Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
How to Build a Computer, Part 1 of N: Silicon
As the illustrious @JohnWalker no longer treads these halls, I figured there was an opportunity to thrust my metaphorical booties into his clodhoppers. I’ve been kicking the idea of this series around for a long time. Broadly speaking it covers everything you need to know to build a computer. Everything. Today, we’re going to learn how to make silicon wafers.
He’s Gone Silicon
Start with silicon. Except we can’t start with silicon; you’ve got to get it from somewhere. Start with sand. (I can hear someone in the back asking where I get the sand. Well, they mine it in Wisconsin these days, but mostly to ship to the North Dakotan oil fields. I don’t know where the computer industry gets its sand, but if you can’t find any you aren’t trying.) Sand, chemically, is SiO2. Close, but we need less O. To render it into pure silicon you heat it up to about a fifteen hundred degrees commie in the presence of coke. Uh, the fuel, not the nose candy. This is the chemical reaction:
SiO2 + 2C -> Si + 2CO
Or possibly this one. Probably both; my sources gave me differing answers.
SiC + SiO2 = Si + SiO + CO
This gets it fairly pure, but you need really really pure. (So pure that only one really won’t cut it.) With the 96% pure stuff you get out of that reaction, you make tetrachlorosilane. That’s one silicon atom with four chlorine atoms stuck to it. You could also make trichlorosilane, which has one hydrogen atom instead of a chlorine. In either case you’ve got a silicon compound that’s readily made into a vapor. With that you can use fractional distillation (like with petroleum, or pappy’s still out back) to increase the purity. This stuff has got to be not just 99.9% pure, but 99.9999999% pure.
Okay, but you can’t make devices out of a gas. You want to reduce that, using something called the Siemens process. You take a rod of (also very pure) silicon. Heat it up to about a thousand degrees C. Then mix your trichlorosilane with hydrogen gas. One more chemical equation.
2SiHCl3 + 2H2 -> 2Si + 6HCl
You get hydrochloric acid out, and silicon. Great! But you can’t just make wafers out of that. I mean you could, but then you’d end up with solar panels, not microchips. What you’ve got now is referred to as ‘polysilicon’ because the crystal structure is all over the place. To make proper transistors you need it to be monosilicon, which is all one big silicon crystal. So we melt the stuff.
One Big Crystal Coming Right Up!
We make a single huge crystal of silicon out of the melt by means of the Czorchalski growth method. Say it with me. “Cho-RAL-ski”. That was pitiful. Try it again. “Cho-RAL-ski”. Eh, close enough. You melt your silicon in a big crucible made of silica. That way when the crucible mixes with the molten silicon, all you’re getting is more silicon. And oxygen. There are reasons why you might want oxygen in your otherwise pure wafer. You want to keep the temperature of your melt just above the melting point. Then you reach down with a seed crystal into the melt and slowly pull it back up.
The surface tension of the silicon will cause the liquid to rise up above the melt. The greater surface area causes it to cool off. By carefully controlling the temperatures you can get the molten silicon to freeze onto the crystal. Because it’s already a single crystal all your solidifying silicon will stick itself into the lattice and extend that crystal structure. The diameter of the cylinder you get depends on how quickly you pull the crystal back out.
There’s another way to get your monocrystal. It’s called the float zone method. You take a column of your very pure polysilicon. You run an infrared heating coil up and down it. The coil heats the silicon in the space directly below it to the melting point. The silicon has enough surface tension to keep the column together. Again, when the molten silicon hardens next to your seed crystal it takes up the already existent lattice and reforms into a single crystal. (So where do they get the seed crystals? Out of previous runs. And where did they get the first one? Good question; I don’t know.)
Breaking Down your Boule
What you get after that is called a ‘boule’. Don’t ask me why. You grow your boule a little wider than you need so you can grind it down afterwards. That lets you control the size much more finely than Czorchalski (“Cho-RAL-ski”) growth does. These things grow anywhere from 75mm to 300mm (roughly three inches to one foot in diameter). They’re also capable of making 450 mm wafers, but the benefits of the larger size aren’t sufficient to upgrade all the machines in your wafer fab to handle the larger size. If it’s one of the smaller ones (up to 150 mm, or six inch wafers) then you grind the edge of your boule to tell the customer what kind of wafer it is. On the larger ones they just grind a notch out of the edge, and scribe that information on with a laser.
You take your long cylinder of silicon and slice it like salami. You use a diamond wire saw to cut it into wafers. The wider you made your boule the thicker you need to make your wafers; they have to support their own weight. You’ve got wafers now, but they’re not ready to ship yet. These things need to be real flat. The smaller the features you’re putting on it the flatter it needs to be. No matter how straight you aligned your saw it isn’t straight enough. You use a process called “lapping” to smooth it out. It’s a planarization process that uses mechanical polishing and some sort of slurry. While you’re at it you should grind that edge down; it makes it harder to break your wafer afterwards.
Lapping your wafer left it with a bunch of micron-sized scratches in the surface. Etch that sucker down with some really nasty acid. acid. It’s still not smooth enough. You can smooth the wafer surface through something called chemical mechanical planarization. Still not flat enough! Nah, I’m kidding. We’re good. Clean ’em and ship ’em.
Now that we’ve got silicon wafers, we can start to build transistors. Exciting! Join us next week when we dive into the intentional and accidental defects in the silicon in “Doping: Semiconductors the Lance Armstrong Way”, or “Gettering R Dunnering”.
This is part one of my ongoing series on building a computer, the Richard Dean Anderson way. You may find all of them under the tag How to Build a Computer. This week’s post has been brought to you by the Society for the Promotion of Chippewa Valley Sand. When you go pound sand, make sure it’s Chippewa Valley Sand.Published in Science & Technology
That’s Frog for “ball.” Sounds more high-tech than “wad.”
Gosh, and I thought you just went over to Fry’s Electronics and bought the finished item.
This is fantastic, just the kind of offbeat Rico post we need more of, a witty and articulate story by a real expert. A worthy successor to anonymous’s fine series of science posts.
Back in the good old days, the flack from the Intel distributor would take everyone in the lab out for lunch. After lunch, we’d all gather around his car and he’d hand out samples and data books.
I’ll get to that part… eventually.
That’s twenty seven hundred degrees freedom, thank you very much.
This post was a bit of a Déjà vu for me. I have a friend that I met right out of college who at that time went to work for Wacker Siltronics which is the division of Wacker Chemie GmbH located in picturesque city of Burghausen situated in Bavaria, Southern Germany that produced those wafers (and you all wonder where I got my predilection for BMW’s). He started as a sales rep for the US northeast and eventually worked his way up to the board (don’t get excited, Germany companies are very egalitarian and they don’t pay like US CEO’s). He has since retired from their executive management and now runs a little outside sales office trading away their “scrap silicone”.
I was treated, while visiting on a long vacation 25 years ago, to the newly established wafer plant located in Bavarian like Portland Oregon. A plant that was operating on the Czochralski method and was getting to kibitz with the production engineers on how to refine the control of the temperature, draw, and “atmosphere” near the boules for what was then going to be state of the art 300mm wafers (which at that time was a BFD). His production crew were sufficiently impressed that he mentioned a few days later that they wanted to offer me a position.
One always wonders about the paths we chose…. Building stuff for space, or helping develop better building blocks for everything that has driven our information culture. At least I had some fabulous vacations in both Southern Europe and the Pacific Northwest, as we annually got together for what became a lifetime of bonding.
Coda to this tale. In August I am going to his baby daughter’s wedding in Portland Oregon. She recently graduating in Engineering from Northwestern University in Illinois, two of his three children landed in Engineering. I like to tickle myself into thinking that being that nerdy pseudo Uncle for so many years had a smidgen to do with it. My wife & I feel a bit honored that we are the only non family adults she invited.
I didn’t think I was gonna read it, but I read it. Fantastic post, Rhody. Looking forward to the next one.
And let that be a lesson to You incels.
Great post and I’m looking forward to reading more. I have worked with “chips” for years, including a job where the design engineers were in the next office, but never got into the real nitty gritty of how they were developed.
So the 99.44% standard is only good for soap?
And Ronnie Milsap.
And bit buckets? Did you get them, too?
yes, I incorrectly edited it.
Or you could just use relays and rotary switches:
The instructions were pretty good, Hank, although I would warn anyone who hasn’t started yet: if you are reading the instructions from your laptop, KEEP THE COMPUTER AWAY from the molten silicon. If it spatters on your keyboard it seems to be almost impossible to clean off. A four inch disk grinder finally did the trick but there are a few keys now where I can’t read the letters, which is slowing me down a little as I type this. I count myself fortunate because my wife can type without looking at the keys, so whenever I get to a letter I can’t find, like A, S, D, W, E, and R, I just ask her to come into the garage and type it (she had kicked me out of the kitchen because tetrachlorosilane seems to give her hives and it killed our goldfish, which I know, we shouldn’t have kept it in the kitchen, how many times did I tell her that and now it’s too late) , and away I go.
The only part I really struggled with was pronouncing that one name, Czorchalski. (It might have been from the monoxide, because I did feel a little woozy). Anyway, after a half-hour or so I gave up and just said “Coach” for short, and my first batch of wafers seem to be okay. I’m hoping there aren’t any, like, hidden defects.
One question: since the next step isn’t coming out till next week, will my wafers keep at room temperature, or do I need to refrigerate them? In which case, will they absorb odors (especially, sauerkraut)?
Fantastic post, Hank! (All attempted humor aside.)
We still did it that way in junior high school in the early Sixties, although on a less elaborate scale. The dials in the old telephones sent numbered pulses to stepping switches, and there were plenty of free plans of how to convert phone switching equipment to primitive computer use. Sympathetic Bell companies often donated Western Electric switch gear to electrical engineering classes; if it made students more interested in Bell careers, so much the better.
One quirk of that time frame: nobody was supposed to be able to own a telephone; in the Bell System they were all rented, all property of the company. As a realistic matter, AT&T knew there were hundreds of thousands of instruments missing from inventory, usually fully amortized ones often left behind in building repossessions or demolitions, and as long as the kids didn’t try to use them to create a competitor to the Bell System, the company was inclined to look the other way.
A lot of interesting things were done with relay logic (and even with mechanical logic) in the pre-computer age: telephone switching systems, industrial control systems, and railroad interlocking and traffic control systems.
I wrote about the introduction of Centralized Traffic Control in one of my Robot Emeritus posts:
The data books all had stickers on them with his name and phone number. I’m pretty sure the anti-static bags had the same stickers. He didn’t really need to hand out anything else.
Stepping switches?! Wow, that brings back memories.
Of course, the dials were Base 10, not binary, a major limitation on most of our amateur designs. Real hotshots bought Nixie tubes for a ten-digit readout. Nixies had just come in somewhere around 1955 and were still exotic high tech magic that cost $8 per tube. But they were “born” Base 10, making the nest of wiring simpler.
That’s when the big boys were still using mercury tubes as delay lines, as well as primitive decay storage with a miniature cathode ray tube and a phototube nose-to-nose. It had a regeneration circuit that would trace a line and be able to read and re-read it until the phosphor faded.
You sure that’s what it was called, and not, you know, just a Mad Magazine parody?
Yeah, there’s a lot of temperature and heat flow stuff that I glossed over pretty hard in this post. Considering too that I’m getting the tech’s eye view of things and my textbooks are glossing over this stuff too… well I had enough thermodynamics in college to know that there’s a great deal going on there that I’m missing.
If you need to you can fall back on “The CZ method”, but that’s just lazy.
Room temperature will be fine for your wafers, but you should probably keep them under HEPA (High efficiency particle absorption) filters. You don’t want to get too much dust on them. You are doing this in a cleanroom right? Uh, with apologies to your wife.
If you’ve finished the wafers entirely they aren’t going to do so well with the sauerkraut odors. In that case you want to stop after lapping and before the etch. The rougher surface will do better at trapping the complex aromatic compounds. In either case expect there to be significant outgassing and hence returend sauerkraout smell next time you heat the wafer up.
Point noted. I’ll be giving conversions in future posts.
Very interesting brute force way to do it.
Each of these things is done much the same way now but with much cheaper, much faster and much smaller and with incremental improvements. Everything is done in stages and Moore’s law driving the show.
I remember watching the movie, “Jobs” and marveling at how much great detail and verisimilitude there was in the electronics store that Jobs sold his first computers to. Genius stuff. I can’t personally vouch for the exact vintage of the things on the shelf but man that art department was killer in that movie. It sure looked like stores I remember from that time.
There is quite a bit of absolutely fascinating stuff in the history of computers. I’d probably go into it, but there’s so many people here who know so much more about it than I do. (For that matter I was expecting to come back to this post and find several I’ve-never-heard-of-them-before Ricochetti saying “well actually the way we do it is…”. That’s the sort of thing you get on Ricochet; people who know absolutely anything.) Heck, I probably wouldn’t be doing this if John I-Ran-My-Own-Computer-Manufactury-In-The-Late-Seventies Walker was still around. Most of this stuff I’ve read about. That guy’s done it.
Fun fact about relays; a lot of industrial control logic was originally done with relays. PLC programming (Programmable Logic Controller; the computers that control industrial machines) is still written out as ladder logic. ‘Course, that’s more @SamRhody ‘s line of work than mine.
Larry, you are as incandescent as the sun on this one, which is to say, I agree. To me, that relatively short, low key scene was the most decisive one in the movie, the one that really determined whether Jobs and Woz and Apple were on a launchpad to the skies, or just another failed small business that never got off the ground. You like to think, “well, talent will always come out on top”, even when we all know that isn’t the case. It wasn’t the cleverness of Apple’s 6502 utilization, but Steve Jobs’ salesmanship that saved the infant company; his first and maybe most significant victory. Because yeah, if you or I promised computers and delivered motherboards we’d expect to get holy hell. This was one of the moments when Ashton Kutcher really came through. I had no special brief for him as an actor, yet he obviously cared deeply about the role and as with the best Hollywood star performances, some part of the real Kutcher wasn’t too far from the real Jobs. He pours on the charm, persuading the owner that not only is this a more than adequate investment of his money, but look at the advantages–the keyboard, case, power supply and disk drives will all, after all, have to be bought from his store.
Bill Gates did the same thing by getting a royalty per copy, instead of the flat $50k that IBM had in mind. One sharp deal at the start sets up everything that follows.