The effect of polyol on multiple ligand capped silver alloyed nanobimetallic particles in tri-n-octylphosphine oxide and oleic acid matrices

All inorganic salts, solvents and chemical reagents used were analytical grade, and were purchased from Sigma-Aldrich Corporation, UK. They are as follow: tri-n-octylphosphine oxide (TOPO), oleic acid (OA), silver nitrate, nickel(II) acetate, cobalt(II) acetate, palladium(II) chloride, platinum(IV) chloride, ruthenium(III) chloride, dioctylether (DOE), glycerol (GLY), ethylene glycol (EG), methanol (99.5%.w/w) and (99.5%.w/w) ethanol.

Fractal monodispersed bimetallic Ag/M (M = Co, Ni, Ru, Pd, Pt) nanostructured particles were prepared by the seed growth or successive addition method [15, 32]. Briefly, 6.162–6.278 g (15.94–16.25 mmol), 99%w/w TOPO (capping agent) was measured into a round bottom flask containing a magnetic stirrer and it was gradually heated to melt. 4.475 g (15.75 mmol) oleic acid (OA) was added as dispersing solvent. The mixture was stirred and gradually heated at the rate of 2.0 °C min⁻¹ to 200 °C followed by the injection of 0.140 g (0.56 mmol) Co(CO₂CH₃)₂.4H₂O dispersed in 5.0 ml dioctylether and 5.0 ml ethylene glycol (EG) or glycerol (GLY) as reducing agent. A rapid colour change to brown was observed. The resultant mixture was left to allow for the growth of Co nanoparticles which also acted as a seed for the next reaction. Then, 0.069–0.144 g (0.41–0.85 mmol) AgNO₃ similarly dispersed in 5.0 ml dioctylether (DOE) was injected into the colloidal mixture, followed by hot injection of 5.0 ml aliquot of ethylene glycol. The reaction continued for 2 h with continuous stirring. After the hot injection of silver nitrate solution, a shiny brown coloured Ag/Co sol was obtained. While hot, the Ag/Co sol was copiously washed with methanol several times, and centrifuged at 4400 rpm for 10–15 min to remove excess unreacted stabilizer. The Ag/Co sol was later redispersed in ethanol.

The procedure was subsequently repeated for other molecular precursors: 0.165 g (0.66 mmol) Ni(CO₂CH₃)₂.4H₂O; 0.048 g (0.23 mmol) RuCl₃; 0.071 g (0.40 mmol) PdCl₂; 0.074 g (0.22 mmol) PtCl₄ to produce Ag/Ni, Ag/Ru, Ag/Pd and Ag/Pt TOPO/OA capped nanocomposites respectively. The centrifuged sol obtained after decantation was redispersed in ethanol and cleaned in ultrasonic bath at 50 °C for 60 min before further characterization.

Optical characterization was done using a UV–vis spectrophotometer (Varian Cary 50 Conc) to carry out the optical measurements and the samples were placed in silica cuvettes (1 cm path length), using toluene as a reference solvent. A luminescence spectrometer (Perkin-Elmer LS 55) was used to measure the photoluminescence of the particles. The samples were placed in a quartz cuvette (1 cm path length).

The crystalline phase was identified by x-ray diffraction (XRD), employing a scanning rate of 0.05° min⁻¹ in a 2θ range from 20° to 80°, using a diffractometer (Bruker AXS D8) equipped with nickel filtered Cu-Kα radiation (λ = 1.5406 Å) at 40 kV, 40 mA and at room temperature. The morphology and particle sizes of the samples were characterized by a JEOL 1010 TEM with an accelerating voltage of 100 kV, Megaview III camera, and Soft Imaging Systems iTEM software. The detail morphological and structural features were investigated using HRTEM images with a JEOL 2010 TEM operated at an accelerating voltage of 200 kV. A cursory survey of surface property and high resolution spectra of nanoparticles were collected using XPS PHI 5000 Versaprobe-Scanning ESCA Microprobe, with 100 μm, 25 W, 15 kV Al monochromatic x-ray beams.

The UV–vis absorption spectra of Ag alloyed nanoparticles capped by TOPO/OA with different metal precursors/ligand mole ratios are shown in figure 1. The optimum stoichiometric ratios for the co-precipitation reactions which produced the TOPO/OA capped Ag alloyed bimetallic sols with different metal precursors/capping agent mixtures are shown in table 1. At the approximate ratios, colloidal dispersions were formed which were easily re-dispersed in ethanol and the optical absorption of Ag alloyed metal sols was conveniently measured.

Zoom In Zoom Out Reset image size

Furthermore, a trend was observed in TOPO/OA stabilized Ag alloyed bimetallic sols (figure 1). The surface plasmon band (SPB) for AgCo, AgNi and AgRu occurred at 447, 462 and 467 nm respectively, while it was almost extinct in the spectra of AgPd and AgPt sols, showing very weak band at 465 and 427 nm, respectively. Moreover, the SPB appeared broad with consequent shift towards the visible region. This was observed as result of contribution from multiple organic capping ligands made up of tri-n-octylphosphine oxide and oleic acid [24]. The phosphinate oxygen has a weaker binding effect, therefore, it is displaced by the carboxyl group of oleic acid [32, 33], thereby, accounting for the observed strong band and red electronic shift in the absorbance spectra of the TOPO/OA capped Ag alloyed bimetallic sols.

The absorption band gaps of the Ag/M sols which were estimated by the direct band gap method [34, 35], using the cut-off band edges from figure 1 are presented in table 1.

The absorption edges were red-shifted from that of Ag, Co, Ni, Ru, Pd and Pt bulk crystals given as 3.99 eV [36], 3.17 eV [37, 38], 5.20 eV [39], 3.2 eV [40], 7.6 eV [41] and 3.0 eV [42, 43].

The photoluminescence measurements of TOPO/OA capped Ag alloyed bimetallic sols was carried out in ethanol, the resulting emission spectrum (figure 2) was analysed to deduce relative position of emission spectra of the different Ag alloyed bimetallic sols. Generally, all the TOPO/OA capped bimetallic sols exhibited strong emission peak between 378 and 385 nm, strongly indicating occurrence of alloy formation with likelihood of surface enrichment by silver (Ag) atoms. Further excitation of the bimetallic sols at 232 and 325 nm showed no significant shift in the emission properties confirming the quantum properties of the as prepared nanosols according to Kasha rule [44] that the emission spectra of potential fluorophores are typically independent of excitation wavelength.

Zoom In Zoom Out Reset image size

In TOPO/OA capped Ag/Ru sols (figure 2), the most intense emission peak was observed at 385 nm for an excitation at 340 nm which occurred at a little relatively lower wavelength than TOPO/OA capped Ag/Co sol, the difference observed can be attributed to the nature of electronic interaction between the metal core and chromophoric group in oleic acid and tertiary alkylphosphine used as capping agents [44, 45]. This effect usually results in fluorescent quenching [46], a phenomenon that leads to decrease in intensity is given by

$$\begin{eqnarray*}&&{F}_{0}/F=1+K[Q]=1+{K}_{{\rm{q}}}{\tau }_{0}[Q],\end{eqnarray*}$$

where F₀/F is decrease in fluorescence intensity, K is the Stern–Volner quenching constant, K_q is the bimolecular quenching constant; τ₀ is the unquenched lifetime, [Q] is the quencher concentration.

It is apparent that fluorescent lifetime and quantum yield efficiency in Ag/Ru TOPO/OA capped sols can be used as a measure of their activity as sensors or actuators paving way for these new materials in effluent gas sensing.

The TEM image (figure 3(a)) and particle size distribution histogram (figure 4(a)) of TOPO/OA capped Ag/Co alloyed nanobimetallic particle reveal particles in the form of clusters monodispersed with a predominantly cubic or close to spherical morphology. The diameter of particles (n = 25) averaged 7.02 ± 1.64 nm, with the lattice quite well-resolved and equally spaced showing their crystalline nature. The lattice spacing was calculated to be 2.30 ± 0.02 Å which is close to bulk Ag which has a lattice space of 2.27 Å. Assuming Vegard's law [47], the observed lattice spacing would yield a nominal composition of 1:2 Ag/Co for a starting composition of 1:1 Ag/Co [48]. This value should be treated with some caution since a little variation in the measurement of the lattice spacing could lead to a large error in the final composition. The corresponding HRTEM image of the TOPO/OA capped Ag/Co (figure 3(c)) also shows lattice fringes. The external shell of Co surrounding the Ag core is evident in the HRTEM image (inset (e) of figure 3(c)) shown by arrows confirming the core–shell structure of the material.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The TEM image of the TOPO/OA capped Ag/Ni nanobimetallic particles (figure 3(d)) reveals their polydispersed nature corroborated by the particle size distribution histogram (figure 4(b)); two different particle sizes were identified. Moreover, the particles showed evidence of cluster in cluster formation (figure 3(d)), in which two Ag cores were in a Ni shell. The diameter of the nanocomposites (n = 25) averaged 22.21 ± 10.27 nm which was also in agreement with Vegard's law [49].

The TEM images of the TOPO/OA capped Ag/Pd nanoparticles (figure 5(a)) showed monodispersed roughened edge cubic structures with the particle diameter (n = 25) averaged 16.50 ± 9.72 nm, which is evident by the particle size distribution histogram (figure 6(a)).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The HRTEM image (figure 5(b)) shows alloy formation with an affinity for anisotropic longitudinal growth [50]. For the formation of alloy, it was necessary that the reduction of both metal ions took place simultaneously. Therefore, different concentrations of metal ions were taken for making the mixed particles. However, the mechanism for alloying of small particles is still not clear. Kinetic consideration indicates that, for the two metals to mix, the diffusion coefficients are expected to be many orders of magnitude larger than those of the bulk materials [51].

The diameters of the TOPO/OA capped Ag/Pt (n = 25) and Ag/Ru nanoclusters (n = 10) averaged 4.70 ± 0.93 and 26.82 ± 6.35 nm respectively. The former consists of monodispersed nanosized particles with well-defined lattice spacing (figures 5(c) and (d)), this deduction is supported by the particle size distribution graph (figure 6(b)). Whereas, the latter is made up of 1D quantized polydispersed nanorods and 2D pseudo nanocubes (figure 7(a)), similar to face-centered cubic (fcc) Ag. The corresponding size distribution histogram is shown in figure 7(b). However, the mechanism of the growth process for this nanohybrid can be regarded as reaction limited, in which, etching by oxygen and Cl⁻ anions present in RuCl₃ promoted the formation of Ag/Ru nanocubes and nanorods [52]. In all, we were able to achieve stable monodispersed silver alloyed NPs with tunable morphology properties as a result of controlled capping by two organic ligands (TOPO/OA), different from our previous work when HDA was used to synthesise silver nanoalloyed particles [14].

Zoom In Zoom Out Reset image size

The p-XRD pattern for TOPO/OA capped Ag/Ru bimetallic nanoparticles is shown in figure 7(c). We carried out p-XRD measurement for Ag/Ru nanoparticles due to its rareness in the literature. The appearance of well resolved reflection peaks in the p-XRD pattern supports the formation of nearly crystalline nanoparticles. However, the observed reflections for the bimetallic cluster were slightly different from that of individual monometallic components, showing the existence of alloy phases corresponding to peaks of fcc Ag, hexagonal close-packed (hcp) Ru and tetragonal (tet) RuO phases which portend the formation of alloyed Ag/Ru particles in agreement with TEM and optical measurement. The fcc {111} Ag plane occurred at 38.34° (2θ) and the {200} plane occurred at 44.91° (2θ) for a mole ratio 1:2 Ag/Ru. There is also an overlap between the hcp {013} Ru plane and the {311} Ag plane at 77.27°–77.69° (2θ) further establishing the presence of an alloy phase and formation of Ag/Ru alloyed nanoparticles. The particle size calculated from the [111] peak at FWHM using the Scherrer equation was 23.07 nm, which is in good agreement with the TEM results.

For the XPS analysis, one of the nanocomposites, Ag/Ru TOPO/OA was considered for characterization with sputtering carried out on the sample at the surface monolayers. The spectrum and the high resolution scans of the core levels of Ag/Ru nanocluster capped with TOPO/OA are presented in figure 8. The high resolution surface scan spectra of Ag/Ru NPs (figure 8(b)) showed the presence of Ag, Ru and O according to their binding energies. The stoichiometric ratio of Ag to Ru in AgRuO_x, determined from the wide area XPS scan was 1:1.8 (figure 8(a)).

Zoom In Zoom Out Reset image size

The deconvolution of Ag (3d_5/2) core levels (figure 8(c)) resulted in a peak at 367.44 eV binding energy (BE) with an auger parameter, AgM₄₅N₂₃N of 1227.9 eV assigned to Ag, while a BE of 366.8 eV corresponding to (3d_5/2) core levels with auger parameter, O KLL of 977.4 eV was assigned to AgM. The chemical shift or change in 3d photoelectron peaks for the Ag core levels was 0.2–0.3 eV, while the 3d spin–orbit split was 6.0 eV. Both the Ag and Ru peaks were integrated as intense peaks in the spectrum. But the deconvolution of Ru (3d_5/2) core levels (figure 8(d)) resulted in a peak at 280.1 eV (BE) which was assigned to RuO [53].

The interaction between Ru and Ag made the Ag (3d) BE to increase by a factor of 0.04 eV, and Ru (3d) BE decreased obviously to 278.6 eV. The presence of Ag withdrew electrons from nearby Ru and thus made Ru electropositive, which was beneficial to produce more active sites where electrophilic oxygen species can be absorbed to increase capping group selectivity [54].

The deconvolution of XPS high-resolution scan for the O (1 s) core level (figure 8(e)) showed the presence of three different peaks. The main peak centered at 531.69 eV has been attributed to RuO–O in the AgRuO_x structure, while the peak at 531.89 eV was associated with the O (1 s) in the Ru–O species. Therefore, in the Ag/Ru nanoparticles charge transfer from ruthenium to oxygen was reduced, thus increasing the shielding effect of the valence electrons in Ru ions which in turn decreased the BE of core electrons in Ru ion. This complex modification of ruthenium binding state was associated with the variation in the number of broken bonds per ruthenium ion due to quantum size effect, which led to the formation of oxygen vacancies in the host [55].

Whereas, AgRu nanoparticles in TOPO/OA matrices indicated significant reduction potential, the energy band gap of AgPt from table 1 points to the fact that it could be a likely better reducing agent for electrocatalytic reaction. This deduction indicating potential application of AgPt nanoparticles as nanocatalysts can be corroborated by its relative nanoscale dimension 4.70 ± 0.93 nm obtained from TEM images. In that order, we consider AgCo nanoparticles in TOPO/OA matrices as having a relatively small reduction potential, with an energy band gap of 2.77 eV and an average particle size of 7.02 ± 1.64 nm.