It would be easier changing the work function of the electrodes. You can choose from the existing list of metals based on the work function you desire.

On the other hand, if you change the energy positions of the material then you need to worry about few things.

- Oxygen vacancies can capture electrons, associated mobility changes

- Oxygen vacancies can lead to device instability

- Is the material well characterized and are the energy levels associated with the material under the conditions you wish to use are they known?

Hi GuO,

I would advise you to read the chapter on Schottky contacts in "Sze, Semiconductor Physics". This is a good start to find ways to create a Schottky contact based on the bandgap of your material and fitting work functions of the metals. Apart from the theory, the way you deposit the material onto IGZO and the cleaning of the substrate (either wet chemically or by plasma sputter etch) can have enormous effects on the Schottky behaviour since the interface will determine for a large part the properties of the contact. Also the stoichiometry of the IGZO and possible vacancies who can act as a dopant can have a large effect on the contact.

Thanks for you guys to explain. Both your advices are so useful. But, sorry, I do not describe my question clearly. I want to make a schottky contact with IGZO on FTO(conductive glass) by sputtering. To make metal(better conductivity) and semiconductor(poor conductivity) by controling concentration of oxygen vacancy. Is it make sence? And how can I increase the quality of IGZO by sputtering ? By the way, my english is poor. Thanks for reading.

Hi Guo,

Thanks for more information on your question;

With your choice of materials fixed the way to influence the schottky contact is by changing the interface between FTO and IGZO.

If you deposit IGZO on FTO you can change the work function by sputter etching the fluorine doped SnO2 and remove fluorine dopant from the interface (this will lower the dopant level and most likely create a schottky contact. But it might also make the SnO2-d substoichiometric which increases dopant level. Anyway, playing with sputter etch power and time will probably effect the contact resistance. On the IGZO side you can play with the Ar/O2 mixture during deposition and you can also make the IGZO loose oxygen (just sputtering in Ar) or loose less oxygen (with higher O2 flows).

You can also play around with the pressure and power of deposition. At low pressure and high power (high voltage) you have more energetic Ar+ atoms impinging on the target, ergo more energetic fragments of IGZO hitting the FTO. So you can get more atom intermixing on the surface. The effect on this intermixing on the schottky contact I cannot predict but trying the opposite (high pressure/low power) should give an indication if this method brings you closer to a contact you wish to obtain. 

Apart from changing the interface with the sputteretch and/or sputter deposition proces you can also try several wet chemical cleans to modify the interface and influence the dopant concentration on the surface. You can think of wet etches with buffered oxide etch (BOE7:1), Piranha or HCl/H2O2 etches to name just a few.

Wow, that a good advices, Thank you so much. I will try it. But if I sputter deposition IGZO twice to make better conductivity and poor conductivity, Does it is a schottky contact?