Prospective use of the potentiostatic triple pulse strategy for the template-free electrosynthesis of metal nanoparticles

Abstract

The electrosynthesis of metal nanoparticles with control over size and dispersion is a challenging task on a macrodisk electrode in the absence of any chemical or physical template. A potentiostatic triple pulse strategy (PTPS) was designed for the electrosynthesis of monodispersed lead nanoparticles (PbNPs) on glassy carbon electrodes. The results were compared to those obtained using the conventional potentiostatic double pulse strategy (PDPS). A switchover of the nucleation mechanism from progressive in PDPS to instantaneous in PTPS was observed for the first time during the final controlled growth step, resulting in smaller and better monodispersed PbNPs (∼8 ± 2 nm).

---

Electrosynthesis is a promising technique for preparing supported metal nanoparticles of controlled size, shape, crystallographic orientation, mass, thickness and morphology, which are important for many applications including diamagnetic fluctuations. Lead nanoparticles (PbNPs) with sizes of less than 87 nm can be employed as superconducting materials to study the quantum size effect on superconducting properties¹ as well as zero-dimensional fluctuation diamagnetism.² Guin et al. reported the preparation of capped hemispherical PbNPs on a template-free gold (Au) electrode at room temperature from an aqueous solution of 1 mM Pb(ClO₄)₂ in 0.1 M HClO₄.³ Potentiostatic triple pulse strategy (PTPS), which is very different from the strategy developed by Adzic et al. and Guin et al.,^4,5 was introduced for the first time in that work, and the effects of pulse potential and duration on particle size and size dispersion were reported. The heights of the capped hemispherical PbNPs varied from 2–18 nm, while the lateral size (i.e., chord of NPs) varied in the sub-micron range. Henceforth, we express the average height of NPs and the variation (i.e., range) in the heights of NPs by the terms “size” and “size dispersion”, respectively. A wider size dispersion is termed as “polydispersion”, while a narrower size dispersion is referred to as “monodispersion”. It is interesting to note that the reason behind the considerable improvement in the particle size and size dispersion in PTPS compared to conventional potentiostatic double pulse strategy (PDPS) is not well understood. Gold could be a suitable substrate for the electrodeposition of lead because the interaction energy of Pb with Au is much stronger than those of Pb with Pb and Au with Au. However, to understand the importance of the pulse strategy, the influence of substrate should be removed. From this perspective, glassy carbon (GC) provides a conductive substrate with a low surface energy and exhibits weak metal–substrate interaction. The nucleation of lead on GC has been investigated for decades by single potentiostatic pulse.⁶ Herein, we first report that the number, potential and sequence of potentiostatic pulses have a strong influence on the mechanism of electrocrystallisation as well as the particle size and size dispersion.

A working solution of 1 mM Pb(ClO₄)₂ in 0.1 M HClO₄ was prepared from Metrohm standard aqueous lead ion solution (C_Pb²⁺ = 0.1000 ± 0.0005 M), 60% HClO₄ and Millipore-MilliQ water. After purging the working solution with high purity nitrogen for a minimum of 15 min, electrochemical experiments were carried out at room temperature using an Autolab PGSTAT-30 electrochemical workstation controlled by GPES software. A three-electrode voltammetric cell with a glassy carbon (GC) working electrode (area = 0.071 cm²), platinum wire counter electrode and Ag/AgCl(saturated) reference electrode was employed. All potentials are reported with respect to the Ag/AgCl(saturated) reference electrode. The electrode was polished in between experiments using the standard protocol.⁵ The polishing of the GC electrode in between experiments did not affect the overall activation and quality of the GC surface (Fig. S1,† ESI). The PbNPs were characterised ex situ by AFM (Nanosurf Easyscan 2) by scanning with a silicon tip (CONTR-10) in contact static mode. The recorded AFM images were studied by the inbuilt programs of the Nanosurf Report 4.1 software in order to obtain the particle size histograms.

Fig. 1A shows the cyclic voltammogram of 1 mM Pb(ClO₄)₂ in 0.1 M HClO₄ on the GC electrode at a scan rate of 10 mV s⁻¹. The potential scan starts from 0.4 V and moves towards the cathodic direction. A surge of cathodic current starts at −0.49 V for the reduction of Pb(II)/Pb, resulting in a cathodic peak at −0.58 V. In the reverse scan direction, two crossovers were observed at −0.55 V and −0.47 V between the forward and reverse scans. This characteristic nucleation loop suggests that the deposition of Pb(II) on GC exhibits a nucleation overpotential, which is the extra potential applied for nucleation over the thermodynamic equilibrium potential of Pb(II)/Pb(0).⁷ Since the activities of Pb(0) and GC in solid state are unity, the nucleation overpotential is an indirect measure of the kinetic hindrance of nucleation of Pb on GC compared to the nucleation of Pb on Pb. Therefore, GC does not favour Pb deposition, i.e., the Pb adatoms–adatoms interaction is stronger than the interaction of Pd adatoms with the GC surface. Thus, Pb nuclei are expected to form on GC followed by three-dimensional diffusion controlled growth.⁸ At more positive potentials, the anodic current increased due to the oxidation of the deposited Pb. A sharp anodic peak was observed with a peak potential at −0.41 V, indicating that the anodic dissolution of deposited lead is quite labile under the present experimental conditions. The open circuit potential (OCP, E₀) of the system was measured as −0.41 V, which is in agreement with the expected value (−0.40 V) calculated from the Nernst equation.⁹ The open circuit potential or equilibrium potential for the overall reaction at room temperature is given by the Nernst equation: OCP = E₀ + 0.0295 log[Pb(II)], where E₀ = −0.312 V (vs. Ag/AgCl(saturated)) and [Pb(II)] = 0.001 M.

Fig. 1 (A) Cyclic voltammogram of 1 mM Pb(ClO₄)₂ in 0.1 M HClO₄ on a GC electrode at a scan rate of 10 mV s⁻¹. (B) Chronoamperograms recorded at −0.15, −0.30, −0.40, −0.49, −0.51, −0.53, −0.55, −0.58, −0.60, −0.62, −0.64 and −0.70 V; preceded pulse: E₁ = 0.4 V, t₁ = 60 s. (C) The dimensionless plot of (I/I_m)² vs. t/t_m along with the theoretical curves for instantaneous and progressive nucleation according to the SH model.

The mechanism of nucleation and growth of lead on GC by PDPS was studied by initially holding the potential of the electrode at +0.40 V (E₁) for 60 s (t₁) followed by a sharp change to a different potential (E₂), as indicated by the circles in Fig. 1A. The current transients were recorded at E₂ for a duration ranging between 10–70 s and are shown in Fig. 1B. No characteristic nucleation and growth feature was observed, with the exception of the discharge current originating from the double layer molecular rearrangements within a few ms (for the current transients recorded up to 70 s at −0.15, −0.3 and −0.4 V). The cathodic current decreased sharply and rapidly at −0.49 V as well due to the rearrangement of electrolyte species at the electrode–electrolyte interface, but it slowly increased after a delay time of ∼22 s. This is attributed to an increase in the overall electroactive area due to an increase in the number of Pb nuclei and/or the growth of Pb nuclei. The spherical diffusion zone around each nucleus grew with time; at a time (t_m) of 50.44 s, which corresponds to the current maximum (I_m; −8.4 μA), these spherical diffusion zones overlapped, and mass transfer became linear to the GC surface. The change in diffusion regime led to a decrease in the current with time following the Cottrell equation. As E₂ became more cathodic, the value of t_m decreased, while the value of I_m increased. At E₂ = −0.62 V, the observed values of t_m and I_m were 0.64 s and 49.2 μA, respectively, indicating that the rate of reduction of Pb(II) on GC increased with increasing cathodic overpotential. For E₂ = −0.64 and −0.70 V, no current maxima could be resolved in the experimental time frame due to the fast nucleation and fast change in the diffusion regime due to Cottrell behaviour. No characteristics of 2D or 2D–2D type transients were observed in any case. The type of three-dimensional multiple nucleation and diffusion controlled growth was qualitatively studied by the dimensionless plot of (I/I_m)² vs. t/t_m for two limiting cases (i.e., instantaneous and progressive nucleation) using the Scharifker–Hills (SH) model (eqn (1) and (2)); the results are shown in Fig. 1C.

It is important to mention that the quantitative evaluation of the electrocrystallisation parameters from the SH model is undisputedly sensitive and controversial.^10–12 However, it is still used in the electrodeposition community for a qualitative understanding of the early stage of nucleation and growth.^13–15 It can be inferred from Fig. 1C that at E₂ = −0.49 V, the nucleation and growth was progressive, i.e., new nuclei were continuously formed at a low nucleation rate during the entire deposition process. As the cathodic overpotential increased to −0.62 V, the nucleation and growth tended towards instantaneous, i.e., all the nuclei were created immediately at a high nucleation rate, and the number of nuclei remained constant during the growth process. From our present understanding, we anticipated that the nucleation and growth of Pb nuclei would be instantaneous at E₂ = −0.7 V. In this study, we have not attempted to extract the nucleation parameters from the current transients using the Scharifker–Mostany (SM), Sluyters-Rehbach, Wijenberg, Bosco, Sluyters (SRWBS) and Heerman–Tarallo (HT) models because our recent study revealed that no correlation exists in the nucleation parameters obtained from the SH, SM, SRWBS, and HT models.¹⁰

Based on the abovementioned analysis combined with the results of our earlier published report,³ we designed a similar PTPS strategy in which the first pulse (E₁) was fixed at 0.4 V for 60 s (t₁) followed by a short second pulse (E₂) at −0.7 V for 0.1 ms (t₂) and a third pulse (E₃) at −0.49 V for 70 s (t₃). The current transient was recorded during the third pulse and is shown in Fig. 2A(i). The current transient recorded during PDPS (ii) at −0.49 V (where E₂ for time t₂ was skipped) is overlaid in Fig. 2A. The time of the current maximum significantly decreased to 0.94 s in PTPS from 50.44 s in PDPS, although the maximum current did not change significantly (PTPS: −7.6 μA; PDPS: −8.4 μA). In PTPS, as t₂ increased from 0.1 ms to 1 ms, t_m increased from 0.94 s to 34.1 s, although I_m did not change much (Fig. 2A(iii)). When E₁ was changed to the OCP value (−0.4 V) keeping all other parameters constant, the t_m increased from 0.94 s to 42.1 s, although I_m did not change much (Fig. 2A(iv)). The dimensionless plots of (I/I_m)² vs. t/t_m (Fig. 2B) show a crucial change in the mechanism of nucleation and growth. The nucleation and growth of Pb nuclei at −0.49 V became instantaneous in PTPS (i), while it was predominantly progressive in PDPS (ii). Moreover, the domination of the instantaneous type of nucleation and growth decreased as t₂ increased (iii). Similar behaviour was also observed when E₁ was changed to the OCP value (−0.4 V; iv).

Fig. 2 (A) Chronoamperograms recorded at −0.49 V for 70 s for (i) PTPS (preceded pulses: E₁ = 0.4 V, t₁ = 60 s, E₂ = −0.7 V, t₂ = 0.1 ms), (ii) PDPS (preceded pulse: E₁ = 0.4 V, t₁ = 60 s), (iii) PTPS (preceded pulses: E₁ = 0.4 V, t₁ = 60 s, E₂ = −0.7 V, t₂ = 1 ms), and (iv) PTPS (preceded pulses: E₁ = −0.4 V, t₁ = 60 s, E₂ = −0.7 V, t₂ = 0.1 ms). (B) The dimensionless plot of (I/I_m)² vs. t/t_m along with the theoretical curves for instantaneous and progressive nucleation according to the SH model.

The final surface topographies of the PbNPs/GC at the end of 70 s of current transients (i–iv) are shown in Fig. 3. The instantaneous nucleation and growth of PTPS resulted in discrete and monodisperse (average size 8 ± 2 nm) capped hemispherical PbNPs (Fig. 3A and B). However, the progressive nucleation and growth and prolonged induction time resulted in small, overlapped and polydisperse (average size 5 ± 5 nm) PbNPs (Fig. 3C and D). Moreover, in PTPS, the dispersity in the size of the PbNPs increased on increasing the duration of the second pulse (Fig. 3E and F). It is interesting to note that both the size and dispersity of the PbNPs increased when the PTPS experiment was started from the OCP (Fig. 3G and H).

Fig. 3 The surface topographies (A, C, E and G) and particle size distributions (B, D, F and H) of PbNPs after (A and B) PTPS (E₁ = 0.4 V, t₁ = 60 s, E₂ = −0.7 V, t₂ = 0.1 ms, E₃ = −0.49 V, t₃ = 70 s), (C and D) PDPS (E₁ = 0.4 V, t₁ = 60 s, E₂ = −0.49 V, t₂ = 70 s), (E and F) PTPS (E₁ = 0.4 V, t₁ = 60 s, E₂ = −0.7 V, t₂ = 1 ms, E₃ = −0.49 V, t₃ = 70 s), and (G and H) PTPS (E₁ = −0.4 V, t₁ = 60 s, E₂ = −0.7 V, t₂ = 0.1 ms, E₃ = −0.49 V, t₃ = 70 s).

The differences between PTPS and PDPS in the mechanism of nucleation and growth as well as in the dispersion in PbNP particle size arose due to the presence of a very short nucleation pulse of high cathodic overpotential in between the start and growth pulses. In the case of PDPS, when the potential was switched from 0.4 V to −0.49 V, the GC surface sites were slowly activated during the prolonged ∼22 s induction period. The nucleation and growth of PbNPs then passed through a slow, progressive type of nucleation that depended on the activation–deactivation dynamics of the surface sites. In PTPS, an enormous number of surface sites were activated in very short time when the potential was switched from 0.4 V to −0.7 V. However, due to the very short imposition time (0.1 ms) of this pulse, a large number of active sites remained unoccupied, and the electrode–electrolyte interface remained in a non-steady state; thus, when the potential was switched back to −0.49 V, instantaneous nucleation occurred at these unoccupied active sites (the dissolution of the subcritical nuclei upon the potential transition replenishes the Pb(II) ions in the double layer). This mechanism is believed to constrain the size dispersion of the PbNPs. Again, the majority of the active sites became occupied upon increasing the duration of the second pulse, and the electrode–electrolyte interface attained a steady state; thus, its influence on nucleation and growth became inferior at the third pulse. This explains why the particle size dispersion exhibited a relative increase when the duration of the second pulse increased. The start potential is crucial to restrict irregular nucleation and growth of the pre-adsorbed Pb(II) on the GC surface. At the OCP, a sufficient amount of Pb(II) ions was adsorbed at the oxygenated functionalities on the GC surface, and irregular nucleation and growth throughout the three pulses resulted in the large particle size distribution and overlapped particles.

Therefore, better monodispersity in the particle size could be achieved by PTPS because of the controlled nucleation and growth at the final (the third) pulse step. The history of the GC surface just before the final pulse was created by the controlled activation and nucleation during the first and second pulses. However, Fig. 3 shows some PbNP aggregates. It is important to note that the active sites are randomly distributed on the electrode in the absence of any physical or chemical templates. Therefore, it is still challenging to avoid particle coalescence using potentiostatic pulse strategies on template-free substrates.⁵ However, the discrete particles could be formed by the multiple galvanostatic pulse strategy on a template-free electrode.⁵

Conclusions

For the first time, the results of the present study reveal the cause of the improvement in particle size and size dispersion by PTPS compared to PDPS. They also highlight the potential of PTPS for the template-free electrosynthesis of metal nanoparticles. The potential of the start (first) pulse is important to restrict the irregular nucleation and growth of the pre-adsorbed ions on the electrode surface, and OCP is not ideal for the start potential. The potential and duration of the seed (second) pulse are the most important features of PTPS as they modulate the number of unoccupied active sites available for the following controlled growth pulse. A better monodispersity in the particle size could be achieved by the controlled nucleation and growth in the final (third) pulse with the pre-history effects of the first and second pulses on the GC surface. It can be noted that both thermodynamic and surface kinetic parameters (of both the substrate and the deposited metals) play a key role in the electrochemical nucleation and growth of metal nanoparticles on the substrate. Therefore, the present work is expected to motivate detailed investigations of the thermodynamics and surface kinetics of PTPS for the electrosynthesis of metal nanoparticles. After a detailed investigation of the metal(deposited)–substrate pair, PTPS may also be applied to metals other than Pb. Furthermore, critically designed PTPS can electrochemically produce discrete and monodisperse capped hemispherical metal nanoparticles on a conductive substrate with a low surface energy. This study is useful for the design of systems with controlled size for fundamental studies and practical purposes in the fields of material science and condensed-matter physics.

Notes and references

1. W.-H. Li, C. C. Yang, F. C. Tsao and K. S. Lee, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 68, 184507.
2. E. Bernardi, A. Lascialfari, A. Rigamonti, L. Romano, V. Lannotti, G. Ausanio and C. Luponio, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 74, 134509.
3. S. K. Guin, H. S. Sharma and S. K. Aggarwal, Electrochim. Acta, 2010, 55, 1245.
4. G. D. Adzic, D. M. Drazic and A. R. Despic, J. Electroanal. Chem., 1988, 239, 107.
5. S. K. Guin, J. S. Pillai, A. S. Ambolikar, A. Saha and S. K. Aggarwal, RSC Adv., 2013, 3, 17977.
6. F. Palmisano, E. Desimoni, L. Sabbatini and G. Torsi, J. Appl. Electrochem., 1979, 9, 517.
7. I. Danaee, J. Ind. Eng. Chem., 2013, 19, 1008.
8. X. Huang, Y. Chen, J. Zhou, Z. Zhang and J. Zhang, J. Electroanal. Chem., 2013, 709, 83.
9. Y. Zhao, F.-X. Deng, L.-F. Hua, Y.-Q. Liua and G.-B. Pana, Electrochim. Acta, 2014, 130, 537.
10. S. K. Guin, R. Phatak, J. S. Pillai, A. Sarkar and S. K. Aggarwal, RSC Adv., 2014 10.1039/c4ra12877j.
11. A. Milchev and L. Heerman, Electrochim. Acta, 2003, 48, 2903.
12. L. Komsiyska and G. Staikov, Electrochim. Acta, 2008, 54, 168.
13. R.-W. Tsai, Y.-T. Hsieh, P.-Y. Chen and I.-W. Sun, Electrochim. Acta, 2014, 137, 49.
14. J. Xu, G. Yua, B. Hub, J. Zhang, Q. Dong and X. Y. Zhang, Powder Technol., 2014, 264, 561.
15. A. G. Oshchepkov, A. N. Simonov, P. A. Simonov, A. N. Shmakov, N. A. Rudina, A. V. Ishchenko, O. V. Cherstiouk and V. N. Parmon, J. Electroanal. Chem., 2014, 729, 34.

---

Footnote

† Electronic supplementary information (ESI) available: (Fig. S1) The cyclic voltammograms of 5 mM K₃[Fe(CN)₆]/K₄[Fe(CN)₆] in 0.1 M KCl at GC electrode polished in between the experiments. See DOI: 10.1039/c4ra07787c

---