Photochemical-mediated solid-state [2+2]-cycloaddition reactions of an unsymmetrical dibenzylidene acetone (monothiophos-dba)

Article abstract of DOI:10.1039/c2ce25480h

A novel unsymmetrical dibenzylidene acetone (monothiophos-dba, compound 3), containing two chemically different alkene moieties, undergoes a selective photochemically-induced [2+2] cycloaddition solid-state transformation (single crystal and powder), as shown by X-ray diffraction methods. In the single crystal the rate constant for the cycloaddition reaction was determined {k = (2.3 ± 0.1) × 10^-4 s^-1}, with a 69% conversion attained, which compares well with a two-photon irradiation experiment using a Nd:YAG laser pumped dye laser at 605 nm radiation (66% conversion). Full conversion was attained using powdered 3 after 43 h. A similar conversion was noted between the single crystal and powder forms (59% and 66% conversion after 5 h irradiation). Irradiation of monothiophos-dba in solution using a 325 nm laser, and monitoring by ¹H NMR spectroscopy, showed that no cycloaddition occurred, but exclusive photochemically-induced trans → cis isomerisation of the alkene bond connecting the unsubstituted aryl group (cis → trans isomerisation in these electron-deficient and polarised CC bonds occurs rapidly by thermal equilibration).

Full text of DOI:10.1039/c2ce25480h

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
CrystEngComm
Cite this: CrystEngComm, 2012, 14, 5564–5571
www.rsc.org/crystengcomm
PAPER
Photochemical-mediated solid-state [2+2]-cycloaddition reactions of an
unsymmetrical dibenzylidene acetone (monothiophos-dba){
A. G. Jarvis,a H. A. Sparkes,*b S. E. Tallentire,b L. E. Hatcher,c M. R. Warren,c P. R. Raithby,*c D. R. Allan,d
A. C. Whitwood,a M. C. R. Cockett,a S. B. Duckett,a J. L. Clarka and I. J. S. Fairlamb*a
Received 2nd April 2012, Accepted 13th June 2012
DOI: 10.1039/c2ce25480h
A novel unsymmetrical dibenzylidene acetone (monothiophos-dba, compound 3), containing two
chemically different alkene moieties, undergoes a selective photochemically-induced [2+2]
cycloaddition solid-state transformation (single crystal and powder), as shown by X-ray diffraction
methods. In the single crystal the rate constant for the cycloaddition reaction was determined
{k = (2.3 ¡ 0.1) 6 1024 s21}, with a 69% conversion attained, which compares well with a two-
photon irradiation experiment using a Nd:YAG laser pumped dye laser at 605 nm radiation (66%
conversion). Full conversion was attained using powdered 3 after 43 h. A similar conversion was
noted between the single crystal and powder forms (59% and 66% conversion after 5 h irradiation).
1
Irradiation of monothiophos-dba in solution using a 325 nm laser, and monitoring by H NMR
spectroscopy, showed that no cycloaddition occurred, but exclusive photochemically-induced
trans A cis isomerisation of the alkene bond connecting the unsubstituted aryl group (cis A trans
isomerisation in these electron-deficient and polarised CLC bonds occurs rapidly by thermal
equilibration).
exceptions, suggesting that not only is the spatial relationship
Introduction
between CLC bonds important, but also the size of the reaction
cavity, the nature and strength of intermolecular interactions5
Solid-state chemical reactions may proceed with a high degree of
chemo-, regio and stereo- selectivity, due to the rigidly-ordered
and the ability of the lattice to tolerate long-range molecular
arrangement of reactant molecules. The high level of control
rearrangements.6
achievable in such systems makes them ideal candidates for a
Since the 1960s several techniques have been used to study the
number of potential applications1 and in particular solid-state
mechanism of solid-state reactions including single crystal X-ray
photochemical reactions, such as [2+2] cycloadditions, have
crystallography,4 solid-state NMR spectroscopy,7 X-ray powder
attracted much interest. Solid-state [2+2] reactions were first
diffraction,8 and atomic force microscopy.9 In all cases the [2+2]
studied by Schmidt et al. in the 1960s,2 who proposed the
cycloaddition of cinnamic acid has been used as a benchmark.
‘‘topochemical principle’’ to determine the outcome of
a
The combination of these complementary techniques has offered
new mechanistic insight and provided further explanation for the
outcome of the reaction.10
reaction.3 More specifically, Schmidt et al. suggested that a
[2+2] cycloaddition is most likely to occur when CLC bonds are
˚
both co-planar and less than 4.2 A apart. While the majority of
During ongoing studies with thiophosphino-alkene ‘E,E-
dibenzylidene acetone’ (dba) ligands by Fairlamb and co-workers,11
a unique single-crystal-to-single-crystal [2+2] cycloaddition of
the dba backbone in ligand 3 was found to occur (vide infra). It
is interesting to note that the parent dba compound does not
crystallise in a geometry that favours a solid-state [2+2] cycloaddi-
tion,12 although it does react in solution.3,13
reactions obey these principles, many exceptions have also been
reported.4 Several research groups have sought to explain these
aDepartment of Chemistry, University of York, Heslington, York, YO10
5DD, UK. E-mail: ian.fairlamb@york.ac.uk; Fax: +44 (0)1904 322516;
Tel: +44 (0)1904 324091
bDepartment of Chemistry, Durham University, South Road, Durham,
DH1 3LE, UK. E-mail: h.a.sparkes@durham.ac.uk
cDepartment of Chemistry, University of Bath, BATH BA2 7AY, UK.
E-mail: p.r.raithby@bath.ac.uk; Fax: +44 (0)1225 386231;
Tel: +44 (0)1225 383183
A symmetrical chlorinated dba undergoes a solid-state [2+2]
cycloaddition involving both CLC bonds to give an exo-exo-
tricyclo[6.2.0.03,6]-decane ring system (Scheme 1, reaction [a]).14
It is postulated that remote ClNNNCl interactions (and/or p-inter-
actions) assist CLC double bond alignment into a viable
geometry for favourable cycloaddition. Structurally-related dba
compounds,15 such as conformationally restricted 2,5-dibenzyli-
denecyclopent-3-ene-1-one {the 1,4-dien-3-one moitey is locked
dDiamond Light Source Ltd, Diamond House, Harwell Science and
Innovation Campus, Didcot, Oxfordshire OX11 0DE, UK
{ Electronic supplementary information (ESI) available: Details of
synthetic chemistry, all X-ray diffraction, laser and NMR spectroscopic
experimental studies. CCDC reference numbers 874829–874853. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/c2ce25480h
5564 | CrystEngComm, 2012, 14, 5564–5571
This journal is ß The Royal Society of Chemistry 2012

View Online
Scheme 1 Exemplar reactions: ‘dba’ solid-state [2+2] cycloadditions.
in a s-cis,s-cis conformation, with non-parallel CLC double
bonds (3.92 A) between molecules, Scheme 1, reaction [b]} and
penta-1,4-diene-3-one-1,5-dicarboxylate16 have been reported to
Fig. 1 X-ray structure of 3 showing close double bonds. Ellipsoids
shown at 50%. Hydrogen atoms removed for clarity.
˚
undergo [2+2] cycloaddition.
Biradha and Santra also showed that 1,5-bis(4-pyridyl)-1,4-
pentadiene-3-one does not crystallise in the required geometry
for both CLC double bonds to react, but a single [2+2]
cycloaddition was found to occur.17A 5-methoxyresorcinol and
phloroglucinol co-crystal leads to a successful [2+2] cycloaddi-
tion involving both CLC double bonds. Hydrogen-bonding to
the pyridyl ring systems align the CLC double bonds favourably
for [2+2] cycloaddition. Dba has also been shown to undergo
solid-state photochemical cycloadditions when co-crystallised
with additives such as UO2Cl2 and SnCl4.18
In this paper, using X-ray diffraction and NMR spectroscopic
methods, a novel unsymmetrical dibenzylidene acetone (mono-
thiophos-dba, compound 3), containing two chemically different
alkene moieties, is investigated for its ability to undergo a
selective photochemically-induced [2+2] cycloaddition transfor-
mation in the solid-state (single crystal and powder).
Fig. 2 Crystal packing of 3 viewed along the a axis of the unit cell.
Black atoms indicate reacting alkene atoms, and dashed lines indicate the
˚
shortest distance between alkenes (3.575 A).
Results and Discussion
Table 1 Unit cell parameters, before irradiation and after 18 h
Time, h
0 (3)
18 (3 and 4)
Compound 3 was synthesised by a Wittig reaction of 2-
(diphenylthiophosphino)benzaldehyde 1 and ylide 2 (Scheme 2).
Single crystals of 3 were obtained by layering a concentrated
solution of 3 in CH2Cl2 with Et2O in the dark. In the single crystal
X-ray structure (Fig. 1, Fig. 2, Table 1) nearest-neighbour
molecules of 3 are positioned across an inversion centre,
generating a ‘‘head-to-tail’’ arrangement with even the most
˚
a/A
9.5167(3)
17.0830(4)
14.5809(6)
90
101.961(3)
90
9.3801(2)
17.2580(2)
14.7370(3)
90
102.273(2)
90
˚
b/A
˚
c/A
a (u)
b (u)
c (u)
3
˚
V/A
2319.0(2)
P21/n
2331.14(7)
P21/n
Space Group
˚
hindered CLC bonds aligned parallel to each other and 3.575 A
apart. This geometry fulfils Schmidt’s criteria for a topochemical
[2+2] cycloaddition. The 1,4-dien-3-one moiety is found in a s-
cis,s-trans conformation (the CLC bonds undergoing cycloaddi-
tion are those adopting the s-cis conformation).
Solid-state UV-visible spectroscopy indicates that the lmax of 3
is 289 nm with a long tail of absorbance across the visible region.
As such, it is unsurprising that natural light is found to mediate a
solid-state intermolecular [2+2] cycloaddition reaction in 3 (vide
infra).
Single crystals of 3 (grown in the dark) were subjected to
incandescent light for 2 h at 110 K, after which time no evidence
of a [2+2] cycloaddition was found by single crystal X-ray
diffraction. Leaving the same crystal in fluorescent light for 48 h
at ambient temperature it was found that a 31% conversion to
the cycloadduct 4 was achieved (Scheme 3).
Scheme 2 Synthesis of 3 by a Wittig reaction.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5564–5571 | 5565

View Online
Scheme 3 Transformations on irradiation of 3 (note: favourable s-cis,s-
trans 1,4-dien-3-one conformation shown).
Further investigation showed that two different transforma-
tions had occurred: firstly one CLC bond had undergone an
intermolecular [2+2] cycloaddition with its nearest neighbour to
form cyclobutane products similar to that shown in Scheme 1;
and secondly a conformational change of the remaining CLC
double bond had occurred from s-trans to s-cis (Scheme 3). It is
not clear whether one of these transformations must precede the
other, but likely there is a geometric requirement to allow the
cycloaddition reaction to take place, whilst ensuring that a
single-crystal’s integrity is maintained.
Fig. 3 Plot of conversion of 3 A 4 as a function of irradiation time.
Least-squares fitting of eqn (3) to data points gave a value of k = (2.3 ¡
0.1) 6 1024 s21 and Avrami exponent n = 0.55 ¡ 0.01 (R2 = 0.996).
Unfortunately, further irradiation caused the crystal to
disintegrate. Possible explanations for this include a phase-
change leading to a loss of single crystal integrity, or damage due
to the cumulative thermal stress induced by the experimental
conditions: while fluorescent irradiation was conducted at room
temperature, the crystal was flash-cooled to 110 K for single
crystal X-ray experiments.
nucleation sites throughout the solid, each with equal probability
of growth. The simple expression (1) assumes a complete (100%)
conversion between the two phases. For the incomplete
photoconversion of 3 to 4, a modified expression (eqn (2)) was
required to describe the progress of the reaction with time.20c
n
y~yf {(yf {yi)e{(kt)
(2)
In order to reduce the likelihood of crystal decomposition due
to thermal stress, an in situ experiment was conducted in which a
single crystal was irradiated with UV LED light whilst mounted
on the diffractometer, using a specially designed LED ring.19 The
ring positions six 400 nm LEDs c.a. 1 cm from the crystal in a
circle, helping to achieve uniform irradiation. The set-up also
allows the crystal to remain in position under the cryostream for
the duration of the experiment, providing greater control of the
temperature throughout.
where yf is the final conversion to 4 at the end of the reaction
and yi is the initial conversion. Subsituting in the values of yf =
69 and yi = 0 gives the final expression fitted in Fig. 3.
n
y~69{69e{(kt)
(3)
Regression analysis shows that eqn (3) fits well with the
observed data (R2 = 0.996). In line with previous studies, a non-
In the first instance, the progress of the cycloaddition reaction
in a single crystal was monitored by X-ray diffraction over time
at 180 K. At the start of the reaction excitation occurs quickly,
with 51% of the crystal converted from 3 to 4 within the first 2 h.
As time progresses the rate of reaction decreases rapidly,
producing a sigmoidal shape reaction profile (Fig. 3) and a
maximum conversion of 69% is achieved after 18 h irradiation.
The curve in Fig. 3 is fairly typical of this type of single crystal
photoreaction and similar profiles have been reported for a
number of other [2+2] systems.20 Due to their sigmoidal shape
profiles have been fitted with expressions derived from Johnson–
Mehl–Avrami–Kologoromov (JMAK) kinetics, which aim to
explain the progress of the photoreaction in terms of a
nucleation and growth model.21 The JMAK model describes
the kinetics of a phase transition in the solid-state using eqn (1):
integer Avrami exponent n = 0.55 ¡ 0.01 was determined.20
A
value of n , 1 suggests that nuclei are distributed homo-
geneously through the solid and that, although new nuclei
continue to form, the rate of nucleation decreases as the
photoreaction progresses.20d In addition the rate constant k =
(2.3 ¡ 0.1) 6 1024 s21 compares well with those obtained for
other solid-state [2+2] reactions.20a,b
After exposure to 18 h of 400 nm LED light at 180 K the
crystal continued to diffract well, showing that single-crystal-to-
single-crystal conversion of 3 to 4 is achievable at least to 69%.
Little change in the unit cell volume is observed following
conversion, with only an increase of 0.5% during the process
(Table 1). Following Schmidt’s requirement that a difference of
less than 4% is required for a topochemical transformation,{ the
69%-complete reaction described here may be considered
topotactic. Reactions that proceed to 100% conversion while
entirely retaining single crystal integrity remain rare,22 although
a few are reported in the literature.23
n
y~1{e{(kt)
(1)
where y is the fraction of the system converted at a time t, k is
the growth rate constant and n is the Avrami exponent. The
value of n indicates the dimensionality of growth from a nucleus;
n = 2 suggests one-dimensional growth, n = 3 two-dimensional
etc. A value of n = 1 represents a homogeneous dispersion of
{ It should be noted that this value is somewhat arbitrary. The use of
topotactic in the literature is varied. Differentiation is also made between
homogenous and heterogenous topotactic reactions. In these cases
heterogenous reactions are often not SCSC transformations.30
5566 | CrystEngComm, 2012, 14, 5564–5571
This journal is ß The Royal Society of Chemistry 2012

View Online
Table 2 Conversion of 3 A 4 with irradiation at different temperatures
T/K
Conversion, %a,b
298
220
decomposition (59)c
70
180
173
69
46
170
160
35
17
150
140 and below
11
negligible
a
b
c
Nearest whole %. after ca. 16 h irradiation. after 5 h, disordered.
In a second series of experiments the LED ring arrangement
was used to determine the maximum conversion achievable at
temperatures between 298 and 100 K, following y 16 h
irradiation (Table 2). At temperatures below 140 K the
photochemical reaction did not occur. The maximum conversion
level was found to increase on raising the temperature, with the
highest 70% conversion achieved at 220 K. At 298 K the crystal
decomposed after just 5 h irradiation, in accordance with
previous observations at ambient temperature. While it must
be noted that a number of possible reasons for the trend in
conversion with temperature exist, detailed analysis of the
resultant crystal structures suggest one explanation: while no
correlation between temperature and the distance between CLC
bond planes was found, it was observed that the lateral shift
between reacting CLC bonds increased on warming (Table 3,
Fig. 4). It may be inferred from this increase that more space
exists in the region of the CLC bond at higher temperatures,
allowing more freedom for the necessary atomic rearrangements
to take place and thus facilitating higher conversion. Conversely,
the observation of lower conversion with lower temperatures
might also suggest both a thermal and photochemical barrier to
the reaction, which must be overcome for the cycloaddition to
occur. From the data it is not possible to rule out the explanation
that conversion from 3 to 4 is simply reduced when there is less
available thermal energy to overcome this barrier.
Fig. 4 Lateral shift between CLC bonds: the shift is calculated as the
offset between centroid positions, taken only in the direction parallel to
the CLC planes.
fluence to promote the inherently inefficient two-photon
absorption process throughout the crystal26 whilst simulta-
neously avoiding crystal damage due to thermal shock.
Using a Nd:YAG laser pumped dye laser, 5 ns pulses of
605 nm radiation was obtained allowing us to probe the reaction
of a single-crystal with two-photon absorption at ambient
temperature. At the lowest laser intensity (0.5 mJ per pulse) no
conversion of 3 was observed by X-ray diffraction analysis of the
single crystal, nor any loss of the crystal quality (R1 = 4.65%
before and 4.61% after). At higher laser intensity (10 mJ per
pulse), the crystal fractured into smaller pieces. Analysis of one
piece showed that 66% conversion of the crystal was attained
after 25 min, the same as observed using the UV-LED light
source at 220 K after 16 h (400 nm).
It was hypothesised that the disintegration could be due to
irradiation approaching from one face of the crystal, leading to
strain on the crystal lattice resulting in fractures. Rotating the
crystal in the laser beam using a mechanical stirrer prevented
crystal degradation, however, only 25% conversion was attained
after 20 min.
In some photochemical [2+2] cycloaddition reactions, conver-
sion has been seen to increase through use of different types of
photoexcitation strategy, for example two-photon irradia-
tion.4d,24,25 However, to obtain high conversion by two-photon
irradiation two criteria must be met: the first is that the
cycloaddition product must not absorb strongly at the irradiat-
ing wavelength, otherwise the back-reaction (the retro-[2+2]
cycloaddition) will be favoured and the photostationary point
will not lie towards the product. The second criterion derives
from the conflicting requirements to ensure sufficiently high laser
There are two plausible explanations for the lower conversion:
i) The rotation of the crystal may not have been absolutely true
causing the crystal to move ‘in and out’ of the path of the laser;
ii) In some cases it has been seen that dimerisation occurs at
defect and dislocation sites leading to autocatalytic reactions.27
The strain from irradiation from one side of the crystal could
have increased the rate of the reaction by causing more defect
sites. However, in these cases mixed-product regioselectivities are
often found which we do not observe.
Table 3 Lateral shift distances at different temperatures (conv. of 3 A
4)
˚
A
T/K
Lateral shift,a
Overall, the two-photon excitation did not appear to increase
the extent of conversion, but the rate of conversion is much
faster than with the one-photon excitation, with 70% conversion
obtained after 25 min (75 ms irradiation), presumably due to the
increased intensity of the laser light compared to the UV-LEDs.
These crystallographic studies suggest that the crystal lattice of
3 is confining the conversion to 4 to ca. 70%. Further evidence of
220
180
173
170
160
150
a
1.655
1.604
1.549
1.362
1.358
1.339
Lateral shift between reacting CLC bonds (see also Fig. 4).
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5564–5571 | 5567

View Online
Fig. 6 13C CPMAS-NMR spectra of a) Compound 3, b) after 2.5 h
irradiation, c) after 5 h irradiation and d) after 43 h irradiation.
the peaks are due to the CLC carbons. Interrupted decoupling
measurements, to remove the CH/CH2 signals, show these to be
sp2 CH carbons. In solution NMR spectra the 13C signals
corresponding to the alkene carbons are seen either side of the
aromatic signals. The 13C signal at 121.5 ppm disappears on
irradiation, which is attributable to the loss of the CLC bond.
The carbonyl signals exhibit more complicated behaviour than
Fig. 5 X-ray structure of 4. Ellipsoids shown at 50%. Selected hydrogen
atoms removed for clarity. Selected atoms labelled. The inset shows the
C-C single bond distances within the cyclobutane ring.
you would expect from a simple ‘A A B’ reaction. The initial 13
C
signal at 188 ppm is accompanied by two further 13C signals (186
and 197 ppm) after 2.5 h irradiation. After 5 h the initial 13C
signal has disappeared and three other signals are present,
including the carbonyl 13C signal of the cycloaddition product 4
(196 ppm). The number of 13C signals indicate that other
intermediates are being observed (Scheme 2), or that poly-
morphism is occurring.
The 31P CPMAS-NMR spectroscopic analysis also proved
inconclusive. The non-irradiated material exhibits a 31P signal at
39.7 ppm with spinning sidebands (Fig. 7). On irradiation the 31P
signal broadened at first, before a new 31P signal was observed at
43 ppm following 5 h irradiation. After 48 h the original signal
disappeared, but with two overlapping 31P signals remaining,
one sharper (43.1 ppm) and one broader (41.6 ppm).
this is provided by the observation that in the amorphous state
the reaction goes to 100% conversion when left in ambient
conditions (and natural light) for 2–3 months, as determined by
1H NMR spectroscopy. Single needle-like crystals of 4, grown
from acetone and dichloromethane, allowed an X-ray structure
to also be determined (Fig. 5). The bond distances of the C–C
bonds in the cyclobutane rings are shown in the chemical
drawing given in the inset.
Powder studies (by X-ray Diffraction and NMR):
Powder X-ray diffraction and solid-state NMR spectroscopic
techniques allow reaction progress to be followed using 3 in
powder form (e.g. when the integrity of a single crystal is not
maintained).8 Powder X-ray diffraction also gives information
about the unit cell dimensions and any structural changes. In
addition, solid-state reactions can be monitored and any phase
changes observed.8 Solid-state NMR spectroscopy can also be
utilised to give information about different polymorphs, and
reaction intermediates.7
The single crystal studies revealed that two different transfor-
mations were occurring during irradiation. Solid-state NMR
experiments were undertaken to see if any further information
about the mechanism could be obtained. Samples for solid-state
NMR spectroscopic analysis were obtained by irradiation of a
powdered sample of 3 using UV LEDs (400 nm). Samples were
taken before irradiation, and after 2.5, 5 and 43 h. The material
was shaken every 30 min to ensure an even irradiation. The 13C
CPMAS-NMR spectra show the formation of the cyclobutane
ring with the appearance of two 13C signals between 30 and
60 ppm (Fig. 6).
It is not possible to assign the 13C signals in the aromatic
region unambiguously. However, the 13C signals either side of
Fig. 7 31P CPMAS-NMR spectra of a) Compound 3, b) after 2.5 h
irradiation, c) after 5 h irradiation and d) after 43 h irradiation.
5568 | CrystEngComm, 2012, 14, 5564–5571
This journal is ß The Royal Society of Chemistry 2012

View Online
product observed was 4. Therefore it is likely that the different
products are species that freely exchange in solution, i.e. the
different conformers around the single bond in the enone
(Scheme 4). After 5 h 66% of the starting material had been
converted to 4. This matches the results obtained from the single-
crystal X-ray diffraction experiments which gave y 59%
conversion after 5 h at 298 K.
The [2+2] cycloaddition was also followed by powder X-ray
diffraction, using the same irradiation source as for the solid-
state NMR experiments. A simulation of the powder X-ray
diffraction pattern from the single crystal structure of 3 and the
crystal at 81% conversion28 indicated that the changes in the
powder diffraction pattern would be observable. This was indeed
found to be the case; the simulated and observed diffraction
patterns match quite well, suggesting that the reaction proceeds
similarly using powdered 3, as in the single crystal. The majority
of the changes occurred within the first 10 h, and the crystal-
linitiy is reduced due to strain broadening in 4 (Fig. 9). This was
expected as the breakdown of long-range order within the
crystals has been seen previously in the single crystal experiments
upon irradiation.
Fig. 8 Overlay of 31P CPMAS NMR spectra of the irradiation of 3.
Further analysis of the powder XRD data using the Rietveld
method29 enabled refinement of the lattice parameters for both
phases with respect to time and the extraction of the %
conversion in the powdered sample (Fig. 10). These data show
that after 20 h irradiation there is 80% conversion. This is in
contrast to the single crystal irradiation also carried out at 298 K
(Table 3) where the single crystal decomposed. It was very
difficult to model the strain broadening in 4 and hence
refinements against the powder data were carried out with a 2h
maximum of 20u to give accurate cell parameters. Refined lattice
parameters and full details of these analyses are given in the
supporting information.
Deconvolution of the bandshape gives 32% at 43.1 ppm and
68% at 41.6 ppm, and closer inspection of the 31P NMR
spectrum shows that there are two singlets of different intensities,
one of which is partially obscured by another broad signal
(Fig. 8).
It is unclear whether the observed spectrum is due to three
separate products or two products, one of which has different
phosphorus chemical environments. Solution-state NMR spec-
troscopic analysis of the solid-state samples at 43 h confirmed
that the reaction had reached 100% conversion, and that the only
During our work with compound 3 no cycloaddition products
were observed in solution in the presence of light. For example,
NMR samples of 3 left for up to a week in the laboratory
(exposed to fluorescent light and direct sunlight) showed no sign
of any cycloaddition reaction.
Irradiation of 3 in CD2Cl2 (0.07 and 0.25 M) using a 325 nm
laser, followed by in situ NMR spectroscopy, also revealed that
no [2+2] cycloaddition occurred even in a saturated solution
(y 0.25 M, Fig. 11). Instead the sample underwent partial
Fig. 9 PXRD patterns of 3 before irradiation (bottom) and after 16 h
Scheme 4 Possible intermediates in the solid-state [2+2] cycloaddition
(top) (300 K).
reaction.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5564–5571 | 5569

View Online
electron-deficient alkene moieties, undergoes a selective photo-
chemically-induced [2+2] cycloaddition in the solid-state. X-ray
diffraction methods have been used to monitor cycloaddition
product 4 conversions. Kinetic analysis of the single crystal
experiments allowed the rate constant for the cycloaddition
reaction to be determined {k = (2.3 ¡ 0.1) 6 1024 s21}, with a
69% conversion attained. This result compares well with related
[2+2] cycloaddition reactions. In addition,
a two-photon
irradiation experiment using a Nd:YAG laser pumped dye laser
at 605 nm radiation also gave the cycloadduct 4 with 66%
conversion. Powder XRD studies showed that full conversion
occurred after 43 h. It is interesting to note the similar
conversion using either single crystals and powder-forms of 3
(with 59% and 66% conversion to 4 after 5 h irradiation). The
[2+2] cycloaddition reaction is unique to the solid-state as
demonstrated by the irradiation of 3 in solution using a 325 nm
laser (monitored by 1H NMR spectroscopy), which showed only
trans A cis isomerisation of the alkene bond connecting the
unsubstituted aryl group.
Fig. 10 Conversion rate of 3 to 4 from Rietveld analysis of the PXRD
patterns. Solution studies by laser-coupled NMR spectroscopy.
Acknowledgements
AGJ was funded by an EPSRC DTA studentship. IJSF thanks
the Royal Society for support (University Research Fellow).
LEH is grateful to the University of Bath for funding, and MRW
and PRR are grateful to the EPSRC for support (and for a
Senior Fellowship to PRR). HAS would like to thank the
Leverhulme Trust and University of Durham for financial
support. We are grateful to the Diamond Light Source Ltd for
the award of beamtime. Dr. David Apperley (Durham Solid-
State NMR service) is thanked for his advice on the deconvolu-
tion of 31P NMR spectra.
Fig. 11 a) Compound 3 before irradiation, b) after 10 h irradiation at
325 nm, c) after 48 h in the dark. * Represents the alkenyl protons whose
13C peaks have C-P coupling.
References
trans-cis isomerisation (approx. 9% by 1H NMR spectroscopy
after 10 h) of one of the CLC double bonds, shown by the
reduction in the 3JHH from 15.8 to 12.7 Hz. Once the sample was
placed back in the dark for 48 h the isomerisation reversed
(thermally equilibrated) to give the original starting compound,
which is expected as this is a highly polarised and electron-
deficient CLC bond {Fig. 11, c)}. 1H-13C HSQC NMR
experiments reveal that the alkenyl protons exhibiting a change
in coupling constant were associated with the alkenyl carbons
without 13C–31P coupling. Therefore, it is the alkene furthest
away from the phosphine sulfide that undergoes CLC bond
isomerisation (Scheme 5).
1 (a) F. Toda, Acc. Chem. Res., 1995, 28, 480–486; (b) K. Tanaka and
F. Toda, Chem. Rev., 2000, 100, 1025–1074; (c) G. Kaupp, J.
Schmeyers and J. Boy, Chemosphere, 2001, 43, 55–61; (d) G. W. V.
Cave, C. L. Raston and J. L. Scott, Chem. Commun., 2001,
2159–2169.
2 (a) M. D. Cohen and G. M. J. Schmidt, J. Chem. Soc., 1964,
1996–2000; (b) M. D. Cohen, G. M. J. Schmidt and F. I. Sonntag,
J. Chem. Soc., 1964, 2000–2013; (c) G. M. J. Schmidt, J. Chem. Soc.,
1964, 2014–2021.
3 G. M. J. Schmidt, Pure Appl. Chem., 1971, 27, 647–678.
4 (a) J. M. Thomas and J. O. Williams, Chem. Commun., 1967,
432–433; (b) V. Ramamurthy and K. Venkatesan, Chem. Rev., 1987,
87, 433–481; (c) G. Kaupp, Advances in Photochemistry, 1995, 19,
119–177; (d) M. F. Mahon, P. R. Raithby and H. A. Sparkes,
CrystEngComm, 2008, 10, 573–576.
5 (a) M. D. Cohen, Angew. Chem., Int. Ed. Engl., 1975, 14, 386–393; (b)
A. Gavezzotti, J. Am. Chem. Soc., 1983, 105, 5220–5225.
6 (a) G. S. Murthy, P. Arjunan, K. Venkatesan and V. Ramamurthy,
Tetrahedron, 1987, 43, 1225–1240; (b) G. Kaupp, Int. J. Photoenergy,
2001, 3, 55–62; (c) G. Kaupp, Comprehensive Supramolecular
Chemistry, 1996, 8, 381–423.
Conclusions
In this paper we have shown that a novel unsymmetrical
dibenzylidene acetone 3, containing two chemically different
7 (a) R. C. Nieuwendaal, M. Bertmer and S. E. Hayes, J. Phys. Chem.
B, 2008, 112, 12920–12926; (b) M. Khan, G. Brunklaus, V.
Enkelmann and H.-W. Spiess, J. Am. Chem. Soc., 2008, 130,
1741–1748; (c) S. E. Lister, A. Soleilhavoup, R. L. Withers, P.
Hodgkinson and J. S. O. Evans, Inorg. Chem., 2010, 49, 2290–2301.
8 (a) J. D. Dunitz, K. D. M. Harris, R. L. Johnston, B. M. Kariuki,
E. J. MacLean, K. Psallidas, W. B. Schweizer and R. R. Tykwinski,
J. Am. Chem. Soc., 1998, 120, 13274–13275; (b) S. E. Lister, V. J.
Rixom and J. S. O. Evans, Chem. Mater., 2010, 22, 5279–5289.
Scheme 5 Trans-cis isomerisation on irradiation of 3 in solution at
325 nm.
5570 | CrystEngComm, 2012, 14, 5564–5571
This journal is ß The Royal Society of Chemistry 2012

View Online
9 (a) G. Kaupp, Int. J. Photoenergy, 2001, 3, 55–62; (b) G. Kaupp,
Comprehensive Supramolecular Chemistry, 1996, 8, 381–423.
10 I. Abdelmoty, V. Buchholz, L. Di, C. Guo, K. Kowitz, V.
Enkelmann, G. Wegner and B. M. Foxman, Cryst. Growth Des.,
2005, 5, 2210–2217.
21 (a) M. Avrami, J. Chem. Phys., 1939, 7, 1103–1112; (b) M. Avrami,
J. Chem. Phys., 1940, 8, 212–224.
22 (a) I. G. Georgiev and L. R. MacGillivray, Chem. Soc. Rev., 2007, 36,
1239–1248; (b) W. L. Leong and J. J. Vittal, Chem. Rev., 2011, 111,
688–764.
11 (a) A. G. Jarvis, A. C. Whitwood and I. J. S. Fairlamb, Dalton
Trans., 2011, 40, 3695–3702; (b) For a review on the use of dba-Z
ligands in Pd catalysis, see: I. J. S. Fairlamb, Org. Biomol. Chem.,
2008, 6, 3645; (c) A. G. Jarvis, PhD thesis (University of York), 2011.
12 It should be noted that on irradiation of single crystals of dba they
lose there diffracting properties:I. Turowska-Tyrk, Chem. Phys.,
2003, 288, 241–247.
13 C. W. Shoppee, Y.-S. Wang, S. Sternhell and G. C. Brophy, J. Chem.
Soc., Perkin Trans. 1, 1976, 1880–1886.
14 B. S. Green and G. M. J. Schmidt, Tetrahedron Lett., 1970, 11,
4249–4252.
15 G. R. Desiraju, J. Bernstein, K. Kishan, K. V. R. Sarma and J. A. R. P.,
Tetrahedron Lett., 1989, 30, 3029–3032.
16 N. E. Rowland, F. Sondheimer, G. A. Bullock, E. LeGoff and K.
Grohmann, Tetrahedron Lett., 1970, 11, 4769–4772.
17 R. Santra and K. Biradha, CrystEngComm, 2011, 13, 3246–3257.
18 (a) N. W. Alcock, N. Herron, T. J. Kemp and C. W. Shoppee,
J. Chem. Soc., Chem. Commun., 1975, 785–786; (b) N. W. Alcock, P.
de Meester and T. J. Kemp, J. Chem. Soc., Perkin Trans. 2, 1979,
921–926.
19 S. K. Brayshaw, J. W. Knight, P. R. Raithby, T. L. Savarese, S.
Schiffers, S. J. Teat, J. E. Warren and M. R. Warren, J. Appl.
Crystallogr., 2010, 43, 337–340.
23 (a) A. M. P. Peedikakkal and J. J. Vittal, Inorg. Chem., 2010, 49,
10–12; (b) A. M. P. Peedikakkal, C. S. Y. Peh, L. L. Koh and J. J.
Vittal, Inorg. Chem., 2010, 49, 6775–6777; (c) J. A. K. Howard, M. F.
Mahon, P. R. Raithby and H. A. Sparkes, Acta Crystallogr., Sect. B:
Struct. Sci., 2009, 65, 230–237; (d) Y. Sonoda, M. Goto, S. Tsuzuki,
H. Akiyama and N. Tamaoki, J. Fluorine Chem., 2009, 130, 151–157;
(e) M. Nagarathinam and J. J. Vittal, Chem. Commun., 2008,
438–440; (f) J. H. Ho, T. I. Ho, T. H. Chen and Y. L. Chow,
J. Photochem. Photobiol., A, 2001, 138, 111–122.
24 J. Harada, H. Uekusa and Y. Ohashi, J. Am. Chem. Soc., 1999, 121,
5809–5810.
25 Two-photon refers to the simultaneous absorption of two photons by
the compound, rather than two sequential one-photon absorption
processes.
26 S. J. Mulligan and B. A. MacVicar, Modern Research and Educational
Topics in Microscopy, Eds A. Me´ndez-Vilas and J. D´ıaz, 2007, 2, 881–
889, Formatex;Spain. Retrieved online from http://www.formatex.-
org on 9th January 2012.
27 (a) J.-P. Desvergne, J. M. Thomas, J. O. Williams and H. Bouas-
Laurent, J. Chem. Soc., Perkin Trans. 2, 1974, 363–368; (b) S.
Ramdas, W. Jones, J. M. Thomas and J. -P. Desvergne, Chem. Phys.
Lett., 1978, 57, 468–470.
28 This crystal was grown in low daylight over 3 weeks.
29 H. M. Rietveld, J. Appl. Crystallogr., 1969, 2, 65–71.
30 (a) K. Novak, V. Enkelmann, G. Wegner and K. B. Wagener, Angew.
Chem., Int. Ed. Engl., 1993, 32, 1614–1616; (b) I. Abdelmoty, V.
Buchholz, L. Di, C. Guo, K. Kowitz, V. Enkelmann, G. Wegner and
B. M. Foxman, Cryst. Growth Des., 2005, 5, 2210–2217; (c) I. Halasz,
Cryst. Growth Des., 2010, 10, 2817–2823.
20 (a) J. B. Benedict and P. Coppens, J. Phys. Chem. A, 2009, 113,
3116–3120; (b) D.-K. Cao, T. V. Sreevidya, M. Botoshansky, G.
Golden, J. Brown Benedict and M. Kaftory, J. Phys. Chem. A, 2010,
114, 7377–7381; (c) R. More´, G. Busse, J. Hallmann, C. Paulmann,
M. Scholz and S. Techert, J. Phys. Chem. C, 2010, 114, 4142–4148;
(d) M. Bertmer, R. C. Nieuwendaal, A. B. Barnes and S. E. Hayes,
J. Phys. Chem. B, 2006, 110, 6270–6273.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5564–5571 | 5571