What is the easiest way of building integrated circuits, MOS or bipolar semiconductors?

What is the easiest way of building integrated circuits, MOS or bipolar semiconductors?

Attached: intel_c4004_b.jpg (646x431, 23K)

Other urls found in this thread:

youtube.com/user/szeloof
amazon.com/Fundamentals-Semiconductor-Fabrication-Gary-May/dp/0471232793/
amazon.com/Physics-Semiconductor-Devices-Simon-Sze/dp/0471143235/
amazon.com/Microchip-Fabrication-Practical-Semiconductor-Processing/dp/0071821015
twitter.com/NSFWRedditImage

installing gentoo

haha fuk off m8

>m8
I'm not your m8, buddy.

okay whatever. so which one is the easiest to implement?

cmos obviously, bjts are too leaky

but weren't early ICs made with BJTs? We aren't talking millions of transistors here, heat dissipation wouldn't be a problem. Isn't the oxide layer need for MOS difficult to deposit?

The 4004 was built using self-aligned gate MOSFETs, so probably MOS.

Bipolar are more robust. I've a ES/9000 with bipolars that still run and it's more than twenty years old.

Silicon naturally reacts with air. It's probably not that difficult.

*crack*
Back in my day, we didn't use that oxide bullshit.
Now BJT, that was a real transistor.
*sip*

Native oxide layer for Si from air exposure is ~1-5 nm and does not further grow. Nitric acid baths are how you might be able to do it at home without a furnace, as dip times give you defined thicknesses. And yes you need an appreciable layer a native one will act as a tunnel junction.

>Nitric acid baths
Yikes
I can already smell my lung cancer 20 years from now

All I said is that natural oxidation occurs. It doesn't seem that much of an extension that there would be another solution.

So would a bipolar junction be easier to pull off?

plz gib advice semiconductor fren

Attached: 1531287249664.png (716x1024, 374K)

How do I do it without the acid eating away everything else? What kind of temperature is enough to make the oxide layer?

The standard fare oxidative bath is 60-80C. You do the oxide formation step before metal patterning. Otherwise you won't have good metal contacts. Alternatively you could solution deposit your dielectric but keep in mind they're not particularly good that way and I can't think of a sol-gel/CBD process for high-k's. For that you could just use tape to pattern.

How do you diffuse dopants? The ghetto way a co-worker of mine did p-n junctions was to e-beam deposit boron and rapid thermal anneal it to avoid using borane or ion implantation.

Too self-limiting was my point

Get your basedboy ass outta semiconductor fab.

Im not your buddy, friend.

I'm not your friend, pal.

>How do you diffuse dopants? The ghetto way a co-worker of mine did p-n junctions was to e-beam deposit boron and rapid thermal anneal it to avoid using borane or ion implantation.
Sounds hard. So I take it CMOS is easier?

I'm not your pal, nigger

CMOS is really hard cause you start with a single dopant type on a wafer and then dope aggressively to generate the complement of the transistor type you would make. You're better off, if you really wanna suffer through this using single logic types (NMOS or PMOS) and even so I can't immediately think of a not-furnace way to dope a wafer. I think I've seen some homebuilt vacuum deposition systems but that's a fucking chore.

I'm not your nigger, kike

I am your kike, bow down to me you fucking nigger cattle.

I'm not your kike, pajeet

how do we weaponize it lads?

Attached: hexbug.jpg (950x950, 100K)

Why are you assuming a hobbyist wouldn't have access to a furnace? It's just a fucking oven

>he doesn't have a space alien
fuck off kike

One that'll hit 1000 centigrade with flow control for gases? If that's available you could in principle thermally grow oxide and etch where needed. I do wonder where to buy HF/NH4F

ebay or amazon.

people are doing everything liquid based at the moment, except the aluminum interconnects

I know the Cu and Ni interconnects are done electroless but I thought that the high-k stuff (Hf/ZrOx) was still vacuum deposited

I mean, by "people" I mean literally the one to three people who are attempting this, and the mosfets they've achieved have bad characteristics compared to commercial ones, but I think they aren't using gasses out of practicality. the chemicals are apparently sourced from the same companies that sell to real fabs.
check out sam zeloof's channel. youtube.com/user/szeloof

Legitimately speaking the coolest thing I've seen on this board. His homebrew asher is honestly nicer than the Tegal I used from the 70s

dafuqs an asher and a tegal

Tegal was (is?) a company that made ICP systems for plasma cleaning of substrates using oxygen or carbon tetrafluoride, colloquially referred to as ashing.

Ah, yeah I just saw the video you're talking about

>What is the easiest way of building integrated circuits, MOS or bipolar semiconductors?

There's a bunch of youtube videos which you can replicate, DIY silicone etching/doping/metallizing to create working semiconductors. Careful with the chemicals, you might go blind.

Obviously not a chip duh.

>Obviously not a chip duh.
eh? also you didn't answer the fucking question pajeet

>DIY silicone
I think you mean silicon.
Silicone is a whole 'nuther.

Fpbp

>he doesn't build comfy silicone waffers in his garage

There's also youtube videos explaining logic gates on chips and how memory keeps the bit and how it's recalled again and such things. Obviously to create something like that which actually does something is near impossible without a laboratory and engineers who know what they're doing. You can create transistors though, the basic building block of a chips logic gates at home and test if they're working and use it in homemade "electronics".

Euro, English is my 4th language, sue me.

install gentoo first and then go read some book, nigger.
Analysis and Design of Analog Integrated Circuits by Paul Gray is the patrician choice.

Attached: 1532247519047.gif (700x921, 53K)

>Careful with the chemicals, you might go blind.
That's the benign ones. Fuck up with the HF etchant and you're running the risk of being very dead.

>design
>analog
yeah no, these are probably more OP's speed:
amazon.com/Fundamentals-Semiconductor-Fabrication-Gary-May/dp/0471232793/
amazon.com/Physics-Semiconductor-Devices-Simon-Sze/dp/0471143235/
amazon.com/Microchip-Fabrication-Practical-Semiconductor-Processing/dp/0071821015
The small amounts of diluted hydrofluoric acid required aren't likely to kill, but yeah, you could go blind if you are stupid enough to splash it in your eyes.

>digital
not specified by the OP, that clearly doesn't know a thing about it and must start from the beginning.

I'll forgive you for your redpilled Lain gif

>forgiveness
i'll forgive your faggotry, because you forgave me.

Attached: 1478977607406.gif (503x354, 1.93M)

>small amounts of diluted hydrofluoric acid required aren't likely to kill
Yes they will. They will literally kill you in your sleep, you'll die of a heart attack because they slowly leach calcium from your bones and put it into your blood stream which will instantly clog up your arteries and give you a heart attack.

The only way to save yourself is to take a glycol injection and go to the hospital immediately.

HF is SERIOUS FUCKING BUSINESS.

wikipedia cites a 1.5 square meter exposure for risk of death for systemic effects from FA

It's not because it clogs up your arteries, it's because the heart needs it's nerve's calcium channels to work.