5-Fluoro- and 5-methyl-1,3-didehydrobenzene - A matrix isolation study

Article abstract of DOI:10.1039/a905189i

The UV photolysis of [2.2]metaparacyclophanediones 4b and 4c leads to α-cleavage and formation of CO, p-xylylene 5, and m-benzynes 2b and c with a fluorine or methyl substituent in the 5-position, respectively. The IR spectra of 2b and 2c were assigned by

Full text of DOI:10.1039/a905189i

View Article Online / Journal Homepage / Table of Contents for this issue
5-Fluoro- and 5-methyl-1,3-didehydrobenzene—a matrix isolation
study
Wolfram Sander* and Michael Exner
Lehrstuhl für Organische Chemie II der Ruhr-Universität, D-44780 Bochum, Germany
Received (in Cambridge, UK) 28th June 1999, Accepted 24th August 1999
The UV photolysis of [2.2]metaparacyclophanediones 4b and 4c leads to α-cleavage and formation of CO, p-xylylene
5, and m-benzynes 2b and c with a fluorine or methyl substituent in the 5-position, respectively. The IR spectra of 2b
and 2c were assigned by comparison with that of the parent m-benzyne 2a and with spectra calculated at the BLYP/6-
311G(d,p) level of theory. Both band positions and intensities of the experimental spectra are in excellent agreement
with the DFT calculations. This method thus provides a reliable tool for the investigation of ground state properties
of m-benzynes. The benzynes 2 exhibit a very strong IR absorption at 545 2 cmϪ1 which is almost not affected by
the substituents and therefore can be used to identify m-benzynes in product mixtures.
Introduction
Dehydrobenzenes 1–3 are highly reactive intermediates of
importance in organic synthesis1 and to the biological activity
of natural products2 and have been investigated by trapping
studies, gas phase spectroscopy, matrix isolation techniques and
numerous theoretical methods.3,4 The understanding of reac-
tion mechanisms involving arynes as reactive intermediates has
profited enormously by the accurate determination of the
thermochemical properties of these species, including the
singlet–triplet splittings (∆EST), published by Robert Squires
and co-workers.5–12
synthesized from quinone diazo compounds 8 via a carbene
rearrangement.16,17 The OH substitution in 7 causes only little
change of the ring geometry compared to 2a. The substitution
results in a stabilization of the benzyne, however, somewhat less
than the stabilization of benzene by OH substitution.17
The properties of the benzynes 1–3 are determined by the
amount of through bond and through space interaction of the
in-plane σ-orbitals at the radical centers. While o-benzyne 1 is
best described as a closed-shell molecule with ∆EST = 37.5 0.3
kcal molϪ1, p-benzyne 3 is a diradical with weakly coupled
radical centers and a singlet–triplet splitting of only 3.8 0.4
kcal molϪ1. m-Benzyne 2a shows intermediate properties with
∆EST = 21.0 0.3 kcal molϪ1 and a heat of formation of 121.9
3.1 kcal molϪ1, half way between that of 1 (105.9 3.3 kcal
The synthesis of matrix-isolated m-benzynes from 8 is
molϪ1) and 3 (137.8 2.9 kcal molϪ1).12 The stabilizing
restricted to dehydrophenols 7, while the problem with per-
through-space interaction between the radical centers in 2a is
oxides of type 6 as precursors is the low volatility in combin-
increased by a strong distortion of the benzene hexagon which
ation with the rapid decomposition of the peroxides during
reduces the distance between the radical centers from 2.4 Å
deposition of the matrix from the gas phase. Derivatives of
(hexagon) to about 2.14 Å (CCSD(T) calculation)13 and leads
metaparacyclophanes 4 are thermally much more stable and
to the large singlet–triplet splitting. Consequently, the gas phase
thus should provide useful precursors for the synthesis of a
reactivity of a derivative of 2 resembles both that of the phenyl
variety of substituted m-benzynes. Here we describe the matrix
radical (although with reduced radical reactivity) and that of
isolation and spectroscopic characterization of two novel m-
o-benzyne 1 (addition reactions).14
benzynes with a fluorine and a methyl substituent, respectively,
m-Benzyne 2a was generated in argon matrices by two
in the 5-position.
independent routes.15 The UV photolysis of matrix-isolated
(Ar, 10 K) [2.2]metaparacyclophane-2,9-dione 4a resulted in
fragmentation into p-xylylene 5, CO, and benzyne 2a. The flash
vacuum pyrolysis of isophthaloyl diacetyl peroxide 6 produced
CO2, methyl radicals, and again 2a. Benzyne 2a was identified
by comparison of the experimental IR spectrum with ab-initio
(CCSD(T)) and DFT (UB3LYP) calculations.
Results
5-Fluoro-1,3-didehydrobenzene 2b
5-Fluoro[2.2]metaparacyclophane-2,9-dione 4b, matrix-isolated
in argon at 10 K, exhibits strong IR absorptions at 1727, 1695
The only derivatives of 2 that have been matrix-isolated and
(C᎐O str. vibrations), and 1296 (C–F str. vibration) cmϪ1. The
᎐
spectroscopically characterized are 2,4-didehydrophenols 7,
carbonyl region of 4b is split to a complex pattern of IR bands,
J. Chem. Soc., Perkin Trans. 2, 1999, 2285–2290
This journal is © The Royal Society of Chemistry 1999
2285

View Article Online
Table 1 IR Spectroscopic data of 5-fluoro-1,3-didehydrobenzene 2b
Argon, 10 Ka BLYP/6-311G(d,p)b
No.c
Sym
ν
Id
ν
Id
Assignmente
22
21
19
15
13
12
11
9
8
7
6
4
a1
a1
a1
a1
b2
a1
b1
b1
b1
a1
b2
b1
a1
3121
1690
1402
1154
965
922
802
747
614
603
535
3
C–H str., sym.
3
100
39
8
13
17
14
1
C–C str. ϩ C–C bend., sym.
C–F str. ϩ C–H bend., sym.
C–C str. ϩ C–H bend. ϩ C–F str., sym
C–H bend. ϩ C–C str.
C–C bend., sym.
C–H wag., o.o. phase
1440
1184
949
884
810
100
46
13
24
18
9
6
6
50
7
767
C–H wag., o.o. phase
C–C wag. ϩ C–H wag., asym.
C–C str. ϩ C–H bend., sym.
C–C rock. ϩ C–C bend., asym.
C–C wag. ϩ C–H wag., asym.
C–C bend., sym.
3
47
6
543
532
514
351
2
8
a IR spectroscopic data of 2b obtained after 12 h irradiation (λ = 254 nm) of 4b, matrix-isolated in argon at 10 K. b Vibrational data of 2b calculated
at the BLYP/6-311G(d,p) level of theory. c Number of vibration according to the calculation. d Relative intensity based on the strongest peak (100%).
e Assignment of calculated and experimental vibrations based on comparison of peak positions and intensities.
presumably due to Fermi resonances. Irradiation with λ > 285
nm for 14 h results in a ca. 80% decrease of all absorptions of
4b and formation of large amounts of CO (2137 cmϪ1). The
photolysis of 4b is almost quantitative after 12 h with λ = 254
nm irradiation, which indicates that the α-cleavage is highly
efficient under these conditions. Simultaneously, p-xylylene 5
with characteristic IR bands at 1740, 1605, 869, and 469 cmϪ1 is
formed. A novel compound with the strongest absorptions at
1440 and 543 cmϪ1 is assigned to 5-fluoro-1,3-didehydro-
benzene 2b (Fig. 1, Table 1). The 543 cmϪ1 absorption is
very close to an intense absorption at 547 cmϪ1 in the parent
m-benzyne, which was assigned to the b1 symmetrical ring
deformation mode.15 Fluorine substitution at C(5) obviously has
only a minor influence on the position of this absorption.
Fig. 1 Bottom: difference IR spectrum (in absorbance) showing the
UV photochemistry of cyclophane 4b. The IR absorptions of the
photoproducts 5 and 2b are pointing upwards. Top: IR spectrum of 2b
calculated at the BLYP/6-311(d,p) level of theory (in absorbance). The
peak numbers refer to Table 1.
with subsequent matrix isolation, which produces the same IR
spectrum as the matrix photolysis of the cyclophane.15
A DFT calculation with the BLYP functional and the
5-Methyl-1,3-didehydrobenzene 2c
6-311G(d,p) basis set nicely reproduces both band positions
and relative intensities of the experimental spectrum (Table 1).
Specifically, the b1 symmetrical mode at 543 cmϪ1 is reproduced
within 8 cmϪ1 (no scaling of vibrations applied). This vibration
changes the bond angles—and thus the hybridization—at the
radical centers C(1) and C(3), and therefore critically depends
on the level of theory used for the calculation. The correspond-
ing a1 symmetrical mode is predicted to be of low intensity at
351 cmϪ1 and not observed experimentally. The most intense
absorption in the spectrum at 1440 cmϪ1 is assigned to a com-
bination of C(5)–F stretching with the symmetrical in plane
bending of the C(4)–H and C(6)–H bonds. In fluorobenzene the
The methyl-substituted benzyne 2c was synthesized analo-
gously to the fluoro derivative 2b. 5-Methyl[2.2]metaparacyclo-
phane-2,9-dione 4c, matrix-isolated in argon at 10 K, exhibits
several intense IR absorptions between 1724 and 1689 cmϪ1 in
the carbonyl stretching region. UV photolysis of 4c produces
the expected fragmentation products CO and p-xylylene 5, and
a series of absorptions assigned to benzyne 2c (Fig. 2). The
most intense IR absorption of 2c is the asymmetrical ring
deformation mode at 543 cmϪ1 at the same position as in 2b,
again close to the corresponding vibration of the parent ben-
zyne 2a (547 cmϪ1) and in good agreement with calculations at
the BLYP/6-311G(d,p) level of theory (531 cmϪ1, Table 2). The
symmetrical ring deformation mode of 2c is calculated as a
weak absorption at 346 cmϪ1 and due to experimental limit-
ations not observed in the experimental spectrum. Similar to
the 1440 cmϪ1 absorption of 2b, the IR spectrum of 2c shows a
medium intensity band at 1462 cmϪ1, assigned to a combination
of C(5)–C(methyl) stretching and in plane C(4)–H and C(6)–H
bending vibration. Due to the lower polarity of the C(2)–
C(methyl) bond compared to the C(5)–F bond in 2c this
vibration is of lower intensity, while the band position is only
slightly shifted.
C–F stretching vibration is found at 1496 cmϪ1 18
.
Since the photolysis of matrix-isolated cyclophane 4b pro-
duces 2b, 5, and CO in close contact in the same matrix cage, a
perturbation of the spectra might occur. However, compared to
an independently synthesized and matrix-isolated sample, the
IR bands of 5 are only slightly shifted and broadened. The
good agreement of the experimental and calculated spectrum
of 2b, and the shape of the IR bands with comparatively
narrow linewidths suggest that there is only a small perturb-
ation of the spectrum of 2b. For the parent m-benzyne 2a
this was confirmed by independent synthesis in the gas phase
2286
J. Chem. Soc., Perkin Trans. 2, 1999, 2285–2290

View Article Online
Table 2 IR Spectroscopic data of 5-methyl-1,3-didehydrobenzene 2c
Argon, 10 Ka BLYP/6-311G(d,p)b
No.c
Sym
ν
Id
ν
Id
Assignmente
33
32
31
30
29
28
27
26
25
23
22
18
16
15
14
13
12
10
9
aЈ
aЈ
aЉ
aЉ
aЈ
aЈ
aЈ
aЉ
aЈ
aЈ
aЈ
aЈ
aЈ
aЉ
aЉ
aЈ
aЈ
aЈ
aЈ
aЈ
aЉ
aЉ
aЈ
aЈ
aЈ
3111
3089
3087
3032
3004
2953
1689
1475
1452
1429
1371
1159
1020
996
925
921
822
745
608
556
531
528
10
C–H str., sym.
C–H str., sym.
C–H str.
C–H str. (CH3)
7
13
12
15
15
4
12
9
53
1
3
7
11
5
3
20
23
2
11
100
2
2
15
2
C–H str. (CH3)
C–H str. (CH3), sym.
C–C bend. ϩ C–C str., sym.
C–H twist (CH3)
C–H bend. (CH3) ϩ C–H wag. (CH3)
C–C bend. ϩ C–C str., sym.
C–H wag. (CH3)
C–C str. ϩ C–H bend., sym.
C–H twist (CH3)
C–H bend. (CH3) ϩ C–C str.
C–H bend. (CH3) ϩ C–C str.
C–C bend. ϩ C–C str., sym.
C–H wag., o.o. phase
C–H wag., i. phase
C–C str. ϩ C–H bend., sym.
C–C wag. ϩ C–H wag., asym.
C–C rock. ϩ C–C bend., asym.
C–C twist., asym.
C–C wag. ϩ C–H wag., asym.
C–C bend., sym.
1570
1533
1462
16
11
42
1000
13
824
768
21
24
8
7
6
5
4
2
570
543
13
100
486
346
176
C–C wag., sym.
a IR spectroscopic data of 2c obtained after 22 h irradiation (λ = 365 nm) of 4c, matrix-isolated in argon at 10 K. b Vibrational data of 2c calculated at
the BLYP/6-311G(d,p) level of theory. c Number of vibration according to the calculation. d Relative intensity based on the strongest peak (100%).
e Assignment of calculated and experimental vibrations based on comparison of peak positions and intensities.
described with the BLYP and BVWN5 functionals.19 Kraka
et al. used UB3LYP and CCSD(T) to calculate the geometry
and the vibrational spectrum of m-benzyne 2a.13 The DFT
results are close to the CCSD(T) calculations, as long as an
unrestricted wave function (UB3LYP) is used, while B3LYP
calculations yield the bicyclic isomer of 2a with spectroscopic
properties in disagreement with the experiment. In a recent
computational study Cramer, Nash, and Squires demonstrated
that thermochemical data of 2a calculated at the BPW91/pVTZ
level of theory come close to experimental data.20 However, the
C(1)–C(3) distance between the radical centers of 1.874 Å cal-
culated with the BPW91 functional is significantly shorter than
the CCSD(T) value of 2.178 Å. With a C(1)–C(3) distance of
2.00 Å the RBLYP structure is closer to the CCSD(T) results
than the BPW91 structure and therefore this method was used
throughout this paper.
Fig. 2 Bottom: difference IR spectrum (in absorbance) showing the
The IR spectra calculated with the RBLYP functional are in
UV photochemistry of cyclophane 4c. The IR absorptions of the
excellent agreement with the experimental data. This method
photoproducts 5 and 2c are pointing upwards. Top: IR spectrum of 2c
thus provides a reliable and, compared to CCSD(T) calcu-
lations, cost efficient method to compute ground state proper-
ties of m-benzynes. The spectra of 2 exhibit two characteristic
vibrations: (i) the ring deformation mode at 543 cmϪ1 in 2b and
2c, at 547 cmϪ1 in 2a, and at 519 cmϪ1 in 7. This vibration is
obviously hardly affected by substitution at C(5), and even the
OH group at C(4), adjacent to one of the radical centers, results
in a red-shift of only less than 30 cmϪ1. Since this vibration is
very strong, it can be used to identify m-benzynes in product
mixtures. (ii) The second characteristic vibration is the combin-
ation of C(5)–X stretching with the symmetrical in plane bend-
ing of the C(4)–H and C(6)–H bonds. This vibration is weak
in 2a (1402 cmϪ1), strong in 2b (1440 cmϪ1), and of medium
intensity in 2c (1462 cmϪ1).
A comparison of the calculated structures of 2a, 2b and 2c
reveals the small influence of substituents at C(5) on the
ring structure (Fig. 3). Both the fluorine and the methyl group
slightly decrease the C(1)–C(3) distance and the C(1)–C(2)–
C(3) bond angle and thus increase the through bond interaction
between the radical centers. The largest effect is, as expected,
calculated at the BLYP/6-311(d,p) level of theory (in absorbance). The
peak numbers refer to Table 2.
Discussion
The UV photolysis of cyclophanes 4 provides a useful method
for the generation of matrix-isolated m-benzynes 2 in high
yields and allows study of the influence of substituents on the
properties of benzynes. One concern using this method is that 2
and the other fragments CO and p-xylylene 5 are formed in the
same matrix cage which might cause perturbations of the
spectra or even secondary thermal or photochemical reactions.
However, such secondary reactions are not observed even with
254 nm irradiation and the experimental IR spectra are in good
agreement with the calculated gas phase spectra, which excludes
large interactions of species trapped in the same matrix cage.
DFT methods were used to calculate the geometries and IR
spectroscopic data of the benzynes 2. Recently Cramer demon-
strated that—despite the single reference character of the DFT
methods—open shell systems such as carbenes are reliably
J. Chem. Soc., Perkin Trans. 2, 1999, 2285–2290
2287

View Article Online
Fig. 3 Structures of m-benzynes 2a, 2b, and 2c calculated at the
BLYP/6-311G(d,p) level of theory.
observed in the local surroundings of the substituent. Thus,
fluorine substitution leads to an increase of the C(4)–C(5)–C(6)
bond angle, while methyl substitution has the opposite effect.
The slightly negative charge located at the radical centers C(1)
and C(3) results in a calculated dipole moment of 0.79 D in
m-benzyne 2a, which is reduced to 0.48 in 2b and increased to
1.42 D in 2c. Thus, the electron withdrawing fluorine substitu-
ent compensates the negative charge at the radical centers while
the electron donating methyl group increases the overall charge
polarization.
In conclusion, we were able to generate the m-benzynes 2b
and 2c by UV photolysis of the corresponding cyclophanes 4b
and 4c, a novel general route to substituted m-benzynes. The
experimental IR spectra of 2b and 2c can be reproduced by
DFT calculations, which reveals the reliability of these methods
for the calculation of ground state properties of m-benzynes.
All solvents were distilled and dried before use. Flash column
chromatography was carried out on ICN silica 32–63 (60 Å).
NMR spectra were recorded on a Bruker DPX 200 instrument,
chemical shifts in δ relative to TMS, J in Hz, d = doublet, t =
triplet, m = multiplet. Infrared spectra were recorded on a
Perkin-Elmer 841 or Bruker IFS 66 IR Spectrometer in the
range 4000–600 cmϪ1. MS data were obtained on a Varian
MAT CH 5 instrument at 70 eV, high resolution mass data
on a VG AutoSpec instrument at 70 eV.
Experimental
Matrix isolation spectroscopy
Matrix isolation experiments were performed by standard tech-
niques with an APD DE-204SL and an APD DE-202 Displex
closed cycle helium cryostat. Matrices were produced by
deposition of argon (Linde, 99.9999%) on top of a CsI (IR) or
sapphire (UV–vis) window with a rate of approximately 0.15
mmol minϪ1. To obtain optically clear matrices the spectro-
scopic window was retained at 30 K during deposition, and the
matrix was subsequently cooled to 10 K. Matrix infrared spec-
tra were recorded by using a Bruker IFS66 FTIR spectrometer
with a standard resolution of 1 or 0.5 cmϪ1 in the range
400–4000 cmϪ1. UV–vis spectra were recorded on a Hewlett-
Packard 8452A diode array spectrophotometer with a reso-
lution of 2 nm. Irradiations were carried out with use of Osram
HBO 500 W/2 or Ushio USH-508SA mercury high-pressure arc
lamps in Oriel housings equipped with quartz optics. IR irradi-
ation from the lamps was absorbed by a 10 cm path of water.
For broad-band irradiation Schott cut-off filters were used
(50% transmission at the wavelength specified). For narrow-
band irradiation interference filters in combination with
dichroic mirrors (“cold mirrors”) or a Gräntzel low pressure
mercury arc lamp (254 nm) were used.
3,5-Bis(dibromomethyl)fluorobenzene 10a
A
mixture of 1,3-dimethyl-5-fluorobenzene 9a (10 ml,
80.7 mmol), NBS (N-bromosuccinimide) (86.13 g, 483.89
mmol) and AIBN (α,αЈ-azoisobutyronitrile) in CCl4 (645 ml)
was heated under reflux for 8 h and the succinimide filtered off.
After removal of the solvent, a yellow oil remained (30.4 g,
86%). ν/cmϪ1 3082, 3012, 1755, 1720, 1599, 1452, 1307, 1216,
1196, 1147, 993, 872, 788, 739, 700, 666, 642; δH(200 MHz,
CDCl3) 7.44 (t, J 1.63, 2 H), 7.29 (dd, J 1.63, J 8.66, 2 H), 6.56
(s,1 H); δC(50.33 MHz, CDCl3) 162.1 (d, J 236.2, CF), 144.2 (d,
J 7.8, Cq), 119.8 (d, J 2.9, CH), 115.7 (d, J 24.3, CH), 38.12
(CH); m/z 443–435 (Mϩ), 363–357 (C8H5Br4F Ϫ Br), 283–279
(C8H5Br3F Ϫ Br), 202–200 (C8H5Br2F Ϫ Br), 120 (C8H5BrF Ϫ
Br).
5-Fluoroisophthalaldehyde 11a
30.4 g of 10a (69.2 mmol) were heated in 150 ml conc. H2SO4
(96%) to 110 ЊC, and the Br2 formed during the reaction
removed by a flow of Ar. The mixture was heated for another 2
h and the reaction mixture poured on ice (1 l). The product was
extracted with TBME, neutralized and dried over MgSO4.
Removing the solvent and re-crystallization from petroleum
ether (bp 60–80 ЊC) yielded a light yellow solid (2.7 g, 17.8 mmol,
26%). Mp: 93 ЊC; ν/cmϪ1 3376, 3074, 2866, 1695, 1595, 1464,
1448, 1383, 1299, 1239, 1138, 988, 964, 893, 753, 672, 654;
δH(200 MHz, CDCl3) 10.06 (d, J 1.76, 2 H), 8.18 (t, J 1.38, 1 H),
7.82 (dd, J 17.91, 1.38, 2 H); δC(50.33 MHz, CDCl3) 189.54
(CO), 163.43 (d, J 252,7, CF), 139.03 (d, J 5.83, Cq), 126.96
(CH), 120.87 (d, J 22.35, CH); m/z 152.059 (calc. for C8H5O2F:
152.0274).
Calculations
Calculations were performed with the GAUSSIAN94 suite of
programs.21 Geometry optimizations and the calculations of
vibrational spectra were performed with a DFT method using
the RBLYP functional22,23 and the 6-311G(d,p) valence triple
zeta basis set.24 The RBLYP wave functions have been tested to
be stable with respect to becoming unrestricted.
Synthesis of cyclophanes
The [2.2]metaparacyclophanes 4 were synthesized according to
a modified procedure by Boekelheide et al.25
2288
J. Chem. Soc., Perkin Trans. 2, 1999, 2285–2290

View Article Online
5-Fluoroisophthalaldehyde bis(propane-1,3-diyl dithioacetal) 12a
δC(50.33 MHz, CDCl3) 139.98 (Cq), 139.11 (Cq), 128.55 (CH),
124.59 (CH), 51.26 (CH), 32.06 (CH2), 25.12 (CH2), 21.22
1.1 ml (4.2 mmol) of BF3ؒEt2O (45%) were added to a solution
of 11a (2.7 g, 17.8 mmol) in 45 ml of acetic acid (99–100%).
Propane-1,3-dithiol (4.5 ml, 44.4 mmol) was added dropwise
under stirring at room temperature. After stirring for another
72 h the colorless precipitate was filtered off and dried in vacuo
(3.45 g, 58.4%). Mp: 137 ЊC; ν/cmϪ1 3066, 2935, 2828, 1769,
1711, 1616, 1592, 1528, 1514, 1465, 1442, 1433, 1423, 1344,
1308, 1274, 1244, 1200, 1180, 1130, 1116, 1012, 987, 905, 884,
826, 762, 723, 695, 675, 666; δH(200 MHz, CDCl3) 7.35
(t, J 1.51, 1 H), 7.14 (dd, J 9.16, 1.63, 2 H), 5.10 (s, 2 H), 3.09–
2.82 (m, 8 H), 2.22–2.08 (m, 2 H), 2.01–1.79 (m, 2 H);
δC(50.33 MHz, CDCl3) 160.24 (d, J 247.8, CF), 141.67 (d,
J 8.75, Cq), 123.29 (CH), 115.05 (d, J 23.33, CH), 50.54 (C–H),
31.79 (CH2), 24.97 (CH2); m/z 332.0197 (calc. for C14H17FS4:
332.0197).
(CH3); m/z 328.0459 (calc. for C15H20S4: 328.0448).
5-Methyl[2.2]metaparacyclophane-2,9-dione bis(propane-1,3-
diyl dithioketal) 13b
The procedure for the synthesis of 13b is analogous to that of
13a described above. 1.75 g (5 mmol) of 12b gave 1.32 g (61%)
of 13b as colorless crystals. ν/cmϪ1 3430, 3033, 2948, 1884,
1734, 1592, 1502, 1455, 1431, 1420, 1346, 1274, 1244, 1191,
1168, 1129, 1107, 1022, 980, 951, 937, 915, 902, 888, 852, 811,
789, 759, 733, 716, 709, 680, 653; δH(200 MHz, CDCl3) 7.51 (m,
J 1.50, 2 H), 7.29 (m, J 1.26, 2 H), 6.63 (1 H), 5.85 (m, J 1.00),
3.44 and 2.74 (AB-quartet, J 13.17, 4 H), 2.94–2.33 (m, 8 H),
2.36 (s, 3 H), 1.95–1.82 (m, 4 H); δC(50.33 MHz, CDCl3) 140.68
(Cq), 137.72 (Cq), 134.50 (Cq), 129.13 (CH), 129.03 (CH),
128.57 (CH), 127.54 (CH), 60.27 (Cq), 53.55 (CH2), 28.99
(CH2), 27.44 (CH2), 24.97 (CH2), 21.84 (CH3); m/z 430.0909
(calc. for C23H26S4: 430.0917).
5-Fluoro[2.2]metaparacyclophane-2,9-dione bis(propane-1,3-diyl
dithioketal) 13a
10 ml of a solution of n-butyllithium in hexane (1.6 M) were
added under argon to a cooled (Ϫ32 ЊC) solution of 12a (1.67 g,
5 mmol) in 250 ml dry THF and stirred for 1.5 h at the same
temperature to generate the dianion. The reddish brown solu-
tion of the dianion (Ϫ32 ЊC) and a solution of α,αЈ-dibromo-p-
xylene (1.32 g, 5 mmol) in 250 ml THF (room temperature)
were simultaneously added under high dilution conditions over
a period of 6 h to refluxing THF (1000 ml). After removing the
solvent colorless crystals were obtained by using flash column
chromatography (CH2Cl2) (0.98 g, 45%). Dp: >225 ЊC; ν/cmϪ1
2926, 2910, 2340, 1735, 1671, 1598, 1579, 1499, 1432, 1422,
1411, 1345, 1281, 1262, 1239, 1194, 1171, 1117, 1102, 1015, 989,
954, 938, 924, 902, 889, 864, 855, 815, 791, 734, 719, 703, 677,
608; δH(200 MHz, CDCl3) 7.47 (dd, J 10.29, 1.76, 2 H), 7.32 (t,
J 1.0, 2 H), 6.66 (t, 1.76, 1 H), 5.97 (t, J 1.0, 2 H), 3.46 and 2.75
(AB quartet, J 13.2, 4 H), 2.90–2.29 (m, 8 H), 1.96–1.83 (m,
4 H); δC(50.33 MHz, CDCl3) 163.5 (d, 248.8, CF), 143.5 (Cq),
134.4 (Cq), 129.2 (CH), 128.8 (CH), 127.7 (CH), 114.0 (d,
J 24.3, CH), 59.7 (Cq), 53.5 (CH2), 28.3 (d, J 73.86, CH2), 24.8
(CH2); m/z 434.0664 (calc. for C22H23FS4: 434.0667).
5-Methyl[2.2]metaparacyclophane-2,9-dione 4c
The procedure for the synthesis of 4c is analogous to that of 4b
described above. 300 mg (0.7 mmol) of 13b gave 40 mg (23%)
of 4c as colorless crystals. ν/cmϪ1 3379, 2929, 1698, 1587, 1501,
1447, 1407, 1296, 1257, 1169, 1134, 1101, 1073, 1035, 994, 916,
885, 871, 817, 785, 741, 720, 605; δH(200 MHz, CDCl3) 7.03
(s, 4 H), 6.92 (m, J 0.52, 2 H), 5.44 (s, 1 H), 3.87 (s, 4 H), 2.29
(s, 3 H); δC(50.33 MHz, CDCl3) 203.21 (CO), 141.47 (Cq),
137.85 (Cq), 135.98 (CH), 132.51 (CH), 125.99 (Cq), 120.75
(CH), 52.55 (CH2), 21.29 (CH3); m/z 250.0992 (calc. for
C17H14O2: 250.0994).
Acknowledgements
Much of our work in the field of the benzynes was inspired by
intense discussions with Robert Squires. His accurate thermo-
chemical data are the basis of experimental and theoretical
investigations in many laboratories throughout the world.
This work was financially supported by the Deutsche
Forschungsgemeinschaft and the Fonds der Chemischen
Industrie.
5-Fluoro[2.2]metaparacyclophane-2,9-dione 4b
A solution of 13a (2.4 g, 5.4 mmol) in 100 ml THF was added
dropwise to a stirred slurry of 3.5 g (16.2 mmol) mercury()
oxide and 4.6 ml (8.8 mmol) of BF3ؒEt2O (45%) in 35 ml THF–
15% water at room temperature. After 2 h the addition was
complete and the mixture was stirred for 12 h at 40 ЊC. The
solution was filtered off (Celite, sand, silica gel) and dried over
MgSO4. After removal of the solvent the cyclophane was puri-
fied using thin layer chromatography (CH2Cl2) to yield colorless
crystals (150 mg, 10.9%). Mp: 88–90 ЊC; ν/cmϪ1 3591, 3531,
3362, 2961, 2931, 2858, 1687, 1616, 1584, 1502, 1451, 1411,
1293, 1260, 1168, 1100, 1048, 919, 891, 869, 814, 779, 738, 609;
δH(200 MHz, CDCl3) 7.07 (s, 4 H), 6.83 (dd, J 8.53, 1.5, 2 H),
5.43 (t, J 1.38, 1 H), 3.91 (s, 4 H); δC(50.33 MHz, CDCl3)
201.18 (CO), 161.40 (d, J 252.7, CF), 143.36 (d, J 5.84, Cq),
135.54 (Cq), 132.60 (CH), 119.37 (d, J 3.89, CH), 112.52 (d,
J 23.32, CH), 52.53 (CH2); m/z 254.0753 (calc. for C16H11O2F:
254.0743).
References
1 C. Grundmann, ‘Arene und Arine’, in Houben Weyl, Methoden der
Organischen Chemie, Vol. 5, Thieme, Stuttgart, 1981, pp. 613–648.
2 K. C. Nicolaou and W. M. Dai, Angew. Chem., 1991, 103, 1453;
Angew. Chem., Int. Ed. Engl., 1991, 30, 1387.
3 R. H. Levin, ‘Arynes’, in Reactive Intermediates, Vol. 1, ed. M. Jones
Jr. and R. A. Moss, Wiley, 1978, pp. 1–26.
4 W. Sander, Acc. Chem. Res., 1999, 32, 669.
5 S. G. Wierschke, J. J. Nash and R. R. Squires, J. Am. Chem. Soc.,
1993, 115, 11 958.
6 P. G. Wenthold, J. A. Paulino and R. R. Squires, J. Am. Chem. Soc.,
1991, 113, 7414.
7 P. G. Wenthold, J. Hu and R. R. Squires, J. Am. Chem. Soc., 1994,
116, 6961.
8 P. G. Wenthold and R. R. Squires, J. Am. Chem. Soc., 1994, 116,
6401.
9 P. G. Wenthold, J. Hu and R. R. Squires, J. Am. Chem. Soc., 1996,
118, 11 865.
10 J. J. Nash and R. R. Squires, J. Am. Chem. Soc., 1996, 118, 11 872.
11 D. Schröder, N. Goldberg, W. Zummack, H. Schwarz, J. C. Poutsma
and R. R. Squires, Int. J. Mass Spectrom. Ion Processes, 1997, 165/
166, 71.
12 P. G. Wenthold, R. R. Squires and W. C. Lineberger, J. Am. Chem.
Soc., 1998, 120, 5279.
13 E. Kraka, D. Cremer, G. Bucher, H. Wandel and W. Sander, Chem.
Phys. Lett., 1997, 268, 313.
14 K. K. Thoen and H. I. Kenttaemaa, J. Am. Chem. Soc., 1999, 121,
800.
5-Methylisophthalaldehyde bis(propane-1,3-diyl dithioacetal)
12b
The reaction of 5-methylisophthalaldehyde26 (3.2 g, 21.3 mmol)
with propane-1,3-dithiol was carried out as described for the
preparation of 12a (5.10 g, 73%). Mp: 163–165 ЊC; ν/cmϪ1 3447,
3022, 2947, 2901, 2824, 1794, 1599, 1451, 1419, 1378, 1290,
1270, 1241, 1201, 1166, 1107, 1083, 998, 973, 905, 878, 822, 757,
726, 703, 675, 657; δH(200 MHz, CDCl3) 7.32 (s, 1 H), 7.21
(s, 2 H), 5.09 (s, 2 H), 3.10–2.82 (m, 8 H), 2.22–1.79 (m, 4H);
15 R. Marquardt, W. Sander and E. Kraka, Angew. Chem., 1996, 108,
825; Angew. Chem., Int. Ed. Engl., 1996, 35, 746.
J. Chem. Soc., Perkin Trans. 2, 1999, 2285–2290
2289

View Article Online
16 G. Bucher, W. Sander, E. Kraka and D. Cremer, Angew. Chem.,
1992, 104, 1225; Angew. Chem., Int. Ed. Engl., 1992, 31, 1230.
17 W. Sander, G. Bucher, H. Wandel, E. Kraka, D. Cremer and
W. S. Sheldrick, J. Am. Chem. Soc., 1997, 119, 10 660.
18 D. C. Smith, E. E. Ferguson, R. L. Hudson and J. R. Nielsen,
J. Chem. Phys., 1953, 21, 1475.
R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker,
J. P. Stewart, M. Head-Gordon, C. Gonzalez and J. A. Pople,
GAUSSIAN94, Revision B.3, Gaussian, Inc., 1995.
22 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter,
1988, 37, 785.
23 A. Becke, J. Chem. Phys., 1993, 98, 5648.
24 R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys.,
1980, 72, 650.
19 S. E. Worthington and C. J. Cramer, J. Phys. Org. Chem., 1997, 10,
755.
20 C. J. Cramer, J. J. Nash and R. R. Squires, Chem. Phys. Lett., 1997,
277, 311.
21 M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G.
Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Petersson,
J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V. G.
Zakrzewski, J. V. Ortiz, J. B. Foresman, C. Y. Peng, P. Y. Ayala,
W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R. Gomperts,
25 V. Boekelheide, P. H. Anderson and T. A. Hylton, J. Am. Chem.
Soc., 1974, 96, 1558.
26 T.-L. Chan, T. C. W. Mak and J. Trotter, J. Chem. Soc., Perkin Trans.
2, 1980, 672.
Paper 9/05189I
2290
J. Chem. Soc., Perkin Trans. 2, 1999, 2285–2290