Poly(acrylic acid)-grafted magnetic nanoparticle for conjugation with folic acid

Article abstract of DOI:10.1016/j.polymer.2010.12.059

Poly(acrylic acid) (poly(AA))-grafted magnetite nanoparticles (MNPs) prepared via surface-initiated atom transfer radical polymerization (ATRP) of t-butyl acrylate, followed by acid-catalyzed deprotection of t-butyl groups, is herein presented. In additio

Full text of DOI:10.1016/j.polymer.2010.12.059

Polymer 52 (2011) 987e995
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Poly(acrylic acid)-grafted magnetic nanoparticle for conjugation with folic acid
*
Metha Rutnakornpituk , Nipaporn Puangsin, Pawinee Theamdee, Boonjira Rutnakornpituk, Uthai Wichai
Department of Chemistry and Center of Excellence for Innovation in Chemistry, Faculty of Science, Naresuan University, Phitsanulok 65000, Thailand
a r t i c l e i n f o
a b s t r a c t
Article history:
Poly(acrylic acid) (poly(AA))-grafted magnetite nanoparticles (MNPs) prepared via surface-initiated
atom transfer radical polymerization (ATRP) of t-butyl acrylate, followed by acid-catalyzed deprotection
of tebutyl groups, is herein presented. In addition to serve as both steric and electrostatic stabilizers,
poly(AA) grafted on MNP surface also served as a platform for conjugating folic acid, a cancer cell
targeting agent. Fourier transform infrared spectroscopy (FTIR) was used to monitor the reaction
progress in each step of the syntheses. The particle size was 8 nm in diameter without significant
aggregation during the preparation process. Photocorrelation spectroscopy (PCS) indicated that, as
increasing pH of the dispersions, their hydrodynamic diameter was decreased and negatively charge
surface was obtained. According to thermogravimetric analysis (TGA), up to 14 wt% of folic acid (about
400 molecules of folic acid per particle) was bound to the surface-modified MNPs. This novel nano-
complex is hypothetically viable to efficiently graft other affinity molecules on their surfaces and thus
might be suitable for use as an efficient drug delivery vehicle particularly for cancer treatment.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 7 October 2010
Received in revised form
22 December 2010
Accepted 24 December 2010
Available online 9 January 2011
Keywords:
Nanoparticle
poly(acrylic acid)
ATRP
1. Introduction
attention in recent years. As compared to a conventional radical
polymerization, surface-initiated ATRP from nanoparticles produced
Synthesis of magnetite nanoparticles (MNPs) coated with a thin
film of organic polymer has recentlyattracted much attention due to
their potential biomedical applications such as magnetic resonance
imaging [1e6], magnetic separation, controlled drug release [7] and
hyperthermia treatment of tumor cells [8]. A thin shell of polymeric
coating on the particle surface is necessary to prevent nanometer-
sized particle aggregation due to their inherent anisotropic dipolar
interaction, resulting in losing the specific properties associated
with their nanometer dimensions [9,10]. In addition, the polymers
on their surface can provide a platform for incorporating biological
functional molecules, such as amino acid [11], protein [12,13] and
DNA [14e16], for particle labeling with fluorescent molecules [10,17]
and for attaching folic acid [18,19], a receptor for tumor cells.
Recently, atom transfer radical polymerization (ATRP) has been
reported as a potential “grafting-from” method for surface modifica-
tion [20,21,27]. ATRP is a living/controlled radical polymerization
method, which does not require stringent experimental conditions
[22,23]. ATRP enables for the polymerization and block copolymeri-
zation of a wide range of functional monomers such as styrene
[24e26], methacrylate [27], acrylate [28,29] and methacrylamide [30],
yielding polymers with narrowly dispersed molecular weights.
Surface modification of nanoparticles via ATRP has attracted a great
polymers with narrow polydispersity index (PDI) and proceeded in
a controlled fashion [31]. In addition, the advantage of ATRP technique
as compared to other controlled radical polymerization (CRP) tech-
niques is that the polymerization can be initiated at low reaction
temperature, while other CRP techniques such as reversible addition-
fragmentation chain transfer (RAFT) and nitroxide-mediated poly-
merizations require relatively high reaction temperature to generate
radicals from azo or peroxide initiators. Moreover, functionalization of
the particle surface with alkyl halide, the ATRP initiating species, can
be easily carried out either by physical absorption of acid-containing
halides [36] or covalent bonding of ATRP initiating halides via silani-
zation [32]. The “grafting from” strategy via ATRP has thus been mostly
adopted for MNP surface modification with a variety of polymeric
surfactants such as polystyrene [27], poly(methyl methacrylate) [33],
poly(ethylene glycol) methacylate [34,35] and poly(acrylamide) [36].
Theaimofthecurrentworkistoadopta“graftingfrom”methodto
modify MNP surfaces with poly(t-butyl acrylate) (poly(t-BA)) via
ATRP, followed by acid-catalyzed deprotection of t-butyl groups to
obtain poly(AA)-grafted MNP. It is thought that ATRP can offer well-
defined water dispersible poly(AA) stabilizers with low molecular
weight distributionon theparticlesurface. Thecarboxylic acidgroups
overexpressed on its surface are readily reactive toward molecules
containing functional groups such as amine and alcohol. It has thus
gained our attention because, not only serving as steric and electro-
staticsurfactants [37], poly(AA) can alsobeused asa keyintermediate
* Corresponding author. Tel.: þ66 5596 3464; fax: þ66 5596 1025.
E-mail address: methar@nu.ac.th (M. Rutnakornpituk).
0032-3861/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2010.12.059

988
M. Rutnakornpituk et al. / Polymer 52 (2011) 987e995
for grafting a large range of functional molecules [38,39]. Folic acid
(FA) is of particular interest in this work because it can specifically
conjugate with folate receptors overexpressed on cancer cell
membranes [40]. Precedents have reported the immobilization of FA
on the outermost surface of MNPs coated with other polymeric
surfactants [41e44]. Therefore, it is expected that the multifunctional
FA-grafted MNPs prepared in this work should bind to cancer cell
membranes specifically and consequently improve uptake efficiency
of the MNP to the cells. The detail studies on the efficiency on treating
cancer cells of this complex are warranted for a future investigation.
In the present work, poly(AA)-coated MNPs were thus prepared
via surface-initiated ATRP of t-BA, followed by acid-catalyzed
hydrolysis of t-butyl groups. FTIR was used to monitor the reaction
progress in each step. Thermogravimetric analysis (TGA) was used
to investigate percent of each composition in the polymer-MNP
complex. Transmission electron microscopy (TEM) technique was
also used to monitor the particle size and the presence of the
polymer in the complex. Vibrating sample magnetometry (VSM)
was performed to reveal their magnetic properties. In combination
with UVevisible spectrophotometry and FTIR, TGA technique was
conducted to evidence the existence of FA in the complexes.
BTPAm, was yellowish thick liquid (78% yield). 1H NMR (400 MHz,
CDCl3) dH: 0.60 [m, 2H, SieCH2,1.20 [t, 9H, OeCH2eCH3],1.65 [m, 2H,
SieCH2eCH2], 1.95, [s, 6H, CH3eCeBr], 3.25 [m, 2H, CH2eNH], 3.80
[m, 6H, CH3eCH2eO]. FT-IR (KBr disc) ymax: 3345 cmꢁ1 (NH stretch-
ing), 2975e2889 cmꢁ1 (CeH stretching), 1738 cmꢁ1 (C]O of acid
bromide stretching), 1658 cmꢁ1 (C]O of amide stretching),
1532 cmꢁ1 (NH bending), 1442 cmꢁ1 (CeN stretching), 1286 cmꢁ1
(CeBr stretching), 1112e1026 (SieO stretching).
2.2.3. Immobilization of 2-bromo-2-methyl-N-(3-(triethoxysilyl)
propanamide (BTPAm)) onto MNP surface (BTPAm-coated MNPs)
(Fig. 1)
To immobilize BTPAm on the oleic acid-coated MNP surface, the
MNP-toluene dispersion (0.1 g of oleic acid-coated MNPs in 5 ml
toluene) (30 ml), BTPAm (0.90 ml) and 2 M TEA in toluene (6 ml)
were added into a round bottom flask. The mixture was stirred for
24 h at room temperature under nitrogen. The particles were
subsequently precipitated in methanol, following by magnet
separation to obtain the BTPAm-modified MNPs. Then, the MNPs
were re-dispersed in toluene and re-precipitated in methanol. This
procedure was repeated several times to completely remove
unreacted BTPAm. The particles were finally dried in vacuo.
2. Experimental section
2.2.4. Synthesis of poly(t-butyl acrylate)-coated MNPs (poly(t-BA)-
coated MNPs) via ATRP reaction
2.1. Materials
To a schlenk tube containing dioxane (1 ml), CuBr (0.3 g,
0.0021 mol), and PMDETA (0.42 ml, 0.0021 mol) were added under
nitrogen blanket. The mixture was stirred until homogenous blue
color was observed. Then, t-butyl acrylate (t-BA) (3 ml, 0.021 mol)
monomer and BTPAm-immobilized MNPs (0.3 g) were added via
a syringe. The mixture was degassed and nitrogen-purged by three
freeze-thaw cycles. The solution was then heated to 90 ꢀC for 24 h
to commence ATRP reaction. At a given time, the reactions were
ceased and poly(t-BA)-grafted MNPs were magnetically separated
and washed thoroughly with methanol and dried in vacuo.
Unless otherwise stated, all reagents were used without further
purification: iron (III) acetylacetonate (Fe(acac)3), 99% (Acros),
benzyl alcohol (Unilab), 3-aminopropyl triethoxysilane (APS), 99%
(Acros), triethylamine (TEA) (Carto Erba), 2-bromoisobutyryl
bromide (BIBB), 98% (Acros), copper (I) bromide (CuBr), 98%
(Acros), N,N,N0,N00,N00-pentamethyldiethylenetriamine (PMDETA),
ethyl-a-bromoisobutyrate (Aldrich), 99% (Acros), folic acid, 97%
(Fluka), N-hydroxyl succinamide (NHS), 98% (Acros), dicyclohexyl
carbodiimide (DCC), 99% (Acros), di-t-butyl dicarbonate (Boc2O),
99% (Aldrich), ethylene diamine (EDA), 99.5% (Fluka), trifluoroacetic
acid (TFA), 99.5% (Fluka). t-Butyl acrylate (t-BA), 99% (Fluka), was
distilled under vacuum prior to use.
2.2.5. Synthesis of poly(acrylic acid)-coated MNPs (poly(AA)-coated
MNPs) via hydrolysis of poly(t-butyl acrylate)-coated MNPs
Poly(t-BA)-coated MNPs were hydrolyzed to obtain acrylic acid
functional groups on MNP surfaces. Briefly, poly(t-BA)-coated
MNPs (0.05 g) were hydrolyzed in a 20-ml TFA solution (0.1 M of
TFA in THF) at room temperature for 24 h. The solution was
concentrated under reduced pressure, diluted with CH2Cl2, and
repeatedly precipitated in cold hexane. The precipitate was sepa-
rated by a permanent magnet and dried in vacuo. The possible
reactions between TFA and polymers coated on MNP surface are
illustrated in supplementary data.
2.2. Synthesis
2.2.1. Synthesis of oleic acid-coated magnetite nanoparticles
(MNPs)
MNPs were prepared via thermal decomposition following the
method previously described [45]. In a typical procedure, Fe(acac)3
(1.0 g, 2.81 mmol) and benzyl alcohol (20 ml) were mixed by
magnetic stirring in a three-neck flask with nitrogen flow. The
mixture was heated to 200 ꢀC for 48 h. The precipitant was then
removed from the dispersion using an external magnet and washed
with ethanol and CH2Cl2 repeatedly to remove benzyl alcohol. The
particles were then dried at room temperature under reduced
pressure. To prepare oleic acid-coated MNPs, the dried MNPs (0.6 g)
were introduced into an oleic acid solution in dried toluene (4 ml
oleic acid in 30 ml THF) and ultrasonicated for 3 h.
2.2.6. Synthesis of N-(2-aminoethyl) folic acid (EDA-FA) (Fig. 2)
2.2.6.1. Protection of an amino group of ethylene diamine (EDA) with
t-butyl carbamate (Boc). A solution of di-t-butyl dicarbonate (Boc2O)
(0.23 ml, 1 mmol) in anhydrous CH2Cl2 (10 ml) was added dropwise
to a cold solution of ethylene diamine (EDA) (0.67 ml, 10 mmol) in
anhydrous CH2Cl2 (10 ml) at 0 ꢀC under nitrogen atmosphere. The
mixture was magnetically stirred at 0 ꢀC for 2 h and at room
temperature for 24 h. Then, distilled water (5 ml) was added into the
mixture to dissolve the precipitate. The organic layer was washed
with brine (15 ml) 5 times, dried over anhydrous Na2SO4, and then
concentrated under reduced pressure to give t-butyl N-(2-amino-
ethyl) carbamate (EDA-Boc), appearing as thick oil (82% yield). 1H
NMR (400 MHz, CDCl3) dH: 1.40 [s, 9H, CH3 Boc], 2.80 [m, 2H,
2.2.2. Synthesis of 2-bromo-2-methyl-N-(3-(triethoxysilyl)
propanamide (BTPAm))
To a stirred solution of 3-aminopropyl triethoxysilane (APS)
(0.18 ml, 0.8 mmol) and triethylamine (TEA) (0.12 ml, 0.8 mmol) in
dried toluene (10 ml), 2-bromoisobutyryl bromide (BIBB) (0.1 ml,
0.8 mmol) in dried toluene (10 ml) was added dropwise at 0 ꢀC for 2 h
under nitrogen. The reaction mixture was warmed to room temper-
ature and stirred for 24 h. The mixture was passed through a filter
paper to remove salts and the filtrate was evaporated to remove the
unreacted TEA under reduced pressure. The resulting product,
CH2eNH2], 3.20[m, 2H, CH2eCH2eNH-Boc]. FTIR (KBr disc) ymax
:
3360 cmꢁ1 (NH stretching), 2955e2923 cmꢁ1 (CeH stretching),
1693 cmꢁ1 (C]O of amide stretching), 1525 cmꢁ1 (NH bending),
1366-1277 cmꢁ1 (CeN bending), 1172 cmꢁ1 (CeO stretching).

M. Rutnakornpituk et al. / Polymer 52 (2011) 987e995
989
2.2.6.2. Coupling folic acid with the amino-protected EDA. To a stir-
red solution of FA (0.275 g, 6.25 10ꢁ4 mol) in anhydrous DMSO
(5 ml) and pyridine (4 ml), the solution of EDA-Boc (0.10 g,
6.25 10ꢁ4 mol) and DCC (0.21 g, 7.5 10ꢁ4 mol) in anhydrous DMSO
(5 ml) were added. The mixture was stirred at room temperature
for 18 h under nitrogen blanket. After the reaction completed, the
mixture was gradually poured into a vigorously stirred diethyl
ether (20 ml) at 0 ꢀC. The yellow precipitate was collected and
washed with cold diethyl ether several times and dried under high
vacuum to obtain {t-butyl N-(2-aminoethyl) carbamate} folic acid
(Boc-EDA-FA), appearing as a yellow solid (85% yield). 1H NMR
(400 MHz, DMSO-d6) dH: 1.40 [s, 9H, CH3 Boc], 2.0 [m, 2H,
CH2eCH2eCOeNH], 2.40 [m, 2H, CH2eCOeNH], 2.90 [m, 2H,
CH2eNHeCO], 3.10 [m, 2H, CH2eNHeBoc], 4.30 [m, 1H,
HOOCeCHeNH], 4.50 [d, 2H, phenyl-NH-CH2 folic acid], 6.60 [d,
J ¼ 8 Hz, 2H, 2CH]CH phenyl folic acid ], 6.90 [t, 1H, phenyl-NH-
CH2], 7.60 [d, J ¼ 8 Hz, 2H, 2CH]CH phenyl folic acid], 8.60 [s, 1H,
N]CH Ar folic acid]. FTIR (KBr disc) ymax: 3360e2600 cmꢁ1 (OH
and NH stretching), 1700 cmꢁ1 (C]O of amide stretching),
1605 cmꢁ1 (CeO of acid stretching), 1168 cmꢁ1 (CeO stretching).
TFA (2 ml) was then added to Boc-EDA-FA and stirred at room
temperature. After 2 h stirring, TFA was removed under reduced
pressure and the resulting residue was dissolved in anhydrous DMF.
Pyridine was added until a formation of yellow precipitate and it was
subsequently washed with diethyl ether and dried to give N-(2-
aminoethyl) folic acid (EDA-FA) (80% yield, Tm 290 ꢀC). 1H NMR
(400 MHz, DMSO-d6) dH: 2.0 [m, 2H, CH2eCH2eCOeNH], 2.40 [m,
2H, CH2eCOeNH], 2.60 [m, 2H, CH2eNHeCO], 3.30 [m, 2H,
CH2eNH2], 4.20 [m, 1H, HOOC-CH-NH], 4.40 [d, 2H, phenyl-NH-CH2
folic acid], 6.60 [d, J ¼ 8 Hz, 2H, 2CH]CH phenyl folic acid], 6.90 [t,
1H, Phenyl-NH-CH2], 7.70 [d, J ¼ 8 Hz, 2H, 2CH]CH phenyl folic
stretching), 1605 cmꢁ1 (CeO of acid stretching), 1532e1335 cmꢁ1
(CeN bending), 1202e1132 cmꢁ1 (CeO stretching).
2.2.7. Immobilization of folic acid on the surfaces of poly(AA)-
coated MNPs
Poly(AA)-coated MNPs were dispersed in a 10 ml aqueous
solution containing NHS (40 mg) and EDC.HCl (20 mg) and the
mixture was kept in a dark place for 2 h. The particles were
recovered, washed with water and dried in vacuo. Then, the parti-
cles were added in a solution of 200 mg of EDA-FA and 50 mg of
EDC in 10 ml anhydrous DMSO. The suspension was agitated
overnight at 37 ꢀC in dark. The particles were then recovered,
washed with DMSO and methanol several times and dried in vacuo.
2.3. Characterization
FTIR was performed on a PerkineElmer Model 1600 Series FTIR
Spectrophotometer. The solid samples were mixed with KBr to
form pellets. Nuclear magnetic resonance spectroscopy (NMR) was
performed on a 400 MHz Bruker NMR spectrometer using CDCl3 as
a solvent. Gel permeation chromatography (GPC) data was con-
ducted on PLgel 10 mm mixed B2 columns and a refractive index
detector. Tetrahydrofuran (THF) was used as a solvent with a flow
rate of 1 ml/min at 30 ꢀC. TEM were performed using a Philips
Tecnai 12 operated at 120 kV equipped with Gatan model 782 CCD
camera. TGA was performed on SDTA 851 Mettler-Toledo at the
temperature ranging between 25 and 600 ꢀC at a heating rate of
20 ꢀC/min under oxygen atmosphere. VSM was performed at room
temperature using a Standard 7403 Series, Lakeshore vibrating
sample magnetometer. The magnetic moment was investigated
over a range of applied magnetic fields from ꢁ10,000 to þ10,000 G
using 30 min sweep time. Hydrodynamic diameter of the particles
was measured via PCS using NanoZS4700 nanoseries Malvern
acid], 8.60 [s, 1H, N]CH Ar folic acid]. FTIR (KBr disc) ymax
:
3600e2800 cmꢁ1 (OH and NH stretching),1684 cmꢁ1 (C]O of amide
Fig. 1. Synthesis of poly(AA)-coated MNPs via ATRP reaction and immobilization of folic acid.

990
M. Rutnakornpituk et al. / Polymer 52 (2011) 987e995
instrument. The sample dispersions were sonicated for 10 min
before the measurement at 25 ꢀC. The presence of FA was investi-
gated using SPECORD S100 UVeVisible spectrophotometer (Ana-
lytikjena AG) coupled with a photo diode array detector at
lmax ¼ 371 nm.
Fig. 3 displays FTIR spectra of poly(t-BA)-coated MNPs with-
drawn from the dispersions at 1, 6, 12 and 24 h of ATRP reaction.
Because ATRP is known as a controlled radical polymerization, the
time period for the ATRP reaction is thus crucial for tuning the
molecular weight of the polymers. A progressive growth of ester
linkage signals (eO(C]O)- stretching, w1724 cmꢁ1 and CeO
stretching, w1147 cmꢁ1) of t-BA repeating units in relative to those
3. Results and discussion
of a SieO signal of the linker (w1100e1020 cmꢁ1 and w800 cmꢁ1
)
The aim of this work is to modify MNP surfaces with poly(AA)
and immobilize folic acid on their surfaces. Poly(AA) grafted on the
particle surfaces is thought to provide steric and electrostatic
stabilizations and dispersibility of the particles in aqueous media.
Another major advantage of this system was that the carboxylic
acid-enriched surfaces of poly(AA)-grafted MNPs provided a plat-
form for efficient surface immobilization of any functional mole-
cules such as DNA, drugs, protein and fluorescent molecules. Hence,
the novelty of this current work is that this is the first report on
synthesizing multifunctional poly(AA)-coated MNPs for attaching
folic acid (FA), a model molecule in this work. Precedents have
reported the immobilization of biomolecules on the distal ends of
MNPs coated with other polymeric surfactant [41e44]. This novel
system is hypothesized to increase the loading efficiency of FA on
the MNP surfaces.
To perform surface-initiated ATRP from MNPs, BTPAm, a mole-
cule containing an ATRP initiating site was first synthesized
through amidization between APS and BIBB, followed by silaniza-
tion of triethoxysilane of BTPAm on MNP surface. The results of the
synthesis of BTPAm including FTIR and 1H NMR are illustrated in
supplementary data. To immobilize BTPAm on MNP surfaces, bare
MNPs were first coated with oleic acid to form well dispersed MNPs
in toluene. The advantage of this procedure was that the MNPs
were well dispersible in the media before reacting with BTPAm,
allowing BTPAm to effectively silanize to their surfaces due to its
greater surface approaching ability in the dispersed MNPs.
indicated that the molecular weights of poly(t-BA) on MNP surfaces
increased as increasing ATRP reaction time. It should be noted that
the signal corresponding to Fe-O bonds from MNP core
(w589 cmꢁ1) were observed throughout the reactions without
significant change in its intensity.
Weight loss from TGA technique of poly(t-BA)-coated MNPs at
various ATRP reaction times was investigated to determine the
relative amount of poly(t-BA) that can be grafted on the particle
surface. It should be noted that the particles were separated from
the uncoordinated species using an external magnet. Using an
assumption that % char yield was the weight of magnetite
remaining at 600 ꢀC, the weight loss of the surface-modified MNPs
was thus attributed to the decomposition of organic components
including BTPAm and poly(t-BA) that complexed to the particle
surface. Hence, percent char yield of bare MNP and MNP coated
with BTPAm were determined to obtain percent of BTPAm in the
complexes in each sample. According to TGA results, percent of
BTPAm in the complexes was about 2 wt%, while percents of poly(t-
BA) were 3 wt%, 15 wt%, 26 wt% and 43 wt% of the complexes at 1, 6,
12 and 24 h ATRP reaction times, respectively (Fig. 4). This was
a supportive result to FTIR that poly(t-BA) chain length was pro-
longed when ATRP reaction time was extended.
To investigate the molecular weight and the molecular weight
distribution of poly(t-BA), small amount of ethyl bromoisobutylate
(EBiB) was added in the dispersion as a “sacrificial initiator” to form
free poly(t-BA) along with poly(t-BA) grafted on MNP. After 24 h of
Fig. 2. Synthesis of N-(2-aminoethyl) folic acid (EDA-FA).

M. Rutnakornpituk et al. / Polymer 52 (2011) 987e995
991
A
1724 cm-1
(-CO-O-)
1609 cm-1
(-NH-CO-)
3400 cm-1
NH stretching
2959-2922 cm-1
CH stretching
802 cm-1
(Si-O stretching)
1148 cm-1
(C-O stretching)
585 cm-1
Fe-O
1090, 1016 cm-1
(Si-O stretching)
B
3401 cm-1
NH stretching
1621 cm-1
(-NH-CO-)
801 cm-1
(Si-O stretching)
2963-2922 cm-1
CH stretching
1725 cm-1
(-CO-O-)
1147 cm-1
589 cm-1
Fe-O
(C-O stretching) 1095, 1019 cm-1
(Si-O stretching)
%T
801 cm-1
(Si-O stretching)
3392 cm-1
NH stretching
2963-2925 cm-1
CH stretching
1725 cm-1
(-CO-O-)
1618 cm-1
(-NH-CO-)
C
1148 cm-1
(C-O stretching)
1101, 1020 cm-1
(Si-O stretching)
589 cm-1
Fe-O
1105, 1023 cm-1
(Si-O stretching)
3368 cm-1
NH stretching
1614 cm-1
(-NH-CO-)
D
2974-2925 cm-1
CH stretching
1726 cm-1
(-CO-O-)
1148 cm-1
(C-O stretching)
800 cm-1
(Si-O stretching)
588 cm-1
Fe-O
4000
3600
3200
2800
2400
2000
1800
1600
1400
1200
1000
800
600
400
cm -1
Fig. 3. FTIR spectra of poly(t-BA)-coated MNP at various ATRP reaction times, A) 1 h, B) 6 h, C) 12 h and D) 24 h.
the reaction, the free poly(t-BA) was removed from the MNP
complex using an external magnet. According to GPC results,
molecular weight of poly(t-BA) was about 18,600 g/mol and its
molecular weight distribution was about 1.22. This narrow molec-
ular weight distribution indicated the living mechanism of
controlled radical polymerization. 1H NMR spectrum of free poly(t-
BA) is shown in supplementary data.
TEM images of MNP complexes at each step of the reaction are
illustrated in Fig. 5. Bare MNPs observed in Fig. 5A were well
organized because they were somewhat uniformed in size, which
was in the range of 6e10 nm in diameter with the average of about
8 nm. Surface modification of the MNPs resulted in a slightly
broader size distribution due to the presence of organic compounds
coated on their surface (Fig. 5BeD). However, the average particle
size was not significant difference from those of bare MNPs. It
should be noted that poly(t-BA)-coated MNPs were well dispersed
in toluene due to the existence of hydrophobic poly(t-BA) on their
surface (Fig. 5C), while poly(AA)-coated MNPs were well dispersed
in water because of the presence of hydrophilic and charge
surfactants of poly(AA) (Fig. 5D).
100
A
80
B
The M-H curves of bare MNP, BTPAm-coated MNP, poly(t-BA)-
coated MNPs and poly(AA)-coated MNP were illustrated in Fig. 6.
They showed superparamagnetic behavior at room temperature as
indicated by the absence of reminance and coercivity upon
removing an external applied magnetic field. According to the
results in Table 1, the decrease of saturation magnetization (Ms)
from 59 emu/g of bare MNPs to 27 emu/g of poly(t-BA)-coated
MNPs was attributed to the presence of the organic surfactant on
their surface, resulting in the decrease of percent of magnetite in
the complexes. After the hydrolysis of poly(t-BA) to form poly(AA)-
coated MNP, its Ms value increased from 27 to 39 emu/g sample due
to the removal of t-BA groups in poly(t-BA), which subsequently
increased percent of magnetite in the complexes. Interestingly,
when taking percent of magnetite in the complex into account, the
Ms values in emu/g magnetite basis of these complexes were not
C
60
D
40
20
0
0
100
200
300
400
500
600
700
Temperature (ºC)
Fig. 4. TGA thermograms of poly(t-BA)-coated MNPs at various ATRP reaction times,
A) 1 h, B) 6 h, C) 12 h and D) 24 h.

992
M. Rutnakornpituk et al. / Polymer 52 (2011) 987e995
Fig. 5. TEM images of A) bare MNPs, B) BTPAm-coated MNPs, C) poly(t-BA)-coated MNPs, D) poly(AA)-coated MNPs. In the TEM sample preparation, MNPs in Figure AeC were
dispersed in toluene and those in Figure D were dispersed in water.
significantly different from each other, indicating that magnetic
properties of the particles were not considerably affected upon
ATRP of poly(t-BA) and hydrolysis to form poly(AA)-coated MNPs.
After the hydrolysis reaction, it was conceived that MNPs having
carboxylic acid-enriched surfaces were obtained. These carboxylic
acid functional groups are readily reactive toward coupling reac-
tions with other molecules having affinity functional groups such
as amine and alcohol. In the current work, folic acid (FA) was
chemically immobilized on the surface-modified MNPs. FA has two
In the grafting reaction between poly(AA)-coated MNPs and
EDA-FA, N-hydroxyl succinimide (NHS) was used to activate the
dangling carboxylic acid groups. FTIR spectra of the products in
each step were thus illustrated in comparison with the starting
compounds (Fig. 7). Fig. 7A showed the FTIR spectrum of poly(AA)-
coated MNPs and those of NHS was depicted in Fig. 7B. In Fig. 7C,
the sharp and strong characteristic signal of ester linkages appeared
at 1723 cmꢁ1, indicating the coupling reaction between carboxylic
acid of poly(AA)-coated MNPs and NHS. In addition, Fe-O linkages
of magnetite core were also observed at 586 cmꢁ1. After the
coupling reaction with EDA-FA (Fig. 7D), the characteristic signals
of FA, such as 1700e1500 cmꢁ1 and 1153e1069 cmꢁ1, appeared in
the FA-bound MNPs (Fig. 7E), indicating the successful conjugation
of FA on the MNP surfaces.
UVevisible spectrophotometry was also applied to confirm the
presence of FA in the conjugated MNP complex. FA showed a lmax
value at 371 nm (Fig. 8A), whilst those of FA-conjugated MNPs also
exhibited a weak absorbance signal at the same wavelength
(Fig. 8B). It is worth to mention that poly(AA)-coated MNPs before
FA loading did not show any absorbance signal at the same wave-
length (Fig. 8C). This result implied that FA was, to some extent,
covalently conjugated to the MNP surfaces.
carboxylic acid groups at the
lently react with amino functional groups of EDA. However, it has
already been verified that -COOH is more accessible to covalently
a and g positions, which can cova-
g
react with amino groups due to its high reactivity [46,47]. FA needs
to be first activated with ethylene diamine (EDA) to obtain primary
amine-terminated FA (N-(2-aminoethyl) folic acid or EDA-FA). This
logical strategy enhanced the reactivity of FA to efficiently react
with carboxylic acid overexpressed on the surface of poly(AA)-
coated MNPs through amidization reaction. Results of the synthesis
of EDA-FA including FTIR and 1H NMR spectra were detailed in
supplementary data.
To determine percentage of magnetite core and organic shell in
the complexes in each step of the reactions, the complexes were
characterized via TGA to investigate their mass loss. Bare MNPs
manifested a drastic weight loss between 200 and 350 ꢀC with 90%
Table 1
Percentage of magnetite in the complex and their magnetic properties.
emu/g samplea
% Fe3O4
emu/g Fe3O4
b
Sample
Bare MNP
59
53
27
39
90
88
45
63
65
60
61
62
BTPAm-coated MNP
Poly(t-BA)-coated MNP
Poly(AA)-coated MNP
a
Estimated from the saturation magnetization (Ms) at 10,000 G from VSM
Fig. 6. M-H curves of A) bare MNP, B) BTPAm-coated MNP, C) poly(t-BA)-coated MNP
and D) poly(AA)-coated MNP.
technique.
b
Estimated from % char yield at 600 ꢀC from TGA technique.

M. Rutnakornpituk et al. / Polymer 52 (2011) 987e995
993
Fig. 7. FTIR spectra of (A) poly(AA)-coated MNP, (B) NHS, (C) NHS-poly(AA)-coated MNP, (D) EDA-FA and (E) FA-poly(AA)-coated MNP.
char yield (Fig. 9A). This was attributable to the decomposition or
desorption of the absorbed ammonium salt at elevated tempera-
ture and eventually loss some weight [48,49]. The weight loss of
MNPs coated with BTPAm, poly(t-BA) and poly(AA) were attributed
to the decomposition of organic components complexing to the
particle surface and % char yields were the weight of magnetite
core. From TGA thermograms in Fig. 9B,C, there was about 2 wt% of
BTPAm and 49 wt% of poly(t-BA) in poly(t-BA)-coated MNPs. The
0.3
100
A
B
0.25
80
A
0.2
D
60
E
0.15
C
B
40
0.1
20
0
0.05
C
0
0
100
200
300
400
500
600
700
320
340
360
380
400
420
440
Temperature (°C)
wavelength (nm)
Fig. 8. UVevisible spectra of (A) folic acid (FA), (B) FA-conjugated MNP and (C) poly
Fig. 9. TGA thermograms of (A) bare MNP, (B) BTPAm-coated MNP, (C) poly(t-BA)
(AA)-coated MNP without FA.
coated MNP, (D) poly(AA)-coated MNP and (E) FA-poly(AA)-coated MNP.

994
M. Rutnakornpituk et al. / Polymer 52 (2011) 987e995
Fig. 10. The effect of pH of the aqueous dispersions containing poly(AA)-coated MNP (A) and FA-poly(AA)-coated MNP (-) on their hydrodynamic diameter and zeta potential. The
experiments were performed at 25 ꢀC.
grafting density of BTPAm, the initiating site for ATRP on the
particle, can be calculated and it was found that there was about 0.8
molecule/nm2 (150 molecules/particle). Examples of the calcula-
tion are illustrated in supplementary data. The grafting density of
poly(t-BA) were comparable to that of BTPAm on the surface.
After the hydrolysis of poly(t-BA), percent of organic components
in the case of poly(AA)-coated MNPs significantly dropped (from 49
to 27 wt%) due to the removal of t-BA groups from the polymeric
layer of the particles (Fig. 9D). The 27 wt% of poly(AA) corresponded
to 23 carboxylic acid/nm2 (4600 acid/particle). From Fig. 9E, there
was about 14 wt% of FA in the complex, corresponding to about 2 FA
molecules/nm2 (400 FA molecules/particle). Therefore, percent
conversion of carboxylic acid to FA was about 8%. The lowering
temperature of TGA curve in Fig. 9E (FA-poly (AA)-coated MNP) as
compared to that in Fig. 9D (poly (AA)-coated MNP) was attributed to
the weight loss of FA component in the complex. The decomposed
TGA thermogram of free FA has been investigated and shown in
supplementary data. In addition, it was also found that there was
about 2.7 FA molecules/site of the ATRP initiator (400 FA molecule/
150 sites of BTPAm in a single particle). The limited number of the %
conversion and grafting density of FA on the particle surface was
attributed to limited accessibility of bulky FA to react with steric poly
(AA). However, the grafting density of FA might be improved by
copolymerization of poly(AA) with other polymers to lessen steric
hindrance of the compact poly(AA), so that FA con be more effec-
tively conjugated. Also, utilization of spacer from the particle surface
is another approach that can diminish steric hindrance on the dense
surface.
Because carboxylic acid functional groups can be easily ionized
in an aqueous solution, it is thus interesting to understand how pH
of the dispersions affect hydrodynamic diameter and surface
charge of poly(AA)-coated MNPs and FA-poly(AA)-coated MNP. pH
of the aqueous dispersions containing the complexes (0.2 mg/ml)
were varied from approximately 1e11 and their hydrodynamic
diameters were determined via PCS technique. In both samples, as
pH of the dispersions increased, their hydrodynamic diameters
rapidly decreased at acidic pH (ranging between pH 1.2e5.4) and
gradually decreased at pH ranging between 5.4 and 11.3 (Fig. 10). It
was hypothesized that as increasing pH of the dispersions, ioniza-
tion of carboxylic acid on the surface of poly(AA)-coated MNPs took
place, resulting in the formation of carboxylate ions on their
surfaces. The negative charges of carboxylate ions led to additional
electrostatic repulsion toward neighboring particles and thus pre-
vented massive flocculation.
other at pH ranging between 1.2 and 6.5. This was attributed to the
presence of amines in FA structure, resulting in protonated amino
groups. Similarly, the enriched amines in the complex might also
influence the lower zeta potential in FA-containing complex at
basic pH.
The large size of the particles in DLS as compared to those from
TEM measurements (8 nm in diameter) might come from the fact
that there were some nano-clusters of particles in the dispersions.
These nano-clusters of the particles can be observed in TEM
measurements from the first step of the particle synthesis (shown
in supplementary data). When poly(AA) was chemically grafted on
their surface, these clusters still presented. Although these nano-
clusters existed in the dispersions, the particles were well
dispersible in aqueous dispersions without macroscopic aggrega-
tion visibly observed because there were poly(AA) coated on their
surface.
Cytotoxicity testings of poly(AA)-coated MNP and FA-poly(AA)-
coated MNP were also performed. According to our preliminary
results, it was found that the dispersions were not toxic against
Vero cell line up to 50
mg/ml concentration of the sample (sulfo-
rhodamine B (SRB) assay method). Detail studies regarding the
toxicity of the magnetite complexes are warranted for future
studies.
4. Conclusions
This work presented a “grafting from” strategy to modify MNP
surfaces with poly(t-BA) via ATRP, followed by a hydrolysis of t-BA
groups to obtain poly(AA) and finally immobilization of folic acid on
their surfaces. The originality of this work is that this is the first
report on modifying MNP surface with poly(AA) which serves as
a platform for folic acid immobilization. Because the folate receptor
is overexpressed on the surface of cancer cells, it is for this reason
that folic acid is of particular interest in the current work in an
attempt to facilitate the intracellular uptake by specific cancer cells
for cancer therapy. Folic acid was successfully activated with
ethylene diamine (EDA) to obtain primary amine-terminated folic
acid derivative. This logical strategy enhanced the reactivity of folic
acid to be efficiently immobilized on MNP surfaces through amid-
ization reaction.
In addition to the use of binding affinity of carboxylic acid
functional groups, poly(AA) on their surface can also provide
stabilization mechanisms through both steric repulsion due to the
long chain polymers and electrostatic repulsion owing to the
formation of negative charges in basic pH dispersions. Furthermore,
poly(AA) on their surfaces also promoted particle dispersibility in
water, which is a minimum requirement for biomedical uses. This
novel magnetically guidable nanocomplex might be suitable for use
as an efficient drug delivery vehicle particularly for cancer
treatment.
The results from zeta potential measurements also supported
this assumption. The surface charges of poly(AA)-coated MNPs
were positive at the pH ranging between 1.2 and 6.5 and negative at
the pH range of 6.5e11.3, implying that point of zero charge (PZC) of
this complex was pH 6.5 (Fig. 10). It was also found that FA-con-
taining complex showed a slightly higher zeta potential than the

M. Rutnakornpituk et al. / Polymer 52 (2011) 987e995
[12] Feng PB, Gao F, Gu HC. J Colloid Interface Sci 2005;284:1e6.
995
Acknowledgement
[13] Liu X, Xing J, Guan Y, Shan G, Liu H. Colloids Surf A Physicochem Eng
2004;238:127e31.
The authors thank the Thailand Research Fund (TRF) and Nar-
esuan University (DBG5380001) for financial funding. The Center of
Excellence for Innovation in Chemistry (PERCH-CIC), Commission
on Higher Education, Ministry of Education is also gratefully
acknowledged for financial support.
[14] Mornet S, Vekris A, Bonnet J, Duguet E, Grasset F, Choy JH, et al. Mater Lett
2000;42:183e8.
[15] Cheng G, Zhao J, Tu Y, He P, Fang Y. Anal Chem Acta 2005;533:11e6.
[16] Lellouche J, Senthil G, Joseph A, Buzhansky L, Bruce I, Bauminger ER, et al. J Am
Chem Soc 2005;127:11998e2006.
[17] Corr SA, Rakovich YP, Gun’ko YK. Nanoscale Res Lett 2008;3:87e104.
[18] Zhang Y, Kohler N, Zhang M. Biomaterials 2002;23:1553e61.
[19] Sonvico F, Mornet S, Vasseur S, Dubernet C, Jaillard D, Degrouard J, et al.
Bioconjug Chem 2005;16:1181e8.
[20] Zhoua Y, Wang S, Ding B, Yang Z. Chem Eng J 2008;138:578e85.
[21] Fischer H. Chem Rev 2001;101:3581e610.
Appendix. Supplementary data
Proposed reaction mechanisms of trifluoroacetic acid (TFA) with
the polymers on MNP surface. FTIR and 1H NMR spectra of 2-
bromo-2-methyl-N-(3-(triethoxysilyl) propanamide (BTPAm)) and
N-(2-aminoethyl) folic acid (EDA-FA). 1H NMR spectrum of poly(t-
BA), TEM images showing some nano-aggregation. Examples of
calculation of grafting density. TGA thermogram of folic acid.
[22] Pyun J, Jia S, Kowalewski T, Patterson GD, Matyjaszewski K. Macromolecules
2003;36:5094e104.
[23] Hu B, Fuchs A, Huseyin S, Gordaninejad F, Evrensel C. Polymer 2006;47:
7653e63.
[24] Zhao H, Shipp DA. Chem Mater 2003;15:2693e5.
[25] Liu G, Yan X, Lu Z, Curda SA, Lal J. Chem Mater 2005;17:4985e91.
[26] Sun Y, Ding X, Zheng Z, Cheng X, Hu X, Peng Y. Euro Polym J 2007;43:762e72.
[27] Hermann High LR, Holder SJ, Penfold HV. Macromolecules 2007;40:7157e65.
[28] Shipp DA, Wang JL, Matyjaszewski K. Macromolecules 1998;31:8005e8.
[29] Xia J, Gaynor SG, Matyjaszewski K. Macromolecules 1998;31:5958e9.
[30] Teodorescu M, Matyjaszewski K. Macromolecules 1999;32:4826e31.
[31] Raghuraman GK, Rühe J, Dhamodharan R. Nanopart Res 2008;10:415e27.
[32] Czaun M, Hevesi L, Takafuji M, Ihara H. Chem Commun; 2008:2124e6.
[33] Marutanil E, Yamamoto S, Ninjbadgar T, Tsujii Y, Fukuda T, Takano M. Polymer
2004;45:2231e5.
Appendix. Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.polymer.2010.12.059.
[34] Zhou Y, Wang S, Ding B, Yang Z. Chem Eng J 2007;40:6217e23.
[35] Fan QL, Neoh KG, Kang ET, Shuter B, Wang SC. Biomaterials 2007;28:5426e36.
[36] Frimpong RA, Hilt JZ. Nanotechnology 2008;19:1e7.
[37] Si S, Kotal A, Mandal TK, Giri S, Nakamura H, Kohara T. Chem Mater
2004;16:3489e96.
References
[1] Pei W, Kumadaa H, Natusmea T, Saitoa H, Ishio S. J Magn Magn Mater
2007;310:2375e7.
[2] Sun S, Murray CB, Weller D, Folks L, Moser A. Science 2000;287:1989e92.
[3] Teng X, Yang H. J Am Chem Soc 2003;125:14559e63.
[4] Casula MF, Floris P, Innocenti C, Lascialfari A, Marinone M, Corti M, et al. Chem
Mater 2010;22:1739e48.
[5] Anders S, Sun S, Murray CB, Rettner CT, Best ME, Thomson T, et al. Micro-
electron Eng 2002;61-62:569e75.
[38] Zhang M, Breiner T, Mori H, Muller AHE. Polymer 2003;44:1449e58.
[39] Wan S, Zheng Y, Liu Y, Yan H, Liu KJ. Mater Chem 2005;15:3424e30.
[40] Sauzedde F, Elaïssari A, Pichot C. Colloid Polym Sci 1999;277:846e55.
[41] Ng V, Lee YV, Chen BT, Adeyeye AO. Nanotechnology 2002;13:554e8.
[42] Hayashi K, Moriya M, Sakamoto W, Yogo T. Chem Mater 2009;21:1318e25.
[43] Kohler N, Fryxell GE, Zhang M. J Am Chem Soc 2004;126:7206e11.
[44] Landmark KJ, DiMaggio S, Ward J, Kelly C, Vogt S, Hong S, et al. ACS Nano
2008;2:773e83.
[6] Woo K, Hong J, Choi S, Lee HW, Ahn JP, Kim CS, et al. Chem Mater
2004;16:2814e8.
[7] Zhang J, Misra RDK. Acta Biomater 2007;3:838e50.
[8] Laurent S, Forge D, Port M, Roch A, Robic C, Elst LV, et al. Chem Rev
2008;108:2064e110.
[9] Veiseh O, Gunn JW, Zhang M. Adv Drug Deliv Rev 2010;62:284e304.
[10] Sahoo Y, Goodarzi A, Swihart MT, Ohulchanskyy TY, Kaur N, Furlani EP. J Phys
Chem 2005;109:3879e85.
[45] Pinna N, Grancharov S, Beato P, Bonville P, Antonietti M, Niederberger M.
Chem Mater 2005;17:3044e9.
[46] Mohapatra S, Mallick SK, Maiti TK, Ghosh SK, Pramanik P. Nanotechnology
2007;18:385102.
[47] Park EK, Lee SB, Lee YM. Biomaterials 2005;26:1053e61.
[48] Rutnakornpituk M, Meerod S, Boontha B, Wichai U. Polymer
2009;50:3508e15.
[11] Hayashi K, Ono K, Suzuki H, Sawada M, Moriya M, Sakamoto W, et al. Chem
Mater 2010;22:3768e72.
[49] Tao K, Dou H, Sun K. Chem Mater 2006;18:5273e8.