Printer Friendly

Thin film semiconductors have proved to be very versatile.

Development of preparation procedures and techniques to control properties

Thin film semiconductors are very versatile materials having a variety of applications. Several, such as amorphous silicon, cadmium telluride and copper indium diselenide, are prime contenders for use in thin film solar cells and modules. Amorphous silicon also is an effective photoreceptor that has found application in photocopiers and laser printers.

Certain metal chalcogenides are sufficiently good photoconductors to be used in video cameras, and others such as cadmium selenide are being studied for use in thin film transistors. Although the list is not inexhaustible, there are many other uses including thin panel displays and sensors. Due to all these applications, many researchers are studying ways to improve the properties of these materials and to develop less expensive preparation processes for them. They also are looking for new materials.

Research in this laboratory involves the preparation and study of metal chalcogenide semiconductor materials that can be used for a variety of applications, ranging from solar energy conversion to thin film transistors.

Tailoring a semiconductor to fit a particular application requires an understanding of the effect of the nature and density of the dopants on various properties. Some key aspects of our work have been to:

* develop inexpensive procedures to form thin films of these semiconductors;

* develop processes to incorporate dopants into these films either during or after film preparation;

* use a variety of techniques to characterize the physical and electronic properties of the undoped and doped films.

Electrodeposition: Until recently, electrochemical procedures have received the most attention in our laboratory. Several criteria have been used to choose the chalcogen precursor for the electrochemical process:

a) the chalcogen compound should be prepared from inexpensive chemicals in one synthetic step;

b) it should contain only one chalcogen atom;

c) the valency of the chalcogen should be as close to zero as possible.

The last two criteria help to simplify the electrochemical process and avoid problems with poor stoichiometry. Up to now, the selenosulfate anion and phosphine chalcogenides have been employed for the aqueous |1~ and non-aqueous |2,3~ processes, respectively. It has been possible to prepare a variety of semiconductors including CdS |3~, CdSe |1,4~, CdTe |2,5~, |Cd.sub.x~|Hg.sub.(1-x)~Se |6~ and |Cd.sub.x~|Hg.sub.(1-x)~Te. |7~ Inclusion of Hg provides a way to lower the band gap and, therefore, extend the absorption spectrum to longer wavelength. |6,7~

The effect of electrochemical conditions on semiconductor composition has been studied using various analytical techniques such as polarography, atomic absorption spectroscopy, Auger profiling, Rutherford back scattering (RBS) spectrometry and proton induced x-ray emission (PIXE) spectrometry. Consequently, it has been possible to determine conditions needed to prepare uniform, smooth and adhesive films over large areas.

Characterization: A number of techniques have been developed for the systematic study of the effect of the nature and density of dopants on properties such as minority carrier diffusion length, conductivity, |4,5~ photoconductivity, |4,8~ majority carrier density and mobility. |9,10~ The last two parameters are determined by measurement of the Hall voltage. Because of the very high resistivities of the undoped films, it was necessary to build a very high impedance differential electrometer amplifier (DEA) to measure the Hall voltage.

For these measurements the films were removed from the conducting electrodes and shaped in the form of a cross; Fig. 2 illustrates the arrangement of the sample in the magnetic field, which could be varied from 0 to 0.8 Tesla.

Earlier work showed that the data could be varied over a very wide range in a systematic fashion and correlated with the densities of the incorporated dopants. Thus for CdSe |10~, carrier concentrations ranged from |10.sup.19~ |cm.sup.-3~ to 109 |cm.sup.-3~, and the electrical resistivities were measured in the range of 1 |ohm~-cm to |is greater than~ 1 x |10.sup.8~ |ohms~-cm. Hall effect mobilities were in the range of 1 to 50 |cm.sup.2~/Vs.

As doping levels increased, both resistivity and mobility decreased. The barrier to carrier transport was obtained from the slope of the temperature dependence of the mobility. 3. In addition, the activation energy for the resistivity depended strongly on the doping density as can be seen from the slopes of the plots in Fig. 4. These changes are consistent with a model in which a barrier to carrier transport results from the charging of grain boundary states. |11,12~.

The recent acquisition of a spreading resistance apparatus now permits the study of the doping profiles of our materials. This technique involves measuring resistance as a function of the depth of the film by stepping across its bevelled surface.

Another recent addition to our list of characterization techniques is thermally stimulated current measurements, which provide information about the energy, density and capture cross section of subband gap states. |13~. We have established that there are a number of traps in polycrystalline CdSe, and now we are investigating the effect of dopants on the energy and density of these traps.

Solid/vapor reactions: During the past year, our electrodeposition approach for making films has been supplemented by a new procedure that involves two steps:

* spray pyrolysis to form a metal oxide;

* solid/vapor reaction between the metal oxide and elementary chalcogen to form the metal chalcogenide semiconductor.

This procedure is very general; sulfides, selenides and tellurides have been prepared. Furthermore, it is not restricted to the formation of binary compounds; ternary and quaternary compound semiconductors have been formed. To date, the details of the preparation of CdSe |14~, |In.sub.2~|Se.sub.3~|15~ and CuIn|Se.sub.2~|16~ have been reported.

Auger profiling data are consistent with a process in which selenium vapor reacts at the surface of the metal oxide to form the metal selenide, which subsequently diffuses into the film. The mechanism is expected to be fairly complicated because it involves reaction and diffusion in the solid. Whatever the details, it seems that the selenium must remove the oxygen by forming an oxide of selenium as well as the metal selenide; a possible overall reaction for CdSe formation might be

3|Se.sub.2~(g) + 4CdO(s) |arrow right~ 4CdSe(s) + 2Se|O.sub.2~.

A very exciting aspect of this new approach is that it permits the incorporation of dopants, which are uniformly dispersed throughout the semiconductor thin film with concentrations that are accurately known. Consequently, it offers the possibility of fine tuning the electrical properties of the films because a variety of dopants can be introduced at the oxide formation stage. The undoped film can be prepared with a minimum of impurities using purified nitrate salts and highly purified water for the spray pyrolysis step, and highly purified chalcogen can be used for the subsequent conversion step. In the near future, this information will be used to correlate electrical properties of CdSe, CdTe and CuIn|Se.sub.2~ with the performance of devices such as solid state photovoltaic cells and photoreceptors.


1. Szabo, J., and Cocivera, M.; Can. J. Chem. 66, 1065 (1988).

2. von Windheim, J., Darkowski, A., and Cocivera, M.; Can. J. Phys. 65, 1053 (1987).

3. Preusser, S.; and Cocivera, M.; Solar Energy Mater. 20, 1 (1990).

4. Wynands, H.; and Cocivera, M.; Chem. Mater. 3, 143 (1991).

5. von Windheim, J.; Renaud, I.; and Cocivera, M.; J. Appl. Phys. 67, 4167 (1990).

6. Weng, S.; and Cocivera, M.; Chem. Mater. 4, 615 (1992).

7. Colyer, C.; and Cocivera, M.; J. Electrochem. Soc. 139, 406 (1992).

8. von Windheim, J.; Wynands, H.; and Cocivera, M.; J. Electrochem. Soc. 138, 3435 (1991).

9. (a) von Windheim, J.; and Cocivera, M.; J. Electrochem. Soc. 138, 250 (1991); (b) von Windheim, J.; and Cocivera, M.; J. Phys. D, 23, 581 (1990); (c) von Windheim, J.; and Cocivera, M.; J. Phys. Chem. Solids, 53, 31 (1992).

10. Wynands, H.; and Cocivera, M.; J. Electrochem. Soc. 139, 2052 (1992).

11. Seto, J.; J. Appl. Phys. 46, 5247 (1975).

12. Baccarani, G.; Ricco, B.; and Spadini, G.; J. Appl. Phys. 49, 5565 (1978).

13. Wynands, H.; and Cocivera, M.; J. Appl. Phys. 1992, submitted.

14. Weng, S.; and Cocivera, M.; J. Electrochem. Soc. 1992, in press

15. Weng, S.; and Cocivera, M.; Chem. Mater, 1992, submitted.

16. Weng, S.; and Cocivera, M.; J. Appl. Phys, 1992, submitted.
COPYRIGHT 1992 Chemical Institute of Canada
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Cocivera, Michael
Publication:Canadian Chemical News
Date:Nov 1, 1992
Previous Article:Molecular materials for nonlinear optics.
Next Article:The opportunities seem endless for advanced materials R&D.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters