Synthesis of biodegradable hydrogel nanoparticles for bioapplications using inverse miniemulsion RAFT polymerization

Article abstract of DOI:10.1021/ma201531w

We report the synthesis of biodegradable hydrogel nanoparticles using a RAFT inverse miniemulsion cross-linking polymerization process with 2-(dimethylamino)ethyl methacrylate (DMAEMA) as monomer. The experimental conditions were optimized to yield a colloidally stable miniemulsion polymerization, which required protonation of DMAEMA using aqueous hydrogen chloride to ensure minimal partitioning to the continuous phase (cyclohexane). The nanoparticles were cross-linked using a disulfide cross-linker, thereby enabling subsequent degradation of the polymer network to its constituent primary chains by exposure to a reductive environment. The molecular weight distributions of the constituent primary chains were consistent with satisfactory control/livingness during the polymerization. These biodegradable hydrogel nanoparticles may have potential application as nanocarriers for encapsulation and controlled release of siRNA.

Full text of DOI:10.1021/ma201531w

ARTICLE
pubs.acs.org/Macromolecules
Synthesis of Biodegradable Hydrogel Nanoparticles for
Bioapplications Using Inverse Miniemulsion RAFT Polymerization
Marco Antonio M. Oliveira,† Cyrille Boyer,‡ Marcio Nele,† Josꢀe Carlos Pinto,† Per B. Zetterlund,*,‡ and
Thomas P. Davis*,†
†Programa de Engenharia Química/COPPE, Universidade Federal do Rio de Janeiro, Cidade Universitꢀaria, CP:68501, Rio de Janeiro,
21941-972 RJ, Brazil
‡Centre for Advanced Macromolecular Design (CAMD), School of Chemical Engineering, The University of New South Wales, Sydney,
New South Wales 2052, Australia
S Supporting Information
b
ABSTRACT: We report the synthesis of biodegradable hydro-
gel nanoparticles using a RAFT inverse miniemulsion cross-
linking polymerization process with 2-(dimethylamino)ethyl
methacrylate (DMAEMA) as monomer. The experimental
conditions were optimized to yield a colloidally stable mini-
emulsion polymerization, which required protonation of
DMAEMA using aqueous hydrogen chloride to ensure minimal
partitioning to the continuous phase (cyclohexane). The nano-
particles were cross-linked using a disulfide cross-linker, thereby
enabling subsequent degradation of the polymer network to its
constituent primary chains by exposure to a reductive environment. The molecular weight distributions of the constituent primary
chains were consistent with satisfactory control/livingness during the polymerization. These biodegradable hydrogel nanoparticles
may have potential application as nanocarriers for encapsulation and controlled release of siRNA.
’ INTRODUCTION
thus providing a route to polymeric nanoparticles comprising
well-defined polymers.7,8 CLRP can also be applied to cross-
linking polymerizations, whereby more well-defined and more
homogeneous polymer networks can be prepared.9ꢀ12 CLRP in
dispersed systems thus offers an attractive route to hydrogel
nanoparticles for drug delivery and gene therapy applications,
consisting of polymer networks built from tailored and well-
defined structures.1ꢀ3 Reversible additionꢀfragmentation chain
transfer (RAFT) polymerization is arguably the most versatile
CLRP method6,13 and appears advantageous for the preparation
of drug delivery systems because of the low toxicity of some
RAFT agents.14ꢀ16
Synthesis of hydrogel nanoparticles by cross-linking hetero-
geneous CLRP requires the use of an inverse system, i.e., a system
where the continuous phase is the oil phase and the dispersed
phase is hydrophilic (a water-in-oil system), stabilized by an oil-
soluble surfactant.7 The heterogeneous polymerization tech-
nique most well-suited for implementation of CLRP, as well as
for synthesis of organic/inorganic hybrid nanoparticles,17ꢀ19 is
miniemulsion polymerization.20,21 Miniemulsion polymerization
is based on the direct transformation of monomer droplets
into polymer particles, thus circumventing diffusion of monomer
Hydrogel nanoparticles (hydrophilic cross-linked polymeric
nanoparticles) possess a number of advantageous properties in
drug delivery applications.1ꢀ4 The particle size can be tuned with
relatively good precision, there is a large total surface area that
can be employed for bioconjugation, and the particle interior can
be used to embed drugs. By controlling the structure of the cross-
linked network density of the hydrogel nanoparticle, it is possible
to create a matrix with a defined porosity, thus allowing precise
control of the loading and release behavior of the polymer
network.3 A number of criteria need to be fulfilled for optimiza-
tion of the performance of hydrogel nanoparticles in drug
delivery applications, including long-term stability for circulation
in vivo, the presence of chemical functionality on the nanoparticle
surface (introduced by postmodification to improve the recogni-
tion of the nanoparticle by receptors on specifics cells), a particle
size less than 200 nm for accumulation by the enhanced
permeation effect (EPR), and the ability of the nanoparticle to
undergo biodegradation on exposure to external stimuli (e.g.,
pH, presence of enzymes). Biodegradation is important to
control the delivery of the drug and also to facilitate removal of
the “empty” nanoparticles or their residues.
Controlled/living radical polymerization (CLRP)5,6 enables
the precise synthesis of polymers with a multitude of architectures
and has now been adapted to be applicable to a wide range of
dispersed systems (e.g., emulsion, miniemulsion, microemulsion),
Received: July 5, 2011
Revised:
August 3, 2011
Published: August 18, 2011
r
2011 American Chemical Society
7167
dx.doi.org/10.1021/ma201531w Macromolecules 2011, 44, 7167–7175
|

Macromolecules
ARTICLE
(and other reactants) across the aqueous phase as in an emulsion
polymerization. Miniemulsions are thermodynamically unstable
and energy is required for their formation (normally provided via
high-energy homogenization), although low-energy approaches
have been reported.22ꢀ24 In an inverse miniemulsion system,25ꢀ27
a hydrophilic monomer is the main component of the dispersed
phase (e.g., acrylamide, acrylic acid, and N-isopropylacrylamide),
while an organic solvent is the main constituent of the contin-
uous phase. To date, there exist only few reports describing
inverse miniemulsion CLRP employing atom transfer radical
polymerization (ATRP)1,28ꢀ30 or RAFT.6,31 For instance, Schork
and co-workers described inverse RAFT miniemulsion polymer-
ization of acrylamide,32ꢀ34 acrylic acid,34 and N-isopropylacrylamide,35
while inverse RAFT microemulsion polymerization of N,N-
dimethylacrylamide has also been reported by McCormick and
co-workers.36
In this paper, we study the inverse miniemulsion polymeriza-
tion of another important hydrophilic monomer, i.e., 2-(di-
methylamino)ethyl methacrylate (DMAEMA).37 This particular
monomer can be used to prepare thermo- and pH-responsive
polymers that can be employed in the preparation of new drug or
gene delivery systems.38ꢀ40 DMAEMA-based polymers have at-
tracted special interest due to their ability to interact with DNA
or siRNA, forming a polymerꢀplasmid complex which can be
taken up by cells.41ꢀ44 Indeed, the amino side groups can be
easily quaternized to yield a positively charged polymer, which
then can be employed to interact with siRNA by electrostatic
interactions.45ꢀ48 It was reported that (co)polymers obtained
from DMAEMA can act as efficient transfection agents. Recently,
CAMD researchers developed several platforms based on poly-
(2-(dimethylamino)ethyl methacrylate (PDMAEMA), such as
biodegradable hyperbranched polymers37,49 or hybrid inorganic/
organic nanoparticles50 for the delivery of DNA or siRNA.
In the present work, we have developed a methodology
for RAFT inverse miniemulsion polymerization of DMAEMA
to produce biodegradable cross-linked PDMAEMA hydrogel
nanoparticles. To our knowledge, this is the first successful de-
scription of the use of DMAEMA in an inverse miniemulsion
RAFT polymerization (or inverse miniemulsion CLRP in
general).
poly(ethylene glycol) methyl ether methacrylate (Mn = 475 g/mol,
Sigma-Aldrich), potassium ferricyanide (III) (98%, Sigma-Aldrich),
silica gel (Grace), sodium chloride (99%, Sigma-Aldrich), sodium hydr-
oxide (97%, Ajax Finechem), sodium methoxide solution 25 wt % in
methanol (Sigma-Aldrich), sodium sulfate (99%, Ajax Finechem), Span
80 (sorbitan oleate; Fluka), tetrahydrofuran (THF, Honeywell,
HPLC grade), and toluene (99.5%, Ajax Finechem). 2,20-Azobis-
(isobutyronitrile) (AIBN, 98%, Sigma-Aldrich) was recrystallized twice
from acetone.
4-Cyano-4-(phenylcarbonothioylthio)pentanoic Acid. This
RAFT agent was prepared based on a method described previously.51
Dithiobenzoic Acid (DTBA). Sodium methoxide (25% solution in
methanol, 108.0 g, 0.5 mol), anhydrous methanol (125.0 g), and ele-
mental sulfur (16.0 g, 0.5 mol) were added to a three-necked round-
bottomed flask. Benzyl chloride (31.5 g, 0.25 mol) was then added
dropwise via addition funnel over a period of 1 h at room temperature
under stirring. The system was heated in an oil bath at 67 °C overnight.
The reaction mixture was cooled to 0 °C using an ice bath, the pre-
cipitated salt was removed by filtration, and the solvent removed by
rotary evaporation. Deionized water (250 mL) was added to the viscous
residue, and the resulting solution was filtered and transferred to a
separation funnel. The sodium dithiobenzoate solution was washed first
with diethyl ether (300 mL) + HCl 1 N (250 mL) and then with
deionized water (100 mL) + NaOH 1 N (250 mL). This washing
process was repeated two more times to finally yield a solution of sodium
dithiobenzoate.
Di(thiobenzoyl) Disulfide. Sodium dithiobenzoate solution
(175 mL) was transferred to a three-necked round-bottomed flask,
and a potassium ferricyanide solution (16.5 g, 0.05 mol in 250 mL of
deionized water) was added dropwise via an addition funnel over a
period of 1 h under vigorous stirring at room temperature. The resulting
red/purple precipitate was filtered and washed with deionized water
until the washings became colorless. The solid was dried in vacuo at
room temperature overnight.
4-Cyanopentanoic Acid Dithiobenzoate. Ethyl acetate
(80.0 mL), 4,40-azobis(4-cyanopentanoic acid) (8.5 g, 0.03 mol), and
di(thiobenzoyl) disulfide (5.7 g, 0.018 mol) were charged into a round-
bottomed flask. The reaction solution was heated and allowed to reflux
overnight. The solvent was removed by rotary evaporation, and the pro-
duct was isolated by column chromatography (silicagel 60 Å, 70ꢀ
230 mesh) using ethyl acetate:petroleum spirit (50/50 v/v) as eluent.
Fractions that were red/pink in color were combined, and the solvent
was removed by rotary evaporation. The final red oil was stored in a
freezer. 1H NMR (300 MHz, CDCl3)/δ ppm: 1.94 (s, 3H, CH3);
2.39ꢀ2.77 (m, 4H, CH2CH2); 7.42 (m, 2H, m-ArH); 7.59 (m, 1H,
p-ArH) and 7.92 (m, 2H, o-ArH).
’ EXPERIMENTAL SECTION
Materials. All chemicals listed below were used as received: 2-(di-
methylamino)ethyl methacrylate (DMAEMA, 98%, Aldrich), 2,20-azobis-
[2-(2-imidazolin-2-yl)propane] dihydrochloride (VA-044, 97%, Waco),
3,30-dithiopropionic acid (99%, Sigma-Aldrich), 4-(dimethylamino)-
pyridine (DMAP, Aldrich, 99%), 4,40-azobis(4-cyanopentanoic acid) (98%,
Fluka), acetonitrile (99.7%, Ajax Finechem), benzyl chloride (99%, Sigma-
Aldrich), cyclohexane (99%, Ajax Finechem), deuterated chloroform
(CDCl3, 99.8%, Cambridge Isotope Laboratories), deuterated dimethyl
sulfoxide (DMSO-d6, 99.9%, Cambridge Isotope Laboratories), deuter-
ium oxide (D2O, 99.9%, Cambridge Isotope Laboratories), dichloromethane
(DCM, 99.5%, Ajax Finechem), diethyl ether (99%, Ajax Finechem),
diphenyl ether (99%, Sigma-Aldrich), DL-dithiothreitol (99%, Sigma-
Aldrich), elemental sulfur (APS, 99.3%), ethyl acetate (99.5%, Ajax
Finechem), hydrochloric acid (HCl, 32%, Ajax Finechem), methanol
anhydrous (99.8%, Sigma-Aldrich), n-hexane (95%, Ajax Finechem),
triethylamine (99%, Sigma-Aldrich), N,N-dicyclohexylcarbodiimide
(DCC, Fluka, 99%), N,N-dimethylacetamide (DMAc; 99.9%, Sigma-
Aldrich), petroleum spirit (BR 40ꢀ60 °C, Ajax Finechem),
poly(ethylene glycol) methacrylate (Mn = 526 g/mol, Aldrich),
MacroRAFT Agent. Poly(ethylene glycol methyl ether) methacry-
late (8.0 g, 16.8 mmol; Mn = 475 g/mol), AIBN (27.2 mg, 0.16 mmol),
4-cyanopentanoic acid dithiobenzoate (232.0 mg, 0.8 mmol), and
acetonitrile (8.0 g, 2 mol) were charged into a round-bottomed flask
which was capped and degassed with nitrogen for 1 h at 15 °C. The sol-
ution was stirred for 6 h at 70 °C, and subsequently the excess solvent
was removed with air at room temperature. The viscous product was
dialyzed at room temperature using a cellulose membrane (MW 3500 g/
mol), first with deionized water and then with methanol. The excess
solvent was removed with air at room temperature, and the final red
viscous product was stored in a freezer. The viscous product was ana-
lyzed by 1H NMR in CDCl3 and D2O. 1H NMR (300 MHz, CDCl3)/
δ ppm: 1.50ꢀ2.00 (m, 3H, CH2ꢀCH); 3.36 (s, 3H, CH3); 3.60 (m, nH,
OCH2CH2O); 4.10 (s, 2H, COOCH2); 7.36 (m, 2H, m-ArH); 7.50 (m,
1H, p-ArH) and 7.87 (m, 2H, o-ArH). Two different batches of
macroRAFT agent were used. MacroRAFT1 (miniemulsion polymer-
izations without cross-linker): Mn = 16 100 g/mol (1H NMR); Mn =
13 650 g/mol, Mw/Mn = 1.11 (GPC). MacroRAFT2 (miniemulsion
7168
dx.doi.org/10.1021/ma201531w |Macromolecules 2011, 44, 7167–7175

Macromolecules
ARTICLE
Table 1. Miniemulsion Recipes for Polymerizations at 60 °C
continuous
phase
dispersed
phase
amount (g)
20
notes
cyclohexane
Span 80
1
HCl(aq)a (1%)
NaCl
2.5
0.05
VA-044
0.0051
[M]:[I] = 400:1
DMAEMA
macroRAFT
DMA-PEOSS
1
[M]:[RAFT] = 100:1
0.87b or 0.90c
based on Mn of macroRAFT
0.05 (if used)
a pH of the dispersed phase was adjusted to 2ꢀ3 by addition of drops of HCl(aq) (20%). b MacroRAFT1 (miniemulsion polymerizations without cross-
linker): Mn = 13 650 g/mol, Mw/Mn = 1.11. c MacroRAFT2 (miniemulsion polymerizations with cross-linker): Mn = 14 200 g/mol, Mw/Mn = 1.09.
polymerizations with cross-linker): Mn = 20 100 g/mol (1H NMR);
Mn = 14 200 g/mol, Mw/Mn = 1.09 (GPC).
Gel Permeation Chromatography (GPC). GPC analyses were
performed in DMAc (0.03% w/v LiBr, 0.05% 2, 6-dibutyl-4-methylphe-
nol (BHT)) at 50 °C and a flow rate of 1 mL/min using a Shimadzu
modular system comprising an LC-20AT pump, a CTO-10A column oven,
and a RID-10A refractive index detector. The system was equipped with a
Polymer Laboratories 5.0 mm bead-size guard column (50 ꢁ 7.8 mm2),
followed by four linear PL (Styragel) columns (105, 104, 103, and 500 Å)
calibrated with polystyrene standards ranging from 500 to 106 g/mol.
Theoretical number-average molecular weights (Mn,th) were calcu-
lated based on eq 1 (without cross-linker) and eq 2 (with cross-linker,
after cleavage of disulfide bonds):
Dithiopropionyl Poly(ethylene glycol) Dimethacrylate
(DMA-PEOSS). This biodegradable cross-linker was prepared as de-
scribed elsewhere,29 with some experimental modifications. First, poly-
(ethylene glycol) methacrylate (Mn = 526 g/mol) was purified as
described elsewhere.29 A solution of 3,30-dithiopropionic acid (2.3 g,
10 mmol) in THF (30 mL) was added dropwise to a solution of the
purified poly(ethylene glycol) methacrylate (10.0 g, 19 mmol) in DCM
(60 mL) containing DCC (3.9 g, 19 mmol) and a catalytic amount of
DMAP (0.5 g) kept in an ice bath over 20 min, and the reaction mixture
was subsequently stirred at room temperature overnight. The formed
solids were removed by vacuum filtration twice, and the solvents were
removed by rotary evaporation at 30 °C. The resulting yellow oil was
stored at in a refrigerator prior to use. 1H NMR (300 MHz, DMSO-d6)/
δ ppm: 1.9 (s, 6H, CH3ꢀ), 2.7 (t, 4H, ꢀC(O)ꢀCH2ꢀCH2ꢀSSꢀ), 2.9
(t,4H,ꢀO(O)CꢀCH2ꢀCH2ꢀSSꢀ),3.3ꢀ3.7 (m, 72H, ꢀ(CH2CH2O)nꢀ),
4.0ꢀ4.3 (m, 8H, ꢀC(O)OꢀCH2ꢀ and ꢀCH2ꢀO(O)Cꢀ), 5.7 (s, 2H,
CH=), 6.0 (s, 2H, CH=).
½Mꢂ0MWmon
R
Mn, th
¼
¼
þ MWmacroRAFT
ð1Þ
½macroRAFTꢂ0
½Mꢂ0MWmon
R
½cross-linkerꢂ0MWcross-linker
R
Mn, th
þ
½macroRAFTꢂ0
½macroRAFTꢂ0
þ MWmacroRAFT
ð2Þ
Miniemulsion Polymerizations. The basic inverse miniemul-
sion recipe was developed based on the work of Schork and co-
workers32,33 (Table 1). First, a solution of all components of the aqueous
phase (monomer, salt, initiator, RAFT agent, and cross-linker (if present))
were dissolved in an aqueous solution of HCl(aq) (1%). The pH of the
dispersed phase was adjusted to 2ꢀ3 by adding a few droplets of HCl(aq)
(20%). In a separated flask, a solution of the continuous organic phase
was prepared by dissolving the surfactant Span 80 in cyclohexane. The
aqueous (dispersed) phase was mixed with the continuous (organic) phase
for 30 min at 15 °C and ultrasonicated (Branson Digital Sonifier model
450) for 10 min at 70% of amplitude, while using an ice bath to cool the
sample. The resulting miniemulsion was transferred to a rounded-bottom
flask and degassed with N2 for 30 min at 15 °C.
The polymerizations were carried out at 60 °C under magnetic
stirring. Samples taken at predetermined times for characterization.
Acetone was added to the miniemulsion, followed by centrifugation for
15 min at 7500 rpm. The acetone/cyclohexane was subsequently
removed, and the precipitate was dried under vacuum at room tem-
perature. 10 mg of the dried samples from the conventional (no RAFT
agent) and RAFT polymerizations was treated with a small amount of
triethylamine for 30 min (in order to neutralize the positive charge of the
polymers) followed by dissolution in DMAc (GPC) or 1H NMR solvent.
Polymer samples from cross-linking polymerizations (RAFT with cross-
linker) were treated with 100 μL of 0.5 M aqueous solution of
DL-dithiothreitol for 1 h to break the disulfide cross-links (resulting in
primary chains remaining, assuming complete cleavage of cross-links).
The resulting aqueous solution was freeze-dried, and the recovered dry
polymer was dissolved in DMAc/1H NMR solvent.
where R denotes the fractional conversion of DMAEMA. Note that two
different batches of macroRAFT agent were used (see macroRAFT
synthesis section above).
Dynamic Light Scattering (DLS). Dynamic light scattering
measurements were conducted using a Malvern Zetasizer Nano ZS
instrument equipped with a 4 mW HeꢀNe laser operating at λ = 633 nm
and with an avalanche photodiode detector with high quantum effi-
ciency. The emulsions were analyzed without dilution.
Particle sizes were also measured for cross-linked particles redis-
persed in aqueous buffer solutions (10 mg in 4 mL): aqueous HCl 20%
(pH = 1ꢀ2), purified water (pH = 7), aqueous Na2CO3 (pH = 12). To
this end, acetone was added to the inverse miniemulsion to break the
emulsion, followed by centrifugation at 7500 rpm for 15 min. The
acetone/cyclohexane was then removed, and the remainder was dried in
vacuo at room temperature. The zeta potential was measured for cross-
linked particles redispersed in purified water.
Gas Chromatography (GC). Monomer conversions were deter-
mined using a Shimadzu 17A gas chromatograph system equipped with
an AT-Wax column (Heliflex capillary, Altech) using H2 as carrier
gas, with injector, column, and detector temperatures at 200, 120, and
250 °C, respectively. Samples of the inverse miniemulsion latex (0.1 g),
taken during the polymerization, were diluted with 2 mL of a GC sol-
ution (cyclohexane:toluene = 1000:2; toluene used as internal standard),
and then injected into the GC system.
Nuclear Magnetic Resonance Spectroscopy (1H NMR).
1H NMR spectra were recorded on a Bruker DPX 300, operating at
300 MHz, using deuterated chloroform (CDCl3), deuterium oxide
(D2O), or deuterated dimethyl sulfoxide (DMSO) as solvents.
7169
dx.doi.org/10.1021/ma201531w |Macromolecules 2011, 44, 7167–7175

Macromolecules
ARTICLE
Scheme 1. Span 80 (Sorbitan Oleate) and Biodegradable Cross-Linker Dithiopropionyl Poly(ethylene glycol) Dimethacrylate
(DMA-PEOSS)
Scheme 2. RAFT Polymerization of Poly(ethylene glycol) Methyl Ether Methacrylate (Mn = 475 g/mol) Employing 4-Cyano-
pentanoic Acid Dithiobenzoate To Generate Poly(ethylene glycol)-Based MacroRAFT Agent, Followed by RAFT Polymerization
of 2-(Dimethylamino)ethyl Methacrylate (DMAEMA) Employing Poly(ethylene glycol)-Based MacroRAFT Agent in Water
(in Inverse Miniemulsion)
’ RESULTS AND DISCUSSION
prevent Ostwald ripening,54 it is necessary to add an osmotic
pressure agent, which is a low molecular weight compound that is
soluble in the dispersed phase but exhibits extremely low
solubility in the continuous phase, e.g., a salt such as NaCl,
Na2SO4, or K2SO4 (equivalent to hexadecane in a normal (not
inverse) miniemulsion).25,27 In the present study, satisfactory
emulsion stability was achieved employing the surfactant Span 80
(Scheme 1) and NaCl (Table 1).
The initial plan was to use the RAFT agent 4-cyanopentanoic
acid dithiobenzoate, but this species proved to partition exces-
sively to the continuous phase. This RAFT agent was therefore
modified into a macroRAFT agent employing a PEG-based
monomer (poly(ethylene glycol) methyl ether methacrylate
(PEG-MA); Mn = 475 g/mol; Scheme 2), thereby increasing
its solubility in the dispersed aqueous phase and reducing its
solubility in the continuous phase (a similar strategy has been
Miniemulsion Formation. It is a fundamental criterion of a
miniemulsion polymerization that the monomer only exhibits
limited solubility in the continuous phase. DMAEMA is miscible
with a wide range of solvents that would normally be prime
candidates as the continuous phase in an inverse miniemulsion
system, e.g., cyclohexane, n-hexane, and heptane. To overcome
this problem, the monomer was protonated using HCl(aq) to
enhance its solubility in a neutral/acidic aqueous phase and
reduce its miscibility with the organic phase.52,53 This approach
ensured that the monomer resided predominantly in the dis-
persed phase (no traces of monomer detected in the continuous
organic phase by 1H NMR analysis).
Inverse miniemulsions can be prepared using a surfactant with
low HLB (hydrophilic lipophilic balance) value.25,27 In order to
7170
dx.doi.org/10.1021/ma201531w |Macromolecules 2011, 44, 7167–7175

Macromolecules
ARTICLE
Figure 1. 1H NMR spectra of (A) the purified poly(PEG-MA) macroRAFT agent and (B) the purified PDMAEMA-b-P(PEG-MA) macroRAFT
agent in D2O.
employed previously55). The synthesis of P(PEG-MA) macro-
RAFT was confirmed by 1H NMR analysis (Figure 1A), where
peaks related to the “Z-group” of the original RAFT agent
(aromatic protons at 7.2ꢀ7.8 ppm) and poly(ethylene glycol)
methyl ether methacrylate (4.1 (ꢀCH2Oꢀ), 3.6 (CH2O), and
3.3 ppm (CH2O)) could be observed (see Experimental Section
for details).
Miniemulsion Polymerization. Inverse miniemulsion RAFT
polymerization of DMAEMA using the macroRAFT agent was
conducted at 60 °C using the aqueous phase (dispersed phase)
initiator VA-044 according to the recipe in Table 1. The
polymerization generates a block copolymer of the structure
shown in Scheme 2. Figure 2 shows photographs of miniemul-
sions before and after polymerization as well as before ultra-
sonication. Identical polymerizations were also carried out in the
absence of the macroRAFT agent. Similar polymerization rates
were observed in the presence and absence of RAFT agent. Both
polymerizations were rapid (close to 70% conversion in 1 h) and
proceeded beyond 80% conversion in 2.5 h (Figure 3). The
RAFT polymerization proceeded to near-complete conversion
7171
dx.doi.org/10.1021/ma201531w |Macromolecules 2011, 44, 7167–7175

Macromolecules
ARTICLE
Figure 2. (a) Reaction mixture (as per recipe in Table 1) for prepara-
tion of emulsion with macroRAFT agent but without cross-linker before
ultrasonication, (b) after ultrasonication, and (c) after polymerization
for 30 min (36% conversion).
Figure 4. Molecular weight distributions obtained in inverse miniemul-
sion polymerizations of DMAEMA initiated by VA-044 at 60 °C (a) with
poly(ethylene glycol)-based macroRAFT agent and (b) with poly-
(ethylene glycol)-based macroRAFT agent and the cross-linker DMA-
PEOSS after degradation by treatment with DL-dithiothreitol to break
the disulfide cross-links.
Figure 3. Conversionꢀtime data for inverse miniemulsion polymeriza-
tions of DMAEMA initiated by VA-044 at 60 °C without RAFT agent/
cross-linker (“conventional”), with poly(ethylene glycol)-based macro-
RAFT agent, and with poly(ethylene glycol)-based macroRAFT agent
and the cross-linker DMA-PEOSS based on recipes in Table 1.
(>98%) in 6 h, whereas the conventional radical polymerization
appeared to reach a limiting conversion below 90%. Interestingly,
the dithioester did not degrade during the polymerization de-
1
spite the low pH. H NMR analysis of the purified copolymer
(Figure 1B) confirms the presence of the RAFT end group at
7.2ꢀ7.8 ppm, and the copolymer is pink in color.
Figure 4a shows molecular weight distributions at different
monomer conversions, revealing how the distributions shift to
higher molecular weight with increasing conversion, consistent
with a controlled/living process. There is however a low molec-
ular weight shoulder, indicative of some fraction of macroRAFT
agents not having participated in the polymerization at a given
conversion. This may be caused by dead chains (i.e., chains not
containing a RAFT end group) of the initial macroRAFT agent,
or it may be related to the heterogeneous polymerization itself.
Figure 5. Mw/Mn and Mn vs conversion for inverse miniemulsion
polymerizations of DMAEMA initiated by VA-044 at 60 °C with poly-
(ethylene glycol)-based macroRAFT agent based on recipe in Table 1.
The Mn values increased linearly with conversion close to Mn,th
,
and Mw/Mn remained below 1.3(Figure 5). The molecular weights
were much higher, as expected, in the absence of RAFT agent (of
the order of 105 g/mol; Table S1 in the Supporting Information).
The molecular weight distributions were slightly bimodal (Figure S1),
probably as a result of dependence of the instantaneous molecular
weight on conversion (reactant concentrations and the termination
rate coefficient varying with conversion).
Figure 6 shows particle diameters vs conversion for miniemulsion
polymerizations with and without RAFT agent. The number-average
7172
dx.doi.org/10.1021/ma201531w |Macromolecules 2011, 44, 7167–7175

Macromolecules
ARTICLE
Figure 6. Particle size measurements for inverse miniemulsion poly-
merizations of DMAEMA initiated by VA-044 at 60 °C without RAFT
agent/cross-linker (“conventional”), with poly(ethylene glycol)-based
macroRAFT agent and with poly(ethylene glycol)-based macroRAFT
agent and the cross-linker DMA-PEOSS based on recipes in Table 1.
The initial value of dw/dn for RAFT with cross-linker was 20 (not shown
in plot).
Figure 7. Particle size distributions by number (—) and volume (
)
3 3 3
for inverse miniemulsion polymerizations of DMAEMA initiated by VA-
044 in at 60 °C (a) without RAFT agent/cross-linker (“conventional”;
82% conversion), (b) with poly(ethylene glycol)-based macroRAFT
agent (92% conversion), and (c) with poly(ethylene glycol)-based
macroRAFT agent and the cross-linker DMA-PEOSS (94% conversion)
based on recipes in Table 1.
particle diameters before polymerization were very similar to with-
out RAFT agent (dn ≈ 100 nm). In the absence of RAFT agent, the
particle size appeared to remain relatively constant with conversion
(the dn data point for conventional radical polymerization and the
RAFT data point at zero conversion are superimposed). In the
presence of RAFT agent, a significant increase in particle size was
observed in the conversion range 0ꢀ30%, beyond which dn
remained in the range 250ꢀ290 nm. It is well established that it
is more difficult to maintain colloidal stability in a CLRP system
relative to its conventional radical polymerization counterpart as a
result of superswelling,7,56 i.e., oligomers (formed by CLRP)
promoting swelling of nucleated monomer droplets by monomer
from not yet nucleated monomer droplets. The particle size
distributions by weight at high conversion (Figure 7) revealed
that a small population of larger particles (micrometer-sized)
were present in the RAFT system, but not in the conventional
system. The values of dw/dn were similar for both systems
throughout the polymerizations, although the system with RAFT
agent had a considerably broader initial droplet size distribution
(higher dw/dn) than the conventional system (Figure 6).
Synthesis of Cross-Linked Biodegradable Particles. Poly-
merizations were carried out as above using the macroRAFT agent
in the presence of the biodegradable cross-linker DMA-PEOSS
according to the recipe in Table 1. The polymerization rate was very
similar to the case without the cross-linker (Figure 3). The number-
average particle sizes were somewhat smaller in the presence of
cross-linker, with dn ≈ 200 nm in the conversion range 30ꢀ100%
conversion, but the particle size distributions were significantly
broader (higher dw/dn with cross-linker; Figure 6). Also in the
presence of cross-linker, a population of larger micrometer-sized
particles were detected in the weight distribution (Figure 7).
The obtained cross-linked networks of the hydrogel nanoparti-
cles were subsequently subjected to induced degradation by treat-
ment with DL-dithiothreitol to reduce the disulfide cross-links. This
would result in formation of the primary chains of the network,
which ideally would be of similar molecular weight distribution to
the polymer formed in the corresponding polymerization in the
absence of cross-linker.10ꢀ12,57 Figure 4B shows molecular weight
distributions of the polymer obtained after degradation, revealing
relatively narrow monomodal distributions that shift to higher
molecular weight with increasing monomer conversion. Interest-
ingly, the fairly prominent low molecular weight shoulder present in
the RAFT polymerizations without cross-linker (Figure 4A) is
absent in this case. This suggests that the low molecular weight
shoulder in the absence of cross-linker has its origin in the
heterogeneous mechanism of the polymerization and is not caused
by dead original macroRAFT agent. The values of Mn increased
with increasing conversion but were somewhat higher than Mn,th
,
and Mw/Mn remained below 1.3 to high conversion (Figure 8).
After isolation, the purified hydrogel nanoparticles can be
readily dispersed in water and exhibit pH-responsive behavior in
aqueous dispersion.38 The particle diameter decreased with
increasing pH: dn = 210 nm at pH = 4, dn = 120 nm at pH = 7,
and dn = 94 nm at pH = 12. The decrease of the size is attributed
to the decrease in water solubility of PDMAEMA at high pH. The
zeta-potential of the purified particles redispersed in pure water
7173
dx.doi.org/10.1021/ma201531w |Macromolecules 2011, 44, 7167–7175

Macromolecules
ARTICLE
’ ACKNOWLEDGMENT
The authors thank the Conselho Nacional de Desenvolvimento
Científico e Tecnolꢀogico (CNPq - Brazil) for the scholarship
for M.A.M.O. In addition, we acknowledge significant research
fellowship funding to T.P.D. (Federation Fellowship) and
C.B. (Australian Postdoctoral Fellowship) from the Austra-
lian Research Council. P.B.Z. is grateful for strategic funding
from UNSW.
’ REFERENCES
(1) Oh, J. K.; Siegwart, D. J.; Lee, H.; Sherwood, G.; Peteanu, L.;
Hollinger, J. O.; Kataoka, K.; Matyjaszewski, K. J. Am. Chem. Soc. 2007,
129, 5939–5945.
(2) Oh, J. K.; Drumright, R.; Siegwart, D. J.; Matyjaszewski, K. Prog.
Polym. Sci. 2008, 33, 448–477.
(3) Hoare, T. R.; Kohane, D. S. Polymer 2008, 49, 1993–2007.
(4) Ryu, J. H.; Chacko, R. T.; Jiwpanich, S.; Bickerton, S.; Babu,
R. R.; Thayumanavan, S. J. Am. Chem. Soc. 2010, 132, 17227–17235.
(5) Braunecker, W. A.; Matyjaszewski, K. Prog. Polym. Sci. 2007,
32, 93–146.
(6) Boyer, C.; Stenzel, M. H.; Davis, T. P. J. Polym. Sci., Part A: Polym.
Chem. 2011, 49, 551–595.
(7) Zetterlund, P. B.; Kagawa, Y.; Okubo, M. Chem. Rev. 2008,
108, 3747–3794.
Figure 8. Mw/Mn and Mn vs conversion for inverse miniemulsion
polymerizations of DMAEMA initiated by VA-044 at 60 °C with
poly(ethylene glycol)-based macroRAFT agent and the cross-linker
DMA-PEOSS after degradation by treatment with DL-dithiothreitol to
break the disulfide cross-links.
(8) Cunningham, M. F. Prog. Polym. Sci. 2008, 33, 365–398.
(9) Ide, N.; Fukuda, T. Macromolecules 1999, 32, 95–99.
(10) Zetterlund, P. B.; Alam, M. N.; Minami, H.; Okubo, M.
Macromol. Rapid Commun. 2005, 26, 955–960.
(11) Saka, Y.; Zetterlund, P. B.; Okubo, M. Polymer 2007, 48,
1229–1236.
(12) Gao, H.; Matyjaszewski, K. Prog. Polym. Sci. 2009, 34, 317–350.
(13) Barner-Kowollik, C. Handbook of RAFT Polymerization; Wiley-
VCH: New York, 2008.
(pH 6.5) was +42 mV. The presence of positive charges at the
surface of the nanogels offers the possibility to conjugate siRNA
or DNA and subsequently exploits these nanoparticles as gene
nanocarriers.
(14) Chang, C.-W.; Bays, E.; Tao, L.; Alconcel, S. N. S.; Maynard,
H. D. Chem. Commun. 2009, 3580–3582.
(15) Boyer, C.; Bulmus, V.; Davis, T. P.; Ladmiral, V.; Liu, J.; Perrier,
S. Chem. Rev. 2009, 109, 5402–5436.
(16) Pissuwan, D.; Boyer, C.; Gunasekaran, K.; Davis, T. P.; Bulmus,
V. Biomacromolecules 2010, 11, 412–420.
(17) Landfester, K. Angew. Chem., Int. Ed. 2009, 48, 4488–4507.
(18) Faucheu, J.; Gauthier, C.; Chazeau, L.; Cavaille, J. Y.; Mellon,
V.; Lami, E. B. Polymer 2010, 51, 6–17.
(19) van Berkel, K. Y.; Hawker, C. J. J. Polym. Sci., Part A: Polym.
Chem. 2010, 48, 1594–1606.
(20) Asua, J. M. Prog. Polym. Sci. 2002, 27, 1283.
(21) Landfester, K. Macromol. Rapid Commun. 2001, 22, 896.
(22) Spernath, L.; Magdassi, S. Polym. Adv. Technol. 2007, 18,
705–711.
(23) Guo, Y.; Liu, J. Q.; Zetterlund, P. B. Macromolecules 2010, 43,
5914–5916.
(24) Cheng, S. Q.; Guo, Y.; Zetterlund, P. B. Macromolecules 2010,
43, 7905–7907.
(25) Landfester, K.; Willert, M.; Antonietti, M. Macromolecules
2000, 33, 2370–2376.
’ CONCLUSIONS
Biodegradable hydrogel nanoparticles with diameters of
∼200 nm have been synthesized via inverse miniemulsion
RAFT polymerization of 2-(dimethylamino)ethyl methacrylate
(DMAEMA). The experimental conditions were optimized so as
to minimize partitioning of both monomer and RAFT agent to
the continuous phase, thereby allowing the synthesis of nano-
particles comprising a well-defined network structure. Cross-
linking (network formation) was achieved by use of a biodegrad-
able disulfide cross-linker, thereby enabling us to subsequently
severe the cross-links by exposure to a reductive environment.
The molecular weight characteristics of the resulting primary
chains were consistent with the cross-linking polymerization
having proceeded via a controlled/living mechanism. These
biodegradable hydrogel nanoparticles are currently being inves-
tigated for encapsulation and controlled release of siRNA.
’ ASSOCIATED CONTENT
(26) Hemingway, M. G.; Gupta, R. B.; Elton, D. J. Ind. Eng. Chem.
Res. 2010, 49, 10094–10099.
S
Supporting Information. Molecular weight data in the
b
(27) Capek, I. Adv. Colloid Interface Sci. 2010, 156, 35–61.
(28) Oh, J. K.; Perineau, F.; Matyjaszewski, K. Macromolecules 2006,
39, 8003–8010.
absence of RAFT agent and cross-linker. This material is available
free of charge via the Internet at http://pubs.acs.org.
(29) Oh, J. K.; Tang, C.; Gao, H.; Tsarevsky, N. V.; Matyjaszewski,
K. J. Am. Chem. Soc. 2006, 128, 5578–5584.
’ AUTHOR INFORMATION
(30) Oh, J. K.; Dong, H.; Zhang, R.; Matyjaszewski, K.; Schlaad, H.
J. Polym. Sci., Part A: Polym. Chem. 2007, 45, 4764–4772.
(31) Moad, G.; Rizzardo, E.; Thang, S. H. Aust. J. Chem. 2005,
58, 379–410.
Corresponding Author
*Tel: +61 2 9385 4331; Fax: +61 2 9385 6250; e-mail:
p.zetterlund@unsw.edu.au (P.B.Z.), t.davis@unsw.edu.au (T.P.D.).
7174
dx.doi.org/10.1021/ma201531w |Macromolecules 2011, 44, 7167–7175

Macromolecules
ARTICLE
(32) Qi, G.; Jones, C. W.; Schork, F. J. Macromol. Rapid Commun.
2007, 28, 1010–1016.
(33) Qi, G.; Eleazer, B.; Jones, C. W.; Schork, F. J. Macromolecules
2009, 42, 3906–3916.
(34) Ouyang, L.; Wang, L.; Schork, F. J. Macromol. Chem. Phys.
2010, 211, 1977–1983.
(35) Lu, F.; Luo, Y.; Li, B.; Zhao, Q.; Schork, F. Macromolecules
2010, 43, 568–571.
(36) Sogabe, A.; McCormick, C. L. Macromolecules 2009, 42,
5043–5052.
(37) Tao, L.; Liu, J.; Tan, B. H.; Davis, T. P. Macromolecules 2009,
42, 4960–4962.
(38) Yuk, S. H.; Cho, S. H.; Lee, S. H. Macromolecules 1997, 30,
6856–6859.
(39) Liu, F.; Urban, M. W. Macromolecules 2008, 41, 6531–6539.
(40) Liu, X.; Ni, P.; He, J.; Zhang, M. Macromolecules 2010,
43, 4771–4781.
(41) Lim, D. W.; Yeom, Y. I.; Park, T. G. Bioconjugate Chem. 2000,
11, 688–695.
(42) van de Wetering, P.; Moret, E. E.; Schuurmans-Nieuwenbroek,
N. M. E.; van Steenbergen, M. J.; Hennink, W. E. Bioconjugate Chem.
1999, 10, 589–597.
(43) Jiang, X.; Lok, M. C.; Hennink, W. E. Bioconjugate Chem. 2007,
18, 2077–2084.
(44) Lin, S.; Du, F.; Wang, Y.; Ji, S.; Liang, D.; Yu, L.; Li, Z.
Biomacromolecules 2007, 9, 109–115.
(45) Wetering, P.; Zuidam, N. J.; Steenbergen, M. J.; Houwen, O.;
Underberg, W. J. M.; Hennink, W. E. Macromolecules 1998,
31, 8063–8068.
(46) Cherng, J.; Wetering, P.; Talsma, H.; Crommelin, D. J. A.;
Hennink, W. E. Pharm. Res. 1996, 13, 1038–1042.
(47) Wetering, P.; Cherng, J.; Talsma, H.; Crommelin, D.; Hennink,
W. J. Controlled Release 1998, 53, 145–153.
(48) Wetering, P.; Cherng, J.; Talsma, H.; Hennink, W. E.
J. Controlled Release 1997, 49, 59–69.
(49) Tao, L.; Chou, W. C.; Tan, B. H.; Davis, T. P. Macromol. Biosci.
2010, 10, 632–637.
(50) Boyer, C.; Priyanto, P.; Davis, T. P.; Pissuwan, D.; Bulmus, V.;
Kavallaris, M.; Teoh, W. Y.; Amal, R.; Carroll, M.; Woodward, R.; St
Pierre, T. J. Mater. Chem. 2010, 20, 255–265.
(51) Mitsukami, Y.; Donovan, M. S.; Lowe, A. B.; McCormick, C. L.
Macromolecules 2001, 34, 2248–2256.
(52) B€ut€un, V.; Billingham, N. C.; Armes, S. P. Chem. Commun.
1997, 671–672.
(53) Lee, A. S.; Gast, A. P.; B€ut€un, V.; Armes, S. P. Macromolecules
1999, 32, 4302–4310.
(54) Taylor, P. Adv. Colloid Interface Sci. 1998, 75, 107–163.
(55) Boyer, C.; Bulmus, V.; Liu, J.; Davis, T. P.; Stenzel, M. H.;
Barner-Kowollik, C. J. Am. Chem. Soc. 2007, 129, 7145–7154.
(56) Luo, Y.; Tsavalas, J.; Schork, F. J. Macromolecules 2001, 34,
5501–5507.
(57) Li, Y.; Armes, S. P. Macromolecules 2005, 38, 8155–8162.
7175
dx.doi.org/10.1021/ma201531w |Macromolecules 2011, 44, 7167–7175