Birch reduction of hexaphenyl- and pentaphenylbenzene and an X-ray crystallography and NMR spectroscopy study of cis- and epi-1,2,3,4,5,6- hexaphenylcyclohexane and of 2,3,5,6-tetraphenyl-1,1′-bicyclohexylidene: Cannizzaro's conundrum revisited

Article abstract of DOI:10.1002/chem.200701306

The Birch reduction of hexaphenylbenzene yields two isomers of 1,2,3,4,5,6-hexaphenylcyclohexane. The X-ray crystal structure of the all-cis isomer, 1, reveals that the severe steric crowding among the three axial phenyls is alleviated by a marked splaying out of those three aryl substituents relative to the positioning in a conventional chair structure. A second product, 2, was identified crystallographically and by NMR spectroscopy as the 1,3-diaxial-2,4,5,6-tetraequatorial (epi) isomer of hexaphenylcyclohexane, in which only five of the six additional hydrogen atoms are positioned on the same face of the C₆Ph₆ precursor. A variable-temperature NMR study of the all-cis isomer 1 yielded a chair-to-chair inversion barrier of ≈ 19 kcal mol^-1, which is somewhat higher than the previously reported values for all-cis-1,2,3,4,5,6-C₆H₆R₆ in which R=Me or CO₂Me. The possible relevance to Cannizzaro's 1854 report of a product with the formula (C₇H₆)_n is discussed. By contrast, Birch reduction of pentaphenylbenzene led to the formation of 2,3,5,6-tetraphenyl-1,1′-bi-cyclohexylidene.

Full text of DOI:10.1002/chem.200701306

DOI: 10.1002/chem.200701306
Birch Reduction of Hexaphenyl- and Pentaphenylbenzene and an X-ray
Crystallography and NMR Spectroscopy Study of cis- and epi-1,2,3,4,5,6-
Hexaphenylcyclohexane and of 2,3,5,6-Tetraphenyl-1,1’-bicyclohexylidene:
Cannizzaroꢀs Conundrum Revisited
John P. Grealis, Helge Müller-Bunz, Yannick Ortin, Mark Condell, Michael Casey,* and
[a]
MichaelJ. McGlinchey*
Abstract: The Birch reduction of hexa-
phenylbenzene yields two isomers of
1,2,3,4,5,6-hexaphenylcyclohexane. The
X-ray crystal structure of the all-cis
isomer, 1, reveals that the severe steric
crowding among the three axial phe-
nyls is alleviated by a marked splaying
out of those three aryl substituents rel-
ative to the positioning in a conven-
tional chair structure. A second prod-
uct, 2, was identified crystallographical-
ly and by NMR spectroscopy as the
1,3-diaxial-2,4,5,6-tetraequatorial (epi)
isomer of hexaphenylcyclohexane, in
which only five of the six additional hy-
drogen atoms are positioned on the
same face of the C6Ph6 precursor. A
variable-temperature NMR study of
the all-cis isomer 1 yielded a chair-to-
chair inversion barrier of
ꢀ19 kcalmolꢁ1, which is somewhat
higher than the previously reported
values for all-cis-1,2,3,4,5,6-C6H6R6 in
which R=Me or CO2Me. The possible
relevance to Cannizzaroꢀs 1854 report
of a product with the formula (C7H6)n
is discussed. By contrast, Birch reduc-
tion of pentaphenylbenzene led to the
formation of 2,3,5,6-tetraphenyl-1,1’-bi-
cyclohexylidene.
Keywords: aromatic interactions ·
Birch reduction
·
conformation
analysis · ring inversion
· steric
hindrance
Introduction
with NaOH and extracted with chloroform, a small amount
of a pure product (m.p. 278–2808C) could be obtained after
recrystallisation from benzene.[2] A molecular-weight deter-
mination by Rastꢀs method (depression of the freezing point
of camphor)[3] gave a value of 556—(C7H6)6 requires 540—
and, in conjunction with the microanalytical results, the data
led the authors to suggest that the material was “one of the
isomeric 1,2,3,4,5,6-hexaphenylcyclohexanes”. Subsequently,
in 1956, Gerrard and Kilburn noted that, when
[PhCH2CH(Ph)O]4Si was treated with HCl, a liquid with a
b.p. of 2298C at 0.03 mm Hg was formed; this was assigned
as “sym-hexaphenylcyclohexane”.[4] However, in the ab-
sence of clear spectroscopic or crystallographic evidence,
these claims for the existence of 1,2,3,4,5,6-hexaphenylcyclo-
hexane remain unproven.
In 1854, Stanislao Cannizzaro reported that the action of a
variety of compounds, including boron trifluoride, sulphuric
acid, phosphorus pentoxide or zinc chloride, on benzyl alco-
hol or benzyl ether gave a hydrocarbon with the formula
(C7H6)n.[1] Over the following 90 years, the question of the
identity of this material was the subject of more than 30
publications, and a comprehensive historical survey has
been provided.[2] We are aware of only 2 substantive reports
of the possible resolution of this problem: in 1941, Shriner
and Berger observed that, when a large quantity of benzyl
alcohol was heated for 12 h at 1808C with boric acid, treated
[a] J. P. Grealis, Dr. H. Müller-Bunz, Dr. Y. Ortin, M. Condell,
Dr. M. Casey, Prof. Dr. M. J. McGlinchey
School of Chemistry and Chemical Biology
University College Dublin
Our own involvement in this topic arose serendipitously
as a continuation of our earlier studies on the syntheses and
dynamic behaviour of sterically crowded organic and organ-
ometallic molecules,[5] in particular [Cr(CO)3
ACHTREUNG
Belfield, Dublin 4 (Ireland)
Fax : (+353)1-716-1178
1-ferrocenyl-2,3,4,5,6-penta(b-naphthyl)benzene.[7]
E-mail: mike.casey@ucd.ie
wished to investigate the possibility of incorporating an or-
ganoruthenium fragment directly into a polyphenylbenzene
michael.mcglinchey@ucd.ie
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
framework,[8,9] and the most common route to such [Ru(h6-
G
1552
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1552 – 1560

FULL PAPER
arene)Cl2]2 complexes involves the Birch reduction of an
arene to the corresponding 1,4-cyclohexadiene and subse-
quent treatment with hydrated ruthenium trichloride.[10] To
this end, hexaphenylbenzene and pentaphenylbenzene were
each subjected to the regular Birch conditions, and the re-
sulting mixtures were separated by careful column chroma-
tography. From the C6Ph6 reaction, two major products were
isolated and identified as isomers of 1,2,3,4,5,6-hexaphenyl-
cyclohexane. These molecules have been unambiguously
characterised by X-ray crystallography, and their dynamic
behaviour has been investigated by NMR spectroscopy.
Results and Discussion
Crystallographic studies: Treatment of hexaphenylbenzene
(HPB) with sodium/liquid ammonia/ethanol, containing
some THF to help solubilise the HPB, furnished two major
products, both of which yielded crystals suitable for X-ray
diffraction studies. As depicted in Scheme 1, the first com-
pound was identified as all-cis-1,2,3,4,5,6-hexaphenylcyclo-
hexane (1), with phenyl groups placed in alternate axial and
equatorial positions on the chair framework. Figure 1 illus-
trates howthe three axial phenyl groups in 1 find them-
selves in a spatially limited environment; to alleviate this
steric crowding, they are markedly splayed out, such that
the bonds linking these phenyl groups to the central ring
make angles of 114.7, 115.6 and 117.28 with the plane de-
fined by their attached cyclohexane ring carbon atoms
(Figure 2). Of course, in a perfect chair, these angles would
ꢁ
be 908. Concomitantly, the axial C H bonds are directed in-
ꢁ ꢁ ꢁ
wards by 78ꢂ38 such that the H C C H dihedral angles
Figure 1. a) Top viewof the molecular structure of all- cis-hexaphenylcy-
clohexane (1), with atom labelling, and b) the corresponding space-filling
model.
average 528. Moreover, the axial-phenyl-ring carbon bonds
(average 1.527 Š) are slightly longer than the corresponding
equatorial-phenyl-ring carbon bonds (average 1.511 Š). It is
also noticeable that the angles within the cyclohexane ring
are very different; for those carbon atoms bearing axial
ꢁ ꢁ
phenyl groups, the C C C angles average 107.38, whereas
ꢁ ꢁ
the C C C angles for ring carbon atoms attached to equa-
torial phenyl groups average 117.38. This latter observation
ꢁ ꢁ
parallels (but is noticeably greater than) the C C C internal
angle differences found in the central ring of all-cis-
1,2,3,4,5,6-hexamethylcyclohexane, which average 109.28
(for carbon atoms bearing axial methyl groups) and 114.78
(for carbon atoms bearing equatorial methyl groups).[11]
A second crystalline product, 2, isolated from the Birch
reduction of C6Ph6 was also characterised by X-ray diffrac-
Figure 2. Side viewof all- cis-hexaphenylcyclohexane (1) with only the
ipso-carbon atoms of the phenyl rings shown.
tion as an isomer of 1,2,3,4,5,6-
hexaphenylcyclohexane. How-
ever, in 2, only five of the six
additional hydrogen atoms are
positioned on the same face of
the cyclohexane ring, thus pro-
ducing a molecule with two
axial and four equatorial
phenyl groups, as shown in
Figure 3.
Scheme 1. Products of the Birch reduction of hexaphenylbenzene.
Chem. Eur. J. 2008, 14, 1552 – 1560
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1553

M. J. McGlinchey, M. Casey et al.
thoxycarbonyl)cyclohexane, C6H6
ACHTREUNG
sponding hexacarboxylic acid C6H6
AHCTREUNG
ꢁ ꢁ
these systems, the C C C angles within the cyclohexane
ring are larger for the carbon atoms bearing the equatorial
substituents. Another isomer of C6H6(CO2Me)6, in which
G
only one of the ester substituents occupies an axial site, has
also been structurally characterised; likewise, in 1-ethoxycar-
bonyl-2,3,4,5,6-penta(methoxycarbonyl)cyclohexane,
only
the ethyl ester is axial.[12] Interestingly, 1,2,3,4,5,6-hexa(iso-
propyl)cyclohexane has been crystallographically character-
ised as the all-trans isomer in which, perhaps surprisingly, all
of the alkyl groups occupy axial sites, probably because of
unacceptable torsional strain in the all-equatorial confor-
mer.[13] However, in all-trans-1,2,3,4,5,6-hexaethylcyclohex-
ane, the six alkyl groups occupy equatorial sites, which re-
sults in an essentially idealised chair structure with cyclohex-
ꢁ ꢁ
ane carbon–carbon distances of 1.552 Š and C C C angles
of 108.48.[14]
1
NMR spectroscopic data: The 600 MHz H NMR spectrum
of 2 in CD2Cl2 (Figure 4) confirms that the 1,3-diaxial-
Figure 3. a) Top viewof the molecular structure of 1,3-diaxial-2,4,5,6-tet-
raequatorial-hexaphenylcyclohexane (2), with atom labelling, and b) the
corresponding side viewwith only the ipso-carbon atoms of the phenyl
rings shown.
Figure 4. The cyclohexane region of the 600 MHz 1H NMR spectrum of
2.
The axial substituents in 2 are disposed in a 1,3 fashion,
thereby generating a system with Cs symmetry on the NMR
timescale. However, in the solid state, this mirror symmetry
is broken as the two axial phenyl groups have quite different
orientations relative to the central ring. Analogously to the
all-cis isomer 1, the bonds in 2 linking the axial phenyl
groups at the C1- and C3-positions on the central ring make
angles of 115.1 and 114.68 with the plane defined by the cy-
clohexane ring carbon atoms C1, C3 and C5. However, over-
all, there appears to be less steric strain in 2 than that en-
2,4,5,6-tetraequatorial conformation is retained in solution.
The single axial proton at the C2-position exhibits a J=
6.0 Hz triplet at d=4.45 ppm, thereby showing coupling to
its two equatorial neighbours, whereas the single axial
proton at the C5-position is a J=12.0 Hz triplet at d=
4.83 ppm coupled to the two adjacent axial hydrogen atoms.
The two equatorial protons H1 and H3 appear as a pseudo-
triplet at d=3.80 ppm with J=6.0 Hz couplings to the axial
hydrogen atoms at the C2/C6- and C2/C4-positions, respec-
tively; finally, the axial protons at the C4- and C6-positions
exhibit, at d=4.13 ppm, a doublet (J=12.0 Hz) of doublets
(J=6.0 Hz) that is appropriate for interactions with one
axial and one equatorial hydrogen atom. The 13C NMR
ꢁ ꢁ
gendered in 1, and the C C C angles within the cyclohex-
ane ring of 2 are somewhat less perturbed (Table 1).
A search of the Cambridge Crystallographic Database re-
vealed only three other hexasubstituted all-cis cyclohexanes:
1,2,3,4,5,6-hexamethylcyclohexane,[11]
1,2,3,4,5,6-hexa(me-
spectrum of 2 was readily assigned by standard H–13C 2D
1
techniques.
The barrier for the chair-to-chair inversion in cyclohexane
itself was evaluated as 10.8 kcalmolꢁ1 in a nowclassic paper
by Anet and Bourn in which coalescence of axial and equa-
torial proton sites in deuterium-decoupled [D11]cyclohexane
was monitored.[15] The corresponding value of 17.3 kcal
molꢁ1 for the ring flip in all-cis-1,2,3,4,5,6-hexamethylcyclo-
hexane was initially obtained from 1H NMR peak coales-
cence of the axial methyl resonances, but no mention was
Table 1. Bond angles in all-cis-1,2,3,4,5,6-hexaphenylcyclohexane (1), 1,3-
diaxial-2,4,5,6-tetraequatorial-hexaphenylcyclohexane (2) and 1,2,3,4,5,6-
hexaethylcyclohexane (7).
Angle [8]
1
2
7
ring C-C-C (axial)
108.2, 107.3,
106.7
109.5, 107.6
–
ꢁ ꢁ
ring C C C (equato-
rial)
117.9, 117.2,
117.1
117.1, 114.7, 114.7,
109.0
108.4
1554
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1552 – 1560

Synthesis of 1,2,3,4,5,6-Hexaphenylcyclohexanes
FULL PAPER
made of the cyclohexyl ring protons.[16] Subsequently, the
magnitude of this barrier was confirmed by 13C NMR spec-
troscopic data on both the methyl and ring carbon atoms,[17]
and values calculated by using empirical force fields have
also been reported.[18] The equilibrium ratios of eight geo-
metric isomers of 1,2,3,4,5,6-C6H6Me6 at 2508C have been
determined by gas chromatography.[19] The barrier for the
corresponding chair-to-chair interconversion process in all-
cis-1,2,3,4,5,6-hexa(methoxycarbonyl)cyclohexane was also
found to be 17 kcalmolꢁ1.[20]
The 1H NMR spectrum of all-cis-hexaphenylcyclohexane
(1) was recorded in deuteriochloroform at 11.7 T (500 MHz)
and exhibited a slightly broadened singlet in the aliphatic
region at room temperature. However, upon cooling of the
solution to ꢁ458C, the singlet gradually decoalescenced into
two triplets at d=4.53 and 4.58 ppm with 3JHH =4.5 Hz, a
value in the normal range for axial–equatorial couplings. By
contrast, the room-temperature 13C NMR spectrum at
125 MHz revealed cyclohexane ring carbon peaks at d=57.3
and 49.4 ppm, each of which broadened and disappeared
into the baseline at 908C; simulation of the variable-temper-
ature spectra yielded a cyclohexane ring inversion barrier of
19ꢂ0.5 kcalmolꢁ1. The relatively small increase in the barri-
er for ring inversion in 1 over the previously reported value
of ꢀ17 kcalmolꢁ1 for both the hexamethyl and hexaester
analogues described above is perhaps somewhat surprising
considering the very severe steric crowding observed in 1.
However, even though the transition state for the chair-to-
chair inversion of 1 will undoubtedly be extremely congest-
ed, we note that the ground-state energy must also be
raised, and the barrier value derived by NMR methods is
merely the difference between them.
Molecular modelling of 1,2,3,4,5,6-hexaphenylcyclohexane
isomers: The structures, relative energies and interconver-
sions of 1,2,3,4,5,6-cyclohexanes bearing six equivalent
groups, C6H6R6, have long fascinated the stereochemical
community.
Pioneering
studies
on the
inositols,
C6H6(OH)6,[21] and on the isomers of hexachlorocyclohex-
ane[22] are nowclassic. Subsequently, elegant work by Farina
and co-workers elucidated the relative stabilities of the anal-
Figure 5. Calculated heats of formation (DH0f ) of isomers of 1,2,3,4,5,6-
hexaphenylcyclohexane.
ogous hexaesters, C6H6(CO2Me)6, and revealed that the
G
base-promoted epimerisation followed the sequence cis!
C6H6Ph6 (which possesses a single axial phenyl substituent)
lies only 7.8 kcalmolꢁ1 above the ground state, and this
structure has been observed by X-ray crystallography in
epi!mucoQchiroQmyoQscyllo.[12c]
In an attempt to estimate the strain inherent in the cis
and epi isomers of 1,2,3,4,5,6-hexaphenylcyclohexane (1 and
2, respectively), the heats of formation of the eight possible
hexahydro derivatives of hexaphenylbenzene were evaluat-
ed.[23] Thus, there is one possible isomer of C6Ph6H6 for ad-
dition of all six hydrogen atoms to the same face, and like-
wise for a 5:1 facial distribution; however, placement of
only two or three hydrogen atoms on the same face gives
rise to three isomers each, as illustrated in Figure 5, which
also lists the appropriate nomenclature for such systems.
Not surprisingly, the scyllo (all-equatorial) structure is the
most favourable, as in C6H6Et6,[14] and is taken as the zero to
which all the others are compared. The myo isomer of
some C6H6
(CO2R)6 cases.[12] Interestingly, in the C6H6Ph6
A
system, it transpires that the muco and allo isomers prefer
to adopt the twist-boat structure. Experimentally, however,
the apparently extremely disfavoured cis structure, which
lies more than 42 kcalmolꢁ1 above the ground state, is the
major product of the Birch reduction of hexaphenylbenzene,
with the epi isomer at almost 23 kcalmolꢁ1 above the
ground state as the only other isolable product.
Other all-cis products, such as the 1,2,3,4,5,6-hexaester-
and hexamethyl-cyclohexane derivatives mentioned previ-
ously, were prepared either from stereochemically rigid pre-
cursors or by catalytic hydrogenation on a metal surface,
Chem. Eur. J. 2008, 14, 1552 – 1560
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1555

M. J. McGlinchey, M. Casey et al.
whereby location of all six hydrogen atoms on the same face
is readily explicable. At first sight, it is difficult to rationalise
the formation of 1 and 2 from hexaphenylbenzene during
the course of a Birch reduction, which presumably proceeds
through successive additions of electrons and protons.[24] As-
suming that the first stereodefining event is the second pro-
tonation reaction leading to either cis- or trans-hexaphenyl-
1,4-dihydrobenzene, one might envisage that the precursor
anion, being extensively delocalised, is a relatively flat spe-
cies and that protonation trans to the phenyl substituent at
the one tetrahedral centre is kinetically favoured because
the phenyl group effectively shields one of the faces. While
it has been reported that Birch reduction of 1,4-diphenyl-
55.88.[30] These values compare favourably with the opti-
mised C3h geometry of the cyclic benzene trimer, which is
calculated to have ring centroid separations of 4.8 Š.[31,32]
Gratifyingly, the corresponding distances between the cent-
roids of the three axial phenyl substituents in 1 are 4.89,
4.90 and 5.15 Š, with dihedral angles of 73.6, 77.4 and 27.48,
values that again suggest an attractive interaction between
the aryl rings.
To probe the generality of this unexpected addition of six
hydrogen atoms to a multiply substituted arene under Birch
conditions, several other aromatic systems were investigated.
Not unexpectedly, the electron-rich systems of 1,3,5-trime-
thoxybenzene and 1,3,5-triferrocenylbenzene[33] were unre-
sponsive and were recovered unchanged. However, as
shown in Scheme 2, Birch reduction of 1,3,5-triphenylben-
zene yielded two known isomeric 1,3,5-triphenylcyclohex-
benzene yields
a substantial proportion of the trans
isomer,[25] the presence of the four additional bulky phenyl
substituents in the present substrate would presumably en-
hance the stereoselectivity for
the cis product. Moreover,
under these low-temperature
conditions and in a strongly
solvating medium, it is likely
that protonation is irreversible,
that is, that the reaction is
under kinetic control. Once
the first cis protonation has oc-
curred, subsequent proton
transfers would be even more
likely 1) to be under kinetic
control and 2) to occur selec-
tively on the more exposed
face, that is, cis to the already
present hydrogen atoms.
An additional factor which
may partly compensate for the
increasing steric crowding that
develops as the reaction pro-
ceeds is the potential for at-
tractive edge-to-face aromatic
interactions. As emphasised by
Scheme 2. Products of the Birch reductions of 1,3,5-triphenylbenzene and of pentaphenylbenzene.
Jennings et al.,[26] this phenomenon may play a greater role
than is generally appreciated in such diverse areas as protein
folding,[27] host–guest binding in supramolecular assem-
blies[28] and other molecular-recognition processes.[29] Experi-
mental and theoretical estimates indicate that these interac-
tions are energetically attractive by approximately 1.5–
2.0 kcalmolꢁ1[26] and are more likely to be evident in solu-
tion at lowtemperatures or in the crystalline state, when
conformational entropy effects are minimised. In support of
such a hypothesis, we note that the space-filling view of 1
(Figure 1b) reveals that the three axial phenyl groups exhib-
it two edge-to-face orientations and one offset-parallel-
stacking interaction.
anes: 1,3,5-e,e,e-C6H9Ph3 (3) was readily identified by the
1
simplicity of its H and 13C NMR spectra, whereas the corre-
sponding 1,3,5-a,e,e isomer 4 was characterised by X-ray
crystallography, the results of which are depicted in
1
Figure 6. The H NMR spectrum of 4 at room temperature
exhibited slightly broadened signals that did not decoalesce
fully even at ꢁ808C, a result suggesting that chair-to-chair
interconversion of the e,e,a and a,a,e isomers has a lowbar-
rier. This observation is consonant with a previous report on
the NMR spectra of 3 and 4 derived from the electrochemi-
cal reduction of 1,3,5-triphenylbenzene.[34]
The Birch reduction of pentaphenylbenzene required the
presence of additional THF to help solubilise the arene. Al-
though the reaction proceeded only to the extent of ꢀ5%
conversion, the product, 5, was obtained in ꢀ95% yield,
based on C6Ph5H consumed. The unambiguous identifica-
tion of 5 was secured by X-ray crystallography: 5 was re-
vealed to be a 1,1’-bicyclohexylidene in which one of the cy-
Interestingly, the very recently reported structure of an
approximately C3-symmetric benzene trimer (as part of a
supramolecular trimeric zinc–porphyrin complex) exhibits
ring centroid separations of 5.04, 4.93 and 5.33 Š, with dihe-
dral angles between the phenyl rings of 82.0, 75.3 and
1556
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1552 – 1560

Synthesis of 1,2,3,4,5,6-Hexaphenylcyclohexanes
FULL PAPER
Figure 6. Top and side views of the structure of a,e,e-1,3,5-triphenylcyclo-
hexane.
Figure 7. Side viewof 2,3,5,6-tetraphenyl-1,1 ’-bicyclohexylidene (5), with
only the ipso-carbon atoms of the phenyl rings shown, and a top view of
the corresponding space-filling model.
clohexylidenes bears four phenyl groups, all on the same
face of the ring (Figure 7). It is evident that the peripheral
ortho- and meta-phenyl groups of the C6Ph5H precursor
have not been reduced, whereas both the central ring and
the para-phenyl substituent each nowhave five additional
hydrogen atoms. As with cis-C6Ph6H6, all of the additional
hydrogen atoms are situated on the same face of the central
ring. The 1H and 13C NMR spectra of 5 were readily as-
signed by conventional two-dimensional techniques.
in 6 (average 1.521 Š). The double bond between the two
cyclohexylidenes has lengthened from 1.339 Š in 6 to
1.346 Š in 5, but the metric parameters of the hydrogenated
cyclohexylidene ring in 5 showlittle deviation from those of
the parent compound.
Conclusion
It is instructive to compare the structure of 5 with that of
the parent 1,1’-bicyclohexylidene, H10C6=C6H10 (6), which
has approximate C2h symmetry.[35] In the tetraphenylated cy-
The Birch reaction of hexaphenylbenzene with sodium in
liquid ammonia does not yield a 1,4-cyclohexadiene but
rather reduces the central ring to yield primarily all-cis-
1,2,3,4,5,6-hexaphenylcyclohexane (1), along with a lesser
amount of the 1,3-diaxial-2,4,5,6-tetraequatorial isomer 2.
The formation of these sterically crowded cis and epi mole-
cules, rather than the thermodynamically favoured all-equa-
torial (scyllo) isomer, raises interesting mechanistic prob-
lems. These products, quite apart from their intrinsic struc-
tural interest, are also noteworthy because they have been
the source of some controversy for more than 150 years.
While the identity of Cannizzaroꢀs original product remains
an enigma, if it were indeed one of the isomers of
1,2,3,4,5,6-hexaphenylcyclohexane, it may have been the en-
ergetically favoured all-equatorial structure, which apparent-
ly is not produced under the low-temperature Birch condi-
tions, in which kinetic factors presumably dominate. In the
total absence of any structural or spectroscopic data from
ꢁ ꢁ
clohexylidene ring of 5, the C C C angles at carbon atoms
C2 and C6, which bear axial phenyl groups, are 108.2 and
109.98; the corresponding angles in 6 are 111.9 and 112.38.
ꢁ ꢁ
Moreover, the C C C angles at carbon atoms C3 and C5,
which bear equatorial phenyl groups, are 113.1 and 113.48,
somewhat larger than the corresponding angles (111.1 and
111.28) found in 1,1’-bicyclohexylidene itself. As in the hexa-
phenylcyclohexanes 1 and 2, the axial phenyl groups in 5 are
splayed out and the angles made by the bonds linking them
to the cyclohexylidene ring with the plane defined by the
ring carbon atoms C2, C4 and C6 are 104.8 and 106.98. In 6,
the corresponding angles for bonds linking hydrogen atoms
to the cyclohexylidene ring are 90.3 and 92.08, values very
close to those expected for the perfect chair. The carbon–
carbon bond lengths in the phenylated bicyclohexylidene
ring in 5 (average 1.538 Š) are somewhat longer than those
Chem. Eur. J. 2008, 14, 1552 – 1560
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1557

M. J. McGlinchey, M. Casey et al.
Data for 2: M.p. 216–2178C; 1H NMR (CD2Cl2, 600 MHz): d=7.38 (2H,
d, J=7.5 Hz; H52, H56), 7.15 (4H, d, J=7.5 Hz; H42, H46, H62, H66),
7.06 (4H, d, J=7.0 Hz; H12, H16, H32, H36), 7.08–7.05 (1H, m; H24),
7.03 (2H, t, J=7.5 Hz; H53, H55), 6.98 (2H, t, J=7.0 Hz; H14, H34),
6.93 (4H, t, J=7.0 Hz; H13, H15, H33, H35), 6.94–6.91 (2H, m; H23,
H25), 6.91 (4H, t, J=7.5 Hz; H43, H45, H63, H65), 6.87 (1H, t, J=
7.5 Hz; H54), 6.81 (2H, t, J=7.5 Hz; H44, H64), 6.41 (2H, d, J=7.5 Hz;
H22, H26), 4.83 (1H, t, J=12 Hz; H5), 4.45 (1H, t, J=6 Hz; H2), 4.13
(2H, dd, J=6, 12 Hz; H4, H6), 3.80 ppm (2H, t, J=6 Hz; H1, H3);
13C NMR (CD2Cl2, 150 MHz): d=144.0 (C51), 142.1 (C41, C61), 142.0
(C21), 139.1 (C11, C31), 133.3 (C12, C16, C32, C36), 131.9 (C22, C26),
130.4 (C42, C46, C52, C56, C62, C66), 127.4 (C43, C45, C63, C65), 127.3
(C23, C25), 126.8 (C13, C15, C33, C35, C53, C55), 126.7 (C24), 126.3
(C14, C34), 126.0 (C54), 125.5 (C44, C64), 57.0 (C4, C6), 56.7 (C2), 54.7
the earlier work, the only viable historical links are the
melting points of the products. In all of the studies over the
past 150 years, the only reported melting point (278–2808C)
is that of the crystalline product reported by Shriner and
Berger;[2] clearly, this does not correlate with those found
here (190–1918C for cis-1, 216–2178C for epi-2), thereby
showing that these products are different from the material
reported in 1941.
Extension of this reaction to 1,3,5-triphenylbenzene and
pentaphenylbenzene reveals that, under these conditions,
the expected 1,4-hexadienes are not the observed products.
Current studies are focussed on Birch reductions of other
hindered arenes to investigate the generality of these obser-
vations.
(C1, C3), 43.7 ppm (C5); elemental analysis: calcd for C42H36·
C 92.04, H 7.02; found: C 91.92, H 7.29; IR (KBr): n˜ =3056, 3023, 2923,
2850, 1598, 1492, 1448 cmꢁ1
G
.
Birch reduction of 1,3,5-triphenylbenzene: In similar fashion to the
method described above, ammonia (100 mL) was condensed onto 1,3,5-
triphenylbenzene (5 g, 16.3 mmol) in EtOH (25 mL) and THF (30 mL).
Sodium (5.25 g, 228.3 mmol) was added over a period of 30 min. The re-
action mixture was allowed to reflux at ꢁ338C for 1 h, warmed to room
temperature, hydrolysed with water and extracted with CH2Cl2. The sol-
vent was then removed, and the residue was recrystallised from hexane/
CH2Cl2 (50:50) to yield a yellowcrystalline product (0.83 g, 2.65 mmol;
16.3%). Further recrystallisations yielded the known compounds 1,3,5-
e,e,e-triphenylcyclohexane (3) and 1,3,5-a,e,e-triphenylcyclohexane (4),
the latter as X-ray quality crystals.
ExperimentalSection
General: 1H and 13C NMR spectra were recorded on Varian Inova 400,
500 or 600 MHz spectrometers. NMR simulations were carried out by
using the multisite EXCHANGE program generously provided by Pro-
fessor R. D. McClung (University of Alberta). All reactions were carried
out under an atmosphere of dry nitrogen. Elemental analyses were car-
ried out by the Microanalytical Laboratory at University College Dublin.
Data for 3: M.p. 74–778C (literature value:[34] 75–778C); 1H NMR
(CDCl3, 400 MHz): d=7.78–7.18 (15H, m; Ph H) 2.97 (3H, t, J=12 Hz;
H1a, H3a, H5a), 2.22 (3H, brd, J=12 Hz; H2e, H4e, H6e), 1.76 ppm
(3H, q, J=12 Hz; H2a, H4a, H6a).
Data for 4: M.p. 45–468C; 1H NMR (CDCl3, 500 MHz): d=7.49 (2H, d,
J=7.5 Hz; H12, H16), 7.38 (2H, t, J=7.5 Hz; H13, H15), 7.29 (4H, t, J=
7.5 Hz; H33, H35, H53, H55), 7.24 (4H, d, J=7.5 Hz; H32, H36, H52,
H56), 7.23 (1H, m; H14), 7.18 (2H, t, J=7.5 Hz; H34, H54), 3.46 (1H,
brs; H1e), 2.94 (2H, t, J=12.4 Hz; H3a, H5a), 2.56 (2H, d, J=13.5 Hz;
Birch reduction of hexaphenylbenzene: In a typical reaction, ammonia
(15 mL) was condensed onto hexaphenylbenzene (HPB) (1600 mg,
2.99 mmol) in EtOH (3 mL) and dry THF (20 mL). When sodium
(350 mg, 15.22 mmol) was added in small pieces over a period of 1 h, the
solution became light purple; it was then maintained at reflux at ꢁ338C
for 2 h before being allowed to warm to room temperature. Addition of
water and filtration led to the formation of a white solid that was washed
with Et2O to yield unreacted HPB (870 mg, 1.63 mmol; 54% recovery).
The organic layer was washed with water and dried over MgSO4 to yield
a white solid (670 mg, 1.24 mmol; 91% based on HPB consumed). Exten-
sive chromatographic investigation
Table 2. Crystallographic data for 1, 2, 4 and 5.
revealed that optimal separation was
achieved on a long, thin silica column
by using exactly 7% diethyl ether in
pentane as the eluent. This procedure
yielded two white crystalline prod-
ucts, 1 (308 mg, 0.57 mmol; 46%) and
1
2
4
5
empirical formula
C42H36
A
C24H24
312.43
C36H36
468.65
M
540.71
crystal system
space group
a [Š]
b [Š]
c [Š]
triclinic
P1 (no. 2)
triclinic
P1 (no. 2)
orthorhombic
Cmc21 (no. 36)
18.529(6)
12.284(4)
7.531(2)
90
triclinic
P1 (no. 2)
¯
¯
¯
2
(214 mg, 0.40 mmol; 32%), each
10.4765(13)
11.9433(15)
13.6128(17)
99.031(2)
109.107(2)
105.647(2)
1492.4(3)
2
1.203
100(2)
0.068
44.00
11.4883(14)
12.1143(15)
12.4502(16)
91.267(2)
92.944(2)
113.853(2)
1580.9(3)
1
8.6361(13)
8.7002(13)
18.212(3)
102.079(3)
91.945(3)
108.257(3)
1263.4(3)
2
suitable for an X-ray crystallography
study.[36]
Data for 1: M.p. 190–1918C;
1H NMR (CDCl3, 500 MHz): d=
7.14–7.07 (12H, m; equatorial Ph
ortho- and meta-H), 7.05–7.01 (3H,
m; axial Ph para-H), 6.94–6.88 (3H,
m; equatorial Ph para-H), 6.79 (6H,
t, J=7.5 Hz; axial Ph meta-H), 6.73
(6H, d, J=7.5 Hz; axial Ph meta-H),
a [8]
b [8]
90
90
g [8]
V [Š3]
Z
1714.1(9)
4
1calcd [gcmꢁ3
T [K]
]
1.214
1.211
1.232
100(2)
0.069
100(2)
0.068
100(2)
0.069
m [mmꢁ1
2qmax [8]
]
4.50 ppm
(6H,
brs;
H1–H6);
52
50
52
13C NMR (CDCl3, 125 MHz): d=
143.0 (axial Ph ipso-C), 139.8 (equa-
torial Ph ipso-C), 133.3 (axial Ph
ortho-C), 130.0 (equatorial Ph ortho-
C), 127.9 (equatorial Ph meta-C),
127.7 (axial Ph para-C), 126.7 (axial
Ph meta-C), 126.5 (equatorial Ph
para-C), 57.3 (C1, C3, C5), 49.3 ppm
(C2, C4, C6); elemental analysis:
reflections measured
8502
12808
3708
10578
reflections used (Rint
parameters
)
3638 (0.0298)
379
6113 (0.0216)
570
841 (0.0502)
115
4927 (0.0278)
469
final R values (I>2s(I)):
R1
wR2
0.0543
0.1278
0.0454
0.1199
0.0401
0.0920
0.0624
0.1467
R values (all data):
R1
0.0703
0.1376
1.081
0.0568
0.1359
1.063
0.0460
0.0950
1.078
0.0763
0.1537
1.079
wR2
calcd for C42H36·
A
GoF on F2
7.68; found: C 89.16, H 7.59.
1558
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1552 – 1560

Synthesis of 1,2,3,4,5,6-Hexaphenylcyclohexanes
FULL PAPER
H2e, H6e), 2.07 (1H, m, J=12.4 Hz; H4e), 2.03 (2H, td, J=13.5, 5 Hz;
H2a, H6a), 1.73 ppm (1H, q, J=12.4 Hz; H4a); 13C NMR (CDCl3,
125 MHz): d=146.0 (C31, C51), 142.8 (C11), 127.6 (C13, C15), 127.5
(C33, C35, C53, C55), 126.6 (C12, C16), 125.4 (C32, C36, C52, C56),
125.2 (C34, C54), 124.7 (C14), 41.32 (C4), 38.23 (C3, C5), 36.44 (C1),
[7] L. E. Harrington, J. F. Britten, M. J. McGlinchey, Can. J. Chem.
2003, 81, 1180–1186.
[8] There is a report of a complex [Ru
A
G
N
the metal is thought to be bonded to the central ring, but no X-ray
crystallographic data are available: M. Crocker, M. Green, A. G.
Orpen, D. M. Thomas, J. Chem. Soc. Chem. Commun. 1984, 1141–
1144.
36.17 ppm (C2,C6); IR (CH2Cl2): n˜ =3064, 3029, 1600, 1494, 1452 cmꢁ1
.
Birch reduction of pentaphenylbenzene: In similar fashion to the method
described above, ammonia (18 mL) was condensed onto pentaphenylben-
zene (4 g, 8.73 mmol) in EtOH (4 mL) and THF (100 mL). The reaction
mixture was then treated with sodium (0.95 g, 41.32 mmol). Hydrolysis
and extraction with CH2Cl2 yielded a yellowsolid (3.59 g). Recrystallisa-
tion of the crude product from Et2O furnished white crystals of 2,3,5,6-
tetraphenyl-1,1’-bicyclohexylidene (5; 0.21 g, 0.44 mmol; 5.1%).
Data for 5: M.p. 236–2378C; 1H NMR (CDCl3, 500 MHz): d=7.45 (4H,
d, J=7.5 Hz; H32, H36, H52, H56), 7.27 (4H, t, J=7.5 Hz; H33, H35,
H53, H55), 7.14 (2H, t, J=7.5 Hz; H34, H54), 6.76 (4H, d, J=6.5 Hz;
H22, H26, H62, H66), 6.68 (2H, t, J=6.5 Hz; H24, H64), 6.63 (4H, t, J=
6.5 Hz; H23, H25, H63, H65), 5.01 (2H, d, J=5.0 Hz; H2e, H6e), 3.49
(2H, td, J=12.0, 4.0 Hz; H3a, H5a), 3.26 (1H, q, J=12.8 Hz; H4a), 2.79
(1H, dt, J=12.7, 3.3 Hz; H4e), 2.66–2.74 (4H, m; H2’a, H2’e, H6’a,
H6’e), 1.78–1.82 ppm (6H, m; H3’a, H3’e, H4’a, H4’e, H5’a, H5’e);
13C NMR (CDCl3, 125 MHz): d=144.2 (C31, C51), 141.2 (C21, C61),
135.5 (C1’), 134.8 (C1), 129.6 (C22, C26, C62, C66), 128.3 (C33, C35,
C53, C55), 127.3 (C23, C25, C63, C65), 126.9 (C32, C36, C52, C56), 125.8
(C34, C54), 124.5 (C24, C64), 46.7 (C3, C5), 44.6 (C2, C6), 32.0 (C2’,
C6’), 28.9 (C3’,C5’), 27.4 (C4’), 26.9 ppm (C4); IR (CH2Cl2): n˜ =3079,
3056, 2983, 1602, 1496 cmꢁ1; elemental analysis: calcd for C36H36: C 92.26,
H 7.74; found: C 92.24, H 7.82.
[9] Other organometallic complexes of hexaphenylbenzene include
a) [Co
Laschi, P. F. Zanazzi, J. Chem. Soc. Chem. Commun. 1989, 405–407;
b) [Mo
(h5-C5H5)(h6-C6Ph6)]: I. C. Quarmby, R. C. Hemond, F. J.
Feher, M. Green, W. E. Geiger, J. Organomet. Chem. 1999, 577,
189–196; c) [V(CO)3
(h6-C6Ph6)]: M. Schneider, E. Weiss, J. Organo-
met. Chem. 1976, 114, C43–C45; d) [Ag2(ClO4)2(C6Ph6)]: G. L. Ning,
A
G
A
ACHTREUNG
AHCTREUNG
A
N
ACHTREUNG
M. Munakata, L. P. Wu, M. Maekawa, Y. Suenega, T. Kuroda-Sowa,
K. Sugimoto, Inorg. Chem. 1999, 38, 5668–5673.
[10] a) R. A. Zelonka, M. C. Baird, Can. J. Chem. 1972, 50, 3063–3072;
b) M. A. Bennett, T.-N. Huang, T. W. Matheson, A. K. Smith, Inorg.
Synth. 1982, 21, 75–78; c) H. Le Bozec, D. Touchard, P. Dixneuf,
Adv. Organomet. Chem. 1989, 29, 163–247.
[11] H. van Koningsveld, J. M. A. Baas, B. van de Graaf, M. A. Hoefna-
gel, Cryst. Struct. Commun. 1982, 11, 1065–1071.
[12] a) S. Brückner, L. Malpezzi, M. Grassi, Cryst. Struct. Commun.
1982, 11, 1043–1048; b) S. Brückner, L. Malpezzi Giunchi, G. Di Sil-
vestro, M. Grassi, Acta Crystallogr. Sect. B: Struct. Crystallogr. Cryst.
Chem. 1981, 37, 586–590; c) M. Farina, M. Grassi, G. Di Silvestro, J.
Am. Chem. Soc. 1985, 107, 5100–5104.
[13] Z. Goren, S. E. Biali, J. Am. Chem. Soc. 1990, 112, 893–894.
[14] A. Immirzi, E. Torti, Atti Accad. Naz. Lincei, Cl. Sci. Fis. Mater.
Nat. Rend. 1968, 44, 98.
[15] F. A. L. Anet, A. J. R. Bourn, J. Am. Chem. Soc. 1967, 89, 760–768.
[16] H. Werner, G. Mann, M. Mühlstädt, H. J. Kçhler, Tetrahedron Lett.
1970, 3563–3566.
[17] H. Werner, G. Mann, H. Jancke, G. Engelhardt, Tetrahedron Lett.
1975, 1917–1920.
[18] B. Van de Graaf, J. M. A. Baas, H. A. Widya, Rec. Trav. Chim. Pays-
Bas 1981, 100, 59–61.
[19] S. S. Berman, V. A. Zakharenko, A. A. Petrov, Neftekhimiya 1969, 9,
500–505.
X-ray crystallography measurements for 1, 2, 4 and 5: Crystal data
(Table 2) were collected by using a Bruker SMART APEX CCD area de-
tector diffractometer, and absorption corrections were performed by
using the program SADABS.[37] The structures were solved by direct
methods with the SHELXS-97[38] software and refined by full-matrix
least-square calculations on F2 for all data with the SHELXL-97 soft-
ware.[39] CCDC 631099 (1), CCDC 631100 (2), CCDC 647380 (4) and
CCDC 647381 (5) contain the supplementary crystallographic data for
this paper. These data can be obtained free of charge from the Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[20] G. Gatti, M. Grassi, G. Di Silvestro, M. Farina, S. Brückner, J.
Chem. Soc. Perkin Trans. 2 1982, 255–258.
[21] a) S. E. Fletcher, Adv. Carbohydr. Chem. 1948, 3, 45–77; b) S. J.
Angyal, Quart. Rev. Chem. Soc. 1957, 11, 212–226; c) T. Posternak,
The Cyclitols, Hermann, Paris, 1965, pp. 8–10; d) S. Brownstein,
Can. J. Chem. 1962, 40, 870–874; e) S. J. Angyal, L. Odier, Carbo-
hydr. Res. 1982, 100, 43–54; f) S. J. Angyal, P. A. J. Gorin, M.
Pitman, Proc. Chem. Soc. 1962, 337–338.
[22] a) Roddꢀs Chemistry of Carbon Compounds, Vol. II (Ed.: S. Coffey),
Wiley, NewYork, 1984, p. 36, and references therein; b) K. Hayami-
zu, O. Yamamoto, K. Kushida, S. Satoh, Tetrahedron 1972, 28, 779–
790.
Acknowledgements
We thank Science Foundation Ireland and University College Dublin for
generous financial support and also the reviewers for their helpful and
perceptive comments.
[23] PC-Model, Version 7.5, Serena Software, Box 3076, Bloomington,
IN 47402–3076, 2000.
[1] a) S. Cannizzaro, Ann. Chem. Pharm. 1854, 90, 252–254; b) S. Can-
nizzaro, Ann. Chem. Pharm. 1854, 92, 113–117.
[24] J. March, Advanced Organic Chemistry: Reactions, Mechanisms and
Structure, 4th ed., Wiley, NewYork, 1992, pp. 781–783.
[25] R. G. Harvey, D. F. Lindlow, P. W. Rabideau, J. Am. Chem. Soc.
1972, 94, 5412–5420.
[26] W. B. Jennings, B. M. Farrell, J. F. Malone, Acc. Chem. Res. 2001, 34,
885–894, and references therein.
[27] a) S. K. Burley, G. A. Petsko, Science 1985, 229, 23–28; b) S. K.
Burley, G. A. Petsko, Adv. Protein Chem. 1988, 39, 125–189; c) E.
Kim, S. Paliwel, C. S. Wilcox, J. Am. Chem. Soc. 1998, 120, 11192–
11193.
[28] M. C. T. Fyfe, J. F. Stoddart, Acc. Chem. Res. 1997, 30, 393–401.
[29] a) C. A. Hunter, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112,
5525–5534; b) C. A. Hunter, Chem. Soc. Rev. 1994, 23, 101–109;
c) F. Cozzi, M. Cinquini, R. Annunziata, J. S. Siegel, J. Am. Chem.
Soc. 1993, 115, 5330–5331; d) S. Paliwel, S. Geib, C. S. Wilcox, J.
Am. Chem. Soc. 1994, 116, 4497–4498; e) F. Cozzi, R. Ponzini, R.
[2] R. L. Shriner, A. Berger, J. Org. Chem. 1941, 6, 305–318.
[3] L. F. Fieser, M. Fieser, Organic Chemistry, 3rd ed., Reinhold, New
York, 1956, p. 11.
[4] W. Gerrard, K. D. Kilburn, J. Chem. Soc. 1956, 1536–1539.
[5] a) S. Brydges, M. J. McGlinchey, J. Org. Chem. 2002, 67, 7688–7698;
b) S. Brydges, L. E. Harrington, M. J. McGlinchey, Coord. Chem.
Rev. 2002, 233–234, 75–105; c) H. K. Gupta, S. Brydges, M. J.
McGlinchey, Organometallics 1999, 18, 115–122; d) S. Brydges, J. F.
Britten, L. C. F. Chao, H. K. Gupta, M. J. McGlinchey, D. L. Pole,
Chem. Eur. J. 1998, 4, 1201–1205; e) M. J. McGlinchey, Adv. Orga-
nomet. Chem. 1992, 34, 285–325; f) B. Mailvaganam, C. S. Framp-
ton, B. G. Sayer, S. Top, M. J. McGlinchey, J. Am. Chem. Soc. 1991,
113, 1177–1185; g) P. A. Downton, B. Mailvaganam, C. S. Frampton,
B. G. Sayer, M. J. McGlinchey, J. Am. Chem. Soc. 1990, 112, 27–32.
[6] B. Mailvaganam, B. G. Sayer, M. J. McGlinchey, J. Organomet.
Chem. 1990, 395, 177–185.
Chem. Eur. J. 2008, 14, 1552 – 1560
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1559

M. J. McGlinchey, M. Casey et al.
Annunziata, M. Cinquini, J. S. Siegel, Angew. Chem. 1995, 107,
1092–1094; Angew. Chem. Int. Ed. Engl. 1995, 34, 1019–1020; f) K.
Nakamura, K. N. Houk, Org. Lett. 1999, 1, 2049–2051.
[35] N. Veldman, A. L. Spek, F. J. Hoogesteger, J. W. Zwikker, L. W. Jen-
neskens, Acta Crystallogr. Sect. C. 1994, 50, 742–744.
[36] As found during the collection of several X-ray data sets, the unit
cells contained nonstoichiometric quantities of diethyl ether solvent
molecules.
[30] T. Morimoto, H. Uno, H. Furuta, Angew. Chem. 2007, 119, 3746–
3749; Angew. Chem. Int. Ed. 2007, 46, 3672–3675.
[31] T. P. Tauer, T. D. Sherrill, J. Phys. Chem. A 2005, 109, 10475–10478.
[32] a) P. Hobza, H. L. Selzle, E. W. Schlag, J. Am. Chem. Soc. 1994, 116,
3500–3506; b) V. Spirko, O. Engkvist, P. Soldan, H. L. Selzle, E. W.
Schlag, P. Hobza, J. Chem. Phys. 1999, 111, 572–582.
[37] G. M. Sheldrick, SADABS, Bruker AXS Inc., Madison, WI 53711,
2000.
[38] G. M. Sheldrick, SHELXS-97, University of Gçttingen, Gçttingen,
1997.
[33] H. K. Gupta, N. Reginato, F. O. Ogini, S. Brydges, M. J. McGlinchey,
Can. J. Chem. 2002, 80, 1546–1554.
[39] G. M. Sheldrick, SHELXL-97–2, University of Gçttingen, Gçttin-
gen, 1997.
[34] H. Lund, Abstracts of Papers, 201st ACS National Meeting, Phila-
delphia, PA, 2002, http://www.electrochem.org/dl/ma/201/pdfs/
1246.pdf.
Received: August 22, 2007
Published online: December 18, 2007
1560
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1552 – 1560