do wafers work even if a human touched it?

[adarw]

if a human touched a wafer after it was made would it still work? would any on the human oils effect the transistors and operation of the microchips?

[Jurrunio]

define "after it's made"? Before it's cut, sealed and packaged onto the substrate or freshly sliced from the original crystal?

[adarw]

1 minute ago, Jurrunio said:

define "after it's made"? Before it's cut, sealed and packaged onto the substrate or freshly sliced from the original crystal?

my bad,  i mean like when its in this state.

 

[BobVonBob]

Wafers only get transistors etched on one side, so if you touched the side without transistors, it would probably still work. This is equivalent to touching the exposed side of your CPU die. On the side with transistors, definitely not. There's a reason chip manufacturers have some of the world's best clean rooms, because even dust particles can ruin a chip. Touching a wafer with your fingers and expecting it to work right would be like emptying an oil drum on your carpet and expecting it not to be stained.

[Roswell]

There are billions upon billions of transistors on an exposed chip. I doubt that you could touch one without taking out half of them since they’re so small and fragile.

 

Then there’s the issue of microscopic debris left behind by your finger. Anything conductive or corrosive and it’s game over. This is why they’re manufactured in crazy strict clean rooms.

[porina]

I'd say it could still work, with certain assumptions. You don't fry it with static. You don't press hard enough to cause mechanical damage. Whatever you touch with is relatively clean. Contaminant conduction wouldn't help, but in itself probably not terminal and could be cleaned, which you'd have to do for further packing and use.

 

On mechanical damage, consider how much pressure it is under when mounted under a heatsink. The cut die will be pressed against whatever they use to hold it to the substrate.

 

The clean room part is primarily needed during manufacturing, when those microscopic structures are being created. It helps to keep it clean while it is tested and packaged, but that is less critical.

[Guest]

Most of the time, humans don't handle these wafers. The one you see is for demo.

[Roswell]

1 hour ago, porina said:

I'd say it could still work, with certain assumptions. You don't fry it with static. You don't press hard enough to cause mechanical damage. Whatever you touch with is relatively clean. Contaminant conduction wouldn't help, but in itself probably not terminal and could be cleaned, which you'd have to do for further packing and use.

 

On mechanical damage, consider how much pressure it is under when mounted under a heatsink. The cut die will be pressed against whatever they use to hold it to the substrate.

 

The clean room part is primarily needed during manufacturing, when those microscopic structures are being created. It helps to keep it clean while it is tested and packaged, but that is less critical.

The pressure doesn’t matter after it’s a completed package because it’s encased in multiple layers of metal at that point. 
 

Touching a bare wafer before that would ruin it.

https://umumble.com/blogs/company_intel/from-sand-to-processor-or-how-a-cpu-is-made/

 

[Jurrunio]

4 hours ago, adarw said:

my bad,  i mean like when its in this state.

Ok so after it has been etched but not yet cut. It definitely cant handle a human's touch.

 

4 hours ago, BobVonBob said:

if you touched the side without transistors, it would probably still work. This is equivalent to touching the exposed side of your CPU die.

There's extra resin coating when it's a CPU leaving the factory, I don't think it's a good idea to touch the silicon directly either... Though I have seen people sanding down their dies for better LN2 cooling, so I guess it's "better not" instead of "hell no"