TEM images of the different gold nanoparticles. a Prepared (sample A). b Cleaned with centrifugation (sample B). c , d Cleaned with acetone (sample C) 

Context in source publication

Context 1

... applications. Optical and catalytic properties observed for such nanoparticles are found to depend mainly on their size, shape, and the interactions between the light and their surface electrons [1 – 3]. This effect, known as surface plasmon resonance, is used in materials science, chemistry, engineering physics, and biotechnology. In these areas, the conservation of the particles shape is important in order to obtain the targeted wavelength suitable for many applications in bio-imaging, cancer treatment, and surface- enhanced Raman spectroscopy or infrared spectroscopy [3 – 9]. Furthermore, the energy level of the surface electrons promotes the affinity of such particles towards the oxygen reduction reaction, leading to surprising catalytic properties [10, 11]. Because of their small size, the surface of such nanoparticles offers interesting structure and morphology which contribute to the enhancement of their catalytic properties. In electrocatalysis, the surface structure in term of surface site, facets, planes, or domains is a determining factor in the material performance. Moreover, the mecha- nism and kinetics of electrochemical reactions greatly depend on this structure and the electrons therein. Synthesis methods of supported catalysts are mostly developed in order to reach high active surface area. For several applications, unsupported catalysts are synthesized using various chemical approaches [12]. The challenge commonly encountered when investigating metallic nanoparticles is to perform their electrochemical characterization in a surfactant-free medium in order to eliminate any interference of the latter [13, 14]. Consequently, the cleaning step in the nanoparticle preparation appears to be critical because it is expected to maintain the particle shape and should not modify their structure or morphology. To date, in the few papers which report on the electrochemical surface characterization of the gold nano- particles [13 17], the exact effect of the surfactant removal has never been mentioned or presented. Generally, it is assumed but without any direct experimental evidence that the nanoparticles maintain their shape after the cleaning step. The aim of this communication is to demonstrate by various characterization methods that the different cleaning procedures used in the Au nanoparticle synthesis have a pronounced impact on their behavior and in some cases might even lead to their agglomeration. The synthesis of gold nanoparticles was performed using the water-in-oil microemulsion of water/poly-ethylenegly- col-dodecylether (Brij® 30)/n-heptane. This method is extensively described by Capek et al. [18]. Two microemulsion solutions were prepared: the first contained the metallic precursor (HAuCl 4 ) and the other the reducing agent (NaBH 4 ). The surfactant in both cases is the Brij® 30. It is well-known that the surfactant permits to control the particles size and shape. When the two solutions are mixed, an intense red solution appears which corresponds to the formation of colloidal gold nanoparticles formed in a reduction reaction. The surface energy favors the formation of spherical particles. Once the gold nanoparticles were obtained, the solution was divided into three parts: the part A was kept as prepared for characterization; the part B was cleaned by centrifugation and water rinsing; and the part C was cleaned using acetone and water. Electrochemical characterization was performed in a supporting electrolyte of 0.1 mol L − 1 NaOH. The latter was prepared using NaOH pellets (99%, Merck) and ultrapure water (MilliQ Millipore water purification system). In the case of the samples B and C, the color of the solution changed after the cleaning, from intense red to light blue to dark blue, respectively. It is well-known that gold colloidal solutions have a specific color, which depends on the size, shape, and agglomeration of nanoparticles. This color results from the interaction between the light and the surface electrons of the nanoparticles. In our case, the color obtained for the samples B and C indicates that there is some anisotropy created inside the colloidal solutions. Figure 1 shows UV – vis spectra of the three samples. Sample A exhibits a maximum absorbance at the wavelength 1 =620 nm, which points to the presence of spherical nanoparticles. According to the literature [19], this 1 value suggests an average particles size greater than 5 nm. The spectrum for the sample B shows a well-defined, large peak around 620 nm. However, in comparison with the sample A, the increase of the full width at half maximum and the shape of the spectrum at 1 >620 nm point to an increase of particles size and the presence of anisotropy in shape. The spectrum for the sample C does not reveal any sharp peak. Thereby, the cleaning step, that involves acetone, leads to a full agglomeration of the particles and the increase of the anisotropy. This observation is supported by our TEM analysis data (Fig. 2) which clearly reveal agglomerated nanoparticles. Indeed, spherical particles are observed in the sample B with little agglomerations being present, while a fully agglomerated structure is observed for the sample C. Thus, the TEM images support our conclusions drawn on the basis of UV – vis experiments, which clearly show that the use of acetone for cleaning unsupported gold nanoparticles leads systematically to their agglomeration. This phenomenon is not observed in the case of carbon- supported gold nanoparticles synthesized by the water-in- oil microemulsion method [20]. The unsupported gold nanoparticles synthesized in this work were also characterized electrochemically. Hernandez et al. [17] showed that in an alkaline medium the voltammogram for gold nanoparticles prepared by the water-in-oil microemulsion and cleaned with acetone was similar to that obtained for the bulk-type Au electrode. Nevertheless, they assumed that the particle shape was maintained after the cleaning step with acetone. In this contribution, we demonstrate for the first time that the cleaning of the nanoparticles under these conditions (i.e., with acetone) leads to their complete agglomeration. Electrochemical characterization of such agglomerated crystallites by cyclic voltammetry in a supporting electrolyte results in a profile characteristic a bulk-type material. In order to advance our analysis, the surface of the unsupported gold nanoparticles was characterized electrochemically using the underpotential deposition (UPD) of lead. Elsewhere [13 – 17], it was shown that the UPD of lead was one of the suitable methods for characterizing the exposed facet of particles at the surface of a gold crystallite because it was a structure-sensitive approach. Specifically, the potential of the Pb UPD desorption is known to depend on the exposed facet. According to the above-cited literature, the potential of the desorption of Pb after its underpotential deposition strongly depends on the crystallite facet orientation. To apply this characterization approach to our samples, a 0.1 mole L − 1 NaOH solution containing 10 − 3 mole L − 1 Pb(NO 3 ) 2 was used. Before each deposition of Pb UPD , a CV profile corresponding to the formation and dissolution of a lead oxide layer was carried out in order to remove any remaining organic compounds on the nanoparticles surface. The shape of the CV profile in the presence of Pb UPD was observed to improve after the electrochemical PbO 2 layer treatment, as previously described by Hernandez et al. [16]. Therefore, we do not present this experiment in this short communication. Figure 3 shows CV desorption profiles of Pb for the three different gold samples in 0.1 mole L − 1 NaOH containing 10 mole L Pb(NO 3 ) 2 and recorded at a scan rate of 50 mV s − 1 and at room temperature (20±2 °C). The desorption of Pb from the (111) facets and from both the (110) facets and defects sites [16] is clearly observed for all samples at 0.425 and 0.575 V vs. RHE, respectively. The CV response related to the Pb UPD desorption from the (110) facet is not observed in the case of the sample A, while in the case of the sample C a small feature appears in the CV profile at 0.47 V vs . RHE. The latter sample presents a well-defined Pb UPD desorption profile which nearly corresponds to the profile regularly observed for a bulk-type polycrystalline gold electrode. Furthermore, the peak corresponding to the Pb desorption from the facet (110) splits (Fig. 3). Our CV characterization leads to the observation that despite the apparent difference in mor- phology observed for the samples B and C, the UPD of lead did not reveal any significant changes, thus suggesting that this approach might not be the most suitable. In summary, gold nanoparticles can be synthesized using a surfactant as stabilizer that determines the particle size distribution. However, it needs to be remembered that the particles are sensitive to the method used to clean them and remove the surfactant from their surfaces. We clearly showed that cleaning the Au nanoparticles with acetone led to their complete agglomeration. Centrifugation seems to be a more suitable approach but requires improvements because at the present stage it does not completely suppress agglomeration. The analysis of nanoparticles ’ orientation using Pb UPD CV desorption profiles has its limitations because similar CV profiles are obtained in the case of either partially or completely agglomerated Au nanoparticles. However, this approach shed light on the existence of specific facets in particles (not necessarily nanometric in size). We believe that a materials science technique, such as TEM, has to be employed in order to prove that agglomeration of nanoparticles does not ...