Friday, April 21, 2017

A Place For Sixty To Reflect

Overheard at Lem's:
Sixty Grit said...
III/V, baby - those were the groups we used. Si, indium, aluminum, gallium. those were the ones I had direct experience with.
April 21, 2017 at 7:31 PM

Carbon, silicon, and germanium align vertically in the Periodic Table like a stack of mid-verse rhymes. I say "mid-verse" because they're not at the end of the periodic verses -- the noble gases are. Further below, the rhyme continues with the heavy metals tin and lead, making a vertical pentad of elements.

Carbon, tin, and lead were long-known but silicon eluded discovery until the early 19th-century. This seems astonishing because silicon is the second most abundant element in the earth's crust behind oxygen. The simple combination of silicon and oxygen, SiO2, is the inorganic analog of CO2 and is otherwise known as sand or, when pure, quartz. Elemental silicon was long suspected and early 19th century chemists were convinced that some new element must be present as an oxide in silica -- the problem was finding something to free it from oxygen. Voltaic piles -- which had then recently helped reveal metallic sodium and potassium -- were too weak. Berzellius used a two-step process, first treating quartz with HF to generate SiF4 and then treating the SiF4 with potassium metal. Recently discovered potassium metal was needed. The Germans still call silicon Silizium, giving it the metallic suffix -ium.

Another German (a very bad one) developed a carbo-centric Periodic Table which contains more chemical subtlety than I can master:


Sixty Grit said...

It had never even occurred to me that Si wasn't one of the first elements discovered. It's everywhere! How numerous are the grains of sand on the beach, etc. Thanks for sharing that history.

But having to bust off the oxygen, well, that's a process best left those capable of doing such things. I would have just used a bigger hammer. Makin' small ones out of big ones. Anybody who can do that by messing around with potassium is O K by me.

To follow up on the previous thread, we made CMOS devices - the N devices sat on the P substrate and the P devices were placed in N wells. As a draftsman it was my job to make sure the design rules were obeyed. The big arguments from those days were "Is poly green or cyan?". Yeah, a high level technical discussion, for sure.

Funny thing, Calma, a local company where I did my first CAD training dreamed up a format for the exchange of graphic databases. Called it GDS, then GDSII. The latter became the industry standard and for years we had to struggle to get all the information we needed through and into GDSII databases. Horrible format. Very limited, but they were there first and became the de facto standard. As with many companies in those days, they were bought by a larger one - in their case, GE, and after mucking around for a few years, it was spun off into what became Cadence, the largest competitor to the various non-Cadence companies I worked for.

Then some smart guy figured out how to analyze the graphics using software and design rules could be checked using DRC software, the connectivity was checked using LVS and from there we went nuts with ERC, antenna rule checking, impedance checking, inductance, parasitic checking, and for the egg heads up at Bell Labs, optical pre-distortion, kind of like what Dolby did in amplifiers.

So I switched jobs from drawing transistors to doing technical support for software that did design checking and that's how I got to travel the world. Good times.

I have been out of the biz for 15 years and have not paid any attention to the advances in technology. I ran into a former coworker the other day and he mentioned finned gates as a way to further reduce gate length while thwarting electron tunneling. Well, what do you know? Me, not much. I have moved on.

Now it is time to go to market to sell carbon-based products to mobile carbon-base units. Just another day in the park.

chickelit said...

I ran into a former coworker the other day and he mentioned finned gates as a way to further reduce gate length while thwarting electron tunneling.

The most cited paper from my publishing days involved proton tunneling. Electrons tunnel with ease -- it's rare when something heavier like a proton gets away with it. And yet, enzymes have evolved to encourage it.

rhhardin said...

Chemistry has never been interesting to me, for some reason. Perhaps very unpleasant labs. I have no intuition for it and no interest.

Physics I took to like a duck out of water or whatever the expression is, until it reached the college level and went all mathematical, meaning pencil and paper calculations.

This was fixed in 1963 when I started programming computers and they did all the nasty work, and physics became possible again.