Carbomethoxychlorocarbene: Spectroscopy, theory, chemistry and kinetics

Article abstract of DOI:10.1021/ja004235m

Photolysis (254 nm) of methyl 8-chloro-3a,7a-methanoindan-8-carboxylate (5) in argon at 14 K produces carbomethoxychlorocarbene (6) as a persistent species. The IR and UV-vis spectra of the carbene were recorded and analyzed with the aid of density functi

Full text of DOI:10.1021/ja004235m

J. Am. Chem. Soc. 2001, 123, 6061-6068
6061
Carbomethoxychlorocarbene: Spectroscopy, Theory, Chemistry and
Kinetics
Igor Likhotvorik, Zhendong Zhu, Eunju Lee Tae, Eric Tippmann, Brian T. Hill, and
Matthew S. Platz*
Contribution from the Department of Chemistry, The Ohio State UniVersity, 100 West 18th AVenue,
Columbus, Ohio 43210
ReceiVed December 11, 2000. ReVised Manuscript ReceiVed May 2, 2001
Abstract: Photolysis (254 nm) of methyl 8-chloro-3a,7a-methanoindan-8-carboxylate (5) in argon at 14 K
produces carbomethoxychlorocarbene (6) as a persistent species. The IR and UV-vis spectra of the carbene
were recorded and analyzed with the aid of density functional calculations (B3-LYP/6-31G*). The IR spectrum
of 6 is consistent with the carbene having a nonplanar singlet ground state, in agreement with the G3(MP2)//
B3-LYP calculations of Scott and Radom (accompanying paper). Irradiation (300 nm) of 5 in solution produces
indane in 97% yield. In cyclohexane, carbene 6 is trapped by insertion into a CH bond, whereas in 2,3-
dimethylbutene it adds to the double bond to form a cyclopropane. Laser flash photolysis of 5 (308 nm, 17 ns,
XeCl excimer) produces carbene 6 which reacts with pyridine to form an ylide (λmax ) 440 nm). It was
possible to resolve the growth of the ylide in Freon-113 (CF2ClCFCl2) to measure the lifetime (τ ) 114 ns,
ambient temperature) of the carbene and the absolute rate constant of its reaction with pyridine (kpyr ) 2 ×
109 M-1 s-1). A plot of log(1/τ) versus 1/T in CF2ClCFCl2 is linear with Arrhenius parameters Ea ) 10.9 (
0.8 kJ/mol and A ) 109.1(0.2 s-1. In perfluorohexane, a less reactive solvent than Freon-113, Ea ) 23.4 ( 1.7
kJ/mol, A ) 1010.6(0. s-1, and τ ) 354 ns at 293 K. It is argued that the activation barrier to carbene disappearance
in perfluorohexane represents the lower limit to the barrier to Wolff rearrangement of the carbene.
I. Introduction
the other hand, the lowest singlet states of the carbonyl carbenes
are predicted to be nonplanar with closed shell configurations.5-8
Formyl carbene has not been generated as a persistent species
in a low-temperature matrix.9 Triplet carbomethoxycarbene,
however, has been detected by ESR spectroscopy at low
temperature.10 This carbene has not been observed by matrix
IR and UV-vis spectroscopy because these techniques have
lower sensitivity than ESR spectroscopy.9 A 20 year old
calculation of Kim and Schaefer indicates that the triplet state
of carbohydroxycarbene is 29.3 kJ/mol more stable than the
singlet. A more recent calculation of Xie and Schaefer predicts
a triplet-singlet spliting of 19.7 kJ/mol.6 Thus, the interpretation
of time-resolved experiments of simple carbonyl carbenes is
complicated by the fact that the ground states of carbenes 1-3
have triplet spin multiplicity,5-8 but WR proceeds through the
closed shell singlet state.11
Carbonylcarbenes are widely used in synthesis and are easily
generated.1 Consequently, the simplest examples such as 1-3
have been extensively studied by classical methods. The
lifetimes of 1-3 in Freon-113 and in cyclohexane are controlled
by intramolecular Wolff rearrangement (WR) and are ≈1 ns at
ambient temperature.2-4 The short lifetimes of 1-3 in solution
prevented direct measurement of the Arrhenius parameters of
the WR of these carbenes by nanosecond spectroscopy.
The ground states of simple carbonyl carbenes are predicted
by theory to have triplet multiplicity and to be planar.5-8 On
Aryl carbonyl carbenes have triplet ground states and have
been characterized in low-temperature matrices by ESR, UV-
vis, and IR spectroscopy.9,12 The aryl group stabilizes the singlet
(1) (a) Rodina, L. L.; Korobitsyna, I. K. Russ. Chem. ReV. 1967, 36,
260. (b) Marchand, A. P.; Brockway, N. M. Chem. ReV. 1974, 74, 431. (c)
Meier, H.; Zeller, K.-P. Angew. Chem., Int. Ed. Engl. 1975, 14, 32. (d)
Toscano, J. P. In AdVances in Carbene Chemistry; Brinker, U. H., Ed.; JAI
Press: Greenwich, CT, 1998; Vol. 7, p 215.
(2) (a) Toscano, J. P.; Platz, M. S.; Nikolaev, V.; Cao, Y.; Zimmt, M.
B. J. Am. Chem. Soc. 1996, 118, 3527. (b) Toscano, J. P.; Platz, M. S.;
Nikolaev, V. J. Am. Chem. Soc. 1995, 117, 4712.
(3) Toscano, J. P.; Platz, M. S.; Nikolaev, V. J. Am. Chem. Soc. 1994,
116, 8146.
(4) Wang, J.-L.; Toscano, J. P.; Platz, M. S.; Nikolaev, V.; Popik, V. J.
Am. Chem. Soc. 1995, 117, 5477.
(6) (a) Kim, K. S.; Schaefer, H. F., III. J. Am. Chem. Soc. 1980, 102,
5389. (b) Xie, Y.; Schaefer, H. F., III Mol. Phys. 1996, 87, 389-397.
(7) Vacek, G.; Galbraith, J. M.; Yamaguchi, Y.; Schaefer, H. F., III;
Nobes, R. H.; Scott, A. P.; Roden, L. J. Phys. Chem. 1994, 98, 8660.
(8) Scott, A.; Radom, L. J. Am. Chem. Soc. 2001, 123, 6069-6076.
(9) Sander, W.; Bucher, G.; Wierlacher, S. Chem. ReV. 1993, 93, 1583.
(10) Hutton, R. S.; Roth, H. D. J. Am. Chem. Soc. 1978, 100, 4321.
(11) Scott, A. P.; Nobes, R. H.; Schaefer, H. F., III; Radom, L. J. Am.
Chem. Soc. 1994, 116, 10159.
(12) (a) Fujiwara, Y.; Tanimoto, Y.; Itoh, M.; Hirai, K.; Tomioka, H. J.
Am. Chem. Soc. 1987, 109, 1942. (b) McMahon, R. J.; Chapman, O. L.;
Hayes, R. D.; Hess, T. C.; Krimmer, H. P. J. Am. Chem. Soc. 1985, 107,
7597. (c) Hayes, R. A.; Hess, T. C.; McMahon, R. J.; Chapman, O. L. J.
Am. Chem. Soc. 1983, 105, 7786.
(5) (a) Baird, N. C.; Taylor, K. F. J. Am. Chem. Soc. 1978, 100, 1333.
(b) Tanaka, K.; Yoshimine; M. J. Am. Chem. Soc. 1980, 102, 7653. (c)
Harding, L. B.; Wagner, A. F. J. Phys. Chem. 1987, 91, 6484. (d) Gosavi,
R. K.; Torres, M.; Strausz, O. P. Can. J. Chem. 1991, 69, 1630.
10.1021/ja004235m CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/25/2001

6062 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
LikhotVorik et al.
state of the carbene which allows relaxation to the triplet state
to compete with WR.
Photolysis of matrix isolated planar triplet carbene 4T
produces orthogonal singlet carbene 4S (Np ) 2-naphthyl). A
small intrinsic barrier, enhanced perhaps by frictional forces in
the matrix, prevents rapid relaxation of 4S to 4T. This allows
the low-lying singlet 4S to be characterized by IR and UV-vis
spectroscopy in frozen argon.13 This is the only case in which
both the lowest singlet and triplet states of a carbene have been
studied by matrix spectroscopy.13
Figure 1. The IR difference spectrum obtained after 254 nm photolysis
of 5 in an Ar matrix at 12 K (top). IR spectra of the two singlet carbene
conformers, Z-16 (middle) and E-16 (bottom), calculated at the B3LYP/
6-31G* level (all frequencies scaled by 0.97).
The equilibrium constant K ) 4S/4T was measured by
Toscano et al. by TRIR spectroscopy in liquid CF2ClCFCl2
(Freon-113) and is approximately 0.7 (∆GST ) 0.2) at ambient
temperature.14 The barrier to disappearance of the spin equili-
brated carbene (4S, 4T) in Freon-113, presumably by Wollf
rearrangement, is 14.2 kJ/mol.15 Again, this result is difficult
to interpret because the triplet is the ground state and the
rearrangement proceeds on the singlet surface.11,14,15
interest of simplicity. Pouring 8 onto dry ice followed by
treatment with methyl iodide formed 5 in 65-76% yield (from
7).
Herein, we are pleased to report that photolysis of 8-chloro-
3a,7a-methanoindane-8-carboxylate (5) produces carbomethoxy-
chlorocarbene (6). The chlorine substituent differentially sta-
bilizes the singlet state and renders the singlet as the ground
state.16 Chlorine, as aryl substitution kinetically stabilizes the
singlet carbene17 and allows for direct observation and study
of the chemistry and spectroscopy of this carbene by a variety
of techniques. It also allows a straightforward mechanistic
analysis of the kinetics of disappearance of a singlet carbon-
ylcarbene in solution without having to consider the influence
of spin state equilibration.
II.2. Matrix Isolation Spectroscopy. Compound 5 was
deposited in argon at 14 K and its IR and UV-vis spectra were
recorded. It was then irradiated at 254 nm (Ray-O-Net) and
spectra were obtained at different intervals. Upon photolysis,
the IR bands of 5 (1750, 1243, 1178, 1047, 742, 728, and 652
cm-1) are reduced and prominent bands at 1270, 1007, and 757
cm-1 as well as a single, broad carbonyl band centered 1702
cm-1 appear in the IR spectrum (Figure 1). Carbon monoxide
(2139 cm-1) is also produced upon photolysis (254 nm) of 5.
One of the anticipated products of photolysis of 5 is indane.
Thus, indane was independently deposited in argon and its
vibrational spectrum was recorded (Figure S7 in the Supporting
Information). The two largest bands in the IR spectrum of indane
are observed at 738 and 751 cm-1. It is difficult to identify
these two bands conclusively in the spectrum obtained upon
photolysis of 5, because of the disappearance of the precursor
band at 742 cm-1. It is clear, however, that while indane may
contribute to the photoproduct band at 757 cm-1, the prominent
photoproduct bands observed at 1702, 1270, and 1007 cm-1
are not due to indane and must originate from another species.
A weak broad band appears at 640 nm in the visible region
along with a stronger band at 320 nm in the ultraviolet region
of the spectrum (Figure 2). These two absorption bands are
formed under the same photolysis conditions which lead to the
changes observed in the IR spectra of Figure 1.
II. Results
II.1. Synthesis. Jones’ dichloride 7 has been shown to be a
convenient photochemical source of dichlorocarbene.18 This
compound was treated with n-butyllithium in THF at -78 °C
to form an organometallic reagent we will designate as 8 in the
(13) Zhu, Z.; Bally, T.; Stracener, L.; McMahon, R. J. Am. Chem. Soc.
1999, 121, 2863.
(14) Wang, Y.; Yuzawa, T.; Hamaguchi, H.; Toscano, J. P. J. Am. Chem.
Soc. 1999, 121, 2875.
(15) Wang, J.-L.; Likhotvorik, I.; Platz, M. S. J. Am. Chem. Soc. 1999,
121, 2883.
(16) (a) Carter, E. A.; Goddard, W. A., III. J. Chem. Phys. 1988, 88,
1752. (b) Shin, S. K.; Goddard, W. A., III; Beauchamp, J. C. J. Chem.
Phys. 1990, 94, 6963. (c) Murray, K. K.; Leopold, D. G.; Miller, T. M.;
Lineberger, W. C. J. Chem. Phys. 1988, 89, 5442.
(17) (a) Bucher, G.; Scaiano, J. C.; Platz, M. S. Kinetics of Carbene
Reactions in Solution, Landolt-Bornstein, Group II, Vol. 18, Subvolume
E2; Springer: Berlin, Germany, 1998; p 141. (b) Ge, C.-S.; Eun, G. J.;
Jefferson, E. A.; Liu, W.; Moss, R. A.; Wlostowska, J.; Xue, S. J. Chem.
Soc., Chem. Commun. 1994, 1479. (c) Moss, R. A. Pure Appl. Chem. 1995,
67, 741.
The vibrational bands at 1702, 1270, and 1007 cm-1 together
with the electronic absorptions at 320 and 640 nm will be
attributed to singlet carbomethoxychlorocarbene 6 on the basis
of the calculations presented in Section II.3 as well as their
behavior upon photolysis and in the presence of carbon
monoxide. Furthermore, it will be shown that the IR spectrum
is consistent with the presence of a single conformer of the
carbene, the lowest energy (Z-conformation) of 6 in the matrix.
(18) Hartwig, J. F.; Jones, M., Jr.; Moss, R. A.; Lawrynowicz, W.
Tetrahedron Lett. 1986, 27, 5907.

Carbomethoxychlorocarbene
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 6063
Figure 3. The B3LYP/6-31G* equilibrium geometries and relative
1
energies (in parentheses) of the two conformers of 6.
Figure 4. The B3LYP/6-31G* equilibrium geometries and relative
energies (in parentheses) of the two most stable conformers of 36
(conformer Z-36a is 9.6 kJ/mol lower in energy than Z-16).
Finally, 6 appears to lose a small amount of carbon monoxide
(2139 cm-1) upon photolysis. Some carbon monoxide is always
produced upon 254 nm photolysis of precursor 5.
Figure 2. UV/vis spectra in an Ar matrix: (a) before the photolysis
of 5; (b) after 254 nm photolysis for 11 min; (c) after subsequent 350
nm for 10 min. The corresponding difference spectra are shown at the
top.
II.3. Density Functional Calculations. B3LYP density
functional calculations20,21 were performed with the 6-31G*
basis set to interpret the above spectroscopic results. The
geometries of various conformers of the singlet and triplet state
of carbene 6 were fully optimized, and the minima were
identified by harmonic vibrational frequency calculations. More
extensive calculations on 6 are reported by Scott et al.8
As in the case of other carbonyl carbenes, the singlet state of
carbomethoxychlorocarbene (16) was found to be nonplanar.5-8
Two stable conformations of 16 were found (Figure 3) whereby
the Z-conformation is 18.6 kJ/mol lower in energy than the
E-conformation as is typical for Z- and E-conformations of
esters.22 A transition structure connecting Z- and E-conforma-
tions was found to lie 46.2 kJ/mol above the anti isomer.
The triplet carbene (36) was optimized within Cs symmetry.
Four planar conformations were found, two of which (Z-36a
and Z-36b) are depicted at the bottom of Figure 4 (the other
two (which are not shown) represent the higher lying E-
conformers E-36c and E-36d). The most stable structure of the
triplet carbene, 36a, is found to be 9.3 kJ/mol more stable than
the lowest energy conformation of the singlet carbene by
B3LYP/6-31G*. However, this theoretical model predicts that
the triplet state of methylene is stabilized by 55.2 kJ/mol relative
to its lowest energy singlet state8 which is 17.5 kcal/mol more
than is found experimentally. On the other hand, the G3(MP2)
method which successfully reproduces the singlet-triplet split-
ting of methylene (calculated, 39.5 kJ/mol;8 experiment, 37.7
Photolysis of 5 in argon containing 4.5% CO again produces
carbene 6, but in reduced yield. The experimental spectra in
the presence and absence of CO are very similar, but two new
bands at 1320 and 2144 cm-1 are observed only in the presence
of CO (cf. Figure S1 in Supporting Information). These two
bands are in good agreement with prominent vibrational
transitions of ketene 9 predicted at 1320 and 2167 cm-1 by
B3LYP/6-31G* (after scaling by a factor of 0.97). This
demonstrates that a CO trappable species (e.g. carbene 6) is
produced upon photolysis of 5.
Photolysis of 6 in argon with either 350 or 650 nm light
bleaches both the UV and the visible absorption bands of the
carbene. This is accompanied by several changes in the IR
spectra (see Figure S9, Supporting Information). From these
we conclude that one significant reaction of excited 6 appears
to be retro-cycloaddition to indane (which is trapped in the same
matrix cavity as 6) to re-form the starting material, 5. Photolysis
of 6 also appears to induce WR to form a small amount of ketene
10 (experiment, 2148 cm-1; calculation, 2167 cm-1; see
Supporting Information) as well as cyclization to a â-lactone,
a process that has precedent.13,17,19 The observed carbonyl
frequency of the â-lactone 11 that arises by cyclization of 6
(1854 cm-1) is in satisfactory agreement with the calculated
(20) (a) Becke, A. D. Phys. ReV. A 1988, 38, 3098. (b) Becke, A. D. J.
Chem. Phys. 1993, 98, 5648.
(21) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (b)
Labanowski, J. W.; Andzelm, J. Density Functional Methods in Chemistry;
Springer: New York, 1991. (c) Parr, R. G.; Yang, W. Density Functional
Theory in Atoms and Molecules; Oxford University Press: New York, 1989.
(22) In each case, the calculation was performed for the Z-conformation
about the CO-OCH3 bond. Experiments (Blom, C. E.; Gu¨nthard, Hs. H.
Chem. Phys. Lett. 1981, 84, 267) and calculations (Wang, X.; Houk, K. N.
J. Am. Chem. Soc. 1988, 110, 1870-1872. Wiberg, K. B.; Laidig, K. E. J.
Am. Chem. Soc. 1988, 110, 1872-1874) likewise show that the Z-
conformations of esters are more stable than the E-conformations by 5-10
kcal/mol, due to unfavorable dipolar interactions in the latter.
value (1899 cm-1). A second species is observed (1877 cm-1
)
that may also be a â-lactone, but is currently unidentified.
(19) Richardson, D. C.; Hendrick, M. E.; Jones, M., Jr. J. Am. Chem.
Soc. 1971, 93, 3790.

6064 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
LikhotVorik et al.
Table 1. B3LYP/6-31G* Harmonic Frequencies of Singlet E- and
Z-6a and the Experimental IR Spectrum of 6 Observed in an Argon
Matrix
calcd for Z-6
cm-1 Ib
0.5
calcd for E-6
mode
cm-1
Ib
obsd
1
2
3
76
95
88
100
157
0.15
0.17
1.3
0.1
0.6
118
4
5
237
245
0.3
2.3
211
233
7.2
3.9
6
7
311
536
15.5
4.4
349
520
9.7
3.5
8
9
732
777
861
18.1
14.3
9.2
33.4
0.3
3.9
100
3.3
2.2
2.3
57.9
6.1
4.1
652
763
857
14.2
737 (w)
757 (w)
883 (w)
1007 (m)
4.7
43.1
22.7
0.65
3.0
100
4.1
5.9
4.5
10
11
12
13
14
15
16
17
18
19
20
21
1001
1144
1180
1297
1445
1468
1476
1702
2991
3069
3101
1045
1143
1156
1311
1456
1473
1481
1706
2968
3042
3090
1270 (vs)
1437 (vw)
1462 (vw)
1485 (vw)
1702 (s)
2997 (vw)
3029 (vw)
79.1
3.9
5.4
2.8
3.3
Figure 5. The molecular orbitals involved in the electronic transitions
of carbomethoxychlorocarbene listed in Table 2.
a All calculated frequencies are scaled by a factor of 0.97. b Intensities
relative to the strongest transition at 1297 cm-1 (Z-6) or 1311 cm-1
(E-6).
electron from the HOMO (in-plane σ orbital) to the LUMO (out-
of-plane pure p orbital).24-26 The absorption maxima of this
transition for a few carbenes are shown below. Thus, we assign
the absorption of 6 at 640 nm to promotion of an electron from
the HOMO to the LUMO of the carbene.
Table 2. Excited States of Singlet Carbomethoxychlorocarbene Z-6
from TD-B3LYP/6-31G* Calculations
excited state
λ (nm)
f
excitationsa
coefficientb
1
2
648
290
0.0001
0.0026
27 f 28
26 f 28
27 f 29
25 f 28
24 f 28
26 f 28
26 f 29
27 f 29
0.51383
0.59061
0.25220
0.69244
0.10695
-0.17986
-0.14777
0.56320
3
4
277
249
0.0001
0.0093
a In terms of molecular orbitals some of which are depicted in Figure
5. b Contribution of excited configuration x f y to the excited-state
wave function.
This assignment is validated by TD-B3LYP/6-31G* cal-
culations on Z-6 (see Table 2). This method predicts a tran-
sition with a very small oscillator strength at 648 nm that is
due to promotion of an electron from the σ HOMO to the π
LUMO. It also predicts two intense transitions at 249 and
289 nm which we cannot clearly resolve in the experimental
spectra. However, it is tempting to associate the predicted
transition at 289 nm with the band observed at 320 nm in argon.
The orbitals involved in these transitions are shown in
Figure 5.
The TD-B3LYP/6-31G* method was also applied to the
lowest energy conformer of the triplet of 6, and it predicts that
the three lowest energy transitions are at 303, 339, and 377 nm
with oscillator strengths of 0.0001, 0.0230, and 0.0004, respec-
tively (full details can be found in the Supporting Information).
The calculated electronic spectrum of singlet 6, especially the
diagnostic HOMO-LUMO transition in the visible region, is
more consistent with the experimental spectrum than the
calculated spectrum of the triplet carbene.
kJ/mol23) predicts that the singlet state of 6 is 16.0 kJ/mol more
stable than its triplet state.8
The harmonic frequencies for the Z- and E-conformers of
singlet carbene 6 computed by B3LYP/6-31G* are given in
Table 1 along with the experimentally observed values. Vibra-
tional frequencies for the lowest energy conformer of triplet
carbene (36a) are listed in the Supporting Information. The
experimental IR spectrum of carbene 6 (Figure 1) is clearly more
consistent with the calculated vibrational spectrum of the singlet,
rather than the triplet state of the carbene.
The very strong band at 1702 cm-1 is attributed to the
carbonyl stretch of singlet 6. This compares to carbonyl
stretching frequencies of 1660 and 1625 cm-1 for naphthyl-
carbene esters 4T and 4S, respectively.13 The IR bands at 1270
and 1007 cm-1 are assigned to C-O single bond stretches of
singlet carbene 6. Thus, although B3-LYP/6-31G* fails to
correctly predict the ground state of 6 it adequately simulates
the experimental IR spectrum of singlet carbomethoxychloro-
carbene.
II.4. Chemical Trapping of Carbomethoxychlorocarbene.
When precursor 5 was dissolved in Freon-113 (CF2ClCFCl2)
and exposed to 300 nm radiation (Quartz, Ray-O-Net), indane
was formed in 97% absolute yield (GC, decane as internal
Singlet carbenes typically have characteristic broad absorption
bands in the visible region associated with the promotion of an
(24) Milligan, D. E.; Jacob, M. E. J. Chem. Phys. 1967, 47, 703.
(25) Pliego, J. R., Jr.; De Almedia, W. B.; Celebi, S.; Zhu, Z.; Platz, M.
S. J. Phys. Chem. A 1995, 103, 7481.
(23) Leopold, D. G.; Murray, K. K.; Stevens-Miller, A. E.; Lineberger,
W. C. J. Chem. Phys. 1985, 83, 4849.
(26) Bally, T.; Matzinger, S.; Truttmann, L.; Platz, M. S.; Morgan, S.
Angew Chem., Int. Ed. Engl. 1994, 33, 19.

Carbomethoxychlorocarbene
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 6065
Table 3. Values of kpyrτ Obtained in Various Solvents at Ambient
Temperature
Scheme 1
solvent
acetonitrile
acetonitrile-d3
cyclopropyl cyanide
pentane
benzene
cyclohexane
cyclohexane-d12
kpyr
τ
τ [ns]a
1.4 ( 0.4
3.5 ( 0.3
11 ( 2
0.7
2
6
45
16
18
47
89 ( 12
32 ( 2
36 ( 6
93 ( 7.4
a Assuming kpyr ) 2 × 109 M-1 s-1 as in Freon-113.
II.5. Laser Flash Photolysis (LFP) Studies. LFP of 5 in
Freon-113 at ambient temperature fails to produce a useful
transient. The broad absorption of 6 detected in argon at 14 K
at 640 nm is too weak to be of value in LFP experiments.
However, LFP of 5 in pentane containing 1.03 M pyridine
produces a transient spectrum with an intense, broad band
peaking at 440 nm (Figure S2 in the Supporting Information),
which is attributed to ylide 15 (Scheme 2).
Scheme 2
It was possible to follow the formation of 15 in Freon-113
(Figure S3, Supporting Information) and in perfluorohexane.
The ylide was formed in an exponential process that could be
analyzed to yield kobs, the observed rate constant of ylide
formation. As expected for a carbene with a singlet ground state,
the magnitude of kobs was not sensitive to the presence of
oxygen.
According to Scheme 2, kobs can be equated to various
elementary rate constants as shown in eq 1, where kWR, kRX
,
kobs ) kWR + kRX[RX] + kpyr[PYR] (1)
standard). Similarly, photolysis in neat cyclohexane or 2,3-
dimethyl-2-butene (tetramethylethylene, TME), respectively, led
to the clean formation of indane and a carbene-solvent adduct.
and kpyr are as defined in Scheme 2, while [RX] and [PYR]
denote the concentrations of the Freon and pyridine, respectively.
A plot of kobs versus [PYR] in Freon-113 is linear (Figure
S5 in the Supporting Information) with a slope that corresponds
to kpyr ) 2.0 × 109 M-1 s-1 at ambient temperature. This is
comparable to other carbene rate constants with pyridine.17 The
intercept of this plot is kWR + kRX[RX], which is 1/τ, where τ
is the carbene lifetime under these conditions (114 ns). These
experiments were repeated in perfluorohexane, in which the
carbene has a lifetime of 354 ns at ambient temperature.
In other, more reactive solvents, the yield of ylide was
measured as a function of pyridine concentration by transient
optical spectroscopy. A double reciprocal treatment of the data
is linear, as shown in the insert to Figure S5 of the Supporting
Information. The ratio of the intercept and the slope of such a
plot is kpyr‚τ. These data are collected in Table 3 and analyzed
using the kpyr value measured in Freon-113, assuming that kpyr
is independent of solvent. In this manner it is found that the
lifetime of 6 in cyclohexane is 2.6 times shorter than that in
cyclohexane-d12. This is consistent with our previously discussed
product study which demonstrates that the lifetime of 6 in
cyclohexane is controlled to some extent by insertion into a
C-H bond of the solvent. The lifetimes of 6 in benzene and in
acetonitrile are both comparatively short. This could be due to
a solvent effect on either kpyr or kWR or more likely due to rapid
chemical reaction of the carbene with the solvent.
Adducts 13 and 14 were formed in 85% and 16% yield,
respectively, relative to indane, as demonstrated by NMR
spectroscopy. Thus, there is little doubt that carbene 6 is formed
from 5 in solution. Although LFP studies (vide infra) suggest
that Freon-113 is not an inert solvent, conclusive evidence for
the formation of any adducts between 6 and this solvent was
not obtained by GC-MS analysis. The most likely mode of
carbene reaction with Freon-113 is chlorine atom abstraction
to form a radical pair,15 as shown in Scheme 2.
The yield of cyclopropane adduct 13 was measured as a
function of the concentration of TME in Freon-113. As expected,
the yield increases initially upon increasing [TME] and then
attains a limiting value. Plotting one over the yield of 13 versus
1/[TME] gives a straight line from which we can obtain kTME
‚
τ, where τ ()1/kOBS) is the lifetime of carbene 6 in Freon-113
in the absence of TME. A plot of ln(kTME/kOBS) versus 1/T
The yield of 15 in Freon-113 containing a constant concentra-
tion [PYR] ) 0.01M is diminished by the presence of a
competitive carbene quencher such as ethanol (EtOH) or 2,3-
dimethyl-2-butene (DMB). Stern-Volmer treatment of the data
0
between 223 and 303 K is also linear and it yields EaTME - Ea
) 3.0 kJ/mol and ATME/A0 ) 2.2, where the superscripts refer
to the Arrhenius parameters for carbene reaction with TME,
and the decay of the carbene in Freon-113 in the absence of
TME by all processes, such as Wolff rearrangement and reaction
with solvent, which consume the carbene in the absence of TME.
(cf. Figure S6 in the Supporting Information) reveals that kEtOH
/
k
pyr is 35 and kTME/kpyr is 23.3. As kpyr ) 2 × 109 M-1 s-1 then
kEtOH )7.0 × 108 M-1 s-1 and kDMB ) 4.1× 108 M-1 s-1 in
Freon-113.

6066 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
LikhotVorik et al.
method.2,3 The lifetimes of both carbenes were found to be ≈1
ns in solution at ambient temperature and they were comparable
in cyclohexane and cyclohexane-d12 which indicated that the
carbene lifetimes are controlled by intramolecular Wolff re-
arrangement rather than by reaction with solvent. It seems likely
that spin state equilibration of formyl and carbomethoxycarbene
is complete within 1 ns, based on analogy to arylcarbenes.27
Because the lifetimes of these carbenes were shorter than the
time resolution of the spectrometer, only the ratio of kWR/kPYR
could be determined. This ratio showed little variation with
temperature. In the present work, the absolute value of the
activation energy of the reaction of 6 with pyridine could be
measured directly and it was found to be 2.9, 4.2, and 5.9 kJ/
mol in Freon-113, perfluorohexane and R,R,R-trifluorotoluene,
respectively. Thus, the barrier to WR of formyl and carbo-
methoxy carbene in solution must lie between 0 and 10 kJ/
mol.17 This value is smaller than that predicted (G3(MP2), 0
K) by Scott et al. for the gas-phase Wolff rearrangement of
formylcarbene (24.9 kJ/mol) and carbomethoxycarbene (25.7
kJ/mol).8,28 The best theoretical estimate of the gas-phase barrier
for formylcarbene using the W1′ method is 18.5 kJ/mol.8
According to SCI-PCM calculations the barrier for formyl-
carbene is reduced in solution by 2 (dielectric constant ) 2) to
6 (dielectric constant )40) kJ/mol in solution thus bringing the
predicted barriers into closer agreement with those estimated
in solution by the LFP experiments.
The experimental barrier to disappearance of spin-equilibrated
carbomethoxy-2-naphthylcarbene in Freon-113 (presumably by
Wolff rearrangement) is 14.2 kJ/mol.15 As the singlet-triplet
splitting of this carbene is only 0.8 kJ/mol in this solvent, nearly
all of this barrier must be on the singlet surface between carbene
and ketene.
The activation barriers for the disappearance of carbomethoxy-
chlorocarbene (6) are 5.4, 10.9, and 24.7 kJ/mol in R,R,R-
trifluorotoluene, Freon-113, and perfluorohexane, respectively.
The barrier is largest in perfluorohexane, the most inert solvent.
In all three solvents the experimental barrier for carbene
disappearance is much lower than that predicted (58.2 kJ/mol)
for isolated 6 at the G3(MP2) level, and it is unlikely that higher
level calculations or the consideration of solvent polarity will
substantially improve the agreement between theory and this
experimental result. The small barriers in R,R,R-trifluorotoluene
and Freon-113 are readily explained by reaction between the
carbene and the solvent.
Figure 6. Arrhenius treatment of the kPYR and 1/τ data of carbo-
methoxychlorocarbene in Freon-113 (top) and Arrhenius treatment of
the kPYR and 1/τ data (bottom).
We also studied the temperature dependence of the observed
rate constant for the formation of ylide 13 in Freon-113 as a
function of the pyridine. Arrhenius treatment of kpyr(T) reveals
that Ea ) 10.9 ( 0.8 kJ/mol and A ) 109.1(0.2 s-1 for the carbene
disappearance in Freon-113 between 243 and 303 K. These data
can be combined with those from the TME trapping study to
deduce EaTME ) 13.8 kJ/mol and ATME ) 1011.3 s-1 which appear
reasonable in comparison with the same parameters measured
for other carbenes.17 A plot of ln(1/τ) and ln kPYR versus 1/T in
perfluorohexane is shown in Figure 6, which yields a barrier to
carbene disappearance of 23.4 ( 1.7 kJ/mol and A ) 1010.6(0.3
s-1 in this solvent.
III. Discussion
Calculations indicate that methylene, formylcarbene, and
carbomethoxycarbene have triplet ground states.5-8 Triplet
ground state multiplicities have been established experimentally
for methylene23 and carbomethoxycarbene.10 The ground-state
multiplicity of formylcarbene has not yet been demonstrated
spectroscopically.9
B3LYP/6-31G* calculations consistently predict triplet states
of carbenes which are too stable relative to their corresponding
singlet states.8 In contrast, the G3(MP2) model successfully
reproduces ST splittings in carbenes where these values have
been experimentally determined.8 This computational method
predicts that singlet carbomethoxychlorocarbene (6) is 16.0 kJ/
mol more stable than the lowest energy triplet species. This is
consistent with the matrix IR and UV-vis spectra of 6, as well
as the low rate constant for its reaction with oxygen in solution.
Thus, 6 is the first ground-state singlet carbonylcarbene to be
studied by fast kinetic methods and, unlike ground-state triplet
carbenes, its kinetics can be analyzed without concerns related
to issues of spin interconversion.
Surely the fraction of 6 decaying by Wolff rearrangement in
perfluorohexane is greater than that in R,R,R-trifluorotoluene
and Freon-113, hence the larger barrier, but we cannot prove
that it is the sole process consuming this carbene under our
experimental conditions. It is not likely that carbene 6 will
abstract a fluorine atom from perfluorohexane. However, 5 is
only sparingly soluble in perfluorohexane and one can speculate
that the precursor is not homogeneously dispersed in this solvent,
so that photogenerated carbene 6 is born in an environment
enriched in precursor, and that part of its decay proceeds by
reaction with precursor. Thus, we can only state with confidence
that the lower limit of the barrier to Wolff rearrangement of
carbomethoxychlorocarbene is g24.7 kJ/mol.
IV. Conclusions
Methyl 8-chloro-3a,7a-methanoindan-8-carboxylate (5) was
deposited in an argon matrix at 14 K. Photolysis (254 nm) of
(27) (a) Sitzman, E. V.; Langan, J.; Eisenthal, K. B. J. Am. Chem. Soc.
1984, 106, 1868. (b) Grasse, P. B.; Braver, B.-E.; Zupancic, J. J.; Kaufmann,
K. J.; Schuster, G. B. J. Am. Chem. Soc. 1983, 105, 6833.
Formylcarbene and carbomethoxycarbene have been studied
previously by transient spectroscopy using the pyridine ylide
(28) Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502.

Carbomethoxychlorocarbene
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 6067
washed with ice cold ether (1 mL) dissolved in DMSO (1.5 mL) and
then iodomethane (0.05 mL, 0.8 mmol) was added. The mixture was
stirred for 0.5 h at room temperature, poured into ice water (15 mL),
extracted with ether (4 × 10 mL), washed with water and brine, dried
over sodium sulfate, decolorized with charcoal, filtered, and evaporated.
The yield was 110 mg (75.7%)of white solid, mp 67-69 °C (2-
this substance led to its disappearance and the formation of
carbomethoxychlorocarbene (6). B3LYP/6-31G* calculations
reveal that there are only two conformations of the nonplanar
singlet (i.e. the E- and Z-conformations of the carboxylate
group), whereas four conformations of the planar triplet carbene
are found. The matrix IR spectrum is consistent with the
presence of a mixture of singlet carbene conformations with a
carbonyl stretch at 1702 cm-1 and prominent bands at 1270,
1007, and 757 cm-1. Indane may contibute to the last band.
There is a broad weak band in the visible at 640 nm corre-
sponding to electron promotion from the HOMO to the LUMO
of the carbene. Photolysis (300 nm) of the precursor in solution
generates indane and carbomethoxychlorocarbene which can be
intercepted with cyclohexane or 2,3-dimethyl-2-butene. Laser
flash photolysis of the precursor in Freon-113 (CF2ClCFCl2)
or perfluorohexane produces carbomethoxychlorocarbene, which
can be trapped with pyridine to form an ylide. It was possible
to measure the growth of the ylide as a function of pyridine
concentration. Analysis of the data indicates that the carbene
lifetime is 114 ns in Freon-113 at ambient temperature and
decays with Ea ) 11.1 kJ/mol and A ) 109.1 s-1. This analysis
was repeated in perfluorohexane to yield Arrhenius parameters
Ea ) 24.7 ( 1.7 kJ/mol and A ) 1010.6(0.3 s-1 (τ ) 354 ns at
293 K). The activation barrier determined in perfluorohexane
is the lower limit of the barrier to Wolff rearrangement of
carbomethoxychlorocarbene.
1
propanol). H NMR (500 MHz, CDCl3, ppm) δ 6.11-6.19 (m, 2H),
5.86-5.95 (m, 2H), 3.83 (s, 3H), 2.42-2.55 (m, 2H), 1.86-2.03 (m,
2H), 1.57-1.70 (m, 1H), 0.86-1.00 (m, 1H); 13C NMR (126 MHz,
CDCl3, ppm) δ 168.14, 124.27, 122.98, 53.06, 45.15, 33.18, 29.66,
19.92; HRMS calcd for C12H13Cl35O2 224.0604, found 224.0604.
Adduct 13. Precursor 5 (100 mg) in neat 2,3-dimethyl-2-butene was
irradiated (300 nm) for 16 h after bubbling with dry argon for several
minutes. The remaining 2,3-dimethyl-2-butene was removed by rotary
evaporation and the mixture was chromatographed (silica, 5% EtOAc
in hexane). 1H NMR (500 MHz, CDCl3, ppm) δ 3.68 (s, 3H), 1.14 (s,
12H); 13C NMR (126 MHz, CDCl3, ppm) δ 169.1, 55.7, 52.8, 29.3,
19.5, 19.4; IR 1736.7 cm-1; MS(EI) m/z (rel intensity) 191(M, 1), 175
(100), 143 (35), 123 (41), 95 (55), 73 (84); HRMS(EI) calcd for
C9Cl35H15O2 (M) 190.0761, found 190.0792.
Adduct 14. Precursor 5 (200 mg) in neat cyclohexane was irradiated
(300 nm) for 16 h after bubbling with dry argon for several minutes.
The remaining cyclohexane was removed by rotary evaporation and
1
the mixture was chromatographed (silica, 5% EtOAc in hexane). H
NMR (500 MHz, CDCl3, ppm) δ 4.02 (d, 1H, J ) 7.0 Hz), 3.71 (s,
3H), 1.86 (m, 2H), 1.70 (m, 2H), 1.59 (m, 2H), 1.19 (m, 2H), 1.07 (m,
3H); 13C NMR (126 MHz, CDCl3, ppm) δ 170.4, 63.4, 53.1, 42.2, 30.3,
29.1, 26.3, 26.1, 26.0; MS(EI) m/z (rel intensity) 191 (M, 1), 155 (2),
131 (3), 108 (100), 55 (29); HRMS(EI) calcd for C9Cl35H15O2 (M)
190.0761, found 190.0717.
V. Experimental Section
Matrix Isolation Spectroscopy. The precursor was put in a glass
U tube that was directly connected to a helium closed cycle cryostat
(Air Products). Argon gas streaming over the precursor was condensed
on the surface of a CsI window held at ca. 20 K. Once formed, the
argon matrix was maintained at 14 K during the entire experiment.
UV-vis spectra were measured with a Perkin-Elmer Lambda 6 UV-
vis spectrophotometer and the IR spectra were recorded on an Perkin-
Elmer FT-IR 2000 interfrerometer at 0.2 cm-1. Ray-O-Net lamps were
used to photolyze the sample and IR and UV-vis spectra were recorded
after each step of the photolysis cycle.
General Methods. 1H NMR spectra were obtained on a Bruker
DRX-500 (500 MHz) spectrometer. IR spectra were recorded on a
Perkin-Elmer 1710 Fourier transform spectrometer interfaced with a
Perkin-Elmer 3700 data station. The GC-MS spectrometer was an HP-
6890 Series GC System with an HP-1 methyl siloxane capillary column
(40.0 m × 100 µm × 0.20 µm). The gas chromatograph was linked to
an HP 5973 mass selective detector.
Tetrahydrofuran, diethyl ether, benzene, and cyclohexane were
purified by distillation from sodium-benzophenone and stored under
an argon atmosphere. Freon-113, perfluorohexane, ethanol, cyclo-
hexane-d12, acetonitrile, and pentane were purchased from Aldrich
Chemical Co. 2,3-Dimethyl-2-butene was dried by passing through a
plug of neutral alumina prior to use. The absolute yields of adducts 13
Density Functional Calculations. Geometries of singlet and triplet
chlorocarbomethoxy carbene were fully optimized at the B3LYP level
using the 6-31G* basis set and harmonic frequencies were calculated
at the same level. Four triplet and two singlet carbene conformers were
identified. A transition state structure connecting the two singlet
conformers was found. Cs symmetry was maintained during the
optimizations for the triplet states. As higher level calculations predict
the singlet to be the ground state of 6,8 the calculated frequencies of
the singlet carbenes were employed in the analysis of IR spectra. All
calculations were carried out with the Gaussian 94 and Gaussian 98
program packages.30
1
and 14 were obtained by H NMR integration using indane (which is
derived from precursor 5) as an internal standard.
Laser Flash Photolysis Studies.29 For LFP studies of precursor 5,
a stock solution in Freon-113 or perfluorohexane was prepared with
an optical density of 0.1-0.3 and placed in a 3 mL cuvette. The LFP
apparatus utilized a Lambda Physik LPX-100 excimer laser (308 nm,
120 mL, 10 ns). LFP experiments required samples to be deoxygenated
by passing a flow of argon through the sample for 2 min. The analysis
of the data was performed with the program Igor Pro by Wavemetrics.
Transient absorption spectra were obtained on an EG&G PARC 1460
optical multichannel analyzer fitted with an EG&G PARC 1304 pulse
amplifier, EG&G PARC 1024 UV detector, and a Jarrell-Ash 1234
grating. Low-temperature LFP experiments were performed using an
NESLAB RTE-110 proportional temperature controller to regulate the
temperature.
(30) (a) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A., Jr.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Peng, C. Y.; Avala, P. Y.; Chen, W.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, Revision D.3;Gaussian,
Inc.: Pittsburgh, PA, 1995. (b) 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.;
Cliford, 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.; Stafanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Goperts, 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, Revision A.6; Gaussian, Inc.: Pittsburgh, PA 1998.
10-exo-Carbomethoxy-10-chlorotricyclo[4.3.1.01,6]decadiene-
2,4 (5). To a stirred solution of 10,10′-dichlorotricyclo[4.3.1.01,6]-
decadiene-2,4 18(130 mg, 0.65 mmol) in 4 mL of THF freshly distilled
from sodium benzophenone ketyl cooled to -78 °C was added 1.6 M
n-butyllithium solution in hexanes (0.42 mL) dropwise under argon.
The dark blue mixture was stirred at -78 °C for 2 h, then poured over
crushed dry ice, allowed to warm to room temperature, and evaporated
with a rotary evaporator. The solid lithium salt thus obtained was
(29) Gristan, N. P.; Zhai, H. B.; Yuzawa, T.; Karweik, D.; Brooke, J.;
Platz, M. S. J. Phys. Chem. A 1997, 101, 2833.

6068 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
LikhotVorik et al.
Acknowledgment. Support of this work by the NSF is
gratefully acknowledged (CHE-9613861). The authors are
indebted to Professors Berson and Wiberg of Yale University
for their generous donation of the matrix isolation equipment
used in this work. The authors gratefully acknowledge a critical
reading of this manuscript by Professor Thomas Bally.
Supporting Information Available: Experimental data
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA004235M