its infrared spectra
consisted of bands assigned to an unreacted 1,2'-dimer, a
Y. Ohashi, Acc. Chem. Res., 21, 268 (1988); Y. T. Osano, A.
Uchida, and Y. Ohashi, Nature, 352, 510 (1991); Y. Ohashi, Acta
Crystallogr. Sect. A, 54, 842 (1998); Y. Ohashi and Y. Sasada,
Nature, 267, 142 (1977).
5
6
7
J. Harada, H. Uekusa, and Y. Ohashi, J. Am. Chem. Soc., 121, 5809
(
1999).
M. Kawano, T. Sano, J. Abe, and Y. Ohashi, J. Am. Chem. Soc., 121,
106 (1999).
8
15
piezodimer of HABI and lophyl radicals. Figure 2 shows the
U. K. Genick, S. M. Soltis, P. Kuhn, I. L. Canestrelli, and D. E.
Getzoff, Nature, 392, 206 (1998); P. Coppens, D. V. Fomitchev, M.
D. Carducci, and K. Culp, J. Chem. Soc., Dalton Trans., 1998, 865.
M. Kawano, A. Ishikawa, Y. Morioka, H. Tomizawa, E. Miki, and
Y. Ohashi, J. Chem. Soc., Dalton Trans., 2000, 2425.
A. B. Cohen and P. Walker, in “Imaging Processes and Materials,”
Neblette’s 8th ed., ed. by J. M. Sturge, V. Walworth, and A. Shepp,
van Nostrand Reinhold, New York (1998), pp. 226–262.
8
9
1
0
K. Maeda, in “Processes in Photoreactive Polymers,” ed. by V. V.
Krongauz and A. D. Trifunac, Chapman & Hall, New York (1995),
pp. 90–110.
1
1
1
2
T. Hayashi and K. Maeda, Bull. Chem. Soc. Jpn., 33, 565 (1960).
HABI was prepared according to ref 11. A thin plate crystal (220 ×
1
40 × 20 µm) of HABI was cooled below 108 K. The 1,2'-dimer
structure of HABI was determined by X-ray diffraction involving the
use of a SMART CCD area detector diffractometry system. With
irradiation, the X-ray data showed no significant crystal decay.
–
C H N , MW = 590.70, triclinic, space group P1, T = 108(2) K, a =
42
30
4
1
1.3298(4) Å, b = 12.8275(4) Å, c = 13.0737(4) Å, α = 60.634(1)°,
3
β = 80.389(1)°, γ = 67.071(1)°, V = 1524.37(9) Å , Z = 2, d
=
calcd
3
1
.287 Mg/m . Structural solution by direct methods and anisotropic
2
least-squares refinement (based on F ) were performed with
SHELXTL (version 5.0). Anisotropic least-squares refinement (416
parameters) on 8853 independent merged reflections (Rint = 0.0439)
converged at R (F) = 0.0462 for 6009 observed data (I > 2σ(I)),
1
GOF = 1.014.
1
3
4
o-Cl-HABI indicates 2,2'-di(o-chlorophenyl)-4,4',5,5'-tetraphenyl
biimidazole. Photolysis of a crystal of o-Cl-HABI produced radical
pairs in 10% yields as reported in ref 6.
IR spectra of 1,2'-dimer (A–D) and piezodimer independently
prepared according to Reference 15. The bands with asterisks of
1
A thin plate crystal (300 × 160 × 20 µm) of HABI was cooled below
1
,2'-dimer in (A) showed salient changes on irradiation. The
1
08 K and irradiated with a high-pressure mercury lamp for 30 min.
42 30 4
–
spectrum of C indicates the appearance of piezodimers and the
decrease of 1,2'-dimer. The IR spectra of A–D show that this
photochemical process is reversible. These facts support that the
piezodimer and lophyl radicals were formed from HABI.
Although the formation of a piezodimer from HABI has been
assumed without knowledge of its structural formula, its struc-
ture has never been determined because of the difficulty in
obtaining single crystals. The recrystallization of the independ-
ently prepared piezodimers led to the formation of stable 1,2'-
dimers owing to the thermodynamical instability; a 2,2'-dimer (a
C H N , MW = 590.70, triclinic, space group P1, T = 108(2) K, a
= 11.3660(7) Å, b = 12.8563(9) Å, c = 13.085(1) Å, α = 60.638(1)°,
3
β = 80.208(1)°, γ = 66.501(1)°, V = 1527.5(2) Å , Z = 2, d
=
calcd
3
1
.287 Mg/m . During the data collection below 108 K, no decay
was observed. Isotropic and anisotropic least-squares refinement
(378 parameters/291 restraints) on 8857 independent merged reflec-
tions (Rint = 0.0783) converged at R (F) = 0.0741 for 4643 observed
1
data (I > 2σ(I)), GOF = 1.031. Refinement: Photolysis of HABI
produced three new molecular groups except the unreacted 1,2'-
dimer (63.8(8)%); that is, lophyl radicals (13.9(2)%), 2,2'-dimer
(
13.9(2)%), and another unreacted 1,2'-dimer (22.3(5)%). The minor
unreacted 1,2'-dimer was located close to the parent unreacted 1,2'-
dimer. The parent 1,2'-dimer was treated as a rigid group that was
modeled from the structural parameters in ref 12. The temperature
parameters of the parent 1,2'-dimer were fixed. The phenyl groups
of lophyl radicals and 2,2'-dimer were restrained due to severe over-
laps with those of the parent 1,2'-dimer. The crystal structure of the
minor 1,2'-dimer is omitted in Figure 1 for clarity.
4
π–4π electron system) is less stable than a 1,2'-dimer (a 4π–6π
electron system) because of lack of resonance effects in the imi-
dazolyl rings. In this study HF/6-31G calculations predicted that
–1
a 1,2'-dimer is more stable by 5.9 kcal mol than a 2,2'-dimer.
This study shows the successful application of a cryotrap-
ping method to capture light-induced unstable intermediates in
a crystal and to characterize their structures by in situ X-ray
crystallography.
15 D. M. White and J. Sonnenberg, J. Am. Chem. Soc., 88, 3825 (1966).
16
T. Shida, K. Maeda, and T. Hayashi, Bull. Chem. Soc. Jpn., 43, 652
(1970); K. Maeda and T. Hayashi, Bull. Chem. Soc. Jpn., 43, 429
(1970).
This work was supported by CREST from JST.