The rearrangement of the cubane radical cation in solution

Article abstract of DOI:10.1002/1521-3765(20010702)7:13<2739::AID-CHEM2739>3.0.CO;2-R

The rearrangement of the cubane radical cation (1^.+) was examined both experimentally (anodic as well as (photo)chemical oxidation of cubane 1 in acetonitrile) and computationally at coupled cluster, DFT, and MP2 [BCCD(T)/cc-pVDZ//B3LYP/6-31G* + ZPVE as well as BCCD(T)/cc-pVDZ//MP2/6-31G* + ZPVE] levels of theory. The interconversion of the twelve C_2v degenerate structures of 1^.+ is associated with a sizable activation energy of 1.6 kcal mol^-1. The barriers for the isomerization of 1^.+ to the cuneane radical cation (2^.+) and for the C-C bond fragmentation to the secocubane-4,7-diyl radical cation (10^.+) are virtually identical (ΔH₀^? = 7.8 and 7.9 kcal mol^-1, respectively). The low-barrier rearrangement of 10^.+ to the more stable syn-tricyclooctadiene radical cation 3^.+ favors the fragmentation pathway that terminates with the cyclooctatetraene radical cation 6^.+. Experimental single-electron transfer (SET) oxidation of cubane in acetonitrile with photoexcited 1,2,4,5-tetracyanobenzene, in combination with back electron transfer to the transient radical cation, also shows that 1^.+ preferentially follows a multistep rearrangement to 6^.+ through 10^.+ and 3^.+ rather than through 2^.+. This was confirmed by the oxidation of syn-tricyclooctadiene (3), which, like 1, also forms 6 in the SET oxidation/rearrangement/electron-recapture process. In contrast, cuneane (2) is oxidized exclusively to semibullvalene (9) under analogous conditions. The rearrangement of 1^.+ to 6^.+ via 3^.+, which was recently observed spectroscopically upon ionization in a hydrocarbon glass matrix, is also favored in solution.

Full text of DOI:10.1002/1521-3765(20010702)7:13<2739::AID-CHEM2739>3.0.CO;2-R

FULL PAPER
The Rearrangement of the Cubane Radical Cation in Solution
[
a]
[a]
[b]
Peter R. Schreiner,* Alexander Wittkopp, Pavel A. Gunchenko,
[b]
[b]
[b]
Alexander I. Yaroshinsky, Sergey A. Peleshanko, and Andrey A. Fokin*
.
‡
Abstract: The rearrangement of the
4,7-diyl radical cation (10 ) are virtually
transient radical cation, also shows that
.
‡
=
� 1
.‡
cubane radical cation (1 ) was exam-
ined both experimentally (anodic as well
as (photo)chemical oxidation of cubane
identical (DH0 ˆ 7.8 and 7.9 kcalmol ,
1
preferentially follows a multistep
.
‡
.‡
respectively). The low-barrier rear-
rearrangement to 6 through 10 and
3
.
‡
.‡
.‡
rangement of 10 to the more stable
syn-tricyclooctadiene radical cation 3
favors the fragmentation pathway that
terminates with the cyclooctatetraene
radical cation 6 . Experimental single-
electron transfer (SET) oxidation of
cubane in acetonitrile with photoexcited
1,2,4,5-tetracyanobenzene, in combina-
tion with back electron transfer to the
rather than through 2 . This was
.
‡
1
in acetonitrile) and computationally at
coupled cluster, DFT, and MP2
BCCD(T)/cc-pVDZ//B3LYP/6-31G*
ZPVE as well as BCCD(T)/cc-
confirmed by the oxidation of syn-tri-
cyclooctadiene (3), which, like 1, also
forms 6 in the SET oxidation/rearrange-
ment/electron-recapture process. In con-
trast, cuneane (2) is oxidized exclusively
to semibullvalene (9) under analogous
[
.
‡
‡
pVDZ//MP2/6-31G* ‡ ZPVE] levels
of theory. The interconversion of the
.
‡
.‡
twelve C degenerate structures of 1 is
conditions. The rearrangement of 1 to
2v
.
‡
.‡
associated with a sizable activation en-
6
via 3 , which was recently observed
�
1
ergy of 1.6 kcalmol . The barriers for
spectroscopically upon ionization in a
hydrocarbon glass matrix, is also favored
in solution.
.
‡
the isomerization of 1 to the cuneane
Keywords: computer chemistry
cubanes ´ oxidation ´ radical ions
´
.
‡
radical cation (2 ) and for the C�C
bond fragmentation to the secocubane-
Introduction
which is probably one of the most rigid and strained hydro-
carbons prepared to date. The exceptional kinetic stability of
1 is presumably due to the fact that breaking just one C�C
While alkane radical cations play an increasingly important
role in understanding some fundamental reaction mecha-
our knowledge about the structures and energies of
these highly reactive intermediates is still rather limited. This
is due to the fast rearrangement or fragmentation of these so-
called s species (C�C or C�H bonds are partially broken),
which makes purely experimental investigations difficult. A
case in point is the radical cation derived from cubane (1),[
bond homolytically[ or heterolytically
5]
[6, 7]
causes only minor
[
1, 2]
nisms,
changes in the cage structure and may simply not be enough
[8]
for opening the entire cage. Hence, the cubane radical cation
.
‡
(1 ) is expected to maintain some of the basic features of the
.
‡
cubane structure, at least at very low temperatures. It is not
.
‡
clear whether 1 actually was observed by ESR spectroscopy;
neither its structure nor the rearrangements of this fascinating
molecule have been resolved. The present paper reports on a
combined experimental/computational study on the struc-
3, 4]
[
a] Prof. Dr. P. R. Schreiner, Dipl. Chem. A. Wittkopp
Institut für Organische Chemie, Georg-August-Universität Göttingen
Tammannstr. 2, 37077 Göttingen (Germany) and
Department of Chemistry, The University of Georgia
Athens, GA 30602-2556 (USA)
.
‡
tures and rearrangements on the C H8 hypersurface starting
8
from cubane.
Paquette et al. first claimed to have observed sharp ESR
Fax : (‡1)706-542-9454
.‡
[9]
signals for 1 in neon matrices at 4 ± 9 K but this finding was
E-mail: prs@sunchem.chem.uga.edu
[10]
refuted later.
Eatonꢀs subsequently recorded ESR spec-
[
b] Prof. Dr. A. A. Fokin, Dr. P. A. Gunchenko, Dr. A. I. Yaroshinsky,
Dipl. Chem. S. A. Peleshanko
trum[ was first interpreted in terms of dynamically inter-
11]
.
‡
converting, Jahn ± Teller distorted C2v structures of 1 but it
was noted later ª...that the cubane radical cation was not
observed previously...º. Strong evidence for 1 comes from
a fluorescence-detected magnetic-resonance (FDMR) study
at 190 K by Eaton et al.; fast equilibration of twelve
equivalent C2v structures (where one bond is suggested to be
Department of Organic Chemistry, Kiev Polytechnic Institute
pr. Pobedy, 37, 252056 Kiev (Ukraine)
Fax : (‡38)044-274-20-04
[10]
.‡
E-mail: aaf@xtf.ntu-kpi.kiev.ua
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/chemistry/or from the author. Table
with relative and absolute energies and xyz-coordinates of all computed
structures.
.
‡
significantly lengthened in 1 ) was indicated based on
Chem. Eur. J. 2001, 7, No. 13
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0713-2739 $ 17.50+.50/0
2739

P. R. Schreiner, A. A. Fokin et al.
FULL PAPER
MINDO/3 calculations; however, no transition structures were
computed, and it was noted that much higher level compu-
[
11]
tations were needed to address this question properly.
.
‡
Two reaction pathways for 1 were previously proposed;
.
‡
that is, isomerization to the cuneane radical cation, 2 , and
C�C bond breaking to give the syn-tricyclooctadiene radical
.
‡
[11, 12]
cation, 3 (Scheme 1).
It was suggested that the isomer-
Abstract in German: Die Umlagerung des Cuban-Radikalkat-
Scheme 1. Two pathways to cyclooctatetraene 6 from the cubane radical
.
‡
ions (1 ) wurde sowohl experimentell (durch anodische bzw.
photochemische Oxidation von Cuban (1) in Acetonitril) als
auch mittels theoretischer Berechnungen auf dem Coupled
.‡
cation 1
.
.
‡
.‡
ization of 1 via 2 to the bicyclooctadienediyl radical cation,
Cluster-,
Dichtefunktionaltheorie-
und
‡
MP2-Niveau
ZPVE und
.
‡
4
, (Scheme 1) occurs under g irradiation of 1 at 77 K in
[BCCD(T)/cc-pVDZ//B3LYP/6-31G*
[11, 12]
.‡
.‡
CF ClCFCl .
In contrast, C�C bond cleavage of 1 to 3
2
2
BCCD(T)/cc-pVDZ//MP2/6-31G* ‡ ZPVE] untersucht. Die
.
‡
.
‡
and further to the bis(cyclobutenylium) radical cation, 5
,
Umwandlung der zwölf entarteten C -Strukturen von 1 hat
2
v
�
1
was recently found under pulse radiolysis in hydrocarbon
glasses at 30 K.[ It was suggested that the latter reaction
involves a [2p‡2p]cycloreversion, because the ring opening of
eine Aktivierungsbarriere von 1.6 kcalmol . Die Barrieren für
die Isomerisierung von 1 in das Cunean- (2 ) bzw. in das
Secocuban-4,7-diyl Radikalkation (10 ) sind praktisch iden-
12]
.
‡
.‡
.
‡
.
‡
.‡
.‡
=
� 1
1
is activated, in contrast to the reaction from 3 to 5
tisch (DH0 ˆ 7.8 bzw. 7.9 kcalmol ). Die ebenfalls von nied-
[12]
.
‡
which occurs spontaneously.
strained radical cation 3 , generated independently from the
The rearrangement of the
rigen Barrieren begleitete Umlagerung von 10 in das stabilere
.
‡
.
‡
syn-Tricyclooctadien-Radikalkation 3 ist etwas bevorzugt
und ergibt schlieûlich das Cyclooctatetraen-Radikalkation 6 .
.
‡
corresponding neutral molecule, to the cyclo-octatetraene
.
‡
.‡
.‡
radical cation (6 ) via 7 and 8 was recently studied
experimentally and computationally by Bally.[
Die Einelektronenoxidation (SET) von Cuban in Acetonitril
13]
(mittels photoangeregtem 1,2,4,5-Tetracyanbenzol, TCB) und
In the present paper we report high-level computational
Elektronen-Rücktransfer zum intermediären Radikalkation
.
‡
.‡
.
‡
.‡
.‡
details for the transformation of 1 to 6 , as well as
zeigt ebenfalls, daû 1 vorzugsweise über 10 und 3 aber
.
‡
.‡
experimental data on the behavior of cubane and some
nicht über 2 zu 6 umlagert. Dies wurde durch die Oxidation
von syn-Tricyclooctadien (3) gezeigt, welches, in Analogie zu
related C H8 hydrocarbons under oxidative conditions in
8
solution. Since Eaton pointed out that the difficulty in
1, ebenfalls 6 in einem SET-/Umlagerungs-/Elektronenein-
.
‡
elucidating the intricacies of the rearrangements of 1 lies
in the identification of the intermediates and products, we
envisaged that combining an oxidative single-electron trans-
fer (SET) step with back electron transfer (ET)[ from the
oxidant to the transient radical cations, would be a useful
approach to this problem.
fangschritt bildet. Im Gegensatz dazu lagert Cunean (2) nach
der SET-Oxidation ausschlieûlich zu Semibullvalen (9) um.
.
‡
.‡
.‡
Die Umlagerung von 1 zu 6 über 3 , die kürzlich auch
spektroskopisch in einer Kohlenwasserstoff-Matrix nachgewie-
sen wurde, ist damit auch in Lösung bevorzugt.
14]
Results and Discussion
[
15, 16]
Although we have had positive experience
with density
functional theory (DFT)[ for treating radical cations, the
17]
applicability of DFT to these species is still under a lot of
discussion.[
3
18±21]
Therefore, we utilized both DFT (B3LYP/6-
1G*) and Mùller± Plesset (MP2/6-31G*)[22] second-order
perturbation theory in combination with coupled-cluster
single-point energies.[
23±28]
The slightly lower absolute
BCCD(T) energies for the DFT structures favor the topology
of the respective potential energy surfaces (PES) at B3LYP
slightly.
.
‡
The interconversion of the degenerate C2v structures of 1
Figure 1) via TS1-1 is associated with a sizable activation
energy of 1.6 kcalmol at 0 K. As a consequence, 1
(
�
1
.‡
undergoes dynamic Jahn ± Teller averaging of static C
2
v
structures as suggested by Eaton et al.[ A comparison of
10]
[
11]
our BCCD(T)-computed and Eatonꢀs published ESR
.
‡
spectra of 1 (Figure 2) strongly supports this analysis. The
.
‡
computed spectra both for 1 and TS1-1 agree well with the
neon-matrix ESR. The ESR spectra may be interpreted on the
2740
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0713-2740 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 13

Cubane Radical Cations
2739±2744
8.
‡
.‡
� 1
Figure 1. B3LYP/6-31G* optimized C
8
H
structures relevant to the rearrangement of the cubane radical cation (1 ); relative energies in kcalmol ; bond
distances in Š; first entry: B3LYP/6-31G* ‡ ZPVE; second entry: BCCD(T)/cc-pVDZ//B3LYP/6-31G* ‡ ZPVE.
.
‡
.‡
In analogy to some other C H8 species, 1 may eventually
8
rearrange (pathway A, Scheme 2) to the cuneane radical
.
‡
.‡
cation, 2 , and then to 4 , which is the computed global
•
+
•+
•+
•+
A
B
•+
2•+
9•+
4•+
•
+
•+
1•+
3•+
6•+
1
0•+
Scheme 2. The two rearrangement pathways for the cubane radical cation
.
‡
(
1 ).
.
‡
.‡
minimum on this part of the C H8 PES. Structure 4 may be
8
considered as the ªopen formº of the semibullvalene radical
.
‡
[29, 30]
� 1
cation, 9 (Figure 1);
a barrier of 35.7 kcalmol pre-
.
‡
.‡
cludes the expansion of 4 to 6 (the reversible reaction is
known).[
13, 30±32]
.
‡
.‡
Figure 2. Comparison of the simulated ESR spectra for 1 , TS1-1, and 10
(
We also computed the C�C bond fragmentation pathway B
.
‡
0 K) at BCCD(T)/cc-pVDZ with the experimental ESR spectrum for 1
77 K, bottom, scanned from ref. [11] with permission). TS1-1 represents the
averaged spectrum for rapidly interconverting, degenerate isomers of 1
.
‡
for the cubane radical cation 1 (Scheme 2 and Figure 3).
(
.
‡
.
‡
Breaking one of the C�C bonds in 1 leads to the
.
.
‡
secocubane-4,7-diyl radical cation, 10 , which is a true and
.
‡
.‡
thus far unrecognized minimum on the C H PES; 10 is
8
8
.
‡
� 1
.‡
basis of the rapidly equilibrating minima of 1 : as the ESR
8.5 kcalmol more stable than 1 . The barrier for this C�C
.
‡
� 1
spectrum of 1 is already narrow and is dominated by a sine-
shaped base peak, the averaged spectrum shows even less
bond cleavage via TS1 (7.8 kcalmol ) is virtually the same
-10
.
‡
.‡
as that for the isomerization of 1
to 2
via TS1-2
�
1
structure and resembles that of TS1-1
.
(7.9 kcalmol ). At the same time pathway B is more
Chem. Eur. J. 2001, 7, No. 13
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0713-2741 $ 17.50+.50/0
2741

P. R. Schreiner, A. A. Fokin et al.
reaction times substantial
FULL PAPER
amounts of syn-tricycloocta-
diene 3 were found together
with 6. Independent oxidation
of 3 with TCB gave only 6.
Conversion of 1 to the neutral
6
, as well as to 3, certainly
involves SET oxidation fol-
lowed by back ET: extended
photo-irradiation of 1 or 3 in
the absence of TCB did not give
6
and the starting hydrocarbons
were recovered quantitatively.
Thus, cubane may undergo
an SET oxidation/photochemi-
cal rearrangement to cycloocta-
.
‡
tetraene 6 via 3 . At the same
8.
‡
Figure 3. B3LYP/6-31G* optimized C
8
H
structures relevant to the C�C bond cleavage of the cubane radical
time, the barriers for the iso-
.
‡
� 1
.‡
.‡
.‡
cation (1 ); relative energies in kcalmol ; bond distances in Š; first entry: B3LYP/6-31G* ‡ ZPVE; second
entry: BCCD(T)/cc-pVDZ//B3LYP/6-31G* ‡ ZPVE.
merization of 1 to 2 and
.‡
fragmentation of 1 to 3 are
similar and two parallel pro-
favorable due to the substantial strain relief in the tricyclo-
cesses cannot be excluded. Moreover, due to fast back ET, we
cannot yet distinguish between pathways A and B (Scheme 2),
because semibullvalene 9 may photochemically rearrange to
.
‡
.‡
octadiene radical cation 3 (the C�C bond cleavage in 10
.
‡
=
that leads to 3 proceeds via the low-lying TS with DH
<
3
-10
0
�
1
.‡
[14]
1
kcalmol ). The ESR spectrum computed for 10 (Fig-
cyclooctatetraene 6 (A):
.
‡
ure 2) is clearly different from that of 1 . Thus, the break
.
‡
down of the cubane cage in 1 occurs in a stepwise fashion.
Despite extensive efforts, we were unable to locate a
transition structure for the concerted [2p ‡ 2p] cycloreversion
e�
hn
.
‡
! 2.‡ ! 4 ! 9 ! 6
!
.‡
1
1
(A)
(B)
�
.‡ e
.‡
! 10.‡ ! 3.‡ ! 6 ! 6
!
.
‡
.‡
of 1 to 3
.
Our computations show that pathway A is thermodynami-
cally less favorable than B due to the substantial strain relief
Trapping of the intermediates was attempted by anodic
oxidation of 1 in acetonitrile, but only a mixture of quite a
.
‡
.‡
� 1
in 3 compared with 2 (18.8kcal mol ). This is in agree-
ment with results obtained for the ionization of 1 in hydro-
number of inseparable C H NHCOCH isomers was observed
8
7
3
(
gas chromatography/mass spectrometry).
To elucidate this point further and to differentiate between
[
12]
carbon glasses under pulse radiolysis at 30 K, where the
.
‡
intermediate bond-cleavage product 5 (Scheme 1) as well as
pathways A and B, we oxidized cuneane (2) with TCB; only
.
‡
the final product radical cation 6 were identified by
.‡
semibullvalene 9 was formed, apparently after back ET to 4
electronic absorption spectroscopy. The unexpected product,
(
Scheme 4). Hydrocarbon 9 was stable under these conditions
and could be isolated.
.
‡
radical cation 4 , which could formally be attributed to
[
12]
pathway A, was observed under g-irradiation of cubane. As
we suspected that this was due to matrix effects or secondary
reactions, we decided to study the SET oxidation of 1 with
1,2,3,5-tetracyanobenzene (TCB) in solution. When hydro-
carbons are oxidized by photoexcited aromatics, back ET is a
[
33±38]
favored process,
trapped efficiently.
and the transient radical cations may be
Scheme 4. Photooxidation of cuneane (2) with TCB.
The photo-oxidation of 1 with TCB in acetonitrile at
408C gave cyclooctatetraene 6 (Scheme 3). At shorter
�
The observed SET rearrangement of 2 to 9 after electron
recapture is clearly different from that of cubane 1. Thus, the
.
‡
rearrangement of the cubane radical cation 1 in solution
.
‡
indeed initially follows fragmentation to 3 , rather than
.
‡
isomerization to 2 . This is similar to the behavior of 1 under
pulse radiolysis in hydrocarbon glasses at 30 K, in which
[
12]
pathway B is apparently followed.
We can therefore
.
‡
conclude that 2 is not involved in the cubane radical cation
rearrangements and that 1 follows the same reaction pattern
for oxidation in matrix conditions and in solution.
Scheme 3. Photooxidation of cubane (1) and syn-tricyclooctadiene (3)
with TCB.
2742
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0713-2742 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 13

Cubane Radical Cations
2739±2744
with water and brine, and were dried over Na
2
SO
4
. The solvent was
Conclusion
removed at atmospheric pressure by using a Vigreux column (200 mm).
Separation of the residue on a silica gel column (40 cm  1.5 cm, Merck
Kieselgel 60, 0.063 ± 0.1 mm) with 2-methylbutane as eluant gave 41 mg
39%) of 1, 18 mg of 3 (17%), and 35 mg of 6 (34%), which are identical
from NMR and MS data to standard samples.
The reaction was carried out as above for 2 h. The H NMR (CDCl
.
‡
As predicted by Eaton et al., the cubane radical cation 1
equilibrates with its degenerate isomers (we computed a
(
�
1
barrier of 1.6 kcalmol ). As the barrier for the rearrange-
.
‡
.‡
� 1
1
ment of 1 to 10 is sizable (about 7.9 kcalmol ) Eaton must
have observed the ESR spectrum of 1 .
The first step in the rearrangement of 1 is either isomer-
3
) of the
.
‡
reaction mixture shows the signals of the cubane 1: 4.03 (s, 91%), cyclo-
octatetraene 6: 5.80 (s, 3%), and syn-tricyclooctadiene 3: 3.21 (m), 6.02
.
‡
(
m), 6%.
Photo-oxidation of syn-tricyclo-octadiene 3[45] with TCB: A solution of 3
52 mg, 0.5 mmol) and TCB (75 mg, 0.4 mmol) in acetonitrile (60 mL) was
irradiated as in A. The H NMR of the reaction mixture shows the signals of
3 (54%) and 6 (46%).
Photo-oxidation of cuneane 2[46] with 1,2,3,4-tetracyanobenzene (TCB): A
solution of 2 (52 mg, 0.5 mmol) and TCB (75 mg, 0.4 mmol) in acetonitrile
.
‡
ization to the cuneane radical cation, 2 (pathway A) or C�C
.
‡
bond breaking to the secocubane-4,7-diyl radical cation, 10
pathway BÐthe first C�C bond cleavage). The latter is
followed by rearrangement to the syn-tricyclooctadiene
(
(
1
.
‡
radical cation, 3 (low-barrier second C�C bond cleavage).
.
‡
Pathway B is favored thermodynamically because 3 is
(
(
60 mL) was irradiated for 4 h as in A. Column chromatography gave 41 mg
�
1
.‡
1
8.8 kcalmol more stable than 2 . While pathway A
terminates with the reduction of the bicyclooctadienediyl
1
79%) of semibullvalene 9. H NMR (CDCl
3
): d ˆ 5.25 (m, 2H), 4.25 (m,
[47]
4
H), 3.04 (m, 2H)Ðidentical to as previously described.
Electro-oxidation of cubane 1: A mixture of 1 (104 mg, 1.0 mmol) in
acetonitrile (75 mL) and NH BF (150 mg, 1.4 mmol) was placed into a
.
‡
radical cation 4 to semibullvalene 9, pathway B gives the
.
‡
cyclooctatetraene radical cation 6 . We identified the two
pathways based on the structural analysis of the isolated
neutrals.
4
4
glass cell with Pt electrodes and subjected to direct current at 2.4 V anode
potential for 36 h. The reaction mixture was diluted with water (5 mL), the
acetonitrile was evaporated and worked up as described in A; 68 mg of 1
were recovered. After evaporation of the solvent, the residue was dissolved
in CH Cl2 (20 mL), washed with water and brine, and dried, and the
solvents were removed in vacuo. The residue (47 mg) was purified by flash
chromatography on silica gel (diethyl ether/methanol 10:1) and analyzed
by GC/MS (HP5890 Series II GC with HP5971A MSD, capillary column
HP Ultra1, 50 m  0.2 mm  0.33 m film, Tˆ 80 ± 2008C, 108Cmin , mass
selective detector) and showed the following main peaks characteristic for
isomeric monoacetamides: 24.74 min (161, 8%; 118, 100%; 104, 3%; 91,
As only 9 was found in the oxidation of 2 with photoexcited
TCB, and as the oxidation of 1 leads exclusively to 6 (via 3),
2
.
‡
.‡
.‡
1
must follow pathway B. Hence, 2 is probably not
.
‡
involved in the rearrangement of 1 to 6
.
Thus, the C�C bond fragmentation of cubane, which was
spectroscopically observed recently for the ionization of 1 in
the solid state, takes place similarly in solution.
�
1
3
2
1
6%; 65, 9%); 25.33 min (161, 15%; 138, 11%; 118, 100%; 91, 27%);
5.69 min (161, 4%; 118, 100%; 91, 30%); 26.35 min (161, 1%; 118, 100%;
02, 15%; 91, 27%).
Computational methods: The B3LYP and MP2 methods with
6
-31G* basis set were used as implemented in Gaussian 98[
39]
for geometry optimizations and harmonic vibrational fre-
quency analyses (NIMAG ˆ 0 for minima and 1 for transition
structures). The reaction pathways along both directions from
the transition structures were followed by the intrinsic
Acknowledgements
[
40]
This work was supported by the Volkswagenstiftung (I/74 614) and the
Fundamental Research Foundation of the Ukraine. We are thankful to the
HRLS (Stuttgart) for generous allotments of computer time and thank
Prof. Dr. A. de Meijere for his continuing support. A.W. thanks Dr. K.
Exner for help with the graphical representation of the computed ESR
spectra.
reaction coordinate method. Coupled-cluster single-point
[41, 42]
energies utilizing Brueckner-type orbitals [BCCD(T)]
with a cc-pVDZ basis set[43, 44] and zero-point B3LYP and
MP2 vibrational energies (ZPVE, unscaled) were used to
improve the energies. Unless noted otherwise, our final
energies refer to BCCD(T)/cc-pVDZ//B3LYP/6-31G* ‡
ZPVE; the MP2 results are available in the Supporting
Information.
[
[
1] M. Schmittel, A. Burghart, Angew. Chem. 1997, 109, 2659 ± 2699;
Angew. Chem. Int. Ed. Engl. 1997, 36, 2550 ± 2589.
2] M. Mella, M. Fagnoni, M. Freccero, E. Fasani, A. Albini, Chem. Soc.
Rev. 1998, 27, 81 ± 89.
[
[
[
[
[
3] P. E. Eaton, T. W. Cole, Jr., J. Am. Chem. Soc. 1964, 86, 962 ± 964.
4] P. E. Eaton, T. W. Cole, Jr., J. Am. Chem. Soc. 1964, 86, 3157 ± 3158.
5] D. S. Reddy, Tetrahedron Lett. 1990, 31, 805 ± 806.
6] T.-Y. Luh, L. M. Stock, J. Am. Chem. Soc. 1974, 96, 3712 ± 3713.
7] P. E. Eaton, C.-X. Yang, Y. Xiong, J. Am. Chem. Soc. 1990, 112, 3225 ±
Experimental Section
All NMR spectra were recorded on a Varian VXR-300 spectrometer at
1
13
3
3
00 MHz ( H NMR) or 75 MHz ( C NMR) in CDCl solutions. The
chemical shifts are given on the d scale in ppm; internal standard: HMDS.
All compounds show adequate IR and distortionless enhancement by
polarization transfer 13C NMR spectra. The GC/MS analyses were carried
out by using a HP5890 Series II gas chromatograph (column HP Ultra1
0 m  0.2 mm  0.33 mm film: cross linked methyl silicone) with
HP5971A mass detector. A standard 150 W UV lamp (maximum emission
00 nm) was used for the photochemical experiments.
Photo-oxidation of cubane 1 with TCB: Method A: A solution of 1
104 mg, 1.0 mmol) and TCB (151 mg, 0.8 mmol) in acetonitrile (120 mL)
3
226.
8] E. W. Della, N. J. Head, P. Mallon, J. C. Walton, J. Am. Chem. Soc.
992, 114, 10730 ± 10738.
[
[
1
9] L. B. Knight, Jr., C. A. Arrington, B. W. Gregory, S. T. Cobranchi, S.
5
a
Liang, L. Paquette, J. Am. Chem. Soc. 1987, 109, 5521 ± 5523.
[
10] Z.-Z. Qin, A. D. Trifunac, P. E. Eaton, Y. Xiong, J. Am. Chem. Soc.
3
1
991, 113, 669 ± 670.
[11] Z.-Z. Qin, A. D. Trifunac, P. E. Eaton, Y. Xiong, J. Am. Chem. Soc.
1990, 112, 4565 ± 4567. Top half of Figure 1 (first spectrum).
[12] A. Marcinek, J. Rogourski, J. Gebicki, G.-F. Chen, F. Williams, J. Phys.
Chem. A 2000, 104, 5265 ± 5268.
(
was irradiated under argon at � 408C with a 150 W lamp (maximum
emission at 300 nm) for 10 h. Acetonitrile was removed in vacuo
(
200 mbar), the residue was then diluted with water (10 mL) and extracted
[13] T. Bally, S. Bernhard, S. Matzinger, J.-L. Roulin, G. N. Sastry, L.
Truttmann, Z. Zhu, A. Marcinek, J. Adamus, R. Kaminski, J. Gebicki,
with 2-methylbutane (5 Â 30 mL). The combined extracts were washed
Chem. Eur. J. 2001, 7, No. 13
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0713-2743 $ 17.50+.50/0
2743

P. R. Schreiner, A. A. Fokin et al.
FULL PAPER
F. Williams, G.-F. Chen, M. P. Fülscher, Chem. Eur. J. 2000, 6, 858 ±
[33] I. R. Gould, D. Ege, J. E. Moser, S. Farid, J. Am. Chem. Soc. 1990, 112,
4290 ± 4301.
[34] A. Albini, M. Mella, M. Freccero, Tetrahedron 1994, 50, 575 ± 607.
[35] M. Mella, M. Freccero, A. Albini, J. Chem. Soc. Chem. Commun. 1995,
41 ± 42.
[36] M. Mella, M. Freccero, A. Albini, Tetrahedron 1996, 52, 5533 ± 5548.
[37] M. Mella, M. Freccero, A. Albini, Tetrahedron 1996, 52, 5549 ± 5562.
[38] J. Heinze, Angew. Chem. 1984, 96, 823 ± 840; Angew. Chem. Int. Ed.
Engl. 1984, 23, 831 ± 847.
[39] Gaussian 98 (Revision A.7), M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A.
Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M.
Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,
S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K.
Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.
Foresman, J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C.
Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen,
M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle, J. A.
Pople, Gaussian, Inc., Pittsburgh PA, 1998.
8
68.
14] H. E. Zimmerman, G. L. Grunewald, J. Am. Chem. Soc. 1966, 88,
83 ± 184.
[
[
[
1
15] A. A. Fokin, P. R. Schreiner, P. v. R. Schleyer, P. A. Gunchenko, J.
Org. Chem. 1998, 62, 6494 ± 6502.
16] A. A. Fokin, P. A. Gunchenko, S. A. Peleshanko, P. v. R. Schleyer,
P. R. Schreiner, Eur. J. Org. Chem. 1999, 855 ± 860.
17] R. G. Parr, W. Yang, Annu. Rev. Phys. Chem. 1995, 46, 701 ± 728.
18] T. Bally, G. N. Sastry, J. Phys. Chem. A 1997, 101, 7923 ± 7925.
19] B. Braida, P. C. Hiberty, A. Savin, J. Phys. Chem. A 1998, 102, 7872 ±
[
[
[
7877.
[
[
[
20] M. Sodupe, J. Bertran, L. Rodriguez-Santiago, E. J. Baerends, J. Phys.
Chem. A 1999, 103, 166 ± 170.
21] J. H. Wang, A. D. Becke, V. H. Smith, J. Chem. Phys. 1995, 102, 3477 ±
3
480.
22] D. Cremer in The Encyclopedia of Computational Chemistry, Vol. 3
Eds.: P. v. R. Schleyer, N. L. Allinger, T. Clark, J. Gasteiger, P. A.
(
Kollman, H. F. Schaefer III, P. R. Schreiner), Wiley, Chichester, 1998,
pp. 1706 ± 1735.
[
[
[
[
[
23] G. E. Scuseria, A. C. Scheiner, T. J. Lee, J. E. Rice, H. F. Schaefer, J.
Chem. Phys. 1987, 86, 2881 ± 2890.
24] M. Urban, J. Noga, S. J. Cole, R. J. Bartlett, J. Chem. Phys. 1985, 83,
[40] C. Gonzalez, H. B. Schlegel, J. Chem. Phys. 1989, 90, 2154 ± 2161.
[41] K. A. Brueckner, Phys. Rev. 1954, 96, 508 ± 516.
[42] N. C. Handy, J. A. Pople, M. Head-Gordon, K. Raghavachari, G. W.
Trucks, Chem. Phys. Lett. 1989, 164, 185 ± 192.
[43] D. E. Woon, T. H. Dunning, Jr., J. Chem. Phys. 1993, 98, 1358 ± 1371.
[44] T. H. Dunning, Jr., J. Chem. Phys. 1989, 90, 1007 ± 1023.
[45] L. Cassar, P. E. Eaton , J. Halpern, J. Am. Chem. Soc. 1970, 92, 3515 ±
3518.
4
041 ± 4046.
25] R. J. Bartlett, J. D. Watts, S. A. Kucharski, J. Noga, Chem. Phys. Lett.
990, 165, 513 ± 522.
1
26] A. C. Scheiner, G. E. Scuseria, J. E. Rice, T. J. Lee, H. F. Schaefer III,
J. Chem. Phys. 1987, 87, 5361 ± 5373.
27] J. F. Stanton, J. Gauss, J. D. Watts, R. J. Bartlett, J. Phys. Chem. 1991,
94, 4334 ± 4345.
[
[
[
28] G. D. Purvis, R. J. Bartlett, J. Chem. Phys. 1982, 76, 1910 ± 1918.
29] H. D. Roth, P. S. Lakkaraju, J. Phys. Chem. 1993, 97, 13403 ± 13407.
30] C. J. Rhodes, C. Glidewell, J. Chem. Soc. Perkin Trans. 2 1992, 1481 ±
[46] L. Cassar, P. E. Eaton, J. Halpern, J. Am. Chem. Soc. 1970, 92, 6366 ±
6368.
[47] H. E. Zimmerman, R. W. Binkley, R. S. Givens, G. L. Grunewald,
M. A. Sherwin, J. Am. Chem. Soc. 1969, 97, 3316 ± 3323.
1
486.
31] T. Bally, L. Truttmann, S. Dai, F. Williams, J. Am. Chem. Soc. 1995, 117,
916 ± 7922.
32] C. J. Rhodes, J. Chem. Soc. Chem. Commun. 1990, 592 ± 593.
[
[
7
Received: November 27, 2000 [F2902]
2744
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0713-2744 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 13