Efficient production of enol ether radical cations by heterolytic cleavage of β-mesylate radicals

Article abstract of DOI:10.1021/ol034061o

(Matrix presented) α-Methoxy-β-mesyloxy radicals were produced in laser flash photolysis reactions, and yields of enol ether radical cations formed by heterolytic fragmentation of the mesylate group were determined. The mesylate heterolysis reaction is fa

Full text of DOI:10.1021/ol034061o

ORGANIC
LETTERS
2003
Vol. 5, No. 6
827-830
Efficient Production of Enol Ether
Radical Cations by Heterolytic Cleavage
of â-Mesylate Radicals
Elsa Taxil, Laurent Bagnol, John H. Horner,* and Martin Newcomb*
Department of Chemistry, UniVersity of Illinois at Chicago, 845 W. Taylor St.,
Chicago, Illinois 60607
men@uic.edu
Received January 13, 2003
ABSTRACT
r-Methoxy-â-mesyloxy radicals were produced in laser flash photolysis reactions, and yields of enol ether radical cations formed by heterolytic
fragmentation of the mesylate group were determined. The mesylate heterolysis reaction is faster than heterolyses of phosphate and bromide
groups in analogous radicals and highly efficient in medium-polarity solvents.
Heterolytic cleavage reactions of radicals with leaving groups
adjacent to the radical center produce radical cations regio-
selectively under nonoxidative reactions, and the method
permits one to generate radical cations from synthetic
intermediates that contain functionality that would not survive
bulk oxidative conditions. Applications of radical heterolysis
reactions require that one controls the rate of the cleavage
reaction that produces the initial ion pair, the fate of the ion
pair (i.e., collapse leading to rearrangement versus diffusional
escape), and, if diffusionally free radical cations are pro-
duced, the rates of reactions of these species.
Radical heterolysis reactions leading either to 1,2-migra-
tion products or to diffusively free radical cations typically
employ carboxylate or phosphate esters as the leaving
groups,1 but other leaving groups have been used. In aryl-
substituted systems, diffusively free styrene radical cations
were formed by heterolytic cleavage of â-halo radicals,2,3
and 1,2-migration reactions of a â-benzenesulfonate radical
and a â-nitrate radical were reported.4 Alkyl â-mesylate
radicals react in water to produce acid, apparently by
heterolytic fragmentations to alkene radical cations and
subsequent deprotonations of the radical cations.5 We report
here laser flash photolysis (LFP) studies of â-mesylate
radicals that give enol ether radical cations and compare the
kinetics and radical cation yields to analogous radicals with
phosphate and bromide leaving groups. The heterolysis
reactions of â-mesylate radicals are quite fast and permit
radical cation formation in good yields in intermediate
polarity solvents.
A radical heterolysis reaction is a composite of a dissocia-
tion step that gives a contact ion pair and an escape step,
and competing reactions in the ion pair, recombination, and
deprotonation of the radical cation, limit the yield of
diffusively free radical cations (Scheme 1). For ion pairs
containing phosphate counterions, we previously demon-
strated that reactions in the ion pair were important in limiting
yields and that increasing the solvent polarity resulted in both
faster heterolysis reactions and higher yields of diffusively
free radical cations.6
(1) Beckwith, A. L. J.; Crich, D.; Duggan, P. J.; Yao, Q. W. Chem. ReV.
1997, 97, 3273-3312.
(2) Cozens, F. L.; O’Neill, M.; Bogdanova, R.; Schepp, N. J. Am. Chem.
Soc. 1997, 119, 10652-10659.
(5) Koltzenburg, G.; Behrens, G.; Schulte-Frohlinde, D. J. Am. Chem.
Soc. 1982, 104, 7311-7312.
(6) Horner, J. H.; Taxil, E.; Newcomb, M. J. Am. Chem. Soc. 2002, 124,
5402-5410.
(3) Bales, B. C.; Horner, J. H.; Huang, X. H.; Newcomb, M.; Crich, D.;
Greenberg, M. M. J. Am. Chem. Soc. 2001, 123, 3623-3629.
(4) Crich, D.; Filzen, G. F. Tetrahedron Lett. 1993, 34, 3225-3226.
10.1021/ol034061o CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/15/2003

Scheme 1
In this work, PTOC esters7 1-4 (Scheme 2) were used as
radical precursors. The PTOC esters were prepared from the
corresponding carboxylic acids, and the syntheses are
described in Supporting Information. In LFP experiments,
flowing solutions of the PTOC esters were irradiated with
355 nm light from a Nd-YAG laser. The photolyses gave
acyloxyl radicals (5) that rapidly decarboxylated to give the
radicals of interest. The pyridine-2-thiyl radical (6) was
formed as a byproduct of the photolysis, and the absorbance
from radical 6 was used as a reference to determine the yields
of the radical cations.
Figure 1. Time-resolved spectrum of radical cation 9 produced
from radical 7a. The traces are at 43 ns (blue), 63 ns (green), and
103 ns (red) after the laser pulse with the trace at 23 ns subtracted
to give a baseline. The inset shows a kinetic trace at 330 nm.
mixtures of acetonitrile containing 2,2,2-trifluoroethanol
(TFE), the observed rate constants for formation of 9 at (20
( 2) °C were in the range of (2.6 ( 0.2) × 107 s-1, which
is the rate constant for cyclization of radical cation 8.6 A
lower limit for the rate constant for mesylate heterolysis in
acetonitrile can be set at khet > 4 × 107 s-1 because no
convolution due to sequential reactions was apparent in the
kinetic traces.8
Scheme 2
Yields of distonic radical cation 9 were determined in
solvent mixtures of varying polarity by comparing the
intensity of the signal at λmax from 9 to the signal intensity
at λ ) 490 nm from byproduct radical 6. The results are
compared with those from radicals 7b and 7c in Figure 2.
Photolysis of PTOC ester 1 gave radical 7a (Scheme 3).
Heterolytic cleavage of the mesyloxy group in 7a gave the
enol ether radical cation 8 that was previously generated by
heterolytic cleavage of the phosphate groups in radicals 7b
and 7c.6 Radical cation 8 cyclizes to distonic radical cation
9, which has a long wavelength absorbance at 335 nm and
is readily detected.6 Figure 1 shows a time-resolved spectrum
of distonic radical cation 9 formed in a typical study of
radical 7a. For formation of distonic radical cation 9 from
7a, the heterolysis reaction that gives 8 or the cyclization
reaction of 8 can be rate limiting. In acetonitrile and in
Figure 2. Yields of radical cation 9 from heterolysis reactions of
radicals 7. TFE is 2,2,2-trifluoroethanol. The yields are listed in
Supporting Information.
Scheme 3
Yields are a function of solvent polarity and the leaving
group. The leaving group effect in acetonitrile was especially
(7) Acronym PTOC is for pyridine-2-thioneoxycarbonyl.
(8) Convolution of two first-order processes with rate constants of 3 ×
107 s-1 would give an apparent first-order rate constant of ca. 1.5 × 107
s-1
.
828
Org. Lett., Vol. 5, No. 6, 2003

striking with the yield of product 9 increasing from 20 to
40 to 100% for the series 7c, 7b, and 7a.
Table 1. Results from Heterolysis Reactions of Radicals 10
The formation of radical cation 9 in 50% yield from radical
7a in the acetonitrile/THF solvent mixture is noteworthy
because diffusively free radical cations were not formed from
the â-phosphate radicals in solvents less polar than aceto-
nitrile.6 Product loss in acetonitrile/THF apparently is due
in part to reaction of the mesylate anion in the ion pair (see
below), but reaction of radical cation 8 with THF also limited
the yield. The observed rate constant for reaction of 8 in
acetonitrile/THF at 20 °C (k ) 3.4 × 107 s-1) was greater
than that found in acetonitrile or in acetonitrile containing
TFE, which requires that another reaction was competing
with the cyclization of radical cation 8. This “other” reaction
is most likely deprotonation of 8 by THF. On the basis of
the observed rate constants, the THF reaction with 8 has a
limiting value of k > 8 × 106 s-1, and the second-order rate
constant for reaction of THF (10 M in 80/20 THF/acetoni-
trile) is k > 8 × 105 M-1 s-1.
b
radical
solvent
ET(30)a
A
640/A490
% yieldc
10a
THF
C6H5CF3
CH2Cl2
CH3CN
CH3CN
0.5% TFEd
37.4
38.5
40.7
45.6
<0.1
0.38
0.83
2.86
0.42
1.89
<3
13
29
100
15
66
10b
a ET(30) solvent polarity parameter; see ref 11. b Ratio of maximum
absorbance at 640 nm (from 12) to initial absorbance at 490 nm (from 6).
c Percent yield of radical cation 12 assuming that the reactions in acetonitrile
were quantitative. d Solvent was 0.5% 2,2,2-trifluoroethanol in acetonitrile.
with radicals 10 are similar to those for radicals 7a and 7c.
The yield of radical cation from 10a increased with increas-
ing solvent polarity as measured by the ET(30) solvent
parameter,11 and the yield of radical cation from reaction of
â-mesyloxyl radical 10a in acetonitrile was much greater
than that from the â-diethylphosphatoxy radical 10b. We
conclude that mesylate deprotonation of radical cation 11
was important in ion pairs produced in the low-polarity
solvents. Reaction of 11 with the solvent THF also is
expected.
The efficiencies of the mesylate and diethyl phosphate
leaving groups also were compared in radicals 10, produced
from PTOC esters 2 and 3. Heterolysis of these radicals gives
enol ether radical cation 11. Relative yields for the heterolysis
reactions were determined by allowing 11 to react with 0.001
M diphenylmethylamine. The enol ether radical cation
oxidizes the amine to the diphenylmethylaminium radical
cation (12) in a diffusion-controlled electron-transfer reaction
(Scheme 4),9 and radical cation 12 is readily detected by its
The mesylate group was compared to diphenyl phosphate
and bromide leaving groups in reactions of radicals 13
(Scheme 5). Previously, heterolysis reactions of radicals 13b
Scheme 5
Scheme 4
long wavelength absorbance. The intermolecular trapping
reaction monitors only diffusively free radical cations and
provides limited kinetic information. However, the kinetics
of reactions with diphenylmethylamine were not convoluted
with another dynamic process, and one can deduce that the
mesylate heterolysis reactions had rate constants of k > 1 ×
107 s-1 in all cases where aminium cation radical 12 was
produced.
In Table 1, we list the ratio of the maximum absorbance
at 640 nm from radical cation 12 to the instantaneous
absorbance at 490 nm from the 2-pyridinethiyl radical (6),
the byproduct of the photolysis reaction. From the results
with â-mesylate radical 7a, we assume that the heterolysis
reaction of radical 10a in acetonitrile is quantitative and
report yields of 12 based on this assumption.10 The results
and 13c were found to give radical cation 14 (λmax at 365
nm) in acetonitrile and allylic radical 15 (λmax at ca. 295
nm) in THF, and radical cation 14 was not observed as a
transient in reactions in THF.3 In the present study, rapid
fragmentation of radical 13a in THF (k > 2 × 108 s-1) gave
radical cation 14 that subsequently was converted to allylic
radical 15 as shown in Figure 3, where the decaying species
gives a negative signal and the growing species gives a
positive signal. The deprotonation reaction that converts 14
to 15 must involve THF as the base because the pseudo-
first-order rate constants for loss of 14 and formation of 15
(k ) 1.2 × 107 s-1) are too large for a bimolecular reaction
of 14 with the mesylate anion, even if that reaction were
diffusion controlled. Using 12.3 M as the concentration of
THF gave the second-order rate constant for the reaction of
THF with radical cation 14 as k ) 1.0 × 106 M-1 s-1.
Mesylate fragmentation from 13a was too fast to measure
in all solvents studied (k > 2 × 108 s-1). Diphenyl phosphate
(9) Newcomb, M.; Miranda, N.; Huang, X. H.; Crich, D. J. Am. Chem.
Soc. 2000, 122, 6128-6129.
(10) Reduced yields of 12 in low-polarity solvents do not necessarily
accurately reflect the yields of 11. Diphenylmethylamine can react with
radical cation 11 as a reductant, a base, or an electrophile, and the relative
rate constants for these processes can be solvent sensitive.
(11) Reichardt, C. Chem. ReV. 1994, 94, 2319-2358.
Org. Lett., Vol. 5, No. 6, 2003
829

Table 2. Results from Heterolysis Reactions of Radical 13a
solvent
ET(30)a
% yield of 14b
THF
C6H5CF3
37.4
38.5
38
49
1% TFE in TFTc
1:1 THF/CH3CN
CH3CN
69
83
100
45.6
a ET(30) solvent polarity parameter; see ref 11. b Yield of 14 determined
from the ratio of the maximum absorbance at 370 nm (from 14) to initial
absorbance at 490 nm (from 6) with the assumption that the reaction in
acetonitrile was quantitative. c Solvent was 1% 2,2,2-trifluoroethanol in
C6H5CF3.
the solvent is moderately polar, but proton transfer becomes
increasingly competitive in less polar solvents.
Figure 3. Time-resolved spectrum from reaction of radical 13a in
THF. The traces are at 58 ns (pink), 78 ns (blue), 138 ns (green),
and 178 ns (red) with signals at 38 ns subtracted to give a baseline.
The species decaying (λmax ) 368 nm) is radical cation 14, and the
species growing (λmax ) 295 nm) is allylic radical 15. Decay of
byproduct radical 6 can be seen in the region λ ) 440-495 nm.
The inset shows kinetic traces at 300 nm (red) and 370 nm (blue);
the initial bleaching is due to destruction of the radical precursor
and laser light scattering in the 370 nm trace.
Our results provide some guidelines for synthetic applica-
tions of enol ether radical cations prepared by the radical
heterolysis pathway. The solvent must be sufficiently polar
to allow efficient escape from the ion pair.12 High-yield
mesylate heterolysis reactions are possible in acetonitrile,
but phosphate (and, presumably, also carboxylate) heterolyses
will require more polar media to prevent proton transfers
within the initially formed ion pair. THF and other ethers
should be avoided because they will react with the cation
radicals. In synthetic applications, where â-mesyloxyl radi-
cals are produced in chain reactions, trifluorotoluene contain-
ing 2,2,2-trifluroethanol or a mixture of trifluorotoluene and
acetonitrile might be attractive solvents if a tin hydride
protocol is used for radical production.
cleavage in radical 13b also was fast in acetonitrile, but the
heterolysis reaction of radical 13b in THF had a rate constant
of 3.4 × 106 s-1.3 Bromide fragmentation in radical 13c was
slower, with rate constants of k ) 8 × 107 s-1 in acetonitrile
and k ) 2.8 × 105 s-1 in THF.3
Table 2 lists the yields of radical cation 14 from reactions
of radical 13a in various solvents. Yields were determined
from the ratio of the maximum absorbance at 370 nm (from
14) to the instantaneous absorbance at 490 nm from the
byproduct of the photolysis reaction, radical 6. As above,
we assumed that the heterolysis in acetonitrile was quantita-
tive.
Acknowledgment. This work was supported by a grant
from the National Science Foundation.
Supporting Information Available: Experimental de-
tails, syntheses of PTOC ester radical precursors 1-4, table
of yields used for Figure 2, and NMR spectra. This material
is available free of charge via the Internet at http://pubs.acs.org.
The critical feature in regard to yields of diffusively free
enol ether radical cations from radical heterolysis reactions
is the competition between proton transfer and ion escape
in the first-formed ion pairs. As shown in this work, enol
ether radical cations react rapidly even with the weak base
THF. The mesylate anion is sufficiently weak as a base that
escape from the ion pair is faster than proton transfer when
OL034061O
(12) Good correlation between kinetics and efficiencies of radical cation
formation with solvent polarities suggests that polarity is more important
in controlling these features than specific interactions of the solvent
molecules with nascent or fully formed radical cations.
830
Org. Lett., Vol. 5, No. 6, 2003