Cyclodextrin-complexed RAFT agents for the ambient temperature aqueous living/controlled radical polymerization of acrylamido monomers

Article abstract of DOI:10.1021/ma2011969

The first aqueous reversible addition-fragmentation transfer (RAFT) polymerization of N,N-dimethylacrylamide (DMAAm), N,N-diethylacrylamide (DEAAm), and N-isopropylacrylamide (NIPAAm) utilizing host/guest complexes of cyclodextrin and hydrophobic chain transfer agents (CTAs) at 25 °C is described. Three novel guest-functionalized CTAs, namely 4-(tert-butyl)phenyl 2-(((ethylthio)carbonothioyl)thio)-2-methylpropanoate, bis(4-tert-butyl)benzyl carbonotrithioate, and benzyl (3-((4-(tert-butyl)phenyl)amino)-3-oxopropyl) carbonotrithioate, were synthesized and employed in aqueous RAFT polymerizations. The presented technique allows for the facile preparation of hydrophilic polymers with hydrophobic end groups in aqueous environments. The living/controlled radical polymerization afforded high molecular masses (7500 ≥ M_n ≥ 116 000 g mol^-1 for poly(DMAAm), 2500 ≥ M_n ≥ 150 000 g mol^-1 for poly(DEAAm), and 4000 ≥ M_n ≥ 50 000 g mol^-1 for poly(NIPAAm)) with low PDIs (1.06 ≥ PDI ≥ 1.54 for poly(DMAAm), 1.05 ≥ PDI ≥ 1.39 for poly(DEAAm), and 1.15 ≥ PDI ≥ 1.46 for poly(NIPAAm)). To confirm the living character of the polymerizations, kinetic measurements were undertaken that evidence a linear evolution of molecular weight with conversion. Furthermore, chain extensions were carried out that indicate a very high reinitiation efficiency (poly(DMAAm): from 10 500 to 97 500 g mol^-1, PDI = 1.08; poly(DEAAm): from 8500 to 83 000 g mol^-1, PDI = 1.13; poly(NIPAAm): from 9000 to 90 000 g mol^-1, PDI = 1.11). The resulting polymers were thoroughly characterized via N,N-dimethylacetamide (DMAc) size exclusion chromatography, ¹H NMR, and electrospray ionization-mass spectrometry (ESI-MS).

Full text of DOI:10.1021/ma2011969

ARTICLE
pubs.acs.org/Macromolecules
Cyclodextrin-Complexed RAFT Agents for the Ambient Temperature
Aqueous Living/Controlled Radical Polymerization of Acrylamido
Monomers
Bernhard V. K. J. Schmidt,† Martin Hetzer,‡ Helmut Ritter,*,‡ and Christopher Barner-Kowollik*,†
†Preparative Macromolecular Chemistry, Institut f€ur Technische Chemie und Polymerchemie, Karlsruhe Institute of Technology (KIT),
Engesserstr. 18, 76128 Karlsruhe, Germany
‡Lehrstuhl f€ur Pr€aparative Polymerchemie, Institut f€ur Organische Chemie und Makromolekulare Chemie, Heinrich-Heine Universit€at,
Universit€atsstrasse 1, Geb. 26.33.00, 40225 D€usseldorf, Germany
S Supporting Information
b
ABSTRACT:
The first aqueous reversible additionꢀfragmentation transfer (RAFT) polymerization of N,N-dimethylacrylamide (DMAAm), N,
N-diethylacrylamide (DEAAm), and N-isopropylacrylamide (NIPAAm) utilizing host/guest complexes of cyclodextrin and
hydrophobic chain transfer agents (CTAs) at 25 °C is described. Three novel guest-functionalized CTAs, namely 4-(tert-
butyl)phenyl 2-(((ethylthio)carbonothioyl)thio)-2-methylpropanoate, bis(4-tert-butyl)benzyl carbonotrithioate, and benzyl
(3-((4-(tert-butyl)phenyl)amino)-3-oxopropyl)carbonotrithioate, were synthesized and employed in aqueous RAFT polymeriza-
tions. The presented technique allows for the facile preparation of hydrophilic polymers with hydrophobic end groups in aqueous
environments. The living/controlled radical polymerization afforded high molecular masses (7500 e Mn e 116 000 g molꢀ1 for
poly(DMAAm), 2500 e Mn e 150 000 g molꢀ1 for poly(DEAAm), and 4000 e Mn e 50 000 g molꢀ1 for poly(NIPAAm)) with
low PDIs (1.06 e PDI e 1.54 for poly(DMAAm), 1.05 e PDI e 1.39 for poly(DEAAm), and 1.15 e PDI e 1.46 for
poly(NIPAAm)). To confirm the living character of the polymerizations, kinetic measurements were undertaken that evidence a
linear evolution of molecular weight with conversion. Furthermore, chain extensions were carried out that indicate a very high
reinitiation efficiency (poly(DMAAm): from 10 500 to 97 500 g molꢀ1, PDI = 1.08; poly(DEAAm): from 8500 to 83 000 g molꢀ1
,
PDI = 1.13; poly(NIPAAm): from 9000 to 90 000 g molꢀ1, PDI = 1.11). The resulting polymers were thoroughly characterized via
N,N-dimethylacetamide (DMAc) size exclusion chromatography, 1H NMR, and electrospray ionization-mass spectrometry (ESI-MS).
’ INTRODUCTION
cyclodextrins including drug delivery,8 supramolecular hydrogels,9
supramolecular polymers,10 polymer/enzyme conjugates,11 or
optical receptors.12 Furthermore, cyclodextrins are obtained
from starch making it a renewable resource, and its applications
are thus highly interesting from the point of sustainability.1
Controlled/living radical polymerization is a versatile tool for
the preparation of complex macromolecular architectures, e.g.,
Cyclodextrins play a significant role in current polymer
research.1,2 The ability of cyclodextrins to form supramolecular
host/guest complexes with hydrophobic recognition sites such as
adamantyl, 4-tert-butylphenyl, or isobornyl in aqueous solution
leads to broad opportunities for the synthesis of complex
macromolecular architectures. Recent examples are supramolecular
block copolymers,3,4 the connection of cyclodextrin-centered
stars with guest endfunctionalized polymers,5 cyclodextrin/guest
networks,6 or supramolecular grafting.7 A variety of applica-
tions are proposed for macromolecular systems containing
Received: May 26, 2011
Revised:
August 12, 2011
Published: August 25, 2011
r
2011 American Chemical Society
7220
dx.doi.org/10.1021/ma2011969 Macromolecules 2011, 44, 7220–7232
|

Macromolecules
ARTICLE
Scheme 1. Procedure for the RAFT Polymerization of Acrylamido Monomers with Cyclodextrin-Complexed CTAsa
a DMAAm: R = R0 = Me; DEAAm: R = R0 = Et; NIPAAm: R = i-Pr, R0 = H: R-approach (CTA1), Z-approach (CTA3), and combined approach (CTA2).
The guest substituent, e.g. 4-tert-butylphenyl, is depicted in blue.
brushes, block copolymers, and bioconjugates. Molecular weight
control and low polydispersity as well as control over the polymer
end groups can be accomplished via nitroxide-mediated radical
polymerization,13,14 atom transfer radical polymerization,15,16 or
the reversible additionꢀfragmentation transfer (RAFT) process.17ꢀ21
In particular, RAFT polymerization has emerged as a very
efficient tool for the synthesis of water-soluble polymers due to
its high tolerance of functional monomers.22ꢀ25 The polymeri-
zation itself can be conducted directly in water with a broad range
of water-soluble monomers that include methacrylates,26 metha-
crylamides,27 styrenics,28 acrylates,29 and acrylamides30ꢀ32 so far.
Furthermore, the utilization of water is interesting from an envi-
ronmental and economic point of view as water is nontoxic,
nonflammable, and readily available. Thus, it is easy to handle
safely and low priced.
The possibility to solubilize hydrophobic molecules in water is
another interesting application of cyclodextrins. In this context
hydrophobic monomers were solubilized in water with cyclodex-
trins, e.g., in radical polymerization,33,34 living radical polymeriza-
tion,35ꢀ37 enzymatic polymerization,38 and rhodium-catalyzed
polymerization.39 To connect the solubilizing effect of cyclo-
dextrins with RAFT polymerization, we investigated the aque-
ous RAFT polymerization of three acrylamido monomers,
namely N,N-dimethylacrylamide (DMAAm), N,N-diethylacryla-
mide (DEAAm), and N-isopropylacrylamide (NIPAAm), in the
presence of a hydrophobic chain transfer agent (CTA) that is
solubilized via a host/guest complex. The employed hydropho-
bic CTAs bear the 4-tert-butylphenyl group that is well-known
for its stable host/guest complexes with β-cyclodextrins.40,41
Three novel CTAs were synthesized, which contain the guest
group in regions of variable reactivity within the molecule. The
guest group can be incorporated in the R group, the Z group, or
both. As the CTA allows control over the chain-end functionality
in RAFT polymerization, hydrophobic chain ends are obtained
directly (as depicted in Scheme 1).
The hydrophobic chain ends may have further use as guests in
complex self-assemblies with cyclodextrin-functionalized poly-
mers or surfaces e.g. to construct supramolecular block copolymers3,4
or to obtain supramolecular grafting, e.g. on cellulose.7 In the
case of thermoresponsive polymers that show lower critical
solution temperature (LCST) behavior such as poly(NIPAAm)
or poly(DEAAm), it is well-known that hydrophobic end groups
can induce a change in the observed LCST.42ꢀ44 Therefore, a
modulation of the thermoresponsivity of these polymers is pos-
sible. Furthermore, most water-soluble CTAs contain carboxylic
acid or sulfonic acid groups which lead to acid-functionalized
polymers.23,30,45 In cases where acid-functionalized water-soluble
polymers—e.g., because of unspecific interactions in biological
systems—are undesirable, a polymer analogous removal/mod-
ification is required which can be complicated depending on the
reaction type. Therefore, it is interesting to study the polymer-
ization via hydrophobic guest-functionalized CTA/cyclodextrin
7221
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
complexes in water. Our current approach provides the oppor-
tunity to create hydrophilic polymers without acidic end groups
in one step.
on silica gel with n-hexane:ethyl acetate 1:2 as eluent. The yellow
fractions were combined and evaporated, and the residue was recrys-
tallized from n-hexane at 40 °C to give the product as yellow crystals
(2.71 g, 12.09 mmol, 74%). 1H NMR (400 MHz, CDCl3): [δ, ppm] =
1.27 (t, 3H, CH3), 1.66 (s, 6H, Cꢀ(CH3)2), 3.23 (q, 2H, CH2). 13C
NMR (100 MHz, CDCl3): [δ, ppm] = 11.9 (CH3), 24.2 (Cꢀ(CH3)2),
30.3 (CH2), 54.6 (Cꢀ(CH3)2), 177.9 (CdO), 219.6 (CdS). ESI-MS:
[M + Na+]exp = 247.09 m/z and [M + Na+]calc = 246.990 m/z.
In the current contribution we thus describe the first aqueous
RAFT-mediated polymerization of hydrophilic monomers em-
ploying a supramolecular cyclodextrin/CTA host/guest complex
utilizing 4-tert-butylphenyl-substituted CTAs. The presented
approach is the first methodology that leads to hydrophilic
polymers with hydrophobic end groups in one step via aqueous
RAFT polymerization. High molecular weights and conversions
were reached at 25 °C with good control over polydispersity and
molecular weight as determined via N,N-dimethylacetamide
(DMAc) size exclusion chromatography. Furthermore, we de-
scribe the first—to the best of our knowledge—living radical
polymerization of DEAAm in aqueous solution. The structure of
the synthesized polymers was confirmed via electrospray ioniza-
Synthesis of 4-(tert-Butyl)phenyl 2-(((Ethylthio)carbono-
thioyl)thio)-2-methylpropanoate (CTA1). In a 50 mL Schlenk
flask 2-(((ethylthio)carbonothioyl)thio)-2-methylpropanoic acid (1.02 g,
4.55 mmol, 1.0 equiv), 4-tert-butylphenol (1.71 g, 11.38 mmol, 2.5
equiv), and DMAP (0.22 g, 1.80 mmol, 0.4 equiv) were dissolved in
anhydrous DCM (20 mL). At 0 °C, a solution of DCC (1.90 g,
9.21 mmol, 2.0 equiv) in anhydrous DCM (12 mL) was added. After
1 h the solution was warmed to ambient temperature, stirred overnight,
filtered, and concentrated under reduced pressure. The residual yellow
oil was purified via column chromatography on silica gel with n-hexane:
ethyl acetate 20:1 as eluent. The product was obtained as yellow oil
1
tion-mass spectrometry (ESI-MS) and H NMR spectroscopy.
The living character of the polymer chains was proven via chain
extension experiments and the recorded evolution of the full
molecular weight distribution with conversion. In addition,
several methods for the postpolymerization removal of the cyclo-
dextrins were studied.
1
which solidified upon cooling (1.54 g, 4.32 mmol, 95%). H NMR
(400 MHz, CDCl3): [δ, ppm] = 1.30 (s, 9H, (CH3)3), 1.34 (t, 3H, J =
7.4 Hz, CH2ꢀCH3), 1.83 (s, 6H, Cꢀ(CH3)2), 3.32 (q, 2H, J = 7.4 Hz,
CH2), 7.00 (d, 2H, J = 8.8 Hz, CHꢀCꢀO), 7.37 (d, 2H, J = 8.8 Hz,
CHꢀCꢀC(CH3)3). 13C NMR (100 MHz, CDCl3): [δ, ppm] = 12.9
(CH3), 25.4 ((CH3)2), 31.3 ((CH3)3), 31.4 (CH2), 34.5 (CꢀCH3)3),
55.8 (Cꢀ(CH3)2), 120.7 (CHꢀCꢀO), 126.2 (CHꢀCꢀCꢀ(CH3)3),
148.7 (CHꢀCꢀO; CꢀC(CH3)3), 171.8 (CdO), 221.1 (CdS). ESI-
MS: [M + Na+]exp = 379.11 m/z and [M + Na+]calc = 379.039 m/z.
Synthesis of Bis(4-tert-butyl)benzyl) Carbonotrithioate
(CTA2). In a 50 mL round-bottom flask, 4-tert-butylbenzylmercaptan
(1.0 mL, 5.36 mmol, 1.0 equiv) was dissolved in a suspension of
’ EXPERIMENTAL PART
Materials. 2-Bromoisobutyric acid (Sigma-Aldrich, 98%), 3-bromo-
propionyl chloride (ABCR, 90%), 4-tert-butylbenzyl bromide (Acros,
97%), 4-tert-butylphenol (Sigma-Aldrich, 99%), 4-tert-butylaniline (Acros,
99%), 4-tert-butylbenzylmercaptan(Sigma-Aldrich, 99%), acetone (VWR,
normapur), benzylmercaptan (Merck, synth. grade), carbon disulfide
(Acros, 99.9%), dichloromethane (DCM, Acros extra dry over mole-
cular sieves), ethanethiol (Acros, 99%), ethyl acetate (VWR, normapur),
K3PO4 H2O (1.39 g, 6.02 mmol, 1.1 equiv) in acetone (20 mL) at
3
ambient temperature. After stirring for 10 min at ambient temperature
carbon disulfide (1.0 mL, 16.56 mmol, 3.1 equiv) was added, and the
solution turned yellow. 4-tert-Butylbenzyl bromide (1.0 mL, 5.44 mmol,
1.0 equiv) was added after 10 min, and the mixture stirred at ambient
temperature overnight. The mixture was filtered, and the filtrate was
concentrated under reduced pressure. The yellow oily residue was puri-
fied via column chromatography on silica gel with n-hexane as eluent. A
yellow oil was obtained which solidified upon cooling (1.82 g, 4.52
mmol, 84%). 1H NMR (400 MHz, CDCl3): [δ, ppm] = 1.30 (s, 18H,
Cꢀ(CH3)3), 4.59 (s, 4H, CH2ꢀS), 7.25ꢀ7.29 (m, 4H, CH), 7.31ꢀ7.36
(m, 4H, CH). 13C NMR (100 MHz, CDCl3): [δ, ppm] = 31.3
(CH3), 34.6 (Cꢀ(CH3)3), 41.3 (CH2), 125.7 (CH), 129.0(CH),
131.8 (CꢀCꢀ(CH3)3), 150.8 (CꢀCH2), 223.2 (CdS). ESI-MS:
[M + Na+]exp = 424.99 m/z and [M + Na+]calc = 425.141 m/z.
hydroquinone (Fluka, 99%), K3PO4 H2O (Sigma-Aldrich, puriss.),
3
randomly methylated β-cyclodextrin (Me-β-CD, average methylation
grade 1.8 per glucose unit, pharmaceutical grade was a gift from Wacker),
n-hexane (VWR, normapur), N,N0-dicyclohexylcarbodiimide (DCC,
ABCR, 99%), N,N-(dimethylamino)pyridine (DMAP, Sigma-Aldrich,
99%), silica gel (Merck, Geduran SI60. 0.063ꢀ0.200 mm), Taka-
Diastase from Aspergillus oryzae (Sigma-Aldrich, 126 u mgꢀ1), tetra-
hydrofuran (THF, Acros extra dry over molecular sieves), triethylamine
(Acros, 99%), and trifluoroacetic acid (ABCR, 99%) were used as
received. 2,20-Azobis[2-(2-imidazolin-2-yl)propane] dihydrochloride
(VA-044, Wako, 99%) was recrystallized twice from methanol. N,N-
diethylacrylamide (DEAAm, TCI, 98%), N,N-dimethylacrylamide
(DMAAm, TCI, 99%), and 1,4-dioxane (VWR, HPLC-grade) were
passed over a short column of basic alumina prior to use. N-isopropy-
lacrylamide (NIPAAm, Acros, 99%) was recrystallized twice from n-
hexane. Acetic acid/acetate buffer had a pH of 5.2 with an acetic acid
(Roth, 99%) concentration of 0.27 mol Lꢀ1 and a sodium acetate (Roth,
Synthesis of3-Bromo-N-(4-(tert-butyl)phenyl)propanamide.
In a 100 mL Schlenk flask 4-tert-butylaniline (1.5 mL, 9.42 mmol, 1.0 equiv)
and triethylamine (1.9 mL, 13.60 mmol, 1.4 equiv) were dissolved in
anhydrous THF (30 mL). At 0 °C, 3-bromopropionyl chloride (1.3 mL,
13.14 mmol, 1.4 equiv) in anhydrous THF (15 mL) was added dropwise
and stirred at ambient temperature overnight. Saturated NaHCO3 solution
(180 mL) was added and extracted with DCM (2 ꢁ 180 mL). The com-
bined organic extracts were washed with deionized H2O (180 mL) and
brine (180 mL), dried over Na2SO4, filtered, and concentrated under
reduced pressure. The solid residue was recrystallized twice from n-hexane:
ethyl acetate 5:1 to give the product as pale yellow crystals (1.37 g, 4.83
99%) concentration of 0.73 mol Lꢀ1
.
Synthesis of 2-(((Ethylthio)carbonothioyl)thio)-2-methyl-
propanoic Acid (EMP). In a 100 mL round-bottom flask ethanethiol
(1.4 mL, 18.91 mmol, 1.2 equiv) was dissolved in a suspension of
K3PO4 H2O (4.27 g, 18.57 mmol, 1.1 equiv) in acetone (60 mL) at
3
ambient temperature. After stirring for 20 min at ambient temperature
carbon disulfide (3.0 mL, 49.69 mmol, 3.0 equiv) was added, and the
solution turned yellow. 2-Bromoisobutyric acid (2.74 g, 16.41 mmol, 1.0
equiv) was added after 20 min, and the mixture was stirred at ambient
temperature overnight. HCl (200 mL, 1 mol Lꢀ1) was added, and the
aqueous phase was extracted with DCM (2 ꢁ 150 mL). The combined
organic extracts were washed with deionized H2O (75 mL) and brine
(75 mL), dried over Na2SO4, and filtered. After evaporation of the
solvent the yellow oily residue was purified via column chromatography
1
mmol, 51%). H NMR (400 MHz, CDCl3): [δ, ppm] = 1.30 (s, 9H,
(CH3)3), 2.92 (t, 2H, J = 6.6 Hz, CH2ꢀCdO), 3.71 (t, 2H, J = 6.6 Hz,
CH2Br), 7.34, (d, 2H, J = 8.5 Hz, CHꢀCꢀ(CH3)3), 7.38 (NH), 7.44 (d,
2H, J = 8.5 Hz, CHꢀCNH). 13C NMR (100 MHz, CDCl3): [δ, ppm] =
27.2 (CH2ꢀBr), 31.3 (Cꢀ(CH3)3), 34.4 (Cꢀ(CH3)3), 40.7
(CH2ꢀCdO), 119.9 (CHꢀCꢀNH), 125.9 (CHꢀCꢀCꢀ(CH3)3),
7222
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
134.8 (CꢀNH), 147.8 (CꢀCꢀ(CH3)3), 167.9 (CdO). ESI-MS: [M +
Na+]exp = 306.12 m/z and [M + Na+]calc = 307.182 m/z.
Synthesis of Benzyl (3-((4-(tert-butyl)phenyl)amino)-3-
oxopropyl) Carbonotrithioate (CTA3). In a 50 mL round-bot-
tom-flask benzylmercaptan (498 μL, 4.23 mmol, 1.0 equiv) was dis-
equiv), VA-044 (0.3 mg, 0.001 mmol, 0.5 equiv), and deionized H2O
(0.6 mL) were added into a Schlenk tube. After three freezeꢀ
pumpꢀthaw cycles the tube was placed into an oil bath at 25 °C and
kept there for 24 h. The reaction mixture was cooled with liquid nitrogen,
subjected to air, and dialyzed for 3 days at ambient temperature.
The solvent was removed in vacuo to give the polymer as a yellow solid
(197 mg, 97%, GPC(DMAc): Mn,GPC = 97 500 g molꢀ1, PDI = 1.08).
Exemplary Procedure for Kinetic Measurements (Refer to
the Supporting Information for the Kinetic Measurements
for DEAAm and NIPAAm). CTA1 (12.2 mg, 0.034 mmol, 1.0 equiv)
and Me-β-CD solution (40 wt % in deionized water, 551 mg,
0.168 mmol, 4.9 equiv) were added to a Schlenk tube. The two-phase
mixture was ultrasonicated until a clear yellow solution was obtained.
Subsequently, DMAAm (1.00 g, 10.10 mmol, 297.1 equiv), VA-044
(2.2 mg, 0.007 mmol, 0.2 equiv), and deionized H2O (2.7 mL) were
added. The solution was separated into several tubes with a stirring bar,
and three freezeꢀpumpꢀthaw cycles were applied. Subsequently, the
tubes were sealed and placed into an oil bath at 25 °C. After specific time
intervals the tubes were cooled with liquid nitrogen. Samples for NMR
analysis were withdrawn, a pinch of hydroquinone (∼5 mg) was added
to inhibit further polymerization, and D2O was added. The conversion
was calculated via comparison of the vinyl proton integrals with the
appropriate integrals of the backbone or side chains of the polymers
(see Characterization Methods for details). The residual sample in each
tube was purified by dialysis for 3 days at ambient temperature and 1 day
at 45 °C. The solvent was removed in vacuo and the polymer subjected
to SEC analysis.
solved in a suspension of K3PO4 H2O (1.08 g, 4.69 mmol, 1.1 equiv) in
3
25 mL of acetone at ambient temperature. After stirring for 10 min at
ambient temperature carbon disulfide (766 μL, 12.69 mmol, 3.0 equiv)
was added, and the solution turned yellow. 3-Bromo-N-(4-(tert-butyl)-
phenyl)propanamide (1.20 g, 4.23 mmol, 1.0 equiv) was added after
10 min, and the mixture was stirred at ambient temperature overnight.
HCl (160 mL, 1 mol Lꢀ1) was added and extracted twice with DCM
(2 ꢁ 160 mL). The combined organic extracts were washed with
deionized H2O (160 mL) and brine (160 mL), dried over Na2SO4,
filtered, and concentrated under reduced pressure. The solid residue was
recrystallized from n-hexane:ethyl acetate 1:1 to give the product as a
yellow solid in two fractions (1.21 g, 3.00 mmol, 71%). 1H NMR (400
MHz, CDCl3): [δ, ppm] = 1.30 (s, 9H, (CH3)3), 1.60 (s, 1H, NH), 2.79
(t, 2H, J = 7.0 Hz, CdOꢀCH2), 3.73 (t, 2H, J = 7.0 Hz, SꢀCH2ꢀCH2),
4.61 (s, 2H, CH2), 7.25ꢀ7.45 (m, 9H, Harom). 13C NMR (100 MHz,
CDCl3): [δ, ppm] = 31.3 (Cꢀ(CH3)3), 32.0 (CH2ꢀCdO), 34.4 (Cꢀ
(CH3)3), 36.1 (CH2ꢀCH2ꢀS), 41.5 (CH2), 119.8 (CHꢀCꢀNH),
125.9 (CHꢀCꢀC(CH3)3), 127.8 (CH), 128.7 (CHꢀCHꢀCꢀCH2),
129.3 (CHꢀCꢀCH2), 134.8, 134.9 (CꢀNH, CꢀCH2), 147.6 (CꢀCꢀ
(CH3)3), 168.5 (CdO), 223.7 (CdS). ESI-MS: [M + Na+]exp
426.27 m/z and [M + Na+]calc = 426.100 m/z.
=
Exemplary Procedure for the Polymerization of DMAAm
(Refer to the Supporting Information for the Polymerization
Procedure of DEAAm and NIPAAm). CTA1 (4.9 mg, 0.014 mmol,
1.0 equiv) and Me-β-CD solution (40 wt % in deionized water, 228 mg,
0.070 mmol, 5.0 equiv) were added into a Schlenk tube. The two-phase
mixture was ultrasonicated until a clear yellow solution was obtained.
Subsequently, a stirring bar, DMAAm (274 mg, 2.77 mmol, 198.9
equiv), VA-044 (1.0 mg, 0.003 mmol, 0.2 equiv), and deionized H2O
(0.8 mL) were added. After three freezeꢀpumpꢀthaw cycles the tube
was sealed and placed into an oil bath at 25 °C and removed after 6 h.
The tube was subsequently cooled with liquid nitrogen to stop the
reaction. An NMR sample was withdrawn for the determination of
conversion, inhibited with a pinch of hydroquinone (∼5 mg) and D2O
was added. A conversion of 88% was calculated based on the NMR data
(see the Characterization Methods section for details of the calculation).
The residue was dialyzed with a SpectraPor3 membrane (MWCO =
3500 Da) for 3 days at ambient temperature and for 2 days at 45 °C. The
solvent was removed in vacuo to yield the polymer as a yellow solid
(96 mg, 39%, GPC(DMAc): Mn,GPC = 17 000 g molꢀ1, PDI = 1.09).
In the case of short-chain polymers (with targeted Mn below 3500 g
molꢀ1) the reaction mixture was dialyzed with a SpectraPor3 membrane
(MWCO = 1000 Da) for 3 days at ambient temperature. The polymer
sample was subsequently diluted with acetic acid/acetate buffer, the
R-amylase containing enzyme mixture Taka-Diastase was added to de-
grade the residual cyclodextrins, and the mixture was incubated at 37 °C
for 24 h and boiled for 10 min.46,47 Finally, the mixture was dialyzed
against water for 3 days at ambient temperature, and the solvent was
removed in vacuo.
Characterization Methods. NMR measurements were con-
ducted on a Bruker AM250 spectrometer at 250 MHz for hydrogen
nuclei for kinetic measurements, a Bruker Avance III 300 spectrometer
at 300 MHz for hydrogen nuclei for kinetic measurements, and a Bruker
AM400 spectrometer at 400 MHz for hydrogen nuclei and at 100 MHz
for carbon nuclei for structure verification. 2D ROESY (rotating frame
nuclear Overhauser effect spectroscopy) NMR spectra were measured
on a Bruker Avance III 300 spectrometer at 300 MHz. For the deter-
mination of the conversion of DMAAm the integrals of one vinylic
proton (5.78ꢀ5.89 ppm) and the methyl side chain protons (2.87ꢀ
3.28 ppm) were employed (refer to the Supporting Information Figure
S11). The conversion of DEAAm was determined with the integral of
one vinylic proton (5.57ꢀ5.73 ppm) and with the integral of the side
chain methyl groups and backbone protons (0.81ꢀ1.97 ppm) (see also
Figure S17 in the Supporting Information). The calculation of the
NIPAAm conversion was carried out with the integrals of one vinylic
proton (5.47ꢀ5.59 ppm) and the integral of the side chain methyl
groups and the backbone protons (0.92ꢀ1.95 ppm) (refer to the
Supporting Information Figure S23).
Size exclusion chromatography (SEC) was performed on a Polymer
Laboratories PL-GPC 50 Plus Integrated System, comprising an auto-
sampler, a PLgel 5 μm bead-size guard column (50 ꢁ 7.5 mm) followed
by three PLgel 5 μm MixedC columns (300 ꢁ 7.5 mm), and a dif-
ferential refractive index detector using N,N-dimethylacetamide (DMAc)
containing 0.3 wt % LiBr as eluent at 50 °C with a flow rate of 1.0 mL
minꢀ1. The SEC system was calibrated against linear poly(styrene)
standards with molecular weights ranging from 160 to 6 ꢁ 106 g molꢀ1
.
For the other polymerizations the Me-β-CD/CTA/initiator ratio was
kept constant at 5/1/0.2. The DMAAm/CTA ratio was altered, and
water was added to keep the concentration constant at 3.0 mol Lꢀ1. The
experiments with CTA2 and CTA3 were conducted in an analogous
fashion.
Exemplary Procedure for the Chain Extension of Poly-
(DMAAm) (Refer to the Supporting Information for the
Chain Extension of Poly(DEAAm) and Poly(NIPAAm)). Poly-
(DMAAm) (Mn,GPC = 10 500 g molꢀ1, PDI = 1.17, 20.0 mg, 0.002
mmol, 1.0 equiv) as macro-CTA, DMAAm (183 mg, 1.85 mmol, 925.0
All SEC calculations were carried out relative to poly(styrene) calibra-
tion (MarkꢀHouwink parameters K = 14.1 ꢁ 10ꢀ5 dL gꢀ1; R = 0.7).48
ESI-MS spectra were recorded on a LXQmassspectrometer(Thermo-
Fisher Scientific, San Jose, CA) equipped with an atmospheric pressure
ionization source operating in the nebulizer-assisted electrospray mode.
The instrument was calibrated in the m/z range 195ꢀ1822 Da using a
standard containing caffeine, Met-Arg-Phe-Ala acetate (MRFA), and a
mixture offluorinatedphosphazenes (Ultramark 1621) (all from Aldrich).
A constant spray voltage of 4.5 kV was used, and nitrogen at a dimen-
sionless sweep gas flow rate of 2 (∼3 L minꢀ1) and a dimensionless
7223
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
Scheme 2. Synthetic Routes to the Utilized Complexable CTAs
sheath gas flow rate of 12 (∼1 L minꢀ1) were applied. The capillary
voltage, the tube lens offset voltage, and the capillary temperature were
set to 60 V, 110 V, and 275 °C, respectively.
synthesis of CTA1). An alternative route was the synthesis of
a guest-functionalized molecule containing a bromine leaving
group (e.g., CTA3). The direct route utilized guest-functionalized
thiols and bromides (e.g., CTA2). The synthesized CTAs were
characterized via NMR spectroscopy (refer to the Supporting
Information Figures S1ꢀS6) and ESI-MS.
Theoretical molecular weights were calculated with the following
equation:
½monomerꢂ0
Complexation of Chain Transfer Agents with Me-β-Cyclo-
dextrin. For the utilization of the guest-functionalized CTAs in
aqueous RAFT polymerizations host/guest complexes have to
be formed with cyclodextrin. Me-β-CD was chosen as host
compound due to its increased water solubility compared to
β-cyclodextrin.51 The complexation was accomplished via mix-
ing the guest-functionalized CTAs with aqueous 40 wt % Me-
β-CD solution and ultrasonication until a clear yellow solution
was obtained. Depending on the structure of the CTA the
suspension had to be ultrasonicated for variable times, as the
ultrasonication time depends on the complex stability and the
possibility to disperse the CTA in the aqueous solution.
Mn;theo ¼ conversion ꢁ MwðmonomerÞ ꢁ
þ MwðCTAÞ
½CTAꢂ0
’ RESULTS AND DISCUSSION
Our approach for the utilization of hydrophobic CTAs in
aqueous RAFT polymerizations via cyclodextrin inclusion com-
plexes includes the synthesis of novel hydrophobic CTAs that
are subjected to complex formation with randomly methylated
cyclodextrin (Me-β-CD) and are subsequently used in the RAFT
polymerization of DMAAm, DEAAm, and NIPAAm.
Figure 1 shows CTA/Me-β-CD solutions and as control
CTA/water solutions before ultrasonication and after ultrasoni-
cation at ambient temperature. From the yellow color in the
solution after ultrasonication it is obvious that the CTA/Me-
β-CD complex was formed. In contrast, the solution in the
control experiment shows no yellow color. To ensure complete
inclusion, a 5-fold excess was used for all CTAs. For CTA1 a
homogeneous solution was obtained after 35 min; for CTA2 and
CTA3 a homogeneous solution was obtained after 100 min of
ultrasonication.
A further method to characterize the CTA/Me-β-CD com-
plexes is the two-dimensional ROESY (rotating frame nuclear
Overhauser effect spectroscopy) NMR technique.3,52 In general,
there are two modes for the formation of the inclusion complex.
The guest can insert into the hydrophobic cavity via the primary,
i.e., the side with the smaller opening, or the secondary face of the
cyclodextrin, i.e., the side with the bigger opening. From the
resonances in the ROESY spectrum it is in principle possible to
assign the formed type of the complex. The complexation of
CTA1 with Me-β-CD could be evidenced via the resonance of
the tert-butyl protons at 1.33 ppm with the inner protons of Me-
β-CD (signal between 3.27 and 3.87 ppm) in D2O at 25 °C as
depicted in Figure 2. Furthermore, the signal of the tert-butyl
protons is shifted from 1.30 to 1.33 ppm, thus changing place
Design and Synthesis of the Chain Transfer Agents.
Several trithiocarbonate-CTAs for the aqueous RAFT-polymer-
ization of hydrophilic monomers are mentioned in the litera-
ture.23,30,45,49 As R-group usually the benzyl or a tertiary-
R-carbonyl group areemployed. The synthesisoftrithiocarbonates
can be carried out via the deprotonation of thiols, their nucleo-
philic attack on carbon disulfide, and nucleophilic attack of the
resulting trithiocarbonate salt on bromo compounds. O’Reilly
and co-workers recently presented an elegant way to synthesize
trithiocarbonate-CTAs utilizing potassium phosphate as base in
acetone.50 This synthetic route was employed for all of the CTAs
described in the current work. As guest group the 4-tert-butyl-
phenyl motif was chosen due to its high complexation constant
with β-cyclodextrin (K ≈ 18 000ꢀ25 000 L molꢀ1).40 From an
earlier publication of Wenz and co-workers a similar complexa-
tion constant of Me-β-CD with the 4-tert-butylphenyl group can
be anticipated from a close analogue (heptakis(2,6-di-O-methyl)-
β-cyclodextrin).41
As depicted in Scheme 2, the synthesis of the guest-function-
alized CTAs was accomplished either directly or in two stages to
include guest groups for β-cyclodextrin complexes into the CTAs.
The guest group was incorporated in the R group (CTA1), Z
group (CTA3), or both (CTA2). One possibility was the use
of DCC-coupling after the synthesis of a precursor CTA (e.g.,
7224
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
Figure 1. Mixtures of CTA1 (a, b), CTA2 (c, d), and CTA3 (e, f) with aqueous 40 wt % Me-β-CD solution (left) and deionized water (right). Top:
before ultrasonication, bottom: after ultrasonication.
Figure 2. 2D ROESY NMR spectrum of a 1:1 molar mixture of CTA1 and Me-β-CD in D2O at 25 °C and a schematic illustration of the hostꢀguest
complex.
with the signal of the methyl protons from the ethyl group (refer
to Figure S1 in the Supporting Information for a H NMR
spectrum of CTA1). Interestingly, the signal for the R-methyl
protons of CTA1 splits into two distinct signals due to chemical
inequivalency after complex formation. This signal shows reso-
nance with the C2-methoxy group which is due to a complex with
an insertion from the secondary side as shown in Figure 2.
Nevertheless, the unspecific resonances between the tert-butyl
1
7225
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
Figure 3. (a) Comparison between the kinetic plots of EMP-mediated polymerization in acetic acid/acetate buffer, EMP-mediated polymerization in
water, or CTA1/cyclodextrin-mediated polymerization at 25 °C with a DMAAm/CTA/I ratio of 300/1/0.2 and a DMAAm concentration of 3.5 mol
L
ꢀ1. (b) Comparison of the obtained molecular weights and PDIs as a function of monomer to polymer conversion.
groups with the C2-methoxy and the C6-methoxy group suggest
a mixture of both inclusion modes.
Second, monomer, water, and initiator were added, the reaction
was degassed, and the polymerization subsequently commenced.
Third, the polymerization mixture was subjected to dialysis to
remove residual monomer, initiator and Me-β-CD.
The 2D ROESY spectra of CTA2 and CTA3 show the
interaction between the CTA and Me-β-CD. The direction of
the inclusion is not clearly assignable, although weak interaction
of the aromatic protons with the C2-methoxy group indicates a
complexation via the secondary face of Me-β-CD (refer to the
Supporting Information Figures S7 and S8). For CTA3 the
resonances of the aromatic protons with the C6-methoxy and
C2-methoxy protons indicate a complexation on both ends.
A comparison between the different CTAs shows several
effects that have an influence on the complex stability upon mono-
mer addition. One matter is the nature of the monomer, e.g.,
hydrophobicity and its ability to act as a guest, which is discussed
below. An additional point is the hydrophobicity of the CTA with
respect to the hydrophobicity of weak guest groups, e.g., dodecyl,
hexyl, or butyl groups.53,54 Although other CTAs exhibit the
possibility to form inclusion complexes with Me-β-CD, e.g., the
analogue of CTA1 with dodecyl instead of ethyl group, addition
of hydrophilic monomers leads to complete demixing due to the
loss of the inclusion complex. Most likely, the hydrophobicity of
the dodecyl group is too weakly masked by a β-cyclodextrin.
Therefore, the complex is lost as competing guest molecules, e.g.
monomers, are added, and the single host/guest complex with
the 4-tert-butyl phenyl group is not strong enough to keep the
whole CTA in solution. It is obvious that the stability of the host/
guest complex is additionally correlated with the number of guest
groups incorporated into the CTA. The more guest groups are
incorporated, the more stable is the complex. As the host/guest
complexes are in equilibrium with the free molecules, it is
advantageous to have two complexed groups in one CTA
molecule. If one of the two host/guest complexes is lost, the
remaining one can keep the entire molecule in solution and
accessible for the recreation of the second complex.
As discussed by McCormick and co-workers, a major concern
in aqueous RAFT polymerization of acrylamides is the hydro-
lysis/aminolysis of the CTA during the polymerization.22,55,56
A solution for hydrolysis suppression/prevention is the use of
acetic acid/acetate buffer as solvent and low-temperature initia-
tors, e.g. VA-044.49,55 Under these conditions, McCormick and
colleagues showed that a controlled polymerization of acryla-
mide, DMAAm, and NIPAAm via the RAFT process in aqueous
media is possible.31,32,49 Furthermore, it is well-known that low
temperatures favor the stability of β-CD complexes. For these
reasons all polymerizations wereconducted at 25 °C. As the CTA/
Me-β-CD complexes were not soluble in acetic acid/acetate
buffer, a series of DMAAm polymerizations—as a test system—
were conducted in variable reaction media with CTA1 or EMP to
determine the effect of the reaction media on the polymerization.
The results are summarized in Figure 3.
Figure 3a shows that the polymerization reaction with the
complexed CTA1 (triangles) proceeds slower than the polymer-
izations with EMP in acetic acid/acetate buffer (filled squares) or
water (open circles), whereas the EMP mediated polymeriza-
tions show comparable reaction rates. Nevertheless, the reaction
with CTA1 leads to a conversion of 70% in 6 h while the reactions
with EMP have a conversion of approximately 80% within 6 h.
We propose that the difference in reaction rate is a result of the
increasing steric hindrance associated with the bulky cyclodextrin
complex. Although a slower reaction is observed a superior con-
trol over the polymerization with the CTA1/cyclodextrin com-
plex can be noted. Lower PDIs were observed in these poly-
merizations as summarized in Figure 3b. In the case of CTA1
(triangles) the observed molecular weight was higher than the
theoretical molecular weight, whereas the molecular weight is
lower than the theoretical molecular weight in the case of EMP
(squares and open circles). As the polymerization has living/
controlled character in both acetic acid/acetate buffer and pure
Polymerization with Complexed Chain Transfer Agents.
For the polymerization of acrylamido monomers with Me-β-CD
complexed CTA three steps were carried out as depicted in
Scheme 1: First, the complex was formed via ultrasonication of
the appropriate CTA in aqueous 40 wt % Me-β-CD solution.
7226
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
Table 1. Results for the Different Purification Methods
Table 2. Results for the Living/Controlled RAFT Polymeri-
zation of DMAAm at 25 °C in Aqueous Solution with CTA1
DMAAm/CTA/I time/h conv Mn,theo/g molꢀ1 Mn,GPC/g molꢀ1 PDI
residual Me-β-CD
[% area]
entry
purification procedure
no purification
107/1/0.2
198/1/0.2
294/1/0.2
585/1/0.2
1018/1/0.2
6
6
65
88
7 300
18 000
29 500
55 000
99 300
10 500
17 000
36 500
62 000
94 000
1.17
1.09
1.08
1.06
1.06
1
2
29.1
7.5
CF3COOH 2 h; dialysis 3 days,
ambient temperature
12
12
12
>99
94
3
4
dialysis 3 days, ambient temperature
dialysis 3 days, ambient temperature;
dialysis 1 day, 45 °C
8.4
5.4
98
5
6
dialysis 3 days, ambient temperature;
dialysis 2 days, 45 °C
dialysis 3 days, ambient temperature;
Taka-Diastase acetate buffer 1 day, 37 °C;
dialysis 3 days, ambient temperature
3.3
4.1
with CTA1 compared to CTA2 and CTA3 (see Table 2 and
Tables S2, S3 in the Supporting Information). This can be
attributed mostly to the complex stability and the tertiary R-ester
R group in CTA1. A monomer concentration of 3.0 mol Lꢀ1 was
chosen and the CTA/initiator ratio was held constant at 1/0.2.
The complex of CTA1 with Me-β-CD was stable throughout the
reaction time and even at high monomer/CTA ratios up to
1000/1. Therefore, molecular weights ranging from 10 000 to
94 000 g molꢀ1 were obtained in good agreement with theore-
tical values. High conversion was reached in short reaction time
even at the low polymerization temperature of 25 °C. The result-
ing PDIs lie between 1.06 and 1.17 (see Table 2 and Figure 4d).
Although the polymers were purified according to the Experi-
mental Part, a small residue of unremoved cyclodextrin (∼0.4ꢀ
5.5%) remained. Apart from residual cyclodextrin, the SEC traces
are unimodal and show only minor low molecular weight tailing
and no high molecular weight coupling products.
Time-resolved experiments with regard to conversion and
molecular weight were carried out to confirm the living character
of the polymerization (refer to Figure S10 in the Supporting
Information for a collection of NMR spectra). As depicted in
Figure 4, the kinetic first-order plot shows linearity, which con-
firms a constant radical concentration during the reaction; only a
short induction period is observed (<30 min). The molecular
weights are increasing linearly with conversion which evidence
the living character of the polymerization, and the PDI is de-
creasing with increasing conversion. Further proof for the living
radical polymerization comes from the chain extension of puri-
fied macro-CTAs with DMAAm. A quantitative reinitiation was
observed that leads to a shift in molecular weight from 10 500 to
97 500 g molꢀ1 with a final PDI of 1.08. Nevertheless, small
amounts of chainꢀchain coupling products were observed in the
high molecular weight region.
water with EMP as CTA at 25 °C, there is no need to carry out the
polymerization at acidic pH in the case of complexed CTAs.
Removal of the Cyclodextrin after the Polymerization. For
the removal of the cyclodextrin several methods were applied. A
test polymerization with DMAAm and CTA1 was carried out,
divided into several samples, and purified in different ways. To
compare the residual amounts of the cyclodextrin in the pro-
ducts, the peak area of the eluting cyclodextrin in the SEC ana-
lysis was calculated relatively to the peak area of the polymer.
Dialysis was performed utilizing dialysis tubing with a MWCO of
3500 Da, but the treatment of the samples was varied. One
purification method was the treatment of the crude polymeriza-
tion solution with trifluoroacetic acid for 2 h as cyclodextrins are
hydrolyzed by strong acids, and the solution was dialyzed for
3 days at ambient temperature afterward (entry 2 in Table 1). For
comparison, the crude polymerization mixture was dialyzed for
3 days at ambient temperature (entry 3 in Table 1). Another
method was dialysis for 3 days at ambient temperature and
subsequently dialysis for 1 day or 2 days at 45 °C (entries 4 and
5 in Table 1). Furthermore, enzymatic treatment with Taka-
Diastase in acetic acid/acetate buffer at 37 °C for 1 day was em-
ployed after dialysis for 3 days at ambient temperature and
another dialysis for 3 days at ambient temperature after the enzy-
matic treatment (entry 6 in Table 1). The enzymatic treatment
with Taka-Diastase from Aspergillus oryzae should lead to degra-
dation of the cyclodextrins as it contains the enzyme R-amylase
which is known to hydrolyze the R-1,4-glucosidic bond of the
cyclodextrins.46,47,57
As listed in Table 1, dialysis at elevated temperatures, e.g.
45 °C, provides the best results with only 3.3% Me-β-CD remain-
ing (entry 5 in Table 1). Nevertheless, enzymatic treatment has a
very similar performance with 4.1% Me-β-CD remaining (entry
6 in Table 1; the corresponding elugrams can be found in the
Supporting Information Figure S9). In the case of poly(DEAAm)
and poly(NIPAAm), dialysis at elevated temperatures leads to
the precipitation of the polymers that supports the removal of
residual cyclodextrin. Generally, it should be noted that dialysis
leads to a loss of low molecular weight polymers and oligomers.
Especially in the case of low target molecular weights, the mole-
cular weight distributions are thus to some extent affected.
Polymerization of N,N-Dimethylacrylamide with Com-
plexed Chain Transfer Agents. DMAAm is very frequently
employed in polymer science. Inthe living/controlled radical poly-
merization with complexed CTAs the best control is obtained
The molecular structure of low molecular weight samples was
1
confirmed via ESI-MS (Mn = 3000 g molꢀ1) and H NMR
(Mn = 10 500 g molꢀ1) (see Figures S12 and S13 in the
Supporting Information), evidencing the incorporation of the
hydrophobic end groups into the hydrophilic polymer. The mass
spectrometric data show no signals from initiator-terminated
polymer which is in accord with highly efficient chain-extension
experiments.
With CTA2 turbidity was observed on monomer addition that
vanished in the beginning of the polymerization. In the poly-
merizations with CTA3 turbidity was observed at monomer/
CTA ratios exceeding 300/1. Nevertheless, molecular weights of
156 000 g molꢀ1 were reached with PDIs ranging from 1.31 to
1.54 (refer to the Supporting Information: Tables S2, S3 and
Figures S14, S15) with high conversions in short reaction times.
As stated above, less control over the polymerization is observed
with CTA2 and CTA3. The less stable complexes of Me-β-CD
with CTA2 and CTA3 lead to partial demixing on monomer and
7227
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
Figure 4. (a) Kinetic plot for the polymerization of DMAAm at 25 °C with CTA1. (b) Evolution of Mn with conversion at 25 °C with DMAAm/CTA1/
I: 297/1/0.2. (c) Chain extension at 25 °C for 24 h. (d) Molecular weight distributions for different monomer/CTA ratios (DMAAm/CTA1
(conversion) from left to right: 107/1 (65%); 198/1 (88%); 294/1 (>99%); 585/1 (94%); 1018/1 (98%)).
water addition as observed via a turbidity of the solution. Such a
demixing explains the higher disagreement of experimental
molecular weights compared to theoretical molecular weights,
as fewer CTA molecules are accessible in the solution in the
case of turbidity, thus leading to higher molecular weights. The
broader polydispersity of the polymers prepared with CTA2 and
CTA3 can be also attributed to the partial demixing as the
undissolved CTA molecules react on a slower time scale. There-
fore, low molecular weight tailing is observed which leads to
higher PDIs. Nevertheless, chain-extension experiments were
conducted that proof the living character of the polymerization
based on a high reinitiation efficiency.
Table 3. Results for the Living/Controlled RAFT Polymeri-
zation of DEAAm at 25 °C in Aqueous Solution with CTA1
DEAAm/CTA/I time/h conv Mn,theo/g molꢀ1 Mn,GPC/g molꢀ1 PDI
12/1/0.2
42/1/0.2
80/1/0.2
163/1/0.2
248/1/0.2
426/1/0.2
813/1/0.2
12
12
12
12
12
18
18
>99
>99
>99
>99
>99
>99
>99
1 900
5 800
2 500
5 000
1.10
1.11
1.09
1.08
1.07
1.06
1.05
10 600
21 100
32 000
54 600
103 800
10 000
18 000
29 000
55 000
87 000
Polymerization of N,N-Diethylacrylamide with Complexed
Chain Transfer Agents. DEAAm is a monomer that exhibits a
very low LCST of around 30 °C.58 At a reaction temperature of
25 °C, it should be possible to polymerize DEAAm in aqueous
media via the RAFT process. Although the living radical polym-
erization in organic media has been described,59 a living/
controlled radical polymerization of this monomer in aqueous
solution has not been accomplished yet. CTA1, CTA2, and CTA3
were employed for the polymerization of DEAAm. The monomer
concentration was held constant at 3.5 mol Lꢀ1, the temperature at
25 °C, and the CTA/initiator ratio at 1/0.2. As with DMAAm,
CTA1 shows the best control over the polymerizations. No
turbidity or demixing is noticed for monomer/CTA ratios up to
200/1, and only a slight turbidity is observed for higher CTA/
monomer ratios in the case of CTA1. Molecular weights from
2500 up to 87 000 g molꢀ1 that were in good agreement with the
theoretical values were reached in short reaction times, e.g., 12 or
18 h, with quantitative conversions and PDIs ranging from 1.05 to
1.11 (see Table 3 and Figure 5d). The resulting molecular weight
distributions are unimodal and display no evidence for high
molecular weight termination products or low molecular weight
7228
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
Figure 5. (a) Kinetic plot for the polymerization of DEAAm at 25 °C with CTA1. (b) Evolution of Mn with conversion at 25 °C with DEAAm/CTA1/I:
241/1/0.2. (c) Chain extension at 25 °C for 24 h. (d) Molecular weight distributions for different monomer/CTA ratios (DEAAm/CTA1 from left to
right (conversion > 99%): 12/1; 42/1; 80/1; 163/1; 248/1; 426/1; 813/1).
Table 4. Results for the Living/Controlled RAFT Polymeri-
zation of NIPAAm at 25 °C in Aqueous Solution with CTA1
NIPAAm/CTA/I time/h conv Mn,theo/g molꢀ1 Mn,GPC/g molꢀ1 PDI
tailing, except for the lowest target molecular weight where low
molecular weight (<1000 g molꢀ1) species are observed.
To confirm the living character of the polymerization, time-
resolved experiments with regard to conversion and molecular
weight were carried out (refer to Figure S11 in the Supporting
Information for a collection of NMR spectra) and chain exten-
sions were performed as depicted in Figure 5. Besides a constant
radical concentration as evidenced by a linear first-order plot with
a short induction period under 30 min, the molecular weight
grows linearly with conversion as expected for polymerizations
with living character. Furthermore, the experimental molecular
weights are in good agreement with the theory, and the PDIs are
decreasing with increasing conversion. The chain extension
affords a very high reinitiation efficiency with a growth in mole-
cular weight from 8500 to 83 000 g molꢀ1 and a PDI of 1.13 for
the resulting polymer with a small amount of chainꢀchain coup-
ling products.
15/1/0.2
39/1/0.2
44/1/0.2
260/1/0.2
18
18
18
20
>99
>99
>99
>99
2100
4800
4000
7500
1.15
1.29
1.15
1.46
5500
9000
30100
50000
there is no indication of initiator derived chains in the ESI-MS
spectrum which explains the high reinitiation efficiency.
For the other CTAs (CTA2 and CTA3) turbid solutions were
observed upon water addition. Nevertheless, high conversions
were reached in short reaction times, and molecular weights
ranging from 4000 to 116 000 g molꢀ1 were reached with PDI
from 1.10 to 1.39 (see also Tables S5 and S6 in the Supporting
Information). Similar to the polymerizations of DMAAm with
CTA2 and CTA3, a disagreement of the experimental molecular
weights with those theoretically predicted was noted that could
be due to the turbid nature of the solution at the beginning of the
polymerization process. It is also worth noting that the polymers
remained in solution throughout the entire reaction time with all
1
ESI-MS and H NMR were recorded of a low molecular
weight sample (Mn = 4000 g molꢀ1) to confirm the structure of
the synthesized polymers. The results are in agreement with the
expected polymer structure proving the incorporation of the
hydrophobic end groups into the hydrophilic polymer (refer to
the Supporting Information: Figures S18 and S19). Furthermore,
7229
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
Figure 6. (a) Kinetic plot for the polymerization of NIPAAm at 25 °C with CTA1. (b) Evolution of Mn with conversion at 25 °C with NIPAAm/CTA1/
I: 44/1/0.2. (c) Chain extension at 25 °C for 24 h. (d) Molecular weight distributions for different monomer/CTA ratios (NIPAAm/CTA1 from left to
right (conversion > 99%): 15/1; 39/1; 44/1; 260/1).
three CTAs although it is known from literature that hydro-
phobic end groups lead to a lower LCST in poly(DEAAm).43
Polymerization of N-Isopropylacrylamide with Com-
plexed Chain Transfer Agents. Another important acrylamido
monomer is NIPAAm. Because of the LCST of poly(NIPAAm)
that is close to human body temperature, many efforts have been
made in its synthesis.25,44,49 As the LCST of poly(NIPAAm) is
close to 31 °C,60 which is comparable to the LCST of poly-
(DEAAm), a controlled polymerization of NIPAAm is feasible at
25 °C in aqueous media.49 Therefore, it should be possible to
polymerize NIPAAm with cyclodextrin-complexed CTAs at
25 °C in aqueous media as well.
CTA/cyclodextrin complex. The obtained polymers show uni-
modal molecular weight distributions with no significant tailing
in the low molecular weight region. No high molecular weight
termination products are visible.
A constant radical concentration was proven by a linear first-
order plot also evidencing a short induction period of ∼30 min. A
confirmation of the living radical polymerization was accom-
plished via a chain-extension experiment and a time-resolved
experiment that indicated a linear increase of molecular weight
with conversion (see Figure 6). Higher experimental molecular
weights compared to the theoretical molecular weights are ob-
served, as discussed above. The chain extension shows that the
amount of dead chains is neglible as chain extension from 9000 to
90000 g molꢀ1 was performed with very high efficiency leading to
a PDI of 1.11 showing only minor chainꢀchain coupling products.
Besides the kinetic studies, ESI-MS was performed to prove
the structure of the obtained polymers. As depicted in Figure
S24, the results fit very well to the expected values, indicating the
incorporation of the hydrophobic end groups into the hydro-
philic polymers. Furthermore, there is no indication of initiator-
derived chains that matches with the observed high reinitiation
In the case of NIPAAm a controlled polymerization was only
possible for CTA1. The other CTAs showed complete demixing
upon addition of the monomer and water, which is further
discussed in the subsequent section. Nevertheless, with CTA1
polymers up to 50 000 g molꢀ1 were synthesized with PDIs
ranging from 1.15 to 1.46 with quantitative conversions in 18 or
20 h. The monomer concentration was held constant at 2.5 mol Lꢀ1
,
the polymerization temperature at 25 °C, and the CTA/initiator
ratio at 1/0.2. The results are summarized in Table 4. A
significant excess of the experimental molecular weights com-
pared to the theoretically predicted ones by a factor of 1.5ꢀ2.0
was evidenced that could be due to a partial disassembly of the
1
efficiency. H NMR was measured (refer to the Supporting
Information Figure S25), indicating the incorporation of the
hydrophobic end group in the aromatic region of the spectrum.
7230
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
Effect of the Monomer Structure on the RAFT Polymeri-
zation with Complexed Chain Transfer Agents. Comparison
of the studied monomers evidences that with increasing guest
character the complexation of the CTAs decreases. With
DMAAm no turbidity was observed for CTA1 and only slight
turbidity for CTA2 and CTA3. This effect is also reflected in the
control over molecular weights and PDIs as the best control is
achieved with CTA1. In the polymerization of DEAAm, only
slight turbidity was observed in the CTA1 solution where
DEAAm/CTA1 ratios exceeded 300/1. With CTA2 and
CTA3 turbidity was observed with DEAAm/CTA ratios from
13/1 to 813/1. In analogy to the polymerization of DMAAm the
CTA1-mediated polymerizations of DEAAm show the best
control over molecular weight and PDI. The overall trend is
continuing in the polymerization with NIPAAm as it could only
be conducted with CTA1 and only short chains up to 50 000 g
molꢀ1 could be synthesized with an increasing PDI toward
longer chains. It appears that the substituents in the acrylamido
monomers have a significant effect on the stability of the CTA/
cyclodextrin complex. DMAAm disturbs the complex only weakly
whereas DEAAm leads to a significant expulsion of CTA mole-
cules from the cyclodextrin cavity. NIPAAm leads to expulsion in
all cases with CTA2 or CTA3, and complexes with CTA1 were
only stable up to a monomer/CTA ratio of 260/1. The complex
stability is increasing from NIPAAm over DEAAm to DMAAm,
which is in contrast to the ability of the substituent in the acry-
lamido monomer to act as a competing guest in the cyclodextrin.
The more hydrophobic and bulky isopropyl group in NIPAAm
has a bigger effect on the CTA/cyclodextrin complex stability
than the ethyl group in DEAAm, which itself has a larger effect on
the CTA/cyclodextrin complex than the methyl group in DMAAm.
We propose that this effect is due to a shift in the host/guest
equilibrium toward disassembly of the complex induced by
additional guest molecules. These can be either monomers or
additional water molecules. As monomers have to be employed
in higher equivalents to synthesize high molecular weight poly-
mers in controlled/living radical polymerizations, the possibility
that the monomer leads to expulsion of the CTA from the Me-
β-CD rises with increasing target molecular weights. An increas-
ing amount of water may also lead to a loss of the CTA/
cyclodextrin complex.61 The amount of water increases with
increasing target chain length as more water is needed to retain a
solution polymerization. Therefore, the decrease in the com-
plexation efficiency in the case of higher targeted molecular
weights can be explained by the increasing employed amount of
monomer and water molecules in these cases.
The polymerization leads to polymers with high molar masses
and low PDIs (7500 e Mn e 116 000 g molꢀ1; 1.06 e PDI e
1.54 for poly(DMAAm), 2500 e Mn e 150 000 g molꢀ1; 1.05 e
PDI e 1.39 for poly(DEAAm) and 4000 e Mn e 50 000 g
molꢀ1; 1.15 e PDI e 1.46 for poly(NIPAAm)). Furthermore,
we described —to the best of our knowledge—the first living
radical polymerization of DEAAm in water. The living character
of the polymerizations was confirmed by a linear increase of the
molecular weight with conversion, and chain extensions showed
very high reinitiation efficiencies. ESI mass spectra and 1H NMR
spectra were in good agreement with the expectations evidencing
the incorporation of hydrophobic end groups in the hydrophilic
acrylamido polymers. Thus, we provide the first living/controlled
polymerization of hydrophilic acrylamido monomers with hy-
drophobic CTA leading directly to hydrophobic end-function-
alized polymers in aqueous solution.
’ ASSOCIATED CONTENT
S
Supporting Information. Analytical data of the chain
b
transfer agents, experimental data on polymerizations with
CTA2 and CTA3, and additional data on polymerizations with
CTA1. This material is available free of charge via the Internet at
http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Tel (+49) 721 608-45642, Fax (+49) 721 608-45740, e-mail
christopher.barner-kowollik@kit.edu (C.B.-K.);Tel (+49) 211 81-
14760, Fax (+49) 211 81-15840, e-mail h.ritter@uni-duesseldorf.de
(H.R.).
’ ACKNOWLEDGMENT
The authors acknowledge financial support for the current
project from the German Research Council (DFG). C.B.-K.
acknowledges financial support from the Karlsruhe Institute of
Technology (KIT) in the context of the Excellence Initiative for
leading German universities and the Ministry for Science and
Arts of the state of Baden-W€urttemberg. The authors are
additionally thankful to Prof. P. Roesky (KIT) for the measure-
ment of a kinetic series of NMR samples.
’ REFERENCES
(1) van de Manakker, F.; Vermonden, T.; van Nostrum, C. F.;
Hennink, W. E. Biomacromolecules 2009, 10, 3157–3175.
(2) Yhaya, F.; Gregory, A. M.; Stenzel, M. H. Aust. J. Chem. 2010,
63, 195–210.
(3) Stadermann, J.; Komber, H.; Erber, M.; Dabritz, F.; Ritter, H.;
Voit, B. Macromolecules 2011, 44, 3250–3259.
(4) Zeng, J.; Shi, K.; Zhang, Y.; Sun, X.; Zhang, B. Chem. Commun.
2008, 3753–3755.
(5) Zhang, Z.-X.; Liu, K. L.; Li, J. Macromolecules 2011, 44,
1182–1193.
(6) Jazkewitsch, O.; Ritter, H. Macromolecules 2010, 44, 375–382.
(7) Zhao, Q.; Wang, S.; Cheng, X.; Yam, R. C. M.; Kong, D.; Li,
R. K. Y. Biomacromolecules 2010, 11, 1364–1369.
(8) Zhou, J.; Ritter, H. Polym. Chem. 2010, 1, 1552–1559.
(9) Kretschmann, O.; Choi, S. W.; Miyauchi, M.; Tomatsu, I.;
Harada, A.; Ritter, H. Angew. Chem. 2006, 118, 4468–4472.
’ CONCLUSIONS
The utilization of cyclodextrins provides new opportunities
for the synthetic methodology of living/controlled radical po-
lymerization. Based on the concept of supramolecular chemistry,
host/guest complexes seem to be attractive as controlling agents
in living radical polymerizations. In the current contribution we
report the first aqueous RAFT polymerization of acrylamido
monomers, e.g. DMAAm, DEAAm, and NIPAAm, with a supra-
molecular complex of cyclodextrin and hydrophobic CTAs. The
solubility of three novel guest-functionalized CTAs in water was
enhanced drastically via a cyclodextrin/CTA inclusion complex.
These complexes were thereafter utilized in living/controlled
radical polymerizations at 25 °C in water, leading to hydro-
philic polymers with hydrophobic end groups in one step.
7231
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232

Macromolecules
ARTICLE
(10) Harada, A.; Hashidzume, A.; Takashima, Y. Cyclodextrin-Based
Supramolecular Polymers. In Supramolecular Polymers Polymeric Betains
Oligomers; Springer: Berlin, 2006; Vol. 201, pp 1ꢀ43.
(11) Felici, M.; Marzꢀa-Pꢀerez, M.; Hatzakis, N. S.; Nolte, R. J. M.;
Feiters, M. C. Chem.—Eur. J. 2008, 14, 9914–9920.
(12) Ng, S. M.; Narayanaswamy, R. Sens. Actuators, B 2009, 139,
156–165.
(43) Li, H.; Yu, B.; Matsushima, H.; Hoyle, C. E.; Lowe, A. B.
Macromolecules 2009, 42, 6537–6542.
(44) Vogt, A. P.; Sumerlin, B. S. Macromolecules 2008, 41, 7368–7373.
(45) Mertoglu, M.; Laschewsky, A.; Skrabania, K.; Wieland, C.
Macromolecules 2005, 38, 3601–3614.
(46) Fetzner, A.; B€ohm, S.; Schreder, S.; Schubert, R. Eur. J. Pharm.
Biopharm. 2004, 58, 91–97.
(13) Hawker, C. J.; Bosman, A. W.; Harth, E. Chem. Rev. 2001, 101,
3661–3688.
(47) Suetsugu, N.; Koyama, S.; Takeo, K. i.; Kuge, T. J. Biochem.
1974, 76, 57–63.
(14) Grubbs, R. B. Polym. Rev. 2011, 51, 104–137.
(15) Ouchi, M.; Terashima, T.; Sawamoto, M. Chem. Rev. 2009,
109, 4963–5050.
(16) Braunecker, W. A.; Matyjaszewski, K. Prog. Polym. Sci. 2007, 32,
93–146.
(48) Kuo, C.; Provder, T.; Koehler, M. E. International GPC
Symposium Proceedings, 1991; pp 147ꢀ159.
(49) 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.
(17) Barner-Kowollik, C.; Perrier, S. J. Polym. Sci., Part A: Polym.
Chem. 2008, 46, 5715–5723.
(50) Skey, J.; O’Reilly, R. K. Chem. Commun. 2008, 4183–4185.
(51) Loftsson, T.; Jarho, P.; Mꢀasson, M.; J€arvinen, T. Expert Opin.
Drug Delivery 2005, 2, 335–351.
(52) Glockner, P.; Schollmeyer, D.; Ritter, H. Des. Monomers Polym.
2002, 5, 163–172.
(18) 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.
(19) Moad, G.; Rizzardo, E.; Thang, S. H. Polymer 2008,
49, 1079–1131.
(53) Funasaki, N.; Yodo, H.; Hada, S.; Neya, S. Bull. Chem. Soc. Jpn.
1992, 65, 1323–1330.
(20) Barner-Kowollik, C. Handbook of RAFT-Polymerization; Wiley-
VCH: Weinheim, Germany, 2008.
(21) Moad, G.; Rizzardo, E.; Thang, S. H. Aust. J. Chem. 2009, 62,
1402–1472.
(22) Lowe, A. B.; McCormick, C. L. Prog. Polym. Sci. 2007, 32,
(54) Rekharsky, M. V.; Inoue, Y. Chem. Rev. 1998, 98, 1875–1918.
(55) Thomas, D. B.; Convertine, A. J.; Hester, R. D.; Lowe, A. B.;
McCormick, C. L. Macromolecules 2004, 37, 1735–1741.
(56) Baussard, J.-F.; Habib-Jiwan, J.-L.; Laschewsky, A.; Mertoglu,
M.; Storsberg, J. Polymer 2004, 45, 3615–3626.
(57) Jodꢀai, I.; Kandra, L.; Harangi, J.; Nꢀanꢀasi, P.; Debrecen; Szejtli, J.
Starch/Staerke 1984, 36, 140–143.
(58) Gan, L. H.; Cai, W.; Tam, K. C. Eur. Polym. J. 2001, 37,
1773–1778.
(59) Delaittre, G.; Rieger, J.; Charleux, B. Macromolecules 2011, 44,
462–470.
283–351.
(23) Smith, A. E.; Xu, X.; McCormick, C. L. Prog. Polym. Sci. 2010,
35, 45–93.
(24) Millard, P.-E.; Barner, L.; Reinhardt, J.; Buchmeiser, M. R.;
Barner-Kowollik, C.; M€uller, A. H. E. Polymer 2010, 51, 4319–4328.
(25) Millard, P.-E.; Barner, L.; Stenzel, M. H.; Davis, T. P.; Barner-
Kowollik, C.; M€uller, A. H. E. Macromol. Rapid Commun. 2006, 27,
821–828.
(60) Heskins, M.; Guillet, J. E. J. Macromol. Sci., Chem. 1968, 2,
1441–1455.
(26) Xiong, Q.; Ni, P.; Zhang, F.; Yu, Z. Polym. Bull. 2004, 53, 1–8.
(27) Scales, C. W.; Vasilieva, Y. A.; Convertine, A. J.; Lowe, A. B.;
McCormick, C. L. Biomacromolecules 2005, 6, 1846–1850.
(28) Mitsukami, Y.; Donovan, M. S.; Lowe, A. B.; McCormick, C. L.
Macromolecules 2001, 34, 2248–2256.
(61) Del Valle, E. M. M. Process Biochem. 2004, 39, 1033–1046.
(29) Garnier, S.; Laschewsky, A. Macromolecules2005, 38, 7580–7592.
(30) Glatzel, S.; Badi, N.; Pach, M.; Laschewsky, A.; Lutz, J. F. Chem.
Commun. 2010, 46, 4517–9.
(31) Convertine, A. J.; Lokitz, B. S.; Lowe, A. B.; Scales, C. W.;
Myrick, L. J.; McCormick, C. L. Macromol. Rapid Commun. 2005, 26,
791–795.
(32) Thomas, D. B.; Convertine, A. J.; Myrick, L. J.; Scales, C. W.;
Smith, A. E.; Lowe, A. B.; Vasilieva, Y. A.; Ayres, N.; McCormick, C. L.
Macromolecules 2004, 37, 8941–8950.
(33) Ritter, H.; Mondrzik, B. E.; Rehahn, M.; Gallei, M. Beilstein
J. Org. Chem. 2010, 6, No. 60.
(34) Schwarz-Barac, S.; Ritter, H. J. Macromol. Sci., Pure Appl. Chem.
2003, A40, 437–448.
(35) K€ollisch, H. S.; Barner-Kowollik, C.; Ritter, H. Chem. Commun.
2009, 1097–1099.
(36) K€ollisch, H.; Barner-Kowollik, C.; Ritter, H. Macromol. Rapid
Commun. 2006, 27, 848–853.
(37) Storsberg, J.; Hartenstein, M.; M€uller, A. H. E.; Ritter, H.
Macromol. Rapid Commun. 2000, 21, 1342–1346.
(38) Pang, Y.; Ritter, H.; Tabatabai, M. Macromolecules 2003,
36, 7090–7093.
(39) Ding, L.; Li, Y.; Deng, J.; Yang, W. Polym. Chem. 2011,
2, 694–701.
(40) Weickenmeier, M.; Wenz, G.; Huff, J. Macromol. Rapid Com-
mun. 1997, 18, 1117–1123.
(41) H€ofler, T.; Wenz, G. J. Inclusion Phenom. Macrocyclic Chem.
1996, 25, 81–84.
(42) Kujawa, P.; Segui, F.; Shaban, S.; Diab, C.; Okada, Y.; Tanaka,
F.; Winnik, F. M. Macromolecules 2005, 39, 341–348.
7232
dx.doi.org/10.1021/ma2011969 |Macromolecules 2011, 44, 7220–7232