Switchable reversible addition-fragmentation chain transfer (raft) polymerization in aqueous solution, n, n -dimethylacrylamide

Article abstract of DOI:10.1021/ma200760q

Reversible addition-fragmentation chain transfer (RAFT) polymerization of a "more activated monomer" (MAM), N,N-dimethylacrylamide (DMAm), has been successfully achieved in aqueous solution using an acid/base switchable pyridyl-substituted dithiocarbamate

Full text of DOI:10.1021/ma200760q

ARTICLE
pubs.acs.org/Macromolecules
Switchable Reversible AdditionꢀFragmentation Chain Transfer
(RAFT) Polymerization in Aqueous Solution, N,N-Dimethylacrylamide
Daniel J. Keddie,* Carlos Guerrero-Sanchez, Graeme Moad,* Ezio Rizzardo, and San H. Thang
CSIRO Materials Science and Engineering, Bag 10, Clayton South, VIC 3169, Australia
S Supporting Information
b
ABSTRACT: Reversible additionꢀfragmentation chain transfer (RAFT)
polymerization of a “more activated monomer” (MAM), N,N-dimethylacryl-
amide (DMAm), has been successfully achieved in aqueous solution using an
acid/base switchable pyridyl-substituted dithiocarbamate RAFT agent. The
effectof strength (pKa) and stoichiometry of the acid used to switchthe RAFT
agent on the molecular weight and dispersity of poly(N,N-dimethylacryl-
amide) (PDMAm) were examined, making use of high-throughput protocol.
Best control was achieved with a stoichiometric amount of the strongest acid investigated, p-toluenesulfonic acid monohydrate. Use of
weaker acids (higher pKa) or less than stoichiometric amounts of acid with respect to RAFT agent resulted in broader molecular
weight distributions. These resultscanbe rationalizedin terms of the extent of protonation of the pyridine nitrogen of the RAFT agent
(pKa ∼ 3.13). The preparation of unimodal low dispersity block copolymers of PDMAm with the “less-activated” monomers (LAMs)
N-vinylcarbazole (NVC), vinyl acetate (VAc), and N-vinylpyrrolidone (NVP) demonstrated the fidelity of the RAFT end group in
these experiments and provides a further example of the utility of switchable RAFT polymerization.
’ INTRODUCTION
dithiocarbamates (Z = NR2), both slows addition of radicals to the
RAFT agent and promotes the subsequent fragmentation such that it
is not the rate determining step in chain transfer. Hence, these RAFT
agents are regularly used for control over LAMs polymerization.
Generally, xanthates and dithiocarbamates are relatively unreactive
toward MAM derived propagating radicals,10 making them ineffective
control agents for these monomers. However, they may be effective
for MAMs when the substituent is part of an aromatic heterocycle10
or when highly electron withdrawing groups are present on the het-
eroatom.11 As the reactivity of commonly used RAFT agents are
tailored to either MAMs or LAMs, preparation of low dispersity
polyMAM-block-polyLAM is not possible using the conventional
RAFT process. While some RAFT agents, such as the N,N-diaryl-
dithiocarbamates, have been reported by Destarac et al.12 and Malepu
et al.13 for the homopolymerization of both MAMs and LAMs, these
RAFT agents give only moderate control with both monomer
classes.12,13 When they were used for the preparation of poly(methyl
acrylate)-block-poly(vinyl acetate), Malepu et al.13 achieved a polymer
of relatively high dispersity (Mw/Mn = 1.37).
Recently, we reported new acid/base “switchable” dithiocar-
bamate14ꢀ16 and trithiocarbonate17 RAFT agents. The dithio-
carbamates offer good control over polymerization of both MAMs
and LAMs and provide a route to polyMAM-block-polyLAM
copolymers of low dispersity. To prepare well controlled copoly-
mers, with unimodal molecular weight distributions, the polyMAM
propagating radical must add effectively to the LAM. In organic
solution, optimum control of MAMs was achieved in the presence
Reversible additionꢀfragmentation chain transfer (RAFT)
polymerization,1ꢀ4 mediated by thiocarbonylthio chain transfer
agents (or RAFT agents, 1) (Scheme 1), possesses several advan-
tages over other forms of reversible deactivation radical polymeriza-
tion (RDRP),5 such as atom transfer radical polymerization
(ATRP)6 and nitroxide mediated polymerization (NMP).7,8 These
include the ability to control the polymerization of a broad range of
functional monomers (including vinyl esters and amides) and the
absence of potentially toxic transition metal catalysts. RAFT poly-
merizations are simple to implement, because experimental condi-
tions can mirror those of conventional radical polymerization;
differing only by the addition of a RAFT agent.2ꢀ4
To achieve optimal control over a RAFT polymerization,
addition of the monomer derived propagating radical (Pn•) to the
thiocarbonyl of a RAFT agent 1 and subsequent fragmentation of
the RAFT intermediate 2 must occur efficiently. To facilitate this,
the selection of a RAFT agent suitable for the monomer system is
critical. Because propagating radicals of “more-activated” monomers
(MAMs) (e.g., methacrylic, acrylic and styrenic monomers) are
somewhat stabilized by conjugation, they are less reactive than propa-
gating radicals derived from the “less-activated” monomers (LAMs)
(e.g., vinyl esters and vinylamides). As such, these two classes of
monomer require RAFT agents that are tailored to their differing
reactivity. For effective control over polymerization of MAMs, dithio-
esters (Z = alkyl or aryl) or trithiocarbonates (Z = SR) are generally
used. When these RAFT agents are used in the polymerization of
LAMs, inhibition/retardation is observed, as fragmentation of the
more reactive LAMs derived propagating radical is slow with respect
to propagation.9 The presence of O or N as the “Z” group adjacent
to the thiocarbonyl, as is the case with xanthates (Z = OR) or
Received: April 1, 2011
Revised:
July 28, 2011
Published: August 15, 2011
Published 2011 by the American Chemical Society
6738
dx.doi.org/10.1021/ma200760q Macromolecules 2011, 44, 6738–6745
|

Macromolecules
ARTICLE
Scheme 1. Simplified Mechanism of RAFT Polymerization
Waters Q-TOF-II using 35 eV cone voltage, employing lock spray and
sodium iodide as a reference sample. High resolution gel permeation
chromatography (GPC) was performed on a system comprising a Waters
590 HPLC pump and a Waters 410 refractive index detector equipped with
3 ꢁ Waters Styragel columns (HT2, HT3, HT4, each 300 mm ꢁ7.8 mm
providing an effective molecular weight range of 100ꢀ600 000). The eluent
was N,N-dimethylformamide (DMF) (containing 0.45% w/v lithium bro-
mide (LiBr)) at 80 °C (flow rate: 1 mL minꢀ1). Number-average (Mn) and
weight-average (Mw) molecular weights were evaluated using Waters
Millennium software. High-throughput GPC measurements were performed
on a Shimadzu system equipped with a CMB-20A controller system, a SIL-
20A HT autosampler, a LC-20AT tandem pump system, a DGU-20A
degasser unit, a CTO-20AC column oven, a RDI-10A refractive index
detector, and a PL Rapide (Varian) column at 70 °C. N,N-Dimethylaceta-
mide(DMAc) (with2.1gLꢀ1 of lithium chloride (LiCl)) was used as eluent
at a flow rate of 1 mL minꢀ1. The GPC columns were calibrated with low
dispersity polystyrene standards (Polymer Laboratories) ranging from 3100
to 650 000 g molꢀ1, and molecular weights are reported as polystyrene
equivalents. A third-order polynomial was used to fit the log Mp vs time
calibration curve, which was linear across the molecular weight ranges.
RAFT Agent Synthesis. Synthesis of 2-(2-(2-(2-Hydroxyethoxy)-
ethoxy)ethoxy)ethyl 2-Bromopropanoate. A solution of tetraethylene glycol
(188 g, 968 mmol, 10.1 equiv) and pyridine (7.90 g, 8.1 mL, 100 mmol, 1.05
equiv) in dry tetrahydrofuran (THF) (100 mL) had a solution of 2-bromo-
propionyl bromide (20.6 g, 10 mL, 95.5 mmol) in dry THF (50 mL) added
dropwise over 1 h after which the reaction mixture was allowed to warm to
room temperature (RT) and stirred for a further 16 h. Aqueous hydrochloric
acid (HCl) (1 M, 500 mL) was added and the resulting mixture extracted
with dichloromethane (DCM, 6 ꢁ 100 mL). The combined organics were
washed with water (2 ꢁ 100 mL) and dried (anhydrous sodium sulfate
(Na2SO4)), and solvent was removed under reduced pressure to give
2-(2-(2-(2-hydroxyethoxy)ethoxy)ethoxy)ethyl 2-bromopropanoate as a
colorless liquid (23.34 g, 70.9 mmol, 74%); 1H NMR (200 MHz, CDCl3)
δ 1.73 (d, J = 6.9 Hz, 3H, CH3), 3.19 (br s, 1H, OH), 3.46ꢀ3.70 (m, 14H,
of 1 equiv of a strong acid (with respect to the RAFT agent), such as
trifluoromethanesulfonic acid or p-toluenesulfonic acid monohy-
drate (TsOH).14 Weaker acids were less effective presumably due to
the weak basicity of the pyridyl functionality (pKa of pyridinium =
5.23).18 This requirement has implications for the use of these RAFT
agents in aqueous polymerization because dissociation of the strong
acid with quantitative formation of hydronium ion (H3O+) in
aqueous solution will decrease the effective acid strength. Further-
more, competitive acidꢀbase equilibria due to the comparatively
high concentration of water has the potential to reduce the efficiency
of “switching” via protonation, when compared with that observed in
organic solution and reduce the level of control achievable.
There has been recent focus on RAFT polymerization per-
formed in aqueous media, partially due to potential economic
benefits.19 The use of water as the reaction solvent also facilitates
control in the polymerization of monomers containing highly polar
functionalities, which include cationic, anionic, zwitterionic and
neutral polar groups,19 which have low solubility in organic media.
To this end, we investigated the applicability of the switchable RAFT
protocol being performed in aqueous media.
In this investigation we illustrate the importance of the
strength (pKa) and stoichiometry of the switching acid for aqueous
polymerization in the presence of switchable N-alkyl-N-pyridyldithio-
carbamate RAFT agents, utilizing the MAM, DMAm, as the example
case. Subsequent neutralization of the PDMAm macro-RAFT agent
and block copolymer formation with the LAMs, NVC, VAc, and
NVP is also presented.
OCH2CH2O), 4.23 (m, 2H, CO2CH2), 4.32 (q, J = 6.9 Hz, 1H, CH); 13
C
NMR (50 MHz, CDCl3) δ 21.5 (CH3), 39.9 (CH), 61.5 (CH2OH), 64.9
(CO2CH2), 68.6 (OCH2), 70.2 (OCH2), 70.4 (OCH2), 70.5 (OCH2), 72.4
(OCH2), 170.1 (CdO). MS (HR-ESI) 351.0417 m/z [M + Na]+
(C11H21BrO6Na requires 351.0419).
’ EXPERIMENTAL SECTION
Synthesis of 2-(2-(2-(2-Hydroxyethoxy)ethoxy)ethoxy)ethyl 2-((Methyl-
(pyridin-4-yl)carbamothioyl)thio)propanoate (6). To a solution of 4-
(methylamino)pyridine (1.08 g, 10 mmol) in dry THF (60 mL) under
nitrogen was added n-BuLi (6.25 mL, 1.6 M, 10 mmol, 1 equiv) in hexanes
dropwise, and the reaction mixture was left to stir at RT for 1 h. Subsequently,
carbon disulfide (CS2) (0.9 mL, 1.14 g, 15 mmol, 1.5 equiv) was added
dropwise, and the reaction mixture was stirred at RT for 2 h under nitrogen.
Afterward, 2-(2-(2-(2-hydroxyethoxy)ethoxy)ethoxy)ethyl 2-bromopro-
panoate (3.62 g, 11 mmol, 1.1 equiv) was added and the resultant solution
was stirred at RT for a further 4 h. Saturated sodium bicarbonate solution
(NaHCO3) was added, and the solution extracted with DCM (2 ꢁ 100 mL)
washed with saturated NaHCO3 (2 ꢁ 100 mL) and brine (1 ꢁ 100 mL).
The combined organics were dried (Na2SO4) and filtered, and the solvent
was removed under reduced pressure to give an orange liquid. Purification by
column chromatography (SiO2, eluent: 95% chloroform (CHCl3)/5%
methanol (MeOH)) gave 2-(2-(2-(2-hydroxyethoxy)ethoxy)ethoxy)ethyl
2-((methyl(pyridin-4-yl)carbamothioyl)thio)propanoate 6 as a viscous yel-
low liquid (3.93 g, 91%); 1H NMR (200 MHz, CDCl3) δ 1.53 (d, J = 7.5 Hz
3H, CH3), 2.68 (br t, 1H, OH), 3.56ꢀ3.79 (m, 14H, OCH2CH2O), 3.71 (s,
1H, NCH3), 4.28 (m, 2H, CO2CH2), 4.64 (q, J = 7.5 Hz, 1H, CH), 7.26 (m,
2H, pyridyl ArH), 8.75 (m, 2H, pyridyl ArH); 13C NMR (50 MHz, CDCl3)
δ 16.8 (CHCH3), 45.2 (NCH3) 49.8 (CHCH3), 61.7 (CH2OH), 64.9
(CO2CH2), 68.9 (OCH2), 70.3 (OCH2), 70.5 (OCH2), 70.6 (OCH2), 70.6
(OCH2), 72.6 (OCH2), 121.8 (pyC-3), 151.6 (pyC-2), 151.9 (pyC-4), 171.8
Materials. All solvents were of analytical reagent (AR) grade unless
otherwise stated. Tetraethyleneglycol, 4-(methylamino)pyridine, bromo-
acetonitrile, 2-bromopropionyl bromide, n-butyllithium (n-BuLi, 1.6 M
in hexanes), TsOH, trifluoroacetic acid (CF3CO2H), chloroacetic acid
(ClCH2CO2H), acetic acid (AcOH), DMAm, VAc, NVP, and NVC
were purchased from Sigma-Aldrich and used as received unless other-
wise stated. 2,20-Azobis[2-methyl-N-(2-hydroxyethyl)propionamide]
(VA 086) was purchased from Wako chemicals and used as received.
DMAm and VAc were filtered through neutral alumina (70ꢀ230 mesh)
and fractionally distilled under reduced pressure immediately before use.
NVP was filtered through neutral alumina and purified by fractional thawing of
frozen sample (zone refining)20 immediately before use. Milli-Q water (18.2
MΩ cm) was used for all aqueous polymerization reactions. Cyanomethyl
methyl(pyridin-4-yl)carbamodithioate (4) and methyl 2-((methyl(pyridin-4-
yl)carbamothioyl)thio)propanoate (5) were prepared by the reported litera-
ture procedures.14 Analytical thin layer chromatography (TLC) was performed
on Merck Silica Gel F254 TLC plates. Preparative column chromatography was
performed using Merck Silica Gel 60 (mesh 230ꢀ400).
Characterization. Nuclear magnetic resonance (NMR) spectra
were obtained with a Bruker Avance 200 or 400 MHz spectrometer (1H
200 or 400 MHz, 13C 50 MHz). High resolution electron impact (HR-
EI) mass spectra recorded using a ThermoQuest MAT95XP operating
at 70 eV using perfluorokerosene (PFK) as a reference. High resolution
electrospray ionization (HR-ESI) mass spectra were recorded using a
6739
dx.doi.org/10.1021/ma200760q |Macromolecules 2011, 44, 6738–6745

Macromolecules
ARTICLE
(CdO), 196.9 (CdS); MS (HR-ESI) 455.1293 m/z [M + Na]+ (C18H28-
N2O6S2Na requires 455.1287).
Polymerization of N,N-Dimethylacrylamide. Ampule Experi-
ments. The polymerization of DMAm was performed by preparing stock
solutions of known concentration, which were placed in ampules, degassed
via three freezeꢀevacuateꢀthaw cycles, flame-sealed, and heated for the
designated time period. An example is shown below.
handling system of the synthesizer at the end of the experiment by
adding the corresponding GPC and NMR solvents. All GPC samples
from the automated synthesizer were measured using a high-throughput
GPC instrument (DMAc) to obtain kinetic trends in data. GPC for the
highest conversion sample of each reaction mixture was also performed
using by high-resolution GPC (DMF).
Block Copolymers. The synthesis of block copolymers of DMAm
and LAMs were performed by preparing stock solutions of known con-
centration, which were placed in ampules, degassed via three freezeꢀ
pumpꢀthaw cycles, flame-sealed and heated for the designated time period.
An example is shown below (see Supporting Information for the experi-
mental conditions for the preparation of the PDMAm macro-RAFT agent B
and additional block copolymers).
Preparation of Poly(N,N-dimethylacrylamide) Macro-RAFT Agent A
Using Cyanomethyl Methyl(pyridin-4-yl)carbamodithioate (4) at
80 °C. A mixture of cyanomethyl methyl(pyridin-4-yl)carbamodithioate
(4) (102 mg), TsOH (88.5 mg), DMAm (4.53 g), VA 086 (13.5 mg),
and water to a volume of 25 mL was prepared in a volumetric flask.
Aliquots of this stock solution were transferred to 2 ampules that were
degassed by three repeated freezeꢀevacuateꢀthaw cycles and flame-
sealed. The ampules were heated at 80 °C for 30 min, cooled, isolated,
and combined. The resultant solution was neutralized with excess
sodium bicarbonate (pH ∼7ꢀ8) and the water removed by lyophiliza-
tion. The isolated residue was dissolved in a minimal amount of DCM
and filtered to remove inorganic salts. Precipitation three times into
diethyl ether gave the purified PDMAm macro-RAFT agent A (2.27 g,
Mn,NMR = 7060, Mn,GPC = 10 000, Mw/Mn = 1.17).
Preparation of Low Dispersity Poly(N,N-dimethylacrylamide)-
block-poly(N-vinylcarbazole) at 60 °C. A stock solution I of 2,20-
azobis(2-methylpropionitrile) (AIBN, 66.5 mg) in a volume of 5 mL
of 1,4-dioxane was prepared in a volumetric flask. A mixture of PDMAm
macro-RAFT agent A (615 mg, Mn,NMR = 7060, Mn,GPC = 10 000,
Mw/Mn = 1.17), NVC (1.67 g), stock solution I (0.5 mL), and 1,4-dioxane
to a volume of 5 mL was prepared in a volumetric flask. Aliquots of this
stock solution were transferred to ampules that were degassed by three
repeated freezeꢀevacuateꢀthaw cycles and sealed. The ampules were
heated at 60 °C for specified times (see Supporting Information for details
of additional polymerizations).
Preparation of Low Dispersity Poly(N,N-dimethylacrylamide) Using
Cyanomethyl Methyl(pyridin-4-yl)carbamodithioate (4) at 80 °C. Stock
solution I of VA 086 (27 mg) in a volume of 5 mL of water was prepared in a
volumetric flask. A mixture of cyanomethyl methyl(pyridin-4-yl)carbamodi-
thioate (4) (40.8 mg), TsOH (35.4 mg), DMAm (1.81 g), stock solution I
(1 mL), and water to a volume of 10 mL was prepared in a volumetric flask.
Aliquots (5 mL) of this stock solution were transferred to ampules that were
degassed by three repeated freezeꢀevacuateꢀthaw cycles and sealed. The
ampules were heated at 80 °C for specified times (see Supporting Informa-
tion for details of additional polymerizations).
High-Throughput Experiments. High-throughput RAFT polymer-
ization experiments were performed using a Chemspeed Swing-SLT auto-
mated synthesizer following procedures similar to those described else-
where.21,22 The synthesizer was equipped with a glass reactor block
consisting of 16 reaction vessels (13 mL) with thermal jackets connected
in series through the reaction block and connected to a heating/cooling
system (H€uber, ꢀ90 to +140 °C). In addition, all reaction vessels were
equipped with coldfinger reflux condensers (∼7 °C). Mixing was achieved
by vortex agitation (up to 1400 rpm). Liquid transfers were handled by a
4-needle head (4-NH) capable of four simultaneous sample transfers. The
4-NH was connected to a reservoir bottle (degassed water) for needle rinsing
after each liquid transfer step. This solvent reservoir was degassed by
continuously sparging with nitrogen and was also utilized to prime the
tubing lines of the 4-NH. When experiments were carried out, the synthesizer
was maintained under an inert atmosphere by supplying a constant flow of
nitrogen into the hood of the synthesizer. A nitrogen atmosphere was also
applied to reactors and stock solutions at all times. Prior to the polymeriza-
tions, an “inertization process” was undertaken in which the reaction vessels
were heated to 135 °C and subjected to 10 cycles of vacuum (2 min) and
filling with nitrogen (2 min) to ensure the elimination of oxygen.
Stock solutions were prepared, degassed by sparging with nitrogen for
15 min, and placed inside the automated synthesizer. Stock solutions of
RAFT agent 6, acid (TsOH, CF3CO2H, ClCH2CO2H, and AcOH) and
VA 086 in water were prepared. Aliquots of stock solutions and water
from the reservoir were transferred into the reactors with the automated
liquid handling system to provide the desired concentrations of reagents
(see Supporting Information for more details). After the liquid transfers
were completed, pH measurements were performed using a pH needle
in the 4-NH of the automated synthesizer. The pH needle was calibrated
with standard buffer solutions (pH = 2, 4, 7, 9, 10). The reaction
solutions were degassed through three freezeꢀevacuateꢀthaw cycles
carried out as follows: The reaction mixtures were cooled to ꢀ35 °C
while vortex was applied to the reaction block (200 rpm, 2 min). Vacuum
(∼10 mbar) was then applied to the reactor block while the reactors
were heated to ꢀ10 °C with vortex (600 rpm, 2 min). The reactors were
then filled with nitrogen. The effectiveness and reproducibility of this
degassing procedure, conducted within this type of automated parallel
synthesizer, is reported elsewhere.23 After the degassing procedure, the
reactors were sealed under a nitrogen atmosphere and heated to desired
reaction temperature (80 °C) while vortex was applied to the reaction
block (150 rpm); the temperature of the reflux condensers on top of the
reactors was set to 7 °C. Time zero (t0) for kinetic data was taken to be
the moment the reactors reached 80 °C. At pre-established times,
monomer conversions and molecular weights of the formed polymers
were followed by sampling 75 μL from the reaction mixtures with the
automated liquid handling system into NMR tubes and GPC vials. GPC
and NMR samples for analysis were prepared with the automated liquid
Measurement of the Acid Dissociation Constant (pKa) of
RAFT Agent 6-H+. Measurements of pH for the 1H NMR titration of
RAFT agent 6 with TsOH were directly obtained from NMR tubes
containing RAFT agent 6 at different levels of protonation, as previously
prepared from stock solutions using the liquid handling system of the
automated synthesizer. The pH needle of the synthesizer was calibrated
with standard buffer solutions (pH = 2, 4, 7, 9, 10) prior to the
1
measurements. H NMR was taken of each solution and the chemical
shift of the hydrogen atoms adjacent to the pyridine nitrogen of 6 were
plotted against pH with the point of inflection giving the pKa (see
Supporting Information for pH and chemical shift data).
Estimation of the Equilibrium Concentration of 6-H+. The
equilibrium concentration of the RAFT agent 6-H+ present in the high-
throughput polymerization reaction mixtures was estimated with CurTi-
Pot pH and AcidꢀBase Titration, version 3.5.4 for Excel,24 using the
initial concentrations of the reagents and literature pKa values of TsOH
(∼ꢀ1.5),25 CF3CO2H (0.52),18 ClCH2CO2H (2.87),18 AcOH
(4.76),18 an approximation of the pKa of DMAm-H+ (∼ꢀ0.5),26 and
the calculated pKa of 6-H+ (see Supporting Information for details).
’ RESULTS AND DISCUSSION
Aqueous Polymerization of N,N-Dimethylacrylamide.
Aqueous RAFT polymerization of DMAm has been previously
reported for the synthesis of homopolymers27 and block
copolymers.28 Polymerization of acrylamido monomers, such
6740
dx.doi.org/10.1021/ma200760q |Macromolecules 2011, 44, 6738–6745

Macromolecules
ARTICLE
as DMAm, are often undertaken using acidic pH in an attempt to
reduce hydrolysis of the acrylamido monomer29 and in turn limit
RAFT agent aminolysis. This has been done via the use of buffered
solutions30 or alternatively by employing RAFT agents28 and/or
initiators27,28 that contain acidic functionality.
For the polymerization of DMAm using switchable RAFT the
reaction conditions were adapted from those reported previously
by the McCormick group.27 The nonacidic VA 086 was used as
the radical initiator so as not to affect the state of the switchable
RAFT agent in solution. Initially, we employed the cyanomethyl
RAFT agent 414 with a polymerization temperature of 80 °C in
water and a target molecular weight of 10 000 (Scheme 2). In the
presence of TsOH ([TsOH]/[RAFT] = 1) polymers of low
dispersity were obtained (entry 1, Table 1).
Polymerizations performed at 70 °C targeting different mo-
lecular weights gave broader dispersity as the targeted molecular
weight increased. This is conceivably due to a decrease in equilib-
rium concentration of the protonated RAFT agent, when the RAFT
agent and acid concentrations in the polymerization solution are
reduced (entries 2 and 3, Table 1). Similar results were obtained
with the secondary ester RAFT agent 514 at either 70 or 80 °C
(entries 4 and 5, Table 1). These RAFT agents (4 and 5) are not
soluble in water in the absence of acid preventing a control reaction
with the neutral nonprotonated RAFT agent.
To allow for aqueous polymerization of DMAm in absence of
acid and to illustrate any differences between the neutral and the
acidified (switched) forms, a new RAFT agent (6) containing a
water-solubilizing “R”group based on tetraethylene glycol (TEG) was
prepared. The TEG was converted to 2-(2-(2-(2-hydroxyethoxy)-
ethoxy)ethoxy)ethyl 2-bromopropanoate (74% yield) by reaction
with the corresponding bromoacyl bromide. This was subse-
quently converted to the desired RAFT agent 6 (91% yield) by
reaction with the methyl(pyridin-4-yl)carbamodithioate salt,
derived from 4-(methylamino)pyridine and CS2, via the
routine methodology.4,14 As expected, the dispersity and the
control of molecular weight for PDMAm synthesis, using the water-
soluble RAFT agent 6, were significantly improved in the presence
of 1 equiv of acid when compared to experiments conducted in the
absence of acid (Table 1, entries 6ꢀ9). These results illustrate the
effect of acidification upon switchable pyridyl-substituted dithiocar-
bamate RAFT agents in relation to their ability to control of the
polymerization of DMAm. Analogous results for the polymerization
of n-butyl acrylate, another MAM, with and without acid were
previously observed in organic solution.14
High Throughput Polymerization. High-throughput experi-
mentation and automated parallel synthetic techniques have
been demonstrated to be valuable tools for rapid kinetic mea-
surements in polymerization reactions (even in oxygen and moist-
ure sensitive reactions, such as anionic polymerization), for the rapid
preparation of well-defined polymeric materials, and for experiments
where a large parameter space must be explored.31ꢀ34 Use of these
experimental techniques has obvious advantages over the “classical”
lab-scale synthesis. These include a reduction in time and resources,
acceleration of research and enhanced accuracy and reproducibility,
as parallel experiments are performed concurrently under the same
conditions. For these reasons we made use of automated parallel
polymerization techniques to systematically study the effect of
different acids and stoichiometries upon the switchable RAFT
polymerization of DMAm.
After the initial success using TsOH as the acid switch, we
undertook experiments with varying acid concentration and acid
strength, as the degree of protonation of the RAFT agent has a
marked effect on the level of control achieved during polymer-
ization. It was postulated that, if proton transfer occurred at a
much higher rate than that of additionꢀfragmentation, then less
than 1 equiv of acid with respect to RAFT agent concentration
may give similar control to that observed when the ratio is 1:1.
This was not the case for these systems. The use of less than
1 equiv of acid ([TsOH]/[RAFT] = 0.5 or 0.2) gave broader
molecular weight distributions than that of 1 equiv. The equilib-
rium between the protonated and neutral forms of the RAFT agent
effectively results in a RAFT agent with activity somewhere between
active and nonactive forms in the presence of less than 1 equiv of
acid. This results in broader molecular weight distributions, as
shown in the example below using TsOH (Figure 1). While the
dispersities of the polymers prepared with less than 1 equiv of acid to
RAFT agent are broader than those prepared with equimolar acid,
there appears to be a preference for the propagating radical to attack
the switched fraction of the RAFT agent in solution in these
reactions, as evidenced by the molecular weight distributions being
significantly narrowed (0.5 equiv, Mw/Mn = 1.21; 0.2 equiv,
Scheme 2. RAFT Polymerization of Dimethylacrylamide in
Water
Table 1. Details of Polymer Syntheses Using Standard Experimental Methodsa
[RAFT] (10ꢀ2 M)
[TsOH] (10ꢀ2 M)
[I] (10ꢀ4 M)
T (°C)
Mn
Mn (theory)c
Mw/Mn
time (h)
conv (%)
b
b
entry
RAFT agent
1
2
3
4
5
6
7
8
9
4-H+
4-H+
4-H+
5-H+
5-H+
6-H+
6
1.83
1.83
1.83
0.915
0.915
0.457
0.915
0
18.3
18.3
18.3
18.3
9.15
18.3
18.3
0.29
0.29
80
70
70
80
70
70
70
80
80
11500
13000
22500
25700
36100
21000
27800
22700
20800
9500
9600
1.11
1.09
1.13
1.15
1.30
1.21
1.48
1.20
1.64
1
3
3
4
3
1
1
6
6
95
96
1.83
0.915
0.915
0.457
0.915
0.915
0.915
0.915
19000
19800
38400
17000
17800
17800
14000
95
>99
96
85
89
6-H+
0.915
0
89
6
70
a [M] = 1.83 M. b GPC DMF eluent, PS equivalents. c Mn (theory) = [([M]0/[RAFT]0) ꢁ MWmonomer] ꢁ % conversion.
6741
dx.doi.org/10.1021/ma200760q |Macromolecules 2011, 44, 6738–6745

Macromolecules
ARTICLE
Figure 2. GPC chromatograms of PDMAm 6-H+ ([TsOH]/[6] = 1;
Mn = 11 500, Mw/Mn = 1.10) (red curve), PDMAm 6-H+ ([CF3CO2H]/
[6] = 1; Mn = 10 000, Mw/Mn = 1.18) (blue curve), PDMAm 6-H+
([ClCH2CO2H]/[6] = 1; Mn = 10 300, Mw/Mn = 1.26) (green curve), and
PDMAm 6-H+ ([AcOH]/[6] = 1; Mn = 11 100, Mw/Mn = 1.38) (black
curve).
Figure 1. GPC chromatograms of PDMAm 6-H+ ([TsOH]/[6] = 1;
Mn = 11 500, Mw/Mn = 1.10) (red curve), PDMAm 6ꢀ0.5H+ ([TsOH]/
[6] = 0.5; Mn = 10 600, Mw/Mn = 1.21) (blue curve), and PDMAm
6ꢀ0.2H+ ([TsOH]/[6] = 0.2; Mn = 10 900, Mw/Mn = 1.29) (green curve).
Mw/Mn = 1.29) from those in the absence of acid (ca. Mw/Mn
∼1.50). This would be expected to occur to some degree if the
rate of proton transfer is indeed faster than that of additionꢀ
fragmentation. While these results may suggest the use of excess acid
([TsOH]/[RAFT] > 1) may further improve the control over the
polymerization of DMAm, preliminary experiments using
RAFT agent 4 showed no significant improvement in disper-
sity, when compared to stoichiometric acid concentrations
([TsOH]/[RAFT] = 1). Further to this, inhibition of polymer-
ization was observed, increasing the probability of unwanted side
reactions (see Supporting Information for details).
Unsurprisingly, polymers of lower dispersity were also obtained
with increasing acid strength, whereby TsOH (pKa ∼ ꢀ1.5)25 gave
PDMAm with the lowest dispersity followed by CF3CO2H (0.52),18
ClCH2CO2H (2.87),18 and AcOH (4.76)18 (see Figure 2).
The combined effects of acid concentration and acid strength
on RAFT polymerization of DMAm can be appreciated by examin-
ing Figure 3 (also Table 2), which clearly shows the trends for lower
dispersity with an increase in either parameter (further details of
polymer syntheses available in Supporting Information).
Figure 3. Dispersity (Mw/Mn) dependence on acid strength and concen-
tration at high monomer conversion (NB: data shown for high-throughput
GPC with DMAc eluent).
Acid Dissociation Constant (pKa) of RAFT Agent 6-H+. The
results obtained when the polymerization of DMAm is performed in
the presence of various acids can be rationalized by calculating the
acid dissociation constant (pKa) of the switchable RAFT agent 6-H+
and therefore the extent of protonation in solution. This was
achieved by 1H NMR titration of 6 with TsOH in 20% dimethyl-
sulfoxide (DMSO)-d6/80% water (v/v) (Scheme 3), following the
protocol of Grycova et al.,35 in which the chemical shift of hydrogen
atoms at the 2 and 6 positions of the pyridine ring is observed with
changing pH. From the titration curve (Figure 4) the pKa of 6-H+
was calculated to be 3.13. It has been previously reported that
concentrations of DMSO need to be upward of 26% (v/v) to see
appreciable changes (>(0.2) in pKa from those in pure water, as the
molar ratio of DMSO to water is significantly lower than the volume
ratio (ca. 6% at 20% v/v).36
[6-H+]eq/[6]0 and therefore the dispersity of the resultant polymer.
This can be seen by the decreasing dispersity of PDMAm with
increasing amounts of acid in the case of TsOH, CF3CO2H, and
ClCH2CO2H. In the case of AcOH there is negligible difference in
the dispersity with increasing acid concentration. This can be
rationalized by the fact that the pKa of AcOH is higher than that
of the RAFT agent 6-H+ and is therefore AcOH is unable to
protonate the RAFT agent 6 to any appreciable amount. This effect
on the apparent transfer constant can also be seen in Figure 5, where
narrowing of dispersity is more pronounced in the presence of
CF3CO2H and ClCH2CO2H when compared with that of TsOH.
A decrease in transfer constant results in addition of more monomer
units per active cycle, induced by less efficient protonation of the
RAFT agent 6. As the polymerization proceeds, the dispersity
decreases as the variation between chains becomes less significant.
In the presence of AcOH, the narrowing of dispersity in less pro-
nounced due to the RAFT agent 6 not being protonated to any
appreciable amount, and hence behaving as a poorer control agent.
Upon protonation of 6, the apparent transfer constant increases,
leading to polymers with lower dispersity in the polymerization of
DMAm. The extent of protonation ([6-H+]eq/[6]0) is dictated by
the pKa of the acid compared to that of the RAFT agent 6-H+
(∼3.13), where dispersity is decreased significantly with increasing
acid strength (refer to Table 2). Acid concentration also affects
6742
dx.doi.org/10.1021/ma200760q |Macromolecules 2011, 44, 6738–6745

Macromolecules
ARTICLE
Table 2. Details of Polymer Syntheses Using High-Throughput Methodsa
c
c
entry
acid
[acid]/[6]
pH (t0)b
Mn
Mn (theory)d
Mw/Mn
time (min)
conv (%)
[6-H+]eq/[6]0 (%)e
1
2
TsOH
1.0
0.5
0.2
1.0
0.5
0.2
1.0
0.5
0.2
1.0
0.5
0.2
2.77
3.47
4.14
1.88
2.92
3.45
3.12
3.48
3.99
3.86
4.17
4.54
11500
10600
10900
10000
10400
10300
10300
11300
10800
11100
11700
11000
8500
8400
8200
8000
8000
8000
8200
8600
8500
8300
8700
7900
1.10
1.21
1.29
1.18
1.19
1.31
1.25
1.23
1.35
1.38
1.37
1.41
74
76
78
66
68
70
66
68
70
71
73
75
85
84
82
80
80
80
82
86
85
83
87
79
78
45
19
77
45
19
56
36
17
13
9
TsOH
3
TsOH
4
CF3CO2H
CF3CO2H
CF3CO2H
ClCH2CO2H
ClCH2CO2H
ClCH2CO2H
AcOH
5
6
7
8
9
10
11
12
AcOH
AcOH
6
a [M] = 1.83 M, [6] = 1.83 ꢁ 10ꢀ2, target MW = 10 000. b pH at t0 measured before degas cycles. c GPC DMF eluent, PS equivalents. d Mn (theory) =
[([M]0/[RAFT]0) ꢁ MWmonomer] ꢁ % conversion. e Calculated using equilibrium concentrations, estimated from pKa values and the initial
concentrations of acid, 6 and DMAm using CurTiPot pH and AcidꢀBase Titration 3.5.4 for Excel.24
1
Scheme 3. H NMR Titration of 6 with TsOH, Illustrating
Hydrogen Atoms at the 2 and 6 Positions on the Pyridine Ring
Figure 5. Progression of dispersity with monomer conversion
(between 55 and 90%) with respect to acid strength ([acid]/[6] = 1):
TsOH (b); CF3CO2H (1); ClCH2CO2H (9); AcOH (2).
terminal NVP unit) in a reaction similar to that described by Pound
et al.37 for the decomposition of a xanthate end group of PNVP. As
this hydrolysis occurs at the terminal carbon of the NVP chain,37
dithiocarbamates, xanthates, and other types of RAFT agents can be
expected to behave similarly in this context. To the best of our
knowledge there are no reports in the literature pertaining to the
RAFT polymerization of LAMs (vinyl esters, vinyl amides, etc.) in
aqueous media. Nevertheless, PDMAm-block-PNVC (Figure 6a),
PDMAm-block-PVAc (Figure 6b), and PDMAm-block-PNVP
(Figure 6c) of low dispersity were successfully prepared from
PDMAm utilizing the switchable RAFT protocol.
For successful preparation of these block copolymers it is
important to purify the initial PDMAm macro-RAFT agent to
remove any residual monomer, as this is incompatible with
the RAFT group upon switching and will lead to poor control
in the polymerization of the second block. Neutralization, by the
addition of excess sodium bicarbonate, is preferred during this
purification process rather than alternately using the addition of
stoichiometric base directly to the block extension reaction mix-
ture containing the protonated macro RAFT agent. This is due
the difficulty of adding exactly 1 equiv of base to switch the RAFT
Figure 4. 1H NMR titration curve of 6 with TsOH in 20% DMSO-d6/
80% water (v/v) (NB: chemical shift of H atoms on 2 and 6 position
observed with changing pH) (solid line) and the second derivative
(dotted line) showing the pKa of 6-H+ as 3.13.
Block Copolymer Synthesis. The significance of switchable
RAFT is the ability to control a much wider range of monomers
than conventional RAFT process. This is exploited to greatest
effect in the preparation of block copolymers incorporating both
MAMs and LAMs. To illustrate this capacity, block copolymers of
PDMAm and the less-activated monomers NVC, VAc, and NVP
were prepared (see Supporting Information for full details of
polymerization conditions). All block copolymers were successfully
prepared in organic solution. Although NVP is water-soluble, a block
copolymer of NVP could not be prepared in aqueous solution. We
attribute this to hydrolytic instability in water of the hemithioaminal
group formed (arising from the dithiocarbamate end-group with a
6743
dx.doi.org/10.1021/ma200760q |Macromolecules 2011, 44, 6738–6745

Macromolecules
ARTICLE
Figure 6. GPC chromatograms of (a) PDMAm 4-H+ (Mn = 10 000, Mw/Mn = 1.17) (red curve) and PDMAm-block-PNVC 4 (Mn = 16 400, Mw/Mn =
1.13) (blue curve), (b) PDMAm 4-H+ (Mn = 10 000, Mw/Mn = 1.17) (red curve) and PDMAm-block-PVAc 4 (Mn = 20 100, Mw/Mn = 1.15) (blue
curve), and (c) PDMAm 4-H+ (Mn = 3660, Mw/Mn = 1.27) (red curve) and PDMAm-block-PNVP 4 (Mn = 16 000, Mw/Mn = 1.19) (blue curve).
end-group, preventing the presence of excess acid or base. Due to
the hydrolytic instability of the RAFT end-group upon incorpora-
tion of LAMs any excess acid (or base) results in diminished control
of the polymerization and broader molecular weight distributions in
the final product. For the polymerization of NVC or NVP block
copolymers, or homopolymers for that matter, removal of all acid is
essential as its presence degrades the monomer to carbazole or
pyrrolidone respectively, with acetaldehyde also being produced
(among other degradation products).38,39
D.J.K. acknowledges the Office of the Chief Executive of CSIRO
for an OCE postdoctoral fellowship.
’ REFERENCES
(1) Chiefari, J.; Chong, Y. K.; Ercole, F.; Krstina, J.; Jeffery, J.; Le,
T. P. T.; Mayadunne, R. T. A.; Meijs, G. F.; Moad, C. L.; Moad, G.;
Rizzardo, E.; Thang, S. H. Macromolecules 1998, 31, 5559–5562.
(2) Moad, G.; Rizzardo, E.; Thang, S. H. Aust. J. Chem. 2005, 58,
379–410.
(3) Moad, G.; Rizzardo, E.; Thang, S. H. Aust. J. Chem. 2006, 59,
669–692.
’ CONCLUSION
(4) Moad, G.; Rizzardo, E.; Thang, S. H. Aust. J. Chem. 2009, 62,
1402–1472.
(5) Jenkins, A. D.; Jones, R. G.; Moad, G. Pure Appl. Chem. 2010, 82,
483–491.
(6) Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921–2990.
(7) Moad, G.; Rizzardo, E.; Thang, S. H. Acc. Chem. Res. 2008, 41,
1133–1142.
We have demonstrated that switchable RAFT polymerization
can be implemented in aqueous media, by the preparation of low
dispersity PDMAm. Acid strength and concentration have a
marked effect on the degree of control over the polymerization.
TsOH, at equimolar concentration to RAFT agent, gave poly-
mers of lowest dispersity in the polymerization of DMAm.
Similar results are expected with the use of any acid with pKa < 0.
As the extent of protonation of the RAFT agent is reduced,
through the use of either a weaker acid or a lower concentration
of acid, the dispersity of the resulting polymer increases. Block
copolymers of PDMAm were successfully prepared with the
LAMs NVC, VAc, and NVP, following neutralization of the
RAFT end group, illustrating the utility of the switchable RAFT
protocol in the preparation of new polymeric materials.
(8) Hawker, C. J.; Bosman, A. W.; Harth, E. Chem. Rev. 2001, 101,
3661–3688.
(9) Chiefari, J.; Mayadunne, R. T. A.; Moad, C. L.; Moad, G.; Rizzardo, E.;
Postma,A.;Skidmore,M.A.;Thang,S.H.Macromolecules 2003,36, 2273–2283.
(10) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Chong, Y. K.;
Moad, G.; Thang, S. H. Macromolecules 1999, 32, 6977–6980.
(11) Destarac, M.; Bzducha, W.; Taton, D.; Gauthier-Gillaizeau, I.;
Zard, S. Z. Macromol. Rapid Commun. 2002, 23, 1049–1054.
(12) Destarac, M.; Charmot, D.; Franck, X.; Zard, S. Z. Macromol.
Rapid Commun. 2000, 21, 1035–1039.
’ ASSOCIATED CONTENT
(13) Malepu, V.; Petruczok., C. D.; Tran, T.; Zhang, T.; Thopasridharan,
M.; Shipp, D. A. In Controlled/Living Radical Polymerization: Progress in RAFT,
DT, NMP & OMRP; Matyjaszewski, K., Ed.; American Chemical Society:
Washington, DC, 2009; ACS Symposium Series Vol. 1024, pp 37ꢀ47.
(14) Benaglia, M.; Chiefari, J.; Chong, Y. K.; Moad, G.; Rizzardo, E.;
Thang, S. H. J. Am. Chem. Soc. 2009, 131, 6914–6915.
(15) Benaglia, M.; Chen, M.; Chong, Y. K.; Moad, G.; Rizzardo, E.;
Thang, S. H. Macromolecules 2009, 42, 9384–9386.
S
Supporting Information. Full details of experimental proce-
b
dures. This material is available free of charge via the Internet at
http://pubs.acs.org.
’ AUTHOR INFORMATION
(16) Moad, G.; Benaglia, M.; Chen, M.; Chiefari, J.; Chong, Y. K.;
Keddie, D. J.; Rizzardo, E.; Thang, S. H. In Non-Conventional Functional
Block Copolymers; Theato, P., Kilbinger, A. F. M., Coughlin, E. B.,
Eds.; American Chemical Society: Washington, DC, 2011; ACS Sym-
posium Series Vol. 1066, pp 81ꢀ102.
Corresponding Author
*E-mail: daniel.keddie@csiro.au (D.J.K.), graeme.moad@csiro.au
(G.M.).
(17) Bicciocchi, E.; Chong, Y. K.; Giorgini, L.; Moad, G.; Rizzardo,
E.; Thang, S. H. Macromol. Chem. Phys. 2010, 211, 529–538.
(18) Lide, D. R., CRC Handbook of Chemistry and Physics, 84th ed.;
CRC Press: Boca Raton, FL, 2003.
’ ACKNOWLEDGMENT
We acknowledge the Capability Development Fund of
CSIRO Materials Science and Engineering for financial support.
6744
dx.doi.org/10.1021/ma200760q |Macromolecules 2011, 44, 6738–6745

Macromolecules
ARTICLE
(19) Lowe, A. B.; McCormick, C. L. Prog. Polym. Sci. 2007, 32,
283–351.
(20) Bamford, C. H.; Schofield, E.; Michael, D. J. Polymer 1985,
26, 945–950.
(21) Guerrero-Sanchez, C.; Abeln, C.; Schubert, U. S. J. Polym. Sci.,
Part A: Polym. Chem. 2005, 43, 4151–4160.
(22) Guerrero-Sanchez, C.; Paulus, R. M.; Fijten, M. W. M.; de la Mar,
M. J.; Hoogenboom, R.; Schubert, U. S. Appl. Surf. Sci. 2006, 252,
2555–2561.
(23) Guerrero-Sanchez, C.; Keddie, D. J.; Martinez, I.; Harrisson, S.;
Saubern, S.; Thang, S. H.; Rizzardo, E.; Chiefari, J.; Moad, G. Manuscript
in preparation, 2011.
(24) Gutz, I. G. R. CurTiPot pH and Acid-Base Titration 3.5.4 for
Excel; http://www2.iq.usp.br/docente/gutz/Curtipot_.html.
(25) Berkowitz, B. J.; Grunwald, E. J. Am. Chem. Soc. 1961, 83, 2956.
(26) March, J. Advanced Organic Chemistry, 3rd ed.; John Wiley &
Sons: New York, 1985.
(27) Donovan, M. S.; Sanford, T. A.; Lowe, A. B.; Sumerlin, B. S.;
Mitsukami, Y.; McCormick, C. L. Macromolecules 2002, 35, 4570–4572.
(28) Convertine, A. J.; Lokitz, B. S.; Vasileva, Y.; Myrick, L. J.; Scales,
C. W.; Lowe, A. B.; McCormick, C. L. Macromolecules 2006, 39, 1724–1730.
(29) Lowe, A. B.; McCormick, C. L. In RAFT polymerization in
homogeneous aqueous media: initiation systems, RAFT agent stability,
monomers and polymer structures; Wiley-VCH Verlag GmbH & Co.
KGaA: Berlin, 2008; pp 235ꢀ284.
(30) Thomas, D. B.; Sumerlin, B. S.; Lowe, A. B.; McCormick, C. L.
Macromolecules 2003, 36, 1436–1439.
(31) Sciannamea, V.; Guerrero-Sanchez, C.; Schubert, U. S.; Catala,
J. M.; Jerome, R.; Detrembleur, C. Polymer 2005, 46, 9632–9641.
(32) Bonilla-Cruz, J.; Guerrero-Sanchez, C.; Schubert, U. S.;
Saldivar-Guerra, E. Eur. Polym. J. 2010, 46, 298–312.
(33) Guerrero-Sanchez, C.; Lohmeijer, B. G. G.; Meier, M. A. R.;
Schubert, U. S. Macromolecules 2005, 38, 10388–10396.
(34) Guerrero-Sanchez, C.; Wouters, D.; Fustin, C.-A.; Gohy, J.-F.;
Lohmeijer, B. G. G.; Schubert, U. S. Macromolecules 2005, 38,
10185–10191.
(35) Grycova, L.; Dommisse, R.; Pieters, L.; Marek, R. Magn. Reson.
Chem. 2009, 47, 977–981.
(36) Holmes, D. L.; Lightner, D. A. Tetrahedron 1995, 51, 1607–22.
(37) Pound, G.; McKenzie, J. M.; Lange, R. F. M.; Klumperman, B.
Chem. Commun. 2008, 3193–3195.
(38) Kurov, G. N.; Afonin, A. V.; Svyatkina, L. I.; Dmitrieva, L. L.;
Pal’chuk, E. G. Russ. Chem. B 1987, 36, 403–405.
(39) Ferguson, J.; Sundar Rajan, V. Eur. Polym. J. 1979, 15, 627–630.
’ NOTE ADDED AFTER ASAP PUBLICATION
This article posted ASAP on August 15, 2011. The second
sentence in the abstract has been revised. The correct version
posted on August 18, 2011.
6745
dx.doi.org/10.1021/ma200760q |Macromolecules 2011, 44, 6738–6745