Well-defined epoxide-containing styrenic polymers and their functionalization with alcohols

Article abstract of DOI:10.1002/pola.27952

Polymers containing electrophilic moieties, such as activated esters, epoxides, and alkyl halides, can be readily modified with a variety of nucleophiles to produce useful functional materials. The modification of epoxide-containing polymers with amines and other strong nucleophiles is well-documented, but there are no reports on the modification of such polymers with alcohols. Using phenyloxirane and glycidyl butyrate as low molecular weight model compounds, it was determined that the acid-catalyzed ring-opening of aryl-substituted epoxides by alcohols to form β-hydroxy ether products was significantly more efficient than that of alkyl-substituted epoxides. An aryl epoxide-type styrenic monomer, 4-vinylphenyloxirane (4VPO), was synthesized in high yield using an improved procedure and then polymerized in a controlled manner under reversible addition-fragmentation chain-transfer (RAFT) polymerization conditions. A successful chain extension with styrene proved the high degree of chain-end functionalization of the poly4VPO-based macro chain transfer agent. Poly4VPO was modified with a library of alcohols and phenols, some of which contained reactive functionalities, e.g., azide, alkyne, allyl, etc., using either CBr₄ (in PhCN at 90°C for 2-3 days) or BF₃ (in CH₂Cl₂ at ambient temperature over 30 min) as the catalyst. The resulting β-hydroxy ether-functionalized homopolymers were characterized using size exclusion chromatography, ¹H NMR and IR spectroscopy, and thermal gravimetric analysis.

Full text of DOI:10.1002/pola.27952

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Well-Defined Epoxide-Containing Styrenic Polymers and their
Functionalization with Alcohols
David C. McLeod, Nicolay V. Tsarevsky
Department of Chemistry and Center for Drug Discovery, Design, and Delivery at Dedman College,
Southern Methodist University, 3215 Daniel Avenue, Dallas, Texas 75275
Correspondence to: N. V. Tsarevsky (E-mail: nvt@smu.edu)
Received 24 September 2015; accepted 10 October 2015; published online 12 November 2015
DOI: 10.1002/pola.27952
ABSTRACT: Polymers containing electrophilic moieties, such as
activated esters, epoxides, and alkyl halides, can be readily
modified with a variety of nucleophiles to produce useful func-
tional materials. The modification of epoxide-containing poly-
mers with amines and other strong nucleophiles is well-
documented, but there are no reports on the modification of
such polymers with alcohols. Using phenyloxirane and glycidyl
butyrate as low molecular weight model compounds, it was
determined that the acid-catalyzed ring-opening of aryl-
substituted epoxides by alcohols to form b-hydroxy ether prod-
ucts was significantly more efficient than that of alkyl-
substituted epoxides. An aryl epoxide-type styrenic monomer,
transfer (RAFT) polymerization conditions. A successful chain
extension with styrene proved the high degree of chain-end
functionalization of the poly4VPO-based macro chain transfer
agent. Poly4VPO was modified with a library of alcohols and
phenols, some of which contained reactive functionalities, e.g.,
4
azide, alkyne, allyl, etc., using either CBr (in PhCN at 90 8C for
2–3 days) or BF (in CH Cl at ambient temperature over 30
3
2
2
min) as the catalyst. The resulting b-hydroxy ether-functional-
ized homopolymers were characterized using size exclusion
1
chromatography, H NMR and IR spectroscopy, and thermal
gravimetric analysis. V
2015 Wiley Periodicals, Inc. J. Polym.
C
Sci., Part A: Polym. Chem. 2016, 54, 1132–1144
4-vinylphenyloxirane (4VPO), was synthesized in high yield
KEYWORDS: controlled/“living” radical polymerization; epox-
using an improved procedure and then polymerized in a con-
trolled manner under reversible addition-fragmentation chain-
ides; post-polymerization modification; RAFT polymerization
INTRODUCTION The systematic development of novel poly-
meric materials with predetermined structures and numer-
ous pendant functional groups has garnered considerable
attention from the scientific community. Unfortunately, the
synthesis of functional materials through direct polymeriza-
tion of analogous monomers can often be difficult due to the
incompatibility of many functionalities with the reactants
and conditions required for some polymerization techniques.
A routine solution to this challenge is to first polymerize
monomers that contain reactive groups, which are compati-
ble with the chosen polymerization method, and then carry
out efficient post-polymerization modification reactions to
obtain the desired functionalities. For example, electrophilic
groups such as epoxides, activated esters, and alkyl halides
are compatible with most radical polymerization techniques
and are capable of undergoing a plethora of synthetically-
containing polymers are very attractive as “universal” precur-
sors of functional materials but their utility is limited by
cross-linking side reactions, which often take place during
modifications. The addition of nucleophiles to epoxide-
functionalized polymers under neutral or basic conditions
generates polymer-bound alkoxide anions, which react with
the epoxide groups of neighboring chains, leading to cross-
linking. Suppression of these undesirable reactions is usually
accomplished by employing large excesses of protic nucleo-
philes, including amines, carboxylic acids, thiols, or azides in
5
–8
the presence of proton sources.
Alcohols are abundantly represented in both natural materials
and synthetic products, but the poor nucleophilicity of the
hydroxy group has thus far precluded their use in the modifica-
tion of epoxide-functionalized polymers. The alcoholysis of
epoxides to form b-hydroxy ethers is an important intermediate
step in the synthesis of many small molecules and is normally
1
useful transformations with nucleophiles.
2
–4
9
Epoxides (oxiranes) are among the most versatile and reac-
tive of the aforementioned electrophilic groups and have
found numerous applications in synthetic chemistry. Epoxide-
catalyzed by Brønsted or Lewis acids. Heterogeneous acidic
catalysts that have been successfully employed for this purpose
1
0,11
include sulfonated polymers (Amberlyst-15 and Nafion-H),
Additional Supporting Information may be found in the online version of this article.
2015 Wiley Periodicals, Inc.
V
C
1
132
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
1
2–14
15–17
neutral alumina,
mesoporous aluminosilicate,
organo-
functionalized polymers synthesized by reversible addition-
fragmentation chain transfer (RAFT) polymerization.
1
8–20
31
tin phosphate condensates (OPCs),
and Fe -montmoril-
2
1
lonite, but it is unlikely that they could achieve sufficient
interaction with the pendant functionalities of polymers to effi-
ciently promote functionalization. Soluble compounds or com-
EXPERIMENTAL
2
2–31
Materials
plexes
metals,
of
main
group
and lanthanides
metals,
transition
are well-known
2
2,25,27,29,32–45
22,27,46–49
4-Vinylphenyloxirane (4VPO) was synthesized as previously
5
6
reported, with the exception that the solvent used in the
for their Lewis acidity and have been investigated specifically
for their ability to catalyze the alcoholysis of epoxides. The cata-
lytic activities of non-metallic Lewis acids like BF ,
hydrazine sulphate, molecular iodine, iminium triflate,
and trimethylsilyl triflate, as well as compounds able to
participate in single-electron reactions like 2,3-dichloro-5,6-
dicyano-p-benzoquinone, tetracyanoethylene, ceric ammo-
nium nitrate, N-methylpyrrolidine-2-one hydrotribromide,
and 1-benzyl-2,4,6-triphenylpyridinium tetrafluoroborate,
have also been explored. Many of these compounds are labeled
as “mild” and “highly-efficient” catalysts that are capable of ren-
dering desired products in excellent yields (>90%) at room
temperature within minutes to a few hours, but a close exami-
nation of the literature reveals that these claims often only hold
true when the nucleophiles are short-chain primary alcohols
first step was changed to CS . The RAFT chain transfer agents
2
5
7
2
9,43
27
(CTAs) butyl 1-phenylethyl trithiocarbonate (BPT), 2-cyano-
2-propyl dodecyl trithiocarbonate (CDT), benzyl dithioben-
CBr4,
3
5
8
5
0
51
29
5
9
2
9
zoate (BDB),
and 2-(butylsulfanylthiocarbonylsulfanyl)-2-
6
0
methyl propionic acid (BMPA) were synthesized according
to literature procedures. Styrene (99%, stabilized with 10–15
ppm 4-tert-butylcatechol, Alfa Aesar) was purified of radical
polymerization inhibitor by passing the neat monomer
through a short column filled with basic alumina (Brockmann
Grade I, 58 angstroms, Alfa Aesar). 3-Azido-1-propanol was
synthesized from 3-chloro-1-propanol and purified as
5
2
53
2
2
54
5
5
6
1
described in the literature. Azobisisobutyronitrile (AIBN,
8%, Aldrich) was purified by recrystallization from methanol.
9
All other reagents were used as received, including (R)-(2)-
glycidyl butyrate (96%, Aldrich), phenyloxirane (971%,
Acros), CBr4 (98%, Acros), allyl alcohol (981%, Alfa Aesar),
phenol (991%, Fisher), L(2)-menthol (99.5%, Acros), 10-
undecen-1-ol (98%, Acros), 2-naphthol (991%, Acros), pro-
pargyl alcohol (99%, Acros), methanol (99.81%, anhydrous,
Aldrich), 1-butanol (BuOH, 99.41%, Baker), isopropanol
(
usually methanol) and are used in large excess (10–100 eq. vs.
epoxide) and without solvent. Complete consumption of epox-
ides with secondary and tertiary alcohols typically requires lon-
ger reaction times and has lower yields. Most reports do not
identify side products or attempt to explain how they may have
2
3,35,41
occurred, but several
note that oligomeric side products
(
99.81%, anhydrous, Acros), 4-chlorobenzyl alcohol (99%,
are generated when fewer equivalents (<10 eq.) of alcohol are
used or when solvent is added to the reaction mixture, an espe-
cially salient observation given that if similar reactions were to
occur during the modification of polymer-bound epoxides, then
Alfa Aesar), 2-octanol (97%, Aldrich), ethanol (100%, anhy-
drous, Koptec), cyclohexanol (99%, Alfa Aesar), 3-chloro-1-
propanol (98%, Acros), benzyl alcohol (99.8%, anhydrous,
Aldrich), 2-bromoethanol (97%, Alfa Aesar), 4-bromophenol
(97%, Acros), furfuryl alcohol (98%, Alfa Aesar), 4-hydroxy-2-
butanone (951%, TCI), 9-anthracenemethanol (97%, Alfa
Aesar), cyclododecanol (99%, Acros), tert-butanol (991%,
Alfa Aesar), 1-decanol (99%, Aldrich), azobiscyclohexanecar-
bonitrile (VAZO88, 98%, Aldrich), benzonitrile (PhCN, 991%,
Acros), CS
(991%, Fisher), toluene (991%, Fisher), dichloromethane
(991%, Fisher), boron trifluoride diethyl etherate (BF -Et O,
981%, Alfa Aesar). The deuterated solvents were purchased
from Cambridge Isotope Laboratories.
3
3
cross-linked products would result. An in-depth study dem-
onstrated how different solvents influence the Lewis acid-
catalyzed alcoholysis of epoxides. It was found that the use of
polar and/or oxygen-containing solvents generally resulted in
poor yields, whereas less-polar, nonoxygenated solvents gave
markedly higher yields over the same length of time. It was rea-
soned that factors such as competitive complexation of the sol-
vent and the Lewis acid or the polarity of the transition states
and their stabilization by the solvent are plausibly responsible
for the observed solvent effects. The complications that arise
from the use of solvents and the nontrivial nature of their selec-
tion are unfortunate given that many polymers are not soluble
in alcohols. Finally, many catalysts suffer from drawbacks,
which include being highly toxic, corrosive, moisture or oxygen-
sensitive, expensive, difficult to synthesize, difficult to solubilize
in nonpolar solvents, or required in greater than catalytic
amounts.
(99.91%, Fisher), anisole (99%, Acros), acetone
2
3
2
Analyses
1
The structures of all compounds were confirmed by H NMR
spectroscopy on either a Bruker Avance DRX 400 or JEOL
ECA-500 spectrometer operating at 400 or 500 MHz, respec-
tively, using samples diluted in deuterated solvents (CDCl or
DMSO-d ) containing tetramethylsilane (TMS) as a chemical
3
6
With these numerous caveats in mind, we were especially
intrigued by the possibility of using commercially-available,
metal-free Lewis acids like CBr4 and BF3 to catalyze the
modification of polymer-bound epoxides with alcohols.
Herein, we disclose our efforts to optimize these reactions
on small molecule epoxides under conditions which are nec-
essary for polymer modification (i.e., using solvent) and then
apply them towards the alcoholysis of well-defined, epoxide-
shift reference. Monomer conversions were determined by
periodically removing samples from the polymerization reac-
tion mixtures and diluting with CDCl for NMR analysis. Appa-
3
rent number-average molecular weights (M ) and molecular
n
weight distribution (MWD) dispersities (Ð 5 M /M ) of the
w
n
polymers were determined by size exclusion chromatography
(SEC) on a Tosoh EcoSEC HLC-8320 system equipped with a
series of four columns (TSK gel guard Super HZ-L, Super
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144
1133

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
HZM-M, Super HZM-N and Super HZ2000) with THF as the
For similar experiments with different RAFT CTAs, the stock
solution was made with 0.171 mmol of either CDT (59.1 mg),
BDB (41.8 mg), or BMPA (43.2 mg) instead of BPT. For experi-
ments with different DPn,targ the amount of CTA stock solution
added to the monomer solution was altered accordingly:
0.2 mL stock solution (0.0342 mmol CTA) for DPn,targ 5 100,
0.05 mL stock solution (0.00855 mmol CTA) for
DPn,targ 5 400, 0.025 mL stock solution (0.00428 mmol CTA)
for DPn,targ 5 800. In all cases, the amount of dry anisole
added to the monomer was changed so that the ratio of
monomer to solvent remained 1:1 (w/v). When differing per-
centages of radical initiator to RAFT agent were desired, the
amount of VAZO88 in the stock solution was changed to
2
1
eluent at a flow rate of 0.35 mL min , at 40 8C. The SEC cali-
bration curve was based on linear polystyrene standards.
Samples were periodically withdrawn from the polymerization
reaction mixtures, diluted with THF, filtered with Acrodisc 0.2
lm PTFE filters, and injected into the SEC system without any
further purification. The b-hydroxy ether-modified polymers
were purified of unreacted alcohol and catalyst by dialysis
against acetone using Spectra/Por dialysis membrane with
2
1
MWCO 5 2000 g mol (wet in 0.05% sodium azide). Ther-
mogravimetric analyses (TGA) were performed on a Perkin-
Elmer TGA 4000 thermogravimetric analyzer. The samples
(
ca. 5 mg) were heated in a ceramic pan from 40 to 900 8C at
2
1
21
4.2 mg (0.0171 mmol for 10 mol % vs. RAFT CTA) or 12.6
a rate of 10 8C min under a nitrogen flow (20 mL min ).
Infrared (IR) spectra were collected on a Thermo Scientific
Nicolet iS10 FT-IR Spectrometer. The samples were prepared
by dissolving 10 mg of polymer in 0.5 mL of chloroform, fol-
lowed by casting of a thin film on a NaCl plate.
mg (0.0513 mmol for 30 mol % vs. CTA). For the experiment
conducted at 70 8C, the radical initiator was AIBN (2.8 mg,
0.0171 mmol; 10 mol % vs. RAFT CTA) instead of VAZO88.
Poly4VPO Macro-CTA Synthesis
In the following procedure, DPn,targ 5 100 with 20% VAZO88
vs. RAFT CTA (BPT). 4VPO (5.00 g, 34.2 mmol) was weighed
out in a 25 mL round bottom flask equipped with a stir bar
and dry anisole (3.0 mL) was added to the monomer. A solu-
tion of RAFT CTA and radical initiator was prepared in a
separate vial by dissolving BPT (92.5 mg, 0.342 mmol) and
VAZO88 (16.7 mg, 0.0684 mmol) in anisole (1.0 mL). The
CTA and initiator solution was then added to the reaction
flask and an additional 1 mL of anisole was used to ensure
complete transfer. The 25 mL round bottom flask was
capped with a rubber septum and sealed with electrical
tape. The reaction mixture was sparged for 1 h with nitro-
gen. The flask was then immersed in an oil bath at 90 8C for
5.75 h (60% monomer conversion) and then the reaction
was stopped by removal from heat and by opening the flask
to air. The reaction mixture was diluted with THF (10 mL)
and the polymer was precipitated by dripping the solution
into 400 mL of vigorously stirred diethyl ether. The soluble
material was decanted and the precipitated polymer was
transferred to 50 mL round bottom flask using THF (10 mL).
The solvent was then removed by rotary evaporation under
vacuum at 40 8C to afford poly4VPO as a light yellow solid
Synthetic Procedures
Model Studies Using CBr
of Epoxides
4
as the Catalyst for Alcoholysis
CBr
4
(16.6 mg, 0.050 mmol) and BuOH (0.7412 g, 10.0
mmol) were weighed out in a 2.5 mL glass vial with screw
top and then diluted with PhCN (1.0 mL). Phenyloxirane (57
mL, 0.50 mmol) was added to the reaction mixture and the
vial was capped. The reaction vial was then immersed to the
bottom of the cap in a heated oil bath (90 8C). Samples (50
mL) were periodically withdrawn using a syringe and diluted
3
with CDCl for analysis by NMR. Similar reactions were car-
ried out using 1 mL of anisole or toluene as the solvent
instead of PhCN. In separate experiments, the amount of
CBr4 was doubled (33.2 mg, 0.100 mmol) or quadrupled
(
66.4 mg, 0.200 mmol) to investigate the effect of catalyst
amount on the reaction rate. To examine the effect of the
alcohol structure on the reaction rate, BuOH was substituted
with 10.0 mmol of cyclohexanol (1.002 g), tert-butanol
(
0.741 g), benzyl alcohol (1.081 g), or phenol (0.941 g).
RAFT Polymerization of 4VPO
In the following procedure, DPn,targ (i.e., [4VPO] /[CTA] ) 5 200
3.06 g, 60% yield). SEC: M 5 7300 g mol , D- 5 1.09. 1
H
n
2
1
(
0
0
NMR (500 MHz, CDCl ): d 5 7.15-6.75 (2H), 6.75-6.2 (2H),
3
with 20 mol % radical initiator vs. the RAFT CTA. 4VPO
0.500 g, 3.42 mmol) was weighed out in a 5 mL glass vial
3
1
.85-3.65 (1H), 3.2-3.0 (1H), 2.9-2.65 (1H), 1.8-1.55 (1H),
.55-1.15 (2H).
(
equipped with a stir bar and dry anisole (0.4 mL) was added to
the monomer. A 103 stock solution of RAFT CTA and radical
initiator was prepared by dissolving BPT (46.3 mg, 0.171
mmol) and VAZO88 (8.4 mg, 0.0342 mmol) in anisole (1.0 mL),
and a 1/10 portion of the stock solution (0.1 mL) was added to
the reaction vial. The vial was capped with a rubber septum
and sealed with electric tape. The reaction mixture was sparged
for 30 min with a steady flow of nitrogen. The vial was then
immersed in an oil bath at 90 8C. As the reaction progressed,
samples (ca. 0.2 mL) were periodically withdrawn using a
nitrogen-purged syringe to determine monomer conversion and
molecular weight characteristics as described in the analytical
procedures.
Chain Extension of Poly4VPO Macro-CTA with Styrene
Poly4VPO macro-CTA (0.210g, 0.0240 mmol CTA) from the
preceding procedure was dissolved in anisole (0.624 mL)
and injected into a 5 mL glass vial equipped with a stir bar.
Styrene (0.500 g, 4.801 mmol; DPn,targ 5 200) was then
added to the reaction vessel. A 103 stock solution of radical
initiator was prepared by dissolving VAZO88 (11.7 mg,
0.0479 mmol) in anisole (1.0 mL) and a 1/10 portion of the
stock solution (0.1 mL) was added to the reaction vial. The
vial was capped, sealed, and sparged for 45 min with nitro-
gen. Samples for analysis of conversion and molecular weight
were periodically withdrawn as previously described.
1
134
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
General Procedure for the Post-Polymerization
Modification of Poly4VPO with Alcohols Using CBr as the
4
Catalyst
The desired alcohol (10.26 mmol, 60 eq. vs. epoxide) was
weighed out in a 4 mL glass vial with screw top and diluted
with PhCN (0.9 mL). A 253 stock solution of epoxide-
functionalized polymer was prepared by dissolving poly4VPO
SCHEME 1 General scheme for the reaction of epoxides with
4
alcohols catalyzed by CBr .
(
0.6250 g, 4.275 mmol epoxide, produced from the macro-
diluted with CH Cl (0.7 mL). A 103 stock solution of
2 2
CTA synthesis procedure) in PhCN (25 mL) and a 1/25 por-
tion (1 mL, containing 25 mg polymer, 0.171 mmol epoxide)
was then added to the alcohol solution. The vial was capped
and vigorously shaken for several seconds to ensure even
mixing. A 253 stock solution of catalyst was prepared by
epoxide-functionalized polymer was prepared by dissolving
poly4VPO (0.250 g, 1.71 mmol of epoxide, produced from
the macro-CTA synthesis procedure) in CH Cl (2.0 mL), and
2 2
a 1/10 portion (0.2 mL, containing 25 mg polymer, 0.171
mmol epoxide) was then added to the alcohol solution. The
vial was sealed with a screw-on cap and vigorously shaken
for several seconds to ensure even mixing. A 1003 stock
solution of catalyst was prepared by dissolving BF -Et O
dissolving CBr (0.1418 g, 0.4275 mmol) in PhCN (2.5 mL),
4
and a 1/25 portion (0.1 mL, containing 5.67 mg, 0.0171
mmol CBr ) was added to the reaction mixture. The vial was
4
3
2
resealed with a screw-on cap and then immersed to the bot-
tom of the cap in a heated oil bath (90 8C). After 72 h (for
secondary alcohols) or 48 h (for all other alcohols and phe-
nols) the reaction mixture was transferred to a dialysis bag
and diluted with acetone (4 mL). The mixture was then
placed in a 500 mL beaker equipped with a stir bar and dia-
lyzed against acetone (300 mL) over 36 h, with the acetone
bath being changed every 12 h. The polymer solution was
then transferred to a round bottom flask and the acetone
was removed by rotary evaporation under vacuum at 40 8C.
The b-hydroxy ether-modified polymer was characterized
according to the procedures listed in the analyses section.
NMR spectra, IR spectra, and TGA curves for all modified
2 2
(120 mg, 0.855 mmol) in CH Cl (10 mL), and a 1/100 por-
tion (0.1 mL, containing 1.2 mg (0.00855 mmol) of BF3-
Et O) was injected into the sealed reaction mixture. After 30
2
min, the modified polymers were analyzed as previously
described.
RESULTS AND DISCUSSION
Model Compound Studies Using CBr
4
CBr is a commercially available, easily handled solid, which
4
is stable to moisture at ambient temperatures and is highly
soluble in most organic solvents. It forms adducts with com-
pounds containing electron-rich heteroatoms by way of halo-
6
2–64
gen bonding,
of acid-labile protecting groups
anhydrides and carboxylic acids with phenols and alco-
and has been used to catalyze the removal
polymers that did not cross-link and were soluble in CDCl
are available in the Supporting Information (SI).
3
6
5–71
and the esterification of
7
2–74
27
Model Compound Studies Using BF as the Catalyst for
hols.
Yadav et al. reported that CBr could be used as
4
3
Alcoholysis of Epoxides
mild and efficient Lewis acid to transform epoxides into
b-hydroxy ethers using alcohols (Scheme 1). When the stud-
ied epoxides were reacted with large excesses (>50 eq.) of
alcohols such as methanol, iso-propanol, or tert-butanol in
BuOH (74.1 mg, 1.00 mmol, 5 eq. vs. epoxide) was weighed
out in a 2.5 mL glass vial with screw top and then diluted
with CH Cl (0.9 mL). Phenyloxirane (23 mL, 0.20 mmol)
2
2
was added to the reaction mixture and the vial was capped.
A 1003 stock solution of catalyst was prepared by dissolv-
ing BF -Et O (141.9 g, 1.00 mmol) in CH Cl (10.0 mL), and
4
the presence of 0.1 eq. of CBr and in the absence of solvent
at 60 8C, 2 to 4.5 h were required to consume all of the
epoxide. Product yields ranged from 70 to 95%, with the
shortest times and highest yields being reported for reac-
tions with the least sterically hindered alcohols.
3
2
2
2
a 1/100 portion (0.1 mL, containing 1.419 mg (0.010
mmol) BF -Et O) was injected into the sealed reaction mix-
3
2
ture. After 30 min, a sample was withdrawn with a syringe
and diluted with CDCl3 for NMR analysis. Similar reactions
were carried out with 1.00 mmol of cyclohexanol (100 mg),
tert-butanol (74.1 mg), benzyl alcohol (108 mg), or phenol
The application of these exact conditions to the modification
of polymer-bound epoxides is not possible due to the poor
solubility of the polymers in alcohols. Given the relative ease
with which the progress of small molecule reactions can be
monitored in comparison to polymer reactions, it was most
economical to first optimize the reaction conditions using
model compounds that closely mimic the structural motifs of
epoxide-containing polymers. Initial experiments were con-
ducted using glycidyl butyrate, a molecule that resembles
the repeat unit of poly(glycidyl methacrylate) (polyGMA),
one of the most easily-accessible epoxide-containing poly-
mers. When methanolysis of this compound was attempted,
the reaction only reached 25% conversion after 16 h. As pre-
viously noted, the presence of extraneous Lewis bases like
the ester carbonyl group in glycidyl butyrate can
(
94 mg) instead of BuOH. Note: Although not problematic
on a small scale, the reactions with primary and benzylic
alcohols were noticeably exothermic, and the reaction vials
became hot to the touch. Similar reactions were carried out
using glycidyl butyrate (28 mL, 0.20 mmol) instead of
phenyloxirane.
General Procedure for the Post-Polymerization
Modification of Poly4VPO with Alcohols Using BF as the
3
Catalyst
The desired alcohol (3.42 mmol, 20 eq. vs. epoxide) was
weighed out in a 2.5 mL glass vial with screw top and
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144
1135

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 1 Conversion of phenyloxirane (0.5 M in the selected solvent) to b-hydroxy ether at 90 8C with 20 eq. of alcohol and vari-
able a) solvent (10 mol % CBr with BuOH as the alcohol), b) amount of catalyst (in PhCN with BuOH as the alcohol), and c) alco-
hol (10 mol % CBr in PhCN).
4
4
dramatically slow or even stop the alcoholysis of the epox-
ides by competitively forming adducts with the acid catalyst.
It is also likely that if these reaction conditions were applied
to polyGMA, transesterification side reactions would occur
between the hydroxy groups of the resulting b-hydroxy ether
products and the ester groups, which link them to the poly-
mer backbone. Thus, the use of glycidyl ester-type polymers
like polyGMA as substrates for modification with alcohols
was considered to be unfeasible.
porting Information Scheme S1, Fig. S1). Switching the sol-
vent to toluene (less polar and less likely to complex with
CBr4 than anisole) or PhCN (more polar than anisole)
reduced the induction period significantly. Unlike the previ-
ously mentioned study on solvent affects, there was no
clear correlation between solvent polarity and reaction rate.
3
3
The presence of induction periods, which have not been pre-
viously reported for the Lewis acid-catalyzed alcoholysis of
epoxides, raised questions about the role of CBr in the reac-
4
There are several examples of epoxide-functionalized poly-
mers in the literature that are not burdened by carbonyl
groups and could therefore serve as alternatives to GMA-
tion mechanism. Although some reports credit the catalytic
2
7,73
activity of CBr to its Lewis acid character,
others claim
It is known that CBr4
reacts with primary alcohols at elevated temperatures to
4
6
6–69
that CBr acts as a source of HBr.
4
7
5
based polymers. Poly(4-vinylphenyloxirane) (poly4VPO) is
particularly interesting because the attached epoxide groups
are similar in structure to phenyloxirane. A substantial num-
ber of publications that examine the acid-catalyzed alcoholy-
sis of epoxides show that phenyloxirane exhibits the fastest
reaction rates and highest yields of regioselective products
as a result of the aromatic group stabilizing the partial posi-
tive charge on the benzylic carbon of the protonated or com-
7
7
produce HBr and the corresponding aldehydes. The mecha-
nism of this reaction has not been studied in detail, but simi-
lar reactions between CCl4 and primary or secondary
alcohols have been reported that involve the formation of
alkyl hypochlorite intermediates, which form HCl and alde-
7
8,79
hydes or ketones.
In the NMR spectra of the CBr -cata-
4
lyzed reactions with BuOH, a triplet associated with the
aldehyde proton of butyraldehyde was seen at d 5 9.72 ppm
7
6
plexed epoxide through resonance.
(
Supporting Information Fig. S1). It is therefore possible that
The methanolysis of phenyloxirane (60 eq. of alcohol and 0.1
HBr, not CBr , is the active catalyst in this system, and the
4
2
7
eq. of CBr at 60 8C) was reported by Yadav et al. to pro-
4
induction period may be attributed to the time required for
a sufficient amount of HBr to accumulate in the reaction
mixture, which is likely to be solvent-dependent. Strong
Brønsted acids often result in large quantities of side prod-
ucts, but the acidity of HBr is much reduced when in organic
solvents (for instance, pK 5 1.06 in DMSO and 5.5 in
MeCN) and is therefore quite mild under these conditions.
Based on these solvent studies, all further reactions were
carried out using PhCN (2 mL per 1 mmol of phenyloxirane)
as the solvent at 90 8C.
ceed to completion in 2 h. To determine if this modification
could be efficiently applied to solvated poly4VPO, the same
reaction was carried out in anisole (2 mL per 1 mmol of
phenyloxirane). It was anticipated that dilution of the reac-
tion mixture would increase the required time for complete
conversion but even after 48 h no methanolysis could be
detected by NMR. The reaction was then attempted at a
higher temperature (90 8C) using 20 eq. of a higher boiling
a
80
(
to avoid evaporation) primary alcohol, BuOH [Fig. 1(a)].
There was a significant induction period observed during the
first 45 h of this reaction where only 10% of the epoxide
was consumed, but in the subsequent 43 h, the reaction rate
greatly increased and an additional 71% of the epoxide was
consumed. The rate of the alcoholysis of phenyloxirane was
monitored by integrating the signals of the epoxide protons
at d 5 2.8, 3.1, and 3.9 ppm and the signal of the benzylidene
proton of the b-hydroxy ether product at d 5 4.4 ppm (Sup-
4
Next, the effect of increasing the amount of CBr on the reac-
tion rate was studied. Experiments using 0.1 eq., 0.2 eq., and
0.4 eq. of CBr4 relative to epoxide were conducted with 20
eq. of BuOH [Fig. 1(b)]. Although the time required to com-
pletely consume the epoxide was slightly reduced as the con-
centration of CBr
4
increased, the improvement was not
significant enough to justify the addition of extra catalyst,
1
136
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 2 General scheme for the polymerization of 4VPO and chain-extension with styrene using various RAFT CTAs. Chain-
extension was only performed with the BPT-functionalized poly4VPO.
and all further studies were conducted using 0.1 eq. of CBr
relative to the epoxide.
4
additional benefit of making all subsequent work-up proce-
dures much easier and increased the overall yield of 4VPO
from 34 to 75%.
Lastly, the effect of alcohol structure on the reaction rate
was investigated [Fig. 1(c)]. As expected, the more sterically
hindered secondary alcohol (cyclohexanol, completed in
RAFT Polymerization of 4VPO
5
6
In a recent publication, we reported that 4VPO could be
polymerized in a controlled manner using low-copper-
catalyst-concentration atom transfer radical polymerization
(ATRP), even in the absence of externally added reducing
agents. Here, we demonstrate that the RAFT polymerization
technique, which is well-known for its tolerance towards
42 h) reacted slower than the primary alcohol (BuOH, com-
pleted in 31 h). Tertiary alcohols are known to react some-
what slower than secondary alcohols, but, interestingly,
under these conditions no reaction whatsoever was observed
between tert-butanol and phenyloxirane, even after several
days. The complete lack of reaction lends further credence to
the possibility that HBr is the active catalyst in this solvated
system, because unlike the hypobromites of primary and sec-
ondary alcohols, the hypobromites of tertiary alcohols cannot
form a carbonyl compound and thus release HBr. Alcohols
that contain electron-rich substituents, such as the benzyl,
propargyl, and allyl moieties, are known to react with epox-
ides at significantly faster (two to sevenfold) rates than pri-
mary alcohols in the presence of acid catalysts. This was
found to be true in the case of benzyl alcohol, which after a
brief induction period of 2 to 3 h, rapidly consumed all of
the phenyloxirane in 2 h. Phenols are significantly more
acidic than alcohols and are known to react with epoxides
without the aid of external catalysts. No significant difference
in the rate of the alcoholysis of phenyloxirane with phenol
8
1
reactive functional groups (including epoxides),
is also
capable of producing homopolymers and block copolymers
from 4VPO using a variety of chain-transfer agents (CTAs),
as shown in Scheme 2.
To begin the investigation, 4VPO was polymerized in anisole at
70 8C using BPT as the CTA with the degree of polymerization
at complete monomer conversion (DPn,targ 5 [4VPO] /[CTA]
0
0
)
set to 200, and with 10 mol % AIBN relative to CTA, acting as
the additional radical source (Fig. 2). The monomer conversion
was periodically determined by integrating the NMR signals of
the vinyl and epoxide hydrogen atoms. The first-order kinetic
plot [Fig. 2(a)] was only linear up to about 40% monomer con-
version, which was reached in about 10 h. The noticeable
decrease in the rate of monomer consumption after that point
was due to an insufficient amount of leftover AIBN (with half-
was seen when CBr
4
was added to the reaction mixture.
82
life time of 5 h at 70 8C). The polymerization was well-
controlled, as indicated by the linear increase of M with con-
Improved Synthesis of 4VPO
n
5
6
We recently reported a simplified version of the Truxa pro-
version [Fig. 2(b)]. The MWDs remained narrow and symmet-
ric, and shifted smoothly towards higher molecular weights
with very little tailing as the reactions proceeded [Fig. 2(c)],
proving that the chosen CTA (BPT) efficiently exchanged
between propagating and dormant chains. Raising the tempera-
ture from 70 to 90 8C and using a radical initiator with longer
life-time than AIBN, VAZO88, increased the polymerization rate
7
5
cedure for synthesizing 4VPO that used less-active reagents
NaBH , potassium tert-butoxide) in the reduction and ring-
(
4
closure steps, and therefore only required a single distillation
at the end of the synthesis to yield very-pure 4VPO. The initial
step of the Truxa procedure, which remained unmodified in
5
6
3
our previous work, employed AlCl in a chlorinated solvent
to effect the Friedel–Crafts acylation of 2-chloroethylbenzene
with chloroacetyl chloride to give the desired aromatic ketone
marginally [Fig. 2(a)], while the evolution of M and MWD dis-
persity [Fig. 2(b)] were not affected significantly.
n
(
4-(2-chloroethyl)phenylchloromethylketone) in 50% yield,
with the remainder being an unidentified polymeric material.
It was ascertained that using CS2 instead of the chlorinated
solvent in the first step suppressed the formation of poly-
meric side product and gave the aromatic ketone in nearly
quantitative yield. The absence of polymeric material had the
Doubling the initial amount of VAZO88 to 20 mol % vs. BPT
increased the polymerization rate [Fig. 3(a)], and a higher
conversion of 77% could be attained at 22 h. The evolution
of M and dispersity [Fig. 3(b)] was nearly identical to that
n
seen for the reactions with 10 mol % AIBN or VAZO88 up to
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144
1137

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
21
FIGURE 2 RAFT of 4VPO (6.84 mol L in anisole) with BPT as the CTA (DPn,targ 5 200) using 10 mol % of AIBN (70 8C) or VAZO88
90 8C) with respect to BPT. a) Kinetics; (b) evolution of molecular weights and dispersities; and c) evolution of SEC traces of the
polymers (for the reactions using 10 mol % AIBN) with monomer conversion (shown at each curve).
(
high monomer conversions (>70%), at which point a grow-
ing amount of high molecular weight polymer was observed
in the SEC traces, which was most likely the result of cou-
pling and/or branching, possibly caused by a small amount
containing monomers, and should be avoided. Based on all of
the above observations, BPT was the most suitable CTA for
the RAFT polymerization of 4VPO.
5
6,83
High molecular weight functional polymers are very desirable
for many applications, and the limitations of the established
optimal conditions for the RAFT polymerization of 4VPO
(20 mol % VAZO88 vs. BPT at 90 8C) were explored next (Fig.
4). At DPn,targ 5 100, relatively high conversion of 76% could
of cross-linker impurities.
The level of control provided by several other readily available
CTAs was also explored under the same conditions (Fig. 3).
Polymerization of 4VPO mediated by CDT reached slightly
higher conversion (85% in 25 h) than BPT but dispersities
were higher than for BPT and the SEC traces (Supporting
be reached in less than 8 h, and the polymers were very well-
defined and had symmetrical MWDs ( D- 5 1.1). Reaching this
Information Fig. S2a) showed a large amount of coupling at
high conversions ( D- 5 3.2 at 85%). BDB-controlled polymer-
conversion was not possible when DPn,targ was increased to
800, even after 48 h, because by then the vast majority of
VAZO88 had decomposed. The SEC traces [Fig. 4(c)] also
revealed that the fraction of undesired poorly defined high
molecular weight polymer became more prevalent at higher
values of DPn,targ. This was probably due to the fact that as the
chain-length of the polymer increased, more units of the poten-
tially present cross-linker impurity were incorporated, which
increased the likelihood of branching to occur. Nevertheless,
well-defined polymers of relatively high molecular weights
could be synthesized under the optimized reaction conditions.
ization of 4VPO also showed higher dispersity values than
BPT, mostly because of tailing in the SEC traces (Supporting
Information Fig. S2b). Well-controlled polymer could not be
obtained when BMPA was utilized as the RAFT agent, likely
due to the ability of the propionic acid moiety in BMPA to
participate in acid-catalyzed opening of the epoxide rings in
4VPO and/or poly4VPO, leading to branching and eventually
cross-linking. Indeed, the SEC traces (Supporting Information
Fig. S2c) showed an increasingly large amount of high molec-
ular weight product with multimodal MWD as higher mono-
mer conversions (>40%) were reached. Clearly, even mildly
acidic or nucleophilic RAFT agents are reactive enough to neg-
atively impact the control over the polymerization of epoxide-
One of the most important advantages that controlled/
“living” radical polymerization techniques like RAFT offer is
the ability to produce well-defined block copolymers through
21
FIGURE 3 RAFT of 4VPO (6.84 mol L in anisole) at 90 8C with CDT, BDB, BMPA, or BPT as CTA (DPn, targ 5 200) using 20 mol %
of VAZO88 with respect to the CTA. a) Kinetics; (b) evolution of molecular weights and dispersities; and c) evolution of SEC traces
of the polymers (for the reaction employing BPT as the CTA) with monomer conversion (shown at each curve).
1
138
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
2
1
FIGURE 4 RAFT of 4VPO (6.84 mol L in anisole) at 90 8C with BPT as the CTA (DPn,targ 5 100, 200, 400, or 800) and using 20 mol
of VAZO88 versus the CTA. a) Kinetics; b) evolution of molecular weights and dispersities; and (c) evolution of SEC traces of the
polymers (DPn, targ 5 800) with monomer conversion (shown at each curve).
%
chain extension reactions. Epoxide-containing segmented
copolymers can be reacted with small molecules to introduce
novel functionalities or can be cross-linked to stabilize previ-
ously established morphologies against changing conditions.
By chain extending butyl trithiocarbonate-end-functionalized
poly4VPO macro-CTA with styrene, well-defined poly(4VPO-
b-styrene) copolymers were produced (Fig. 5). As with all
previous polymerizations, the chain-extension was very well-
controlled up to comparatively high conversions, indicating
that the x-chain-end functionality of the macro-CTA was
preserved.
( D- 5 1.79) and bimodal [Fig. 6(a)], indicating that both linear
and branched polymers were produced. No alkoxide anions
are generated by the ring-opening of epoxides under acidic
conditions, and ether linkages are inert towards further sub-
stitutions, so the coupling of the polymers is likely caused by
reactions between the primary hydroxy groups of the
b-hydroxy ether products and unreacted epoxides on other
polymer chains (Scheme 3, bottom left). It was also noted in
this experiment and in all other post-polymerization modifi-
cations that if the modified polymer was isolated by precipi-
tation or dialysis and all solvent was removed by
evaporation before complete conversion of the epoxide had
occurred, then the partially modified polymer would rapidly
cross-link to form a gel for the reasons already discussed.
Modification of Poly4VPO with Alcohols Using CBr
as the Catalyst
4
An advantage of the post-polymerization modification tech-
nique is that many different polymers can be prepared in a
short amount of time from a “universal” precursor. Most pro-
cedures for the modification of epoxide-functionalized poly-
mers use excess nucleophile relative to epoxide (usually 2–
Increasing the amount of BuOH relative to the epoxide greatly
lowered the MWD dispersities of the modified polymers [Fig.
6(b)]. At 20 eq. of BuOH, the SEC trace had a clearly-visible
shoulder of high-molecular-weight polymer, which became less
apparent as the amount of BuOH was increased. Modifications
of poly4VPO using 60 eq. of 21 different alcohols and 3 phe-
nols (Scheme 3, right) were carried out, and 16 new b-hydroxy
ether-functionalized polymers were successfully prepared. To
ensure that complete conversion of the epoxide to b-hydroxy
ether products had occurred before purification was attempted,
secondary alcohols were reacted for 72 h, and all other alco-
hols and phenols were reacted for 48 h. The modified polymer
20 eq., and sometimes more if the nucleophile is an inexpen-
sive, low molecular weight compound) to suppress cross-
linking side reactions. First, the modification of well-defined
2
1
poly4VPO (M 5 7300 g mol , D- 5 1.09) was attempted
n
with only 5 eq. of BuOH using PhCN as the solvent, and 0.1
eq. of CBr as the catalyst at 90 8C (Scheme 3). After 36 h,
4
no NMR peaks associated with unreacted the epoxides could
be seen. SEC traces of the resulting polymers were broad
21
FIGURE 5 RAFT of styrene (6.63 mol L in anisole) at 90 8C with butyl trithiocarbonate-functionalized poly4VPO as CTA (DPn,
targ 5 200) and using 20 mol % of VAZO88 versus the macro-CTA. a) Kinetics; b) evolution of molecular weights and dispersities;
c) evolution of SEC traces of the polymers with monomer conversion (shown at each curve).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144
1139

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 3 General scheme for the post-polymerization modification of poly4VPO with alcohols (top left). Cross-linking side reac-
tion between the b-hydroxy ether products and unreacted epoxide groups (bottom left). List of all alcohols used in the modifica-
tion of poly4VPO (right). Names of the alcohols that either showed no reaction or formed cross-linked polymers are boxed.
solutions were dialyzed against acetone to remove the excess
alcohol and, after evaporation of the acetone, the modified poly-
mers were characterized, where possible, by SEC, NMR, IR, and
TGA, as shown in the Supporting Information.
polymers were completely soluble in methanol. Polymers
modified with longer alkyl chain alcohols, e.g., 1-decanol or
2-octanol, could be solubilized in isopropanol. Cyclohexanol-
and phenol-modified polymers were soluble in far fewer sol-
vents, and once most of the acetone from dialysis had been
removed, could only be resolubilized in DMSO.
In most cases, polymers modified by primary, secondary, and
benzylic alcohols, and phenols were successfully produced with
minimal branching even when reactive functionalities such as
halides, azides, and unsaturated moieties were present (Sup-
porting Information Table S1). The least amount of branching
was observed when benzylic alcohols were employed, probably
because they react with the epoxides much faster than the pri-
mary alcohol groups of the b-hydroxy ether products.
When relatively large and/or sterically hindered reagents like
menthol, cyclododecanol, or 2-naphthol were used, the modified
polymers underwent extensive cross-linking upon purification
and removal of solvent due to the presence of unreacted epoxide
groups in the products. This problem persisted even when reac-
tion times were extended to 120 h. The modification of polyGMA
with large nucleophiles has been reported to be inefficient,
owing to the bulkiness of the newly attached groups, which pre-
vents access of the nucleophile to the still present unreacted
Most of the modified polymers were more polar than
poly4VPO and could be solubilized in a wider range of sol-
vents. Methanol-, ethanol-, isopropanol-, and BuOH-modified
84
epoxide groups. Several of the alcohols showed little or no
FIGURE 6 SEC traces of poly4VPO and the product of its CBr
4
-catalyzed modification with BuOH at different of equivalents of alco-
hol relative to epoxide. The dispersities are shown at each curve.
1
140
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
3
FIGURE 7 Molecular structures and NMR spectra (in CDCl ) of a) poly4VPO, b) BuOH-modified polymer, c) benzyl alcohol-
modified polymer, and d) allyl alcohol-modified polymer.
reaction with the poly4VPO. For instance, 9-anthracenemethanol
could not be solubilized in PhCN or toluene, even when the num-
ber of equivalents were reduced sixfold and higher temperature
was applied. 4-Hydroxy-2-butanone contains a carbonyl group,
which may compete with the epoxide for the acid catalyst, and
thereby prevent activation of the epoxide towards substitution.
As in the model compound studies, tert-butanol showed no reac-
tion with poly4VPO. There is no clear rationalization for why the
modification of the epoxide-containing polymer with 10-
undecen-1-ol failed in light of the fact that poly4VPO could be
successfully modified with 1-decanol, another long chain pri-
mary alcohol, and with allyl alcohol, which also contains a vinyl
group. It is possible that impurities in the alcohol deactivated the
catalyst. 2-Bromoethanol rapidly decomposed at 90 8C and only
partially modified the poly4VPO, resulting in cross-linking after
purification. The furfuryl alcohol solution underwent macrogela-
tion in situ (the only alcohol to do so), either because the alcohol
itself polymerized under the acidic, high temperature conditions,
ing Information Fig. S3). In all cases, the signals of the epoxide
protons in poly4VPO at d 5 2.8, 3.1, and 3.8 ppm [Fig. 7(a)] com-
pletely disappeared and were replaced by the signals of the ben-
zylidene proton centered at d 5 4.4 ppm and -CH -OH protons
2
at d 5 3.6 ppm of the b-hydroxy ether products [Fig. 7(b–d)].
Signals from the newly attached substituents could also be eas-
ily identified. For example, the benzyl alcohol-modified polymer
[Fig. 7(c)] had a strong signal centered at d 5 7.25 ppm from the
aromatic protons of the benzyl group, and the vinyl protons of
allyl alcohol modified polymer [Fig. 7(d)] had signals at d 5 5.1
and 5.9 ppm. Additional NMR spectra are presented in the SI
(Supporting Information Fig. S3).
FT-IR spectroscopy was useful for detecting the presence of
functional groups, which do not have protons or whose pro-
ton NMR signals are difficult to distinguish. The epoxide moi-
8
6
ety has two characteristic adsorption bands, the first at
2
1
9
15 cm , which is caused by stretching of the CAO bond,
21
85
and the second at 3050 cm , which is caused by the CAH
stretch. These bands are easily seen in the IR spectrum of
poly4VPO [Fig. 8(a)]. In all of the IR spectra of modified
polymers [Fig. 8(a–c)], these peaks were replaced with a
or because Diels-Alder dimers were formed that served as
bridges between polymer chains.
The composition of the modified polymers was confirmed by
NMR spectroscopy (Fig. 7; additional spectra presented Support-
21
strong band from 1000 to 1100 cm , which is typical of
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144
1141

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 8 IR spectra of thin films (from chloroform) of a) poly4VPO, b) BuOH-modified, c) 3-azido-1-propanol-modified, and d) pro-
pargyl alcohol-modified poly4VPO. CBr was used as the catalyst in all cases.
4
ether or alcohol CAO stretches, and a broad adsorption cen-
in the model compound study) to determine the effect of alco-
hol concentration on the extent of modification and branching.
SEC traces of the resulting modified polymers (Fig. 9) showed
that only 10 eq. of BuOH vs. epoxide was required to produce
polymers with characteristics similar to those observed in the
2
1
87
tered at 3350 cm , typical of an alcohol OAH stretch. The
2
1
particularly strong adsorption at 2096 cm
for 3-azido-1-
propanol-modified polymer [Fig. 8(c)] is caused by the asym-
8
8
metric vibration of the N moiety. The sharp peak at
3
2
1
3288 cm
caused by the stretching of the terminal C-H
4
CBr -catalyzed reaction using 60 eq. of BuOH. Similar reac-
8
7
bond of alkynes is recognizable in the spectrum of propar-
gyl alcohol-modified polymer [Fig. 8(d)]. Additional IR spec-
tra are presented in the Supporting Information Fig. S4.
tions were carried out using 20 eq. of cyclohexanol, tert-buta-
nol, benzyl alcohol, propargyl alcohol, and allyl alcohol, and in
all cases well-defined b-hydroxy ether-functionalized polymers
were obtained (Supporting Information Table S2). When 20
The thermal stability of the modified polymers that could be
obtained as solids was analyzed by TGA. All polymers were
ground into powders except for 1-decanol-modified polymer,
which had a wax-like consistency. These data are available in
the Supporting Information Fig. S5.
Modification of Phenyloxirane and Poly4VPO with
Alcohols Using BF As the Catalyst
3
BF is a relatively strong Lewis acid, which can serve as an
3
alternative to CBr for the ring-opening of epoxides by alco-
4
hols. Phenyloxirane (0.2 M in CH Cl ) was used as a model
2
2
compound in BF
opening reactions, employing 5 eq. of hydroxy compounds
BuOH, cyclohexanol, tert-butanol, benzyl alcohol, or phenol)
3
-catalyzed (0.05 eq. vs. epoxide) ring-
(
at ambient temperature. In all cases, NMR analysis showed
that the epoxide was completely consumed within 30 min.
The signals in the NMR spectra of the b-hydroxy ether prod-
ucts of phenyloxirane and the alcohols were easily recog-
nized but the NMR spectrum of the reaction with phenol
showed none of the desired product.
FIGURE 9 SEC traces of poly4VPO and BuOH-modified poly-
mers produced using different amounts of BuOH relative to
epoxide (with dispersities shown at each curve).
Poly4VPO was then modified using 5, 10, 20, 40, or 60 eq. of
BuOH (under conditions similar to those previously described
1
142
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
eq. of phenol was used in this reaction, extensive branching
was observed in the SEC traces after 30 min ( D- 5 2.28) and
after several hours insoluble polymer began to precipitate out
of the mixture.
3 G. Dake, In Comprehensive Heterocyclic Chemistry III; A. R.
Taylor, C. A. Katritzky, E. F. V. Ramsden, R. J. K. Scriven, Eds.;
Elsevier: Oxford, 2008; pp 173–233.
4
5
R. E. Parker, N. S. Isaacs, Chem. Rev. 1959, 59, 737–799.
ꢀ
J. Kalal, F. Svec, V. Marou ꢀs ek, J. Polym. Sci. Part C: Polym.
Although BF is not as easily handled as CBr and is suscep-
Symp. 1974, 47, 155–166.
3
4
tible to degradation by moisture, it appears to be superior
from every other viewpoint as a catalyst for the alcoholysis
6 O. I. Strube, L. Nothdurft, Z. Abisheva, G. Schmidt-Naake,
Macromol. Chem. Phys. 2012, 213, 1274–1284.
7
M. Benaglia, A. Alberti, L. Giorgini, F. Magnoni, S. Tozzi,
Polym. Chem. 2013, 4, 124–132.
N. V. Tsarevsky, S. A. Bencherif, K. Matyjaszewski, Macromo-
lecules 2007, 40, 4439–4445.
B. Rickborn, In Comprehensive Organic Synthesis; B. M. T.
Fleming, Ed.; Pergamon: Oxford, 1991; pp 733–775.
of the phenyloxirane moiety in poly4VPO. Unlike CBr , BF3
4
can catalyze the reaction of tertiary alcohols with epoxides
because pre-catalytic reactions between the acid catalyst
and alcohols are not required. For the same reason, induc-
8
9
tion periods were not observed with BF . The other advan-
3
tages of BF compared to CBr are faster reactions at lower
3
4
1
2
0 Y. H. Liu, Q. S. Liu, Z. H. Zhang, J Mol. Catal. A: Chem.
008, 296, 42–46.
temperatures are possible and fewer equivalents of alcohol
relative to epoxide are required to achieve efficient
modification.
1
1 G. A. Olah, A. P. Fung, D. Meidar, Synthesis 1981, 1981,
2
80–282.
1
2 G. H. Posner, D. Z. Rogers, C. M. Kinzig, G. M. Gurria, Tetra-
hedron Lett. 1975, 16, 3597–3600.
CONCLUSIONS
1
3 G. H. Posner, D. Z. Rogers, J. Am. Chem. Soc. 1977, 99,
8
208–8214.
Aryl-substituted epoxides like phenyloxirane are far more
susceptible to acid-catalyzed alcoholysis than alkyl-
substituted epoxides. In this work, the conditions were opti-
mized for reacting phenyloxirane with various alcohols using
CBr4 as the catalyst. A procedure for the synthesis of an
aryl-epoxide-functionalized monomer, 4-vinylphenyloxirane
1
4 G. H. Posner, M. Hulce, R. K. Rose, Synth. Commun. 1981,
1, 737–741.
1
1
5 L. Saikia, J. K. Satyarthi, D. Srinivas, P. Ratnasamy, J. Catal.
007, 252, 148–160.
2
1
6 M. W. C. Robinson, R. Buckle, I. Mabbett, G. M. Grant, A. E.
Graham, Tetrahedron Lett. 2007, 48, 4723–4725.
7 M. W. C. Robinson, A. M. Davies, R. Buckle, I. Mabbett, S. H.
Taylor, A. E. Graham, Org. Biomol. Chem. 2009, 7, 2559–2564.
8 J. Otera, Y. Yoshinaga, K. Hirakawa, T. Nakata, Tetrahedron
Lett. 1985, 26, 3219–3222.
(
4VPO), in high yield was developed, and well-defined homo-
1
polymers and block copolymers were produced from 4VPO
using RAFT polymerization. Poly4VPO was modified in a
combinatorial manner using a library of alcohols and phe-
nols, and the resulting b-hydroxy ether-functionalized poly-
mers were characterized. For modifications catalyzed by
1
1
9 J. Otera, Y. Niibo, N. Tatsumi, H. Nozaki, J. Org. Chem.
988, 53, 275–278.
1
CBr , reactions were most efficient when carried out at high
4
20 J. Otera, N. Yoshihisa, N. Hitosi, Tetrahedron 1991, 47,
temperature over several days in benzonitrile, and branch-
ing/cross-linking side reactions were prevented by using rel-
atively large excess of alcohol. BF3 was a more efficient
catalyst for the alcoholysis of poly4VPO, and modification
reactions could be carried out at ambient temperature
within 30 min using fewer equivalents of alcohol to obtain
well-defined b-hydroxy ether-modified polymers with virtu-
ally no branching.
7625–7634.
21 B. M. Choudary, Y. Sudha, Synth. Commun. 1996, 26, 2989–2992.
22 J. M. Chapuzet, S. Beauchemin, B. Daoust, J. Lessard, Tetra-
hedron 1996, 52, 4175–4180.
23 D. B. G. Williams, M. Lawton, Org. Biomol. Chem. 2005, 3,
3
269–3272.
2
4 H. Firouzabadi, N. Iranpoor, A. A. Jafari, S. Makarem, J.
Mol. Catal. A: Chem. 2006, 250, 237–242.
5 A. R. Pourali, S. Ghayeni, F. Afghahi, Bull. Korean Chem.
Soc. 2013, 34, 1741–1744.
6 I. Mohammadpoor-Baltork, S. Tangestaninejad, H. Aliyan, V.
Mirkhani, Synth. Commun. 2000, 30, 2365–2374.
7 J. S. Yadav, B. V. S. Reddy, K. Harikishan, C. Madan, A. V.
Narsaiah, Synthesis 2005, 17, 2897–2900.
8 B. H. Kim, F. Piao, E. J. Lee, J. S. Kim, Y. M. Jun, B. M. Lee,
Bull. Korean Chem. Soc. 2004, 25, 881–888.
9 G. Prestat, C. Baylon, M. P. Heck, C. Mioskowski, Tetrahe-
dron Lett. 2000, 41, 3829–3831.
0 M. Moghadam, S. Tangestaninejad, V. Mirkhani, I.
Mohammadpoor-Baltork, S. A. Taghavi, Catal. Commun. 2007,
2
2
ACKNOWLEDGMENTS
2
The authors thank Dr. Yannick Borguet and Houliang Tang for
the synthesis of the RAFT CTAs. Perkin-Elmer is acknowledged
for lending the TGA instrument. The authors also gratefully
acknowledge the financial support provided by Southern Meth-
odist University in the form of start-up funds for NVT.
2
2
3
8
, 2087–2095.
REFERENCES AND NOTES
3
1 C. Moberg, L. R aꢁ kos, L. Tottie, Tetrahedron Lett. 1992, 33,
1
K. A. G u€ nay, P. Theato, H. A. Klok. J Polym. Sci. Part A:
Polym. Chem. 2013, 51, 1–28.
2191–2194.
2 K. Venkatasubbaiah, X. Zhu, E. Kays, K. I. Hardcastle, C. W.
3
2
J. Gorzynski Smith, Synthesis 1984, 629–656.
Jones, ACS Catal. 2011, 1, 489–492.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144
1143

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
3
3 A. S. Capes, A. T. Crossman, L. A. Webster, M. A. J.
62 S. C. Blackstock, J. P. Lorand, J. K. Kochi, J Org. Chem.
1987, 52, 1451–1460.
Ferguson, I. H. Gilbert, Tetrahedron Lett. 2011, 52, 7091–7094.
3
4 R. P. Hanzlik, A. Hamburg, J Am. Chem. Soc. 1978, 100,
745–1749.
63 J. Murray, K. Paulsen, P. Politzer, Proc. Indian Acad. Sci.
(Chem. Sci.). 1994, 106, 267–275.
1
3
5 J. Barluenga, H. V aꢁ zquez-Villa, A. Ballesteros, J. M.
64 W. S. Zou, J. Han, W. J. Jin, J. Phys. Chem. A 2009, 113,
Gonz aꢁ lez, Org. Lett. 2002, 4, 2817–2819.
10125–10132.
36 N. Iranpoor, H. Adibi, Bull. Chem. Soc. Jpn. 2000, 73, 675–680.
65 A. S. Y. Lee, F. Y. Su, Y. C. Liao, Tetrahedron Lett. 1999, 40,
1
323–1326.
37 P. Salehi, B. Seddighi, M. Irandoost, F. K. Behbahani, Synth.
Commun. 2000, 30, 2967–2973.
66 A. Shih-Yuan Lee, Y. J. Hu, S. F. Chu, Tetrahedron 2001, 57,
2
6
6
6
121–2126.
3
3
4
4
4
8 G. D. Yadav, S. Singh, Tetrahedron Lett. 2014, 55, 3979–3983.
9 N. Iranpoor, P. Salehi, Synthesis 1994, 1994, 1152–1154.
0 K. Jeyakumar, D. K. Chand, Synthesis 2008, 2008, 807–819.
1 K. Fagnou, M. Lautens, Org. Lett. 2000, 2, 2319–2321.
2 M. Caron, K. B. Sharpless, J. Org. Chem. 1985, 50, 1557–1560.
7 M. Y. Chen, A. S. Y. Lee, J. Org. Chem. 2002, 67, 1384–1387.
8 A. S. Y. Lee, C. L. Cheng, Tetrahedron 1997, 53, 14255–14262.
9 M. Y. Chen, L. N. Patkar, K. C. Lu, A. S. Y. Lee, C. C. Lin, Tet-
rahedron 2004, 60, 11465–11475.
0 M. Y. Chen, K. C. Lu, A. Shih-Yuan Lee, C. C. Lin, Tetrahe-
dron Lett. 2002, 43, 2777–2780.
7
4
3 P. N. Guivisdalsky, R. Bittman, J. Am. Chem. Soc. 1989, 111,
077–3079.
3
7
1 A. S. Y. Lee, F. Y. Su, Tetrahedron Lett. 2005, 46, 6305–6309.
4
4 N. Iranpoor, B. Zeynizadeh, Synth. Commun. 1999, 29,
017–1024.
7
2
2 A. Shih-Yuan Lee, H. C. Yang, F. Y. Su, Tetrahedron Lett.
001, 42, 301–303.
1
4
5 S. A. Taghavi, M. Moghadam, I. Mohammadpoor-Baltork, S.
7
3 L. Zhang, Y. Luo, R. F. Wu, J. Green Chem. 2007, 9, 1022–1025.
4 S. V. Chankeshwara, R. Chebolu, A. K. Chakraborti, J. Org.
Tangestaninejad, V. Mirkhani, A. R. Khosropour, V. Ahmadi,
7
Polyhedron 2011, 30, 2244–2252.
Chem. 2008, 73, 8615–8618.
46 N. Iranpoor, M. Shekarriz, F. Shiriny, Synth. Commun. 1998,
7
5 R. Truxa, M. Suchop aꢁ rek, Die Makromol. Chem. 1990, 191,
28, 347–366.
1
931–1939.
47 V. Mirkhani, S. Tangestaninejad, B. Yadollahi, L. Alipanah,
7
6 D. L. Whalen, In Advances in Physical Organic Chemistry; J.
Tetrahedron 2003, 59, 8213–8218.
8 R. Dalpozzo, M. Nardi, M. Oliverio, R. Paonessa, A.
Procopio, Synthesis 2009, 2009, 3433–3438.
P. Richard, Ed.; Academic Press, 2005; Vol. 40, pp 247–298.
4
7
7 T. Bolas, C. E. Groves, J Chem. Soc. 1870, 23, 161–165.
8 R. I. Khusnutdinov, N. A. Shchadneva, R. Y. Burangulova, Z.
7
49 E. Schoffers, L. Kohler, Tetrahedron: Asymmetry 2009, 20,
S. Muslimov, U. M. Dzhemilev, Russ. J. Org. Chem. 2006, 42,
1897–1902.
1
615–1621.
5
0 A. J. L. Leit a~ o, J. A. R. Salvador, R. M. A. Pinto, M. L. S aꢁ e
7
9 J. W. Bae, E. J. Jang, D. H. Jo, J. S. Lee, K. H. Lee, J. Mol.
Melo, Tetrahedron Lett. 2008, 49, 1694–1697.
Catal. A: Chem. 2003, 206, 225–238.
0 K. Izutsu, Acid-Base Dissociation Constants in Dipolar
Aprotic Solvents; Blackwell Science: Oxford; 1990.
1 M. A. Gauthier, M. I. Gibson, H. A. Klok, Angew. Chem. Int.
Ed. Engl. 2009, 48, 48–58.
2 D. H. Solomon, G. Moad, The Chemistry of Radical Polymer-
ization, 2nd ed.; Elsevier: Amsterdam, 2006.
3 N. V. Tsarevsky, W. Jakubowski, J. Polym. Sci. Part A:
Polym. Chem. 2011, 49, 918–925.
4 D. Navarro-Rodriguez, F. J. Rodriguez-Gonzalez, J. Romero-
5
1 N. Iranpoor, B. Tamami, K. Niknam, Can. J. Chem. 1997, 75,
913–1919.
8
1
5
2 N. Iranpoor, I. Mohammadpour Baltork, Tetrahedron Lett.
990, 31, 735–738.
8
1
5
3 C. Uyanik, J. R. Hanson, P. B. Hitchcock, M. A. Lazar, Tetra-
8
hedron 2005, 61, 4323–4327.
5
4 S. Singhal, S. L. Jain, B. Sain, Synth. Commun. 2011, 41,
829–1837.
8
1
5
5
5
5 H. R. Memarian, A. Saffar-Teluri, Beilstein J. Org. Chem. 2007, 3, 2.
6 D. C. McLeod, N. V. Tsarevsky, Polym. Int. 2014, 63, 868–875.
7 A. Postma, T. P. Davis, R. A. Evans, G. Li, G. Moad, M. S.
8
Garcia, E. J. Jimenez-Regalado, D. Guillon, Eur. Polym. J. 1998,
34, 1039–1045.
O’Shea, Macromolecules 2006, 39, 5293–5306.
8 Y. K. Chong, G. Moad, E. Rizzardo, S. H. Thang, Macromole-
cules 2007, 40, 4446–4455.
9 Y. K. Chong, J. Krstina, T. P. T. Le, G. Moad, A. Postma, E.
Rizzardo, S. H. Thang, Macromolecules 2003, 36, 2256–2272.
0 K. Skrabania, A. Laschewsky, H. Berlepsch, C. B o€ ttcher,
Langmuir 2009, 25, 7594–7601.
1 B. S. Sumerlin, N. V. Tsarevsky, G. Louche, R. Y. Lee, K.
8
5 A. P. Bapat, J. G. Ray, D. A. Savin, E. A. Hoff, D. L. Patton,
B. S. Sumerlin, Polym. Chem. 2012, 3, 3112–3120.
6 M. G. Gonz aꢁ lez, J. C. Cabanelas, J. Baselga, In Infrared
5
8
5
Spectroscopy - Materials Science, Engineering and Technol-
ogy; T. Theophanides, Ed.; Intech, 2012; pp 261–284.
6
87 L. J. Bellamy, The Infra-Red Spectra of Complex Molecules;
Methuen & Co: London, 1964.
6
88 J. E. Gurst, In The Chemistry of the Azido Group; Patai, S., Ed.;
Interscience: London, 1971; pp 191–202.
Matyjaszewski, Macromolecules 2005, 38, 7540–7545.
1
144
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1132–1144