Cis-Trans Isomerization of 1,2-Diaryl Olefins in the Solid State
examples provided above, this is exactly what happens. Cohen
defines a reaction cavity as the space occupied by the reactant
molecules plus the empty space surrounding it. In the crystalline
state, unlike in biological media, the reaction cavity is rigid and
the space needed for a given reaction must be built in.11
Therefore, in loosely packed crystals, the occurrence of geo-
metric isomerization requiring large motions should not be a
surprise.12
Experimental Section
The synthesis and characterization of 1-3 have been reported
previously.5
Photoreaction and Analysis Procedure. Crystals 1-3 (∼5 mg),
either as ground single crystals or in polycrystalline form (powder),
were sandwiched between two Pyrex plates and spread to cover a
surface area of 2-3 cm2. The plates were sealed with Parafilm on
all the sides before the irradiation process. After the irradiation,
FIGURE 5. Crystal structure of cis-3 showing short contacts.
1
the solid was dissolved in 2 mL of CDCl3 and analyzed using H
absence of a large globular (three-dimensional) free volume,
this process would be prevented in a crystal by the walls of the
reaction cavity. An alternative mechanism known as the “hula-
twist” is being currently considered as a possibility in a
constrained medium.9 The hula-twist process (HT), unlike
conventional one-bond processes, requires less volume change
during cis-trans conversion. The hula-twist is more of a two-
dimensional rather than a three-dimensional process. During the
cis-trans conversion via a one-bond flip mechanism, one-half
of the molecule undergoes a 180° flip; i.e., half of the molecule
rises from the molecular plane and sweeps a 180° motion before
it rests on the same plane in a different geometry. On the other
hand, a hula-twist process which involves a simultaneous
rotation of two adjacent bonds (a single and a double bond) or
a 180° translocation of one C-H unit results in cis-trans
isomerization. In the hula-twist process, only one C-H unit of
the molecule rises above the molecular plane during cis-trans
conversion. Therefore, the volume demand on this process is
much less than that during the one-bond flip. The point to be
noted is that, between the one-bond flip and the hula-twist, the
latter would be preferred under conditions where the reaction
cavity has only a limited free volume. Although the suggestion
has been made that the geometric isomerization within a crystal
may occur via a hula-twist process, concrete evidence in favor
of this is still lacking.10 The occurrence of geometric isomer-
ization in crystalline cis-1 and -2a-c in this study can be
understood on the basis of the conventional one-bond flip on
the phenyl side of the alkenes. However, the possibility of the
hula-twist mechanism could not be fully ruled out. Molecules
in crystalline cis-3 must be so tightly packed so that neither the
one-bond flip nor the hula-twist is able to generate the trans
isomer.
NMR. The melting points of the substrates were determined and
are uncorrected. Irradiations were performed using a 450 W
medium-pressure mercury arc lamp in a water-cooled immersion
well. Photoreactions were conducted using appropriate filters (350
nm, Corning-0-52 cutoff filter for cis-1 and cis-2a-c; and 425 nm,
Corning-3-73 cutoff filter for cis-3).
Crystal Structure Determination. Full spheres of data were
collected using 606 scans in ω (0.3° per scan) at æ ) 0, 120, and
240°.13a,b The raw data were reduced to F2 values using the SAINT+
software,13c and global refinements of unit-cell parameters employ-
ing approximately 2000-5000 reflections chosen from the full data
sets were performed. Multiple measurements of equivalent reflec-
tions provided the basis for empirical absorption corrections as well
as for corrections for any crystal deterioration during the data
collection (SADABS13d). All structures were completed by suc-
cessive cycles of full-matrix, least-squares refinement followed by
the calculation of difference maps. All hydrogen atoms attached to
carbon were placed in calculated positions with isotropic displace-
ment parameters 20% larger than those of the attached atoms and
allowed to ride. All computations associated with structure solution,
refinement, and presentation were performed with the SHELXTL13b
package. The cavity plots were obtained using the PLATON8
program. For crystal structure visualization, MERCURY (CSD
software)7 was used.
Acknowledgment. V.R. thanks the NSF (CHE-0212042) for
financial support, and J.T.M. thanks the Louisiana Board of
Regents through the Louisiana Educational Quality Support
Fund (Grant No. LEQSF (2003-2003)-ENH-TR-67) for the
purchase of the diffractometer.
Supporting Information Available: Dihedral angles, packing
diagrams generated through the Mercury program showing the short
contacts for 1-3, and the stereoscopic view of the packing diagrams
of 1-3. This material is available free of charge via the Internet at
Conclusion
JO0520644
There is a general belief that reactions that require large
motions would not occur in the solid state. This is based on the
assumption that molecules are densely packed in the crystalline
state. If for any reason the packing is such that it leaves void
space near the reacting molecules, this space could be utilized
during the transformation of a reactant to a product. In the
(11) (a) Ramamurthy, V.; Weiss, R. G.; Hammond, G. S. AdV. Photo-
chem. 1993, 18, 67. (b) Weiss, R. G.; Ramamurthy, V.; Hammond, G. S.
Acc. Chem. Res. 1993, 26, 530. (c) Liu, R. S. H.; Hammond, G. S. Acc.
Chem. Res. 2005, 38, 396.
(12) For a summary of photoreactions of olefins in crystals, see:
Natarajan, A.; Ramamurthy, V. In The Chemistry of Cyclobutanes; Rapport,
Z., Liebman, J. F., Eds.; John Wiley: New York, 2005; pp 807-872.
(13) (a) SMART, version 5.625; Bruker-AXS: Madison, WI, 2000. (b)
SHELXTL, version 6.10; Bruker-AXS: Madison, WI, 2000. (c) SAINT+,
version 7.03; Bruker-AXS: Madison, WI, 2004. (d) Sheldrick, G. M.
SADABS, version 2.05; University of Gottingen: Germany, 2002.
(9) (a) Liu, R. S. H.; Hammond, G. S. Chem.-Eur. J. 2001, 7, 2. (b)
Liu, R. S. H.; Hammond, G. S. Acc. Chem. Res. 2001, 34, 555.
(10) Kaupp, G. Photochem. Photobiol. 2002, 76, 590.
J. Org. Chem, Vol. 71, No. 3, 2006 1059