Question: how are computer micro chips made

Simone Sturniolo answered on 18 Jun 2013:

This is an interesting question with a long answer!
Computer microchips are made of lots and lots of small electronic components called “transistors”. A transistor, in a nutshell, is a switch. The only difference from your normal switches (the ones you use to turn on or off the lights etc.) is that, instead of being activated by a movement (your finger pushing a button), it’s activated from another electric signal. So a transistor is a switch that has three electric wires connected to it: one for the electricity going in, one for the electricity coming out (only if the switch is “on”, of course), and one to tell if the switch has to be “on” or “off”.
The fun thing with transistors is that they allow us to make mathematical operations. For example, imagine having one transistor, and we decide that for each of the wires attached to it, if there is current, it means “1”, while if there is no current, it means “0”.
So let’s see the possible combinations:
There’s no current going in (zero) X There’s no current in the ‘switch’ wire (zero) = there’s no current going out (zero)
There is current going in (one) X There’s no current in the ‘switch’ wire (zero) = there’s no current going out (zero, because the switch is ‘off’ so the current does not pass!)
There’s no current going in (zero) X There is current in the ‘switch’ wire (one) = there’s no current going out (zero, because the switch is ‘on’ but there’s no current that can pass through anyway)
There is current going in (one) X There is current in the ‘switch’ wire (one) = there is current going out (one)
So with a single transistor we can represent a simple operation: multiplication. By combining many transistors, linking them to one another etc., and by using a system called “binary code”, which allows us to represent all numbers as a series of 0 and 1, we can use transistors to make all operations that we want to. We’ve just designed a micro chip!
Now the problem is how to make it small. At the beginnings of computers, there were no micro chips, and calculators would take up entire rooms of space (while being a lot less powerful than a Playstation or a cellphone). Nowadays, microchips are very small thanks to a technique called “photoetching”. It’s a fabrication technique that borrows a bit from old photography (the one with actual film instead of digital cameras) and allows the fabrication of tiny chips with billions of transistors. It works like this: the engineers design a new microchip and print its scheme on a transparent film. The scheme then is projected on a “wafer” made of silicon. The projection works in reverse compared to the one you see in cinemas: instead of making the image bigger, it makes it smaller. At this point, a new trick comes in play! The silicon wafer is covered in some substance that we’ll call, familiarly, strong-ite (don’t tell this to your chemistry teacher or he’ll kill me) and that is very resistant to acids, but weak to light, that makes it transform into not-so-strong-ite (again, not exactly a scientific name). When the projected image hits the wafer, it will transform the strong-ite into not-so-strong-ite only in certain parts – and basically draw the scheme of the chip on it! At that point, acid is poured on the wafer. The strong-ite, being strong, resists it, while not-so-strong-ite is not so lucky and is washed away, exposing the silicon. In this way it’s possible to lay down, for example, metal wires, or perform other operations only on certain parts of the wafer (the exposed ones) while leaving the others intact. And so a chip is drawn and created. The original scheme is very big, of course, but it allows to create a lot of microchips each of the size of a fingernail!