The answer to your questions depends on a number of factors: the properties of your semiconducting material, what you want to do with the FETs, what equipment you have access to etc.

Here is a link to an early semester thesis of mine showing one or two ways to do it (you might be most interested in sections 2 and 3)

https://drive.google.com/uc?export=download&id=11a2m_Yxy-wKVC5om6G5rSOU6yNVFdNaK

Edit: fix link that broke due to Google Drive permission updates

Thanks Leopold Talirz for your kind reply. Firstly, we synthesized new semiconducting material and we need to check its applicability as OFET/TFT. Now, we don't have access to sophisticated instruments for fabricating OFET device. However, we would like to move on, hence, I would like to know:

Thanks in advance. 

1. Spin-coating can, in principle, be used to deposit both the dielectric and the semiconductor. However, the solvent used for the semiconductor must not damage the dielectric (I don't have experience with doing both from solution - perhaps someone else would like to chip in?)

2. Yes. In my thesis I spin-coated the Cytop dielectric on ITO to verify that the mixture was fine (section 4.2).

3. Ag or Cu have also been used (see e.g. http://pubs.rsc.org/en/content/articlepdf/2008/jm/b808435a?page=search). Which contacts to use depends again on several factors, such as the type of transport (n vs p). See 2.3.2 in the semester thesis and reference 7 therein.

Thanks Leopold Talirz. Your suggestions are highly helpful. 

As you can read in a lot of articles on OFET, the easiest way to test a new organic semiconductor is to use a heavily doped silicon wafer with thermal oxide (100 nm is enough). The silicon wafer is the gate of the devices and the oxide is the gate dielectric. On this substrate the organic semiconductior can be deposited by thermal evaporation or by spin coating, depending on the formulation of the material. Source and drain contact can be evaporated through a shadow mask. Gold is the best choice for S/D contact of p-type semiconductor, but can be also used for n-type. Before the electric measurement the devices must be isolated to avoid leakage current. This can be done by mechanically scratching the semiconductor around the contact. 

Dear Luigi Mariucci,

Thanks for your kind information. Your suggestions are highly helpful to us.