No abstract

The kinetics and initial stages of Bi 2 Te 3 electrodeposition on gold from an aqueous electrolyte composed of bismuth and telluryl ions and nitric acid were investigated using voltammetry and chronopotentiometry. For tellurium alone, it is found that the reduction of tellurium ions to tellurium occurs with four exchanging electrons at −0.05 V vs Ag/AgCl ͑3 M KCl͒. At −0.6 V, tellurium reduces to H 2 Te under kinetic control and the following simultaneous diffusion-controlled reaction occurs with one exchange electron per tellurium atom: HTeO 2 + + H 2 Te + H + + 2e − → 2Te + 2H 2 O. Bismuth telluride deposition starts at more positive potential than bismuth and tellurium deposit individually. It is suggested that the electrodeposition of Bi 2 Te 3 proceeds via a Stransky-Krastanov mechanism. From the Bard correction to the Sand equation, the diffusion coefficients of bismuth and telluryl ions were determined to be 1.90 ϫ 10 −5 and 1.39 ϫ 10 −6 cm 2 /s, respectively, in the investigated solutions. Due to the low diffusivity and solubility of telluryl ions, the limiting current density for Bi 2 Te 3 deposition in the investigated solution is low; it is only 2 mA/cm 2 . Above the limiting current density, a smooth deposit cannot be obtained without agitation. A smooth deposit can be achieved when the depletion of anions is eliminated.

The influence of current density and temperature on the macrotexture, the orientation and size of grains, and the corrosion resistance of tin deposits was studied. Tin coatings with two different textures, (100) and (301) fiber textures were produced by electrodeposition at 20°C by varying current density. At a lower current density of 100A/m2, (301) fibre was obtained. At the current densities of 100 and up to 400 A/m2, only (100) fibre texture was observed. An increase in current density leads to a decrease in grain size. At the same current density, the grain size of tin coatings increases with increased temperature. The influence of temperature (20, 40, 60 and 80 °C) on texture is relatively negligible. The corrosion resistance of tin coatings increases with a decrease in grain size. The corrosion resistance of tin coating with (301) fibre is higher than that of tin coating with (100) fibre texture. The results suggest that texture and microstructure play an important role in controlling corrosion rate of tin based coatings.

Cathodic potential oscillations of Sn͑II͒ reduction and hydrogen evolution in acid stannous sulfate solutions were investigated by cyclic voltammetry under galvanostatic conditions and chronopotentiometry. The potential oscillations are observed when the applied current density is above the limiting value, and they are observed only at low current scan rates and at constant current density. In the absence of gelatin, damped potential oscillations are evident. This is attributed to a dendritic growth, which increases the surface area and reduces the current density below the limiting value. Sustained potential oscillation can be achieved with gelatin, where a smooth deposit is expected due to an inhibitory effect and/or absorption of gelatin on the cathode surface. It has been found that gelatin has the effect of extending the lifespan of the potential oscillations. In order to have sustained potential oscillations, the deposits on the cathode should be smooth so that the effective current density remains higher than the limiting value.Oscillatory behaviors in electrochemical reacting systems can be observed during the anodic dissolution and cathodic deposition of metals, and during eletrocatalytic reactions. 1 Almost all electrochemical reactions are observed to have oscillating behavior in a certain range of the external parameters ͑for example, voltage or bulk concentrations of the reactants͒. 2,3 In potentiostatic ͑constant potential͒ systems the current might oscillate, and in galvanostatic ͑constant current͒ systems the potential might oscillate. The cathodic potential oscillations have been observed in the reduction of Zn͑IV͒, 4-9 Fe͑III͒, 10 Sn͑II͒, [11][12][13][14][15]16,17 Sn͑II͒ + Cu͑II͒,18,19 Ag͑I͒, 20 Cd͑II͒, 21 Fe͑CN͒ 6 ͑III͒, 22 and H 2 O 2 . 23,24 To explain the potential oscillations of zinc deposition that were observed when the applied current density was above the limiting current density, St-Pierre and Piron 4,5 proposed a dual depletion/replenishment mechanism. This proposal involves: ͑i͒ depletion of Zn͑IV͒ ions at the cathode that causes the decrease of the cathodic potential to a very negative value which in turn allows appreciable hydrogen evolution to occur; and ͑ii͒ the replenishment of Zn ions in the cathode vicinity by the bubbling of hydrogen, marking the beginning of a new cycle of potential oscillation.Li et al. 22 proposed a similar mechanism for the potential oscillations in Fe͑CN͒ 6 3− reduction: a combination of electrochemical reactions and diffusion-convection mass transfer accounts for the potential oscillations that appear under conditions of transportlimited current. Fukami et al. 6 have related the potential oscillation during zinc electrodeposition to the dendritic and autocatalytic Zn growth, and to the Mullins-Sekerka instability. The spherical diffusion layer for Zn͑IV͒ ions at the growth front makes zinc deposit faster at places with high surface curvature than those with small surface curvature. The potential oscillation also accompanies the oscillation of the diffu...

This paper describes our submission to ICASSP 2022 Multi-channel Multi-party Meeting Transcription (M2MeT) Challenge. For Track 1, we propose several approaches to empower the clustering-based speaker diarization system to handle overlapped speech. Front-end dereverberation and the direction-of-arrival (DOA) estimation are used to improve the accuracy of speaker diarization. Multi-channel combination and overlap detection are applied to reduce the missed speaker error. A modified DOVER-Lap is also proposed to fuse the results of different systems. We achieve the final DER of 5.79% on the Eval set and 7.23% on the Test set. For Track 2, we develop our system using the Conformer model in a joint CTC-attention architecture. Serialized output training is adopted to multi-speaker overlapped speech recognition. We propose a neural front-end module to model multi-channel audio and train the model end-to-end. Various data augmentation methods are utilized to mitigate over-fitting in the multi-channel multi-speaker E2E system. Transformer language model fusion is developed to achieve better performance. The final CER is 19.2% on the Eval set and 20.8% on the Test set.