Vibrational spectrum of m-benzyne: A matrix isolation and computational study

Article abstract of DOI:10.1021/ja012686g

m-Benzyne (2) was generated in low-temperature matrices and IR spectroscopically characterized from four different precursors. To assign the IR absorptions, the perdeuterated derivative 2-d₄ was also investigated. By comparison with CCSD(T) cal

Full text of DOI:10.1021/ja012686g

Published on Web 10/15/2002
Vibrational Spectrum of m-Benzyne: A Matrix Isolation and
Computational Study†
Wolfram Sander,*,‡ Michael Exner,‡ Michael Winkler,‡ Andreas Balster,‡
Angelica Hjerpe,§ Elfi Kraka,§ and Dieter Cremer*,§
Contribution from Lehrstuhl fu¨r Organische Chemie II der RuhrsUniVersita¨t,
D-44780 Bochum, Germany, and Department of Theoretical Chemistry, Go¨teborg UniVersity,
Reutersgatan 2, S-41320 Go¨teborg, Sweden
Received December 10, 2001
Abstract: m-Benzyne (2) was generated in low-temperature matrices and IR spectroscopically characterized
from four different precursors. To assign the IR absorptions, the perdeuterated derivative 2-d4 was also
investigated. By comparison with CCSD(T) calculations all vibrations between 200 and 2500 cm-1 with a
predicted relative intensity >2% could be assigned. All experimental and theoretical results are in accordance
with a biradicaloid structure for 2, while there is no evidence for a bicyclic closed-shell structure. While
benzyne 2 is stable under the conditions of matrix isolation at low temperature, flash vacuum pyrolysis at
high temperatures or UV irradiation results in the rearrangement to cis-enediyne. A mechanism involving
ring opening accompanied by hydrogen migration is proposed.
Scheme 1
Introduction
The didehydrobenzenes 1-3 (benzynes) are important reac-
tive intermediates that have also found multiple applications in
synthesis.1 The heat of formations ∆Hf,298 of 1-3 were
determined by Squires et al. to be 105.1 ( 3.2, 121.9 ( 3.1,
and 137.8 ( 2.9 kcal/mol, respectively,2 the latter value being
in excellent agreement with the measurement of Roth and co-
workers.3 Among these species m-benzyne (2) is the least
investigated derivative.1a,4 The radical centers in 2 are coupled
substantially by through-space and through-bond interactions5
which lead to a singlet-triplet splitting of 21.0 ( 0.3 kcal/
mol.6 Thus, the reactivity of 2 is expected to reflect both
biradicaloid and closed shell character.
formation of p-xylylene (6), CO, and 2. Isophthaloyl diacetyl
peroxide (7), on the other hand, is a thermal precursor that on
(4) (a) Nelson, E. D.; Artau, A.; Price, J. M.; Tichi, S. E.; Jing, L.; Kentta¨maa
H. I. J. Phys. Chem. A 2001, 105, 10155. (b) Thoen, K. K.; Kentta¨maa H.
I. J. Am. Chem. Soc. 1999, 121, 800. (c) Thoen, K. K.; Kentta¨maa H. I. J.
Am. Chem. Soc. 1997, 119, 3832. For computational studies on 2 and its
derivatives, see for example: (d) Kraka, E.; Anglada, J.; Hjerpe, A.; Filatov,
M.; Cremer, D. Chem. Phys. Lett. 2001, 348, 115. (e) Winkler, M.; Sander,
W. J. Phys. Chem. A 2001, 105, 10422. (f) Johnson, W. T. G.; Cramer, C.
J. J. Am. Chem. Soc. 2001, 123, 923. (g) Cramer, C. J.; Thompsom, J. J.
Phys. Chem. A 2001, 105, 2091. (h) Johnson, W. T. G.; Cramer, C. J. J.
Phys. Org. Chem. 2001, 14, 597. (i) Hess, B. A., Jr. Eur. J. Org. Chem.
2001, 2185. (j) Hess, B. A., Jr. Chem. Phys. Lett. 2002, 352, 2185. (k)
Lindh, R.; Bernhardsson, A.; Schu¨tz, M. J. Phys. Chem. A 1999, 103, 9913.
(l) Cramer, C. J.; Nash, J. J.; Squires, R. R. Chem. Phys. Lett. 1997, 277,
311. (m) Kraka, E.; Cremer, D.; Bucher, G.; Wandel, H.; Sander, W. Chem.
Phys. Lett. 1997, 268, 313. (n) Nash, J. J.; Squires, R. R. J. Am. Chem.
Soc. 1996, 118, 11872. (o) Lindh, R.; Schu¨tz, M. Chem. Phys. Lett. 1996,
258, 409. (p) Lindh, R.; Lee, T. J.; Bernhardsson, A.; Persson, B. J.;
Karlstro¨m, G. J. Am. Chem. Soc. 1995, 117, 7186. (q) Lindh, R.; Persson,
B. J. J. Am. Chem. Soc. 1994, 116, 4963. (r) Kraka, E.; Cremer, D. J. Am.
Chem. Soc. 1994, 116, 4929. (s) Kraka, E.; Cremer, D. Chem. Phys. Lett.
1993, 216, 333. (t) Wierschke, W. S.; Nash, J. J.; Squires, R. R. J. Am.
Chem. Soc. 1993, 115, 11958. (u) Nicolaides, A.; Borden, W. T. J. Am.
Chem. Soc. 1993, 115, 11951. (v) Debbert, S. L.; Cramer, C. J. Int. J.
Mass Spectrom. 2000, 201, 1. (w) Nash, J. J.; Squires, R. R. J. Am. Chem.
Soc. 1996, 118, 11872.
During the past years we developed several precursors of 2
and derivatives that are suitable for generating m-benzynes in
low-temperature matrices (Scheme 1).7 UV photolysis of matrix-
isolated [2.2]metaparacyclophanediketone (5) leads to the
* To whom correspondence should be addressed.
† Dedicated to Prof. Waldemar Adam on the occasion of his 65th
birthday.
‡ RuhrsUniversita¨t.
§ Go¨teborg University.
(1) (a) Sander, W. Acc. Chem. Res. 1999, 32, 669. (b) Kauffman, T.; Wirthwein,
R. Angew. Chem., Int. Ed. Engl. 1971, 10, 20. (c) Hoffman, R. W.
Dehydrobenzene and Cycloalkynes; Academic Press: New York, 1967.
(2) (a) Wenthold, P. G.; Squires, R. R. J. Am. Chem. Soc. 1994, 116, 6401.
(b) Wenthold, P. G.; Hu, J.; Squires, R. R. J. Am. Chem. Soc. 1994, 116,
6961. (c) Wenthold, P. G.; Paulino, J. A.; Squires, R. R. J. Am. Chem.
Soc. 1991, 113, 7414.
(5) Hoffmann, R.; Imamura, A.; Hehre, W. J. J. Am. Chem. Soc. 1968, 90,
1499.
(6) Wenthold, P. G.; Squires, R. R.; Lineberger, W. C. J. Am. Chem. Soc.
1998, 120, 5279.
(3) Roth, W. R.; Hopf, H.; Horn, C. Chem. Ber. 1994, 127, 1765.
9
13072
J. AM. CHEM. SOC. 2002, 124, 13072-13079
10.1021/ja012686g CCC: $22.00 © 2002 American Chemical Society

Vibrational Spectrum of m-Benzyne
A R T I C L E S
flash vacuum pyrolysis (FVP) at 300 °C yields methyl radicals,
CO2, and again 2.7d
The structure of 2smonocyclic biradical 2a vs anti-Bredt
olefin 2bshas not yet been discussed in depth on the basis of
experimental data.8 On the basis of B3LYP calculations, Hess
recently argued that our IR spectrum published in 1996 might
well be interpreted in favor of a bicyclic structure (bicyclo-
[3.1.0]hexatriene (2b)).4i,j,9 It should be emphasized, however,
that Hess assigned several absorptions in the experimental
spectrum leading to a pattern similar to that calculated at the
B3LYP level for 2b. Here we will show that this assignment is
not backed by experimental evidence and that the theoretical
level used by this author is not suitable for describing the
biradicaloid structure 2a.9
Figure 1. IR spectra showing the photochemical formation and decay of
m-benzyne (2). (a) Difference spectrum showing the photochemistry of
cyclophane (5) on 335 nm irradiation. Bands pointing upward are assigned
to 2 and p-xylylene (6). (b) Difference spectrum showing the photochemistry
of the same matrix after photolysis at 305 nm. (c) IR spectrum of an
independently sythesized sample of cis-4. (d) BLYP/6-311G(d,p) calculated
spectrum of cis-4. (e) Vibrational spectrum of 2 calculated at the CCSD-
(T)/6-311G(2d,2p) level of theory.
Another interesting point is the thermal or photochemical
rearrangement of 2 to give cis-enediyne 4.10 This rearrangement
requires a change of the CH connectivity and thus involves the
migration of a hydrogen atom. The hydrogen migration in
benzynes 1 and 2 to give 3 has recently been postulated by
Zewail et al.11 The three isomeric benzynes 1-3 were generated
by femtosecond photolysis of 1,2-, 1,3-, and 1,4-dibromoben-
zene, respectively. Femtosecond time-resolved mass spectrom-
etry revealed that both CBr bonds are broken within less than
100 fs. The benzyne molecules then decay with a lifetime of
around 400 ps. From the similar lifetimes of all three benzynes
it was concluded that 1 and 2 rearrange to 3, which subsequently
ring opens to 4.11 In this paper we show that 2 indeed rearranges
to 4, both thermally and photochemically; however, for this
rearrangement the hydrogen migration in 2 to give 3 as the
primary precursor of 4 is not necessary.
with subsequent trapping of the products in low-temperature
matrices with a large excess of an inert gas should produce 2
and the byproducts in separate matrix cages and thus minimize
unwanted intermolecular interactions.
Broad-band irradiation of 5, matrix isolated in argon at 10
K, for several hours with λ > 335 nm (Hg high-pressure arc
lamp in combination with a cutoff filter) results in the decrease
of the IR absorptions of 5 and formation of 2, CO, and 6, as
has been described before (Figure 1).7d If the monochromatic
light of a low-pressure mercury arc lamp (λ ) 254 nm) is used
to irradiate 5, additional absorptions at 1833, 1127, 693, and
677 cm-1 were observed which reach a maximum intensity after
several hours of irradiation. These absorptions disappear on
broad-band irradiation with λ > 305 nm. A strong IR absorption
around 1830 cm-1 is characteristic of acyl radicals,12 and it is
thus tempting to assign the photolabile species the structure of
biradical 10. The 305 nm photolysis of mixtures of 5 and 10
yields the same products as the direct 305 nm photolysis of 5;
however, since the concentration of 10 (estimated from the IR
intensities) is always low compared to that of 5, it is not possible
to directly follow the photochemistry of 10 (Scheme 2).
Alternative intermediates that could explain the 1833 cm-1
absorption are the acylradicals 11 and 12. Although these
structures cannot be rigorously excluded, we expect them to be
less stable than 10.
Results and Discussion
Matrix Isolation of m-Benzyne (2). Four different precursors
of 2 were investigated in our laboratory: cyclophane 5 as a
photochemical precursor and peroxide 7, 1,3-diiodobenzene (8),
and 1,3-dinitrobenzene (9) as thermal precursors. The photo-
chemical precursor yields 2 by UV irradiation of 4 in solid argon
or neon at temperatures between 3 and 12 K. Byproducts of
this photolysis are CO and 6, which are formed in the same
matrix cage. Since this might result in secondary thermal
reactions and in shifts of the IR absorptions, independent thermal
sources of 2 were developed. FVP of the thermal precursors
(7) (a) Wenk, H. H.; Sander, W. Chem. Eur. J. 2001, 7, 1837. (b) Sander, W.;
Exner, M. J. Chem. Soc., Perkin Trans. 2 1999, 2285. (c) Sander, W.;
Bucher, G.; Wandel, H.; Kraka, E.; Cremer, D.; Sheldrick, W. S. J. Am.
Chem. Soc. 1997, 119, 10660. (d) Marquardt, R.; Sander, W.; Kraka E.
Angew. Chem., Int. Ed. Engl. 1996, 35, 746. (e) Bucher, G.; Sander, W.;
Kraka, E.; Cremer, D. Angew. Chem., Int. Ed. Engl. 1992, 35, 746.
(8) Early trapping experiments have been interpreted in favor of a bicyclic
structure bicyclo[3.1.0]hexatriene instead of a monocyclic biradical: (a)
Washburn, W. N.; Zahler, R. J. Am. Chem. Soc. 1976, 98, 7827. (b)
Washburn, W. N. J. Am. Chem. Soc. 1975, 97, 1615.
(9) For a detailed computational treatment of this point, see refs 4d,e.
(10) Sander, W.; Marquardt, R.; Bucher, G.; Wandel, H. Pure Appl. Chem. 1996,
68, 353.
(11) Diau, E. W. G.; Casanova, J.; Roberts, J. D.; Zewail, A. H. Proc. Natl.
Acad. Sci. U.S.A. 2000, 97, 1376.
The absorptions of 2 and 6 both disappear on prolonged
irradiation with λ > 305 nm, which allows one to clearly assign
(12) See, for example: Neville, A. G.; Brown, C. E.; Rayner, D. M.; Lusztyk,
J.; Ingold, K. U. J. Am. Chem. Soc. 1991, 113, 1869.
9
J. AM. CHEM. SOC. VOL. 124, NO. 44, 2002 13073

A R T I C L E S
Sander et al.
Scheme 2
even weak IR absorptions of 2 (Figure 1). At least two products
are formed; however, cis-4 is the only product that could be
unequivocally identified by comparison of the IR spectra with
that of an independently synthesized sample. Since 6 in the
absence of 2 is photochemically completely stable under the
same irradiation conditions, the other photoproduct most likely
is a reaction product between 2 and 6. Due to its low yield, this
product could not be identified. This result clearly demonstrates
that 2 photochemically rearranges to 4 (Figure 1). The mech-
anism of this rearrangement is discussed below.
Figure 2. (a) IR spectrum of an argon matrix containing the FVP products
of 8. Bands denoted by circles belong to 2; bands labeled by crosses are
assigned to 13. (b) IR spectrum of an independently sythesized sample of
cis-4. (c) IR spectrum of an independently sythesized sample of trans-4.
(d) IR spectrum of 8.
changes in the IR spectrum, which indicates that molecular
iodine is formed primarily in the gas phase or, more likely,
during the deposition process of the matrix by recombination
of iodine atoms (dark brown matrix).
Aromatic diiodo compounds have been used both as thermal
and photochemical precursors of biradicals.13 Recently it was
demonstrated that photolysis of 1,3-diiodo-2,4,5,6-tetrafluo-
robenzene, matrix isolated in neon at 3 K, produces 1,3-dide-
hydro-2,4,5,6-tetrafluorobenzene (the perfluorinated m-benzyne)
in reasonable yields via the corresponding monoradical.7a
Interestingly, irradiation in argon at 10 K did not result in the
loss of iodine atoms in amounts high enough to be detected by
IR spectroscopy (traces of the radical could be observed by the
more sensitive and selective EPR spectroscopy).
Photolysis of 8 in neon at 3 K produced a new set of bands
that was assigned to the 3-iodo phenylradical 13 by comparison
with the spectrum simulated at the UB3LYP/6-311G(d,p) level
of theory. Prolonged irradiation at λ ) 254 nm gave small
amounts of 2. In the IR spectrum the most intense absorption
of 2 at 547 cm-1 was detected. Obviously, the loss of the second
iodine atom in 13 is much less efficient than in the correspond-
ing fluorinated radical. The influence of the matrix (neon vs
argon), temperature, and fluorination on the product yields is
not yet understood.
The FVP of 8 was investigated first by Fischer and Lossing
in 1963 using mass spectrometry.13 High temperatures (960 °C)
were employed in these experiments, and only the enediynes
cis-4 and trans-4 were detected in the mass spectra. Therefore,
we reinvestigated the FVP of 8 carefully at various temperatures
by trapping of the products of the FVP in solid argon at 10 K.
The decomposition of 8 starts at 500 °C, producing small
amounts of cis-4 and trans-4 as well as 13 and 2. At 600 °C
the decomposition of 8 is still not complete, but substantial
amounts of all four products are detected. At temperatures above
650 °C decomposition of 8 is almost complete, but the yield of
2 is low, and apart from the enediynes additional fragmentation
products were obtained. Figure 2 shows the spectrum of an argon
matrix containing the products of the FVP of 8 at 600 °C.
Annealing of the matrix up to 38 K does not result in significant
Peroxide 7, isolated in argon at 10 K, produces small amounts
of 2 upon photolysis (λ ) 254 nm), but a very complex product
mixture is formed by in cage recombination reactions. FVP at
300 °C, however, allows the isolation of larger quantities of 2.
Other products that can be identified are CO2 and methyl
radicals. No enediynes 4 are formed under these conditions.
Annealing the matrix at 35 K for several minutes causes all
absorptions of 2 and CH3 to decrease in intensity, which
indicates reactions between the benzyne and CH3 radicals as
soon as the matrix is soft enough to allow diffusion of the
trapped species. The disappearance of the IR bands of 2 during
this annealing process again allows the assignment of even weak
absorptions (see below).
Aromatic dinitro compounds have been reported to be suitable
thermal precursors of biradicals.14 FVP of 9 at 600 °C with
subsequent trapping of the products in argon at 10 K produces
2 in low yields. All absorptions in the region 400-1600 cm-1
assigned to 2 by using the other precursors could also be
detected. Only very small amounts of the enediynes are formed
from this precursor, which may be due to the fact that the C-N
bond is much stronger than the C-I bond, thus leaving less
excess energy for fragmentation of 2 formed from 9 compared
to 8.
Ab Initio and DFT Calculations. The geometry and the
vibrational spectrum of 2 were calculated at the CCSD(T) level
of theory15 using an augmented valence triple-ú basis of the
6-311G(2d,2p) type.16 Use of DFT turns out to be rather
problematic in the case of 2: Depending on the functional,
(14) (a) Gru¨tzmacher, H.-F.; Lohmann, J. Liebigs Ann. Chem. 1970, 733, 88.
(b) Gru¨tzmacher, H.-F.; Lohmann, J. Liebigs Ann. Chem. 1975, 2023.
(15) Raghavachari, K.; Trucks, G. W.; Pople, J. A.; Head-Gordon, M. Chem.
Phys. Lett. 1989, 157, 479.
(13) Fisher, I. P.; Lossing, F. P. J. Am. Chem. Soc. 1963, 85, 1018.
9
13074 J. AM. CHEM. SOC. VOL. 124, NO. 44, 2002

Vibrational Spectrum of m-Benzyne
A R T I C L E S
Table 1. IR Spectroscopic Data of m-Benzyne (2)
argon, 10 K
Ib
BLYP 6-311++G(3df,3pd)
CCSD(T) 6-311G(2d,2p)
-1
-1
-1
viba
ν˜/cm
ν˜i/ν˜c
ν˜/cm
Ib
ν˜i/ν˜c
ν˜/cm
Ib
ν˜i/ν˜c
sym
assignmtd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
362
367
358
364
531
535
545
748
819
818
824
6
15
0
100
10
35
17
0
2
0
22
1
0
0
1
0
0
8
8
0
26
6
0.86
0.98
0.95
0.97
0.86
0.81
0.79
0.80
0.97
0.82
0.82
0.79
0.74
0.91
0.78
0.86
0.98
0.93
0.97
0.98
0.74
0.74
0.74
0.75
370
405
502
535
541
748
827
831
852
5
6
0
10 0
7
49
10
0
1
0
27
1
0
0
1
0
0
12
16
2
18
7
0.87
0.98
0.96
0.97
0.88
0.80
0.79
0.79
0.97
0.81
0.82
0.80
0.75
0.88
0.77
0.89
0.97
0.93
0.96
0.98
0.74
0.74
0.74
0.74
B1
A1
A2
B2
B1
B1
B1
A2
A1
B1
B2
A1
A1
B2
B2
B2
B2
A1
B2
A1
A1
A1
B2
A1
oop ring deformn, sym (boat)
ip ring deformn, sym
547
561
751
824
100
10
45
0.97
0.87
0.81
0.79
ip ring deformn, asym
oop ring deformn, asym (boat)
C-H wag, 4 H, in phase
C-H wag, 4 H, oop
20
ip ring deformn, sym (breath)
938
914
933
944
936
25
0.83
C-H bend, 2 H, in phase + C-C str
1029
1075
1091
1207
1271
1364
1381
1468
1680
3064
3107
3112
3124
1034
1094
1122
1261
1290
1385
1425
1504
1666
3152
3192
3197
3232
1402
1486
15
15
0.93
0.98
C-H bend, 2 H, in phase + C-C str. sym
C-H bend + C-C str
C-H str, sym
C-H str, asym
C-H str + C-C str, sym
C-H str, sym
11
4
6
2
a Number of the vibrational mode based on the CCSD(T) calculation. b Relative intensity based on the most intense absorption (100%). c Ratio of the
frequencies of 2 vs 2-d4. d The assignment of experimental and calculated IR bands is based on comparison of band positions and relative intensities and is
only tentative. oop ) out of plane; ip ) in plane.
strongly differing structures are obtained (the distances of the
radical centers are as follows: CCSD(T)/6-311G(2d,2p), 2.101
Å; RBLYP/6-311++G(3df,3pd), 1.991 Å; RB3LYP/6-311++G-
(3df,3pd), 1.606 Å).9 The surprisingly good performance of
BLYP17a,b in contrast to B3LYP17c to reproduce the structure
and IR spectrum of m-benzyne9 (and several derivatives
thereof)7b has recently been explained by the self-interaction
error of the B exchange functional, which mimics long-range
(nondynamic) electron correlation, thus leading to fortuitous
error cancellations for m-benzynes for the price of describing 2
as a bicyclic structure without any biradical character and an
Figure 3. Structure of m-benzyne calculated at the CCSD(T)/6-311G-
unusually long C1C3 bond.18 The biradical character of 2 is
(2d,2p) level of theory.
not very large (19-32%), however,4e,r which has been explained
by the strong σ-allylic interaction with the geminal C2H7-σ*
3) rather than bicyclus 2b contrary to claims made in the
literature.
antibonding orbital.4e,5
Assignment of IR Vibrations and Implications on the
Structure of 2. The assignment of IR bands to 2 in the
experimental spectra is based on the following arguments: (i)
The same set of IR absorptions is formed from four independent
precursors 5 and 7-9 under different experimental conditions
(matrix photolysis or FVP with subsequent matrix isolation).
For each signal there is at least one spectrum in which the
intensity of the respective absorption of 2a is not obscured by
partial overlap with other nearby signals of the precursor or
byproducts. The highest yield of 2 was obtained by UV
photolysis of 5. (ii) The availability of four different precursors
also allowed determination of the relative intensities of the IR
absorptions with a reasonable accuracy. (iii) Photolysis of 2
produces 4; thus all vibrations assigned to 2 decrease on
subsequent photolysis. (iv) Annealing of a matrix containing 2
produced by FVP of 7 results in a decrease of all absorptions
assigned to 2 and of the methyl radical, which is formed as a
byproduct. This is presumably caused by the reaction of 2 with
methyl radicals. (v) The assignment is further confirmed by
On the basis of the B3LYP results, it has been argued that 2
might exist in the form of the bicyclic structure 2b rather than
the biradical 2a.4i,j However, reliable CCSD(T), CAS(8,8)-
CISD+Q, and CASPT2 calculations refute this claim: The
bicyclic structure does not exist; it is just a transient form of
higher energy not occupying a stationary point on the potential
energy surface.9 Kraka and co-workers showed that apart from
the quantum chemical evidence there is also chemical evidence
that speaks against the existence of 2b.4d In the current work,
we now present spectroscopic evidence that clearly confirms
that m-benzyne possesses the structure of biradical 2a (Figure
(16) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980,
72, 650.
(17) (a) Becke, A. D. Phys. ReV. A 1988, 38, 3098. (b) Lee, C.; Yang, W.; Parr,
R. G. Phys. ReV. B 1988, 37, 785. (c) Becke, A. D. J. Chem. Phys. 1993,
98, 5648.
(18) (a) Cremer, D. Mol. Phys. 2001, 23, 1899. (b) Polo, V.; Kraka, E.; Cremer,
D. Theor. Chem. Acc. 2002, 107, 291. (c) Polo, V.; Gra¨fenstein, J.; Kraka,
E.; Cremer, D. Chem. Phys. Lett. 2002, 352, 469. (d) Cremer, D.; Filatov,
M.; Polo, V.; Kraka, E.; Cremer, D. Int. J. Mol. Sci. 2002, 3, 604. (e)
Polo, V.; Kraka, E.; Cremer, D. Mol. Phys. 2002, 11, 1771.
9
J. AM. CHEM. SOC. VOL. 124, NO. 44, 2002 13075

A R T I C L E S
Sander et al.
Figure 4. IR spectra of 2 calculated at the (a) B3LYP/6-311++G(3df,3pd),
(b) BLYP/6-311++G(3df,3pd), and (c) CCSD(T)/6-311G(2d,2p) level of
theory. (d) The experimental IR spectrum of 2 is shown schematically for
comparison.
Figure 5. IR spectra of m-benzyne prepared from four different precursors.
(a) Difference spectrum of 5 before and after 335 nm irradiation. Bands
that form during photolysis are pointing upward. (b) IR spectrum of the
FVP products of 8. (c) Difference spectrum of the FVP products of 7 before
and after annealing the matrix to 40 K. Bands pointing upward decrease in
intensity upon annealing. (d) IR spectrum of the FVP products of 9.
comparison of the experimental spectroscopic data with results
from calculations. The frequencies, relative band intensities, and
isotopic shifts on deuteriation (2-d4) extracted from the experi-
ments were compared to the data calculated at various levels
of theory.
in the region 1200-1600 cm-1 (Figures 4 and 5). For the
bicyclic form four intense bands are predicted, while for the
monocyclic biradical just two absorptions are calculated. Only
two absorptions are found in the experimental spectrum of 2
(Table 1), again in accordance with the assignment of the
biradicaloid structure 2a rather than the bicyclic structure 2b
to our experimental data.
In agreement with all available facts, matrix-isolated 2 has
the structure of biradical 2a, while there is no evidence for the
bicyclic structure 2b. This is also in agreement with the
conclusion from reliable high-level calculations (in contrast to
B3LYP calculations) that only one minimum energy structure
exists for m-benzyne with the equilibrium distance between the
radical centers around 2.05 Å.9
Mechanism of the Rearrangement of Benzyne 2 to
Enediyne 4. The pyrolysis of 2 to 4 can involve a variety of
rearrangement and bond breaking processes. We considered five
different reaction paths that are shown in Figure 6 (reaction
and activation enthalpies in Figure 6 are all given relative to
the enthalpy of 2). The benzvalene mechanism (I) leads via an
intermediate 14 possessing the benzvalene structure to biradical
3 which can open by a retro-Bergman cyclization to 4. The latter
process requires an activation energy of just 22.3 (activation
enthalpy at 298 K: 20.2) kcal/mol,19 which at 600 °C can be
considered to be a relatively small barrier, and therefore it is
not decisive for the feasibility of reaction path I. However, 14
does not represent a stationary point and rearranges via a capped
annulene structure (C3 on top of a cyclopentadienylidene
structure) to vinylidene 20. The latter compound, however, can
be formed from 2 in a more direct, less energy consuming
manner (V in Figure 6; see below). Hence, path I can be
excluded as a likely reaction path.
The frequencies and relative intensities of the IR absorptions
that are unambiguously assigned to 2 are given in Table 1. This
clearly reveals that the additional two peaks in the region 1200-
1600 cm-1 of our original experimental spectrum7d that were
assigned to 2 by Hess4i do not belong to 2. The only bands that
were not observed in our experiments are those predicted by
the CCSD(T) calculation to be of zero or very low intensity
(e2% of the strongest peak at 547 cm-1).
In Figure 4, the B3LYP/6-311++G(3df,3pd) IR spectrum
of 2b is compared with the CCSD(T)/6-311G(2d,2p) and BLYP/
6-311++G(3df,3pd) IR spectra of 2a. If m-benzyne would adopt
the bicyclic structure 2b, it should possess the second strongest
IR band at 344 cm-1 (Figure 4a) which involves 1,3 stretching
and bending of the ring framework. However, the measured IR
spectrum clearly reveals that there are just two IR bands below
500 cm-1 (362 and 367 cm-1, Table 1, Figure 1), which due to
their very low intensity can only tentatively be assigned to
m-benzyne. This is in accordance with the prediction for the
biradicaloid structure 2a and disagrees with a bicyclic structure
2b.
A characteristic feature of the IR spectrum of 2 is the strong
absorption at 547 cm-1, which exhibits only a small isotopic
shift on deuteration (Table 1). CCSD(T)/6-311G(2d,2p) (535
cm-1), CCSD(T)/6-31G(d,p) (545 cm-1), and BLYP/6-311++G-
(3df,3pd) (535 cm-1) suggest this band to be due to the b2-
symmetrical in-plane ring-deformation mode no. 4. The weak
absorption at 561 cm-1 can now be assigned to vibration no. 5
of 2 (Table 1, Figures 1 and 2). Signals at 751 (no. 6), 824 (no.
7), 936 (no. 11), 1402 (no. 18), and 1486 cm-1 (no. 19) are
observed in the spectra of 2 produced from all four precursors
(Figure 5). Band positions, relative intensities, and isotopic shifts
nicely agree with the calculated data (Table 1). Other low-
intensity peaks in the region between 700 and 1600 cm-1 are
due to noise or byproducts and definitely do not belong to 2.
Also possible is a 1,2-carbon shift mechanism (II) which
corresponds to an intramolecular retro-carbene rearrangement
(see IIa) and leads to carbene 16. Since carbene rearrangements
are accompanied by the conversion of a single to a double bond,
they are normally strongly exothermic and possess a low
reaction barrier.20 However, singlet carbene 16 is also a biradical
A distinguishing feature between the IR spectra of 2a and
2b that was proposed by Hess4i is the number of intense bands
(19) Gra¨fenstein, J.; Hjerpe, A. M.; Kraka, E.; Cremer, D. J. Phys. Chem. A
2000, 104, 1748. See also ref 3.
9
13076 J. AM. CHEM. SOC. VOL. 124, NO. 44, 2002

Vibrational Spectrum of m-Benzyne
A R T I C L E S
to biradical 18 (activation enthalpy: 90.7 kcal/mol) and the
rearrangement from 18 to 15 (activation enthalpy: 96.2 kcal/
mol) require too much energy to make path IIIb likely.
Mechanism IV involves a breaking of either C1C2 or C2C3,
yielding vinylidene 19, which by a [1,2]-H shift can rearrange
to 4. The investigation of IV revealed that the breaking of the
C-C bond is already accompanied by a H shift so that 4 is
formed in a one step mechanism requiring an activation energy
of 51.9 kcal/mol (activation enthalpy: 48.5 kcal/mol). All
attempts to find the all-cis vinylidene 19a failed. Only the cis-
trans form of 19b could be located, which turned out to be 2
kcal/mol lower in energy (49.9 kcal/mol) than the TS of reaction
path IV. Vinylidene 19b rearranges with a small barrier (0.1
kcal/mol) to 4, which after the inclusion of zero-point energy
and thermal corrections vanishes; i.e., neither 19a nor 19b exist
according to theory.
Reaction path V may be considered to start from a bicyclic
form 2b. Although it does not exist, it may be passed as a
transient point during C1C2 bond cleavage accompanied by
[1,2]-H shift. The product of this reaction, which possesses a
reaction enthalpy of 17.0 kcal/mol and requires a relatively low
activation enthalpy of 59.6 kcal/mol, is vinylidene 20. Cleavage
of the (formal) single bond C1C3 in 20 followed by a [1,5]-H
shift can lead to 4.
In summary, the most likely mechanism for the pyrolysis of
2 is reaction IV requiring an activation enthalpy of just 48.5
kcal/mol (B3LYP/6-31G(d,p)). Paths V (activation enthalpy:
59.6 kcal/mol), IIb (activation enthalpy: 71.9 kcal/mol), and
IIIa (activation enthalpy: 64.6 kcal/mol) cannot be fully
excluded but will not play an important role. All other paths
considered in this work (Figure 6) are unlikely even at elevated
temperatures.
Figure 6. Reaction mechanism for the pyrolysis of 2. Numbers in normal
print represent B3LYP/6-31G(d,p) enthalpies at 298 K given in kilocalories
per mole relative to the enthalpy of 2 where enthalpies connected with a
reaction arrow correspond to activation enthalpies while enthalpies connected
with a structure correspond to reaction enthalpies. Structures in parentheses
were not found to occupy a stationary point. Small numbers indicate the
numbering of the C atoms to follow rearrangements of the molecular
framework.
and therefore represents a high-energy form (reaction enthalpy
2 f 16: 119.8 kcal/mol), implying a reaction barrier of at least
120 kcal/mol.21 Accordingly, reaction IIa will play no role even
at 600 °C. However, there is the possibility that the formation
of the highly unstable carbene 16 is avoided in mechanism IIb
and biradicals such as 15 are generated. The relative energy of
15 is 44.8 kcal/mol (relative enthalpy: 43.7 kcal/mol) at the
BS-UB3LYP/6-31G(d,p) level of theory. Ring opening of 15
to yield 4 was found to require 50.7 kcal/mol (activation
enthalpy; 48.3 kcal/mol). However, the rearrangement from 2
to 15 is characterized by an activation energy of 74.9 kcal/mol
(activation enthalpy: 71.9 kcal/mol); i.e., path IIb is not very
likely even under pyrolysis conditions.
Reaction paths IIIa and IIIb belong to the [1,2]-hydrogen shift
mechanism. Transfer of a H atom should lead to either 1 or 3,
where in the latter case again 4 can be formed via retro-Bergman
cyclization. The [1,2]-H shift TS leading from 2 to 3 (reaction
path IIIa) has a relative energy of 68.3 kcal/mol (activation
enthalpy: 64.6 kcal/mol). Reaction IIIb, however, leads to an
unusual carbene (17) that according to the calculated geometry
possesses besides biradical also allene character thus increasing
its ring strain. Reaction enthalpy and activation enthalpy are
62.4 and 73.6 kcal/mol, respectively. Subsequent rearrangement
Conclusions
m-Benzyne (2) was matrix isolated and spectroscopically
characterized from four different precursors. A major problem
for the generation of 2 in the gas phase is the rapid rearrange-
ment to enediyne 4. Only precursors that decompose at
comparatively low temperatures allow production of 2 in
quantities large enough for IR spectroscopic detection. The
mechanism for this rearrangement, which requires ring opening
accompanied by hydrogen migration, has been investigated by
theoretical methods. The most reasonable reaction path starts
with the ring opening of 2 to give vinylidene 19. This vinylidene
is highly labile and with a very low (or even absent) barrier
rearranges to 4. The rearrangement of 2 to 4 is also induced by
UV irradiation; however, our experiments do not allow any
distinction between a photochemical or a hot ground-state
reaction.
The structure of 2, biradicaloid 2a vs bicyclic 2b, was
investigated both by spectroscopic means and by using high
level ab initio calculations. The spectroscopic evidence collected
in this work unequivocally confirms that 2 adopts the structure
of biradical 2a. This is in agreement with our first report and
rejects all speculations that m-benzyne adopts the bicyclic
structure 2b. According to high-level calculations 2b is not even
a minimum on the C6H4 potential energy surface. However,
there are considerable interactions between the formally unpaired
electrons, and the energy required for reducing the distance
between the radical centers from approximately 2 Å in the
(20) Nickon, A. Acc. Chem. Res. 1993, 26, 84.
(21) The B3LYP activation enthalpy is 108.8 kcal/mol. However, due to the
fact that the chemical processes of bond cleavage/bond formation in reaction
2 f 16 involve an odd number of electrons, the DFT description of the
transition state suffers from a large self-interaction error, leading to a
significant underestimation of the rection barrier.
9
J. AM. CHEM. SOC. VOL. 124, NO. 44, 2002 13077

A R T I C L E S
Sander et al.
All solvents were distilled and dried before use. Column chroma-
tography was carried out on ICN silica 32-63 (60 Å), preparative thin-
layer chromatography on Merck precoated PLC plates (silica gel 60,
equilibrium structure to 1.6 Å in the bicyclic structure is only
a few kilocalories per mole.
F254, 2 mm). NMR spectra were recorded on a Bruker DPX 200 or on
Experimental Section
a Bruker DRX 400; chemical shifts are given relative to TMS: d )
doublet, t ) triplet, and m ) multiplet. Infrared spectra were recorded
on a Perkin-Elmer 841 IR spectrometer in the range of 600-4000 cm-1
or FT-IR Bruker Equinox 55 and FT-IR Bruker IFS 66 in the range of
250-4000 cm-1. MS data were obtained on a Varian MAT CH 5 and
for high resolution on a VG AutoSpec at 70 e.V.
Matrix Isolation Spectroscopy. Matrix isolation experiments were
performed by standard techniques with an APD DE-204SL and an APD
DE-202 Displex closed cycle helium cryostat. Matrices were produced
by co-deposition of the compound with a large excess of argon (Messer
Griesheim, 99.9999%) on top of a cold CsI window with a rate of
approximately 0.15 mmol/min. To obtain optically clear matrices, the
spectroscopic window was retained at 30 K during deposition and
subsequently cooled to 10 K. FVP experiments were carried out without
additional heating of the spectroscopic window (temperature around
15 K).
3,5-Bis(dibromomethyl)tetradeuteriobenzene. A mixture of tetra-
deutero-m-xylene (5.5 g, 50.0 mmol), NBS (N-bromosuccinimide) (53.4
g, 300.0 mmol), and AIBN (R,R′-azoisobutyronitrile) (0.5 g, 3.0 mmol)
in CCl4 (645 mL) was heated under reflux for 12 h and the succinimide
filtered off. After removal of the solvent, the product crystallized in
Matrix infrared spectra were recorded by using a Bruker Equinox
55 FTIR spectrometer with a standard resolution of 0.5 cm-1 using a
N2(l) cooled MCT detector in the range of 500-4000 cm-1 and a Bruker
IFS 66 FTIR spectrometer with the same resolution using a DTGS (FIR)
detector in the range 250-500 cm-1. 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 and
dichroic mirrors (280-400 nm). IR irradiation from the lamps was
absorbed by a 10-cm path of water. For broad-band irradiation Schott
cutoff filters were used (50% transmission at the wavelength specified).
For narrow-band irradiation a Gra¨ntzel low-pressure mercury arc lamp
(254 nm) was used.
yellowish needles (17.5 g, 41.1 mmol, 86%): mp 93-94 °C; IR (cm-1
)
3005, 2365, 2336, 2285, 1721, 1428, 1371, 1224, 1213, 1195, 1146,
1054, 988, 976, 934, 842, 824, 808, 777, 716, 655, 610; 1H NMR (200
MHz, CDCl3) δH 6.63; 13C NMR (50.33 MHz, CDCl3) δC 142.14,
129.01, 39.55; MS (EI, 70 eV) (m/z, %) 429 (2.3) [M+], 427 (8.8),
425 (14.4), 423 (10.2), 421 (2.8), 349 (29.4), 347 (91.8), 345 (100),
343 (37.9), 268 (27.5), 266 (56.6), 265 (30.5), 187 (9.8), 186 (14.5),
185 (13.4), 133 (28.2), 106 (65.5), 105 (40.4), 78 (20.7), 77 (21.8), 53
(53.1), 52 (26.5), 51 (26.1). Calcd: 425.738 800. Obsd: 425.741 309.
2,4,5,6-Tetradeuterioisophthalaldehyde. A 20.6 g amount of 3,5-
bis(dibromomethyl)tetradeuteirobenzene (48.0 mmol) was added to 150
mL of H2SO4 (96%) at 110 °C. The bromine formed during the reaction
was removed by a flow of argon. The reaction mixture was poured on
crushed ice (1000 mL). The brownish solid precipitate was filtered off
and dissolved in TBME (150 mL). The liquid layer was extracted with
TBME (5 × 150 mL), neutralized, dried over MgSO4, and evaporated.
The residue was purified chromatographically (silica gel; CH2Cl2) to
furnish 3.1 g (22.3 mmol, 46.4%) of a light yellow solid: mp 86-87
°C; IR (cm-1) 3362, 2864, 1699, 1573, 1419, 1404, 1372, 1357, 1312,
1239, 1218, 1040, 1000, 935, 839, 777, 719, 650, 633; 1H NMR (200
MHz, CDCl3) δH 10.09; 13C NMR (50.33 MHz, CDCl3) δC 190.97,
136.83, 129.65; MS (EI, 70 eV) (m/z, %) 138 (76.7) [M+], 137 (100),
136 (33.4), 109 (50.1), 108 (23.3), 81 (35.5), 80 (23.1), 78 (11.0), 54
(18.6), 53 (31.8), 52 (25.9), 51 (14.4), 29 (8.0). Calcd: 138.061 887.
Obsd: 138.061 800.
Theoretical Methods. Three different levels of theory were applied.
(i) First calculations were carried out with DFT using both the BLYP17a,b
and the B3LYP17c functional and a variety of basis sets ranging from
6-31G(d,p) to 6-311G(d,p), 6-311G(2d,2p), and 6-311++G(3df,3pd).16
22
Apart from this, Dunning’s cc-pVDZ and cc-pVTZ basis sets were
used to verify the results. Geometries and the vibrational spectra were
calculated to characterize stationary points and to determine IR spectra
for comparison with experimental data. (ii) In the second step, restricted
DFT (RDFT) results were checked for internal and external stability.23
At large 1,3 distances (> 2.017 Å) the RB3LYP solution of 2a turned
out to become unstable, and a more stable broken symmetry (BS)-
UB3LYP solution was found by mixing of HOMO and LUMO.24
However, the BS-UDFT minimum of 2a vanishes for larger basis sets,
and even for the smaller basis sets it is separate by a barrier of just
0.03 kcal/mol from the more stable bicyclic form 2b. The barrier
vanishes when ZPE corrections are applied. Hence, for none of the
B3LYP/basis set combinations used a true minimum was found for
2a, which may be rationalized by the large S-T splitting of 2a and
strong T contamination of the UDFT description of 2a. (iii) In the third
step, geometry and the vibrational spectrum were calculated at the
CCSD(T) level of theory15 using the 6-311G(2d,2p) basis set.
Calculations were performed with Gaussian 98,25 COLOGNE2000,26
and a local version of the ACES II program.27
Bis(1,3-propanedithioacetal) of 2,4,5,6-Tetra-deuterioisophthal-
aldehyde. A 3.9 g (28.0 mmol) amount of 2,4,5,6-tetradeuterioiso-
phthalaldehyde was dissolved in 70 mL of acetic acid (99-100%). A
1.4 mL aliquot (5.3 mmol) of BF3-Et2O (45%) and 7.1 mL of propane-
1,3-dithiol (44.4 mmol) were added dropwise under stirring at room
(22) (a) Woon, D. E.; Dunning, T. H. J. Chem. Phys. 1993, 98, 1358. (b) Kendall,
R. A.; Dunning, T. H.; Harrison, R. J. J. Chem. Phys. 1992, 96, 6796. (c)
Dunning, T. H. J. Chem. Phys. 1989, 90, 1007.
(23) Bauernschmitt, R.; Ahlrichs, R. J. Chem. Phys. 1996, 104, 9047.
(24) Gra¨fenstein, J.; Kraka, E.; Filatov, M.; Cremer, D. Int. J. Mol. Sci. 2002,
3, 360.
(25) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98; Gaussian, Inc.:
Pittsburgh, PA, 1998.
General Procedure. The [2.2]m,p-cyclophane 5 and its perdeuterated
isotopomer 5-d4 were synthesized according to a modified procedure
by Boekelheide et al.28
(26) Kraka, E.; Gra¨fenstein, J.; Gauss, J.; He, Y.; Reichel, F.; Olsson, L.; Konkoli,
Z.; He, Z.; Cremer, D. COLOGNE2000; Go¨teborg University: Go¨teborg,
Germany, 2000.
(27) Stanton, J. F.; Gauss, J.; Watts, J. D.; Lauderdale, W. J.; Bartlett, R. J.
ACES II; Quantum Theory Project, University of Florida: Gainesville, FL,
1992.
(28) Boekelheide, V.; Anderson, P. H.; Hylton, T. A. J. Am. Chem. Soc. 1974,
96, 1558.
9
13078 J. AM. CHEM. SOC. VOL. 124, NO. 44, 2002

Vibrational Spectrum of m-Benzyne
A R T I C L E S
temperature. After the mixture was stirred for another 11 days, the
colorless precipitate was filtered off and dried in vacuo (7.48 g, 23.5
mmol, 84.2%): mp 126-127 °C; IR (cm-1) 2934, 2899, 2826, 1737,
1579, 1423, 1381, 1342, 1276, 1252, 1224, 1184, 1116, 999, 949, 910,
873, 840, 808, 793, 742, 730, 693, 674, 650, 605; 1H NMR (400 MHz,
CDCl3) δH 5.14 (s), 3.06-2.98 (m), 2.90-2.85 (m), 2.17-2.10 (m),
1.96-1.85 (m); 13C NMR (100.66 MHz, CDCl3) δC 139.98, 128.88,
51.09, 31.96, 25.05; MS (EI, 70 eV) (m/z, %) 318 (100) [M+], 317
(50.0), 245 (15.4), 244 (65.5), 243 (36.2), 211 (15.7), 210 (15.4), 170
(53.6), 169 (53.9), 168 (15.9), 157 (16.5), 139 (12.6), 138 (13.4), 125
(12.6), 119 (15.2), 106 (91.7), 105 (31.8), 74 (39.8), 73 (31.7), 58 (15.2),
46 (21.0), 45 (41.2), 40 (22.0). Calcd: 318.054 245. Obsd: 318.053 200.
4,5,6,8-Tetradeuterio[2.2]metaparacyclophanebis(1,3-propane-
dithioketal). A 10 mL aliquot of a solution of n-butyllithium in hexane
(1.6 N) was added under argon to a cooled (-32 °C) solution of 2,4,5,6-
tetradeuterioisophthalaldehydebis(1,3-propanedithioacetal) (1.60 g, 5
mmol) in dry THF (250 mL), stirred for 1.5 h at the same temperature,
and then pumped into a cooled (-32 °C) dropping funnel with an excess
pressure of argon. This reddish brown solution of the dianion (-32
°C) and a solution of R,R′-dibromo-p-xylene (1.32 g, 5 mmol) in THF
(250 mL, room temperature) were simultaneously added under high
dilution conditions over a period of 6 h to refluxing THF (1000 mL).
After removal of the solvent, the residue was absorbed on silica gel
and separated from polymeric materials by eluation with CH2Cl2.
Colorless crystals were obtained by using thin layer column chroma-
tography (silica gel, CH2Cl2) (1.0 g, 2.5 mmol, 49%): DP > 220 °C;
IR (cm-1) 3035, 2902, 2352, 1882, 1725, 1593, 1501, 1424, 1357, 1327,
1275, 1243, 1200, 1117, 991, 934, 923, 904, 850, 835, 816, 798, 748,
4,5,6,8-Tetradeuterio[2.2]metaparacyclophane-2,9-dione-d4 (5).
A solution of 4,5,6,8-tetradeuterio[2.2]m,p-cyclophanebis(1,3-pro-
panedithioketal) (1.4 g, 3.4 mmol) in THF (50 mL) was added dropwise
to a stirred slurry of 2.2 g (10.2 mmol) of mercury(II)oxide and 2.9
mL (10.3 mmol) of BF3-Et2O (45%) in THF/15% (25 mL) water at
room temperature. The mixture was stirred for another 7.5 h at 40 °C
after the complete addition. The cooled mixture was filtered off (Na2-
SO4, Celite, sand, silica gel) at room temperature and dried over MgSO4.
After removal of the solvent the cyclophane was purified by thin-layer
chromatography (silica gel, CH2Cl2) to yield pale yellowish crystals
(90 mg, 0.4 mmol, 11.0%): IR (10 K, cm-1) 3032, 3016, 1889, 1709,
1691, 1570, 1541, 1505, 1413, 1405, 1371, 1338, 1315, 1287, 1273,
1216, 1137, 1103, 1022, 989, 939, 929, 904, 871, 854, 837, 830, 823,
812, 799, 788, 768, 759, 746, 732, 679, 668, 590, 587, 555, 490, 480,
1
422, 370, 294; H NMR (200 MHz, CDCl3) δH 7.04, 3.90; 13C NMR
(50.33 MHz, CDCl3) δC 202.9, 141.02, 136.02, 132.60, 127.60, 52.57;
MS (EI, 70 eV) (m/z, %) 240 (59.8) [M+], 239 (26.5), 184 (36.6), 183
(22.2), 108 (100.0), 107 (46.6), 104 (11.6), 80 (78.9), 79 (36.4), 78
(14.6), 77 (12.6), 52 (23.0), 51 (11.5). Calcd: 240.108 837. Obsd:
240.110 300.
Acknowledgment. At Bochum, this work was financially
supported by the Deutsche Forschungsgemeinschaft and the
Fonds der Chemischen Industrie and at Go¨teborg by the Swedish
Natural Science Research Council (NFR). Calculations were
done on the supercomputers of the Nationellt Superdatorcentrum
(NSC), Linko¨ping, Sweden. D.C. and E.K. thank the NSC for
a generous allotment of computer time.
1
729, 679, 665, 645, 617; H NMR (400 MHz, CDCl3) δH 7.32, 5.84,
3.46 (d), 2.75 (d), 2.89-2.37 (m), 1.97-1.85 (m); 13C NMR (100.66
MHz, CDCl3) δC 140.66, 134.55, 129.15, 128.62, 60.26, 53.55, 28.95,
27.44, 24.99; MS (EI, 70 eV) (m/z, %) 420 (29.4) [M+], 419 (12.4),
318 (19.2), 317 (30.5), 316 (100.0), 315 (76.6), 314 (21.3), 283 (12.8),
242 (32.3), 241 (15.9), 168 (41.9), 167 (22.1), 158 (21.7), 149 (10.1),
124 (32.9), 123 (14.8), 121 (16.4), 120 (11.5), 86 (36.2), 84 (61.2), 49
(12.0), 47 (14.7), 41 (10.5), 36 (14.9). Calcd: 420.101 195. Obsd:
420.101 500.
Supporting Information Available: Figures 1S-6S detailing
IR spectra of argon matrices at 10 K containing m-benzyne
generated from three different precursors in the ranges 500-
700, 700-900, 900-1100, 1100-1300, 1300-1500, and
1500-1650 cm-1 (PDF).This material is available free of charge
via the Internet at http://pubs.acs.org.
JA012686G
9
J. AM. CHEM. SOC. VOL. 124, NO. 44, 2002 13079