Crystalline-state Z,E-photoisomerization of a series of (Z,E,Z)-1,6-diphenylhexa-1,3,5-triene 4,4′-dicarboxylic acid dialkyl esters. Chain length effects on the crystal structure and photoreactivity

Article abstract of DOI:10.1021/jo051137g

Crystalline-state Z,E-photoisomerization of a series of (Z,E,Z)-1,6-diphenylhexa-1,3,5-triene 4,4′-dicarboxylic acid dialkyl (R) esters [(Z,E,Z)-1a, R = Me; (Z,E,Z)-1b, R = Et; (Z,E,Z)-1c, R = n-Pr; (Z,E,Z)-1d, R = n-Bu] was investigated. All Z,E,Z isomer

Full text of DOI:10.1021/jo051137g

Crystalline-State Z,E-Photoisomerization of a Series of
(Z,E,Z)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic Acid Dialkyl
Esters. Chain Length Effects on the Crystal Structure and
Photoreactivity†
Yoriko Sonoda,*,‡ Yuji Kawanishi,‡ Seiji Tsuzuki,§ and Midori Goto|
Nanotechnology Research Institute and Technical Center, National Institute of Advanced Industrial
Science and Technology (AIST), Higashi 1-1-1, Tsukuba, Ibaraki 305-8565, Japan, and Research Institute
of Computational Sciences, National Institute of Advanced Industrial Science and Technology (AIST),
Umezono 1-1, Tsukuba, Ibaraki 305-8568, Japan
y.sonoda@aist.go.jp
Received June 6, 2005
Crystalline-state Z,E-photoisomerization of a series of (Z,E,Z)-1,6-diphenylhexa-1,3,5-triene 4,4′-
dicarboxylic acid dialkyl (R) esters [(Z,E,Z)-1a, R ) Me; (Z,E,Z)-1b, R ) Et; (Z,E,Z)-1c, R ) n-Pr;
(Z,E,Z)-1d, R ) n-Bu] was investigated. All Z,E,Z isomers underwent one-way isomerization to
the corresponding E,E,E isomers. The reaction efficiency was strongly enhanced as the length of
the alkyl chain increased. Single-crystal X-ray analyses of (Z,E,Z)-1a-d showed that the alkyl
chain part of the crystals became larger as the chain length increased. The conformational flexibility
of the alkyl chains made the large change in the triene geometry in the lattice possible, leading to
the enhancement of the photoreactivity in the crystalline state.
Introduction
crystalline-state photoisomerization is relatively rare.
The reaction has been reported for only a few kinds of
compounds7 such as (Z)-1,2-di(1-naphthyl)ethylene,8,9
aromatic retinoids,10 and (Z,Z)-muconates11 up to now.
It is obvious that the large changes in the molecular
Organic photoreactions in the solid state are attracting
increased attention at the present time. The crystalline-
state photopolymerization of diacetylenes1 and diolefins2
are well-known, and more recently, the polymerization
of conjugated dienes has been reported.3 For olefinic
compounds in the crystalline state, one of the most
common photoreactions would be [2+2] cycloaddition.4,5
The Z,E-photoisomerization of olefins in restricted media
has also received considerable attention;6 however, the
(4) (a) Ramamurthy, V.; Venkatesan, K. Chem. Rev. 1987, 87, 433.
(b) Ito, Y. Synthesis 1998, 1, 1. (c) Sonoda, Y. In CRC Handbook of
Organic Photochemistry and Photobiology, 2nd ed.; Horspool, W., Lenci,
F., Eds.; CRC Press: Boca Raton, FL, 2004; Chapter 73.
(5) (a) Singh, A. K.; Krishina, T. S. R. J. Phys. Chem. A. 1997, 101,
3066. (b) Sonoda, Y.; Miyazawa, A.; Hayashi, S.; Sakuragi, M. Chem.
Lett. 2001, 410. (c) Gao, X.; Fri_s_cic´, T.; MacGillivray, L. R. Angew.
Chem., Int. Ed. 2004, 43, 232. (d) Caronna, T.; Liantonio, R.; Logoth-
etis, T. A.; Metrangolo, P.; Pilati, T.; Resnati, G. J. Am. Chem. Soc.
2004, 126, 4500. (e) Papaefstathiou, G. S.; Zhong, Z.; Geng, L.;
MacGillivray, L. R. J. Am. Chem. Soc. 2004, 126, 9158.
(6) (a) Liu, R. S. H. Acc. Chem. Res. 2001, 34, 555. (b) Liu, R. S. H.;
Hammond, G. S. Chem. Eur. J. 2001, 7, 4536. (c) Liu, R. S. H.
Photochem. Photobiol. 2002, 76, 580.
(7) (a) Scheffer, J. R.; Garcia-Garibay, M.; Nalamasu, O. In Organic
Photochemistry; Padwa, A., Ed.; Marcel Dekker: New York, 1987; Vol.
8, Chapter 4. (b) Scheffer, J. R.; Pokkuluri, P. R. In Photochemistry in
Organized and Constrained Media; Ramamurthy, V., Ed.; VCH: New
York, 1991; Chapter 5.
(8) Alfimov, M. V.; Razumov, V. F. Mol. Cryst. Liq. Cryst. 1978, 49
(Lett.), 95.
† Y.S. dedicates this article to Prof. Michinori Oh ki on the occasion
of his 77th birthday.
* Author to whom correspondence should be addressed. Fax: +81-
29-861-4673.
‡ Nanotechnology Research Institute, National Institute of Advanced
Industrial Science and Technology (AIST).
§ Research Institute of Computational Sciences, National Institute
of Advanced Industrial Science and Technology (AIST).
| Technical Center, National Institute of Advanced Industrial
Science and Technology (AIST).
(1) Sarkar, A.; Okada, S.; Matsuzawa, H.; Matsuda, H.; Nakanishi,
H. J. Mater. Chem. 2000, 10, 819.
(2) (a) Coates, G. W.; Dunn, A. R.; Henling, L. M.; Ziller, J. W.;
Lobkovsky, E. B.; Grubbs, R. H. J. Am. Chem. Soc. 1998, 120, 3641.
(b) Takahashi, S.; Miura, H.; Kasai, H.; Okada, S.; Oikawa, H.;
Nakanishi, H. J. Am. Chem. Soc. 2002, 124, 10944.
(3) (a) Matsumoto, A.; Odani, T. Macromol. Rapid Commun. 2001,
22, 1195. (b) Matsumoto, A. Polymer J. 2003, 35, 93.
(9) Aldoshin, S. M.; Alfimov, M. V.; Atovmyan, L. O.; Kaminsky, V.
F.; Razumov, V. F.; Rachinsky, A. G. Mol. Cryst. Liq. Cryst. 1984, 108,
1.
(10) Pfoertner, K.-H.; Englert, G.; Schoenholzer, P. Tetrahedron
1987, 43, 1321.
10.1021/jo051137g CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/22/2005
J. Org. Chem. 2005, 70, 9755-9763
9755

Sonoda et al.
SCHEME 1. Preparation of (Z,E,Z)-1a-d by Wittig
Reactions and One-Way Z,E-Photoisomerization of
(Z,E,Z)-1a-d in the Crystalline State
geometry required by Z,E-photoisomerization are difficult
to occur in the limited space of the rigid crystal lattice,
since they usually accompany the large changes in
molecular size and volume. This means that the easiness
of the movements of the double-bond carbon atoms in the
lattice plays a key role in determining the Z,E-photoi-
somerization reactivity in the crystalline state. It is
therefore expected that, if an olefinic molecule has large
conformational flexibility, it can easily isomerize in the
crystal lattice without the large change in the total
molecular size. The easiness of the atomic movements
in the lattice is, of course, closely related to the crystal
structure. However, our knowledge about the crystalline-
state photoisomerization is highly fragmental. Because
of lack of systematic study, the relationship between the
crystal structure and the photoisomerization reactivity
is not clear at present.
During our study on the Z,E-photoisomerization of
various kinds of ring-substituted (E,E,E)-1,6-diphenyl-
hexa-1,3,5-trienes [(E,E,E)-DPHs] in solution,12 we have
accidentally found that the dimethyl ester of (Z,E,Z)-DPH
4,4′-dicarboxylic acid, (Z,E,Z)-1a, underwent photoisomer-
ization in the solid state. The ester derivatives of the acid
having longer alkyl chains than methyl were expected
to be easily obtained by Wittig reactions from corre-
sponding alkyl 4-formylbenzoates. Since the easiness of
the atomic movements of the triene carbons in the crystal
should strongly be affected by the alkyl chain length, the
investigation of the Z,E-photoisomerization behaviors
and crystal structures for these esters would afford new
insight into the structure-property relationship for the
crystalline-state Z,E-isomerization. In this study, single
crystals of a series of (Z,E,Z)-DPH 4,4′-dicarboxylic acid
dialkyl esters, (Z,E,Z)-1a-d (Scheme 1), were prepared
and their crystal structures and photoisomerization
behaviors were systematically investigated.
Although aldehydes 2a-d had no ring substituent at
the ortho positions to the formyl group, the products of
the Wittig reactions were rich in Z isomers. It is known
that the Wittig reactions of reactive aldehydes and
phosphonium salts tend to give increased amount of Z
isomers in the products.13 In the present cases, the
reactions were considerably fast because of the high
reactivity of 2a-d having electron-withdrawing alkyl
ester groups. The enhanced reaction rates probably led
to the favorable formation of (Z,E,Z)-1a-d by kinetic
control.
2. Crystalline-State Z,E-Photoisomerization of
(Z,E,Z)-1a-d. Irradiation of the single crystals of (Z,E,Z)-
1a-d in air at room temperature induced Z,E-isomer-
ization. The structures of the photoproducts were deter-
mined to be their E,E,E isomers from NMR, IR, and UV-
vis spectra (Scheme 1). The photoreaction of (Z,E,Z)-1a
was very clear. Compound (E,E,E)-1a was shown to be
an exclusive product by HPLC. No [2+2] cycloaddition
was observed. For (Z,E,Z)-1b-d, trace amounts of un-
known compounds were detected in the photoproducts.
Figure 1 shows the solid-state UV-vis spectra before
and after irradiation of (Z,E,Z)-1a. A weak and broad
absorption was seen in the region of 340-450 nm before
irradiation, whereas after irradiation, a strong and
structured absorption was observed around 420 nm. It
is clear that photoproduct (E,E,E)-1a absorbed the ir-
radiation light of wavelength λ > 340 nm. However, the
E,E,E isomer was photostable in the solid state. This led
to Z,E,ZfE,E,E “one-way” isomerization.
Results and Discussion
1. Preparation of (Z,E,Z)-1a-d. Compounds (Z,E,Z)-
1a-d were prepared by the Wittig reactions of aldehydes
2a-d and bis-phosphonium salt 3 (Scheme 1). The
reactions gave Z-E isomeric mixtures of 1a-d, from
which crude (Z,E,Z)-1a-d were isolated (Experimental
section). Single crystals of (Z,E,Z)-1a suitable for X-ray
analysis were obtained by recrystallization of the crude
compound from acetonitrile, and the crystals of (Z,E,Z)-
1b-d were grown from their acetonitrile solutions by
slow evaporation at room temperature. The structures
of (Z,E,Z)-1a-d were confirmed by NMR, IR, and UV-
vis absorption spectroscopy. The Z,E,Z isomers of DPHs
have been unknown compounds as far as we know. They
are not photochemically produced from the E,E,E isomers
in solution.12 Compounds (Z,E,Z)-1a-d were stable in the
solid state at least for a year in the dark at room
temperature.
The photoproducts from (Z,E,Z)-1a-d were analyzed
by HPLC after various irradiation times. No Z,E,E isomer
was found in the reaction mixtures even in the very early
stages of the reactions. This shows that the reactions
are not stepwise Z,E,ZfZ,E,EfE,E,E but concerted
Z,E,ZfE,E,E isomerization, in which simultaneous ZfE
isomerization occurs around the two Z double bonds in
the Z,E,Z isomers. Irradiation of (Z,E,Z)-1a with light of
(11) Odani, T.; Matsumoto, A.; Sada, K.; Miyata, M. J. Chem. Soc.,
Chem. Commun. 2001, 2004.
(12) (a) Sonoda, Y.; Suzuki, Y. J. Chem. Soc., Perkin Trans. 2 1996,
401. (b) Sonoda, Y.; Suzuki, Y. Chem. Lett. 1996, 659. (c) Sonoda, Y.;
Morii, H.; Sakuragi, M.; Suzuki, Y. Chem. Lett. 1998, 349. (d) Sonoda,
Y.; Kawanishi, Y.; Sakuragi, M. Chem. Lett. 1999, 587. (e) Sonoda, Y.;
Kwok, W. M.; Petrasek, Z.; Ostler, R.; Matousek, P.; Towrie, M.; Parker,
A. W.; Phillips, D. J. Chem. Soc., Perkin Trans. 2 2001, 308. (f) Sonoda,
Y.; Kawanishi, Y. Chem. Lett. 2003, 32, 978.
(13) Maercker, A. Org. React. 1965, 14, 270.
9756 J. Org. Chem., Vol. 70, No. 24, 2005

Chain Length Effects on Crystal Structure
attributed to the fact that the pale yellow, transparent
needles of (Z,E,Z)-1a became yellow powder on irradia-
tion; the irradiation light did not reach the inner part of
the crystals at the later stages of the reaction. The
photoisomerization of (Z,E,Z)-1a occurred efficiently
when its solution cast film was irradiated. No introduc-
tion time was seen for the film, and the reaction was
completed after 4 h.
In contrast to the case of (Z,E,Z)-1a, the conversion
reached 100% for (Z,E,Z)-1b-d (Figure 2). This corre-
sponds to the fact that the (Z,E,Z)-1b-d crystals stayed
almost transparent and therefore the irradiation light
was able to reach the inner part of the crystals during
the reactions. The introduction time for (Z,E,Z)-1b was
ca. 0.5 h, which was considerably shorter than that for
(Z,E,Z)-1a. No introduction time was observed for (Z,E,Z)-
1c and (Z,E,Z)-1d.
3. Crystal Structures. Crystal Structures of (Z,E,Z)-
1a-d. Table 1 shows the major structural parameters
of (Z,E,Z)-1a-d obtained by X-ray crystal structure
analyses. Their crystal data are summarized in Support-
ing Information Table S1.
FIGURE 1. UV-vis absorption spectra for the cast film of
(Z,E,Z)-1a (a) before and (b) after irradiation. Irradiation
time: 240 min.
Single-crystal X-ray analyses showed that molecules
of (Z,E,Z)-1a-d were centrosymmetric. Figure 3 shows
their ORTEP drawings. The structures of DPH moieties
for (Z,E,Z)-1a-d were similar. The triene chains and
aromatic rings were each planar for conjugation. The
most characteristic features of the structures of these
molecules are the twists around C3-C4 single bonds. The
torsion angles (φ) of C2-C3-C4-C5 were (40-43°
(Table 1). The twists around C3-C4 bonds largely
contribute to the decrease in the steric hindrance between
C1-H and C5-H hydrogens. The angles φ in (Z,E,Z)-
1a-d are similar to the corresponding values in other
(Z)-arylethylenes such as (Z)-1,2-di(1-naphthyl)ethylene
(φ ) 44.1°)9 and (Z)-dibenzo[e,e′]-3,3′-diazastilbene (39°).14
Thus, the molecules of (Z,E,Z)-1a-d are twisted and
bulky in the crystalline state, similar to other compounds
showing the crystalline-state Z,E-photoisomerization.7
The conformation of the methylene chains were all-trans
in (Z,E,Z)-1b-d.
FIGURE 2. Time profile of the conversion for the photoi-
somerization of (Z,E,Z)-1a-d single crystals. (9) 1a, (2) 1b,
(O) 1c, and (b) 1d.
reduced intensity using a neutral density filter resulted
in the considerable decrease in the rate of the reaction.
Also in this case, however, no (Z,E,E)-1a was found in
the reaction mixture.
When the crystals of (Z,E,Z)-1a were heated above its
mp, thermal isomerization to (E,E,E)-1a occurred very
slowly. When the crystals of (Z,E,Z)-1b-d were heated
slightly above their mps, on the other hand, no amount
or only a trace amount of the corresponding E,E,E
isomers was detected. For (E,E,E)-1a-d, no thermal
isomerization was observed even after heating above
their mps.
Figure 2 shows the relationship between the irradia-
tion time and the conversion for the reactions of (Z,E,Z)-
1a-d single crystals. It is found that the reaction
efficiency was strongly enhanced as the alkyl chain
length increased.
Although the isomerization of (Z,E,Z)-1a was inef-
ficient at its early stage, at the later stages, the reaction
occurred more efficiently. The “introduction time” of 1-2
h was thus observed for the reaction. The conversion was
lower than 50% even after being irradiated for 4 h or
longer. This incompleteness of the photoreaction can be
The structures of (Z,E,Z)-1a and (Z,E,Z)-1d were
optimized by the HF/6-31G* level ab initio calculations.
The calculated geometrical parameters including the
torsion angles φ agree fairly well with those in the
crystals.
Figure 4 shows the projections from c axes for (Z,E,Z)-
1a-d. In these crystals, the molecules were stacked
parallel along b axes with π-conjugation planes. The
packing patterns of the alkyl chains in (Z,E,Z)-1a and
(Z,E,Z)-1c were different from those in (Z,E,Z)-1b and
(Z,E,Z)-1d. The crystal systems of (Z,E,Z)-1a and (Z,E,Z)-
1c were P21/c, whereas those of (Z,E,Z)-1b and (Z,E,Z)-
1d were C2/c (Table S1). In Figure 4, two crystallographic
repeating units were shown for (Z,E,Z)-1a and (Z,E,Z)-
1c to compare with the cases of (Z,E,Z)-1b and (Z,E,Z)-
1d. The lattice constant along the long molecular axis
increased as the alkyl chain length increased; 1a: 2a )
31.7 Å, 1b: a ) 35.3 Å, 1c: 2a ) 39.3 Å, and 1d: a )
44.9 Å. The values of b and c were similar for all four
compounds. Although the crystal systems of (Z,E,Z)-1a
(14) Vansant, J.; Smets, G.; Declercq, J. P.; Germain, G.; Van
Meerssche, M. J. Org. Chem. 1980, 45, 1557.
J. Org. Chem, Vol. 70, No. 24, 2005 9757

Sonoda et al.
TABLE 1. Major Structural Parametersa for (Z,E,Z)-1a-d Obtained from X-ray Crystal Structure Analyses
parameter
(Z,E,Z)-1a
(Z,E,Z)-1b
(Z,E,Z)-1c
(Z,E,Z)-1d
C1-C2
1.440(2)
1.469(2)
1.348(3)
1.342(2)
1.397(2)
1.396(2)
117.8(1)
123.5(2)
128.4(1)
127.9(1)
-1.5(3)
-41.0(2)
-179.9(2)
-0.2(2)
-1.8(2)
178.3(1)
1.445(3)
1.472(2)
1.358(3)
1.344(3)
1.399(2)
1.395(2)
118.1(2)
123.1(2)
128.0(2)
127.8(2)
1.2(3)
1.444(2)
1.470(2)
1.340(3)
1.344(2)
1.402(2)
1.396(2)
117.8(1)
123.7(2)
128.3(1)
127.7(1)
1.3(2)
1.447(3)
1.473(3)
1.351(4)
1.349(3)
1.404(2)
1.402(2)
117.6(2)
123.5(2)
128.3(2)
128.0(2)
-1.3(3)
C3-C4
C1-C1i
C2-C3
C4-C9
C4-C5
C5-C4-C9
C1i-C1-C2
C1-C2-C3
C2-C3-C4
C1-C2-C3-C4
C2-C3-C4-C5
C1i-C1-C2-C3
C6-C7-C10-O1
O1-C10-O2-C11
C7-C10-O2-C11
C10-O2-C11-C12
O2-C11-C12-C13
C11-C12-C13-C14
42.2(3)
42.6(2)
-40.8(3)
179.8(2)
-4.9(3)
-179.9(2)
5.8(3)
0.0(4)
178.0(2)
89.8(3)
180.0(2)
6.8(2)
1.2(2)
-179.5(1)
91.8(2)
177.2(1)
-2.3(3)
-179.2(2)
-104.6(2)
-175.9(2)
172.0(3)
1
1
a In Å and degrees; for the numbering of atoms, see Figure 3. Symmetry codes: 1a(i) -x, -y + 2, -z + 2; 1b (i) -x + /2, -y + /2, -z
+ 2; 1c (i)-x + 1, -y + 2, -z + 1; 1d (i) -x + 1/2, -y + 1/2, -z + 2.
Powder X-ray Diffraction (XRD) Patterns before
and after Irradiation. Figure 5a shows the powder
XRD pattern of (Z,E,Z)-1a before irradiation. The pattern
was in good agreement with the one calculated from the
single-crystal data. On irradiation of (Z,E,Z)-1a, the
intensity of the original peaks decreased and some new
peaks appeared. The reaction was thus found to be
crystal-to-crystal transformation. The positions of the
new peaks of the photoproduced (E,E,E)-1a were identical
to those for the “solution-grown” (E,E,E)-1a obtained by
crystallization from toluene (see Experimental section)
(Figure 5b). At present, however, the single-crystal X-ray
analysis of neither the photoproduced nor the solution-
grown (E,E,E)-1a is unsuccessful. Figure 5c shows the
XRD pattern of the authentic (Z,E,E)-1a. The character-
istic peak of (Z,E,E)-1a at 2θ ) 7.62° was not observed
during the isomerization, which is in accordance with the
observation in the HPLC analysis.
Figure 6a shows the XRD pattern calculated from the
single-crystal data of (Z,E,Z)-1d before irradiation. Figure
7 shows the pattern of (E,E,E)-1d measured after ir-
radiation. For the photoproduced (E,E,E)-1d, the lattice
constants of a and b were estimated from the figure to
be 44 and 4.4 Å, respectively. Interestingly, they were
near the values of a ) 44.904(4) Å and b ) 4.6750(4) Å
FIGURE 3. Views of the molecular structures of (Z,E,Z)-1a-
for (Z,E,Z)-1d before irradiation. Figure 6b shows the
calculated pattern based on the single-crystal data of the
solution-grown (E,E,E)-1d obtained by recrystallization
from toluene (see Experimental section). In contrast to
the case of (E,E,E)-1a, the XRD pattern of the photopro-
duced (E,E,E)-1d was clearly different from that of its
solution-grown crystals.
Crystal Structure of (E,E,E)-1d. Although the single-
crystal X-ray analysis of the photoproduced (E,E,E)-1d
was unsuccessful, the crystal data were able to be
obtained for the solution-grown (E,E,E)-1d. Figures 8 and
9 show the ORTEP drawing and the packing diagram
for the solution-grown (E,E,E)-1d, respectively. The
crystal data and major structure parameters are listed
in Table 2 and Supporting Information Table S2. The
molecules were shown to be nearly planar, and the
conformation of the methylene chains was all-trans.
d. (a) 1a, (b) 1b, (c) 1c, and (d) 1d. Displacement ellipsoids
are drawn at the 50% probability level and H atoms are shown
as small spheres of arbitrary radii. [Symmetry code: (*) -x+1,
-y, -z+1.]
and (Z,E,Z)-1c were different from those of (Z,E,Z)-1b
and (Z,E,Z)-1d, no even-odd effect was observed in their
photoreactions.
The alkyl chain part of the crystals became larger as
the chain lengths increased. For (Z,E,Z)-1a, a large part
of the crystals was occupied by the π-stacked rigid
conjugated triene and aromatic rings, whereas for (Z,E,Z)-
1d, the flexible alkyl chains occupied a large part of the
crystals. Values of mp and density decreased as the chain
length increased, indicating that the strength of inter-
molecular interactions in the crystals decreased as the
chain length increased (Table S1).
9758 J. Org. Chem., Vol. 70, No. 24, 2005

Chain Length Effects on Crystal Structure
FIGURE 4. Packing diagrams for (Z,E,Z)-1a-d along the c axes. (a) 1a, (b) 1b, (c) 1c, and (d) 1d.
These structural features of the solution-grown (E,E,E)-
1d are in good agreement with those predicted by ab
initio calculation. The lattice constants for the solution-
grown (E,E,E)-1d (Table S2) were very different from the
value of a ) 44 Å for the photoproduced (E,E,E)-1d. The
crystal structure of the photoproduced (E,E,E)-1d was
thus entirely different from that of the solution-grown
(E,E,E)-1d.
4. Relationship between the Crystal Structure
and the Photoisomerization Reactivity. Since the
large part of the (Z,E,Z)-1a crystals was occupied by rigid
conjugated double bonds and aromatic rings (Figure 4a),
molecules were difficult to move in the lattice. As a result
of this, the photoreaction was inefficient at least at its
earlier stage. However, once one of the (Z,E,Z)-1a mol-
ecules isomerized to (E,E,E)-1a in the lattice, the large
change in molecular structure was expected to lead to
the formation of structural defects in its neighborhood.
The molecular motions became easier by such defects,
and the reaction occurred more easily at its later stage.
The enhanced photoreaction efficiency relative to that for
the crystals and no introduction time for the cast film
are easily understood by assuming that the crystal lattice
of (Z,E,Z)-1a was partially destroyed before irradiation
by dissolving the compound to prepare the solution cast
film.
J. Org. Chem, Vol. 70, No. 24, 2005 9759

Sonoda et al.
FIGURE 6. Powder XRD patterns calculated from single-
crystal X-ray data. (a) (Z,E,Z)-1d before irradiation and (b)
solution-grown (E,E,E)-1d obtained by recrystallization from
toluene.
FIGURE 5. Powder XRD patterns of (a) Z,E,Z, (b) E,E,E, and
(c) Z,E,E isomers of 1a.
In the case of (Z,E,Z)-1d, the flexible n-butyl chains
occupied a large part of the crystals (Figure 4d). The
flexibility of the alkyl chains made the large change in
the molecular geometry required by the isomerization
possible, leading to the enhanced photoreactivity in the
crystalline state. The fact that the lattice constant a )
44.904(4) Å for (Z,E,Z)-1d was near the value of a ) 44
Å for the photoproduced (E,E,E)-1d suggests that the
molecular length along the long axis did not change
largely before and after the photoreaction. It is probable
that the conformational changes in the alkyl chains play
a key role in this case. The rigid DPH framework is
expected to be nearly planar in the crystals of the
photoproduced (E,E,E)-1d as well in those of the solution-
grown (E,E,E)-1d (Figure 9). On the other hand, the
methylene chains are flexible, and its conformation can
at least partially be gauche in the photoproduced (E,E,E)-
1d.
FIGURE 7. Powder XRD patterns measured after irradiation
of (Z,E,Z)-1d.
conformers for n-butyl benzoic acid were calculated at
the HF/6-311G* level. The calculations showed that some
conformers having gauche bonds were only 1-2 kcal/mol
less stable than the most stable all-trans conformer. It
is very probable that the n-butyl chains in (E,E,E)-1d
have some gauche bonds after the photoisomerization in
the crystal, as the energy change associated with the
conformational change of the n-butyl chains is not very
large.
To evaluate the energy differences among various
n-butyl conformers of (E,E,E)-1d, relative energies of
The molecular lengths along the long axes for (Z,E,Z)-
1d and (E,E,E)-1d with all-trans butyl groups were
9760 J. Org. Chem., Vol. 70, No. 24, 2005

Chain Length Effects on Crystal Structure
FIGURE 8. A view of the molecular structure of solution-grown (E,E,E)-1d obtained by recrystallization. Displacement ellipsoids
are drawn at the 50% probability level and H atoms are shown as small spheres of arbitrary radii.
FIGURE 9. A packing diagram along the b axis for solution-grown (E,E,E)-1d obtained by recrystallization.
TABLE 2. Major Structural Parametersa for (E,E,E)-1db
Obtained from X-ray Crystal Structure Analysis
C8-C9
1.428(3) C4-C7-C8-C9
1.459(3) C5-C4-C7-C8
1.339(3) C7-C8-C9-C10
1.338(3) C6-C1-C19-O1
1.397(3) O1-C19-O2-C20
1.402(3) C1-C19-O2-C20
-179.4(2)
3.9(4)
-177.3(2)
-11.2(4)
1.4(3)
-178.2(2)
-171.7(2)
-179.9(2)
C4-C7
C9-C10
C7-C8
C3-C4
C4-C5
C3-C4-C5
117.7(2)
C19-O2-C20-C21
O2-C20-C21-C22
C8-C9-C10 124.9(2)
FIGURE 10. One of the possible gauche conformers of
(E,E,E)-1d; compare with its all-trans butyl conformer shown
in Figure 8.
C7-C8-C9
C4-C7-C8
123.8(2)
126.7(2)
C20-C21-C22-C23 -177.1(2)
a In Å and degrees; for the numbering of atoms, see Figure 8.
b Obtained by recrystallization from toluene.
Experimental Section
Chemicals. Methyl 4-formylbenzoate (2a) was purchased
and used without further purification. Ethyl, n-propyl, and
n-butyl 4-formylbenzoates (2b, 2c, and 2d, respectively) were
synthesized by similar procedures to that for n-octyl 4-formyl-
benzoate.15 The purity of 2b-d was checked by 1H NMR
spectra.16,17 Bis-triphenylphosphonium salt 3 was prepared
according to the literature.18
(Z,E,Z)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic
Acid Dimethyl Ester [(Z,E,Z)-1a]. To a solution of methyl
4-formylbenzoate 2a (1.50 g, 9.14 mmol) and salt 3 (2.97 g,
4.57 mmol) in ethanol (35 mL) was added a solution of sodium
ethoxide in ethanol (0.30 M, 30 mL). The mixture was stirred
under nitrogen atmosphere at room temperature. After stirring
for 3 days, the resulting pale yellow precipitate was filtered
off, washed with water (50 mL), and dried. The single crystals
of (Z,E,Z)-1a were obtained by recrystallization of the crude
product from acetonitrile. Yield 38%. Mp 172-173 °C; 1H NMR
(CDCl3) δ 8.03 (4H, d, J ) 8.6 Hz, arom.), 7.41 (4H, d, J ) 8.2
Hz, arom.), 6.87-6.99 (2H, m, triene 3-H and 4-H), 6.52 (2H,
apparently (app.) d, J ) 11.5 Hz, triene 1-H and 6-H), 6.32-
estimated from ab initio calculations to be 25.87 Å and
27.95 Å, respectively. Thus, the increase in the molecular
length that accompanied the triene Z,E,ZfE,E,E isomer-
ization was predicted to be ca. 2.1 Å. On the other hand,
the conformational analyses of n-butyl benzoic acid
showed that the decrease in the molecular length that
accompanied the transfgauche conformational change
in the butyl group was ca. 1 Å for one of the possible
gauche conformers, whose energy was higher by only 1
kcal/mol than the all-trans conformer. It can therefore
be expected that the transfgauche changes in the two
n-butyl groups of (E,E,E)-1d decrease the total molecular
length of ca. 2 Å (Figure 10). Thus, the decrease in
molecular length because of the butyl conformational
changes can fully compensate for the increase in molec-
ular length caused by the triene geometrical isomeriza-
tion. Although s-transfs-cis conformational changes in
the triene can also lead to the decrease in the total
molecular length of (E,E,E)-1d, the observed strong
dependence of the photoreactivity on the chain length
strongly suggests that the conformational changes in the
alkyl groups are responsible for the highly efficient
photoisomerization of (Z,E,Z)-1d in the limited space of
the crystal lattice.
(15) Shao, X.-B.; Jiang, X.-K.; Wang, X.-Z.; Li, Z.-T.; Zhu, S.-Z.
Tetrahedron 2003, 59, 4881.
(16) Konya, K. G.; Paul, T.; Lin, S.; Lusztyk, J.; Ingold, K. U. J.
Am. Chem. Soc. 2000, 122, 7518.
(17) Lindsey, J. S.; Schreiman, I. C.; Hsu, H. C.; Kearney, P. C.;
Marguerettaz, A. M. J. Org. Chem. 1987, 52, 827.
(18) McDonald, R. N.; Campbell, T. W. J. Org. Chem. 1959, 24, 1969.
J. Org. Chem, Vol. 70, No. 24, 2005 9761

Sonoda et al.
6.48 (2H, m, triene 2-H and 5-H), 3.93 (6H, s, -CH3); 13C NMR
(CDCl3) δ 166.9 (CdO), 142.0 (arom. C-1), 132.1 (triene C-2
and C-5), 131.8 (triene C-3 and C-4), 130.2 (triene C-1 and C-6),
129.6 (arom. C-3 and C-5), 128.9 (arom. C-2 and C-6), 128.6
(arom. C-4), 52.1 (CH3); IR (KBr) νmax 1718, 1605, 1415, 1282,
1170, 1110, 1001, 950, 869, and 784 cm-1; UV-vis (MeCN)
λmax (ꢀ) 353 nm (41400); HR-MS calcd for C22H20O4: 348.1361,
found: 348.1375. Anal. calcd for C22H20O4: C, 75.84; H, 5.79.
Found: C, 76.01; H, 5.49.
C-3 and C-5), 129.0 (arom. C-4), 128.9 (arom. C-2 and C-6),
61.0 (CH2), 14.3 (CH3); IR (KBr) νmax 2986, 1723, 1607, 1277,
1170, 1104, 1002, 950, 870, 792, and 738 cm-1; UV-vis (MeCN)
λmax (ꢀ) 352 nm (54100); HR-MS calcd for C24H24O4: 376.1672,
found 376.1630.
(E,E,E)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic
Acid Diethyl Ester [(E,E,E)-1b]. Compound (E,E,E)-1b was
obtained by recrystallization of the crude product from 2b and
3, as described above. Mp 184-186 °C; 1H NMR (CDCl3) δ 8.00
(4H, d, J ) 8.2 Hz, arom.), 7.47 (4H, d, J ) 8.2 Hz, arom.),
6.91-7.08 (2H, m, triene 2-H and 5-H), 6.65 (2H, app. d, J )
15.2 Hz, triene 1-H and 6-H), 6.54-6.65 (2H, m, triene 3-H
and 4-H), 4.38 (4H, q, J ) 7.1 Hz, -CH2-), 1.40 (6H, t, J )
7.1 Hz, -CH3); 13C NMR (CDCl3) δ 166.4 (CdO), 141.5 (arom.
C-1), 134.6, 132.4, and 131.2 (triene), 130.0 and 126.2 (arom.
C-2, C-3, C-5, and C-6), 129.3 (arom. C-4), 60.9 (CH2), 14.3
(CH3); IR (KBr) νmax 2918, 1717, 1603, 1416, 1282, 1178, 1107,
997, 876, 837, 765, and 697 cm-1; UV-vis (MeCN) λmax (ꢀ) 376
nm (81500); HR-MS calcd for C24H24O4: 376.1672, found
376.1670.
(Z,E,Z)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic
Acid Di(n-propyl) Ester [(Z,E,Z)-1c]. To a solution of
n-propyl 4-formylbenzoate 2c (1.00 g, 5.20 mmol) and 3 (1.76
g, 2.71 mmol) in ethanol (15 mL) was added a solution of
sodium ethoxide in ethanol (0.30 M, 17.3 mL). The mixture
was stirred under nitrogen atmosphere at room temperature.
After stirring for 20 h, the resulting yellow precipitate was
filtered off, washed with water (30 mL), and dried. The
ethanol-insoluble yellow solid was shown to be a 1:1 mixture
of (Z,E,Z)-1c and (E,E,E)-1c by NMR. The isomeric mixture
was partially dissolved in a small portion of ethanol and was
filtered. The ethanol filtrate was cooled to 0 °C (ice bath), and
the resulting yellow solid ((Z,E,Z)-1c) was filtered off and
dissolved in acetonitrile. Single crystals of (Z,E,Z)-1c were
grown from this solution by very slow evaporation at room
temperature in dark. Yield 15%. Mp 100-101 °C; 1H NMR
(CDCl3) δ 8.04 (4H, d, J ) 8.2 Hz, arom.), 7.41 (4H, d, J ) 8.2
Hz, arom.), 6.87-6.99 (2H, m, triene 3-H and 4-H), 6.52 (2H,
app. d, J ) 11.2 Hz, triene 1-H and 6-H), 6.32-6.48 (2H, m,
triene 2-H and 5-H), 4.30 (4H, t, J ) 6.8 Hz, -OCH2-), 1.75-
1.88 (4H, m, -CH2Me), 1.05 (6H, t, J ) 7.4 Hz, -CH3); 13C
NMR (CDCl3) δ 166.4 (CdO), 141.9 (arom. C-1), 132.1 (triene
C-2 and C-5), 131.7 (triene C-3 and C-4), 130.2 (triene C-1 and
C-6), 129.6 (arom. C-3 and C-5), 129.0 (arom. C-4), 128.9 (arom.
C-2 and C-6), 66.6 (OCH2), 22.1 (CH2Me), 10.5 (CH3); IR (KBr)
νmax 2966, 1723, 1606, 1276, 1170, 1103, 1003, 950, 869, and
785 cm-1; UV-vis (MeCN) λmax (ꢀ) 353 nm (58000); HR-MS
calcd for C26H28O4: 404.1986, found 404.2003.
(1Z,3E,5E)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicar-
boxylic Acid Dimethyl Ester [(Z,E,E)-1a]. To the ethanol-
soluble part of the crude product from 2a and 3 was added
water to give a 60% v/v aq ethanol solution. After this solution
had been vigorously stirred, the resulting precipitate was
filtered, washed with aq ethanol (60% v/v), and dried at room
temperature. The Z,E,E isomer of 1a was obtained as yellow
solid. The purity was checked by HPLC. Yield 21%. Mp 87-
89 °C; 1H NMR (CDCl3) δ 8.04 (2H, d, J ) 8.2 Hz, arom.), 7.98
(2H, d, J ) 8.6 Hz, arom.), 7.46 (2H, d, J ) 8.6 Hz, arom.),
7.42 (2H, d, J ) 8.2 Hz, arom.), 6.89-7.02 (2H, m, triene 3-H
and 5-H), 6.65 (1H, d, J ) 15.5 Hz, triene 6-H), 6.59-6.65 (1H,
m, triene 4-H), 6.53 (1H, d, J ) 12.2 Hz, triene 1-H), 6.43 (1H,
app. t, J ) 11.1 Hz, triene 2-H), 3.94 (3H, s, -CH3), 3.91 (3H,
s, -CH3); 13C NMR (CDCl3) δ 166.9 and 166.8 (CdO), 142.1
and 141.6 (arom. C-1), 136.3, 132.6, 131.8, 131.4, 130.5, and
129.9 (triene), 130.0, 129.7, 128.9, and 126.3 (arom. C-2, C-3,
C-5, and C-6), 129.0 and 128.6 (arom. C-4), 52.14 and 52.08
(CH3); IR (KBr) νmax 1714, 1602, 1434, 1279, 1177, 1106, 993,
960, 871, and 784 cm-1; UV-vis (MeCN) λmax (ꢀ) 367 nm
(55700); HR-MS calcd for C22H20O4: 348.1361, found 348.1382.
Compound (Z,E,E)-1a was somewhat unstable in air at room
temperature; its UV-vis spectrum significantly changed after
standing for 2-3 months in the dark.
(E,E,E)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic
Acid Dimethyl Ester [(E,E,E)-1a]. A saturated solution of
(Z,E,Z)-1a or (Z,E,E)-1a in toluene was irradiated with Pyrex-
filtered light. The resulting solution-grown crystals of (E,E,E)-
1a were collected and dried at room temperature. Mp 250-
1
251 °C (lit.,19 256-259 °C); H NMR (CDCl3) δ 7.99 (4H, d, J
) 8.6 Hz, arom.), 7.47 (4H, d, J ) 8.2 Hz, arom.), 6.91-7.09
(2H, m, triene 2-H and 5-H), 6.66 (2H, app. d, J ) 15.5 Hz,
triene 1-H and 6-H), 6.53-6.65 (2H, m, triene 3-H and 4-H),
3.92 (6H, s, -CH3); 13C NMR (CDCl3) δ 166.8 (CdO), 141.7
(arom. C-1), 134.6, 132.4, and 131.3 (triene), 130.0 and 126.3
(arom. C-2, C-3, C-5, and C-6), 128.9 (arom. C-4), 52.1 (CH3);
IR (KBr) νmax 2921, 1720, 1602, 1282, 1178, 1108, 995, 878,
838, 765, and 699 cm-1; UV-vis (MeCN) λmax (ꢀ) 376 nm
(78600); HR-MS calcd for C22H20O4: 348.1361, found 348.1391.
(Z,E,Z)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic
Acid Diethyl Ester [(Z,E,Z)-1b]. To a solution of ethyl
4-formylbenzoate 2b (1.00 g, 5.61 mmol) and 3 (1.82 g, 2.80
mmol) in ethanol (30 mL) was added a solution of sodium
ethoxide in ethanol (0.30 M, 18.7 mL). The mixture was stirred
under nitrogen atmosphere at room temperature. After stirring
for 24 h, the resulting pale yellow precipitate was filtered off,
washed with water (50 mL), and dried. The ethanol-insoluble
yellow solid was shown from NMR spectra to be a 6:4 mixture
of (Z,E,Z)-1b and (E,E,E)-1b. This solid was recrystallized from
acetonitrile to give pure (E,E,E)-1b. The mother liquid mainly
contained (Z,E,Z)-1b, from which the solvent acetonitrile
evaporated very slowly at room temperature in dark to give
the single crystals of (Z,E,Z)-1b. Yield 21%. Mp 140-142 °C;
1H NMR (CDCl3) δ 8.04 (4H, d, J ) 8.2 Hz, arom.), 7.41 (4H,
d, J ) 8.2 Hz, arom.), 6.86-6.99 (2H, m, triene 3-H and 4-H),
6.52 (2H, app. d, J ) 11.5 Hz, triene 1-H and 6-H), 6.31-6.47
(2H, m, triene 2-H and 5-H), 4.39 (4H, q, J ) 7.0 Hz, -CH2-),
1.41 (6H, t, J ) 7.3 Hz, -CH3); 13C NMR (CDCl3) δ 166.4
(CdO), 141.9 (arom. C-1), 132.1 (triene C-2 and C-5), 131.7
(triene C-3 and C-4), 130.2 (triene C-1 and C-6), 129.6 (arom.
(E,E,E)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic
Acid Di(n-propyl) Ester [(E,E,E)-1c]. The isomeric mixture
of (Z,E,Z)-1c and (E,E,E)-1c was irradiated in toluene with
Pyrex-filtered light. The resulting crystals of (E,E,E)-1c were
collected by filtration, recrystallized from toluene, and dried
at room temperature. Mp 143-145 °C; 1H NMR (CDCl3) δ 8.00
(4H, d, J ) 8.6 Hz, arom.), 7.47 (4H, d, J ) 8.2 Hz, arom.),
6.90-7.08 (2H, m, triene 2-H and 5-H), 6.65 (2H, app. d, J )
15.5 Hz, triene 1-H and 6-H), 6.53-6.65 (2H, m, triene 3-H
and 4-H), 4.28 (4H, t, J ) 6.6 Hz, -OCH2-), 1.73-1.87 (4H,
m, -CH2Me), 1.04 (6H, t, J ) 7.4 Hz, -CH3); 13C NMR (CDCl3)
δ 166.4 (CdO), 141.5 (arom. C-1), 134.6, 132.4, and 131.2
(triene), 130.0 and 126.2 (arom. C-2, C-3, C-5, and C-6), 129.3
(arom. C-4), 66.5 (OCH2), 22.1 (CH2Me), 10.5 (CH3); IR (KBr)
νmax 2974, 1714, 1602, 1415, 1280, 1177, 1109, 1000, 877, 837,
767, 698 cm-1; UV-vis (MeCN) λmax (ꢀ) 377 nm (94600);
HR-MS calcd for C26H28O4: 404.1986, found 404.2006.
(Z,E,Z)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic
Acid Di(n-butyl) Ester [(Z,E,Z)-1d]. To a solution of n-butyl
4-formylbenzoate 2d (2.00 g, 9.70 mmol) and 3 (4.50 g, 6.93
mmol) in DMF (60 mL) was added a solution of sodium
ethoxide in ethanol (0.60 M, 23 mL). The mixture was stirred
under nitrogen atmosphere at room temperature. After stirring
for 24 h, water (60 mL) was added to the reaction mixture,
(19) Mitsudo, T.; Fischetti, W.; Heck, R. F. J. Org. Chem. 1984, 49,
1640.
9762 J. Org. Chem., Vol. 70, No. 24, 2005

Chain Length Effects on Crystal Structure
and the resulting pale yellow precipitate was filtered off. The
precipitate was washed with aq ethanol (60-70% v/v, 600 mL)
to remove any triphenylphosphine oxide and was dried at room
temperature. The crude product was shown by NMR to be a
2:1 mixture of (Z,E,Z)-1d and (E,E,E)-1d. The isomeric mixture
was partially dissolved in a small portion of acetonitrile. Single
crystals of (Z,E,Z)-1d were grown from this acetonitrile
solution by very slow evaporation of the solvent at room
temperature in dark. Yield 7%. Mp 96-97 °C; 1H NMR (CDCl3)
δ 8.03 (4H, d, J ) 8.2 Hz, arom.), 7.41 (4H, d, J ) 8.2 Hz,
arom.), 6.87-6.99 (2H, m, triene 3-H and 4-H), 6.52 (2H, app.
d, J ) 11.5 Hz, triene 1-H and 6-H), 6.33-6.47 (2H, m, triene
2-H and 5-H), 4.34 (4H, t, J ) 6.6 Hz, -OCH2-), 1.72-1.82
(4H, m, -CH2Et), 1.45-1.60 (4H, m, -CH2Me), 0.99 (6H, t, J
) 7.4 Hz, -CH3); 13C NMR (CDCl3) δ 166.4 (CdO), 141.9
(arom., C-1), 132.1 (triene C-2 and C-5), 131.7 (triene C-3 and
C-4), 130.2 (triene C-1 and C-6), 129.6 (arom. C-3 and C-5),
129.0 (arom. C-4), 128.9 (arom. C-2 and C-6), 64.9 (OCH2), 30.8
(CH2Et), 19.3 (CH2Me), 13.8 (CH3); IR (KBr) νmax 2960, 1722,
was proportional to the product of the isomer content and the
extinction coefficient at monitor wavelengths.
Single-Crystal X-ray Structure Analyses. The single-
crystal XRD measurements of the Z,E,Z isomers of 1a-d and
recrystallized (E,E,E)-1d were performed at 183 K using a
CCD area-detector diffractometer with graphite monochro-
mated MoKR radiation (λ ) 0.71073 Å). Data collection,
reduction, and empirical absorption correction were carried
out using SMART, SAINTPLUS, and SADABS.20 The struc-
ture was solved by direct methods using SIR9221 and was
refined by full matrix least squares on F2 with SHELXTL.22
The non-hydrogen atoms were refined anisotropically. Hydro-
gen atoms were placed in geometrically calculated positions
and were refined by a riding model.
Powder XRD Pattern Measurements. The powder X-ray
analysis data of (Z,E,Z)-1a, (Z,E,E)-1a, (E,E,E)-1a, and the
photoproduced (E,E,E)-1d were obtained with a diffractometer
with graphite monochromated Cu KR radiation.
Computational Method. The Gaussian 9823 program was
used for the ab initio molecular orbital calculations. The
6-31G*, 6-311G** basis sets were used for the calculations.
1607, 1281, 1171, 1104, 1003, 950, 869, 783, and 729 cm-1
;
UV-vis (MeCN) λmax (ꢀ) 351 nm (63600); HR-MS calcd for
C28H32O4: 432.2299, found 432.2352.
Acknowledgment. We thank Dr. Y. Nagawa (AIST)
for the measurements of two-dimensional NMR spectra
of (Z,E,Z)-1a. We also thank Dr. M. Sasaki (Rigaku) for
the measurement of powder XRD pattern of the photo-
produced (E,E,E)-1d.
(E,E,E)-1,6-Diphenylhexa-1,3,5-triene 4,4′-Dicarboxylic
Acid Di(n-butyl) Ester [(E,E,E)-1d]. A saturated solution
of an isomeric mixture of (Z,E,Z)-1d and (E,E,E)-1d in toluene
was irradiated with Pyrex-filtered light. The resulting solution-
grown crystals (predominantly E,E,E) were collected and dried
at room temperature. Recrystallization from toluene gave pure
1
Supporting Information Available: Crystallographic
data for (Z,E,Z)-1a-d and (E,E,E)-1d in CIF format; general
experimental information; analytical and spectroscopic data
(E,E,E)-1d. Mp 133-134 °C; H NMR (CDCl3) δ 7.99 (4H, d,
J ) 8.2 Hz, arom.), 7.47 (4H, d, J ) 8.2 Hz, arom.), 6.91-7.08
(2H, m, triene 2-H and 5-H), 6.65 (2H, app. d, J ) 15.5 Hz,
triene 1-H and 6-H), 6.53-6.65 (2H, m, triene 3-H and 4-H),
4.32 (4H, t, J ) 6.6 Hz, -OCH2-), 1.71-1.81 (4H, m, -CH2-
Et), 1.42-1.56 (4H, m, -CH2Me), 0.99 (6H, t, J ) 7.3 Hz,
-CH3); 13C NMR (CDCl3) δ 166.4 (CdO), 141.5 (arom. C-1),
134.6, 132.4, and 131.2 (triene), 130.0 and 126.2 (arom. C-2,
C-3, C-5, and C-6), 129.3 (arom. C-4), 64.8 (OCH2), 30.8 (CH2-
Et), 19.3 (CH2Me), 13.8 (CH3); IR (KBr) νmax 2961, 1712, 1600,
1417, 1283, 1176, 1111, 998, 879, 833, 767, and 697 cm-1; UV-
vis (MeCN) λmax (ꢀ) 377 nm (103000); HR-MS calcd for
C28H32O4: 432.2299, found 432.2319.
1
for 2b-d; H NMR spectra of (Z,E,Z)-1a-d, (Z,E,E)-1a, and
(E,E,E)-1a-d; crystal and structure refinement data for
(Z,E,Z)-1a-d and (E,E,E)-1d; results of the ab initio calcula-
tion for (Z,E,Z)-1d and (E,E,E)-1d, and the conformational
analysis of n-butyl benzoate. This material is available free of
charge via the Internet at http://pubs.acs.org.
JO051137G
(20) SMART, version 5.625; Bruker AXS: Madison, WI, 2001.
SAINTPLUS, version 6.22; Bruker AXS: Madison, WI, 2001; Sheld-
rick, G. M. SADABS; Program for scaling and correction of area,
detector data; University of Go¨ttingen: Go¨ttingen, Germany, 1998.
(21) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;
Brula, M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27,
435.
Irradiation Experiments and Photoproduct Analyses.
Irradiation experiments were carried out using a high-pressure
mercury lamp (500 W) as a light source. The light was filtered
with glass filters (λ > 340 nm). The irradiation was interrupted
every 50-60 min by standing in the dark for 10 min to prevent
any thermal reaction. The single crystals of (Z,E,Z)-1a-d were
placed between quartz plates (10 × 40 mm2) and were
irradiated. The film sample of (Z,E,Z)-1a was prepared by
casting its concentrated solution in dichloromethane on a
quartz plate in the dark and by drying at room temperature
in air. The irradiated samples were dissolved in acetonitrile
in the dark and were submitted to HPLC analysis.
The photoproducts were analyzed using an HPLC monitored
by a multichannel photodetector. The eluent was acetonitrile.
The isomer ratio in the reaction mixture was determined from
the ratio of HPLC peak area and the extinction coefficients of
the Z,E,Z and E,E,E isomers of 1a-d at the monitor wave-
length. It was confirmed by NMR that the HPLC peak area
(22) Sheldrick, G. M. SHELXTL, version 6.12; Bruker AXS: Madi-
son, WI, 2000.
(23) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. Gaussian 98, revision A. 11; Gaussian, Inc.: Pittsburgh,
PA, 1998.
J. Org. Chem, Vol. 70, No. 24, 2005 9763