Competition of mechanisms in the photochemical cleavage of the C-X bond of aryl-substituted vinyl halides

Article abstract of DOI:10.1021/jo981164p

The photolysis of aryl halides caused homolysis of the carbon-halogen bond and formation of aryl radicals. In contrast, photolysis of vinyl halides can induce both heterolysis of the C-X bond, thereby generating vinyl cations, and homolysis, giving vinyl radicals. Examples of this competition among pathways is reported here for three vinylic precursors, namely, 1,2,2- triphenylbromoethene (1), 1-phenyl-2,2-bis(o-methoxyphenyl)-1-bromoethene (11), and β-bromostyrene (19). Incursion of the photoinduced S(RN)1 process, through the intermediacy of the vinyl radical, is verified in the presence of reducing nucleophiles, such as the enolate ions of ketones, and in part with (EtO)₂PO^-. Conversely, incursion of the heterolytic path, and intermediacy of the vinyl cation, occurs in the presence of weak electron-donor anions, such as NO₂^-, N₃^-, and Cl^-. The vinyl cation produced from 19, which is less stable than those derived from 1 and 11, gives phenylacetylene via an E1-type elimination. An estimate is provided for the intramolecular rate of interception of the vinyl cation derived from 11 by the ortho-methoxy groups of the β-o-anisyl substituents. Finally, evidence against a photoinduced electron transfer from RO^- ions to vinyl halide 1 is presented.

Full text of DOI:10.1021/jo981164p

9292
J . Org. Chem. 1998, 63, 9292-9299
Com p etition of Mech a n ism s in th e P h otoch em ica l Clea va ge of th e
C-X Bon d of Ar yl-Su bstitu ted Vin yl Ha lid es
Carlo Galli,*,† Patrizia Gentili,† Alessandra Guarnieri,† Shinjiro Kobayashi,‡ and
Zvi Rappoport§
Dipartimento di Chimica, Universita` “La Sapienza”, 00185 Roma, Italy, Institute for Fundamental
Research of Organic Chemistry, Kyushu University, Fukuoka 812, J apan, and Department of Organic
Chemistry, The Hebrew University, J erusalem 91904, Israel
Received J une 17, 1998
The photolysis of aryl halides causes homolysis of the carbon-halogen bond and formation of aryl
radicals. In contrast, photolysis of vinyl halides can induce both heterolysis of the C-X bond, thereby
generating vinyl cations, and homolysis, giving vinyl radicals. Examples of this competition among
pathways is reported here for three vinylic precursors, namely, 1,2,2-triphenylbromoethene (1),
1-phenyl-2,2-bis(o-methoxyphenyl)-1-bromoethene (11), and â-bromostyrene (19). Incursion of the
photoinduced SRN1 process, through the intermediacy of the vinyl radical, is verified in the presence
of reducing nucleophiles, such as the enolate ions of ketones, and in part with (EtO)2PO-. Conversely,
incursion of the heterolytic path, and intermediacy of the vinyl cation, occurs in the presence of
weak electron-donor anions, such as NO2-, N3-, and Cl-. The vinyl cation produced from 19, which
is less stable than those derived from 1 and 11, gives phenylacetylene via an E1-type elimination.
An estimate is provided for the intramolecular rate of interception of the vinyl cation derived from
11 by the ortho-methoxy groups of the â-o-anisyl substituents. Finally, evidence against a
photoinduced electron transfer from RO- ions to vinyl halide 1 is presented.
Sch em e 1
The photolysis of aryl halides causes homolysis of the
C-X bond.1 In contrast, for vinyl halides (VyX) the
heterolysis of the C-X bond is reported to occur under
comparable photochemical conditions (Scheme 1).1
Certainly, this scheme delineates two extreme types
of behavior, and less clear-cut situations are also docu-
mented.1,2 In any event, it was recently learned that
photochemically excited vinyl halides partition between
homolysis and heterolysis of the C-X bond (Scheme 2),3
but evidence was also provided that the latter pathway
in general prevails for aryl-substituted vinylic structures
in solution.1,3,4
Sch em e 2
Ch a r t 1. Releva n t Th er m och em ica l Da ta (in
k ca l/m ol) of P h en yl a n d Vin yl Br om id es a n d th e
Der ived Ra d ica l a n d Ca tion In ter m ed ia tes
This significant difference in the photochemical reac-
tion pathways of ArX and VyX species is certainly
puzzling, in view of the similarity of their functional
group, i.e., the C(sp2)-halogen bond. From a thermo-
chemical point of view, on comparing the gas-phase
homolytic5 and heterolytic6 bond dissociation energy
values of the two typical parent precursors, i.e., Ph-Br
and CH2dCH-Br (Chart 1), or the stability of the
* Corresponding
author.
Fax:
+39-06-490421.
E-mail:
cgalli@uniroma1.it.
† Universita´ “La Sapienza”.
‡ Kyushu University.
§ The Hebrew University.
(1) Lodder, G.; Cornelisse, J . In The Chemistry of Functional Groups.
Supplement D2: The Chemistry of Halides, Pseudo-halides and
Azides: Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, U.K., 1995;
Chapter 16.
possibly derived reactive intermediates,6 only minor
differences exist.
(2) Kropp, P. J . Acc. Chem. Res. 1984, 17, 131.
(3) (a) van Ginkel, F. I. M.; Cornelisse, J .; Lodder, G. J . Am. Chem.
Soc. 1991, 113, 4261. (b) Verbeek, J .-M.; Stapper, M.; Krijnen, E. S.;
van Loon, J .-D.; Lodder, G.; Steenkeen, S. J . Phys. Chem. 1994, 98,
9526.
(4) (a) Lodder, G. In Dicoordinated Carbocations; Rappoport, Z.,
Stang, P. J ., Eds.; Wiley: Chichester, U.K., 1997; Chapter 8. (b) Suzuki,
T.; Kitamura, T.; Sonoda, T.; Kobayashi, S.; Taniguchi, H. J . Org.
Chem. 1981, 46, 5324.
The D(R-X) values derive from the heats of formation
of the two intermediates resulting from homolytic or
heterolytic cleavage of the C-X bond.6 Unfortunately, the
dynamic of the photocleavage of these two chromophores1,3
has not yet been investigated as thoroughly as in the case
of the C-X bond of diphenylmethyl halides, where the
(5) Galli, C.; Pau, T. Tetrahedron 1998, 54, 2893.
10.1021/jo981164p CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/24/1998

Photochemical Cleavage of Vinyl Halides
J . Org. Chem., Vol. 63, No. 25, 1998 9293
Ch a r t 2. Rela tive Ca r boca tion Sta biliza tion
Sch em e 3
+
En er gy to C2H3 (in k ca l/m ol) fr om Differ en ces in
H- Affin ity
partitioning between radical pair (R•X•) and contact ion
pair (R+X-) could be followed kinetically.12 However, we
call the attention upon a structural difference of the
phenyl vs vinyl halide: the parent linear13 vinyl cation
(Vy+), in comparison with the necessarily bent14 phen-
ylium ion (Ph+), is provided with an “extra hand”
enabling additional stabilization by substituents (Chart
conjugation becomes then possible, and it is calculated
(see Experimental Section) to give an additional stabili-
zation of ca. 12 kcal/mol to the 1,3-butadien-2-yl cation.15
Analogously, the R-styryl cation is expected and calcu-
lated15 to gain π-p(C+) stabilization of ca. 24 kcal/mol
whenever full conjugation of the positive charge with the
aromatic ring is conformationally allowed. It is not
unlikely, even though not straightforward, that this
structural effect, namely, the absence of conjugative
stabilization for the Ph+ ion, can play a role in the
dynamic of the photocleavage process, so to help to
explain why photoheterolysis of vinyl halides, bearing
appropriate substituents, may take over their photo-
homolysis (Scheme 1) in solution. An investigation in this
direction is desirable and planned.
+
2). For example, when the R C-H bond of C2H3 is
substituted with a methyl, a remarkable stabilization of
ca. 22 kcal/mol results for CH2dC(+)CH3.15
A similar gain in stabilization is not possible for the
phenylium ion. An even stronger effect results if the R
C-H of the ethenyl cation is substituted with a group
capable of π-conjugation, as for a vinyl substituent being
coplanar with the empty p orbital and perpendicular to
+ 15,16
We became interested in this dichotomy in the behavior
of the vinyl halides in connection with our studies on the
vinylic SRN1 reaction,17 whose propagation cycle is re-
ported in Scheme 3.
the other double bond in C4H5
.
An allylic-type
(6) The homolytic D(R-Br) values given in Chart 1 have already
been reported.5 The heterolytic D(R-Br) values are derived from the
following thermochemical equations, where IE are the ionization
potentials of the radicals: D298(R-Br) ) ∆fH298(R+) + ∆fH298(Br-) -
∆fH298(RBr) (eq a) and ∆fH298(R•) + IE f ∆fH298(R+) (eq b). The IE of
Ph• is 8.32 eV,7 while its ∆fH298 is 81.1 kcal/mol;8 from eq b one
accordingly obtains 272.8 kcal/mol for ∆fH298(Ph+). Analogously, the
IE of CH2dCH• is 8.25 eV,7 and ∆fH298(CH2dCH•) is 71.7 kcal/mol,9
from which ∆fH298(CH2dCH+) ) 261.9 kcal/mol. From ∆fH298(Br-) )
-50.9 kcal/mol,10 ∆fH298(PhBr) ) 25.2 kcal/mol,11 and ∆fH298(CH2d
CHBr) ) 18.9 kcal/mol,11 eq a gives the heterolytic D(R-Br) values of
Ph-Br and of CH2dCHBr as 196.7 and 192.1 kcal/mol, respectively
(Chart 1).
(7) Lias, S. G.; Rosenstock, H. M.; Draxl, K.; Steiner, B. W.; Herron,
J . T.; Holmes, J . L.; Levin, R. D.; Liebman, J . F.; Kafafi, S. A. Ionization
Energetics Data. In NIST Standard Reference Database Number 69;
Mallard, W. G., Linstrom, P. J ., Eds.; National Institute of Standards
and Technology: Gaithersburg, MD, 1997 (http://webbook.nist.gov).
(8) Davico, G. E.; Bierbaum, V. M.; DePuy, C. H.; Ellison, G. B.;
Squires, R. R. J . Am. Chem. Soc. 1995, 117, 2590.
Evidence for the intermediacy of a vinyl radical in this
nucleophilic substitution process was provided,17d in
analogy with the intermediacy of aryl radical in the
reaction of aryl halides.18 In keeping with the aromatic
case, the vinylic SRN1 process is likely to be initiated by
photostimulated electron transfer from a nucleophile19
to the vinyl halide, forming an intermediate radical anion
that then fragments to the vinyl radical. Could the
photochemical induction directly provoke also the het-
erolysis of the C-X bond of the substrate, giving the
corresponding vinyl cation? The product of the nucleo-
philic substitution could then (also) arise from a cation-
to-anion coupling (eq 1).
(9) Ervin, K. M.; Gronert, S.; Barlow, S. E.; Gilles, M. K.; Harrison,
A. G.; Bierbaum, V. M.; DePuy, C. H.; Lineberger, W. C.; Ellison, G.
B. J . Am. Chem. Soc. 1990, 112, 5750.
To better appreciate this subtle feature of the vinylic
SRN1, which is not shared by its aromatic counterpart,
(10) Bartmess, J . E. NIST Negative Ion Energetics Database, Version
3.0, Standard Reference Database 19B; National Institute of Standards
and Technology: Gaithersburg MD, 1993.
(11) Pedley, J . B.; Rylance, J . Sussex NPL Computer Analysed
Thermochemical Data: Organic and Organometallic Compounds;
University of Sussex: Sussex, U.K., 1977.
(12) (a) Lipson, M.; Deniz, A. A.; Peters, K. S. J . Phys. Chem. 1996,
100, 3580. (b) Dreyer, J .; Peters, K. S. J . Phys. Chem. 15156. (c) Lipson,
M.; Deniz, A. A.; Peters, K. S. Chem. Phys. Lett. 1998, 288, 781. (d)
Bartl, J .; Steenken, S.; Mayr, H.; McClelland, R. A. J . Am. Chem. Soc.
1990, 112, 6918.
(13) (a) Mayr, H.; Schneider, R.; Wilhelm, D.; Schleyer, P. v. R. J .
Org. Chem. 1981, 46, 5336. (b) Rappoport, Z. In Reactive Intermediates;
Abramovitch, R. A., Ed.; Plenum Press: New York, 1983; Vol. 3, pp
427-615. (c) Berkowitz, J .; Mayhew, C. A.; Ruscic, B. J . Chem. Phys.
1988, 88, 7397. (d) Lindh, R.; Rice, J . E.; Lee, T. J . J . Chem. Phys.
1991, 94, 8008.
(16) Apeloig, Y.; Mu¨ller, T. In Dicoordinated Carbocations; Rap-
poport, Z.; Stang, P. J ., Eds.; Wiley: Chichester, U.K., 1997; Chapter
2.
(17) (a) Galli, C.; Gentili, P. J . Chem. Soc., Chem. Commun. 1993,
570. (b) Galli, C.; Gentili, P.; Rappoport, Z. J . Org. Chem. 1994, 59,
6786. (c) Amatore, C.; Galli, C.; Gentili, P.; Guarnieri, A.; Schottland,
E.; Rappoport, Z. J . Chem. Soc., Perkin Trans. 2 1995, 2341. (d) Galli,
C.; Gentili, P.; Guarnieri, A.; Rappoport, Z. J . Org. Chem. 1996, 61,
8878.
(18) (a) Bunnett, J . F. Acc. Chem. Res. 1978, 11, 413. (b) Rossi, R.
A.; De Rossi, R. H. Aromatic Substitution by the SRN1 Mechanism; ACS
Monograph 178; American Chemical Society: Washington, DC, 1983.
(c) Rossi, R. A.; Pierini, A. B.; Pen˜e´n˜ory, A. B. In The Chemistry of
Functional Groups. Supplement D2: The Chemistry of Halides, Pseudo-
halides and Azides; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester,
U.K., 1995; Chapter 24.
(14) (a) Tasaka, M.; Ogata, M.; Ichikawa, H. J . Am. Chem. Soc. 1981,
103, 1885. (b) Angelini, G.; Fornarini, S.; Speranza, M. J . Am. Chem.
Soc. 1982, 104, 4773. (c) Filippi, A.; Lilla, G.; Occhiucci, G.; Sparapani,
C.; Ursini, O.; Speranza, M. J . Org. Chem. 1995, 60, 1250.
(15) Aue, D. H. In Dicoordinated Carbocations; Rappoport, Z., Stang,
P. J ., Eds.; Wiley: Chichester, U.K., 1997; Chapter 3.
(19) (a) Hoz, S.; Bunnett, J . F. J . Am. Chem. Soc. 1977, 99, 4690.
(b) Fox, M. A.; Younathan, J .; Fryxell, G. E. J . Org. Chem. 1983, 48,
3109. (c) Tolbert, L. M.; Nesselroth, S. M.; Netzel, T. L.; Raya, N.;
Stapleton, M. J . Phys. Chem. 1992, 96, 4492. (d) Ahbala, M.; Hapiot,
P.; Houmam, A.; J ouini, M.; Pinson, J .; Save´ant, J .-M. J . Am. Chem.
Soc. 1995, 117, 11488.

9294 J . Org. Chem., Vol. 63, No. 25, 1998
Galli et al.
Sch em e 4
(1)
we have carried out an investigation, whose results are
herein described.
Resu lts
cally induced SRN1 nucleophilic substitution in DMSO
solution.17b On irradiation at 350 nm in the presence of
the enolate ion from pinacolone (Me3CCOCH2-), which
is an electron donor species under photostimulation,19,23
the nucleophilic substitution product 3 was obtained in
a fairly good yield (eq 4; expt 2). It was accompanied by
Com p etition a m on g Heter olysis, Hom olysis, a n d
SRN1 w ith 1,2,2-Tr ip h en ylbr om oeth en e (1). The tri-
phenyl-substituted vinyl bromide 1 gives an “SN1” sol-
volytic reaction in ethanol (eq 2); this occurs either, and
(2)
(4)
more slowly, upon heating at high temperature20 or, and
more easily, on irradiation with UV lamps.1,21 Interme-
diacy of the R-phenyl substituted vinyl cation 1+ was
proposed. Structurally similar compounds, where one or
more of the phenyl substituents of 1 are replaced by
anisyl groups (An), were also investigated (eq 3).1,3a On
minor amounts of the hydrodehalogenated product 4,
presumably derived from the vinyl radical intermediate
by H-atom abstraction from the solvent. An SRN1 process
was analogously substantiated for a photostimulated
reaction of 1 with the enolate ion from acetophenone17b
and with the conjugate base of nitromethane under
entrainment.24
Evidence that the substitution product 3 derives from
the vinyl radical intermediate within an SRN1 cycle was
that formation of the substitution product was almost
completely suppressed (eq 5; expt 3) when the photo-
(3)
(5)
irradiation, solvolysis products were consistently ob-
served, involving also those derived from rearrangements
(e.g., 1,2-aryl shift) governed by the tendency to afford
the more stabilized vinyl cation,1,4 accompanied by lower
amounts of hydrodehalogenated product(s) deriving from
the vinyl radical intermediate (Scheme 2). Whenever the
reactivity and selectivity of the photochemically produced
vinyl cation was studied, it resembled those of the same
species generated by thermal solvolysis, thereby suggest-
ing the intermediacy of a ground-state vinylic ion under
photochemical induction.3a Capture of this vinylic ion by
a nucleophile purposely added in solution was also
reported (Scheme 4).1,3a,4a,22
In keeping with a photoinduced heterolytic cleavage
of the C-Br bond of 1, we have confirmed (Table 1, expt
1) that a significant yield of the ethoxy-derivative 2 is
obtained by irradiation in EtOH at 350 nm and room
temperature in our photochemical reactor. This should
be compared with the very slow solvolysis reaction of 1
even at a much higher temperature.20a However, 1 had
also provided an unambiguous example of a photochemi-
-
stimulated reaction of 1 with Me3CCOCH2 was run in
the presence of a 20% molar amount of an electron
scavenger, such as p-dinitrobenzene. In this case, minor
amounts of two previously unobserved products were
instead identified (5 and 6); 5 is likely to be formed from
capture of the intermediate vinyl cation by traces of water
present in the reaction environment, while 6 should be
a cleavage product from 5.3a,20d
A control experiment indicated that, in the photolysis
of 1 in 95% DMSO-5% water (v/v) without added
nucleophile, 5 and 6 are indeed produced and that the
amount of 5 decreases, while that of 6 increases, for a
prolonged irradiation. We conclude that photoejected
electrons from a reducing nucleophile convert 1 into the
vinyl radical intermediate 1• through the 1•- fleeting
intermediacy (see eq 4 and Scheme 3), and the SRN
1
substitution product is obtained. In contrast, when the
purposely added electron scavenger prevents ET events,
the SRN1 cycle is suppressed so that the direct photohet-
erolysis of the C-Br takes over (as in eq 5). Under these
circumstances, only the formation of the “solvolytic
product” 5 (and 6) is observed, with no traces of the
nucleophilic substitution product that could conceivably
result from coupling of the vinyl cation with the enolate
ion (eq 1). The latter process appears therefore unlikely
(20) (a) Rappoport, Z.; Gal, A. J . Org. Chem. 1972, 37, 1174. (b)
Rappoport, Z. Recl. Trav. Chim. Pays-Bas 1985, 104, 309. (c) Kitamura,
T.; Taniguchi, H.; Tsuno, Y. In Dicoordinated Carbocations: Rappoport,
Z.; Stang, P. J ., Eds.; Wiley: Chichester, U.K., 1997; Chapter 7. (d)
Rappoport, Z.; Gal, A. J . Am. Chem. Soc. 1969, 91, 5246.
(21) (a) Verbeek, J . M.; Cornelisse, J .; Lodder, G. Tetrahedron 1986,
42, 5679. (b) Kitamura, T.; Kobayashi, S.; Taniguchi, H.; Fiakpui, C.
Y.; Lee, C. C.; Rappoport, Z. J . Org. Chem. 1984, 49, 3167.
(22) Kitamura, T.; Kobayashi, S.; Taniguchi, H. Chem. Lett. 1984,
1523.
(23) (a) Galli, C.; Gentili, P. Acta Chem. Scand. 1998, 52, 67. (b)
Galli, C.; Gentili, P.; Guarnieri, A. Gazz. Chim. Ital. 1997, 127, 159.
(24) Santiago, A. N.; Lassaga, G.; Rappoport, Z.; Rossi, R. A. J . Org.
Chem. 1996, 61, 1125.

Photochemical Cleavage of Vinyl Halides
Ta ble 1. Rea ction Con d ition s a n d Yield s of th e P h otoin d u ced Rea ction sa
products, yield (%)
J . Org. Chem., Vol. 63, No. 25, 1998 9295
expt no. subst RX conditions
solvent
reagent Y
reactn time (min)
RY
RH
others
recov RX
1
1
1
1
1
1
1
1
1
1
1
1
8
1
1
1
1
hν, 350 nm EtOH
20
20
20
20
20
20
20
60
20
20
20
60
20
20
20
20
20
2, 33
3, 55
0
7, 8
0
0
4, 3
4, 4
4, 2
4, 38
4, 2
4, 0.7
4(-d ), 12 5, 0
4, 20
4, 0.2
4, 4
4, 2
42
24
80
35
75
20
38
28
41
48
45
10
-
2
hν, 350 nm DMSO
hν, 350 nm DMSO
hν, 350 nm DMSO
hν, 350 nm DMSO
hν, 254 nm DMSO
Me3CCOCH2-
3b
4
Me3CCOCH2
(EtO)2PO-
(EtO)2PO-
(EtO)2PO-
5, 4; 6, 2
5, 1
5, 2
5b
6
5, 1.5
7
8
9
10
hν, 350 nm DMSO-d6 (EtO)2PO-
7, 8
2, 15
hν, 350 nm EtOH
hν, 350 nm DMSO
hν, 350 nm DMSO
hν, 350 nm DMSO
hν, 350 nm DMSO
hν, 350 nm DMSO
hν, 350 nm DMSO
hν, 254 nm DMSO
hν, 419 nm DMSO
hν, 254 nm DMSO
hν, 419 nm DMSO
hν, 350 nm DMSO
hν, 254 nm DMSO
hν, 419 nm DMSO
hν, 350 nm C6H6
hν, 350 nm C6H6
hν, 350 nm DMSO
(EtO)2PO-
-
NO2
N3
5, 9; 6, 0.1
5, 13; 6, 0.1
5, 12
-
11
12
13d
14
15
16
17
18
19
20
21
22f
23
24
25
26
27
28
29
30
31
32
PhCH2Et3NCl
PhCH2Et3NCl
9, traces
e
c, 21
NO2- and Me3CCOCH2
4, 3
4, 1
4, 0.5
4, 0.5
4, 3
-
6, 12
6, 35
6, 0
62
42
95
17
51
45
22
70
-
1
1
Me3CCOCH2-
3, 69
3, 29
12, 33
12, 65
12,6
Me3CCOCH2
20
4, 1
11
11
11
E-13
Z-13
11
11
11
11
19
22
1
120
120
120
180
180
60
300
360
240
60
14, major <1
g
g
14, major <1
-
Me3CCOCH2
17, 40
18, 3
30
30
5
10
51
37
35
42
64
80 °C
C6H6
Bu3SnH/AIBN
Bu3SnH/AIBN
Bu3SnH/AIBN
18, 60
18, 27
18, 19
21, 11
21, 0
4, 23
4, 0.7
4, 0.5
hν, 350 nm C6H6
hν, 350 nm DMSO
hν, 254 nm EtOH
hν, 254 nm EtOH
hν, 350 nm EtOH
hν, 350 nm t-BuOH
hν, 350 nm t-BuOH
12, 21
12, 11
20, 5
20, 35; 23, 5
60
60
80
80
EtO-
2, 12
1
1
t-BuO-
a
Photolyses were run at room temperature under argon. Typical conditions: substrate, 60 µmol; nucleophile, 200 µmol; solvent, 3 mL.
In the presence of 15 µmol of p-dinitrobenzene. c 4,4′-Dimethoxybenzophenone. Entrainment experiment: 84 µmol of Me3CCOCH3
b
d
and 423 µmol of KNO2. e Ph2CdC(Ph)CH2COCMe3 (20%) and traces of Ph2CdC(Ph)NO2 (by GC-MS). f Reported in ref 27. Traces of
g
benzofuranophenanthrene 15 (see ref 27).
under our conditions, at least at the concentration of the
anion investigated.
Rea ction s of Oth er Nu cleop h iles w ith 1. A photo-
stimulated SRN1 reaction of 1 was analogously attempted
with (EtO)2PO- as the nucleophile (eq 6). The substitu-
with heterolysis contributing more at the shorter irradia-
tion wavelength. This wavelength effect on the efficiency
of the photoheterolysis has already been documented.1,4a
Reaction of 1 and (EtO)2PO- in DMSO-d6, under
otherwise identical conditions, gave a lower amount of
deuterated reduction product 4-d (compare expts 4 and
7). This evidence for a kinetic isotope effect in the
H-abstraction event points to the solvent as the source
of hydrogen (or, deuterium) in the reduction product.
Finally, an SRN1 reaction with (EtO)2PO- ion in EtOH
solution was unsuccessfully attempted (expt 8). Either
depression of nucleophilic reactivity of (EtO)2PO- by
hydrogen bonding, or preferential solvation of the vinyl
cation by the solvent, prevented the SRN1 process in favor
of both the direct heterolysis (with formation of 2) and
homolysis of the C-Br bond (with formation of 4).
No SRN1 process was obtained in photostimulated
experiments of 1 with NO2-, N3-, or Cl- as nucleophiles,
and only the “solvolytic product” 5 (and 6) was identified
(expts 9-11; see eq 5). This is in keeping with photohet-
erolysis of 1 prevailing over the ET SRN1 process for these
anions, which are indeed weaker reductants than
(EtO)2PO- and, even more so, than Me3CCOCH2- (Table
2).23,25 Only from An2CdC(Br)An (8), which produces a
vinyl cation stabilized by an R-anisyl group, photolysis
with Cl- provided a trace of the chloro derivative An2Cd
C(Cl)An (9; identified by GC-MS); 9 is presumably formed
(6)
tion product 7 was obtained in low yield, and it was
accompanied by major amounts of reduced product 4
(expt 4). In agreement with the occurrence of a SRN
1
process, the formation of both 7 and 4 was inhibited in
the presence of p-dinitrobenzene (expt 5). When compared
-
to the case of Me3CCOCH2 as the nucleophile (expt 2),
the lower efficiency of (EtO)2PO- in the SRN1 process (see
expt 4) results not only from obtaining a higher amount
of 4, via radical 1•, but also from the formation of the
“solvolytic” product 5 via ion 1+. On irradiation at “254
nm” in a quartz vessel (expt 6), the formation of 7 is
completely suppressed, while 5 and 4 are still observed,
even though in low yield. This supports a competition
between photoejection of electrons from (EtO)2PO- ion
and direct heterolysis (and homolysis) of the C-Br bond,
(25) (a) Eberson, L. In Electron-Transfer Reactions in Organic
Compounds; Springer-Verlag: Berlin, 1987. (b) Acta Chem. Scand.
1984, B38, 439.

9296 J . Org. Chem., Vol. 63, No. 25, 1998
Galli et al.
Ta ble 2. Oxid a tion P oten tia ls of Som e Releva n t An ion s
Y- (E° in V vs SCE)a
Y-
E°
Y-
E°
-
Me3CCOCH2
PhS-
-0.15b
0.1
MeO-
0.6
0.7
1.3
1.9
-
NO2
-
(EtO)2PO-
PhO-
0.34b
0.25
N3
Cl-
a
b
Data in DMSO or MeCN; see refs 23a and 25. Ep values at
500 mV/s on a Pt electrode.
by capture of the intermediate vinyl cation An2CdC(+)-
An by Cl- (Table 1, expt 12; compare with Scheme 4).
An entrainment experiment of nucleophiles with 1 was
(8)
-
-
attempted between excess NO2 and Me3CCOCH2
,
under irradiation at 350 nm. Only traces of a product
that could be 10 were detected by GC-MS (eq 7; expt 13),
cationic intermediate (11+), which is efficiently trapped
by the “internal” nucleophile, that is, the methoxy group
of the â ortho-anisyl moiety. This process had already
been documented for 11 in benzene solution.1,4,20,27
Photolysis in benzene of the two related isomeric vinyl
bromides E-13 and Z-13 consistently afforded 14 (expts
22 and 23). This supports a direct heterolysis of the C-Br
bond even in this poorly solvating medium, which is
followed by intramolecular capture of the linear vinyl
cation intermediate by the ortho-anisyl group (eqs 9 and
10). A subsequent photochemical process gave 15, as we
(7)
along with the prevailing formation of 3. This would
endorse a slight occurrence of the SRN1 propagation cycle
with NO2-, once the process is initiated by the competing
and more efficient reductant Me3CCOCH2- anion, which
makes the vinyl radical 1• available. In the absence of
-
Me3CCOCH2-, the poor reducing character of NO2
makes inefficient the initiation of the vinylic SRN1 process
(see expt 9). Examples of entrainment between nucleo-
philes are well documented both for aromatic and vinylic
SRN1 processes.18,24,26 The NO2 anion was selected for
-
this entrainment experiment since recent calculations
show that in an aromatic SRN1 process the coupling of
NO2 with Ph• is thermodynamically favored.23a
-
Effect of th e Ir r a d ia tion Wa velen gth . This effect
was investigated in the case of 1 in DMSO solution. In
the absence of the nucleophile (i.e., Me3CCOCH2-), ir-
radiation at 350 nm provides a minor amount of 6 (expt
14); production of 6 is instead substantial (35%) for
irradiation at 254 nm in a quartz vessel, while at 419
nm 6 is absent (expts 15 and 16). This confirms that the
photoheterolytic cleavage of the C-Br bond is more
efficient at a shorter wavelength, since the absorption of
1 is likely to decrease strongly in the visible range.1,3b
On addition of the enolate ion, the SRN1 process takes
place and overwhelms the heterolytic route (expt 2), even
at 254 nm (expt 17). Even for the SRN1 process the
efficiency decreases gradually on shifting to a longer
wavelength, but formation of the substitution product 3
is still appreciable at 419 nm (expt 18). No measurement
of the absorption spectra of the reagents nor of the
possible charge-transfer complex (i.e., Y- plus substrate)19a
has however been carried out so far.
observed a gradual conversion of 14 into 15 when
samples of the reaction mixture were withdrawn after
1, 3, and 5 h of photolysis. Such conversion had already
been documented.4b
Formation of 14 from Z-13 strongly argues against a
possible intramolecular assistance of the ortho-anisyl
group to the departure of the halide concerted with the
C-Br bond heterolysis. Such a neighboring group par-
ticipation could in principle be stereoelectronically pos-
sible for the E-isomer, in a single-step intramolecular SN2
fashion but not for the Z-isomer. Previous work on
structurally similar compounds had ensured that no E/ Z
isomerization of the precursors takes place under this
conditions.17d
Com p etition of P a th w a ys w ith Su bstr a te 11. Pho-
tolysis of 11 in DMSO at 350 nm gave a fairly good
conversion to the benzofuran derivative 12 (eq 8; expt
19). Conversion was higher at 254 nm and much less so
at 419 nm (expts 20 and 21). Formation of 12 supports
the heterolysis of the C-Br bond and the formation of a
A photochemical reaction of 11 in the better solvating
EtOH solvent still does not produce even traces of the
intermolecular capture product of the intermediate vi-
nylic cation by the solvent, but only 12 is obtained
(27) Sonoda, T.; Kobayashi, S.; Taniguchi, H. Bull. Chem. Soc. J pn.
1976, 49, 2560.
(26) Galli, C.; Gentili, P. J . Chem. Soc., Perkin Trans. 2 1993, 1135.

Photochemical Cleavage of Vinyl Halides
J . Org. Chem., Vol. 63, No. 25, 1998 9297
(experiment not shown). This high selectivity underlines
the enormous efficiency of the intramolecular process of
eq 8, which leads to a five-membered cyclic product. The
reported intermolecular rate constant for capture of the
vinylic carbocation by EtOH for the structurally similar
in boiling benzene (expt 25; eq 13). However, when the
(13)
compound 16 under flash-photolysis (eq 11)3b is kinter
)
AIBN/Bu3SnH reaction of eq 13 was conducted by pho-
tochemically inducing the cleavage of AIBN at 350 nm
at room temperature, both 18, derived from 11•, and 12,
derived from 11+, were obtained in comparable amounts,
both in benzene and in DMSO (expts 26 and 27). Thus,
(11)
9.7 × 107 M-1 s-1. For structurally similar compounds,
kinter values in the range of 105-106 M-1 s-1 were also
reported.4,28 We tentatively assume that a kinter value of
107 M-1 s-1 also holds for a hypothetical intermolecular
capture of the vinyl cation 11+ by EtOH. However, from
the absence of even traces of the intermolecular capture
product vs that of intramolecular capture (i.e., 12) in the
reaction of 11 in EtOH (as in eq 8), we evaluate that the
intramolecular process must be g100 times more efficient
than the intermolecular one. Thus, since the concentra-
tion of neat EtOH is 21.7 M, the kintra value in eq 8 would
be 1010 s-1. Regardless of the approximate nature of this
the absence of 12 in expt 24 does suggest that the SRN
1
route overshadows even a highly efficient photohetero-
lytic process.
Rea ction s of â-br om ostyr en e (19). The lack of an
R-aryl substituent in 19 should reduce the stabilization
of the intermediate â-phenylethenyl cation, as compared
with the corresponding intermediates derived from 1 and
11 (i.e., 1+ and 11+), thus making the photoheterolysis
less competitive. Indeed, in the photolysis of 19 in EtOH
at 350 nm the starting material was quantitatively
recovered. Only at 254 nm in a quartz vessel a minor
conversion (ca. 5%) into phenylacetylene (20) was de-
tected, the latter being the likely product of â-elimination
from the intermediate â-phenylethenyl carbocation. No
“SN1” solvolytic product was instead detected, while 11%
of styrene (21), i.e., the reduction product from the
homolytic cleavage of the C-Br bond, was observed (expt
28; eq 14).
evaluation, we point out that the effective molarity (EM
29
) kintra/kinter
)
for the formation of 12 comes to be 103 M,
which is a reasonable value among those related to the
efficiency of closure of five-membered rings, where typical
EM values cluster around 104-105 M or are even
higher.29b,c Hence, our evaluation of 1010 s-1 as the kintra
for 11 certainly represents a lower limit, and the value
could even be 1 power of 10 higher.
Since 11 is the precursor of 11+, which is endowed with
such a high intramolecular reactivity, we wondered if an
SRN1 process could efficiently compete with the photo-
(14)
-
heterolytic route for 11. Reaction of 11 with Me3CCOCH2
under irradiation at 350 nm in DMSO gave a fairly good
conversion into the SRN1 substitution product 17, which
was accompanied by minor amounts of the hydrodeha-
logenation product 18, without traces of benzofuran 12
(expt 24; eq 12).
The homolytic cleavage appears to prevail in this
system, thereby underlying the importance of having an
R-aryl substituent for a successful occurrence of the
photoheterolysis. Indeed, photolysis of R-bromostyrene
(22) at 254 nm in EtOH gave a higher conversion (ca.
35%) to 20, accompanied by 5% of acetophenone (23), the
tautomer of the capture product of the R-phenylethenyl
cation by adventitious water, but no traces of 21 (expt
29; eq 15).
(15)
(12)
This confirms the prevalence of heterolysis over ho-
molysis for an R-aryl substituted vinyl halide under
irradiation.1,3,4 A search for the possible occurrence of a
SRN1 process between an enolate ion and 19 indicated
that the presence of the neighboring C-H and C-Br
bonds allows instead the incursion of an ionic elimina-
tion-addition process.17a,b
Ar e th e Alk oxid e Ion s Red u cta n t Sp ecies? The
reported reaction of the vinyl bromide 24 in MeOH offers
further evidence for a competition between photoinduced
heterolysis and homolysis of a C-Br bond, since both
Product 18 was independently obtained as the only
product on reacting 11 with the AIBN/Bu3SnH system
(28) (a) Kobayashi, S.; Hori, Y.; Hasako, T.; Koga, K.; Yamataka,
H. J . Org. Chem. 1996, 61, 5274. (b) Kobayashi, S.; Schnabel, W. Z.
Naturforsch. 1992, 47b, 1319.
(29) (a) J encks, W. P. In Catalysis in Chemistry and Enzymology;
McGraw-Hill: New York, 1969. (b) Kirby, A. J . Adv. Phys. Org. Chem.
1980, 17, 183. (c) Mandolini, L. Adv. Phys. Org. Chem. 1986, 22, 1. (d)
Illuminati, G.; Mandolini, L. Acc. Chem. Res. 1981, 14, 95. (e) Winnik,
M. A. Acc. Chem. Res. 1977, 10, 173.

9298 J . Org. Chem., Vol. 63, No. 25, 1998
Galli et al.
solvolytic (25) and hydrodehalogenated (26) products
were obtained at 350 nm (eq 16).21a
(19)
(16)
easily produced from an alkoxide ion under radical
conditions,31 and this technique was even recently ex-
ploited by us for obtaining an ET-induced hydrodehalo-
genation from a vinyl halide.17d
In contrast to present knowledge,25 a recent report
indicates the tert-butoxide ion as a rather strong reduc-
tant in solution,32 on the basis of an extrapolation of its
oxidation potential (i.e., E° -0.23 V vs SCE in DMF) from
gas-phase data. The suggestion that t-BuO- can behave
as an electron donor32 would support the explanation21a
delineated in eq 17 for a photostimulated process. To
clarify this point, we have first irradiated 1 in neat
t-BuOH and then in t-BuO-/t-BuOH. Only the reduced
product 4 was formed, albeit in low yield (expts 31 and
32), with no traces of the solvolytic substitution product;
this may be due to the steric bulk of t-BuOH compared
with EtOH. Although the yields of expts 32 vs 31 are
disappointly low, it is however possible to state that there
is no evidence for an enhancement of the amount of 4
upon the addition of t-BuO-, in contrast to the increase
of 4 in the corresponding expt 30 vs 1. According to our
explanation, this negative outcome is due to the lack of
R-CH bond in t-BuO-, which makes impossible the
formation of a ketyl species31a (as in eq 18). Hence, the
latter species would appear to be the likely electron donor
(see eq 19), and the hypothesis of a photostimulated ET
step from an alkoxide ion, as being responsible for the
induction of the hydrodehalogenation reaction (eq 17),
seems to lack support, at least for substrates having an
electron affinity comparable with that of 1 (Ep -1.86 V
vs SCE in THF).17c
Surprisingly, on addition of MeO- ion, the formation
of the reduction product 26 was enhanced. This was
attributed to MeO- behaving as an electron donor
toward 24 (eq 17).21a The radical anion 24•- produced
(17)
from this ET step will generate the corresponding vinyl
radical 27 by loss of bromide ion. Hence, incursion of the
photostimulated ejection of electrons from MeO- would
spur the production of 27, from which the reduction
product 26 originates, as compared to the formation of
27 by the direct photohomolysis of the C-Br bond (see
eq 16) whenever the alkoxide is not present, with that
accounting for the experimental finding. In keeping with
the evidence delineated in eq 16,21a we have found that
in the photolysis of 1 in EtOH the formation of reduction
product 4 is indeed enhanced when the reaction is
conducted in the presence of EtO- (compare expt 30 with
1). However, we believe that the explanation given
above21a is somewhat unlikely, since the alkoxide ions are
weak electron donors (Table 2).23,25 For example, we did
not observe any reactivity for PhO- or t-BuO- ions in an
aromatic SRN1 reaction, not even under photostimula-
tion,23 even though electron transfer from a nucleophile
is known to become easier under suitable irradiation.19c,d
Therefore, it seems more likely to suggest that the vinyl
radicals, which are produced from 1 or from 24 by the
photohomolysis of the C-Br bond (see Scheme 2), ab-
stract hydrogen atom not only from C2H5OH but also
from C2H5O-, whenever it is present (eq 18). From the
Exp er im en ta l Section
In str u m en ta tion . Photochemical reactions were usually
conducted in a Rayonet RPR-100 reactor equipped with a set
of 16 “350 nm” lamps and occasionally with sets of 16 “419” or
“254 nm” lamps. Reactions at 254 nm were conducted in a
quartz cuvette, provided with a magnetic ministirrer; other-
wise, a small Pyrex flask was used. Characterization of the
structure of the reaction products was by conventional NMR
techniques (data in CDCl3), while HRMS determinations were
carried out on a Bruker Apex TM47e FTMS.
Ma ter ia ls. Reagent grade (Aldrich) R- and â-bromostyrene
(22 and 19) were distilled prior to use. 1,2,2-Triphenylbromo-
ethene (1), 1,2,2-triphenylethoxyethene (2), 1,1,2-triphen-
ylethene (4), R,R-diphenylacetophenone (5), 1,2,2-trianisylbro-
moethene (8), 1-phenyl-2,2-bis(o-methoxyphenyl)-1-bromo-
ethene (11), 2-phenyl-3-(o-methoxyphenyl)benzofuran (12), E-
and Z-1-phenyl-2-(o-methoxy-phenyl)-2-(p-methoxyphenyl)bro-
moethene (E-13 and Z-13), 2-phenyl-3-(p-methoxyphenyl)-
benzofuran (14), 2-methoxy-9,10-benzofuranophenanthrene
(15), and 1-phenyl-2,2-bis(o-methoxyphenyl)ethene (18) were
available from previous investigations.4b,17,20,27 The solvents
were reagent grade commercial samples; benzene was dried
over sodium wires, while DMSO (ERBA RPE) was distilled
from CaH2 under vacuum and stored over molecular sieves.
(18)
latter a ketyl species would be generated, i.e., the radical
anion of acetaldehyde CH3CHO•-, a powerful reductant,30
which could efficiently transfer an electron to 1 to produce
1•- (eq 19), and subsequently Ph2CdC(Ph)•.
In this way the formation of reduced product 4 could
indeed be fostered by the presence of the alkoxide ion.
There is circumstantial evidence that a ketyl species is
(31) (a) Bunnett, J . F.; Wamser, C. C. J . Am. Chem. Soc. 1967, 89,
6712. (b) Boyle, W. J .; Bunnett, J . F. J . Am. Chem. Soc. 1974, 96, 1418.
(c) Amatore, C.; Badoz-Lambling, J .; Bonnel-Huyghes, C.; Pinson, J .;
Save´ant, J .-M.; Thie´bault, A. J . Am. Chem. Soc. 1982, 104, 1979. (d)
Bunnett, J . F. Acc. Chem. Res. 1992, 25, 2.
(32) Workentin, M. S.; Maran, F.; Wayner, D. D. M. J . Am. Chem.
Soc. 1995, 117, 2120.
(30) Wardman, P. J . Phys. Chem. Ref. Data 1989, 18, 1637.

Photochemical Cleavage of Vinyl Halides
J . Org. Chem., Vol. 63, No. 25, 1998 9299
Ta ble 3. Isod esm ic Rea ction s
Freshly sublimed t-BuOK was employed in the generation of
enolate anions.
conventional workup with CHCl3, a crude oil was obtained.
Column chromatography (silica gel) with 6:1 toluene/diethyl
ether as the eluent gave 6 mg of 7 (9% yield). 1H NMR (δ,
CDCl3): 7.2-6.7 (m, 15H), 3.8 (q, 4H), 1.2 (t, 6H). HRMS: m/e
392.4366; C24H25O3P requires 392.4384.
P h otoch em ica l Exp er im en ts. With Solven t On ly. In a
typical experiment, 19 mg of 1 (57 µmol) and the chosen solvent
(3 mL), previously purged with argon, were placed in a suitable
flask and irradiated at the selected wavelength under argon.
Depending on the solvent, either direct injection (for reactions
in benzene or EtOH) into GC or GC-MS columns was carried
out, after the addition of an internal standard, or a conven-
tional workup (for DMSO) preceded the injection. The GC yield
of the products was determined by calibrating the response
factors of pure samples of the compounds. In some reactions
of 1, where 5 is obtained (expts 3-7, 9-11), subsequent
conversion of 5 to 6 should occur.33 This would require
concurrent formation of benzoic acid.20d In one attempt we were
able to detect it by MS analysis of the water extracts.
In th e P r esen ce of a Nu cleop h ile. The enolate ion
2,2-Dim eth yl-5-ph en yl-6,6-bis(2-m eth oxyph en yl)-5-h ex-
en -3-on e (17). Reaction of 0.49 mL of pinacolone (3.9 mmol),
0.56 g of t-BuOK (5 mmol), and 0.4 g of 11 (1 mmol) in 24 mL
of DMSO was conducted under photostimulation at 419 nm
for 3 h. The crude material obtained as above (0.58 g) was
chromatographed on silica gel with toluene as eluent. A pure
1
fraction of 17 was collected (130 mg, 43% yield). H NMR (δ,
CDCl3): 7.3-6.6 (m, 13H), 3.65 (s, 6H), 3.50 (s, 2H), 0.92 (s,
9H). 13C NMR (δ, CDCl3): 213, 157, 143, 135, 132, 131, 128-
125, 120, 119, 111, 110, 56, 55, 44, 43, 26. HRMS: 414.5428;
C
28H30O3 requires 414.5420.
Ca lcu la tion s. An evaluation of the stabilizing effect from
R-substituents in ethenyl cation structures (R+) is presented,
to be compared with the literature values quoted in Chart 2,15
which are derived from H- affinity data in the reaction: H-
+ R+ f RH. We have instead evaluated the R-substituent
effect from the isodesmic reactions (∆H° reaction in kcal/mol)
reported in Table 3. The numbers under the structures are
∆fH298 values in kcal/mol. For the uncharged species, the
parent vinyl cation, and the 2-propenyl cation,7 these values
are taken from the literature.11 For the other structures, we
have calculated the ∆fH298’s by the Hyperchem program34 (AM1
level), imposing the desired conformation (single point). We
have checked the reliability of these calculations by calculating
also the ∆fH298’s of vinyl (261 kcal/mol) and 2-propenyl (234
kcal/mol) ions: the agreement of these calculated values with
those of the literature (261.77 and 235.0,15 respectively), which
are reported in Table 3, is remarkable and gives us confidence
as to the reliability of our other calculations. From inspection
of Table 3, it clearly emerges that R-π-substituents do stabilize
the vinyl cation, particularly when they are coplanar with the
empty p orbital of the ion and, therefore, able to conjugate with
it.
-
Me3CCOCH2 was generated in situ by addition of 28 mg of
t-BuOK (0.25 mmol) to pinacolone (3,3-dimethyl-2-butanone,
25 µL; 0.2 mmol) in 3 mL of DMSO under argon; 20 mg of
substrate 1 (60 µmol) was then added and irradiation followed.
Similarly, (EtO)2PO- was generated from (EtO)2PHO (Fluka)
with t-BuOK. In the attempt to run a reaction of (EtO)2PO-
in EtOH (expt 8), a stoichiometric amount of the base EtONa
was produced in situ by the addition of a small piece of sodium
of appropriate weight to the solvent, and then (EtO)2PHO was
added. Sublimed t-BuOK was added to t-BuOH in expt 32.
Anhydrous commercial salts (KNO2, NaN3, PhCH2Et3NCl)
were used as source of the nucleophiles in expts 9-13. In the
entrainment expt 13, 423 µmol of KNO2 and 84 µmol of
-
Me3CCOCH2 in a photostimulated reaction with 83 µmol of
1 in 4 mL DMSO were employed. Traces of product 10 (m/e
301) were detected, alongside the prevailing formation of 3.
Hyd r od eh a logen a tion w ith Bu 3Sn H/AIBN. Previously
reported conditions of homolytic cleavage of AIBN in boiling
benzene were followed,17d in order to generate the tributyltin
radical in situ. Alternatively, photochemical cleavage of AIBN
at 350 nm in either benzene or DMSO was used.
Syn th esis of P r od u cts. Dieth yl (1,2,2-tr ip h en yleth en -
yl)p h osp h on a te (7). Reaction of 90 mg of (EtO)2PHO (0.65
mmol), 80 mg of t-BuOK (0.71 mmol), and 61 mg of 1 (0.18
mmol) in 8 mL of DMSO was conducted under photostimula-
tion for 20 min; brine was then added and, following a
Ack n ow led gm en t. We thank the Italian MURST,
the USA-Israel Binational Science Foundation (BSF),
and the Ministry of Education Science and Culture of
J apan for financial support of this work.
J O981164P
(33) (a) Ruberu, S. R.; Fox, M. A. J . Am. Chem. Soc. 1992, 114, 6310.
(b) Doering, W. v. E.; Haines, R. M. J . Am. Chem. Soc. 1954, 76, 482.
(c) Gersmann, H. R.; Bickel, A. F. J . Chem. Soc. B 1971, 2230.
(34) HyperChem is a trademark of Autodesk, Inc., 2320 Marinship
Way, Sausalito, CA 94965.