Evidence for divalent iodine (9-I-2) radical intermediates in the thermolysis of (tert-butylperoxy)iodanes. An unusually efficient deiodination of o-iodocumyl alcohols by cyclohexyl radicals

Article abstract of DOI:10.1021/ja9636404

1-(tert-Butylperoxy)-3,3-dimethyl-1H-1,2-benziodoxoles (2a and 2b) and 1-(tert-butylperoxy)-3,3-bis(trifluoromethyl)-5-methyl-1H-1,2-benziodo xole (2c) were prepared from chloroiodanes 1a-c and tert-butylhydroperoxide in the presence of potassium tert-butoxide in tetrahydrofuran. Products, kinetic data for the decomposition of 2 in cyclohexane, benzene, toluene, and acetonitrile (E(a) = 31.0 ± 1.0 kcal/mol, log A = 17.0 ± 0.5; 35-70 °C), and the increased rate of decomposition of 2c in benzene-d₆ in the presence of a magnetic field (7 T) indicate that homolytic cleavage of the I-O bond in 2 with the formation of iodanyl (9-I-2) and tert-butylperoxyl radicals is the primary decomposition step. The nearly quantitative formation of iodocyclohexane during the decomposition of 2c in cyclohexane is due to the unexpected reaction of cyclohexyl radicals with 2-(2-iodo-5-methylphenyl)-1,1,1,3,3,3-hexafluoro-2-propanol, a primary decomposition product of 2c. The results of a separate study of the deiodination of o-iodocumyl alcohols (3) by cyclohexyl radicals are consistent with an S(H)2 type mechanism.

Full text of DOI:10.1021/ja9636404

2628
J. Am. Chem. Soc. 1997, 119, 2628-2632
Evidence for Divalent Iodine (9-I-2) Radical Intermediates in
the Thermolysis of (tert-Butylperoxy)iodanes. An Unusually
Efficient Deiodination of o-Iodocumyl Alcohols by Cyclohexyl
Radicals1
Darko Dolenc and Bozˇo Plesnicˇar*
Contribution from the Department of Chemistry, UniVersity of Ljubljana, P.O. Box 537,
1000 Ljubljana, SloVenia
ReceiVed October 18, 1996. ReVised Manuscript ReceiVed January 21, 1997X
Abstract: 1-(tert-Butylperoxy)-3,3-dimethyl-1H-1,2-benziodoxoles (2a and 2b) and 1-(tert-butylperoxy)-3,3-bis-
(trifluoromethyl)-5-methyl-1H-1,2-benziodoxole (2c) were prepared from chloroiodanes 1a-c and tert-butyl
hydroperoxide in the presence of potassium tert-butoxide in tetrahydrofuran. Products, kinetic data for the
decomposition of 2 in cyclohexane, benzene, toluene, and acetonitrile (Ea ) 31.0 ( 1.0 kcal/mol, log A ) 17.0 (
0.5; 35-70 °C), and the increased rate of decomposition of 2c in benzene-d6 in the presence of a magnetic field (7
T) indicate that homolytic cleavage of the I-O bond in 2 with the formation of iodanyl (9-I-2) and tert-butylperoxyl
radicals is the primary decomposition step. The nearly quantitative formation of iodocyclohexane during the
decomposition of 2c in cyclohexane is due to the unexpected reaction of cyclohexyl radicals with 2-(2-iodo-5-
methylphenyl)-1,1,1,3,3,3-hexafluoro-2-propanol, a primary decomposition product of 2c. The results of a separate
study of the deiodination of o-iodocumyl alcohols (3) by cyclohexyl radicals are consistent with an SH2 type
mechanism.
Introduction
benziodoxol-3(1H)-one, by Lewis acid-catalyzed (BF3-Et2O)
exchange of the hydroxy group of 1-hydroxy-1,2-benziodoxol-
3(1H)-one with a tert-butylperoxy group.7-9 They also provided
evidence for the involvement of iodanyl radicals in its
decomposition.7b
As a part of our continuing interest in divalent iodine radicals
and in systems capable of providing such intermediates, we
report a convenient synthesis of other members of the family
of relatively stable (alkylperoxy)iodanes. The decomposition
of these peroxides was studied. The results of a study of an
unusually effective deiodination of o-iodocumyl alcohols by
cyclohexyl radicals are also reported.
Although diaryl- and dialkyliodonium ions are well-estab-
lished,2 relatively little is known about divalent iodine (9-I-2)
radicals.3 Such intermediates have been proposed in a number
of synthetically and theoretically4 important reactions, but no
direct ESR evidence for their existence has as yet been reported.
Strong support for a divalent iodine radical was reported by
Tanner et al. in a study of the kinetics for the transfer of the
iodine atom from an aryl iodide to an aryl radical.5 Most
recently, Scaiano et al. reported evidence for the intermediacy
of a cyclic hypervalent iodine radical in the laser flash and laser-
drop photolysis of 1,5-diiodo-1,5-diphenylpentane.6a,6
Results and Discussion
Ochiai et al. have recently prepared and characterized the
first stable (alkylperoxy)iodane, i.e., 1-(tert-butylperoxy)-1,2-
(tert-Butylperoxy)iodanes. Treatment of 1-chloro-3,3-dim-
ethyl-1H-1,2-benziodoxoles (1a and 1b) or 1-chloro-3,3-bis-
(trifluoromethyl)-5-methyl-1H-1,2-benziodoxole (1c)4d with tert-
butyl hydroperoxide in the presence of potassium tert-butoxide
X Abstract published in AdVance ACS Abstracts, March 1, 1997.
(1) Dolenc, D.; Plesnicˇar, B. Presented in part at the 7th International
Symposium on Organic Free Radicals; Bardolino (Lake Garda), Italy, June
1996. Book of Abstracts, p 71.
(2) (a) Stang, P. J. Angew. Chem., Int. Ed. Engl. 1992, 31, 274. Stang,
P. J.; Zhdankin, V. V. Chem. ReV. 1996, 96, 1123. (b) Koser, G. F. In The
Chemistry of Functional Groups, Suppl. D; Patai, S., Rappoport, Z., Eds.;
Wiley: New York, 1983; Chapter 18. (c) Moriarty, R. M.; Vaid, R. K.
Synthesis 1990, 431. (d) Varvoglis, A. The Chemistry of Polycoordinated
Iodine; VCH: Weinheim, 1992.
(3) For a nomenclature, see: Perkins, C. W.; Martin, J. C.; Arduengo,
A. J.; Lau, W.; Alegria, A.; Kochi, J. K. J. Am. Chem. Soc. 1980, 102,
7753. Martin, J. C. Science 1983, 221, 509.
(6) (a) Banks, J. T.; Garcia, H.; Miranda, M. A.; Perez-Prieto, J.; Scaiano,
J. C. J. Am. Chem. Soc. 1995, 117, 5049. (b) For a previous attempt to
detect such an intermediate, see: Chateauneuf, J.; Lusztyk, J.; Ingold, K.
U. J. Org. Chem. 1990, 55, 1061.
(7) (a) Ochiai, M.; Ito, T.; Masaki, Y.; Shiro, M. J. J. Am. Chem. Soc.
1992, 114, 6269. (b) Ochiai, M.; Ito, T.; Takahashi, H.; Nakanishi, A.;
Toyonari, M.; Sueda, T.; Goto, S.; Shiro, M. J. Am. Chem. Soc. 1996, 118,
7716. (c) See also: Ochiai, M.; Ito, T.; Shiro, M. J. Chem. Soc., Chem.
Commun. 1993, 218.
(4) For thermally or photoinitiated chlorinations by (dichloro)iodoben-
zenes, see: (a) Russell, G. A. J. Am. Chem. Soc. 1958, 80, 4987. (b) Banks,
D. F.; Huyser, E. S.; Kleinburg, J. J. Org. Chem. 1964, 29, 3692 (c)
Chlorinations by [chloro(tert-butoxy)iodo]benzene: Tanner, D. D.; Gidley,
G. C. J. Am. Chem. Soc. 1968, 90, 808. (d) Chlorinations by the Martin’s
chloroiodanes: Amey, R. L.; Martin, J. C. J. Am. Chem. Soc. 1979, 101,
3060. J. Org. Chem. 1979, 44, 1779. (e) For the Breslow’s template-directed
chlorination of steroids, see: Breslow, R.; Corcoran, R.; Dale, J. A. Liu,
S.; Kalicky, P. J. Am. Chem. Soc. 1974, 96, 1973. White, P.; Breslow, R.
J. Am. Chem. Soc. 1990, 112, 6842, and references therein. (f) For
anchimeric acceleration of ortho-iodo-substituded peroxides homolysis, see,
for example: Martin, J. C. ACS Symp. Ser. 1978, 69, 71, and references
cited therein.
(8) Moss et al. have prepared 1-(tert-butylperoxy)-1,2-benziodoxol-3(1H)-
one by a ligand exchange of a labile (phosphoryloxy)iodane (prepared in
situ by the reaction of 1-chloro-1,2-benziodoxol-3(1H)-one with silver
diphenyl phosphate in DMSO) with tert-butyl hydroperoxide. Moss, R. A.;
Zhang, H. J. Am. Chem. Soc. 1994, 116, 4471.
(9) [Bis(tert-butylperoxy)iodo]benzene was proposed as an intermediate
in the low-temperature reaction of iodosobenzene with tert-butyl hydrop-
eroxide, and its decomposition to iodobenzene and tert-butylperoxyl radicals
has been reported. When the reaction was run in diethyl ether, good yields
of tert-butylperoxy acetal were isolated. Milas, N. A.; Plesnicˇar, B. J. Am.
Chem. Soc. 1968, 90, 4450. See, also: Traylor, T. G.; Xu, F. J. Am. Chem.
Soc. 1987, 109, 6201. (b) [Bis(benzoylperoxy)iodo]benzenes were isolated
as amorphous, labile (rather hazardous) materials: Plesnicˇar, B.; Russell,
G. A. Angew. Chem., Int. Ed. Engl. 1970, 9, 797. Plesnicˇar, B. J. Org.
Chem. 1975, 40, 3267.
(5) Tanner, D. D.; Reed, D. W.; Setiloane, B. P. J. Am. Chem. Soc. 1982,
104, 3917, and references cited therein.
S0002-7863(96)03640-2 CCC: $14.00 © 1997 American Chemical Society

DiValent Iodine (9-I-2) Radical Intermediates
J. Am. Chem. Soc., Vol. 119, No. 11, 1997 2629
Scheme 1
Table 1. Kinetic and Activation Parameters for the Decomposition
of (tert-Butylperoxy)iodanes in Various Solventsa
(alkylperoxy)-
iodane
105k[55 °C],
Ea,
solvent
s-1
kcal mol-1
log A
2a
2a
2a
2a
2b
2c
2cd
c-C6D12
20.7
18.3
107
44.7
48.4
1.42c
1.27c
31.4 ( 1.0 17.3 ( 0.1
31.9 ( 1.0b 17.5 ( 0.1
30.2 ( 1.0 17.2 ( 0.1
31.7 ( 1.0 17.8 ( 0.1
30.8 ( 1.7 17.3 ( 0.2
31.4 ( 1.0 15.2 ( 0.1
30.1 ( 2.0 14.3 ( 0.1
b
c-C6D12
CD3CN
C6D6
C6D6
C6D6
C6D6
a Kinetic measurements were carried out in sealed NMR tubes in
thermostated NMR probe (Bruker AVANCE 300 DPX, B ) 7 T) unless
stated otherwise. The disappearance of H7 absorption in the starting
compound was in all cases measured by 1H NMR. b One equiv of 2,6-
di-tert-butyl-4-methylphenol was added. c At 70 °C. d Water-bath ther-
mostat.
Figure 1. Arrhenius plots for the decomposition of 2a in various
solvents: A, CD3CN; B, C6D6; C, Cyclohexane-d12; and D, cyclohex-
ane-d12 (with added 2,6-di-tert-butyl-2-methylphenol).
Table 2. Products of the Thermal Decomposition of 2a and 2c in
Various Solventsa,b
in tetrahydrofuran at 0 °C gave 1-tert-butylperoxy-1H-1,2-
benziodoxoles (2a-c) in yields ranging from 50 to 60% (after
crystallization). (Scheme 1). All (alkylperoxy)iodanes were
fully characterized by elemental and spectral analysis. The pure
crystalline solids have an indefinite shelf life when kept in the
dark.
The decomposition of the (alkylperoxy)iodanes in four
different solvents (cyclohexane, toluene, benzene, acetonitrile)
typically in the temperature range from 35 to 70 °C was studied,
by following the decay of the H7 (1H NMR) absorptions of 2
as a function of time. The disappearance of the (alkylperoxy)-
iodanes obeyed first-order kinetics over several half-lives in all
solvents investigated. Kinetic and activation parameters for the
decomposition of 2 are collected in Table 1 (Figure 1).
Solvent effects on the decomposition are relatively small; i.e.,
comparable rate constants (within a factor of 4-5) were obtained
for the decomposition of the (alkylperoxy)iodanes in all solvents
investigated. A study of the decomposition of 2a in the presence
of 2,6-di-tert-butyl-4-methylphenol showed a relatively small
reduction in rate and nearly the same activation parameters.
Although the effect of substituents on the decomposition has
not been studied systematically, it is interesting to note that
electron-donating p-OCH3 group somewhat accelerates decom-
position of 2a.
The decomposition products of peroxyiodanes are collected
in Table 2. The observed products and activation parameters
are indicative of homolytic cleavage of the I-O bond to form
iodanyl (9-I-2) and tert-butylperoxyl radicals in the primary
decomposition step (Scheme 2). The somewhat faster decom-
position of 2c (by a factor 1.18 ( 0.05, from the Arrhenius
plot) in the cavity of the NMR instrument (7 T; singlet-triplet
rephasing) compared to the decomposition run in a thermostated
bath indicates that the iodanyl radical is stable enough to
undergo geminate recombination with the tert-butylperoxyl
radical in the solvent cage to regenerate the original precursor.10
Two surprising observations regarding the decomposition
products of the (alkylperoxy)iodanes in cyclohexane deserve
comment. In the decomposition of the gem-dimethyl derivative
2a, large amounts of tert-butyl cyclohexyl peroxide were found
a Reactions were run in ampoules (0.05 mmol of iodane, 1 mL of
solvent, Ar atmosphere) in a boiling hexane (2a) or cyclohexane bath
(2c). b Yields of products in mol %, based on the initial amount of
iodane. c Yields in the presence of 2 equiv of TEMPO. 3, 4, 5a: R1
) CH3, R2 ) H; 3, 4, 5c: R1 ) CF3, R2 ) CH3. Besides the products
quoted in the table, considerable amounts of tert-butyl alcohol and
acetone were produced as well as 20-40% of alcohol and ketone,
derived from solvent (in toluene and cyclohexane). The decomposition
of 2a in benzene gave mainly ketone 4a (56%) and alcohol 3a (26%).
among the products (40%). Although this peroxide might result
from SH2 attack of the cyclohexyl radical on the oxygen atom
next to iodine in 2a, we believe that it is formed predominantly
by coupling of the tert-butylperoxyl and the cyclohexyl radicals.
In the presence of TEMPO (2,2,6,6-tetramethyl-1-piperidiny-
loxyl), an efficient radical scavenger, besides the iodo alcohol
3a, tert-butyl hydroperoxide, was the main decomposition
product (41%). Even larger amounts of dialkyl peroxide, i.e.,
benzyl tert-butyl peroxide, were found when toluene was the
solvent.
On the other hand, the decomposition of the trifluoromethyl
substituted (alkylperoxy)iodane 2c in cyclohexane gave only

2630 J. Am. Chem. Soc., Vol. 119, No. 11, 1997
Dolenc and Plesnicˇar
Scheme 2
Figure 2. A Hammett plot for the deiodination of substituted
2-iodophenyl-2-propanols by cyclohexyl radicals.
A Hammett plot (σ) (Figure 2) gave a fair correlation with F )
1.10 (r ) 0.97). (2) Experiments with the OD deuterated
derivative of 3a gave unlabeled iodocyclohexane and cumyl
alcohol. At the same time, the reaction of unlabeled iodoalcohol
with labeled cyclohexyl radicals in cyclohexane-d12 resulted in
incorporation of deuterium into the ipso position of the product.
(3) The absence of the cyclohexyl substituted benzylic alcohols
among the reaction products and the inability of benzyl radicals
(in toluene) to perform the same reaction seem to rule out the
SRN1 mechanism.11 (4) The trifluoromethyl substituted iodo
alcohol 3c is more reactive than the gem-dimethyl derivative
3a by a factor of 46 ( 3. (5) Phenyl and n-heptyl radicals (57
and 25% conversions, respectively, in benzene) and tert-butyl
radicals (25%, in isobutane) were found to be less efficient than
cyclohexyl radicals (in cyclohexane) in these reactions. How-
ever, the reaction conditions in these cases were not completely
comparable, so that the differences in reactivity were difficult
to evaluate.
Table 3. Reactions of Substituted Bromo- and Iodobenzenes
(X-Ar-I) with Di-tert-butylperoxyoxalate (DBPO) in
Cyclohexanea,b
c
Based on the above evidence, we propose an SH2 type
mechanism (Scheme 3) involving electron transfer from the
cyclohexyl radical to the substituted iodobenzene, via a transition
state resembling an iodobenzene radical anion-cyclohexyl cation
pair, to give an aryl radical and iodocyclohexane in an essentially
concerted process (pathway A). The possibility of the involve-
ment of a discrete ion pair intermediate cannot be excluded.
An alternative mechanism (pathway B) involving a divalent
iodine (9-I-2) radical species (Ar-I-R), already proposed as
intermediates in thermoneutral deiodinations of aryl iodides by
phenyl radicals,5 and subsequent decomposition to an aryl radical
and iodocyclohexane also cannot be ruled out. However, on
the basis of the somewhat greater sensitivity of reactions with
cyclohexyl radicals to electronic effects, compared to that with
phenyl radicals (F ) 0.474, r ) 0.820),5 this mechanistic option
appears to be less likely.
a Iodobenzene (0.1 mmol) and 0.1 mmol of DBPO in 4 mL of
cyclohexane (boiling cyclohexane, 2 h, argon atmosphere). b Compo-
sition of the reaction mixture in mol % per mol of the starting
halobenzene. c Bromocyclohexane.
small amounts of the iodo alcohol 3c and nearly quantitative
yields of the deiodinated cumyl alcohol 5c and iodocyclohexane.
Deiodination of o-Iodocumyl Alcohols by Cyclohexyl
Radicals. Since we suspected that cyclohexyl radicals initiated
the abstraction of iodine from the iodo alcohol, this reaction
was studied separately. When iodo alcohols 3a and 3c were
allowed to react with cyclohexyl radicals, generated by the
decomposition of di-tert-butyl peroxyoxalate (DBPO) in cy-
clohexane under argon, quantitative yields of iodocyclohexane
and the corresponding benzylic alcohols were formed (Table
3). Considerably lower yields of iodocyclohexane were obtained
with other electronegatively 2-substituted iodobenzenes.
Since the direct abstraction of the iodine atom by the
cyclohexyl radical seemed unlikely (endothermic by 11 ( 1
kcal/mol), further experiments to elucidate the mechanism of
this reaction were performed. The results of these studies can
be summarized as follows: (1) The relative reactivities,
determined in competition experiments of meta- and para-
substituted 3a, showed that electron-withdrawing groups ac-
celerate deiodination while electron-releasing groups retard it.
The question arises as to why o-iodocumyl alcohols are so
much more reactive toward deiodination than other electrone-
gatively ortho-substituted iodobenzenes studied. We believe
that the enhanced reactivity is due to intramolecular hydrogen
(10) Turro, N. J.; Kraeutler, B. Acc. Chem. Res. 1980, 13, 369.
Buchachenko, A. L.; Khudyakov, I. V. Acc. Chem. Res. 1991, 24, 177.
(11) (a) Russell, G. A.; Danen, W. C. J. Am. Chem. Soc. 1966, 88, 5663.
(b) Kornblum, N.; Michel, R. E.; Kerber, R. C. J. Am. Chem. Soc. 1966,
88, 5662. (c) Bunnett, J. F. Acc. Chem. Res. 1978, 11, 413; 1992, 25, 2. (d)
Russell, G. A. AdV. Phys. Org. Chem. 1987, 23, 271.

DiValent Iodine (9-I-2) Radical Intermediates
J. Am. Chem. Soc., Vol. 119, No. 11, 1997 2631
Scheme 3
Therefore, the difference in reactivity is not caused by the
difference in the electron donor ability of these two radicals
but is most likely due to the difference in bond dissociation
energies (BDE) of the carbon-iodine bonds in the respective
iodoalkanes (∆BDE (cy-C6H11I - PhCH2I) ) 11.5 ( 1.0 kcal/
mol).17
Preliminary results show that cyclohexyl radicals are also
capable of debromination of o-bromohexafluorocumyl alcohol
(see Table 3).
Further work on these interesting reactions (also involving
still some other alkyl radicals) is in progress, and the results of
these studies will be reported in due course.
Experimental Section
General Methods. NMR spectra were recorded on a Bruker Avance
300 DPX spectrometer. Chemical shifts are reported relative to internal
standards Me4Si or CCl3F. IR spectra were measured on a Perkin-
Elmer FTIR 2000 instrument. Melting points are uncorrected. Mi-
croanalyses were performed on a Perkin-Elmer 2400 CHN Analyzer.
GC analyses were carried out on a Hewlett-Packard 6890 gas
chromatograph or GC-MS combination, using 30 m capillary HP5 or
Innowax columns. Mass spectra and high resolution mass measure-
ments were performed on a VG-Analytical Autospec EQ instrument.
Materials. Solvents for the decomposition and iodine atom abstrac-
tion studies were column distilled before use. THF was freshly distilled
from LiAlH4 in an argon atmosphere. In kinetic measurements,
deuterated NMR solvents (Fluka), benzene-d6 (>99.95%D), acetonitrile-
d3 (>99.8%D) and cyclohexane-d12 (∼99.5%D), were used as received.
tert-Butyl hydroperoxyde was purified by the literature method,18 di-
tert-butyl peroxyoxalate (DBPO),19 dioctanoyl peroxide,20 1-chloro-
3,3-dimethyl-1H-1,2-benziodoxole (1a), and 1-chloro-5-methyl-3,3-
bis(trifluoromethyl)-1H-1,2-benziodoxole (2c)4d,15 as well as alcohols
3a and 3c4d were prepared according to the literature procedures.
1-tert-Butylperoxy-1H-1,2-benziodoxoles (2). A Typical Pro-
cedure: 1-tert-Butylperoxy-3,3-dimethyl-1H-1,2-benziodoxole (2a).
A mixture of t-BuOK (5.0 mmol) and t-BuOOH (6.0 mmol) in 10 mL
of THF was slowly added to a stirred THF solution of 1a in an ice
bath. The reaction mixture was diluted with diethyl ether-pentane
(1:1), washed three times with water, and dried over Na2SO4, the solvent
evaporated, and 1.51 g of the crude product was obtained. Recrystal-
lization from hexane yielded 0.88 g (50%) of 2a, mp 92-95 °C. 1H
NMR (CDCl3) δ: 1.29 (s, 9H, t-Bu); 1.48 (s, 6H, Me3); 7.18 (dd, J )
1.5, 7.4 Hz, 1H, H4); 7.44 (dt, J ) 1.3, 7.3 Hz, 1H, H5); 7.50 (dt, J )
1.6, 7.6 Hz, 1H, H6); 7.94 (dd, J ) 1.2, 8.0 Hz, 1H, H7).
1-tert-Butylperoxy-5-methoxy-3,3-dimethyl-1H-1,2-benziodox-
ole (2b). Mp 65-68 °C. 1H NMR (CDCl3) δ: 1.26 (s, 9H, t-Bu);
1.46 (s, 6H, Me3); 3.85 (s, 3H, OMe); 6.72 (d, J ) 2.7 Hz, 1H, H4);
7.05 (dd, J ) 8.9, 2.7 Hz, 1H, H6); 7.77 (d, J ) 9.0 Hz, 1H, H7).
1-tert-Butylperoxy-5-methyl-3,3-bistrifluoromethyl-1H-1,2-ben-
ziodoxole (2c). Yield (57%) mp 131-132.5 °C. 1H NMR (CDCl3)
δ: 1.29 (s, 9H, t-Bu); 2.51 (s, 3H, Me5); 7.50 (bs, 1H, H4); 7.58 (d,
J ) 8.5 Hz, 1H, H6); 7.92 (d, J ) 8.4 Hz, 1H, H7).
Thermal Decomposition of (tert-Butylperoxy)iodanes. A Typical
Procedure. 2a or 2c (0.05 mmol) was dissolved in 1 mL of the
appropriate solvent, and the solution was placed in an ampoule, purged
with argon, and sealed. The ampoules were heated in a boiling hexane
(69 °C, for 2a) or cyclohexane (81 °C, for 2c) bath for 3 and 30 h,
respectively. The ampoules were opened, and a weighted amount of
bromobenzene as the internal standard was added to the reaction
bonding.12 It perhaps renders these systems more susceptible
to acceptance of an incoming electron into an antibonding σ*
MO of the C-I bond13 and/or stabilizes the proposed iodine
centered radical intermediate, i.e., ArI•- (or Ar-I-R), since
considerably lower yields of deiodinated products were obtained
when other electronegatively 2-substituted iodobenzenes were
employed. A further observation that the deiodination reactions
proceed much less efficiently in cyclohexane with added “basic”
hydrogen bond donors (B) (B ) tert-butyl methyl ether (â =
0.47), pyridine (0.642), N,N-dimethylacetamide (0.756);14 cy-
clohexane:B ) 4:1, v/v), capable of interrupting intramolecular
hydrogen bonding by forming o-iodocumyl alcohol-B adducts,
seems to support this presumption. With increasing hydrogen
bond donor ability (â) of the added B, the efficiency of
deiodination was decreasing. The fact that the gem-bis-
(trifluoromethyl) derivative 3c is more reactive (more “acidic”)15
than the dimethyl analogue 3a, is also indicative in this respect.
The dramatic difference in reactivity of cyclohexyl and benzyl
radicals also deserves comment. Both radicals must have
comparable adiabatic ionization potentials (cyclopentyl, 7.21 eV
(IP for cyclohexyl is not available); benzyl, 7.20 eV).16
(12) The strength of the intramolecular hydrogen bonding in o-iodophenol
was estimated by IR (CCl4) to be 1.55 ( 0.15 kcal/mol (Bourassa-Bataille,
H.; Sauvageau, P.; Sandorfy, C. Can. J. Chem. 1963, 41, 2240. Carlson,
G. L.; Fateley, W. G.; Manocha, A. S.; Bentley, F. F. J. Phys. Chem. 1972,
76, 1553). Theoretical studies have indicated that the intramolecular
interactions in the cis conformer of o-halophenols are due to a competition
between the attractive and repulsive H‚‚‚halogen interactions, as well as
the O‚‚‚halogen repulsions. An important factor is also the O-H‚‚‚X angle
(Dietrich, S. W.; Jorgensen, E. C.; Kollman, P. A.; Rothenberg, S. J. Am.
Chem. Soc. 1976, 98, 8310). Since this angle is near to 180° in the six-
membered hydrogen-bonded ring in o-iodocumyl alcohols, i.e., the optimal
value for such interactions (it is around 141° in o-iodophenol), it seems
safe to predict that the strength of the intramolecular hydrogen bonding in
these compounds must be g2 kcal/mol.
(13) Fukui, K.; Morokuma, K.; Kato, H.; Yonezava, T. Bull. Chem. Soc.
Jpn. 1963, 36, 217. Nagai, S.; Gillbro, T. J. Phys. Chem. 1977, 81, 1793.
Mishra, S. P.; Symons, M. C. R. J. Chem. Soc., Perkin Trans. 2 1981, 185.
(14) Kamlet, M. J.; Jones, M. E.; Taft, R. W.; Abboud, J.-L. J. Chem.
Soc., Perkin Trans. 2 1979, 342.
(15) Moss, R. A.; Wilk, B.; Krogh-Jespersen, K.; Blair, J. T.; Westbrook,
J. D. J. Am. Chem. Soc. 1989, 111, 250.
(17) (a) Wentrup, C. ReactiVe Molecules; Wiley: New York, 1984; pp.
28-29. (b) Handbook of Chemistry and Physics; Lide, D. R., Ed.; CRC
Press: Boca Raton, 1995. (c) Denisov, E. T. Zh. Fiz. Khim. 1995, 69, 436.
(d) Prof. S. W. Benson, personal communication.
(18) Hill, J. G.; Rossiter, B. E.; Sharpless, K. B. J. Org. Chem. 1983,
48, 3607.
(19) Bartlett, P. D.; Benzing, E. P.; Pincock, R. E. J. Am. Chem. Soc.
1960, 82, 1762.
(20) Cooper, W. J. Chem. Soc. 1951, 3106.
(16) (a) Hayashibara, K.; Kruppa, G. H.; Beauchamp, J. L. J. Am. Chem.
Soc. 1986, 108, 5441, and references cited therein. (b) Wayner, D. D. M.;
McPhee, D. J.; Griller, D. J. Am. Chem. Soc. 1988, 110, 132.
(21) Moss, R. A.; Alwis, K. W.; Shin, J. J. Am. Chem. Soc. 1984, 106,
2651.
(22) Johnstone, R. A. W.; Rose, M. E. Tetrahedron 1979, 35, 2169.

2632 J. Am. Chem. Soc., Vol. 119, No. 11, 1997
Dolenc and Plesnicˇar
mixture, which was subsequently analyzed by GC. The identity of
the products was determined by GC-MS and by matching with authentic
samples.
probehead of the NMR spectrometer (B ) 7 T) or in a water bath
thermostat. The temperature in the NMR spectrometer was calibrated
by means of the methanol sample. The temperature was held within
(0.1 °C. The integral of H7 was followed usually to 50-70%
conversion and compared with a standard, which was an integral of
residual protons in the deuterated solvent (benzene and acetonitrile) or
the methyl signal of added anisole in cyclohexane. The measurements
in the thermostat were carried out by a rapid transfer of the NMR tubes
from the thermostat to the NMR probehead, heated to the same
temperature. After measurements, the tubes were quickly returned to
the water bath. Control experiments with five tubes, containing the
aliquots of the same initial solution and taken one by one from the
water bath in 30 min intervals, yielded the same result within the
experimental error.
Deiodinations by Alkyl Radicals. A Typical Procedure: Substi-
tuted iodobenzene (0.10 mmol) and 0.10 mmol of DBPO, together with
one drop of bromobenzene (internal standard), were dissolved in 4 mL
of deaerated solvent. The solution was heated in an argon atmosphere
at 70-80 °C (when in cyclohexane or benzenesunder reflux; in the
case of isobutane, the reaction mixture was heated in a sealed heavy
wall tube) for 2-4 h and analyzed by GC. When radical sources other
than DBPO were used, the reaction times were adjusted according to
their decomposition rates.
The Reaction of 2-(2-Iodophenyl)-2-propanol with DBPO in
Cyclohexane-d12. 3a (37 mg, 0.14 mmol) and 25 mg (0.11 mmol) of
di-tert-butyl peroxyoxalate (DBPO) was dissolved in 1 mL of cyclo-
hexane-d12 (Fluka, 99.5%D), the solution was flushed with argon, and
heated for 2 h under reflux in an argon atmosphere. After chroma-
tography on silica gel with diethyl ether-pentane (1:3) as the eluent,
15 mg of an oily product was isolated. GC analysis showed a pure
compound with a retention time identical to that of 2-phenyl-2-propanol.
Spectroscopic data correspond to 2-(2-deuterophenyl)-2-propanol: 1H
NMR (CDCl3) δ: 1.59 (s, 6H, H1, H3); 1.75 (s, OH); 7.25 (ddd, J )
7.61; 6.93; 1.33 Hz, 1H, H4′); 7.31-7.38 (m, 2H, H3′, H5′); 7.50 (dm,
J ) 8.5 Hz, 1H, H6′). 13C NMR (CDCl3) δ: 31.77 (C1, C3); 72.53
Kinetic and activation parameters were obtained by standard
procedures.
Relative Rates of Deiodination of Substituted 2-(2-Iodophenyl)-
2-propanols by Cyclohexyl Radicals. A solution of substituted 2-(2-
iodophenyl)-2-propanol (5-OMe, 5-Me, 5-Cl, and 4-Cl; 0.10 mmol),
3a (0.10 mmol), and DBPO (0.05 mmol) in 5 mL of cyclohexane was
heated under reflux in an inert atmosphere for 2 h and analyzed by
GC. The relative rates kX/kH were calculated by using the integrated
rate equation, kX/kH ) ln[(A - X)/A]/ln[(B - Y)/B], where A is the
amount of the substituted alcohol, B is the amount of 3a at the beginning
of the reaction, and X and Y are the amounts of the corresponding
deiodinated alcohols at the end of the reaction.
1
(C2); 124.08 (t-1:1:1, JC-D ) 24.0 Hz, C2′); 124.37 (C6′); 126.70
(C4′); 128.12 (C3′); 128.23 (C5′); 149.05 (C1′). MS m/z (%): 137
(M+, 4); 122 (100); 121 (12); 106 (7); 92 (8); 78 (23).
The Reaction of O-Deutero-2-(2-iodophenyl)-2-propanol with
DBPO in Cyclohexane. The solution of 3a (0.27 mmol) in 3 mL of
degassed cyclohexane and 0.50 mL of deuterium oxide (Fluka,
99.95%D) was irradiated in an ultrasound bath for 30 min, the layers
were separated, and the organic layer was dried over Na2SO4. The
degree of the isotopic exchange was estimated by 1H NMR to be about
60%. To this solution was added 0.10 mmol of DBPO, and the reaction
mixture was heated in an argon atmosphere under reflux for 2 h and
analyzed by GC-MS. No incorporation of deuterium into iodocyclo-
hexane or 2-phenyl-2-propanol was found.
The Reaction of 2-(2-Iodophenyl)-2-propanol with Dibenzoyl
Peroxide in Benzene. The solution of 0.115 mmol of 3a, 0.111 mmol
of dibenzoyl peroxide, and 0.118 mmol of bromobenzene as the internal
standard in benzene was heated for 24 h under reflux in an argon
atmosphere. GC analysis showed partial (64%) disappearance of 3a
and the formation of 57% of iodobenzene and 32% of 2-phenyl-2-
propanol, together with biphenyl and some benzoic acid. (The
discrepancy between the percentage of iodobenzene and 2-phenyl-2-
propanol in the reaction mixture is probably due to formation of
nonvolatile byproducts.)
Acknowledgment. We wish to thank Prof. Glen A. Russell
for very stimulating discussions. This research was funded by
the Ministry of Science and Technology of the Republic of
Slovenia and (in part) by the U.S.-Slovene Joint Fund (NSF
No. 95-455).
Supporting Information Available: Additional spectro-
scopic (IR, 13C and 19F NMR, MS) data and elemental analyses
of (tert-butylperoxy)iodanes 2a, 2b, and 2c, full details of
syntheses and characterization data for 1-chloro-5-methoxy-3,3-
dimethyl-1H-1,2-benziodoxole (1b), 2-(2-iodophenyl)-2-meth-
oxypropane, 2-(2-iodo-5-methoxyphenyl)-2-propanol (3b), 2-(2-
iodo-5-methylphenyl)-2-propanol, 2-(5-chloro-2-iodophenyl)-2-
propanol, 2-(4-chloro-2-iodophenyl)-2-propanol, and 2-(2-bromo-
5-methylphenyl)-1,1,1,3,3,3-hexafluoro-2-propanol, infrared OH
stretching frequencies for selected cumyl alcohols in CCl4 (7
pages). See any current masthead page for ordering and Internet
access instructions.
Kinetic Measurements. Kinetic measurements were carried out
in NMR tubes, flushed with argon, and sealed, in a thermostated
JA9636404