Rate constants for anilidyl radical cyclization reactions

Article abstract of DOI:10.1021/jo051953o

N-Aryl-5,5-diphenyl-4-pentenamidyl radicals (3) were produced by 266 nm laser-flash photolysis of the corresponding N-(phenylthio) derivatives, and the rate constants for the cyclizations of these radicals were measured directly. The 5-exo cyclization rea

Full text of DOI:10.1021/jo051953o

Rate Constants for Anilidyl Radical Cyclization Reactions
E. Martinez, II and M. Newcomb*
Department of Chemistry, UniVersity of Illinois at Chicago, 845 West Taylor Street,
Chicago, Illinois 60607
men@uic.edu
ReceiVed September 17, 2005
N-Aryl-5,5-diphenyl-4-pentenamidyl radicals (3) were produced by 266 nm laser-flash photolysis of the
corresponding N-(phenylthio) derivatives, and the rate constants for the cyclizations of these radicals
were measured directly. The 5-exo cyclization reactions were fast (kc > 2 × 105 s-1), and radicals 3
generally behaved as electrophilic reactants with a Hammett correlation of F ) 1.9 for five of the six
radicals studied. However, the p-methoxyphenyl-substituted radical 3f cyclized much faster than expected
from the Hammett analysis. Variable temperature studies of parent radical 3a (aryl ) phenyl) gave an
Arrhenius function with log k ) 9.2 - 4.4/2.3RT (kcal/mol). The rate constant for the reaction of
p-ethylphenyl-substituted anilidyl radical 3b with Bu3SnH at 65 °C was kT ) 4 × 105 M-1 s-1.
Introduction
received increased attention. As part of a broad survey of
hypervalent iodine reagents in synthesis, Nicolaou and co-
workers introduced methodology that involved oxidations of
anilides to anilidyl radicals by hypervalent iodine reagents;12-17
reactions of several N-aryl amides, carbamates, and ureas
containing remote unsaturation with o-iodoxybenzoic acid (IBX)
gave cyclic products, apparently formed via 5-exo radical
cyclization reactions. In related works, Studer and Janza reported
that IBX oxidation of acylated alkoxyamines gave alkoxyamidyl
radicals in good yields,18 and Sua´rez and co-workers studied
intramolecular hydrogen atom transfer reactions to amidyl
radicals produced by reactions of amides with (diacetoxyiodo)-
benzene and iodine under irradiation.19 The generally high yields
and the facility of the IBX-promoted anilidyl radical reactions
Applications of radical reactions for the synthesis of small
molecules gained popularity in the past decade largely in the
context of carbon-centered radicals.1-5 Heteroatom-centered
radicals are less common in synthesis, in part because of tedious
preparations and instabilities of heteroatom radical precursors.
Oxidative entries to heteroatom-centered radicals are attractive
because they start from a primary reactant such as an alcohol
or amine instead of a derivative. The use of hypervalent iodine
reagents for the production of alkoxyl radicals from alcohols
has developed into an attractive synthetic method.6,7
A number of entries to nitrogen-centered radicals are
known,8-11 but oxidative entries to these transients have recently
(1) Giese, B. Radicals in Organic Synthesis: Formation of Carbon-
Carbon Bonds; Pergamon: Oxford, England, 1986.
(11) Stella, L. In Radicals in Organic Synthesis; Renaud, P., Sibi, M.
P., Eds.; Wiley-VCH: Weinheim, Germany, 2001; Vol. 2, pp 407-426.
(12) Nicolaou, K. C.; Zhong, Y. L.; Baran, P. S. Angew. Chem., Int. Ed.
2000, 39, 622-625.
(2) Curran, D. P. Synthesis 1988, 417-439.
(3) Curran, D. P. Synthesis 1988, 489-513.
(4) Jasperse, C. P.; Curran, D. P.; Fevig, T. L. Chem. ReV. 1991, 91,
1237-1286.
(13) Nicolaou, K. C.; Zhong, Y. L.; Baran, P. S. Angew. Chem., Int. Ed.
2000, 39, 625-628.
(5) Radicals in Organic Synthesis; Renaud, P., Sibi, M. P., Eds.; Wiley-
VCH: Weinheim, Germany, 2001.
(14) Nicolaou, K. C.; Baran, P. S.; Kranich, R.; Zhong, Y. L.; Sugita,
K.; Zou, N. Angew. Chem., Int. Ed. 2001, 40, 202-206.
(15) Nicolaou, K. C.; Baran, P. S.; Zhong, Y. L.; Sugita, K. J. Am. Chem.
Soc. 2002, 124, 2212-2220.
(16) Nicolaou, K. C.; Sugita, K.; Baran, P. S.; Zhong, Y. L. J. Am. Chem.
Soc. 2002, 124, 2221-2232.
(17) Nicolaou, K. C.; Baran, P. S.; Zhong, Y. L.; Barluenga, S.; Hunt,
K. W.; Kranich, R.; Vega, J. A. J. Am. Chem. Soc. 2002, 124, 2233-2244.
(18) Janza, B.; Studer, A. J. Org. Chem. 2005, 70, 6991-6994.
(19) Martin, A.; Pe´rez-Martin, I.; Sua´rez, E. Org. Lett. 2005, 7, 2027-
2030.
(6) Sua´rez, E. In Radicals in Organic Synthesis; Renaud, P., Sibi, M.
P., Eds.; Wiley-VCH: Weinheim, Germany, 2001; Vol. 2, pp 440-454.
(7) Hartung, J. In Radicals in Organic Synthesis; Renaud, P., Sibi, M.
P., Eds.; Wiley-VCH: Weinheim, Germany, 2001; Vol. 2, pp 427-439.
(8) Stella, L. Angew. Chem., Int. Ed. Engl. 1983, 22, 337-350.
(9) Esker, J. L.; Newcomb, M. In AdVances in Heterocyclic Chemistry;
Katritzky, A. R., Ed.; Academic Press: San Diego, CA, 1993; Vol. 58, pp
1-45.
(10) Fallis, A. G.; Brinza, I. M. Tetrahedron 1997, 53, 17543-17594.
10.1021/jo051953o CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/20/2005
J. Org. Chem. 2006, 71, 557-561
557

Martinez and Newcomb
TABLE 1. Rate Constants for the Cyclizations of Radicals 3 in
THFa
SCHEME 1
radical
substituent
σp
kcyc (s-1
)
% yieldb
3a
3b
3c
3d
3e
3f
H
0
(7.8 ( 0.3) × 105
(2.5 ( 0.1) × 105
(1.3 ( 0.1) × 106
(3.0 ( 0.1) × 106
(1.4 ( 0.1) × 107
(7.7 ( 0.8) × 107
8
13
46
34
34
82
CH2CH3
F
C(O)CH3
CN
-0.13
0.15
0.47
0.70
OCH3
-0.12
a Rate constants determined by laser-flash photolysis in THF at (22 (
2) °C; errors at 2σ. b Isolated percent yields of lactams 5 in nonoptimized
preparative reactions.
suggest good synthetic utility for this method, but limited kinetic
information is available for amidyl radical reactions in gen-
eral,20,21 especially for anilidyl radicals. In this work, we report
kinetic results for anilidyl radical reactions that can be used for
synthetic planning.
FIGURE 1. Time-resolved spectrum from the reaction of radical 3a
in THF. The time slices are at 0.07, 0.22, 0.47, 0.78, 1.25, and 1.65
µs. The inset shows the kinetic trace at 335 nm; the decay of product
radical 4a due to radical-radical reactions is apparent by 3 µs.
again, could cyclize to radicals 4. Both radicals 3 and radicals
4 react with the tin hydride by hydrogen atom transfer reactions
that give acyclic anilides 1 and lactams 5, respectively. In
practice, lactams 5 were isolated from all of the precursors 2
studied in this work. Yields of lactams 5, which were not
optimized, showed an interesting apparent correlation with the
cyclization rate constants (Table 1).
Figure 1 shows the results of a typical LFP study with
precursor 2a. The phenylthiyl radical, produced instantly during
photolysis, has a strong absorbance with λmax ≈ 290 nm and a
broad, weaker absorbance with λmax ≈ 450 nm; signals from
this radical are decaying with time in Figure 1. Anilidyl radicals
3 were expected to have no appreciable absorbances at
wavelengths greater than 300 nm, and this was confirmed by
the production of the acetanilidyl radical from the photolysis
of N-(thiophenyl)aceteanilide. Diphenylalkyl radical 4a has a
long-wavelength absorbance at λmax ≈ 335 nm, and the signal
from this radical is growing with time in Figure 1. The inset in
Figure 1 shows a kinetic trace at λ ) 335 nm.
Radicals 3 were studied at ambient temperature, and the rate
constants determined in LFP studies are listed in Table 1. The
reactions were fast, and most of the radicals behaved predictably.
Exceptional behavior was observed with p-methoxy-substituted
radical 3f, which cyclized quite rapidly. Hammett analysis of
the cyclization rate constants is shown in Figure 2. Excluding
the results for radical 3f, the plot gives F ) 1.9, indicating a
relatively large increase in negative charge (or decrease in
positive charge) at nitrogen in the transition states for the
cyclization reactions.
Results and Discussion
We measured rate constants for anilidyl radical cyclization
reactions directly via laser flash photolysis (LFP) methods. The
design for the kinetic study is illustrated in Scheme 1. Anilides
1 were prepared and converted to the corresponding N-(phen-
ylthio)amide derivatives 2 that could serve as radical precursors
both for laser flash photolysis (LFP) experiments and chain
reactions. Precursors 2b-2f were prepared by reactions of the
corresponding anilides with NaH, and the resulting sodium salts
were treated with 1.0 equiv’s of PhSCl at 0 °C. This method,
which was used previously for preparations of N-(phenylthio)-
amides,20 failed for the preparation of 2a. We speculate that
the sodium salt of anilide 1a was only slightly soluble in THF
and that the PhSCl reagent partially reacted with the alkene
moiety. Precursor 2a was prepared successfully by reaction of
the lithium anilide salt, from the reaction of anilide 1a with
BuLi, with PhSCl. The lithium salt method was also used for
the preparation of N-(phenylthio)acetanilide, which served as a
reference compound in the LFP studies.
In the LFP studies, irradiation of precursors 2 with 266 nm
light cleaved the N-S bond to give the phenylthiyl radical and
the desired anilidyl radicals 3. Radicals 3 cyclized to give lactam
radicals 4 with first-order rate constants kc. Radicals 4 contain
the diphenylalkyl radical moiety, which has a strong chro-
mophore in its UV-visible spectrum, and the kinetics of the
cyclization reactions of 3 were measured directly.
In preparative reactions20 with tin hydride present, radical
chain reactions were propagated by reactions of the stannyl
radical with N-(phenylthio)anilides 2 to give radicals 3 that,
The Hammett study can be compared to that conducted by
Nicolaou and co-workers for IBX-promoted anilide cycliza-
tions.17 In this work, the F value was negative, indicating that
the positive charge developed on nitrogen during the rate-
(20) Esker, J. L.; Newcomb, M. Tetrahedron Lett. 1993, 34, 6877-6880.
(21) Horner, J. H.; Musa, O. M.; Bouvier, A.; Newcomb, M. J. Am.
Chem. Soc. 1998, 120, 7738-7748.
558 J. Org. Chem., Vol. 71, No. 2, 2006

Anilidyl Radical Cyclization Reactions
FIGURE 2. Hammett plot for the cyclizations of radicals 3. Radical
3f was not included in the regression line.
FIGURE 3. Arrhenius plot for the cyclization of radical 3a in THF.
SCHEME 2
limiting step in the reaction, and the conclusion was that the
electron-transfer reaction between IBX and the anilide (i.e.,
oxidation of the anilide) was the slow step in the reaction
sequence.17 The demonstration here that the anilidyl radicals
display an increase in negative charge on nitrogen in the
transition states for cyclization reactions is consistent with
previous conclusions; the rate-limiting process in the earlier
study17 could not be the radical cyclization reactions, and
oxidations of the anilides were the slow steps in the IBX-
promoted reactions.
The cyclization rate constant for radical 3f was greater than
that for any other radical in the series. The value is especially
large when the electronic effect of the methoxy substituent is
considered, as is apparent in Figure 2 where we used the
standard σ value for the p-methoxy group. From the Hammett
analysis, the expected rate constant for 3f would be similar to
that found for 3b or more than 2 orders of magnitude smaller
than that observed, and the σ+ value for the methoxy group of
-0.87 would predict an even smaller rate constant. The
reasonably good correlation for the rate constants for radicals
3a-3e and the obviously poor fit for radical 3f suggest that the
mechanism for the reaction of radical 3f is not the same as those
for the other radicals 3. In principle, it is possible that the
photolysis reaction of 2f gave ionic products (i.e., a thiophe-
noxide anion and an anilide cation),22 but the UV-visible
spectrum of the product was that of a diphenylalkyl radical with
this system was excellent. These interesting observations
prompted us to attempt to study the 6-exo cyclization of
p-methoxy-substituted anilidyl radical 6. In pairs of analogous
radicals that cyclize to give five- and six-membered rings, the
5-exo cyclization typically is about 2 orders of magnitude faster
than the 6-exo cyclization.26,27 Thus, it appeared possible that
6-exo cyclizations of methoxy-substituted anilidyl radicals might
be fast enough for synthetic utility. In practice, however, we
did not observe signal growth at 335 nm in the LFP reaction of
6. Instead, the large signal centered at about 300-305 nm was
observed to decay without the apparent formation of any new
signal. From previous studies, we expected 1,1-diphenylallyl
radical 7 to have a long-wavelength λmax at ca. 305 nm,21 and
we speculated that the hydrogen atom transfer of 6 to give 7
would be faster than the cyclization of 6 to give 8 (Scheme 2).
For the parent system, that is, anilidyl radical 6 without the
phenyl groups on the alkene, the activation energies for 6-exo
cyclization and 1,5-H-atom transfer are computed to be similar,28
and efficient 1,5-H-atom transfers in amidyl radicals are well
known.9,21,29 Given the overlap in absorbance of the phenylthiyl
radical with the expected absorbance of 7, it is possible that 7
was formed with a rate constant smaller than the kinetic limit
of our instrument but still remained undetectable.
λ
max at ca. 335 nm and not that of a diphenylalkyl cation, which
would have had a very strong signal with λmax at about 450
nm.23,24 The fast reaction of radical 3f might reflect its behavior
as a nucleophilic radical instead of an electrophilic radical as a
result of the electron-releasing substituent, but it is possible that
the kinetic value for 3f results from a profoundly different
reaction mechanism that we have not envisioned.25
In addition to the large rate constant for the cyclization of
radical 3f, the yield of lactam 5f in the preparative reaction for
Variable-temperature studies of the cyclization of radical 3a
in THF were performed between of 5 and 59 °C, and the results
are shown in Figure 3. The Arrhenius function for these data is
log k ) (9.2 ( 0.5) - (4.4 ( 0.7)/θ, where θ is 2.3RT in kcal/
mol (errors at 2σ). The entropy term, log A, is similar to those
found for 5-exo cyclizations of a number of carbon-centered
radicals,27,30 and we used this log A value to estimate rate
constants for radical 3b at 65 °C for the tin hydride trapping
kinetic studies discussed below.
(22) Kropp, P. J.; Fryxell, G. E.; Tubergen, M. W.; Hager, M. W.; Harris,
G. D.; McDermott, T. P.; Tornerovelez, R. J. Am. Chem. Soc. 1991, 113,
7300-7310.
(23) Faria, J. L.; Steenken, S. J. Phys. Chem. 1993, 97, 1924-1930.
(24) Horner, J. H.; Bagnol, L.; Newcomb, M. J. Am. Chem. Soc. 2004,
126, 14979-14987.
(25) A referee of an early draft suggested that the diphenylethene moiety
in 3f might be participating in an internal electron transfer reaction for this
radical. Consistent with that conjecture, the relative rate constant for a 5-exo
cyclization reaction versus the tin hydride trapping of a p-methoxy-
substituted anilidyl radical lacking the diphenylethene moiety was smaller
than that of the unsubstituted parent.17 If the rates of the tin hydride trapping
of the two radicals are comparable, then the cyclization of the p-methoxy-
substituted radical in that study was slower than the cyclization of the parent
radical, as expected from the results in Figure 2.
(26) Newcomb, M. Tetrahedron 1993, 49, 1151-1176.
(27) Newcomb, M.; Horner, J. H.; Filipkowski, M. A.; Ha, C.; Park, S.
U. J. Am. Chem. Soc. 1995, 117, 3674-3684.
(28) Chen, Q.; Shen, M. H.; Tang, Y.; Li, C. Z. Org. Lett. 2005, 7, 1625-
1627.
(29) Esker, J. L.; Newcomb, M. J. Org. Chem. 1993, 58, 4933-4940.
J. Org. Chem, Vol. 71, No. 2, 2006 559

Martinez and Newcomb
With the cyclization rate constants for anilidyl radicals 3
available, these species can be employed as radical clocks31,32
to time competing radical reactions by indirect kinetic studies.26
We applied this method to determine the rate constant for the
tin hydride reaction with radical 3b, which should be useful
for synthetic planning purposes. Radical 3b was the slowest
reacting radical in the series we studied, which permitted
reasonably efficient tin hydride trapping.
and cyclization of 9 was kT/kc ) 0.66 M.14,17 Using the rate
constant for the Bu3SnH reaction with 3b, we found the
estimated rate constant for the cyclization of radical 9 at 65 °C
to be kc ≈ 3 × 105 s-1. This value is 1 order or magnitude
smaller than the rate constant we found for the cyclization of
radical 3a at 65 °C (kc ) 2 × 106 s-1). The difference in the
kinetics of 9 and 3a is quite consistent with other radical
cyclization rate constants. For example, the diphenyl substitution
on the alkene moiety in 3a is expected to result in radical 3a
cyclizing about 100 times faster than that of hypothetical radical
10 as estimated from the kinetics of many carbon radical
pairs,27,30 and the ring and alkyl substitution in radical 9 is
predicted to result in about a 10-fold acceleration in cyclization
in comparison to that for 10, which is again based on the kinetics
of carbon radical pairs.26
The tin hydride trapping study was conducted at 65 °C.
Solutions of radical precursor 2b and Bu3Sn3H in benzene-d6
were equilibrated in a temperature-controlled bath, and the
reactions were initiated by the addition of a solution containing
AIBN.33 The reactions required several hours for completion
as determined by TLC, which indicates that the chain lengths
were quite short. After the reactions were complete, the solutions
were concentrated and analyzed by NMR spectroscopy to
determine the ratios of acyclic anilide 1b to lactam 5b. A plot
of [1b]/[5b] versus tin hydride concentration had a slope of 0.55
( 0.13 M-1, which is the ratio kT/kc, where kT is the second-
order rate constant for the reaction of the radical with Bu3SnH
and kc is the cyclization rate constant (Scheme 1). To determine
kc for the cyclization of radical 3b at 65 °C, we estimated the
Arrhenius function for this cyclization reaction using the entropic
log A term for radical 3a and the observed rate constant for the
cyclization of radical 3b at 22 °C. The approximate Arrhenius
function for the cyclization of 3b thus determined is log k )
9.2 - 5.1/2.3RT (kcal/mol), which gives a rate constant for
Anilidyl- and related aryl-substituted amidyl radicals were
shown to be generally useful for synthesis, especially in the
context of 5-exo cyclizations, and the oxidative entries to these
radicals are very attractive.12-17 The rate constants for anilidyl
radical reactions found in this work are consistent with previous
results and can be used for some aspects of synthetic planning.
In terms of reactivities in both cyclization reactions and reactions
with tin hydride, anilidyl radicals are 3 to 4 orders of magnitude
less reactive than analogous amidyl radicals, a kinetic effect
that is somewhat attenuated from that of phenyl group substitu-
tion on a carbon-centered radical.26,36 In general, 5-exo cycliza-
tion reactions of anilidyl radicals are fast enough that interfering
radical-radical coupling reactions can be avoided in typical
synthetic applications, but attempts at 6-exo cyclizations in
analogous anilidyl radicals are expected to be problematic on
the basis of our failure to observe a cyclization reaction of
radical 6 as well as observations from the synthetic studies by
Nicolaou’s group.12-17 The relatively large effect of the aryl
group substituents on the kinetics of the anilidyl radical
cyclization reactions is noteworthy and should be considered
in synthetic planning.
cyclization of 3b at 65 °C of kc ) 8 × 105 s-1
.
Assuming a minor solvent effect on the cyclization reaction
between benzene and THF,34 the rate constant kc determined
above can be combined with the ratio of rate constants from
the competing kinetic study to give a rate constant of kT ) (4
( 1) × 105 M-1 s-1 at 65 °C for the reaction of radical 3b
with tin hydride. This rate constant is approximately 1 order of
magnitude smaller than that for the reaction of tin hydride with
a primary alkyl radical at the same temperature.35,36 The high
reactivity of tin hydride with the relatively stable anilidyl radical
is not unexpected because a favorable polarity match exists
between the electron-deficient nitrogen radical in 3b and the
electron-rich hydride atom in tin hydride. A similar situation
exists with simple amidyl radicals, which react with tin hydride
at ambient temperature nearly 3 orders of magnitude faster than
do alkyl radicals.21,36
The tin hydride trapping results provide an interesting link
with a kinetic result found by Nicolaou and co-workers in their
studies of anilidyl radical cyclization reactions. When the
cyclization of radical 9 was studied in the presence of Bu3SnH
in toluene at 65 °C, the ratio of rate constants for the trapping
Experimental Section
Details for the preparation of anilides 1 and the properties of
compounds 1, 2, and 5 are in Supporting Information.
Sulfenamide Preparation. The following method (method B)
was used for the preparation of sulfenamides 2b-2f. Into a flame-
dried, round-bottomed flask equipped with a stirbar, bubbler, and
reflux condenser under static nitrogen in a 60 °C water bath was
placed 1.0 mmol of anilide 1 and 1.1 equiv of sodium hydride (60%
dispersion in oil). To this mixture, 10 mL of THF was added
dropwise via a syringe. The mixture was stirred for 40 min or until
bubbling ceased. The flask was cooled to -78 °C, and ca. 1.1 equiv
of phenylsulfenyl chloride37 was added dropwise (the yellow color
of the reagent persisted for approximately 1 min). The mixture was
stirred for 10 min and quenched by the addition of 5 mL of saturated
aqueous ammonium chloride solution. The crude mixture was
diluted with 50 mL of ether, extracted with 1 × 40 mL of saturated
sodium bicarbonate solution, 2 × 40 mL of water, and 1 × 40 mL
(30) Ha, C.; Horner, J. H.; Newcomb, M.; Varick, T. R.; Arnold, B. R.;
Lusztyk, J. J. Org. Chem. 1993, 58, 1194-1198.
(31) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317-323.
(32) Newcomb, M. In Radicals in Organic Synthesis; Renaud, P., Sibi,
M. P., Eds.; Wiley-VCH: Weinheim, Germany, 2001; Vol. 1, pp 317-
336.
(33) Photochemically initiated preparative reactions could be performed,
but photochemically initiated kinetic studies were not possible. The
phenylthiyl radical formed as a byproduct in the photolysis reaction will
react with tin hydride to give highly reactive thiophenol.
(34) Cursory studies showed that a change in solvent from THF to hexane
had at most a minor effect on the rate constants for the cyclization of radical
3a.
(35) Chatgilialoglu, C.; Ingold, K. U.; Scaiano, J. C. J. Am. Chem. Soc.
1981, 103, 7739-7742.
(36) Chatgilialoglu, C.; Newcomb, M. AdV. Organomet. Chem. 1999,
44, 67-112.
(37) Harpp, D. N.; Mathiaparanam, P. J. Org. Chem. 1972, 37, 1367-
1374.
560 J. Org. Chem., Vol. 71, No. 2, 2006

Anilidyl Radical Cyclization Reactions
of saturated brine and dried over MgSO4. The concentration of
the organic layer yielded oils, which were purified by chromatog-
raphy on silica gel using ethyl acetate/hexanes mixtures to yield
N-(phenylthio)-amides 2. Sulfenamide 2a was prepared by a similar
method with the exception of the use of BuLi as the base
(Supporting Information).
absorbance at 266 nm was ca. 0.5 AU. Solutions were sparged with
helium and allowed to flow through a 1 cm × 1 cm flow cell.
Laser irradiation was accomplished with the fourth harmonic of a
Nd:YAG laser (266 nm). All samples showed the growth of a signal
with λmax at ca. 335 nm that was reasonably well fit to a first-order
exponential function.
Preparation of Lactams 5. Solutions of sulfenamide 2 and tri-
n-butyl tin hydride were prepared as follows. Into a volumetric flask
was placed a fixed volume of a stock solution of sulfenamide 2 in
THF and a variable amount of distilled Bu3SnH. The mixture was
diluted with the appropriate volume of THF. The solution was
transferred to a Spectrosil quartz fluorometer cuvette equipped with
a Teflon-coated stirbar and a Teflon-coated screw-cap. Larger
preparative reactions were performed in a 3 cm × 30 cm quartz
tube. The solution was degassed with nitrogen for 2 min, capped,
and irradiated in a photoreactor using full-spectrum, medium-
pressure mercury lamps for 2.0 h. Consumption of the sulfenamide
was followed by TLC and was typically completed within 1.5 h.
The reaction mixture was transferred to a round-bottomed flask,
and the solvent was removed under vacuum. The residue was
dissolved in 5 mL of acetonitrile and extracted with 3 × 5 mL of
hexanes to remove excess Bu3SnH. The acetonitrile was re-
moved under reduced pressure yielding an oily residue, which
was dried under high vacuum and then dissolved in acetone-d6 for
NMR analysis. Preparative reactions were then subjected to col-
umn chromatography on silica gel using ethyl acetate/hexanes
mixtures. Isolated yields of 5, which were not optimized, are listed
in Table 1.
Competition Kinetics. A stock solution of 2b in benzene-d6 (69
mg/mL) was prepared, and 1.0 mL of this solution was added to
three 4-mL glass vials. To these vials was added 0.8, 0.4, and
0.2 mL of Bu3SnH. The volume in each vial was adjusted to
2.8 mL with benzene-d6. The vials were warmed in a bath at 65
°C, and 0.2 mL of a solution containing 25 mg of AIBN in 1 mL
of benzene-d6 was added to each. The final concentrations of
Bu3SnH were 1.0, 0.5, and 0.25 M. After 1.5 h, a second aliquot
of AIBN solution (0.2 mL) was added. Consumption of 2b was
monitored by TLC. After 2b was consumed, the reaction mixtures
were cooled to room temperature, concentrated to 1 mL volume
under a stream of nitrogen, and analyzed by NMR spectroscopy to
determine the ratio of 1b to 5b.
Acknowledgment. This work was supported by a grant from
the National Science Foundation. E.M. thanks the NIH for
fellowship support. We thank an anonymous referee for the
mechanistic suggestion noted in ref 25.
Supporting Information Available: Characterization data for
compounds 1, 2, and 4 and NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org.
Laser flash photolysis studies were performed as described
previously.24 Solutions of 2 in THF were prepared such that the
JO051953O
J. Org. Chem, Vol. 71, No. 2, 2006 561