Substituent effects on the photogeneration and selectivity of triaryl vinyl cations

Article abstract of DOI:10.1021/jo800687z

(Chemical Equation Presented) The photochemical reactions of a series of triaryl vinyl halides IX in acetic acid and in acetonitrile have been studied using product analysis as a function of the time of irradiation. The quantum efficiencies of formation of the products derived from the photogenerated vinyl cations 1⁺ depend on the α-aryl substituent, the β-aryl substituent, the leaving group X (= bromide or chloride), and the temperature at which the irradiations are carried out. Hammett correlation or noncorrelation of the α-aryl substituent effects with (excited-state) substituent constants indicates that the ions 1⁺ are formed directly from the excited states of 1X by heterolytic cleavage of the carbon-halogen bond. Homolytic cleavage, yielding radicals 1^?, is a parallel process: the partitioning into ion and radical occurs in the excited state. This conclusion is corroborated by the leaving group effect and the temperature effect. By performing the photochemical reactions of IX in the presence of HOAc and/or NaOAc as well as the labeled common halide ion ⁸²Br ^- or ³⁶Cl^-, the relative reactivities of the cations 1⁺ toward these nucleophiles were determined. The selectivities follow the Reactivity-Selectivity Principle. The temperature effect data show that the reactions of the cations with the anionic nucleophiles are (de)solvation controlled and their reactions with the neutral nucleophile activation controlled.

Full text of DOI:10.1021/jo800687z

Substituent Effects on the Photogeneration and Selectivity of Triaryl
Vinyl Cations
Jan Willem J. van Dorp and Gerrit Lodder*
Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden UniVersity,
P.O. Box 9502, 2300 RA Leiden, The Netherlands
lodder@chem.leidenuniV.nl
ReceiVed March 27, 2008
The photochemical reactions of a series of triaryl vinyl halides 1X in acetic acid and in acetonitrile have
been studied using product analysis as a function of the time of irradiation. The quantum efficiencies of
formation of the products derived from the photogenerated vinyl cations 1+ depend on the R-aryl
substituent, the ꢀ-aryl substituent, the leaving group X () bromide or chloride), and the temperature at
which the irradiations are carried out. Hammett correlation or noncorrelation of the R-aryl substituent
effects with (excited-state) substituent constants indicates that the ions 1+ are formed directly from the
excited states of 1X by heterolytic cleavage of the carbon-halogen bond. Homolytic cleavage, yielding
radicals 1•, is a parallel process: the partitioning into ion and radical occurs in the excited state. This
conclusion is corroborated by the leaving group effect and the temperature effect. By performing the
photochemical reactions of 1X in the presence of HOAc and/or NaOAc as well as the labeled common
halide ion 82Br- or 36Cl-, the relative reactivities of the cations 1+ toward these nucleophiles were
determined. The selectivities follow the Reactivity-Selectivity Principle. The temperature effect data
show that the reactions of the cations with the anionic nucleophiles are (de)solvation controlled and their
reactions with the neutral nucleophile activation controlled.
Introduction
electron-donating to electron-withdrawing groups and with a
small number of ꢀ-substituents. Triaryl vinyl cations have been
at the forefront of this research. These ions are also thermally
accessible, albeit only at high reaction temperatures.2 The nature
of photogenerated triaryl vinyl cations has been shown to be
the same as that of the ions involved in thermal nucleophilic
Photolysis of vinyl halides in appropriate media has become
a versatile method to generate vinyl cations under mild reaction
conditions.1 The method has been used to prepare a score of
vinyl cations with a variety of R-substituents ranging from
5416 J. Org. Chem. 2008, 73, 5416–5428
10.1021/jo800687z CCC: $40.75 2008 American Chemical Society
Published on Web 06/24/2008

Photogeneration and SelectiVity of Triaryl Vinyl Cations
SCHEME 1
SCHEME 2
substitution reactions, i.e., free (not ion-paired), “cold” (ther-
mally relaxed, i.e., solvated) vinyl cations: the extent of ꢀ-aryl
rearrangement3 as well as the selectivity toward different
nucleophiles are the same in photochemically4 and thermally5
induced reactions. Also, triaryl vinyl cations have been observed
in flash photolysis studies by both optical absorption and
electrical conductivity measurements.6,7
Upon irradiation of vinyl halides, in general, both ion- and
radical-derived products are produced. In the course of time,
several mechanisms have been proposed for their formation
(Scheme 1).
Initially,theserouteswere(A)homolysisofthecarbon-halogen
bond yielding a radical pair (pathway 2), followed either by
electron transfer yielding an ion pair (pathway 3) or by
dissociation into free radicals (pathway 6),8 and (B) heterolysis
of the carbon-halogen bond forming an ion pair directly
(pathway 1), in parallel with homolysis (pathway 2).4a Recently,
(C) initial heterolysis of the carbon-halogen bond forming an
ion pair (pathway 1), followed either by dissociation into ions
(pathway 5) or by electron transfer yielding a radical pair
(pathway 4), has been proposed.9 In the presence of a suitable
electron donor, vinyl radicals can also be formed by photoin-
duced electron transfer (pathway 7) followed by halide ion loss
from the radical anion (pathway 8).10
Thus, the partitioning into ion and radical occurs in the radical
pair in mechanism A, in the excited state in mechanism B, and
in the ion pair in mechanism C.
(1) (a) Kropp, P. J. In CRC Handbook of Organic Photochemistry and
Photobiology; Horspool, W. M., Song, P.-S., Eds.; CRC Press: Boca Raton, FL,
1995; Chapter 84. (b) Kitamura, T. In CRC Handbook of Organic Photochemistry
and Photobiology; Horspool, W. M., Song, P.-S., Eds.; CRC Press: Boca Raton,
FL, 1995; Chapter 85. (c) Lodder, G. ; Cornelisse, J. In The Chemistry of
Functional Groups: Supplement D2; Patai, S., Rappoport, Z., Eds.; Wiley:
Chichester, UK, 1995; Chapter 16. (d) Lodder, G. In Dicoordinated Carbocations;
Rappoport, Z., Stang, P. J., Eds.; Wiley: Chichester, UK, 1997; Chapter 8. (e)
Okuyama, T.; Lodder, G. AdV. Phys. Org. Chem. 2002, 37, 1–56.
(2) (a) Rappoport, Z. Acc. Chem. Res. 1976, 9, 265–273. (b) Stang, P. J.;
Rappoport, Z.; Hanack, M.; Subramanian, L. R. Vinyl Cations; Academic Press:
New York, 1979. (c) Rappoport, Z. In ReactiVe Intermediates; Abramovitch,
R. A., Ed.; Plenum Press: New York, 1983; Vol. 3, pp 427-615.
(3) Kitamura, T.; Kobayashi, S.; Taniguchi, H.; Fiakpui, C. Y.; Lee, C. C.;
Rappoport, Z. J. Org. Chem. 1984, 49, 3167–3170.
In this paper, we report substituent effects on the photo-
chemical generation and selectivity of the triaryl vinyl cations
1+ produced from the triaryl vinyl halides 1X depicted in
Scheme 2. The compounds 1aX-1eX were selected because
(a) their E a Z isomerization is a degenerate process and
(b) in their cations 1,2-aryl shifts across the double bond
are either degenerate or do not occur.2,3,4a This ensures that
the reaction mixtures are not unnecessarily complex. The
series 1aBr-1bBr-1cBr gives information about R-aryl
substituent effects, the series 1cBr-1dBr-1eBr about ꢀ-aryl
substituent effects, and the series 1aBr-1aCl and 1dBr-
1dCl about leaving group effects. All compounds were
irradiated in two solvents, acetic acid and acetonitrile/1 M
(4) (a) van Ginkel, F. I. M.; Cornelisse, J.; Lodder, G. J. Am. Chem. Soc.
1991, 113, 4261–4272. (b) van Ginkel, F. I. M.; Cornelisse, J.; Lodder, G. J.
Photochem. Photobiol. A: Chem. 1991, 61, 301–315.
(5) van Ginkel, F. I. M.; Hartman, E. R.; Lodder, G.; Greenblatt, J.;
Rappoport, Z. J. Am. Chem. Soc. 1980, 102, 7514–7519.
(6) (a) Schnabel, W.; Naito, I.; Kitamura, T.; Kobayashi, S.; Taniguchi, H.
Tetrahedron 1980, 36, 3229–3231. (b) Kobayashi, S.; Zhu, Q. Q.; Schnabel, W.
Z. Naturforsch. 1988, 43b, 825–829. (c) Kobayashi, S.; Schnabel, W. Z.
Naturforsch. 1992, 47b, 1319–1323.
(9) (a) Gronheid, R.; Zuilhof, H.; Hellings, M. G.; Cornelisse, J.; Lodder,
G. J. Org. Chem. 2003, 68, 3205–3215. (b) van Alem, K.; Belder, G.; Lodder,
G.; Zuilhof, H. J. Org. Chem. 2005, 70, 179–190.
(7) van Ginkel, F. I. M.; Visser, R. J.; Varma, C. A. G. O.; Lodder, G. J.
Photochem. 1985, 30, 453–473.
(8) Kropp, P. J.; McNeely, S. A.; Davis, R. D. J. Am. Chem. Soc. 1983,
105, 6907–6915.
(10) Verbeek, J. M.; Cornelisse, J.; Lodder, G. Tetrahedron 1986, 42, 5679–
5684.
J. Org. Chem. Vol. 73, No. 14, 2008 5417

van Dorp and Lodder
SCHEME 3
acetic acid. Also, the effect of the temperature at which the
irradiations are carried out (23 vs 60 °C) was investigated
for compounds 1aBr and 1dBr. Together, these effects allow
discrimination between mechanisms A, B, and C.
tetraethylammonium halide or in acetonitrile/acetic acid/labeled
tetraethylammonium halide gives the mixture of products
depicted in Scheme 3: the isotopically labeled 1-halo-1,2,2-
triarylethenes 1X#, the 1-acetoxy-1,2,2-triarylethenes 2, the
benzophenones 3, the triarylethenes 4, the 9-halo-10-arylphenan-
threnes 5X, the 9-acetoxy-10-arylphenanthrenes 6, and the
9-arylphenanthrenes 7.
Almost all irradiations were performed in the presence of
HOAc and/or NaOAc and the labeled (#) common halide ion
Br-[82Br] or Cl-[36Cl] as nucleophiles. This procedure probes
the selectivity of the photochemically formed triaryl vinyl
cations 1+ toward selected pairs of nucleophiles, including the
(labeled) common halide ion. The method was initially intro-
duced5 to deal with the common ion rate depression which often
hampersthermalkineticstudiesoftriarylvinylcationformation.2a,11
It has become a powerful supplement to the more common
mechanistic tools to investigate the mechanistic pathway from
substrate to nucleophilic photosubstitution product. It also
provides insight into the selectivity of carbocations of the type
1+ toward different kinds of nucleophiles.
Examination of the product concentrations as a function of
time of irradiation revealed that 1X#, 2, 3, 4, and 5X are primary
photoproducts, while 6 and 7 are secondary ones. Products 1X#
and 2 are formed through the intermediacy of the vinyl cations
1+ reacting with (labeled) X- and HOAc/OAc-, respectively.
Products 3 and 4 are formed through the intermediacy of the
vinyl radicals 1• reacting with O2 (with the vinyl peroxyl radical
1O2 as a further intermediate12) and with a H-atom donor,
•
respectively. 5X is the (cis-stilbene-type) photocyclization13
product of 1X. The secondary products 6 and 7 result from
photocyclization of the primary products 2 and 4. Formation
Results and Discussion
(11) Rappoport, Z.; Gal, A. Tetrahedron Lett. 1970, 3233–3236.
(12) Verbeek, J. M.; Stapper, M.; Krijnen, E. S.; van Loon, J.-D.; Lodder,
G.; Steenken, S. J. Phys. Chem. 1994, 98, 9526–9536.
Photoproducts. Irradiation (λexc ) 313 nm) of the 1-halo-
1,2,2-triarylethenes 1X in acetic acid/sodium acetate/labeled
5418 J. Org. Chem. Vol. 73, No. 14, 2008

Photogeneration and SelectiVity of Triaryl Vinyl Cations
TABLE 1. Quantum Yields of the Photoreactions of the Vinyl Bromides 1aBr-1eBr in Acetic Acid and in Acetonitrile (λexc ) 313 nm; T )
23 °C) and Selectivities of the Vinyl Cations 1a+-1e+ toward Br- and OAc- or HOAc
entry
substrate
solvent
HOAca
102Φd
102Φc
102Φox
102ΦOAc
102Φex
102Φion
SNaOAc
SHOAc
1
2
3
4
5
6
7
8
9
1aBr
1bBr
1cBrb
1dBrb
1eBr
1aBr
1bBr
1cBr
1dBr
1eBr
11.2
8.6
5.6
2.9
4.8
17.3
17.2
6.0
2.9
5.4
3.0
1.8
2.0
0.6
2.8
2.3
1.3
1.9
0.39
3.0
0.45
0.41
0.1c
0.15c
0.56
1.2
6.8
5.6
1.5
0.97
2.6
2.0
1.5
0.30
0.145
0.24
7.5
11.9
28
20
27
34
66
81
47.6
50
14.3
17.5
29.5
21
29.6
36
68
81
48
50
1.4
2.6
23
25.5
13
280
520
4600
5100
2600
240
CH3CNd
1.1
620
0.71
0.21
0.83
3800
4700
2900
10
a Added nucleophiles present: [Et4NBr#] ) 71 × 10-3 M; [NaOAc] ) 87 × 10-3 M. b Reported in ref 4a. c Φred; Φox not reported. d added
nucleophiles present: [Et4NBr#] ) 71 × 10-3 M; [HOAc] ) 1.0 M.
of 6 from 5X by nucleophilic aromatic photosubstitution14 is a
relatively insignificant reaction route,4a and so is the formation
of 7 from 5X by reductive dehalogenation.1c All irradiations
were performed in the presence of atmospheric oxygen, which
is necessary to quantitatively convert the dihydrophenanthrenes
formed initially in the photocyclization process of 1X and 2
into the easily monitorable phenanthrenes 5X and 6. Under these
conditions, almost the only products derived from the vinyl
radicals 1• are the benzophenones 3; the reduction products 4
(and 7) are at best observed in trace amounts.
In the irradiations in acetonitrile/1.0 M HOAc, no Ritter
products resulting from the reaction of 1+ with acetonitrile are
detected. Also, no products resulting from ipso-substitution, in
which the p-methoxy group is replaced by one of the nucleo-
philes,15 are found in the irradiations of the R-(p-methoxyphenyl)
compounds 1cBr, 1dBr, 1eBr, and 1dCl.
Photoreactivity. (a) Kinetics. The rates of disappearance of
starting material and of formation of products as a function of
the intensity of the absorbed light were determined under the
same reaction conditions as used for the product studies. The
kinetics for photochemically induced simultaneous isotope
exchange and product-forming reactions are represented by eqs
1–3.4b
algebraic magnitude which does not include the isotopic
exchange and other (formally) degenerate processes); Φd is the
quantum yield of disappearance of starting material, and I0 is
the intensity of the light absorbed by 1X at time t ) 0. Φp is
the quantum yield of product formation. Φex is the quantum
yield for the isotope exchange, and nt and n∞ are the specific
radioactivities of 1X# at time t and at equilibrium.
The quantum yields of disappearance of starting material and
of formation of products under various conditions are listed in
Tables 1, 4, and 5. In these tables, Φd is the quantum yield of
disappearance of starting material (as above); Φc the quantum
yield of formation of the photocyclization product 5X; Φox the
quantum yield of formation of the (substituted) benzophenone
3; Φred the quantum yield of formation of the reduction products
4 and 7 combined; ΦOAc the quantum yield of formation of
compounds 2, obtained by monitoring the formation of both
the vinyl acetate 2 itself and its secondary photoproduct, the
phenanthrene acetate 6; and Φex the quantum yield of the
isotope-exchange reaction, i.e., the formation of the labeled
product 1X# (as above). Φion is the quantum yield of formation
of the photoproducts derived from 1+, i.e., 1X#, 2 (and 6), and
thus Φion ) Φex + ΦOAc
.
Also listed are two selectivity constants: SNaOAc is the
selectivity of the vinyl cations 1+ in their reactions with
the common halide ion and acetate ion (eq 4a), and SHOAc is
the selectivity of the cations 1+ in their reactions with the
common halide ion and acetic acid (eq 4b).
[RX]t
[RX]0
I0
ln
) -kdt ) -Φdt
(1)
(2)
(
)
[RX]0
I0
-kdt
(
)
[product]t ) Φ
1 - e
pkd
kex
Φex [NaOAc]
SNaOAc
)
)
)
(4a)
(4b)
kOAc ΦOAc [Et4NX]
{n∞ ⁄ n∞ - nt}
[RX]0 + [Et4NX]0
[RX]0[Et4NX]0
(3)
ln
) ΦexI0t
tkd
tkd
kex
Φex [HOAc]
ΦOAc [Et4NX]
(
)
e
- {n∞ ⁄ n∞ - nt}{e - 1}
SHOAc
)
kOAc
Table 1 shows the quantum yields of the photoreactions of
the vinyl bromides 1aBr-1eBr. The data show that the
formation of ion-derived products as expressed by Φion is more
efficient in the more polar16 medium acetonitrile than in acetic
acid. This solvent effect does not discriminate between mech-
anisms A and B in Scheme 1. Polar solvents favor electron
transfer within the radical pair (mechanism A) but also direct
heterolysis of the electronically excited state (mechanisms B
and C). They disfavor electron transfer within the ion pair
initially formed in mechanism C. Therefore, the larger values
of Φox in acetonitrile than in acetic acid accompanying the larger
values of Φion do not agree with the latter mechanism.
In these equations, [RX]t and [RX]0 represent the concentra-
tion of vinyl halide 1X in the photolysis mixture at times t and
0; kd stands for the rate constant of disappearance of 1X (an
(13) (a) Laarhoven, W. H. Recl. TraV. Chim. Pays-Bas 1983, 102, 185–204,
and 241-254. (b) Mallory, F. B.; Mallory, C. W. Org. React. (N.Y.) 1984, 30,
1–456. (c) Laarhoven, W. H. Org. Photochem. 1989, 10, 163–308.
(14) (a) Cornelisse, J.; Havinga, E. Chem. ReV. 1975, 75, 353–388. (b)
Havinga, E.; Cornelisse, J. Pure Appl. Chem. 1976, 47, 1–10. (c) Cornelisse, J.;
Lodder, G.; Havinga, E. ReV. Chem. Intermed. 1979, 2, 231–265. (d) Terrier, F.
Nucleophilic Aromatic Displacement. The Influence of the Nitro Group; VCH
Publishers: Weinheim, 1991; Chapter 6.
(15) (a) Kitamura, T.; Murakami, M.; Kobayashi, S.; Taniguchi, H. Tetra-
hedron Lett. 1986, 3885–3888. (b) Kitamura, T.; Kabashima, T.; Kobayashi, S.;
Taniguchi, H. Chem. Lett. 1988, 1951–1954. (c) Kitamura, T.; Kabashima, T.;
Nakamura, I.; Fukuda, T.; Taniguchi, H. J. Am. Chem. Soc. 1991, 113, 7255–
7261.
(b) r-Aryl Substituent Effects. Within the series 1aBr,
1bBr, and 1cBr (R1 ) H, CH3, OCH3), a clear increase in Φion
,
J. Org. Chem. Vol. 73, No. 14, 2008 5419

van Dorp and Lodder
TABLE 2. Correlations and Noncorrelations of Quantum Yields of Ion Formation from 1aBr, 1bBr, and 1cBr with Ground- And
Excited-State Substituent Constants
a
c
d
e
f
g
•
solvent
σex
σ*b
σhν
σ•hν
σp
σ+
σjj
subst const
p-H
0
0
0
0
0
0
0
p-CH3
p-OCH3
F
standard dev
corr coeff R2
F
standard dev
corr coeff R2
Fτ
-0.13
-0.39
-0.80
0.04
0.998
-0.81
0.47
0.750
-3.10
0.25
0.994
-3.10
0.26
-0.08
-0.78
-0.36
0.07
0.963
-0.31
0.30
0.522
-1.42
0.23
0.974
-1.37
0.46
-0.53
-1.17
-0.27
0.05
0.967
-0.29
0.12
0.857
-1.03
0.23
0.954
-1.05
0.06
0.74
0.33
0.10
0.42
0.049
0.36
0.35
0.509
0.32
1.63
0.036
0.58
1.57
0.120
-0.17
-0.27
-1.08
0.44
0.858
-1.33
0.25
0.966
-4.13
1.84
0.835
-4.38
1.15
-0.31
-0.78
-0.40
0.05
0.986
-0.42
0.21
0.810
-1.55
0.23
0.977
-1.57
0.02
0.15
0.23
1.25
0.55
0.840
1.57
0.25
0.974
4.78
2.27
0.816
5.09
1.47
0.922
Φion
HOAc
CH3CN
HOAc
CH3CN
Φion
kion ) Φion/τ
kion ) Φion/τ
standard dev
corr coeff R2
Fτ
standard dev
corr coeff R2
0.993
0.899
0.997
0.936
0.999
a Taken from ref 20b. b Taken from ref 20d. c Taken from ref 21. d Taken from ref 23. e Taken from ref 26. f Taken from ref 22. g taken from ref 24.
FIGURE 1. Hammett plot of the efficiency (×) and rate (∆) of photolytic formation and selectivity (x) of the vinyl cations 1a+, 1b+, and 1c+ in
acetic acid.
the quantum yield of formation of ion-derived products, with
increasing electron-donating ability of the R-aryl substituent is
observed in both media (Table 1, entries 1-3 and 6-8). The
formation of radical-derived photoproducts as expressed by Φox
is in all cases a minor process with, if any, a small oppositely
directed R-aryl substituent effect. There is no complementarity
reactions occurring from the lowest excited singlet state.20 The
inherent problems of this approach have been reviewed by
McEwen and Yates,21 who proposed a σhν scale based on the
effects of aryl substituents on the rates of photohydration of
styrenes and phenylacetylenes. This concept was tested against
literature data and gives good correlations for photoreactions
for which actual rates are known, but not if only quantum yields
are available. This σhν scale is an excited-state equivalent of
the Brown-Okamoto ground-state σ+ scale,22 which reflects
between the decrease in Φox and the increase in Φion
.
The R-aryl substituent effects on Φion (and on Φox) invite a
Hammett treatment.17 This is, however, not as straightforward
for photochemical reactions as it is for thermal ones because
the usual substituent constants represent electronic effects in
the ground rather than in the excited state.18 Moreover, the
Hammett equation pertains to rate (or equilibrium) constants
and not to quantum yields. The relationship between the rate
constant kr of a photochemical reaction and its quantum yield
Φr is given by eq 5, with τ the lifetime of the reactive excited
state. Therefore, ideally lifetimes should be taken into account.
As for all other vinyl halides investigated,1d the reactive excited
state in the present case is a singlet species.19
(16) Murov, S. L. ; Carmichael, I. ; Hug, G. L. Handbook of Photochemistry
2nd ed.; Marcel Dekker: New York, 1993; p 284.
(17) Hammett, L. P. J. Am. Chem. Soc. 1937, 59, 96–103.
(18) Ziffer, H.; Sharpless, N. E. J. Org. Chem. 1962, 27, 1944–1946.
(19) The photoreactions of 1aBr in acetic acid are neither sensitized by
xanthone nor quenched by 1,3-cyclohexadiene. In the presence of the latter (0.5
M), instead significant amounts of the otherwise rarely found reduction products
4a and 7a (Φred ) 9.7 × 10-2) are produced. Probably 1,3-cyclohexadiene first
acts as electron donor (Scheme 1, pathways 7 and 8) and next as hydrogen-
atom donor.
(20) (a) Sengupta, D.; Lahiri, S. C. Z. Phys. Chem. (Leipzig) 1977, 258, 1097–
1106. (b) Baldry, P. J. J. Chem. Soc., Perkin Trans. 2 1979, 951–953. (c) Shim,
S. C.; Park, J. W.; Ham, H. S. Bull. Korean Chem. Soc. 1982, 3, 13–18. (d)
Shim, S. C.; Park, J. W.; Ham, H. S.; Chung, J. S. Bull. Korean Chem. Soc.
1983, 4, 45–47.
kr ) Φr ⁄ τ
(5)
Two sets of photochemical substituent constants based on
pKa* values (σex and σ*) have been suggested to correlate
(21) McEwen, J.; Yates, K. J. Phys. Org. Chem. 1991, 4, 193–206.
(22) Brown, H. C.; Okamoto, Y. J. Am. Chem. Soc. 1958, 80, 4979–4987.
5420 J. Org. Chem. Vol. 73, No. 14, 2008

Photogeneration and SelectiVity of Triaryl Vinyl Cations
FIGURE 2. Hammett plot of the efficiency (×) and rate (∆) of photolytic formation and selectivity (x) of the vinyl cations 1a+, 1b+, and 1c+ in
acetonitrile.
direct resonance interaction with a cationic reaction center.
Fleming and Jensen subsequently proposed a σ•hν scale for
excited-state radical cleavage reactions based on the photo-
chemical fragmentation of benzyl phenyl thioethers.23 This σ•hν
scale is an excited-state equivalent of the ground-state σ•
scales.24
Again, the best correlations are found with σ+, σex, and σhν,
now in both media employed (see Table 2). There is no good
correlation with σp or σ• and not even a semblance of a
correlation with σ•hν. In Figures 1 and 2, the effect of the lifetime
correction on the quantum yields is shown for the correlations
with σ+.
Several authors have tested the kinetic data they obtained
for various photoreactions against excited-state as well as against
ground-state substituent constants and have drawn their mecha-
nistic conclusions.25
In Table 2, we report the correlations of the kinetic data in
Table 1 for the nucleophilic photosubstitution reactions of the
compounds 1aBr, 1bBr, and 1cBr with the excited-state
substituent constants σex, σ*, σhν, and σ•hν as well as with the
ground-state substituent constants σp, σ+, and σ•.27 Considering
that our data refer to three compounds only, the values of Φion
in acetic acid correlate well with σ+, σex, σ*, and σhν but not
with σp, σ•, or with σ•hν. In acetonitrile, the correlation
It was also tested whether any Hammett correlation is present
with Φox, the quantum yield of vinyl radical formation. Neither
Φox nor Φox/τ correlates with σ+, σex, σhν, σp, or σ•. The only
correlations found were Fox(σ*) ) +0.29 ( 0.02 (R2 ) 0.996)
and Fox,τ(σ•hν) ) +0.24 ( 0.02 (R2 ) 0.995) in acetonitrile.
Thus, no connection exists between the (lack of) Hammett
correlations for the ion- and radical-formation process.
Almost the same R-aryl substituent effects as for compounds
1a-cBr have been observed for the nucleophilic photosubsti-
tution reactions of a series of 1-aryl-2-(2,2′-biphenyldiyl)vinyl
bromides (8Br) in deoxygenated methanol12,29 (Scheme 4, Table
3).
coefficients are smaller: the lines level off because with Φion
0.81 for 1cBr the limiting value of 1.0 is approached.
)
In contrast to the vinyl cations 1+, the vinyl cations 8+ are
not subject to 1,2-aryl shifts across the double bond. These shifts
are prevented by the rigid fluorenyl moiety at the ꢀ-carbon
atom.10 In the series of compounds 8Br, therefore, also the
effects of electron-withdrawing R-aryl substituents such as
p-bromo- and p-cyanophenyl could be and were investigated.
A Hammett treatment of the values of Φion of the five
compounds 8Br using the same seven sets of substituent
constants as used with the compounds 1a-cBr shows only
correlation with σ+ and σ*. Singlet excited-state lifetimes are
not available for p-bromo- and p-cyanostyrene. Therefore Φion/τ
) kion can only be calculated for 8aBr, 8bBr, and 8cBr but not
for 8dBr and 8eBr. The values of Φion/τ for 8a-cBr correlate
with σex and σhν as well as with σ+ and σ*. No correlation exists
Regrettably, the correlations of kion () Φion/τ according to
eq 5) with the various substituent constants cannot be studied
directly. Triaryl vinyl halides do not fluoresce, and consequently,
their singlet excited-state lifetimes elude measurement. Correc-
tion of the quantum yields of ion formation from 1aBr, 1bBr,
and 1cBr with the lifetimes of the corresponding, structurally
related, styrenes (styrene, τ ) 7.5 ns, p-methylstyrene, τ ) 4.5
ns, and p-methoxystyrene, τ ) 1.2 ns)21 allows us to study the
correlations of kion with the various substituent constants
indirectly.28 This approach gives values of log{(Φion,R1/Φion,H)(τH/
τR1)}, serving as approximation for log{kion,R1/kion,H}.
(23) Fleming, S. A.; Jensen, A. W. J. Org. Chem. 1996, 61, 7040–7044.
(24) Jiang, X.-K. Acc. Chem. Res. 1997, 30, 283–289.
with σp and certainly not with σ• or σ•hν
.
(25) (a) Baldry, P. J. J. Chem. Soc., Perkin Trans. 2 1980, 805–808, and
809-815. (b) Givens, R. S.; Matuszewski, B.; Athey, P. S.; Stoner, M. R. J. Am.
Chem. Soc. 1990, 112, 6016–6021. (c) Boyd, M. K.; Lai, H. Y.; Yates, K. J. Am.
Chem. Soc. 1991, 113, 7294–7300. (d) DeCosta, D. P.; Bennett, A.; Pincock,
A. L.; Pincock, J. A.; Stefanova, R. J. Org. Chem. 2000, 65, 4162–4168.
(26) Hansch, C.; Leo, A.; Taft, W. Chem. ReV. 1991, 91, 165–195.
(27) Several σ• scales exist.24 The data in Table 2 and the text pertain to
Overall, only with respect to σ+ is a correlation found for
both Φion and Φion/τ for the compounds 1a-cBr as well as for
the compounds 8Br. The F(σ+) values for the vinyl bromides
8Br in methanol are F(σ+) ) -0.58 ( 0.09 (R2 ) 0.927) and
Fτ(σ+) ) -1.69 ( 0.31 (R2 ) 0.967), respectively. These values
almost equal those of F(σ+) ) - 0.40(HOAc), - 0.42(CH3CN),
24
•
Jiang’s recently established σjj scale.
(28) The quite similar substituent effects on the singlet lifetimes of excited
phenylacetylenes21 and (2-arylcyclopropyl) methylacetates support the use of
these model compounds: Hixson, S. S.; Franke, L. A.; Gere, J. A.; Xing, Y. D.
J. Am. Chem. Soc. 1988, 110, 3601–3610.
(29) Verbeek, J. M.; Lodder, G. Unpublished results.
J. Org. Chem. Vol. 73, No. 14, 2008 5421

van Dorp and Lodder
SCHEME 4
TABLE 3. Quantum Yields of the Photoreactions of the Vinyl
6 τp-methoxystyrene) offsets the about 50% higher values of Φion
for 1cBr than for 1dBr. This brings the ꢀ-aryl substituent effect
on the rates of the photochemical cation formation in the same
direction as the R-aryl substituent effect: a slight increase with
increasing electron-donating abilities within the series 1c-eBr.
Thermal ꢀ-aryl substituent effects on vinyl cation formation
are also small;2a e.g., at both 120.3 and 141.5 °C 1dBr
acetolyzes thermally about 3 times faster than 1cBr.11 Still, that
thermal substituent effect is larger than the photochemical one,
reflecting again the late transition state in the former and the
early transition state in the latter process.
Bromides 8Br in Deoxygenated Methanol (λexc ) 254 nm; T ) 0 °C)
a
a
a
R
p-H
p-CH3
p-OCH3
p-Brb
p-CNb
c
103Φion
103Φred
35
15
42
7
109
10
20
17
16
12
d
a Data from ref 12. b Data from ref 29. c Formation of 8OCH3.
d Formation of 8H.
and Fτ(σ+) ) -1.55(HOAc), -1.57(CH3CN) for the vinyl
bromides 1a-cBr listed in Table 2.
The correlation of the quantum yields (lifetime-uncorrected
and -corrected) of ion formation from both 1a-cBr and from
8Br with σ+, and in some cases with σex, σ*, and σhν, but not
with σp, σ•, or σ•hν, indicates that both the vinyl cations 1+ and
8+ are formed directly from the excited states of 1Br and 8Br
by heterolytic cleavage of the C-Br bond and not by homolytic
cleavage giving initially the vinyl radical(pair)s 1• or 8•, which
subsequently give electron transfer to the ions (Scheme 1:
pathway 1 and not pathways 2 and 3). This is in accordance
with mechanisms B and C, but not with mechanism A. The
overall noncorrelation of the quantum yields (lifetime-uncor-
rected and -corrected) of radical formation with any set of
substituent constants shows that radical formation and ion
formation are not consecutive processes. This points to mech-
anism B rather than to mechanism C (or A).
The Fτ(σ+) values of -1.6 and -1.7 for the photochemical
formation of the vinyl cations 1+ and 8+ are small compared to
the F∆(σ+) value of -4.6 for the thermal acetolysis of triaryl
vinyl bromides.30 This difference is easily explained in terms
of the transition states involved: early, with little charge
separation, in the photochemical process (“on the descent” from
the excited state) and late, with more charge separation, in the
thermal process (“on the ascent” from the ground state).
(c) ꢀ-Aryl Substituent Effects. In the series 1cBr, 1dBr,
and 1eBr (R2 ) H, OCH3, Cl) the Φion values do not increase
with the electron-donating ability of the ꢀ-aryl substituent (Table
1, entries 3-5 and entries 8-10). Here too, however, rate
constants (kr) rather than quantum yields matter. Although the
substituent effects on the lifetimes of R-aryl-substituted styrenes,
which we used in the discussion of the R-aryl substituent effects,
can obviously not be used quantatively to correct the ꢀ-aryl
(d) Leaving Group Effects. Table 4 lists the quantum yields
of the photoreactions of the vinyl chlorides 1aCl and 1dCl.
Comparison of Φion for the formation of cation 1a+ from 1aBr
(Table 1, entries 1 and 6) with those from 1aCl (Table 4, entries
1 and 3) as well as of Φion for the formation of cation 1d+
from 1dBr (Table 1, entries 4 and 9) with those from 1dCl
(Table 4, entries 2 and 4) shows that the bromine compounds
form vinyl cations more efficiently than the chlorine compounds,
both in acetic acid and in acetonitrile.31 This leaving group effect
is larger for the formation of the less stable triphenyl vinyl cation
1a+ (4-6-fold) than for the formation of the more stable tris(p-
methoxyphenyl) vinyl cation 1d+ (1-2-fold). Thus, the leaving
group effect on Φion is attenuated by the stability of the vinyl
cations produced.
Thermally, at 141.5 °C, 1dBr acetolyzed about 10 times faster
than 1dCl.11 Once more, the effect of a structural variation is
stronger for the thermal than for the photochemical process:
the transition state for the latter reaction is earlier than that for
the former.
The same leaving group effect on Φion (Br > Cl) has been
found in kinetic studies of the photosolvolysis of compounds
8a-cBr/Cl12 and of R-alkenyl-substituted vinyl bromides and
chlorides.32 The effect parallels the anionic leaving group
abilities of the halides and indicates that CsX bond cleavage
occurs heterolytically, as in mechanisms B and C. For mech-
anism A, the reverse effect (Cl > Br) would be expected, in
parallel with the electron affinities of the halogen atoms.
(31) Leaving group effects Br > Cl can be discussed qualitatively in terms
of Φion. Singlet lifetimes of bromine compounds are smaller than those of
corresponding chlorine compounds (see, for instance: Durand, A.-P.; Brown,
R. G.; Worrall, D.; Wilkinson, F. J. Chem. Soc., Perkin Trans. 2 1998, 365–
370. ). Therefore, the leaving group effect on Φion /τ ) kion should be larger
than that on Φion, but in the same direction.
substituent effects on Φion, their relative magnitude (τstyrene
)
(32) Krijnen, E. S.; Zuilhof, H.; Lodder, G. J. Org. Chem. 1994, 59, 8139–
8150.
(30) Rappoport, Z.; Gal, A. J. Org. Chem. 1972, 37, 1174–1181.
5422 J. Org. Chem. Vol. 73, No. 14, 2008

Photogeneration and SelectiVity of Triaryl Vinyl Cations
TABLE 4. Quantum Yields of the Photoreactions of the Vinyl Chlorides 1aCl and 1dCl in Acetic Acid and in Acetonitrile (λexc ) 313 nm; T
) 23 °C) and Selectivities of the Vinyl Cations 1a+ and 1d+ toward Cl- and OAc- or HOAc
entry
substrate
solvent
HOAca
102Φd
102Φc
102Φox
102ΦOAc
102Φex
102Φion
SNaOAc
SHOAc
1
2
3
4
1aCl
1dCl
1aCl
1dCl
5.4
8.4
8.6
4.1
3.1
5.1
3.4
3.2
2.6
3.3
0.78
0.32
0.80
15.9
5.4
3.4
19
6.2
26
0.35
5.9
70
1200
100
0.7
trace
0.67
CH3CNb
25.8
1130
b
a Added nucleophiles present: [Et4NCl#] ) 71 × 10-3 M; [NaOAc] ) 87 × 10-3 M. Added nucleophiles present: [Et4NCl#] ) 71 × 10-3 M;
[HOAc] ) 1.0 M.
In early reports on the photoreactivity of vinyl bromides and
chlorides, the ratio of ion-derived vs radical-derived products
was used to discuss leaving group effects on the photochemistry
of vinyl halides. This ratio was found to be larger for the
bromine than for the chlorine compounds in the case of
halomethylene cyclohexanes8,33 but smaller in the case of diaryl
vinyl halides.34 Still, both effects were considered to be in line
with mechanism A (Scheme 1) in which first the carbon-halogen
bond cleaves homolytically and vinyl cations are formed
subsequently by electron-transfer within the initially formed
radical pair.
In the case of compounds 1X, the ratio of ion- to radical-
derived products is larger for 1dBr than for 1dCl (229 vs 39 in
acetonitrile), but the opposite is true for 1aBr and 1aCl (30 vs
almost infinite; 1aCl forms virtually no radical-derived product).
Clearly, great care should be taken in attributing mechanistic
consequences to such product ratios.
The (virtual) nonformation of radical-derived products from
1aCl does not support mechanism A, in which vinyl cation
formation occurs through the intermediacy of a vinyl radical
(Scheme 1, pathways 2 and 3), but is in accordance with
mechanisms B and C, in which vinyl cation formation occurs
directly from the excited state (pathway 1). The formation of
radical-derived product from 1dCl, where electron transfer in
the ion pair [1d+ Cl-], forming the radical pair [1d• Cl•]
(pathway 4), should be less favorable than in the ion pair [1a+
Cl-], (not) forming the radical pair [1a• Cl•], points at
mechanism B, in which the formation of ion and radical
formation are parallel processes, rather than at mechanism C.
(e) Temperature Effects. In Table 5, the effect on the
quantum yields of the photoreactions of the temperature at which
the irradiations were carried out (60 vs 23 °C) is reported for
compound 1aBr with (entry 1 vs 2) and without (entry 6 vs 5)
added nucleophiles as well as for compound 1dBr with (entry
8 vs 7) and without (entry 10 vs 9) added nucleophiles, always
in acetic acid. At 60 °C, as at 23 °C, no thermal reaction occurs.
The dependence of the quantum yields on the reaction tem-
parature provides the activation parameters given in Table 6.
Because quantum yields are ratios of reaction rate constants
(see eq 5), the calculated activation parameters are differences
of activation energies/enthalpies, not absolute values. However,
temperature hardly affects the photophysical decay of excited
states,35 and the photochemical decay only constitutes a small
contribution to the lifetimes τ.36 Therefore, the overall lifetimes
are only weakly affected by the temperature and the calculated
activation energies/enthalpies effectively pertain to the reaction
rates kr.
None of the photoreactions of the triaryl vinyl halides 1X is
a single-step process: the vinyl cation 1+ is an intermediate in
the nucleophilic photosubstitution, the vinyl radical 1• in the
photooxidation, and the dihydrophenanthrene in the photocy-
clization.13 Consequently, the activation parameters pertain to
the rate determining step of each reaction.
The enthalpy of activation for the photoacetolysis of 1aBr
() formation of 1a+) and its activation energy for photooxi-
dation () formation of 1a•) in acetic acid without any additives
are both zero (Table 6, entry 2). As it is difficult to envision a
cation a radical transition to be entirely entropy controlled
without any enthalpy contribution, this lack of an activation
barrier shows that both the vinyl cation 1a+ and the vinyl radical
1a• are formed directly from the excited state of 1aBr without
either one of them being an intermediate in the formation of
the other (intermediate defined by Jencks as a species with a
significant lifetime and having barriers for its breakdown to both
reactants and products37). This disagrees with mechanism A,
which involves a radical f ion conversion (Scheme 1, pathway
3), as well as with mechanism C, which involves an ion f
radical conversion (pathway 4). It supports mechanism B, in
which cation (pathways 1 and 5) and radical formation
(pathways 2 and 6) are parallel processes.
As the triphenyl vinyl cation 1a+ is formed photochemically
with zero enthalpy of activation, the same surely also applies
to the more stabilized vinyl cation 1d+. Indeed, determination
of the capturing nucleophile in the following section shows that
all other activation parameters for the ionic photoprocesses of
1aBr and 1dBr (ΦOAc and Φex) listed in Table 6 (entries 1, 3,
and 4) pertain to the second step of the product formation
process, the (thermal) reaction of the vinyl cation with a
nucleophile.
In the case of the photolysis of (E)-bromostyrene in methanol,
for which mechanism C has been established by means of
isotope effects, the activation energies for the formation of ion-
and radical-derived photoproducts were also quite similar,
butsdue to the absence of a carbocation-stabilizing R-aryl
(33) Sonawane, H. R.; Nanjundiah, B. S.; Panse, M. D. Tetrahedron Lett.
1985, 3507–3510.
(34) (a) Kitamura, T.; Kobayashi, S.; Taniguchi, H. J. Org. Chem. 1982, 47,
ˇ
2323–2328. (b) Sket, B.; Zupan, M. J. Chem. Soc., Perkin Trans. 1 1979, 752–
756.
substituentsmuch larger (6.7 and 6.1 kcal ·mol-1 9a than here.
)
(35) Birks, J. B. In Organic Molecular Photophysics; Birks, J. B., Ed.;
Wiley: London, 1973; Vol. 1, Chapter 1.
It was argued that two entirely different photoprocesses from
the same excited state are unlikely to have virtually the same
activation energies. The activation energies (6.7 ≈ 6.1
kcal·mol-1) were attributed to the ion-pair formation (Scheme
1, pathway 1), and the activation energy for the electron-transfer
process converting the ion pair into a radical pair (pathway 4)
was considered to be lower than that. Valid as the argument
(36) For instance, in the case of 1aBr (Φd + Φex)23 ) 0.187, i.e. more than
80% of the decay is photophysical. Therefore, the overall lifetime is not
(significantly) affected by the temperature. As (Φd + Φex)60/(Φd + Φex)23 ) 1.4
the effect of increasing the temperature from 23 to 60 °C on τ is estimated to be
less than 10% (40% of 20%), i.e., τ23/τ60 < 1.1. Insertion of that information,
eq 5, and the temperatures into the Eyring equation shows that the contribution
of the lifetimes to the calculated activation energies/enthalpies is less than 0.5
kcal · mol-1, i.e., within the random error.
(37) Jencks, W. P. Acc. Chem. Res. 1980, 13, 161–169.
J. Org. Chem. Vol. 73, No. 14, 2008 5423

van Dorp and Lodder
TABLE 5. Quantum Yields of the Photoreactions of 1aBr and 1dBr under Various Conditions in Acetic Acid (λexc ) 313 nm) and Selectivities
of the Vinyl Cations 1a+ and 1d+ toward Br- and OAc- or HOAc
entry
substrate
additives
T (°C)
102Φd
102Φc
102Φox
102ΦOAc
102Φex
102Φion
SNaOAc
SHOAc
1
2
3
4
5
6
7
8
9
1aBr
a,b
a,b
a
60
23
23
23
23
60
23
60
23
60
15.1
11.2
11.7
9.6
8.2
11.2
2.9
5.1
2.7
3.6
4.5
3.0
3.4
2.8
3.1
0.56
0.45
0.53
0.33c
0.56
0.56
8.1
6.8
5.7
4.5
4.0
4.5
0.97
1.58
0.15
0.57
11.0
7.5
7.8
19
1.7
1.4
340
280
340
14.3
13.5
4.5
4.0
4.5
21
33
0.15
0.57
b
3.7
e
1dBr
a,b,d
a,b,d
0.15f
20
31
25.5
23.7
5100
4700
e
e
0.5
0.6
0.42
0.55
10
b
a [Et4NBr#] ) 71 × 10-3 M. [NaOAc] ) 87 × 10-3 M. c Traces of the reduction products 4a and 7a were also found. d Reported in ref 4a. e Not
reported. f Φred; Φox not reported.
TABLE 6. Activation Parameters ((0.5 kcal·mol-1) for the Photoreactions of 1aBr and 1dBr in Acetic Acid
ΦOAc
Φex
Φox
Φc
entry
substrate
additives
Ea
∆H‡
Ea
∆H‡
Ea
∆H‡
Ea
∆H‡
1
2
3
4
1aBr
a
0.9
0.6
2.6
7.1
0.3
0.0
2.0
6.5
2.1
1.5
1.1
0.1
c
0.5
-0.5
c
2.2
0.9
c
1.6
0.3
c
1dBr
a,b
2.3
1.7
1.4
0.8
1.0
0.4
a [Et4NBr#] ) 71 × 10-3 M; [NaOAc] ) 87 × 10-3 M. b Reported in ref 4a. c Not reported.
about the equality of the activation energies for the formation
of ion- and radical-derived photoproducts is for measurable
activation energies, it cannot be applied when these activation
enthalpies/energies are 0, as is the case here for the photochemi-
cal formation of the vinyl cations 1+ and vinyl radicals 1•.
Selectivity. (a) Capturing Nucleophile. The selectivity for
the irradiation of 1aCl in acetic acid/sodium acetate/tetraethy-
lammonium chloride[36Cl] is less than unity: SNaOAc ) 0.35
(Table 4, entry 1). This value catches the eye because it is most
unlikely that OAc- is more reactive than Cl- toward 1a+; that
would constitute a reversal of the normal order of nucleophilicity
(Cl- > OAc-).38 It suggests that the triphenyl vinyl cation 1a+
reacts efficiently with acetic acid even in the presence of acetate.
This is in contrast to the more stable 1-(p-methoxyphenyl)-2,2-
diphenylvinyl cation 1c+, for which it has been shown that the
capturing nucleophile in the (photo)acetolysis reaction is acetate
ion rather than acetic acid.4a That acetic acid and not acetate
ion is the capturing nucleophile in the photoacetolysis of 1aBr
in acetic acid is confirmed by irradiations of 1aBr with only
one or neither of the nucleophiles NaOAc and Et4NBr[82Br]
present (Table 5, entries 2 vs 3 and 4 vs 5): ΦOAc is virtually
independent of the presence of NaOAc.
The change from acetic acid as capturing nucleophile in the
photoacetolysis of the R-phenyl-substituted compound 1aBr to
NaOAc in the photoacetolysis of the R-(p-methoxyphenyl)-
substituted compounds 1cBr and 1dBr is reflected in the values
of the activation parameters listed in Table 6. Because the vinyl
cations 1+ are formed with zero ∆Hq (vide supra) all activation
parameters for the acetolysis and exchange processes listed in
Table 6 pertain to the reaction of the vinyl cations with the
nucleophiles Br-, OAc-, and HOAc. The values of the enthalpy
of activation (∆Hq) for the reactions with the anionic nucleo-
philes Br- (1.5 kcal ·mol-1 for 1aBr and 1.7 kcal ·mol-1 for
1dBr) and OAc- (2.0 kcal ·mol-1 for 1dBr) are virtually
independent of the nature of the vinyl cation (1a+ or 1d+). The
∆Hq values for the reaction with the solvent acetic acid,
however, show a pronounced substituent effect: 0.0 kcal ·mol-1
for the triphenyl vinyl cation 1a+ and 6.5 kcal ·mol-1 for the
more stable tris(p-methoxyphenyl) vinyl cation 1d+. That is
consistent with the reaction of the vinyl cations 1+ with the
solvent acetic acid being activation-controlled. The activation
energies of the reactions of the vinyl cations 1+ with the anionic
nucleophiles Br- and OAc- on the other hand are at 2.1-2.6
kcal·mol-1 almost equal and within the 2-4 kcal·mol-1 range39
associated with Ingold’s “collapse theory”. These small but
distinct activation energies represent the need for (partial)
desolvation of the ions to get sufficiently close to each other to
commence the release of bond energy in excess.39a,40a That is
consistent with the ionic combination reactions being (de)sol-
vation-controlled.
The thermal nucleophilic substitution of 1dBr5 occurs via
the SN2 (C+) mechanism,40b,41 a rarely observed variation on
the SN1 mechanism in which the cation-nucleophile reaction
is rate determining under solvolytic conditions. The activation
data (∆Hq is larger for the reaction of the stabilized tris(p-
methoxyphenyl) vinyl cation 1d+ with the weakly nucleophilic
solvent acetic acid than for the ionic combination reactions of
1d+ with Br- and OAc-) show that the same happens photo-
chemically. The reactions of the less stable triphenyl vinyl cation
1a+ are the second steps of an SN1 process: ∆Hq is larger for
the ionic combination reactions of 1a+ with Br- and OAc- than
for its reaction with the solvent HOAc.42
(b) Selectivity. When the selectivities S (SHOAc and SNaOAc
only differ by a constant concentration factor) of the vinyl
cations 1a-c+ listed in Table 1 are subjected to Hammett
treatment with reference to σ+, the same FS values are obtained
(38) Richard, J. P.; Toteva, M. M.; Crugeiras, J. J. Am. Chem. Soc. 2000,
122, 1664–1674.
(39) (a) Hawdon, A. R.; Hughes, E. D.; Ingold, C. K. J. Chem. Soc. 1952,
2499–2503. (b) Bunton, C. A.; del Pesco, T. W.; Dunlop, A. M.; Yang, K. U.
J. Org. Chem. 1971, 36, 887–897.
(40) Ingold, C. K. Structure and Mechanism in Organic Chemistry, 2nd ed.;
Cornell University Press: Ithaca, 1969; Chapter 7, (a) pp 494-496; (b) p 470.
(41) Kinetic evidence for such a process has recently been obtained in the
case of the bis(p-methoxyphenyl)methyl cation: Mayr, H.; Minegishi, S. Angew.
Chem., Int. Ed. 2002, 41, 4493–4495.
(42) Minegishi, S.; Loos, R.; Kobayashi, S.; Mayr, H. J. Am. Chem. Soc.
2005, 127, 2641–2649.
5424 J. Org. Chem. Vol. 73, No. 14, 2008

Photogeneration and SelectiVity of Triaryl Vinyl Cations
in acetic acid (FS,HOAc(σ+) ) -1.60 ((0.31; R2 ) 0.963)) and
in acetonitrile (FS,CH3CN(σ+) ) -1.55 ((0.10; R2 ) 0.996))
(Figures 1 and 2). The presence of OAc- in acetic acid but not
in acetonitrile makes no difference in this respect, even though
OAc- is the main capturing nucleophile leading to 2c for the
vinyl cation 1c+, whereas HOAc is the main capturing nucleo-
phile leading to 2a for the vinyl cation 1a+. The (de)solvation-
controlled reaction with OAc- occurs with almost constant
efficiency over the investigated range, about 1.1 × 10-2 for
1a+ (Table 5, the difference between ΦOAc in entries 2 and 3)
and about 1.5 × 10-2 for 1c+ (Table 1, entry 3), and
consequently does not contribute significantly to the FS(σ+)
values. These are solely determined by the reactions with Br-
and HOAc.
Principle (RSP)46,47 both within the series 1aBr, 1bBr, 1cBr
(R-aryl substituent effect) and 1cBr, 1dBr, 1eBr (ꢀ-aryl
substituent effect): the better stabilized vinyl cation is the more
selective in both media employed.
The selectivity data in Table 4 (last two columns) show that
the same applies for the selectivities S of the vinyl cations 1a+
and 1d+ (photogenerated from 1aCl and 1dCl) for their
reactions with Cl- vs NaOAc/HOAc: 1d+ is more selective than
1a+.
-/Cl-
On the other hand, for the selectivities SBr
of the vinyl
cations 1a+ and 1d+ toward the two anionic nucleophiles Br-
and Cl- (calculated by dividing the SHOAc values obtained in
the photolyses of 1aBr and 1aCl, respectively, 1dBr and
1dCl),48 a different picture emerges. In acetic acid SBr
) 4
-/Cl-
-/Cl-
for both 1a+ and 1d+. Similarly, in acetonitrile SBr
) 4 for
Vinyl cations have also been generated by laser flash
photolysis, and actual rates of their reactions with several
1d+ and 2.5 for 1a+. This appears to be an adherence to
Ritchie’s constant selectivity rule recalled in eq 6, with kNu the
reaction rate of the cation with the nucleophile, k0 a parameter
only depending on the identity of the cation (usually the rate of
the reaction of the cation with the solvent, originally kH2O), and
N+ a parameter depending only on the identity of the nucleophile
and the solvent.49
nucleophiles have been measured.6,7,12 According to eq 4b,
+
FS(σ+) ) FBr (σ ) s FHOAc(σ+), and, of course, one wishes to
-
compare the FS(σ+) values obtained in the continuous irradiation
experiments of the vinyl bromides 1a-cBr reported here with
+
values of FBr (σ ) s FHOAc(σ+) for the reactions of the vinyl
-
cations 1+ with bromide ion and acetic acid, respectively. Using
+
-
the flash photolysis method, a value of FBr (σ ) ) +2.0 (cation-
reactivity results in sign inversion compared with cation
formation) was obtained for the reaction of ions 1b+ and 1c+
with Br- in acetonitrile.43a A value of FHOAc(σ+), however, could
not be determined:43 Stern-Volmer plots for the reaction of
1c+ with acetic acid are nonlinear, and low concentrations of
HOAc in acetonitrile actually increase the lifetime of 1c+ rather
than decrease it.44 The value of FHOAc(σ+) can, however, be
kNu
log
) N+
(6)
[
]
k0
The Ritchie equation has recently been extended by Mayr
and co-workers for kNu < 108 M-1 s-1 to eq 7, with kNu the
reaction rate (in M-1 s-1) of the cation with the nucleophile at
20 °C, sE an electrophile-specific slope parameter (sE ) 1 for
reactions of carbocations with nucleophiles), sN a nucleophile-
specific slope parameter, E a nucleophile-independent electro-
philicity parameter, and N an electrophile-independent nucleo-
philicity parameter.42,46g,50
+
estimated as +3.6 () -FS(σ+) + FBr (σ )). This agrees well
with the value of FEtOH(σ+) ) +3.4 reported by Kobayashi and
Schnabel for the reaction of a series of multisubstituted triaryl
vinyl cations with another protic nucleophile, ethanol.6c
-
The values of FS(σ+) for the cations 1a-c+ are the same as
those listed in Table 2 for the R-aryl substituent effect on their
rate of photochemical formation Φion/τ, Fτ(σ+) (Figures 1 and
2). The same equality, FS(σ+) ) Fτ(σ+), has been found in
irradiations of vinyl bromides 8Br in methanol in the presence
of Et4NBr[82Br].12,29 Apparently, the difference in energy
content of the transition states for the reactions of the vinyl
log kNu ) sEsN(E + N)
(7)
The Ritchie N+ values for Br- and Cl- are 2.2 and 1.2,
respectively;38 i.e., N+ - N+ ) 1.0. According to eq 6
Br-
Cl-
-/Cl-
-/Cl-
this should give a value of log SBr
agreement with the value of log SBr
the selectivities in our competition experiments, taking the
) 1.0, which is in good
) 0.6 we calculate from
-
difference in solvent (H2O vs HOAc) into account. Mayr’s NBr
and NCl values are not available for the solvents we use, but
cations 1+ with Br- and HOAc (FS(σ+) ) FBr (σ ) - FHOAc(σ+))
is subject to the same electronic substituent effect as the energy
content of the transition states for the photochemical formation
of the vinyl cations 1+ (Fτ(σ+)). In the case of diarylmethyl
cations, it has been shown that bond making in the transition
state is 50-65% advanced for the reaction with the protic
nucleophile water and ∼25% advanced for the reaction with
Br-.45 If similar values would apply for the reaction of the vinyl
cations 1+ with Br- and HOAc the equality of FS(σ+) and Fτ(σ+)
would put the (early) transition state of the photochemical
formation of the cations 1+ at 25-40% on the reaction
coordinate.
+
-
-
-
-
in the solvents for which they have been reported, NBr - NCl
≈ 1.5,42 still within one log unit of the number (0.6) we find.
The selectivity behavior of carbocations in general and the
discrepancy between the RSP and Ritchie behavior in particular
have been the subject of considerable research, and perhaps more
exceptions than adherences to the RSP have been reported.
(46) (a) Pross, A. AdV. Phys. Org. Chem. 1977, 14, 69–132. (b) Giese, B.
Angew. Chem. 1977, 89, 162–173. (c) Johnson, C. D. Tetrahedron 1980, 36,
3461–3480. (d) Arnett, E. M.; Molter, K. E. Acc. Chem. Res. 1985, 18, 339–
346. (e) Richard, J. P. Tetrahedron 1995, 51, 1535–1573. (f) Richard, J. P.;
Amyes, T. L.; Toteva, M. M. Acc. Chem. Res. 2001, 34, 981–988. (g) Mayr, H.;
Ofial, A. R. Angew. Chem., Int. Ed. 2006, 45, 1844–1854.
(47) (a) Johnson, C. D.; Stratton, B. J. Chem. Soc., Perkin Trans. 2 1988,
1903–1907. (b) Denton, P.; Johnson, C. D. J. Chem. Soc., Perkin Trans. 2 1995,
477–481.
(c) Reactivity-Selectivity. The selectivities S for the reac-
tions of the vinyl cations 1+ with Br- vs NaOAc/HOAc listed
inTable1(lasttwocolumns)followtheso-calledReactivity-Selectivity
(48) This approach presumes that ΦOAc is independent of the leaving group
(Br/Cl).
(43) (a) Lodder, G. Unpublished results. (b) Krijnen, E. S. Ph.D. Thesis,
University of Leiden, The Netherlands, 1992.
(49) (a) Ritchie, C. D. Acc. Chem. Res. 1972, 5, 348–354. (b) Ritchie, C. D.
Can. J. Chem. 1986, 64, 2239–2250.
(44) A lifetime extension by acetic acid has also been observed in a flash
photolysis study of the vinyl cation 8c+.12 Alcohols have been reported to exert
a similar effect on the retinyl cation: Pienta, N. J.; Kessler, R. J. J. Am. Chem.
Soc. 1992, 114, 2419–2428.
(50) (a) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36, 66–
77. (b) Minegishi, S.; Mayr, H. J. Am. Chem. Soc. 2003, 125, 286–295. (c)
Minegishi, S.; Kobayashi, S.; Mayr, H. J. Am. Chem. Soc. 2004, 126, 5174–
5181. (d) Phan, T. B.; Breugst, M.; Mayr, H. Angew. Chem., Int. Ed. 2006, 45,
3869–3874.
(45) Van Pham, T.; McClelland, R. A. Can. J. Chem. 2001, 79, 1887–1897.
J. Org. Chem. Vol. 73, No. 14, 2008 5425

van Dorp and Lodder
Therefore, the term “Reactivity-Selectivity Principle” is ques-
tionable as the word “principle” is misleading, but this is the
term used by most authors.46,47,49–53 Ta-Shma and Rappoport
have first shown that in many cases which follow the RSP the
reaction with the less reactive nucleophile is activation-
controlled and the reaction with the more reactive nucleophile
is diffusion-controlled.51 Jencks, Richard, and co-workers have
reported several more cases following that pattern52 but have
also pioneered the concept that in other cases that follow
the RSP the reaction with the less reactive nucleophile is
activation-controlled and the reaction with the more reactive
nucleophile is (de)solvation-controlled.53
We propose that our selectivity results are fully consistent
with the reaction of the vinyl cations 1+ with acetic acid being
activation controlled and that with the anionic nucleophiles Br-
and OAc- being (de)solvation controlled. This is based on the
activation data in Table 6 and the analysis of the capturing
nucleophile and supported by the R-aryl substituent effect on
the selectivities. Flash photolysis studies have shown that the
reaction of the 1-(p-methoxyphenyl)vinyl cation 1c+ with Br-
in acetonitrile is not diffusion controlled: the reaction rate (k )
2 × 109 M-1 s-1)7 is an order of magnitude lower than that for
the reaction of the vinyl cation 8c+ with Br- (k ) 1.7 × 1010
M-1 s-1).12 McClelland, Steenken, and co-workers have ob-
served the influence of nucleophile solvation on the rates of
cation-nucleophile reactions in flash photolysis studies.45,54a,b
Moreover, they measured a rate 3-4 orders of magnitude below
the diffusion-limited value for the reaction of 1c+ with Br- (k
) 6 × 106 M-1 s-1) in TFE and also found that that even 1a+,
the least stabilized vinyl cation in the series 1+, reacts with Br-
at a rate (k ) 8.2 × 108 M-1 s-1) at least an order of magnitude
below the diffusion-limited value in 1,1,1,3,3,3-hexafluoro-2-
propanol.54c
which ion-pair and radical-pair formation are parallel processes:
the partitioning into ion and radical occurs in the excited state.
Mechanistically, the photochemistry of R-aryl-substituted vinyl
halides such as 1X thus occupies a position in between benzyl
halides, which follow mechanism A or B,55 and R-H- and
R-alkyl-substituted vinyl halides, which follow mechanism C.9
This reflects the both benzylic and vinylic nature of the reaction
center of the compounds 1X. The mechanism directing the
photoreactivity of vinyl compounds depends on structural effects
as it does for aromatic compounds.14
Selectivity. The activation data for the reactions of the vinyl
cations 1+ with Br-, OAc-, and HOAc are consistent with
Ingold’s “collapse theory”.39a,40a The reactions of the vinyl
cations 1+ with the anionic nucleophiles Br- and OAc- are
(de)solvation controlled and their reaction with acetic acid is
activation controlled. Therefore, the selectivities of the vinyl
cations 1+ toward bromide ion and acetic acid follow the
Reactivity-Selectivity Principle.46,47
Experimental Section
Procedure. The experimental setup as well as the sampling and
workup procedure have been described previously,4a and so have
the kinetics4b and the 3-nitroanisole actinometry.56,57 Irradiation
at λexc ) 313 nm was achieved using a Hanau TQ 81 high-pressure
mercury arc in combination with a filter solution.4a
The concentration of the vinyl halides 1X was always 4.7 ×
10-3 M. If NaOAc was present in the reaction mixture it was so in
a concentration of 87 × 10-3 M. The (labeled) tetraethylammonium
halide Et4NBr[82Br] or Et4NCl[36Cl] used in the irradiation mixtures
was invariably that with the halide in common with the substrate;
if present it was so in a concentration of 71 × 10-3 M. In the
irradiation mixtures of compounds 1X in acetonitrile spiked with
acetic acid, the latter was present in a concentration of 1.0 M.
For all systems studied, control experiments in the dark showed
that neither product formation nor radioactive halide exchange
occurs thermally under the reaction conditions employed. The
counting procedures for 82Br4a and 36Cl58 have been reported
previously, and so has the analysis of the disappearance of starting
material and the formation of products by means of HPLC
(Chrompack Si 60-5 4.6 mm × 25 cm column, eluents 0-7.5%
THF in hexane) calibrated with standard solutions of the starting
material and products.4a,58
All irradiations were performed in triplicate. The values of the
quantum yields and selectivity constants reported are average values;
the error (i.e., deviation from the mean) is 5-10%. An error of
5% in the quantum yields leads to an error of 0.5 kcal ·mol-1 in
the activation parameters.
Purification of Reagents and Solvents. Sodium acetate, tetra-
ethylammonium bromide, and tetraethylammonium chloride were
analytical grade and used as such. Acetonitrile was spectroscopic
grade. Hexane (technical quality) and THF (from CaH2) were
distilled before use. Acetic acid was prepared by refluxing analytical
glacial acid with 1% acetic anhydride and distilling and collecting
the middle fraction.59
Conclusions
In this study, we have been concerned with the photoreactivity
of a series of triaryl vinyl halides 1X, as well as with the
selectivity of the photogenerated vinyl cations 1+.
Photoreactivity. The R- and ꢀ-aryl substituent effects, the
leaving group effect, and the temperature effect on the photo-
reactivity of the compounds 1X were investigated and inter-
preted in terms of the mechanisms A, B, and C (Scheme 1).
The ꢀ-aryl substituent effect does not allow mechanistic
conclusions, but the R-aryl substituent effect, the leaving group
effect, and the temperature effect all argue against mechanism
A, which involves initial homolytic CsX bond scission to a
radical pair followed by electron tranfer to form an ion pair,
and against mechanism C, which involves initial heterolytic
CsX bond scission to an ion pair followed by electron tranfer
to form a radical pair. The effects support mechanism B, in
(51) (a) Ta-Shma, R.; Rappoport, Z. J. Am. Chem. Soc. 1983, 105, 6082–
6095. (b) Ta-Shma, R.; Rappoport, Z. AdV. Phys. Org. Chem. 1992, 27, 239–
291.
Synthesis of Starting Materials. 1-Bromo-1,2,2-triphenylethene
(1aBr) was commercially available and used as such. The other vinyl
(52) (a) Richard, J. P.; Rothenberg, M. E.; Jencks, W. P. J. Am. Chem. Soc.
1984, 106, 1361–1372. (c) Richard, J. P.; Jencks, W. P. J. Am. Chem. Soc. 1984,
106, 1373–1383.
(55) (a) Cristol, S.; Bindel, T. H. Org. Photochem. 1983, 6, 327–415. (b)
Pincock, J. A. Acc. Chem. Res. 1997, 30, 43–49. (c) Fleming, S. A.; Pincock,
J. A. Org. Mol. Photochem. 1999, 3, 211–281.
(53) (a) Jencks, W. P.; Brent, S. R.; Gandler, J. R.; Fendrich, G.; Nakamura,
C. J. Am. Chem. Soc. 1982, 104, 7045–7051. (b) Jencks, W. P.; Haber, M. T.;
Herschlag, D.; Nazaretian, K. L. J. Am. Chem. Soc. 1986, 108, 479–483. (c)
Richard, J. P. J. Chem. Soc., Chem. Commun. 1987, 1768–1769. (d) Richard,
J. P.; Amyes, T. L.; Vontor, T. J. Am. Chem. Soc. 1992, 114, 5626–5634.
(54) (a) McClelland, R. A.; Kanagasabapathy, V. M.; Banait, N. S.; Steenken,
S. J. Am. Chem. Soc. 1992, 114, 1816–1823. (b) McClelland, R. A. Tetrahedron
1996, 52, 6823–6858. (c) Cozens, F. L.; Kanagasabapathy, V. M.; McClelland,
R. A.; Steenken, S. Can. J. Chem. 1999, 77, 2069–2082.
(56) van Riel, H. C. H. A.; Lodder, G.; Havinga, E. J. Am. Chem. Soc. 1981,
103, 7257–7262.
(57) (a) Havinga, E.; de Jongh, R. O. Bull. Soc. Chim. Belg. 1962, 71, 803–
810. (b) de Jongh, R. O.; Havinga, E. Recl. TraV. Chim. Pays-Bas 1966, 85,
275–283.
(58) Lodder, G.; van Dorp, J. W. J.; Avramovitch, B.; Rappoport, Z. J. Org.
Chem. 1989, 54, 2574–2580.
(59) Rappoport, Z.; Apeloig, G. J. Am. Chem. Soc. 1969, 91, 6734–6742.
5426 J. Org. Chem. Vol. 73, No. 14, 2008

Photogeneration and SelectiVity of Triaryl Vinyl Cations
bromides (1bBr-1eBr) were prepared by bromination-dehy-
drobromination60 in CCl4 of the corresponding ethylene compounds
4b-e, which in turn were synthesized by the procedure described in
ref 61. The vinyl chlorides were also prepared from the corresponding
ethylene compounds, 1aCl according to ref 62 and 1dCl according to
ref 61. The spectral data of 1-chloro-1,2,2-triphenylethene (1aCl),63
1-chloro-1,2,2-tris(p-methoxyphenyl)ethene (1dCl),64 1-bromo-1-(p-
methoxyphenyl)-2,2-diphenylethene (1cBr),59 and 1-bromo-1,2,2-
tris(p-methoxyphenyl)ethene (1dBr)59 have been reported.
intensity) 328 (M+, 11), 286 (M+ - ketene, 100); HRMS (EI, 70
eV) m/z 328.1468 (C23H20O2 requires 328.1463).
1-Acetoxy-1-(p-methoxyphenyl)-2,2-bis(p-chlorophenyl)ethene (2e):
UV (hexane) λmax 239 nm (ꢀ 19800), 263 (8800), 266 (9500), 301
(16300); 1H NMR (CDCl3, 400 MHz) δ 2.00 (s, 3H), 3.77 (s, 3H),
6.72 (d, 2H, J ) 9 Hz), 6.98 (d, 2H, J ) 9 Hz), 7.12-7.16 (m,
6H), 7.26-7.30 (m, 2H); 13C NMR (CDCl3, 100 MHz) δ 20.9,
55.2, 113.7, 128.5, 128.6, 130.3, 130.5, 132.1 (APT -); δ 127.5,
128.6, 133.1, 133.3, 138.1, 138.3, 144.7, 159.6 169.6 (APT +);
MS (EI, 70 eV) m/z (relative intensity) 416, 414, 412 (M+, 2, 8,
14), 374, 372, 370 (M+ - ketene, 10, 63, 100); HRMS (EI, 70
eV) m/z 412.0638 (C23H18O3Cl2 requires 412.0633).
1-Bromo-1-(p-methylphenyl)-2,2-diphenylethene (1bBr)65: UV
(hexane) λmax 215 nm (ꢀ 13500), 234 (22800), 267 (7000), 298
1
(10100); H NMR (CDCl3, 400 MHz) δ 2.26 (s, 3H), 6.95-6.97
(m, 4H), 7.04-7.07 (m, 3H), 7.20 (d, 2H, J ) 8 Hz), 7.27-7.31
(m, 1H), 7.36 (d, 4H, J ) 4 Hz); 13C NMR (CDCl3, 100 MHz) δ
21.2, 126.8, 127.4, 127.8, 128.1, 128.7, 129.5, 130.2, 130.3
(Attached Proton Test66 (APT) -); δ 122.5, 137.8, 138.1, 141.2,
143.0, 143.9 (APT +); MS (EI, 70 eV): m/z (relative intensity)
350, 348 (M+, 39, 38), 269 (M+ - Br, 100); HRMS (EI, 70 eV)
m/z 348.0517 (C21H17Br requires 348.0514).
1-Bromo-1-(p-methoxyphenyl)-2,2-bis(p-chlorophenyl)ethene
(1eBr): UV (hexane) λmax 214 nm (ꢀ 19800), 220 (21100), 245
(25400), 307 (12900); 1H NMR (CDCl3, 400 MHz) δ 3.77 (s, 3H),
6.71-7.06 (AA′BB′, 4H), 6.84-7.23 (AA′BB′, 4H), 7.26-7.35
(AA′BB′, 4H); 13C NMR (CDCl3, 100 MHz) δ 55.2, 113.6, 128.2,
128.5, 131.1, 131.5, 131.6 (APT -); δ 123.6, 132.8, 132.9, 133.5,
139.4, 140.2, 141.9, 159.4 (APT +); MS (EI, 70 eV) m/z (relative
intensity) 438, 436, 434, 432 (M+, 5, 22, 45, 30), 357, 355, 353
(M+ - Br, 11, 67, 100), 320, 318 (M+ - Br - Cl, 3, 14); HRMS
(EI, 70 eV) m/z 431.9686 (C21H15OBrCl2 requires 431.9684).
Tetraethylammonium bromide[82Br] was prepared by neutron
irradiation of Et4NBr,5 tetraethylammonium chloride[36Cl] according
to the method described in ref 67.
Identification of Products. The photoproducts were separated
from the irradiation mixtures according to the method reported
in ref 4a. The photoproducts were identified on the basis of their
NMR, mass, and UV spectra and their HPLC retention times
and by comparison with authentic samples. Of the photoproducts,
benzophenone (3a), 4,4′-dimethoxybenzophenone (3d), 4,4′-
dichlorobenzophenone (3e), and triphenylethene(4a) were com-
mercially available. 1,1-Diphenyl-2-(p-methoxyphenyl)ethene
(4c) and tris(p-methoxyphenyl)ethene (4d) were intermediate
products in the synthesis of the starting materials 1cBr and 1dBr/
Cl.
1-Acetoxy-1,2,2-triphenylethene (2a) and 1-acetoxy-1,2,2-tris(p-
methoxyphenyl)ethene (2d) were prepared according to ref 68. The
remaining vinyl acetates 2 were synthesized in an analogous fashion.
The spectral data of 2a and 2d68 as well as of 1-acetoxy-1-(p-
methoxyphenyl)-2,2-diphenylethene (2c)69 have been reported.
1-Acetoxy-1-(p-methylphenyl)-2,2-diphenylethene (2b): UV (hex-
ane) λmax 229 nm (ꢀ 20600), 257 (7800), 260 (7600), 267 (9500),
290 (15000); 1H NMR (CDCl3, 400 MHz) δ 1.97 (s, 3H), 2.26 (s,
3H), 6.96 (d, 2H, J ) 8 Hz), 7.07-7.16 (m, 7H), 7.22-7.33 (m,
5H); 13C NMR (CDCl3, 100 MHz) δ 20.9, 21.3, 127.0, 127.3, 128.0,
128.1, 128.7, 128.8, 129.0, 130.8 (APT -); δ 131.4, 132.9, 138.0,
139.9, 140.2, 144.0, 169.8 (APT +); MS (EI, 70 eV) m/z (relative
The compounds 5X, 6, and 7 were prepared by independent
photocyclization13 in acetic acid of the corresponding compounds
1X, 2, and 4. For some of the phenanthrene derivatives 5X, 6, and
7, spectral data have already been reported:4a 10-bromo-3-methoxy-
9-phenylphenanthrene (5cBr), 9-bromo-3,6-dimethoxy-10-(p-meth-
oxyphenyl)phenanthrene (5dBr), 10-acetoxy-3-methoxy-9-phe-
nylphenanthrene (6c), 9-acetoxy-3,6-dimethoxy-10-(p-methoxy-
phenyl)phenanthrene (6d), 3-methoxy-9-phenylphenanthrene (7c),
3,6-dimethoxy-9-(p-methoxyphenyl)phenanthrene (7d), and 9-phe-
nylphenanthrene (7a).70
9-Bromo-10-phenylphenanthrene (5aBr)71: UV (hexane) λmax
201 nm (ꢀ 22700), 225 (16500), 251 (32100), 258 (39700), 271
(19000), 281 (15100), 290 (13700), 303 (11800); 1H NMR (CDCl3,
400 MHz) δ 7.34–7.74 (multiple multiplets (mm), 10H), 8.53 (d,
1H, J ) 8 Hz), 8.71-8.75 (dd, 2H); 13C NMR (CDCl3, 100 MHz)
δ 122.6, 126.8, 127.0, 127.4, 127.6, 127.7, 127.9, 128.4, 129.0,
130.0 (APT -); δ 123.6, 129.7, 130.5, 131.0, 132.6, 139.7, 141.0
(APT +); MS (EI, 70 eV) m/z (relative intensity) 334, 332 (M+,
35, 36), 253 (M+ - Br, 25), 252 (M+ - Br - H, 38), 51 (C4H3,
100); HRMS (EI, 70 eV) m/z 332.0207 (C20H13Br requires
332.0201).
10-Bromo-3-methyl-9-phenylphenanthrene (5bBr): UV (hexane)
λmax 208 nm (ꢀ 39100), 230 (34700), 252 (58300), 259 (76400),
1
274 (24700), 282 (16000), 293 (16000), 305 (17400); H NMR
(CDCl3, 400 MHz) δ 2.66 (s, 3H), 7.33-7.65 (mm, 9H), 8.40 (d,
1H, J ) 8 Hz), 8.52 (s, 1H), 8.71 (d, 1H, J ) 8 Hz); 13C NMR
(CDCl3, 100 MHz) δ 21.9, 122.4, 122.6, 126.6, 126.9, 127.6, 127.9,
128.4, 128.8, 129.4, 130.1 (APT -); δ 123.5, 128.5, 131.0, 132.7,
137.3, 138.6, 141.1 (APT +); MS (EI, 70 eV) m/z (relative intensity)
348, 346 (M+, 90, 92), 267 (M+ - Br, 43), 252 (M+ - Br - CH3,
100); HRMS (EI, 70 eV) m/z 346.0354 (C21H15Br requires
346.0357).
9-Bromo-3-chloro-6-methoxy-10-(p-chlorophenyl)phenan-
threne (5eBr): UV (hexane) λmax 243 nm (ꢀ 34100), 253 (39100),
261 (40000), 284 (13600), 304 (10200), 316 (10700), 349 (1500),
366 (1400); 1H NMR (CDCl3, 400 MHz) δ 4.04 (s, 3H), 7.22-7.78
(mm, 8H), 7.92 (d, 1H, J ) 2 Hz), 8.39 (d, 1H, J ) 9 Hz), 8.56 (d,
1H, J ) 2 Hz); 13C NMR (CDCl3, 100 MHz) δ 55.6, 104.0, 118.3,
122.4, 127.6, 128.8, 129.1, 130.8, 131.6 (APT -); δ 123.7, 125.2,
130.2, 130.9, 131.3, 132.7, 133.9, 135.4, 138.9, 159.3 (APT +);
MS (EI, 70 eV) m/z (relative intensity) 436, 434, 432, 430 (M+, 4,
26, 57, 33), 419, 417, 415 (M+ s CH3, 4, 8, 5), 391, 389, 387
(M+ - CH3 - CO, 5, 10, 7), 303, 301 (M+ - CH3 - Br - Cl, 7,
20), 275, 273 (M+ - CH3 - Br - Cl - CO, 33, 100); HRMS (EI,
70 eV) m/z 429.9532 (C21H13OBrCl2 requires 429.9527).
9-Chloro-10-phenylphenanthrene (5aCl). UV (hexane) λmax 210
nm (ꢀ 31900), 224 (24400), 251 (46200), 258 (57700), 272 (18300),
(60) Koelsch, C. F. J. Am. Chem. Soc. 1932, 54, 2045–2048.
(61) Shelton, R. S.; van Campen, M. G., Jr.; Meisner, D. F.; Parmerter, S. M.;
Andrews, E. R.; Allen, R. E.; Wyckoff, K. K. J. Am. Chem. Soc. 1953, 75,
5491–5495.
1
280 (11900), 291 (11200), 303 (12200); H NMR (CDCl3, 400
(62) Grovenstein, E., Jr J. Am. Chem. Soc. 1957, 79, 4985–4990.
(63) Crenshaw, M. D.; Zimmer, H. J. Org. Chem. 1983, 48, 2782–2784.
(64) Mitchell, R. H.; Mazuch, L.; Shell, B.; West, P. R. Can. J. Chem. 1978,
56, 1246–1252.
MHz) δ 7.33-7.35 (m, 2H), 7.41-7.43 (m, 2H), 7.44-7.59 (mm,
4H), 7.65-7.69 (m, 2H), 8.47 (q, 1H), 8.63-8.69 (m, 2H); 13C
NMR (CDCl3, 100 MHz) δ 122.5, 122.6, 125.9, 126.6, 127.0, 127.3,
127.4, 127.5, 127.7, 128.4, 130.1 (APT -); δ 129.3, 129.4, 129.7,
130.9, 132.3, 136.7, 138.6 (APT +); MS (EI, 70 eV) m/z (relative
(65) Benjamin, B. M.; Collins, C. J. J. Am. Chem. Soc. 1956, 78, 4952–
4956.
(66) Patt, S. L.; Shoolery, J. N. J. Magn. Reson. 1982, 46, 535–539.
(67) Hayami, J.; Tanaka, N.; Kurabayashi, S.; Kotani, Y.; Kaji, A. Bull. Chem.
Soc. Jpn. 1971, 44, 3091–3095.
(68) Rappoport, Z.; Gal, A. J. Am. Chem. Soc. 1969, 91, 5246–5254.
(69) Rappoport, Z.; Gal, A.; Houminer, Y. Tetrahedron Lett. 1973, 641–
644.
(70) Koussini, R.; Lapouyade, R.; Fornier de Violet, P. J. Am. Chem. Soc.
1978, 100, 6679–6683.
(71) Koelsch, C. F. J. Am. Chem. Soc. 1934, 56, 480–484.
J. Org. Chem. Vol. 73, No. 14, 2008 5427

van Dorp and Lodder
intensity) 290, 288 (M+, 31, 93), 253 (M+ - Cl, 78), 252 (M+
-
127.6, 128.2, 128.9, 130.3 (APT -); δ 124.3, 128.3, 131.2, 132.1,
135.7, 136.8, 137.0, 142.4, 169.5 (APT +); MS (EI, 70 eV) m/z
(relative intensity) 326 (M+, 14), 284 (M+ - ketene, 100); HRMS
(EI, 70 eV) m/z 326.1303 (C23H18O2 requires 326.1307).
Cl - H, 100); HRMS (EI, 70 eV) m/z 288.0714 (C20H13Cl requires
288.0706).
9-Chloro-3,6-dimethoxy-10-(p-methoxyphenyl)phenanthrene
(5dCl): UV (hexane) λmax 207 nm (ꢀ 20900), 214 (17800), 240
(23900), 253 (31000), 260 (30700), 289 (11100), 304 (6500), 314
9-Acetoxy-3-chloro-6-methoxy-10-(p-chlorophenyl)phenan-
threne (6e): UV (hexane) λmax 235 nm (ꢀ 26000), 251 (26100),
257 (25800), 271 (13200), 282 (12300), 298 (9100), 310 (8500),
1
(6900), 354 (1100), 372 (1100); H NMR (CDCl3, 400 MHz) δ
1
3.86 (s, 3H), 3.98 (s, 3H), 4.03 (s, 3H), 6.92–7.40 (mm, 7H), 7.97
(q, 1H), 8.05 (d, 1H, J ) 9 Hz), 8.37 (d, 1H, J ) 9 Hz); 13C NMR
(CDCl3, 100 MHz) δ 55.2, 55.4, 55.5, 104.5, 104.6, 113.7, 116.4,
116.9, 127.6, 129.1, 131.4 (APT -); δ 121.6, 124.5, 127.4, 127.7,
130.1, 131.6, 133.8, 158.0, 158.6, 159.0 (APT +); MS (EI, 70 eV)
m/z (relative intensity) 380, 378 (M+, 34, 100), 363 (M+ - CH3,
12); HRMS (EI, 70 eV) m/z 378.1021 (C23H19O3Cl requires
378.1023).
346 (1500), 363 (1100); H NMR (CDCl3, 400 MHz) δ 2.11 (s,
3H), 4.03 (s, 3H), 7.25-7.49 (mm, 7H), 7.78 (d, 1H, J ) 9 Hz),
7.96 (d, 1H, J ) 2 Hz), 8.57 (s, 1H); 13C NMR (CDCl3, 100 MHz)
δ 20.4, 55.6, 104.2, 118.3, 122.4, 124.0, 127.5, 128.3, 128.7, 131.7
(APT -); δ 121.1, 125.4, 129.7, 130.5, 132.2, 133.7, 133.9, 142.7,
159.3, 169.2 (APT +); MS (EI, 70 eV) m/z (relative intensity) 414,
412, 410 (M+, 2, 14, 19), 372, 370, 368 (M+ - ketene, 12, 68,
100); HRMS (EI, 70 eV) m/z 410.0485 (C23H16O3Cl2 requires
410.0476).
9-Acetoxy-10-phenylphenanthrene (6a): UV (hexane) λmax 211
nm (ꢀ 29000), 220 (19300), 248 (38000), 254 (44900), 270 (13100),
276 (10100), 286 (6200), 299 (8500); 1H NMR (CDCl3, 400 MHz)
δ 2.06 (s, 3H), 7.30-7.89 (mm, 11H), 8.70-8.73 (dd, 2H, J ) 7
Hz); 13C NMR (CDCl3, 100 MHz) δ 20.3, 122.2, 122.7, 122.9,
126.3, 126.9, 127.1, 127.2, 127.7, 128.3, 130.2 (APT -); δ 126.4,
129.2, 129.4, 131.1, 131.9, 135.6, 142.3, 169.4 (APT +); MS (EI,
70 eV) m/z (relative intensity) 312 (M+, 17), 270 (M+ - ketene,
100); HRMS (EI, 70 eV) m/z 312.1146 (C22H16O2 requires
312.1150).
10-Acetoxy-3-methyl-9-phenylphenanthrene (6b): UV (hexane)
λmax 235 nm (ꢀ 16700), 250 (28600), 256 (34200), 272 (11700),
279 (9600), 289 (7700), 300 (8100); 1H NMR (CDCl3, 400 MHz)
δ 2.12 (s, 3H), 2.69 (s, 3H), 7.39-7.66 (mm, 9H), 7.83 (d, 1H, J
) 8 Hz), 8.59 (s, 1H), 8.77 (d, 1H, J ) 8 Hz); 13C NMR (CDCl3,
100 MHz) δ 20.4, 22.1, 122.1, 122.6, 122.7, 126.1, 126.7, 127.1,
Acknowledgment. We thank Professor Jan Cornelisse for
useful discussions, Fons Lefeber for the measurement of the
NMR spectra, Jaap van Houte for the measurement of the mass
spectra, Marcel Plugge for technical assistance, and Dr. Ir. Peter
Bode of the Reactor Institute Delft, The Netherlands, for the
generous and continuous supply of Et4NBr[82Br].
Supporting Information Available: 1H and 13C NMR (APT)
spectra of compounds 1bBr, 1eBr, 2b, 2e, 5aBr, 5bBr, 5eBr,
5aCl, 5dCl, 6a, 6b, and 6e. This material is available free of
charge via the Internet at http://pubs.acs.org.
JO800687Z
5428 J. Org. Chem. Vol. 73, No. 14, 2008