Oxide thin films are done through techniques like e-beam evaporation and CVD etc. If the compound is low melting and can easily evaporate/sublime using resistive heating then a boat/crucible of some high melting point metal can be used for the purpose.
Yes you can do that it depend on the melting point of the oxide layer. Like MOO3,WO3, GeO, have low melting point and can be easly evaporated but oxide like TiO2, have high melting point so need e-beam evaporation.
Indeed, sputtering or e-beam are the preferred methods. Sputtering is preferred over e-beam to get correct chemistry. In e-beam, while the energetic beam may have no problem vaporizing the high Tm compound, the chemistry may come out a bit off in the product. Generally erring towards oxygen deficient.
Alternatively, you may use thermal evaporation in a reactive environment. For example, SiO2 may be realized evaporating an SiO or Si source in a partial pressure of O2 impinging on the substrate. Further, some companies make "baffle boxes" to deposit oxides like SiO2.
Due to the reactive nature of Fe, an oxide component will generally be present unless deposited under UHV conditions. Further, once exposed to ambient, depending on the thickness of your film, oxygen will diffuse into your film. You may be able to get close to your desired chemistry by depositing the Fe in an O rich environment and then post deposition annealing in ambient.
Depending on the affinity of metal to be coated ebeam and thermal evaporation can be use with oxygen gas injection in the chamber at required level near deposition substrate during the evaporation.
Yes, it is possible to get thin films via thermal evaporation route, even nanostructures (like nanowires) can also be synthesized using thermal evaporation. You just need to control the experiment conditions and parameters.
In a first approach, the answer is yes by using a certain oxygen pressure during the evaporation of the metal or semiconductor. However some considerations should be taken. First, it is important to know that metallic or semiconductor oxides usually growth by this method with oxygen deficiency. This can be solved by heating the substrate or/ and by using some ionic bombardment to improve the oxidation reaction. Thermal evaporation is a thermodynamic quasi-equilibrium process and in compound oxides it could appear in the substrate some phase separations or metallic aggregations. These problems are very difficult to overcome. Other important issue when using reactive thermal evaporation is the changes in the deposited film morphology. A gas pressure in excess can make the surface of the deposited film very rough.
Yes, it can be done, but the high melting point materials are particularly challenging. I have done TiO2 by evaporating Ti metal with oxygen present (not accurate stoichiometry, but sufficient for many applications). You have to raise the current to a high level very quickly in order to melt the Ti and begin to evaporate before it partially oxidizes. Iron is of a similar melting temperature, though I have never tried it. As Stephen mentioned, SiO can be evaporated to produce something close to SiO2 (again the stoiciometry is generally off a bit). Certainly other methods are preferred, but it is not completely out of the question to evaporate thermally (just very difficult!).