Radical crossover in nitroxide mediated 'living' free radical polymerizations

Article abstract of DOI:10.1021/ja9624228

The efficiency of exchange between the mediating nitroxide moieties at the termini of growing polymer chains during 'living' free radical polymerizations has been probed by a series of crossover experiments using functionalized unimolecular initiators. The design of appropriately substituted initiators permitted the synthesis of specifically functionalized model polymers which could be readily distinguished using high-performance liquid chromatography (HPLC). Using these models, the mixture of macromolecules obtained from a 1:1 combination of disparate initiators was separated and identified. The results reveal that exchange of the mediating nitroxide free radicals is a facile process and at essentially all stages of the polymerization a nearly statistical mixture of crossover products is obtained. The HPLC techniques developed are also useful in evaluating the extent of chain termination in nitroxide mediated 'living' free radical polymerizations.

Full text of DOI:10.1021/ja9624228

J. Am. Chem. Soc. 1996, 118, 11467-11471
11467
Radical Crossover in Nitroxide Mediated “Living” Free Radical
Polymerizations
,†
,‡
†
Craig J. Hawker,* George G. Barclay,* and Julian Dao
Contribution from the IBM Research DiVision, Almaden Research Center, Center for Polymeric
Interfaces and Macromolecular Assemblies, 650 Harry Road, San Jose, California 95120-6099,
and Shipley Company, 455 Forest Street, Marlborough, Massachusetts 01752-3092
X
ReceiVed July 15, 1996
Abstract: The efficiency of exchange between the mediating nitroxide moieties at the termini of growing polymer
chains during “living” free radical polymerizations has been probed by a series of crossover experiments using
functionalized unimolecular initiators. The design of appropriately substituted initiators permitted the synthesis of
specifically functionalized model polymers which could be readily distinguished using high-performance liquid
chromatography (HPLC). Using these models, the mixture of macromolecules obtained from a 1:1 combination of
disparate initiators was separated and identified. The results reveal that exchange of the mediating nitroxide free
radicals is a facile process and at essentially all stages of the polymerization a nearly statistical mixture of crossover
products is obtained. The HPLC techniques developed are also useful in evaluating the extent of chain termination
in nitroxide mediated “living” free radical polymerizations.
Introduction
lymerization of reactive monomers, such as sodium styrene
1
0
sulfonate, to give narrow polydispersity homopolymers with
The field of “living” free radical polymerizations has
witnessed explosive growth recently due to the great promise
shown by these synthetically robust and simple procedures.
Of these procedures, those mediated by stable nitroxide free
radicals, such as 2,2,6,6-tetramethylpiperidinyloxy (TEMPO),
are the most widely studied and a variety of authors have
shown that the polymerization of styrene based monomers can
be controlled to levels previously only obtained using anionic
or cationic procedures.4 Narrow polydispersity materials with
6
molecular weights approaching 10 amu.
The origin of this unprecedented control in free radical
polymerizations is believed to be the reversible termination of
the growing polymeric radical by the stable nitroxide free
radical. This gives a dormant, or inactive species, 1, in which
the nitroxide is covalently bound to the polymer chain end.
Under the polymerization conditions the carbon-oxygen bond
is homolytically unstable and can undergo fragmentation to give
a stable nitroxide free radical and the polymeric radical, 2.
Depending on the rate of recombination of this radical pair the
polymeric radical, 2, can then undergo chain extension with
monomer to yield a similar polymeric radical, 3, in which the
degree of polymerization has increased. Recombination of 3
with the nitroxide then gives the dormant species, 4, which has
essentially the same structure as 1 and the cycle of homolysis-
monomer addition-recombination can be repeated (Scheme 1).
While this is the generally accepted blueprint for nitroxide
mediated “living” free radical polymerizations a number of
mechanistic questions remain unanswered. For example, it is
not known if the nitroxide counter-radicals are associated with
the same polymeric chain end during the course of the
polymerization or if they are able to diffuse freely to the reaction
1
2
3
5
6
controlled molecular weights, chain ends, and chain architec-
7
tures can be readily prepared by heating the neat monomer
with a nitroxide based initiator at 120-130 °C under an inert
atmosphere. Interestingly, well-defined random and block
copolymers can be readily prepared from a variety of monomers,
while the mild polymerization conditions also allow the po-
8
9
†
Center for Polymeric Interfaces and Macromolecular Assemblies.
Shipley Company.
Abstract published in AdVance ACS Abstracts, November 1, 1996.
‡
X
(1) Georges, M. K.; Veregin, R. P. N.; Kazmaier, P. M.; Hamer, G. K.
Macromolecules 1993, 26, 2987. Hawker, C. J. J. Am. Chem. Soc. 1994,
1
5
16, 11314. Wang, J. S.; Matyjaszewski, K. J. Am. Chem. Soc. 1995, 117,
614. Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura, T. Macro-
molecules 1995, 28, 1721. Percec, V.; Barboiu, B.; Neumann, A.; Ronda,
J. C.; Zhao, M. Macromolecules 1996, 29, 3665.
1
1
(
2) Hawker, C. J. Trends Polym. Sci. 1996, 4, 183. Georges, M. K.;
Veregin, R. P. N.; Kazmaier, P. M.; Hamer, G. K. Trends Polym. Sci. 1994,
, 66.
3) Moad, G.; Rizzardo, E. Macromolecules 1995, 28, 8722. Kazmaier,
medium. In this paper we wish to report radical crossover
experiments designed to probe the exchange of nitroxide free
radicals at various stages during “living” free radical polymer-
izations.
2
(
P. M.; Moffat, K. A.; Georges, M. K.; Veregin, R. P. N.; Hamer, G. K.
Macromolecules 1995, 28, 1841. Li, I.; Howell, B. A.; Matyjaszewski,
K.; Shigemoto, T.; Smith, P. B.; Priddy, D. B. Macromolecules 1995, 28,
Results and Discussion
6
692. Hawker, C. J.; Carter, K. R.; Hedrick, J. L.; Volksen, W. Polym.
Prepr. 1995, 36, 110. Puts, R. D.; Sogah, D. Y. Macromolecules 1996,
Previously, it has been demonstrated that the chain ends of
vinyl polymers can be accurately controlled using nitroxide
2
2
9, 3323. Catala, J. M.; Bubel, F.; Hammouch, S. O. Macromolecules 1995,
8, 8441.
6,13
(
4) Webster, O. W. Science 1991, 251, 887. Fr e´ chet, J. M. J. Science
994, 263, 1710.
5) Hammouch, S. O.; Catala, J. M. Macromol. Rapid Commun. 1996,
7, 149.
mediated “living” free radical polymerizations.
This feature
1
(
(9) Yoshida, E.; Ishizone, T.; Hirao, A.; Nakahama, S.; Takata, T.; Endo,
T. Macromolecules 1994, 27, 3119. Fukuda, T.; Terauchi, T.; Goto, A.;
Tsujii, Y.; Miyamoto, T.; Shimizu, Y. Macromolecules 1996, 29, 3050.
(10) Keoshkerian, B.; Georges, M. K.; Boils-Boissier, D. Macromolecules
1995, 28, 6381.
(11) Matyjaszewski, K. Polym. Prepr. 1996, 37(1), 325.
(12) Hawker, C. J.; Barclay, G. G.; Orellana, A.; Dao, J.; Devonport,
W. Macromolecules 1996, 29, 5245.
1
(
(
6) Hawker, C. J.; Hedrick, J. L. Macromolecules 1995, 28, 2993.
7) Hawker, C. J. Angew Chem., Int. Ed. Engl. 1995, 34, 1456. Hawker,
C. J.; Fr e´ chet, J. M. J.; Grubbs, R. B.; Dao, J. J. Am. Chem. Soc. 1995,
17, 10763.
8) Hawker, C. J.; Elce, E.; Dao, J.; Russell, T. P.; Volksen, W.; Barclay,
G. G. Macromolecules 1996, 29, 2686.
1
(
S0002-7863(96)02422-5 CCC: $12.00 © 1996 American Chemical Society

11468 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Hawker et al.
Scheme 1
Scheme 2
is exploited in the design of crossover experiments to probe
the potential diffusion of the mediating radical from the
propagating chain end during “living” free radical polymeriza-
tions. The design of an appropriate crossover experiment is
depicted in Scheme 2 and involves the use of two structurally
similar unimolecular initiators which differ only in their
substitution pattern. One derivative is unfunctionalized, 5, while
the unimolecular initiator, 6, contains two hydroxy groups, one
attached to the TEMPO unit and the second located at the
â-carbon atom of the ethylbenzene unit. If a 1:1 mixture of 5
and 6 is used to initiate the “living” free radical polymerization
of styrene, homolysis of the carbon-oxygen bond of 5 and 6
will lead to four radical species. Each of the radicals produced
are chemically different and comprise a pair of initiating or
propagating radicals, 7 and 8, which are both based on
ethylbenzene except that 8 has a hydroxy functionality attached
to the â-carbon atom. Similarly, a pair of structurally similar
mediating nitroxide radicals are produced, TEMPO, 9, and
detailing the separation of polymeric materials using chromato-
graphic techniques that are based on polarity, and not size,
differences. For example, the purification and identification of
4
-OH-TEMPO, 10, which again differ by the presence of a
single hydroxy group in 10. If no escape of the mediating
nitroxide radical from propagating chain end occurs, only two
polystyrene derivatives will therefore be formed; 11, which is
derived from 5, will have no hydroxy groups at the chain ends,
while 12 contains a single hydroxy group at both chain ends.
In contrast, if radical crossover does occur and the mediating
nitroxide radicals are free to diffuse to the polymerization
medium, four polystyrenes having different substitution patterns
will be obtained. In addition to the “non-crossover” products,
1
4
dendrimers has relied heavily on flash chromatography and
high-performance liquid chromatography (HPLC), while Mc-
15
Carthy has examined the separation of chain end functionalized
polystyrene derivatives by thin-layer chromatography (TLC).
To investigate the application of these separation techniques to
the products in Scheme 2 the synthesis of all four possible model
polymers, 11, 12, 13, and 14, was undertaken. This was
accomplished by the synthesis of four different unimolecular
initiators, 5, 6, 15, and 16, which vary in the number and
placement of hydroxy functionalities (Figure 1). The prepara-
tion of the derivatives based on ethylbenzene, 5 and 15,
employed the synthetic approach recently described by Priddy
1
1 and 12, two monohydroxy functionalized polystyrenes, 13
and 14, are obtained. Both 13 and 14 result from exchange, or
crossover, of the nitroxides and differ in the placement of the
single hydroxy chain end. Reaction of 4-OH-TEMPO, 10, with
the polymer chain derived from 7 gives 14 in which the hydroxy
group is at the same chain end as the nitroxide unit. Alterna-
tively, reaction of TEMPO, 9, with the polymer chain derived
from 8 gives 13 where the hydroxy group is at the opposite
chain end to the nitroxide unit.
In either of the above scenarios a mixture of functionalized
polystyrenes, having similar molecular weights but different
chain end functional groups, is obtained. The separation and
analysis of such a mixture is not a standard technique in polymer
chemistry. However, a limited number of papers have appeared
(14) Jansen, J. F.; de Brabander van den Berg, E. M.; Meijer, E. W.
Science 1994, 266, 1226. Tomalia, D. A.; Baker, H.; Dewald, J.; Hall, M.;
Kallos, G.; Martin, R.; Ryder, J.; Smith, P. Polym. J. 1985, 17, 117.
Newkome, G. R.; Yao, Z.; Baker, G. R.; Gupta, V. K.; J. Org. Chem. 1985,
50, 2003. Hawker, C. J.; Frechet, J. M. J. J. Chem. Soc., Chem. Commun.
1990, 1010. Hawker, C. J.; Wooley, K. L.; Frechet, J. M. J. J. Am. Chem.
Soc. 1993, 115, 4375. Jansen, J. F.; Meijer, E. W.; de Brabander van den
Berg, E. M. J. Am. Chem. Soc. 1995, 117, 4417. Hawker, C. J.; Fr e´ chet,
J. M. J. J. Am. Chem. Soc. 1990, 112, 7638. Xu, Z.; Moore, J. S. Angew.
Chem., Int. Ed. Engl. 1993, 32, 1354. Bhyrappa, P.; Young, J. K.; Moore,
J. S.; Suslick, K. S. J. Am. Chem. Soc. 1996, 118, 5708. Percec, V.; Chu,
P.; Ungar, G.; Zhou, J. J. Am. Chem. Soc. 1995, 117, 11441.
(15) Iyengar, D. R.; McCarthy, T. J. Macromolecules 1990, 23, 4344.
(16) Li, I.; Howell, B. A.; Ellaboudy, A.; Kastl, P. E.; Priddy, D. B.
Polym. Prepr. 1995, 36(1), 469.
(13) Hedrick, J. L.; Hawker, C. J.; Dipietro, R.; Jerome, R.; Charlier, Y.
Polymer 1995, 36, 4855. Barclay, G. G.; Orellana, A.; Hawker, C. J.; Elce,
E.; Dao, J. Polym. Mater. Sci. Eng. 1996, 74, 311. Frank, B.; Gast, A. P.;
Russell, T. P.; Brown, H. R.; Hawker, C. J. Macromolecules 1996, 29, 6531.

Radical CrossoVer in Free Radical Polymerizations
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11469
Scheme 4
Figure 1. Structure of functionalized unimolecular initiators, 5, 6, 15,
and 16, used for the preparation of the model chain end functionalized
polystyrenes.
Scheme 3
Scheme 5
ends. Near baseline separation was observed between all four
polymers with the shortest retention time peak (2.90 min)
corresponding to the unfunctionalized polymer, 11, while the
peak with the longest retention time (5.80 min) corresponds to
the polymer, 12, with hydroxy groups at both chain ends.
Surprisingly, the monohydroxy terminated polystyrenes, 13 and
and Howell.16 For example, a mixture of di-tert-butyl peroxide
and 4-benzoyloxy-TEMPO, 17, was heated at reflux in ethyl-
benzene for 16 h. This gave the alkylated TEMPO derivative,
1
4, displayed different elution times even though they both
contain a single hydroxy group and a single TEMPO unit. The
presence of the hydroxy group and TEMPO unit at the same
chain end of 14 results in a longer retention time (4.70 min)
when compared to 13 (3.40 min). A possible reason for this
difference is that the presence of the polar hydroxy and TEMPO
functionalities at the same chain end results in increased
interaction with the stationary phase when compared to the case
where the hydroxy and TEMPO functionalities are at different
chain ends. To determine the effect of molecular weight on
elution times, model polystyrene macromolecules, 11-14, with
molecular weights of approximately 5000 and 7500 amu were
prepared and examined under the same conditions. Near
baseline was again observed between all the samples and only
a slight reduction in retention times was observed. Therefore,
slight variation in molecular weights for the model polymers,
or the polystyrene derivatives produced during the crossover
experiments, will not affect the analysis or identification of
mixtures containing 11-14 at molecular weights less than 7500
amu. It should be noted that as the molecular weight increases
above 7500 amu separation becomes progressively more difficult
and above 10 000 amu is no longer possible. This is due to
the influence of the chain ends becoming minimal at higher
molecular weights, and the elution time of all the samples
eventually becomes the same. For example, the elution time
for the monohydroxy-terminated polystyrene, 14, decreases from
4.70 min to 2.55 min as the molecular weight increases from
2 900 to 45 000 amu. A similar effect has been observed by
1
8, in 41% yield after purification which was then hydrolyzed
with potassium hydroxide to give the desired monohydroxy
derivative, 15, in which the hydroxy group is located on the
nitroxide ring (Scheme 3). In contrast, the derivatives in which
the hydroxy group is located on the â-carbon, 6 and 18, were
both prepared by reaction of benzoyl peroxide with styrene and
the appropriately functionalized TEMPO derivative. In this
case, a mixture of benzoyl peroxide and 4-benzoyloxy-TEMPO
is heated at 85 °C in excess styrene to give the diester, 19, which
on hydrolysis yields the desired dihydroxy derivative, 6, in 34%
yield over both steps (Scheme 4).
The preparation of the model polystyrene derivatives from
the appropriately functionalized unimolecular initiators was then
performed using standard “living” free radical polymerization
conditions. The molecular weights of the model polymers, 11-
1
4, were controlled by the addition of specified equivalents of
styrene to the corresponding unimolecular initiators and were
approximately 3000 amu in each case. For example, reaction
of 6 with 35 equiv of styrene at 125 °C for 48 h gave the
dihydroxy-terminated polystyrene derivative, 12, in 85% yield
which was shown to have a molecular weight, Mn, of 2900 and
a polydispersity of 1.09 (Scheme 5).
Significantly, examination of the model polystyrene macro-
molecules, 11-14, by HPLC revealed substantially different
elution times depending on the functional group/s at the chain

11470 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Hawker et al.
Scheme 6
Figure 2. HPLC trace of the product mixture obtained from the
polymerization of 35 equiv of styrene using a 1:1 combination of
unimolecular initiators, 5 and 6.
results at conversions ranging from 15% to 80%. From these
results it can be concluded that the nitroxide free radicals are
free to diffuse out of the reaction cage during “living” free
radical polymerization and that statistical exchange of the chain
ends occurs at an early stage of the polymerization.
To confirm the above results a modified crossover experiment
was performed using a 1:1 mixture of the unfunctionalized
initiator, 5, and the monohydroxy derivative, 15. In this case
homolysis generates only three radical species, an initiating
radical, 7, and two different mediating nitroxide radicals, 9 and
1
0. The commonality of 7 to both systems means that only
two different polystyrene macromolecules, 11 and 14, should
be obtained (Scheme 6). Therefore, polymerization of 40 equiv
of styrene with a 1:1 mixture of 5 and 15 gives a narrow
polydispersity product (Mn ) 3500, PD ) 1.10) that was shown
to contain only two peaks by HPLC analysis (Figure 3). These
peaks correlate with the expected products, 11 and 14, and the
absence of other peaks or unknown products demonstrates the
validity of the above crossover experiments.
Figure 3. HPLC trace of the product mixture obtained from the
polymerization of 40 equiv of styrene using a 1:1 combination of
unimolecular initiators, 5 and 15.
McCarthy15 for the analysis of functionalized polystyrenes by
thin-layer chromatography.
Crossover experiments were then performed using a 1:1
mixture of the unfunctionalized unimolecular initiator, 5, and
the dihydroxy initiator, 6. As detailed above, if no radical
exchange, or crossover, occurs HPLC analysis should show only
two products, 11 and 12. However, if crossover does occur, a
statistical mixture of all four possible polystyrene derivatives
should be obtained. The polymerization of 35 equiv of styrene
with 5 and 6 was therefore conducted under standard “living”
free radical polymerization conditions with samples being
removed at 50 and 90% conversion. The mixture of polystyrene
macromolecules obtained was found to have molecular weights,
Mn, of 1600 (PD ) 1.08) and 3100 (PD ) 1.13), respectively,
which indicates that controlled growth is occurring from the
mixture of initiators. Interestingly, HPLC analysis of these
samples showed that all four possible products were present in
an approximately statistical ratio (Figure 2). Repetition of this
experiment using 100 equiv of styrene gave essentially the same
In conclusion, a radical crossover experiment to probe the
mobility of mediating nitroxide free radicals during “living” free
radical polymerizations has been designed. To investigate the
validity of this concept a series of unimolecular initiators having
different hydroxy group substitution patterns as well as the
corresponding polystyrene model polymers were synthesized.
A HPLC technique was then developed which allowed the
separation and identification of polystyrene macromolecules
having different substitution patterns at the chain ends. Com-
parison of these model compounds with the product mixture
obtained from the crossover experiments demonstrated conclu-
sively that migration of the mediating nitroxide radicals during
“living” free radical polymerizations is a facile process. The
versatility and application of HPLC to other issues in “living”
free radical polymerizations and telechelic polymers is currently
being investigated.

Radical CrossoVer in Free Radical Polymerizations
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11471
Experimental Section
solution was evaporated to dryness and the reaction mixture purified
by flash chromatography column eluting with 1:1 hexane/dichlo-
romethane gradually increasing to dichloromethane to give the modified
diester, 19, as a pale yellow oil (3.70 g, 39%); IR (neat) 3100-2850,
Nuclear magnetic resonance spectroscopy was performed on a Bruker
AM 200 FT-NMR spectrometer using deuterated chloroform as solvent
and tetramethylsilane as internal reference. Gel permeation chroma-
tography was carried out on a Waters chromatograph connected to a
Waters 410 differential refractometer with THF as the carrier solvent.
Differential scanning calorimetry was performed on a Perkin Elmer
DSC-7 calorimeter using a scanning rate of 10 °C/min under a nitrogen
atmosphere. The glass transition temperature was defined as the half-
way point of transition heat flow. Analytical TLC was performed on
commercial Merck plates coated with silica gel GF254 (0.25 mm thick).
Silica gel for flash chromatography was Merck Kieselgel 60 (230-
-1
1
1
720, and 1200 cm ; H NMR (CDCl
3
) δ 0.70, 1.10, 1.24, 1.33 (each
); 4.45 (ABq, J ) 6 Hz, 1 H,
br s, 12H, CH ), 1.35-1.95 (m, 4H, CH
3
2
CHH), 4.76 (ABq, J ) 6 Hz, 1 H, CHH), 5.00 (ABq, J ) 3 Hz, 1 H,
CH), 5.16 (complex m, 1 H, CH), 7.25-7.50 (m, 11H, ArH), 7.84 (B
of ABq, J ) 6 Hz, 2H, ArH), and 7.93 (B of ABq, J ) 6 Hz, 2H,
13
ArH); C NMR (CDCl
3
) δ 20.90, 21.22, 32.87, 34.10, 44.16, 44.80,
5
1
9.80, 60.63, 66.61, 67.22, 84.24, 127.71, 127.89, 128.18, 128.33,
28.42, 129.50, 129.71, 130.10, 132.88, 140.25, 166.13, and 166.327;
mass spectrum (EI) m/z 501.
4
00 mesh). High-pressure liquid chromatography experiments were
4-Hydroxy-1-((2′-hydroxy-1′-phenylethyl)oxy)-2,2,6,6-tetrameth-
conducted using a Waters 510 pump connected to a HP 1050 variable
ylpiperidine (6). To a solution of diester 19 (2.50 g, 5.00 mmol) in
ethanol (50 mL) was added a solution of potassium hydroxide (800
mg, 14.0 mmol) in water (5 mL). The reaction mixture was then heated
at reflux under nitrogen for 16 h, cooled, and evaporated to dryness.
The residue was partitioned between water (150 mL) and dichlo-
romethane (150 mL) and the aqueous layer extracted with dichlo-
romethane (2 × 50 mL). The combined organic extracts were then
dried and evaporated to dryness. The crude product was purified by
flash chromatography eluting with dichloromethane gradually increasing
to 1:2 ether/dichloromethane to give the dialcohol, 6, as a waxy solid
wavelength UV-vis detector and a Waters Microporosil 60A column
(
4.4 × 300 mm) using a 1:1 mixture of THF and isooctane as eluent.
All solvents used for synthesis were dried and distilled in the appropriate
manner before use; the commercial reagents were obtained from Aldrich
and used without further purification. The unimolecular initiators, 5
and 16, were prepared and purified as previously reported.12
4-(Benzoyloxy)-1-((1′-phenylethyl)oxy)-2,2,6,6-tetramethylpiperi-
dine (18). A solution of 4-benzoyloxy-TEMPO, (17, 8.00 g, 29.0
mmol) and di-tert-butyl peroxide (4.23 g, 29.0 mmol) in ethylbenzene
(250 mL) was heated at reflux for 16 h, cooled, and evaporated to
-
1 1
(
1.29 g, 88%). IR (neat) 3250, 2970, 1450, 1380, and 1040 cm ; H
NMR (CDCl ) δ 1.10, 1.20, 1.25, and 1.41 (each br s, 3H, CH ), 1.30-
.90 (complex m, 4H), 3.66 (complex m, 1H, CH), 3.93 (complex m,
H, CH), 4.14 (ABq, J ) 7 Hz, 1H, CH), 5.19 (ABq, J ) 7 Hz, 1H,
dryness. The crude product was purified by flash chromatography
eluting with 1:4 hexane/dichloromethane gradually increasing to
dichloromethane to give the ester, 18, as a waxy solid (4.53 g, 41%).
3
3
1
1
-
1 1
IR (neat) 2950, 1710, 1270, and 1095 cm ; H NMR (CDCl
.16, 1.31, and 1.36 (each br s, 3H, CH ), 1.49 (d, J ) 7 Hz, 3H,
CH ), 1.40-2.00 (complex m, 4H), 4.78 (q, J ) 7 Hz, 1H, CH), 5.21
complex m, 1H, CH), 7.20-7.55 (complex m, 8 H, ArH), and 7.99 (d
3
) δ 0.68,
1
3
CH), and 7.25-7.33 (complex m, 5 H, ArH); C NMR (CDCl
2
8
3
) δ
1
3
1.27, 21.57, 33.08, 34.74, 48.75, 49.93, 60.85, 61.84, 62.53, 69.00,
3.97, 126.91, 128.29, 128.67, and 138.70; mass spectrum (FAB) 293.
3
(
13
Radical Crossover Experiments. A mixture of the unfunctionalized
of d, 2H, ArH); C NMR (CDCl
3
) δ 21.16, 23.32, 34.10, 34.40, 44.67,
0.45, 60.73, 67.47, 83.39, 126.69, 127.01, 128.08, 128.31, 129.49,
30.61, 132.83, 145.13, and 166.17; mass spectrum (FAB) 381.
-Hydroxy-1-((1′-phenylethyl)oxy)-2,2,6,6-tetramethylpiperi-
dine (15). To a solution of ester 18 (3.81 g, 10.0 mmol) in ethanol
100 mL) was added a solution of potassium hydroxide (1.12 g, 20.0
initiator (5, 261 mg, 1.0 mmol) and the dihydroxy initiator (6, 293
mg, 1.0 mmol) was dissolved in styrene (8.32 g, 80.0 mmol) and heated
at 125 °C under argon for 48 h. During this time, samples of the
polymerization mixture were withdrawn at periodic intervals and
evaporated to dryness. The polymeric mixture obtained was then
analyzed by HPLC eluting with a 7:3 mixture of isooctane and
tetrahydrofuran; the individual peaks were then compared with known,
independently synthesized samples.
6
1
4
(
mmol) in water (5 mL). The reaction mixture was then heated at reflux
under nitrogen for 16 h, cooled, and evaporated to dryness. The residue
was partitioned between water (150 mL) and dichloromethane (150
mL) and the aqueous layer extracted with dichloromethane (2 × 50
mL). The combined organic extracts were then dried and evaporated
to dryness. The crude product was purified by flash chromatography
eluting with dichloromethane gradually increasing to 1:19 ether/
dichloromethane to give the alcohol, 15, as a waxy solid (2.23 g, 81%).
General Procedure for the Preparation of Specifically Function-
alized Polystyrenes. The hydroxy functionalized initiator (16, 276
mg, 1.0 mmol) was dissolved in styrene (4.16 g, 40.0 mmol) and heated
at 125 °C under argon for 48 h. The polymerization mixture, which
solidified after ca. 24 h, was then dissolved in dichloromethane and
precipitated into methanol (500 mL). The resulting solid was then
collected by vacuum filtration and dried to give the hydroxy-terminated
polystyrene, 13, as a white solid (4.03 g, 90%): IR (neat) 3250, 3020,
-
1 1
IR (neat) 3250, 2970, 1450, 1380, and 1040 cm ; H NMR (CDCl
δ 0.65, 1.08, 1.23, and 1.31 (each br s, 3H, CH ), 1.48 (d, J ) 7 Hz,
), 1.20-1.90 (complex m, 4H), 3.92 (complex m, 1H, CH),
.758 (q, J ) 7 Hz, 1H, CH), and 7.20-7.33 (complex m, 5 H, ArH);
3
)
3
3
4
H, CH
3
-
1
1
2980, 1605, 1360, and 1050 cm ; H NMR (CDCl
) δ 0.15-0.40,
3
1
3
1.00-1.80, 2.50 (CH-TEMPO), 3.70 (CH OH), and 6.50-7.10.
C NMR (CDCl
3
) δ 21.26, 23.39, 34.14, 34.45, 48.78, 48.88, 59.99,
2
60.20, 63.32, 83.30, 126.65, 126.95, 128.06, and 145.45; mass spectrum
(
FAB) 277.
-(Benzoyloxy)-1-((2′-(benzoyloxy)-1′-phenylethyl)oxy)-2,2,6,6-tet-
ramethylpiperidine (19). To a solution of benzoyl peroxide (2.29 g,
.50 mmol) in distilled styrene (100 mL) was added 4-benzoyloxy-
,2,6,6-tetramethyl-1-piperidinyloxy (5.00 g, 18.9 mmol) and the
Acknowledgment. The authors gratefully acknowledge the
financial support of the NSF Center for Polymeric Interfaces
and Macromolecular Assemblies, IBM Corp., and the Shipley
Company.
4
9
2
solution heated at 90 °C under nitrogen for 20 h. After cooling the
JA9624228