Stannane-Mediated Radical Addition to Arenes. Generation of Cyclohexadienyl Radicals and Increased Propagation Efficiency in the Presence of Catalytic Benzeneselenol

Full text of DOI:10.1021/jo972197s

J . Org. Chem. 1998, 63, 2765-2770
2765
Sch em e 1
Sta n n a n e-Med ia ted Ra d ica l Ad d ition to
Ar en es. Gen er a tion of Cycloh exa d ien yl
Ra d ica ls a n d In cr ea sed P r op a ga tion
Efficien cy in th e P r esen ce of Ca ta lytic
Ben zen eselen ol
Some support for this hypothesis may be drawn from the
work of Engel,6 in which it is demonstrated that benzhy-
dryl radicals reduce azo compounds. More recent work
by Rosa et al. with labeled compounds mitigates against
this mechanism, at least for AIBN.7 Bowman and co-
workers have advanced a further hypothesis, related to
the SRN1 type reaction, in which the adduct radical
undergoes deprotonation to give a radical anion which,
in turn, transfers an electron to the alkyl halide and so
achieves aromaticity.8 As written by Bowman, this
mechanism, which has found some support,9,10 uses the
stannane as a base to perform the proton abstraction
leading to the formation of molecular hydrogen gas and
a stannyl radical.
Others have taken advantage of the ability of cyclo-
hexadienyl radicals to expel stabilized heteroatomic and
carbon-centered radicals to design radical chain se-
quences leading to inter- and intramolecular ipso-
substitution reactions of arenes.11 Walton has advanced
such a fragmentation as a new means of entry into carbon
radicals.12
David Crich* and J ae-Taeg Hwang
Department of Chemistry, University of Illinois at Chicago,
845 West Taylor Street, Chicago, Illinois 60607-7061
Received December 3, 1997
In tr od u ction
In preparative free radical chemistry, the highest
yielding, cleanest reactions are typically chain sequences
in which each of the individual propagation steps is
rapid.1 Adherance to this paradigm enables radical
concentration to be maintained at a minimum which in
turn reduces the possibility of debilitating radical-
radical reactions. Inefficiency in any one of the several
propagation steps in a given chain sequence will lead to
a build up of radical concentration, the consequent
formation of dimerization and disproportionation prod-
ucts, and a shortening of the kinetic chain length. This
latter phenomenon in turn requires the use of abnormally
high amounts of chain initiator if the substrate is to be
fully consumed. A case in point, and the focus of this
study, is the stannane-mediated addition of radicals to
arenes. Here (Scheme 1), a cyclohexadienyl radical is
generated which is reluctant to propagate the chain by
hydrogen abstraction from the stannane. In this chem-
istry the ultimate fate of the cyclohexadienyl radical is
usually rearomatization to a substituted arene,2,3 but the
mechanism by which this oxidation step takes place is
not at all well understood and is the subject of debate in
the literature. Capture by the stannane to give regio-
siomeric mixtures of cyclohexadienes which are oxidized
on work up or chromatography is often assumed. How-
ever, the ease of formation and isolation of cyclohexa-
dienes from Birch type reductions4 of arenes belies this
argument, at least as the major pathway. The poor
propagation and short kinetic chain lengths of such
reactions with the consequent need for disproportionately
large amounts of initiator have led to the suggestion that
the cyclohexadienyl radicals may be oxidized by the
initiator,5 usually AIBN, or an initiator-derived radical.
We have been interested in the catalysis of stannane-
mediated radical chain reactions by selenols, a process
which in its simplest form may be formulated as the three
propagation step sequence of eqs 1-3.13
R• + PhSeH f RH + PhSe•
(1)
(2)
PhSe• + Bu3SnH f PhSeH + Bu3Sn•
Bu3Sn• + RX f Bu3SnX + R•
(3)
This sequence functions because of the 1000-fold dif-
ference in rate constants for hydrogen atom abstraction
from stannanes14 and selenols15 by alkyl radicals. Its
utility is enhanced by the ability to introduce benzene-
selenol as diphenyl diselenide which is reduced in situ
(6) Engel, P. S.; Wu, W.-X. J . Am. Chem. Soc. 1989, 111, 1830-
1835.
(7) Rosa, A. M.; Lobo, A. M.; Branco, P. S.; Prabhakar, S.; Pereira,
A. M. D. L. Tetrahedron 1997, 53, 269-284.
(8) (a) Bowman, W. R.; Heaney, H.; J ordan, B. M. Tetrahedron 1991,
47, 10119-10128. (b) Aldabbagh, F.; Bowman, W. R. Tetrahedron Lett.
1997, 38, 3793-3794. (c) Aldabbagh, F.; Bowman, W. R.; Mann, E.
Tetrahedron Lett. 1997, 38, 7937-7940.
(9) Moody, C. J .; Norton, C. L. J . Chem. Soc., Perkin Trans. 1 1997,
2639-2643.
(10) Beckwith, A. L. J .; Storey, J . M. D. J . Chem. Soc., Chem.
Commun. 1995, 977-978.
(11) (a) da Mata, M. L. E. N.; Motherwell, W. B.; Ujjainwalla, F.
Tetrahedron Lett. 1997, 38, 137-140. (b) da Mata, M. L. E. N.;
Motherwell, W. B.; Ujjainwalla, F. Tetrahedron Lett. 1997, 38, 141-
144. (c) Aboutayab, K.; Caddick, S.; J enkins, K.; J oshi, S.; Khan, S.
Tetrahedron 1996, 52, 11329-11340. (d) Giraud, L.; Lacote, E.;
Renaud, P. Helv. Chim. Acta 1997, 80, 2148-2156.
(12) (a) Baguley, P. A.; Binmore, G.; Milne, A.; Walton, J . C. J . Chem.
Soc., Chem. Commun. 1996, 2199-2200. (b) Binmore, G.; Cardellini,
L.; Walton, J . C. J . Chem. Soc., Perkin Trans. 2 1997, 757-762.
(13) (a) Crich, D.; Yao, Q. J . Org. Chem. 1995, 60, 84-88. (b) Crich,
D.; J iao, X.-Y.; Yao, Q.; Harwood: J . S. J . Org. Chem. 1996, 61, 2368-
2373. (c) Crich, D.; Hwang, J .-T.; Liu, H. Tetrahedron Lett. 1996, 37,
3105-3108.
(1) Motherwell, W. B.; Crich, D. Free Radical Chain Reactions in
Organic Synthesis; Academic Press: London, 1992.
(2) Beckwith, A. L. J .; Ingold, K. U. In Rearrangements in Ground
and Excited States; de Mayo, P., Ed.; Academic Press: New York, 1980;
Vol. 1, pp 162-310.
(3) For example see: (a) Estevez, J . C.; Villaverde, M. C.; Estevez,
R. J .; Castedo, L. Tetrahedron 1995, 51, 4075-4082. (b) J ones, K.; Ho,
T. C. T.; Wilkinson, J . Tetrahedron Lett. 1995, 36, 6743-6744. (c)
Murphy, J . A.; Sherburn, M. S. Tetrahedron 1991, 47, 4077-4088. (d)
Narashiman, N. S.; Aidhen, I. S. Tetrahedron Lett. 1988, 29, 2987-
2988. (e) Rosa, A. M.; Lobo, A. M.; Branco, P. S.; Prabhakar, S.
Tetrahedron 1997, 53, 285-298. (f) Rosa, A. M.; Lobo, A. M.; Branco,
P. S.; Prabhakar, S.; Sa´-da-Costa, M. Tetrahedron 1997, 53, 299-306.
(g) Togo, H.; Kikuchi, O. Tetrahedron Lett. 1988, 29, 4133-4134.
(4) Birch, A. J .; Subba Rao, G. In Advances in Organic Chemistry;
Taylor, E. C., Ed.; Wiley-Interscience: New York, 1972; Vol. 8, pp 1-65.
(5) (a) Curran, D. P.; Yu, H.; Liu, H. Tetrahedron 1994, 50, 7343-
7366. (b) Curran, D. P. J . Chem. Soc., Perkin Trans. 1 1994, 1377-
1393.
S0022-3263(97)02197-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/28/1998

2766 J . Org. Chem., Vol. 63, No. 8, 1998
Notes
by the stannane (eq 4), thereby eliminating the need to
handle the noxious selenol itself.13
R3SnH + PhSeSePh f R3SnSePh + PhSeH (4)
This catalytic cycle has been found to be extremely
efficient at preventing slow to moderate rearrangments
of the initial alkyl radical.13,16 More recently, we have
found that benzeneselenol is also effective in promoting
stannane-mediated chain reactions involving resonance
stabilized allylic radicals.17 Here, we report that ben-
zeneselenol may also be used to catalyze the additions
of aryl halides to arenes by facilitating the chain transfer
step involving hydrogen atom donation to the intermedi-
ate cyclohexadienyl radicals. Moreover, the process is
shown to be reductive overall and to enable the isolation
of cyclohexadienes, even after extensive silica gel chro-
matography.
recovered substrate (1), 23% of the reduction product (4),
and 12% of the phenanthridone (9). 1H NMR spectro-
scopic investigation of the crude reaction mixture indi-
cated that the spirocyclic product (7) was present in only
trace amounts, certainly <5% (Table 1, entry 1). These
results reflect reasonably well those of Bowman and co-
workers, which are characterized by the presence of large
quantities of recovered substrate, indicative of poor
propagation, and the isolation of the thermodynamic
product 9. Unlike Bowman, we isolated a significant
quantity of the reduced substrate 4. This, we believe, is
again a function of poor chain propagation and the
dropwise addition of the stannane. At the beginning of
the reaction the concentration of the stannane is ex-
tremely low and, added to the poor propagating abilities
of 5 and 6, this has the effect of effectively preventing
any reaction from occurring. Over time the concentration
of Bu3SnH builds up until it is sufficient to sustain the
reaction but also to quench the initial aryl radical 3 before
cyclization, giving 4. This type of phenomenon is a
relatively frequent, but often unrecognized, occurrence
when syringe pump additions are used. Next, the
experiment was repeated with the difference that diphen-
yl diselenide (15 mol %) was added to the initial reaction
mixture. On reduction by the stannane this gives a 0.004
M solution of PhSeH. As seen from Table 1 (entry 2) this
modification provoked a significant change in the course
of the reaction. First, the reaction ran smoothly and all
of the substrate was consumed. Second, the major
product isolated was now the spirocycle 7. Moderate
amounts of the reduced substrate 4 and the phenanthri-
done 9 were also isolated. Thus, we conclude that
benzeneselenol does catalyze the chain propagation by
quenching of the kinetically cyclized radical 5, before it
undergoes rearrangement to 6.
We next turned to 10, the ester analogue of 1. Drop-
wise addition of Bu3SnH and AIBN to a 0.02 M solution
of 10 in benzene at reflux provided the results in entry 3
of Table 1. These results, like those of entry 1, are
characterized by the return of a significant amount of
substrate (10) and formation of direct reduction product
(11). We also isolated from this reaction mixture 5% of
benzo[c]chromanone (14) and 3% of biphenyl (15). In the
presence of 20 mol % (0.004 M) diphenyl diselenide a
much smoother reaction ensued with complete conversion
of the substrate. Moreover, two new, inseparable prod-
ucts were formed with a combined yield of 40% (Table 1,
entry 4). These products were assigned structures 12
and 13 and derive from addition of the initial aryl radical
to the solvent benzene followed by trapping with the
Resu lts a n d Discu ssion
To probe our hypothesis we first selected the cyclization
of iodide 1 for study. This system had previously been
studied by Kharasch and,18 subsequently, Hey when the
aryl radical was generated by simple photolysis of the
iodide as well as by copper-catalyzed decomposition of
the corresponding diazonium salt 2.19 Later, it was
reinvestigated by Bowman and co-workers under tin
hydride-mediated conditions.8 The initial nonchain work
of the Kharasch and Hey laboratories established that
the aryl radical 3 cyclizes in the 5-exo mode, giving the
cyclohexadienyl radical 5. In the absence of efficient
radical traps this species undergoes rearrangement to the
thermodynamic product radical 6. Thus, operating in the
presence of HI as an excellent radical trap Hey and co-
workers obtained the spirocyclic cyclohexadiene 7.19 In
the presence of oxygen the cross-conjugated ketone 8 was
obtained. In the absence of efficient radical traps the
phenanthridone 9 and assorted dimers of radicals 5 and
6 were obtained.19 Working with a 10% excess of Bu3-
SnH, but of unspecified concentration, in toluene at reflux
with 30 mol % of AIBN, Bowman and co-workers ob-
tained a 15% yield of 9 from iodide 1 together with 38%
recovered 1. It was stated that the low yields were due
to difficulties in the removal of organotin byproducts and,
importantly, that no spirocyclic products were observed
in the crude reaction mixture.
First we studied the reaction of 1 with Bu3SnH in the
absence of selenol. A solution of Bu3SnH (120 mol % vs
1) and AIBN (15 mol % vs 1) was added dropwise over
20 h to a 0.01 M solution of 1 in benzene at reflux under
nitrogen. Chromatographic purification of the relatively
complex reaction mixture enabled isolation of 37% of the
(14) At 25 °C the rate constant for hydrogen abstraction from Bu3-
SnH by a primary alkyl radical is 2.4 × 106 M-1 s-1: Chatgilialoglu,
C.; Ingold, K. U.; Scaiano, J . C. J . Am. Chem. Soc. 1981, 103, 7739-
7742.
(15) At 25 °C the rate constant for hydrogen abstraction from PhSeH
by a primary alkyl radical is 2.1 × 109 M-1 s-1: Newcomb, M.; Varick,
T. R.; Ha, C.; Manek, M. B.; Yue, X. J . Am. Chem. Soc. 1992, 114,
8158-8163.
(16) Newcomb, M.; Musa, O. M.; Martinez, F. N.; Horner, J . H. J .
Am. Chem. Soc. 1997, 119, 4569-4577.
(17) Crich, D.; Mo, X.-S. J . Org. Chem. 1997, 62, 8624-8625.
(18) Thyagarajan, B. S.; Kharasch, H. B.; Lewis, H. B.; Wolf, W. J .
Chem. Soc., Chem. Commun. 1967, 614-614.
(19) (a) Hey, D. H.; J ones, G. H.; Perkins, M. J . J . Chem. Soc., Chem.
Commun. 1969, 1375-1376. (b) Hey, D. H.; J ones, G. H.; Perkins, M.
J . J . Chem. Soc. (C) 1971, 116-122.

Notes
J . Org. Chem., Vol. 63, No. 8, 1998 2767
Ta ble 1. Ad d ition of Ar yl Ra d ica ls to Ar en es
but has precedent in the work of Barton, when sufficent
thermodynamic driving force, such as gain of the aro-
matic stabilization energy20 or relief of ring strain,17 is
provided.
substrate [PhSeH] recovered
entry [conc, M]
(M)%
substrate
products (% yield)
1
2
3
4
1 [0.01]
1 [0.01]
10 [0.02]
10 [0.02]
0
1 (37)
1 (0)
10 (27)
10 (0)
4 (23), 7 (<5), 9 (12)
4 (22), 7 (43), 9 (22)
11 (25), 14 (5), 15 (3)
11 (17), 12 + 13
(40; 6.8:1) + 14 (12),
15 (21)
Finally, we examined 21, the regiosiomer of 10. In the
absense of benzeneselenol (Table 1, entry 5) chain
propagation was again poor and a substantial amount
of substrate was recovered. The major product from this
uncatalyzed reaction was the rearomatized benzene
adduct 26, which was isolated in 37% yield. In the
presence of 0.004 M diphenyl diselenide, all of the
substrate was consumed and the yield of 26 reduced in
favor of two new products, the isomeric cyclohexadienes
24 and 25, which were isolated in 77% combined yield
(Table 1, entry 6).
0.004
0
0.004
5
6
21 [0.02]
21 [0.02]
0
0.04
21 (32)
21 (0)
22 (8), 26 (37)
22 (12), 24 +
25 (77; 2.5:1), 26 (8)
Sch em e 2
The most important conclusion to be drawn from the
above results is that benzeneselenol is indeed effective
as a catalyst for the quenching of cyclohexadienyl radicals
by stannanes. Furthermore, any cyclohexadienes so-
formed are relatively stable to the reaction conditions and
may be isolated by silica gel chromatography without
difficulty.
Selenol quenching of the various cyclohexadienyl radi-
cals may in principle occur at either of the terminii of
the pentadienyl system or at the central carbon. In
practice, we observe moderate selectivity for quenching
at the central carbon with preferential formation of the
skipped dienes. These (7, 12, and 24) are readily
discernible from their respective conjugated isomers
owing to the symmetry of the 1,4-cyclohexadienyl system
which is reflected in the various NMR spectra. Beckwith
has previously recorded a similar phenomenon in the
quenching of related cyclohexadienyl radicals by oxygen,
which he attributed to the reduced steric strain in the
skipped as opposed to conjugated cyclohexadiene prod-
ucts.21 It seems likely that a similar effect is operative
on the transition states for quenching by benzeneselenol,
there being a considerable steric interaction between the
incoming selenol and the adjacent substituent for quench-
ing at the terminus of the pentadienyl system. It is
possible that the use of a more bulky selenol could
improve the regioselectivity of the quenching somewhat
and so render the present reactions preparatively useful.
A further point of interest concerns the widely differing
propensity toward cyclization or intermolecular addition
selenol. Inspection of the NMR spectra of the crude
reaction mixtures showed that in the absence of Ph-
SeSePh (Table 1, entry 3) these products were formed in
only trace amounts. However, the same NMR spectra
indicated numerous related products, which were not
isolated and characterized, but which must arise from
dimerization of the adduct cyclohexadienyl radical. These
purported dimers, which account for much of the missing
material in the uncatalyzed reactions, are more or less
absent from the catalyzed process, when the mass bal-
ance is also good. We suggest that the formation of all
products may be rationalized as in Scheme 2, in which
it is seen that the somewhat unexpected formation of
biphenyl arises from a 5-exo-trig cyclization to give 18,
with subsequent expulsion of the carboxyl radical (19),
and eventual decarboxylation. This process is related to
the biaryl synthesis developed by Motherwell.11 The
â-elimination of carboxyl radicals is a very rare process
(20) Barton, D. H. R.; Dowlatshahi, H. A.; Motherwell, W. B.;
Villemin, D. J . Chem. Soc., Chem. Commun. 1980, 732.
(21) Beckwith, A. L. J .; O’Shea, D. M.; Roberts, D. H. J . Am. Chem.
Soc. 1986, 108, 6408-6409.

2768 J . Org. Chem., Vol. 63, No. 8, 1998
Notes
to benzene of the three systems studied. With amide 1,
we found no conclusive evidence for addition to solvent.
Given that the rate constant for addition of a phenyl
radical to benzene is 4.5 × 105 M-1 s-1 at 25 °C22 and
taking a conservative ratio of 10:1 for the cyclization to
addition products in this reaction, we estimate a mini-
mum rate constant of 5 × 107 s-1 for this particular
cyclization. On the other hand, with 21 no evidence for
cyclization was observed. Here, taking a similarly con-
servative estimate of 86:223 for the ratio of addition to
cyclization products, we estimate an absolute maximum
rate constant of 1 × 105 s-1 for this cyclization. Some-
where between the two extremes we have the radical 16.
In this case, using a ratio of 33:40 for total cyclization to
total addition derived products, we arrive at a cyclization
rate constant of 4 × 106 s-1. These numbers take no
account of the effect of the substituents on the rate of
aryl radical addition to benzene, nor of the discrepancy
in temperature between the experiments described and
that at which the clock reaction was measured. Never-
theless, substituent effects for the σ-type aryl radicals
are relatively small,24 and the numbers should serve as
a reasonable first approximation. We also note that the
numbers are for total 5-exo plus 6-endo cyclization as the
present experiments do not permit us to comment on
whether the 6-endo products arise directly or via rear-
rangment of a kinetically 5-exo cyclized radical.
The approximately 1 order of magnitude difference in
cyclization rate constants between 1 and 10 can be
confidently ascribed to differing rotamer populations.
Unlike simple aliphatic amides, anilides and N-alkyl
anilides are known to prefer the anti-conformation
depicted.25 Radical 3 is therefore generated in a confor-
mation somewhat prearranged for cyclization. On the
other hand, O-aryl esters, such as phenyl benzoate, adopt
the standard syn-conformation about the C-O bond both
in the crystal26 and in solution.27 Therefore any radical
(16) generated from 10 will need to undergo rotation
about the C(dO)O bond to the higher energy anti-
conformer (16′) before cyclization can occur.
ring has rotated out of the plane of the carbonyl group.
Evidently, these dynamics are so unfavorable as to
effectively preclude the cyclization of 21.
We now return to the question of the catalytic effect
of benzeneselenol on the propagation sequence. The
effect is real, whether the additions are intra- or inter-
molecular, as is clearly seen from the effect on chain
propagation, as reflected in substrate consumption, in
each of the three examples studied. Moreover, the much
higher yields of cyclohexadienes isolated in the presence
of benzeneselenol strongly suggest that the effect does
arise from quenching of cyclohexadienyl radicals by the
selenol. The C-H bond dissociation energy for 1,4-
cyclohexadiene is 73 ( 2 kcal mol-1 28 and that for Sn-H
in trimethyltin hydride 74 kcal mol-1 29
from which it is
,
seen that propagation by tin hydride would be essentially
thermoneutral. Estimates for the Se-H bond dissocia-
tion energy in benezeneselenol have varied widely from
67 to 74 kcal mol-1 13,30
but it has recently been deter-
,
mined, at least for the gas phase, to be 78 ( 4 kcal mol-1
using a mass spectrometric method.31 Evidently, the
reaction in question is at best borderline thermoneutral
and likely slightly exothermic, yet it apparently proceeds
and much better than chain transfer with tin hydride.
Fully conscious of the controversy surrounding Robert’s
concept of polarity reversal catalysis,32 we suggest that
the rate of hydrogen atom transfer from the selenol is
accelerated by substantial polarization at the transition
state as depicted in 32 and correspondingly that abstrac-
tion of hydrogen from the stannane by the cyclohexadi-
enyl radical is retarded by unfavorable polar effects. The
whole sequence is driven in the forward direction by the
combination of eq 2, which regenerates the selenol, and
by the irreversible removal of the stannyl radical in the
form of a stannyl halide (eq 3). Whatever the reason for
the rate acceleration, it is evident that any further
stabilization of the cyclohexadienyl radical will render
the chain transfer step with the selenol significantly
endothermic and therefore less likely. Thus, those cy-
Structural studies of phenyl benzoate further tell us
that it not only prefers the syn-ester conformation but
also that the plane of the O-Ph ring is twisted some 65°
out of that of the rest of the molecule as in 27.26,27 If
this twist persists in the higher energy anti-conformation,
then once radical 16 is in the anti-conformation (16′), it
will be ideally set up for cyclization as in 28. However,
such a twist would mean that the radical (23) derived
from 21 is not poised for cyclization even when in the
anti-conformation (29). Indeed, for the cyclization of 21
to occur the derived radical 23 must adopt either con-
formation 30, a high-energy conformation on the pathway
for interconversion of the syn- and anti-ester minima, or
an anti-ester conformation 31 in which the benzoate Ph
(22) (a) Scaiano, J . C.; Stewart, L. C. J . Am. Chem. Soc. 1983, 105,
3609. (b) Levy, A.; Meyerstein, D.; Ottolenghi, M. J . Phys. Chem. 1973,
77, 3044-3047.
(23) 86% total addition products; only 2% of material unaccounted
for which may be cyclization derived materials.
(24) Migita, T.; Takayama, K.; Abe, Y.; Kosugi, M. J . Chem. Soc.,
Perkin Trans. 2 1979, 1137-1142.
(28) Burkey, J . J .; Majewski, M.; Griller, D. J . Am. Chem. Soc. 1986,
108, 2218-2221.
(29) Griller, D.; Kanabus-Kaminska, J . M.; Maccoll, A. J . Mol.
Struct. (THEOCHEM) 1988, 163, 125-131.
(30) Newcomb, M.; Manek, M. B.; Glenn, A. G. J . Am. Chem. Soc.
1991, 113, 949-958.
(31) Leeck, D. T.; Li, D. T.; Chyall, L. J .; Kenttamaa, H. I. J . Phys.
Chem. 1996, 100, 6608-6611.
(32) (a) Allen, R. P.; Roberts, B. P.; Willis, C. R. J . Chem. Soc., Chem.
Commun. 1989, 1387-1388. (b) Roberts, B. P. J . Chem. Soc., Perkin
Trans. 2 1996, 2719-2725. (c) Roberts, B. P.; Steel, A. J . J . Chem.
Soc., Perkin Trans. 2 1994, 2155-2162. (d) Zavitsas, A. A.; Chatgil-
ialoglu, C. J . Am. Chem. Soc. 1995, 117, 10645-10654. (e) Zavitsas,
A. A. J . Chem. Soc., Perkin Trans. 2 1996, 391-393.
(25) (a) Bourn, A. J . R.; Gillies, D. G.; Randall, E. W. Tetrahedron
1966, 22, 1825-1829. (b) Nagarajan, K.; Nair, M. D.; Pillai, P. M.
Tetrahedron 1967, 23, 1683-1690. (c) Azumaya, I.; Kagechika, H.;
Yamaguchi, K.; Shudo, K. Tetrahedron 1995, 51, 5277-5290.
(26) Adams, J . M.; Morsi, S. E. Acta Crystallogr. Sect B 1976, B32,
1345-1347.
(27) Le Fevre, R. J . W.; Sundaram, A. J . Chem. Soc. 1962, 3904-
3915.

Notes
J . Org. Chem., Vol. 63, No. 8, 1998 2769
h. After a further 1 h at reflux, the reaction mixture was cooled
to room temperature and the solvent removed under reduced
pressure. The crude reaction mixture was taken up in aceto-
nitrile and washed with petroleum ether and then evaporated
under reduced pressure and purified by chromatography on silica
gel (eluent hexanes:ether 10:1) to give the reduced product 11
(64 mg, 25%), the 6-endo product 14 (13 mg, 5%), and biphenyl
15 (6 mg, 3%), together with recovered starting material (10)
(113 mg, 27%). Phenyl benzoate (11): mp 66-67 °C (lit.38 mp
clohexadienyl radicals arising from the formal 6-endo
pathway (6 and 20), which benefit from extra stabiliza-
tion by the heteroatom, are apparently not quenched by
the selenol but eventually undergo oxidation with the
formation of the aromatized products 9 and 14, respec-
tively.
Finally, we draw attention to the fact that the inclusion
of catalytic benzeneselenol in these reaction mixtures
does not significantly increase the proportion of the direct
reduction products (4, 11, and 22). This is in line with
our previous observations when we noted that the stan-
nane-mediated trapping of neither aryl nor vinyl radicals
is accelerated by the presence of catalytic selenol.13 This
may be indicative of unfavorable polar effects at the
transition state for hydrogen abstraction from selenols
by σ-type radicals.13 Alternatively, it may simply reflect
the fact that the rate constant for the quenching of aryl
radicals by tributyltin hydride already approaches the
diffusion controlled limit33 and so leaves little room for
further acceleration.13
1
66-68 °C); H NMR δ 7.20-7.32 (3H, m), 7.42 (2H, app t, J )
7.4 Hz), 7.53 (2H, app t, J ) 7.5 Hz), 7.65 (1H, 2H, tt, J ) 1.3
and 7.4 Hz), and 8.24 (2H, dt, J ) 7.1 and 1.7 Hz); 13C NMR δ
121.9 (2C), 125.8, 128.5 (2C), 129.4, 130.1 (2C), 133.5, 150.9, and
165.1. Benzo[c]chromen-6-one (14): mp 91-92 °C (lit.38 mp 89-
1
92.5 °C); H NMR δ 7.35 (2H, m), 7.50 (1H, dt, J ) 1.4 and 7.1
Hz), 7.60 (1H, t, J ) 8.0 Hz), 7.81 (1H, dt, J ) 1.3 and 7.7 Hz),
8.04 (1H, dd, J ) 1.4 and 8.1 Hz), 8.10 (1H, d, J ) 8.0 Hz), and
8.38 (1H, d, J ) 7.8 Hz); 13C NMR δ 117.7, 117.9, 121.1, 121.6,
122.7, 124.5, 128.8, 130.4, 130.5, 134.7, 134.8, 151.2, and 161.2.
Biphenyl (15): mp 69-70 °C; 1H NMR δ 7.35 (2H, tt, J ) 7.3
and 1.4 Hz), 7.46 (4H, t, J ) 7.1 Hz), and 7.60 (4H, d, J ) 7.3
Hz); 13C NMR δ 127.1 (2C), 127.2, 128.7 (2C), and 141.2.
Rea ction of P h en yl 2-Iod oben zoa te (10) w ith Bu 3Sn H
a n d P h SeSeP h . The above experiment was repeated with the
difference being that PhSeSePh (80 mg, 0.26 mmol) was added
to the initial solution of 10 in benzene before addition of Bu3-
SnH commenced. Isolation and chromatographic purification
as above provided an inseparable mixture of 12 and 13 (143 mg,
40%, 6.8:1), 11 (44 mg, 17%), 14 (31 mg, 12%), and biphenyl 15
(42 mg, 21%). Phenyl 2-(cyclohexadienyl)benzoate (12 and 13):
1H NMR, major isomer (12) δ 2.78 (2H, m), 5.03 (1H, m), 5.85
(4H, m), 7.2-7.42 (4H, m), 7.42-7.6 (4H, m), and 8.12 (1H, dd,
J ) 7.4 and 1.4 Hz); minor isomer (13) δ 2.38 (1H, m), 2.78 (1H,
m), 4.63 (1H, m), and 5.68-6.15 (4H, m), 8.25 (1H, d, J ) 7.5
Hz); 13C NMR, major isomer (12) δ 25.7, 37.9, 121.7 (2C), 124.0
(2C), 125.9, 126.2, 128.3 (2C), 128.5, 129.5 (2C), 130.4, 130.6,
132.8, 146.9, 150.8, and 166.1; minor isomer (13) δ 31.3, and
35.9; HRMS calcd for C13H11O (M+• - PhO•) 183.0810, found
183.0756.
Exp er im en ta l Section 34
Reaction of o-Iodo-N-m eth ylben zan ilide (1) with Bu 3Sn H.
To a solution of o-iodo-N-methylbenzanilide8 1 (0.5 g, 1.48 mmol)
in benzene (150 mL) at reflux under Ar was added a solution of
Bu3SnH (0.48 mL, 1.8 mmol) and AIBN (32 mg, 0.22 mmol) in
benzene (60 mL) dropwise by means of a syringe pump over 20
h. After a further 1 h at reflux, the reaction mixture was cooled
to room temperature and the solvent removed under reduced
pressure. The crude reaction mixture, taken up in acetonitrile,
was washed with petroleum ether, evaporated, and purified by
chromatography on silica gel (eluent hexanes:ether 3:1) to give
4 (71 mg, 23%), 9 (37 mg, 12%), and recovered 1 (184 mg, 37%).
N-Methylbenzanilide 4: mp 183-186 °C (lit.35 mp 185-187 °C);
1H NMR δ 3.49 (3H, d, J ) 1.3 Hz), 7.02 (2H, app d, J ) 7.0
Hz), 7.09-7.23 (6H, m), and 7.28 (2H, m); 13C NMR δ 38.3, 126.4,
126.8 (2C), 127.6 (2C), 128.6 (2C), 129.0 (2C), 129.5, 135.8, 144.8,
and 170.6. N-Methyl-6(5H)-phenanthridinone 9: mp 104-106
°C (lit.8 mp 108.5 °C): 1H NMR δ 3.79 (3H, s), 7.30 (1H, t, J )
8.0 Hz), 7.38 (1H, d, J ) 8.2 Hz), 7.55 (2H, m), 7.74 (1H, dt, J )
1.3 and 8.3 Hz), 8.24 (2H, d, J ) 8.67 Hz), and 8.53 (1H, d, J )
7.9 Hz); 13C NMR δ 20.9, 115.0, 119.1, 121.5, 122.4, 125.5, 127.8,
128.8, 129.5, 132.3, 133.4, 137.9, and 161.5.
Reaction of o-Iodo-N-m eth ylben zan ilide (1) with Bu 3Sn H
a n d P h SeSeP h . The above experiment was repeated with the
difference being that PhSeSePh (92 mg, 0.30 mmol) was added
to the initial solution of 1 in benzene before addition of Bu3SnH
commenced. Isolation and chromatographic purification as
above provided 4 (74 mg, 22%), 9 (68 mg, 22%), and 7 (134 mg,
43%). Spirolactam 7, an oil, had spectral characteristics in full
agreement with the literature:36 1H NMR δ 2.90 (5H, m), 5.27
(2H, dt, J ) 10.3 and 2 Hz), 6.26 (2H, dt, J ) 10.2 and 3.3 Hz),
7.25 (1H, dd, J ) 6.9 and 0.8 Hz), 7.4-7.53 (2H, m), and 7.8
(1H, dd, J ) 6.8 and 0.6 Hz); 13C NMR δ 25.0, 25.9, 63.2, 123.0,
123.1, 125.6 (2C), 128.3, 129.5 (2C), 131.2, 131.5, 149.7, and
167.5.
Rea ction of 2-Iod op h en yl Ben zoa te (21) w ith Bu 3Sn H.
To a solution of 2-iodophenyl benzoate 2139 (0.43 g, 1.3 mmol)
in benzene (65 mL) at reflux under Ar was added a solution of
Bu3SnH (0.45 mL, 1.69 mmol) and AIBN (22 mg, 0.13 mmol) in
benzene (20 mL) dropwise by means of a syringe pump over 15
h. After a further 1 h at reflux, the reaction mixture was cooled
to room temperature and the solvent removed under reduced
pressure. The crude reaction mixture, taken up in acetonitrile,
was washed with petroleum ether and then evaporated under
reduced pressure and purified by chromatography on silica gel
(eluent hexanes:ether 10:1) to give the reduced product 22 (21
mg, 8%), 26 (131 mg, 37%), and recovered starting 21 (138 mg,
32%). o-Biphenyl benzoate (26): mp 76-77 °C (lit.40 73-76 °C);
1H NMR δ 7.27-7.50 (11H, m), 7.59 (1H, tt, J ) 0.6 and 7.4
Hz), and 8.05 (2H, td, J ) 0.7 and 7.24 Hz); 13C NMR δ 123.0,
126.3, 127.3, 128.2 (2C), 128.38 (2C), 128.44, 128.9 (2C), 129.3,
130.0 (2C), 130.9, 133.3, 134.9, 137.4, 147.7, and 165.0.
Rea ction of 2-Iod op h en yl Ben zoa te (21) w ith Bu 3Sn H
a n d P h SeSeP h . The above experiment was repeated with the
difference that PhSeSePh (80 mg, 0.26 mmol) was added to the
initial solution of 21 in benzene before addition of Bu3SnH
commenced. Isolation and chromatographic purification as
above provided an inseparable mixture of 24 and 25 (276 mg,
77%), 22 (31 mg, 12%), and 26 (28 mg, 8%). 24 and 25: 1H NMR,
major isomer (24) δ 2.7 (2H, m), 4.25 (1H, m), 5.70-5.87 (4H,
m), 7.2-7.45 (4H, m), 7.54 (2H, t, J ) 6.6 Hz), 7.66 (1H, t, J )
7.5 Hz), and 8.24 (2H, dd, J ) 0.9 and 7.0 Hz); minor isomer
(25) δ 2.35-2.42 (1H, m), 2.50-2.62 (1H, m), 3.89 (1H, m), 5.70-
6.10 (4H, m), and 7.44 (2H, t, J ) 8.1 Hz); 13C NMR, major
isomer (24): δ 25.6, 35.7, 122.4, 124.3 (3C), 126.5, 127.2 (2C),
127.4, 128.6 (2C), 130.0,130.1(2C), 133.6, 134.8, 148.3, and 165.2;
minor isomer (25) δ 30.1, 33.6,121.7 (2C), 122.6, 123.7, 124.8,
Rea ction of P h en yl 2-Iod oben zoa te (10) w ith Bu 3Sn H.
To a solution of phenyl 2-iodobenzoate 1037 (0.42 g, 1.3 mmol)
in benzene (65 mL) at reflux under Ar was added a solution of
Bu3SnH (0.45 mL, 1.69 mmol) and AIBN (22 mg, 0.13 mmol) in
benzene (20 mL) dropwise by means of a syringe pump over 15
(33) At ambient temperatures the rate constant for hydrogen
abstraction from Bu3SnH by a phenyl radical is 7.8 × 108 M-1 s-1. At
80 °C it is estimated to be ∼1.0 × 109 M-1 s-1: Garden, S. J .; Avila, D.
V.; Beckwith, A. L. J .; Bowry, V. W.; Ingold, K. U.; Lusztyk, J . J . Org.
Chem. 1996, 61, 805-809.
(34) For the general experimental part, see footnote 13a.
(35) von Braun, J .; Weisbach, K. Chem. Ber. 1930, 63, 489-497.
(36) Hey, D. H.; J ones, G. H.; Perkins, M. J . J . Chem. Soc. (C) 1971,
116-122.
(37) Singh, A.; Andrews, L. J .; Keefer, R. M. J . Am. Chem. Soc. 1962,
84, 1179-1185.
(38) Heacock, R. A.; Hey, D. H. J . Chem. Soc. 1954, 2481-2484.
(39) Buchan, S.; McCombie, H. J . Chem. Soc. 1931, 137-144.
(40) Wenkert, E.; Barnett, B. F. J . Am. Chem. Soc. 1960, 82, 4671-
4675.

2770 J . Org. Chem., Vol. 63, No. 8, 1998
Notes
125.4, 125.9, 126.2, 128.9, 129.2, 129.3, 129.5 (2C), 133.5, 137.4,
149.7, and 164.5; HRMS calcd for C19H16O2 276.1150, found
276.1133.
Su p p or tin g In for m a tion Ava ila ble: copies of 1H and 13C
NMR spectra for the mixture of 12 and 13 and of 24 and 25 (4
pages). This material is contained in libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
Ack n ow led gm en t . We thank the NSF (CHE
9625256) for support of this work and Dr. Qingwei Yao
for a preliminary experiment.
J O972197S