Laser flash photolysis study of arylcyclopropylcarbenium ions: Cation stabilizing abilities of cyclopropyl and phenyl groups

Article abstract of DOI:10.1021/ja960268j

Arylcyclopropylcarbenium ions, Ar(c-Pr)CH⁺, were generated as transient intermediates by laser flash photolysis (LFP) of trans-2,3-diphenylaziridinimines of aryl cyclopropyl ketones in 2,2,2-trifluoroethanol (TFE). The carbocations are thought to arise by way of diazo compounds and carbenes. Rate constants for the unimolecular decay in TFE and for the bimolecular reaction with methanol in TFE were obtained for Ar(c-Pr)CH⁺ and for analogous arylphenylcarbenium ions, ArPhCH⁺. Within these series, the cation stabilizing abilities of cyclopropyl and phenyl groups are found to be similar in magnitude. However, cyclopropyl responds more strongly than phenyl to increasing electron demand. Hence cyclopropyl is superior to phenyl in cation stabilizing ability for Ar = Ph but inferior to phenyl for Ar = 4-MeOC₆H₄.

Full text of DOI:10.1021/ja960268j

+
+
J. Am. Chem. Soc. 1996, 118, 7473-7477
7473
Laser Flash Photolysis Study of Arylcyclopropylcarbenium
Ions: Cation Stabilizing Abilities of Cyclopropyl and Phenyl
Groups
Wolfgang Kirmse,*,† Birgit Krzossa,† and Steen Steenken‡
Contribution from the Fakulta¨t fu¨r Chemie, Ruhr-UniVersita¨t Bochum, D-44780 Bochum,
Germany, and Max-Planck-Institut fu¨r Strahlenchemie, D-45413 Mu¨lheim, Germany
ReceiVed January 25, 1996X
Abstract: Arylcyclopropylcarbenium ions, Ar(c-Pr)CH+, were generated as transient intermediates by laser flash
photolysis (LFP) of trans-2,3-diphenylaziridinimines of aryl cyclopropyl ketones in 2,2,2-trifluoroethanol (TFE).
The carbocations are thought to arise by way of diazo compounds and carbenes. Rate constants for the unimolecular
decay in TFE and for the bimolecular reaction with methanol in TFE were obtained for Ar(c-Pr)CH+ and for analogous
arylphenylcarbenium ions, ArPhCH+. Within these series, the cation stabilizing abilities of cyclopropyl and phenyl
groups are found to be similar in magnitude. However, cyclopropyl responds more strongly than phenyl to increasing
electron demand. Hence cyclopropyl is superior to phenyl in cation stabilizing ability for Ar ) Ph but inferior to
phenyl for Ar ) 4-MeOC6H4.
Introduction
to give eventually carbocations. The reported transient spectra
indicate that 3a and 4a were in fact generated by this technique.
However, solvents (1,2-dichloroethane or dichloromethane) and
nucleophiles (trialkylamines) had to be used which led to near-
diffusional rates (k ranging from 9 × 108 to 3 × 109 M-1 s-1).
Thus, differences in reactivity tend to be blurred, as was
observed for 4a and benzyl cations.
The effects of cyclopropyl and phenyl groups on the kinetic
and thermodynamic stability of carbocations have been a matter
of some dispute.1 The solvolysis rates of 1 exceed those of 2
In view of the limitations inherent to pulse radiolysis, an
alternative method, applicable in protic media, was clearly
desirable. The present work describes the flash photolytic
generation of arylcyclopropylcarbenium ions 3. The reaction
rates of 3 are compared with those of diarylcarbenium ions 4.
The k4/k3 rate ratios thus obtained are close to unity, with a
tendency to increase as the aryl group Ar becomes more electron
donating.
by factors of 3 × 102 (R ) Me) to 1 × 104 (R ) H).2 Ion
cyclotron resonance data in the gas phase3 and pKR values in
+
solution4 also support the notion that cyclopropyl is superior to
phenyl in stabilizing the positive charge. On the other hand,
the 13C chemical shifts of the cationic centers for phenyl- and
cyclopropyl-substituted carbocations (e.g., Ph2CH+ δ ) 198.7,
c-PrCH+Ph δ ) 225.7) indicate better delocalization by phenyl
than by cyclopropyl groups.5
Solvolysis rates are influenced by ground state effects of the
substrates (1 and 2). Charge distributions, as reflected by 13C
chemical shifts, may not correlate with thermodynamic and/or
kinetic stabilities of the carbocations. The discrepancies caused
by such effects should be resolved by measuring the absolute
rates for electrophilic reactions of the carbocations 3 and 4,
respectively. This approach was pioneered by Dorfman et al.
in their pulse radiolysis studies of 1-OH and 2-OH.6 The
primary species formed in pulse radiolysis is the solvent cation
radical which reacts with solutes in a dissociative charge transfer
† Ruhr-Universita¨t Bochum.
Results and Discussion
‡ Max-Planck-Institut fu¨r Strahlenchemie.
X Abstract published in AdVance ACS Abstracts, July 15, 1996.
(1) For a review, see: Olah, G. A.; Reddy, V. P.; Prakash, G. K. S.
Chem. ReV. 1992, 92, 69 (Section X).
(2) (a) R ) H: Shono, T.; Oku, A.; Oda, R. Tetrahedron 1967, 24, 421.
(b) R ) Me: Brown, H. C.; Ravindranathan, M.; Peters, E. N. J. Org.
Chem. 1977, 42, 1073; Brown, H. C.; Peters, E. N. J. Am. Chem. Soc. 1977,
99, 1712.
(3) Wolf, J. F.; Harch, P. G.; Taft, R. W.; Hehre, W. J. J. Am. Chem.
Soc. 1975, 97, 2902.
(4) Deno, N. C.; Richey, H. G., Jr.; Liu, J. S.; Lincoln, D. N.; Turner, J.
O. J. Am. Chem. Soc. 1965, 87, 4533.
A variety of diarylcarbenium ions have been generated by
laser flash photolysis (LFP) of diarylcarbinyl esters and ethers,
via photoheterolysis of the C-O bond.7 We used acetate
precursors to generate 4a-e in 2,2,2-trifluoroethanol (TFE). The
spectroscopic and kinetic data for 4a-c were in excellent
agreement with those reported in the literature (Table 1).
Unfortunately, analogous precursors failed to produce transient
(6) (a) DePalma, V. M.; Wang, Y.; Dorfman, L. M. J. Am. Chem. Soc.
1978, 100, 5416. (b) Dorfman, L. M.; DePalma, V. M. Pure Appl. Chem.
1979, 51, 123.
(5) (a) Olah, G. A.; White, A. M. J. Am. Chem. Soc. 1969, 91, 5801. (b)
Olah, G. A.; Prakash, G. K. S.; Liang, G. J. Org. Chem. 1977, 42, 2666.
S0002-7863(96)00268-5 CCC: $12 00 © 1996 American Chemical Society

+
+
7474 J. Am. Chem. Soc., Vol. 118, No. 32, 1996
Kirmse et al.
Table 1. Spectroscopic and Kinetic Data for Arylcyclopropylcarbenium Ions 3 and Arylphenylcarbenium Ions 4 in Oxygen-Saturated TFEa
3
4
Ar
λ
max (nm)
kTFE (s-1
)
kMeOH (M-1 s-1
3.2 × 107
)
λ
max (nm)
435
kTFE (s-1
)
kMeOH (M-1 s-1
4.2 × 107
)
a Ph
345
1.6 × 106
2.2 × 105
1.4 × 104
3.4 × 106
(3.2 × 106)b
2.8 × 105
b 4-MeC6H4
370
385
5.1 × 106
9.4 × 104
450
455
5.8 × 106
8.1 × 104
(2.7 × 105)c
1.5 × 103
c 4-MeOC6H4
(1.2 × 103)b
2.3 × 105
d 4-PhC6H4
e 1-naphthyl
440
500
2.2 × 105
5.1 × 106
515
575
5.2 × 106
1.9 × 105
4.8 × 106
2.2 × 105
4.9 × 106
a The precision of rate constants is ∼5%, the reproducibility ∼10%. b Kirmse, W.; Kilian, J.; Steenken, S. J. Am. Chem. Soc. 1990, 112, 6399.
c Reference 6a.
Scheme 1
nitrogen, followed by proton transfer from ROH to the carbenes
8, and (ii) protonation of 7 to give diazonium ions 9 which
lose nitrogen with formation of 3. The carbene route was
recently confirmed for the diaryl series where the lifetime of
the diazo compounds (∼102 s) excludes significant protonation
by ROH in the microsecond regime of LFP experiments.14 In
the case of 7, there is also evidence for intervention of the
carbenes 8 (see below). However, since the reaction rate of 7
in TFE is not known, the diazonium route to 3 remains a viable
alternative.
Preparative photolyses of 5 in MeOH or TFE afforded nearly
quantitative yields of E-stilbene (6), which undergoes light-
induced E h Z isomerization, cyclization, reduction, and
addition of ROH on prolonged irradiation. Decomposition of
the diazo compounds 7 leads to the ethers Ar(c-Pr)CHOR (57-
68% in MeOH, 41-57% in TFE), along with small amounts
(e5%) of 1-arylcyclobutenes. The latter are known to arise
by rearrangement of arylcyclopropylcarbenes 8.15 Photolysis
of 5c in cyclohexane afforded 43% of 1-(p-methoxyphenyl)-
cyclobutene. In protic solvents, reaction of the carbene 8 with
ROH predominates over the relatively “slow” ring expansion.16
LFP of 5 in TFE produced the permanent absorption of 6 at
300 nm as well as transient spectra at longer wavelengths, as
exemplified in Figure 1. The transient absorptions are assigned
to the carbocations 3 on the basis of the following evidence:
(i) The absorption maxima of 3a (345 nm) and 3b (370 nm) in
TFE are in excellent agreement with data from the literature
(343 and 366 nm, respectively, in fluorosulfonic acid at -75
°C).17 (ii) The transients are are not affected by the presence
of oxygen but are effectively quenched by nucleophiles accord-
ing to second-order kinetics. Even the most stable species, 3c,
was found to react with azide at a diffusion-controlled rate
(k ) 1.0 × 1010 M-1 s-1 in acetonitrile-water 2:1).
spectra of the arylcyclopropylcarbenium ions 3. (A weak signal
of 3c, the most stable species, was obtained by LFP of the
appropriate acetate.) In the diarylmethyl series, the quantum
yields of carbocations were found to increase linearly with their
8
+
pKR values. As a rule, replacement of phenyl with cyclopropyl
4,9
+
enhances the pKR of carbocations by ca. 2 units. Hence the
reluctance of 1-OAc (R ) H) to undergo photoheterolysis comes
as a surprise. We speculate that excitation of 1-OAc may lead
to rupture of C-C rather than C-O bonds. Cyclopropylmeth-
ylbenzene has long been known as a photolabile compound.10
For an alternative source of carbocations 3, we resorted to
the trans-2,3-diphenylaziridinimines 5 (Scheme 1). Analogues
of 5 were widely employed as diazo precursors in thermal
reactions (Eschenmoser fragmentation)11 but have rarely been
photolyzed.12 Precedent with N-alkyl- and N-arylaziridines
suggests the intermediacy of azomethine ylides.13 However,
no evidence for a two-step mechanism was found in the case
of 5.
In protic solvents (ROH), carbocations 3 can arise from diazo
compounds 7 by two routes: (i) light-induced extrusion of
As to the origin of 3, the following observations are
relevant: (i) The decay of 3 (inset a in Figure 1) indicates that
the cation is fully formed during the laser pulse (20 ns). There
is no evidence for a dark reaction, such as 7 f 9 f 3, after the
laser pulse. (ii) The yield of 6, as measured by the optical
density (OD) at 300 nm, increases linearly with the laser dose
(7) (a) McClelland, R. A.; Kanagasabapathy, V. M.; Steenken, S. J. Am.
Chem. Soc. 1988, 110, 6913. (b) McClelland, R. A.; Kanagasabapathy, V.
M.; Banait, N.; Steenken, S. J. Am. Chem. Soc. 1989, 111, 3966. (c)
McClelland, R. A.; Chan, C.; Cozens, F.; Modro, A.; Steenken, S. Angew.
Chem. 1991, 103, 1389; Angew. Chem., Int. Ed. Engl. 1991, 30, 1337. (d)
Cozens, F. L.; Mathivanan, N. McClelland, R. A.; Steenken, S. J. Chem.
Soc., Perkin Trans. 2 1992, 2083. (e) For a review, see: Das, P. K. Chem.
ReV. 1993, 93, 119.
(8) Bartl, J.; Steenken, S.; Mayr, H.; McClelland, R. A. J. Am. Chem.
Soc. 1990, 112, 6918.
(9) (a) Kerber, R. C.; Hsu, C.-M. J. Am. Chem. Soc. 1973, 95, 3239. (b)
Moss, R. A.; Shen, S.; Krogh-Jespersen, K.; Potenza, J. A.; Schugar, H. J.;
Munjal, R. C. J. Am. Chem. Soc. 1986, 108, 134.
(10) (a) Leermakers, P. A.; Vesley, G. F. J. Org. Chem. 1965, 30, 539.
(b) Turro, N. J.; Neckers, D. C.; Leermakers, P. A.; Seldner, D.; D’Angelo,
P. J. Am. Chem. Soc. 1965, 87, 4097.
(11) (a) Felix, D.; Mu¨ller, R. K.; Horn, U.; Joos, R.; Schreiber, J.;
Eschenmoser, A. HelV. Chim. Acta 1972, 55, 1276. (b) Felix, D.; Wintner,
C.; Eschenmoser, A. Org. Synth. 1988, Collect. Vol. 6, 679. (c) Kim, S.;
Cho, C. M. Tetrahedron Lett. 1994, 35, 8405; 1995, 36, 4845.
(12) (a) Maier, G.; Gebhardt, J. Results reported in Angew. Chem. 1988,
100, 317; Angew. Chem., Int. Ed. Engl. 1988, 27, 309. (b) Chen, N.; Jones,
M., Jr.; White, W. R.; Platz, M. S. J. Am. Chem. Soc. 1991, 113, 4981.
(13) For reviews, see: (a) Trozzolo, A. M.; Leslie, T. M.; Sarpotdar, A.
S.; Small, R. D.; Ferraudi, G. J.; DoMinh, T.; Hartless, R. L. Pure Appl.
Chem. 1979, 51, 261. (b) Nastasi, M.; Streith, J. In Rearrangements in
Ground and Excited States; Mayo, P., Ed.; Academic Press: New York,
1980; Vol. 3, p 451. (c) DoMinh, T. Res. Chem. Intermed. 1989, 12, 125.
See also: (d) Hermann, H.; Huisgen, R.; Ma¨der, H. J. Am. Chem. Soc.
1971, 93, 1779. (e) Bhattacharyya, K.; Ramaiah, D.; Das, P. K.; George,
V. M. J. Phys. Chem. 1986, 90, 3221.
(14) Kirmse, W.; Krzossa, B.; Steenken, S. Tetrahedron Lett. 1996, 37,
1197.
(15) (a) Petrellis, P. C.; Griffin, G. W.; Hendrick, M. E.; Jones, M., Jr.
J. Chem. Soc., Chem. Commun. 1972, 1002. (b) Moss, R. A.; Wetter, W.
P. Tetrahedron Lett. 1981, 22, 997.
(16) For kinetics of carbenic 1,2-shifts, see: (a) Liu, M. T. H. Acc. Chem.
Res. 1994, 27, 287. (b) Moss, R. A. Pure Appl. Chem. 1995, 67, 741.
(17) Sekuur, T. J.; Kranenburg, P. Spectrochim. Acta A 1973, 29, 803.

+
+
Laser Flash Photolysis of Arylcyclopropylcarbenium Ions
J. Am. Chem. Soc., Vol. 118, No. 32, 1996 7475
4d leads to F+ ) 2.48 (r ) 0.977) for 3a-d and F+ ) 4.14 (r
) 0.983) for 4a-d. The F+ values for nucleophilic capture of
3 and 4 differ in sign and ranking order from those reported for
solVolyses of 1 (R ) H, X ) OPNB, F+ ) -3.61)2a and 2 (R
) H, X ) OPNB, F+ ) -3.18),18 respectively. Since the less
reactive carbocations are formed faster, these changes are in
accordance with expectation.
Rates constants for nucleophilic capture (knuc) and solvolysis
(ksolv) can lead to divergent estimates for the relative stability
of carbocations. In terms of eqs 1-3, ∆Gqsolv is fully affected
TS2
S {TS } R+
8 P
(1)
(2)
1
∆Gqsolv ) Gq1 - GS
∆Gqnuc ) Gq2 - GR
Figure 1. Time-dependent absorption spectra obtained after laser
excitation (248 nm, 20 ns, 95 mJ/pulse) of 5b (0.10 mM) in oxygen-
saturated TFE. Inset a: decay of 3b, recorded at 370 nm. Inset b: yields
of 6 and of 3b (scaled by a factor of 5) as a function of the laser dose
(mJ/pulse), recorded 70 ns after LFP of 5b (0.097 mM) in oxygen-
saturated TFE.
(3)
+
by substrate (S) energies while product (P) energies influence
∆Gq but slightly. Therefore, the relative stability of car-
nuc
bocations is mirrored more accurately by ∆∆Gq than by
nuc
∆∆Gq
.
solv
According to Table 1, ∆Gqnuc is nearly the same for 3a (9.0
kcal/mol) and 4a (8.5 kcal/mol) whereas ∆Gq
for 1 (24.8
solv
kcal/mol)19 differs substantially from that for 2 (29.1 kcal/mol)20
(these data refer to R ) H, X ) ODNB, 80% acetone, 25 °C).
For an explanation, we focus on the great demand of carboca-
tions for conjugative stabilization which is not shared by the
neutral substrates. Optimal geometries, with bisected cyclo-
propane rings and coplanar phenyl groups, are rarely attained,
due to adverse steric interactions. The crystal structure of 10
shows that the plane of the cyclopropane ring bisects the
carbocation plane perfectly whereas the plane of the phenyl ring
is twisted by ∼27°.21 The phenyl twist angle (ΦPh) of 10 should
be regarded as an upper limit for 3a which lacks the electron-
donating HO group of 10. Analogously, we assume twist angles
of e27° for both phenyl groups of 4a. Some support comes
from the crystal structure of 11, which shows ΦClPh ) 41° and
ΦPh ) 16°, despite electron donation from R-Cl.22
Figure 2. Hammett plots of log kTFE versus σ+. The correlation lines
shown are for 3a-c (b, s) and 4a-c (0, - - -).
up to ∼40 mJ and then levels off, due to depletion of 5 (inset
b in Figure 1). Analogous plots for the yield of 3 are upward
curved at low laser doses, pointing to the biphotonic sequence
7 f 8 f 3.
The ionization of covalent substrates involves conformational
changes and bond breaking which occur in concert but can be
formally dissected. Simplified pictures for 1 and 2 are shown
in Scheme 2. Since the relative energies of the carbocations
3a and 4a agree closely, conformational changes, 1 f 1′ and 2
f 2′, must account for the difference in ∆Gqsolv of 1 and 2 (R
) H). Promotion of cyclopropylmethylbenzene (1, R ) X )
H) from its ground state (Φc-Pr ) 29°, ΦPh ) 91°) to the
geometry of 10 requires 9.5 kcal/mol, as calculated with the
MMX force field. The X-ray structure of diphenylmethane (ΦPh
) 64° and 71°)23 is only 0.3 kcal/mol above the MMX
In order to assess the reactivity of carbocations 3 and 4, rate
constants for the pseudo-unimolecular decay in TFE and for
the bimolecular reaction with methanol (e0.2 M) in TFE were
estimated (Table 1). For [MeOH] e 0.2 M, kobs correlates
linearly with [MeOH] while upward curvature is found at higher
concentrations. The rate constant ratios k4/k3 are close to unity
for most of the cations studied. With increasing rate of solvent
capture, k4/k3 increases, more strongly for the weak nucleophile
TFE than for the good nucleophile MeOH (the reactivity of
MeOH toward 3 and 4 exceeds that of TFE by factors of g102
if the molarity of neat TFE, 13.84 M, is taken into account).
These data indicate that cyclopropyl is superior to phenyl in
stabilizing “electron-poor” cations (3a vs 4a) while the reverse
holds for “electron-rich” cations (3c vs 4c). The different
electron-donating tendencies of cyclopropyl and phenyl groups
are also reflected by the reaction constants for 3a-c (F+ ) 2.64,
r ) 0.999) and 4a-c (F+ ) 4.34, r ) 0.996), obtained by
plotting log(kTFE) vs σ+ (Figure 2). The data for both 3d and
4d deviate from the correlation lines thus defined (a tentative
explanation is given below). Inclusion of the data for 3d and
(18) McLennan, D. J.; Martin, P. L. Aust. J. Chem. 1979, 32, 2361.
(19) Friedrich, E. C.; Taggart, D. B.; Saleh, M. A. J. Org. Chem. 1977,
42, 1437.
(20) Mindl, J.; Pivonka, P.; Vecera, M. Coll. Czech. Chem. Commun.
1972, 37, 2568.
(21) Childs, R. F.; Faggiani, R.; Lock, C. J. L.; Mahendran, M.; Zweep,
S. D. J. Am. Chem. Soc. 1986, 108, 1692.
(22) Laube, T.; Bannwart, E.; Hollenstein, S. J. Am. Chem. Soc. 1993,
115, 1731.
(23) Barnes, J. C.; Paton, J. D.; Damewood, J. R., Jr.; Mislow, K. J.
Org. Chem. 1981, 46, 4975.

+
+
7476 J. Am. Chem. Soc., Vol. 118, No. 32, 1996
Kirmse et al.
(OD) at 248 nm of 0.8-1.5/cm. The solutions were flowed through a
2 × 4 mm Suprasil quartz cell. The light-induced optical transmission
changes were digitized in parallel by Tektronix 7612 and 7912 transient
recorders interfaced with a DEC LSI11/73+ computer which also
process-controlled the apparatus and preanalyzed the data. Final data
analysis was performed on a Microvax I connected to the LSI.
Scheme 2
Preparation of trans-2,3-Diphenylaziridinimines 5. To a solution
of arylmagnesium bromide, obtained from the appropriate aryl bromide
(79 mmol) and magnesium turnings (2.0 g, 83 mmol) in ether (100
mL), was added dropwise a solution of cyclopropanecarbonitrile (5.0
g, 74.5 mmol) in ether (40 mL). The mixture was heated at reflux for
5 h. Methanol (18 mL) was then added at room temperature, and
stirring was continued for 30 min. With exclusion of moisture, the
precipitate of methoxymagnesium bromide was filtered off, and the
solution was concentrated in vacuo. The imine thus obtained was used
without purification.
minimum. Promotion of 2 (R ) X ) H) to phenyl twist angles
of 27° requires 12.3 kcal/mol, ca. 3 kcal/mol more than is needed
for 1. The difference in promotion energies increases, in favor
of 1, for smaller twist angles (e.g., ca. 6 kcal/mol for 25°). In
To a solution of the arylcyclopropylmethanimine (2.0 mmol) in
anhydrous benzene (10 mL) was added 1-amino-trans-2,3-diphenyl-
aziridine25 (0.42 g, 2.0 mmol). The mixture was stirred at room
temperature (for 5a-c) or at 35 °C (for 5d,e). When the evolution of
ammonia had ceased (2-4 h), the solution was concentrated in vacuo.
The residue was flash-chromatographed (neutral Al2O3, ether), and the
ether was evaporated in vacuo. HPLC (polygosil 60-5-NO2, ether-
hexane) afforded two isomers, A and B (presumably syn/anti), of each
summary, the difference in ∆Gq
of 1 and 2 (R ) H) is
solv
attributed to the following trends: (i) The bisected conformation
of the cyclopropane ring in 1 is easily attained. (ii) The
approach of phenyl groups to coplanarity is energetically less
demanding for 1 than for 2.
1
aziridinimine. The isomers are distinguished mainly by the H NMR
As discussed above, ksolv and knuc do not correlate for
substrates whose ionization involves variable changes in con-
formational energy. This principle should also apply to the
solvolysis of p-substituted cumyl chlorides from which the
signals assigned to 2-H and 3-H (δ 3.1-4.1).
5aA: 38%; oil; IR (film) 1603, 1495 cm-1; UV (MeCN) λmax
)
1
238 nm (ꢀ ) 7500); H NMR (CDCl3) δ 0.4-1.0 (m, 4 H), 1.5-1.9
(m, 1 H), 3.2 (br. s, 1 H), 3.6 (br. s, 1 H), 7.0-7.4 (m, 15 H). 5aB:
41%; oil; IR (film) 1602, 1495 cm-1; UV (MeCN) λmax ) 235 nm (ꢀ
) 7200); 1H NMR (CDCl3) δ 0.1-0.9 (m, 4 H), 2.2-2.5 (m, 1 H), 3.8
(br. s, 2 H), 6.9-7.5 (m, 15 H).
+
values of σp were derived.24 In order to achieve enhanced
conjugation, the biphenyl group of the cation 13 will be less
twisted than that of the substrate 12. As a consequence, the
log knuc data for biphenyl-4-ylcarbenium ions, such as 3d and
5bA: 44%; mp 87 °C; IR (KBr) 1601, 1450 cm-1; UV (MeCN)
+
4d, are poorly accomodated by σp (Figure 2).
1
λmax ) 249 (ꢀ ) 16 300), 308 nm (ꢀ ) 1900); H NMR (CDCl3) δ
0.4-1.0 (m, 4 H); 1.4-1.8 (m, 1 H), 2.30 (s, 3 H), 3.20 (br. s, 1 H),
3.50 (br. s, 1 H), 6.9-7.4 (m, 14 H). 5bB: 38%; oil; IR (film) 1603,
1
1451 cm-1; λmax ) 249 (ꢀ ) 17 200), 308 nm (ꢀ ) 2200); H NMR
(CDCl3) δ 0.0-0.9 (m, 4 H); 1.1-1.5 (m, 1 H), 2.30 (s, 3 H), 3.80 (br
s, 2 H), 6.9-7.5 (m, 14 H).
5cA: 29%; mp 58 °C; IR (KBr) 1601, 1507, 1448 cm-1; UV
1
Conclusion
(MeCN) λmax ) 247 nm (ꢀ ) 16 200); H NMR (CDCl3) δ 0.5-1.2
(m, 4 H), 1.5-1.8 (m, 1 H), 3.35 (br s, 1 H), 3.65 (br s, 1 H), 3.82 (s,
3 H), 6.80 (d, J ) 9 Hz, 2 H), 7.1-7.4 (m, 10 H), 7.45 (d, J ) 9 Hz,
2 H). 5cB: 59%; oil; IR (film) 1607, 1511, 1452 cm-1; UV (MeCN)
The flash photolytic generation and study of arylcyclopro-
pylcarbenium ions 3 has been achieved, using the aziridinimines
5 as precursors. In the series of secondary cations 3 and 4, the
stabilizing abilities of cyclopropyl and phenyl groups are found
to be similar in magnitude. Some variation in k4/k3 is observed,
however, since cyclopropyl responds more strongly to increasing
electron demand than phenyl. Divergent rate ratios for the
solvolytic generation and nucleophilic capture of 3 and 4 are
explicable in terms of the conformational changes that occur
on ionization of the substrates 1 and 2, respectively.
1
λmax ) 245 nm (ꢀ ) 15 800); H NMR (CDCl3) δ 0.1-1.0 (m, 4 H),
2.1-2.5 (m, 1 H), 3.80 (s, 3 H), 3.90 (br s, 2 H), 6.75 (d, J ) 9 Hz,
2 H), 7.08 (d, J ) 9 Hz, 2 H), 7.2-7.5 (m, 10 H).
5dA: 31%, mp 102 °C; IR (KBr) 1596, 1486, 1448 cm-1; UV
1
(MeCN) λmax ) 256 nm (ꢀ ) 24 000); H NMR (CDCl3) δ 0.5-1.1
(m, 4 H), 1.6-1.9 (m, 1 H), 3.30 (br s, 1 H), 3.50 (br s, 1 H), 7.1-7.6
(m, 19 H). 5dB: 37%; mp 51 °C; IR (KBr) 1600, 1486, 1449 cm-1
UV (MeCN) λmax ) 257 nm (ꢀ ) 23 400); H NMR (CDCl3) δ 0.0-
0.9 (m, 4 H), 2.1-2.5 (m, 1 H), 3.85 (br s, 2 H), 7.1-7.6 (m, 19 H).
;
1
5eA: 44%; oil; IR (film) 1602, 1497, 1451 cm-1; UV (MeCN) λmax
) 223 nm (ꢀ ) 53 600); 1H NMR (CDCl3) δ 0.5-1.0 (m, 4 H), 1.7-
2.0 (m, 1 H), 3.1 (br s, 2 H), 6.4-7.8 (m, 17 H). 5eB: 40%, mp 50
°C; IR (KBr) 1587, 1498, 1450 cm-1; UV (MeCN) λmax ) 221 nm (ꢀ
Experimental Section
General Methods. Melting points were determined on a Kofler
hot-stage apparatus and are uncorrected. 1H NMR spectra were
obtained at 80 (Bruker WP 80) and 400 MHz (Bruker AM-400).
Chemical shifts are reported in δ relative to tetramethylsilane. IR
spectra were recorded on a Perkin-Elmer 881 instrument. Gas
chromatography (GC) was performed by the use of a Siemens Sichromat
equipped with glass capillary columns. High-pressure liquid chroma-
tography (HPLC) was carried out with LDC (Milton Roy) chromato-
graphs and refractometric detection. 2,2,2-Trifluoroethanol (TFE) was
dried with Na2SO4 and then distilled over NaHCO3 to remove traces
of acid impurities.
1
) 54 600); H NMR (CDCl3) δ 0.0-1.0 (m, 4 H), 2.6-3.0 (m, 1 H),
3.8 (br d, J ) 5 Hz, 1 H), 4.1 (br d, J ) 5 Hz, 1 H), 6.72 (dd, J ) 7
and 2 Hz, 1 H), 7.2-7.8 (m, 16 H).
Photolyses of Aziridinimines 5. Solutions of 5 (0.75 mM) in
methanol or TFE were irradiated (120 W medium-pressure mercury
arc, quartz vessel, 20 °C). The product distributions were monitored
by GC (6.6 m OV1, 80 f 220 °C). At low conversions, the major
products were trans-stilbene (6) and the ether 14 or 15, depending on
the solvent (Table 2). After 1 h, ∼93% of 6 had been converted into
cis-stilbene (38%), phenanthrene (38%), 2-methoxy-1,2-diphenylethane
(6%),26 trans,trans,trans-tetraphenylcyclobutane (7%),27 and cis,trans,-
For the laser flash photolysis (LFP) experiments, we used a Lambda
Physik EMG103MSC excimer laser (KrF*) which emitted ≈20 ns
pulses of 248 nm light. The substrate solutions had optical densities
(24) Okamoto, Y.; Inukai, T.; Brown, H. C. J. Am. Chem. Soc. 1958,
80, 4972.
(25) Mu¨ller, R. K.; Joos, R.; Felix, D.; Schreiber, J.; Wintner, C.;
Eschenmoser, A. Org. Synth. 1988, Collect. Vol. 6, 56.

+
+
Laser Flash Photolysis of Arylcyclopropylcarbenium Ions
J. Am. Chem. Soc., Vol. 118, No. 32, 1996 7477
Table 2. Products (%) Obtained on Photolysis of
Aziridinimines 5a
(s, 3 H), 3.51 (d, J ) 7 Hz, 1 H), 3.82 (s, 3 H), 6.85 (d, J ) 9 Hz, 2
H), 7.22 (d, J ) 9 Hz, 2 H). 1H NMR (CDCl3) of 14d: δ 0.2-0.8 (m,
4 H), 1.1-1.5 (m, 1 H), 3.31 (s, 3 H), 3.60 (d, J ) 7 Hz, 1 H), 7.3-
7.7 (m, 9 H). 1H NMR (CDCl3) of 14e: δ 0.2-0.8 (m, 4 H), 1.3-1.7
(m, 1H), 3.29 (s, 3 H), 4.32 (d, J ) 8 Hz, 1 H), 7.4-8.3 (m, 7 H). 1H
NMR (CDCl3) of 15b: δ 0.2-0.7 (m, 4 H), 1.1-1.3 (m, 1 H), 2.35 (s,
3 H), 3.61 (dq, J ) 12.2 and 8.6 Hz, 1 H), 3.72 (dq, J ) 12.2 and 8.6
Hz, 1 H), 3.80 (d, J ) 7.8 Hz, 1 H), 7.16 (d, J ) 8.2 Hz, 2 H), 7.20
(d, J ) 8.2 Hz, 2 H). 1H NMR (CDCl3) of 15c: δ 0.2-0.7 (m, 4 H),
0.85 (m, 1 H), 3.61 (dq, J ) 12.5 and 8.5 Hz, 1 H), 3.69 (dq, J ) 12.5
and 8.5 Hz, 1 H), 3.77 (d, J ) 8.1 Hz, 1 H), 3.80 (s, 3 H), 6.89 (d, J
) 8.7 Hz, 2 H), 7.24 (d, J ) 8.7 Hz, 2 H). 1H NMR (CDCl3) of 15d:
δ 0.3-0.8 (m, 4 H), 1.2-1.3 (m, 1 H), 3.71 (dq, J ) 12.2 and 8.8 Hz,
1 H), 3.79 (dq, J ) 12.2 and 8.8 Hz, 1 H), 3.89 (d, J ) 8.0 Hz, 1 H),
7.3-7.5 (m, 4 H), 7.6 (m, 5 H). 1H NMR (CDCl3) of 15e: δ 0.2-0.8
(m, 4 H), 1.4-1.8 (m, 1 H), 3.7 (q, J ) 9 Hz, 2 H), 4.58 (d, J ) 8 Hz,
1 H), 7.4-8.4 (m, 7 H).
Minor amounts of the ketones 16a, 16b,29 16c,30 16d [mp 103 °C;
IR (KBr) 1660 cm-1; 1H NMR (CDCl3) δ 0.9-1.4 (m, 4 H), 2.5-2.9
(m, 1 H), 7.3-7.8 (m, 7 H), 8.10 (d, J ) 8 Hz)], and 16e31 were also
detected (Table 2). 1-Phenylcyclobutene (17a) was identified by
comparison with an authentic sample.32 Photolysis of 5c in cyclohexane
afforded 43% of 1-(4-methoxyphenyl)cyclobutene (17c): 1H NMR
(CDCl3) δ 2.50 (m, 2 H), 2.78 (m, 2 H), 3.82 (s, 3 H), 6.13 (t, J ) 2
Hz, 1 H), 6.85 (d, J ) 8 Hz, 2 H), 7.30 (d, J ) 8 Hz, 2 H). The
remaining cyclobutenes were tentatively assigned on the basis of their
retention times.
a Extrapolated to t ) 0.
cis-tetraphenylcyclobutane (4%).28 1H NMR (CDCl3) of 14a: δ 0.1-
1.4 (m, 5 H), 3.25 (s, 3 H), 3.55 (d, J ) 8 Hz, 1 H), 7.35 (br s, 5 H).
1H NMR (CDCl3) of 14b: δ 0.2-0.8 (m, 4 H), 1.0-1.4 (m, 1 H),
2.34 (s, 3 H), 3.23 (s, 3 H), 3.52 (d, J ) 8 Hz, 1 H), 7.1-7.4 (m, 4 H).
1H NMR (CDCl3) of 14c: δ 0.1-0.8 (m, 4 H), 1.0-1.3 (m, 1 H), 3.22
JA960268J
(29) Close, W. J. J. Am. Chem. Soc. 1957, 79, 1455.
(30) Bennett, J. G.; Bunce, S. C. J. Org. Chem. 1960, 25, 73.
(31) Blake, K. W.; Gillies, I.; Denney, R. C. J. Chem. Soc., Perkin Trans.
1 1981, 700.
(26) Woning, J.; Oudenampsen, Y.; Laarhoven, W. H. J. Chem. Soc.,
Perkin Trans. 2 1989, 2147.
(27) Fulton, J. D.; Dunitz, J. D. Nature 1947, 160, 161.
(28) Pailer, M.; Mu¨ller, U. Monatsh. Chem. 1948, 79, 615.
(32) Wilt, J. W.; Roberts, D. D. J. Org. Chem. 1962, 27, 3430.