Tris(1-naphthyl)- and tris(2-naphthyl)methyl cations: Highly crowded triarylmethyl cations

Article abstract of DOI:10.1039/a804505d

The highly hindered tris(1-naphthyl)methyl and tris(2-naphthyl)methyl cations 3 and 5 have been prepared under long lived stable ion conditions and characterized by ¹³C NMR spectroscopy at low temperatures. The latter can abstract hydride from cycloheptatriene to afford tris(2-naphthyl)methane 7, but the reaction of cycloheptatriene with the more crowded cation 3 failed to give the corresponding hydrocarbon. A primary kinetic isotope effect, k_H/k_D = 7.1 ± 0.5, found in the former case supports a colinear hydride abstraction mechanism. The ions 3 and 5 upon quenching rearranged to give a mixture of isomeric 13-(1-naphthyl)-1,2,7,8-dibenzofluorene 4 and 13-(2-naphthyl)-2,3,6,7-dibenzofluorene 6, respectively.

Full text of DOI:10.1039/a804505d

View Article Online / Journal Homepage / Table of Contents for this issue
Tris(1-naphthyl)- and tris(2-naphthyl)methyl cations: highly
crowded triarylmethyl cations1
George A. Olah,*a Qimu Liao,a J. Casanova,b Robert Bau,a Golam Rasula and
G. K. Surya Prakash*a
a Donald P. and Katherine B. Loker Hydrocarbon Research Institute, Department of Chemistry,
University of Southern California, Los Angeles, CA 90089-1661, USA
b Department of Chemistry, California State University, Los Angeles, CA 90032-8202, USA
Received (in Gainesville, FL) 12th June 1998, Accepted 22nd July 1998
The highly hindered tris(1-naphthyl)methyl and tris(2-naphthyl)methyl cations 3 and 5 have been prepared under
long lived stable ion conditions and characterized by 13C NMR spectroscopy at low temperatures. The latter can
abstract hydride from cycloheptatriene to afford tris(2-naphthyl)methane 7, but the reaction of cycloheptatriene
with the more crowded cation 3 failed to give the corresponding hydrocarbon. A primary kinetic isotope effect,
kH/kD = 7.1 0.5, found in the former case supports a colinear hydride abstraction mechanism. The ions 3 and
5 upon quenching rearranged to give a mixture of isomeric 13-(1-naphthyl)-1,2,7,8-dibenzofluorene 4 and
13-(2-naphthyl)-2,3,6,7-dibenzofluorene 6, respectively.
not give the desired salt of 3, tris(1-naphthyl)methyl fluoro-
borate or perchlorate. Instead 13-(1-naphthyl)-1,2,7,8-dibenzo-
fluorene 4 was obtained in quantitative yield.
Successful preparation of tris(1-naphthyl)methyl cation 3
was subsequently achieved under stronger superacidic condi-
tions by the reaction of tris(1-naphthyl)methanol 1 with
FSO3H–SO2ClF at low temperature (Ϫ78 to Ϫ20 ЊC). When 1
was treated with a mixture of FSO3H and SO2ClF at Ϫ78 ЊC, a
dark blue solution resulted.
Introduction
Stable triarylmethyl cations have been widely studied. Previ-
ously known triarylmethyl cations are quite stable, even at
room temperature.2 The structures of several such cations
have been established by X-ray crystallography and by NMR
studies.3,4 The only example, however, of a sterically over-
crowded triarylmethyl cation known to us is tris(1-azulenyl)-
methyl cation,5 which is stable enough to be isolated at room
temperature.
With our interest in the chemistry and structure of carbo-
cations, we have sought to investigate crowded triarylmethyl
carbocations that may exhibit unusual stability or reactivity. So
far the most highly crowded trialkylmethyl cation reported, the
tris(1-adamantanyl)methyl cation, was prepared under stable
ion conditions in our laboratory.6 Despite extensive studies on
triarylmethyl cations, no investigation of tris(naphthyl)methyl
cations has yet been reported. It was of interest to us to estab-
lish the stability of tris(naphthyl)methyl cations in relation to
triphenylmethyl cation and determine whether overcrowding
would significantly diminish aromatic delocalization through
rotation of the aryl π-system out of the carbocationic plane. In
continuation of our studies, we wish to report here the prepar-
ation of tris(1-naphthyl)methyl and tris(2-naphthyl)methyl
cations and two very crowded triarylmethyl cations. The low
temperature 13C NMR studies on both of them, their reactivity,
and their tendency for rearrangement proved to be unusual and
interesting.
FSO3H / SO2ClF
-78 to -20oC
3
3
C+
COH
1
3
4
Scheme 1
The 13C NMR spectrum at Ϫ40 ЊC showed that ion 3 had
been formed. The chemical shift of the characteristic carbo-
cationic center was δ 13C 198.6, shielded from Cϩ of the
triphenylmethyl cation by 14 ppm. This indicates enhanced
charge dispersal into the electron-rich naphthalene rings. In
addition to the cationic carbon resonance, a total of 29 13C
NMR peaks (see experimental section for multiplicity assign-
ment) were observed between δ 13C 124.2 and 146.6. These data
clearly indicate that carbocation 3 exists as two conformational
isomers with overlapping carbocationic centers (at least at 50
MHz 13C spectrometer frequency). We suggest that these iso-
mers are the symmetrical and unsymmetrical helical conform-
ational isomers. The MMX8 calculated energy barrier between
them is substantial, and they would not be expected to inter-
convert at a measurable rate, even at room temperature.
Results and discussion
The preparation of tris(1-naphthyl)methyl and tris(2-naph-
thyl)methyl cations was undertaken by the ionization of the
corresponding alcohols, tris(1-naphthyl)- and tris(2-naphthyl)-
methanols 1 and 2. Alcohols 1 and 2 were obtained by the reac-
tion of 1-naphthoyl chloride and 1-naphthyllithium and the
reaction of 2-naphthoyl chloride and 2-naphthyllithium,
respectively.
Initially the usual methods used for obtaining triphenyl-
methyl cation7 were attempted to prepare the tris(1-naphthyl)-
methyl cation from the alcohol 1. When 1 was treated with
HBF4 or HClO4 in acetic anhydride at 20 ЊC, the reaction did
Carbocation 3 is considerably less stable than the triphenyl-
J. Chem. Soc., Perkin Trans. 2, 1998, 2239–2242
2239

View Article Online
Fig. 2 X-Ray structure of 4sp. Hydrogen atoms are omitted for clarity.
Bond distances (Å) and bond angles (Њ): C1–C2, 1.383 (17), C2–C3, 1.406
(16), C3–C4, 1.431 (17), C4–C5, 1.438 (18), C5–C6, 1.444 (15), C6–C7,
1.404 (17), C7–C8, 1.437 (16), C8–C9, 1.400 (20), C9–C10, 1.381 (22), C10–
C5, 1.406 (18), C11–C12, 1.379 (17), C12–C13, 1.439 (17), C13–C14, 1.362
(18), C14–C15, 1.436 (21), C15–C16, 1.452 (18), C16–C17, 1.435 (18), C17–
C18, 1.413 (16), C18–C19, 1.477 (21), C19–C20, 1.382 (23), C20–C15, 1.444
(16), C21–C22, 1.390 (16), C22–C23, 1.406 (16), C23–C24, 1.358 (18), C24–
C25, 1.415 (18), C25–C26, 1.457 (14), C26–C27, 1.422 (15), C27–C28, 1.386
(16), C28–C29, 1.447 (18), C29–C30, 1.386 (16), C30–C25, 1.422 (15), C1–
C31, 1.546 (14), C11–C31, 1.507 (15), C1–C2–C3, 123.3 (1.2), C2–C3–C4,
117.1 (1.2), C3–C4–C5, 120.3 (1.2), C4–C5–C6, 120.1 (1.2), C5–C6–C7,
120.8 (1.2), C6–C7–C8, 118.6 (1.2), C7–C8–C9, 118.9 (1.4), C8–C9–C10,
123.2 (1.4), C9–C10–C5, 119.2 (1.4), C10–C5–C6, 119.3 (1.3), C10–C5–C4,
120.4 (1.2), C1–C6–C7, 121.3 (1.1), C11–C12–C13, 124.7 (1.1), C11–C12–C2,
108.5 (1.0), C12–C13–C14, 115.6 (1.3), C13–C14–C15, 123.1 (1.2), C14–C15–
C16, 119.9 (1.2), C14–C15–C20, 120.4 (1.4), C20–C15–C16, 119.7 (1.5), C15–
C16–C11, 116.9 (1.3), C15–C16–C17, 119.5 (1.1), C11–C16–C17, 123.5 (1.2),
C16–C17–C18, 121.0 (1.3), C17–C18–C19, 118.0 (1.5), C18–C19–C20, 122.1
(1.4), C19–C20–C15, 119.6 (1.5), C6–C1–C31, 128.0 (1.0), C2–C1–C31, 110.7
(0.9), C1–C2–C12, 107.7 (1.1), C3–C2–C12, 128.6 (1.2), C31–C21–C22, 117.9
(1.1), C26–C21–C22, 120.9 (1.0), C21–C22–C23, 120.7 (1.2), C22–C23–C24,
121.4 (1.2), C23–C24–C25, 119.0 (1.1), C24–C25–C26, 121.3 (1.2), C26–C25–
C30, 117.6 (1.2), C25–C26–C21, 116.6 (1.1), C25–C26–C27, 118.7 (1.1), C26–
C27–C28, 121.9 (1.2), C27–C28–C29, 121.0 (1.3), C28–C29–C30, 117.4 (1.3),
C29–C30–C25, 123.3 (1.2), C1–C31–C11, 100.9 (0.8), C11–C31–C21, 113.7
(0.9), C1–C31–C21, 108.9 (0.8).
Fig. 1 Space-filling structures of 3–6 computed using MMX in PC
Model 4.3 and drawn using Chem 3D plus 2.0.1.
The striking contrast of intrinsic stability between the tri-
phenylmethyl cation and cation 3 suggests significant steric
deformation in the latter from the carbocationic plane. We cal-
culated the preferred geometry of the conformers using the
molecular mechanics program.8 Two conformers were indicated
as the most stable forms of cation 3, a symmetrical helical con-
former (C3v symmetry) and an unsymmetrical helical con-
former (C1 symmetry). Examination of the bond rotation
necessary to interconvert these conformers using a rigid sphere
approximation with the molecular mechanics calculation leads
to the conclusion that the conformers cannot interconvert at or
near room temperature (Ea in excess of 100 kcal molϪ1). Hence
this pair of cations [3(C3v) and 3(C1)] constitute a pair of con-
formational isomers. Molecular mechanics calculations for the
helical conformations led us to estimate a p–pπ interorbital
angle of 50Њ between the contiguous cationic and the aromatic p
orbitals, suggesting poor overlap and thus explaining the
decreased stability with respect to the triphenylmethyl cation.
Interconversion of the α- and β-helical conformers requires
cooperative rotation and may itself be a highly unfavorable
process, probably inaccessible at room temperature. Regarding
the formation of 4 upon quenching ion 3, it is reasonable to
suggest that 2(C3v) and 2(C1) are converted stereoselectively
to 4sp and 4ap, respectively. A mechanism involving Friedel–
Crafts cycloalkylation and 1,3-sigmatropic rearrangement
would rationalize the transformation. Attempts to quench 3
with hydride donors such as cycloheptatriene or triethyl-
methyl cation; when the solution of 3 was quenched with
ice–water, it gave 13-(1-naphthyl)-1,2,7,8-dibenzofluorene 4
through 1,5-cyclization. Such cyclization in triphenylmethyl
cation to yield 9-phenylfluorene has been previously recog-
nized.9 Quenching of the cation 3 at Ϫ78 ЊC also resulted in 4
(56% yield) along with the starting alcohol 1. To rule out any
contamination of isomeric protonated 13-(1-naphthyl)-1,2,7,8-
dibenzofluorene 4 in the ionization of 1, we independently
protonated 4 in FSO3H–SO2ClF at Ϫ78 ЊC. Under these condi-
tions no discernible peaks were observed in the 13C NMR spec-
trum indicating extensive decomposition. Thus we believe that
1,5-cyclization of 3 to give 4 takes place only during quenching.
Attempted hydride abstraction of cycloheptatriene with 3 to
give tropylium ion was unsuccessful as no reaction was
observed (vide infra).
Two conformational isomers, 4ap and 4sp of 13-(1-naphthyl)-
1,2,7,8-dibenzofluorene 4 (Fig. 1), were observed in the 1H
NMR spectrum of the mixture at room temperature. The
MMX calculated energy barrier between them is extremely high
and thus they cannot interconvert at room temperature.
The ratio of 4sp to 4ap is about 3:1. Isomer 4sp was success-
fully isolated as single crystals by recrystallization of the mix-
ture of 4 from acetone. The structure of 4sp was confirmed by
single crystal X-ray crystallographic analysis and is shown in
Fig. 2.
2240
J. Chem. Soc., Perkin Trans. 2, 1998, 2239–2242

View Article Online
silane to obtain the parent tris(1-naphthyl)methane were not
successful.
Tris(2-naphthyl)methyl cation 5 was also successfully pre-
large primary isotope effect10 supports the colinear nature of
the hydride transfer reaction and is in accord with Karabatsos’
studies on the related hydride transfer of triarylmethanes.11
pared by the ionization of tris(2-naphthyl)methanol 2 in
Ar
+
FSO3H / SO2ClF
-78 to -20oC
H
H
Ar
Ar
C+
COH
3
3
2
5
Experimental
All melting points are uncorrected. The NMR spectra were
recorded on a 200 MHz NMR spectrometer. Chemical shifts
are given in ppm. Mass spectra were recorded using a Finnigan
Incos-50 GC-MS instrument. FT-IR spectra were obtained
with a Perkin-Elmer 1550 spectrometer. The molecular mechan-
ics calculations were carried out with PC Model 4.3, Serena
Software.8
6
Preparation of tris(1-naphthyl)methanol 1
Scheme 2
To a stirred suspension of lithium powder (40 mmol) in dry
ether (10 ml), a solution of 1-naphthyl bromide (40 mmol) in
dry ether (30 ml) was added dropwise. The mixture reacted
exothermically and refluxed spontaneously. After the addition
was complete, the mixture was refluxed for another 0.5 h, gen-
erating a dark blue solution. The mixture was then cooled to
Ϫ78 ЊC and 1-naphthoyl chloride was added dropwise without
solvent. The mixture was refluxed for 10 min, and then poured
into ice–water. The organic layer was dried (MgSO4), filtered,
and evaporated. The residual solid was purified by column
chromatography on silica gel (2:1 hexane–CH2Cl2). Crystalliz-
ation from carbon tetrachloride gave 1 as white solid (50%
overall), mp 218–219 ЊC. δH(CDCl3–TMS) 4.35 (s, 1H), 7.05–
8.26 (m, 21H). δC(CDCl3–TMS) 87.6, 124.5, 125.2, 125.4, 128.6,
128.9, 129.0, 129.3, 131.6, 135.2, 141.7. IR (KBr): 3609, 3087,
3045, 1635, 1618, 1598, 1577, 1559, 1541, 1522, 1507, 1005, 782,
777 cmϪ1; MS: m/e = 410 (Mϩ). Analysis, Found: C, 90.45, H,
5.46%; C31H22O, Calc.: C, 90.70, H, 5.40%.
FSO3H– SO2ClF at Ϫ78 ЊC. However, attempts to isolate either
the perchlorate or tetrafluoroborate salt of 5 at room temper-
ature were unsuccessful. Cation 5 shows the carbocationic cen-
ter peak at δ 13C 203.5 (deshielded from 3 by 4.9 ppm) indicat-
ing slightly less charge delocalization into the naphthalene
skeleton (in agreement with the 2-substitution). In addition to
the cationic center, 23 more peaks were observed for the skeletal
carbons. These data again suggest that 5 exists as symmetrical
and unsymmetrical helical conformers (Fig. 1). The MMX cal-
culated energy barrier between them is also high (ca. 85 kcal
molϪ1 using a rigid sphere model). Hence, the conformers
cannot interconvert either at modest or room temperature.
Quenching of 5 gave 13-(2-naphthyl)-2,3,6,7-dibenzofluorene
6 by cycloalkylation reaction. Again independent protonation
of 6 in FSO3H–SO2ClF at Ϫ78 ЊC did not give any discernible
protonated species ruling out any of its contamination in the
ionization of 2. Moreover quenching of 5 with water produces
the starting alcohol.
Preparation of tris(2-naphthyl)methanol 2
Unlike 4sp and 4ap, the rotamers 6sp and 6ap (Fig. 1) cannot
be separated and distinguished by NMR at room temperature.
The MMX calculated rotational energy barrier between them is
very low, 6–8 kcal molϪ1. Therefore it is suggested that the
rotamer interconversion of 6 is rather fast in solution at ambi-
ent temperature.
The procedure used was the same as for 1 except that 2-
naphthyl bromide and 2-naphthoyl chloride were used. Yield:
56%, mp 217–218 ЊC. δH(CDCl3–TMS) 2.91 (s, 1H), 7.47–7.92
(m, 21H). δC(CDCl3–TMS) 82.6, 126.2, 126.3, 126.4, 126.6,
126.8, 127.5, 127.9, 128.5, 132.6, 132.8, 143.8. IR (KBr): 3469,
3053, 2926, 1599, 1273, 1120, 821, 798 cmϪ1. MS: m/e: 410
(Mϩ). Analysis, Found: C, 90.37, H, 5.70%; C31H22O, Calc.: C,
90.70, H, 5.40%.
Interestingly, quenching of 5 in HBF4 in ether with cyclo-
heptatriene clearly gave tris(2-naphthyl)methane 7a in 90%
X
X
X
X
X
Preparation of tris(1-naphthyl)methyl cation 3
X
X
To a 5-mm NMR tube containing a cold (Ϫ78 ЊC) mixture of 1
(40 mg) in SO2ClF was slowly added a cold (Ϫ78 ЊC) solution
of FSO3H in SO2ClF. The 13C NMR spectrum was recorded at
Ϫ40 ЊC, using capillary acetone-d6 as the deuterium lock, as
well as the reference. δC 124.2 (d), 124.7 (d), 125.7 (d), 126.4 (d),
126.7 (d), 128.3 (d), 128.5 (d), 128.9 (d), 129.8 (d), 129.9 (d),
130.2 (d), 130.3 (d), 130.8 (d), 131.0 (d), 131.5 (s), 133.5 (s),
133.7 (s), 134.1 (s), 134.3 (s), 136.7 (s), 141.7, 142.6 (s), 142.7 (s),
142.9 (s), 143.4 (d), 144.4 (d), 146.6 (d), 146.2 (d), 146.6 (d),
198.6 (s).
X
–40 to 0oC
C+
3
3
CX
7a X = H; 7b X = D
5
Scheme 3
yield. This indicates less steric hindrance in 5 compared to 3 for
the approach needed for hydride abstraction. Furthermore,
upon treatment with perdeuterium labeled cycloheptatriene,
cation 5 gave monodeuterated 7b which was characterized by
1H and 2H NMR as well as mass spectrometry. This experiment
proves that the methine hydrogen of 7a comes directly from
cycloheptatriene. In order to better understand the nature of
the hydride abstraction, we also measured the kinetic deuterium
isotope effect by treating 5 with an equimolar mixture of
cycloheptatriene and cycloheptatriene-d8. Analysis of the
Preparation of tris(2-naphthyl)methyl cation 5
The procedure used was similar to that of 3 except that 2 was
used. δC 128.0 (d), 128.4 (d), 128.6 (d), 129.4 (d), 129.6 (d),
131.9 (d), 132.1 (s), 132.3 (s), 132.4 (s), 133.0 (d), 134.2 (d),
134.6 (d), 136.9 (s), 137.6 (s), 138.0 (s), 138.2 (s), 138.5 (s), 138.7
(s), 139.0 (s), 148.2 (d), 148.3 (d), 149.0 (d), 149.4 (d), 203.5 (s).
1
product mixture by H NMR revealed kH/kD = 7.1 0.5. This
J. Chem. Soc., Perkin Trans. 2, 1998, 2239–2242
2241

View Article Online
Preparation of 4
128.2, 132.3, 133.5, 141.1. IR (KBr): 3049, 1627, 1592, 1500,
1265, 1123 cmϪ1. MS: m/e: 395 (Mϩ).
A cold (Ϫ78 ЊC) solution of 3 in FSO3H–SO2ClF prepared
as above was warmed to Ϫ20 ЊC and quenched by plunging
it into ice–water. After workup, an isomeric mixture of 13-(1-
naphthyl)-1,2,7,8-dibenzofluorene (4) was obtained as white
solid, yield 98%, mp 236–238 ЊC. δH(CDCl3–TMS) 5.70 (s, 1/4
H), 6.08 (s, 3/4 H), 6.47–8.04 (m, 19H). δC(CDCl3–TMS) 47.0,
57.0, 118.5, 118.6, 123.0, 123.6, 124.1, 125.0, 125.5, 125.6,
125.7, 126.0, 126.2, 126.5, 127.0, 127.1, 128.3, 128.5, 128.7,
129.0, 129.9, 130.1, 130.6, 131.2, 131.8, 133.3, 133.4, 134.4,
137.1, 138.6, 139.4, 143.3, 146.4. IR (KBr): 3053, 3010, 1696,
1622, 1591, 1577, 1560, 1517, 1467, 1436 cmϪ1. MS: m/e: 392
(Mϩ), 265, 188. Analysis, Found: C, 94.64, H, 5.26%; C31H20,
Calc.: C, 94.86, H, 5.14%.
Kinetic isotope effect experiment
To a cold (Ϫ40 ЊC) suspension of 2 (20 mg, 0.05 mmol) in acetic
anhydride (5 ml) were added two drops of 48% aqueous HBF4
solution. The resulting deep blue mixture was treated with a
mixture of cycloheptatriene (5 mmol) and [1,2,3,4,5,6,7,7-
2H8]cycloheptatriene (5 mmol). The stirred mixture was kept at
Ϫ10 ЊC until the characteristic blue color of tris(2-naphthyl)-
methyl cation 5 vanished. After workup, the purified mixture of
1
7a and 7b was investigated by H NMR. The integration ratio
of the nonaromatic protons to the aromatic protons was found
to be 24:1. This translates to kH/kD of 7.1 0.5.
Quenching cation 3 with aqueous sodium hydroxide
X-Ray crystallography
A cold (Ϫ78 ЊC) solution of 3 in FSO3H–SO2ClF prepared as
above was quenched with 20% aqueous sodium hydroxide. The
resulting mixture was stirred at Ϫ20 ЊC for 30 min and warmed
to room temperature. After workup, 1 and 5 were isolated in 32
and 56% yield, respectively.
Single crystals of compound 4sp were obtained upon slow
evaporation of an acetone solution of 4 in a 5 mm NMR tube
for about one week. The X-ray data were collected at room
temperature using a Nicolet/Syntex P21 diffractometer and are
reported in Fig. 2.
Preparation of 6
Acknowledgements
To a cold (Ϫ20 ЊC) suspension of 2 (41 mg, 0.1 mmol) in diethyl
ether (5 ml) was added dropwise 10 drops of 50% HBF4 ether
solution. The resulting dark blue mixture was stirred at Ϫ10 ЊC
for 30 min, then warmed to room temperature. The character-
istic blue color vanished in 2 h. After workup, 13-(2-naphthyl)-
2,3,6,7-dibenzofluorene 6 was isolated as white solid, yield
100%, mp 222–224 ЊC, δH(CDCl3–TMS) 5.33 (s, 1H), 6.78–8.85
(m, 19H). δC(CDCl3–TMS) 56.0, 123.0, 125.0, 125.7, 126.1,
126.2, 127.1, 127.6, 127.7, 127.9, 128.3, 128.6, 128.8, 128.9,
132.7, 133.3, 134.3, 137.8, 137.9, 147.6. IR (KBr): 3050,
1690, 1610, 1570, 1425 cmϪ1. MS: m/e: 392 (Mϩ), 293, 265.
Analysis, Found: C, 94.70, H, 3.31%; C31H20, Calc.: C, 94.86,
H, 5.14%.
Support of our work by the National Science Foundation is
gratefully acknowledged. We thank Dr Robert Aniszfeld for his
help in MS analysis.
References
1 Stable carbocations. 307. For Part 306 see: G. K. S. Prakash,
G. Rasul, G. A. Olah, R. Liu and T. T. Tidwell, Can. J. Chem.,
in the press.
2 G. A. Olah and P. v. R. Schleyer, Carbonium Ions, Interscience,
New York, 1968, vol. I, p. 2.
3 A. H. Gomes de Mesquita, C. H. MacGillavry and K. Eriks, Acta
Crystallogr., 1965, 18, 437.
4 R. B. Moodie, T. M. Connor and R. Stewart, Can. J. Chem., 1959,
37, 1402; R. Dehl, W. R. Vaughan and R. S. Berry, J. Org. Chem.,
1959, 24, 1616; R. S. Berry, R. Dehl and W. R. Vaughan, J. Chem.
Phys., 1961, 34, 1460; D. E. O’Reilly and H. P. Leftin, J. Phys.
Chem., 1960, 64, 1555; G. A. Olah, J. Am. Chem. Soc., 1964, 86, 932;
G. A. Olah, E. B. Baker and M. B. Comisarom, ibid. 1964, 86, 1265;
I. I. Schuster, A. K. Colter and R. J. Kurland, J. Am. Chem. Soc.,
1968, 90, 4679.
5 S. Ito, N. Morita and T. Asao, Tetrahedron Lett., 1991, 32, 773.
6 G. A. Olah, G. K. S. Prakash and R. Krishnamurti, J. Am. Chem.
Soc., 1990, 112, 6422.
7 H. J. Dauben, L. R. Honnen and K. M. Harmon, J. Org. Chem.,
1960, 25, 1442.
Preparation of tris(2-naphthyl)methane 7a
To a cold (Ϫ40 ЊC) suspension of 2 (41 mg, 0.1 mmol) in acetic
anhydride (2 ml) were added two drops of 48% aqueous HBF4
solution. The resulting deep blue mixture was treated with
cycloheptatriene (0.15 mmol). The stirred mixture was kept
at Ϫ10 to 0 ЊC until the characteristic blue color of tris-
(2-naphthyl)methyl cation 5 vanished and precipitation of
tropylium fluoroborate commenced in 1 h. After workup, tris-
(2-naphthyl)methane (7a) was obtained as white solid (36 mg),
yield 90%, mp 56–58 ЊC. δH(CDCl3–TMS) 6.03 (s, 1H), 7.37–
7.85 (m, 21H). δC(CDCl3–TMS) 57.2, 125.8, 126.1, 127.6, 127.9,
128.0, 128.1, 128.2, 132.3, 133.5, 141.2. IR (KBr): 3048, 1620,
1600, 1504 cmϪ1. MS: m/e: 394 (Mϩ), 267, 133. Analysis, Found:
C, 94.21, H, 5.80%; C31H22, Calc.: C, 94.38, H, 5.62%.
8 PC Model 4.3, Serena Software, Box 3076, Bloomington, IN 47402-
3076, USA.
9 G. A. Olah and P. v. R. Schleyer, Carbonium Ions, Interscience,
New York, 1970, vol. II, p. 83.
10 T. H. Lowry and K. S. Richardson, Mechanism and Theory in
Organic Chemistry, Harper & Row, New York, 3rd edn., 1987,
p. 333.
Preparation of [1-2H]tris(2-naphthyl)methane 7b
11 E. J. Karabatsos and M. Tornaritis, Tetrahedron Lett., 1989, 30,
5733.
The same procedure was used as for 7a except that [1,2,3,4,
5,6,7,7-2H8]cycloheptatriene12 was used. Yield 92%, mp 57–
59 ЊC. δH(CDCl3–TMS) 7.38–7.85 (m, 21H). δD(CHCl3–CDCl3)
6.0 (s). δC(CDCl3–TMS) 125.7, 126.1, 127.6, 127.9, 128.0, 128.1,
12 C. Engdahl, G. Jonsall and P. Ahlberg, J. Am. Chem. Soc., 1983, 105,
891.
Paper 8/04505D
2242
J. Chem. Soc., Perkin Trans. 2, 1998, 2239–2242