Monitoring structural transformations in crystals. 13. on photocyclization in 2,3,4,5,6-penta-methyl-benzophenone, 1,3-diphenyl-butan-1-one and 2,4,6-tri-isopropyl-4′-methoxy-benzophenone

Article abstract of DOI:10.1107/S0108270109052858

The geometrical parameters governing the potential for the photocyclization reaction occurring in crystals of 2,3,4,5,6-penta-methyl-benzophenone, C ₁₈H₂₀O, (I), 1,3-diphenyl-butan-1-one, C ₁₆H₁₆O, (II), and 2,4

Full text of DOI:10.1107/S0108270109052858

organic compounds
Acta Crystallographica Section C
Crystal Structure
Communications
influence of UV–vis radiation. However, it has also been
proposed that the mechanism of formation of benzocyclo-
butenols can have a more complicated character, namely the
reaction can proceed not only via a biradical directly but,
additionally, via an enol form (Ito et al., 2009). Nevertheless, it
was stated that, in the case of large substituents and steric
crowding, formation of enols in a crystalline state can be
neglected (Ito et al., 2009; Moorthy et al., 2004). ortho-Alkyl-
phenyl ketones can also undergo a ꢁ-abstraction reaction (Ito
et al., 2009) as follows:
ISSN 0108-2701
Monitoring structural transformations
in crystals. 13. On photocyclization in
2,3,4,5,6-pentamethylbenzophenone,
1,3-diphenylbutan-1-one and 2,4,6-
triisopropyl-40-methoxybenzophenone
Julia Ba˛kowicz and Ilona Turowska-Tyrk*
.
´
Faculty of Chemistry, Wrocław University of Technology, 27 Wybrzeze Wyspian-
skiego, 50-370 Wrocław, Poland
Correspondence e-mail: ilona.turowska-tyrk@pwr.wroc.pl
Some 2,4,6-triisopropylbenzophenones are photoactive in
the crystalline state and undergo the photocyclization reaction
according to Scheme 1 (Fukushima et al., 1998; Ito et al., 2009;
Ito, Kano et al., 1998). However, some 2,4,6-triisopropyl-
benzophenones are photoinert (Fukushima et al., 1998; Ito,
Kano et al., 1998; Ito, Yasui et al., 1998). Moreover, 2,4,6-
trimethylbenzophenone is photoinert in the crystalline state
(Ito et al., 2009), but mesitylaldehyde in a solid inclusion
compound is photoactive (Moorthy et al., 2001).
In this paper, we present and discuss the structures of the
following compounds: 2,3,4,5,6-pentamethylbenzophenone,
(I), 1,3-diphenylbutan-1-one, (II), and 2,4,6-triisopropyl-40-
methoxybenzophenone–3,5-diisopropyl-7-(4-methoxyphenyl)-
8,8-dimethylbicyclo[4.2.0]octa-1,3,5-trien-7-ol (9/1), (III),
which is the product of the photocyclization of 2,4,6-triiso-
propyl-40-methoxybenzophenone, (IV) (see Scheme 1). The
intramolecular photocyclization is one of the reactions moni-
tored by us recently (Turowska-Tyrk, Ba˛kowicz et al., 2006;
Turowska-Tyrk, Ba˛kowicz et al., 2007; Turowska-Tyrk,
Łabe˛cka et al., 2007; Turowska-Tyrk, Trzop et al., 2006).
Received 27 October 2009
Accepted 8 December 2009
Online 12 December 2009
The geometrical parameters governing the potential for the
photocyclization reaction occurring in crystals of 2,3,4,5,6-
pentamethylbenzophenone, C18H20O, (I), 1,3-diphenylbutan-
1-one, C16H16O, (II), and 2,4,6-triisopropyl-40-methoxybenzo-
phenone, C23H30O2, (IV), have been evaluated. Compound
(IV) undergoes photocyclization but (I) and (II) do not,
despite the fact that their geometrical parameters appear
equally favourable for reaction. The structure of the partially
reacted crystal of the photoactive compound, i.e. 2,4,6-
triisopropyl-40-methoxybenzophenone–3,5-diisopropyl-7-(4-
methoxyphenyl)-8,8-dimethylbicyclo[4.2.0]octa-1,3,5-trien-7-ol
(9/1), 0.90C23H30O2ꢀ0.10C23H30O2, (III), was also determined,
providing structural evidence for the reactivity of the
compound. It has been found that the carbonyl group of the
photoactive compound reacts with one of the two o-isopropyl
groups. The study has shown that the intramolecular geo-
metrical parameters are not the only factors influencing the
reactivity of compounds in crystals.
Comment
Compounds containing a carbonyl group and an Hꢀ atom can
potentially undergo a photocyclization reaction. ortho-Alkyl-
phenyl ketones are an example of such compounds. A
mechanism of this reaction is presented in the scheme below
(hereafter referred to as Scheme 1).
There exist geometrical parameters describing conditions
that must be fulfilled for a photocyclization reaction to
proceed in crystals (Natarajan et al., 2005; Xia et al., 2005).
They are as follows (see the scheme below): the (C)Oꢀ ꢀ ꢀHꢀ
As can be seen, the reaction leads to the formation of a
cyclobutane ring from a biradical formed previously under the
Acta Cryst. (2010). C66, o29–o32
doi:10.1107/S0108270109052858
# 2010 International Union of Crystallography o29

organic compounds
distance, d, the (O)Cꢀ ꢀ ꢀCꢀ distance, D, the deviation of Hꢀ
from the mean plane of the carbonyl group, !, the C Oꢀ ꢀ ꢀHꢀ
angle, Á, and the Cꢀ—Hꢀꢀ ꢀ ꢀO angle, Â. The ideal and
average literature values of these parameters for photoactive
compounds are given in Table 1. However, it should be
pointed out that good values of these parameters are not the
only necessary condition for photocyclization to proceed in
crystals (Ba˛kowicz & Turowska-Tyrk, 2009; Ito, Yasui et al.,
1998; Moorthy et al., 2006; Zouev et al., 2006).
are given in Table 1. In the case of (I) and (II), only the values
for the Hꢀ atom closer to the carbonyl group are given, but for
(IV), the values for both o-isopropyl groups are presented. As
can be seen, not all of the geometrical parameters for (I), (II)
and (IV) are appropriate for the photocyclization reaction in
crystals. For (I), the value of the ! parameter is greater (by
10.5ꢁ) than the largest literature value for compounds under-
going the Yang photocyclization (Turowska-Tyrk, Ba˛kowicz et
˚
al., 2007). In the case of (II), D is slightly worse (by 0.02 A)
than the limit known in the literature, but ! is slightly better
(by 3.8ꢁ). For the first o-isopropyl group of (IV), the d and !
parameters are larger and exceed the literature values (by
ꢁ
˚
0.13 A and 15.6 , respectively). For the second o-isopropyl
ꢁ
group, d, ! and  are worse (by 0.05 A, 10.9 and 0.7ꢁ,
˚
respectively).
In the case of (I)–(III), there are no ꢂ–ꢂ interactions
between neighbouring naphthalene rings. The lack of ꢂ–ꢂ
stacking can increase the reactivity of compounds (Fukushima
et al., 1998; Ito et al., 2009).
However, despite the above considerations, (I) and (II) are
photoinert and (IV) photoactive. Irradiation of crystals of (I)
and (II) for periods of 7 and 6 h, respectively, did not cause the
photocyclization reaction. We did not observe any changes in
the cell constants over the irradiation time and the structures
determined after the irradiation revealed only reactant mol-
ecules.
In order to make the photoreaction easier, (I) (m.p. 410–
411 K) and (II) (m.p. 343–345 K) were irradiated at elevated
temperatures: at 373 K for 4 h and at 313 K for 5 h, respec-
tively. Nevertheless, it did not help to induce the photoreac-
tion.
The structure of (IV) before irradiation, i.e. containing only
reactant molecules, has been reported previously (Fukushima
et al., 1998). Nevertheless, we redetermined it in order to have
confidence that we have the proper structure for further
experiments. The molecular geometries for both structures
were very similar. The information that (IV) undergoes the
photocyclization reaction according to Scheme 1 was also
given previously (Fukushima et al., 1998; Ito & Matsura, 1988),
but structural evidence, i.e. a structure of a crystal containing
product molecules, was not supplied. The photoreaction of
(IV) does not proceed in a single-crystal-to-single-crystal
manner. Crystals under prolonged influence of UV–vis
One of the reasons for photochemical inactivity of
compounds can be the presence of intermolecular ꢂ–ꢂ inter-
actions. They hinder shifts of atoms and in this manner a whole
reaction (Fukushima et al., 1998; Ito et al., 2009). Too small a
reaction cavity can be another reason for photochemical
inactivity (Fukushima et al., 1998; Ito, Kano et al., 1998; Ito,
Yasui et al., 1998; Moorthy et al., 2006; Zouev et al., 2006). In
these cases, despite the formation of biradicals, compounds do
not cyclize but return to substrates (Ito, Yasui et al., 1998). In
order to make molecular and atomic movements easier,
reactions were sometimes conducted at elevated temperatures
(Fukushima et al., 1998; Ito, Yasui et al., 1998).
Figs. 1 and 2 present the structures of (I) and (II), respec-
tively. The bond lengths in the molecules of both compounds
are typical. The dihedral angles between the planes of the two
benzene rings are 82.98 (5) and 62.36 (10)ꢁ for (I) and (II),
respectively. The above-mentioned geometrical parameters
influencing the photochemical reaction of the title compounds
Figure 1
The molecular structure of (I), showing the atom-numbering scheme.
Displacement ellipsoids are drawn at the 30% probability level.
Figure 2
The molecular structure of (II), showing the atom-numbering scheme.
Displacement ellipsoids are drawn at the 30% probability level.
ꢂ
o30 Ba˛kowicz and Turowska-Tyrk
C18H20O, C16H16O and 0.90C23H30O2ꢀ0.10C23H30O2
Acta Cryst. (2010). C66, o29–o32

organic compounds
Table 1
Values of geometrical parameters describing photocyclization.
! (ꢁ)
Á (ꢁ)
 (ꢁ)
˚
d (A)
˚
D (A)
Ideal value
<2.7
0
90–120
82 (8)
2.49–2.82 2.82–3.12 49.0–67.5 52.9–88.0 111.0–128.0
180
116 (3)
Average literature valuea 2.64 (8) 3.00 (9) 54 (10)
Rangeb
(I) (this work)
(II) (this work)
(IV)c
2.71
2.58
2.95
2.89
2.885 (3) 78.0
3.142 (4) 45.2
2.931 (5) 83.1
2.928 (5) 77.9
61.5
87.6
53.5
58.5
121.8
117.8
122.8
108.9
Notes: (a) the mean values of d, ! and Á are given for 57 and  for 40 aromatic ketones
undergoing photocyclization (Natarajan et al., 2005), and D for 53 structures (Xia et al.,
2005); (b) the range of the parameters is given on the grounds of 47 compounds for d, !,
Á and  (Chen et al., 2005; Ihmels & Scheffer, 1999; Leibovitch et al., 1998; Natarajan et
al., 2005; Turowska-Tyrk, Ba˛kowicz et al., 2007; Turowska-Tyrk, Łabe˛cka et al., 2007;
Turowska-Tyrk, Trzop et al., 2006; Vishnumurthy et al., 2002) and 15 compounds for D
(Leibovitch et al., 1998; Turowska-Tyrk, Ba˛kowicz et al., 2007; Turowska-Tyrk, Łabe˛cka et
al., 2007; Turowska-Tyrk, Trzop et al., 2006); (c) on the basis of the literature data
(Fukushima et al., 1998) – the first and the second lines are for the reacting and unreacting
isopropyl groups, respectively.
Figure 3
A view of the photoproduct molecule (empty bonds) superimposed on
the reactant molecule (filled bonds) for the crystal of (III). H atoms have
been omitted for clarity. Displacement ellipsoids are drawn at the 10%
probability level.
Data collection
Kuma KM-4 CCD diffractometer
7742 measured reflections
2553 independent reflections
1751 reflections with I > 2ꢆ(I)
Rint = 0.036
radiation lose their diffracting properties. Nevertheless, we
were able to determine the crystal structure containing
89.6 (7)% of reactant and 10.4 (7)% of product molecules. The
structure is presented in Fig. 3. It is very interesting that the
carbonyl group reacts with one of two o-isopropyl groups, at
least at the beginning of the photoreaction. Unfortunately,
because the crystal loses its diffracting properties, it is not
possible to see the behaviour of the compound during the rest
of the photoreaction.
Refinement
R[F2 > 2ꢆ(F2)] = 0.057
wR(F2) = 0.177
S = 1.04
177 parameters
H-atom parameteꢃrs3 constrained
˚
Áꢇmax = 0.17 e A
Áꢇmin = ꢃ0.21 e A
ꢃ3
˚
2553 reflections
Compound (II)
In many cases, intramolecular parameters describing
geometrical demands for a chemical reaction work very well.
In the scientific literature, many such examples are known (for
instance: Chen et al., 2005; Ihmels & Scheffer, 1999; Leibovitch
et al., 1998; Natarajan et al., 2005; Turowska-Tyrk, Ba˛kowicz et
al., 2007; Turowska-Tyrk, Łabe˛cka et al., 2007; Turowska-Tyrk,
Trzop et al., 2006; Vishnumurthy et al., 2002; Xia et al., 2005).
However, there are also situations where predictions about
reactivity on the basis of such parameters fail (for instance:
Fukushima et al., 1998; Ito, Kano et al., 1998). Such a situation
concerns also the title compounds.
Crystal data
3
˚
V = 1261.4 (3) A
Z = 4
C16H16O
Mr = 224.29
Orthorhombic, Pca21
Mo Kꢄ radiation
ꢅ = 0.07 mmꢃ1
T = 299 K
0.30 ꢄ 0.15 ꢄ 0.10 mm
˚
a = 10.7926 (18) A
˚
b = 14.931 (2) A
˚
c = 7.8275 (11) A
Data collection
Kuma KM-4 CCD diffractometer
6471 measured reflections
1201 independent reflections
1050 reflections with I > 2ꢆ(I)
Rint = 0.029
Refinement
Experimental
R[F2 > 2ꢆ(F2)] = 0.038
wR(F2) = 0.115
S = 1.05
1201 reflections
155 parameters
1 restraint
H-atom parameteꢃrs3 constrained
Compounds (I) and (II) were purchased from Alfa Aesar and
compound (IV) from Sigma–Aldrich. Compounds (I) and (II) were
recrystallized from acetone and ethanol, respectively; (IV) was not
recrystallized.
˚
Áꢇmax = 0.11 e A
ꢃ3
˚
Áꢇmin = ꢃ0.11 e A
Compound (I)
Compound (III)
Crystal data
Crystal data
3
3
˚
˚
V = 2093.7 (4) A
Z = 4
C18H20O
Mr = 252.34
Monoclinic, P21=n
V = 1458.1 (6) A
Z = 4
0.90C23H30O2ꢀ0.10C23H30O2
Mr = 338.47
Monoclinic, P21=c
Mo Kꢄ radiation
ꢅ = 0.07 mmꢃ1
T = 299 K
0.60 ꢄ 0.30 ꢄ 0.13 mm
Mo Kꢄ radiation
ꢅ = 0.07 mmꢃ1
T = 299 K
0.40 ꢄ 0.30 ꢄ 0.15 mm
˚
a = 6.3784 (15) A
˚
a = 9.2288 (10) A
˚
˚
b = 12.0541 (14) A
b = 12.793 (2) A
˚
c = 18.014 (5) A
˚
c = 18.821 (2) A
ꢃ = 97.28 (2)ꢁ
ꢃ = 90.021 (9)ꢁ
ꢂ
Acta Cryst. (2010). C66, o29–o32
Ba˛kowicz and Turowska-Tyrk
C18H20O, C16H16O and 0.90C23H30O2ꢀ0.10C23H30O2 o31

organic compounds
tion, 2003); data reduction: CrysAlis RED; program(s) used to solve
structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine
structure: SHELXL97 (Sheldrick, 2008); molecular graphics:
ORTEP-3 for Windows (Farrugia, 1997); software used to prepare
material for publication: SHELXL97.
Data collection
Kuma KM-4 CCD diffractometer
9964 measured reflections
3564 independent reflections
1609 reflections with I > 2ꢆ(I)
Rint = 0.040
Refinement
R[F2 > 2ꢆ(F2)] = 0.066
wR(F2) = 0.232
S = 1.00
3564 reflections
328 parameters
224 restraints
H-atom parameteꢃrs3 constrained
Supplementary data for this paper are available from the IUCr electronic
archives (Reference: GZ3170). Services for accessing these data are
described at the back of the journal.
˚
Áꢇmax = 0.26 e A
ꢃ3
˚
Áꢇmin = ꢃ0.15 e A
References
Methyl H atoms of (I) and of the C23 methyl group of (III) were
located in difference Fourier maps and refined as part of rigid
rotating groups. The remaining H atoms of (I) and (III) and all H
atoms of (II) were positioned geometrically and treated as riding.
Ba˛kowicz, J. & Turowska-Tyrk, I. (2009). Acta Cryst. C65, o377–o380.
Chen, S., Partick, B. O. & Scheffer, J. R. (2005). Can. J. Chem. 83, 1460–
1472.
Farrugia, L. J. (1997). J. Appl. Cryst. 30, 565.
Fukushima, S., Ito, Y., Hosomi, H. & Ohba, S. (1998). Acta Cryst. B54, 895–
906.
˚
C—H distances were fixed at 0.93–0.98 A and Uiso(H) values were
fixed at 1.5Ueq(C) for methyl H atoms and at 1.2Ueq(C) for other H
atoms.
Ihmels, H. & Scheffer, J. R. (1999). Tetrahedron, 55, 885–907.
Ito, Y., Kano, G. & Nakamura, N. (1998). J. Org. Chem. 63, 5643–5647.
Ito, Y. & Matsura, T. (1988). Tetrahedron Lett. 25, 3087–3090.
Ito, Y., Takahashi, H., Hasegawa, J. & Turro, N. J. (2009). Tetrahedron, 65, 677–
689.
Ito, Y., Yasui, S., Yamauchi, J., Ohba, S. & Kano, G. (1998). J. Phys. Chem. A,
102, 5415–5420.
Leibovitch, M., Olovsson, G., Scheffer, J. R. & Trotter, J. (1998). J. Am. Chem.
Soc. 120, 12755–12769.
Moorthy, J. N., Mal, P., Natarajan, R. & Venugopalan, P. (2001). J. Org. Chem.
66, 7013–7019.
Moorthy, J. N., Mal, P., Singhal, N., Venkatakrishnan, P., Malik, R. &
Venugopalan, P. (2004). J. Org. Chem. 69, 8459–8466.
Moorthy, J. N., Venkatakrishnan, P., Savitha, G. & Weiss, R. G. (2006).
Photochem. Photobiol. Sci. 5, 903–913.
Natarajan, A., Mague, J. T. & Ramamurthy, V. (2005). J. Am. Chem. Soc. 127,
3568–3576.
Oxford Diffraction (2003). CrysAlis CCD and CrysAlis RED. Versions 1.170.
Oxford Diffraction Ltd, Wrocław, Poland.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112–122.
Turowska-Tyrk, I., Ba˛kowicz, J. & Scheffer, J. R. (2007). Acta Cryst. B63, 933–
940.
Turowska-Tyrk, I., Ba˛kowicz, J., Scheffer, J. R. & Xia, W. (2006). CrystEng-
Comm, 8, 616–621.
Turowska-Tyrk, I., Łabe˛cka, I., Scheffer, J. R. & Xia, W. (2007). Pol. J. Chem.
81, 813–824.
Turowska-Tyrk, I., Trzop, E., Scheffer, J. R. & Chen, S. (2006). Acta Cryst. B62,
128–134.
Vishnumurthy, K., Cheung, E., Scheffer, J. R. & Scott, C. (2002). Org. Lett. 7,
1071–1074.
Xia, W., Scheffer, J. R., Botoshansky, M. & Kaftory, M. (2005). Org. Lett. 7,
1315–1318.
Zouev, I., Lavy, T. & Kaftory, M. (2006). Eur. J. Org. Chem. pp. 4164–4169.
The crystal of (IV) was irradiated for 4 h using a 100 W Hg lamp
equipped with a water filter. For (III), the first atoms of the minor
component (product) were found in a difference Fourier map and the
remaining atoms were located geometrically. The difference peaks
were seen near one o-isopropyl group and not near the other. The
major component (reactant) was refined anisotropically and the
minor component was refined isotropically. H atoms in –OH and
–OCH3 groups of the minor component were omitted. The percen-
tage of product molecules was determined by refinement of the site-
occupation factor. The R1 value improved from 0.075 to 0.066 after
inclusion of the minor component. Owing to a reactant–product
disorder, which is always a feature of partially reacted crystals, the
following weak restraints from SHELXL97 (Sheldrick, 2008) were
applied for the minor component: DFIX, DANG, FLAT and SIMU.
The DFIX and DANG commands restrained bond lengths and
valence angles to target values. The target values were taken from the
structures of the pure reactant (IV) and the pure photoproduct of a
similar compound. FLAT restrained some atoms to be coplanar.
SIMU restrained the displacement parameters of atoms of the
photoproduct.
For (II), the absolute structure could not be determined and an
arbitrary choice was made for the enantiomer (C3-R) chosen for the
asymmetric unit.
For all compounds, data collection: CrysAlis CCD (Oxford
Diffraction, 2003); cell refinement: CrysAlis RED (Oxford Diffrac-
ꢂ
o32 Ba˛kowicz and Turowska-Tyrk
C18H20O, C16H16O and 0.90C23H30O2ꢀ0.10C23H30O2
Acta Cryst. (2010). C66, o29–o32