Factors affecting the rates of addition of free radicals to alkenes - Determination of absolute rate coefficients using the persistent aminoxyl method

Article abstract of DOI:10.1021/ja025730g

The rate of coupling of alkyl radicals with the persistent aminoxyl radical 1,1,3,3-tetramethyl-isoindolin-N-oxyl (1) has been used as a kinetic probe to determine absolute rate coefficients for the addition of alkyl radicals to methyl acrylate. The results are discussed in terms of the role of the structure and functionalization of the attacking radical on the rates of addition, particularly as they affect steric, polar, and enthalpic factors. The aminoxyl method is assessed against other methods for determining free radical addition rate coefficients.

Full text of DOI:10.1021/ja025730g

Published on Web 07/17/2002
Factors Affecting the Rates of Addition of Free Radicals to
AlkenessDetermination of Absolute Rate Coefficients Using
the Persistent Aminoxyl Method
Athelstan L. J. Beckwith and James S. Poole*
Contribution from the Research School of Chemistry, Australian National UniVersity,
Australian Capital Territory, Australia 0200
Received January 28, 2002
Abstract: The rate of coupling of alkyl radicals with the persistent aminoxyl radical 1,1,3,3-tetramethyl-
isoindolin-N-oxyl (1) has been used as a kinetic probe to determine absolute rate coefficients for the addition
of alkyl radicals to methyl acrylate. The results are discussed in terms of the role of the structure and
functionalization of the attacking radical on the rates of addition, particularly as they affect steric, polar,
and enthalpic factors. The aminoxyl method is assessed against other methods for determining free radical
addition rate coefficients.
8
Introduction
ideal potential “clocks” for free radical reactions. They have
been used successfully in competitive kinetic experiments to
determine the rates of unimolecular free radical reactions. For
Persistent aminoxyl radicals have found widespread use in
both chemistry and biology as mechanistic and kinetic probes
for reactions believed to proceed via carbon-centered free radical
intermediates. The utility of species such as 1,1,3,3-tetrameth-
ylisoindolin-N-oxyl (1)1,2 arises from the fact that they selec-
9
bimolecular free radical reactions the major application of
aminoxyl couplings has been the identification of product
radicals. In particular, the groups of Rizzardo and Solomon and
of Jenkins and Busfield have identified the products arising from
tively couple with carbon-centered free radicals, generating
1
0,11
alkoxylamine adducts.3
the early stages of polymerization
and the reactions of
1
2,13
alkoxyl radicals with various organic substrates.
Given the
obvious success of these groups in isolating and identifying the
products of telomerization reactions, it is perhaps surprising that
the use of persistent aminoxyl radicals as “clocks” for free
1
4
radical addition reactions is not more widespread.
In the current work we have determined the rate coefficients
and Arrhenius parameters for the addition of a number of free
radicals to methyl acrylate using the aminoxyl coupling method
and have compared the results obtained with those derived from
(
7) Sobek, J.; Martschke, R.; Fischer, H.; J. Am. Chem. Soc. 2001, 123, 2849-
857.
2
The coupling reaction is rapid: rate coefficients for coupling
(
8) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317-323.
8
9
-1
kC with small radicals generally range between 10 and 10 M
s .
(9) Examples are (a) Beckwith, A. L. J.; Bowry, V. W. J. Am. Chem. Soc.
1994, 116, 2710-2716. (b) Engel, P. S.; He, S.-L.; Banks, J. T.; Ingold,
K. U.; Lusztyk, J. J. Org. Chem. 1997, 62, 1210-1214.
-1 4-7
The availability of rate coefficients for coupling of
carbon-centered radicals with 1 and other aminoxyls makes them
(10) (a) Moad, G.; Rizzardo, E.; Solomon, D. H. Macromolecules 1982, 15,
9
09-914. (b) Rizzardo, E.; Serelis, A. K.; Solomon, D. H. Aust. J. Chem.
1
982, 35, 2013-2024. (c) Grant, R. D.; Griffiths, P. G.; Moad, G.; Rizzardo,
*
To whom correspondence should be addressed. Current address:
Department of Chemistry, Ohio State University, Columbus, OH 43210-
156. E-mail: jpoole@chemistry.ohio-state.edu.
E.; Solomon, D. H. Aust. J. Chem. 1983, 36, 2447-2454. (d) Cuthbertson,
M. J.; Rizzardo, E.; Solomon, D. H. Aust. J. Chem. 1983, 36, 1957-1973.
(11) (a) Busfield, W. K.; Jenkins, I. D.; Thang, S. H.; Rizzardo, E.; Solomon,
D. H. Aust. J. Chem. 1985, 38, 689-698. (b) Busfield, W. K.; Jenkins, I.
D.; Thang, S. H.; Rizzardo, E.; Solomon, D. H. J. Chem. Soc., Perkin Trans.
1 1988, 485-490. (c) Busfield, W. K.; Grice, I. D.; Jenkins, I. D. Aust. J.
Chem. 1995, 48, 625-634. (d) Busfield, W. K.; Jenkins, I. D.; Monterio,
M. J. Aust. J. Chem. 1997, 50, 1-7.
1
(
1) (a) Rozantsev, E. G.; Sholle, V. D. Synthesis 1971, 190-202. (b) Keana,
J. F. W. Chem. ReV. 1978, 78, 37-64.
(
(
(
(
2) Griffiths, P. G.; Moad, G.; Rizzardo, E.; Solomon, D. H. Aust. J. Chem.
1
983, 36, 397-401.
3) Murayama, K.; Morimura, S.; Yoshioka, T. Bull. Chem. Soc. Jpn. 1969,
2, 1640-43.
4) Chateauneuf, J.; Lusztyk, J.; Ingold, K. U. J. Org. Chem. 1988, 53, 1629-
2.
5) (a) Beckwith, A. L. J.; Bowry, V. W.; Moad, G. J. Org. Chem. 1988, 53,
4
(12) Grant, R. D.; Rizzardo, E.; Solomon, D. H. J. Chem. Soc., Perkin Trans.
2 1985, 379-84.
3
(13) (a) Busfield, W. K.; Jenkins, I. D.; Thang, S. H.; Rizzardo, E.; Solomon,
D. H. Tetrahedron Lett. 1985, 26, 5081-5084. (b) Busfield, W. K.; Grice,
D. I.; Jenkins, I. D.; Thang, S. H. Aust. J. Chem. 1991, 44, 1407-1415.
(c) Bottle, S. E.; Busfield, W. K.; Jenkins, I. D. J. Chem. Soc., Perkin
Trans. 2 1992, 2145-2150. (d) Busfield, W. K.; Jenkins, I. D.; Thang, S.
H.; Rizzardo, E.; Solomon, D. H. Eur. Polym. J. 1993, 29, 397-400.
(14) Nakamura, T.; Busfield, W. K.; Jenkins, I. D.; Rizzardo, E.; Thang, S. H.;
Suyama, S. J. Org. Chem. 1997, 62, 5578-5582.
1
632-1641. (b) Beckwith, A. L. J.; Bowry, V. W.; Ingold, K. U. J. Am.
Chem. Soc. 1992, 114, 4983-4992. (c) Bowry, V. W.; Ingold, K. U. J.
Am. Chem. Soc. 1992, 114, 4992-4996.
(
6) Ingold, K. U.; Walton, J. C. Radical Reaction Rates in Liquids, SubVolume
c. Nitrogen Centred Radicals, Aminoxyls and Related Radicals; Springer-
Verlag: Berlin, 1994; Vol. 18.
10.1021/ja025730g CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 9489-9497
9
9489

A R T I C L E S
Beckwith and Poole
Scheme 1
[
3]t
[1] - [1]t
other methods. The resultant data are discussed in terms of the
effect of the structure and functionalization of the attacking
radical on the rate of addition to olefins.
0
)
- 1
(2a)
(2b)
{[5] + [7]}t
R + [1]0
R ln
{
}
R + [1]t
Results and Discussion
kadd,2[MA]
R )
The Persistent Aminoxyl Method. Scheme 1 shows the
reactions that occur under our conditions when alkyl radicals
are generated in the presence of methyl acrylate (MA) and the
aminoxyl 1, which was chosen because it is crystalline, is easily
purified, and affords adducts that are readily quantified by
HPLC/UV. In theory the reaction should produce a mixture of
higher oligomers, but in practice the coupling constants kC are
so much larger than the addition constants kadd that the formation
of products arising from further addition reactions of radical 6
is negligible. Under these circumstances the only stable products
formed in detectable concentrations are 3, 5, and 7. The usual
treatment of the reactions of Scheme 1 gives the kinetic
expressions
kC
[5]t
[1] - [1]t
0
)
- 1
[
7]t
â + [1]0
â ln
{
}
â + [1]t
k
add,4[MA]
â )
kC
The absolute rate coefficients for addition, kadd, may be
calculated from the relative rate coefficients using published
rate coefficients for the coupling of free radicals with 1, kC.
The radicals 2b-f were generated in the presence of known
amounts of 1 and an internal standard (naphthalene) in monomer
between 298 and 355 K. Radicals 2b-d were generated from
the corresponding diacyl peroxides and 2d-f from the tert-
butyl peresters. The precursors for 2d were chosen on the basis
of the temperature employed. Crossover experiments performed
at 315 and 333 K demonstrated that identical relative rate
coefficients were obtained from both precursors. The use of tert-
butyl peresters as radical precursors leads to the generation of
tert-butoxyl radicals, the reactions of which are discussed below.
The reactions were carried out in duplicate in neat methyl
acrylate at three concentrations of 1 covering an order of
magnitude, each at four temperatures within the range 298-
355 K. Product analysis was performed by reverse-phase HPLC
with detection by UV absorption, and the final concentration
d{[5] + [7]}
d[3]
k
add,2[MA]
kC[1]
)
(1a)
(1b)
d[7]
d[5]
k
add,4[MA]
kC[1]
)
The difference in the magnitudes of kC and kadd requires the
reactions to be conducted with MA in relatively high concentra-
tions and in large excess. Under these circumstances, [MA]
remains effectively constant throughout the course of reaction.
The observed kinetics are therefore pseudo first order in nature
and, thus, allow the relative rate coefficients to be determined
analytically from the integrated rate equations (2a) and (2b).
9490 J. AM. CHEM. SOC.
9
VOL. 124, NO. 32, 2002

Rates of Addition of Free Radicals to Alkenes
A R T I C L E S
Table 1. Kinetic Data for Addition of Organic Free Radicals to Methyl Acrylatea
rel kinetic datab
competing reacn
abs kinetic data for addition
add − E c
E
E
298
k
add
E
C
C
add
(kJ mol-1
e,f
-1 f
log Aadde
(M-1 s-1
)
radical
methyl (2a)
pentyl (2b)
-methylbutyl (2c)
cyclohexyl (2d)
log (Aadd/A
C
)c
(kJ mol-1
)
Nd
log A
9.7
9.7
9.7
9.7
9.7
9.7
C
(kJ mol
)
)
5
5
6
6
6
8
4
-1.02 ( 0.52
-1.32 ( 0.08
-1.72 ( 0.08
-1.39 ( 0.07
-1.99 ( 0.09
-1.00 ( 0.19
-5.13 ( 0.31
-6.0 ( 1.1
14.0 ( 3.2
11.1 ( 0.4
7.2 ( 0.4
6.4 ( 0.4
2.6 ( 0.6
-2.6 ( 1.1
-30.0 ( 1.9
-24 ( 7
12.6 ( 2.3
9.1 ( 2.8
25
30
35
36
28
16
25
20
10
25
2a + 1 f 3a
2b + 1 f 3b
2c + 1 f 3c
2d + 1 f 3d
2e + 1 f 3e
2f + 1 f 3f
2g + 1 f 3g
3.8
3.8
3.8
3.8
3.8
3.8
8.7
8.4
8.0
8.3
7.7
8.7
7.6 ( 0.3
6.7 ( 1.1
7.2
17.8
3.8 × 10
6.2 × 10
1.2 × 10
3.3 × 10
3.8 × 10
3.0 × 10
3.4 × 10
14.9
11.0
10.2
6.4
1
1
1
,1-dimethylpropyl (2e)
-adamantyl (2f)
1.2
g
12.7 ( 0.3g
12.7 ( 0.3
47.5 ( 1.1g
47.5 ( 1.1
tert-butoxy (2g)
tert-butoxy (2g)
4
17.5 ( 1.9
23 ( 7
16.4
g,h
i
g
g
2
2g + 1 f 9c
5 × 10
i
4
4
b
-2.53 ( 0.38
-2.55 ( 0.46
4b + 1 f 5b
4g + 1 f 5 g
9.7
9.7
3.8
3.8
4.2 × 10
j
4
g
7.1
12.9
6.9 × 10
a
Tail addition except as otherwise noted. Determined in the temperature range 298-355 K. c Errors quoted are 90% confidence intervals for the line
b
d
e
-1
of best fit. Number of experiments used to construct the relevant Arrhenius curves. Data shown are numerical values only, equivalent to log (A/M s ).
Data from ref 5a unless otherwise specified. As these data were quoted without any experimental uncertainty, the experimental uncertainties of the absolute
f
g
kinetic data cannot be determined and are not included in this table. Data from ref 18. The competing reaction is unimolecular â-fission. Therefore, the
units for AC and Aadd/AC in this case are s and M, respectively. h Data for head addition to methyl acrylate. 1-(Methoxycarbonyl)heptyl radical. j 1-
-
1
i
(Methoxycarbonyl)-2-tert-butoxyethyl radical.
of 1 was determined relative to naphthalene internal standard.
The compounds 3b-f and 5b-f were all isolated and character-
ized; the compounds 7b-f and 5b(head), arising from “head”
addition of the pentyl radical (addition to the substituted end
of the monomer), were identified on the basis of HPLC retention
times and electrospray mass spectra (ESMS) of the collected
fractions. The potential reversibility of the coupling reaction
under the reaction conditions, particularly with 2e, was inves-
tigated by obtaining the time-resolved product analysis. Each
time point in these experiments yielded the same (within
experimental uncertainty) relative rate coefficients, with no
apparent increase in R over time. Thus, we conclude that the
coupling reaction may be considered effectively irreversible
under the conditions used.
Figure 1. Relative Arrhenius plots of kadd/kC for reactions of radicals 2b
triangles), 2c (squares), 2d (circles), 2e (crosses), and 2f (diamonds) by
addition to methyl acrylate and by aminoxyl coupling. log(kadd/kC) data for
(
2f are shifted by -2.5.
Substitution into eq 2a of the experimentally determined
concentrations of 1 and addition products yielded the reduced
concentration parameter R. Correction of R for the effective
concentration of methyl acrylate (equivalent to the value for
neat solvent, 11.1 M, with a small dilution factor due to the
addition of initiator solution to the reaction mixture) yielded
the relative rate coefficients kadd/kC. Arrhenius curves for the
relative rate coefficients were constructed for the additions of
The tert-Butoxyl Radical. Thermal decomposition of tert-
butyl peresters in the experiments described above for 2d-f
generates the highly reactive tert-butoxyl radical 2g. The various
reactions of 2g relevant to the present work are shown in Scheme
2
.
•
â-Fission of 2g generates CH3 (2a), which adds to methyl
acrylate to afford the radical 4a. Radical 2g also reacts directly
with methyl acrylate by hydrogen atom abstraction to afford 8
or by addition to afford 4g and 9, which couple with 1 to afford
5g and 10, respectively. Under the conditions of the present
work 4g also undertakes secondary addition to methyl acrylate,
thus generating radical 6g, which couples with 1 to afford 7g.
Since the reactions of 2g with methyl acrylate in the presence
2
b-f to methyl acrylate. These are shown in Figure 1. A
summary of relative Arrhenius parameters is shown in Table
1
.
At low concentrations of 1, the alkoxylamine product 7b
arising from secondary addition reactions could be reliably
identified and quantified in product mixtures. This allowed
relative rate coefficients and Arrhenius parameters for the
addition of 4b to methyl acrylate to be estimated (Table 1).
These parameters are subject to a relatively high degree of
experimental uncertainty, because only small amounts of the
coupling product 7b are observable at low concentrations of 1.
At higher concentrations of 1, the observed levels of 7b fall
below the limit of quantification of the method.
1
0a,b
of 1 have been previously studied,
the final products (3a,
5a, 5g, 7g, and 10) could be identified simply on the basis of
chromatographic retention and ESMS. The compound 5g was
isolated and characterized as a confirmation of the retention
order. The resolution of peaks corresponding to products arising
from 8 was insufficient to provide reliable quantification of these
products, and thus these addition reactions could not be studied.
J. AM. CHEM. SOC.
9
VOL. 124, NO. 32, 2002 9491

A R T I C L E S
Beckwith and Poole
Scheme 2
Because radical 2g does not react directly with 1, the usual
aminoxyl coupling cannot be used to determine the relative rates
of reactions of 2g with methyl acrylate. Fortunately, the rate
coefficients and Arrhenius parameters for the â-fission reaction
is probably under entropic control and that the order of reactivity
for alkyl radicals is 1° > 2° > 3° but covers a narrow range
9
-1 -1
8
-1 -1
from kC ) 1.3 × 10 M
s
for pentyl to 9.1 × 10 M
s
for tert-butyl at 298 K. Because of the lack of more reliable
2
g f 2a are known,15 and this reaction therefore serves as a
data, we decided to use the general expression obtained from
5a
clock for reactions of 2g. Kinetic analysis yields expressions
similar to eq 1 for determination of the kadd for “head” and “tail”
indirect methods and recommended by Beckwith et al. for the
rate coefficient for coupling of carbon-centered free radicals
t
•
-1
-1
addition of Bu O to methyl acrylate relative to the corresponding
constant (kâ) for its â-fission. The relative rate coefficients and
Arrhenius parameters for reactions leading to the formation of
with 1 (kC; R in J mol K ):
log (k (M s)) ) 9.7 - (3800/2.303RT)
(3)
10
C
4
a, 4g, 6g, and 9 are summarized in Table 1.
Absolute Rate Data for the Addition of Free Radicals to
It should be noted that although our measurements are subject
to a small degree of uncertainty, the use of a single expression
for kC, for which no experimental uncertainties were cited, does
provide a significant (but unknown) degree of uncertainty in
our measurements. For this reason, we have not included
uncertainties for values of log Aadd and Eadd in Table 1. Given
the observations of previous workers, we expect that absolute
rate constants for addition to methyl acrylate obtained from eq
3 will tend to be low for methyl radical and high for tertiary
radicals, with variations mainly reflected in a narrow range of
values for log Aadd. Eadd would also be expected to vary only
within a relatively narrow range of values. Thus, the discrep-
ancies arising from this approach should be small and should
not affect the general trends revealed by our experiments.
Methyl Acrylate. Absolute values of the rate coefficients and
Arrhenius parameters for the various reactions discussed above
were determined by comparison of the relative values with the
corresponding absolute values for the competing reaction. The
choice of appropriate kinetic data for coupling of carbon-
7
centered radicals with 1 was problematic, due to recent work
which has demonstrated that many reactions of this type do not
4,5a
exhibit linear Arrhenius behavior. However, earlier work has
indicated that that coupling of simple alkyl radicals to 1 gives
a good linear correlation of kC with 1/T over the temperature
range studied, and this is supported by the linear behavior of
the relative data shown in Figure 1. Unfortunately, the limited
amount of data available from direct methods (LFP) is restricted
to single temperatures.4,5 They indicate that aminoxyl coupling
t
•
For the “head” and “tail” additions of Bu O to methyl acrylate
the kinetic data for the â-fission reaction were used as reference
-
1
-1 15
(15) Weber, M.; Fischer, H. J. Am. Chem. Soc. 1999, 121, 7381-7388.
(R in J mol K ):
9
492 J. AM. CHEM. SOC. VOL. 124, NO. 32, 2002
9

Rates of Addition of Free Radicals to Alkenes
A R T I C L E S
Table 2. Comparison of Kinetic Data Obtained by the Persistent
Aminoxyl and Other Methods for the Addition of Alkyl Radicals to
Methyl Acrylate
log (k (s)) ) (12.7 ( 0.3) - (47.5 ( 1.1) ×
1
0
â
3
-1
1
0 /2.303RT), R in Jmol K-1 (4)
2
98
log (Aadd
E
add
k
add
(M s)) (kJ mol-1)
(M-1 s-1
)
The derived rate coefficient data for addition of carbon-centered
free radicals to methyl acrylate are summarized in Table 1.
With the data from Table 1 in hand, the results obtained by
the persistent aminoxyl method can be compared with data
obtained from direct kinetic EPR (KEPR) measurements
and, to a lesser extent, from pulsed laser polymerization (PLP)
studies.19 There is only a relatively limited data set available
entry
radical
method
aminoxyl 8.7
KEPR 8.5
aminoxyl 8.4
3.8 × 105
1
2
3
4
methyl (2a)
17.8
16.9c
14.9
15.5c
methyl (2a)a
1-pentyl (2b)
b
b
3.4 × 10
5
5
6.2 × 10
2-(tert-butoxycarbonyl)- KEPR1
8.4
4.9 × 10
5
1
6-18
methyla
4
5
1-(methoxycarbonyl)-
aminoxyl 7.2
16.4
2.1 × 10
heptyl (4b)
butyl acrylate polymerd PLP
7.2
1,1-dimethylpropyl (2e) aminoxyl 7.7
17.3
6.4
7.2
1.5 × 10
4
6
7
8
for comparison, and the useful data is summarized in Table 2.
Generally there is good agreement between the data obtained
6
3.8 × 10
a
c
6
tert-butyl
KEPR
7.3
1.1 × 10
1
6
by the KEPR and aminoxyl methods for methyl radical (entries
and 2) and primary (entries 3-4) and secondary radicals
a
Data from ref 16. b Values used are an average value for log A
1
1
6-18,22
determined for the reaction of the target radical with several olefins.
The activation energies are calculated from the experimental rate coef-
(entries 5 and 6).
c
16-18,22 d
ficients and the average Arrhenius preexponentials.
Kinetic data
Comparisons of the values for tertiary radicals obtained from
KEPR18 and persistent aminoxyl-based methods (entries 7 and
obtained from pulsed laser polymerization.19
8
) exhibit a greater disparity. This is presumably due to the
the electron affinity of the olefin), consistent with a charge-
transfer FMO approach to polar effects.
fact that a single expression for kC is unlikely to hold for all
carbon-centered radicals. In addition, the quoted value of Eadd
is that calculated by Fischer et al.18 from an average value of
Aadd observed for a number of other olefins and may, therefore,
be subject to some uncertainty.
Factors Affecting the Rates of Addition of Free Radicals
to Olefins. The factors affecting the rate and selectivity of free
radical addition to olefins in solution have received significant
attention over time and have been extensively reviewed by
More recently, Fischer and Radom22 have demonstrated that
the activation energy for these reactions may be modeled by
an Evans-Polyani-Semenov approach, which provides an
upper limit for Eadd based on the enthalpy of reaction. The
Evans-Polyani-Semenov limiting value for Eadd is then cor-
rected by polar effects, parametrized in terms of multiplicative
polar factors (MPFs). It is clear, therefore, that bond strength,
stabilization, and polar effects are not readily separable in these
reactions.
2
0
21
Tedder, Walton, and Giese. Most recently, Fischer and
2
2
Radom have extended and improved the treatment of radical
additions by earlier qualitative models and have developed a
quantitative, parametrized model for radical additions to alkenes.
All workers have agreed that no single interaction adequately
describes the variation in rate and selectivity (chemo-, regio-,
and stereoselectivity) observed in these systems: i.e., these
reactions are governed by a “complex interplay of polar, steric
In this study, we have attempted to determine the effect of
•
changing the structure and functionalization of R on the rate
2
1a
and bond strength terms”.
Fischer et al.1
of addition to a single olefin (methyl acrylate). Fischer’s results
indicate that polar effects should be important, and a number
6-18,23-25
have demonstrated in a number of
•
studies that, for the addition of a radical species R with a series
of functionalized olefins, the rate coefficients could be modeled
using a single value of log Aadd (dependent on the structure of
of previous workers have demonstrated the significance of steric
20,21
effects on these reactions.
We have, therefore, attempted to
rationalize our observed data by reference to such effects.
•
R ). The activation energies observed exhibited good correlation
Steric Effects. It is widely accepted that for the Arrhenius
model the preexponential factor reflects the change of motional
degrees of freedom accompanying the formation of the transition
structure. For the addition of a series of neutral radicals to a
given olefin in a single solvent the effect should become
increasingly unfavorable with increasing steric bulk of the
radical. Hence, in the present work the value of log Aadd is
expected to decrease down the series primary > secondary >
tertiary. This is clearly shown in Table 1 for the reactions of
2a-d. For rigid or otherwise constrained radicals the effect is
expected to be less than that for more flexible analogues; thus,
log Aadd for cyclohexyl is greater than that for the secondary
radical 2d, while log Aadd for 1-adamantyl is the same as that
for methyl.
with the energy differences between SOMO and LUMO of the
radical-olefin pair (the ionization potential of the radical minus
(
16) (a) Zytowski, T.; Fischer, H. J. Am. Chem. Soc. 1996, 118, 437-439. (b)
Zytowski, T.; Fischer, H. J. Am. Chem. Soc. 1997, 119, 12869-12878.
17) (a) Fischer, H. In Substituent Effects in Radical Chemistry; NATO ASI
Series C; Viehe, H. G., Janousek, Z., Merenyi, R., Eds.; D. Reidel:
Dordrecht, The Netherlands, 1986; Vol. 189, pp 123-142. (b) M u¨ nger,
K.; Fischer, H. Int. J. Chem. Kinet. 1985, 17, 809-829.
(
(
18) Wu, J. Q.; Beranek, I.; Fischer, H. HelV. Chim. Acta 1995, 78, 194-214.
19) Lyons, R. A.; Hutovic, J.; Piton, M. C.; Clay, P. A.; Manders, B. G.; Kable,
S. H.; Gilbert, R. G. Macromolecules 1996, 29, 1918-27.
(
(
20) (a) Tedder, J. M.; Walton, J. C. Acc. Chem. Res. 1976, 9, 183-191. (b)
Tedder, J. M.; Walton, J. C. Tetrahedron 1980, 36, 701-707. (c) Tedder,
J. M. Angew. Chem., Int. Ed. Engl. 1982, 21, 401-410.
(
21) (a) Giese, B. Angew. Chem., Int. Ed. Engl. 1983, 22, 753-764. (b) Giese,
B.; Mehl, W. Tetrahedron Lett. 1991, 32, 4275-4278. (c) Zipse, H.; Jianing,
H.; Houk, K. N.; Geise, B. J. Am. Chem. Soc. 1991, 113, 4324-4325.
22) Fischer, H.; Radom, L.; Angew. Chem., Int. Ed. Engl. 2001, 40, 1340-
(
(
1
371.
23) (a) Walbiner, M.; Wu, J. Q.; Fischer, H. HelV. Chim. Acta 1995, 78, 910-
For a more quantitative approach, we expect that the observed
values of Aadd in Table 1 should show a correlation with an
appropriate steric factor. Two possible steric parameters have
9
8
7
24. (b) Martschke, R.; Farley, R. D.; Fischer, H. HelV. Chim. Acta 1997,
0, 1363-1374. (c) Weber, M.; Fischer, H. HelV. Chim. Acta 1998, 81,
70-780.
(
(
24) Fischer, H.; Paul, H. Acc. Chem. Res. 1987, 20, 200-206.
26
been considered: the traditional Taft steric parameter, ES, and
25) Zytowski, T.; Knuhl, B.; Fischer, H. HelV. Chim. Acta 2000, 83, 658-
2
7
6
75.
the ΩS scale of Sakakibara et al. (Table 3).
J. AM. CHEM. SOC.
9
VOL. 124, NO. 32, 2002 9493

A R T I C L E S
Beckwith and Poole
Figure 2. Correlation of observed Arrhenius preexponentials for addition of free radicals to methyl acrylate with the Taft steric parameter ES, and the ΩS
parameter of Sakakibara et al. Data shown are from this work (O) and from the literature (0, refs 18 and 23a). The values of ES and ΩS for 2g were assumed
to be equivalent to those of a neopentyl group.
Table 3. Values of log Aadd, ES, and ΩS for Selected Organic Free
universally for all radical-olefin pairs, particularly those for
which there are large polar effects, further examination of the
use of steric parameters as predictors of log Aadd in radical
reactions appears to be warranted.
Radicals
c
Ω d
S
radical
log Aadda,b
E
S
methyl (2a)
8.7
8.4
8.0
8.3
7.7
7.3
7.6
0
0.205
0.269
0.329
0.309
0.377
0.352
0.332e
0.281
0.259
1
1
-pentyl (2b)
-methylbutyl (2c)
-0.40
-1.13
-0.79
-1.54
-1.54
-1.74
-0.38
-0.94
Steric factors are also expected to affect the magnitude of
Eadd, since for bulky radicals the formation of the transition
structure should involve an increase in strain energy. The data
in Table 1 which show that Eadd decreases with increasing bulk
down the series 2a-e indicate that this is not the case for this
type of reaction. Clearly other factors come into play (see
below). However, the value of Eadd for adamantyl radical,
although suspiciously low, may reflect the fact that the SOMO
of this radical has a greater degree of s character than acyclic
alkyl radicals. Consequently, the reaction of 2f with methyl
acrylate should have a particularly early transition state and will
be especially insensitive to steric effects, leading also to a higher
cyclohexyl (2d)
,1-dimethylpropyl (2e)
1
tert-butyl
tert-butoxy (2g)
benzyl
e
f
8.5
8.4
g
cyanomethyl
a
b
From present work unless otherwise specified. Footnote c in Table
1
. c Reference 26. Reference 27a. Approximated by the value for the
neopentyl group. Reference 23a. Reference 18.
d
e
f
g
The derivation of these parameters is distinct: The Taft
parameter ES is defined in terms of strain energy (an enthalpic
property) but should still provide some indication of the extent
to which steric hindrance will affect the loss of translational
and rotational degrees of freedom in the transition structure. In
contrast, ΩS is obtained from molecular mechanics optimized
structures and represents the restriction of approach of the
reacting moieties due to the steric size of substituents (calculated
by combined van der Waals radii of attached groups). The
correlation of Aadd values obtained in this work, and those values
for benzyl and cyanomethyl radicals obtained by Fischer et al.,
with ES and ΩS are shown in Figure 2.
2
8
than expected value of log Aadd.
The selectivity of addition also reflects steric phenomena,
consistent with the qualitative observations of Tedder, Walton,
2
0,21
and Giese.
Addition adjacent to the functional group (so-
called “head” addition) is believed to be disfavored on steric
grounds, and in this study only the relatively sterically unde-
manding 1-pentyl radical was observed to add to methyl acrylate
in the “head” mode, although in insufficient yield to allow the
rate to be measured. The tert-butoxyl radical was also observed
to add in this mode, but this is due in large part to polarity
effects (see discussion below).
The correlations obtained are quite good, when one considers
that (a) the steric factor for 2g was approximated as being
equivalent to that of a neopentyl group and (b) both steric scales
are intended to model the effects of substituents attached to
carbons R to the reactive site, rather than substituents directly
attached to the active site. It should be noted that recently ΩS
values were calculated for alkyl radicals27b and that the small
Bond Strength and Stabilization Effects. The recent review
22
of Fischer and Radom has demonstrated that the activation
energies for the addition of a free radical to a range of olefins
may be well described by consideration of the components of
a state correlation model. We therefore seek to determine
whether the activation energies for addition of a range of free
radicals to a single olefin show similar correlations. As a first
approximation, we consider a Evans-Polyani-Semenov ap-
proach, where Eadd is plotted against ∆Hr for the addition
reaction. ∆Hr is calculated for the gas-phase reaction using a
group additivity approach to ∆Hr analogous to that of Fischer
•
number of species considered above for which ΩS values have
been calculated show a qualitatively similar correlation. Al-
though it is unlikely that such a close correlation will hold
(
26) (a) Exner, O. Correlation Analysis of Chemical Data; Plenum Press: New
York, 1988. (b) Taft, R. W. In Steric Effects in Organic Chemistry;
Newman, M. S., Ed.; Wiley: New York, 1956; pp 556-675.
2
2
and Radom. Thus ∆Hr(R,MA) is estimated by
(27) (a) Isizawa, J.; Sakakibara, K.; Hirota, M. Bull. Chem. Soc. Jpn. 1996, 69,
1
003-1015. (b) Iwao, K.; Sakakibara, K.; Hirota, M. J. Comput. Chem.
998, 19, 215-221.
1
(28) Walton, J. C. Chem. Soc. ReV. 1992, 105-112.
9494 J. AM. CHEM. SOC.
9
VOL. 124, NO. 32, 2002

Rates of Addition of Free Radicals to Alkenes
A R T I C L E S
•
Table 4. Bond Dissociation Enthalpies (BDE) and Ionization Potentials (IP(R )) for Selected Organic Free Radicals and Their Heats of
Reaction (∆Hr(R,MA)) and Activation Energies (Eadd) for Addition to Methyl Acrylate
a
BDEb
(kJ/mol)
(RH)b
IP(R•)c
(eV)
E
add
∆H°
(kJ/mol)
f
∆H°
f
(RCH
2
CH
3
)
r
∆H (R,MA)
(kJ/mol)
radical
(kJ/mol)
(kJ/mol)b
methyl (2a)
pentyl (2b)
17.8
14.9
11.0
10.2
6.4
439
420
411
400
404
403d
440
400
370
389
-74.9 ( 0.5
-146.8 ( 0.6
-146.8 ( 0.6
-123.1 ( 0.8
-153.7 ( 0.6
-134.4 ( 2.3
-312.6 ( 0.9
-438g
-104.7 ( 0.5
-187.8 ( 0.8
-192.3 ( 1.3
-171.8 ( 1.5
-201.2
-117
-109
-105
-97
9.84
7.94
7.41
7.66
6.65
6.21
1-methylbutyl (2c)
cyclohexyl (2d)
1
1
,1-dimethylpropyl (2e)
-adamantyl (2f)
-100
-108
-89
1.2
-190.4 ( 3.0e
-350.8 ( 2.6
-488h
tert-butoxy (2g)
-(methoxycarbonyl)heptyl (4b)
17.5
16.4
26.6j
f
7.70i
7.20
10.87
1
-98
benzyl
cyanomethyl
50.0 ( 0.6
-7.8 ( 0.8
-5.8 ( 1.0
-61
k
17.7
74.1 ( 0.4
-48
a
b
c
Data from the present study unless otherwise specified. Data from ref 30 otherwise specified. All data are calculated for gas phase and were obtained
from ref 29 unless otherwise indicated. Reference 31. Reference 32. Reference 33. g Modeled by methyl propanoate, heat of formation calculated at
HF/6-31G*.
d
e
f
3
4 h
Modeled by methyl isovalerate, heat of formation calculated at HF/6-31G*. Reference 25. Reference 23a. k Reference 18.
34 i
j
∆
H (R,MA) ) ∆H (RCH CH ) - ∆H (CH CH CH ) -
r f 2 3 f 3 2 3
∆
H (RH) + ∆H (CH ) - BDE(RH) + BDE(CH ) (5)
f f 4 4
where BDE is the C-H bond dissociation enthalpy for each
species. The value of ∆Hr for methyl radical addition to methyl
-
1
acrylate has been determined as -117 kJ mol , and the
remaining thermochemical data are available from NIST
databases29 or the literature.
in Table 4.
30-34
The relevant data are shown
Comparison of the data for Eadd and ∆Hr(R,MA) demonstrates
that no simple correlation exists between the activation energy
and the enthalpy of reaction and that the simple Evans-
26
Polyani-Semenov model does not apply. Fischer and Radom
have found that, for many addition reactions, an accurate model
for the rate coefficients cannot be constructed without consider-
ing polar effects.
Figure 3. Correlation of the observed activation energies for addition of
free radicals to methyl acrylate with the charge transfer state parameter
Polar Effects. Initially, the charge-transfer interactions for
•
26
IP(R ) - Eea(MA), with Eea(MA) ) -0.5 eV: (O) this work; (0) Fischer
et al. Values for cyanomethyl radical and 4b are gas-phase ionization
potentials calculated by ab initio MO methods. No experimental error bars
have been included for Eadd (see footnote f in Table 1).
such a system were described by a frontier molecular orbital
3
5
(
FMO) model, as developed by Fukui and successfully applied
21a
16-18,23-25
to similar systems by Giese and Fischer et al.
This
model indicates that the delocalization stabilization of a nu-
cleophilic radical-olefin system as it evolves from reactant to
transition state will depend on the energies of the olefin LUMO
and the radical SOMO, determined experimentally as the
electron affinity (Eea) and ionization potential (IP) of the olefin
and radical, respectively.
Because methyl acrylate is an olefin with a strongly electron
withdrawing group, the “tail” end of the olefin may be con-
sidered relatively electron poor. Stabilizing interactions will be
observed for nucleophilic radicals attacking the tail position and
electrophilic radicals attacking the head position. We thus expect
•
a strong correlation of Eadd with IP(R ) - Eea(MA) for
The state correlation model uses the same parameters to
model the energies of the charge transfer statessthe lower the
energy of the charge transfer states, the greater the configuration
mixing and stabilization of the ground state, and the favorable
interactions of the charge transfer states are those characterized
nucleophilic radicals, and such behavior is observed in Figure
3 for simple alkyl radicals. On the basis of the Fukui FMO
formalism, we would expect to see an even better correlation
with the inverse of the above charge transfer function, and this
is indeed observed for nucleophilic radicals. For electrophilic
radicals, the correlation is less marked as expected and is due
to the fact that the charge transfer states for electrophile-
electrophile systems are of higher energy. The same phenom-
enon may be alternatively described as a change in the exchange
•
by large values of the parameters IP(MA) - Eea(R ) or
•
IP(R ) - Eea(MA).
(
29) Thermochemical data may be found at the National Institute of Standards
and Technology (NIST) Chemistry WebBook: http//webbook.nist.gov.
30) (a) Kerr, J. A In CRC Handbook of Chemistry and Physics, 75th ed.; Lide,
D. R., Ed.; CRC Press: Boca Raton, FL, 1994; pp 9/63-69. (b) Wayner,
D. D. M.; Griller, D. AdV. Free Radical Chem. 1990, 1, 159-192.
3
5
integral â in the older Fukui FMO formalism.
The data for addition of the benzyl radical to methyl acrylate
(
2
3a
(
(
(
31) Aubry, C.; Holmes, J. L.; Walton, J. C. J. Phys. Chem. A 1998, 102, 1389-
measured by Fischer et al.
indicates the benzyl radical
1
393.
represents a special case with regard to polar effects, presumably
due to its high degree of stabilization, as indicated by Fischer
32) Melkhanova, S. V.; Pimenova, S. M.; Kolesov, V. P.; Pimerzin, A. A.;
Sarkisova, V. S. J. Chem. Thermodyn. 2000, 32, 1311-1317.
33) (a) Welle, F. M.; Beckhaus, H.; R u¨ chardt, C. J. Org. Chem. 1997, 62, 552-
22
and Radom.
5
58. (b) Brocks, J. J.; Beckhaus, H.-D.; Beckwith, A. L. J.; R u¨ chardt, C.
J. Org. Chem. 1998, 63, 1935-1943.
The correlation observed for the effect of philicity on the
rate of addition is almost certainly characteristic for methyl
(
34) Allinger, N. L.; Schmitz, L. R.; Motoc, I.; Bender, C.; Labanowski, J. K.
J. Comput. Chem. 1992, 13, 838-841.
35) Fukui, K. Fortschr. Chem. Forsch. 1970, 15, 1-85.
17,18,23-25
(
acrylate. Fischer et al.
have demonstrated that changes
J. AM. CHEM. SOC.
9
VOL. 124, NO. 32, 2002 9495

A R T I C L E S
Beckwith and Poole
in correlation between the electron affinities of olefins and the
observed activation energies for addition of a specific radical
In summary, the persistent aminoxyl method involving 1
provides reliable kinetic data for the addition of a variety of
simple and substituted alkyl radicals to methyl acrylate.
Furthermore, the results suggest that the method could be
particularly useful when more direct experimental methods are
not readily available and might be applicable to kinetic studies
of a much wider range of radical reactions, thus affording data
for testing the utility of modern theoretical approaches.
(in particular, the slope of the line of best fit for these
parameters) reflect the philicity of that radical. It is logical that
the converse should hold true. In addition, we have not
considered the steric component of Eadd in this analysis. It is
apparent, however, that the component will not be large for these
systems and that Eadd is dominated largely by polar effects in
the system under investigation. This is to be expected, as there
is significant separation (approximately 2.15 Å from MO
Experimental Section
2
1c,36
calculations)
of attacking radical and the olefin in the
Materials and Instrumentation. All solvents and synthetic precur-
sors, unless otherwise described, were obtained from commercial
sources and were used as received without further purification. Water
used in RPHPLC analysis was deionized by a Milli-Q filtration
apparatus. RPHPLC analysis was performed using an Alltech Alltima
C18 5 µm column (250 mm × 4.6 mm i.d.) with Waters 510 HPLC
pumps and a Waters 486 UV-visible detector set at 270 nm.
transition state. Of course, for truly bulky radicals such as
triphenylmethyl, a significant steric component in the observed
Eadd is expected.
The selectivity of addition is also affected by polar factors.
For example, the electrophilic tert-butoxyl radical exhibits
significant “head” addition, despite the relatively unfavorable
steric interactions. In addition, the head addition of pentyl radical
was observed (but could not be quantified). Unfortunately,
neither the thermochemical (ground state) or polar (charge
transfer states) correlations are capable of addressing the
question of regioselectivity in their current forms. It is possible,
however, that the regioselectivity of additions could be predicted
once the thermochemistry (∆HR) of head addition of methyl
radical to methyl acrylate could be determined.
1
13
H (300 MHz) and C (75 MHz) NMR spectra were obtained with
a Varian Gemini 300 FT instrument. FTIR spectra were obtained using
a Perkin-Elmer 1600 laser FTIR from solutions of the compounds in
CDCl , unless otherwise specified. UV-visible spectra were collected
3
digitally with a Shimadzu UV-2101 PC spectrometer. Mass spectra
were obtained with a Fisons VG Autospec mass spectrometer. Elec-
trospray MS were obtained using a Fisons VG Quattro II mass
spectrometer.
2
1
was synthesized as previously described and recrystallized twice
from hexane prior to use. Diacyl peroxides were obtained by standard
methods,37 and tert-butyl peresters were prepared by one of two
methods.
Conclusions
We have conducted a series of experiments designed to test
the use of radical coupling reactions of the aminoxyl 1 as kinetic
yardsticks for determining the rates of addition of free alkyl
radicals to olefins. The results demonstrate that when coupling
rates are calibrated by reference to the best available data, the
method is capable of affording rate coefficients and Arrhenius
parameters comparable with reliable literature values.
Method A. A solution of pyridine (1.7 mol equiv) and tert-butyl
hydroperoxide (1.2 mol equiv) in pentane (10 mL) was cooled to 4
°
C, whereupon a solution of the appropriate acyl chloride (1.0 mol
equiv) in pentane (10 mL) was added dropwise with vigorous stirring.
The reaction mixture was warmed to room temperature and stirred for
a further 2 h. The pentane solution was washed with water, 2 M H
SO solution, and saturated NaHCO
solution (1 × 10 mL each) and
dried (MgSO ) and the solvent removed under reduced pressure.
2
-
4
3
Although the set of data obtained in this work cannot be
readily tested against the kinetic model recently proposed by
4
Method B. Sodium tert-butylperoxide (2.0 mol equiv) was suspended
in dry ether (40 mL), and the resultant mixture cooled to 0 °C,
whereupon a solution of the appropriate acyl chloride (1.0 mol equiv)
in dry ether (10 mL) was added dropwise with vigorous stirring. The
solution was warmed to room temperature and stirred for a further 5
h. Water (10 mL) was added to the reaction mixture, which was stirred
22
Fischer and Radom, it appears that the results conform to long-
held views of the influence of polar, steric, and enthalpic effects
on the rates of radical addition reactions. Thus, for the addition
of nonbenzylic radicals to methyl acrylate, the activation energy
exhibits a strong correlation with parameters describing charge
transfer states (polar interactions), reflecting the nucleophilicity
of the attacking species. The observed values of log Aadd for
the addition reaction depend on radical structural features
expected to affect the change in motional degrees of freedom
upon formation of the transition structure. Interestingly, a good
correlation was observed with appropriate values of the Taft
steric parameter ES and, to a lesser extent, with the Sakakibara
ΩS parameter. Although these parameters have previously been
considered to reflect mainly enthalpic factors (e.g. strain energy),
it now appears that they may be useful measures of entropic
effects. Their potential as indicators of relative values of log A
for radical reactions in general and addition reactions in
particular deserves further exploration. Interestingly, kinetic data
available in the literature from work involving different
experimental approaches conform well to the same correlations
of log Aadd and Eadd as those that apply to our own results. This
provides further illustration of the reliability of our method.
38
overnight at 4 °C. The workup is identical to that of method A.
Aminoxyl Experiments. The following procedure for generation
of 2b in the presence of methyl acrylate and 1 is typical: a solution of
bis(hexanoyl)peroxide in n-heptane (10-50 µL) was added to a solution
of known concentrations of 1 and naphthalene (internal standard) in
methyl acrylate (5 mL). The combined solution was degassed using a
freeze-pump-thaw cycle (five cycles), sealed, and immersed in a
constant temperature oil bath for the requisite period of time, which
varied from 2 h at 353 K to 20 days at 298 K. The reaction mixture
was then subjected to RPHPLC analysis. [1]
0
and [1]final were
determined relative to naphthalene, and [MA] was assumed to be
equivalent to the value of neat solvent, corrected by a dilution factor
due to the addition of initiator solution (assuming that the final volume
of solution is equivalent to the sum of the volumes of the component
solutions). Finally, with the ratios of product alkoxylamines (3a, 5a,
and 7a) in hand, the roots of eqs 2a and 2b (R and â, as and where
applicable) were determined by Newton-Raphson techniques to yield
values for the effective relative rate coefficients. Where tert-butyl
peresters were used, the product distributions due to 2g were fitted
independently of the product distributions due to 2d-f to yield relative
(
36) (a) Wong, M. W.; Pross, A.; Radom, L. J. Am. Chem. Soc. 1994, 116,
6
284-6292. (b) Wong, M. W.; Pross, A.; Radom, L. J. Am. Chem. Soc.
(37) Kochi, J. K.; Mocadlo, P. E. J. Org. Chem. 1965, 30, 1134-1141.
(38) Bartlett, P. D.; Pincock, R. E. J. Am. Chem. Soc. 1962, 84, 2445.
1
994, 116, 11938-11943.
9496 J. AM. CHEM. SOC.
9
VOL. 124, NO. 32, 2002

Rates of Addition of Free Radicals to Alkenes
rate coefficients. However, [1] and [MA] were determined
A R T I C L E S
0
, [1]fina
,
l
(2 × s, 3H + 9H) overlying 1.20-1.55 (m, 4H); 0.97, 0.96 (2 × d,
exactly as described above for both initiator fragments.
J ) 6.7, 6.4 Hz, 3H); 0.89 (t, J ) 6.8 Hz, 3H) ppm. 13C NMR (CDCl
,
3
Authentic samples of alkoxylamine adducts were obtained from
75 MHz): δ 174.4, 174.2, 144.9, 144.6, 127.2, 127.2, 127.1, 127.1,
121.5, 121.3, 83.6, 83.2, 68.2, 68.1, 67.6, 67.5, 51.3, 39.6, 39.4, 39.4,
39.2, 30.4, 29.5, 25.5, 25.0, 28.9, 28.9, 19.8, 19.7, 14.2, 14.1 ppm.
semipreparative HPLC of combined reaction mixtures.
1
2
-Pentoxy-1,1,3,3-tetramethylisoindoline (3b). H NMR (CDCl
3
,
-
1
3
1
00 MHz): δ 7.23 (m, 2H); 7.10 (m, 2H); 3.92 (t, J ) 6.6 Hz, 2H);
.63 (br. tt, J ) 6.6, 7.1 Hz, 2H); 1.43 (br. s, 18H); 0.94 (t, J ) 6.9
FTIR (CDCl
332 (39), 190 (100), 176 (34), 160 (29), 158 (23), 145 (23). HRMS:
27NO requires m/z 347.2462, found m/z 347.2460.
2-[2-Cyclohexyl-1-(methoxycarbonyl)ethoxy]-1,1,3,3-tetrameth-
3
): ν 2958, 2930, 2872, 1741 cm . EIMS: m/z 347 (16),
1
3
Hz, 3H) ppm. C NMR (CDCl
3
, 75 MHz): δ 145.2, 127.0, 121.4,
7.4, 67.0, 28.9, 28.6, 22.6, 14.0 ppm. FTIR (CDCl ): ν 2963, 2933,
872 cm . EIMS: m/z 261 (18), 246 (100), 190 (22), 176 (82), 160
17
C H
7
2
(
3
-
1
1
ylisoindoline (5d). H NMR (CDCl , 300 MHz) δ 7.24 (m, 2H); 7.08
3
23), 158 (27), 145 (28). HRMS: C17H27NO requires m/z 261.2093,
(m, 2H); 4.62 (t, J ) 6.9 Hz, 1H); 3.75 (s, 3H); 1.59 (s, 3H), overlying
1.57-1.83 (m, 7H); 1.44, 1.42, 1.41 (3s, 9H); 1.25 (m, 4H); 0.94 (m,
found m/z 261.2091.
1
13
2-(1-Methylbutoxy)-1,1,3,3-tetramethylisoindoline (3c). H NMR
2H) ppm. C NMR (CDCl , 75 MHz): δ 174.2, 144.8, 144.5, 127.2,
3
(
CDCl
3
, 300 MHz): δ 7.24 (m, 2H); 7.11 (m, 2H); 3.93 (tq, J ) 6, 6
127.1, 121.5, 121.2, 82.9, 68.3, 67.7, 51.3, 39.8, 33.8, 33.4, 33.3, 30.4,
Hz, 1H); 1.72 (m, 1H); 1.53, 1.50, 1.38, 1.32 (4 × br s, 12H, overlying
29.4, 26.3, 26.1, 26.0, 25.5, 25.0 ppm. FTIR (CDCl ): ν 2926, 2853,
1744 cm . EIMS: m/z 359 (16), 344 (40), 190 (100), 176 (37), 160
3
-
1
1
.50-1.40 (m, 3H)); 1.25 (d, J ) 6.1 Hz, 3H); 0.97 (t, J ) 7.2 Hz)
1
3
ppm. C NMR (CDCl
3
, 75 MHz): δ 145.4, 145.4, 126.9, 121.4, 78.6,
(32), 158 (23), 145 (25). HRMS: C H NO requires m/z 359.2460,
22
33
3
6
7.5, 67.1, 38.3, 30.5, 30.0, 25.2, 25.2, 19.8, 18.9, 14.3 ppm. FTIR
found m/z 359.2466.
-[1-(Methoxycarbonyl)-3,3-dimethylpentoxy]-1,1,3,3-tetrameth-
-
1
(CDCl
3
): ν 2966, 2932, 2873 cm . EIMS: m/z 261 (17), 246 (8),
2
1
91 (37), 176 (100), 160 (24), 158 (26), 145 (23). HRMS: C17
H27NO
1
ylisoindoline (5e). H NMR (CDCl
3
, 300 MHz): δ 7.23 (m, 2H); 7.08
requires m/z 261.2093, found m/z 261.2086.
-(Cyclohexyloxy)-1,1,3,3-tetramethylisoindoline (3d). H NMR
CDCl , 300 MHz): δ 7.22 (m, 2H); 7.10 (m, 2H); 3.70 (m, 1H); 2.10
m, 2H); 1.77 (m, 2H); 1.50 (s, 6H); 1.35 (s, 6H), overlying 1.24-
(
m, 2H); 4.48 (dd, J ) 5.6, 7.4 Hz, 1H); 3.71 (s, 3H); 1.80 (d, J ) 7.5
1
2
Hz, 1H); 1.78 (d, J ) 5.5 Hz, 1H); 1.59 (s, 3H); 1.40, 1.38, 1.35 (3s,
underlain by multiplet 1.35-1.26 ppm, 11H); 0.92, 0.90 (2s, 6H); 0.86
(
1
3
2
1
found m/z 347.2461.
-[2-(1-Adamantyl)-1-(methoxycarbonyl)ethoxy]-1,1,3,3-tetra-
methylisoindoline (5f). H NMR (CDCl
(
(
3
13
t, J ) 7.4 Hz) ppm. C NMR (CDCl
3
, 75 MHz): δ 174.3, 144.8,
13
1
8
2
1
.44 (m, 6H) ppm. C NMR (CDCl
3
, 75 MHz): δ 145.3, 127.0, 121.5,
): ν 2975,
934, 2857 cm . EIMS: m/z 273 (22), 258 (9), 191 (56), 177 (61),
44.6, 127.2, 127.1, 121.4, 121.3, 81.6, 68.0, 67.7, 51.3, 42.9, 34.6,
1.6, 67.2, 32.5, 30.3, 25.9, 25.2, 24.4 ppm. FTIR (CDCl
3
2.0, 30.9, 29.2, 26.8, 25.6, 25.1, 8.3 ppm. FTIR (CDCl ): ν 2969,
3
-
1
-1
933, 2880, 1741 cm . EIMS: m/z 347 (10), 332 (46), 190 (100),
76 (100), 160 (26), 158 (39), 145(40), 144 (37). HRMS: C18
H27NO
3
76 (47), 160 (29), 145 (19). HRMS: C21H33NO requires m/z 347.2460.
requires m/z 273.2093, found m/z 273.2091.
1
2
-(1,1-Dimethylpropoxy)-1,1,3,3-tetramethylisoindoline (3e). H
2
NMR (CDCl
3
, 300 MHz): δ 7.23 (m, 2H); 7.09 (m, 2H); 1.62 (q, J )
1
3
, 300 MHz): δ 7.23 (m, 2H);
.07 (m, 2H); 4.53 (dd, J ) 8.2, 4.7 Hz, 1H); 3.72 (s, 3H); 1.95 (m,
H); 1.49-1.76 (m, 14H); 1.41, 1.39 (2s, 6H); 1.34 (s, 3H); 1.26 (s,
7
.6 Hz, 2H); 1.48 (s, 6H); 1.32 (s, 6H); 1.25 (s, 6H); 0.96 (t, J ) 7.6
7
3
3
1
2
1
3
Hz, 3H) ppm. C NMR (CD
7
2
2
Cl
2
, 75 MHz): δ 145.9, 127.1, 121.8,
): ν 2975,
930 cm . EIMS: m/z 261 (13), 245 (11), 190 (43), 175 (100), 157
8.8, 68.2, 35.5, 30.6, 26.2, 25.7, 9.2 ppm. FTIR (CDCl
3
13
H) ppm. C NMR (CDCl
27.1, 121.4, 121.3, 80.6, 68.0, 67.7, 51.3, 46.3, 42.3, 36.7, 31.5, 30.9,
9.1, 28.5, 25.6, 25.1 ppm. FTIR (CDCl ): ν 2973, 2905, 2849, 1739
3
, 75 MHz): δ 174.3, 144.8, 144.5, 127.2,
-
1
(19), 145 (18). HRMS: C17H27NO requires m/z 261.2093, found m/z
3
2
61.2089.
cm-1. EIMS: m/z 411 (13), 396 (28), 190 (100), 175(23), 160 (21),
1
2-(1-Adamantyloxy)-1,1,3,3-tetramethylisoindoline (3f). H NMR
1
4
35 (28). HRMS:
11.2774.
26 3
C H37NO requires m/z 411.2773, found m/z
(
CDCl
3
, 300 MHz): δ 7.23 (m, 2H); 7.07 (m, 2H); 2.18 (br s, 3H);
1
1
7
2
.87 (br d, J ) 3.2 Hz, 6H); 1.65 (br d, J ) 2.8 Hz, 6H); 1.49 (s, 6H);
2
-[4,4-Dimethyl-1-(methoxycarbonyl)pent-1-oxy]-1,1,3,3-tetra-
1
3
.34 (s, 6H) ppm. C NMR (CDCl
3
, 75 MHz): δ 145.6, 126.9, 121.6,
): ν
976, 2914, 2853 cm . EIMS: m/z 325 (2), 176 (10), 135 (100).
31NO requires m/z 325.2406, found m/z 325.2406.
-[1-(Methoxycarbonyl)heptanoxy]-1,1,3,3-tetramethylisoindo-
1
methylisoindoline (5g). H NMR (CDCl
H); 7.08 (m, 2H); 4.59 (dd, J ) 5.2, 7.2 Hz, 1H); 3.77 (s, 3H); 3.71
dd, J ) 7.2, 9.4 Hz, 1H); 3.60 (dd, J ) 5.2, 9.4 Hz, 1H); 1.57 (s, 3H);
.42 (s, 6H); 1.39 (s, 3H); 1.21 (s, 9H) ppm. 13C NMR (CDCl
, 75
MHz): δ 172.53, 144.8, 127.1, 121.5, 121.3, 85.8, 73.4, 68.2, 67.5,
3
, 300 MHz): δ 7.22 (m,
5.2, 67.8, 46.3, 43.2, 36.4, 31.0, 30.6, 25.7 ppm. FTIR (CDCl
3
2
(
1
-
1
HRMS: C22
H
3
2
1
line (5b). H NMR (CDCl
3
, 300 MHz): δ 7.22 (m, 2H); 7.09 (m, 2H);
.43 (t, J ) 6.6 Hz, 1H); 3.76 (s, 3H); 1.95-1.65 (m., 2H);
.56,1.43,1.41 (3 × s, 12H) overlying 1.20-1.55 (m, 8H)); 0.91 (t,
, 75 MHz): δ 174.0, 144.9,
44.6, 127.2, 127.1, 121.5, 121.2, 85.0, 68.0, 67.6, 51.3, 32.2, 31.5,
9.5, 25.4, 25.0, 29.1, 25.2, 22.5, 14.0 ppm. FTIR (CDCl ): ν 2956,
928, 2858, 1741 cm . EIMS: m/z 347 (12), 332 (58), 190 (100),
76 (38), 160 (28), 158 (24), 145 (23). HRMS: C17 27NO requires
6
2
1.9, 51.5, 29.7, 29.5, 27.3, 25.1, 25.0 ppm. FTIR (CDCl
933, 2874, 1744 cm . EIMS: m/z 349 (5), 348 (4), 335 (39), 334
3
): ν 2977,
4
1
-
1
(90), 191 (47), 190 (93), 176 (96), 160 (100), 145 (47), 144 (43).
1
3
J ) 6.7 Hz, 3H) ppm. C NMR (CDCl
3
HRMS: C20
NO
H
31NO
4
requires m/z 349.2253, found m/z 349.2240; C20
30
H -
1
2
2
1
+
4
(M - 1) requires m/z 348.2175, found m/z 348.2173.
3
-
1
Acknowledgment. J.S.P. wishes to thank the Commonwealth
Government of Australia for funding in the form of an
Australian Postgraduate Research Award.
H
m/z 347.2460, found m/z 347.2464.
-[1-(Methoxycarbonyl)-3-methylhexanoxy]-1,1,3,3-tetramethyl-
2
isoindoline (5c). The obtained product was a mixture of diastereomers,
Supporting Information Available: 1H and 13C NMR spectra
of isolated alkoxylamines. This material is available free of
charge via the Internet at http://pubs.acs.org.
not readily separable under the chromatographic conditions used for
1
kinetic analysis of the reaction mixtures. H NMR (CDCl
3
, 300 MHz):
δ 7.22 (m, 2H); 7.09 (m, 2H); 4.43 (2 × overlying dd, J ) 7 Hz, 1H);
3
.74 (2 × s, 3H); 1.95-1.82 (m, 1H); 1.80-1.60 (m, 2H); 1.55, 1.38
JA025730G
J. AM. CHEM. SOC.
9
VOL. 124, NO. 32, 2002 9497