Local Conditions Influencing In Situ Formation of ... - ACS Publications

Article pubs.acs.org/JPCC

Local Conditions Influencing In Situ Formation of Different Shaped Silver Nanostructures and Subsequent Reorganizations in Ionomer Membrane Sabyasachi Patra,† Debasis Sen,‡ Ashok K. Pandey,*,† Chhavi Agarwal,† Shobha V. Ramagiri,§ Jayesh R. Bellare,§ S. Mazumder,‡ and A. Goswami† †

Radiochemistry Division, and ‡Solid State Physics Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India § Department of Chemical Engineering, I.I.T. Bombay, Powai, Mumbai 400 076, India ABSTRACT: Ionomer membranes are promising platforms for the metal nanostructures and provide possibilities for utilizing unique properties of the nanostructures. However, it is not known how local environment prevailing in the membrane matrix plays a role in the process of formation of different dimensional metal nanostructures, and the subsequent fate of the mesoscopic physical architecture of the matrix. Therefore, different dimensional silver nanostructures were formed in the Nafion-117 membrane by manipulating local environment using carefully selected reductants, varying temperature, and controlled loading of Ag+ ions in the matrix by ion-exchange process. Ag nanostructures thus formed had different sizes and shapes, spherical nanoparticles, nanorods, and nanosponge depending upon in situ reduction and growth conditions. These nanocomposites were studied by combination of transmission electron microscopy (TEM) and small-angle Xray scattering (SAXS) for understanding the formation and subsequent modifications in the shape of the metal nanostructure as well as in the self-assembling morphology of the matrix in different post reduction counterionic environment. Under specific conditions, silver nanorods having ∼8 nm mean diameter and ∼40 nm length were formed. The embedded Ag nanostructures of different sizes and shapes were found to affect the self-diffusion mobility of Na+ and Cs+ counterions in the nanocomposite matrixes differently due to variation in reorganizations of the ionomer matrix.

■

INTRODUCTION

incorporating desirable nanostructures having appropriate shape and size for a given application. Nafion has a high degree of disorder in physical organization of its matrix. To obtain information on the structural morphology of Nafion under different environment, several experimental techniques such as small-angle X-ray scattering (SAXS), wide-angle X-ray scattering (WAXS), small-angle neutron scattering (SANS), atomic force microscopy (AFM), transmission electron microscopy (TEM), and nuclear magnetic resonance (NMR) spectroscopy have been used over the years.19−27 On the basis of these studies, different models describing the self-assembling morphology of Nafion under different physicochemical environments have been proposed.28−46 The first important morphological description of Nafion was based on the “cluster-channel network model” proposed by Hsu and Gierke.19,20 This model qualitatively explained the origin of the ionomer peak (due to water clusters) in SAXS profile, high permselectivity, and ionic conductivity of the Nafion membrane. However, the connecting cylindrical nano-

Polymer nanocomposites contain nanoparticles of various types dispersed in polymer matrix,1−4 and are of immense technological importance because of their extensive use in homogeneous and heterogeneous catalysis, sensorics, electrochemical devices, optoelectronics, and artificial muscles.5−12 There are various ways of synthesizing metal−polymer nanocomposites. Ionomer membranes, because of having unique morphology in a combination of hydrophilic and hydrphobic domains,13 serve as a platform for in situ synthesis of nanoparticle using appropriate precursor metal ions. Nafion117 is a cation-exchange ionomer membrane consisting of polytetrafluoroethylene (PTFE) backbone and randomly spaced long perfluorovinyl ether side chains terminated by sulfonate ionic groups. The presence of self-assembled water clusters in Nafion-11714 has prompted many researchers to use it as a nanoreactor for the synthesis of nanosized metal and ceramic particles.15−18 In most of these studies, spherical nanoparticles having size much bigger than the size of the water clusters have been formed. This suggests that formation of the nanoparticles is a multistep process that may be controlled by unknown matrix parameters and the nature of reductants.16 The understanding of these parameters is important in © XXXX American Chemical Society

Received: October 24, 2012 Revised: May 17, 2013

A

dx.doi.org/10.1021/jp402155m | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

and nature of the reductant in controlling sizes and shapes of the nanostructures. The size dispersions of nanostructures and modifications in the self-assembling morphology of membrane have been studied by SAXS and TEM of the cross-sections of membrane. The effects of incorporation of silver nanostructures on transport properties of the membrane have also been studied by measuring the self-diffusion coefficients (SDC) of Na+ and Cs+ ions using radiotracers. Na+ and Cs+ ions were chosen as they constitute two extremes with respect to the extent of their hydrations that result in a large difference in water contents of the membrane in these two ionic forms.

channels proposed in this model have never been directly evidenced by scattering techniques and microscopy. This model has been found to be quantitatively inconsistent with data over a wide range of scattering angles37,38 and from membranes oriented by tensile draw.39 Rubatat et al. suggested the “polymeric bundle model” consisting of fluorocarbon chains surrounded by ionic groups and water.38,40 In recent studies, Schmidt-Rohr and Chen have quantitatively simulated previously published small-angle scattering data of hydrated Nafion and have proposed the “parallel cylinder model”.37 They described the characteristic “ionomer peak” to be arising from long parallel but otherwise randomly packed water channels surrounded by partially hydrophilic side branches, forming inverted-micelle cylinders. More recently, a unified morphological description of organization of water swollen Nafion has been proposed.30 This model is based on both statistical (maximum entropy approach) and thermodynamic (dissipative particle dynamics) descriptions, and suggests a bicontinuous network of ionic clusters embedded in a matrix of fluorocarbon chains. This model accepts the existence of ionic clusters network without the need of extended parallel channels. Small-angle scattering (SAS) is a well-established nondestructive technique.46−52 SAXS is also capable of providing information on size distributions of nanostructures embedded in the polymer matrix. However, SAXS analyses require an assumption of shape of the mesoscopic structures. The shape and spatial distribution of metal nanostructures embedded in the matrix can be obtained by transmission electron microscopy (TEM) of cross-sections of the nanocomposite samples.15,16 Thus, the combination of SAXS and TEM can be very effective for studying mesoscopic structures in the ionomer nanocomposites. The major objective of this work is to study the formation of different dimensional silver nanostructures in the Nafion membrane matrix, and subsequent reorganizations of physical structure of the membrane in different ionic forms. It has been observed in our previous work that reduction of Ag+ ions with BH4− ions occurs at the surface of Nafion membrane.15 This is attributed to the fact that anions are excluded from cationexchange matrix. During BH4− reduction, Na+ ions displace Ag+ ions from ion-exchange sites, and thus reduction occurs at the membrane interface. To form Ag nanoparticles within the membrane matrix, the nonionic reductants such as formamide (HCONH2) and dimethyl formamide (HCON(CH3)2) have been used to form Ag nanoparticles uniformly distributed in the Nafion matrix.16 However, these nonionic reducntants swell the Nafion matrix and destroy original physical architecture of the membrane during reduction. In this work, silver nanostructures have been synthesized by controlled loading of Ag+ ions by ionexchange process and subsequent in situ chemical reduction by selected reductants at varying temperature. The choice of reductant with suitable reactivity is important to prepare silver nanoparticles of different sizes and shapes. Ascorbic acid used in the present work is a well-known chemical reductant for nanoparticle synthesis.53 Ascorbic acid does not affect water clusters network in the membrane. To understand the formation of nanoparticles in Nafion without water clusters, we have included formamide reduction also as formamide swells Nafion matrix resulting in destruction of the water clusters network. Thus, the matrix effects on the formation of Ag nanoparticles are expected to be different during formamide reduction from that during ascorbic acid reduction. In this work, we demonstrate the role of the membrane morphology

■

EXPERIMENTAL SECTION Materials and Reagents. Nafion-117 ionomer membrane with an equivalent weight of 1100 g/SO3H was obtained from Ion Power Inc. Analytical Reagent grade chemicals AgNO3, CsCl, NaCl, NaNO3, formamide, and ascorbic acid were purchased from BDH (Poole, England). Deionized water (18 MΩ/cm) purified by model Quantum from Millipore (Mumbai, India) was used throughout the experiments. The radiotracers 22Na and 137Cs were procured from the Board of Radiation & Isotope Technology (Mumbai, India). Sample Preparation. Nafion-117 membrane was cut into a few 2 cm × 2 cm pieces, and conditioned by conventional treatments described elsewhere.54 The conditioned pieces of Nafion were equilibrated with freshly prepared 0.25 mol L−1 AgNO3 solution for 30 min with continuous stirring, ensuring almost all ion-exchange sites occupied with Ag+ ions. These membrane samples were subjected to chemical reduction for obtaining nanocomposites. The conditioned pristine and nanocomposite membranes were converted to Na+ and Cs+ ionic forms by immersing the membrane pieces into 25 mL of 0.5 mol L−1 NaCl and 0.5 mol L−1 CsCl solutions, respectively, for about 24 h. The conditions used for conversion of one ionic form to another ionic form of the membrane sample under stirring and nonstirring equilibrations were based on our previous work.55 Reduction of Ag+-Loaded Membrane with Formamide. The membrane samples fully loaded with Ag+ ions were chemically reduced with formamide at temperatures of 40, 50, and 65 °C. Constant temperature was maintained during the reduction process by a water bath. Reduction was done for 30 min for each of the samples. The samples were coded as AFT40, AFT50, and AFT65 for identification of the formamide reduced samples at 40, 50, and 65 °C, respectively. The membrane samples were equilibrated with 0.25 mol L−1 NaNO3 to remove unreduced Ag+ ions, if any. Reduction Ag+-Loaded Membrane with Ascorbic Acid. Ag+-loaded membrane samples were equilibrated with well-stirred 0.25 mol L−1 NaNO3 solution for times of 30, 45, and 60 s to prepare the membrane samples having 25%, 10%, and 2% of Ag+ loading, respectively. The time required for preparing membranes having a desired extent of loading of Ag+ ions was fixed based on experimental information given in our earlier publication.15 The membrane samples were then reduced with well-stirred 0.1 mol L−1 ascorbic acid solution for 5 min at room temperature. A thin metallic layer formed was removed by gentle scratching of surfaces of the membrane samples with wet adsorbent paper. The samples were coded as A1, A2, and A3 for the identification of ascorbic acid reduced samples containing 25%, 10%, and 2% of Ag+ ions occupying the ion-exchange sites. Finally, post reduction conditioning of the membrane samples was carried out by equilibrating these in B



Article

appropriate salt solutions as described for the formamide reduced nanocomposite samples. Measurements of Counterion Self-Diffusion Coefficients. The self-diffusion coefficients of Na+ and Cs+ ions were obtained from the analyses of the isotopic-exchange rate profiles as described in our previous publication.54 The measurements of isotopic-exchange rates of (*Mn+)mem ⇌ (Mn+)aq (*tagged with radiotracer) were carried out using 2 cm × 2 cm pieces of the membrane samples. Na+ and Cs+ counterions were tagged with 22Na and 137Cs radiotracers, respectively. These radiotracer-loaded membrane samples were equilibrated with well-stirred 0.25 mol L−1 salt solution having the same ions as counterions in the membrane samples. The amount of radiotracer counterions exchanged between membrane sample and salt solution was monitored by taking out the membrane sample from equilibrating solution at regular time intervals and counting radioactivity using a γ-ray spectrometer. The membrane sample was again placed in equilibrating solution after counting. The actual residence time of the membrane in equilibrating solution was considered as time for isotopic-exchange of counterions between membrane sample and salt solution. Transmission Electron Microscopy. Transmission electron microscopy (TEM) of cross-sections across thickness of the membrane samples was carried out after post reduction neutralization with Na+ ions. The membrane samples were sectioned under cryogenic environment in Leica ultramicrotome to 70 nm thickness. The sections were picked on 200 mesh Formvar and carbon-coated Cu grids. The grids were examined in an FEI Technai G2 electron microscope at IIT Mumbai, at 120 keV without any external treatment. Small-Angle X-ray Scattering. SAXS experiments have been performed using a laboratory-based SAXS instrument with Cu Kα as probing radiation. Radial averaged scattering intensity (I(q)) was obtained within a wave vector transfer (q = 4π sin(θ)/λ, where λ is the wavelength and 2θ is the scattering angle) range from ∼0.1 to 2.5 nm−1.

I1(q) = INS(q) = (

∞

PNS(q , R )R6D NS(R ) dR )

(2)

where PNS(q,R) represents the form factor of a nanostructure. For spherical nanostructures of radius R: PNS(q , R ) = 9

(sin(qR ) − qR cos(qR ))2 (qR )6

(3)

DNS(R) represents the nanostructure size distribution; that is, DNS(R) dR indicates the probability of having size R to R + dR. In the present case, standard log-normal distribution of the following type: ⎡ [ln(R /R )2 ] ⎤ 0 ⎥ exp⎢ − 2 2 2 σ 2 ⎦ ⎣ 2πσ R 1

D NS(R ) =

(4)

was considered. R0 represents the median radius, and σ represents the polydispersity index of the distribution being obtained by fitting the SAXS profiles. ii. Local Monodisperse Spherical Particle Model. The second term (I2(q)) of eq 1 under local monodisperse approximation is expressed as I2(q) = IWC(q) =(

∫0

∞

PWC(q , R )R6D WC(R )S WC(q , R ) dR )

(5)

where PWC(q,R) represents the form factor of a water cluster. DWC(R) represents the water cluster size distribution and is defined in the same manner as DNS(R). SWC(q,R) represents the inter water cluster interaction. It is worthy to mention that the peak-like feature at higher q in the scattering profile appears due to such correlation between the water clusters. A hard sphere-type interaction potential was considered in the present case between the water clusters.56 In some sample, another extra length scale existed in the system due to agglomeration of nanoparticles or formation of nonspherical anisotropic particles. To account for the scattering contribution from such inhomogeneities, which is predominant only at very small q regime, a term similar to that of eq 2 was added to eq 1 for fitting the whole scattering profile in a simpler way. In addition, the scattering profiles for the nanorod sample (A1) in the Na+ and Cs+ ionic forms were fitted using the form factor of cylinder.57 A log-normal distribution for the radius polydispersity was assumed. Radius distribution and length of the nanorods were estimated by fitting the scattering data with the model. For cylindrical nanostructures of radius “R” and length “L”, the term “PNS” (form factor) of eq 2 was expressed as:

■

THEORETICAL SECTION SAXS Analyses. Because of the existence of ionic water clusters as well as silver nanostructures, the whole scattering profile was broadly subdivided into two zones, that is, water clusters and Ag nanostructures embedded in fluorocarbon matrix. The presence of hydrophobic fluorocarbon structures along with silver nanostructures was not considered separately for simplicity. In the entire analysis and discussion, the silver nanostructures profile actually represents the profile due to nanostructures in hydrophobic fluorocarbon matrix. As the scattering space and the real space are connected by Fourier transform, the information about the smaller length scale (water cluster in this case) is primarily manifested at relatively higher q region. Similarly, the scattering signal at lower q region is primarily due the nanoparticles and their agglomerated structures. The scattering profiles were analyzed using the following model. Total scattering intensity I(q) may be represented as I(q) = CII1(q) + C2I2(q)

∫0

PNS(q) =

∫0

π /2

⎡ 2J (qR sin(α)) sin(qL cos(α)/2) ⎤2 ⎢ 1 ⎥ sin(α) dα (qL cos(α)/2) ⎦ ⎣ qR sin(α)

(6)

where J1(x) represents the Bessel function of first order. The integration over π/2 takes care of all possible orientations of the cylinders. Analysis of Isotopic-Exchange Profiles. The experimentally measured profiles of the isotopic-exchange rates of (*Mn+)mem ⇌ (Mn+)aq systems (*tagged with radiotracer) were analyzed to obtain self-diffusion coefficients of counterions. The isotopic-exchange rates profiles were least-squares fitted using an analytical solution of Fick’s second law given as:54,58

(1)

The first and second terms in eq 1 correspond to scattering from Ag nanostructures and water clusters, respectively. C1 and C2 are proportional to respective number densities. i. Polydisperse Spherical Particle Model. For such a case, I1(q) may be approximated as: C



Article

Figure 1. SAXS profiles of formamide (65 °C) (a) and ascorbic acid (27 °C) (b) treated Nafion-117 samples. ⎡ ⎤ ⎛8⎞ 1 n(tk) = n*⎢1 − ⎜ 2 ⎟ exp(− Dπ 2tk /L2) + exp( −9Dπ 2tk /L2) + ... ⎥ ⎝π ⎠ ⎣ ⎦ 9

{

leads to metallization. It was observed that all Ag+ ions reduced to Ag were fixed in the membrane as a color change of ascorbic acid solution in contact with membrane sample was not observed. Shapes and Spatial Distributions of Nanostructures. Formamide Reduced Samples (AFT65, AFT50, and AFT40). Formamide reduction produces spherical silver nanostructures with significant size dispersion as known from our earlier work.16 Reduction of Ag+ ions with formamide at elevated (40−65 °C) temperature generated nanoparticles uniformly spread throughout the matrix. The representative TEM image of cross-section of a formamide reduced membrane sample is shown in Figure 2. It was observed that the size of the nanostructures increased with increase in reduction temperature. During formamide reduction, the nucleation of silver seeds occurred in formamide swelled matrix that resulted in 80% loss of silver seeds. The retained seeds in the matrix selfagglomerated to form a bigger nanosphere due to close spacing and mobility of Ag seed in soft domain. Therefore, the sizes of silver nanoparticles were larger than water clusters in the membrane. Because the network of water clusters was disrupted by formamide, diffusion of Ag+ ions was blocked during reduction process. This resulted in uniform reduction of Ag+ ions throughout the matrix. However, microinhomogeneity in spatial distribution Ag nanoparticles was observed in the TEM image shown in Figure 2. This could be attributed to high density fluorocarbon domains that did not contain a significant number of ionic clusters. It is noteworthy that the water clusters network regenerated after equilibration of formamide reduced sample in salt solution as indicated by SAXS analyses. The SAXS profiles for the “formamide reduced samples” named AFT65, AFT50, and AFT40 after post reduction neutralization with Na+ and Cs+ ions are shown in Figure 3a and b, respectively. SAXS profiles for pristine Nafion in the respective ionic forms (Unmod) were also plotted to show the changes after incorporation of nanostructure in the membrane matrix. The ionomer peak near a q value of 1.7 nm−1 signifies scattering from the water clusters. The peak around q ≈ 0.4 nm−1 is associated with the fluorocarbon matrix region of unmodified Nafion matrix and the embedded silver nanostructures along with fluorocarbon structures in Nafion nanocomposites. The variation in position and intensity of the scattering maxima at ionomer peaks indicated dimensional changes of the spherical water clusters. For Cs+ loaded samples,

}

(7)

where D is the self-diffusion coefficient of counterions in the membrane, L is the thickness of the membrane, and tk is equilibration time. n* is the total amount of radiotracer ions in the membrane at equilibrium (t = ∞). In case of radiotracer tagged ions diffusing out of the membrane into solution, the symbol n* in eq 7 represents total radiotracer ions in salt solution at equilibrium (t = ∞). n(tk) is radioactivity of the radiotracer in equilibrating salt or membrane sample at a fixed equilibration time tk. The initial and boundary conditions are described in our earlier publication.54,59

■

RESULTS AND DISCUSSION Formamide is a mild nonionic reducing agent and can swell PTFE backbone resulting in disruption of the water cluster network in the membrane. To ensure this, water-swollen Nafion sample was equilibrated with formamide and subjected to SAXS analyses. It is seen from Figure 1a that the typical ionomer peak (water cluster) of Nafion at higher scattering vector (q ≈ 1.7 nm−1) was not seen in the SAXS profile after equilibration of the membrane sample in formamide. The reduction with formamide was carried out with fully Ag+ loaded membrane samples as only 8−16% of Ag+ ions occupying ionexchange sites were finally locked in the membrane as nanoparticles.16 The bulk of the nanoparticle was lost to formamide solution due to swelling of the ionomer matrix. Controlled loading of Ag+ ions could not be used for varying extent of Ag+ reduction in the membrane. Therefore, temperature was kept as a variable parameter during formamide reduction to vary reduction rates. The selection of ascorbic acid was based on considerations that (i) it is water-soluble, (ii) room temperature reduction is possible, and (iii) it can enter in the ionomer matrix as neutral molecules unlike BH4− reductant that is excluded from the cation-exchange membrane due to Donnan exclusion of anions.15 The presence of ionomer peak in SAXS profile confirmed that the water clusters network morphology of pristine Nafion existed during reduction with ascorbic acid; see Figure 1b. Thus, ascorbic acid is expected to reduce Ag+ ions in water clusters domain of the ionomer matrix. The reduction with ascorbic acid was carried out using membrane samples containing 25%, 10%, and 2% of ion-exchange sites occupied by Ag+ ions. The ascorbic acid reduction of membrane samples having more than 25% ion-exchange sites occupied by Ag+ ions D



Article

Figure 4. Size distributions of spherical Ag nanoparticles in the formamide reduced Nafion samples obtained from analyses of SAXS profiles in Na+ and Cs+ post reduction neutralized forms.

distribution for the same sample (Figure 2b) showed reasonably good agreement. This indicated that SAXS profiles can be used for identifying changes in bulk size dispersions of the nanoparticles if assumption for shape factor is valid. From the size distribution plots of Na+ neutralized samples, it is seen that the peak diameter and size dispersion of the spherical Ag nanoparticles increased with increase in temperature during formamide reduction. With increase in temperature, the kinetics of growth process became faster due to softening of matrix as well as increase in reduction rate. This led to the formation of bigger size nanoparticles. Thus, lower temperature is preferable for the formation of spherical nanoparticles with narrower size dispersion. The size dispersion of nanoparticles in the samples post neutralized with Cs+ ions shifted to higher peak diameter value, indicating further reorganization of Ag nanostructures. Water clusters size distributions in the nanocomposite samples having Na+ and Cs+ ionic forms were compared to those in the pristine samples; see Figure 5. In the Na+ neutralized samples, the peak diameters and size diameters and size dispersions of water clusters were increased with an increase in the peak size of spherical Ag nanostructures. As can be seen from Figure 5a, the peak diameter of water clusters followed the trend AFT65 > AFT50 > AFT40 and lied in the range of 2−4 nm. The size of water clusters in the unmodified

Figure 2. (a) Representative TEM image of the cross-section of membrane showing spatial distribution of silver nanospheres produced in sample AFT65 after post reduction neutralization with Na+ ions, and (b) corresponding size distribution histogram.

the ionomer peak was weaker due to less hydration of the membrane in Cs+ form. It can be seen from Figure 3 that the SAXS pattern changed significantly depending upon ionic form of the membrane. Fitting of the obtained SAXS profiles with appropriate models described in the Theoretical Section gave the size distribution profiles for both Ag nanostructures and water cluster in Nafion matrix. Figure 4 shows the size distribution profiles for spherical Ag nanoparticles in Nafion matrix in Na+ and Cs+ ionic forms. The comparison of diameter distribution profile obtained by SAXS analyses (plot 3 of Figure 4) with TEM histogram of size

Figure 3. SAXS profiles for “formamide reduced samples” in Na+ (a) and Cs+ (b) post reduction counterionic forms. E



Article

Figure 5. Water cluster size distribution plots for “formamide reduced samples” in Na+ (a) and Cs+ (b) ionic forms of membrane.

Figure 6. TEM images showing shapes and spatial distributions of Ag nanocylinders near surface (a), and nanorods across thickness (b) of the ionomer nanocomposite (A1) prepared by ascorbic acid reduction of the ionomer sample having 25% of ion-exchange sites occupied with Ag+ ions and the rest with Na+ ions.

Nafion in Na+ form lies between AFT65 and AFT50. The water cluster size distribution for the Cs+ neutralized samples indicated no significant difference in peak diameters of water clusters. The size distributions of water clusters in pristine and nanostructures embedded samples were almost similar in Cs+ neutralized forms. Ascorbic Acid Reduced Samples (A1, A2, and A3). The representative TEM images showing shapes and spatial distributions of ascorbic acid reduced sample (A1) formed using sample containing 25% Ag+ ions in ion-exchange sites (IES) are given in Figure 6. It is seen from Figure 6 that silver nanorods were formed. This is unusual as water clusters (reverse micelle) in the ionomer matrix are expected to be sites for growth of controlled size spherical Ag nanoparticles. At the surface, thin metallization occurred due to a high reduction rate. This thin sheet was removed by gentle scrubbing of the surface. In a few places just below the surface, hollow cylinders were formed as shown in Figure 6a. However, needle-like Ag nanorods were uniformly formed at the interior of the ionomer matrix. The formation of hollow cylinders may be due to defects in matrix near the surface. In a sample containing 10% Ag+ ions in ion-exchange sites (A2), the sponge-like bigger Ag nanostructures were formed instead of nanorods (Figure 7). These bigger nanostructures appear to be embedded in localized area of PTFE matrix. These

nanostructures have a central dense region surrounded by clustering of spherical nanoparticles. On further decreasing the concentration of Ag+ ions to 2% (A3 sample), the TEM images given in Figure 8 clearly show the concentration-dependent formation of clusters of nanoparticles during reduction with ascorbic acid. It appears in this case that number density of nanoparticles was not sufficient for formation of bigger particles, as the nanoparticles remained separated or occasionally formed clusters. It can be seen that separate nanoparticles have a diameter range of 4−5 nm, which matches well with the diameter of water clusters in Nafion. The SAXS profiles of ascorbic acid reduced samples A1, A2, and A3 after post reduction neutralization with Na+ and Cs+ ions are compared to SAXS profiles of pristine Nafion in respective ionic forms in Figure 9. The SAXS profiles of “ascorbic acid reduced samples” were significantly different from “formamide reduced samples”. Appearance of a new, relatively more intense scattering maximum at lower q region, especially at q < 0.2 nm−1, corroborated the existence of extra length scale in the system as observed in the TEM images. This extra length scale could be attributed to the presence of relatively large rod-like nanostructures in A1 sample and large nanosponge and nanoclusters in A2 and A3 samples. The size dispersion in cross-sectional diameter of nanostructures in the A1, A2, and A3 samples having Na+ and Cs+ ionic forms was obtained by F



Article

Figure 7. TEM images of cross-section of ionomer nanocomposite (A2) showing clustering of nanoparticles to nanosponge as big nanostructures at the surface (a) as well as the interior matrix (b−d). The samples were prepared by ascorbic acid reduction of the ionomer sample loaded with 10% of Ag+ ions followed by neutralization with Na+ ions.

Figure 8. TEM images of cross-section of nanocomposite (A3) showing the formation of nanoparticle clusters in ionomer matrix. The nanocomposite samples were prepared by ascorbic acid reduction of ionomer sample loaded with 2% of Ag+ ions followed by neutralization with Na+ ions.

fitting profiles as described in the Theoretical Section using a spherical model. The diameter and length distributions of Ag nanorods in the A1 samples obtained by SAXS analyses are compared to the TEM size histogram in Figure 10. It is seen from the plots given in Figure 10 that there was reasonably good agreement between

distribution profiles obtained from SAXS analyses with TEM histograms. As size distribution profiles obtained from SAXS analyses represent bulk distributions, the nanorods having diameter ≈ 8 nm and length ≈ 40 nm were formed in bulk matrix. It is also seen from SAXS distribution profiles that G



Article

Figure 9. SAXS profiles for ascorbic acid reduced samples after post reduction neutralization with Na+ (a) and Cs+ (b) ions.

Figure 10. Comparison of diameter and length distributions of nanorods in A1 samples obtained by SAXS analyses (a) with TEM histograms (b).

dimensions of the nanorods were not affected significantly by change in ionic form of Nafion matrix. The size distributions of nanorods given in Figure 10 were obtained by fitting SAXS profiles with the form factor for sphere. To ensure the validity of size distributions given in Figure 10, the SAXS profiles for the sample A1 (nanorods) in two different counterionic forms (Na+ and Cs+) were also analyzed using the form factor of cylinder. In this case, simultaneous determination of log-normal distribution for both the diameter and the length of the nanorods is ambiguous.

Thus, the better way to analyze profile is by assuming one parameter as monodisperse (e.g., length of the nanorod) and then determining the log-normal distribution of the other parameter (e.g., diameter of the nanorod). In the fitting SAXS profiles, the length was assumed to be monodispersed with size 30 nm, and the scattering profiles were fitted using the form factor of cylinder by varying the three parameters representing number density, polydispersity index, and the median of the distribution of nanorod diameter. It is seen from Figure 11 that the diameter distribution of nanorods obtained by fitting H



Article

Figure 11. Comparison of experimental SAXS profile of sample A1 (in Na+ form) with that fitted with form factor of cylinder (a), and thus obtained diameter distributions of nanorods in the membrane samples having Na+ and Cs+ ionic forms (b).

Figure 12. Ag nanostructures size distributions of the ascorbic acid reduced samples A2 (a) and A3 (b) obtained by SAXS analyses. The samples were converted to Na+ and Cs+ ionic forms after reduction.

Figure 13. Water cluster size distribution plots for “ascorbic acid reduced samples” in Na+ (a) and Cs+ (b) ionic forms.

The diameter distributions of nanoparticles aggregated as nanosponges and nanoclusters in samples A2 and A3 having different ionic forms are given in Figure 12. It is seen from Figure 12 that nanoparticles size distributions drastically changed with change in Na+ to Cs+ ionic form. Water content

scattering profiles using the form factor of model did not show a significant difference from that shown in Figure 10. This provides validity to analyzing SAXS profiles of nanorods in A1 sample using the spherical model. I



Article

Figure 14. Fractional attainment of isotopic-exchange equilibrium as a function of equilibration time in nanocomposite samples having nanorods (A1) in Na+ ionic form (a), and spherical nanoparticles (AFT65) in Cs+ ionic form (b).

Table 1. Self-Diffusion Coefficient Values (Dion) of Na+ and Cs+ Ions in Pristine and Nanostructure Embedded Nafion-117 Membranesa Na+ form sample Pristine AFT50 AFT65 A1 A2 a

Cs+ form

Ag nanostructure shape and diameter (dion)

median diameter and spread (s) of water clusters (nm)

Dion (×10−6, cm2 s−1)

median diameter and spread (s) of water clusters (nm)

Dion (×10−6, cm2 s−1)

none spherical, dNa+ = 7.52 nm, dCs+ = 10.92 nm spherical, dNa+ = 8.72 nm, dCs+ = 9.82 nm nanorod (L = 36.4 nm, d = 8 nm) bigger sized nanosponge and cluster

2.92 (s = 3.56) 2.80 (s = 5.00)

1.03b 0.88

1.62 (s = 1.96) 1.70 (s = 2.6)

0.19b 0.21

3.26 (s = 5.50)

1.07

1.64 (s = 2.10)

0.19

3.44 (s = 3.86)

1.14

1.76 (s = 2.92)

0.33

3.40 (s = 3.60)

1.21

1.70 (s = 2.92)

0.33

The D values are within 5% error limit. bTaken from ref 54.

size and polydispersity of water clusters in the nanoparticles embedded membranes increased with respect to pristine membrane (unmodified) in Na+ ionic form. There was not a significant change in the peak diameters of water clusters of the nanoparticles loaded membranes in Cs+ ionic form as shown in Figure 13b. However, a larger dispersion of size of water clusters in the nanoparticles loaded samples clearly suggested agglomeration of water clusters to some extent. Diffusion Studies of Na+ and Cs+ Ions in Nanocomposite Membranes. Transport of ions in the ionomer membrane is highly dependent on connectivity of the water clusters. Measurement of self-diffusion coefficient (Dion) of ions provides information on the mobility of the ions, which are highly dependent on the water clusters network in the membrane. Therefore, DNa+ and DCs+ have been measured by analyzing experimental isotopic-exchange profiles as described in the Theoretical Section. The representative experimental and fitted (eq 7) isotopic-exchange rates profiles are shown in Figure 14. Fractional attainment of isotopic-exchange equilibrium has been obtained from the ratio of radioactivity of radiotracer in membrane at time tk to that at equilibrium (t = ∞). The values of self-diffusion coefficients thus obtained are given in Table 1. For comparison, literature values of the selfdiffusion coefficient of pristine membrane are also given in Table 1.54 It is seen from Table 1 that DNa+ > DCs+ in silver nanocomposite membranes that is similar to pristine membrane. In formamide reduced samples, the SDC values for Na+ and Cs+ ions are found to be close to that in pristine

in Nafion matrix is highly dependent on its ionic form; that is, water contents in Na+ and Cs+ forms of Nafion-117 are 17.2 and 8.2 wt %, respectively.54 In Na+ form, the nanoparticles having peak diameter 8−10 nm were formed in both A2 and A3 samples. This size of nanoparticles is greater than size water clusters but similar to diameters of nanorods (A1) and nanoparticles in formamide reduced samples. Thus, all nanostructures appear to be formed from Ag seeds (small particles and clusters) ejected from water clusters into surrounding low density domains. When water content reduced in the Cs+ form, the nanoparticles agglomerated due to strain caused by shrinking of surrounding low density fluorocarbon matrix on reduction of the water content. Thus, the size of agglomerated particles increased considerably. It is to be noted that the size distributions given in Figure 12 are only for nanostructures having sizes less than 100 nm. The bigger nanostructures (≫100 nm) were also seen in TEM images of samples A2 and A3, which were not detected by SAXS. The formation of Ag nanosponges and clusters in the samples having lower content of Ag seems to suggest diffusion of Ag+ ions and Ag atoms through connected clusters to localized reduction zone. The water cluster size distribution plots of “ascorbic acid reduced samples” after post reduction neutralization with Na+ and Cs+ ions are given in Figure 13a and b, respectively. It is seen from Figure 13 that there were no significant changes in peak diameter of the spherical water clusters in all three A1, A2, and A3 samples in the same ionic (Na+/Cs+) form. Because of comparatively large size of incorporated Ag nanostructures, the J



Article

(9) Ramaraj, R. Nanostructured Metal Particle-Modified Electrodes for Electrocatalytic and Sensor Applications. J. Chem. Sci. 2006, 118, 593−600. (10) Selvaraju, T.; Sivagami, S.; Thangavel, S.; Ramaraj, R. Electrochemical and In Situ Spectroelectrochemical Studies of Gold Nanoparticles Immobilized Nafion Matrix Modified Electrode. Bull. Mater. Sci. 2008, 31, 487−494. (11) Xing, S.; Xu, H.; Chen, J.; Shi, G.; Jin, L. Nafion Stabilized Silver Nanoparticles Modified Electrode and Its Application to Cr(VI) Detection. J. Electroanal. Chem. 2011, 652, 60−65. (12) Hirano, L. A.; Escote, M. T.; Martins-Filho, L. S.; Mantovani, G. L.; Scuracchio, C. H. Development of Artificial Muscles Based on Electroactive Ionomeric Polymer-Metal Composites. Artif. Organs 2011, 35, 478−483. (13) Tant, M. R.; Mauritz, K. A.; Wilkes, G. L. Ionomers: Synthesis, Structure, Properties and Applications; Blackie Academic & Professional: Glasgow, 1997. (14) Gierke, T. D.; Munn, G. E.; Wilson, F. C. The Morphology in Nafion Perfluorinated Membrane Products as Determined by Wideand Small-Angle X-Ray Studies. J. Polym. Sci., Part B: Polym. Phys. 1981, 19, 1687−1704. (15) Sachdeva, A.; Sodaye, S.; Pandey, A. K.; Goswami, A. Formation of Silver Nanoparticles in Poly (Perfluorosulfonic) Acid Membrane. Anal. Chem. 2006, 78, 7169−7174. (16) Kumar, R.; Pandey, A. K.; Dhara, S.; Misra, N. L.; Ramagiri, S. V.; Bellare, J. R.; Goswami, A. Inclusion of Silver Nanoparticles in Host Poly (Perfluorosulfonic) Acid Membrane Using Ionic and Nonionic Reductants. J. Membr. Sci. 2010, 352, 247−254. (17) Sun, Y.-P.; Atorngitjawat, P.; Lin, Y.; Liu, P.; Pathak, P.; Bandara, J.; Elgin, D.; Zhang, M. Nanoscale Cavities in Ionomer Membrane for the Formation of Nanoparticles. J. Membr. Sci. 2004, 245, 211−217. (18) Hasegawa, T.; Strunskus, T.; Zaporotjenko, V.; Faupel, F.; Mizuhata, M. Preparation of Silver Nanoparticles-Nafion Membrane Composite by Photoreduction Process. ECS Trans. 2012, 41, 9−18. (19) Hsu, W. Y.; Gierke, T. D. Ion Transport and Clustering in Nafion Perfluorinated Membranes. J. Membr. Sci. 1983, 13, 307−326. (20) Hsu, W. Y.; Gierke, T. D. Elastic Theory for Ionic Clustering in Perfluorinated Ionomers. Macromolecules 1982, 15, 101−105. (21) Gierke, T. D.; Munn, G. E.; Wilson, F. C. In Morphology of Perfluorosulfonated Membrane Products. Perfluorinated Ionomer Membranes; Eisenberg, A., Yeager, H. L., Eds.; ACS Symposium Series No. 180; American Chemical Society: Washington, DC, 1982; pp 195− 216. (22) Lee, E. M.; Thomas, R. K.; Burgess, A. N.; Barnes, D. J.; Soper, A. K.; Rennie, A. R. Local and Long Range Structure of Water in a Perfluorinated Ionomer Membrane. Macromolecules 1992, 25, 3106− 3109. (23) Rollet, A.-L.; Gebel, G.; Simonin, J.-P.; Turq, P. A SANS Determination of the Influence of External Conditions on the Nanostructure of Nafion Membrane. J. Polym. Sci., Part B: Polym. Phys. 2001, 39, 548−558. (24) Rollet, A.-L.; Diat, O.; Gebel, G. A New Insight into Nafion Structure. J. Phys. Chem. B 2002, 106, 3033−3036. (25) Lehmani, A.; Durand-Vidal, S.; Turq, P. Surface Morphology of Nafion 117 Membrane by Tapping Mode Atomic Force Microscope. J. Appl. Polym. Sci. 1998, 68, 503−508. (26) Xue, T.; Trent, J. S.; Osseo-Asare, K. Characterization of Nafion Membranes by Transmission Electron Microscopy. J. Membr. Sci. 1989, 45, 261−271. (27) MacMillan, B.; Sharp, A. R.; Armstrong, R. L. An NMR Investigation of the Dynamical Characteristics of Water Absorbed in Nafion. Polymer 1999, 40, 2471−2480. (28) McLean, R. S.; Doyle, M.; Saner, B. B. High Resolution Imaging of Ionic Domains and Crystal Morphology in Ionomers using AFM Techniques. Macromolecules 2000, 33, 6541−6550. (29) Mauritz, K. A.; Moore, R. B. State of Understanding of Nafion. Chem. Rev. 2004, 104, 4535−4585.

membrane. However, the SDC values are significantly enhanced in ascorbic acid reduced samples. This enhancement is due to incorporation of bigger nanostructures (nanorod and nanosponge) in the membrane matrix. The data given in Table 1 show that self-diffusion mobility increased approximately 17% and 70% for Na+ and Cs+ counterions, respectively, in A2 samples containing big nanosponge-like Ag nanostructures.

■

CONCLUSIONS The studies carried out in the present work indicated that local conditions during reduction and subsequent reorganizations of matrix affect shape, size dispersion, and distribution of Ag nanostructures formed in Nafion. The monodisperse spherical nanoparticles (5 nm) were formed by formamide reduction at 45 °C and Na+ post reduction counterionic form of Nafion. The shape, size, and distribution of Ag nanostructures formed by ascorbic acid reduction were found to be highly dependent on loading of Ag+ ions in Nafion matrix. Above the critical concentration of Ag+ ions, one-dimensional growth in the form of the nanorods having ∼8 nm diameter and varying length was observed, which was not expected in Nafion matrix. At lower critical concentration of Ag+ ions, the Ag seeds (atoms or clusters) agglomerated in localized zones resulting in the formation of big sponge-like nanostructures. In the absence of chemical cross-linking, Nafion matrix could reorganize to attain original water clusters network in a given counterionic form. However, the water clusters network in matrix was modified significantly after incorporation of bigger Ag nanoparticles and nanorods. These led to 17% and 70% enhancements of diffusion mobility of Na+ and Cs+ counterions, respectively, in Nafion matrix.

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +91-22-25594566. Fax: +91-22-25505150/25505151. Email: [email protected]. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS The authors thank the CryoTEM Facility of IIT Bombay for TEM access. REFERENCES

(1) Krishnamoorti, R.; Vaia, R. A. Polymer Nanocomposites. J. Polym. Sci., Part B: Polym. Phys. 2007, 45, 3252−3256. (2) Vaia, R. A.; Maguire, J. F. Polymer Nanocomposites with Prescribed Morphology: Going Beyond Nanoparticle-Filled Polymers. Chem. Mater. 2007, 19, 2736−2751. (3) Paul, D. R.; Robenson, L. M. Polymer Nanotechnology: Nanocomposites. Polymer 2008, 49, 3187−3204. (4) Schaefer, D. W.; Justice, R. S. How Nano are Nanocomposites? Macromolecules 2007, 40, 8501−8517. (5) Sanchez, C.; Julian, B.; Belleville, P.; Popall, M. Applications of Hybrid Organic-inorganic Nanocomposites. J. Mater. Chem. 2005, 15, 3559−3592. (6) Corma, A.; Garcia, H. Supported Gold Nanoparticles as Catalysts for Organic Reactions. Chem. Soc. Rev. 2008, 37, 2096−2126. (7) Wang, J.; Musameh, M.; Lin, Y. Solubilisation of Carbon Nanotubes by Nafion Toward the Preparation of Amperometric Biosensors. J. Am. Chem. Soc. 2003, 125, 2408−2409. (8) Selvaraju, T.; Ramaraj, R. Nanostructured Copper ParticlesIncorporated Nafion-Modified Electrode for Oxygen Reduction. Pramana 2005, 65, 713−722. K



Article

(30) Elliott, J. A.; Wu, D.; Paddison, S. J.; Moore, R. B. A Unified Morphological Description of Nafion Membranes from SAXS and Mesoscale Simulations. Soft Matter 2011, 7, 6820−6827. (31) Beers, K. M.; Hallinan, D. T., Jr.; Wang, X.; Pople, J. A.; Balsara, N. P. Counterion Condensation in Nafion. Macromolecules 2011, 44, 8866−8870. (32) Kusoglu, A.; Modestino, M. A.; Hexemer, A.; Segalman, R. A.; Weber, A. Z. Subsecond Morphological Changes in Nafion During Water Uptake Detected by Small-Angle X-ray Scattering. ACS Macro Lett. 2012, 1, 33−36. (33) Mistry, M. K.; Choudhury, N. R.; Dutta, N. K.; Knott, R. Nanostructure Evolution in High-Temperature Perfluorosulfonic Acid Ionomer Membrane by Small-Angle X-ray Scattering. Langmuir 2010, 26, 19073−19083. (34) Gebel, G.; Moore, R. B. Small-Angle Scattering Study of Short Pendant Chain Perfluorosulfonated Ionomer Membranes. Macromolecules 2000, 33, 4850−4855. (35) MacKnight, W. J.; Taggart, W. P.; Stein, R. S. Model for the Structure of Ionomers. J. Polym. Sci., Polym. Symp. 1974, 45, 113−128. (36) Roche, E. J.; Stein, R. S.; Russell, T. P.; MacKnight, W. J. SmallAngle X-ray Scattering Study of Ionomer Deformation. J. Polym. Sci., Polym. Phys. Ed. 1980, 18, 1497−1512. (37) Schmidt-Rohr, K.; Chen, Q. Parallel Cylindrical Water Nanochannels in Nafion Fuel-cell Membranes. Nat. Mater. 2008, 7, 75−83. (38) Rubatat, L.; Rollet, A. L.; Gebel, G.; Diat, O. Evidence of Elongated Polymeric Aggregates in Nafion. Macromolecules 2002, 35, 4050−4055. (39) Rollet, A. L.; Diat, O.; Gebel, G. Transport Anisotropy of Ions in Sulfonated Polyimide Ionomer Membranes. J. Phys. Chem. B 2004, 108, 1130−1136. (40) Rubatat, L.; Gebel, G.; Diat, O. Structure of Nafion: Matching Fourier and Real Space Studies of Corresponding Films and Solutions. Macromolecules 2004, 37, 7772−7783. (41) Haubold, H.-G.; Vad, Th.; Jungbluth, H.; Hiller, P. Nanostructure of NAFION: A SAXS Study. Electrochim. Acta 2001, 46, 1559−1563. (42) Starkweather, H. W., Jr. Crystallinity in Perfluorosulfonic Acid Ionomers and Related polymers. Macromolecules 1982, 15, 320−323. (43) Heitner-Wirguin, C. Recent Advances in Perfluorinated Ionomer Membranes: Structure, Properties and Applications. J. Membr. Sci. 1996, 120, 1−33. (44) Fujimura, M.; Hashimoto, T.; Kawai, H. Small-Angle X-ray Scattering Study of Perfluorinated Ionomer Membranes. 1. Origin of Two Scattering Maxima. Macromolecules 1981, 14, 1309−1315. (45) Fujimura, M.; Hashimoto, R.; Kawai, H. Small-Angle X-ray Scattering Study of Perfluorinated Ionomer Membranes. 2. Models for Ionic Scattering Maximum. Macromolecules 1982, 15, 136−144. (46) Gebel, G. Evidence Structural Evolution of Water Swollen Perfluorosulfonated Ionomers from Dry Membrane to Solution. Polymer 2000, 41, 5829−5838. (47) Bahadur, J.; Sen, D.; Mazumder, S.; Paul, B.; Bhatt, H.; Singh, S. G. Control of Buckling in Colloidal Droplets During EvaporationInduced Assembly of Nanoparticles. Langmuir 2012, 28, 1914−1923. (48) Sen, D.; Bahadur, J.; Mazumder, S.; Verma, G.; Hassan, P. A.; Bhattacharya, S.; Vijai, K.; Doshi, P. Nanocomposite Silica Surfactant Microcapsules by Evaporation Induced Self-Assembly: Tuning the Morphological Buckling by Modifying Viscosity and Surface Charge. Soft Matter 2012, 8, 1955−1963. (49) Bahadur, J.; Sen, D.; Mazumder, S.; Bhattacharya, S.; Frielinghaus, H.; Goerigk, G. Origin of Buckling Phenomenon During Drying of Micrometer-Sized Colloidal Droplets. Langmuir 2011, 27, 8404−8414. (50) Sen, D.; Melo, J. S.; Bahadur, J.; Mazumder, S.; Bhattacharya, S.; D’Souza, S. F.; Frielinghaus, H.; Goerigk, G.; Loidl, R. Arrest of Morphological Transformation During Evaporation-Induced SelfAssembly of Mixed Colloids in Micrometric Droplets by Charge Tuning. Soft Matter 2011, 7, 5423−5429.

(51) Sen, D.; Khan, A.; Bahadur, J.; Mazumder, S.; Sapra, B. K. Use of Small-Angle Neutron Scattering to Investigate Modifications of Internal Structure in Self-Assembled Grains of Nanoparticles Synthesized by Spray Drying. J. Colloid Interface Sci. 2010, 347, 25−30. (52) Sen, D.; Mazumder, S.; Melo, J. S.; Khan, A.; Bhattacharya, S.; D’Souza, S. F. Evaporation Driven Self-Assembly of a Colloidal Dispersion During Spray Drying: Volume Fraction Dependent Morphological Transition. Langmuir 2009, 25, 6690−6695. (53) Qin, Y.; Ji, X.; Jing, J.; Liu, H.; Wu, H.; Yang, W. Size Control Over Spherical Silver Nanoparticles by Ascorbic Acid Reduction. Colloids Surf., A 2010, 372, 172−176. (54) Goswami, A.; Acharya, A.; Pandey, A. K. Study of Self-Diffusion of Monovalent and Divalent Cations in Nafion-117 Ion-Exchange Membrane. J. Phys. Chem. B 2001, 105 (38), 9196−9201. (55) Sodaye, S.; Agarwal, C.; Goswami, A. Study on Multicomponent Diffusion of Ions in Poly (perfluorosulfonated) Ion-Exchange Membrane Using Radiotracers. J. Membr. Sci. 2008, 314, 221−225. (56) Pedersen, J. S. Determination of Size Distributions from SmallAngle Scattering Data for Systems with Effective Hard-Sphere Interactions. J. Appl. Cryst. Sci. 1994, 7, 595−608. (57) Pedersen, J. S. Analysis of Small-Angle Scattering Data From Colloids and Polymer Solutions: Modeling and Least Squares Fitting. Adv. Colloid Interface Sci. 1997, 70, 171−210. (58) Suresh, G.; Pandey, A. K.; Goswami, A. Self-Diffusion Coefficients of Water in Nafion-117 Membrane with Multivalent Counterions. J. Membr. Sci. 2006, 284, 193−197. (59) Suresh, G.; Scindia, Y. M.; Pandey, A. K.; Goswami, A. Isotopic and Ion-Exchange Kinetics in the Nafion-117 Membrane. J. Phys. Chem. B 2004, 108, 4104−4110.

L


Local Conditions Influencing In Situ Formation of ... - ACS Publications

Recommend Documents