High-purity tellurium is a supporting material for contemporary high-tech industries. Zone refining is commonly used as a reliable final purification step to achieve higher purity levels of tellurium. However, the removal of Selenium (Se)—the most troublesome impurity in tellurium often occurring in quite high amounts—stays as the challenge during zone refining, due to its very unfavorable distribution coefficient. In addition, there is an important factor, essential to predict the zone refining efficiency (required number of passes), which is still seldom explored. Therefore, in the present work, the effect of different number of passes on the purification effect of tellurium was investigated by using 4N tellurium as raw material. Firstly, the possibility of removing vaporized Se through introducing an reductive gas flux was studied by testing initial concentrations of Se (13 ppm) and undergoing different process parameters. The experimental results showed that when the number of passes was 3, the impurity removal rate was low and the purification effect was poor; when the number of zone refining reached 9, the removal rate of most impurities in tellurium reached 100 pct, the removal rate of impurity Se reached 80.15 pct, and the total removal rate reached 89.49 pct, and the purity of tellurium ingot was 5N7.
Macroporous adsorption resin was used as the sorbent for solid phase extraction and determination of the trace Se content in high purity tellurium prior to hydride generation atomic fluorescence spectrometry analysis. Selenium was converted into an organic Se chelate using 3,3′-diaminobenzidine and was separated from the tellurium matrix by solid phase extraction . The resin was packed as a column for solid phase extraction. Under optimum conditions, trace Se can be quantitatively extracted and the tellurium matrix can be removed. The Se in the eluate was determined by hydride generation atomic fluorescence spectrometry . The limit of detection (3σ) of this method was 0.22 ng g − 1 and the relative standard deviation (RSD, n = 5) ranged from 2.0 to 2.5% for the three investigated tellurium samples. The proposed method was successfully applied for the determination of the trace Se content in high purity tellurium samples.