Electrochemical Behavior of Ti in Molten Fluoride-Oxide System

INTRODUCTION

An innovative process of Ti metal production is earnestly required to supply Ti metal sufficiently at a moderate price. Some new production processes have been proposed. Our group has been studying the direct electrolytic production of liquid Ti metal by using a direct-current electro-slag remelting (DC-ESR) unit in a CaF₂-CaO-TiO₂ melt, and showed that Ti metal in liquid were obtained. In the previous researches, the electro-deposition of Ti by this process was strongly influenced by the electrolytic bath composition.

In this study, the cathodic reaction of Ti in a similar fluoride-oxide melt containing TiO₂ was studied to clarify the influence of the bath composition. Cyclic voltammetry was applied in the bath of various compositions, and potentio-static electrolysis was performed to analyze the electrodeposit at the specific potential.

EXPERIMENTAL

The apparatus is schematically illustrated in Fig.1. A eutectic mixture of CaF₂-MgF₂ (55:45 in mole) was used as an electrolytic bath. CaO and TiO₂ were added to control the molar ratio of CaO to TiO₂, R_CaO/TiO2, between 0~2. The mixture was fused under a pure Ar atmosphere, and kept at 1273~1373K. Some kinds of crucible material, such as graphite, Mo and BN, were used because there was a possibility that the crucible reacted with the bath.

The cathodic reaction was investigated by cyclic voltammetry (CV). A Mo wire (f1mm) was used as a working electrode, and immersed in the bath at a depth of 5~10mm. A graphite rod (f5mm) was used as a counter electrode. Since no stable reference electrode was reported in this bath, another Mo wire was used as a quasi-reference electrode. The potential of the quasi-reference electrode was calibrated by the zero-current potential determined by Mg deposition and dissolution in the CV. All the potentials are represented versus this potential. Since the current density and the electrode potential should be contained inaccuracy, the CV shape is discussed qualitatively.

Based on the results of the CV measurement, potentio-static electrolysis was performed using a Mo plate electrode, and the electrodeposit was analyzed by XRD.

RESULTS and DISCUSSION

Fig. 2 shows the change in the CV with R_CaO/TiO2. Reduction current increased below 0.4V in all the baths. The cathodic current in this potential region increased with TiO₂ addition, so it was thought to correspond to the reduction of Ti ion. In the bath of R_CaO/TiO2 = 1.0, a cathodic current hump was seen around 0.65V.  A small hump still appeared around 0.75V in the bath of R_CaO/TiO2 = 1.5, while it disappeared in the bath of R_CaO/TiO2 = 2.0. Although residual current seems un-negligible in this potential region, the change in the CV shape suggests that the reduction sequence of a Ti ion strongly depends on R_CaO/TiO2.

No remarkable change in the CV shape with the crucible materials was observed at 1373K. The reaction of the crucible materials with TiO₂ in the bath seems negligible at this temperature.

The Mo plate electrode was covered with grey deposit after potentio-static electrolysis at 0.4V in the bath of R_CaO/TiO = 1.5. Fig.3 shows a typical XRD pattern of the deposit after rinse with water. Ti metal was detected in the deposit though the ingredients of the bath were contained. This result shows that the reduction to Ti metal occurs below 0.5V.

The results of this study clearly indicate that Ti metal can be electrodeposited in the CaF₂-MgF₂-CaO-TiO₂ bath, and that the reduction mechanism is influenced by the bath composition, that is the molar ratio of CaO to TiO₂.