Langmuir 2007, 23, 1523-1529
1523
Electrolytic Processes in Various Degrees of Dispersion Andrey Korshunov,† Michael Heyrovsky´,*,‡ Snejana Bakardjieva,§ and Libor Brabec‡ Department of General and Inorganic Chemistry, Tomsk Polytechnic UniVersity, Lenin AVenue 30, 634050 Tomsk, Russian Federation, J. HeyroVsky´ Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, DolejsˇkoVa 3, 182 23 Prague 8, Czech Republic, and Institute of Inorganic Chemistry, Academy of Sciences of the Czech Republic, 250 68 Rˇ ezˇ, Czech Republic ReceiVed August 29, 2006. In Final Form: October 25, 2006 While following voltammetric behavior of ultrafine metallic powders, we realized that the results obtained with six metals (Al, Fe, Ni, Cu, Mo, and W) were providing us with material for treating the connection between electroactivity and the state of dispersion of matter. The electroactive species were metallic oxides formed spontaneously on surfaces of the metallic powder particles, and we could follow their electrochemical reactivity in the states of coarse and fine suspensions, colloids, and true solutions. Each state of dispersion can be characterized by a distinctive form of electroactivity, which we illustrated by experimental results with all six metals.
Introduction Electrochemical methods have already been used in powder metallurgy several times to provide information about various properties of nanoparticles of metals (e.g., see refs 1-5). In Tomsk Polytechnic University metallic powders have been prepared for the past 3 decades by “electric explosion of wires (EEW)6 ”, and after a set of various powder samples of six different metals had been offered to us from there, we decided to test them by voltammetry with mercury electrodes.7 In our study of voltammetry of the fine metallic powders we found that, in general, the behavior of an electroactive species dispersed in a given medium depends on its degree of dispersion. The electroactive species in our case was metallic oxides, reducible at a mercury cathodesthey had formed spontaneously on the surface of individual grains of the metallic powder samples during the course of their storage in access of air. The degree of oxidation of the metallic surface depends on the time of storagesthe initial few islets of oxides on surfaces of freshly prepared powder particles extend with time, for each metal at a different rate, into a layer, and further, contingently, into several oxide layers. For a rigorous quantitative study of the topic of the present article our metallic powders are not very well suited, as each sample consists of a heterogeneous mixture of particles of dimensions between tens of nanometers and tens of micrometers, depending on conditions of its preparation. At any rate, the electrochemical testing provides a useful contribution to qualitative characterization of any given powder dispersion. In the present report we give a systematic account of selected experiments with aqueous dispersions of metallic powders and discuss the results in detail. * To whom correspondence should be addressed. † Tomsk Polytechnic University. ‡ J. Heyrovsky ´ Institute of Physical Chemistry, Academy of Sciences of the Czech Republic. § Institute of Inorganic Chemistry, Academy of Sciences of the Czech Republic. (1) Ga´l, M. F.; Ga´lova´, M.; Turonˇova´, A. Collect. Czech Chem. Commun. 2000, 65, 1515. (2) Taillefert, M.; Bono, A. B.; Luther, G. W., III. EnViron. Sci. Technol. 2000, 34, 2169. (3) Friedrich, K. A.; Henglein, F.; Stimming, U.; Unkauf, W. Electrochim. Acta 2000, 45, 3283. (4) McKenzie, K. J.; Marken, F. Pure Appl. Chem. 2001, 73, 1885. (5) Cepria´, G.; Bolea, E.; Laborda, F.; Castillo, J. R. Anal. Lett. 2003, 36, 921. (6) Kotov, Yu. A. J. Nanoparticle Res 2003, 5, 539. (7) Korshunov, A. V.; Heyrovsky´, M. Electroanalysis 2006, 18, 423.
The degrees of dispersion in liquid media of electroactive species, differing in electrolytic behavior, are basically four: (1) the molecular dispersion, or the true solution, where the dimensions of individual molecules or ions in general do not exceed 1 nm, (2) the colloidal dispersion, consisting of mostly lyophilic particles of dimensions between about 1 nm and 1 µm, (3) the fine suspension with particles mostly lyophobic, of dimensions between approximately 100 nm and 1 µm, and (4) the coarse suspension of lyophobic particles larger than 1 µm. The latter particles are relatively heavy, undergoing sedimentation; to produce electrolytic current, they have to be brought into contact with the surface of the working electrode by stirring the suspension.8-10 Such accidental brief contacts generate irregular flares of electrolytic current, typical for electrode reactions of particulate mechanism. The fine suspensions, on the other hand, have certain stabilitysthe particles are kept aloft in the liquid medium by Brownian motion, which brings about their fortuitous impingements at the electrode surface and thereby again irregular flares of voltammetric current; the electrode reaction mechanism is still particulate, this time without the need of stirring. Due to the statistical nature of the particulate mechanism, the reproducibility of results in voltammetry of suspensions is poor, especially of the values of the current. In contrast, the colloidal dispersions behave in certain aspects as true solutionssthe transport of particles into the medium occurs mostly by diffusion, and in this the homodispersed colloids11 behave like big molecules. However, the particulate as well as electronic structure of colloids, containing tens or hundreds of individual molecules in one dispersed particle, are the cause of the mechanism of electron-transfer reactions basically different from that observed with species in true solutions. Polydispersed colloids, in addition, bring in the factor of statistical distribution of sizes and thereby of different rates of motions of individual particles in transport as well as in chemical reactions.12,13 Hence, when electrolytic processes are being interpreted, the state of dispersion of the electroactive species has to be taken into consideration. (8) Micka, K. Collect. Czech Chem. Commun. 1956, 21, 647. (9) Micka, K. In AdVances in Polarography; Longmuir, I. S., Ed.; Pergamon Press: Oxford, 1960; p 1182. (10) Songina, O. A.; Dausheva, M. R. Elektrokhimiya 1965, 1, 1464. (11) Matijevic’, E. Langmuir 1994, 10, 8. (12) Heyrovsk’y, M.; Jirkovsky´, J. Langmuir 1995, 11, 4288. (13) Kolárˇ, M.; Meˇsˇt’a´nkova´, H.; Jirkovsky´, J.; Heyrovsky´, M.;Subrt, J. Langmuir 2006, 22, 598.
10.1021/la062536p CCC: $37.00 © 2007 American Chemical Society Published on Web 12/02/2006
1524 Langmuir, Vol. 23, No. 3, 2007
Figure 1. SEM image of a Fe sample: polydispersed powder containing nanoparticles and aggregates.
In the present article we demonstrate how the degree of dispersion in aqueous solutions affects the behavior of ultrafine powders of six different metals in voltammetry with a hanging mercury drop electrode. Materials and Methods The metallic powders were prepared in Tomsk Polytechnic University by the method of “electric explosion of wires”.6 In a special chamber filled with inert atmosphere an impulse of strong current density (up to 1010 A/m2) heats up a selected metallic wire of a particular length (in cm) and thickness (of the order of tens of µm), at the fast rate of up to 107 K/s, to temperatures as high as 104 K. The wire instantaneously “explodes”( i.e., it melts and evaporates), and the thus momentarily produced gaseous and liquid phases of the metal cool rapidly down, condense, and solidify. After the “explosion” the powder is left for some days to rest in the inert atmosphere, into which the air is gradually let in. Inert gas is often used, argon, at a pressure of 200 kP. The actual size, shape, and state of individual particles and their scatter depend on experimental “explosion conditions”: on the properties of the metal, on the thickness and length of the metallic wire, on the voltage of the initial electric impulse (up to 40 kV), and on the type and pressure of the inert atmosphere in the chamber. The resulting metallic powder particles produced by the electric explosion have special features compared to powders prepared by other methods. The powders are in the ideal case initially of regular spherical shape with diameters up to tens of micrometers; the real, more or less aged powder samples are individual heterogeneous mixtures of basic fine grits with their agglomerates and some spheres (typical morphology is shown in Figure 1). Besides the common morphology, particles of all the metallic powders obtained by such technique have similar structure. Each particle consists of a metallic core and of an oxide layer on its surface. In Figure 2 is shown a TEM micrograph of Mo particles. On TEM images the metallic cores and the oxide layers can be clearly distinguished from each other because of their significant difference in density and hence in permeability for the electron beam. In the course of storage the superficial oxide layer increases in its thickness. The dynamics of that process depends on several conditions, among them on the nature of the metal, on its dispersion, on the nature and structure of the oxide film (continuous or not continuous, dense or friable, self-limiting or not, etc.), and on conditions of storage. It was found out that the number and sizes of agglomerates increase in time, obviously due to increase of the oxide layer thickness. TEM micrograph (Figure 2) shows that oxide layers, when grown, unite the neighboring particles in agglomerates. There are some more common regularities for the “electroexploded” metals: first, the superficial oxide layers are mostly in the amorphous state, especially for freshly prepared powders, and hence detection
KorshunoV et al.
Figure 2. TEM image of the Mo particles occluded by a common oxide layer. of their phase composition is a relatively difficult procedure. Second, because of the conditions of the formation process being far from equilibrium, the obtained powders are in a metastable state of matter. The crystal lattice of such metals has usually many defects of different kinds, which we have observed with our samples in studying their phase composition by X-ray diffraction: the peaks in the XRD patterns have increased widths, the relationship between integral intensities for planes (111)/(200) (for cubic lattice) often does not coincide with the standard ones and is less than unity, and the peaks corresponding to the metallic phase are split.14 Detailed characteristics of our samples were explored using several physical methods. Phase composition of the powders was studied by X-ray analysis (XRD) using a DRON-3 diffractometer operating in the reflection mode with Cu KR radiation (35 kV, 25 mA) with the step 0.07°/2θ and 2.4 s/step. Diffraction patterns of powders were compared with reference to the PDF-2 database. Morphology and elementary composition of the samples were examined by a scanning electron microscope (SEM) JSM-5500 (voltage 20 kV) with quantitative EDAX analysis using an EDX detector attached to the SEM. High-resolution transmission electron microscope (HRTEM) JEOL JEM-3010 (LaB6 cathode, accelerating voltage 300 kV) equipped with an EDS detector served to study the structure and composition of the powder particle surfaces in microanalytical regime. The particle sizes were assessed on Nanosizer ZS (Malvern Instruments, UK; 0.6 nm-6 µm) at 25 °C in ethylene glycol medium. Statistical analysis was performed using nonparametric tests. Of the metallic powders prepared in the above way we tested the voltammetric behavior of Al, Fe, Ni, Cu, Mo, and W.
Experimental Section The aqueous dispersions of metallic powders were prepared by introducing weighed amounts of powder samples (of the order of tens of milligrams) into 10 mL of solutions of supporting electrolytes in bidistilled water. Electrolytes were used, NaClO4, NaOH, and acetate, phosphate, and borate buffers, basically in 0.1 M concentration; all the chemicals used were of analytical purity. Voltammograms of the dispersions were recorded shortly after their preparation; fine suspensions and colloids possessed good stability, so there was no need to add stabilizers. Although it was verified that the electroreduction of air oxygen initially present in the medium does not affect the electrochemical behavior of metallic powders, the prepared dispersions were always deaerated by pure nitrogen and electrolyzed in its atmosphere. In definite cases stirring was realized by passing a mild stream of fine nitrogen bubbles through the dispersion. The electrochemical research was carried out by recording voltammetric curves with the hanging mercury drop electrode (14) Thompson, C. V.; Carel, R. J. Mech. Phys. Solids 1996, 44, 657.
Electrolytic Processes and Degrees of Dispersion
Figure 3. Cyclic dc voltammograms of 10 mL deaerated Al powder dispersions in 0.1 M NaClO4; Ei ) -1.2 V; Erev ) -1.9 V; V ) 50 mV/s; weight of the powder: 1, 10 mg; 2, 100 mg. Curve of blank solution merges with potential axis. (HMDE) using the pen-recording polarograph PA4 and the computeroperated polarograph ETP, both made in the Czech Republic. As we studied not very precisely defined systems, we chose HMDE as the working electrode with its best fitting potential window and its high reproducibility to avoid introducing additional uncertainty of results. Of the voltammetric methods available we used cyclic dc voltammetry with linear potential scan. In the three-electrode systems, besides the drops of mercury as working electrodes, platinum strips served as auxiliary electrodes and reference electrodes were either SCE or (Ag/AgCl/KCl saturated) electrodes, respectively. The values of the potentials are given with respect to the latter (SAgClE). When repeated with the same solution, the parameters (current and potential) of the voltammetric peaks showed considerable scatter in the case of suspensions: the peak potentials varied by up to 60 mV and the peak heights up to 100%. With colloids the scatter was less: the half-wave, or peak, potentials varied by
