Synthesis and fluorescence emission behavior of anti-[2.3](3,10)phenanthrenophane: Overlap between phenanthrene rings required for excimer formation

Article abstract of DOI:10.1246/cl.2001.970

Novel anti-[2.3](3,10)phenanthrenophane 3, prepared by the intramolecular [2 + 2] photocycloaddition of 1,3-bis(3-vinyl-10-phenanthryl)propane, exhibited monomer fluorescence. This is in a remarkable contrast with the excimer fluorescence emission from an

Full text of DOI:10.1246/cl.2001.970

970
Chemistry Letters 2001
Synthesis and Fluorescence Emission Behavior of anti-[2.3](3,10)Phenanthrenophane:
Overlap between Phenanthrene Rings Required for Excimer Formation
Yosuke Nakamura, Takahiro Fujii, and Jun Nishimura*
Department of Chemistry, Gunma University, Tenjin-cho, Kiryu, Gunma 376-8515
(Received June 21, 2001; CL-010590)
in the literature,5,6,9 to afford a mixture of syn- and anti-3 in
42% yield. The syn:anti-isomer ratio was determined as ca. 6:1
on the basis of the peak areas of H NMR spectra. Both iso-
mers were successfully separated by repeated GPC (polystyrene
column, chloroform), and no interconversion between them was
observed at room temperature for at least several months.
Novel anti-[2.3](3,10)phenanthrenophane 3, prepared by the
intramolecular [2 + 2] photocycloaddition of 1,3-bis(3-vinyl-10-
phenanthryl)propane, exhibited monomer fluorescence. This is
in a remarkable contrast with the excimer fluorescence emission
from anti-[2.3](3,9)phenanthrenophane 2 with slightly larger
overlap between phenanthrene rings than in anti-3.
1
Among a series of aromatic hydrocarbons, phenanthrene is
unique in its failure to give excimer fluorescence in solution at
room temperature.1 Even 1,3-diphenanthrylpropanes mainly
afford monomer fluorescence,2 on the contrary to Hirayama’s
rule.3 We have succeeded in the first observation of excimer flu-
orescence almost free from monomer fluorescence at room tem-
perature for syn-[2.2](1,6)phenanthrenophanes whose phenan-
threne rings are held almost in parallel.4 For phenanthrene, how-
ever, the relationship between the arrangement of aromatic nuclei
and fluorescence behavior has been hardly clarified, mainly due
to the difficulty in the synthesis and separation of appropriate
model compounds. The fluorescence for anti-phenanthreno-
phanes with partially overlapped phenanthrene rings had been
unknown before our recent research on anti-[2.3](2,7)phenan-
threnophane 15 and anti-[2.3](3,9)phenanthrenophane 2,6
although Staab et al. observed excimer fluorescence from a mix-
ture of syn- and anti-[2.2](2,7)phenanthrenophanes in a fluorene
host crystal at 4.2 K.7,8 Intriguingly, even anti-2 where the two
phenanthrene rings overlap mainly at the single six-membered
ring afforded broad and structureless excimer fluorescence blue-
shifted relative to that of the syn-isomer.6 This result raised a
question what degree of overlap is essential or sufficient for the
excimer formation of phenanthrene. In order to answer the ques-
tion, we have designed anti-[2.3](3,10)phenanthrenophane 3,
which is expected to have further small overlap than anti-2,
according to MM2 calculation. Here, we report the preparation
and fluorescence behavior of anti-3.
The structures of syn- and anti-3 were determined mainly
1
by H NMR spectroscopy in a manner similar to 2.10 In syn-3,
only eight sets of aromatic proton peaks were observed and
generally high-field shifted compared to those of precursor 4.
The two methine protons of the cyclobutane ring appeared as an
equivalent peak. These results apparently indicate the syn-con-
formation where the two chromophores are well overlapped.
The configuration of cyclobutane ring was determined as
depicted in Scheme 1 on the basis of an NOESY experiment;11
NOE interaction was detected between the cyclobutane methyl-
ene protons and H4 protons of the phenanthrene ring.
1
In the H NMR spectrum of anti-3, sixteen peaks for aro-
matic protons and two peaks for the methine protons were
observed, suggesting a lower symmetrical structure. Among
the aromatic protons, H1, H2, and H4 are high-field shifted rel-
ative to 4, while H5–9 are hardly shifted. Especially, the H2 (δ
5.75, 6.13) and H1 protons (δ 6.28, 6.34) in anti-3 resonate at
much higher fields than those in syn-3 (H2: δ 6.67; H1: δ 7.42),
indicating that these protons are located above the opposite
phenanthrene ring. On the contrary, the H4 protons (δ 8.15,
8.48) of anti-3 are resonated at lower fields than those in syn-3
1,3-Bis(3-vinyl-10-phenanthryl)propane 4, precursor of 3,
was prepared as shown in Scheme 1. The oxidative photocy-
clization of stilbene derivative 6 toward 7 was employed as a
key step. The photoreaction of 4 was carried out in benzene (2
mM) using a 400-W high-pressure mercury lamp through a
Pyrex filter under a nitrogen atmosphere in the manner reported
Copyright © 2001 The Chemical Society of Japan

Chemistry Letters 2001
971
(δ 7.88), though higher fields than those in 4 (δ 8.67). Thus,
the H4 protons of anti-3 suffer from much less shielding effect
by the opposite phenanthrene ring than H1 and H2. These
observations are rather different from those for anti-2, where
the H4 as well as H1 and H10 resonates at high fields relative to
syn-2 and the two phenanthrene rings overlap mainly at the sin-
gle six-membered ring on the cyclobutane side. These results
obviously indicate that anti-3 possesses a less overlapped struc-
ture than anti-2; the overlap in anti-3 is only about half of the
six-membered ring on the cyclobutane side. Although single
crystals suitable for X-ray crystallographic analysis have not
been obtained, MM2 calculations demonstrated such overlap
between the two phenanthrene rings, which are arranged almost
in parallel with a distance of ca. 3.5 Å.12
difference in the extent of overlap between anti-3 and anti-2.
Therefore, it is definitely concluded that the excimer formation
of phenanthrene requires at least the overlap at one six-mem-
bered ring of two phenanthrene nuclei, provided that they are
arranged almost in parallel with a distance of ca. 3.5 Å.
In summary, novel anti-[2.3](3,10)phenanthrenophane 3
was prepared by the intramolecular [2 + 2] photocycloaddition
of 4 and successfully isolated from the syn-isomer.
Intriguingly, anti-3 exhibited monomer-like emission due to the
smaller overlap between phenanthrene rings than in anti-2.
This work was financially supported by the Grant-in-Aid
from the Ministry of Education, Culture, Sports, Science, and
Technology, Japan.
The absorption spectra of syn- and anti-3 in cyclohexane
exhibited considerable broadening and a red shift relative to
that of phenanthrene, though their shapes were rather different
from each other.13
Dedicated to Prof. Hideki Sakurai on the occasion of his
70th birthday.
References and Notes
The fluorescence spectra of syn- and anti-3 were measured
upon 280 nm excitation in cyclohexane at room temperature
(Figure 1). The spectrum of syn-3 is composed of a broad
structureless and red-shifted emission with a maximum at 420
nm similar to syn-2, though weak emission is also observed
around 370 nm. This broad emission is reasonably assignable
to the excimer fluorescence, as in the other cases.4–6 The latter
weak emission is probably due to photodecomposition prod-
ucts, since the relative intensity increased with repeated meas-
urements. On the other hand, anti-3 affords vibrational struc-
tures characteristic of monomer fluorescence, which shows the
mirror image of the longest absorption band. Although the con-
tribution of excimer fluorescence cannot be completely exclud-
ed, it is reasonable to assume that the observed emission is
derived predominantly from monomer fluorescence. This
behavior is in a remarkable contrast with that in anti-2, which
afforded distinct excimer fluorescence without vibrational
structures. Such difference is apparently ascribed to the slight
1
E. A. Chandross and H. Thomas, J. Am. Chem. Soc., 94, 2421
(1972).
K. A. Zachariasse, R. Busse, U. Schrader, and W. Kühnle, Chem.
Phys. Lett., 89, 303 (1982).
F. Hirayama, J. Chem. Phys., 42, 3163 (1965).
Y. Nakamura, T. Tsuihiji, T. Mita, T. Minowa, S. Tobita, H.
Shizuka, and J. Nishimura, J. Am. Chem. Soc., 118, 1006 (1996).
Y. Nakamura, T. Fujii, S. Sugita, and J. Nishimura, Chem. Lett.,
1999, 1039.
Y. Nakamura, T. Fujii, and J. Nishimura, Tetrahedron Lett., 41,
1419 (2000).
2
3
4
5
6
7
8
H. A. Staab and M. Haenel, Chem. Ber., 106, 2190 (1973).
D. Schweitzer, J. P. Colpa, J. Behnke, K. H. Hausser, M. Haenel,
and H. A. Staab, Chem. Phys., 11, 373 (1975).
a) J. Nishimura, H. Doi, E. Ueda, A. Ohbayashi, and A. Oku, J. Am.
Chem. Soc., 109, 5293 (1987). b) J. Nishimura, M. Takeuchi, M.
Koike, H. Sakamura, O. Yamashita, J. Okada, and N. Takaishi, Bull.
Chem. Soc. Jpn., 66, 598 (1993).
9
10 The spectral data of syn- and anti-3 are as follows. syn-3: mp 255
˚C (dec.); 1H NMR (500 MHz, CD2Cl2) δ 8.03 (2H, dd, J = 7.8 and
0.8 Hz, H5), 7.88 (2H, d, J = 1.6 Hz, H4), 7.42 (2H, d, J = 8.5 Hz,
H1), 7.26 (2H, dd, J = 7.6 and 1.5 Hz, H8), 7.10 (2H, s, H9), 7.04
(4H, m, H6 and H7), 6.67 (2H, dd, J = 8.6 and 2.0 Hz, H2), 4.45
(2H, m, cyclobutane methine), 3.56 (2H, m, one of ArCH2CH2),
2.94 (3H, m, one of ArCH2CH2 and one of ArCH2CH2), 2.16 (1H,
m, one of ArCH2CH2); 13C NMR (125.65 MHz, CD2Cl2) δ 131.87,
134.80, 131.78, 131.10, 129.38, 129.14, 128.88, 127.66, 126.81,
126.06, 125.33, 124.60, 122.12, 121.83, 47.44, 35.97, 25.95, 20.84;
HRMS (FAB) found: m/z 448.2192. Calcd for C35H28: M+,
448.2191. anti-3: mp 255 ˚C (dec.); 1H NMR (500 MHz, CDCl3) δ
8.74 (1H, dd, J = 7.7 and 0.6 Hz, H5), 8.60 (1H, dd, J = 7.9 and 1.8
Hz, H5), 8.48 (1H, s, H4), 8.15 (1H, d, J = 1.2 Hz, H4), 7.92 (1H,
dd, J = 8.4 and 1.5 Hz, H8), 7.87 (1H, dd, J = 8.2 and 2.1 Hz, H8),
7.71 (1H, s, H9), 7.67 (1H, s, H9), 7.62 (4H, m, H6, H6, H7 and
H7), 6.34 (1H, d, J = 8.5 Hz, H1), 6.28 (1H, d, J = 8.6 Hz, H1), 6.13
(1H, dd, J = 8.7 and 1.4 Hz, H2), 5.75 (1H, dd, J = 8.6 and 1.5 Hz,
H2), 4.51 (1H, m, one of cyclobutane methine), 4.23 (1H, m, one of
cyclobutane methine), 2.78 (7H, m), 2.52 (3H, m); 13C NMR
(125.65 MHz, CDCl3) δ 139.11, 138.07, 135.99, 135.85, 131.94,
131.91, 130.01, 129.96, 129.83, 129.63, 129.53, 129.50, 128.54,
127.92, 127.86, 126.10, 125.51, 125.47, 124.41, 124.12, 123.80,
123.47, 123.26, 122.88, 122.72, 122.65, 119.56, 47.54, 46.70,
31.69, 24.38, 24.19, 21.48, 19.44; HRMS (FAB) found: m/z
448.2174. Calcd for C35H28: M+, 448.2191.
11 The isomer possessing a cyclobutane ring directed to the H2-side
seems to be only slightly formed, but its isolation and characteriza-
tion has been unsuccessful.
12 MM2 calculations were performed by CS Chem 3D Pro Version 4.0
(Cambridge Soft Corporation).
13 Absorption and fluorescence spectra were measured for the solu-
tions in the range of 10–5–10–4 M. The fluorescence excitation
spectra were in good agreement with the corresponding absorption
spectra in all cases.