# Would it be possible to make a solid state transistor by hand?

I am curious, would it be possible to fabricate a modern BJT/FET transistor with hobby/garage tools only if size and aesthetics was not important?

Which materials would one use and how would one acquire them?

• I think this question has come up before. A point-contact transistor can be done by hand, but a modern BJT or FET can't. May 21 '15 at 21:12
• Which aspects are modern if not the size (or aesthetics)? May 21 '15 at 21:12
• @Samuel: The materials and working principles? May 21 '15 at 21:14
• A transistor is difficult. But maybe you could try making vacuum tubes? May 21 '15 at 22:46
• Lots of DIY projects on this site (including FET). sparkbangbuzz.com/index.html May 21 '15 at 23:53

Well, yes and no.

You can make a transistor by hand in your kitchen. It will involve some nasty chemicals, but it is doable. A friend of mine made a LED a couple of years ago. Not a transistor, sure, but the same process was involved. (She was studying semiconductors at that time, so she had help. The process itself was quite similar to producing transistors).

However, It is very unlikely that you'll ever be able to make a transistor in your kitchen that even comes close to modern mass produced transistors. You wouldn't even come close to what has been produced in the 70th.

Nonetheless you may end up with a working transistor that does amplification. If you want to do this, and you're willing to invest some time and money into it, please do so! That would be really cool.

• She made a diode, not a transistor. Big difference (the process is MUCH simpler to make a diode). I would be FLABBERGASTED if you could make an operable BJT that provides amplification. How do you make the base thin in your kitchen? May 21 '15 at 22:07
• @crgrace first working transitors have been created about 100 years ago. Back then this was a curiosity and they didn't knew much about the process. Nowadays we have much better tools, material and so on. It is doable in your kitchen if you really want to. May 21 '15 at 22:16
• @crgrace the patent was applied 1925, so it's 90 years ago that the idea came up. Yes, it took a couple of years before the first transistor has been produced, but that first transistor was a huge lump of germanium. Larger than your thump. I repeat: building primitive transistors is possible. You can do that. They won't perform well, but there is exactly zero magic involved. Just physics. May 21 '15 at 22:51
• There you go: A couple of freaks cooking their own FET in their kitchen: santarosa.edu/~yataiiya/E45/PROJECTS/… May 21 '15 at 23:18
• @crgrace, you can get a 1000°C kiln on Amazon for $400. – Zulu May 22 '15 at 1:33 Sure it can be done, if you have ninja skillz: Homebrew NMOS Transistor Step by Step - So Easy Even Jeri Can Do It Making Microchips at Home - Cooking with Jeri Part 1 I wanted to post more links but I don't have enough reputation points to allow that. Just search for the followup videos to the 2 links I have provided. A very similar question has been asked before. My answer there included this link to the website of a someone who has actually made thin film transistors out of zinc oxide. That site also includes links to otehr people doing similar work - at home and in the garage. So, it is possible to make transistors at home, but from what I see it is more of a learning experience as the produced transistors have really poor performance. It's possible, but not so easy and the result is not going to be anywhere near the quality of components that you can buy. The main issue is that in order to actually form the semiconductor junctions inside of the transistor, you need to carefully change the properties of a piece of semiconductor in a controlled manner. This requires rather specialized tools as well as some pretty nasty chemicals. Namely, you at least need a quite high temperature (1000C) oven. See https://www.youtube.com/watch?v=w_znRopGtbE for a high level explanation of what you would need to do to make a FET. To make something even remotely 'modern', you basically have to build your own clean room and acquire used semiconductor manufacturing equipment. It is possible to make some very primitive diodes, such as what is used in a crystal radio, but this is more 'finding' a diode than making one as the actual junction is already in the crystal, you just need to go poking around with a piece of wire until you find one. • The junction isn't in the crystal; the semiconductor is. The rectifying junction is formed when the "cat's whisker" touches the surface of the semiconducting material. May 21 '15 at 21:45 • So it's a semiconductor-metal junction, then? Interesting. For some reason I thought it had something to do with the surface of the crystal and the wire was just required for the electrical contact. May 21 '15 at 21:53 • @alex.forencich yep. It's actually a Schottky diode (although obviously the theory hadn't been worked out yet). May 21 '15 at 22:31 If I recall it right, Jeri Ellsworth made a few transistors, depletion-mode NMOS be exact, and recorded videos logging the process of making it. You can check YouTube for how she made it and her tests. Your question is wrong, because size is one of the key attributes that defines a "modern" BJT or MOSFET. Given that you really mean "is it possible to make a BJT or MOSFET using obsolete techniques in your garage" I would say it would be so difficult as to be almost impossible both from a technical and financial standpoint. Using just used, refurbished equipment you would at the very least need a diffusion furnace, a spinner to apply resist, a few masks, a light source for photolithography, and some way to metallize the wafer (presumably using an evaporator). You'll also need a bath for wet etch. Then you have to have the ability to test your structures (they won't work for a while...) Good luck getting all that to work. In addition, this all needs to be done in a clean environment, so you'll have to seal your garage and set up a laminar air flow and install air scrubbers. This could EASILY run over$150k even using surplus gear. It just isn't a good hobby. There are a lot of fun things you can do in electronics. Transistor fabrication isn't one of them.

You could make capacitors or resistors or inductors. That could be fun.

• Notwithstanding your list, which I understand the reasons for, I suspect one could do it for under \$1000, given enough dedication. Jun 20 '15 at 5:48