Molecule-induced alkane homolysis with dioxiranes

Article abstract of DOI:10.1021/ja0158096

The mechanisms of C-H and C-C bond activations with dimethyldioxirane (DMD) were studied experimentally and computationally at the B3LYP/6-311+G**//B3LYP/6-31G* density functional theory level for the propellanes 3,6-dehydrohomoadamantane (2) and 1,3-dehydroadamantane (3). The σ_C-C activation of 3 with DMD (ΔG? = 23.9 kcal mol^-1 and ΔG_r = -5.4 kcal mol^-1) is the first example of a molecule-induced homolytic C-C bond cleavage. The C-H bond hydroxylation observed for 2 is highly exergonic (ΔG_r = -74.4 kcal mol^-1) and follows a concerted pathway (ΔG_r = 34.8 kcal mol^-1), in contrast to its endergonic molecule-induced homolysis (ΔG? = 28.8 kcal mol^-1 and ΔG_r = +9.2 kcal mol^-1). The reactivities of 2 and 3 with CRO₂Cl₂, which follow a molecule-induced homolytic activation mechanism, parallel the DMD results only for highly reactive 3, but differ considerably for more stable propellanes such as 4-phenyl-3,6-dehydrohomoadamantane (1) and 2.

Full text of DOI:10.1021/ja0158096

11248
J. Am. Chem. Soc. 2001, 123, 11248-11252
Molecule-Induced Alkane Homolysis with Dioxiranes
,
†
†
†
Andrey A. Fokin,* Boryslav A. Tkachenko, Oleg I. Korshunov,
†
,‡
Pavel A. Gunchenko, and Peter R. Schreiner*
Contribution from the Department of Organic Chemistry, KieV Polytechnic Institute, 37 Pobeda AVenue,
0
3056 KieV, Ukraine, and Department of Chemistry, UniVersity of Georgia, Athens, Georgia 30602
ReceiVed March 13, 2001. ReVised Manuscript ReceiVed August 30, 2001
Abstract: The mechanisms of C-H and C-C bond activations with dimethyldioxirane (DMD) were studied
experimentally and computationally at the B3LYP/6-311+G**//B3LYP/6-31G* density functional theory level
for the propellanes 3,6-dehydrohomoadamantane (2) and 1,3-dehydroadamantane (3). The σC-C activation of
q
-1
-1
3
with DMD (∆G ) 23.9 kcal mol and ∆Gr ) -5.4 kcal mol ) is the first example of a molecule-induced
homolytic C-C bond cleavage. The C-H bond hydroxylation observed for 2 is highly exergonic (∆Gr )
-
1
q
-1
-
74.4 kcal mol ) and follows a concerted pathway (∆G ) 34.8 kcal mol ), in contrast to its endergonic
q
-1
-1
molecule-induced homolysis (∆G ) 28.8 kcal mol and ∆Gr ) +9.2 kcal mol ). The reactivities of 2 and
with CrO2Cl2, which follow a molecule-induced homolytic activation mechanism, parallel the DMD results
3
only for highly reactive 3, but differ considerably for more stable propellanes such as 4-phenyl-3,6-dehydro-
homoadamantane (1) and 2.
Introduction
Scheme 1. Bifurcation of the Reaction Pathway in the C-H
Bond Activation with DMD
Highly reactive three-membered ring peroxides, i.e., diox-
iranes, are able to hydroxylate unactivated C-H bonds in
aliphatic hydrocarbons under mild conditions1 with remark-
ably high tertiary/secondary positional selectivities, high stereo-
selectivities, and in good preparative yields.4 Relatively stable
dimethyldioxirane (DMD) is one of the key reagents for alkane
oxidations which can easily be prepared from acetone and
-3
-9
[
(KHSO5)2‚KHSO4‚K2SO4].10 Despite a number of efforts, the
mechanism for the C-H activation step is not entirely clear,
as there is evidence for both radical1
1-13
14-17
reaction path (Scheme 1) which involves either concerted
and nonradical
1
6
insertion (A) giving hydroxy products directly, or a radical
pathways. A number of puzzling and controversial experimental
results first were rationalized18 in terms of a bifurcation of the
pair (B) via a common transition structure: The bifurcation point
-
1
is located about 2 kcal mol below and after the transition
†
Kiev Polytechnic Institute.
University of Georgia.
structure. Radical pairs thus formed either recombine (rebound)11
‡
(
11.
1) Adam, W.; Curci, R.; Edwards, J. O. Acc. Chem. Res. 1989, 22, 205-
or escape from the solvent cage, giving rise to free radicals
2
12
(molecule-induced homolysis).
(
(
22.
2) Murray, R. W. Chem. ReV. 1989, 89, 1187-1201.
3) Curci, R.; Dinoi, A.; Rubino, M. F. Pure Appl. Chem. 1995, 67, 811-
The mechanistic dichotomy in the reactions of DMD with
1
9
alkanes also may arise from the potential multistate reactivity
8
5
1
1
6
1
(
4) Teager, D. S.; Murray, R. K., Jr. J. Org. Chem. 1993, 58, 5548-
of DMD. The closed-shell A1 singlet ground state is only 11.5
kcal mol more stable than the biradical open-shell OS- A1
550.
-1
1
(5) Mello, R.; Fiorentino, M.; Fusco, C.; Curci, R. J. Am. Chem. Soc.
singlet state at B3LYP/6-311+G**//B3LYP/6-31G* and 11.1
989, 111, 6749-6757.
-
1
20
(6) Murray, R. W.; Jeyaraman, R.; Mohan, L. J. Am. Chem. Soc. 1986,
kcal mol at CCSD(T)/VTZ2P+f,d. Computations show that
three distinctly different pathways on the singlet PES could take
place, as shown in Scheme 2 for the reaction of DMD with
08, 2470-2472.
7) Fusco, C.; Fiorentino, M.; Dinoi, A.; Curci, R. J. Org. Chem. 1996,
1, 8681-8684.
8) Mello, R.; Cassidei, L.; Fiorentino, M.; Fusco, C.; Curci, R.
Tetrahedron Lett. 1990, 31, 3067-3070.
9) Adam, W.; Asensio, G.; Curci, R.; Gonzalez-Nu n˜ ez, M. E.; Mello,
R. J. Org. Chem. 1992, 57, 953-955.
10) Curci, R.; Fiorentino, M.; Troisi, L.; Edwards, J. O.; Pater, R. H. J.
Org. Chem. 1980, 45, 4758-4760.
11) Minisci, F.; Zhao, L.; Fontana, F.; Bravo, A. Tetrahedron Lett. 1995,
6, 1697-1700.
12) Bravo, A.; Fontana, F.; Fronza, G.; Minisci, F.; Zhao, L. J. Org.
Chem. 1998, 63, 254-263.
13) Vanni, R.; Garden, S. J.; Banks, J. T.; Ingold, K. U. Tetrahedron
Lett. 1995, 36, 7999-8002.
14) Asensio, G.; Mello, R.; Gonzalez-Nu n˜ ez, M. E.; Boix, C.; Royo, J.
Tetrahedron Lett. 1997, 38, 2373-2376.
15) Curci, R.; Dinoi, A.; Fusco, C.; Lillo, M. A. Tetrahedron Lett. 1996,
7, 249-252.
(
1
(
isobutane. Insertion of A1-DMD into a C-H bond occurs in a
concerted fashion (TS1) as found by a number of groups.2
1-23
(
(16) Adam, W.; Curci, R.; D’Accolti, L.; Dinoi, A.; Fusco, C.; Gasparrini,
F.; Kluge, R.; Paredes, R.; Schulz, M.; Smerz, A. K.; Veloza, L. A.;
Weink o¨ tz, S.; Winde, R. Chem. Eur. J. 1997, 3, 105-109.
(17) Simakov, P. A.; Choi, S.-Y.; Newcomb, M. Tetrahedron Lett. 1998,
39, 8187-8190.
(
(
3
(
(18) Shustov, G. V.; Rauk, A. J. Org. Chem. 1998, 63, 5413-5422.
(19) Shaik, S.; Filatov, M.; Schr o¨ der, D.; Schwarz, H. Chem. Eur. J.
1998, 4, 193-199.
(
(20) Cremer, D.; Kraka, E.; Szalay, P. G. Chem. Phys. Lett. 1998, 292,
97-109.
(
(21) Glukhovtsev, M. N.; Canepa, C.; Bach, R. D. J. Am. Chem. Soc.
1998, 120, 10528-10533.
(
3
(22) Du, X.; Houk, K. N. J. Org. Chem. 1998, 63, 6480-6483.
1
0.1021/ja0158096 CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/17/2001

Molecule-Induced Alkane Homolysis with Dioxiranes
J. Am. Chem. Soc., Vol. 123, No. 45, 2001 11249
Scheme 2. Possible Reaction Pathways for the Isobutane/DMD System Computed at B3LYP/6-311+G**//B3LYP/6-31G* (∆E,
-
1 a
kcal mol )
a
These data are compiled from our own computations and from the respective literature.21,23
The barrier for this reaction at B3LYP/6-311+G**//B3LYP/6-
question of concerted vs molecule-induced homolytic/rebound
mechanism unanswered.
-1
3
1G* is 24.2 kcal mol . The H-abstraction via linear TS2 with
formation of the radical pair exemplifies the molecule-induced
It should be emphasized that the mechanisms giving open-
shell species (i.e., molecule-induced homolysis) from closed-
shell reactants are rare. This scenario is, however, typical for
23
homolytic pathway. While it is accompanied by a lower barrier
-1
26
(
20.0 kcal mol ), this reaction is endothermic; this low-barrier
reaction could, however, be driven by the exothermic radical
recombination or other transformations of the two incipient
alkane activations utilizing metal-oxo reagents where two
radicals can form from the reaction of two closed-shell
1
27-30
products. The barrier for the radical reaction with OS- A1-DMD
molecules. Concerted insertion
vs hydrogen abstraction/
-
1
31-33
as well as molecule-induced homolytic31,34 mech-
is even lower (TS3, 15.8 kcal mol ) and may also lead to
rebound
radical pair formation, although this pathway is hampered by
the high barrier for DMD homolysis (OS- A1-DMD is the
anisms are discussed vividly; the reactivity of CrO2Cl2 is most
1
34-38
well studied,
and a molecule-induced homolytic mechanism
36,37
product of this process); the O-O bond breaking requires an
was unequivocally shown for the C-H activation step.
-
1
20
activation of 23.0 kcal mol at CCSD(T)/cc-VTZ2P+f,d, and
Although there are some obvious common experimental
findings in alkane activations with dioxiranes and metal-oxo
species (high tertiary selectivities, stereoselectivities, and sen-
sitivity to radical traps), this connection has been made only
-
1
20
2
3.1 kcal mol at B3LYP/6-31G**. H-abstraction can also
occur on the ground-state triplet surface: The A2-DMD
biradical lies 5.5 kcal mol above OS- A1-DMD, and H-
abstraction with A2-DMD via TS4 forming the radical pair is
3
-
1
1
3
32
vaguely, and there are no decisive mechanistic comparisons.
virtually barrierless.
This is mostly due to the lack of relevant model hydrocarbons
that are able to distinguish the mechanistic steps cleanly. In the
present paper we describe the preparative transformations of
three well-chosen strained propellanes (1-3) which allow us
Dioxiranes are also characterized by a number of low-lying
2
4
exited states which may participate in the C-H activation
reactions; these channels are supported by recent observation
of weak chemiluminescence in the reaction of certain dioxiranes
with hydrocarbons (for instance, with the significantly more
reactive methyl(trifluoromethyl)dioxirane, TDA, reacting with
(
26) R u¨ chardt, C.; Gerst, M.; Ebenhoch, J. Angew. Chem., Int. Ed. Engl.
1
997, 36, 1407-1430.
(27) Lee, D. G.; Chen, T. J. Am. Chem. Soc. 1993, 115, 11231-11236.
(28) Filatov, M.; Shaik, S. J. Phys. Chem. 1998, 102, 3835-3846.
2
5
adamantane). Thus, the activation of C-H bonds with DMD
is indeed very complex: traces of radical starters may cause
free-radical processes, and depending on the reaction media, as
well as the structure of the hydrocarbon, different reaction
pathways may be followed. Hypersensitive radical clocks with
rearrangement times close to the lifetime of the transition
(29) Newcomb, M.; Le Tadic, M.-H.; Putt, D. A.; Hollenberg, P. F. J.
Am. Chem. Soc. 1995, 117, 3312-3313.
(30) Newcomb, M.; Le Tadic-Biadatti, M. H.; Chestney, D. L.; Roberts,
E. S.; Hollenberg, P. F. J. Am. Chem. Soc. 1995, 117, 12085-12091.
(
31) Gardner, K. A.; Kuehnert, L. L.; Mayer, J. M. Inorg. Chem. 1997,
3
6, 2069-2078.
(32) Strassner, T.; Houk, K. N. J. Am. Chem. Soc. 2000, 122, 7821-7822.
(33) Wang, K.; Mayer, J. M. J. Org. Chem. 1997, 62, 4248-4252.
1
6,17
states
do not support free-radical pathways but leave the
(
34) Mayer, J. M. Acc. Chem. Res. 1998, 31, 441-450.
(
23) Freccero, M.; Gandolfi, R.; Sarzi-Amad e` , M.; Rastelli, A. Tetra-
hedron Lett. 2001, 42, 2739-2742.
24) Anglada, J. M.; Bofill, J. M.; Olivella, S.; Sol e´ , A. J. Phys. Chem.
A 1998, 102, 3398-3406.
25) Kazakov, D. V.; Barzilova, A. B.; Kazakov, V. P. Chem. Commun.
001, 191-192.
(35) Limberg, C. Chem. Eur. J. 2000, 6, 2083-2089.
(36) Cook, G. K.; Mayer, J. M. J. Am. Chem. Soc. 1994, 116, 1855-
(
1868.
(37) Cook, G. K.; Mayer, J. M. J. Am. Chem. Soc. 1995, 117, 7139-
7156.
(38) Shaik, S.; Shurki, A. Angew. Chem., Int. Ed. 1999, 38, 587-625.
(
2

11250 J. Am. Chem. Soc., Vol. 123, No. 45, 2001
Fokin et al.
Scheme 3. Experimental Results on the Reactions of
Propellanes 1-3 with DMD and CrO2Cl2
to study the competition between the C-H and the C-C bond
activations in the reactions with DMD and CrO2Cl2 and verify
the differences in their reactivities. These investigations also
reveal important mechanistic aspects of the reactions of ali-
phatics with uncharged closed-shell electrophiles, in particular
with DMD.
Figure 1. B3LYP/6-31G* optimized geometries for the transition
structures TS5 and TS6 (selected bond distances in angstroms).
relative to pure aliphatics.53 The cage isomerization of 1 upon
treatment with CrO2Cl2 is very similar to the rearrangement of
4
-substituted 3,6-dehydrohomoadamantanes in its reactions with
43
Results and Discussion
electrophiles/oxidizers. The oxidation of 3 is the first example
of an alkane σC-C bond activation with dioxirane. Nonetheless,
this is not observed for 2, although both hydrocarbons are highly
strained (vide infra). To elucidate the mechanistic details and
the nature of such apparent reactivity differences, we studied
certain aspects of these reactions computationally (Table 1,
Supporting Information; see Computational Methods for details)
at the B3LYP/6-311+G**//B3LYP/6-31G* level of theory
As the ring size controls their strain and thus their reactivity,
propellanes3 (in particular, 1-3 ) have proven to be relevant
9
40,41
and highly valuable substrates in our4
2-48
and other
49,50
groups
for modeling alkane activation mechanisms. Our results for the
reactions of 1-3 with DMD and CrO2Cl2 are summarized in
Scheme 3. These reactions were run at low conversions to avoid
side reactions (see Experimental Section for details).
(Scheme 4).
In the reactions of 1 and 244 with CrO2Cl2, we only found
C-C bond addition products (4, 5, 7, and 8), while the
transformations with DMD exclusively give C-H substitution
products (6 and 9). At the same time, 3 forms only C-C addition
products with both DMD and CrO2Cl2. Dihydroxy derivative
Transition structures TS5 and TS6 (Figure 1) for the C-H
activation of 2 and 3, respectively, describe the insertion of the
q
oxygen into the C-H bonds; the ∆G298 values are similar (34.8
-
1
and 30.5 kcal mol ) for both hydrocarbons. These highly
-
1
exergonic (∆G298 ) -74.4 and -75.2 kcal mol ) insertions
1
0 is produced with both DMD and CrO2Cl2, while cage
occur as one-step concerted processes terminating with alcohols
fragmentation product 11 is isolated from the reaction of 3 with
DMD; conversely, chlorohydroxy product 12 forms with CrO2-
Cl2. The cage fragmentation is in marked contrast to the addition
reactions previously found for the reactions of 3 with a number
9
and 13 after acetone elimination, respectively. Although the
barrier for H-abstraction with DMD on the open-shell singlet
-
1
manifold is lower (28.7 kcal mol ) than for the concerted
51
insertion, dissociation into the radical 14 is endergonic (9.2 kcal
of radical reagents such as halogens, halomethanes, and
-
1
oxygen;52 radical oligomerizations dominate in the latter case.
mol ) and cannot effectively compete with the concerted
hydroxylations. However, this does not rule out fast radical
recombination which may be exothermic. The activation enthal-
The reactivity difference of 1 and 2 toward DMD is not
surprising because the benzylic hydrogen is more reactive
q
pies for the concerted insertions via TS5 and TS6 (∆H ) 27.4
(
39) Wiberg, K. B. Chem. ReV. 1989, 89, 975-983.
-1
and 22.3 kcal mol , respectively) are very close to those com-
(40) Krasutsky, P. A.; Rodionov, V. N.; Chesskaya, N. S.; Yurchenko,
2
1,23
puted previously
for the activation of the tertiary C-H bond
A. G. Zh. Org. Khim. 1985, 21, 1684-1691.
41) Yurchenko, A. G.; Voroshenko, A. T.; Stepanov, F. N. Zh. Org.
Khim. 1970, 5, 188-193.
42) Fokin, A. A.; Gunchenko, P. A.; Yaroshinsky, A. I.; Krasutsky, P.
A.; Yurchenko, A. G. Tetrahedron Lett. 1995, 4479-4482.
43) Fokin, A. A.; Gunchenko, P. A.; Kulik, N. I.; Iksanova, S. V.;
Krasutsky, P. A.; Gogoman, I. V.; Yurchenko, A. G. Tetrahedron 1996,
857-5866.
44) Fokin, A. A.; Gunchenko, P. A.; Tkachenko, B. A.; Butova, E. D.;
Yurchenko, A. G. Tetrahedron Lett. 1997, 38, 639-642.
45) Fokin, A. A.; Schreiner, P. R.; Schleyer, P. v. R.; Gunchenko, P.
A. J. Org. Chem. 1998, 62, 6494-6502.
46) Fokin, A. A.; Gunchenko, P. A.; Peleshanko, S. A.; Schleyer, P. v.
R.; Schreiner, P. R. Eur. J. Org. Chem. 1999, 855-860.
47) Fokin, A. A.; Peleshanko, S. A.; Gunchenko, P. A.; Gusev, D. V.;
Schreiner, P. R. Eur. J. Org. Chem. 2000, 3357-3362.
48) Fokin, A. A.; Lauenstein, O.; Gunchenko, P. A.; Schreiner, P. R. J.
Am. Chem. Soc. 2001, 123, 1842-1847.
49) Warner, P.; LaRose, R.; Schleis, T. Tetrahedron Lett. 1976, 4443-
446.
-
1
(
of isobutane in its reaction with DMD (24.2 kcal mol , TS1,
Scheme 2). Thus, the experimentally observed hydroxylation
of the C-H bond of propellane 2 is likely to occur via a
nonradical concerted pathway; molecule-induced homolysis is
thermodynamically unfavorable.
(
(
5
While the similarities of the computational results obtained
for the C-H hydroxylations of 2 and 3 are not surprising, the
reactivities of their C-C bonds toward DMD are distinctly
different. The addition of DMD to the C-C bond of 2 resulting
(
(
(
-
1
in biradical 15 is exergonic (-6.8 kcal mol ) and is ac-
-
1
(
companied by a high barrier (51.8 kcal mol , TS7, Figure 2);
the alternative barrier for C-H hydroxylation via TS5 is 17.0
(
-1
kcal mol lower, and this path is highly exergonic (-74.4 kcal
-
1
(
mol ), which agrees well with our experiments (Scheme 3).
The reactivity of the C-C vs C-H bonds of 3 toward DMD is
quite different: Formation of biradical 16 occurs via low-lying
4
(
50) Milne, I. R.; Taylor, D. K. J. Org. Chem. 1998, 63, 3769-3771.
(51) Fokin, A. A.; Schreiner, P. R.; Gunchenko, P. A.; Peleshanko, S.
q
-1
-1
TS8 (∆G298 ) 23.9 kcal mol ) and is -5.4 kcal mol
A.; Shubina, T. E.; Isaev, S. D.; Tarasenko, P. V.; Kulik, N. I.; Schiebel,
H.-M.; Yurchenko, A. G. J. Am. Chem. Soc. 2000, 122, 7317-7326.
(52) Wiberg, K. B. Acc. Chem. Res. 1984, 17, 379-386.
(53) Murray, R. W.; Gu, H. J. Org. Chem. 1995, 60, 5673-5677.

Molecule-Induced Alkane Homolysis with Dioxiranes
J. Am. Chem. Soc., Vol. 123, No. 45, 2001 11251
Scheme 5. Isodesmic Equation To Evaluate the Relative
Reactivities of the Propellanes 2 and 3 at B3LYP/
6
-311+G**//B3LYP/6-31G*
Scheme 6. Molecule-Induced C-C Bond Homolysis of
Propellane 3 with CrO2Cl2
Figure 2. B3LYP/6-31G* optimized geometries for the transition
structures TS7 and TS8 (selected bond distances in angstroms).
forms. The latter can only result from capture of 16 with DMD,
which is an efficient radical trap. Note that the strain energy
of hydrocarbon 3 is not very much higher than that of 2 (41.1
Scheme 4. B3LYP/6-311+G**//B3LYP/6-31G* Computed
3
4
-
1
Pathways (∆G298, kcal mol ) for the Reactions of
Propellanes 2 (A) and 3 (B) with DMD
-
1
-1
45
kcal mol for 2 vs 55.5 kcal mol for 3), and strain release
in TS8 vs TS7 cannot be the only reason for such pronounced
reactivity differences. A possible explanation is the difference
in the propellanic C-C bonds of these hydrocarbons. In contrast
5
2
to 2, the quaternary carbons in 3 are inverted, and the central
C-C bond has some biradical character. We estimated this
through the differences in singlet-triplet energies (∆∆EST ) 8
-
1
-1
kcal mol ) which are much lower for 3 (30 kcal mol ) than
-1
for 2 (42 kcal mol , B3LYP/6-31G*) and via the homodesmic
equation (Scheme 5), which shows that 3 is 24.9 kcal mol
-
1
less stable (and thus more reactive) than 2, assuming that
adamantane and homoadamantane both have low strain ener-
54
gies. Thus, both computational and experimental results show
that molecule-induced homolysis of DMD is possible only for
very weak σ-bonds (like in 3).
On the basis of the experimentally found product distribution
for the reaction of 3 with CrO2Cl2, the molecule-induced
homolysis also seems operative in this case (Scheme 6).
V
Biradical 18 formed after the activation step (Cr alkoxides are
5
5
well-characterized intermediates ) could be trapped either by
the CrdO group or through Cl-abstraction from CrO2Cl2, giving
alkoxides 19 and 20 and alcohols 10 and 12 after aqueous
workup. Dihydroxy derivative 10 was found in the reaction of
3
with DMD (vide supra) through an analogous radical trapping
reaction of 16.
Conclusions
The competitive C-H and C-C bond reactivities with DMD,
verifying concerted vs homolytic reaction mechanisms, were
studied experimentally and computationally for a set of pro-
pellanes. Although the molecule-induced homolytic C-H reac-
tion is generally characterized by a slightly lower barrier than
the concerted insertion, only the latter reaction is exergonic.
The C-C bond addition of DMD is more favorable than the
C-H insertion (the ∆G298 values are 23.9 vs 30.5 kcal mol ,
respectively) only for highly reactive propellane 3 with an
inverted geometry at the quaternary carbons. The molecule-
induced homolytic mechanism was also found experimentally
for propellanes 1-3 in their reactions with CrO2Cl2. The
reactivities of propellanes 1 and 2 toward DMD (C-H activa-
exergonic. In stark contrast to 2, propellane 3 can form biradical
q
-1
1
6 easily. This is the first clear-cut example for molecule-
induced alkane C-C bond homolysis with DMD. The elimina-
tion of acetone from 16 is facile as it does not require much
atomic movement (consequently, the exothermicity is extraor-
dinarily high), but merely electron rearrangements as indicated
by the half-arrows on 16 in Scheme 4.
The computations nicely complement our experiments (see
details below). Despite highly exergonic fragmentation of
biradical 16 to ketone 11, some (8%) dihydroxy product (10)
(
54) Engler, E. M.; Andose, J. D.; Schleyer, P. v. R. J. Am. Chem. Soc.
973, 95, 8005-8025.
55) Wistuba, T.; Limberg, C.; Kircher, P. Angew. Chem., Int. Ed. 1999,
38, 3037-3039.
1
(

11252 J. Am. Chem. Soc., Vol. 123, No. 45, 2001
Fokin et al.
1
tion) and CrO2Cl2 (C-C activation) are clearly different, and
only relatively unstable 3 reacts with both CrO2Cl2 and DMD
similarly via thermodynamically favored C-C bond molecule-
induced homolysis. Hence, this mechanism, which is typical
for metal-oxo reagents in general, and for CrO2Cl2 in particular,
plays at best a minor role in alkane activations with dioxiranes.
In stark contrast to the reactions with metal-oxo reagents, only
very weak σ-bonds (like the central C-C bond in 3) can be
homolyzed by ground-state singlet DMD.
1-hydroxy-3,6-dehydrohomoadamantane (9), mp ) 171-172 °C. H
NMR (δ, ppm, CDCl
3
): 1.55-1.62 m (4H), 1.64-1.70 s (4H), 1.79-
1
3
1
.86 m (1H), 1.88-2.00 m (4H), 2.01-2.08 d (2H), 2.39 bs (1H). C
3
NMR (δ, ppm, CDCl ): 80.03, 54.88, 48.01, 46.82, 42.87, 39.30, 28.37.
MS (70 eV): m/z ) 164, 146, 136, 121, 109, 106, 95, 91, 81, 67, 55.
Anal. Calcd for C11 16O: C, 80.44; H, 9.82. Found: C, 80.29; H, 9.71.
C. Reaction of 1 with CrO Cl . Into a stirred solution of 400 mg
1.79 mmol) of 1 in 11 mL of CCl was added a solution of 1.08 g of
CrO Cl (7.01 mmol) in 10 mL of CCl dropwise at 0 °C under argon.
The reaction mixture was quenched with aqueous saturated Na S O ,
H
2
2
(
4
2
2
4
2
2
5
the organic layer was separated, and the aqueous part was extracted
Computational Methods
with CHCl (3 × 15 mL). The combined extracts were washed with
3
2 4
water and brine and were dried over Na SO . After removal of the
Geometries were fully optimized utilizing the density functional
5
6,57
solvent, column chromatography (ether:n-hexane ) 2:1) of the residue
gave 113 mg (0.38 mmol, 21%) of 1-chloro-3-(1-chlorobenzyl)ada-
mantane (4) and 366 mg (1.33 mmol, 75%) of phenyl-3-chloroadamant-
three-parameter hybrid B3LYP functional
in conjunction with
-31G* and 6-311+G** basis sets as implemented in Gaussian 98.5
8
6
Temperature corrections (298 K) were derived from B3LYP/6-31G*
frequency computations; these corrections were used to correct the
higher quality single-point energies at B3LYP/6-311+G**. The
1
-yl ketone (5).
Compound 4. H NMR (δ, ppm, CDCl
1H), 7.15 m (5H). 13C NMR (δ, ppm, CDCl
127.77, 77.32, 68.32, 48.48, 46.73, 42.36, 37.34, 37.12, 34.59, 31.11.
Anal. Calcd for C17 : C, 69.16; H, 6.83; Cl, 24.02. Found: C,
69.23; H, 6.81; Cl, 24.13.
1
3
): 1.1-2.3 m (14H), 4.3 s
3
): 137.61, 128.82, 128.06,
(
“
guess)mix” option was used for open-shell singlet state computations.
59,60 61
While this is an approximate treatment, others
and we have had
H20Cl
2
very positive experience with this type of approach; notably, Bach et
al. observed RHFfUHF instabilities for some dioxirane structures.21
Although DFT is, in principle, a single-determinant method, very often
obvious multireference states can be described adequately if one
dominant configuration yields an acceptable density. Hence, the results
of such DFT computations must be gauged against higher level methods.
As the state splittings for DMD agree well at MRD-CI, CCSD(T), and
1
Compound 5. H NMR (δ, ppm, CDCl
3
): 7.76 m (1H), 7.53 m
(
1H), 7.30 m (3H), 2.70 s (2H), 2.33 s (2H), 2.12-2.15 m (2H), 1.7-
1
3
20
1.92 m (8H). C NMR (δ, ppm, CDCl
1
3
3
): 200.66, 150.32, 133.56,
32.39, 126.73, 126.18, 125.62, 52.12, 51.32, 49.33, 46.63, 45.09, 35.99,
3.56. Anal. Calcd for C17 19ClO: C, 74.31; H, 6.97; Cl, 12.90.
H
Found: C, 74.24; H, 6.95; Cl, 12.82.
D. Reaction of 3 with DMD. Following procedure A, from 105
mg (0.78 mmol) of 1,3-dehydroadamantane (3) in 10 mL of CH Cl
2 2
B3LYP levels of theory, the approach used here seems justified for
making qualitative arguments and comparisons with experiment.22 The
reaction pathways along both directions from the transition structures
were followed by the IRC method. xyz coordinates and energies for
all optimized species are summarized in the Supporting Information.
6
2
and 12.5 mL of a 0.063 M acetone solution of DMD (0.79 mmol) were
obtained 55 mg of 3-methylenebicyclo[3.3.1]nonan-7-one (11) (0.36
mmol, 46%) and 11 mg of 1,3-dihydroxyadamantane (10) (0.06 mmol,
6
3,64
8
%), identical from MS and NMR data to standard samples.
The
Experimental Section
conversion of 3 was virtually complete as determined after quenching
the reaction mixture with water (only trace amounts of 1-hydroxyada-
mantane as a result of the addition of water to 3 were found). In a
series of control experiments, we also ran this reaction with only 10%
DMD under argon. After quenching of the reaction with water, the
conversion of 3 was found to be close to 10%; the product distributions
were identical. Running this reaction in the presence of oxygen (bubbled
through the solution), we obtained the same results as under argon in
the absence of oxygen. Hence, 3 reacts much faster with DMD than
with oxygen, in contrast to the slow reactions of 1 and 2 where DMD
decomposes much faster relative to the hydrocarbon oxidation reaction.
E. Reaction of 3 with CrO Cl . Following procedure C, from 500
A. Reaction of 1 with DMD. Into a degassed solution of 312 mg
(
1.39 mmol) of 4-phenyl-3,6-dehydrohomoadamantane (1) in 3 mL of
dry acetone was added 21 mL of a 0.073 M acetone solution of DMD
1.54 mmol) under argon. After 20 h the solvents were removed under
reduced pressure. Column chromatography (silica gel Merck 60, ether:
n-hexane ) 2:1) of the residue gave 180 mg of unreacted 1 (conversion
2%) and 77 mg (0.32 mmol, 23%) of 4-hydroxy-4-phenyl-3,6-
(
4
1
dehydrohomoadamantane (6). H NMR (δ, ppm, CDCl
3
): 1.30-1.78
m (7H), 1.86 s (2H), 2.13 AB (15 Hz, 1H), 2.24 bs (2H), 2.47 bs (1H),
3
2
.34 m (1H), 2.77 AB (15 Hz, 1H), 7.25-7.38 (5H). 1 C NMR (δ,
ppm, CDCl ): 145.25, 128.27, 127.02, 126.64, 76.20, 58.56, 49.66,
9.15, 46.59, 42.40, 42.36, 42.23, 41.59, 41.14, 35.81. MS (70 eV):
m/z ) 240, 222, 197, 183, 179, 167, 147, 129, 105, 91, 77, 65, 51.
Anal. Calcd for C17 20O: C, 84.96; H, 8.39. Found: C, 85.05; H, 8.26.
3
2
2
4
2 2
mg (3.73 mmol) of 3 in 15 mL of CH Cl and a solution of 700 mg
of CrO Cl2 (4.52 mmol) in 5 mL of CCl4 was obtained 445 mg
2
H
(2.39 mmol, 64%) of 1-chloro-3-hydroxyadamantane (12) after column
chromatography, identical to the standard sample65 from MS and
NMR data. Changing the solvent to methanol gave 150 mg (0.89 mmol,
24%) of 1,3-dihydroxyadamantane (10), also identical to the standard
sample.64
B. Reaction of 2 with DMD. Following procedure A, to 300 mg
2.03 mmol) of 3,6-dehydrohomoadamantane (2) in 2 mL of acetone
was added 44 mL of a 0.047 M acetone solution of DMD (2.07 mmol),
(
to give 270 mg of unreacted 2 and 20 mg (0.12 mmol, 6%) of
(
(
(
56) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft, the Fonds der Chemischen Industrie,
the University of Georgia, and the Fundamental Research
Foundation of the Ukraine.
57) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
58) 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. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian,
Inc.: Pittsburgh, PA, 1998.
Supporting Information Available: xyz coordinates and
absolute energies (Table 1) of all optimized species (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA0158096
(
59) Gr a¨ fenstein, J.; Kraka, E.; Cremer, D. Chem. Phys. Lett. 1998, 288,
(63) Stetter, H.; Tacke, P. Ber. Dtsch. Chem. Ges. 1963, 96, 694-697.
(64) Moiseev, I. K.; Belyaev, P. G.; Barabanova, I. V.; Bardyug, O. P.;
Vishnevskii, E. H.; Novatskaya, N. I.; Golod, E. L.; Gidaspov, B. V. Zh.
Org. Khim. 1975, 11, 214-215.
(65) Moiseev, I. K.; Klimochkin, Y. N.; Zemcova, M. N.; Trahtenberg,
P. L. Zh. Org. Khim. 1984, 20, 1435-1438.
5
93-602.
(
(
60) Cramer, C. J. J. Am. Chem. Soc. 1998, 120, 6261-6269.
61) Schreiner, P. R.; Prall, M. J. Am. Chem. Soc. 1999, 121, 8615-
8
627.
62) Fukui, K. J. Phys. Chem. 1970, 74, 4161.
(