ARTICLE pubs.acs.org/Langmuir
Controlling Nanoparticle Formation via Sizable Cages of Supramolecular Soft Materials Jing-Liang Li,†,‡ Xiang-Yang Liu,*,†,§ Xun-Gai Wang,‡ and Rong-Yao Wang†,|| †
Department of Physics and Department of Chemistry, National University of Singapore, 2 Science Drive 3, Singapore 117542 Centre for Material and Fiber Innovation, Institute for Technology Research and Innovation, Deakin University, Waurn Ponds, Victoria 3127, Australia § College of Material Science and Engineering & State Key Laboratory for Modification of Chemical Fibers and Polymer Materials, Donghua University, Shanghai 201620, P.R. China Key Laboratory of Cluster Science of Ministry of Education, Beijing Institute of Technology, Beijing 100081, P R. China
)
‡
ABSTRACT: We present a new generic strategy to fabricate nanoparticles in the “cages” within the fibrous networks of supramolecular soft materials. As the cages can be acquired by a designand-production manner, the size of nanoparticles synthesized within the cages can be tuned accordingly. To implement this idea, both selenium and silver were chosen for the detailed investigation. It follows that the sizes of selenium and silver nanoparticles can be controlled by tuning the pore size of the fiber networks in the material. When the concentration of the gelator is high enough, monodisperse nanoparticles can be prepared. More interestingly, the morphology of the nanoparticles can be altered: silver disks can be formed when the concentrations of both the gelator and silver nitrate are sufficiently low. As the fiber network serves as a physical barrier and semisolid support for the nanoparticles, the stability in the aqueous media and the ease of application of these nanoparticles can be substantially enhanced. This robust surfactant-free approach will not only allow the controlled fabrication of nanoparticles, but also can be applied to the fabrication of composite materials for robust applications.
T
he synthesis of nanoparticles normally involves the use of surfactants or other types of stabilizer such as polymers, which are hard to remove completely and impose some problems in applications. In addition, the controlled formation of nanoparticles with monodisperse and tunable (and thus tenable) sizes remains a challenge. Moreover, the low stability of nanoparticles prepared in solutions especially those that are light or temperature sensitive makes their long-term storage difficult.1 Developing an approach that addresses these problems is significant considering the important applications of nanoparticles in many fields. Supramolecular gels have been a research focus of recent years because of their important applications for drug delivery,2 templates for the fabrication of nanostructures,3 scaffolds for tissue engineering,4 self-supporting porous materials for the separation of macromolecules,5,6 and materials for art conservation,7etc. Nowadays, it is possible to engineer the fibrous network of the soft materials based on a design-and-production approach.8 This research is to explore a generic caging approach to fabricate nanoparticles utilizing the micro/nanosized pores of the network as the cages or “containers” to control the formation of nanoparticles (Figure.1a). As the pore size of the networks in such a material is relatively regular, and can be tuned based on soft materials engineering developed recently via r 2011 American Chemical Society
thermodynamic, sound, or additive-mediated approaches,8 it is expected that nanoparticles can be prepared in the fibrous network of such a material in a controllable and predictable manner. The nanoparticles formed within the cages can also be relatively monodisperse and sizable (Figure 1b). As the cages can keep the synthesized nanoparticles physically apart and reduce their exposure to the environment, the particles can be stored in the caging materials for long-term use without aggregation. In addition, the in situ formation of particles will open a new route of engineering new advanced materials with important applications in many fields such as chemical sensing and therapeutics,9 in which the gel serves as a stabilizer/ supporter. Although using the fibers in such a material as templates to prepare nanostructures such as nanofiber, nanowire, and lamellar structures has received great interest,3,10,11 their application in controlling the formation of nanoparticles is rather limited.1214 To implement the nanocaging approach, the following criteria should be met: (1) The medium to synthesize nanoparticles should be compatible with the supramolecular soft materials. For Received: January 16, 2011 Revised: May 6, 2011 Published: May 31, 2011 7820
dx.doi.org/10.1021/la200196k | Langmuir 2011, 27, 7820–7827
Langmuir
ARTICLE
nanoparticles have also attracted extensive interest due to their important applications in optics, catalysis, surface-enhanced Raman spectroscopy, etc. These properties are strongly dependent on the size and especially the shape of the particles.1923 However, silver nanoparticles are also quite unstable and are sensitive to the storage conditions such as light and temperature.24 Developing an approach to preparing monodisperse, size tenable, and stable selenium and silver nanoparticles is practically significant. The results to be discussed below will show that tunable and tenable mondisperse nanoparticles selenium and silver nanoparticles can be fabricated with the caging effect of agarose fiber network.
Figure 1. (a) Scanning electronic microscopic image of fibrous networks of agarose networks acquired by freeze-drying technology. (b) Illustration of size-controlled nanoparticle formation in the pores of a gel fiber network. The size of nanoparticles can be manipulated by controlling the pore size of the fiber network.
instance, for the synthesis of nanoparticles in aqueous solutions, water-based gels should be adopted. Otherwise, nonaqueous solvent-based gels should be adopted. (2) To ensure the formation of nanoparticles, the interaction between the reacting agent molecules/ions and the fibers should be avoided or minimized. Otherwise, more complicated structures, i.e., hybrid structures, will be acquired.10,11 In other words, such a material should only provide cages for the particle formation and incorporation. In this work, we will start with metal salts as a reactive agent. Agarose hydrogel, an essentially uncharged marine polysaccharide, is selected as the caging material. Because the solvent phase is water, it is easy for the metal salts to be dissolved in the gel. The uncharged nature, good biocompatibility of agarose, and tunable pore size15 of the gel contribute to its general application in biomedical fields such as in electrophoresis for the separation of biomolecules. This feature also makes this gel a good candidate for the controlled formation of nanoparticles by the aforementioned caging approach.16 We will use selenium and silver nanoparticles as examples to demonstrate this robust approach. The choice of these nanoparticles is based on the following facts: Selenium nanoparticles have promising applications in fabrication of photonic crystals, photodetectors or sensors, electrical rectifiers, and cancer therapy.1,17,18 However, due to the low glass transition temperature (31 °C), the shape of selenium nanoparticles is hard to maintain, limiting their long-term storage.17 Silver
’ RESULTS AND DISCUSSION It has been proven that the gelator concentration determines the uniformity and pore size of fiber networks in a gel.25,26 Therefore, it may affect the formation of nanoparticles. The SEM images of selenium (Se) nanoparticles and silver (Ag) nanoparticles formed in agarose gels with different agarose concentrations, as well as the corresponding agarose fiber networks (xerogel), are shown in Figure 2. In agarose gels, spherical selenium nanoparticles were obtained. Due to its low glass transition temperature, selenium normally precipitates from a solution phase as amorphous solid at ambient temperatures in the presence of a reducing agent. The amorphous nature of the Se nanoprticles is characterized by the brick red color of the particlegel composite (cf., inset of Figure 3f). At the low agarose concentration of 0.25 wt %, the size of Se nanoparticles formed showed a broad distribution, with the largest around 1 μm in diameter and the smallest around 100 nm (Figure 2a-1). This means that the agarose fiber network plays a minimal role in regulating the nanoparticle size. With the increase of agarose concentration to 1 and 2 wt %, smaller and more homogeneously size-distributed particles formed (cf., Figure 2b-1,c-1, respectively). The polydispersity of the nanoparticles is within 5%, meaning that homogeneous nanoparticles can be formed. The mean size of Se nanoparticles as a function of agarose concentration ranging from 0.5 to 3 wt % is shown in Figure 3a. It follows that the agarose concentration is significant in controlling the particle size. With the increase of agarose concentration, the diameter of the nanoparticles drops significantly and then levels off to values around 100 nm when the agarose concentration is above 1 wt %. To understand the effect of fiber network on particle formation, the xerogels formed at different agarose concentrations were obtained. The SEM images of the fiber networks are shown in the bottom row of Figure 2. When the concentration of agarose is 0.25 wt % (Figure 2a-3), the pore size of the fiber network is very large (several micrometers), which explains the possibility to form larger particles at this low agarose concentration. With an increase in agarose concentration to 1.0 wt % (Figure 2b-3), the network becomes denser and more homogeneous. When the concentration is above 1.0 wt %, noticeable difference was not observed. It is worth mentioning that obtaining an intact xerogel has been a technical challenge. Large pores with sizes around 1 μm can be identified from the SEM images of the xerogels even at agarose concentrations of 1 and 2 wt % (Figure 2b-3,c-3), which could be due to the partial breaking of the fiber network during the freeze-drying process. Most of the small pores, which are most probably intact, in these two xerogels 7821
dx.doi.org/10.1021/la200196k |Langmuir 2011, 27, 7820–7827
Langmuir
ARTICLE
Figure 2. SEM images of selenium nanoparticles (first row) and silver nanoparticles (second row) formed in agarose gels with different agarose concentrations, and the corresponding fiber networks of agarose (the bottom row). The concentrations of Na2SeO3 and AgNO3 are 0.1 and 0.005 M, respectively. The concentration of ascorbic acid is fixed at 1 and 0.1 M for the preparation of Se and Ag nanoparticles, respectively. The scale bars are 1 μm.
have sizes from 100 to 200 nm, which corresponds to the size of the Se nanoparticles formed in the gels. Formation of nanoparticles is a process consisting of two main phases: nucleation of molecules to form clusters of critical size (nuclei) and growth of the particles. With the increase of agarose concentration, the fiber network becomes denser, and a larger number of smaller pores are created in a fixed gel volume, forming smaller and more uniform cages. Each cage can be a potential reactor. The fibers provide a soft physical barrier that limits the growth of nanoparticles in the cages. The pore sizes (diameters) of agarose gels formed at different concentrations were calculated from the UVvis light absorption of the gels, following a reported method.15 The results together with the corresponding sizes of Se nanoparticles are given in Figure 3c, which shows the corresponding reduction in particle size when the size of the cages becomes smaller with increase in agarose concentration. When the agarose concentration is larger than 0.5 wt %, the size of the nanoparticles correlates well with the size of the cages. This is consistent with the conclusion drawn from SEM images of the particles and gel fiber networks (Figure 2). To further demonstrate the caging effect of the fiber networks on nanoparticle formation, the concentrations of the precursors were varied. Figure 3b gives the size of Se nanoparticles formed in 2 wt % agarose gels as a function of Na2SeO3 concentration. The size of the nanoparticles increases sharply from 50 to 90 nm when the concentration of Na2SeO3 is increased from 0.001 to 0.01 M, and then levels off with further increase of the precursor concentration to 0.1 M (diameter of nanoparticles is 110 nm). The small change in particle size at higher Na2SeO3
concentration indicates that, in a fixed gel volume, more nanoparticles form with increase in precursor concentration. Here we use number density of particles to characterize the effect of agarose network on nucleation of selenium. The number density is defined as Nd ¼ M=ð4π=3 R 3 dÞ
ð1Þ
where Nd is the number density of particles (mol/L), M is the molar concentration of the precursor (mol/L), R is the median radius of the particles formed, and d is the mass density of nanoparticles. For selenium, d is 4.82 10 kg/m3. Number density has been proven to be a useful way to characterize the nucleation of nanoparticles such as zinc oxide nanoparticles.27 The number density of nanoparticles is proportional to M/R3, in which M is molar concentration of the precursor and R is the radius of the nanoparticles. On the basis of this information, the ratio of the number density of nanoparticles at different precursor concentrations can be estimated. For example, in the slow growth stage (Figure 3b, from 0.01 to 0.1 M), the number density of the nanoparticles at the Na2SeO3 concentration of 0.1 M is about 5.5 times that when its concentration is reduced to 0.01 M. In comparison, in the steep growth stage of the curve in Figure 3b (from 0.001 to 0.01 M), the number density is only increased 1.7 times when the precursor concentration is increased also 10 times. While it is expected that a higher supersaturation of Na2SeO3 can enhance the nucleation rate, when the gel network is present, this enhancement is more significant compared with the nucleation in aqueous phase (to be discussed later). This observation indicates that the modulation of particle size (caging 7822
dx.doi.org/10.1021/la200196k |Langmuir 2011, 27, 7820–7827
Langmuir
ARTICLE
Figure 4. Ag nanoparticles formed within agarose networks at the agarose concentration of 0.25 and 1 wt % with different concentrations of AgNO3. The concentration of ascorbic acid is fixed at 0.1 M.
Figure 3. Dependence of size of Se nanoparticles on the (a) agarose concentration and (b) Na2SeO3 concentration. (c) Particle size and gel pore size at different agarose concentrations. Number density of particles as a function of (d) agarose concentration and (e) Na2SeO3 concentration. (f) UVvis absorption spectra of the gels loaded with Se nanoparticles. The inset is a photo of agarose gels (2 wt %) without and loaded with Se nanoparticles; the concentration of Na2SeO3 initially dissolved in the gels in the four vials from left to right is 0, 0.001, 0.05, and 0.1 M, respectively. The corresponding UVvis absorption spectra of the nanoparticles formed are shown in part f (the curves from bottom to top); the gel without nanoparticle loading is used as baseline. The concentration of ascorbic acid is fixed at 1 M.
effect) by agarose fiber network is accompanied by promoted primary nucleation. The increase in the number density of Se nanoparticles with increase in precursor concentration is indicated by the thicker coloration of the nanoparticle-loaded gels (inset of Figure 3f). The change in particle size with Na2SeO3 concentration has also been verified by measuring the optical absorption of the nanoparticles. Figure 3f shows that the maximal absorption wavelength of the Se nanoparticles redshifted from 260 to 280 nm when the particle size is increased from 50 (0.001 M Na2SeO3) to 105 nm (0.05 M Na2SeO3). While a further shift of only about 5 nm is observed when the concentration of Na2SeO3 is increased to 0.1 M, indicating a small increase in the size of Se nanoparticles. On the basis of mass balance and the particle size, the number densities at different agarose concentrations and Na2SeO3 concentration are calculated. At a fixed Na2SeO3 concentration of 0.1 M, the number density of selenium nanoparticles increases linearly when the agarose concentration is increased to 2 wt % (Figure 3d). Above this value, the increase of number density is not significant. In the low concentration range of agarose, it is expected that an increase in agarose concentration will lead to a
corresponding increase in the number of pores (cages), which contributes to the mass (precursor ions) distribution into larger numbers of smaller cages (reactors). When the agarose concentration is high enough, the change in pore size and hence the number of cages in a fixed volume becomes less significant (Figure 3c). This leads to an increase in Nd to a lesser extent. It is interesting to note that, at the fixed agarose concentration of 2 wt %, the number density of Se nanoparticles increases linearly with the concentration of Na2SeO3 (Figure 3d). It is worth mentioning that although each cage is a potential reactor, the chance for nanoparticle formation in a cage is also dependent on the nucleation rate (defined as the number of nuclei occurring per unit volume and time). Keeping all other parameters constant (only the precursor concentration is changed for Figure 3d), a higher thermodynamic driving force (precursor concentration) leads to a higher nucleation rate. That is, more cages will have nanoparticles with increase in precursor concentration. That is, there should be some empty cages, particularly when the precursor concentration is low. Nevertheless, the uniform coloring of the gels loaded with Se nanoparticles (inset of Figure 3f) indicates the distribution of nanoparticles in the gel volume is homogeneous, even at the lowest precursor concentration of 0.001 M. The size control of silver nanoparticles using agarose fiber networks is also demonstrated by Figure 2a-2,b-2,c-2. However, what is more interesting is that silver nanoparticles of different shapes can be produced by varying the agarose concentration. At the low agarose concentration of 0.25 wt %, most silver particles are in the form of disks/plates. The size of the disks is around 800 nm. Gels have been used to control the growth of crystals of sparingly soluble salts. Control over crystal size and morphology has been achieved by controlling the gel density.28 In a recent work, Mantion et al. also reported the formation of silver nanoparticles of different shapes including triangle plates in organogels.14 Traditionally, to control the size and shape of silver particles, templates or shape-inducing agents such as surfactants or polymers are always needed.2931 Furthermore, we notice that the disk-shaped particles are of potentially important applications in polymer-based conductive composites 7823
dx.doi.org/10.1021/la200196k |Langmuir 2011, 27, 7820–7827
Langmuir
ARTICLE
Figure 5. Sizes of silver nanoparticles obtained at different concentrations of (a) agarose and (b) silver nitrate.
and elelctromagnetic shielding materials.32 Anisotropic silver nanoparticles such as nanodisks can be normally obtained by introducing silver seeds and in the presence of structure directing agents.32 The unseeded production of silver disks in the gel fiber networks without using any structure directing agent provides a convenient approach to the preparation of this type of particle. It was observed that, at the agarose concentration of 0.5 wt %, a mixture of disks and irregularly shaped silver nanoparticles (not shown) were produced. The size of the disks is about 500 nm which is smaller than those formed in the 0.25% gel network. With further increase in agarose concentration to 1 and 2 wt %, smaller silver nanoparticles (∼200 nm) are formed (Figures 2b-2,c-2). The silver nanoparticles formed at different concentrations of AgNO3 were also examined. Since it has been observed that agarose fiber networks can affect not only the size but also the shape of silver nanoparticles, the effects of fiber network at both the low concentration of 0.25 wt % and the higher concentration of 1.0 wt % were further investigated. The SEM images of silver nanoparticles formed are given in Figure 4. It is observed that further reducing the concentration of AgNO3 to 0.0005 M in the 0.25 wt % agarose gel leads to the formation of silver disks about 1 μm in diameter, which are larger than those formed at a lower precursor concentration of 0.005 M (Figure 2a-2). However, with an increase in AgNO3 concentration to 0.01 M, Ag nanoparticles with an average size of 200 nm were produced. In brief, at the agarose concentration of 0.25 wt %, the concentration of the silver ions has a significant effect on the shape and size of the Ag nanoparticles formed. The slower nucleation kinetics at a very low concentration of silver ion may contribute to the preferable growth of the silver nanoparticles along a certain crystal facet, leading to the formation of silver disks. At this stage, it is still unclear to us why the growth of large silver disks occurs at the low concentrations of both precursor and agarose. In fact, gel matrix has been proven to be effective for
Figure 6. (a) TEM image and (b) electron diffraction of silver disks formed at 0.0005 M AgNO3 in 0.25 wt % agarose gel.
growing crystals of appreciable size and of high quality.33 Unfortunately, a clear understanding of the mechanisms has not been obtained. The decrease in size of Ag nanoparticles with the increase of AgNO3 concentration is contrary to the size increase of Se nanoparticles as the concentration of Na2SeO3 increases. This means that the nucleation rate of silver is more affected by the concentration of AgNO3. Figure 4b-1,b-2 shows the SEM images of silver nanoparticles formed in 1.0 wt % agarose gel when the AgNO3 concentration is 0.01 and 0.05 M, respectively. A comparison of these two images with Figure 2b-2 indicates that when the precursor concentration is varied from 0.005 to 0.05 (1 order of magnitude difference), the size and shape of silver nanoparticles change very little, except that, at low concentrations of AgNO3, a small population of silver disks formed. This observation indicates that size controllable formation of Ag nanoparticles is achievable by the denser fiber networks of 1.0 wt % gel. Figure 5 gives the average sizes of silver nanoparticles obtained at different concentrations of agarose and silver nitrate. In 0.25 wt % gel, the particle size decreases almost linearly with an increase in precursor concentration. The size of the nanoparticles fits the pore size of the agarose fiber network very well when the agarose concentration is 1 wt % and above (Figure 5a), and as shown in Figure 5b, the particle size is almost constant when the concentration of AgNO3 varies from 0.001 to 0.05 M when the agarose concentration is fixed at 1 wt %. According to eq 1, the number density of silver nanoparticles is also linearly dependent on the precursor concentration, which is similar to what is observed for Se nanoparticles. Due to the limitation of the absorption method for calculating the pore size when the pore size is extremely large (>2 μm), the pore size of the fiber network at 0.25 wt % agarose concentration is not given in Figure 5a. 7824
dx.doi.org/10.1021/la200196k |Langmuir 2011, 27, 7820–7827
Langmuir
Figure 7. SEM images of silver nanoparticles obtained at different concentrations of ascorbic acid in 1 wt % agarose gel. The concentration of AgNO3 is fixed at 0.005 M. The scale bars are 1 μm.
The SEM image of Figure 3a-1 indicates that most of the pores at this agarose concentration are several micrometers, which is a few times the average size of the corresponding silver particles. The silver disks obtained were further characterized with transmission electron microscopy (TEM). Figure 6a shows a TEM image of a single disk, and Figure 6b gives the electron diffraction pattern of the disk with the electron beam irradiated perpendicular to the flat surface. The diffraction pattern is typical of a single crystal silver disk.34 It is worth mentioning that although the reductant, ascorbic acid (Ac) diffused from the top of the gel, which can lead to concentration gradient through the volume of the gel, noticeable difference in size and shape of selenium and silver nanoparticles formed at different depth of gel was not observed. This could be due to the high concentration of the ascorbic acid, which is 20 and 5 times of its stoichiometric concentrations even for the maximum concentrations of AgNO3 and Na2SeO3 used in the work, respectively. To demonstrate the effects of ascorbic concentration, silver nanoparticles were prepared in 1 wt % agarose at varying ascorbic concentrations (0.025 and 1 M). No noticeable difference between the silver nanoparticle was observed (Figure 7). The nanoparticles formed in gels were compared with those formed in water with the surfactant Tween 20 as a stabilizer. The sizes of the surfactant-stabilized Se particles formed are 37 and 75 nm at the Na2SeO3 concentration of 0.005 and 0.05 M, respectively (Figure 8a,b). On the basis of mass balance, the number density of particles at the higher precursor concentration (0.05 M) is 1.2 times that at the lower concentration (0.005 M). In the agarose gel network, this ratio is 4.4, indicating promoted nucleation by the fiber network. Figure 8d,e shows the SEM images of Ag nanoparticles formed in aqueous solution with Tween 20 as a stabilizer. Two concentrations of AgNO3, 0.005 and 0.01 M, were used. The concentration of ascorbic acid is the same as that used for preparation of Ag nanoparticles in agarose gel. A comparison between Figures 2, 4, and 8d,e shows that, at the same concentrations of reactants, the shapes of the particles formed in aqueous phase are quite different from those formed in gels. In aqueous phase, agglomerates of Ag nanoparticles formed.
ARTICLE
Although similar to the case of gel with low concentration of agarose (0.25 wt %), the sizes of the particles are also reduced with an increase in silver nitrate concentration, and the controlled mass transfer in gel networks slows down the reaction, contributing to the formation of silver crystals with smooth surfaces rather than agglomerates. When the fiber network becomes denser, it provides cages to effectively control/limit the formation of particles, making the size tenable irrespective of the precursor concentration. This makes the preparation of nanoparticles with a desired size easily achievable. To acquire a better understanding of the effect of fiber networks on the formation of nanoparticles, Se and Ag particles were also prepared in water without using any stabilizer (Figure 8c,f). At a Na2SeO3 concentration of 0.1 M, large Se nanoparticles with sizes around 500 nm formed. The sizes of the nanoparticles are much larger than those formed in gels (cf., Figures 2 b-1,c-1 and 3a), which again proves the caging effect of the fiber networks. It is noticed that, in the absence of any stabilizer, spheres of selenium are hard to obtain. Due to the low glass transition temperature, in aqueous solution, a-Se is rather sensitive to thermal fluctuation and is easy to transform into irregular colloids or trigonal selenium due to Rayleigh instability (Figure 8c).2 On the contrary, the presence of agarose molecules can effectively stabilize the particles. It was observed that the size and shape of both Se and Ag particles in the gel did not undergo observable change even after a year. Therefore, the gel caging technique can also be used for the long-term storage of thermal sensitive particles. Caging of particles into the gel phase can also reduce their potential health hazards. Figure 8f shows Ag nanoparticles formed in water without the presence of any stabilizer. A mixture of larger agglomerates (1.2 μm) and smaller agglomerates (250 nm) of silver particles formed. (Figure 8f). It has been shown from the above examples that although other factors also influence the size and shape of the resulting nanoparticles, the pore size of supramolecular soft materials plays a key role in this nanocaging approach. Apart from the concentration of gelators, the pore size of the fibrous network within the caging materials can be altered by the chemical, thermal, ultrasound, etc. stimuli.8 In this sense, we can adopt different stimuli in engineering the nanoparticles by controlling the pore size of the caging materials to suit the requirement of various applications. This will then broaden the applications of this technology. It is noteworthy to mention that the technique developed in this work is not limited to the controlled preparation of nanoparticles. The in situ controlled formation of nanoparticles in supramolecular materials with 3D fiber networks also provides an approach to the fabrication of nanocomposite materials. In such a material, the fiber network provides not only the cages to isolate and disperse the nanoparticles, but also a “solid” support to facilitate the applications of the nanoparticles. This class of semisolid soft functional materials can have important applications in many fields. For example, silver nanoparticles have been demonstrated to have antimicrobial property, gelAg nanoparticle composite film can thus be used for wound healing.9,35 Due to the anticancer property of selenium, the composite may be applicable to cancer treatment. Applications can also be found in many other fields such as catalysis, chemical sensing, and controlled drug delivery, all of which are interesting areas to explore. In addition, as gels can be water-based or oil-based, and the networks within the materials can be physically or chemically linked, the principles acquired here can be applicable to the systems other than aqueous gels. 7825
dx.doi.org/10.1021/la200196k |Langmuir 2011, 27, 7820–7827
Langmuir
ARTICLE
Figure 8. Se and Ag nanoparticles formed in water with Tween 20 as a stabilizer (a, b, d, and e) and without any stabilizer (c and f).
’ CONCLUSION In summary, size controlled formation of Se and Ag nanoparticles is achieved in the sizable micro/nano pores (cages) of agarose gel fiber networks. At a certain precursor concentration, the size of the nanoparticles can be reduced by increasing the agarose concentration (reducing the pore size of fiber network). Nanoparticles with tunable sizes can be prepared by varying the concentrations of the precursors. Due to the caging effect, the size of nanoparticles can be tenable, facilitating the fabrication of nanoparticles with prescribed sizes. When both the silver ion and agarose concentrations are very low, silver disks can be prepared. More detailed work will be carried out on the formation mechanisms of silver disks in the fiber network.
’ EXPERIMENTAL SECTION Chemicals. Agarose (low melt, preparative grade) was obtained from Bio-Rad. Silver nitrate (g98%), sodium selenite, Tween 20, and L-ascorbic acid (g98%) were from Sigma-Aldrich. Synthesis Procedures. To prepare agarose gels, weighed amounts of agarose were dissolved in 20 mL glass tubes at 110 °C, and the solutions were left to cool down naturally to room temperature (20 °C). The concentration of agarose is from 0.25 to 2.0 wt %. To obtain the fiber networks (xerogels), pieces of agarose gels were immersed in liquid nitrogen to freeze the structures, and then the water contained in the gel was removed by freeze-drying. Synthesis of Nanoparticles in Agarose Gel. To prepare selenium or silver nanoparticles in agarose gel, certain volumes of concentrated Na2SeO3 or AgNO3 aqueous solution were added into solution of agarose before gel forms. The mixed solutions were homogenized using a vortex mixture. Then, the sample tubes were left statically at room temperature for 48 h for gel formation. A simple diffusion method was used to synthesize nanoparticles in agarose hydrogel network. A 2 mL portion of 1 M (for Se nanoparticles) or 0.1 M (for Ag nanoparticles) ascorbic acid aqueous solution was put on top of 5 mL agarose gels dissolved with sodium selenite or silver nitrate at different concentrations. The ascorbic acid molecules diffused gradually into the gel, which was evidenced by the propagation of red (Se nanoparticle) or gray (Ag nanoparticle) color from the surface to the bottom of the gel. After 48 h, the top ascorbic solution was removed. The gels were dissolved, and the solutions were centrifuged at 10 000 rpm to remove agarose molecules and unreacted ascorbic acid. The particles
were resuspended in deionized water. Finally, 10 μL of the particle suspension was put on a small piece of clean glass and was dried naturally at room temperature overnight before a SEM analysis. The particle size was analyzed using dynamic light scattering.
Synthesis of Surfactant-Stabilized Ag and Se Nanoparticles. To prepare surfactant-stabilized nanoparticles, aliquots of concentrated solution of ascorbic acid were dropped into 5 mL of silver nitrate or sodium selenite solutions containing Tween 20 at a concentration of 10 times its critical micelle concentration (CMC, 0.06 mM) under magnetic stirring. After 2 h, aliquots of particle dispersion were centrifuged, and the particles were resuspended in water for sample preparation. Characterization of Nanoparticles. The nanoparticles were characterized by SEM (field emission scanning electron microscope, JEOL JSM6700F), TEM (Philips CM120 BioTWIN transmission electron microscope), and a high performance particle sizer.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT This work is supported by Singapore AcRF funding (R-144000-264-112) and Chinese MOE Chang Jiang Scholarship Chair Professorship (Dong Hua University). ’ REFERENCES (1) Li, J. L.; Liu, X. Y. J. Nanosci. Nanotechnol. 2008, 8, 2488–2491. (2) Kim, H. K.; Shim, W. S.; Kim, S. E.; Lee, K. H.; Kang, E.; Kim, J. H.; Kim, K.; Kwon, I. C.; Lee, D. S. Tissue Eng., Part A 2009, 15, 923–933. (3) Jung, J. H.; Kobayashi, H.; van Bommel, K. J. C.; Shinkai, S.; Shimizu, T. Chem. Mater. 2002, 14, 1445–1447. (4) Shaikh, F. M.; Brien, T. P. O.; Callanan, A.; Kavanagh, E. G.; Burke, P. E.; Grace, P. A.; Badylak, S. F.; McGloughlin, T. M. Tissue Eng. 2007, 13, 1696–1697. (5) Atmeh, R. F.; Massad, T. T.; Kana’an, B. M.; Abu-Alrob, A. A. Anal. Biochem. 2008, 373, 307–312. (6) Braz, V. A.; Howard, K. J. Anal. Biochem. 2009, 388, 170–172. 7826
dx.doi.org/10.1021/la200196k |Langmuir 2011, 27, 7820–7827
Langmuir
ARTICLE
(7) Carretti, E.; Bonini, M.; Dei, L.; Berrie, B. H.; Angelova, L. V.; Baglioni, P.; Weiss, R. G. Acc. Chem. Res. 2010, 43, 751–760. (8) Li, J. L.; Liu, X. Y. Adv. Funct. Mater. 2010, 20, 3196–3216. (9) Jain, J.; Arora, S.; Rajwade, J. M.; Omray, P.; Khandelwal, S.; Paknikar, K. M. Mol. Pharmaceutics 2009, 6, 1388–1401. (10) Xue, P. C.; Lu, R.; Li, D. M.; Jin, M.; Tan, C. H.; Bao, C. Y.; Wang, Z. M.; Zhao, Y. Y. Langmuir 2004, 20, 11234–11239. (11) Yang, Y. G.; Suzuki, M.; Shirai, H.; Kurose, A.; Hanabusa, K. Chem. Commun. 2005, 2032–2034. (12) Love, C. S.; Chechik, V.; Smith, D. K.; Wilson, K.; Ashworth, I.; Brennan, C. Chem. Commun. 2005, 1971–1973. (13) Huang, X.; Weiss, R. G. Tetrahedron 2007, 63, 7375–7385. (14) Mantion, A.; Guex, A. G.; Foelske, A.; Mirolo, L.; Fromm, K. M.; Painsi, M.; Taubert, A. Soft Matter 2008, 4, 606–617. (15) Xiong, J. Y.; Narayanan, J.; Liu, X. Y.; Chong, T. K.; Chen, S. B.; Chung, T. S. J. Phys. Chem. B 2005, 109, 5638–5643. (16) Xiong, J. Y.; Liu, X. Y.; Li, J. L.; Vallon, M. W. J. Phys. Chem. B 2007, 111, 5558–5563. (17) Jeong, U. Y.; Xia, Y. N. Adv. Mater. 2005, 17, 102–106. (18) Chen, T.; Wong, Y. S.; Zheng, W.; Bai, Y.; Huang, L. Colloids Surf., B 2008, 67, 26–31. (19) Cobley, C. M.; Skrabalak, S. E.; Campbell, D. J.; Xia, Y. N. Plasmonics 2009, 4, 171–179. (20) Pietrobon, B.; Kitaev, V. Chem. Mater. 2008, 20, 5186–5190. (21) Pietrobon, B.; McEachran, M.; Kitaev, V. ACS Nano 2009, 3, 21–26. (22) Singh, A. K.; Senapati, D.; Neely, A.; Kolawole, G.; Hawker, C. J.; Ray, P. C. Chem. Phys. Lett. 2009, 481, 94–98. (23) Nelayah, J.; Kociak, M.; Stephan, O.; Geuquet, N.; Henrard, L.; de Abajo, F. J. G.; Pastoriza-Santos, I.; Liz-Marzan, L. M.; Colliex, C. Nano Lett. 2010, 10, 902–907. (24) Pinto, V. V.; Ferreira, M. J.; Silva, R.; Santos, H. A.; Silva, F.; Pereira, C. M. Colloids Surf., A 2010, 364, 19–25. (25) Li, J. L.; Liu, X. Y.; Wang, R. Y.; Xiong, J. Y. J. Phys. Chem. B 2005, 109, 24231–24235. (26) Li, J. L.; Wang, R. Y.; Liu, X. Y.; Pan, H. H. J. Phys. Chem. B 2009, 113, 5011–5015. (27) Taubert, A.; Glasser, G.; Palms, D. Langmuir 2002, 18, 4488–4494. (28) Petrova, R. I.; Swift, J. A. Cryst. Growth Des. 2002, 2, 573–578. (29) Fan, L.; Guo, R. Cryst. Growth Des. 2008, 8, 2150–2156. (30) Tsuji, M.; Maeda, Y.; Hikino, S.; Kumagae, H.; Matsunaga, M.; Tang, X. L.; Matsuo, R.; Ogino, M.; Jiang, P. Cryst. Growth Des. 2009, 9, 4700–4705. (31) Xu, Z. C.; Shen, C. M.; Xiao, C. W.; Yang, T. Z.; Zhang, H. R.; Li, J. Q.; Li, H. L.; Gao, H. J. Nanotechnology 2007, 18, Article 115608. (32) Duan, J. Y.; Zhang, Q. X.; Wang, Y. L.; Guan, J. G. Acta PhysicoChimica Sinica 2009, 25, 1405–1408. (33) Lorber, B.; Sauter, C.; Ng, J. D.; Zhu, D. W.; Giege, R.; Vidal, O.; Robert, M. C.; Capelle, B. J. Cryst. Growth 1999, 204, 357–368. (34) Jin, R. C.; Cao, Y. W.; Mirkin, C. A.; Kelly, K. L.; Schatz, G. C.; Zheng, J. G. Science 2001, 294, 1901–1903. (35) Varaprasad, K.; Mohan, Y. M.; Ravindra, S.; Reddy, N. N.; Vimala, K.; Monika, K.; Sreedhar, B.; Raju, K. M. J. Appl. Polym. Sci. 2010, 115, 1199–1207.
7827
dx.doi.org/10.1021/la200196k |Langmuir 2011, 27, 7820–7827
