FUKUSHIMA, ITO, HOSOMI AND OHBA
901
ꢂ
Table 4. Dihedral angles (ꢂ) between the carbonyl O C<
plane and phenyl rings A and B
Ê
Table 3. Selected geometric parameters (A, ) for (X)
O1ÐC10
C4ÐC5
C4ÐC9
C4ÐC17
C5ÐC6
C6ÐC7
C6ÐC20
C7ÐC8
C8ÐC9
1.211 (2)
1.395 (2)
1.394 (2)
1.520 (2)
1.378 (2)
1.383 (2)
1.518 (2)
1.388 (2)
1.394 (2)
C8ÐC23
C9ÐC10
C10ÐC11
C17ÐC18
C17ÐC19
C20ÐC21
C20ÐC22
C23ÐC24
C23ÐC25
1.517 (2)
1.513 (2)
1.490 (2)
1.516 (2)
1.512 (2)
1.509 (2)
1.507 (2)
1.516 (2)
1.518 (2)
A is the triisopropylphenyl group and B the 30- or 40-substituted phenyl
group. The standard uncertainties of the dihedral angles are
ꢀ1ꢂ.
'
'
B
A
(I)
(II)
(III)
(IV)
(V)
(VI)
(VII)
(VIII)
(IX)
(X)
86.4
78.5
88.6
87.2
88.5
80.8
81.8
83.7
86.4
89.0
87.2
3.5
5.1
11.1
7.4
9.7
7.5
14.8
5.5
6.8
8.1
9.1
C5ÐC4ÐC9
C5ÐC4ÐC17
C9ÐC4ÐC17
C4ÐC5ÐC6
C5ÐC6ÐC7
C5ÐC6ÐC20
C7ÐC6ÐC20
C6ÐC7ÐC8
C7ÐC8ÐC9
C7ÐC8ÐC23
C9ÐC8ÐC23
C4ÐC9ÐC8
C4ÐC9ÐC10
118.0 (1)
120.2 (1)
121.7 (1)
122.1 (1)
118.1 (1)
121.7 (1)
120.2 (1)
122.5 (1)
117.8 (1)
121.0 (1)
121.2 (1)
121.5 (1)
119.5 (1)
C8ÐC9ÐC10
O1ÐC10ÐC9
O1ÐC10ÐC11
C9ÐC10ÐC11
C4ÐC17ÐC18
C4ÐC17ÐC19
C18ÐC17ÐC19
C6ÐC20ÐC21
C6ÐC20ÐC22
C21ÐC20ÐC22
C8ÐC23ÐC24
C8ÐC23ÐC25
C24ÐC23ÐC25
119.0 (1)
120.3 (1)
120.1 (1)
119.6 (1)
110.5 (1)
112.1 (1)
110.6 (2)
110.5 (1)
113.6 (1)
111.9 (1)
113.2 (1)
110.7 (1)
111.0 (1)
(XI)
carbonyl plane is almost perpendicular to the triiso-
propylphenyl ring, the dihedral angle 'A being 79 (1)±
89 (1)ꢂ (Table 4). The methine H atoms of the isopropyl
groups at positions 2 and 6 are directed towards the
carbonyl side, as the result of minimizing the steric
repulsions between the neighbouring substituents. The
distances between the isopropyl methine H and the
Although ꢂ is 90.00 (1)ꢂ for (VII), the Laue group is
apparently not mmm, but 2/m. Rotational disorder of
the isopropyl groups was observed in most of the
compounds. As seen in Fig. 1, the ellipsoids of the
isopropyl terminal C atoms are large and elongated. The
split-atom model was tried for some compounds, but
remarkable improvement of R values was not obtained.
The average R value for the 11 compounds studied
(0.052 < R < 0.081) was 0.071, which is rather high and
may be due to the disorder of the substituents. In (IX)
rotational disorder of the chloroformyl group was taken
into account. The COCl group is almost coplanar with
the phenyl ring and has two possible orientations. The
split-atom model was not applicable to the present case
owing to the overlapping of the diffuse electron densi-
ties of the disordered Cl and O atoms. By using the
approximation that the positions of Cl and O in the
orientation of ClÐC O are the same as those of O and
Cl in the other orientation of O CÐCl, and assuming
that the probabilities of the two orientations are 50%
each, the arti®cial occupancy factors of Cl1 and O3
atoms were estimated to be (17 Â 0.5 + 8 Â 0.5)/17 and
(17 Â 0.5 + 8 Â 0.5)/8, respectively, based on the number
of electrons of the elements, following the treatment of
orientational disorder of thiophene rings (Pelletier &
Brisse, 1994). The re®ned Ueq values of Cl1 and O3
Ê
carbonyl O atoms are 2.7±3.3 A. Fig. 2 shows the
variation of the intramolecular distances between the
carbonyl O and the central C atoms of the ortho-
isopropyl groups along with the angle 'A. The greater
the deviation of 'A from 90ꢂ, the shorter one of the
OÁ Á ÁC(isopropyl) distances becomes, which seems
favourable for hydrogen abstraction by the excited
carbonyl O atom in photoirradiation. Indeed, (II), (VI)
and (VII) have 'A = 79 (1)±82 (1)ꢂ and show high
photoreactivity. However, the dihedral angle 'A is not
directly correlated with the reactivity, since 'A of the
other highly photoreactive compounds (IV) and (V)
range between 87 (1) and 89 (1)ꢂ. There is an orienta-
tional disorder of the isopropyl group at position 4.
Fig. 3 shows the correlation between the CÐCH3 bond
distances and the CH3ÐCÐCH3 bond angles. The
shorter the CÐCH3 bond distances, which are artefact
owing to the disorder, the larger the CH3ÐCÐCH3
bond angle. In the photostable crystals (IX) and (X)
there is a small degree of disorder, indicating a good
ef®ciency of molecular packing.
3.2. Reaction cavity
2
Ê
positions were similar, 0.133 (1) and 0.121 (1) A , indi-
cating that the model of disorder is reasonable.
Since the key factor for photoreactivity was not found
in the molecular structures, the environment of the
molecules in the crystals was analysed using the program
CAVITY (Ohashi et al., 1981). The reaction cavity for
the triisopropylphenylcarbonyl moiety was calculated
assuming that neighbouring atoms have spheres of radii
greater than the corresponding van der Waals radii by
3. Discussion
3.1. Molecular structure
0
0
Ê
The conformations of the molecules are essentially 1.2 A. The intramolecular 3 - or 4 -substituted phenyl
the same for all the compounds, as shown in Fig. 1. The group was treated as constituting barrier atoms to clarify