Mode attack of atomic carbon on benzene rings

Article abstract of DOI:10.1021/ja980710t

Reaction of are generated carbon atoms with tert-butylbenzene, 4, gives 3-methyl-3-phenyl-1-butene, 5, and 1,1-dimethylindane, 6. Labeling studies and isotope effects demonstrate that 5 results from an insertion into a methyl C-H bond followed by 1,2 hydrogen shift while 6 arises from initial ortho C-H insertion followed by intramolecular insertion into a methyl C-H bend. When fluoroboric acid is added to the 77 K matrix of 4 + C, tert-butyltropylium fluoroborate, 18, is formed. Labeling studies indicate that 18 results from initial insertion of C into meta and para C-H bonds of 4 followed by ring expansion to cycloheptatetraenes which are subsequently protonated. The reaction of C with benzene gives similar results, indicating that initial C-H insertion is the preferred mode of attack of atomic carbon on benzene rings.

Full text of DOI:10.1021/ja980710t

J. Am. Chem. Soc. 1998, 120, 6007-6011
6007
Mode of Attack of Atomic Carbon on Benzene Rings
Brian M. Armstrong, Fengmei Zheng, and Philip B. Shevlin*
Contribution from the Department of Chemistry, Auburn UniVersity,
Auburn UniVersity, Alabama 36849-5312
ReceiVed March 3, 1998
Abstract: Reaction of arc generated carbon atoms with tert-butylbenzene, 4, gives 3-methyl-3-phenyl-1-butene,
5, and 1,1-dimethylindane, 6. Labeling studies and isotope effects demonstrate that 5 results from an insertion
into a methyl C-H bond followed by 1,2 hydrogen shift while 6 arises from initial ortho C-H insertion
followed by intramolecular insertion into a methyl C-H bond. When fluoroboric acid is added to the 77 K
matrix of 4 + C, tert-butyltropylium fluoroborate, 18, is formed. Labeling studies indicate that 18 results
from initial insertion of C into meta and para C-H bonds of 4 followed by ring expansion to cycloheptatetraenes
which are subsequently protonated. The reaction of C with benzene gives similar results, indicating that initial
C-H insertion is the preferred mode of attack of atomic carbon on benzene rings.
Introduction
studied the reaction of nucleogenic 11C atoms with toluene. In
this case the methyl group traps an initially formed o-
tolylcarbene as benzocyclobutene. A partial degradation of the
benzocyclobutene and an examination of the label distribution
indicated that 43% of the benzocyclobutene arose from o-
tolylcarbene formed by an initial C-H insertion (eq 1). The
remainder of the label in the benzocyclobutene was in the ring,
indicating the initial formation of the m- and p-tolylcarbenes.
Although the reaction of atomic carbon with benzene and its
derivatives has been the subject of many investigations,1 the
site of initial attack by carbon remains an unanswered question.
It is reasonable to assume that this reaction, which provides an
entry to the C7H6 energy surface, proceeds either by a C-H
insertion to generate phenylcarbene, 1, or by double bond
addition (DBA) to give an initial bicycloheptadienylidene, 2,
which may be expected to ring open to cycloheptatetraene, 3.
Although investigations of the reaction of C with benzene have
We have recently reported several investigations in which
we have used 13C enriched arc generated carbon vapor to
elucidate the mechanisms of reaction of atomic and molecular
carbon.5,6 We now report a study of the reaction of 13C atoms
with benzene and tert-butylbenzene in which we demonstrate
that products result from a phenylcarbene formed by an initial
C-H insertion.
invariably led to products ascribable to the intermediacy of 1
and 3,1 the situation is complicated by the fact that these species
are interconvertable through the extensively investigated phen-
ylcarbene rearrangement.2,3 It is clear that isotopic labeling
studies are required to solve this problem and that a fruitful
approach will involve the use of a molecule with an available
intramolecular trap.
Results and Discussion
Such a study is reported in the elegant work of Gaspar,
Berowitz, Strongin, Svoboda, Tuchler, Ferrieri, and Wolf,4 who
Reaction of Atomic Carbon with tert-Butylbenzene. Since
the reaction of arc generated C with toluene gives an extremely
complex product mixture,7 our initial studies have focused on
the reaction of C with tert-butylbenzene, 4, which gives
3-methyl-3-phenyl-1-butene, 5, and 1,1-dimethylindane, 6, as
the only major products in a 2:1 ratio. When 13C enriched C
(1) (a) Wolf, A. P.; Redvanly, C. S.; Anderson, R. C. Nature 1955, 176,
831. (b) Schrodt, A. G.; Libby, W. F. J. Am. Chem. Soc. 1956, 78, 1267.
(c) Suryanarayana, B.; Wolf, A. P. J. Phys. Chem. 1958, 62, 1369. (d)
Sprung, J. L.; Winstein, S.; Libby, W. F. J. Am. Chem. Soc. 1965, 87, 1812.
(e) Rose, T.; MacKay, C.; Wolfgang, R. J. Am. Chem. Soc. 1967, 89, 1529.
(f) Williams, R. L.; Voigt, A. F. J. Phys. Chem. 1969, 73, 2538. (g) Lemmon,
R. M. Acc. Chem. Res. 1973, 6, 65 and references therein. (h) Biesiada, K.
A.; Koch, C. T.; Shevlin, P. B. J. Am. Chem. Soc. 1980, 102, 2098.
(2) For reviews of experimental work with many references see: (a)
Jones, W. M. In Rearrangements in Ground and Excited States; de Mayo,
P., Ed.; Academic: New York, 1980; Vol. 1. (b) Wentrup, C. In Methoden
der Organischen Chemie (Houben-Weyl); Regitz, M., Ed.; G. Thieme:
Stuttgart, 1989; Vol. E19b, pp 824-1021. (c) Gaspar, P. P.; Hsu, J.-P.;
Chari, S.; Jones, M. Tetrahedron 1985, 41, 1479-1507. (d) Platz, M. S.
Acc. Chem. Res. 1995, 28, 487-492.
(4) Gaspar, P. P.; Berowitz, D. M.; Strongin, D. R.; Svoboda, D. L.;
Tuchler, M. B.; Ferrieri, R. A; Wolf, A. P. J. Phys. Chem. 1986, 90, 4691.
(5) (a) Emanuel. C. J.; Shevlin P. B. J. Am. Chem. Soc. 1994, 116, 5991.
(b) Pan, W.; Shevlin, P. B. J. Am. Chem. Soc. 1996, 118, 10004. (c) Pan,
W.; Balci, M.; Shevlin, P. B. J. Am. Chem. Soc. 1997, 119, 5035. (d) Pan,
W.; Shevlin, P. B. J. Am. Chem. Soc. 1997, 119, 5091.
(6) For reviews of the chemistry of atomic carbon see: (a) Skell, P. S.;
Havel J. J.; McGlinchey, M. J. Acc. Chem. Res. 1973, 6, 97. (b) MacKay,
C. In Carbenes; Moss, R. A., Jones, M., Jr., Eds.; Wiley-Interscience: New
York, 1975; Vol. II, pp l-42. (c) Shevlin, P. B. in “Reactive Intermediates”
Abramovitch, R. A., Ed.; Plenum Press: New York, 1980; Vol. I, pp 1-36.
(7) Although this reaction does give benzocyclobutene, the complexity
of the reaction mixture is due in part to secondary reactions of the styrene
produced.
(3) Recent theoretical work includes: (a) Matzinger, S.; Bally, T.;
Patterson, E. V.; McMahon R. J. J. Am. Chem. Soc. 1996, 118, 1535. (b)
Wong. M. W.; Wentrup, C. J. Org. Chem. 1996, 61, 7022. (c) Schreiner,
P. R.; Karney, W. L.; Schleyer, P. v. R.; Borden, W. T.; Hamilton, T. P.;
Schaefer, H. F., III J. Org. Chem. 1996, 61, 7030. (d) Cramer, C. J.; Dulles,
F. J.; Falvey, D. E. J. Am. Chem. Soc. 1994, 116, 9787.
S0002-7863(98)00710-0 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/05/1998

6008 J. Am. Chem. Soc., Vol. 120, No. 24, 1998
Armstrong et al.
Scheme 1
The production of 6 by initial insertion into a meta or para C-H
bond to give 10 or 11 followed by phenylcarbene rearrangement
to 9 would place the label on the aromatic methyne carbons of
6. An initial DBA followed by ring expansion to 12a-c and
phenylcarbene rearrangement would have resulted in 6′, which
is not observed.
However, analogy with bicycloheptanylidene, 13, which
reacts by intramolecular C-H insertion,9 and bicyclohepten-
ylidene, 14, which undergoes a Skattebøl rearrangement,10
indicates that an initially formed tert-butylbicycloheptadi-
enylidene, 15a or 15b, could undergo intramolecular C-H
insertion to generate 6 rather than ring expand (eq 5).
If this is the case, initial reaction by carbon would not involve
the breaking of a C-H bond and little discrimination between
attack on a C-H or C-D bond would be expected. Accord-
ingly, we have reacted a mixture of 4 and 4-d14 with carbon
and have found that 5 and 6 are formed with observed kH/kD
values of 1.59 and 1.77, respectively. Although these isotope
effects are rather small, they are similar to those observed for
carbene C-H insertions11,12 and characteristic of nonlinear
transition states of the type expected here.13 To determine the
isotope effect on a C atom reaction which must involve an initial
C-H insertion, carbon was reacted with a mixture of cyclo-
hexane and cyclohexane-d12. This reaction led to the formation
of methylenecyclohexane and methylenecyclohexane-d12 with
an observed kH/kD ) 1.47 (eq 6). The observed kH/kD when C
reacts with a mixture of pyrrole and pyrrole-d5, a process known
to involve an initial DBA,5d is 1.03.14
vapor was used, an examination of the 13C NMR spectra of the
products indicated that 5 was labeled in the 1-position and 6 in
the 3-position (eq 2). It is likely that 5 results from an initial
insertion into a methyl C-H bond to give carbene 7 followed
by a 1,2-hydrogen shift (eq 3). The possibility that 6 results
from an intramolecular insertion of 7 into an ortho C-H bond
is ruled out by generating 7 by the C atom deoxygenation of
aldehyde 8 and observing that 5 is the only product (eq 4).8
These results indicate that both 5 and 6 are formed with
similar primary isotope effects which are indicative of an initial
C-H insertion and rule out the formation of 6 by the
mechanisms in eq 5. When tert-butyl-d9-benzene is reacted with
C, the ratio of 5-d9 to 6-d9 is 1.2:1 as compared to 2:1 in the
undeuterated case.
Although these experiments lead to the conclusion that 6 is
the exclusive result of an initial insertion by carbon into an ortho
C-H bond of 4, it is unreasonable to assume that carbon would
be selective in its C-H insertions and we are left to wonder
Scheme 1, which shows the anticipated label distribution in
6 resulting from all possible C-H insertions and DBAs of 13
C
(9) (a) Moore, W. R.; Ward, H. R.; Merritt, R. F. J. Am. Chem. Soc.
1961, 83, 2019. (b) Ko¨brich, G.; Goyert, W. Tetrahedron 1968, 24, 4327.
(10) Warner, P.; Chu, I.-S. J. Org. Chem. 1984, 49, 3666.
(11) Franzen, V.; Edens, R. Justus Liebigs Ann. Chem. 1969, 729, 33.
(12) Kirmse, W. Carbene Chemistry, 2nd ed.; Academic Press: New
York, 1971; p 240.
with 4, indicates that the most reasonable path to 6-3-13C is
ortho C-H insertion to give o-tert-butylphenylcarbene, 9, which
undergoes intramolecular insertion into a methyl C-H bond.
(8) Carbon atom deoxygenation of carbonyl compounds is a well
documented route to carbenes. See: Armstrong, B. M.; McKee, M. L.;
Shevlin, P. B. J. Am. Chem. Soc. 1995, 117, 3685 and references therein.
(13) More O’Ferrall, R. A. J. Chem. Soc. B 1970, 785.
(14) Pan, W. Ph.D. Thesis, Auburn University, December 1997.

Mode of Attack of Atomic Carbon on Benzene Rings
J. Am. Chem. Soc., Vol. 120, No. 24, 1998 6009
about the fate of the products of insertion into meta and para
C-H bonds, 10 and 11. The labeling distribution in 6 tells us
that 10 and 11 do not undergo phenylcarbene rearrangement to
9. If 10 and 11 lack the energy to undergo any intramolecular
reaction, we may expect decay to triplet ground states and
subsequent hydrogen abstraction. However, examination of the
reaction products does not show any that could be attributed to
H abstraction by 10 and 11.
The results of these experiments indicate that atomic carbon
reacts with benzene rings to generate phenylcarbenes by C-H
insertion. Since this reaction is exothermic by ∼115 kcal/mol,16
there is enough energy in the product to overcome the ∼13
kcal/mol required for rearrangement to the more stable cyclo-
heptatetraenes.3 However, in the reaction of C with 4, it appears
that the tert-butyl substituent on the cycloheptatetraene (16a,
16b, or 17) lends sufficient degrees of freedom to allow energy
dissipation such that the molecule is not able to surmount the
∼30 kcal/mol barrier for rearrangement back to an isomeric
phenylcarbene. We are currently examining the effect of smaller
benzene ring subtitutents on the products of this reaction.19 The
fact that rearrangement of 9 to 19a,b is not observed indicates
that the barrier to intramolecular C-H insertion to give 6 is
lower than that for ring expansion to 19a,b.
To test our postulate that the exothermicity of the carbon atom
reaction is sufficient to bring about the ring expansion of 1 to
3, we have generated 1 by the C atom deoxygenation of
benzaldehyde at 77 K. Since this reaction is estimated to be
exothermic by ∼116 kcal/mol, we again expect rearrangement
to 3 under the reaction conditions. This expectation is confirmed
by adding HBF4 to the 77 K C + benzene matrix and observing
the formation of 20-d0 (eq 9).
Recent computational results indicate that the first step of
the phenylcarbene rearrangement, ring expansion of 1, must
surmount a barrier of ∼13 kcal/mol to give 3.3 However,
rearrangement back to 1 (in this case interconverting 9, 10, and
11 as shown in Scheme 1) must cross a barrier of ∼30 kcal/
mol, and it may be that initial ring expansion of 10 and 11 leads
to cycloheptatetraenes 16a,b and 17 which, lacking the energy
to rearrange further, are polymerized during workup. In support
of this assumption, we find that addition of an ethereal solution
of HBF4 to the 77 K reactor bottom results in the formation of
tert-butyltropylium fluoroborate, 18, along with 5 and 6 (eq 7).
It is likely that 18 results from protonation of an extremely
reactive cycloheptatetraene. When the HBF4 is added to the
reactor after warming to room temperature, no 18 is formed.
Addition of HBF4 to the cold products of the reaction of 13C
enriched carbon vapor with 4 results in 18a and 18b which
contain excess 13C in the 3 and 4 positions in a 1:2 ratio and no
excess 13C in the 2 position (Scheme 1). This result indicates
that 10 and 11 ring expand to 16a,b and 17, respectively. Since
10 can ring expand to ultimately yield 18 labeled in the 3 and
4 positions while 11 gives only 4-labeled 18, statistical insertion
of C into the ortho and para positions of 4 will yield the observed
label distribution in 18. The fact that the 2 position in 18 is
unlabeled and that the 3 and 4 positions are labeled in a 1:2
ratio indicates that 9 does not ring expand to 19a,b. Instead, 9
is trapped by intramolecular C-H insertion to give 6.
1
3
Summary. The examples presented here along with previous
work4 demonstrate that carbon atoms react with benzene rings
by C-H insertion rather than by DBA. These results are in
contrast to the reaction of C atoms with pyrrole in which DBA
is the exclusive process5a,d and with thiophene which shows a
mixture of C-H insertion, DBA, and attack at sulfur.5c While
the reason for the preference for C-H insertion by carbon in
its reaction with benzene rings must await a detailed examination
of the potential energy surface by sophisticated computational
techniques, we note that a thermodynamic preference for the
formation of 1 over the highly strained bicycloheptadienylidene,
2, is expected and it is possible that these thermodynamics are
reflected in the relative rates of formation of the two species.
These reactions, which provide alternate routes to the C7H6
energy surface, may facilitate the elucidation of additional
chemistry of the elusive cycloheptatetraene and its derivatives.
When a double labeling experiment is carried out by adding
DCl to the cold matrix of 4 + 13C (eq 8), the resultant tropylium
salt shows a deuterium signal in the 2H NMR spectrum but no
evidence of deuterium attachment to the labeled carbons in the
13C spectrum.15 This result rules out an initial DBA to form
Experimental Section
General Information. All 1H and 13C NMR were recorded on
Bruker AM 250 or Bruker AM 400 spectrometers. Infrared spectra
were obtained with an IR-44 IBM FT-IR. GC analysis were carried
out on a Shimadzu GC 14-A instrument with a 30 m × 0.25 mm o.d.
DB-5 column. The GC/MS were recorded on a Fisons Trio 2000
12b and 12c which would be subsequently deuterated on the
labeled carbon.
Reaction of C with Benzene. To confirm the generality of
this reaction, we have reacted 13C atoms with benzene-d6 and
added HBF4 to the cold matrix generating tropylium fluorobo-
rate, 20. The 13C NMR spectrum of 20 shows only a triplet
(J ) 25.5 Hz) due to C-D coupling.
(16) A ∆Hf of 103 kcal/mol17 and a singlet-triplet splitting of 2.3 kcal/
mol18 for 1 were used to determine this value.
(17) Poutsma, J. C.; Nash, J. J.; Paulino, J. A.; Squires, R. R. J. Am.
Chem. Soc. 1997, 119, 4686.
(15) In a typical reaction, the enriched C vapor contains approximately
6% 13C. Thus we estimate that the ratio of 13CH:13CD at C3 ) 4.5 and at
C4 ) 6.45. Since the 13CD signal is split into a triplet, further decreasing
the relative intensities, the triplet for the13CD signal, resulting from the
natural abundance of 13C in 4, is not observable.
(18) Admasu, A.; Gudmundsdo´ttir, A. D.; Platz, M. S. J. Phys. Chem.
A 1997, 101, 3832.
(19) We have reported interconversion of the tolylcarbenes resulting from
tolylaldehyde deoxygenation by atomic carbon. Rahman, M.; Shevlin, P.
B. Tetrahedron Lett. 1985, 26, 2959.

6010 J. Am. Chem. Soc., Vol. 120, No. 24, 1998
Armstrong et al.
quadrupole spectrometer interfaced with a Hewlett-Packard Series II
5890 gas chromatograph equipped with 30 m × 0.25 mm o.d. DB-5
column. tert-Butylbenzene, tropylium hexachlorophosphate, trityl
fluoroborate, ethereal fluoroboric acid, and tert-butyl chloride were used
as received from Aldrich. Benzene-d6, cyclohexane-d12, and tert-butyl
chloride-d9 were used as received from Cambridge Isotopes. 3-Methyl-
3-phenyl-1-butene, 5,20 1,1-dimethylindane, 6,21 and tert-butyltropylium
fluoroborate, 18,22 were prepared by literature methods.
(75 mmol) and the products isolated as described above. The products
were analyzed by GC and GC/MS. In all cases the d14 and h14 substrates
and products were >95% of the isotopomers as determined by MS
and could be separated by GC. 3-Methyl-3-phenyl-1-butene-d0 and
-d14 were formed in a 1.59:1 ratio. 1,1-Dimethylindane-d0 and -d14
were formed in a 1.77:1 ratio. 3-Methyl-3-phenyl-1-butene-d0 and 1,1-
dimethylindane-d0 were formed in 2:1 ratios. 3-Methyl-3-phenyl-1-
butene-d14 and 1,1-dimethylindane-d14 were formed in 2.2:1 ratios.
Reaction of Atomic Carbon with tert-Butylbenzene. The carbon
arc reactor was modeled after that described by Skell, Wescott, Golstein,
and Engel.23 The quantities of carbon given below are calculated from
the loss of weight of the graphite. Since most of the carbon is lost in
macroscopic pieces, yields based on carbon are not meaningful. Carbon
(125 mmol) was vaporized by striking an intermittent arc between two
graphite rods attached to water-cooled brass electrodes and condensed
on the 77 K walls of the reactor with tert-butylbenzene (11.2 mmol) at
5 × 10-5 Torr. When the reaction was over, the reactor bottom was
extracted with methylene chloride and the extract filtered and concen-
trated by rotary evaporation. The resulting products were identified
by GC and GC/MS: 3-methyl-3-phenyl-1-butene, 5, and 1,1-dimeth-
ylindane, 6, were formed in a 2.0:1.0 ratio. The yield of 5 was 3.2 ×
10-2 mmol.
Preparation of tert-Butylbenzene-d9. Benzene (0.084 mol), an-
hydrous aluminum chloride (0.012 mol), and tert-butyl-d9 chloride (0.02
mol) were reacted as described above. The fraction which distilled at
164-168 °C was collected as tert-butylbenzene-d9 (2.1 g, 73.0%): MS
m/e (rel intensity) 143 (11), 142 (25), 141 (19), 140 (41), 139 (23),
138 (16), 137 (11), 127 (3), 126 (26), 125 (72), 124 (91), 123 (100),
122 (53), 121 (35) 95 (16), 94 (56), 93 (93), 92 (76), 91 (11), 78 (26),
77 (17), 51 (20), 45 (14), 44 (14), 43 (8), 42 (12), 41 (12), 40 (10), 39
(8).
Reaction of Atomic Carbon with tert-Butylbenzene-d9. tert-
Butylbenzene-d9 (6.5 mmol) was reacted with atomic carbon (91 mmol)
and the products isolated as described above. The products were
analyzed by GC and GC/MS. 3-Methyl-3-phenyl-1-butene-d9 and
1,1-dimethylindane-d9 were formed in a 1.2:1 ratio and were >95% of
the isotopomers.
Reaction of 13C-Enriched Carbon Vapor with tert-Butylbenzene.
The normal graphite rods which were used in these experiments were
Reaction of Atomic Carbon with a 1:1 Ratio of Cyclohexane-d12
and Cyclohexane-d0. An equimolar mixture of cyclohexane-d0 and
-d12 (11.9 mmol) was condensed with atomic carbon (134 mmol) and
the products isolated as described above. The products were analyzed
by GC and GC/MS. The d12 and d12 substrates and products could be
separated by GC. Methylenecyclohexane-d0 and -d12 were formed in
a 1.47:1 ratio and were >95% of the isotopomers.
Reaction of Atomic Carbon with tert-Butylbenzene followed by
the Addition of HBF4. tert-Butylbenzene (6.5 mmol) was reacted with
atomic carbon (91 mmol) as described above. Before the reactor was
allowed to warm, 1 mL of an ethereal solution of HBF4 (7.3 mmol)
was distilled in. The reactor was allowed to warm to room temperature
and extracted with acetonitrile. The extract was filtered, the volatiles
evaporated, and the residue taken up in 0.5 mL of D2O or CD3CN,
and the resultant tert-butyltropylium fluoroborate, 18 (the only com-
pound detectable by 13C NMR) was analyzed by NMR: 1H NMR
(δ, 400 MHz, CD3CN) δ 1.58 (s, CMe3), 9.05 (m, H3), 9.08 (m, H4),
9.35 (m, H2). 13C NMR (δ, 100 MHz, CD3CN) 31.45 (CH3), 153.99
(C2), 154.19 (C4), 154.44 (C3), 184.14 (C1). Protons were assigned
from the proton-proton correlation spectrum and the carbons from the
carbon-proton correlation spectrum.
1
altered by drilling a /8 in. hole into each individual rod to a depth of
0.75 in. The resulting holes were packed with amorphous 13C powder
(0.18 g). During this tedious process, small amounts of the powder
were added to the hole by a specially made funnel. After each small
quantity of powder was added to the hole in the rod, it was packed
firmly with a small dowel. This process was repeated until both
graphite rods were tightly filled with 13C powder. These rods were
installed on brass electrodes and the reactor was assembled and
evacuated as usual. After the reactor had reached its ultimate vacuum,
the two graphite rods were firmly placed head to head (packed hole
against packed hole) and the power was turned on. The two rods started
to glow red and absorbed gases were evolved. Power was turned off
for brief periods to let the rods cool. The heating of the rods to degas
the packed amorphous carbon-13 was carried out until no more gas
was evolved.
tert-Butylbenzene (11.2 mmol) was condensed on the walls of the
reactor bottom at 77 K with 13C enriched atomic carbon (65.3 mmol)
and the products isolated as above and identified by 1H and 13C NMR,
GC, and GC/MS. 3-Methyl-3-phenyl-1-butene, 5, and 1,1-dimethylin-
dane, 6, were formed in a 2:1 ratio. A comparison of the intensities of
the 13C NMR signals of 5 and 6 with those of 5 and 6 from the reaction
of unlabeled carbon with 4 (taken under identical NMR conditions)
revealed that the signal for C1 had increased by a factor of 11.0 in 5
and the signal for C3 in 6 had increased by a factor of 10.0. No other
enrichments were observed.
Reaction of 13C-Enriched Carbon Vapor with tert-Butylbenzene
followed by the Addition of HBF4. tert-Butylbenzene (11.2 mmol)
was condensed on the walls of the reactor bottom at 77 K with 13C
enriched atomic carbon (36.7 mmol) and 18 isolated as above and
identified by 1H and 13C NMR. Compound 18 was the only compound
detectable by 13C NMR. 13C enrichments were calculated assuming
Preparation of tert-Butylbenzene-d14. Benzene-d6 (0.084 mol) was
added to anhydrous aluminum chloride (0.012 mol). The flask was
kept at 0 °C with an ice bath and all inlets were protected by drying
tubes.
that no 13C was incorporated into the methyl groups of 18.
A
comparison of the intensity ratios of the ring carbons to the methyl
carbons of 18 with those of 18 from the reaction of unlabeled carbon
with 4 (taken under identical NMR conditions) revealed a 36.3%
increase at C3 and a 66.2% increase at C4 and no increase at C1 or C2.
tert-Butyl-d9 chloride (0.02 mol) was added dropwise over 3 h and
the mixture was allowed to stir for another hour. D2O (5 mL) was
added to the flask and the mixture allowed to stir for 10 min. The
mixture was extracted with ether and dried with magnesium sulfate.
The fraction which distilled at 163-166 °C was collected as tert-
butylbenzene-d14 (2.0 g, 68.0%): MS m/e (rel intensity) 149 (5), 148
(100), 131 (31), 130 (97), 129 (15), 110 (26), 99 (14), 98 (97), 84
(19), 82 (48), 70 (21), 66 (16), 54 (33), 52 (12), 46 (41), 42 (29).
Reaction of 13C-Enriched Carbon Vapor with tert-Butylbenzene
followed by the Addition of DCl. tert-Butylbenzene (11.2 mmol) was
condensed on the walls of the reactor bottom at 77 K with 13C enriched
atomic carbon (30.0 mmol), gaseous DCl (0.5 mmol) was added before
warming. After warming, an ethereal solution of HBF4 (7.3 mmol)
was added and 18 was isolated as above. The 2H NMR spectrum
showed a peak at δ 9.2 and the 13C NMR showed only singlets for the
ring carbons. A comparison of the intensity ratios of the ring carbons
to the methyl carbons of 18 with those of 18 from the reaction of
unlabeled carbon with 4 (taken under identical NMR conditions)
revealed a 77.6% increase at C3 and a 168.2% increase at C4 and no
increase at C1 or C2.
Reaction of Atomic Carbon with a 1:1 Mixture of tert-Butyl-
benzene-d14 and tert-Butylbenzene-d0. An equimolar mixture of tert-
butylbenzene-d0 and -d14 (6.6 mmol) was condensed with atomic carbon
(20) Abramovici, M.; Pines, H. J. Org. Chem. 1969, 34, 266.
(21) Bright, S. T.; Coxon, J. M.; Steel, P. J. J. Org. Chem. 1990, 55,
1338.
(22) Heyd, W. E.; Cupas, C. A. J. Am. Chem. Soc. 1971, 93, 6086.
(23) Skell, P. S.; Wescott, L. D., Jr.; Golstein, J. P.; Engel, R. R. J. Am.
Chem. Soc. 1965, 87, 2829.
Reaction of 13C-Enriched Carbon Vapor with Benzene-d6 fol-
lowed by the Addition of HBF4. The reaction was carried out as

Mode of Attack of Atomic Carbon on Benzene Rings
J. Am. Chem. Soc., Vol. 120, No. 24, 1998 6011
described above by co-condensing carbon (35.0 mmol) with benzene-
d6 (11.2 mmol) and adding HBF4 (7.3 mmol). Tropylium fluoroborate-
d6 was isolated by using the methods described above. The 13C NMR
spectrum showed a triplet centered at δ ) 155.36 ppm with a splitting
of 25.5 Hz.
Reaction of Atomic Carbon with Benzaldehyde followed by the
Addition of HBF4. Benzaldehyde (18.0 mmol) was reacted with
atomic carbon (52 mmol) as described above. Before the reactor was
allowed to warm, 1 mL of an ethereal solution of HBF4 (7.3 mmol)
was distilled in. The reactor was allowed to warm to room temperature
and extracted with acetonitrile. The extract was filtered, the solvent
evaporated, and the residue taken up in CD3CN, and the resultant
tropylium fluoroborate was analyzed by 1H NMR (δ, 250 MHz, CD3-
CN): δ 9.2 (s).
Acknowledgment. Support of this work through National
Science Foundation Grant No. CHE-9508570 is gratefully
acknowledged. We thank Dr. Partha Sarathy for assistance in
assigning the NMR signals of 18.
JA980710T