PhSeSiR3-catalyzed group transfer radical reactions

Article abstract of DOI:10.1021/jo000128z

A novel design for initiating radical-based chemistry in a catalytic fashion is described. The design of the concept is based on the phenylselenyl group transfer reaction from alkyl phenyl selenides by utilizing PhSeSiR₃ (1) as a catalytic reagent. The reaction is initiated by the homolytic cleavage of -C-Se- bond of an alkyl phenyl selenide by the in situ generated alkylsilyl radical (R₃Si), obtained by the mesolysis of PhSeSiR₃](·)^- (1(·)^-). The oxidative dimerization of counteranion PhSe^- to PhSeSePh functions as radical terminator. The generation of 1(·)- is achieved by the photoinduced electron transfer (PET) promoted reductive activation of 1 through a photosystem comprising of a visible-light (410 nm)-absorbing electron rich DMA as an electron donor and ascorbic acid as a co-oxidant (Figure 1). The optimum mole ratio between the catalyst 1 and alkyl phenyl selenides for successful reaction is established to be 1:10. The generality of the concept is demonstrated by carrying out variety of radical reactions such as cyclization (10, 15-18), intermolecular addition (25), and tandem annulations (32).

Full text of DOI:10.1021/jo000128z

J . Org. Chem. 2000, 65, 4309-4314
4309
P h SeSiR
3
-Ca ta lyzed Gr ou p Tr a n sfer Ra d ica l Rea ction s
Ganesh Pandey,* K. S. S. P. Rao, and K. V. Nageshwar Rao
Division of Organic Chemistry (Synthesis), National Chemical Laboratory, PUNE-411 008, India
pandey@ems.ncl.res.in
Received J anuary 31, 2000
A novel design for initiating radical-based chemistry in a catalytic fashion is described. The design
of the concept is based on the phenylselenyl group transfer reaction from alkyl phenyl selenides by
utilizing PhSeSiR
3
(1) as a catalytic reagent. The reaction is initiated by the homolytic cleavage of
•
-
C-Se- bond of an alkyl phenyl selenide by the in situ generated alkylsilyl radical (R Si ), obtained
3
-
•-
•-
3
by the mesolysis of PhSeSiR ] (1 ). The oxidative dimerization of counteranion PhSe to PhSeSePh
•
-
functions as radical terminator. The generation of 1 is achieved by the photoinduced electron
transfer (PET) promoted reductive activation of 1 through a photosystem comprising of a visible-
light (410 nm)-absorbing electron rich DMA as an electron donor and ascorbic acid as a co-oxidant
(Figure 1). The optimum mole ratio between the catalyst 1 and alkyl phenyl selenides for successful
reaction is established to be 1:10. The generality of the concept is demonstrated by carrying out
variety of radical reactions such as cyclization (10, 15-18), intermolecular addition (25), and tandem
annulations (32).
In tr od u ction
of the reagents. Since there is growing demand to reduce
the amount of toxic wastes and byproducts arising out
of chemical reactions, increasing emphasis is laid on
The use of tin-based reagents in free radical chemistry
has dominated the scene ever since the original discovery
of the radical generation by organotin hydrides (X
this dominance has been rightly referred to, recently, as
Tyranny of Tin”, despite their drawbacks such as cost,
instability, toxicity, loss of valuable functionality due to
10
the invention and development of a catalytic and envi-
ronmentally compatible strategy for initiating radical-
based chemistry owing to its ever increasing popularity
among synthetic chemists. Significant progress has also
been made, recently, toward developing newer method-
ologies for initiating radical-based reactions in catalytic
manner, utilizing either tin hydride reagents catalyti-
SnH);1
3
2
“
3
the termination of radical sequence by irreversible H-
4
abstraction, and the difficulty encountered in removing
5
the tin byproducts from the residue whose disposal poses
11
12
cally or its in situ generation or use of titanocene-
hazardous problem. Although partial solution to some of
these problems have been addressed by introducing
13
catalyzed reductive ring opening of an epoxide; however,
considering the importance of radical reactions in organic
synthesis, introduction of another catalytic strategy could
be a welcome addition in the repertoire of organic
chemists.
many new reagents,2 atom/group transfer reactions,
,6
7,8
9
and several other approaches, the initiation step in all
these approaches invariably requires stoichiometric use
We have reported14 an efficient strategy for the gen-
*
To whom correspondence should be addressed. FAX: (91) 20-
893153.
1) (a) Van der Kerk, G. J . M.; Noltes, J . G.; Luijten, J . G. A. J .
Appl. Chem. 1957, 7, 356. (b) Kuivila, H. G. Acc. Chem. Res. 1968, 1,
-
eration of phenyl selenide anion (PhSe ) and alkylsilyl
5
•
15
•-
radical (R
by the visible-light (410 nm)-initiated photoinduced
electron transfer (PET) activation of PhSeSiR
3 3
Si ) by the mesolysis of PhSeSiR , produced
(
2
99.
(
072.
3
(R ) tert-
2) Baugley, P. A.; Walton, J . C. Angew. Chem., Int. Ed. 1998, 37,
butyldiphenyl) through the photosystem as shown in
3
(
3) (a) De Mora, S. J . Tributyltin: Case Study of an Environmental
Figure 1. The significantly higher rate constant (9.6 ×
Contaminant; Cambridge University Press: Cambridge, UK, 1996. (b)
Boyer, I. J . Toxicology 1989, 55, 253.
7
-1 -1 6b
•
1
0 M
s
)
for the reaction of R
3
Si with alkyl phen-
-
ylselenides and fast oxidative dimerization of PhSe to
(4) (a) Chatgilialogulu, C.; Ingold, K. U.; Scaiano, J . C. J . Am. Chem.
PhSeSePh, an excellent radical scavenger16 (the rate
Soc. 1981, 103, 7739. (b) J ohnston, L. J .; Wayner, D. D. M.; Abeywick-
reyma, A. N.; Beckwith, A. L. J .; Scaiano, J . C.; Ingold, K. U. J . Am.
Chem. Soc. 1985, 107, 4594.
(
5) For a succinct overview, see: Crich, D.; Sun, S. J . Org. Chem.
996, 61, 7200.
6) (a) Chatgilialogulu, C.; Ferreri, C.; Lucarini, M. J . Org. Chem.
993, 58, 249. (b) Ballestri, M.; Chatgilialogulu, C.; Clark, K. B.; Griller,
(10) (a) Amato, J . Science 1993, 259, 1538. (b) Illman, D. L. Chem.
Eng. News 1993, 71 (13), 5; 1994, 72 (36), 22.
(11) Malaeczka, R. E.; Terstiege, I. J . Org. Chem. 1999, 64, 342 and
the references therein.
1
(
1
D.; Giese, B.; Kopping, B. J . Org. Chem. 1991, 56, 678. (c) Lusztyk, J .;
Maillard, B.; Lindsay, D. A.; Ingold, K. U. J . Am. Chem. Soc. 1983,
(12) Tormo, J .; Hays, D. S.; Fu, G. C. J . Org. Chem. 1998, 63, 5296.
(13) (a) Gansauer, A.; Bluhm, H.; Pierobon, M. J . Am. Chem. Soc.
1998, 120, 12849. (b) Gansauer, A.; Pierobon, M.; Bluhm, H. Angew.
Chem., Int. Ed. 1998, 37, 101.
1
05, 3578. (d) Boivin, J .; Camara, J .; Zard, S. Z. J . Am. Chem. Soc.
992, 114, 7909.
1
(
7) (a) Curran, D. P.; Chen, M. H.; Kim, D. J . Am. Chem. Soc. 1986,
(14) (a) Pandey, G.; Rao, K. S. S. P. Angew. Chem., Int. Ed. Engl.
1995, 34, 2669. (b) Pandey, G.; Rao, K. S. S. P. J . Org. Chem. 1998,
63, 3968.
1
2
08, 2489. (b) Curran, D. P.; Chang, C. T. Tetrahedron Lett. 1987, 28,
477.
(
8) (a) Curran, D. P.; Eichenberger E.; Collis, M.; Roepel, M. G.;
(15) (a) Maslak, P.; Vallombroso, T. M.; Chapman, W. H.; Narvaez,
J . N., J r. Angew. Chem., Int. Ed. Engl. 1994, 33, 73. (b) Maslak, P.
Top. Curr. Chem. 1993, 168, 1.
Thoma, G. J . Am. Chem. Soc. 1994, 116, 4279. (b) For leading
references, see: reference 40a and the references therein.
(
9) For leading references, see: reference 2 and the references
(16) (a) Newcomb, M.; Marquardt, D. J .; Kumar, M. U. Tetrahedron
1990, 46, 2345. (b) Newcomb, M. J . Am. Chem. Soc. 1992, 114, 8158.
therein.
1
0.1021/jo000128z CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/17/2000

4
310 J . Org. Chem., Vol. 65, No. 14, 2000
Pandey et al.
Sch em e 1. P r ep a r a tion of 10 a n d Its P ET
Activa tion
3
F igu r e 1. PET reductive activation of PhSeSiR .
through Weller equation19 employing the values of E1/2
ox
DMA) as 0.98 eV,20
E
red
of 10 as -1.4 eV, estimated
by cyclic voltammetry utilizing an identical experimental
setup as described elsewhere and E0,0 of DMA as 3.21
eV.21 Since the selectivity of the radical ion generation
from a mixture of potential electron donors/acceptors
(
1/2
1
4
depend on the magnitude of ∆Get values associated with
the electron-transfer processes,2
2,23
the large difference
between the ∆Get values for the formation of 1•- vs 10
•-
indicated that there will be selectivity in the formation
•
-
of 1 if a mixture of 1 and 10 were activated. This study
1
4
•-
coupled with the known mesolytic characteristics of 1
•
-
(producing R
3
Si and PhSe ) and significantly higher
quantum yield of disappearance of 1 (φdis ) 0.223) in
comparison to φdis for -C-Se- bonds (≈0.054) con-
vinced us that PET activation of a mixture of 1 and 10
through the photosystem as shown in Figure 1 would
2
4
2
5
F igu r e 2. A catalytic strategy for the group transfer radical
reactions.
initiate a radical reaction from 10 by chalcogen transfer
•
to R
3
Si while regenerating 1 in the process as shown in
constant for the S
PhSeSePh is calculated to be 1.2 × 10 M
H
2 attack of 5-hexenyl radical upon
Figure 2. The reorganization of the radical followed by
its termination by PhSeSePh, produced by the oxidative
7
-1 -1 17
s
), led us
to envision that a catalytic strategy for phenylselenyl
group transfer radical-based reactions might be possible
to effect from a precursor of type 2 through a cycle as
shown in Figure 2. Additional advantage of this strategy
was perceived by the stability of the precursors as well
as products and the various avenues available for the
transformation of the products. The success of our effort
is delineated in this article.
-
dimerization of PhSe , would set off a catalytic cycle for
the formation of 11 through the steps as shown in Figure
2
. With this premise, we began our study first by
establishing the catalytic role of 1 in the above reacti-
on.Toward this endeavor, PET activation of 10 at differ-
ent concentrations (0.18, 0.27, and 0.35 mmol) with a
fixed concentration of 1 (0.018 mmol), DMA (0.11 mmol),
and H
2
A (0.32 mmol) was studied by irradiating 50 mL
solution of each in Pyrex test tubes at 410 nm (450-W
Resu lts a n d Discu ssion
26
Hanovia lamp, NH
3
4
-CuSO solution). Progress of the
reaction was monitored by HPLC analysis. After 45 min
of irradiation (consumption of 10, ≈55-60%), the solu-
tions in the test tubes were analyzed by HPLC which
showed negligible change in the concentration of 1 as well
as DMA in the tube having 10 and 1 in 10:1 mole ratio.
This study was not performed at higher consumption of
Design , Op tim iza tion a n d Gen er a liza tion of th e
Con cep t. To introduce a concept of catalytic group
transfer radical reaction in organic synthesis, a design
as depicted in Figure 2 was visualized. Initially, we
decided to investigate the phenylselenyl group transfer
radical cyclization reaction of 10, prepared in 80% yield
by the reaction of PhSeBr (7) with ethyl vinyl ether (6)
in the presence of allyl alcohol (9).18 To ensure that the
PET activation of a mixture containing 1 and 10, utilizing
the photosystem as shown in Figure 1, selectively gener-
(
19) Rhem, D.; Weller, A. Isr. J . Chem. 1970, 8, 259.
(20) Zweig, A.; Maurer, A. H.; Roberts, B. G. J . Org. Chem. 1967,
32, 1322.
(
(
(
21) Lee, G. A.; Israel, S. H. J . Org. Chem. 1983, 48, 4557.
22) Mattes, S. L.; Farid, S. Science 1984, 226, 917.
23) Pandey G.; Sudharani, K.; Lakshmaiah, G. Tetrahedron Lett.
•
-
•-
ates 1 , ∆Get values for the formation of 1 (-181 kJ
-
1
14
as well as 10•- (-80 kJ M ) were compared. The
-1
M
)
1992, 35, 5107.
•
-
(24) Rao, K. S. S. P., Ph.D. Thesis, submitted to University of Pune,
∆Get value for the formation of 10 was estimated
1
998.
25) Pandey, G.; Rao, K. S. S. P.; Rao, K. V. N. J . Org. Chem. 1996,
61, 6799.
(26) Murov, S. L. Handbook of Photochemistry; Marcel Dekker: New
York, 1973; p 124.
(
(
17) (a) Russell, G. A.; Tashtoush, H. I. J . Am. Chem. Soc. 1983,
05, 1398.
18) Petrzilka, M. Helv. Chim. Acta 1978, 61, 2286.
1
(

PhSeSiR
3
-Catalyzed Group Transfer Radical Reactions
J . Org. Chem., Vol. 65, No. 14, 2000 4311
1
0 to avoid competitive reaction by the product 11. In
Sch em e 2. Selen osilyla tion Rea ction of
Cycloh exen on e
other test tubes, the consumption of 10 was not found
linearly related with the concentration of 1, probably due
to the interference of 10 with the light absorbance of
DMA. Therefore, based on the above study, it may be
suggested that the present catalytic group transfer
radical reaction strategy can be best performed at 10:1
mole ratio of the substrate (10) to catalyst (1).
Preparative PET reaction by irradiating (410 nm,
4 3
50-W Hanovia lamp, CuSO :NH
filter)26 a mixture
4
containing 10 (0.76 mmol), DMA (0.30 mmol), ascorbic
acid (0.88 mmol), and 1 (0.08 mmol), followed by usual
workup and column chromatographic purification of the
residue, gave 11 in 75% yield (Scheme 1). Although,
quantitative estimation of recovered DMA and 1 was not
made after column chromatography, negligible change
in their concentrations, after the photolysis was com-
pleted, was established by comparing the HPLC analysis
of the photolyzate before and after the irradiation. The
accumulated concentration of PhSeSePh in the reaction
mixture at a given time, also determined by HPLC
analysis of the aliquots withdrawn at different intervals
of time during photolysis, was found to be very small,
indicating that the combined rates of generation and
cyclization27 of 3-oxa-5-hexenyl radical of type 3 from 10
is possibly slower than the rate of oxidative dimerization
molecular oxygen is diffusion-controlled (k ) 4.9 ( 0.6
9
-1 -1
31
×
10 M
s
for cyclopentyl radical), and the rate-
determining step is the termination of the resultant
•
32
peroxy radical (ROO ) by hydrogen abstraction and not
the reaction of alkyl radical with oxygen. Therefore, the
reaction of 4 with oxygen, if it occurs at all, could be
33
reversible, as there is no possibility of H-abstraction by
the corresponding peroxy radical. Moreover, we were
unable to detect any product related to the peroxy radical
derived from 10. Since, no kinetic data, to the best of our
•
knowledge, is available on the reaction of R
3
Si with O
2
in the solution phase (the rate constant for the reaction
-
28
of PhSe to PhSeSePh.
•
34
10
-1 -1
of Me
3
Si with O
2
is reported to be ≈1.0 × 10
M
s
To provide some conclusive support to the mechanism
for the catalytic cycle as proposed in Figure 2, the
possibilities of other competing reaction pathways avail-
able to the radical intermediates must be eliminated. For
example, the alkyl radical 4 could possibly terminate by
chalcogen transfer either from PhSeSePh or from 2.
However, based on the comparative rate constant values
in gas phase), no definite statement could be made about
the rate constants for the reaction of tert-butyldiphenyl-
silyl radical with oxygen in comparison to the oxidative
-
28
dimerization rate constant of PhSe to PhSeSePh.
35
However, in a separate study, where we were attempt-
ing to study the selenosilylation of an enone 12 through
the route as shown in Scheme 2, tert-butyldiphenylsilanol
7
of chalcogen transfer from PhSeSePh (k ) 2.6 × 10
3
(R SiOH, 14) was obtained as the major product. The
absence of 14 during the transformation of 10 f 11
-1
-1
5
-1 -1
M
s
) and alkyl selenide (k ) 1.0 × 10 M
s ) to an
2
9
octyl radical, reported by Curran et al. and their
conclusion that diselenide would usually be the preferred
reagent for the termination of an alkyl radical in such
cases, the involvement of the later possibility can easily
be ignored. Furthermore, if 4 is presumed to be terminat-
ing by chalcogen transfer from 2, a radical chain reaction
would have set in for the transformation of 10 f 11.
However, our observation that the conversion of 10 f
suggests that the rate of chalcogenide transfer from alkyl-
•
SePh to R
3
Si could be much higher than the reaction rate
•
3
of R Si with molecular oxygen.
To compare the efficiency of this methodology over the
25
one earlier reported by us by the direct PET activation
of -C-Se- bond compounds, a control experiment was
performed in a similar manner as described above but
without having 1 in the reaction mixture. The compara-
tive results indicated that, within the constant period of
irradiation, the efficiency for the formation of 11 in the
tube containing 1 was at least 4-5 times higher than
the one without 1. This observation was expected due to
1
1 is very much dependent on the irradiation time
supports the termination of 4 by chalcogen transfer by
•
PhSeSePh and not by alkyl-SePh. The PhSe , generated
after the termination of cyclized alkyl radical derived
from 10 by PhSeSePh, dimerizes efficiently (k
r
) 7.0 ×
the higher quantum efficiency of -Se-Si- bond dissocia-
9
-1 -1 30
1
0 M
s
)
to PhSeSePh as the slower rate of reaction
•-
tion from its corresponding radical anion (1 , φdis
0
)
•
30
of PhSe with an olefinic bond and its known revers-
ibility rules out any other competing decay mode.
24
25
.223) in comparison to φdis of -C-Se- bonds (≈0.054)
and high rate constant of -C-Se- bond dissociation by
The possible decay of alkyl radical 4 as well as tert-
•
7
-1 -1 6b
R
3
Si (9.6 × 10 M
s ).
•
butyldiphenylsilyl radical (R
3
Si ) by the reaction of mo-
To establish the generality of the catalytic phenylse-
lecular oxygen could also be ruled out on the following
considerations. The reaction of an alkyl radical with
lenyl group transfer radical cyclization reactions, a
number of substrates (15-18) were studied, and the
results are given in Table 1. The starting substrates (15-
(
27) The rate constant for the cyclization of 3-oxa-5-hexenyl radical
6
-1 -1
is reported to be 9.0 × 10
M
s ; Newcomb, M.; Filipkowski, M. A.;
J ohnson, C. C. Tetrahedron Lett. 1995, 36, 3643.
(31) Rabani, J .; Pick, M.; Simic, M. J . Phys. Chem. 1974, 78, 1040.
(32) Scaiano, J . C.; Ingold, k. U.; Maillard, B. J . Am. Chem. Soc.
1983, 105, 5095.
(33) (a) Chan, H. W. S.; Levett, G.; Matthew, J . A. J . Chem. Soc.,
Chem. Commun. 1978, 756. (b) Porter N. A.; Weber, B. A.; Weenen,
H.; Khan, J . A. J . Am. Chem. Soc. 1980, 102, 5597. (c) Benson, S W.
J . Am. Chem. Soc. 1965, 87, 972.
(
28) Although, the rate constant for the oxidative dimerisation of
-
PhSe to PhSeSePh is not known, it is understood to be much higher
than the oxidative dimerization of PhSeH. Encyclopaedia of Reagents
in Organic Synthesis; Paquette, L. A., Ed.; J ohn Wiley & Sons: New
York, 1995; Vol. 1, p 270.
(
29) Curran, D. P.; Martin-Esker, A. A.; Bo Ko, S.; Newcomb, M. J .
Org. Chem. 1993, 58, 4691.
30) Ito, O. J . Am. Chem. Soc. 1983, 105, 850.
(34) Niiranen, J . T.; Gutman, D. J . Phys. Chem. 1993, 97, 4106.
(35) Pandey, G.; Rao, K. V. N. Unpublished result.
(

4
312 J . Org. Chem., Vol. 65, No. 14, 2000
Pandey et al.
Ta ble 1. P h en ylselen yl Gr ou p Tr a n sfer Ra d ica l
Rea ction
F igu r e 3. Catalytic intermolecular phenylselenyl group ad-
dition.
Sch em e 3. Iod in e Atom Tr a n sfer Str a tegy for th e
Con str u ction of F ive-Mem ber ed Rin gs.
i
See Supporting Information for the preparation and charac-
ii
1
13
terizations of 17 and 18. Characterized by H NMR, C NMR,
and mass spectral analyses. Isolated yield, unoptimized.
iii
1
8) were obtained by following the reported procedures
3
6
with little or no modification.
In ter m olecu la r a d d ition s a n d Ta n d em An n u la -
tion Rea ction s. The spectacular success of the catalytic
phenylselenyl group transfer radical cyclization reaction
of 10 through the cycle, as shown in Figure 2, encouraged
us to evaluate further the scope of this strategy for other
contemporary radical reactions such as intermolecular
additions and tandem annulations. Intermolecular radi-
cal additions, though, are powerful tools in preparative
organic chemistry, have difficulty in their execution by
classical radical-based methodology due to competing
bimolecular side reactions.37 Moreover, with stannous-
based reactions the difficulty lies, apart from being
ecologically noncompatible, in preventing premature
H-atom transfer to the radical before it adds to the
olefin.38 The other alternative approaches known in this
would be diminished due to fast termination of the adduct
7
-1
radical by phenylselenyl group transfer (k ) 3 × 10 M
-1
41
s
for secondary alkyl radicals). Toward this endeavor,
2
we studied the addition of PhSeCH COOEt (24) onto the
4-tert-butyldimethylsilyloxy-1-decene (23). Considering
the favorable electron pairing (electron-poor radical/
electron-rich acceptor) PET reaction of a mixture con-
taining 1 (0.18 mmol), 24 (1.85 mmol), 23 (1.85 mmol),
DMA (0.63 mmol), and ascorbic acid (1.57 mmol) followed
context also lack flexibility in the choice of radical
by workup and purification gave 25 as a yellow oil in 61%
precursors.3
9,40
Therefore, it occurred to us that applica-
1
yield (Figure 3). Product 25 was characterized by
H
NMR, 13C NMR, and mass spectral data.
tion of our methodology, through the sequences as shown
in Figure 3, could provide an attractive route in catalytic
manner for the group transfer intermolecular radical
addition reaction where bimolecular radical reactions
The exo-mode cyclization property of 5-hexenyl radicals
have been widely utilized for the construction of five-
3
7
membered carbocyclic rings. Curran et al. have also
reported42 in situ generation and iodine atom transfer
cyclization of 5-hexenyl radicals via addition/cyclization
sequence (annulation) from 26 to a variety of simple
olefins (28) to construct five-membered ring systems
(Scheme 3). However, construction of six-membered
carbocyclic ring systems, also widely distributed in
numerous biologically active molecules, by endo-mode
(36) (a) For the preparation of compound 15 and 16, see: Engman,
L.; Gupta, V. J . Org. Chem. 1997, 62, 157 whereas compound 17 was
prepared by the nucleophilic displacement of phenylselenyl anion on
the corresponding bromide (cf: Sharpless, K. B. J . Am. Chem. Soc.
1
973, 95, 2697). (b) For the preparation of compound 18, see: Padwa,
A.; Nimmesgern, H.; Wong, G. S. K. J . Org. Chem. 1985, 50, 5620.
37) (a) Giese, B. Radicals in Organic Synthesis: Formation of
(
Carbon-Carbon Bonds; Pergamon: Oxford, 1986. (b) Curran, D. P.
In Comprehensive Organic Synthesis; Trost, B. M., Flemming, I., Eds.;
Pergamon: Oxford, 1991; Vol. 4, p 715. (c) Curran, D. P. Synthesis
3
7
radical cyclizations has been known to be difficult.
The presence of silicon atom R as well as â to a carbon-
centered radical is known43 to reverse the regioselectivity
of radical cyclizations (favoring the endo-mode) due to
1
988, 417, 489. (d) Giese, B.; Kopping, B.; Gobel, B. T.; Dickhaut.;
Thoma, G.; Kulicke, K. J .; Trach, F. Org. React. 1996, 48, 301.
38) (a) Giese, B. Angew. Chem., Int. Ed. Engl. 1983, 22, 753. (b)
(
Giese, B. Angew. Chem., Int. Ed. Engl. 1985, 24, 553. (c) Also see:
reference 37c.
(39) Curran, D. P.; Thoma, G. J . Am. Chem. Soc. 1992, 114, 4436.
40) (a) Byers, J . H.; Lane, G. C. J . Org. Chem. 1993, 58, 3355. (b)
(41) Ogawa, A.; Ogawa, I.; Obayashi, R.; Umezu, K.; Doi, M.; Hirao,
T. J . Org. Chem. 1999, 64, 86.
(
Byers, J . H.; Harper, B. C. Tetrahedron Lett. 1992, 33, 6953.
(42) Curran, D. P.; Seong, C. M. Tetrahedron 1992, 48, 2157.

PhSeSiR
3
-Catalyzed Group Transfer Radical Reactions
J . Org. Chem., Vol. 65, No. 14, 2000 4313
Irradiations were performed in a specially designed double-
walled photoreactor. The photoreactor consisted of three
chambers. The first and outermost chamber contained irradia-
tion solution while the second one was charged with CuSO
4
‚
5
H
2
O:NH filter solution. This filter solution allowed only 410
3
2
6
nm wavelength light to pass through. A 450-W Hanovia
medium-pressure mercury vapor lamp was used as light source
that was housed in a water-circulated double-jacketed chamber
immersed into the second chamber, maintaining a 1 cm path
length of the filter solution. The whole photoreactor was made
of Pyrex glass.
Eva lu a tion of Ca ta lytic P r op er ty of 1. A 200 mL stock
solution in acetonitrile containing 1 (0.07 mmol), DMA (0.42
mmol), and ascorbic acid (0.31 mmol) was prepared. Fifty
milliliter amounts of this solution were distributed into three
test tubes made up of Pyrex glass, and 0.05 g (0.176 mmol),
0
.075 g (0.265 mmol), and 0.1 g (0.35 mmol) of 10 were
introduced into each test tube resulting in mole ratios of 10
with respect to 1 as 10:1, 15:1, and 20:1, respectively. One
milliliter each of this solution was analyzed before irradiation
by HPLC after adding 0.5 mL of solution of Ph
an internal standard, and the area ratios of 10:Ph
:Ph As were recorded. These tubes were irradiated externally
at 410 nm wavelength light coming out of a 450-W Hanovia
lamp after passing through a CuSO :NH filter solution.
3
As (0.05 M) as
3
As and
1
3
4
3
Aliquots were analyzed time to time by HPLC in the similar
manner as described above, and the area ratios were com-
pared. After 45 min, the irradiation was discontinued. The
HPLC analysis of the test tube containing 10 and 1 in 10:1
mole ratio (first tube) indicated negligible change in the
concentration of 1. Formation of 11 as the only product was
noticed by HPLC analysis. The other two tubes showed no
correlation between the conversion of 10 and the concentration
of 1.
F igu r e 4. Strategy for endo-trig cyclization in radical reac-
tions.
the involvement of seemingly less dipolar transition state
structure compared to normal carbon-centered radical
cyclizations. Though this study has been shown only with
an example where cyclization places a silicon atom in the
ring, it occurred to us that if these observations are
general, a tandem addition/cyclization reaction between
P r ep a r a tive P ET Cycliza tion of 10. A dilute solution of
CH CN (500 mL) containing a mixture of 1 (0.07 g, 0.17 mmol),
3
2
9 and allyltrimethylsilane (30) should provide six-
membered carbocyclic ring system 32 by endo-cyclization
of the intermediate 31 via expected transition state
structure 31a . The success of this concept was also
expected to provide an option to organic chemists for
manipulation of the regiochemistry of radical reactions
during carbocyclization reactions. With this background,
we studied the PET activation of a mixture containing 1
10 (0.5 g, 1.74 mmol), DMA (0.15 g, 0.63 mmol), and ascorbic
acid (0.28 g, 1.62 mmol) was irradiated in the special photo-
reactor (as described in the general Experimental Section) with
a 450-W Hanovia medium-pressure mercury lamp at room
temperature without removing dissolved oxygen from the
solution. The progress of the reaction was monitored by HPLC.
When substantial consumption of 10 was noticed, the irradia-
tion was discontinued. Solvent was removed under vacuum
and the crude photolyzate was purified by silica gel column
chromatography to give a yellow oily product 11 (0.37 g, 75%
(0.15 mmol), 29 (1.4 mmol), DMA (0.63 mmol), ascorbic
acid (1.62 mmol), and allyltrimethylsilane (30) (3.5 mmol)
which gave, to our pleasant surprise, only compound 32
in good yield (65%) (Figure 4). All the spectral charac-
terizations of 32 indicated it to be a pure diastereomer;
however, we did not try to ascertain its exact stereo-
chemistry.
In conclusion, a conceptually new and ecologically
compatible44 approach for initiating a catalytic phenyl-
selenyl group transfer radical reaction, in general, has
been designed and developed. The development of this
strategy is expected to add new dimensions to radical-
based chemistry.
yield).
1
1
1: H NMR (200 MHz, CDCl
3
) δ 1.15-1.30 (m, 3H), 1.60-
1
3
5
.80 (m, 1H), 2.15-2.60 (m, 2H), 2.90-3.17 (m, 2H), 3.30-
.50 (m, 1H), 3.57-3.80 (m, 2H), 3.95-4.10 (m, 1H), 5.10-
1
3
.17 (m, 1H), 7.20-7.30 (m, 3H), 7.45-7.55 (m, 2H); C NMR
(
50 MHz, CDCl
3
) δ 132.8, 131.5, 129.9, 128.9, 127.6, 126.9,
1
04.0, 103.8, 72.0, 63.0, 62.8, 39.9, 39.6, 38.6, 37.9, 32.3, 31.5,
+
15.4, 15.3; MS m/e (relative intensity) 286 (M , 28), 240 (15),
157 (23), 91 (42), 83 (100).
Identical irradiation procedures were adopted for the PET
activation of 15-18, 23, and 29, and the spectral characteriza-
tion of products 19-22, 25 and 32 are given as follows:
1
1
9: yield: 77%; H NMR (200 MHz, CDCl
3
) δ 1.45-1.90
(
2
4
m, 6H), 2.00-2.20 (m, 1H), 2.55-2.75 (m, 1H), 2.85-3.15 (m,
Exp er im en ta l Section
H), 3.30-3.50 (m, 1H), 3.95 (dd, J ) 7.8, 13.5 Hz, 1H), 4.45-
Gen er a l. DMA45 and PhSeSiR 46
purified by following the literature procedures.
were synthesized and
13
3
.60 (m, 1H), 7.20-7.30 (m, 3H), 7.45-7.55 (m, 2H); C NMR
(
3
50 MHz, CDCl ) δ 132.9, 132.6, 130.3, 130.1, 129.0, 126.9,
1
3
2
26.8, 86.2, 85.1, 73.3, 72.2, 50.1, 48.1, 47.1, 43.4, 34.4, 34.0,
(43) (a) Wilt, J . W. J . Am. Chem. Soc. 1981, 103, 5251. (b) Wilt, J .
3.0, 30.5, 26.4, 25.9, 25.2, 24.0; MS m/e (relative intensity)
W.; Lusztyk, J .; Peeran, M.; Ingold, K. U. J . Am. Chem. Soc. 1988,
+
82 (M , 10), 157 (15), 124 (17), 95 (78), 67(100).
1
1
10, 281. (c) Koreeda, M.; Hamann, L. G. J . Am. Chem. Soc. 1990,
12, 8175. (d) Wilt, J . W. Tetrahedron 1985, 41, 3979.
1
20: yield: 82%; H NMR (200 MHz, CDCl
3
) δ 1.35-1.85
(44) This strategy could be considered ecologically compatible as in
(m, 4H), 2.00-2.15 (m, 1H), 2.50-2.75 (m, 1H), 2.80-3.05 (m,
H), 3.65-3.85 (m, 3H), 4.05 (t, J ) 7.1 Hz, 1H), 5.25 (d, J )
this approach the only molecule which undergoes chemical destruction
is the ascorbic acid. No selenium-containing compound is produced as
byproduct. Moreover, organoselenium compounds are known to be less
toxic than their inorganic counterparts. Aromatic selenium compounds
seems to be still less toxic than aliphatic selenium compounds. For
detailed toxicological data on selenium compounds see: In Selenium;
Zingaro, R. A., Cooper, C. W., Ed.; Van Nostrand Reinhold Company:
New York, 1974; p 669.
2
(45) (a) Yeung, C. K.; J asse, B. Makromol. Chem. 1984, 185, 541.
(b) Also see references 21 and 22. (c) J ohn S. Meek, Pearle A. Monroe,
Constantine J . Bouboulis, J . Org. Chem. 1963, 28, 2572.
(46) (a) Detty, M. R. J . Org. Chem. 1981, 46, 1283. (b) For a modified
synthesis, see: reference 14b.

4
314 J . Org. Chem., Vol. 65, No. 14, 2000
Pandey et al.
+
3
.7 Hz, 1H), 7.20-7.35 (m, 3H), 7.45-7.60 (m, 2H); 13C NMR
MS m/e (relative intensity) 514 (M , 4), 457 (22), 244 (100),
(50 MHz, CDCl
3
) δ 133.1, 129.5, 129.0, 127.2, 101.8, 70.1, 61.0,
171 (30), 157 (20).
3
+
4
6
1.2, 37.4, 25.8, 22.9, 19.2; MS m/e (relative intensity) 298 (M ,
1
2: yield: 65%; H NMR (200 MHz, CDCl
3
) δ -0.05-0.05
), 197 (20), 157 (15), 141 (50), 116 (42).
(
s, 9H), 0.30-0.70 (m, 2H), 1.25 (t, J ) 7.3 Hz, 6H), 1.87-2.05
1
2
1: yield: 73%; H NMR (200 MHz, CDCl
3
) δ 1.25 (t, J )
(
1
m, 1H,), 2.10-2.50 (m, 5H), 2.65-2.83 (m, 1H), 2.95-3.05 (m,
7
.3 Hz, 6H), 1.35-1.55 (m, 1H), 1.85-2.05 (m, 2H), 2.10-2.40
H), 4.2 (q, J ) 7.3 Hz, 4H), 7.20-7.35 (m, 3H), 7.45-7.55
13
(
m, 3H), 2.45-2.60 (m, 1H), 2.95 (d, J ) 8.2 Hz, 2H), 4.20 (q,
(m, 2H); C NMR (75 MHz, CDCl
3
) δ 172.6, 132.9, 130.4, 128.9,
1
3
J ) 7.3 Hz, 4H), 7.20-7.30 (m, 3H);7.45-7.55 (m, 2H);
NMR (50 MHz, CDCl
C
1
26.7, 61.3, 58.9, 43.9, 40.6, 38.8, 38.7, 29.1, 16.2, 13.9, -0.1;
) δ 172.6, 133.0, 129.3, 127.1, 61.6, 60.4,
+
3
MS m/e (relative intensity) 470 (M , 100), 455 (34), 425 (16),
97 (16), 337 (4), 313 (57), 239 (20), 157 (10); HRMS for
SiSe calculated 470.139159, found 470.137444.
4
3
0.9, 40.2, 34.1, 33.5, 32.6, 14.3; MS m/e (relative intensity)
3
+
84 (M , 20), 227 (52), 119 (27), 153 (100).
22 34 4
C H O
1
2
2: yield: 64%; H NMR (200 MHz, CDCl
3
) δ 1.50-1.65
(
2
7
m, 1H), 1.90-2.05 (m, 1H), 2.21-2.37 (m, 1H), 2.60-2.85 (m,
Ack n ow led gm en t. Financial support by BRNS,
H), 2.95-3.05 (m, 1H), 3.15-3.55 (m, 3H), 7.20-7.30 (m, 3H),
.35-7.65 (m, 5H), 7.70-7.85 (m, 2H); 1 C NMR (75 MHz,
) δ 133.0, 132.5, 129.0, 128.9, 129.4, 127.2, 53.2, 47.3,
3
Bombay, India, is gratefully acknowledged. One of us
(K.V.N.R.) thanks CSIR, New Delhi, for awarding a
research fellowship.
CDCl
3
2
3
+
9.1, 31.6, 30.6, 29.6; MS m/e (relative intensity) 381 (M , 4),
2
23 (100), 209 (13), 141 (25); HRMS calculated for C17 19NO -
H
SSe 381.030171, found 381.032655.
Su p p or tin g In for m a tion Ava ila ble: Preparation and
characterization details of the compounds 17, 18, 24, and 29
1
2
5: yield: 61%; H NMR (200 MHz, CDCl
H), 0.85-0.97 (br s, 13H,), 1.25-1.55 (m, 12H), 1.75-2.05
m, 2H), 2.40-2.65 (m, 2H), 3.15-3.30 (m, 1H), 4.10 (q, J )
.1 Hz, 2H), 7.20-7.35 (m, 3H), 7.45-7.57 (m, 2H); 13C NMR
75 MHz, CDCl ) δ 173.0, 135.6, 127.3, 70.5, 60.2, 43.5, 42.5,
2.1, 37.8, 29.4, 25.9, 25.2, 24.7, 22.5, 18.0, 14.1, -4.1, -4.3;
3
) δ 00-0.10 (m,
9
(
8
(
1
13
and H NMR and C NMR spectra of compounds 11, 19-22,
2
5, 29, and 32. This material is available free of charge via
the Internet at http://pubs.acs.org.
3
4
J O000128Z