Norbornene-containing dithiocarbamates for use in reversible addition-fragmentation chain transfer (RAFT) polymerization and ring-opening metathesis polymerization (ROMP)

Article abstract of DOI:10.1016/j.polymer.2015.10.028

Two new dithiocarbamate chain transfer agents (CTAs) with norbornene-containing Z-groups were prepared for use in reversible addition-fragmentation chain transfer (RAFT) polymerization. CTA 1b, which contains an electron deficient norbornene imide Z-group, was found to effectively mediate RAFT polymerization of 2° more activated monomers (MAMs) but did not facilitate RAFT of 3° MAMs or less activated monomers (LAMs). In contrast, CTA 2, derived from a norbornene amine, was well suited for the polymerization of LAMs, but did not control RAFT of MAMs. Poly(vinyl acetate) (PVAc), prepared by RAFT polymerization mediated by CTA 2, possessed the expected CTA-derived α- and ω-end groups. Ring-opening metathesis polymerization (ROMP) of 2 was carried out, and full conversion to polymer was achieved within 20 min. Based on this result, ROMP grafting-through of a PVAc macromonomer derived from CTA 2 was carried out, resulting in the formation of a well-defined PVAc bottlebrush polymer with a narrow molecular weight distribution.

Full text of DOI:10.1016/j.polymer.2015.10.028

Polymer 79 (2015) 205e211
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Norbornene-containing dithiocarbamates for use in reversible
additionefragmentation chain transfer (RAFT) polymerization and
ring-opening metathesis polymerization (ROMP)
1
1
*
Jeffrey C. Foster , Scott C. Radzinski , Sally E. Lewis, Matthew B. Slutzker, John B. Matson
Department of Chemistry and Macromolecules and Interfaces Institute, Virginia Tech, Blacksburg, VA 24061, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 September 2015
Received in revised form
Two new dithiocarbamate chain transfer agents (CTAs) with norbornene-containing Z-groups were
prepared for use in reversible additionefragmentation chain transfer (RAFT) polymerization. CTA 1b,
which contains an electron deficient norbornene imide Z-group, was found to effectively mediate RAFT
polymerization of 2 more activated monomers (MAMs) but did not facilitate RAFT of 3 MAMs or less
activated monomers (LAMs). In contrast, CTA 2, derived from a norbornene amine, was well suited for
the polymerization of LAMs, but did not control RAFT of MAMs. Poly(vinyl acetate) (PVAc), prepared by
9
October 2015
ꢀ
ꢀ
Accepted 11 October 2015
Available online 19 October 2015
RAFT polymerization mediated by CTA 2, possessed the expected CTA-derived
a- and u-end groups.
Keywords:
RAFT polymerization
graft polymer
Ring-opening metathesis polymerization (ROMP) of 2 was carried out, and full conversion to polymer
was achieved within 20 min. Based on this result, ROMP grafting-through of a PVAc macromonomer
derived from CTA 2 was carried out, resulting in the formation of a well-defined PVAc bottlebrush
polymer with a narrow molecular weight distribution.
poly(vinyl acetate)
©
2015 Elsevier Ltd. All rights reserved.
1. Introduction
used to prepare macromonomers for use in bottlebrush synthesis
[23e25], a topic of interest to our group and others.
Reversible deactivation radical polymerization (RDRP) tech-
niques are versatile tools for the preparation of a variety of poly-
mers with controlled molecular weights (MWs), narrow and
symmetrical molecular weight distributions, and well-defined
macromolecular architectures [1e3]. Examples of RDRPs include
atom transfer radical polymerization (ATRP) [4,5], nitroxide-
mediated radical polymerization (NMRP) [6], single electron
transfer polymerization (SET-LRP) [7], and reversible additione-
fragmentation chain transfer polymerization (RAFT) [8]. In recent
years, RAFT polymerization has been utilized extensively due its
applicability to a wide range of monomers and reaction conditions.
RAFT polymerization has been employed to functionalize bio-
molecules with stabilizing polymers [9e12], mediate the formation
of complex morphologies in suspension polymerizations [13e16],
prepare multiblock polymers containing up to 20 blocks [17], and
prepare advanced drug-releasing scaffolds [18e21], among many
other diverse applications [22]. RAFT polymerization has also been
RAFT polymerization relies on reversible deactivation of the
propagating or “active” radical species by a thiocarbonyl-thio-
containing chain transfer agent (CTA) [26]. Tremendous effort
over the past two decades has been dedicated toward the design
and synthesis of RAFT CTAs to improve their selectivity towards
certain monomer types, to diversify the routes of their synthesis, to
achieve specific built-in functionality that can be utilized in
downstream processes, and to realize new classes of CTAs capable
of mediating the polymerization of monomers across the radical
stability spectrum (e.g., universal or switchable CTAs) [27]. The
structure of the dithio compound [ZeC(S)eSeR] determines the
types of monomers that can be polymerized by a given CTA. Spe-
cifically, the Z-group modulates the reactivity of the thiocarbonyl
bond toward radicals, with more stable monomer radicals requiring
more electron withdrawing Z-groups, in general [28]. The relative
radical stability of the leaving R-group determines the partition
coefficient for the monomer/CTA pair [29]. The partition coefficient
is a term that describes the propensity of the CTA R-group to
fragment with respect to the polymer radical. If the difference in
relative radical stability between the R-group-derived radical and
the polymer-derived radical is large, the polymerization can be
*
Corresponding author.
E-mail address: jbmatson@vt.edu (J.B. Matson).
These authors contributed equally to this work.
1
http://dx.doi.org/10.1016/j.polymer.2015.10.028
032-3861/© 2015 Elsevier Ltd. All rights reserved.
0

206
J.C. Foster et al. / Polymer 79 (2015) 205e211
either uncontrolled (if the polymer-derived radical is more stable)
or inhibited (if the fragmented R-group radical is more stable). The
rational selection of R and Z-groups allows chemists to design CTAs
that mediate polymerization of a wide variety of monomer types.
Structural variation among RAFT CTAs allows for modulation of
reactivity towards certain monomers as well as for tuning of non-
RAFT specific parameters such as solubility, stability/lability (use-
ful for aminolysis, as an example), and built-in functionality.
Dithiobenzoate, trithiocarbonate, xanthate, and dithiocarbamate
CTAs have been prepared from aryl Gringard, thiol, alcohol, and
amine starting materials, respectively [26,30,31]. In the case of
dithiocarbamate CTAs, the electron withdrawing ability of the ni-
trogen atom plays a critical role in the reactivity of the resulting
CTA. More electron-deficient Z-groups such as those derived from
amides, pyrrole, or imidazole effectively mediate the polymeriza-
tion of the broad class of more-activated monomers (MAMs), while
electron donating Z-groups allow for polymerization of less-
activated monomers (LAMs) [32]. We recently reported on the
synthesis of a norbornene-imide-derived dithiocarbamate CTA
with a benzyl R-group (CTA 1a, Fig. 1) [33]. This novel dithio
compound was shown to effectively mediate the polymerization of
styrene and n-butyl acrylate. In addition, the norbornene-
containing Z-group was readily polymerized in the presence of a
Ru-based metathesis catalyst, undergoing ring-opening metathesis
polymerization (ROMP) to form a poly(CTA) with pendant chain
transfer functionality on each repeat unit. This poly(CTA) was uti-
lized subsequently in RAFT transfer-to polymerization to produce
monodisperse bottlebrush polymers with molecular weights
novel CTAs to mediate RAFT polymerization of 5 different mono-
mersdMMA, styrene, acryloylmorpholine (ACMO), methyl acrylate
ꢀ
(MA), and VAcdencompassing common monomer types from 3
MAMs, 2 MAMs, and LAMs.
ꢀ
2. Materials and methods
2.1. Materials
All reagents were obtained from commercial vendors and used
as received unless otherwise stated. MMA, styrene, ACMO, MA, and
VAc were passed through small columns of basic alumina prior to
0
0
use. 2,2 -Azobis(2-methylpropionitrile) (AIBN) and 1,1 -azobis(cy-
clohexanecarbonitrile) (ACHN) were recrystallized from methanol
prior to use. ROMP catalyst (H
obtained as a generous gift from Materia. ROMP catalyst (H
(pyr) (Cl)
2
IMes) (Cl)
2
(PCy
3
)Ru¼CHPh (G2) was
2
IMes)
2
2
Ru¼CHPh (G3) was prepared from G2 according to
literature procedures [34,35]. exo-Norbornene imide (NI) was
prepared according to a previous report [33].
2.2. Methods
NMR spectra were measured on Agilent 400 MHz or Bruker
1
13
500 MHz spectrometers. H and C NMR chemical shifts are re-
ported in ppm relative to internal solvent resonances. Yields refer
to chromatographically and spectroscopically pure compounds
unless otherwise stated. Size exclusion chromatography (SEC) was
carried out in THF on two Agilent PLgel 10 mm MIXED-B columns
(
MWs) in excess of 1 MDa. CTA 1a was also utilized to prepare
connected in series with a Wyatt Dawn Helios 2 light scattering
detector and a Wyatt Optilab Rex refractive index detector. No
calibration standards were used, and dn/dc values were obtained
by assuming 100% mass elution from the columns. High-resolution
mass spectra were taken on an Agilent Technologies 6230 TOF LC/
MS mass spectrometer. UV-Vis absorbance spectra were recorded
on a Cary 5000 UV-Vis (Agilent) from 550 to 300 nm.
macromonomers for use in bottlebrush synthesis by the grafting-
through approach.
The norbornene-derived CTA previously reported by our group
ꢀ
was incapable of mediating RAFT polymerization of the 3 MAM
methyl methacrylate (MMA) or the LAM vinyl acetate (VAc). RAFT
polymerization of 3 MAMs such as MMA requires a CTA with a
deactivated, 3 R-group such as eC(CH
ꢀ
ꢀ
3
)
2
CN. Therefore, the poor
control over the polymerization of MMA was attributed to the low
partition coefficient of the benzyl R-group with respect to MMA.
Polymerization of VAc in the presence of 1a was completely
inhibited, with no polymer product detected even after 48 h. Here,
it was speculated that the electron withdrawing norbornene imide
2.3. Synthesis of CTA 1b
KOH (0.360 g, 6.43 mmol) was ground to a fine powder with a
mortar and pestle and placed in a 100 mL round bottom flask. To
the flask was added NI (1.00 g, 6.13 mmol) followed by 15 mL of
DMF. This mixture was stirred for 5 min, followed by dropwise
(NI) Z-group renders the C]S bond of the CTA derivative unreac-
tive towards LAM radicals. To expand the scope of monomers that
could be polymerized by norbornene-based CTAs, we sought to
prepare new dithio compounds with different R- and Z-groups.
Based on our previous work, we hypothesized that replacing the
benzyl R-group of 1a with an R-group more suitable for the poly-
merization of MMA would yield a CTA suitable for mediating RAFT
polymerization of methacrylate and methacrylamide monomers
while still utilizing the ROMP-active norbornene-containing Z-
group. We also envisioned a norbornene-derived CTA with a less
active Z-group and a less stable homolytic leaving group (i.e.,
2
addition of CS (1.11 mL, 18.4 mmol). The solution developed a deep
red color. After an additional 3 h of stirring, (1-bromoethyl)benzene
(1.67 mL, 12.3 mmol) was added dropwise, and the reaction
mixture was stirred at rt for 12 h. The reaction mixture was then
diluted with CH
and brine. The organic layer was dried over Na
was removed by rotary evaporation. The crude product was puri-
fied on a silica gel column, eluting with 1:1 CH Cl /hexanes, to give
2
Cl
2
(50 mL) and washed with H
2
O (3 ꢁ 150 mL)
2
SO
4
, and the solvent
2
2
0.59 g of 1b as a yellow solid (28% yield, 8:2 mixture of S- and N-
alkylated products that could not be separated via column chro-
1
R ¼ eCH
2
CN) might facilitate the RAFT polymerization of LAMs
matography). H NMR (CDCl
3
):
d
7.5-7.2 (m, 5H), 6.32 (t, 2H), 4.96
13
such as VAc. In this contribution, we describe the synthesis and
characterization of new dithiocarbamate CTAs with norbornene-
containing Z-groups. In addition, we evaluate the ability of these
(q, 1H), 3.38 (s, 2H), 2.79 (s, 2H), 1.81 (d, 3H), 1.57 (m, 2H). C NMR
(CDCl 198.63, 173.76, 139.61, 138.32, 128.93, 128.24, 127.90,
3
) d
52.30, 48.20, 46.49, 43.24, 20.46. HRMS (m/z): calculated 344.0773,
found 344.0757.
2.4. Synthesis of reduced imide CTA 2
4
A 3-necked round bottom flask was charged with LiAlH (2.09 g,
5
5.2 mmol). The flask was placed in an ice bath. To the flask was
atmosphere. NI (3.00 g,
8.4 mmol) was dissolved in 70 mL of dry THF in a second flask. The
imide solution was transferred to an addition funnel and was added
slowly added 30 mL of dry THF under N
1
2
Fig. 1. CTA evaluated in our previous work [33].

J.C. Foster et al. / Polymer 79 (2015) 205e211
207
dropwise to the LiAlH
4
slurry. After the addition was complete, the
in a second vial by dissolving an amount of G3 in dry CH
2
Cl
2
. 0.5 mL
reaction mixture was allowed to warm to rt and then was heated at
reflux for 12 h. To quench the reaction, the reaction mixture was
of this G3 solution was added rapidly to the first vial. After 20 min, a
few drops of ethyl vinyl ether were added to terminate the poly-
merization, and the reaction mixture was stirred for an additional
5 min. The polymer was isolated via precipitation from hexanes.
placed in an ice bath, and 2 mL of H
by 2 mL NaOH solution (10% w/v in water) and then 6 mL H
2
O was added slowly, followed
O. Upon
2
warming to rt the solids became white. The quenched reaction
mixture was diluted with diethyl ether (~200 mL) and filtered
through celite. This solution was concentrated and washed with
brine, dried over Na SO , and rotovapped. The resulting yellow oil
2 4
norbornene amine (NA) was used in the next step without further
purification.
2
.6.1. Synthesis of PVAc bottlebrush
Macromonomer NA-PVAc (40 mg) was dissolved in 0.3 mL of
dry CH
of G3 was prepared in a second vial by dissolving an amount of G3
in dry CH Cl . 0.1 mL of this G3 solution was added rapidly to the
first vial. After 60 min, a few drops of ethyl vinyl ether were added
to terminate the polymerization, and the reaction mixture was
stirred for an additional 5 min. The polymer was isolated via pre-
cipitation from diethyl ether.
2 2
Cl in a vial equipped with a stir bar. A 2.3 mg/mL solution
2
2
NA was dissolved in 15 mL of EtOH in a round bottom flask. The
flask was placed in an ice bath. To the flask was added CS
2
(1.23 mL,
2
0.3 mmol) followed by dropwise addition of 2.3 mL of 50% wt/v
aqueous KOH (1.1 equiv). The reaction mixture was allowed to
warm to room temperature and was stirred for 12 h. The reaction
2 4
mixture was then diluted with MeOH (~40 mL), dried over Na SO ,
and rotovapped. The residue was used in the next step without
further purification.
The crude dithiocarbamate salt was dissolved in 40 mL of
acetone in a round bottom flask. The flask was placed in an ice bath.
To the flask was added bromoacetonitrile (1.40 mL, 20.1 mmol)
dropwise. The reaction mixture was allowed to warm to room
temperature and was stirred for 12 h. The reaction mixture was
3
. Results and discussion
3.1. R-group modification
To prepare NI-derived CTAs with various R-groups, we initially
adopted a strategy that involved the in situ generation of a
norbornene-imide dithiocarbamate intermediate that could be
transformed into the desired CTA via addition of the appropriate
alkyl-halide, as shown in Scheme 1. Similar to our previous report
in which we obtained 1a in moderate yields (30e40%) using a facile
then rotovapped and the residue was dissolved in CH
CH Cl solution was washed with H O and brine, dried over Na
and rotovapped. The crude product was purified on a silica gel
column, eluting with 2:3 CH Cl /hexanes to give the pure product
2) as an off white powder (2.80 g, 61% yield). H NMR (CDCl
2
Cl
2
. The
2
2
2
2
4
SO ,
1-pot procedure [33], we attempted to synthesize additional
2
2
1
norbornene-containing dithiocarbamates (1a-1f) by first dissolving
the nucleophilic exo-norbornene imide (NI) in DMF, then adding
finely ground potassium hydroxide, and then dropping in carbon
(
3
):
6.13 (q, 2H), 4.19 (s, 2H), 4.06 (m, 1H), 3.76 (m, 2H), 3.40 (m, 1H),
.74 (d, 2H), 2.44 (m, 2H), 1.46 (d, 1H), 1.32 (d, 1H). C NMR (CDCl ):
3
d
2
d
4
2
1
3
2
disulfide (CS ). Finally, the appropriate alkyl halide was added after
187.20, 137.79, 137.20, 116.11, 60.41, 54.98, 47.69, 47.62, 45.36,
2.86, 42.20, 22.29. HRMS (m/z): calculated 251.0671, found
51.0671.
a period of 1e2 h to form the desired product, which we would
attempt to isolate using standard aqueous workup and/or column
chromatography. Benzyl bromide (Table 1, entry a), (1-bromoethyl)
benzene (b), N-(chloromethyl)phthalimide (c), 2-bromo-2-
2.5. Representative RAFT polymerization procedure
methylpropionic acid (d), ethyl
a-bromoisobutyrate (e), or 2-
bromopropionic acid (f), was utilized to attempt to alkylate the
intermediate in the final step. Of these, only a and b were found to
yield the desired product in an appreciable quantity (see Table 1).
Starting materials or N-alkylated byproducts were recovered from
reactions with the other alkylating agents.
A typical polymerization procedure is as follows: To an oven-
dried Schlenk tube equipped with a magnetic stir bar was added
CTA, azo initiator, monomer, and solvent (if applicable). The tube
was deoxygenated by subjecting the contents to three freeze-
ꢂpumpꢂthaw cycles. The Schlenk tube was then backfilled with N
and submerged in an oil bath maintained at the appropriate tem-
perature. Samples were removed periodically by N -purged syringe
2
To explain these results, a product analysis of the synthesis of 1a
1
was performed. H NMR spectroscopy of a mixture of NI, KOH, and
2
2 6
CS that had been allowed to react in DMSO-d revealed an
to monitor molecular weight evolution by SEC and conversion by
H NMR spectroscopy.
important insight: the equilibrium between the deprotonated
imide and the dithiocarbamate intermediate depicted in Scheme 1
is strongly biased in favor of the reactants (Keq ¼ 0.024 by relative
integration). We attribute this observation to both the poor
nucleophilicity of the NI nitrogen and the instability of the
dithiocarbamate salt, which may decompose via a zwitterionic in-
termediate formed through protonation of the N atom in a similar
1
2
.6. Synthesis of poly(2)
2 2
CTA 2 (50 mg) was dissolved in 1.5 mL of dry CH Cl in a vial
equipped with a stir bar. A 5.8 mg/mL solution of G3 was prepared
Scheme 1. Preparation of dithiocarbamate CTAs from NI.

208
J.C. Foster et al. / Polymer 79 (2015) 205e211
Table 1
Table 2
Isolated yields from alkylation of NI-derived dithiocarbamate intermediate using
Polymers prepared by RAFT polymerization using CTA 1b.
various alkylating agents.
Monomer
% Conversiona
M
n
(Da)b
M
n, expected (Da)c
Ðb
Entry
Alkylating agent
Product
Isolated yield
40%
MMA
ACMO
MA
Styrene
VAc
14
99
70
40
0
62,000
15,000
6100
1800
17,500
7500
5200
1.40
1.09
1.03
1.01
a
Br
1a
4400
b
c
1b
1c
28%
d
d
d
e
e
e
Br
a
Determined using 1H NMR spectroscopy.
b
c
ꢀ
O
<1%
Measured by SEC-MALLS in THF at 30 C.
Calculated based on monomer conversion measured by 1H NMR spectroscopy.
CI
N
d
No polymer was detected by SEC. Data represent the ultimate timepoint taken
O
from kinetic analyses.
a
d
e
f
O
O
O
1d
1e
1f
e
Br
Br
Br
OH
O
ꢀ
0
ACMO and at 75 C for styrene in the presence of 2,2 -azobis(2-
methylpropionitrile) (AIBN) to provide initiating radicals by ther-
mal decomposition. For each monomer type a [M]/[1b]/[AIBN] ratio
of 100: 1: 0.1 was employed. Table 2 summarizes a series of poly-
mers prepared by RAFT polymerization using CTA 1b. In general,
the polymers possessed low dispersities and MWs that were close
to expected values. Kinetic analysis of these polymerizations
revealed propagation rate profiles that were first order in monomer
consumption (Fig. 2A). As shown in Fig. 2B, plots of number average
a
e
a
e
OH
a
Compounds not isolated from reaction mixtures.
manner to related compounds [36]. After 2 h benzyl bromide was
n
molecular weight (M ) (determined by size exclusion chromatog-
added, and the composition of the reaction mixture was re-
evaluated after 12 h by 1H NMR spectroscopy. By comparing in-
raphy (SEC)) vs. conversion were linear, indicating well controlled
polymerizations. SEC traces of aliquots taken at specific timepoints
during kinetic analysis were generally narrow and monomodal (as
in Fig. 3). In general, polymerizations using CTA 1b progressed more
rapidly than those previously reported for CTA 1a. We attribute this
phenomenon to the relatively more stable homolytic leaving group
of CTA 1b, which likely accelerates consumption of the RAFT agent
during the pre-equilibrium phase compared with CTA 1a [29]. RAFT
polymerization of MMA and VAc were also attempted. As was the
case with CTA 1a, polymerization of MMA mediated by CTA 1b was
uncontrolled, while RAFT of VAc using this CTA was completely
inhibited for >48 h.
tegrations of the various benzyl methylene resonances, and
knowing the identity of each of these species by matching against
pure spectra of molecules isolated from the crude reaction mixture,
it was determined that the N-alkylated norbornene imide was in
fact the major product, with the desired dithiocarbamate (CTA 1a)
and a trithiocarbamate byproduct forming to a lesser extent
Scheme 2). Increasing the equivalents of CS to attempt to drive
2
the equilibrium toward the product had a negligible effect on the
isolated yield.
(
Given this analysis, it is not surprising that alkylating agents
with carboxylic acid functionalities (d and f) could not be employed
to synthesize dithiocarbamate CTAs with the NI Z-group. Alkylation
using ethyl
fruitless, likely due to the fact that S
hindered and S 1 using this particular alkyl bromide is unlikely due
a
-bromoisobutyrate (e, the ethyl ester of d) also proved
3.2. Z-group modification
N
2 at the -carbon is sterically
a
N
To access the class of LAMs, we pursued an alternative strategy
in which NI was reduced to the corresponding secondary amine
(Scheme 3)da significantly more reactive nucleophile than the
imide. The increased electron density of the nitrogen atom of the
resulting dithiocarbamates is not desirable for mediation of the
polymerization of MAMs; however, this change in electronics was
expected to enable RAFT of LAMs such as VAc. The reduction of NI
to the destabilization of the carbocation adjacent to the electron-
withdrawing carbethoxy group. Neutralization of d with triethyl-
amine prior to alkylation was also attempted without success.
Based on these observations, we speculate that the structures of
CTAs with the norbornene imide Z-group may be limited to those
prepared using alkylating agents that are sterically unhindered and
do not contain acidic functionalities that may catalyze CTA
decomposition (e.g., a or b). These limitations are at odds with the
was accomplished using LiAlH [37], giving norbornene amine (NA)
in nearly quantitative yield. Treatment of this compound with CS2
4
ꢀ
structural requirements of a good R-group for 3 MAMs.
and aqueous KOH gave the dithiocarbamate potassium salt, which
could be isolated from excess reagents. Alkylation was conducted in
We previously reported on the successful RAFT polymerization
ꢀ
ꢀ
of styrene and n-butyl acrylate (2 MAMs) mediated by CTA 1a. We
acetone at 0 C using bromoacetonitrile. Using this strategy, a
therefore expected that CTA 1b would effectively mediate RAFT
second generation norbornene-containing CTA (CTA 2) was pre-
pared in 3 steps from NI in 61% overall yield.
ꢀ
polymerization of styrene, MA, and ACMO (2 MAMs). The poly-
ꢀ
merizations were carried out in THF at 50 v/v% at 60 C for MA and
The dialkyl dithiocarbamate CTA 2 possesses an electron rich
Scheme 2. Product analysis of synthesis of CTA 1a.

J.C. Foster et al. / Polymer 79 (2015) 205e211
209
dialkyl amine Z-group. Typically, the electron-donating character of
the N atom in dithiocarbamates renders the C]S bond less reactive
toward radicals than CTAs that contain more electron-withdrawing
Z-groups [28]. This has the effect of decreasing the chain transfer
efficiency of the CTA, thereby increasing the active radical con-
centration. As a result, dialkyl dithiocarbamate CTAs tend to
mediate the polymerization of LAMs effectively but lead to un-
controlled polymerization of MAMs. To assess the quality of the
polymerization of various MAMs by RAFT using CTA 2, reactions
were conducted for 2 h using conditions similar to those employed
when using CTA 1b. Our evaluation is summarized in Table 3. SEC
1
and H NMR analysis of the crude reaction mixtures showed that
RAFT polymerizations of MMA, MA, styrene, and ACMO were
poorly controlled, giving polymers of moderate to broad dispersity
with MWs significantly higher than expected. In addition, the
anticipated CTA-derived polymer end groups were absent from the
1
H NMR spectrum of a purified polystyrene sample polymerized in
the presence of CTA 2 (see Fig. S8), supporting the expected low
propensity of CTA 2 to react with MAM radicals. This result is
further corroborated by comparing the UV-Vis absorption spectrum
of CTA 2 to that of CTA 1b (Fig. S10). The decreased lmax of CTA 1b
relative to CTA 2 is consistent with increased electron density on
C]S [38].
RAFT polymerization of VAc mediated by CTA 2 was more suc-
cessful than the other monomers tested. The dialkyl amine Z-group
has been shown to mediate (to a degree) RAFT polymerization of
LAMs such as VAc [32]. The reduced reactivity of the C]S bond of
alkyl dithiocarbamates and xanthates in particular promotes suf-
ficient fragmentation of the CTA-polymer adduct, allowing for
polymerization of LAMs to progress without retardation. VAc po-
ꢀ
0
lymerizations using CTA 2 were conducted neat at 80 C using 1,1 -
azobis(cyclohexanecarbonitrile) (ACHN) as the radical source. As
expected, the formation of polymer was observed when CTA 2 was
used in the polymerization of VAc. However, the reaction exhibited
a long incubation period of >7 h, after which propagation pro-
ceeded rapidly. An isolated polymer prepared via RAFT of VAc
n
mediated by CTA 2 was monomodal by SEC with an M that agreed
with the expected value (Table 3). The relatively high Ð of this
polymer likely resulted from two independent phenomena: (1)
intermolecular termination by coupling (also supported by the
higher than expected MW); and (2) the presence of a population of
high MW species that typically form early in the polymerization of
Fig. 2. A) Kinetic analysis of RAFT polymerization of styrene (squares), MA (diamonds),
and ACMO (circles) using 1b. B) Plot of M vs. conversion for the same polymerizations.
n
The solid lines represent the theoretical M
n
values calculated from conversions as
1
determined by H NMR spectroscopy.
1
VAc [39]. H NMR spectroscopy of the polymer sample showed the
appropriate CTA-derived
a- and u-chain ends (Fig. 4), and end
group analysis gave a MW value that was in relatively good
agreement with those determined by SEC and monomer conver-
sion (5200 Da). Taken together, these data suggest that CTA 2
effectively mediates the polymerization of VAc.
CTAs with norbornene Z-groups can be utilized in the synthesis
of bottlebrush polymers by both the transfer-to and grafting-
through approaches. Both strategies rely on ring-opening metath-
esis polymerization (ROMP) to generate the bottlebrush backbone.
ROMP of CTA 2 was performed to prepare poly(2) as shown in Fig. 5
and S9. Polymerizations by the modified 2nd generation Grubbs'
catalyst (G3) at a [M]/[G3] ratio of 50: 1 achieved quantitative
conversion in <20 min. The SEC trace of the resulting polymer
1
(
Mn ¼ 12,700 Da) was monomodal with low Ð (1.01). H NMR
spectroscopy of the crude reaction mixture confirmed quantitative
consumption of the olefin functionality, evidenced by the disap-
pearance of the resonance at 6.13 ppm. A pair of new peaks be-
tween 5.2 and 5.6 ppm were assigned to the backbone olefins of
poly(2).
Based on this positive result, we considered whether ROMP
grafting-through of PVAc prepared using 2 could be accomplished
to form a bottlebrush polymer (as shown in Scheme 4). The
Fig. 3. SEC traces of aliquots removed during the kinetic evaluation of the polymeri-
zation of styrene mediated by CTA 1b.

210
J.C. Foster et al. / Polymer 79 (2015) 205e211
Scheme 3. Preparation of dithiocarbamate CTA, 2, from NA.
Table 3
polymerization mediated by CTA 2 (NA-PVAc) (M
¼ 6600 Da). The
n
Polymers prepared by RAFT polymerization using CTA 2.
resulting bottlebrush polymer had a MW of 110 kDa with a narrow
and symmetrical molecular weight distribution (Ð ¼ 1.26). Full
conversion of norbornene-functionalized MMs was confirmed by
Monomer
% Conversiona
M
n
(Da)b
M
n, expected (Da)c
Ðb
MMA
ACMO
MA
Styrene
VAc
25
>99
75
16
48
168,000
61,800
17,300
37,600
6600
2500
14,000
6500
1700
4100
1.63
2.05
1.58
1.31
1.43
1
H NMR spectroscopy. A second population of macromolecules
centered on 17 min in the SEC trace in Fig. 6 corresponds to residual
PVAc MM that was not consumed during ROMP. These “dead
chains” arise primarily from termination reactions during RAFT
a
Determined using 1H NMR spectroscopy.
polymerization that leave the polymer chain without the desired u-
b
c
ꢀ
Measured by SEC-MALLS in THF at 30 C.
end group. The minimal fraction of residual polymer after ROMP
grafting-through further supports the capability of CTA 2 to control
RAFT polymerization of VAc, as the majority of the macro-
Calculated based on monomer conversion measured by 1H NMR spectroscopy.
monomers possess the norbornene
. Conclusions
In summary, we have prepared two new dithiocarbamate CTAs
u-end group.
grafting-through technique involves the polymerization of a
macromolecule functionalized with a polymerizable group on one
chain end (macromonomer, MM). Bottlebrush polymers prepared
using this method are considered to have perfect grafting density,
as each repeat unit bears a pendant polymer side chain. Bottlebrush
polymers possessing PVAc side chains have been synthesized pre-
viously via radical grafting-through using a vinyl-functionalized
PVAc MM [40], through a tandem ATRP/RAFT method by grafting-
from [41], and employing an all-RAFT method using the R-group
4
with Z-groups derived from NI and the corresponding reduced
secondary amine. Based on the electronics of NI, it is apparent that
(
grafting-from) and Z-group (transfer-to) approaches [42]. These
macromolecules are precursors to poly(vinyl alcohol) (PVOH)-
based materials that have interesting industrial and biomedical
applications [43]. In addition, branched PVOHs possess different
physical properties than linear ones, such as decreased solution
viscosity [44].
As shown in Fig. 4, the norbornene functionality of CTA 2 is
retained on the
u-chain end after the polymerization of VAc. This
ROMP-active moiety can be utilized to form a bottlebrush polymer
via grafting-through. ROMP grafting-through was carried out at a
[MM]/[G3] ratio of 25:1 using a PVAc MM prepared via RAFT
Fig. 5. 1H NMR spectra of CTA 2 (top spectrum) and a crude spectrum of a polymer
prepared via ROMP of CTA 2 (bottom spectrum) highlighting the quantitative con-
sumption of olefin protons. Spectra were taken in DMSO-d .
6
Fig. 4. 1H NMR spectrum of a PVAc MM prepared using CTA 2 highlighting the CTA-
derived - and -chain ends (protons a and c). Spectrum was taken in CDCl
a
u
3
.
Scheme 4. Preparation of PVAc bottlebrush by ROMP grafting-through.

J.C. Foster et al. / Polymer 79 (2015) 205e211
211
References
[
[
[
[
[
[
[
[
1] K. Matyjaszewski, Macromolecules 45 (2012) 4015.
2] G. Moad, E. Rizzardo, S.H. Thang, Aust. J. Chem. 65 (2012) 985.
3] S. Perrier, P. Takolpuckdee, J. Polym. Sci. Part A Polym. Chem. 43 (2005) 5347.
4] J.-S. Wang, K. Matyjaszewski, Macromolecules 28 (1995) 7901.
5] J.-S. Wang, K. Matyjaszewski, J. Am. Chem. Soc. 117 (1995) 5614.
6] V. Sciannamea, R. Jerome, C. Detrembleur, Chem. Rev. 108 (2008) 1104.
7] N. Zhang, S.R. Samanta, B.M. Rosen, V. Percec, Chem. Rev. 114 (2014) 5848.
8] J. Chiefari, Y.K. Chong, F. Ercole, J. Krstina, J. Jeffery, T.P.T. Le, R.T.A. Mayadunne,
G.F. Meijs, C.L. Moad, G. Moad, E. Rizzardo, S.H. Thang, Macromolecules 31
(
1998) 5559.
9] V. V ꢀa zquez-Dorbatt, Z.P. Tolstyka, H.D. Maynard, Macromolecules 42 (2009)
650.
[
7
[
[
[
[
[
[
10] V. Bulmus, Polym. Chem. 2 (2011) 1463.
11] P. De, M. Li, S.R. Gondi, B.S. Sumerlin, J. Am. Chem. Soc. 130 (2008) 11288.
12] B.S. Sumerlin, ACS Macro Lett. 1 (2012) 141.
13] L.P.D. Ratcliffe, A.J. Ryan, S.P. Armes, Macromolecules 46 (2013) 769.
14] N.J. Warren, S.P. Armes, J. Am. Chem. Soc. 136 (2014) 10174.
15] C.A. Figg, A. Simula, K.A. Gebre, B.S. Tucker, D.M. Haddleton, B.S. Sumerlin,
Chem. Sci. 6 (2015) 1230.
Fig. 6. SEC traces of PVAc MM prepared using CTA 2 (NA-PVAc) (longer retention time)
and bottlebrush polymer prepared using this MM via ROMP grafting-through
shorter retention time).
[16] Y. Kang, A. Pitto-Barry, A. Maitland, R.K. O'Reilly, Polym. Chem. 6 (2015) 4984.
[17] G. Gody, T. Maschmeyer, P.B. Zetterlund, S. Perrier, Nat. Commun. 4 (2013)
2505.
a
(
[
[
18] J.C. Foster, J.B. Matson, Macromolecules 47 (2014) 5089.
19] L. Zhang, W. Liu, L. Lin, D. Chen, M.H. Stenzel, Biomacromolecules 9 (2008)
3
321.
only certain alkylating agentsdsterically unhindered alkyl halides
lacking acidic functionalitiesdcan be employed to prepare CTAs
derived from this Z-group. Nevertheless, CTA 1b was prepared in
moderate yields and was found to efficiently mediate RAFT poly-
merization of 2 MAMsdstyrene, MA, and ACMO. Reduction of NI
was conducted to reduce the electron withdrawing ability of the Z-
[
20] A.W. York, S.E. Kirkland, C.L. McCormick, Adv. Drug. Deliv. Rev. 60 (2008)
1018.
[
[
[
21] S. Cerritelli, D. Velluto, J.A. Hubbell, Biomacromolecules 8 (2007) 1966.
22] G. Moad, E. Rizzardo, S.H. Thang, Chem. An Asian J. 8 (2013) 1634.
23] Y.C. Teo, Y. Xia, Macromolecules 48 (2015) 5656.
[24] Z. Li, K. Zhang, J. Ma, C. Cheng, K.L. Wooley, J. Poly. Sci. Part A, Polym. Chem. 47
(2009) 5557.
25] X. Li, H. ShamsiJazeyi, S.L. Pesek, A. Agrawal, B. Hammouda, R. Verduzco, Soft
Matter 10 (2014) 2008.
26] G. Moad, E. Rizzardo, S.H. Thang, Polymer 49 (2008) 1079.
[27] M.R. Hill, R.N. Carmean, B.S. Sumerlin, Macromolecules 48 (2015) 5459.
28] J. Chiefari, R.T.A. Mayadunne, C.L. Moad, G. Moad, E. Rizzardo, A. Postma,
S.H. Thang, Macromolecules 36 (2003) 2273.
29] Y.K. Chong, J. Krstina, T.P.T. Le, G. Moad, A. Postma, E. Rizzardo, S.H. Thang,
Macromolecules 36 (2003) 2256.
30] D.J. Keddie, G. Moad, E. Rizzardo, S.H. Thang, Macromolecules 45 (2012) 5321.
ꢀ
[
group. Novel dithiocarbamate CTA 2 prepared from norbornene
ꢀ
ꢀ
amine did not control the polymerization of 2 or 3 MAMs;
however, it provided moderate control over the polymerization of
VAc, yielding polymers with the appropriate CTA-derived end
groups. ROMP of a PVAc MM prepared using CTA 2 was carried out
to give a bottlebrush polymer with PVAc side chains using the
grafting-through technique. To the best of our knowledge, this
work represents the first preparation of a bottlebrush polymer with
PVAc side chains by the grafting-through method. We expect that
this contribution will assist in the rational design of CTAs capable of
mediating RAFT polymerization to form polymers functionalized
with ROMP-active norbornene moieties.
[
[
[
[
[
[
31] J. Skey, R.K. O'Reilly, Chem. Commun. (2008) 4183.
32] R.T.A. Mayadunne, E. Rizzardo, J. Chiefari, Y.K. Chong, G. Moad, S.H. Thang,
Macromolecules 32 (1999) 6977.
33] S.C. Radzinski, J.C. Foster, J.B. Matson, Polym. Chem. 6 (2015) 5643.
34] J.A. Love, J.P. Morgan, T.M. Trnka, R.H. Grubbs, Angew. Chem. Int. Ed. 41 (2002)
[
[
4
035.
[35] J. Liu, A.X. Gao, J.A. Johnson, J. Vis. Exp. (2013) e50874.
36] E. Humeres, N.A. Debacher, M. M. d. S. Sierra, J.D. Franco, A. Schutz, J. Org.
Chem. 63 (1998) 1598.
37] L.D. Van Vliet, T. Ellis, P.J. Foley, L. Liu, F.M. Pfeffer, R.A. Russell, R.N. Warrener,
F. Hollfelder, M.J. Waring, J. Med. Chem. 50 (2007) 2326.
[
[
Acknowledgments
This work was supported by the Army Research Office (W911NF-
4-1-0322) and the American Chemical Society Petroleum
[38] T.G. McKenzie, Q. Fu, E.H.H. Wong, D.E. Dunstan, G.G. Qiao, Macromolecules
8 (2015) 3864.
4
1
[39] M.H. Stenzel, L. Cummins, G.E. Roberts, T.P. Davis, P. Vana, C. Barner-Kowollik,
Research Fund (54884-DNI7). We thank Materia for catalyst as well
as Hannah Tecson and Kyle Moran for experimental assistance. We
also thank Prof. Amanda J. Morris (Virginia Tech) for use of
instruments.
Macromol. Chem. Phys. 204 (2003) 1160.
[40] T. Ohnaga, T. Sato, Polymer 37 (1996) 3729.
41] A. Nese, Y. Kwak, R. Nicolaÿ, M. Barrett, S.S. Sheiko, K. Matyjaszewski, Mac-
romolecules 43 (2010) 4016.
42] J. Bernard, A. Favier, T.P. Davis, C. Barner-Kowollik, M.H. Stenzel, Polymer 47
[
[
[
[
(2006) 1073.
43] M.I. Baker, S.P. Walsh, Z. Schwartz, B.D. Boyan, J. Biomed. Mater. Res. Part B
Appl. Biomater. 100B (2012) 1451.
44] R. Baudry, D.C. Sherrington, Macromolecules 39 (2006) 5230.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.polymer.2015.10.028.