Polymer 76 (2015) 271e279
Contents lists available at ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
Facile synthesis of gold/polymer nanocomposite particles using polymeric amine-based particles as dual reductants and templates Noel Peter Bengzon Tan a, Cheng Hao Lee a, Lianghui Chen a, Kin Man Ho a, Yan Lu b, Matthias Ballauff b, Pei Li a, * a b
Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hung Hom, Kowloon, PR China F-I2 Soft Matter and Functional Materials, Helmholtz-Zentrum Berlin für Materialien und Energie GmbH, Hahn-Meitner-Platz 1, 14109 Berlin, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 June 2015 Received in revised form 31 August 2015 Accepted 7 September 2015 Available online 9 September 2015
Herein we report a facile synthesis of gold nanoparticle/polymer nanocomposite particles through a spontaneous reduction of tetrachloroauric (III) acid and encapsulation of resultant gold nanoparticles using amine-rich polymeric particles in water. The particle consisted of coreeshell nanostructure with a dense layer of polyethyleneimine (PEI) shell. The hydrophilic PEI shell acted as a reductant to generate gold nanoparticles (AuNPs), while the particle as a polymer template to in-situ encapsulate and stabilize the resultant AuNPs, giving a stable gold/polymer nanocomposite particle in water. The PEI-based core eshell particles with different degree of softness were found to have little inﬂuence on the reduction ability of the gold salt, but considerablely affect the encapsulating capability. Increasing the level of softness of particle core gave higher encapsulation efﬁciency to the gold nanoparticles. A soft type of core eshell microgel, namely poly(N-isopropyl acrylamide)/polyethyleneimine (PNIPAm/PEI) was further investigated with respect to the reduction rate, encapsulation efﬁciency, as well as nanocomposite stability and properties. Results indicated that reduction rate of gold salts using the PNIPAm/PEI microgels as the reductant was two orders of magnitude faster than that of the native PEI. Solution pH and amine to gold salt ratio also affected the formation of [email protected]
/PEI composite particles. Our results demonstrate that the use of polymeric amine-based particles is a simple and green synthesis of [email protected]
nanocomposite particles in aqueous system which is free from organic solvent, reducing and stabilizing agents. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Nanocomposites Gold nanoparticle Core-shell microgel
1. Introduction Gold/polymeric composite particles have attracted much research interests because of their synergistic effects of the gold nanoparticles on optical and electronic properties, biological compatibility and nano-scale size, as well as the polymer which enables the control of the composite particles through inducing chemical and physical changes of the polymer and surface functionality. The combined chemical and physical properties of the gold/polymer composite particles have opened up many potential applications in various emerging ﬁelds including sensing [1,2], biomedicine [3,4], and catalysis . The gold/polymer composite particles can be synthesized using different types of polymers such
* Corresponding author. E-mail addresses: [email protected]
(Y. Lu), [email protected]
(P. Li). http://dx.doi.org/10.1016/j.polymer.2015.09.015 0032-3861/© 2015 Elsevier Ltd. All rights reserved.
as homopolymer , block copolymers [7e9], and colloidal particles [10e14]. Among the various polymer mediated synthesis, amphiphilic particles with polyelectrolyte brushes are of particular interest because the hydrophilic brush shells can act as both nanoreactors and templates for gold nanoparticle formation and insitu encapsulation of the resultant metallic nanoparticles. Lu et al. have reported a type of coreeshell particle composing of a solid polystyrene core grafted with long cationic polyelectrolyte chains of poly(2-aminoethyl methacrylate hydrochloride) (PAEMH) [15,16]. When dispersing such type of particles in water, the polyelectrolyte shells become highly swollen, giving dense layer of polymer chains on the particle surface. Their counterions could reside within the shell layer of the coreeshell particles. When [AuCl4] ions are added into the particle dispersion, they could undergo metal ion exchange with the counterions, resulting in conﬁning the negatively charged gold ions within the cationic brush-like shells. This type of colloidal particles was synthesized via ﬁrst producing a hydrophobic core, e.g. polystyrene (PS)
N.P.B. Tan et al. / Polymer 76 (2015) 271e279
through a conventional emulsion polymerization, followed by attaching the photo initiator on the surface of the core. The watersoluble macromonomer was then polymerized from the surface of the core to form tree-like structures. However, such method of producing gold/polymer composite particles can be tedious and expensive. They also need to use stabilizer as well as toxic and highly reactive reducing agents to generate gold nanoparticles. These drawbacks would limit many potential applications of the gold/polymer composite particles. Recently, Agrawal et al. reported a reducing agent free synthesis of gold nanoparticles using poly(Nvinyl caprolactam-co-acetoacetoxyethyl methacrylate-co-acrylic acid) P(VCL-AAEM-AAc) microgels as the template . Gold nanoparticles were conﬁned within the core region of the template particles. It was speculated that b-diketone groups of AAEM located in the core was responsible for the reduction of gold salt to nanoparticles due to its electron rich atoms from keto-enol tautemerization. Amines are a particularly attractive class of compounds because of their almost universal presence in many biological and environmental systems. Amine-containing polyelectrolytes have been reported to act as dual reductants and stabilizers in gold nanoparticle synthesis [18e23]. Mechanistic studies suggest that tertiary amines are most effective in reducing Au3þ, while secondary amines are more effective in providing the necessary coordination for stabilizing gold nanoparticles as they are formed. Primary amines can form quaternary ammonium sites in water, thus capable of attracting negative gold salt precursors through electrostatic interaction . Apart from functioning as reductant and source of binding sites, polymeric amines can also act as capping agent to limit the growth and agglomeration of the metallic nanoparticles. Furthermore, polymeric amines are pH-sensitive, thus inter-particle distance between those immobilized gold nanoparticles can be altered through varying solution pH, giving different optical property and morphology of the nanocomposite particles. Among various types of polymeric amines, branched polyethyleneimine (PEI), a cationic polymer comprising 25, 50 and 25% primary, secondary and tertiary groups, respectively, has received much attention in the past two decades because it is one of the most potent non-viral polymeric vectors for gene delivery . Our group has already synthesized a variety of PEI-based coreeshell particles and studied their formation mechanism [26e28]. We have also developed diverse applications using these coreeshell particles because of their unique chemical and physical properties such as high surface area to volume ratio, well-deﬁned coreeshell nanostructure with tunable composition, surface functionality, and good water dispersibility [29,30]. As part of our continuous development in coreeshell particle platform technology, we envision that the amine-rich coreeshell particles could act as multifunctional polymer particles for gold nanoparticle synthesis and spontaneous immobilization. Thus the gold/polymer composite particles could be generated in the absence of any toxic reducing and stabilizing agents. Therefore, this work aimed at using PEI-based coreeshell particles as a green and efﬁcient platform for generating gold/polymer composite particles in water. The coreeshell particles in this system play dual functions: as nanoreactor and reductant to generate AuNPs, and as a polymer template to spontaneously encapsulate and stabilize the resulting AuNPs, forming stable gold/polymer composite particles in an aqueous. 2. Experimental section 2.1. Materials Branched poly(ethyleneimine) (PEI) (50% solution in water with
a weight average molecular weight of 750,000), N,N0 -methylenebisacrylamide (MBA) and tert-butyl hydroperoxide (TBHP, 70% solution in water) were all purchased from Sigma Aldrich Chemical Co., and used without further puriﬁcation. Methyl methacrylate (MMA, Aldrich) was puriﬁed by washing it three times with 10% sodium hydroxide solution (volume ratio of MMA to NaOH solution was 10: 1), followed by repeated washing with deionized water (volume ratio of MMA to water was 5: 1) until the pH of the aqueous phase dropped to neutral. Spindle-crystals of N-isopropylacrylamide (NIPAm, Aldrich) were puriﬁed by repeated recrystallization of the NIPAm monomer using a mixture of toluene and n-hexane (1:5 v/v). Hydrogen tetrachloroaurate (III) trihydrate (HAuCl4$3H2O) was purchased from AldricheSigma and used as received. Deionized water or Milli-Q water was used for dilution and dispersion medium. 2.2. Synthesis of gold/polymer composite particles Three types of PEI-based particles were used to prepare Auloaded composite particles through a simple mixing with hydrogen tetrachloroaurate(III) trihydrate (HAuCl4$3H2O) solution according to the following general procedure: Hydrogen tetrachloroaurate (III) trihydrate stock solution (1.317 103 M) was purged with N2 for 30 min. This gold salt solution (1 mL) was added dropwise into the PNIPAm/PEI microgel dispersion (20 mL, 400 ppm, molar ratio of N/Auþ3 was 28). The mixture was stirred at 250 rpm for 4 h at 25 C. Kinetic experiments were carried out for 96 h to monitor the reduction rate proﬁle of the gold ions with either PEI or PEI-based microgels with the same PEI content. The resultant gold-loaded microgels were puriﬁed by a single cycle of centrifugation at 12,000 rpm for 1 h at 10 C. The suspended pink product was redispersed into deionized water under sonication. A similar procedure was used for other types of PEI-based particles such as PMMA/PEI coreeshell particles and PEI-g-PMMA hollow particles using amine to Au3þ molar ratios of 19:1 and 30:1, respectively. 2.3. Measurement and characterization 2.3.1. Particle size and surface charge measurement The hydrodynamic diameter and zeta-potential of the template particles and gold-loaded composite particles were determined with a Beckman Coulter Delsa Nanoparticle analyzer using a photon correlation spectroscopy with electrophoretic dynamic light scattering (a two-laser diode light source with a wavelength of 658 nm at 30 mW). Hydrodynamic diameter, Dh, was obtained from the Einstein Stokes equation, Dh ¼ kT/3phD, where k is the Boltzmann constant, h is the dispersant viscosity, T is the temperature (K), and D is the diffusion coefﬁcient obtained from the decay rate of the intensity correlation function of the scattered light (i.e., correlogram), G(t) ¼ !I(t)I(t þ t)dt. Surface charges of particles were measured based on their electrophoretic mobility. Samples for zetaepotential measurement were diluted to 100e200 ppm with 1 mM NaCl solution and measured at 25 C. Each value was an average of triplicate measurement. 2.3.2. Scanning electron microscopy Scanning electron microscope (SEM) samples were examined with a JEOL-JSM 6335 ﬁeld emission scanning electron microscope at an accelerating voltage of 5 kV. A diluted sample (around 200 ppm) was spread onto a glass substrate, followed by air-dried in a covered, free-dust environment. A thin layer of gold ﬁlm with 2e5 nm thickness was sputtered on the dried sample under vacuum.
N.P.B. Tan et al. / Polymer 76 (2015) 271e279
2.3.3. Transmission electron microscopy Transmission electron microscopy (TEM) images of pure PNIPAm/PEI microgel and gold-loaded PNIPAm/PEI composite particles were observed using a transmission electron microscope (JEOL 100 CX) at an accelerating voltage of 100 kV. The TEM images of goldloaded PMMA/PEI composite particles were observed by another TEM (Philips CM30) with an acceleration voltage of 300 kV. The high resolution TEM micrographs of single Au nanoparticle were characterized by a JEOL 2010 TEM at an accelerating voltage of 200 kV. The samples were generally prepared by wetting carboncoated grids with 10 mL of the diluted particle dispersion, followed by drying at room temperature prior to TEM analysis. For gold-loaded composite samples, no staining treatment of the samples were preformed, while pure PEI-based templates were all stained with 2 wt. % phosphotungstic acid (PTA) for 1 min, followed by drying at room temperature. Particle sizes of gold nanoparticles imbedded in the polymer template were estimated based on TEM images of the composite particles using image processing software (Image J v. 1.47, NIH, USA). Detailed procedure is described in the Supporting Information. 2.3.4. X-ray photoelectron spectroscopy X-ray photoelectron spectroscopy (XPS) data were recorded on a multi-surface analyses system (PHI 5600, Physical Electronics) with a monochromatic AlKa X-ray source (1486.6 eV) under various spot sizes from 200 to 800 mm in diameter. The pass energies of exciting radiations were set at 187 and 45 eVs for survey and elemental scans, respectively. The energy and emission currents of the electrons were 4 eV and 0.35 mA, respectively. Energy resolution was at 0.7 eV with a chamber pressure of 5 1010 torr. Spectral calibration was established by setting the C 1s component at 285.0 eV. All data acquisition was processed with a PC-based Avantage software (version 1.85). The surface composition was determined using the manufacturer's sensitivity factors. Curve ﬁtting of the spectrum was accomplished using a nonlinear least-squares method. A Gaussian function was assumed for the curve-ﬁtting. The deconvolution of nitrogen peak was processed with MagicPlot software (version 2.5.1). 2.3.5. UVeVis spectroscopy UVeVis spectra of gold/polymer composite samples were recorded on a Varian Cary 4000 Spectrophotometer using wavelengths ranging from 350 to 900 nm with an absorbance set from 0 to 0.70 a.u. Samples were diluted to appropriate concentrations and measured in a 5 mL cuvette. The kinetics of gold nanoparticle formation was monitored by a UVeVis spectroscopy at the absorbance wavelength of 525 nm. The actual absorbance intensity was plotted and a ﬁtted curve was derived from an analysis-ﬁtting function of Origin Pro 8.0 software. The equation of ﬁtted linear curve gave the rate through its slope (dy/dx), which is the change in absorption per hour (DA/hr). 2.3.6. X-ray diffraction The X-ray diffraction (XRD) patterns of the gold nanoparticles were collected from a high resolution X-ray diffractometer (HRXRD) (Rigaku SmartLab) using a CuKa radiation (l ¼ 1.541 Å) with a scan rate of 2 /min. The X-ray tube generator was operated at 45 kV and 200 mA with a graphite monochromator. Sample was prepared by a repeated deposition of [email protected]
/PEI composite solution on a glass substrate and air dried for 24 h. 3. Results and discussion Three types of PEI-based coreeshell particles with hard, soft and hollow cores were ﬁrst investigated in order to determine the most
appropriate template, which is capable of effective reduction of gold salt with high loading capacity of the resultant gold nanoparticles. The three types of PEI-based templates are: a) coreeshell particles with hard poly(methyl methacrylate) (PMMA) cores (PMMA/PEI); b) ultra-soft hollow particles consisting of PEI-gPMMA copolymer prepared fabricated from the same type of PMMA/PEI coreeshell particles except removing the PMMA homopolymer from the particle core; and c) coreeshell microgels with temperature-sensitive poly(N-isopropylacrylate) (PNIPAm) cores (PNIPAm/PEI). These particles were synthesized according to our previously established methods [26,27,31]. A schematic diagram of the synthesis of coreeshell PNIPAm/PEI is presented in Date in Brief . Fig. 1a,b,c show particle morphologies of the three types of templates which have similar charge density and PEI content. They were all synthesized using the same polymer to monomer weight ratio. The only difference between them is their softness property. The formation of gold/polymer composite particles was simply carried out by mixing appropriate concentrations of PEI-based coreeshell particle dispersion and hydrogen tetrachloroaurate(III) solution at room temperature. It was found that all these PEIcontaining particle templates could effectively reduce the H [AuCl4] to gold nanoparticles in the absence of any external reducing agent in aqueous solution. These results suggest that the reduction of gold salt mainly depends on the PEI shell, not much on the core property. However, the encapsulation efﬁciency of the resultant gold nanoparticles was strongly inﬂuenced by the degree of softness of the particle template. Fig. 1d shows that loading capacity of the PMMA/PEI template containing a hard core was low. Even with increasing gold salt concentrations from 5 to 15% of the particle (equivalent to N/Au3þ molar ratios of 37:1 to 9:1) were still unable to enhance the loading capacity. This type of hard template could only hold up gold nanoparticles with sizes less than 10 nm in diameter. Larger gold nanoparticles were ejected from the particle shell and dispersed in solution (see Figure S1, Supporting Information). When using ultra-soft PEI-g-PMMA hollow particle containing similar PEI content and charge density to the hard PMMA/PEI coreeshell particle, loading capacity of the gold nanoparticles was signiﬁcantly enhanced (Figure S2 in Supporting Information). Fig. 1e illustrates that sizes of the gold nanoparticles being immobilized onto the ultra-soft hollow particles are up to 25 nm in diameter. The increase of gold nanoparticle size was evidenced by the appearance of a deep purple dispersion. The reason of increasing the level of particle softness increases the encapsulation efﬁciency may be explained by the fact that the PMMA-g-PEI hollow particle has very soft texture in the core region, allowing deep entrapment of the gold nanoparticles inside the template which are very difﬁcult to escape. However, the PMMA/PEI template contains a rigid core. The gold nanoparticle could not be entrapped deeply. As a result, the bigger gold nanoparticles can easily escape from the template while those small gold nanoparticles with diameters less than 10 nm could be encapsulated. When using PNIPAm/PEI microgel particles as a soft microgel template, loading capacity was considerably increased as illustrated in Fig. 1f. The PNIPAm/PEI template was able to entrap most of gold nanoparticles generated in situ. There were almost no unbound gold nanoparticles in solution. Those entrapped gold nanoparticles were homogeneously distributed in the shell region of the templates. Since the shell of PNIPAm/PEI microgel contained both hydrophilic PEI and grafted PNIPAm chains, the gold nanoparticles were deeply buried in very dense hydrophilic shells. This is evident by the blurry TEM image of the entrapped gold nanoparticles, while those unbound gold nanoparticles appeared much darker. To verify the sizes of the Au nanoparticles immobilized in
N.P.B. Tan et al. / Polymer 76 (2015) 271e279
Fig. 1. Morphologies of different particle templates and their corresponding [email protected]
The particle templates were stained with 0.5 w% phosphotungstic acid solution for appropriate time, while composite particles were observed directly without staining: a) PMMA/PEI particles with PMMA hard core (light core region is PMMA while gray shell is PEI); b) Ultra-soft hollow particles of PEI-g-PMMA copolymer; and c) Soft PNIPAm/PEI microgel templates with darker PNIPAm cores and fuzzy shells containing both PEI and grafted PNIPAm chains; d) [email protected]
/PEI composite particles with an average size of gold nanoparticles of 6.2 ± 1.9 nm (N/Au3þ molar ratio ¼ 19); e) [email protected]
ultra-soft composite particles with an average size of gold nanoparticles of 25.0 ± 8.2 nm (N/Au3þ molar ratio ¼ 30); and f) [email protected]
/PEI soft composite particles with an average size of gold nanoparticles of 15.2 ± 4.0 nm (N/Au3þ molar ratio ¼ 28).
the PEI shells, a TEM image of the composite particles was processed with Image J software (Java-based image processing program developed by National Institute of Health). Sizes of the gold nanoparticles were then determined as described in the Supporting Information. These gold nanoparticles were in the range of 6e24 nm in diameter with average size of 15.2 nm ± 4.0 nm (Figure S3, Supporting Information).
3.1. Kinetic study of gold nanoparticle formation using PNIPAm/PEI microgel as both reductant and template The TEM image shown in Fig. 1c clearly reveals that the PNIPAm/ PEI migrogel particle contains a very dense hydrophilic shell, where the PEI macromolecules are covalently linked to the PNIPAm core. Thus, the local amine concentration of the shell is much higher than that of the free PEI in water. The high local amine content might increase the reduction rate of gold ions to generate gold nanoparticles. To verify this effect, PNIPAm/PEI microgel particle was selected as a model template to study kinetics of gold nanoparticle formation under different reaction conditions. The reduction mechanism involved an electrostatic interaction between the positively charged amine groups of the PEI and the negatively charged [AuCl4] ions. Such interaction attracted the gold ions into the PEI shells where neutral amines could reduce Au3þ ions to Au0 via an electron transfer reaction þ (HAuCl4 þ 3NR3 / Au0 þ 3NRþ 3 þ H þ 4Cl ) [24,32]. Scheme 1 shows the synthesis of the [email protected]
/PEI coreeshell composite particles through a simple mixing of the gold salt precursor solution, H[AuCl4], with the microgels solution at room temperature under gentle stirring. During the reaction, the mixture changed its color from yellowish to light pink within 30 min, indicating the formation of gold nanoparticles.
The kinetics of gold nanoparticle formation was monitored by a UVeVis spectroscopy at absorbance wavelength of 525 nm. This direct evidence on the formation of gold nanoparticles is based on the collective oscillation of free electrons of the gold nanoparticles excited through an incident electromagnetic radiation. Fig. 2a shows the time dependence of absorbance intensity of gold nanoparticles generated in the presence of PNIPAm/PEI microgels. The absorbance intensity increased rapidly in the initial 30 min, followed by only a little increase up to 23 h reaction. The UVeVis spectra show almost no shift in wavelength throughout the whole reaction, suggesting that the gold nanoparticles generated during the reaction have comparable size and shape. According to the literature, absorbencies in the range of 520e525 nm are characteristic wavelengths of spherical gold nanoparticles ranging from 15 to 30 nm . This size range is consistent with our TEM observation as shown in Fig. 1f. Reduction rates of Au3þ to Au0 using PNIPAm/PEI microgel and native PEI in water under the same amine to Au3þ molar ratio are compared in Fig. 2b. In the case of free PEI, the initial reduction rate of 9 h was 1.74 103 (Figure S4b, Supporting Information). When using the PNIPAm/PEI microgels as the reductant, an abrupt change in absorbance intensity of the gold nanoparticles at 525 nm occurred at the initial 30 min. The slope of this linear line was 0.38 (Figure S4a, Supporting Information), suggesting that the reduction rate of PNIPAm/PEI microgel was approximately 218 times faster than that of free PEI. Such signiﬁcant enhancement in reduction rate with the PEI-based coreeshell particles was attributed to the very high local amine concentration due to the conﬁnement of high molecular weight of PEI molecules by the particle. Thus the reduction of gold ions could effectively take place within the dense amine-riched shell. On the contrary, the free PEI is water-soluble and highly swelling in water. Thus the local amine concentration
N.P.B. Tan et al. / Polymer 76 (2015) 271e279
Scheme 1. Schematic representation of the synthesis of [email protected]
/PEI coreeshell composite particles.
Fig. 2. a) UVeVis spectra of the formation of gold nanoparticles at 525 nm wavelength using PNIPAm/PEI microgels as reductant at different reaction times. b) Comparison of kinetics of Au salt reduction with free PEI and PNIPAm/PEI microgels. Open circles (B) represent the reaction kinetics with free PEI. The initial rate of reaction up to 9 h is 1.74 103 (DA/hr). Solid circles (C) represent the reaction kinetics using PNIPAm/PEI microgels. Slope of initial reaction up to 30 min is 0.38 (DA/hr), (reaction conditions for both reactions: 25 C, pH 5.6, 200 rpm, N/Au3þ ¼ 28).
is much lower than that of the PEI shell, giving very slow reduction rate. These results suggest that the PEI-based coreeshell particle containing high local amine concentration can act as both efﬁcient reductant and nanoreactor to generate gold nanoparticles from gold salt within the PEI shell. 3.2. Effect of solution pH and temperature on reduction rate and formation of [email protected]
/PEI composite particles To further verify the effect of high local concentration of amine groups on the efﬁcient reduction of gold ions, we have investigated the formation of gold nanoparticles under various solution pHs using PNIPAm/PEI microgels as both the reluctant and template. The rationale behind this study is that varying solution pH could alter the protonation degree of amino groups and charge density of the PEI molecule, thus affecting the compactness of the particle shell and local amine concentration. Fig. 3a shows kinetic results of gold nanoparticle formation at different pHs. It was found that increasing solution pH from 3 to 11 could considerably enhance the rate of gold nanoparticle formation. This effect may attribute to the increase of the number of neutral amine groups available for reduction of gold ions. It has been reported that percentage of unprotonated amines of the native PEI molecule in pH between 3 and 4 is around 30%, while nearly 100% at pH 10 . Therefore, higher solution pH leads to higher local amine concentration, thus increasing the rate of gold nanoparticle formation. However, the
composite particles obtained under alkaline pH were not as stable as those produced under acidic pH due to their lack of electrostatic stabilization. They easily formed aggregates after storage at room temperature for a few days. Since the reaction temperature might affect both reducing ability of the PEI and solubility of the PNIPAm in water, effect of reaction temperature from 15 to 35 C on reaction kinetics was also examined. Fig. 3b shows that increasing reaction temperature could generally enhance the rate of gold nanoparticle formation. Even though the reduction rate at 35 C was the fastest among those four temperatures studied, the resultant composite particles were unstable during the reaction. The poor particle stability at 35 C was caused by the shrinkage of PNIPAm chains above their phase transition temperature of 32 C. Thus the composite particles lost their steric stabilization ability which were provided by the PNIPAm chains on the particle shells. Supplemental TEM images on the effect of temperature using PNIPAm/PEI template are shown in Figs. 2 and 3 in Date in Brief . Besides the reduction on gold ions and stability of the composite particles, the solution pH was found to affect the in situ encapsulation efﬁciency of the gold nanoparticles. TEM images shown in Fig. 4a and b reveal that gold nanoparticles were effectively generated and entrapped within the polymer templates under both acidic and neutral conditions. The blurry image of the gold nanoparticles in the TEM images indicated that the gold nanoparticles were entrapped within the polymer. On the other hand, Fig. 4c and
N.P.B. Tan et al. / Polymer 76 (2015) 271e279
Fig. 3. Time and temperature dependence of absorbance intensity of gold nanoparticles using PNIPAm/PEI microgels as the reductant: (a) Under different solution pHs. (Conditions: [microgel] ¼ 400 ppm; molar ratio of N/Au3þ ¼ 28; at 25 C); (b) Under different reaction temperatures. (Conditions: [microgel] ¼ 400 ppm; molar ratio of N/Au3þ ¼ 28; at pH 5.6).
Fig. 4. TEM images of AuNPs generated with PNIPAm/PEI templates at (a) pH 3.4; (b) pH 7.3; (c) pH 9.0 and (d) pH 12.0 (reaction conditions: 25 C, 250 rpm, 2 h with N/Au3þ molar ratio of 28).
d illustrate that most of template particles were empty when the reactions were carried out under pH 9 and 12. Majority of the gold nanoparticles (dark dots) were aggregated and localized outside of the templates. The poor encapsulation efﬁciency under alkaline condition may be due to dense and ﬂexible polymer chains of the particle shell, which did not have sufﬁcient space and chain rigidity to accommodate the gold nanoparticles. The TEM image from Fig. 4b also illustrates that most of the gold nanoparticles are evenly immobilized onto the particle shells, forming highly homogeneous,
stable, and uniform [email protected]
/PEI composite particles. Therefore, the optimal pH to generate composite particles using PNIPAm/ PEI as both reductant and template is under neutral pH. 3.3. Effect of amine to gold ion ratio The effect of amine to gold salt ratio was studied because concentration of the gold salt would affect the size, size distribution, shape, and stability of the resultant gold nanoparticles as well as
N.P.B. Tan et al. / Polymer 76 (2015) 271e279
light to dark pink (Figure S5, Supporting Information). The red-shift of UV absorbance wavelength indicated the presence of gold nanoparticles with different size and shape. TEM images of the gold nanoparticles generated at N/Au3þ of 9 revealed the formation of gold nanoparticles with different morphologies including spherical, triangular, diamond and pentagonal shapes (Figure S6, Supporting Information). Spherical gold nanoparticles produced under this condition had an average size of 30.5 nm. This size of the gold nanoparticles was too big to be entrapped by the template, thus residing outside of the microgel templates. 3.4. Properties of [email protected]
/PEI composite particles
Fig. 5. UVeVis spectra of gold nanoparticle formation using PNIPAm/PEI microgels through a ten-stage gold salt reduction. Inset: Concentration dependence of absorbance intensity of gold nanoparticles as function of gold salt concentration. (All reactions were conducted in pH 5.6 at 25 C).
loading capacity of the composite particles. Fig. 5 displays UVeVis absorbance of gold nanoparticles with gold salt concentrations varying from 0.07 to 0.51 mM through a multi-stage reduction experiment. Increasing the gold salt concentrations to 0.07, 0.125, 0.182 and 0.236 mM could increase the intensity of absorbance linearly, but further increase of the gold salt concentrations resulted in little changes in the absorbance intensity (Fig. 5, inset ﬁgure). These results suggested that the microgel template reached its maximum reducing capacity at a gold salt concentration of 0.236 mM, which is equivalent to N/Au3þ ¼ 7. The concentration of gold salt was also found to affect the morphology of the gold nanoparticles. Comparing with the absorbance wavelengths of gold nanoparticles generated at N/Au3þ ratio of 28 (gold salt conc. 0.07 mM) and 14 (gold salt conc. 0.125 mM), similar absorption wavelengths around 521 nm were obtained. But the wavelength shifted to 532 nm when increasing gold salt concentration to up to N/Au3þ ratio of 9 (gold salt conc. 0.182 mM). The change in wavelength was also evident by their color changes from
3.4.1. Size and distribution of the composite particles Fig. 6a shows the particle size and size distribution of the microgel templates before and after loading with gold nanoparticles. The original PNIPAm/PEI microgels had an average hydrodynamic diameter of 409 nm, but it decreased to 298 nm after forming the [email protected]
/PEI composite particles. Particle size distribution of the composite particles also became narrower. Increasing gold salt concentrations from N/Au3þ molar ratio of 28, 14 to 9 could further reduce the size of the resultant composite particles as shown in Fig. 6b. This effect might mainly attributed to the strong binding of gold nanoparticles with amine groups, leading to tightening of the PEI shell . 3.4.2. Chemical composition of [email protected]
/PEI composite particles In order to gain an in-depth understanding of the formation of Au nanoparticles and their interaction with the nitrogencontaining polymers, we have examined the surface composition of [email protected]
/PEI microgls with X-ray Photoelectron Spectroscopy (XPS) analysis. This technique allows detection of elements at a depth of 10 nm, and provides surface chemical information based on photoemission of electrons induced by X-rays. The survey scan in Fig. 7a shows characteristic binding energies of C, O, N, Au. Deconvoluted C 1s and O 1s are shown in Figure S7 in Supporting Information. The deconvoluted C 1s peaks at 285.0 and 287.7 eV are assigned to CeC/CeH and eNHeC]O bonds. The deconvoluted O 1s peaks at 531.6 and 533.3 eV are assigned to the amide and hydroxyl (OeH) groups which come from physically absorbed H2O . The presence of the gold nanoparticles is conﬁrmed based on two characteristic peaks at 84.3 and 88.0 eV (Fig. 7a inset) .
Fig. 6. (a) Particle size and size distribution of pure microgel templates and [email protected]
/PEI composite particles synthesized at 25 C and pH 7.3. Line with circles: Pure PNIPAm/PEI microgels having average particle diameter of 409 nm; Solid line: [email protected]
/PEI microgel synthesized with N/Au3þ mole ratio of 28, particle diameter ¼ 298 nm; (b) Particle sizes of pure PNIPAm/PEI microgels and gold loaded microgels with different amine to gold ion ratios.
N.P.B. Tan et al. / Polymer 76 (2015) 271e279
Their peak position and observed spineorbit splitting of 3.7 eV agree well with the literature value, and the 4f doublet is the characteristics of the Au0 oxidation state . These results suggest that the composite particle shell contains both PEI and PNIPAm chains as well as gold nanoparticles.
To further verify the interaction between gold nanoparticles and nitrogen-containing functional groups, XPS spectra of N 1s peak before and after gold loading onto the microgels were compared (Figure S8 in Supporting Information). It was found that the binding energy of the N 1s peak shifted from 398.60 to 399.68 eV after immobilizing gold nanoparticles. The changes in its binding energy and the broadening of N1s peak proﬁle suggest that chemical bonding environment of the nitrogen has been altered after encapsulating gold nanoparticles. Fig. 7b shows the deconvoluted N 1s peak of the native PNIPAm/PEI particles. Both amine and amide binding energies are identiﬁed at 398.4 and 399.3 eV, respectively. Fig. 7c shows the deconvoluted N 1s peak of the composite particles. Three peaks at 401.07, 399.6 and 398.6 eV were resolved. The large peak at 399.6 eV is the binding energy of Au-amine, while the small peak at 398.6 eV is the unbound amine peak. Based on the integrated peak area of the deconvoluted components, the atomic concentration ratio of Au-bound amine nitrogen and free amine nitrogen is 5.13:1. This value implies that 84% nitrogen atoms were bounded to gold atoms. The amide binding energy has also increased from 399.3 eV to 401.07 eV, indicating that the gold nanoparticles also coordinate to amide groups. These XPS results evidently suggest that there are strong interactions between gold nanoparticles and nitrogen-containing functional groups such as amine and amide groups [35,39]. 3.5. Characterization of the gold nanoparticles Properties of the gold nanoparticles were investigated with a high resolution TEM. Fig. 8 shows a gold nanoparticle of ca. 22.5 nm in diameter. Fig. 8a shows a 5-fold twinned boundary at the center of an Au nanocrystal, indicating the formation of a multiply twinned particle (MTP) which is closely related to the irregular stacking of atoms during nucleation stage. Fig. 8b displays lattice fringe having a d-spacing of 0.236 nm. Thus gold nanoparticle is mainly composed of (111) lattice plane. The fuzzy region of the Au nanocrystal is probably due to the coverage of dense polymeric materials. Selected area of electron diffraction (SAED) pattern over several Au nanoparticles shows four obvious ring patterns of (111), (200), (220), (311) and (331) lattice plane (Figure S9a, Supporting Information), indicating that the gold nanoparticles have a facecentered cubic (fcc) gold crystal lattice with polycrystalline nature. Crystallinity of Au nanoparticles embedded in the microgel template was also analyzed by an X-ray Diffractometer (XRD) (Figure S9b, Supporting Information). Sharp peaks at 2q ¼ 38, 44, and 65 can be assigned as reﬂection from (111), (200) and (220) planes of the gold nanoparticle in face-centered cubic (fcc) crystal structure. 4. Conclusions
Fig. 7. a) XPS survey scans of [email protected]
/PEI composite particles. Inset diagram is the Au 4f core-level spectra obtained from [email protected]
/PEI composite particles. b) N 1s XPS peak of native NIPAm/PEI microgels with deconvoluted amine and amide components. c) N 1s peak of Au-loaded PNIPAm/PEI template with deconvoluted Au-bound amine and free amine components.
This work demonstrates a facile synthesis of gold/polymer composite particles in aqueous system using amine-rich coreeshell polymer particles as both nanoreactor and reductant to generate gold nanoparticles (AuNPs) and as a polymer template to encapsulate and stabilize the resultant gold nanoparticles. Results suggest that PEI-based particle containing a ﬂexible core structure is a more appropriate type of template for encapsulating gold nanoparticles. The PEI shell is capable of efﬁcient reduction of gold ions because of the high local amine concentration. This synthesis of gold/polymer composite particles is simple, very efﬁcient, and free from any organic solvent, reducing agent and stabilizer. Therefore, it is a simple and safe synthetic route to prepare gold-based composite particles, which may ﬁnd important applications in diverse ﬁelds, especially for biological systems.
N.P.B. Tan et al. / Polymer 76 (2015) 271e279
Fig. 8. HRTEM image of a Au nanoparticle from the [email protected]
/PEI template: a) 5-fold twinned boundary at the center of a Au nanocrystal; b) d-spacing of (111) lattice plane.
Acknowledgments This work was supported by the Hong Kong Polytechnic University and the Research Grants Council Hong Kong PhD Fellowship Scheme (HKPF11-07529). We also gratefully acknowledge the Germany/Hong Kong Joint Research Scheme (G_HK035/09) for supporting a travel grant for the collaboration. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.polymer.2015.09.015. References              
N. Uehara, Anal. Sci. 296 (2010) 1219e1228. Y. Liu, X. Feng, J. Shen, J.J. Zhu, W.J. Hou, Phys. Chem. B 112 (2008) 9237e9242. L.E. Strong, J.L. West, WIREs Nanomed. Nanobiotechnol. 3 (2011) 307e317. H. Tang, S. Shen, J. Guo, B. Chang, X. Jiang, W.J. Yang, Mater. Chem. 22 (2012) 16095e16103. Y. Lu, S. Proch, M. Schrinner, M. Dreschsler, R. Kempe, M. Ballauff, J. Mater. Chem. 19 (2009) 3955e3961. D. Li, Q. He, Y. Cui, J. Li, Chem. Mater. 19 (2007) 412e417. T. Sakai, P. Alexandridis, J. Phys. Chem. B 109 (2005) 7766e7777. A. Meristoudi, S. Pispas, Polymer 50 (2009) 2743e2751. H. Yabu, K. Koike, K. Motoyoshi, T. Higuchi, M. Shimomura, Macromol. Rapid Comm. 31 (2010) 1267e1271. J.H. Kim, B.W. Boote, J.A. Pham, J. Hu, H. Byun, Nanotechnology 23 (2012) 275606. M. Karg, T. Hellweg, Curr. Opin. Colloid Interface Sci. 14 (2009) 438e450. Y. Dong, Y. Ma, T. Zhai, Y. Zeng, H. Fu, J. Yao, Nanotechnology 18 (2007) 455603. D. Suzuki, H. Kawaguchi, Langmuir 22 (2006) 3818e3822. I. Gorelikov, L.M. Field, E. Kumacheva, J. Am. Chem. Soc. 126 (2004) 15938e15939.
 M. Schinner, F. Polzer, Y. Mei, Y. Lu, Macromol. Chem. Phys. 208 (2007) 1542e1547.  Y. Lu, Y. Mei, R. Walker, M. Ballauff, M. Dreschler, Polymer 47 (2006) 4985e4995.  G. Agrawal, M.P. Schürings, P. vanRijn, A. Pich, J. Mater. Chem. A 1 (2013) 13244e13251.  P. Kuo, C. Chen, M. Jao, J. Phys. Chem. B 109 (2005) 9445e9450.  X. Sun, S. Dong, E. Wang, Mater. Chem. Phys. 96 (2006) 29e33.  R. Sardar, N.S. Bjorge, J.S. Shumaker-Parry, Macromolecules 41 (2008) 4347e4352.  M.J. Richardson, J.H. Johnston, T. Borrmann, Eur. J. Inorg. Chem. (2006) 2618e2623.  H. Sun, Z. Gao, L. Gao, K. Hou, J. Macromol. Sci. A Pure Appl. Chem. 48 (2011) 291e298.  M. Aslam, L. Fu, M. Su, K. Vijayamohanan, V.P. Dravid, J. Mater. Chem. B 14 (2004) 1795e1797.  J.D.S. Newman, G.J. Blanchard, Langmuir 22 (2006) 5882e5887.  M. Neu, D. Fischer, T. Kissel, J. Gene. Med. 7 (2005) 992e1009.  P. Li, J. Zhu, P. Sunintaboon, F.W. Harris, Langmuir 18 (2002) 8641e8646.  M.F. Leung, J. Zhu, F.W. Harris, P. Li, Macromol. Rapid Commun. 25 (2004) 1819e1823.  K.M. Ho, W.Y. Li, C.H. Lee, C.H. Yam, R.G. Gilbert, P. Li, Polymer 51 (2010) 3512e3519.  M. Feng, P. Li, J. Biomed. Mater. Res. 80 (2007) 184e193.  M. Mimi, K.M. Ho, Y.S. Siu, A. Wu, P. Li, J. Control. Release 158 (2012) 123e130.  C.H. Lee, K.M. Ho, F.W. Harris, S.Z.D. Cheng, P. Li, Soft Matter 5 (2009) 4914e4921.  F.R. Keene, Coord. Chem. Rev. 187 (1999) 121e149.  W. Haiss, N.T.K. Thanh, J. Aveyard, D.G. Fernig, Anal. Chem. 79 (2007) 4215e4221.  J. Suh, S.H. Lee, S.M. Kim, S.S. Hah, Bioorg. Chem. 25 (1997) 221e231.  G. Lee, H. Lee, K. Nam, J.H. Han, J. Yang, S.W. Lee, D.S. Yoon, K. Eom, T. Kwon, Nanoscale Res. Lett. 7 (2012) 608.  A.P. Grosvenor, B.A. Kobe, N.S. McIntyre, Surf. Sci. 572 (2004) 217e227.  D.V. Leff, L. Brandt, J.R. Heath, Langmuir 12 (1996) 4723e4730.  X. Yang, M. Shi, R. Zhou, X. Chen, H. Chen, Nanoscale 3 (2011) 2596e2601.  F. Zhang, M.P. Srinivasan, Langmuir 23 (2007) 10102e10108.  N.P.B. Tan, C.H. Lee and P. Li Data in Brief “submitted”.