Long-range polar and steric effects in propionate-SG1-type alkoxyamines (SG1-CHMeCOOX): A multiparameter analysis

Article abstract of DOI:10.1002/poc.1035

The effects of the substituent X on the homolysis rate constants (k _d) of SG1-propionate type alkoxyamines (SG1-CHMeCOOX) are analyzed by a multiparametric equation with v, the steric constant and σ_I, the polar inductive/field Hammett constant of X. An influence of long-range polar and steric effects on k_d was observed, that is, decrease in k_d with increasing size of the X group and increase in k_d with increasing polarity of the X group. Copyright

Full text of DOI:10.1002/poc.1035

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2006; 19: 269–275
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.1035
Long-range polar and steric effects
in propionate-SG1-type alkoxyamines
(SG1-CHMeCOOX): a multiparameter analysisy
Gennady Ananchenko,1 Emmanuel Beaudoin,2 Denis Bertin,2 Didier Gigmes,2 Pierre Lagarde,2
2
2
Sylvain R. A. Marque,2 Eve Revalor and Paul Tordo
*
1NRC Steacie Institut for Molecular Sciences, 100 Sussex Dr., Ottawa, ON, Canada
2UMR 6517 case 542, Universite´ d’Aix—Marseille I–III, Avenue Escadrille Normandie-Niemen, 13397 Marseille Cedex 20, France
Received 8 November 2005; revised 16 January 2006; accepted 6 February 2006
ABSTRACT: The effects of the substituent X on the homolysis rate constants (kd) of SG1-propionate type
alkoxyamines (SG1-CHMeCOOX) are analyzed by a multiparametric equation with y, the steric constant and sI,
the polar inductive/field Hammett constant of X. An influence of long-range polar and steric effects on kd was
observed, that is, decrease in kd with increasing size of the X group and increase in kd with increasing polarity of the X
group. Copyright # 2006 John Wiley & Sons, Ltd.
Supplementary electronic material for this paper is available in Wiley Interscience at http://www.interscience.
wiley.com/jpages/0894–3230/suppmat/
KEYWORDS: alkoxyamines; multiparameter analysis; long-range effects; polar and steric effects; homolysis
INTRODUCTION
(heteroatom bonded to the carbon) and polar11,22,26,28,29
(electron withdrawing groups (EWG) bonded to the
nitrogen atom) effects or weakened by the stabiliz-
ation11,17,21,23,24,26,29 of the released alkyl and nitroxyl29
(intramolecular hydrogen bond IHB) radicals, and by
the steric strain and polar effects of both alkyl and
nitroxyl fragments.11,13–29 Recently, Ananchenko et al.30,31
observed a decrease in kd with bulkier alkyl groups. Thus,
we investigated the influence of this long range steric
effect and observed that the values of kd decreased linearly
(Eqns 1(a) and 1(b)) with the increasing size of the
substituent X of SG1-CHMeCOOX alkoxyamines 1–9
(Scheme 2).32 Furthermore, it was noted that alkoxya-
mines 13 and 14 (SG1-CHMeCOOPh-4-Z type, Scheme 2)
exhibited long-range polar effect, that is kd increased with
the increasing electron-withdrawing capacities of the
Z groups.32 A thorough study of the series 10–17 showed a
clear influence of the polarity of the benzene ring para-
substituent on kd but the importance of the effect, that is the
slope of the regression, was different for each isomer
(Eqns 2(a) and 2(b)).33 Such difference can be ascribed
either to a different steric effect of the aromatic ring,34,35
for each isomer or to a polar effect depending on the
conformation36–38 of the alkoxyamine, or even to
stereoelectronic effects.39 Aiming to discriminate between
these three possibilities, in this work we prepared
alkoxyamines 18–22 containing halogenoalkyl groups
(CH2CH3ꢀnXn, X ¼ F, Cl, and Br, Scheme 2). The
homolysis rate constants kd of both isomers were measured
A decade ago, Rizzardo1 and Georges2 showed that it was
possible to prepare well-defined living/controlled poly-
mers using nitroxyl radicals as controllers. Nitroxide
mediated polymerization (NMP) was born3,4 and follow-
ing this pioneering work, numerous studies have been
carried out to elucidate the mechanism5 and the kinetics
of NMP processes,6–8 to prepare new materials3,9,10 and
to develop efficient initiators/controllers.11–16 Scheme 1
displays the ideal NMP process, with kd the rate constant
for the homolysis of the C—ON bond of the alkoxyamine
(so-called dormant species), kc the rate constant for the
reformation of the alkoxyamine, kp the propagation rate
constant of the polymerization and kt the self-termination
rate constant of the reactions yielding dead polymers.
Alkoxyamines (R1R2NOR3) are the key intermediates5
of a NMP process and the strength of the C—ON bond is a
crucial parameter to control.6,11,15,16 It has been shown that
the activation energy (Ea) of their homolysis is a good
estimate of the value of the bond dissociation energy
(BDE) of the C—ON bond.17,18 We15,17,19–22 and
others11,23–29 have shown that the C—ON bond of
alkoxyamines was either strengthened by anomeric26,29
*Correspondence to: Sylvain R. A. Marque, UMR 6517 case 542,
´
Universite d’Aix—Marseille I–III, Avenue Escadrille Normandie-
Niemen, 13397 Marseille Cedex 20, France.
E-mail: Sylvain.marque@up.univ-mrs.fr
yIn memoriam of Prof. Hanns Fischer, deceased 22 February 2005
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 269–275

270
G. ANANCHENKO ET AL.
For RS/SR isomers:
kd
sꢀ1
log
¼ ꢀ3:6ðꢁ0:1Þ þ 2:3ðꢁ0:3Þ ꢂ sI;4ꢀXC H (2b)
6
4
EXPERIMENTAL SECTION
Scheme 1
Solvents for synthesis, triethylamine, 4-(dimethylami-
no)pyridine (DMAP), thionyl chloride (SOCl2), and
halogenoalcohols were purchased from Aldrich and used
as received. TEMPO was purchased from Acros and
sublimed. tert-Butylbenzene was purchased from Aldrich
and purified by a conventional procedure.40 Amine 23
was prepared as previously described12 and used as
internal standard for kinetic experiments. Nitroxyl radical
24 (SG1, Scheme 2) was kindly provided by Arkema and
alkoxyamine 25 (Scheme 2) was prepared as already
described.32 Reactions were monitored by TLC (60 F 240
silica gel plates, eluent ethyl acetate/pentane 1/1), with
UV and phosphomolybdic acid detection. Alkoxyamines
were purified by chromatography (60 Silica gel, 70–230
and correlated to the steric (Es) and polar (sI) Hammet
constants.
For RR/SS isomers:
kd
log
¼ ꢀ3:07ðꢁ0:04Þ þ 0:13ðꢁ0:02Þ ꢂ Es (1a)
sꢀ1
For RS/SR isomers:
kd
log
¼ ꢀ2:59ðꢁ0:03Þ þ 0:14ðꢁ0:02Þ ꢂ Es (1b)
sꢀ1
For RR/SS isomers:
kd
sꢀ1
log
¼ ꢀ4:3ðꢁ0:1Þ þ 4:3ðꢁ0:2Þ ꢂ sI;4ꢀXC H (2a)
6
4
Scheme 2
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 269–275

LONG-RANGE POLAR AND STERIC EFFECTS
271
mesh, Merck), eluent: ethyl acetate/pentane 3/1. NMR
experiments were performed in CDCl3 on a 300 Avance
washed three times with 15 mL of NH4Cl 5% aqueous
solution, three times with 15 mL of a saturated sodium
carbonate aqueous solution and with water to neutral pH.
The organic layer was dried over anhydrous MgSO4 and
the solvent removed to yield oil which was purified by
column chromatography to afford alkoxyamines 18–22
(see SI).
Bruker spectrometer (1H 300 MHz, 13C 75.48 MHz and
31
ˆ
P 121.59 MHz) in the Spectropole (Marseille) where the
elemental analyses were also performed. Chemical shifts
were given with TMS as internal reference for 1H NMR,
CDCl3 (internal reference) for 13C NMR and H3PO4 85%
(external reference) for 31P NMR. The experimental
procedure for kinetic measurements by 31P-NMR
spectroscopy has been described earlier,41tert-butylben-
zene was used as a solvent and a twofold excess of
TEMPO was employed as radical trap.
RESULTS
Alkoxyamines 18–22 were prepared as depicted in
Scheme 3.32,33 Since each of these alkoxyamines exhibits
two chiral centers (see Scheme 2), two isomers were
obtained for each molecule. Their 31P NMR signals were
not significantly different from those of series 1–17.
Therefore, with the absolute configurations attributed to 9
and 1742 given by the X-ray structures, the 31P NMR
shifts of ca. 24.7 and 24.0 ppm correspond to the RR/SS
and RS/SR diastereoisomers, respectively (See Exper-
imental and Table 1).32,33
General procedure
A solution of 25 (4.0 g, 11.0 mmol) in CH2Cl2 was
degassed by nitrogen bubbling for 10 min. Then, 3
equivalents SOCl2 (2.4 mL, 33.0 mmol) were added under
nitrogen atmosphere. The mixture was stirred for 45 min
at room temperature and the excess of SOCl2 was
removed under vacuum (0.1 mb) to yield alkoxyamine 29.
Crude 29 was diluted in ether and then a 20 mL ether
solution of alcohol (2 eq.), Et3N (1 eq., 1.5 mL, 11 mmol)
and DMAP (0.4 eq., 0.3 g, 2.4 mmol) was added under
nitrogen atmosphere. A white solid precipitated and the
mixture was stirred for 4 h at room temperature. The
white solid was filtered off and the solvent removed to
yield oil. The oil was dissolved in 30 mL of ether, then
The kd of both diastereoisomers were measured in a
single 31P NMR by monitoring the decay of the
alkoxyamine, and given by Eqn (3).41
½alkoxyamineꢃt
ln
¼ ꢀkd ꢄ t
(3)
½alkoxyamineꢃ0
Activation energies Ea were estimated using the
average frequency factor 2.4 ꢂ 1014 sꢀ1 defined in the
Scheme 3
Table 1. 31P NMR shifts d, homolysis rate constants kd, activation energies Ea for alkoxyamines 18–22
d (ppm)a kd (10ꢀ4 sꢀ1 c Ea (kJ/mol)d
)
kd (10ꢀ4 sꢀ1
)
120
RR/SS
SR/RS
24.2
24.2
24.0
24.6
24.1
T (8C)b
RR/SS
SR/RS
RR/SS
SR/RS
127.6
126.7
125.3
126.4
127.3
RR/SS
10.2
17.7
22.6
16.6
12.6
SR/RS
26.3
34.6
53.2
38.0
28.8
18
19
20
21
22
24.6
24.4
24.3
24.9
24.5
110
111
110
107
111
109
109
107
111
109
3.4
4.2
6.1
4.8
10.5
6.2
5.3
4.2
5.0
3.9
8.4
13.0
12.6
9.0
23.8
16.1
13.3
9.4
130.7
128.9
128.1
129.1
130.0
12.1
9.0
a Amine 23 as internal standard in C6D6/tert-butylbenzene d ¼ 31.0 ppm.
b Temperature ꢁ 18C.
c Errors were less than 5%.
d Ea ꢁ 2 kJ molꢀ1
.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 269–275

272
G. ANANCHENKO ET AL.
literature,17,29 and then all the kd were re-estimated at
120 8C to facilitate the comparison with published data.
There are more than 40 Arrhenius parameters available in
the literature, and most of them are in the region
dependent only on the position of the COOR group and
not on the position of the CH2CH3ꢀnXn groups. To avoid
using too many parameters, and because the Es and y
values are highly correlated (Eqn (6)),49 the y values were
used to define the new set of steric constants. Thus, new
constants yCHMeCOOX and sI,CHMeCOOX have to be
developed to account for the steric and polar influence
of the X groups of the ester moiety (Eqns (8) and (9)) with
j and k the correction terms for the steric and polar effects
of the X group, respectively.
5 ꢂ 1013 sꢀ1–5 ꢂ 1014 sꢀ1 43
.
Because of the A–Ea com-
pensation error effect,44 it is very difficult to determine
small changes in A values; therefore, it is easier to use an
averaged frequency factor to determine Ea values, and
then to estimate kd at the suitable temperature for
comparison. Although all the Ea values shown in Table 1
are within the experimental error (ꢅ2 kJ molꢀ1), the
values of kd are accurate enough (less than 10% error) for
the discussion. Furthermore, for alkoxyamines 10–17, it
was shown that the plot kd versus sI at 120 8C afforded
similar regression coefficients than the plot at 110 8C,
log kd ¼ ꢀ14:3ðꢁ1:3Þ þ 15:3ðꢁ2:2Þ ꢂ sRS
þ 19:5ðꢁ3:0Þ ꢂ sI þ 7:0ðꢁ1:1Þ ꢂ y (7)
R2 ¼ 0:85
N ¼ 19
s ¼ 0:80
F99;99 ¼ 29
supporting the use of A ¼ 2.0 ꢂ 1014
s
ꢀ1 to estimate kd at
temperature close to the experimental one.33 The plot of
kd versus (sI, Es) for 1–9, and 18–22 (not reported), at
110 8C, afforded close regression coefficients. Values of
kd, 31P NMR shifts, temperatures, kd at 120 8C and Ea
values of 18–22 are gathered in Table 1. The polar45 (sI)
and steric35 (Es) constants of CH2CH2F, CH2CHF2,
CH2CF3, CH2CCl3, and CH2CBr3 were estimated with
Eqns (4)46 and (5),35 respectively. They are stored in
Table 1SI along with the Charton47,48 (y) steric constants,
and the missing y for CH2CH2F, CH2CHF2, CH2CF3,
CH2CCl3, and CH2CBr3 groups are given by Eqn (6).49
Assuming that Eqn (7) holds for the series 1–22 and
that the stabilizing effect is not affected by the ester
substituent,33 and using alkoxyamines 2 and 9 (same
polar effect) with yCHMeCOOMe (1.0) and sI,CHMeCOOMe
(0.09) as references, the values of ꢀ0.103 and ꢀ0.095
for j of RR/SS and RS/SR diastereoisomers, respectively,
were estimated with Eqns 10–14. Because the slopes of
log (kd) versus sI for the two isomers are too different
from one another (Eqns 2(a) and 2(b)) and because the
values of Es and y for the phenyl group are not clearly
established,34,35,47,49 the series 10–17 was not used to
determine the values of k. The values of 0.283 and 0.251
for k of RR/SS and RS/SR diastereoisomers, respectively,
were estimated with Eqns (15)–(18), when applied to
alkoxyamines 9 and 20. For the sake of simplicity,
averaged values of ꢀ0.099 and 0.267 were used for j
and k, respectively.
sIðCH2RÞ ¼ 0:416 ꢂ sIðRÞ ꢀ 0:0103
(4)
EsðCR1R2R3Þ ¼ ꢀ 2:104 þ 3:429 ꢂ EsðR1Þ þ 1:978
ꢂ EsðR2Þ þ 0:649 ꢂ EsðR3Þ
(5)
y ¼ 0:502 ꢀ 0:477 ꢂ Es
(6)
Fujita et al.35 showed that the steric effect can be
represented by a linear combination (Eqn (5)) of the
individual Taft steric constants for a group R, noted
Es(Ri), with i being the rank related to the value of
Es(Ri), that is its size (the more negative Es(Ri) is, the
larger Ri is).
yCHMeCOOX ¼ yCHMeCOOMe þ j ꢂ yX ꢀ j ꢂ yMe (8)
sI;CHMeCOOX
¼ sI;CHMeCOOMe þ k ꢂ sI;X ꢀ k ꢂ sI;Me
(9)
log kd;9 ¼ log k0 þ rRS ꢂ sRS;CHMeCOOMe þ rI
ꢂ sI;CHMeCOOMe þ d ꢂ yCHMeCOOMe
Ea values of alkoxyamines 1–17 were determined in
our previous works.32,33Ea values, Taft (Es) and Charton
(y) steric constants, polar (sI and sI,Ph-4-Z) Hammett
(10)
constants, steric (yCHMeCOOX) and polar (sI,CHMeCOOX
)
log kd;2 ¼ log k0 þ rRS ꢂ sRS;CHMeCOOtꢀBu þ rI
constants of the CHMeCOOX groups in alkoxyamines
1–22 are listed in Table 1SI. In a recent paper,21 we
provided an analysis of the kd values in terms of
stabilizing (sRS), polar (sI) and steric (y) effects
(Eqn (7)). It is generally accepted that the size and the
polarity of the alkoxy group of the ester function exert
only a weak influence – if any – onto the reactive center.50
Despite the presence of long-range (4–7 s bonds) polar
and steric effects, it would be interesting to correlate
series 1–22 to Eqn (7). It was assumed that the
halogenoalkyl groups exhibited the same steric effect
for both diastereoisomers and that the polar effect was
ꢂ sI;CHMeCOOtꢀBu þ d ꢂ yCHMeCOOtꢀBu (11)
D log kd
¼ yCHMeCOOMe ꢀ yCHMeCOOtꢀBu
(12)
d
yCHMeCOOtꢀBu
¼ yCHMeCOOMe þ j ꢂ ytꢀBu ꢀ j ꢂ yMe
(13)
(14)
D log kd
j ¼
dðyMe ꢀ ytꢀBu
Þ
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 269–275

LONG-RANGE POLAR AND STERIC EFFECTS
273
log kd;20 ¼ log k0 þ r ꢂ s
mind that the distance between the halogen atom(s) and
the reactive center is 5 s-bonds, the comparison of these
estimated values with the experimental values (Table 1SI)
shows that the polar effect increases from weak in 18 (kd
measured ca. 1.5 times as high as kd estimated) to strong
in 20 (kd measured ca. 4.5 times as high as kd estimated).
This increase is even more striking (8 and 4 times
stronger) when the estimated kd of the isomers of 17 (5.0
and 14.0 ꢂ 10ꢀ4 sꢀ1) was compared with the experimen-
tal kd values (38.0 and 56.4 ꢂ 10ꢀ4 sꢀ1). Thus, the values
of kd were analyzed as a linear combination of the long-
range polar and steric effects (Eqn (19)).
RS;CHMeCOOCH2CF3
RS
þ rI ꢂ sI;CHMeCOOCH CF þ d
3
2
ꢂ yCHMeCOOCH CF
(15)
(16)
2
3
D log kd ꢀ djðyMe ꢀ yCH CF
Þ
2
3
r1
¼ sI;CHMeCOOMe ꢀ sI;CHMeCOOCH CF
3
2
s
I;CHCOOCH2CF3
¼ sI;CHCOOMe þ k ꢂ sI;CH CF ꢀ k ꢂ sMe
(17)
(18)
3
2
logðkd=sꢀ1
Þ
D log kd ꢀ djðyMe ꢀ yCH CF
Þ
2
3
k ¼
¼ log k0 þ rI ꢂ sI þ d ꢂ Esðor d0 ꢂ yÞ
(19)
rIðsMe ꢀ sI;CH CF
Þ
3
2
Because the phenyl ring is a steric ‘‘Janus’’ group,34,35
the value of Es (Es ¼ ꢀ1.8, Table 1SI) was chosen to yield
the best fit (Eqns 23(a) and 23(b), Table 2).
As expected, the plots log (kd/sꢀ1) versus sI (not
shown, Eqns 20(a) and 20(b) in Table 2) and log (kd/sꢀ1
)
versus Es (Fig. 1, Eqns 21(a) and 21(b) in Table 2) display
scattered dots for the alkoxyamines of series 1–9 and
18–22. On the other hand, a biparameter (Es, sI)
correlation exhibits good statistical output (Eqns 22(a)
and 22(b), Table 2) and accounts for the influence of
the sizes and the polarities of the ester groups for 1–9 and
18–22 (Fig. 2). The coefficients rI obtained in Eqns 22(a)
and 22(b) are different for each isomer and different from
those in Eqns 2(a) and 2(b) while coefficients d are the
same in Eqns 1(a) and 1(b), and Eqns 22(a) and 22(b). The
close values of DrI for the series 1–9 and 18–22
(DrU ¼ 0.82) and for the series 10–17 (DrU ¼ 1.0) leads
to discard the possibly enhanced polar effect due to the
DISCUSSION
When the kd of 18–22 are compared to those of 9
(Table 1SI), the importance of the polar effect is not
striking, that is 20 decomposes roughly two times faster
than 9. On the other hand, assuming the absence of any
polar effect, Eqns 1(a) and 1(b) give 6.8, 5.9, 5.1, 4.3, and
4.0 ꢂ 10ꢀ4 sꢀ1
, and 20.3, 17.5, 14.8, 12.2, and
11.3 ꢂ 10ꢀ4 sꢀ1 as values of kd for RR/SS and RS/SR
isomers of alkoxyamines 18–22, respectively. Keeping in
Table 2. Coefficients rI, d, and d0 for the linear combinations of kd (Eqn (19)) at 1208C with the molecular descriptors for the
polar (sI) and the steric (Es or y) effects for both isomers (a for RR/SS and b for RS/SR) of series 1–22, and the statistical outputs
c
Equations
log kd,0
rIa
db
d0
nd
se
R2f
tg
Fh
20a
21a
22a
23a
24a
25a
26a
20b
21b
22b
23b
24b
25b
26b
ꢀ3.20 (ꢁ0.06)
ꢀ2.96 (ꢁ0.11)
ꢀ3.04 (ꢁ0.04)
ꢀ3.04 (ꢁ0.03)
ꢀ2.99 (ꢁ0.17)
ꢀ2.93 (ꢁ0.06)
ꢀ2.93 (ꢁ0.05)
ꢀ2.73 (ꢁ0.06)
ꢀ2.50 (ꢁ0.09)
ꢀ2.57 (ꢁ0.03)
ꢀ2.56 (ꢁ0.03)
ꢀ2.50 (ꢁ0.14)
ꢀ2.44 (ꢁ0.05)
ꢀ2.45 (ꢁ0.05)
2.98 (ꢁ0.72)
—
3.53 (ꢁ0.34)
3.66 (ꢁ0.19)
—
4.02 (ꢁ0.44)
4.05 (ꢁ0.24)
2.16 (ꢁ0.71)
—
2.71 (ꢁ0.30)
2.61 (ꢁ0.16)
—
—
0.08 (ꢁ0.06)
—
—
—
—
14
14
14
22
14
14
22
14
14
14
22
14
14
22
0.18
0.26
0.09
0.07
0.28
0.10
0.08
0.18
0.21
0.07
0.06
0.23
0.09
0.08
0.59
0.54
0.92
0.95
0.02
0.89
0.94
0.44
0.26
0.91
0.94
0.10
0.88
0.90
99.87
81.25
99.99
99.99
40.00
99.99
99.99
99.00
94.00
99.99
99.99
81.00
99.99
99.99
—
—
61
191
0.13 (ꢁ0.02)
0.13 (ꢁ0.02)
—
—
ꢀ0.08 (ꢁ0.14)
ꢀ0.31 (ꢁ0.06)
ꢀ0.31 (ꢁ0.04)
—
43
147
—
—
59
—
—
0.10 (ꢁ0.05)
0.13 (ꢁ0.02)
0.13 (ꢁ0.01)
—
—
—
—
140
ꢀ0.13 (ꢁ0.12)
ꢀ0.31 (ꢁ0.05)
ꢀ0.29 (ꢁ0.04)
3.23 (ꢁ0.38)
—
—
41
87
2.98 (ꢁ0.23)
a Coefficient for the polar effect sI.
b Coefficient for the steric effect Es.
c Coefficient for the steric effect y.
d Number of data.
e Standard deviation.
f Square of the linear regression coefficient.
g Student t-test given in per cent.
h F-test at 99.99% confidence.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 269–275

274
G. ANANCHENKO ET AL.
(series 18–22) exhibit steric and polar effects, whereas for
Eqn (23), the number of molecules (9 alkoxyamines)
exhibiting a pure steric effect is balanced by the number
of molecules (8 alkoxyamines, series 10–17) exhibiting a
pure polar effect. All the comments made on and from the
correlations using Es as steric molecular descriptors hold
for the correlations using y but the statistical output are
slightly poorer (Eqns (24)–(26a and b), Table 2).
Because the statistical output and weight coefficients
(Table 2SI) of Eqn (28) (three parameters) are very close
to those of Eqn (27) (five parameters), there is no need to
use more than three parameters to fit the whole set of data.
When the two isomers of each of the alkoxyamines 1–22
are fitted with Eqn (28) (Fig. 2SI), the RR/SS isomer data
are closer to the straight line than the RS/SR isomer data.
This result could mean that the RR/SS and the RS/SR
isomers exhibit a normal and an enhanced polar effect,
respectively. The enhanced polar effect could be due to
the RS/SR configuration which allows either a preferred
conformation, in which the distance and the angle
between the reactive center and the polar group are
optimum, or a possible interaction between the ester
group and the phosphoryl group.42
Figure 1. Plot of log (kd/sꢀ1) at 120 8C versus Es for alkox-
yamines 1–9 (filled symbols) and 18–22 (open symbols).
(&, &) RR/SS isomers, (*, *) RS/SR isomers. The straight
lines are for Eqns 1(a) and 1(b) and for data 1–9
kd
sꢀ1
log
¼ ꢀ14:62ðꢁ0:74Þ þ 15:75ðꢁ1:25Þ ꢂ sRS
þ 19:75ðꢁ1:65Þ ꢂ s0I þ 7:28ðꢁ0:61Þ
ꢂ y0 þ 3:49ðꢁ0:91Þ ꢂ sI
ꢀ 0:27ðꢁ0:12Þ ꢂ y
(27)
F99;99 ¼ 65
R2 ¼ 0:85
s ¼ 0:42
N ¼ 61
kd
sꢀ1
log
¼ ꢀ14:03ðꢁ0:73Þ þ 15:38ðꢁ1:30Þ ꢂ sRS
þ 18:83ðꢁ1:45Þ ꢂ s0I þ 6:79ðꢁ0:60Þ
Figure 2. Plot of log (kd/sꢀ1) at 120 8C versus Eqns 23(a)
(square) and 23(b) (circle), (&, &, ) RR/SS isomers, (*, *
)
ꢂ y0
(28)
RS/SR isomers, for alkoxyamines 1–9 (filled symbols), 10–17
(crossed symbols), and 18–22 (open symbols)
R2 ¼ 0:84
s ¼ 0:44
N ¼ 61
F99;99 ¼ 97
We have already discussed the short- and long-range
polar effect,20,21,33 and ascribed it to the change in
electronegativity difference between the oxygen and the
carbon atoms of the C—ON bond. This involves
stabilization or destabilization (small Dx) of the
alkoxyamine, that is decrease or increase of kd,
respectively. However, the presence of a long-range
effect is better accounted for by through-space trans-
mission (field effect) of the substituent polarity to the
reactive center rather than by through s-bond trans-
mission (inductive effect) of this polarity to the reactive
center.36,42,52 Such through-space transmission of the
polarity may account for the kd increase with the
increasing length of the polybutylacrylate-SG1 polymer,
as observed by the authors.53 That is, assuming a folded
polymer chain, the longer the chain, the more numerous
position of the aromatic ring.36–38 When all the
alkoxyamines 1–22 are taken into account, biparameter
Eqns 23(a) and 23(b) well account for the polar and steric
effects of the ester group. The weaker polar effect for the
RS/SR than for the RR/SS isomer is due to the dependence
of sI on the geometry of the molecule (the anchimeric
effect, see Ref. 42). It is noteworthy that the long-range
steric effect does not depend on the configuration of the
alkoxyamine, that is both isomers exhibit the same slope
(Eqns (1) and (21)–(26) in Table 2). With weighting
equations,51 it appears that the polar effect is the major
contribution (54%–70%, Table 2SI) to the cleavage. The
steric contribution is larger in Eqn (22) than in Eqn (23)
(Table 2SI). In fact, for Eqn (22), 9 alkoxyamines (series
1–9) exhibit a pure steric effect and 5 alkoxyamines
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 269–275

LONG-RANGE POLAR AND STERIC EFFECTS
275
the polar units close to the reactive center and therefore
the higher the polarity at the carbon atom of the cleaved
C—ON bond, the larger the values of kd.
In a recent work,30 we showed that the kds of TEMPO-
based alkoxyamines did not exhibit such dependence on
the size of the alkyl ester group. This absence of
dependence could be due either to a very fast exchange
process or to the alkyl fragment pre-set in the right
conformation for the one-step homolysis pathway.
16. Couturier J-L, Guerret O, Gigmes D, Marque S, Chauvin F, Dufils
P-E, Bertin D, Tordo P. Patent no FR2843394 (2004), WO
2004014926.
17. Marque S, Le Mercier C, Tordo P, Fischer H. Macromolecules
2000; 33: 4403–4410.
18. Marsal P, Roche M, Tordo P, de Sainte Claire P. J. Phys. Chem. A
1999; 103: 29899–29905.
19. Le Mercier C, Lutz J-F, Marque S, Le Moigne F, Tordo P, Lacroix-
Desmazes P, Boutevin B, Couturier JL, Guerret O, Martschke R,
Sobek J, Fischer H. ACS symp. ser. 2000; 768: 108–122.
20. Bertin D, Gigmes D, Le Mercier C, Marque SRA, Tordo P. J. Org.
Chem. 2004; 69: 4925–4930.
21. Bertin D, Gigmes D, Marque SRA, Tordo P. Macromolecules
2005; 38: 2638–2650.
22. Marque S. J. Org. Chem. 2003; 68: 7582–7590.
23. Moad CL, Moad G, Rizzardo E, Thang SH. Macromolecules 1996;
29: 7717–7726.
CONCLUSION
24. Skene WG, Belt ST, Connolly TJ, Hahn P, Scaiano JC. Macro-
molecules 1998; 31: 9103–9105.
25. Rizzardo E, Chieffari J, Chong BYK, Ercole F, Krstina J, Jeffery J,
Le TPT, Mayadunne RTA, Meijs GF, Moad CL, Moad G, Thang
SH. Macromol. Symp. 1999; 143: 291–307.
Our results support and exemplify the presence of long-
range steric and polar effects in ester SG1-based
alkoxyamines 1–22. The long-range steric effect is
accounted for either by the presence of a conformer or
26. Ciriano MV, Korth H-G, van Scheppingen WB, Mulder P. J. Am.
Chem. Soc. 1999; 121: 6375–6381.
¼
by an activation entropy effect (DS ). The long-range
polar effect is configuration dependent and thus resembles
the field effect which depends on the proximity of the
reaction center to the polar group. This long-range polar
effect occurs in the initial state and was ascribed
previously to the change in electronegativity difference
between the atoms forming the cleaved bond.
27. Cameron NR, Reid AJ, Span P, Bon SAF, van Es JJGS, German
AL. Macromol. Chem. Phys. 2000; 201: 2510–2518.
28. Goto A, Fukuda T. Macromol. Chem. Phys. 2000; 201: 2138–2142.
29. Marque S, Fischer H, Baier E, Studer A. J. Org. Chem. 2001; 66:
1146–1156.
30. Ananchenko GS, Marque S, Gigmes D, Bertin D, Tordo P. Org.
Biomol. Chem. 2004; 2: 709–7014.
31. Ananchenko GS, Matyjaszewski K. Macromolecules 2002; 35:
8323–8329.
32. Bertin D, Gigmes D, Maurin R, Tordo P. J. Polym. Sci.: Part A:
Polym. Chem. 2004; 42: 3504–3515.
33. Bertin D, Gigmes D, Marque SRA, Milardo S, Peri J, Tordo P. Coll.
Czec. Chem. Commun. 2004; 69: 2223–2238.
Acknowledgements
34. Kutter E, Hansch C. J. Med. Chem. 1969; 647–652.
35. Fujita T, Takayama C, Nakajima M. J. Org. Chem. 1973; 38: 1623–
1631.
The authors are grateful to the University of Provence,
CNRS, and Arkema for their financial support.
36. Charton M, Charton BI. J. Chem. Soc., Perkin Trans. 2: 1999;
2203–2211.
37. Bowden K, Grubbs EJ. Progr. Phys. Org. Chem. 1992; 19: 183–
224.
REFERENCES
38. Bowden K, Grubbs EJ. Chem. Soc. Rev. 1996; 25: 171–182.
39. Juaristi E, Cuevas G. The Anomeric Effect. CRC Press: Boca
Raton, FL, 1995.
40. Armarego WLF, Chai CLL. Purification of Laboratory Chemicals
(5th edn). Butterworth Heinemann: Amsterdam, 2003.
41. Bertin D, Gigmes D, Marque S, Tordo P. e-Polymers 2003;
paper 2.
42. Beaudoin E, Bertin D, Gigmes D, Marque SRA, Siri D, Tordo P.
Eur. J. Org. Chem. 2006; 1755–1768.
43. This issue is addressed in a forthcoming review Bertin D, Gigmes
D, Marque SRA. Recent Res. Develop. Org. Chem. in press.
1. Solomon DH, Rizzardo E, Cacioli P. US Patent 4,581,429; Eur. Pat.
Appl. 135280. Chem. Abstr. 1985; 102: 221335q.
2. Georges MK, Veregin RPN, Kazmaier PM, Hamer GK. Macro-
molecules 1993; 26: 2987–2988.
3. Hawker CJ. Acc. Chem. Res. 1997; 30: 373–382.
4. Hawker CJ, Bosman AW, Harth E. Chem. Rev. 2001; 101: 3661–
3688 and references cited therein.
5. Greszta D, Matyjaszewski K. Macromolecules 1996; 29: 7661–
7670 and references cited therein.
6. Fischer H, Souaille M. Chimia 2001; 55: 109–113.
7. Fischer H. Chem. Rev. 2001; 101: 3581–3610 and references cited
therein.
´
´
44. Heberger K, Kemeny S, Vidoczy T. Int. J. Chem. Kin. 1987; 19:
171–181.
45. Charton M. Progr. Phys. Org. Chem. 1981; 13: 119–251.
46. Charton M. In Advances in Quantitatives Structure-Property
Relationship. The Estimation of Electrical Effect Substituent Con-
stants for Correlation Analysis. M. Charton (ed.). Jai Press Inc.:
Greenwich CT, 1996; 1: 171–219.
47. Charton M. Stud. Org. Chem. 1991; 42: 629–692.
48. Charton M. Top. Curr. Chem. 1983; 114: 57–91.
49. Gallo R. Progr. Phys. Org. Chem. 1983; 14: 115–163.
50. Charton M. Adv. Mol. Struc. Res. 1999; 5: 25–88.
51. Shorter M. In Correlation Analysis of Organic Reactivity. J. Wiley
& Sons: New York, 1982, 73–126.
8. Goto A, Fukuda T. Prog. Polym. Sci. 2004; 29: 329–385.
9. Matyjaszewski K (ed). Controlled-Living Radical Polymerization:
Progress in ATRP, NMP, and RAFT. American Chemical Society:
Washington, DC, Vol. 768, 2000;
10. Matyjaszewski K. Macromol. Symp. 2001; 174: 51–68.
11. Moad G, Rizzardo E. Macromolecules 1995; 28: 8722–8728.
12. Grimaldi S, Le Moigne F, Finet J-P, Tordo P, Nicol P, Plechot M.
International Patent WO 96/24620. 15 August 1996.
13. Benoit D, Chaplinski V, Braslau R, Hawker CJ. J. Am. Chem. Soc.
1999; 121: 3904–3920.
14. Zink M-O, Kramer A, Nesvadba P. Macromolecules 2000; 33:
8106–8108.
15. Le Mercier C, Acerbis S, Bertin D, Chauvin F, Gigmes D, Guerret
O, Lansalot M, Marque S, Le Moigne F, Fischer H, Tordo P.
Macromol. Symp. 2002; 182: 225–247.
52. Charton M. J. Phys. Org. Chem. 1999; 12: 275–282.
53. Guerret O, Couturier JL, Chauvin F, El-Bouazzy H, Bertin D,
Gigmes D, Marque S, Fischer A, Tordo P. ACS symp. ser. 2003;
854: 412–423.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 269–275