Following a photoinduced reconstructive phase transformation and its influence on the crystal integrity: Powder diffraction and theoretical study

Article abstract of DOI:10.1002/anie.201402515

In the course of solid-state photoreactions, a single crystal (SC) of the reactant can be transformed into an SC of the product or it can lose crystallinity and become amorphous. In-between these two scenarios exist the reconstructive phase transformations, where upon irradiation, the reactant SC becomes a powder or an SC with increased mosaicity. We present a detailed description of reconstructive photodimerization, where the structural changes are directly correlated with the disintegration process. The kinetics of the reaction is explained by two kinetic regimes, forming an autocatalytic autoinhibition photoreaction set with high quantum yield. In addition, the photoreaction pathways were studied theoretically. Changing phases: In the course of solid-state photoreactions, the single crystal of a reactant can be transformed into a single crystal of the product or it can become amorphous. Between these scenarios exist the reconstructive phase transformations, where the single crystal becomes a powder. A detailed description of the latter is given.

Full text of DOI:10.1002/anie.201402515

Angewandte
Chemie
DOI: 10.1002/anie.201402515
Photoreconstructions
Following a Photoinduced Reconstructive Phase Transformation and
its Influence on the Crystal Integrity: Powder Diffraction and
Theoretical Study**
ˇ
ˇ
Tomce Runcevski,* Marina Blanco-Lomas, Marco Marazzi, Marcos Cejuela, Diego Sampedro,*
and Robert E. Dinnebier
Abstract: In the course of solid-state photoreactions, a single
crystal (SC) of the reactant can be transformed into an SC of
the product or it can lose crystallinity and become amorphous.
In-between these two scenarios exist the reconstructive phase
transformations, where upon irradiation, the reactant SC
becomes a powder or an SC with increased mosaicity. We
present a detailed description of reconstructive photodimeri-
zation, where the structural changes are directly correlated with
the disintegration process. The kinetics of the reaction is
explained by two kinetic regimes, forming an autocatalytic
autoinhibition photoreaction set with high quantum yield. In
addition, the photoreaction pathways were studied theoret-
ically.
lattice points to coincide with the lattice points of the
reactant.[2] Most of these reactions are reversible and proceed
in a single-crystal-to-single-crystal (SCSC) manner, which
represents a promising platform for the design of photo-
switches and molecular motors.[3] There are examples, how-
ever, of irreversible photoamorphization and/or melting
reactions.[4] In between these two reaction types exist the
so-called reconstructive phase transformations (RPTs).[4b]
When the geometrical movements during the reaction are
significant (in the sense of different crystal packings, accom-
modated in different unit cells), the crystal cannot withstand
the stress and therefore disintegrates.[5] The crystal disinte-
gration and irreversibility of the reaction hamper the
perspective applications of RPT systems in molecular machi-
neries. Contrary, in terms of synthetic photochemistry,
irreversible high-yield RPT reactions are very favorable.
Thus, tuning the reaction fashion stands at the frontiers of
modern crystal engineering. The borders between the realms
of SCSC and RPT reactions are not impassable, for instance
dilution of the crystal packing with single water molecules can
change the type of the reaction from RPT to SCSC.[5b,c]
In order to better understand RPTs, knowledge on the
structure before and after the photoreaction is crucial.
Generally, small organic molecules build SCs suitable for
single-crystal X-ray diffraction (SCXRD). In RPTs, however,
the crystal disintegrates upon irradiation, thwarting the
structure solution of the products. For example, it was
recently reported that crystals of molecules resembling the
green fluorescent protein (GFP) chromophore, disintegrate
when they dimerize under exposure to UV light.[5a] Recrys-
tallization of stable photoproducts, followed by ex situ
analysis is not always applicable, as it often leads to different
polymorphs. There are reported strategies for favoring SCSC
reactions and enabling the solution of the crystal structures of
the photoproducts (for example using reduced flux of
irradiation, sometimes at low temperature). Those photo-
products generally have a different crystal structure com-
pared to the photoproducts obtained by RPT. The former
products are obtained through a topotactic reaction and have
similar unit cells as the reactant, while the latter have
significantly different unit cells (and that difference effec-
tively causes the crystal disintegration). In general, disinte-
gration of a crystal means reduction to sizes not suitable for
the current SC diffractometers and/or a substantial increase
of the mosaicity. These serious problems for SCXRD can be
overcome by using X-ray powder diffraction (XRPD). Beside
the advances in instrumentation and software, crystal-struc-
T
he first report on photoinduced reactions of organic crystals
dates back to the 19th century; the topochemical principle was
postulated based on the intensive research on solid-state
photoreactions that followed.[1] This very intuitive and
exceedingly useful concept employs the necessity of minimum
geometrical changes in constrained environments such as the
crystal packing. Topotactic reactions are those in which the
product forms in an oriented fashion, requiring at least some
ˇ
[*] T. Runcevski, Prof. R. E. Dinnebier
Max-Planck-Institute for Solid State Research
Heisenbergstrasse 1, 70569 Stuttgart (Germany)
E-mail: t.runcevski@fkf.mpg.de
Dr. M. Blanco-Lomas, M. Cejuela, Dr. D. Sampedro
Departamento de Quꢀmica, Universidad de La Rioja
Madre de Dios, 54, 26006 La Rioja (Spain)
E-mail: diego.sampedro@unirioja.es
Dr. M. Marazzi
Department of Theoretical Chemical Biology
Institute of Physical Chemistry, KIT
Kaiserstrasse 12, 76131 Karlsruhe (Germany)
and
Departamento de Quꢀmica Analꢀtica, Quꢀmica Fꢀsica e Ingenierꢀa
Quꢀmica, Universidad de Alcalꢁ
28871 Alcalꢁ de Henares (Spain)
[**] D.S. acknowledges the Spanish Ministerio de Ciencia e Innovaciꢂn
(MICINN)/Fondos Europeos para el Desarrollo Regional (FEDER)
(CTQ2011-24800). M.B.-L. thanks the Spanish Ministerio de
Educaciꢂn y Ciencia (MEC) for a grant. M.M. acknowledges the
UAH (Universidad de Alcalꢁ) and the Alexander von Humboldt
Foundation. T.R. thanks the IMPRS-CMS. Andrea Knçller is
acknowledged for recording the SEM images and Frank Adams for
designing the photochamber.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201402515.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
ture solution by XRPD appears to be a niche technique. It
should be noted that while the deviation of precise intra-
molecular parameters by XRPD is hampered by its intrinsic
technical problems,[6] intermolecular features are certainly
much better defined, as they depend on collective properties
of the molecular models, and on the lattice symmetry and
periodicity, which in fact are more relevant to the study of
RPT. Moreover, XRPD, unlike SCXRD, provides informa-
tion on the average structure of the bulk, and provides
additional information, such as domain size, the presence of
defects, faults, and strain, among other.[6]
Benzylidene oxazolones are photoactive compounds that,
surprisingly, have been scarcely explored in terms of photo-
reactivity.[7] A member of this family is the (Z)-4-(4-bromo-
benzylidene)-2-methyloxazol-5(4H)-one system (Z-1). As
evident from the UV spectrum (Figure S1 in the Supporting
Information), a range of wavelengths can be used to promote
its photoreactivity. When irradiated in solution under a variety
of conditions, Z–E photoisomerization was the main photo-
reaction (Scheme 1a). However, when placed on a plate and
Figure 1. SEM images of crystals of Z-1 a) before irradiation, and after
b) 3 days and c) 5 days of irradiation. d) Respective XRPD patterns.
ples through the iterative use of singular value decomposition,
followed by a Pawley fitting and a simulating annealing
approach to solve the crystal structures. The structure models
were refined by the Rietveld method (for details see the
Supporting Information). Z-1 crystallizes in an orthorhombic
unit cell (space group Pbc21) with two molecules in the
asymmetric unit (denoted A and B). The main difference
between the symmetrically independent molecules is the
torsion angle between both rings (N-C4-C5-C6 in A = 3.848
and in B = ꢀ16.678, see CIF in Supporting Information). The
two molecules are closely head-to-tail packed, where the
distances between the double bond carbon atoms are CA4–
CB5 = 3.375 ꢀ and CA5–CA4 = 3.625 ꢀ, making the pairs
aligned parallel to the a-axis (Figure 2a). The Hirshfeld
surfaces around the C4 and C5 atoms of both molecules show
Scheme 1. a) Solution photoisomerization and b) solid-state photodi-
merization.
irradiated at 350 nm using a photoreactor, no isomerization
was detected by 1H NMR spectroscopy. Instead, dimer 2,
a product of a [2+2] photocycloaddition, was observed
(Scheme 1b). After five days of irradiation, 2 was found in
the reaction crude in a 90% yield. The resulting crude was
purified by chromatography on silica gel and the product was
confirmed to be 2 by 1H and 13C NMR spectroscopy and
HRMS (Figures S2 and S3 in the Supporting Information). To
check the stability of 2, several trials were conducted with
samples that were obtained directly from the reaction crude
after irradiation, and no decomposition of 2 was found (see
the Supporting Information).
Crystals of Z-1 were monitored on the course of the
irreversible photoreaction with scanning transmission micros-
copy (SEM) and XRPD. Figure 1 shows the irreversible
crystal disintegration as the reaction proceeds, followed by
changes of the diffraction patterns. Accordingly, the photo-
reaction outlined in Scheme 1b turns out to be a perfect
model for studying of RPT reaction by correlating the
macroscopic effects on the crystals with molecular move-
ments in the structure during the reaction.
ꢀ
intensive contacts, which together with the C C distances
smaller than 4.2 ꢀ (the Schmidt criterion for photocycload-
dition)[1] explain the photoreactivity of Z-1 and formation of
dimer 2. The dimer crystallizes in the low symmetry, triclinic
ꢀ
P1 space group with half a molecule (one respective
monomer) in the asymmetric unit. Figure 2b shows the
necessary molecular movements of the A and B monomers
in order to form the dimer molecule. The five-membered
rings perform remarkable bending, whereas the heavier six-
membered ring is slightly rotated.
In order to accommodate the new geometry of the dimer
product, the crystal packing significantly changes. In the
reactant, the monomers are packed forming two layers
(depicted in green and blue, Figure 2d), whereas the dimers
in the product become ordered in one layer (Figure 2e). For
this transformation to happen, significant molecular move-
ments (such as in-layer rotation and shifting) must occur.
Those movements generate immense internal stress which,
when released, triggers disintegration of the crystal. The
result of this disintegration is visible in Figure 1. We
Attempts to grow crystals of Z-1 suitable for SCXRD
resulted in microcrystalline powders. The in situ synthesis of 2
led to an even more “powdered” sample. Therefore, the
crystal structures were determined using high-resolution
laboratory XRPD. Indexing was performed from first princi-
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Figure 3. a) 2D projection of the observed X-ray intensity as a function
of diffraction angle and time. c) Sharp–Hancock plot of the reaction.
more favorable crystal packing. This reconstruction transition
is assumed to happen during the second regime in the kinetics
plot; ksecond = 1.68(4) ꢁ 10ꢀ5 minꢀ1 and nsecond = 0.33(6). The
dimensionality of the process has a negative value, which can
be described by including a negative autocatalytic step,
termed autoinhibition.[10] We assume that the disintegration
of the crystal acts as an inhibition step to the photoreaction by
cracking and reducing the domains, thus influencing the
nucleation rate and growth of the dimer phase in the reacting
matrix of the monomer phase. However, the fragmentation of
the particles enables easier penetration of photons inside the
reacting crystal, making the bulk material accessible to
irradiation. Thus, it can be suggested that the fragmentation
leads to a fine balance between kinetics inhibition and
fragmentation-assisted bulk phototransformation. The inter-
play between the homogenously reacting first regime and
fragmentation-effect-dominated second regime leads to
unique photoreaction set with (theoretically) 100% yield.
In order to gain insights into the processes before the
SCSC and RPT reactions, a series of theoretical calculations
on the photoreaction were performed (Figure 4). The exper-
imental absorption spectrum of Z-1 in the solid state was used
to validate the model by time-dependent density functional
theory (TD-DFT) at the CAM-B3LYP/6-31 + G* level.[9]
Calculations were performed on the whole unit cell with the
Gaussian 09 program package (see the Supporting Informa-
tion for the reference). For the S0!S1 and S0!S2 transitions,
more than one electronic configuration is involved per
electronic state (Table S2 in in the Supporting Information).
In the case of the optically bright S2 state, most of the
configurations describe an excitation localized on four head-
to-tail facing molecules, that is, two dimers, with some
contributions from single-monomer excitations (Figure S7 in
the Supporting Information).
Figure 2. a) Head-to-tail packed monomers along the a-axis and the
Hirshfeld surface around the reactive bonds. Hirshfeld surfaces of
b) the monomers in Z-1, and c) the dimer in 2. Crystal packing
diagrams of d) Z-1 and e) 2.
calculated the domain size as the reaction proceeds and found
that the domain size of the product particles reduces by
approximately 50% compared to the domain size of the
reactant (see the Supporting Information).
To explain the nature of the photoreaction and crystal
disintegration, information on the nucleation and the sub-
sequent growth of the product phase is needed. There are
many cases where the kinetics of such reactions is explained
using the JMAK model.[8] The model is centered around the
equation ln(ꢀln(1ꢀx)) = nln(t) + nln(k). The Sharp–Han-
cock plot is representable for such a model and gives a straight
line with a slope n and intercept nln(k), where t is the time to
collapse, x(t) is the evolved fraction of the product phase, k
(sꢀ1) is the reaction rate constant, and n is the Avrami
exponent. The rate constant k depends on the nucleation and
growth rates and is also sensitive to temperature. The Avrami
exponent represents the order of the reaction and describes
the dimensionality of the growth of the product phase. It
relates to the dimensionality (dim) of the growth of the dimer
with the equation dim = nꢀ1. Interestingly, the Sharp–Han-
cock plot of the collected data of the Z-1 dimerization
exhibits two different growth regimes (Figure 3).
The first regime explains the kinetic behavior at the
beginning of the reaction. Inspecting the slope and intercept
of the plot gives kfirst = 5.9(5) ꢁ 10ꢀ4 minꢀ1 and nfirst = 1.1(2).
These values indicate a fast reaction in connection with
homogeneous nucleation.[8] We believe that during this
regime, photocycloaddition takes place in an SCSC manner.
With minimal molecular movements, the monomers dimerize.
Upon further irradiation, the product is formed and accom-
modated in the unit cell of the reactant, thus forming a solid-
solution mixture. The molecular geometries of the reactant
and the product (Figure 2b and c) significantly differ.
Accordingly, soon after its formation, the product molecules
collectively move, accommodating in a thermodynamically
This delocalization over the unit cell is in agreement with
the experimental results, considering that 2 is promoted by
a dimer initially arranged in a head-to-tail conformation. In
order to reliably take into account the multiconfigurational
nature of the observed electronic excitations, the MS-
CASPT2//SA-CASSCF methodology[10] was applied with an
affordable ANO-S basis set to one of the solved dimer
structures. The computed absorption spectrum using the
MOLCAS 7.6 package[11] (see the Supporting Information) is
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
initial structure (i.e. internal conversion). Considering the
steepness of the S0, and especially the high vibrational excess
calculated from the Franck–Condon region, the photochem-
ical pathway should preferentially lead to the formation of 2,
thus expecting high quantum yield (i.e. no side reactions, no
fluorescence and almost no internal conversion). This mech-
anism is in agreement with the experimental findings and
suggests that a theoretical yield of 100% is possible.
To summarize, the photoreaction proceeds through a min-
imum of three consecutive steps upon photoirradiation. The
first one is photoexcitation. The physics of this process, with
the accent on the excited-state pathways, is explained by using
quantum mechanical calculations. During the second one,
photodimerization of the monomer molecules takes place,
assuming an SCSC reaction manner. During the last process,
a collective molecular movement happens, causing an RPT,
which results in irreversible disintegration of the crystals. For
the first time, the latter, important macroscopic event is
correlated with the significant changes on molecular level in
the crystal structures before and after the photoreaction.
These (in situ followed) changes of the crystal packing (in-
layer rotations and molecules shifting) create immense
internal stress, which when released causes the disintegration
of the crystals. The kinetics of this photodimerization reaction
supports the postulated reaction model and is explained by
two different kinetic regimes upon photoexcitation, the
second one representing an interesting autocatalytic auto-
inhibition RPT photoreaction set, disintegrating the crystal
and giving dimer products with high quantum yields (the
experimental quantum yield after five days of irradiation
reached 90%).
Figure 4. Excited-state pathways as a function of the most relevant
coordinates. Top: Initial relaxation on S2 and S1 states. The central-
ꢀ
bonds-lengthening coordinate is defined as the average of C1 C2 and
ꢀ
C2 C3 bond lengths of both molecules, A and B. Bottom: Continua-
ꢀ
tion of the pathway, as a function of the two C2 C3 intermolecular
distances, including S2/S1 energy degenerate region and S1/S0 conical
intersection (CI), which funnels the formation of the photoproduct on
the ground state (blue arrows), the less favored route is represented
by a dashed blue line.
in accordance with the one calculated at the TD-DFT level
for the whole unit cell. The selected active space comprises,
for each of the two molecules, one set of p and p* orbitals
Received: February 17, 2014
Published online: && &&, &&&&
ꢀ
centered on the C2 C3 formal double bond (Figure 4), one
Keywords: excited states · oxazolones · phase transformations ·
photodimerization · powder diffraction
set of p and p* orbitals centered on the Ph-Br moiety, and one
set of p and p* orbitals delocalized over the chromophore,
overall including 12 electrons in 12 orbitals. At this level of
theory, the dimer absorption properties qualitatively match
with TD-DFT results, therefore allowing us to study the
formation of the photoproduct through multiconfigurational
quantum chemistry (Table S3 in the Supporting Information).
After absorption at the Franck–Condon region, minimum-
energy paths were calculated from S2 (Figure 4). The overall
mechanism can be separated into two parts. The first step
consists of bond lengthening of the A and B molecules, with
almost identical energy relaxation in S2 and S1 states, both
1(p,p*) states (Figure 4, top). This first step increases the
structural flexibility. Then, the orbital overlapping resulting
from p–p stacking generates attraction between A and B
molecules. This therefore allows breaking the degeneracy in
energy between S2 and S1, further dissipating the excitation
energy in S1 up to a conical intersection with the ground state
(S1/S0 CI), where the dimer has an almost symmetrical
geometry, with A facing B and an equivalent intermolecular
distance between C2 and C3 atoms of 2.13 ꢀ (Figure 4,
bottom). This type of CI is similar to those found for
photochemical pericyclic reactions.[12] From the S1/S0 CI, two
pathways are available: formation of 2 or recovery of the
.
[1] a) S. Trommsdorff, Ann. Chem. Pharm. 1834, 11; b) H. I.
Bernstein, W. C. Quimby, J. Am. Chem. Soc. 1943, 65, 1845;
c) M. D. Cohen, G. M. J. Schmidt, J. Chem. Soc. 1964, 1996.
[2] H. Morawetz, S. Z. Jakabhazy, J. B. Lando, J. Shafer, Proc. Natl.
Acad. Sci. USA 1963, 49, 789 – 793.
[3] a) S.-Y. Yang, P. Naumov, S. Fukuzumi, J. Am. Chem. Soc. 2009,
131, 7247; b) K. Tanaka, F. Toda, E. Mochizuki, N. Yasui, Y. Kai,
I. Miyahara, K. Hirotsu, Angew. Chem. 1999, 111, 3733; Angew.
Chem. Int. Ed. 1999, 38, 3523; c) N. L. Toh, M. Nagarathinam,
J. J. Vittal, Angew. Chem. 2005, 117, 2277; Angew. Chem. Int. Ed.
2005, 44, 2237; d) M. H. Mir, L. L. Koh, G. K. Tan, J. J. Vittal,
Angew. Chem. 2010, 122, 400; Angew. Chem. Int. Ed. 2010, 49,
390; e) S. D. Karlen, H. Reyes, R. E. Taylor, S. I. Khan, M. F.
Hawthorne, M. A. Garcia-Garibay, Proc. Natl. Acad. Sci. USA
2010, 107, 14973; f) M. Nagarathinam, A. M. P. Peedikakkal, J. J.
Vittal, Chem. Commun. 2008, 5277; g) S.-Y. Yang, X.-L. Deng,
R.-F. Jin, P. Naumov, M. K. Panda, R.-B. Huang, L.-S. Zheng,
B. K. Teo, J. Am. Chem. Soc. 2014, 136, 558; h) V. Enkelmann, G.
Wegner, K. Novak, K. B. Wagener, J. Am. Chem. Soc. 1993, 115,
10390.
ˇ
ˇ
[4] a) P. Naumov, P. Makreski, G. Petrusevski, T. Runcevski, G.
Jovanovski, J. Am. Chem. Soc. 2010, 132, 14973; b) D. de Loera,
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
A. Stopin, M. A. Garcia-Garibay, J. Am. Chem. Soc. 2013, 135,
6626.
[5] a) P. Naumov, J. Kowalik, K. M. Solntsev, A. Baldridge, J.-S.
Moon, C. Kranz, L. M. Tolbert, J. Am. Chem. Soc. 2010, 132,
5845; b) N. K. Nath, K. Manoj, A. S. Gac, P. Naumov, Chem. Eur.
J. 2013, 19, 8094; c) Water molecules also influence the
Angew. Chem. Int. Ed. 2014, DOI: 10.1002/anie.201402040;
e) A. F. Mabied, M. Mueller, R. E. Dinnebier, S. Nozawa, M.
Hoshino, A. Tomita, T. Sato, S. Adachi, Acta Crystallogr. Sect. B
2012, 68, 424; f) R. Morꢂ, G. Busse, J. Hallmann, C. Paulmann,
M. Scholz, S. Techert, J. Phys. Chem. C 2010, 114, 4142; g) J. B.
Benedict, P. Coppens, J. Phys. Chem. A 2009, 113, 3116.
[9] a) T. Yanai, D. P. Tew, N. C. Handy, Chem. Phys. Lett. 2004, 393,
51; b) V. Voliani, R. Bizzarri, R. Nifosꢃ, S. Abbruzzetti, E.
Grandi, C. Viappiani, F. Beltram, J. Phys. Chem. B 2008, 112,
10714.
[10] K. Andersson, P.-ꢀ. Malmqvist, B. O. Roos, A. J. Sadlej, K.
Wolinski, J. Phys. Chem. 1990, 94, 5483 – 5488.
[11] F. Aquilante, L. De Vico, N. Ferrꢂ, G. Ghigo, P.-ꢀ. Malmqvist, P.
Neogrꢄdy, T. B. Pedersen, M. Pitonak, M. Reiher, B. O. Roos, L.
Serrano-Andrꢂs, M. Urban, V. Veryazov, R. Lindh, J. Comput.
Chem. 2010, 31, 224 – 247.
ˇ
crystallinity during the crystallization processes, T. Runcevski,
ˇ
G. Petrusevski, P. Makreski, S. Ugarkovic, R. E. Dinnebier,
Chem. Commun. 2014, DOI: 10.1039/c4cc01430h.
[6] R. E. Dinnebier, S. J. L. Billinge, Powder Diffraction: Theory
and Practice, RSC Publishing, Cambridge, 2008.
[7] a) M. Blanco-Lomas, P. J. Campos, D. Sampedro, Org. Lett. 2012,
14, 4334; b) I. Funes-Ardoiz, M. Blanco-Lomas, P. J. Campos, D.
Sampedro, Tetrahedron 2013, 69, 9766.
[8] a) M. Avrami, J. Chem. Phys. 1939, 7, 1103; b) W. A. Johnson,
P. A. Mehl, Trans. AIME 1939, 135, 416; c) A. N. Kolmogorov,
Bull. Acad. Sci. USSR Phys. Ser. 1937, 1, 355; d) R. Medishetty,
[12] F. Bernardi, M. Olivucci, M. A. Robb, J. Photochem. Photobiol.
A 1997, 105, 365.
ˇ
A. Husain, Z. Bai, T. Runcevski, R. E. Dinnebier, P. Naumov,
J. J. Vittal, Angew. Chem. 2014, DOI: 10.1002/ange.201402040;
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Photoreconstructions
ˇ
T. Runcevski,* M. Blanco-Lomas,
M. Marazzi, M. Cejuela, D. Sampedro,*
R. E. Dinnebier
&&&& — &&&&
Following a Photoinduced Reconstructive
Phase Transformation and its Influence
on the Crystal Integrity: Powder
Changing phases: In the course of solid-
state photoreactions, the single crystal of
a reactant can be transformed into
a single crystal of the product or it can
become amorphous. Between these sce-
narios exist the reconstructive phase
transformations, where the single crystal
becomes a powder. A detailed description
of the latter is given.
Diffraction and Theoretical Study
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!