Functionalization and kinetic stabilization of the [4]paracyclophane system and aromaticity of its extremely bent benzene ring

Article abstract of DOI:10.1021/ja021206y

Kinetic stabilization of the [4]paracyclophane skeleton by the introduction of substituents, which serve to sterically hinder reactions at the reactive bridgehead sites, and properties of the resultant [4]-paracyclophanes are investigated in this study. Modification of the property of [4]paracyclophane by functionalization is also intended. [4]Paracyclophanes are designed to be derived from the corresponding Dewar benzene isomers via their photochemical aromatization, and the requisite 1,4-bridged Dewar benzenes bearing sterically demanding functional groups are prepared. Irradiation of these precursors under matrix isolation at 77 K leads to the formation of [4]paracyclophanes, which exhibit characteristic electronic absorption spectra. The half-lives of the generated species vary widely from less than 1 min at -90° C to 0.5 h at -20° C, depending on the type of substituents and the pattern of substitution. One of the derivatives, 24, is stable enough and its content in the irradiated mixture is high enough to permit the measurement of the ₁H NMR spectrum. The recorded spectrum, which is reproduced very well by theoretical calculations using the GIAO method at the hybrid HF-DFT (B3LYP/6-31+G^*) level, suggests the sustenance of rather strong diatropicity in its severely bent benzene moiety. Calculations on the bent benzene whose geometry is constrained to that calculated for 24 support that aromaticity is retained to a significant extent as compared to that of planar benzene, as judged by the magnetic criteria of aromaticity, that is, diamagnetic susceptibility exaltation and nucleus-independent chemical shift. The reason for the retention of aromaticity despite the severe bending of the benzene ring is discussed. Cyclophane 24 is so strained that it exceeds the corresponding Dewar benzene precursor in energy and thermally reverts to the latter with a half-life of 15 ±5 min at -20° C (ΔG^? = 18.3 ±0.3 kcal mol-¹).

Full text of DOI:10.1021/ja021206y

Published on Web 01/01/2003
Functionalization and Kinetic Stabilization of the
[4]Paracyclophane System and Aromaticity of Its Extremely
Bent Benzene Ring
Takashi Tsuji,* Masahiro Okuyama, Masakazu Ohkita, Hidetoshi Kawai, and
Takanori Suzuki
Contribution from the DiVision of Chemistry, Graduate School of Science, Hokkaido UniVersity,
Sapporo 060-0810, Japan
Received September 23, 2002 ; E-mail: tsuji@sci.hokudai.ac.jp
Abstract: Kinetic stabilization of the [4]paracyclophane skeleton by the introduction of substituents, which
serve to sterically hinder reactions at the reactive bridgehead sites, and properties of the resultant [4]-
paracyclophanes are investigated in this study. Modification of the property of [4]paracyclophane by
functionalization is also intended. [4]Paracyclophanes are designed to be derived from the corresponding
Dewar benzene isomers via their photochemical aromatization, and the requisite 1,4-bridged Dewar
benzenes bearing sterically demanding functional groups are prepared. Irradiation of these precursors under
matrix isolation at 77 K leads to the formation of [4]paracyclophanes, which exhibit characteristic electronic
absorption spectra. The half-lives of the generated species vary widely from less than 1 min at -90 °C to
0.5 h at -20 °C, depending on the type of substituents and the pattern of substitution. One of the derivatives,
24, is stable enough and its content in the irradiated mixture is high enough to permit the measurement of
1
the H NMR spectrum. The recorded spectrum, which is reproduced very well by theoretical calculations
using the GIAO method at the hybrid HF-DFT (B3LYP/6-31+G*) level, suggests the sustenance of rather
strong diatropicity in its severely bent benzene moiety. Calculations on the bent benzene whose geometry
is constrained to that calculated for 24 support that aromaticity is retained to a significant extent as compared
to that of planar benzene, as judged by the magnetic criteria of aromaticity, that is, diamagnetic susceptibility
exaltation and nucleus-independent chemical shift. The reason for the retention of aromaticity despite the
severe bending of the benzene ring is discussed. Cyclophane 24 is so strained that it exceeds the
corresponding Dewar benzene precursor in energy and thermally reverts to the latter with a half-life of 15
( 5 min at -20 °C (∆Gq ) 18.3 ( 0.3 kcal mol-1).
initially with semiempirical methods4 and later by ab initio
computations,5,6 along with the experimental investigations.
Introduction
The research on the preparation of paracyclophanes with ever
shorter bridges has made remarkable progress in the past three
decades1 and has been driven by the interest in the properties
of strained molecules and culminated in the successful genera-
tion of [4]paracyclophane.2,3 [4]Paracyclophane is, however, a
highly labile species, rapidly decaying in fluid solution even
below -130 °C,3 and the experimental exploration of its
properties has been impeded because of the instability. Several
theoretical studies of [4]paracyclophane have been conducted,
Grimme has suggested, on the basis of a high-level ab initio
study, that [4]paracyclophane is best classified as a strained
molecule with partial diradicaloid character.5 He has also pointed
out that (i) the small degree of bond alternation in the distorted
benzene ring and the benzene-type excited state shows that some
benzene-like character is retained, (ii) the strain energy is
estimated to be 91 kcal mol-1, and the largest contribution (87%)
arises from the deformation of the benzene skeleton, and (iii)
the energy difference between [4]paracyclophane and the
corresponding Dewar benzene isomer is estimated to be 0 ( 3
kcal mol-1. More recently, Schaefer and co-workers have
performed a theoretical investigation on the energetics and
magnetic susceptibility of [4]paracyclophane and the related
species.6 [4]Paracyclophane is now predicted to lie 9 ( 4 kcal
mol-1 higher in energy than the Dewar benzene isomer, and
(1) (a) Bickelhaupt, F.; de Wolf, W. H. Recl. TraV. Chim. Pays-Bas 1988,
107, 459-478. (b) Bickelhaupt, F.; de Wolf, W. H. In AdVances in Strain
in Organic Chemistry; Halton, B., Ed.; JAI Press: Greenwich, CT, 1993;
Vol. 3, pp 185-227. (c) Tobe, Y. In Topics in Current Chemistry; Weber,
E., Ed.; Springer: Berlin, 1994; Vol. 172, pp 1-40. (d) Kane, V. V.; de
Wolf, W. H.; Bickelhaupt, F. Tetrahedron 1994, 50, 4575-4622. (e) de
Meijere, A.; Ko¨nig, B. Synlett 1997, 1221-1232. (f) Tsuji, T. In AdVances
in Strained and Interesting Organic Molecules; Halton, B., Ed.; JAI Press:
Stamford, CT, 1999; Vol. 7, pp 103-152. (g) Tsuji, T.; Ohkita, M.; Kawai,
H. Bull. Chem. Soc. Jpn. 2002, 75, 415-433.
(2) (a) Kostermans, G. B. M.; Bobeldijk, M.; De Wolf, W. H.; Bickelhaupt,
F. J. Am. Chem. Soc. 1987, 109, 2471-2475. (b) Bickelhaupt, F. Pure
Appl. Chem. 1990, 62, 373-382.
(3) (a) Tsuji, T.; Nishida, S. J. Chem. Soc., Chem. Commun. 1987, 1189-
1190. (b) Tsuji, T.; Nishida, S. J. Am. Chem. Soc. 1988, 110, 2157-2164.
(c) Tsuji, T.; Nishida, S. J. Am. Chem. Soc. 1989, 111, 368-369. (d) Tsuji,
T.; Nishida, S.; Okuyama, M.; Osawa, E. J. Am. Chem. Soc. 1995, 117,
9804-9813.
(4) (a) Kostermans, G. B. M.; Kwakman, P. J.; Pouwels, P. J. W.; Somsen,
G.; de Wolf, W. H.; Bickelhaupt, F. J. Phys. Org. Chem. 1989, 2, 331-
348. (b) Jenneskens, L. W.; Louwen, J. N.; de Wolf, W. H.; Bickelhaupt,
F. J. Phys. Org. Chem. 1990, 3, 295-300. (c) Bockisch, F.; Rayez, J. C.;
Liotard, D.; Duguay, B. J. Comput. Chem. 1992, 13, 1047-1056.
(5) Grimme, S. J. Am. Chem. Soc. 1992, 114, 10542-10547.
(6) Ma, B.; Sulzbach, H. M.; Remington, R. B.; Schaefer, H. F., III. J. Am.
Chem. Soc. 1995, 117, 8392-8400.
9
10.1021/ja021206y CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 951-961
951

A R T I C L E S
Tsuji et al.
Scheme 1 a
Scheme 2 a
a (a) hν/TPP/CCl4, O2, 95%; (b) Et3N/CH2Cl2, 95%; (c) MnO2, 92%;
(d) Bu3SnH, Ph(PPh3)4, 84%; (e) PCC, 89%.
the activation energy for the isomerization of the former to the
latter is estimated to be 22 kcal mol-1. They also showed that
boat-shaped benzene with the same geometry as in [4]paracy-
clophane has the same magnetic susceptibility as hypothetical
cyclohexatriene with alternating double and single bonds, and,
therefore, a weak ring current is expected.
a (a) CH2(CN)2, â-alanine, 54%; (b) C5H5N‚HBr3/C5H5N, 73%; (c)
CH2(CO2t-Bu)2, TiCl4/C5H5N, 26%; (d) Zn/AcOH, 76% (10:11 ) 1:3); (e)
i-Pr3SiOTf/i-Pr2NEt, 82%.
Scheme 3 a
Steric strain in [4]paracyclophane mainly arises from the
deformation of the benzene skeleton, as pointed out in those
theoretical investigations, and its known reactivity suggests that
the extreme lability arises from the susceptibility of bridgehead
carbon atoms toward the addition of diverse reagents.2,3 This
implies that the system may be kinetically stabilized by
introducing substituents that specifically shield the bridgehead
carbon atoms from access by other reagents. A number of
substituted [4]paracyclophanes were designed on the basis of
such consideration, and their generation and kinetic stabilities
were investigated in this study. One of the derivatives was stable
a 16: (a) hν, 1,4-bis(trimethylsilyl)-2-butyne, 18%; (b) HCO2Et, EtONa;
TsN3, Et3N, 58%; hν, MeOH, 80%; LDA, PhSeBr/HMPA, 79%; H2O2,
53%. 17: (a) hν, 1,4-bis(tri-i-propylsilyl)-2-butyne, 3%; (b) HCO2Et, EtONa;
TsN3, Et3N, 55%; hν, Me2NH, 66%; LDA, PhSeBr/HMPA, 31%; mCPBA,
60%.
dition of photochemically generated singlet oxygen9,10 to 1 smo-
othly proceeded to afford 2 as colorless crystals in high yield,
which provided 3 upon subsequent treatment with triethylamine.9
Oxidation of 3 with manganese dioxide in dichloromethane
delivered the ene dione derivative 4, while the chemoselective
reduction of the enone moiety of 311 followed by PCC oxidation
furnished 6. The Knoevenagel condensation of 6 with malono-
nitrile readily proceeded in the presence of â-alanine12 to afford
7, from which 8 was obtained via treatment with pyridinium
bromide perbromide in pyridine. The condensation of 6 with
di-tert-butyl malonate was more difficult but somehow pro-
ceeded under the catalysis of TiCl4 in pyridine to produce 9
directly in reasonable yield.13 Reduction of 9 with zinc in acetic
acid14 provided 10 as a minor product, together with 11. Treat-
ment of 6 with tri-i-propylsilyl triflate in i-Pr2NEt furnished
the corresponding bis(enol silyl ether) 12. On the other hand,
the Dewar benzene precursors 16 and 17 were prepared from
tetrahydroindanone 13 following essentially the same procedure
as that reported for the preparation of the corresponding
unsubstituted ester,3b,8 except that 1,4-bis(trialkylsilyl)but-2-ynes
were used in place of acetylene (Scheme 3).
1
enough to permit the measurement of the H NMR spectrum
for the first time for [4]paracyclophane, which suggested the
sustenance of a significant diamagnetic ring current in its
extremely bent benzene ring. In this paper, we present details
of the investigation including (i) the preparation of 1,4-bridged
Dewar benzenes bearing a variety of sterically demanding
functional groups which serve to hinder reactions at the
bridgehead carbon atoms, (ii) their photochemical transformation
into the corresponding [4]paracyclophanes, which exhibit
characteristic electronic absorption spectra, (iii) degrees of
kinetic stability of the resultant [4]paracyclophanes, (iv) suc-
1
cessful measurement of the H NMR spectrum of one of the
[4]paracyclophanes, (v) theoretical investigation on the aroma-
ticity of [4]paracyclophane, and (vi) experimental verification
for the reversal of stability order, benzene form > Dewar
benzene form, in [4]paracyclophane.7
Results and Discussion
Preparation of 1,4-Bridged Dewar Benzenes Bearing
Functional Groups. The generation of functionalized [4]-
paracyclophanes herein studied was achieved exclusively by the
photochemical aromatization of the corresponding Dewar ben-
zene derivatives. The preparation of 4, 6-10, and 12 was accom-
plished through the transformation starting from tetraene 13d,8
as illustrated in Schemes 1 and 2. Thus, the [2 + 4] cycload-
(9) (a) Su¨tbeyaz, Y.; Sec¸en, H.; Balci, M. J. Chem. Soc., Chem. Commun.
1988, 1330-1331. (b) Wasserman, H. H.; Murray, R. W. Singlet Oxygen;
Academic Press: New York, 1979.
(10) Foster, C. H.; Berchtold, G. A. J. Org. Chem. 1975, 40, 3743-3746 and
references cited therein.
(11) Keinan, E.; Gleize, P. A. Tetrahedron Lett. 1982, 23, 477-480.
(12) Crawford, R. J. J. Org. Chem. 1983, 48, 1366-1368.
(13) (a) Lehnert, W. Tetrahedron Lett. 1970, 4723-4724. (b) Iio, H.; Isobe,
M.; Kawai, T.; Goto, T. J. Am. Chem. Soc. 1979, 101, 6076-6081.
(14) Elks, J.; Evans, R. M.; Long, A. G.; Thomas, G. H. J. Chem. Soc. 1954,
451-462.
(7) For a preliminary account of this work, see: (a) Okuyama, M.; Tsuji, T.
Angew. Chem., Int. Ed. Engl. 1997, 36, 1085-1087. (b) Okuyama, M.;
Ohkita, M.; Tsuji, T. Chem. Commun. 1997, 1277-1278.
(8) (a) Tsuji, T.; Komiya, Z.; Nishida, S. Tetrahedron Lett. 1980, 21, 3583-
3586. (b) Tsuji, T.; Nishida, S. Tetrahedron Lett. 1983, 24, 3361-3364.
9
952 J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003

Functionalization of the [4]Paracyclophane System
A R T I C L E S
Scheme 4
Figure 1. UV-vis absorption (a) and difference spectra (b-d) in ether-
2-methylbutane (1:1) glass at 77 K: (a) for 7;15 (b) (spectrum after irradiating
254 nm light for 30 s) minus (spectrum a); (c) (spectrum after irradiating
254 nm light for 1 min) minus (spectrum a); (d) (spectrum after subsequently
irradiating >440 nm light for 1 min) minus (spectrum c).
accommodated by postulating the initial formation of 18, and,
accordingly, the characteristic absorption spectrum extending
to 480 nm was assigned to 18. The prepared Dewar benzenes
are more or less prone to undergo irreversible photochemical
transformation into the corresponding naphthalene derivatives.
Thus, extended irradiation of matrix containing 7 led to the
steady accumulation of 21 after the generation of 18 reached a
plateau. The deviation of the difference spectrum d in Figure 1
from the mirror image of the spectrum c in the 300-340 nm
region is primarily due to the formation of 21 during the
irradiation of 254 nm light.16
Scheme 5
Electronic absorption spectra for 4, 10, 12, 16, and 17 and
difference spectra derived from their irradiation are shown in
Figure 2. While the photochemical behavior of 4 and 10 is
similar to that of 7, compounds 8, 16, and 17 are less prone to
isomerize to the naphthalene derivatives, and a photostationary
state consisting of the Dewar benzene precursor and the
corresponding [4]paracyclophane was reached after a short time
of irradiation of each substrate. Dione 6 is at the other end of
the reactivity spectrum and appeared to produce 20 almost
exclusively upon irradiation under matrix isolation. Although
the mechanism of the formation of the naphthalene derivatives
is not clear at present, the exclusive formation of 20 from 6
suggests that they are derived from the triplet excited states of
the Dewar benzene precursors, while the aromatization to the
[4]paracyclophanes occurs in their singlet excited states. The
photochemical reaction of 9 was complex, producing thermally
labile products whose properties were consistent neither with
the corresponding [4]paracyclophane 25 nor with the 1,4-
naphthoquinone derivative 28, and hence was not investigated
further. The difference spectrum for 16, that is, (c) in Figure
2D, is closely similar in shape to that reported for 33,3a,b except
for the bathochromic shifts of absorption bands by 20-25 nm.
The bending of the benzene ring leads to the rise of HOMO as
well as the lowering of LUMO in energy.5 Parent [4]paracy-
clophane, as a result, exhibits absorption maxima in the 330-
340 and 370-380 nm regions.3b The observed spectra demon-
strate that electronic interactions between those frontier molecular
orbitals and the substituents give rise to the pronounced further
bathochromic shifts of absorption bands.
Photochemical Behavior of the Dewar Benzene Precur-
sors. Reactivity common to the 1,4-C4-bridged Dewar benzenes
studied so far is clean, reversible photochemical transformation
into the corresponding [4]paracyclophanes.3 Thus, as reported
previously, irradiation of 30 with 254 nm light under matrix
isolation at 77 K leads to the formation of 33 which cleanly
reverts to 30 upon secondary irradiation with 365 nm light with
which 33, but not 30, is electronically excited (Scheme 5).3a,b
Compounds 4, 7, 8, 10, 12, 16, and 17 photochemically behaved
similarly, and their irradiation under matrix isolation at low
temperature led to the development of new absorption bands at
long wavelength regions, respectively, which rapidly decayed
to restore the original spectra upon secondary irradiation with
light of wavelengths selectively absorbed by the products,
suggesting the transient generation of [4]paracyclophanes
(Schemes 4 and 5). As described later, the photochemical
generation of 24 from 8 and the ready reverse transformation
1
upon secondary irradiation are indeed supported by H NMR
measurements.
Figure 1 shows difference UV-vis spectra between the
absorptions before and after the irradiation of 7 with 254 nm
light in ether-2-methylbutane glass at 77 K (spectra b and c)
and that between those before and after the subsequent irradia-
tion with light of wavelength >440 nm (spectrum d), together
with the absorption spectrum for 7 (spectrum a). The difference
spectrum d in Figure 1 unambiguously demonstrates the
regeneration of 7 at the expense of the product generated in the
prior irradiation with 254 nm light. These observations are best
(15) The weak absorption in the 300-380 nm region is due to the contaminant
8 (<2%).
(16) The formation of 21 was confirmed by 1H NMR. See the Experimental
Section.
9
J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003 953

A R T I C L E S
Tsuji et al.
Figure 2. UV-vis absorption (a) and difference spectra (b-e) in ether-2-methylbutane (1:1) glass at 77 K. (A) (a) For 4; (b) (spectrum after irradiating
254 nm light for 3 min) minus (spectrum a); (c) (spectrum after irradiating 254 nm light for 5 min) minus (spectrum a); (d) (spectrum after irradiating 254
nm light for 7 min) minus (spectrum a); (e) (spectrum after subsequently irradiating 365 nm light for 30 s) minus (spectrum d). (B) (a) For 10; (b) (spectrum
after irradiating 254 nm light for 10 s) minus (spectrum a); (c) (spectrum after irradiating 254 nm light for 20 s) minus (spectrum a); (d) (spectrum after
subsequently irradiating >400 nm light for 30 s) minus (spectrum c). (C) (a) For 12; (b) (spectrum after irradiating 313 nm light for 5 min) minus (spectrum
a); (c) (spectrum after subsequently irradiating 365 nm light for 8 min) minus (spectrum b). (D) (a) For 16; (b) (spectrum after irradiating 254 nm light for
1 min) minus (spectrum a); (c) (spectrum after irradiating 254 nm light for 3 min) minus (spectrum a); (d) (spectrum after subsequently irradiating >470 nm
light for 5 min) minus (spectrum c). (E) (a) For 17; (b) (spectrum after irradiating 254 nm light for 5 min) minus (spectrum a); (c) (spectrum after subsequently
irradiating >440 nm light for 6 min) minus (spectrum b).
Kinetic Stabilities of [4]Paracyclophanes. [4]Paracyclo-
phane, for example, 33, remains unchanged indefinitely under
matrix isolation at 77 K but is highly labile in fluid solution,
polymerizing almost instantaneously even at -130 °C,3 unless
some stabilizing measures are undertaken. The lability of [4]-
paracyclophane undoubtedly arises from its susceptibility at the
bridgehead carbon atoms toward the addition of diverse reagents,
by which means steric strain inherent to the structure is largely
relieved. [4]Paracyclophanes preferentially react indeed with
trapping reagents at their bridgehead carbon atoms.2,3 This
implies that the ring system may be kinetically stabilized by
introducing sterically demanding substituents which serve to
sterically hinder reactions at the reactive bridgehead sites. The
[4]paracyclophane derivatives herein studied were designed on
the basis of such consideration.
Table 1. Kinetic Stabilities of the Substituted [4]Paracyclophanes
in Ether-2-Methylbutane (1:1)
temp
(°C)
t1/2
(min)
content
(%)a
temp
(°C)
t1/2
(min)
content
(%)a
compd
compd
18
19
24
-90 10 ( 3
-20 30 ( 5
-20 12 ( 2
e6b
e3b
29
31
32
-50
30 ( 5
20 ( 5
e2d
e15b
e5b
-130
e10c
-50 100 ( 10
a At a quasi-photostationary state reached upon irradiation of the
corresponding Dewar benzene isomer. b Under irradiation of 254 nm light.
c Under irradiation of 365 nm light. d Under irradiation of 313 nm light.
their decay spectrometrically. The results summarized in Table
1 show that significant improvement in kinetic stability has been
achieved in 19, 24, 29, and 32. The stabilities of 19 and 24 are,
while comparable to each other, much superior to those of the
other [4]paracyclophanes examined. Cyclophane 18 is much less
stable than 24 rather unexpectedly and disappears fairly rapidly
even at -90 °C. Although the cause of the poor stability of 18
Glass matrixes containing the [4]paracyclophanes were
thawed, and their kinetic stabilities were evaluated by monitoring
9
954 J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003

Functionalization of the [4]Paracyclophane System
A R T I C L E S
Chart 1. Experimental and Calculated (in Parentheses, GIAO/
is not entirely clear, examination of molecular models of 18
and 24 is revealing. While the geometry-optimized structure of
24 is C2V symmetric (vide infra) and the cyano substituents
effectively hinder destructive reactions at the bridgehead posi-
tions, the -CH2CH2- moiety of 18 preferentially adopts a
staggered conformation as is the case with parent [4]para-
cyclophane,4-6 thereby displacing the dicyanomethylene groups
from the plane bisecting the bent aromatic ring and, thus, making
the access to the bridgehead carbon atoms less sterically
encumbered. The stabilities of 29 and 32 are only intermediate
among the derivatives investigated, because of the conforma-
tional flexibility which allows the substituents to deviate from
the desired, bridgehead-protecting arrangements.
B3LYP/6-31+G*) Proton Chemical Shifts for 8, 24, and 34, and
Chemical Shift Changes (∆δ) Accompanying Rearrangement of 8
to 24 and 34
1H NMR Spectrum of 24. The measurement of the 1H NMR
spectrum of [4]paracyclophane ought to provide not only
definitive evidence for its formation but also valuable informa-
tion concerning the aromaticity of its severely bent benzene ring,
but has been thwarted because of the extreme thermal instability.
The improved thermal stability of 24, however, is high enough
benzene ring of the latter despite its extreme bending. The
exhibition of significant diatropicity by such a severely bent
benzene ring was rather unexpected6,19 and seemed to deserve
further investigation. In the following section, the aromaticity
of 24 was investigated theoretically. The calculations of
chemical shifts for 8 and 24 at the RHF/6-31G* and RHF/6-
31+G* levels were less satisfactory, for example, providing δ
7.19 and 7.02 and δ 5.52 (∆δ -1.67) and 8.50 (∆δ 1.48) for
the Ha and Hb of 8 and 24, respectively, at the latter level. The
theoretical examination on 24 described below is, therefore,
based on the calculations at the B3LYP/6-31+G* level unless
otherwise noted.
1
to permit the measurement of the H NMR spectrum. Another
requisite to be fulfilled to measure an NMR spectrum of [4]-
paracyclophane is its content at the quasi-photostationary state
reached upon irradiation of the Dewar benzene precursor.
Fortunately, the proportion of 24 to 8 at the quasi-photoequi-
librium under irradiation of 365 nm light was sufficiently high
as listed in Table 1.
1
The H NMR spectrum of 8 consists of a pair of singlets at
Aromaticity of the Extremely Bent Benzene Moiety of 24.
The principal geometrical parameters for geometry-optimized
24 are summarized in Table 2. The structure of 24 is C2V
symmetric,20 and the planar bis(dicyanomethylene)buteno bridge
bisects the bent benzene ring. The deformation angles R and â,
which give a measure of the degree to which the benzene ring
is bent, are 28.6° and 43.9°, respectively, and the sum (R + â)
amounts to 72.5°. As a result, the bridgehead carbon atoms are
markedly pyramidalized, decreasing the sum of the bond angles
around each of them to 341.5° from 360° for a planar trivalent
carbon atom. The residual aromatic carbon atoms are also
pyramidalized (the sum of the bond angles is 356.0°) as
illustrated in Table 2, so as better to maintain π-orbital overlap
with the adjacent bridgehead carbon atom.21
δ 6.93 and 7.20 in an intensity ratio of 2:1 in CD2Cl2. Irradiation
of 8 with 365 nm light at -90 °C led to the development of a
pair of weak singlet signals with an intensity ratio of ca. 2:1 at
δ 7.97 and 5.85, which rapidly decayed upon subsequent
irradiation with light of wavelength >400 nm to restore the
original two line spectrum.7a The observed photochemical
behavior of the transient and its simple 1H NMR spectrum are
consistent with the formation of 24. Corroborating evidence for
this assignment was provided by the theoretical calculations of
proton chemical shifts.17 Thus, the calculations for 8 and 24
using the gauge-independent atomic orbital (GIAO) method18
at the hybrid HF-DFT (B3LYP/6-31+G*) level of theory
satisfactorily reproduce the experimental spectra as summarized
in Chart 1. The experimental and computed chemical shift
changes (∆δ) accompanying the rearrangement of 8 to 24 agree
particularly well with each other. Prismane derivative 34 is
apparently incompatible with the observed spectrum. The
pronounced upfield shift of Ha signal (∆δ ) -1.35 ppm)
accompanied by the marked downfield shift of Hb (∆δ ) 1.04
ppm) upon the isomerization of 8 to 24 strongly suggests the
sustenance of a substantial diamagnetic ring current in the
A number of definitions or criteria have been considered for
characterizing aromaticity.22 The majority are, however, ap-
plicable only for nearly planar aromatic compounds and often
disturbed by strain effects and other contributing factors.
Schleyer and co-workers recently proposed two probes, the
diamagnetic susceptibility exaltation (Λ)23 and the nucleus-
(19) For the diatropicity of [5]paracyclophanes which bear less strained benzene
rings than [4]paracyclophane, see: (a) Jenneskens, L. W.; de Kanter, F. J.
J.; Kraakman, P. A.; Turkenburg, L. A. M.; Koolhaas, W. E.; de Wolf, W.
H.; Bickelhaupt, F.; Tobe, Y.; Kakiuchi, K.; Odaira, Y. J. Am. Chem. Soc.
1985, 107, 3716-3717. (b) Tobe, Y.; Kaneda, T.; Kakiuchi, K.; Odaira,
Y. Chem. Lett. 1985, 1301-1304. (c) Kostermans, G. B. M.; de Wolf, W.
H.; Bickelhaupt, F. Tetrahedron Lett. 1985, 27, 1095-1098. (d) Koster-
mans, G. B. M.; de Wolf, W. H.; Bickelhaupt, F. Tetrahedron 1987, 43,
2955-2966.
(17) All calculations were performed with the programs implemented in the
Gaussian 98 program package. 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, M.; Mennucci, B.; Pomelli, C.;
Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui,
Q.; Morokuma, K.; Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.;
Baboul, A. G.; 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.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.11.1;
Gaussian, Inc.: Pittsburgh, PA, 2001.
(20) The geometry optimizations of 24 starting from the various less symmetrical
geometries converged to the C2V symmetric one.
(21) Haddon, R. C. Acc. Chem. Res. 1988, 21, 243-249 and references therein.
(22) (a) Garrat, P. J. Aromaticity; Wiley & Sons: New York, 1986. (b) Minkin,
V. I.; Glukhovtsev, M. N.; Simkin, B. Y. Aromaticity and Antiaromatic-
ity: Electronic and Structural Aspects; Wiley & Sons: New York, 1994.
(23) Schleyer, P. v. R.; Jiao, H. Pure Appl. Chem. 1996, 68, 209-218. This
criterion was first proposed by Dauben et al.: Dauben, H. J., Jr.; Wilson,
J. D.; Laity, J. L. J. Am. Chem. Soc. 1968, 90, 811. Dauben, H. J., Jr.;
Wilson, J. D.; Laity, J. L. J. Am. Chem. Soc. 1969, 91, 1991.
(18) Wolinski, K.; Hinton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112, 8251-
8260.
9
J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003 955

A R T I C L E S
Tsuji et al.
Table 2. Selected Geometrical Parameters for C2v Symmetric 24
Optimized at the B3LYP/6-31+G* Level of Theory
Table 3. Magnetic Susceptibility Data (ppm cgs with Sign
Reversed)a
IGLO/TZP
øzz øav
CSGT/B3LYP/6-311+G**
øip øzz øav ∆ø
compd
øip
∆ø
benzeneb,c
46.8 110.2 68.0 63.4 30.4 97.9 52.9 67.5
45.0 100.5 63.5 55.5
bent benzened
35
30.6 86.6 49.3 56.0
hypothetical
cyclohexatrienee
46.1 100.6 64.2 54.6 30.8 93.2 51.6 62.4
Bond Lengths and Nonbonding Interatomic Distances (Å)
C1-C2
C2-C3
C3-C4
C1-C8
C7-C8
C2-C11
C11-C13
1.4993
1.4859
1.3868
1.4122
1.3921
1.3784
1.4325
C11-C14
C13-N
C14 -N
C3-H
1.4340
1.1639
1.1638
1.0892
1.0876
2.6756
2.4165
a øip is the in-plane component: øip ) (øxx + øyy)/2. øzz is the out-of-
plane component. øav is the isotropic part: øav ) (2øip + øzz)/3. ∆ø is the
anisotropic part: ∆ø ) øzz - øip. b IGLO/TZP: ref 27. CSGT/B3LYP/6-
311+G**: ref 28. c Experimental values: øip ) 34.9 ( 2.0, øzz ) 94.6 (
2.5, øav ) 54.8, and ∆ø ) 59.7 ppm cgs (with sign reversed): ref 31. d With
the same geometry as in the DZ+d SCF optimized [4]paracyclophane: ref
6. e With alternating single and double bonds: r (C-C) ) 1.501 Å, r (CdC)
) 1.336 Å, r (C-H) ) 1.083 Å, and all angles equal to 120°. IGLO/TZP:
ref 27. CSGT/B3LYP/6-311+G**: this work.
C7-H
C1‚‚‚C6
C7‚‚‚C10
Bond, Torsional, and Bending Angles (deg)
C1-C2-C3
C1-C2-C11
C1-C8-C7
C2-C3-C4
C3-C2-C11
C2-C1-C8
C8-C1-C9
C2-C11-C13
C2-C11-C14
C2-C3-H
114.944
124.595
117.030
137.527
120.461
111.908
117.657
120.477
123.358
111.244
C1-C8-H
120.429
118.518
157.565
67.253
-99.164
32.503
28.597
43.874
147.703
22.435
C7-C8-H
C1-C8-C7-H
C3-C2-C1-C8
C7-C8-C1-C2
C7-C8-C1-C9
R
â
γ
δ
independent chemical shift (NICS),24 as effective aromaticity/
antiaromaticity criteria, applicable not only for planar molecules
but also for nonplanar ones including spherical systems.
According to these definitions, significantly exalted (negative)
Λ and negative NICS denote aromaticity, whereas positive Λ
and NICS indicate antiaromaticity. NICS value calculated at
the center (mean of the carbon atom coordinates) of the C6 ring
moiety of 24 is -8.1 (GIAO/B3LYP/6-31+G*) as compared
with -9.7 for planar benzene, and Λ computed for the bent
benzene 3525 whose geometry was constrained to that present
in 24 is -11.5 ppm cgs26 (CSGT29/B3LYP/6-311+G**) as
compared with -15.1 ppm cgs for planar benzene,28 suggesting
that aromaticity is retained to a considerable extent in 24. NICS
value at the center of 35 is -7.7, and the chemical shift for the
protons on the edge carbon atoms is δ 7.92,30 in good agreement
with those calculated for 24. Thus, 35 seems to be an adequate
model for the examination of the aromaticity of 24. Schaefer
and co-workers have shown that the boat-shaped benzene with
the same geometry as that in [4]paracyclophane has almost the
same magnetic susceptibility as the hypothetical cyclohexatriene
with alternating single and double bonds (Table 3), and,
therefore, a weak ring current is expected in [4]paracyclophane.6
The magnetic susceptibility data for 35 computed using the
Figure 3. Plots of calculated magnetic shielding effect in ppm (GIAO/
B3LYP/6-31+G*) versus distance from the ring center along the z axis for
(a) planar benzene and (b) the bent benzene 35 with the same geometry as
that in 24. Negative shielding denotes diamagnetic effect.
CSGT/B3LYP/6-311+G** are listed in Table 3, together with
those for the hypothetical cyclohexatriene.
In Figure 3 are illustrated plots of absolute magnetic
shieldings (NICS) computed for planar benzene and 35 along
their z axes. The maximum diamagnetic shieldings found above
and below the ring center of benzene reflect the π electron toroid
density.24 The magnitude of diamagnetic shielding effect in 35
is markedly unsymmetrical with respect to the plane of bent
benzene ring, and the effect in the region of space above the
concave surface is significantly stronger than that below the
convex surface. As described above, all of the carbon atoms of
the boat-shaped 35 are extensively pyramidalized so as better
to maintain the π-orbital overlaps, and the axes of the atomic
π-orbitals, those at the bow and stern in particular, are tilted
toward the concave side. The π-orbitals may, as a result, overlap
with each other more effectively above the concave surface than
below the convex surface, and the diamagnetic ring current
induced above the former may consequently be stronger than
that below the latter. It should also be noted that the diamag-
(24) Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao, H.; van Eikema
Hommes, N. J. R. J. Am. Chem. Soc. 1996, 118, 6317-6318.
(25) The benzene ring was frozen in the conformation present in 24, two
hydrogen atoms were placed in the same direction as the benzylic carbon
atoms, and the C-H bond lengths were optimized at the B3LYP/6-31+G*
level.
(26) Calculated using the increment system derived from the computed magnetic
susceptibility data for ethylene and 1,3-butadiene. For derivation of Λ, see
refs 27 and 28.
(27) Fleischer, U.; Kutzelnigg, W.; Lazzeretti, P.; Mu¨hlenkamp, V. J. Am. Chem.
Soc. 1994, 116, 5298-5306.
(28) Simion, D. V.; Sorensen, T. S. J. Am. Chem. Soc. 1996, 118, 7345-7352.
(29) Continuous set of gauge transformations method: (a) Keith, T. A.; Bader,
R. F. W. Chem. Phys. Lett. 1992, 194, 1-8. Keith, T. A.; Bader, R. F. W.
Chem. Phys. Lett. 1993, 210, 223-231. (b) Cheeseman, J. R.; Frisch, M.
J.; Trucks, G. W.; Keith, T. A. J. Chem. Phys. 1996, 104, 5497-5509.
(30) Calculated chemical shift for the protons at the bow and stern positions of
35 is δ 5.85.
9
956 J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003

Functionalization of the [4]Paracyclophane System
A R T I C L E S
netically shielded region of space above the concave surface
seems to be restricted to concentrate the effect within a relatively
narrow space along the z coordinate because of the tilting of
the atomic π-orbitals, as compared to that below the reverse
surface which appears to be extended to the wider space. The
stronger diamagnetic shielding effect above the concave surface
may, partially at least, be due to reinforcement by the local
shielding effect by the tilted π-orbitals at the bow and stern of
35. The magnitude of this local effect may be estimated by
calculating magnetic shielding for 1,3-butadiene, whose geom-
etry is constrained to that of the C3-C2-C1-C6 moiety of
35.32 The results suggest that the diamagnetic shielding in the
concave region 0.6-2.0 Å above the ring center of 35 is
enhanced by this local effect, but the enhancement will be less
than 3 ppm. Because the magnetic shielding effect calculated
for nonconjugated, planar 1,4-cyclohexadiene along the z axis
decays monotonically with increasing distance from the ring
center, the appearance of maxima in magnetic shielding above
and below the bent benzene ring of 35 is also consistent with
the maintenance of the aromatic conjugation. Thus, 35 and,
accordingly, 24 seem to sustain substantial aromaticity despite
the extreme bending of their C6 rings, as judged by the magnetic
criteria of aromaticity, Λ and NICS.
Figure 4. UV-vis absorption (a) and difference spectra, (b) and (c), in
ether-2-methylbutane (1:1) glass at 77 K: (a) for 8; (b) (spectrum after
irradiating 365 nm light for 7 min) minus (spectrum a); (c) (spectrum after
briefly heating the resulting mixture to ambient temperature in the dark)
minus (spectrum b).
relative to 24 is -20.3 kcal mol-1 at the B3LYP/6-31+G* level,
supporting the substantial exothermicity of the isomerization
process. The agreement of this energy barrier with a value of
22 kcal mol-1 computed for the rearrangement of parent [4]-
paracyclophane to the Dewar benzene isomer6 is extremely
good, especially when a higher exothermicity for the process
from 24 to 8 is taken into account.
The aromatic stabilization energy of benzene has been
evaluated at 25 ( 5 kcal mol-1 22
while the energy difference
,
Steric strain in paracyclophane is rapidly accumulated as a
bridging side chain is shortened, whereas the strain in the
corresponding Dewar benzene form remains largely unaf-
fected.4,5 Thus, it is expected that the thermodynamic stability
order, benzene form > Dewar form, is reversed at a certain
chain length and the former becomes more energetic than the
latter. This reversion of the stability order has been predicted
by theoretical calculations for a pair of [4]paracyclophane/1,4-
(CH2)4-bridged Dewar benzene,6 but it remained experimentally
unverified because of the extreme instability of [4]paracyclo-
phane. The thermal rearrangement of 24 to 8 represents the first
experimental verification of the theoretical prediction. Thus, the
kinetic stability of the [4]paracyclophane system is defined not
only by the susceptibility of its bridgehead carbon atoms toward
the addition of diverse reagents, but also by the intrinsic thermal
reactivity to rearrange to the Dewar form. Further shortening
of the side chain will certainly result in a lowered energy barrier
to isomerization to the Dewar form and hence in a reduced
lifetime of the benzene form. Compound 24 seems to represent
almost a limiting case of experimentally characterizable strained
paracyclophanes.
between planar benzene and 35 is 87 kcal mol-1 at the B3LYP/
6-31+G* level. The aromatic stabilization energy of 35 thus
compensates only a fraction of its strain energy at best. In
benzene, distortion into a nonplanar geometry unavoidably gives
rise to the torsion of π-bonds, which is not suppressible by the
localization of π-bonds. Thus, the carbon atoms of benzene
undergo extensive rehybridization-pyramidalization upon severe
bending to avoid the rupture of π-bonds leading to degeneration
into a biradical(oid), at the expense of energy even much greater
than its aromatic stabilization energy, and, consequently, the
cyclic conjugation seems to be retained. The sustenance of
aromatic conjugation in 24 is, thus, basically not attributable to
the aromatic stabilization effect. Accordingly, the benzene ring
moiety of 24 is extremely bent and strained, yet it sustains rather
strong diatropicity and only a small degree of bond alternation.
Thermal Cycloreversion of 24 to 8. The investigation of
the thermal behavior of 24 revealed that its insufficient stability
for isolation at ambient temperature was due to the reactivity
to undergo spontaneous reversion to 8 below 0 °C rather than
the insufficient protection of its reactive bridgehead sites toward
intermolecular reactions. The difference UV-vis spectrum
between the spectra taken before and after a cycle of thawing
an ether-2-methylbutane matrix containing 24-brief warming
to room temperature-refreezing at 77 K clearly displayed the
regeneration of 8 at the expense of 24 (Figure 4). This
observation indicates that 24 is higher in energy than 8 and
thermally reverts to the latter rapidly at room temperature. The
measured half-life of 24 is 15 ( 5 min at -20 °C, and,
accordingly, the activation free energy for the isomerization of
24 to 8 is 18.3 ( 0.3 kcal mol-1. The calculated energy of 8
Conclusion
Irradiation of 1,4-bridged Dewar benzenes bearing a variety
of functional groups led more or less successfully to the
formation of the corresponding [4]paracyclophanes, which
underwent efficient cycloreversion upon subsequent electronic
excitation. [4]Paracyclophane is a highly labile species, but could
be stabilized kinetically by introducing substituents which
sterically hinder access to the reactive bridgehead carbon atoms
by other reagents to an extent to allow the measurement of the
1H NMR spectrum. The recorded spectrum suggested that the
[4]paracyclophane derivative sustains significant diatropicity
despite the extreme bending of its benzene ring. Theoretical
calculations also support the retention of substantial aromaticity
in the bent benzene moiety as judged by the magnetic criteria
of aromaticity, that is, diamagnetic susceptibility exaltation and
(31) (a) Hoarau, J.; Lumbroso, N.; Pacult, N. C. R. Acad. Sci. 1956, 242, 1702.
(b) Flygare, W. H. Chem. ReV. 1974, 74, 653-687.
(32) Under the constraint of all of the bond and dihedral angles to the
corresponding angles in the C3-C2-C1-C6 moiety of 35, all of the bond
lengths were optimized. The magnitude of magnetic shielding in the region
of space corresponding to 0.6-2.0 Å above the ring center of 35 is -0.6
to -1.5 ppm.
9
J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003 957

A R T I C L E S
Tsuji et al.
temperature, the resulting mixture was filtered and subjected to
NICS. [4]Paracyclophane is so strained that the [4]paracyclo-
phane derivative 24 exceeds the corresponding Dewar benzene
form 8 in energy and undergoes thermal cycloreversion to 8
with a half-life of 15 ( 5 min at -20 °C (∆Gq ) 18.3 ( 0.3
kcal mol-1). Thus, the kinetic stability of [4]paracyclophane is
defined not only by the propensity to undergo destructive
intermolecular reactions at the bridgehead sites, but also by the
intrinsic thermal reactivity to rearrange to the Dewar benzene
form.
chromatography on silica gel eluted with ether-hexane (1:1) to afford
1
12 mg of 4 (92%) as pale yellow crystals. H NMR: δ 6.59 (s, 2H),
6.70 (s, 4H). 13C NMR: δ 65.91, 140.09, 140.38, 195.21. IR (KBr):
3064, 1676, 1658, 1592, 1300, 1100, 1014, 808, 752 cm-1. EI-MS:
m/z 158 (M, 2), 130 (25), 104 (70), 102 (80), 76 (100), 50 (51). EI-
HRMS for C10H8O2: calcd 158.0368, found 158.0365. UV-vis λmax
(ether): 229 (ꢀ 14 000), 367 (113), 380 (113), 400 (sh, 75), 423 (sh,
19) nm.
Reduction of 3 to 5. To a solution of 87 mg of 3 (0.54 mmol) and
19 mg of Pd(PPh3)4 (16 µmol) in 29 mL of THF was added 177 mg of
Bu3SnH (0.61 mmol) under argon at 0 °C. After being stirred 0.5 h,
the mixture was concentrated and subjected to chromatography on silica
gel eluted with ether to afford 74 mg of 5 (84%) as a colorless oil. 1H
NMR: δ 1.78 (br s, 1H), 1.9-2.1 (m, 2H), 2.3-2.6 (m, 2H), 4.43 (t,
J ) 4.5 Hz, 1H), 6.55-6.75 (m, 4H). 13C NMR: δ 28.97, 36.54, 65.20,
66.61, 68.22, 140.22, 141.35, 142.03, 144.42, 208.45. IR (neat): 3450
(br), 1705, 1295, 1220, 1110, 1075, 918, 800, 749 cm-1. EI-MS: m/z
162 (M, 0.4), 131 (17), 115 (100), 104 (25), 77 (54), 51 (37). FD-
HRMS for C10H10O2: calcd 162.0681, found 162.0690.
Experimental Section
1
General. H and 13C NMR spectra were recorded on a JEOL EX-
400 (1H/400 MHz, 13C/100 MHz) spectrometer in CDCl3 unless
otherwise indicated. IR spectra were taken on a Hitachi Model 215
grating spectrometer. Mass spectra were recorded on JEOL JMS-DX
300 (EI) and JMS-01SG-2 (FD) spectrometers. UV-vis spectra were
taken on a Hitachi U-4000 spectrophotometer. Column and thin-layer
chromatography (TLC) were performed on silica gel 60 (Merck) of
particle size 63-200 and 5-20 µm, respectively. HPLC was carried
out on LiChrosorb Si60 (Merck, 7 µm). Elemental analyses were
performed by the Center for Instrumental Analysis of Hokkaido
University. Halos (Eiko-sha, Japan) 450 W high-pressure and 120 W
low-pressure mercury lamps were employed as the light sources of
photochemistry. A 450 W high-pressure mercury lamp fitted with a
jacket containing an aqueous K2CrO4 solution and a Corning 7-54 glass
filter, that with a Corning 0-52 filter, and that with Corning 0-52 and
7-60 filters were used as 313, >340, and 365 nm light sources,
respectively. A 500 W xenon lamp fitted with a Corning 3-71 glass
filter, that with a Corning 3-72 filter, and that with a Corning 3-73
filter were used as >470, >440, and >400 nm light sources,
respectively. Tricyclo[4.2.2.01,6]deca-2,4,7,9-tetraene,3d 1,4-bis(trim-
ethylsilyl)-2-butyne,33 and 4,5,6,7-tetrahydro-1-indanone were prepared
following the known procedures. Other reagents and solvents were
obtained from commercial sources and purified prior to use.
Oxidation of 5 to 6. To a solution of 74 mg of 5 (0.46 mmol) in 2
mL of CH2Cl2 was added 148 mg of PCC (0.69 mmol) at 0 °C. After
being stirred 3 h at room temperature, the mixture was filtered through
a short column of Florisil, and the filtrate was subjected to chroma-
tography on silica gel eluted with ether-hexane (1:2) to afford 65 mg
1
of 6 (89%) as a colorless oil. H NMR: δ 2.76 (s, 4H), 6.69 (s, 4H).
13C NMR: δ 37.80, 68.60, 140.71, 204.21. IR (neat): 1702, 1424, 1300,
1278, 1134, 800 cm-1. EI-MS: m/z 160 (M, 0.97), 104 (98), 76 (100),
50 (54). FD-HRMS for C10H8O2: calcd 160.0525, found 160.0528.
Condensation of Malononitrile to 6. To a mixture of 23 mg of 6
(0.14 mmol) and 135 mg of malononitrile (2.0 mmol) was added a
solution of 0.75 mg of â-alanine (8.4 µmol) in 0.15 mL of water. After
being stirred 90 h at room temperature, the resulting mixture was diluted
with 10 mL of CH2Cl2, washed with 5% Na2CO3 (×5) and brine, dried
(MgSO4), and concentrated to give 31 mg of crude product, which was
crystallized from CH2Cl2-hexane to afford 22 mg of 7 (61%) as
Preparation of 2 from 1. A solution of 103 mg of 1 (0.80 mmol)
and 5 mg of 5,10,15,20-tetraphenyl-21H,23H-porphine (TPP, 8 µmol)
in 36 mL of CCl4 was irradiated under gentle bubbling of O2 with a
500 W Xe lamp fitted with a Corning 3-73 filter at 0 °C for 5 h. The
resulting mixture was concentrated in vacuo and subjected to chroma-
tography on silica gel eluted with CH2Cl2-hexane (1:1) to afford 123
1
colorless crystals. H NMR: δ 2.97 (s, 4H), 6.94 (s, 4H). IR (KBr):
2232, 1600, 1592, 1424, 1252, 784, 734, 560 cm-1. FD-MS: m/z 257
(M + 1, 20), 256 (M, 100). FD-HRMS for C16H8N4: calcd 256.0749,
found 256.0733; UV λmax (ether): 219 (sh, ꢀ 11 000), 258 (28 000)
nm. Compound 7 is susceptible to oxidation to produce 8, and the
samples of 7 thus prepared were invariably contaminated by a small
amount of 8.
1
mg of 2 (95%) as colorless crystals. H NMR: δ 5.10-5.17 (m, AA′
part of AA′XX′, 2H), 6.30-6.38 (s at δ 6.34 and XX′ part of AA′XX′,
4H), 6.87 (s, 2H). 13C NMR: δ 52.78, 78.31, 126.81, 142.70, 144.15.
IR (KBr): 3052, 2948, 1365, 1285, 1252, 1140, 1057, 962, 935, 900,
810, 780, 752, 717, 700 cm-1. EI-MS: m/z 160 (M, 0.42), 128 (100),
102 (45), 77 (26), 51 (41). EI-HRMS for C10H8O2: calcd 160.0525,
found 160.0518.
Conversion of 2 to 3. To a solution of 2 (39 mg, 0.24 mmol) in 3
mL of CH2Cl2 was added Et3N (61 mg, 0.60 mmol) under argon at 0
°C. After being stirred 2 h at 0 °C and then 12 h at 5 °C, the resulting
mixture was concentrated and subjected to chromatography on silica
gel eluted with ether to produce 37 mg of 3 (95%) as a colorless oil.
1H NMR: δ 1.94 (d, J ) 8.4 Hz, 1H), 4.90 (br d, J ) 8.4 Hz, 1H),
5.86 (dd, J ) 10.2, 2.3 Hz, 1H), 6.52-6.77 (m, 5H). 13C NMR: δ
61.58, 65.82, 66.51, 126.57, 140.12, 141.26, 143.78, 144.13, 147.57,
197.29. IR (neat): 3600-3200 (br), 1674, 1380, 1302, 1214, 1086,
1064, 844, 794, 738 cm-1. EI-MS: m/z 160 (M, 0.33), 131 (20), 115
(100), 104 (31), 77 (47), 51 (31). EI-HRMS for C10H8O2: calcd
160.0525, found 160.0501. UV λmax (ether): 203 (sh, ꢀ 9000), 335.0
(39), 345.5 (39) nm.
Conversion of 7 to 8. To a solution of 22 mg of 7 (87 µmol) in 0.6
mL of CH3CN were added 30 mg of pyridinium bromide perbromide
(94 µmol) and, after 10 min, 15 mg of pyridine (0.19 mmol) under
argon at 0 °C. After being stirred 1.5 h at 0 °C, the mixture was diluted
with 20 mL of ether, washed with 1 N HCl (×3), water, and brine,
dried (MgSO4), and filtered through a short column of neutral alumina.
The filtrate was concentrated, and the residue was crystallized from
1
CH2Cl2-hexane to afford 16 mg of 8 (73%) as colorless crystals. H
NMR: δ 6.90 (s, 4H), 7.14 (s, 2H). 13C NMR: δ 162.36, 142.92, 131.35,
111.43, 110.11, 89.16, 58.03. IR (KBr): 2228, 1556, 806, 758, 526
cm-1. FD-MS: m/z 255 (M + 1, 20), 254 (M, 100). FD-HRMS for
C16H6N4: calcd 254.0592, found 254.0603. UV λmax (ether): 314 (sh,
ꢀ 66 000), 332 (110 000), 346 (140 000), 361 (110 000) nm.
Condensation of t-Butyl Malonate to 6. To 2.1 mL of dry THF
were added a solution of 351 mg of TiCl4 (1.86 mmol) in 0.21 mL of
CCl4 and, after 15 min at 0 °C, a solution of 17 mg of 6 (0.11 mmol),
220 mg of tert-butyl malonate (1.02 mmol), and 632 mg of pyridine
(7.99 mmol) in 1.04 mL of THF. After being stirred 32 h at room
temperature, the resulting mixture was diluted with 30 mL of ether,
filtered, and concentrated in vacuo to remove pyridine and the unreacted
tert-butyl malonate. The crude crystalline residue was subjected to
chromatography on silica gel eluted with ether-hexane, and the
separated product was crystallized from hexane to afford 16 mg of 9
Oxidation of 3 to 4. To a solution of 13 mg of 3 (81 µmol) in 2 mL
of CH2Cl2 was added 85 mg of MnO2. After being stirred 2 h at room
(33) (a) Guijarro, A.; Yus, M. Tetrahedron 1995, 51, 231-234. (b) Reich, H.
J.; Reich, I. L.; Yelm, K. E.; Holladay, J. E.; Gschneidner, D. J. Am. Chem.
Soc. 1993, 115, 6625-6635.
9
958 J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003

Functionalization of the [4]Paracyclophane System
A R T I C L E S
(26%) as pale yellow crystals. 1H NMR: δ 1.53 (s, 36H), 6.63 (s, 4H),
6.76 (s, 2H). IR (KBr): 1722, 1594, 1372, 1246, 1156, 1132, 850,
800, 742 cm-1. FD-MS: m/z 555 (M + 1, 80), 554 (M, 37), 57 (100).
FD-HRMS for C32H42O8: calcd 554.2880, found 554.2847. UV λmax
(ether): 325.5 (ꢀ 28 000) nm.
stirred 17.5 h at 24 °C, the mixture was treated with 30 mL of saturated
NH4Cl to make the mixture slightly acidic and extracted with ether (5
× 15 mL). The ethereal extracts were combined, dried (MgSO4), and
concentrated to give 213 mg of crude R-formylated ketone as a pale
yellow solid. 1H NMR (90 MHz): δ 0.04 (s, 9H), 0.06 (s, 9H), 1.25-
1.90 (m, 12H), 2.20 (d, J ) 14 Hz, 1H), 2.48 (d, J ) 14 H, 1H), 7.18-
7.22 (m, 1H). IR (neat): 3600-2500 (br), 1690, 1666, 1598, 1250,
1182, 860, 842 cm-1. FD-MS: m/z 363 (M + 1, 35), 362 (100).
To a mixture of the above crude product and 231 mg of TsN3 (1.17
mmol) in 23 mL of CH2Cl2 was added 237 mg of Et3N (2.34 mmol) at
0 °C, and the resulting mixture was stirred for 22 h at 24 °C before the
reaction was quenched with 17 mL of 7% KOH. After the resulting
mixture was diluted with 80 mL of CH2Cl2, the organic layer was
separated from the aqueous layer, washed with water (×3) and brine,
and dried (Na2SO4). After the solvent was removed, the residue was
subjected to chromatography on silica gel eluted with benzene to afford
123 mg of the R-diazoketone (58% from 14) as a reddish brown oil.
1H NMR: δ 0.04 (s, 9H), 0.08 (s, 9H), 1.35-1.58 (m, 5H), 1.44 (d, J
) 14.7 Hz, 1H), 1.47 (s, 2H), 1.52 (d, J ) 14.7 Hz, 1H), 1.65-1.80
(m, 3H), 2.63 (d, J ) 13.7, 1H), 2.90 (d, J ) 13.7 Hz, 1H). IR (neat):
2072, 1666, 1320, 1248, 856, 694 cm-1. FD-MS: m/z 361 (M + 1,
32), 364 (100).
Reduction of 9 with Zn in AcOH. A mixture of 21 mg of 9 (38
µmol), 126 mg of Zn powder (1.93 mmol), and 1.26 mL of AcOH
was heated at 110 °C for 1.5 h, diluted with 15 mL of ether, washed
with water (×5), and dried (MgSO4). After the solvent was removed,
the residue was subjected to preparative TLC on silica gel developed
with ether-hexane (1:9) to afford 4 mg of 10 (Rf ) 0.25, 20%) and 12
mg of 11 (Rf ) 0.11, 57%) both as colorless crystals. 10, 1H NMR: δ
1.50 (s, 18H), 1.52 (s, 18H), 2.62 (s, 4H), 6.63 (s, 4H). IR (KBr): 1722,
1630, 1372, 1286, 1250, 1156, 1134, 852, 800, 742 cm-1. FD-MS:
m/z 557 (M + 1, 17), 556 (M, 20), 57 (100). FD-HRMS for C32H44O8:
calcd 556.3036, found 556.3040. UV λmax (ether): 235 (ꢀ 12 000) nm.
11, 1H NMR: δ 1.46 (s, 36H), 4.00 (s, 2H), 5.60 (s, 2H), 6.50 (s, 4H).
IR (KBr): 1746, 1730, 1372, 1312, 1256, 1162, 1142, 790 cm-1. FD-
MS: m/z 557 (M + 1, 41), 556 (M, 100), 57 (40). FD-HRMS for
C32H44O8: calcd 556.3036, found 556.3027. UV λmax (ether): 295.5 (ꢀ
5600) nm.
Preparation of 12 from 6. To a solution of 13 mg of 6 (81 µmol)
in 0.65 mL of CH2Cl2 were added successively 77 mg of i-Pr2NEt (0.60
mmol) and 108 mg of i-Pr3SiOTf (0.35 mmol) under argon at 0 °C.
After being stirred 26 h at 0 °C, the resulting mixture was diluted with
10 mL of pentane, washed with water (×5), and dried (Na2SO4). After
removal of the volatile components in vacuo, the residue was subjected
to chromatography on neutral alumina eluted with benzene to afford
(Photo-Wolff rearrangement) A solution of 60 mg of the R-diaz-
oketone (0.22 mmol) and 10 mg of Et3N (0.10 mmol) in 4 mL of dry
MeOH was irradiated through Pyrex with a 450 W high-pressure
mercury lamp under argon at 5 °C until gas evolution ceased (3 h).
After the solvent was removed in vacuo, the residue was subjected to
preparative TLC on silica gel developed with benzene to afford 64 mg
1
1
31 mg of 12 (82%) as a colorless oil. H NMR (500 MHz, THF-d8):
of the methyl ester (80%) as an ca. 9:1 mixture of stereoisomers. H
δ 1.11 (d, J ) 4 Hz, 36H), 1.15-1.25 (m, 6H), 4.59 (s, 2H), 6.60 (s,
4H). FD-MS: m/z 473 (M + 1, 41), 472 (M, 100). FD-HRMS for
C28H48O2Si2: calcd 472.3193, found 472.3145.
NMR (the major component): δ 0.04 (s, 9H), 0.08 (s, 9H), 1.12 (d, J
) 14.7 Hz, 1H), 1.22-1.44 (m, 5H), 1.38 (d, J ) 14.7 Hz, 1H), 1.46
(d, J ) 14.7 Hz, 1H), 1.55 (d, J ) 14.7 Hz, 1H), 1.60-1.68 (m, 1H),
1.76-1.88 (m, 2H), 1.90 (dd, J ) 12.2, 9.8 Hz, 1H), 2.27 (dd, J )
12.2, 6.3 Hz, 1H), 2.93 (dd, J ) 9.8, 6.4 Hz, 1H), 3.62 (s, 3H). IR
Photocycloaddition of 1,4-Bis(trimethylsilyl)-2-butyne to 13. A
solution of 0.89 g of 13 (6.5 mmol) and 11.8 g of 1,4-bis(trimethylsilyl)-
2-butyne (60 mmol) in 160 mL of acetone was irradiated through Pyrex
with a 450 W high-pressure mercury lamp under argon at -20 °C until
13 was largely consumed (2.5 h) and then concentrated. After the
unreacted 2-butyne was recovered by distillation in vacuo (∼100 °C/
17 mmHg), the residue was subjected to chromatography on silica gel
eluted with benzene-hexane (1:1) to afford 392 mg of 14 (18%) as a
colorless oil. 1H NMR: δ 0.03 (s, 9H), 0.09 (s, 9H), 1.35 (d, J ) 15.6
Hz, 1H), 1.41 (d, J ) 15.6 Hz, 1H), 1.47 (s, 2H), 1.28-1.62 (m, 7H),
1.71 (m, 2H), 1.89 (br dd, J ) 13.2, 9.3 Hz, 1H), 2.08 (br dd, J )
17.6, 7.8 Hz, 1H), 2.76 (ddd, J ) 17.6, 12.2, 9.3 Hz, 1H). 13C NMR:
δ -0.31, 0.02, 16.22, 17.13, 20.28, 20.50, 22.48, 26.30, 31.88, 35.50,
50.09, 57.97, 138.06, 144.71, 220.39. IR (neat): 1724, 1654, 1250,
856, 846, 838, 694 cm-1. FD-MS: m/z 335 (M + 1, 32), 334 (100).
Photocycloaddition of 1,4-Bis(tri-i-propylsilyl)-2-butyne to 13. A
solution of 175 mg of 13 (1.29 mmol) and 2.36 g of 1,4-bis(tri-i-
propylsilyl)-2-butyne in 18 mL of acetone was irradiated through Pyrex
with a 450 W high-pressure mercury lamp under argon at -20 °C until
13 was largely consumed (10 h) and then concentrated. After the
unreacted 2-butyne was recovered by distillation in vacuo (∼120 °C/2
× 10-3 mmHg), the residue was subjected to preparative TLC on silica
gel developed with benzene-hexane (1:1) to afford 21 mg of 15 (3.2%)
as a colorless oil. 1H NMR: δ 1.06-1.09 (m, 42H), 1.39 (d, J ) 14.6
Hz, 1H), 1.47 (d, J ) 14.6 Hz, 1H), 1.54 (s, 2H), 1.24-1.78 (m, 9H),
1.89 (br dd, J ) 13.2, 8.8 Hz, 1H), 2.10 (br dd, J ) 17.8, 8.0 Hz, 1H),
2.75 (ddd, J ) 17.8, 12.0, 9.6 Hz, 1H). 13C NMR: δ 8.94, 9.46, 12.07,
12.35, 18.98, 19.83, 20.19, 23.41, 25.06, 30.82, 36.07, 50.33, 58.82,
138.66, 146.18, 219.84. IR (neat): 1726, 1466, 882, 662 cm-1. FD-
MS: m/z 503 (M + 1, 35), 502 (100).
(neat): 1738, 1644, 1448, 1436, 1250, 1200, 1170, 858, 840, 696 cm-1
.
FD-MS: m/z 365 (M + 1, 35), 364 (100).
(R-Selenenylation of the ester) A solution of 38 mg of i-Pr2NH (0.38
mmol) in 0.4 mL of THF was treated with 190 µL of 1.56 M BuLi in
hexane (0.30 mmol) at -78 °C to afford a LDA solution, to which a
solution of 54 mg of the above ester (150 µmol) in 0.4 mL of THF
was added. The resulting mixture was warmed to -40 °C over a period
of 45 min, stirred 20 min, cooled again to -78 °C, and treated with a
solution of 90 mg of PhSeBr (0.38 mmol) and 54 mg of HMPA (0.30
mmol) in 0.5 mL of THF. The mixture was warmed to room
temperature over a period of 3 h, diluted with 10 mL of ether, washed
with saturated NaHCO3 (×3), and dried (MgSO4). After the solvent
was removed, the residue was subjected to preparative TLC on silica
gel developed with benzene-hexane (1:1) to give 60 mg of the
R-selenenylated ester (79%) as an ca. 10:1 mixture of stereoisomers.
1H NMR (the major component, 90 MHz): δ 0.01 (s, 9H), 0.04 (s,
9H), 1.20-2.10 (m, 12H), 2.08 (d, J ) 13.2 Hz, 1H), 2.87 (d, J )
13.2 Hz, 1H), 3.53 (s, 3H), 7.20-7.38 (m, 3H), 7.45-7.60 (m, 2H).
FD-MS: m/z 522 (42), 521 (49), 520 (100), 519 (24), 518 (57), 517
(24), 516 (24).
(Conversion of the R-selenenylated ester to 16) To a solution of 30
mg of the R-selenenylated ester (156 µmol) in 0.5 mL of CH2Cl2 cooled
to 0 °C were added 20 mg of pyridine (0.26 mmol), 82 µL of water,
and 82 µL of 30% H2O2. The mixture was warmed to 24 °C, stirred
2.5 h, diluted with 10 mL of ether, washed with water, saturated
NaHCO3, and water, and dried (MgSO4). To the ethereal solution
containing MgSO4 was added 40 µL of pyridine, and the mixture was
refluxed for 2.5 h. The resulting mixture was washed with water (×5),
dried (MgSO4), concentrated, and subjected to preparative TLC on silica
gel developed with benzene-hexane (1:1) to afford 28 mg of 16 (67%)
Conversion of 14 into 16. (R-Diazotization of 14) To a mixture of
218 mg of HCO2Et (2.95 mmol) and 160 mg of MeONa (2.96 mmol)
in 2 mL of dry benzene cooled to -5 °C was added a solution of 197
mg of 14 (0.59 mmol) in 1.5 mL of benzene under argon. After being
1
as a colorless oil. H NMR: δ 0.05 (s, 9H), 0.07 (s, 9H), 1.32-1.42
(m, 4H), 1.42 (d, J ) 14.2 Hz, 1H), 1.54 (s, 2H), 1.56 (d, J ) 14.2 Hz,
9
J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003 959

A R T I C L E S
Tsuji et al.
1H), 1.83-1.90 (m, 2H), 1.90-1.98 (m, 1H), 1.99-2.08 (m, 1H), 3.76
(s, 3H), 7.31 (s, 1H). IR (neat): 1720, 1650, 1300, 1248, 1074, 856,
840 cm-1. FD-MS: m/z 363 (M + 1, 38), 362 (M, 100). FD-HRMS
for C20H34O2Si2: calcd 362.2098, found 362.2126.
CH2Cl2. The mixture was stirred 30 min at -78 °C, treated with 20
mg of Et3N (0.20 mmol), warmed to room temperature over a period
of 2 h, and stirred 3 h. The resulting mixture was diluted with 10 mL
of CH2Cl2, washed with saturated NaHCO3 (×2) and brine, dried
(MgSO4), and filtered. To the filtrate was added 0.10 mL of pyridine,
and the mixture was refluxed for 6 h. The resulting mixture was washed
with saturated NaHCO3 (×5), dried (MgSO4), concentrated, and
subjected to preparative TLC on silica gel developed with ether-hexane
Conversion of 15 into 17. (R-Diazotization of 15) To a mixture of
87 mg of 15 (0.17 mmol) and 66 mg of HCO2Et (0.89 mmol) in 1 mL
of benzene was added 48 mg of MeONa (0.89 mmol) under argon at
5 °C. After being stirred 24 h at 24 °C, the mixture was treated with
10 mL of saturated NH4Cl to make the mixture slightly acidic and then
it was extracted with ether (5 × 5 mL). The ethereal extracts were
combined, dried (MgSO4), and concentrated to give 55 mg of crude
1
(3:7) to afford 6 mg of 17 (60%) as a colorless oil. H NMR (500
MHz, CD2Cl2, 23 °C): δ 1.00-1.20 (m, 36H), 1.32-1.46 (m, 6H),
1.48-1.72 (m, 4H), 1.83-2.08 (m, 8H), 3.03 (s, 6H), 6.62 (s, 1H);
(-60 °C) δ 0.90-1.15 (m, 36H), 1.18-1.38 (m, 6H), 1.44 (d, J ) 15
Hz, 1H), 1.56 (d, J ) 15 Hz, 1H), 1.59 (s, 2H), 1.75-1.92 (m, 8H),
2.92 (s, 3H), 3.07 (s, 3H), 6.59 (s, 1H). FD-MS: m/z 544 (M + 1, 50),
543 (M, 100). FD-HRMS for C33H61ONSi2: calcd 543.4291, found
543.4266.
1
R-formylated ketone as a pale yellow solid. H NMR: δ 0.93-1.07
(m, 42H), 1.10-1.68 (m, 12H), 2.14 (dd, J ) 15.6, 2.0 Hz, 1H), 2.38
(d, J ) 15.6 Hz, 1H), 7.08 (br s, 1H), 7.16 (s, 1H). IR (Nujol): 3600-
2600 (br), 1668, 1600, 1162, 1120, 1070, 882 cm-1. FD-MS: m/z 531
(M + 1, 45), 530 (100).
Photolysis of the Dewar Benzene Derivatives and Measurement
of Electronic Absorption Spectra of [4]Paracyclophanes. A 1:1
mixture of ether and 2-methylbutane was used as solvent unless
otherwise indicated. A solution of Dewar benzene precursor in 5 mL
of solvent was placed in a quartz cuvette and degassed by freeze-
pump-thaw cycles. After the cuvette was sealed under vacuum, the
solution was frozen in liquid nitrogen to give a clear colorless glass
and irradiated with a light source of specified wavelength.
(a) Photolysis of 4. Irradiation of a 0.9 M solution of 4 with 254
nm light led to the development of an absorption with λmax 330 and
410 nm (relative intensity: ca. 3:2), which was efficiently bleached
upon secondary irradiation with >400 nm light to restore the original
spectrum. The species generated by the irradiation of 254 nm light
was also thermally highly labile and had suffered complete decomposi-
tion when the glass was thawed below -120 °C and immediately
refrozen in liquid nitrogen. On the basis of these observations, the
absorption was assigned to 23. The absorption ceased to grow in a
short time of irradiation, and the extended irradiation led to the
formation of second product which was stable thermally as well as
photochemically and exhibited an absorption with λmax 255 and 335
nm. This absorption was similar in shape to that reported for 2634and
was accordingly assigned tentatively to 26.
To a mixture of the above crude product and 40 mg of TsN3 (0.20
mmol) in 4 mL of CH2Cl2 was added 42 mg of Et3N (0.42 mmol) at
0 °C, and the mixture was stirred for 24 h at 24 °C before quenching
the reaction with 3 mL of 7% KOH. The resulting mixture was diluted
with 20 mL of CH2Cl2, and the organic layer separated from the aqueous
layer was washed with water (×3) and brine, and dried (Na2SO4). After
the solvent was removed, the residue was subjected to preparative TLC
on silica gel developed with ether-hexane (1:19) to afford 29 mg of
the R-diazoketone (55% from 15) as a reddish brown oil. 1H NMR: δ
1.08 (br s, 42H), 1.20-1.85 (m, 8H), 1.48 (d, J ) 14 Hz, 1H), 1.50 (d,
J ) 15 Hz, 1H), 1.51 (d, J ) 15 Hz, 1H), 1.60 (d, J ) 14 Hz, 1H),
2.69 (d, J ) 13.7, 1H), 2.93 (d, J ) 13.7 Hz, 1H). 13C NMR: δ 8.60,
8.93, 11.94, 12.11, 18.86, 19.02, 24.42, 28.42, 30.23, 58.07, 60.94,
66.77, 140.86, 144.57, 201.08. IR (neat): 2072, 1666, 1464, 1322, 1188,
1178, 1014, 918, 882, 662 cm-1. FD-MS: m/z 529 (M + 1, 48), 528
(100).
(Photo-Wolff rearrangement) A solution of 29 mg of the R-diaz-
oketone (55 µmol) in 1.2 mL of Me2NH was irradiated through Pyrex
with a 450 W high-pressure mercury lamp for 6 h under argon at -20
°C. The resulting mixture was concentrated and subjected to preparative
TLC on silica gel developed with ether-hexane (1:1) to afford 19 mg
of carboxamide (66%) as a stereochemically nearly homogeneous,
1
(b) Photolysis of 6. Irradiation of a 0.3 M solution of 6 in EPA (a
5:5:2 mixture of ether, pentane, and ethanol) with 254 nm light led to
the slow development of an absorption with λmax 260 and 300 nm
(relative intensity: ca. 3:1), which remained unchanged when the
resulting glass was irradiated with 365 nm light or thawed and warmed
to room temperature and, thus, appeared incompatible with the
corresponding [4]paracyclophane. In a separate experiment, a solution
of 5 mg of 6 in 6 mL of EPA was degassed and frozen in liquid
nitrogen. After repeating an irradiation (254 nm)-thaw-freeze cycle
colorless oil. H NMR: δ 1.08 (br s, 36H), 1.04-1.18 (m, 6H), 1.21
(d, J ) 14.5 Hz, 1H), 1.22-1.78 (m, 7H), 1.58 (s, 2H), 1.68 (d, J )
14.5 Hz, 1H), 1.86 (dd, J ) 11.7, 9.8 Hz, 1H), 2.05 (m, 1H), 2.49 (dd,
J ) 6.4, 11.7 Hz, 1H), 2.86 (s, 3H), 2.99 (s, 3H), 3.05 (dd, J ) 9.8,
6.4 Hz, 1H). 13C NMR: δ 8.83, 9.40, 12.09, 12.16, 18.89, 18.97, 19.15,
19.22, 21.78, 21.82, 29.52, 31.95, 32.17, 35.37, 37.27, 42.34, 43.05,
49.25, 137.95, 144.42, 173.21. IR (neat): 1730, 1652, 1466, 1392, 1150,
882, 660 cm-1. FD-MS: m/z 546 (M + 1, 47), 545 (100).
(R-Selenenylation of the carboxamide) A solution of 129 mg of i-Pr2-
NH (1.28 mmol) in 0.7 mL of THF was treated with 0.64 mL of 1.56
M BuLi in hexane (1.00 mmol) at -78 °C to afford a LDA solution,
to which a solution of 57 mg of the above amide (0.10 mmol) in 1 mL
of THF was added. The resulting mixture was warmed to -40 °C over
a period of 1 h, stirred 3 h, cooled again to -78 °C, and treated with
a solution of 302 mg of PhSeBr (1.28 mmol) and 179 mg of HMPA
(1.00 mmol) in 1 mL of THF. The mixture was warmed to room
temperature over a period of 3 h, diluted with 10 mL of ether, washed
with saturated NaHCO3 (×3), and dried (MgSO4). After the solvent
was removed, the residue was subjected to preparative TLC on silica
gel developed with ether-hexane (3:7) to give 22 mg of the R-sele-
1
several times, the resultant mixture was analyzed with H NMR and
UV. These spectra indicated the formation of 2035 in g60% yield as a
1
single detectable product. H NMR (90 MHz): δ 3.90 (s, 4H), 7.70-
7.80 (m, AA′ part of AA′XX′, 2H), 8.00-8.10 (m, XX′ part of AA′XX′,
2H).
(c) Photolysis of 7. Irradiation of a 5 mM solution of 7 with 254
nm light led to development of an absorption with λmax 327 and 440
nm (relative intensity: ca. 1:1), which was efficiently bleached upon
secondary irradiation with >440 nm light. Difference spectra for the
initial and secondary irradiation demonstrated that 7 was regenerated
at the expense of the product upon the secondary irradiation, and, thus,
the developed absorption was assigned to 18. The absorption remained
unchanged indefinitely at 77 K in the dark, but decayed with a half-
life of 10 ( 3 min at -90 °C. The extended irradiation with 254 nm
1
nenylated amide (31%). H NMR (the major component): δ 1.12 (br
s, 42H), 1.20-2.20 (m, 12H), 2.47 (d, J ) 13.5 Hz, 1H), 2.78 (s, 3H),
3.07 (s, 3H), 3.36 (d, J ) 13.5 Hz, 1H), 7.1-7.4 (m, 5H). FD-MS:
m/z 703 (36), 702 (53), 701 (100), 700 (29), 699 (57), 698 (24), 697
(19).
(Conversion of the R-selenenylated amide to 17) To a solution of
18 mg of the R-selenenylated amide (20 µmol) in 1.5 mL of CH2Cl2
cooled to -78 °C was added 185 µL of 0.12 M mCPBA (22 µmol) in
(34) (a) Singh, I.; Ogata, R. T.; Moore, R. E.; Chang, C. W. J.; Scheuer, P. J.
Tetrahedron 1965, 24, 6053. (b) Anderson, L. C.; Roedel, M. J. J. Am.
Chem. Soc. 1945, 67, 955.
(35) (a) Cooper, M. A.; Manatt, S. L. J. Am. Chem. Soc. 1970, 92, 1605-1614.
(b) Pearson, M. S.; Jensky, B. J.; Greer, F. X.; Hagstrom, J. P.; Wells, N.
M. J. Org. Chem. 1978, 43, 4617-4622.
9
960 J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003

Functionalization of the [4]Paracyclophane System
A R T I C L E S
light led to the formation of the second product, which was thermally
stable and exhibited an absorption with λmax 323 nm. In a separate
experiment, a solution of 4 mg of 7 in 5 mL of ether-2-methylbutane
(1:1) was degassed and frozen in liquid nitrogen. After an irradiation
(254 nm)-thaw-freeze cycle was repeated several times, the resultant
mixture was analyzed with 1H NMR and UV, which showed the
formation of 21 as a single detectable product. 1H NMR (90 MHz): δ
3.21 (s, 4H), 7.75-7.85 (m, AA′ part of AA′XX′, 2H), 8.25-8.35 (m,
XX′ part od AA′XX′, 2H).
(d) Photolysis of 8. Irradiation of a 0.5 mM solution of 8 with 365
nm light led to the development of a broad absorption band in the range
280-410 nm at the expense of 8. The process was reversed upon
secondary irradiation with >400 nm light as demonstrated by the
difference spectra recorded for the initial and secondary irradiation.
The newly developed absorption decayed with a half-life of 12 ( 2
min at -20 °C. The generated species was confirmed as 24 by the 1H
NMR measurement described below.
340, 380 (sh), and 450 nm (relative intensity: ca. 8:3:2.5:1), which
was cleanly bleached upon secondary irradiation with >470 nm light
to restore the original spectrum. The photoreaction of 16 reached a
quasi-photostationary state at ca. 15% conversion. On the basis of this
photoreversibility and similarity of the spectrum in shape to that of 33
reported previously, the developed absorption was assigned to 31. The
species was thermally unstable and suffered decomposition with a half-
life of 20 ( 5 min at -130 °C.
(h) Photolysis of 17. Irradiation of a 20 mM solution of 17 with
254 nm light led to the development of an absorption with λmax 302
and 390 nm (relative intensity: ca. 3:1), which was cleanly bleached
upon secondary irradiation with >440 nm light to restore the original
spectrum. The photoreaction of 17 reached a quasi-photostationary state
at ca. 5% conversion. On the basis of this photoreversibility, the
developed absorption was assigned to 32. The absorption decayed with
a half-life of 100 ( 10 min at -50 °C.
1
Measurement of the H NMR Spectrum of 24. A solution of 8
(e) Photolysis of 10. Irradiation of a 2 mm solution of 10 with 254
nm light led to the development of absorption bands at 289, 352, and
428 nm. The latter two bands were bleached upon secondary irradiation
with >400 nm light, while the band at 289 nm remained intact.
Difference spectra recorded for the initial and secondary irradiation
showed that 10 was regenerated at the expense of the product exhibiting
the bands at 352 and 428 nm, to which structure 19 was thus assigned.
The absorption due to 19 decayed with a half-life of 30 ( 5 min at
-20 °C. The absorption due to 19 ceased to grow at ca. 3% conversion,
and the continued irradiation with 254 nm light led to continuous growth
of the band at 289 nm. The second product was stable thermally as
well as photochemically and was tentatively assigned structure 22.
(f) Photolysis of 12. Irradiation of a 15 mM solution of 12 with
313 nm light led to the development of an absorption with λmax 269,
278 (sh), and 360 nm (relative intensity: ca. 10:9:1), which was
efficiently bleached upon secondary irradiation with >340 nm light.
Difference spectra recorded for the initial and secondary irradiation
demonstrated that 12 was regenerated at the expense of the photoproduct
upon the secondary irradiation. Thus, structure 29 was assigned to the
latter. The absorption due to 29 decayed with a half-life of 30 ( 5 min
at -50 °C. The absorption ceased to grow at ca. 2% conversion, and
the continued irradiation with 313 nm light induced decomposition of
12/29 to an unidentifiable mixture.
(ca. 2 mg) in 0.6 mL of CD2Cl2 was placed in a Pyrex NMR tube and
degassed by freeze-pump-thaw cycles. The tube was sealed under
vacuum and connected to a mechanical stirrer to be rotated during
1
irradiation. The H NMR spectrum (400 MHz) recorded at -75 °C
after irradiating the sample with 365 nm light at ca. -90 °C exhibited
two new signals at δ 5.85 and 7.97 with an intensity ratio of ca. 2:1,
besides the signals due to 8 at δ 6.93 and 7.20. The intensities of the
former signals ceased to increase at ca. 6% conversion. Upon subsequent
irradiation with a >400 nm light source, the signals due to the product
cleanly decayed to restore the original two line spectrum.
Acknowledgment. This research was supported by Grant-
in-Aids for Scientific Research (06453031 and 08454192) from
the Ministry of Education, Science, Sports, and Culture of Japan.
This paper is dedicated to Emeritus Professor Soichi Misumi
on the occasion of his 77th birthday.
Supporting Information Available: Absolute energies and
Cartesian coordinates specifying the positions of atoms for
geometry-optimized 8, 24, and 34 at the B3LYP/6-31+G* level
of theory (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
(g) Photolysis of 16. Irradiation of a 15 mM solution of 16 with
254 nm light led to the development of an absorption with λmax 302,
JA021206Y
9
J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003 961