1 Introduction

Owing to 4f electrons, rare earth elements have specific chemical, electronic and optical characteristics, which make them suitable to be used in high-performance luminescent devices and catalyst formulations (such as petrochemical products, refining and treatment of exhaust gases from the mobile sources) [1]. They were also used in high-efficiency phosphors and magnetic to dielectric formulations in the production of ceramic capacitors and superconductors (specially, dysprosium oxide with the cubic structure) [2]. Furthermore, dysprosium has some specific uses which make it as an important rare earth element in various industries. Dysprosium oxide crystallizes as a cubic structure below 1,870 °C. At higher temperatures, it has monoclinic or hexagonal structures. It is highly insoluble and thermally stable. It can be used as either magnetic resonance or optical imaging agents. It has the highest magnetic moments in lanthanide ion, so nanosize dysprosium is a good candidate in T2 magnetic resonance (MR) imaging [3]. It was proposed to use dysprosium oxide admixed and pelletized with Zr2O3 in the advanced heavy-water reactor (AHWR) to maintain the negative void coefficient in the reactor and also to slow down the fast neutrons [4]. In the above-mentioned application and many others, particles with nanosize enhanced the physical and chemical properties compared with bulk materials. This is because of locating high proportion of atoms at their surface. Such nanosize materials can be synthesized by various methods such as precipitation, hydrothermal, spray method and reverse micellar system [1, 5–9].

Precipitation, thermal decomposition and sol–gel are the most studied methods. An example for the precipitation method is homogenous precipitation by urea. Particle growth in the process was described by the short nucleation time model of LaMer. In this model, the formation of nuclei is at super-saturation point, and then, nuclei will grow to form larger particles by the addition of solute species. But later, aggregation mechanism was used to describe the process of homogeneous precipitation by urea [10–12]. Several parameters controlling the size and size distribution in this method should be adjusted carefully to have the desirable size and distribution (such as aging time, concentration, temperature, etc.). These parameters will have great effect on the reproducibility of the method. Consequently, the method may practically go out of control.

Thermal decomposition of various inorganic dysprosium precursors in air or nitrogen atmosphere at very high temperatures is another way to produce Dy2O3 nanoprticles. Dy2(C2O4)3·5H2O, Dy(CH3COO)3·5H2O, Dy(NO3)3 and Dy2(OH)2(CO3)2·2H2O are some precursors used to prepare Dy2O3. But high temperatures used in this process lead to high degree of agglomeration which results in large particles [13].

In sol–gel method, an organic compound of dysprosium such as dysprosium n-butoxide (DB) was dissolved in organic solvent and the sol was formed by hydrolysis of the DB precursor. The gel was obtained by aging at a proper temperature, and then it was calcined to remove the organic components. However, the method would produce narrow distribution of dysprosium oxide nanoparticles, but the precursors were expensive and also the process was time consuming [13, 14]. Electrodeposition, a powerful and interesting process, can be applied in numerous fields [15]. The cathodic electrodeposition of Dy2O3 by electro-generation of base was rarely investigated and so it is an interesting subject to be studied. Here, the preparation of Dy2O3 nanoparticles via cathodic electrodeposition method was reported.

2 Experimental

All chemicals were used as received without further purification. Dy(NO3)3 was purchased from Merck company. Steel plates (316 L) with size of 20.0 nm × 20.0 nm × 0.5 mm were used as the cathode substrate. Before deposition, the steel plates were cleaned by abrasive papers and subsequently washed with diluted hydrochloric acid and distilled water. In the electrochemical cell for deposition, the steel cathode was centered between the two parallel graphite anodes. The electrodeposition was carried out by applying constant current density (0.5, 3.0 and 6.0 mA·cm−2) for 600 s. Thermal annealing was conducted in air between the room temperature and 700 °C at heating rate of 10 °C·min−1. The X-ray diffraction (XRD) powder patterns were obtained on a STOE Model STADI MP diffractometer. The X-ray was generated by a tube with radiation of Cu Kα (0.154 nm). The operational settings for the XRD scans were: voltage of 40 kV, current of 30 mA and scan range of 3°–100°. The morphologies were measured by transmission electron microscopy (TEM, Phillips EM 2085). The specific surface area for the prepared sample was obtained through measuring N2 adsorption–desorption isotherms at 77 K with Quanta-chrome NOVA-2200e system.

3 Results and discussion

3.1 Deposition mechanism

Figure 1 shows the curves of potential versus time recorded during galvanostatc deposition of the samples in 0.005 mol·L−1 Dy(NO3)3·6H2O solution. The change of the shape of the polarization curves with the applied current densities can be related to the reduction reactions at the electrode [16]. The change of shape of polarization curves only at low current density (<1 mA·cm−2) and concentrated electrolyte can be seen clearly in the initial stage of electrodeposition. The transient curves of potential versus time were characterized by the relaxation of cathodic deposition potential toward a constant value. The value of the stabilized cathodic deposition potential decreases toward more negative values as the applied current density becomes more negative as seen in Fig. 1. A continuous variation of potential with time is typically ascribed to a single reaction, whereas sudden changes are the indication of a change in the main reaction occurring at the electrode. Under the applied current density, the potential can be sustained by the following reduction processes:

Fig. 1
figure 1

Potential versus time for deposition at 0.5, 3.0 and 6.0 mA·cm−2 from 0.005 mol·L−1 concentrations of Dy(NO3)3·6H2O solutions

Full size image
$$ 2{\text{H}}_{2} {\text{O}} + 2{\text{e}}^{ - } \to {\text{H}}_{2} + 2{\text{OH}}^{ - } $$
(1)
$$ 2{\text{H}}_{3} {\text{O}}^{ + } + 2{\text{e}}^{ - } \to {\text{H}}_{2} + 2{\text{H}}_{2} {\text{O}} $$
(2)
$$ {\text{O}}_{2} + 2{\text{H}}_{2} {\text{O}} + 4{\text{e}}^{ - } \to 4{\text{OH}}^{ - } $$
(3)
$$ {\text{O}}_{2} + 2{\text{H}}_{2} {\text{O}} + 2{\text{e}}^{ - } \to 2{\text{OH}}^{ - } + {\text{H}}_{2} {\text{O}}_{2} $$
(4)
$$ {\text{NO}}_{3}^{ - } + {\text{H}}_{2} {\text{O}} + 2{\text{e}}^{ - } \to {\text{NO}}_{2}^{ - } + 2{\text{OH}}^{ - } $$
(5)

As shown, all these half-cell reactions produce OH− and cause the increase in pH at the cathode surface (Fig. 2). The reduction in nitrates and oxygen species has a minor role in increasing pH due to the great number of solvent molecules (water) in media. Therefore, the role of the reduction in nitrates and oxygen can be disregarded. The electro-generation of the base at the interface increases the local pH and can lead to the development of dysprosium hydroxide Dy(OH)3 according to the following reaction:

Fig. 2
figure 2

Formation mechanism of Dy2O3 nanoparticles by galvano static technic

Full size image
$$ {\text{Dy}}^{3 + } + 3{\text{OH}}^{ - } \to {\text{Dy}}({\text{OH}})_{3} $$
(6)

In summary, in cathodic deposition technique, hydrolysis reactions result in the accumulation of colloidal particles near the electrode and the formation of a deposit is caused by flocculation introduced by the electrolyte. The deposit forming on electrode surface can be explained by DLVO (Derjaguin, Landau, Verwey and Overbeek) theory [17]. The DLVO theory states that the total pair interaction between colloidal particles consists of two parts: the columbic double-layer repulsion and van der Waals’ attraction. The total energy (E T) of interaction of two isolated and identically charged particles is defined as:

$$ E_{\text{T}} = E_{\text{A}} + E_{\text{R}} $$
(7)

where E A is attractive energy, and E R is the repulsive energy. The attractive energy (E A) of the London-van der Waals’ interaction between the two spherical particles can be expressed by:

$$ E_{\text{A}} = - \frac{A}{6}\left(\frac{2}{{s^{2} - 4}} + \frac{2}{{s^{2} }} + \ln \frac{{s^{2} - 4}}{{s^{2} }}\right) $$
(8)

where A is the Hamaker constant and s = 2 + H/a (where H is the shortest distance between the two spheres, and a is the particle radius). If H ≪ a, Eq. (2) can be simplified in the following formula:

$$ E_{\text{A}} = - A\frac{a}{12H} $$
(9)

The repulsive energy (E R) is:

$$ E_{\text{R}} = 2\uppi \varepsilon \varepsilon_{0} a\psi^{2} \ln (1 + {\text{e}}^{ - kH} ) $$
(10)

where ε is the dielectric constant of the solvent, ε 0 is the vacuum dielectric permittivity, ψ is the surface potential, and 1/k is the Debye length:

$$ k = \left(\frac{{{\text{e}}_{0}^{2} \sum {n_{i} z_{i}^{2} } }}{{\varepsilon \varepsilon_{0} 'k'T}}\right)^{1/2} $$
(11)

where ε 0′ is the electron charge, k′ is the Boltzmann constant, T is the absolute temperature, and n i is the concentration of ions with valence z i . The repulsion between colloidal particles is directly related to the diffuse layer charge on the particles.

When the diffuse layer repulsion is sufficiently high compared with the van der Waals’ attraction, the total energy of particle interaction exhibits the maximum value. This is the energy barrier to particle coagulation.

The DLVO theory explains the existence of critical electrolyte concentration (flocculation value) for coagulation which decreases with the valence of the electrolyte ions of a charge opposite to that of the colloidal particles. As the electrolyte concentration increases, the energy barrier disappears, and coagulation becomes possible (in other words, with the progress of electrodeposition process, the electrolyte concentration near the electrode surface will increase and coagulation will take place).

3.2 Current efficiencies

The cathodic current efficiencies were calculated on the basis of the mass of the deposit and the coulometric data. As shown in Fig. 3, with the increase in current density, the efficiencies decrease. This result is inconsistent with Faraday’s law, according to which the deposit weight is proportional to the passed charge. The possible reasons for the deviation of experimental deposit weights from the Faraday’s law are clear. When the possible factors that can influence the deposition rate are considered, it seems most reasonable to conclude that such deposit weight behavior is attributed to the low concentration of the electrolyte used for the electrodeposition [18]. Indeed, for electrolytes with low concentration, the rate of cathodic reactions depends on the rate of diffusion of the reacting species, detracting from the deposition process efficiency. Another reasonable factor for the deviation from Faraday’s law can be described on the basis of hydrogen evolution at electrode surface. The limitation of the rate of dysprosium ion diffusion toward the cathode surface with current density increasing leads to the subsequent decrease in efficiency. With the increase in current density, the rate of hydrogen evolution increases, and the hydrogen bubbles limit the dysprosium diffusion rate toward the electrode surface. Also, the hydrogen bubbling at the electrode surface would cause the deposit to splash into solution and reduce product on the electrode surface; high current density leads to high rate hydrogen evolution and more splashing.

Fig. 3
figure 3

Current efficiency versus current density

Full size image

3.3 Morphologies

The surface morphologies of the sample were investigated by TEM with different applied current densities. Figure 4 shows the TEM images of as-deposited samples at the current densities of 0.5, 3.0 and 6.0 mA·cm−2, respectively. It can be clearly seen that the surface morphology of the samples is significantly affected by the applied current density. As shown in Fig. 4, the decrease in the particle size with the increase in current density is clearly evident.

Fig. 4
figure 4

TEM images of as-deposited samples at a 0.5 mA·cm−2, b 3.0 mA·cm2, c 6.0 mA·cm2 for 10 min in water Dy(III) nitrate solution and annealed samples at 1,000 °C at d 0.5 mA·cm2, e 3.0 mA·cm2, f 6.0 mA·cm2

Full size image

The morphologies seem to result from the competition between nucleation and growth process at the electrode surface during electro-base generation. At high current density, an instantaneous nucleation mode is favorable and the rate of nucleation is very fast in comparison with the rate of crystal growth [19]. Nuclei form within a very short time, so the coverage rate of electrode surface sites by hydroxide deposition should be very fast. Particles with low growth and small size are hence preferable. In addition, at low current density such as 0.5 mA·cm−2, a particle growth process is preferred, because the rate of crystal growth is higher than that of nucleation, and the nuclei continuously form as the crystals grow. Therefore, thick particles are obtained. A similar finding was reported by Eliaz and Eliyahu [19]. The cathodic deposition of hydroxides/oxide involves an electrochemical reduction step (i.e., H2O reduction) followed by a chemical precipitation step (i.e., Dy(OH)3 formation); the influence of the chemical precipitation kinetics on the microstructures of deposits is acceptable. The heat treatment at 1,000 °C causes the decrease in particle size (Table 1), which can be attributed to the liberation of electrolyte ions, water and oxygen molecules from the as-prepared sample at high temperature.

Table 1 Particle size of samples
Full size table

3.4 XRD results

Figure 5 shows the XRD pattern of the samples obtained at different current densities. The as-prepared sample is found to be amorphous (inset in Fig. 5), and at 1,000 °C, the samples display well-defined peaks. The peaks of dysprosium oxide observed in this pattern are based on ICSD No. 061269. It has cubic structure with space group Ia-3 (lattice constant of cubic unit cell: 1.0670 nm). The curve at 0.5 mA·cm−2 has sharp peaks with high intensity, which shows that the sample has high crystallinity and the particle sizes are not very small. The Scherrer equations [20] for the estimation of particle size are used:

$$ D_{hkl} = 0.89\lambda /\beta_{hkl} \cos \theta $$
(12)
$$ \beta_{hkl} = \, \Delta 2\theta \uppi /180^\circ $$
(13)

where D hkl denotes the average crystal size, λ is the wavelength of the incident X-rays (0.15406 nm), θ is the diffraction angle, and β hkl is the full width at half maximum (FWHM). The results are summarized in Table 1. For curves at 3.0 and 6.0 mA·cm−2, the peak intensities are low, which can be concluded that the particles are smaller than those of previous sample. The particle size calculated in these samples is 27.8 nm based on Scherrer equation. The areas of the curves show the formation of amorphous phase in these samples.

Fig. 5
figure 5

XRD patterns of sample synthesized at 6.0, 3.0 and 0.5 mA·cm−2 and annealed at 1,000 °C. Insert being XRD patterns of as-prepared samples

Full size image

The crystal size estimated from XRD Scherrer’s equation is slightly smaller than that estimated from TEM. This is because XRD gives the information of crystalline region only, and the amorphous region does not contribute to this process. Moreover, TEM gives the complete images of the nanoparticles.

3.5 BET results

For the determination of the specific surface area, the pore size distribution and the pore volume, the sample with the smallest particle sizes was selected (synthesized at 6 mA). An adsorption isotherm is obtained by measuring the amount of gas adsorbed within a wide range of relative pressures at constant temperature (typically liquid N2, 77 K). Conversely, desorption isotherms are achieved by measuring the gas removed as the pressure reduces. The surface area of sample annealed at 1,000 °C and synthesized at 6 mA·cm−2 (sample with the smallest particle size) was measured using the Brunauer–Emmett–Teller (BET) method. The N2 adsorption–desorption isotherm and BJH pore size distribution by Barrett–Joyner–Halenda (BJH) method at 77 K are shown in Fig. 6. Before adsorption measurements, the sample was activated by heating to 423 K at heating rate of 0.5 K·min−1 and keeping at this temperature under the turbo molecular pump vacuum. The amount of N2 adsorbed at p/p 0 of 0.995 is 125 cm3·g−1. The isotherm has Type III form that is characterized by the heat of adsorption, less than the heat of liquification for adsorbate; adsorption proceeds as the adsorbate interaction with the adsorbed layer is greater than the interaction with the adsorbent surface, and the nitrogen hysteresis loop is Type H3 according to the International Union of Pure and Applied Chemistry (IUPAC) classification [21], which indicates the existence of slit-shaped pores between the crystallites. The BET surface area of the sample is found to be 91 m2·g−1. The pore size distribution of Dy2O3 is shown in Fig. 6b. According to IUPAC nomenclature, the size of micropores is <2 nm in diameter, that of mesopores is between 2 and 50 nm in diameter, and that of macropores are >50 nm in diameter. For our sample, a narrow peak with mean pore widths of around 6 nm emerges, indicating that the sample contains micropores.

Fig. 6
figure 6

N2 adsorption (down curve)-desorption (up curve) isotherms a and BJH pore size distribution curve b of sample synthesized at 6 mA·cm−2

Full size image

4 Conclusion

Dy2O3 powders were produced by electrochemical deposition. The growth of the nanostructures was associated with the precipitation of Dy(OH)3 induced by the increase in pH on the cathode surface caused by the reduction processes. Owing to low concentration compared with water molecules, it seems that the reduction in the dissolved oxygen and nitrates ions do not increase the interfacial pH sufficiently, which is necessary to form colloidal Dy(OH)3 particles. The size variation of Dy2O3 as a function of current density was studied. The cathodic current density varies from 0.5 to 6.0 mA·cm−2. The result shows that the morphology is markedly affected by current density and as the cathodic current density increases, the mean nanocrystal size decreases. The amount of the deposited material (current efficiency) as a function of current density was studied. The results show that with the increase in the current density, the efficiencies decrease; consequently, a significant deviation from Faraday’s law can be observed. This behavior is attributed to the low concentration of the electrolyte used for the electrodeposition and hydrogen bubbling at the electrode surface.