Absolute rate constant for the reaction of aryl radicals with tri-n-butyltin hydride1

Full text of DOI:10.1021/jo951245a

J . Org. Chem. 1996, 61, 805-809
805
measured in the dibenzoyl peroxide/tin hydride system,
viz.,7 5.9 × 108 M-1 s-1 at 30 °C, and had assigned to
hydrogen atom abstraction by phenyl radicals, reaction
5, had to be reassigned to the benzoyloxyl radical/tin
hydride reaction.
Absolu te Ra te Con sta n t for th e Rea ction of
Ar yl Ra d ica ls w ith Tr i-n -bu tyltin Hyd r id e1
S. J . Garden,2 D. V. Avila,2 A. L. J . Beckwith,3
V. W. Bowry,3 K. U. Ingold,* and J . Lusztyk*
kH
9
Steacie Institute for Molecular Sciences, National Research
Council of Canada, Ottawa, Ontario, Canada K1A 0R6
Ar• + n-Bu3SnH
8 ArH + n-Bu3Sn•
(5)
We now describe the determination, by two indepen-
dent methods, of kH and also of kD, the rate constant for
deuterium abstraction from tri-n-butyltin deuteride by
aryl radicals. One method is based on new LFP data for
the reaction of aryl radicals with tetrahydrofuran (THF)
and its perdeutero derivative (TDF) in aqueous solution.
The other utilizes LFP data reported earlier by Scaiano
and Stewart12 for the reaction of phenyl radicals with
methyl methacrylate in Freon 113. The two independent
values which can be calculated for kH (R• ) aryl) are in
reasonable agreement with one another. At last, there-
fore, it is possible to apply conventional competitive
kinetic techniques to determine the magnitude of the rate
constants for reactions of aryl radicals with other com-
pounds of chemical7,17 and biochemical18 importance.
Received J uly 12, 1995
“A knowledge of the rates with which a radical R• is
generated, rearranged and trapped by tin hydride now
permits the rational planning of such free-radical reac-
tions”.4
For over a quarter of a century an important activity
at the National Research Council of Canada (NRCC) has
been the measurement of the absolute rate constants for
the reactions of carbon-centered radicals with tri-n-
butyltin hydride.5-8
kH
9
R• + n-Bu3SnH
8 RH + n-Bu3Sn•
(1)
The availability of kH values for a very wide variety of
R• structures has undoubtedly contributed to the popu-
larity of rational synthetic strategies which make use of
free-radicals and tin hydride.4,9-11 However, when we
Resu lts
Mea su r em en t of kH a n d kD by In ter m olecu la r
Com p etition s w ith TDF a n d THF . A major difficulty
associated with kinetic studies on aryl radicals is their
high reactivity toward all common organic solvents. We
have recently demonstrated21 that this problem can be
overcome by 308 nm LFP of the sodium salt of 4-iodo-
benzoic acid (1.55 × 10-2 M)22 in water. The kinetic
behavior of the resultant aryl radical was monitored at
320 nm via its reaction with the water-soluble “probe”23,24
molecule, the sodium salt of 4-styrenesulfonic acid, which
was employed at a concentration of 2.0 × 10-3 M in the
probe experiments. The global rate constants for reaction
of this aryl radical with the probe,24 kprobe, and with
perdeuterotetrahydrofuran, tetrahydrofuran, and a few
other compounds, kgl, were measured in the usual way
(incremental additions of the substrate at a constant
concentration of the probe), i.e.,
attempted to measure kH for R• ) C6H5 by our (then)
•
new laser flash photolysis (LFP) technique7 we were
misled by the view prevailing at that time that the
photolysis of dibenzoyl peroxide gave phenyl radicals
“instantaneously”.12 However, we later demonstrated13-16
(PhCO2)2
9
hν8 2Ph• + 2CO2
(2)
that the 308 nm LFP of diaroyl peroxides (including
dibenzoyl peroxide14) gave aroyloxyl radicals which sub-
sequently undergo a relatively slow decarboxylation. It
hν8 2 ArCO2
(3)
•
(ArCO2)2
9
ArCO2 slow8 Ar• + CO2
(4)
•
appeared, therefore, that the rate constant which we had
kexptl ) kprobe[probe] + kgl[substrate]
(I)
(1) Issued as NRCC No. 39084.
(2) NRCC Research Associate, 1992-94.
(3) Research School of Chemistry, Australian National University,
Canberra, ACT 0200, Australia.
(4) J asperse, C. P.; Curran, D. P.; Fevig, T. L. Chem. Rev. 1991, 91,
1237-1286.
(5) Carlsson, D. J .; Ingold, K. U. J . Am. Chem. Soc. 1968, 90, 1055.
Carlsson, D. J .; Ingold, K. U. J . Am. Chem. Soc. 1968, 90, 7047-7055.
(6) Chatgilialoglu, C.; Ingold, K. U.; Scaiano, J . C. J . Am. Chem.
Soc. 1981, 103, 7739-7742.
(7) J ohnston, L. J .; Lusztyk, J .; Wayner, D. D. M.; Abeywickreyma,
A. N.; Beckwith, A. L. J .; Scaiano, J . C.; Ingold, K. U. J . Am. Chem.
Soc. 1985, 107, 4594-4596.
(8) Avila, D. V.; Ingold, K. U.; Lusztyk, J .; Dolbier, W. R., J r.; Pan,
H.-Q.; Muir, M. J . Am. Chem. Soc. 1994, 116, 99-104.
(9) Curran, D. P. Synthesis 1988, 417-439. Curran, D. P. Synthesis
1988, 489-513.
These rate constants are given in Table 1.
A radical chain reaction between iodobenzene and tri-
n-butyltin deuteride (nominally 97 atom % D) at ca. 1:10
mole ratio was carried out in THF as solvent at 31 °C
(17) Glover, S. A.; Warkentin, J . J . Org. Chem. 1993, 58, 2115-
2121.
(18) The metabolism of certain exogenous compounds which may
give rise to phenyl radicals in vivo, e.g., PhNHNH2, PhI, and PhBr,
causes the peroxidation of hepatic cellular lipids with consequent
cellular damage.19 Peroxidation is undoubtedly a free radical process
since, in at least one case, the liver can be protected by dosing the
animal with the water-soluble radical-trapping antioxidant, Trolox.20
(19) Halliwell, B.; Gutteridge, J . M. C. Free Radicals in Biology and
Medicine, 2nd ed.; Clarendon Press: Oxford, U.K., 1989.
(20) Casini, A. F.; Pompella, A.; Comporti, M. Am. J . Pathol. 1985,
118, 225-237.
(21) Sommeling, P. M.; Mulder, P.; Louw, R.; Avila, D. V.; Lusztyk,
J .; Ingold, K. U. J . Phys. Chem. 1993, 97, 8361-8364.
(22) This concentration gives an optical density, OD ∼ 0.3 at 308
nm in our 7-mm LFP cell.
(23) Paul, H.; Small, R. D., J r.; Scaiano, J . C. J . Am. Chem. Soc.
1978, 100, 4520-4527.
(10) Neumann, W. P. Synthesis 1987, 665-683.
(11) Giese, B. Radicals in Organic Synthesis: Formation of Carbon-
Carbon Bonds. Pergamon: Oxford, U.K., 1986.
(12) Scaiano, J . C.; Stewart, L. C. J . Am. Chem. Soc. 1983, 105,
3609-3614.
(13) Chateauneuf, J .; Lusztyk, J .; Ingold, K. U. J . Am. Chem. Soc.
1987, 109, 897-899. Chateauneuf, J .; Lusztyk, J .; Ingold, K. U. J . Am.
Chem. Soc. 1988, 110, 2877-2885.
(14) Chateauneuf, J .; Lusztyk, J .; Ingold, K. U. J . Am. Chem. Soc.
1988, 110, 2886-2893.
(24) In Scheme 1 this reaction is written as an addition to form a
benzylic radical which is the species presumed to be mainly responsible
for the grow-in of the 320 nm absorption. However, the global rate
constant measured with this substrate includes all modes of reaction
including, for example, addition to the aromatic ring.
(15) Aroyloxyl radicals have also been detected by conventional EPR
spectroscopy.16
(16) Korth, H.-G.; Mu¨ller, W.; Lusztyk, J .; Ingold, K. U. Angew.
Chem., Int. Ed. Engl. 1989, 28, 183-185.
0022-3263/96/1961-0805$12.00/0 Published 1996 by the American Chemical Society

806 J . Org. Chem., Vol. 61, No. 2, 1996
Notes
Ta ble 1. Absolu te Ra te Con sta n ts for Rea ction of th e
that the n-Bu3SnD was not 100% deuterated. The
claimed isotopic purity was 96.6 atom % D which could
imply as much as 3.4% n-Bu3SnH which would serve as
an “extra” source of C6H6. In addition, there was the
possibility that some of the phenyl radicals which react
with n-Bu3SnD would abstract hydrogen from the butyl
groups rather than deuterium from the SnD moiety thus
giving additional “extra” C6H6. The correction for C6H6
4--O2CC6H4 Ra d ica l w ith som e Su bstr a tes in Wa ter a t 25
•
( 2 °C
substrate
kgl/107 M-1 s-1 a
32.5 ( 1.50 (kprobe
-
H2CdCHC6H4SO3
THFb
)
0.375 ( 0.030 (kTHF
0.132 ( 0.003 (kTDF
3.72 ( 0.23
)
)
TDFc
2-cyclohexen-1-ol
H2CdCHC(CH3)2OH
HCtCC(CH3)2OH
furan
2.96 ( 0.16
•
formation in the C6H5 /n-Bu3SnD reaction was checked
1.46 ( 0.11
experimentally by photolysis (4 min, λ > 300 nm) of
phenyl iodide in neat n-Bu3SnD (ca. 1:10 mole ratio). In
three separate experiments the mean actual yield of C6H6
was found to be 2.8 ( 0.5% of the total yield of benzene
(C6H5D + C6H6) s which implies that there is little or
no hydrogen abstraction from the butyl groups by the
phenyl radical. The actual yield of C6H6 (calculated as
described in the Experimental Section) must therefore
be reduced by 2.8% before the true magnitude of the rate
constant ratio, kTHF/kD, can be determined. If all of these
corrections are applied, we have:
2.35 ( 0.31
a
Errors are expressed as 2σ (95% confidence limit) but include
only random errors. Rate constants which are employed in the
text are given in parentheses. Tetrahydrofuran. c Perdeuterotet-
b
rahydrofuran.
(see Scheme 1). The chain reaction was initiated by
photolysis of the deoxygenated mixture (see Experimen-
tal Section) in a Pyrex reaction vessel with light having
λ > 300 nm. The iodobenzene was completely consumed
within 15 min and at this time a sample of the reaction
mixture was analyzed by GC/MS. The relative abun-
dance of the ions at m/z ) 76, 77, 78, 79, and 80 were
determined by selective ion monitoring and the actual
relative yields of benzene, (C6H6)actual, and benzene-d1
(C6H5D)actual, (i.e., concentrations of benzene corrected for
the contribution of the M-1 ion from benzene-d1 and
concentrations of benzene-d1, corrected for contributions
from natural abundance 13C and 2H in benzene) were
calculated as described in the Experimental Section.
kTHF (C6H6)actual - 0.028 (C6H5D)actual
)
×
kD
(C6H5D)actual
(0.960 × 0.972) (n-Bu3SnD)nominal
(II)
[THF]
The results of five separate experiments and the values
calculated for kTHF/kD are summarized in Table 2.
In order to obtain the true rate constant ratio, kTHF
/
kD, two corrections must be applied. First, the nominal
concentration of n-Bu3SnD must be reduced by 4% to
allow for the fact that the purity of this material was
only 96.0%. The second correction arises from the fact
In a second series of experiments the rate constant,
kTDF, for aryl radical attack on perdeuterotetrahydrofuran
(TDF, > 99.5 atom % D) was measured by LFP (vide
supra, Table 1 and Scheme 1). A radical chain reaction
Sch em e 1

Notes
J . Org. Chem., Vol. 61, No. 2, 1996 807
Ta ble 2. Rea ction of Iod oben zen e w ith Tr i-n -bu tyltin
Ta ble 3. Rea ction of Iod oben zen e w ith Tr i-n -bu tyltin
Hyd r id e in P er d eu ter iotetr a h yd r ofu r a n a t 31 °Ca
Deu ter id e in Tetr a h yd r ofu r a n a t 31 °Ca
reactants (mol)
reactants (mol)
105
PhI
×
104
×
102
THF/
C6H6/
102
×
105
PhI
×
105
×
102
TDFc n-Bu3SnH C6H5Dd
×
TDF/
C6H6/
102
kTDF/kH
×
e
n-Bu3SnDb × THF n-Bu3SnDb C6H5Dc
k
THF/kD
n-Bu3SnHb
d
1.13
1.42
2.40
2.01
1.08
1.08
1.35
1.89
1.60
2.06
1.22
1.23
1.24
1.22
1.22
113
91.1
65.6
76.3
59.2
0.970
0.838
0.597
0.735
0.588
0.858
0.920
0.910
0.963
0.993
1.03
1.03
1.08
0.931
1.08
16.9
1.57
3.53
6.83
2.47
1.27
1.27
1.27
1.28
1.27
75.1
809
360
187
514
7.74
0.506
1.41
2.84
0.936
0.172
0.244
0.197
0.188
0.208
a
a
Experiments are listed in the order in which they were carried
Experiments are listed in the order in which they were carried
b
b
out. The moles of n-Bu3SnD have been corrected for the purity
out. Since the n-Bu3SnH was supplied as 97% purity, the moles
and atom % in the following manner: (n-Bu3SnD)actual ) (n-
Bu3SnD)nominal × 0.960 × 0.972, where the n-Bu3SnD was supplied
as 96.0% purity and the atom % was determined from the mean
values of three experiments where iodobenzene was photolyzed
in neat n-Bu3SnD. c This ratio was calculated from the ion
intensities of m/z ) 78 and 79 and solution of the two simultaneous
equations as detailed in the Experimental Section. The intensity
data for C6H6 has been corrected for C6H6 formation from the
n-Bu3SnH present in the n-Bu3SnD sample: (C6H6)actual ) (C6H6)
of n-Bu3SnH have been reduced by 3%. c The perdeuteriotetrahy-
drofuran (TDF) was supplied as 99.5+ atom % deuterium and
hence no correction was applied for incompletely deuterated TDF.
d
This ratio was calculated from the ion intensities of m/z ) 78
and 79 and solution of the two simultaneous equations as detailed
e
in the Experimental Section. kTDF/kH ) 1/{(C6H6/C6H5D) × (TDF/
n-Bu3SnH)}.
Ta ble 4. Rea ction of p-Iod otolu en e (∼0.5 M) w ith
Tr i-n -bu tyltin Hyd r id e (Deu ter id e) a n d Meth yl
Meth a cr yla te in Cycloh exa n e a t 25 °C
- [0.028 × 0.96 × (C6H5D)]. kTHF/kD ) (C6H6/C6H5D)c/(THF/n-
d
Bu3SnD)b.
[ArH]b
[ArD(H)]b
[Add-D]
[n-Bu3SnH]a [MMA]
[n-Bu3SnD]a [MMA]
of iodobenzene and n-Bu3SnH in TDF was carried out in
the manner described above (see also Scheme 1). The
results of these experiments are summarized in Table 3.
In order to obtain the true rate constant ratio, kTDF/kH,
where kH refers only to abstraction of the hydrogen atom
from tin, i.e., from the SnH moiety, it would be necessary
to correct the actual yield of C6H6 for C6H6 formed by
hydrogen atom abstraction from the butyl groups of
n-Bu3SnH. Since this reaction is unimportant with
n-Bu3SnD we can safely ignore it with n-Bu3SnH. Fur-
thermore, since the TDF contained 99.5 + atom % D no
(M)
(M)
[Add-H]
(M)
(M)
0.12
0.22
0.40
0.72
1.31
1.42
1.35
1.20
0.61
0.81
1.7
0.14
0.25
0.42
0.74
1.31
1.45
1.40
1.25
0.54
0.69
1.25
2.2
3.1
a
Average concentration of n-Bu3SnH or n-Bu3SnD during the
reaction. b Ratio of yields of toluene (ArH or ArD + ArH) to reduced
first adduct (Add-H or Add-D) as measured by GC.
the reactant mixture.25 Under pseudo-first-order condi-
tions, i.e., low conversion of n-Bu3SnH, this equation
approximates to:
•
correction is necessary for C6H6 formed in the C6H5 /TDF
reaction. Since the n-Bu3SnH was only 97% pure the
equation used to derive the kTDF/kH ratios given in Table
3 was:
[MMA]([ArH]/[Add-H]) )
k
SH[SH]/kMMA + kH[n-Bu3SnH]/kMMA (V)
kTDF (C6H5D)actual 0.97(n-Bu3SnH)nominal
Thus, a plot of [MMA]([ArH]/[Add-H]) against [n-Bu3-
SnH] should give a straight line with a slope equal to
)
×
(III)
kH
[TDF]
(C6H6)actual
kH/kMMA
relevant equation is:
. For the reaction of ArI with n-Bu3SnD the
Mea su r em en t of kH a n d kD by In ter m olecu la r
Com p etition w ith Meth yl Meth a cr yla te (MMA). The
radical chain reduction of p-iodotoluene, ArI, with n-Bu3-
SnH(D) in the presence of MMA affords mixtures of
toluene, ArH(D), and the adduct, ArCH2CH(Me)CO2Me
(Add-H) or ArCH2CD(Me)CO2Me (Add-D):
[MMA]([ArD]/[Add-D]) )
k
SH[SH]/kMMA + kD[n-Bu3SnD]/kMMA (VI)
This equation applies because only the total yield of
toluene was measured (by GC), not the actual yield of
ArD (by, e.g., GC/MS), and because it seems safe to
assume that the Add• radical will react only with n-Bu3-
SnD.
n-Bu3Sn• + ArI f n-Bu3SnI + Ar•
(6)
kH(D)
Ar• + n-Bu3SnH(D)
8 ArH(D) + n-Bu3Sn• (5)
The experiments were conducted by adding n-Bu3SnH-
(D) and a catalytic amount of di-tert-butyl hyponitrite to
a solution containing about 1 mol equiv of p-iodotoluene
and an excess of MMA in cyclohexane. The reactions
were stopped after 2 h by which time they had proceeded
about 5-10% toward completion. The mixtures were
then quenched by the addition of carbon tetrachloride and
analyzed by GC.26 Four experiments were carried out
with n-Bu3SnH and four with n-Bu3SnD (see Table 4).
Plots according to eqs V and VI yield kH/kMMA ) 5.0 (
0.6 with kSH[SH]/kMMA ) 0.17 ( 0.24, and kD/kMMA ) 3.5
kMMA
Ar• + CH2dC(Me)CO2Me
8
ArCH2C˙ (Me)CO2Me (Add•) (7)
Add• + n-Bu3SnH(D) f Add-H(D) + n-Bu3Sn• (8)
The relative rate of formation of ArH and Add-H is
given by:
d[ArH]/d[Add-H] )
(25) The potential H-atom donors include MMA, the benzylic methyl
group of p-iodotoluene and the cyclohexane solvent.
(kSH[SH] + kH[n-Bu3SnH])/kMMA[MMA] (IV)
(26) Polymerization (i.e., Add• + MMAf) and stannylation (i.e.,
n-Bu3Sn• + MMAf) afforded minor byproducts under our conditions
(i.e., [n-Bu3SnH(D)]/[MMA] g 0.05, [ArI]/[MMA] g 0.3).
where kSH represents the total H-atom donor ability of

808 J . Org. Chem., Vol. 61, No. 2, 1996
Notes
( 0.4 with kSH[SH]/kMMA ) 0.21 ( 0.16, where the errors
correspond to 2σ.
more than our 95% (random error) confidence limits.
Hopefully, their mean value, viz., 5.2 × 108 M-1 s-1 will
be fairly reliable.34 More gratifyingly, the kH values
derived from the TDF experiments, viz., (6.5 ( 1.7) ×
108 M-1 s-1, and from the MMA experiments, viz., (9.0 (
1.1) × 108 M-1 s-1, are in agreement within our 95%
confidence limits. Our recommended value for the rate
constant for the reactions of aryl radicals with n-Bu3SnH
at ambient temperatures is the mean of the two mea-
sured values,35 viz.,
Discu ssion
Because the SOMO in aryl radicals lies in the plane of
the aromatic ring, i.e., these are “isolated” σ-radicals,27,28
substituents on the ring are expected (and have been
shown)29 to have only a relatively minor effect on the
rates of reactions of these radicals.30 We can therefore
assume that the two aryl radicals generated in the
kH ) 7.8 × 108 M-1 s-1
present work, viz., 4--O2CC6H4• and 4-CH3C6H4 , should
•
have rather similar reactivities. For the reactions in-
vestigated, solvent effects are also expected to be rela-
tively unimportant despite the wide range of solvents
employed.31 The small difference in the temperatures
employed in the product studies (31 °C) and in the LFP
measurements (25 ( 2 °C) can also be ignored.
The kinetic picture which emerges from the experi-
ments involving intermolecular competition reactions
with THF and TDF is (i) kTHF ) (3.75 ( 0.30) × 106 M-1
s-1 (Table 1) and kTHF/kD ) (0.929 ( 0.093) × 10-2 (Table
2), from which:
At 80 °C (a temperature at which a great deal of
synthetic chemistry is carried out with tin hydride) we
estimate that kH ≈ 1.0 × 109 M-1 s-1
.
Exp er im en ta l Section
Ma ter ia ls. The following compounds were purchased from
Aldrich and were generally used without further purification:
(34) In this connection, we note that some years ago we attempted
to measure kD using o-(3-phenylpropyl)iodobenzene as the source of
an aryl radical which could rearrange by an intramolecular abstraction
of a benzylic hydrogen atom (eq i). In isooctane at 25 °C an LFP study
gave kexptl ) (1.4 ( 0.2) × 107 s-1 for the growth of the rearranged
radical but this rate constant is not kr since it also included contribu-
tions from any first- and pseudo-first-order processes which the radical
undergoes, including H-atom abstraction from the solvent, rate
constant, kOCT. A photoinitiated radical chain reaction of this iodide
with n-Bu3SnD in isooctane at 25 °C followed by 2H NMR analysis
showed only an aromatic and a benzylic deuterium. From the ratio of
these two products at known concentrations of n-Bu3SnD we obtained
kD/kr ) 35 ( 4 M-1, and hence kD e 4.9 × 108 M-1 s-1 which implies
that this ortho-substituted phenyl radical is somewhat less reactive
than the 4-CH3C6H4• and 4--O2CC6H4• radicals. The rough magnitude
of the correction to kr for reaction with the solvent was determined by
photolyzing PhI (0.02 M) and n-Bu3SnD (0.1 M) in isooctane and
analyzing for C6H6 and C6H5D. After applying all corrections we found
that 4.8% of the benzene was due to phenyl radical attack on the
isooctane. Thus, for the phenyl radical itself, kOCT[OCT] ) 0.048 kD
[n-Bu3SnD] ) 0.0048 k D ) 4.8 × 10-3 × 5.2 × 108 ) 2.5 × 106 s-1, and
kD ) (4.04 ( 0.73) × 108 M-1 s-1
and (ii) kTDF ) (1.32 ( 0.03) × 106 M-1 s-1 (Table 1) and
k
TDF/kH ) (0.202 ( 0.048) × 10-2 (Table 3), from which:
kH ) (6.53 ( 1.70) × 108 M-1 s-1
where all errors have been expressed as 2σ but include
only random errors. These data give a deuterium kinetic
isotope effect, DKIE, kH/kD ) 1.6, which is somewhat
larger than previously reported values for aryl radicals,
viz., 1.432 and 1.3.33
hence kOCT ≈ 4 × 105 M-1 s-1
.
The kinetic picture which emerges from the intermo-
•
kr
•
lecular competition reactions with MMA is (iii) kMMA
)
(CH2)3
(CH2)2CH
(1)
1.8 × 108 M-1 s-1 12 and kD/kMMA ) 3.5 ( 0.4 (vide supra
and Table 4) from which:
(35) Serious attempts were made to improve the accuracy of kH by
measuring kH/kD in direct competition experiments using n-Bu3SnH/
n-Bu3SnD mixtures and iodobenzene or bromobenzene and analyzing
for the C6H6/C6H5D ratios. To avoid complications due to H-atom
abstraction from the solvent the reactions were photo-initiated using
a deficiency of the halobenzene in mixtures of (neat) n-Bu3SnH and
(neat) n-Bu3SnD (ca. 1:10:10 mole ratio). After applying all necessary
corrections (see Experimental Section) we discovered to our surprise
that the measured kH/kD ) 1.0 (see Table 5 in the Supplementary
Material for typical results with iodobenzene). This was demonstrated
to be a “peculiarity” of the phenyl radical since using haloalkanes which
would yield primary alkyl radicals we obtained kH/kD (1-pentyl) ) 2.1
( 0.1 by direct competition in neat n-Bu3SnH/n-Bu3SnD, and kH/kD
(5-hexenyl) ) 2.3 ( 0.1 by inter/intramolecular competition, both
reactions being run in decane at 25 °C. The literature gives kH/kD (1-
butyl) ) 2.3 by LFP.6 Our initial “explanation” for these results with
the phenyl halides was that H(D)-atom abstraction by phenyl radicals
in neat n-Bu3SnH/n-Bu3SnD was a diffusion-controlled process. Un-
fortunately, this simple “explanation” cannot be correct because the
viscosity of neat n-Bu3SnH is only 1.5 cP at 20 °C36 and a diffusion-
controlled radical-molecule reaction in a solvent of this viscosity would
kD ) (6.3 ( 0.7) × 108 M-1 s-1
and (iv) kH/kMMA ) 5.0 ( 0.6 (vide supra and Table 4)
from which:
kH ) (9.0 ( 1.1) × 108 M-1 s-1
and hence kH/kD ) 1.4.
The kD values derived from the THF experiments, viz.,
(4.0 ( 0.7) × 108 M-1 s-1, and from the MMA experi-
ments, viz., (6.3 ( 0.7) × 108 M-1 s-1 differ by slightly
(27) Barclay, L. R. C.; Griller, D.; Ingold, K. U. J . Am. Chem. Soc.
1974, 96, 3011-3012. Brunton, G.; Griller, D.; Barclay, L. R. C. J . Am.
Chem. Soc. 1976, 98, 6803-6811. Brunton, G.; Gray, J . A.; Griller, D.;
Barclay, L. R. C.; Ingold, K. U. J . Am. Chem. Soc. 1978, 100, 4197-
4200.
(28) Porter, G.; Ward, B. Proc. R. Soc., London, Ser. A 1965, 287,
457-470. Bennet, J .; Mile, R.; Thomas, A. Proc. R. Soc., London, Ser.
A 1966, 293, 246-258.
(29) Mertens, R.; von Sonntag, C. Angew. Chem., Int. Ed. Engl. 1994,
33, 1262-1264. von Sonntag, C. Private communication.
(30) Except when the radical center is protected by two bulky ortho-
substituents.27
37
have a rate constant of about 4 × 109 M-1 s-1 which is at least five
times as large as our measured value for kH. The reason for the peculiar
behavior of phenyl radicals in neat n-Bu3SnH/n-Bu3SnD mixtures is
not clear. However, we tentatively suggest that we were not irradiating
the phenyl halide per se, but rather we were irradiating phenyl halide/
tin hydride charge transfer complexes and that this process gave
benzene and tin halide directly and immediately with no possibility of
H/D discrimination by free phenyl radicals.
(31) These two assumptions receive support from the agreement
between the values of kH/kMMA and kD/kMMA calculated from the data
in Table 4 via eqs V and VI, viz., 5.0 ( 0.6 and 3.5 ( 0.4, respectively
(vide infra).
(32) Strong, H. L.; Brownawell, M. L.; San Filippo, J ., J r. J . Am.
Chem. Soc. 1983, 107, 6526-6528.
(33) Abeywickrema, A. N.; Beckwith, A. L. J . J . Chem. Soc., Chem.
Commun. 1986, 464-465.
(36) We thank Witco GmbH, Bergkamen, Germany, for providing
this data and an unknown referee for encouraging us to pursue the
origin of the peculiarity of the neat n-Bu3SnH/n-Bu3SnD system.
(37) (kdiff/M-1 s-1) ) 2 × 107 T/η with η in cP, see: Turro, N. J .
Modern Molecular Photochemistry; Benjamin/Cummings Publ. Co.:
Menlo Park, California, 1978; p 314.

Notes
J . Org. Chem., Vol. 61, No. 2, 1996 809
tri-n-butyltin hydride, (97% pure), tri-n-butyltin deuteride (96.0%
pure, 96.6 atom % D), 4-iodobenzoic acid, 4-iodotoluene (colorless
crystals, 99.8% pure by GC), sodium salt of 4-styrene sulfonic
acid, tetrahydrofuran (THF purified by distillation from Na/K
alloy), perdeuterotetrahydrofuran (99.5+ atom % D) and methyl
methacrylate (MMA purified by distillation on a Dufton column,
the 99.9-100.0 °C fraction was 99.8% pure by GC and was stored
in a brown bottle at -20 °C).
(n-Bu3SnD)actual )0.960 × 0.972 (n-Bu3SnD)nominal
) 0.933 (n-Bu3SnD)nominal
Furthermore, because 2.8% of “extra” C6H6 is formed in the
C6H5•/n-Bu3SnD reaction, the actual yield of C6H6 must be
corrected in order to obtain the quantity of C6H6 formed by
phenyl radical abstraction from THF:
P u r ifica tion of Iod oben zen e. Iodobenzene (10.0 g, Aldrich)
in ether (20 mL) was treated with an aqueous solution of sodium
thiosulfate (0.10 M, 30 mL). The ether solution was separated
and passed through a column of basic alumina. The elution was
completed using an additional quantity of ether (20 mL). The
ether was removed at room temperature on a rotary evaporator,
and the residue was fractionally distilled under reduced pres-
sure. Pure, colorless iodobenzene was collected at 74 °C/15-20
Torr (8.9 g, 89% yield). It was stored in the dark over copper
wire (which had previously been treated with a strong aqueous
nitric acid solution, rinsed, and dried).
(C6H6)THF ) (C6H6)actual - 0.028 (C6H5D)actual
We have employed the above corrections to calculate the kTHF
kD ratios given in Table 2. That is,
/
kTHF (C6H6)actual - 0.028 (C6H5D)actual
)
×
kD
(C6H5D)actual
0.933 (n-Bu3SnD)nominal
La ser F la sh P h otolysis (LF P ). The general procedures
have been adequately described in numerous other publications
from this laboratory.6-8,12-14,21
[THF]
Rea ction of Iod oben zen e w ith Tr i-n -bu tyltin Hyd r id e
in P er d eu ter otetr a h yd r ofu r a n . Experiments were conducted
in an identical manner to those employed to study the n-Bu3-
SnD/THF competition. Since the TDF contained 99.5 + atom
% D no correction was considered to be necessary for C6H6
formed in the C6H5•/TDF reaction. Furthermore, because the
C6H6 formed in the C6H5•/n-Bu3SnD reaction appears to come
entirely from a n-Bu3SnH impurity (vide supra) no correction
would appear to be necessary for C6H6 formed by C6H5• abstrac-
tion from the butyl groups of n-Bu3SnX (X ) H, D, I). Thus,
the only correction required is to the quantity of n-Bu3SnH
employed because this material was only 97% pure:
Rea ction of Iod oben zen e w ith Tr i-n -bu tyltin Deu ter id e
in Tetr a h yd r ofu r a n . A typical experiment was carried out as
follows. Into a rubber septum-capped Pyrex tube were added,
via syringe, iodobenzene (2.3 mg, 11.3 µmoles), tetrahydrofuran
(THF, 1 mL, 0.880 g, 12.2 mmol) and n-Bu3SnD (33.9 mg, 116
µmol). This mixture was cooled to -78 °C (acetone, dry ice) and
purged with a slow, steady stream of nitrogen for 15 min. The
reaction mixture was then warmed to room temperature and
was photolyzed for 15 min in a Rayonet reactor (at 31 °C) with
λ > 300 nm light. A sample of the reaction mixture (5 µL) was
then immediately injected onto a Hewlett Packard HP-1 column
of a GC/MS (Hewlett-Packard 5890 gas chromatograph combined
with a 5970 series mass selective detector) and the relative
abundance of the ion peaks at m/z ) 76, 77, 78, 79, and 80 was
determined by selective ion monitoring. Under these conditions
all the iodobenzene was consumed.
kTDF (C6H5D)actual 0.97(n-Bu3SnH)nominal
)
×
kH
[TDF]
(C6H6)actual
The corrections to the apparent C6H6/C6H5D ratios (i.e., m/z
78/79 ratios) for natural abundance 13C and 2H and for the (M
- 1)+ peak were determined on the same GC/MS instrument
using 5 µL aliquots of solutions of benzene (2.0 mg, 3.5 mg and
5.0 mg) in THF (1 mL). The intensities of the m/z ) 77, 78, and
79 peaks were corrected for the background signal, this correc-
tion being based on a 2 min portion of the GC/MS spectrum
devoid of any true signal. As a result of the three independent
experiments with C6H6 the mean contributions to m/z ) 78 and
79 are 93.55 ( 0.06% and 6.448 ( 0.056%, respectively, where
the errors are 2σ. For C6H5D the corresponding figures for m/z
) 78 and 79 are 24.60 ( 0.19% and 75.40 ( 0.19%, respectively.
This equation was used to derive the final column in Table 3.
Rea ction of p-Iod otolu en e w ith Tr i-n -bu tyltin Hyd r id e
(Deu ter id e) a n d Meth yl Meth a cr yla te in Cycloh exa n e. In
a typical experiment, n-Bu3SnH(D) (50 µL, 185 µmol) and di-
tert-butyl hyponitrite (0.6 mg, 4 µmol) were added to a solution
of p-iodotoluene (∼40 mg, ∼200 µmol) and MMA (50 µL, 425
µmol) in cyclohexane (200 µL). The mixture was then freeze-
pump-thaw degassed and flame-sealed in vacuo in an 800 µl
Carey tube. After standing for ∼2 h at ambient temperature,
the ampoule was opened and the contents were added to 400
µL of argon-purged CCl4. The solution was warmed to 43 °C
for 0.5 h and then analyzed by capillary GC (Varian 6500
instruments/25 m BP1 column) using the pure addends for
reference.38 The mean concentration of n-Bu3SnH(D) (92- 95%
of the initial value) was calculated from the ratio n-Bu3Snl/n-
Bu3SnCl (GC) in the quenched mixture. The results are
tabulated in Table 4.
The observed abundances of the peaks at m/z ) 78 and 79
were proportioned into the contributions actually due to C6H6
and to C6H5D by solving the two following equations:
m/z (78)intensity ) 93.55 (C6H6)actual + 24.60 (C6H5D)actual
m/z (79)intensity ) 6.45 (C6H6)actual + 75.40 (C6H5D)actual
Ack n ow led gm en t. We thank Mr. A. Wallner for
assistance in conducting the experiments described in
footnote 34. We also thank Professor C. von Sonntag
for communicating some of his results to us prior to
publication and the Association for International Cancer
Research and the National Foundation for Cancer
Research for partial support of this work.
Although these two equations yield the (C6H6)actual/(C6H5D)actual
ratios there are two additional corrections which must be applied
before the true rate constant ratio, kTHF/kD, can be calculated.
The first involves correcting the nominal quantity of n-Bu3SnD
used in each experiment for the fact that this material was only
96.0% pure. The second arises from potential sources of C6H6
•
in the C6H5 /n-Bu3SnD reaction. There are two potential sources
Su p p or tin g In for m a tion Ava ila ble: Table 5 with prod-
uct data for the reaction of iodobenzene with (neat) n-Bu3SnH/
(neat) n-Bu3SnD mixtures (1 page). 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 informa-
tion.
of C6H6 in this reaction: (i) a n-Bu3SnH impurity, (ii) hydrogen
•
abstraction from the butyl groups of n-Bu3SnD by the C6H5
radical. With regard to (i) the claimed isotopic purity was 96.6
atom % deuterium which could imply as much as 3.4% n-Bu3-
SnH as an impurity. To check on the real relative yield of C6H6
in the C6H5•/n-Bu3SnD reaction a solution of phenyl iodide in
n-Bu3SnD (ca. 1:10 mole ratio) was photolyzed (4 min, λ > 300
J O951245A
nm). Product analysis showed that (C6H6)actual
) 0.028
[(C6H5D)actual + (C6H6)actual] which probably means that there is
about 2.8% n-Bu3SnH present in the n-Bu3SnD (rather than
3.4%). Thus,
(38) Bordwell, F. G.; Carlson, M. W. J . Am. Chem. Soc. 1970, 92,
3370-3377.