A [2+2] cross-photodimerisation of photostable olefins via a three-component cocrystal solid solution

Article abstract of DOI:10.1039/c2cc15453f

A ditopic hydrogen-bond-donor template in the form of resorcinol facilitates a [2+2] cross-photodimerisation of 4-Cl-stilbazole and 4-Me-stilbazole in a rare cocrystal solid solution. A photoreaction does not proceed with the olefins individually or as a solid solution composed solely of the two olefins.

Full text of DOI:10.1039/c2cc15453f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 1790–1792
www.rsc.org/chemcomm
COMMUNICATION
A [2+2] cross-photodimerisation of photostable olefins via a
three-component cocrystal solid solutionw
Dejan-Kresimir Bucar, Arundhuti Sen, S. V. Santhana Mariappan and
ˇ ˇ
Leonard R. MacGillivray*
Received 1st September 2011, Accepted 25th November 2011
DOI: 10.1039/c2cc15453f
A ditopic hydrogen-bond-donor template in the form of resorcinol
facilitates a [2+2] cross-photodimerisation of 4-Cl-stilbazole and
4
-Me-stilbazole in a rare cocrystal solid solution. A photoreaction
does not proceed with the olefins individually or as a solid solution
composed solely of the two olefins.
[
2+2] Cross-photodimerisation reactions (CPRs) in the solid
state are a challenging task, with a relatively small number
1
–11
having been reported.
The main approach to conduct a
1
solid-state CPR has involved the use of solid solutions
–8
wherein two structurally-related molecules react in a lattice that
is isostructural with at least one of the pure olefins. Success
of the approach has, thus, relied on at least one of the olefins
crystallising to be reactive as a pure solid. Olefins do not
necessarily crystallise, however, to adopt the geometry criteria
Scheme 1
9
two stilbazoles wherein the two alkenes are preorganised in a
1
of Schmidt for a photodimerisation in the solid state, which
4–16
rare cocrystal solid solution
via hydrogen bonds to react to
limits the olefins that react in solids. A second approach to
give a head-to-head cross photoproduct (Scheme 1). The organi-
sation of the olefins and reactivity of the cross-reacting partners
would, thus, rely on hydrogen-bond-donor abilities of the template
and not intrinsic packing of the olefins. We show that cocrystalli-
sation of res with a combination of trans-1-(4-chlorophenyl)-2-
CPRs has involved the use of stoichiometric solids in the form
10–13
of cocrystals.
Although less explored, two-component solids
1
based on either charge-transfer or perfluorobenzene–benzene
0
12
interactions have been used to achieve CPRs with reasonable
success. A particularly successful strategy in this regard has been
the recent work of Wheeler et al. that has shown how hydrogen-
bond-mediated self-assembly of J-shaped olefins can overcome
effects of crystal packing to achieve a CPR in a two-component
(
(
4-pyridyl)ethylene (Cl-PE) and trans-1-(4-methylphenyl)-2-
4-pyridyl)ethylene (Me-PE) (1 : 1 : 1 ratio) affords a three-
component, or ternary, cocrystal solid solution of composition
(res)ꢀ(Cl-PE) ꢀ(Me-PE) (x E 1) (1). The olefins react to
13
cocrystal based on a quasiracemate dimeric scaffold.
(x)
(2ꢁx)
yield the expected heterodimer (ꢂ)-rctt-1-(4-chlorophenyl)-2-
4-methylphenyl)-3,4-bis(4-pyridyl)cyclobutane (2a,b), as well
Here, we report a strategy that combines the solid solution
and cocrystal approaches to facilitate a CPR in the solid state.
The approach achieves a CPR of olefins that are photostable
both individually and when mixed together as a solid solution.
During our efforts to employ small-molecule templates based
on resorcinol (res) to direct [2+2] photodimerisations in the
solid state, we reasoned that a res—despite being a homoditopic
receptor—could be used to direct a CPR of two different olefins
in a solid. In such a design, a res would be co-crystallised with
(
as the homodimers rctt-1,2-(4-chlorophenyl)-3,4-bis(4-pyridyl)-
cyclobutane (2c) and rctt-1,2-(4-methylphenyl)-3,4-bis(4-pyridyl)-
cyclobutane (2d) in near quantitative conversion (i.e. B96%)
(
Scheme 1). We show that a CPR involving Cl-PE and Me-PE
is not achieved in a two-component solid solution of the
1
7
olefins in the absence of the res template. The approach,
thus, provides access to a product that is, otherwise, difficult to
achieve using previous methodologies.
The choices of reactants and template for a CPR are based
on three factors. First, we have utilised templates based on
res to assemble stilbazoles for [2+2] photodimerisations that
Department of Chemistry, University of Iowa, Iowa City, IA, USA.
E-mail: len-macgillivray@uiowa.edu; Fax: +1-319-335-1270;
Tel: +1-319-335-0563
w Electronic supplementary information (ESI) available: Synthetic
procedures (e.g. crystal growth, photoreactivity studies), HPLC
product separation, ESI mass spectra, NMR spectra and crystallo-
graphic data. CCDC 809484–809488. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c2cc15453f
1
8
proceed stereospecifically and quantitatively in solids. The
templates decouple the organisation of olefins from vexatious
effects of crystal packing. Second, early work of Kitaigorodskii
proposed facile interchangeability of Cl-atoms and Me-groups
1
790 Chem. Commun., 2012, 48, 1790–1792
This journal is c The Royal Society of Chemistry 2012

View Article Online
Fig. 1 X-Ray structures: (a) Me-PE along the crystallographic b axis
inset: crystallographic (0 1 0) and (1 0 3) planes) and (b) Cl-ME along
(
the crystallographic (1 0 ꢁ2) and (4 ꢁ1 0) planes (inset: b axis).
in the solid state owing to similarities in shape and size (19.9 vs.
3
3.5 A , respectively). Jones and Desiraju et al. later demon-
19
˚
2
strated the utility of the interchangeability in the crystal engineer-
4
,5
ing of two-component photoreactive solid solutions. Third,
recent reports have described the design and synthesis of rare
1
4–16
13
and quaternary solid solutions of cocrystals.
Fig. 2 X-Ray structure of hydrogen-bonded assemblies within the
ternary cocrystal solid solution 1 (inset: overlap of –Cl and –Me groups).
ternary
Whereas fundamental factors that govern the formation of such
materials are emerging, applications to affect properties (e.g.
reactivity) have yet to be reported. Collectively, the findings
suggested that res, Cl-PE, and Me-PE would be viable candidates
to achieve a CPR in a cocrystalline solid composed of the three
components.
A single-crystal X-ray diffraction analysis revealed the three
components to crystallise in the triclinic space group P 1% as
1
4–16
a rare ternary cocrystal solid solution
of composition
(res)ꢀ(Cl-PE)( ꢀ(Me-PE)
x)
(2ꢁx)
(x E 1) (1). The components of 1
form discrete three-component hydrogen-bonded assemblies
˚
In our initial experiments, we determined both Cl-PE and
Me-PE to be photostable as individual pure solids. The lack of
reactivity was attributed to crystal packings unfavourable
for photodimerisation (Fig. 1). An X-ray structure study of
single crystals of Me-PE, obtained via slow evaporation from
acetonitrile (MeCN), showed the olefin to crystallise in the
(d(Oꢀ ꢀ ꢀN): 2.741(3), 2.733(3) A) with two crystallographically
independent stilbazoles stacked in head-to-head geometries.
The X-ray data are consistent with stacking of either different
or same stilbazoles (Cl-PE/Me-PE occupancies: 0.48 : 0.52 and
0.17 : 0.83) in each assembly in the form of (res)ꢀ(Me-PE)ꢀ(Cl-PE)
(1a,b), (res)ꢀ2(Cl-PE) (1c) and (res)ꢀ2(Me-PE) (1d) (Fig. 2). Within
the assembly, the CQC bonds lie crisscrossed, being separated
1 1 1
chiral orthorhombic space group P2 2 2 with one molecule of
˚
Me-PE in the asymmetric unit (Fig. 1a).z The stilbazoles lie
stacked offset in a head-to-head manner, with neighbour-
ing molecules interacting via C–Hꢀ ꢀ ꢀN and C–Hꢀ ꢀ ꢀp forces.
Within each stack, the carbon–carbon double (CQC) bonds
by 4.06 A. Adjacent assemblies stack anti-parallel, being
˚
separated by 4.70 A (Fig. 2). A [2+2] photodimerisation
was, thus, expected to occur within the hydrogen-bonded
assemblies, with both the heterodimers 2a,b and homodimers
2
2c,d forming as products (Fig. 2).
0
˚
are separated by 6.03 A, which does not conform to the criteria
of Schmidt for photoreaction. A crystallisation of Cl-PE from
To evaluate the photoreactivity of the cocrystal solid solution
MeCN revealed the olefin to crystallise in the achiral triclinic
%
space group P1, with the asymmetric unit consisting of two
1, a powdered crystalline sample was exposed to UV irradiation
1
for ca. 40 h. From H NMR spectroscopy, the generation of
molecules of Cl-PE (Fig. 1b). Adjacent olefins lie organised
head-to-tail, being sustained by ClꢀꢀꢀN(pyridyl) forces. The CQC
cyclobutane products occurred in up to 96% yield. The formation
of a mixture of photoproducts was evidenced by the nearly
complete disappearance of the olefin peaks (d = 7.4–7.7 ppm
range) and the appearance of a broad and unsymmetrical multiplet
in the cyclobutane-proton region (d = 4.4–4.7 ppm range) (Fig. S3,
ESIw). The broad shape of the cyclobutane peak was consistent
with the olefins having reacted to form multiple products.
˚
bonds lie separated by 6.10 and 6.18 A. Upon UV-irradiation
(broadband, 450 W Hg-medium-pressure lamp) for a period of
ca. 10 h, both Me-PE and Cl-PE were found to be photostable.
Given that Cl-PE and Me-PE are photostable, we turned to
develop a solid solution of the olefins. Thus, equimolar amounts
of Cl-PE and Me-PE were refluxed in MeCN. Yellow crystalline
The generation of a heterodimer was confirmed by 2D
1
1
1
13
H– H homonuclear and H– C heteronuclear NMR experi-
plates of composition (Me-PE) ꢀ(Cl-PE)
(where x E 0.5)
formed upon slow cooling after a period of ca. 1 h, as determined
x
(1ꢁx)
ments (vide infra) performed upon isolation of the products
from reacted 1 via an HPLC analysis. More specifically,
through-bond J-coupling correlations (COSY) identified pyridyl,
toluyl, chlorophenyl, and cyclobutyl groups, while through-space
1
by H NMR spectroscopy. An X-ray powder diffraction analysis
showed the solid to resemble the structure of photostable Me-PE
(
Fig. S1, ESIw). In line with the structure of Me-PE, the solid
1
13
solution upon UV irradiation for a period of ca. 10 h was
determined to be photostable. Longer exposure times resulted
in partial decomposition of the components (Fig. S2, ESIw).
To facilitate a CPR of Cl-PE and Me-PE, we next attempted
a cocrystallisation of res, Cl-PE and Me-PE. Specifically,
equimolar amounts of res, Cl-PE and Me-PE were refluxed in
MeCN. Yellow single crystals suitable for X-ray diffraction
formed upon slow cooling from MeCN after a period of ca. 5 h.
NOE (NOESY) and H– C single- and multiple-bond (HSQC
and HMBC) correlations demonstrated connectivity patterns
characteristic of 2a,b (see ESIw). HPLC chromatograms showed
2a,b, 2c, and 2d to be generated in B57%, 29% and 14% yields,
respectively, in the bulk solid (Fig. S13, ESIw). The yields
of the photoproducts differed from the occupancies of the
X-ray experiment, which suggests the composition of 1 to vary
from sample crystal to crystal. Electrospray ionisation (ESI)
This journal is c The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1790–1792 1791

View Article Online
ꢁ
1
C
33
.
34
H
30
.
03Cl
0
.
66
N
2
O
2
,
M
r
=
514.05
˚
g
mol
,
triclinic, P 1% ,
˚
˚
a = 9.9438(11) A, b = 10.1775(11) A, c = 14.0261(15) A, a =
05.642(5)1, b = 98.420(5)1, g = 92.492(5)1, V = 1347.0(3) A , T =
3
˚
1
1
4
wR
1
C
1
1
ꢁ
1
50(2) K, Z = 2, m(MoKa) = 0.142 mm , 8856 reflections measured,
679 independent reflections (Rint = 0.02), R (obs) = 0.0434,
(obs) = 0.1079, R (all) = 0.0585, wR (all) = 0.1148, S =
.067. Crystal data for (296 K) (CCDC 809484):
05Cl , M
= 513.89 g mol , triclinic, P 1% , a =
1
1
2
2
1
ꢁ1
33
.
35
H
30
.
0
.
65
N
2
O
2
r
˚
˚
˚
0.0708(11) A, b = 10.4551(11) A, c = 13.9967(15) A, a =
3
˚
05.885(5)1, b = 99.037(5)1, g = 92.659(5)1, V = 1393.6(3) A ,
ꢁ1
T = 296(2) K, Z = 2, m(MoKa) = 0.136 mm , 8611 reflections
measured, 4843 independent reflections (Rint = 0.0253), R (obs) =
.0674, wR (obs) = 0.1941, R (all) = 0.1138, wR (all) = 0.2142,
1.087. Crystal data for
05Br Cl 63, M
, b = 9.8023(11) A
a = 901, b = 93.253(5)1, g = 901, V = 2581.9(5) A , T = 200(2) K,
1
2 2
Fig. 3 X-Ray structure of 3 showing H -2a,b and homodimers H -2c,d.
0
S
C
1
2
2
=
3
(CCDC 809486):
mass-spectrometry data of fractions from the HPLC analysis
displayed peaks at m/z 411, 431 and 391, which are consistent
with the heterodimers 2a,b, as well as homodimers 2c and 2d.
ꢁ1
26
.
93
H
26
.
2
1
.
07
N
2
O
0
˚
.
r
= 585.54 g mol , monoclinic,
˚ ˚
, c = 14.3984(15) A,
3
P2 /c, a = 18.3233(19) A
1
˚
ꢁ
1
21
Z = 4, m(MoKa) = 3.271 mm , 11967 reflections measured, 4489
independent reflections (Rint = 0.0385), R (obs) = 0.0484, wR (obs) =
.1023, R (all) = 0.0733, wR (all) = 0.1096, S = 1.025.
To confirm stereochemistries of the photoproducts, we
managed to grow single crystals of an HBr salt of the cyclo-
butane photoproducts once isolated from the res template. Single
1
1
0
2
2
2+
ꢁ
1 J. D. Hung, M. Lahav, M. Luwisch and G. M. J. Schmidt, Isr. J.
Chem., 1972, 10, 585–599.
crystals of [(H -2x) ][Br ] ꢀ0.63(H O) (3) (where x = a–d) were
2
2
2
grown from MeCN to which few drops of HBr(aq) were added.
Red-brown single crystals grew over a period of ca. 4 days.
A single-crystal X-ray structure analysis of 3 was consistent
with the solid being composed of protonated cyclobutanes
of 2a–d (Fig. 3). The salt 3 crystallised in the triclinic space
2
3
4
M. D. Cohen, R. Cohen, M. Lahav and P. L. Nie, J. Chem. Soc.,
Perkin Trans. 2, 1973, 1095–1100.
A. Elgavi, B. S. Green and G. M. J. Schmidt, J. Am. Chem. Soc.,
1973, 95, 2058–2059.
W. Jones, C. R. Theocharis, J. M. Thomas and G. R. Desiraju,
J. Chem. Soc., Chem. Commun., 1983, 23, 1443–1444.
5 C. R. Theocharis, G. R. Desiraju and W. Jones, J. Am. Chem. Soc.,
1984, 106, 3606–3609.
%
group P1, with the asymmetric unit consisting of diprotonated
2
+
cyclobutanes [(H
2
-2x) ]. The photoproducts exhibit a head-
6
K. Kinabara, Y. Adegawa, K. Saigo and M. Hasegawa, Bull.
Chem. Soc. Jpn., 1993, 66, 1204–1210.
to-head geometry, which agrees with the structure of
(
res)ꢀ(Cl-PE)(x)ꢀ(Me-PE)(2ꢁx) (x E 1) (Fig. 3). The cyclo-
7 T. Suzuki, T. Fukushima, Y. Yamashita and T. Miyashi, J. Am.
Chem. Soc., 1994, 116, 2793–2803.
8 Y. Maekawa, S. Kato and M. Hasegawa, J. Am. Chem. Soc., 1991,
113, 3867–3872.
9 G. M. J. Schmidt, Pure Appl. Chem., 1971, 27, 647–678.
butanes are intercalated, being sustained by a combination
ꢁ
of C–Hꢀ ꢀ ꢀBr , Clꢀ ꢀ ꢀp, C–Hꢀ ꢀ ꢀp, and pꢀ ꢀ ꢀp forces. The included
water molecules partially occupy cavities formed between
2
pairs of [(H -2x) ][Br ] . ESI mass spectrometry and HPLC
+
ꢁ
10 G. R. Desiraju and C. V. K. M. Sharma, J. Chem. Soc., Chem.
Commun., 1991, 1239–1241.
2
2
analyses revealed the product distribution in 3 to compare
favourably to the bulk photoreacted solid (res)ꢀ(Cl-PE)(x)
Me-PE)(2ꢁx) (x E 1) (Fig. S20, ESIw).
In conclusion, we have described a CPR developed in a three-
1
1 (a) I. Weissbuch and M. Lahav, Chem. Rev., 2011, 111, 3236–3267;
b) J. Bernstein, B. S. Green and M. Rejto, J. Am. Chem. Soc.,
1980, 102, 323–328.
ꢀ
(
(
1
2 G. W. Coates, A. R. Dunn, L. M. Henling, J. W. Ziller, E. B.
Lobkovsky and R. H. Grubbs, J. Am. Chem. Soc., 1998, 120,
component cocrystal solid solution involving two stilbazoles
and a res template. A solid solution of the two olefins alone was
photostable, which underscores the utility of the approach. In
3
641–3649.
13 (a) R. C. Grove, S. H. Malehorn, M. E. Breen and K. A. Wheeler,
Chem. Commun., 2010, 46, 7322–7324; (b) K. A. Wheeler, J. D.
Wiseman and R. C. Grove, CrystEngComm, 2011, 13, 3134–3137.
2
2
addition to being applicable to other olefins, we expect the
approach here to be promising for asymmetric CPRs where a
source of chirality (e.g. auxiliary, lattice) is integrated into the
reactive solid. Factors such as the influence of concentration
variation of the olefins are also under investigation.
1
4 M. Dabros, P. R. Emery and V. R. Thalladi, Angew. Chem.,
Int. Ed., 2007, 46, 4132–4135.
15 M. A. Oliveira, M. L. Peterson and D. Klein, Cryst. Growth Des.,
008, 8, 4487–4493.
2
1
6 S. Chen, H. Xi, R. F. Henry, I. Marsden and G. G. Z. Zhang,
CrystEngComm, 2010, 12, 1485–1493.
We thank the National Science Foundation (LRM DMR-
1
7 Extensive polymorph screening could, in principle, be applied to
a solid solution of the type described here to discover a reactive
phase.
1
104650) for financial support. We also thank Prof. Amnon
Kohen (University of Iowa) for use of HPLC instrumentation.
1
8 L. R. MacGillivray, J. L. Reid, J. A. Ripmeester and G. S.
Papaefstathiou, Ind. Eng. Chem. Res., 2002, 41, 4494–4497.
9 A. I. Kitaigorodskii, Molecular Crystals and Molecules, Academic
Press, London, 1973.
Notes and references
1
z Crystal data for Cl-PE (CCDC 809487): C13
H
10ClN, M
r
= 215.67 g
ꢁ1
mol , triclinic, P 1% , a = 9.4815(10) A, b = 9.6019(12) A, c =
˚
˚
20 The cocrystal solid solution 1 is isostructural with (res)ꢀ2(Cl-PE)
wherein a cyclobutane forms stereospecifically and in quantitative
yield (see ref. 18). Cocrystallisation of 1 and Me–PE afforded a
solid that exhibits broad peaks in the PXRD diffractogram and is
not isostructural with 1 (see ESIw). Experiments are underway to
determine the structure and reactivity of the resulting solid.
21 The stereochemistry of the cyclobutane ring could not be unequivocally
determined from the suite of NMR data since the resonances of the
cyclobutyl protons were not resolved and unique NOEs could not be
identified even at a magnetic field of 13.8 tesla (600 MHz).
˚
1
1
4.2885(18) A, a = 71.745(5)1, b = 76.746(6)1, g = 62.801(5)1, V =
3
ꢁ1
˚
093.0(2) A , T = 298(2) K, Z = 4, m(MoKa) = 0.312 mm , 6753
reflections measured, 3773 independent reflections (Rint = 0.0641),
(obs) = 0.1168, wR (obs) = 0.2928, R (all) = 0.2106, wR (all) =
.3451, S = 1.061. Crystal data for Me-PE (CCDC 809488): C14 13N,
R
0
M
7
1
1
1
2
2
H
ꢁ1
˚
r
= 195.25 g mol , orthorhombic, P2
˚
1
2
1
2
1
, a = 6.0339(7) A, b =
˚
.6092(9) A, c = 23.135(3) A, a = 901, b = 901, g = 901, V =
3
ꢁ1
˚
062.2(2) A , T = 150(2) K, Z = 4, m(MoKa) = 0.071 mm , 6496
reflections measured, 1117 independent reflections (Rint = 0.0379),
(obs) = 0.038, wR (obs) = 0.0929, R (all) = 0.0486, wR (all) =
.0973, S = 1.151. Crystal data for 1 (150 K) (CCDC 809485):
R
0
1
1
2
2
22 M. Yoshizawa, Y. Takeyama, T. Okano and M. Fujita, J. Am.
Chem. Soc., 2003, 125, 3243–3247.
1
792 Chem. Commun., 2012, 48, 1790–1792
This journal is c The Royal Society of Chemistry 2012