6 ms·
The science of ultra pure silicon
- qume 8y agoIf pure silicon is 100% silicon atoms by definition, what is ultra pure silicon? 101%? Cool!
- qume 8y agoHey respected downvoters - pure means homogeneous. Can you please explain to be what ultra homogeneous means? When discussing scientific concepts being haphazard with language seems like it should be called out, especially when the source is a respected publication like wired. Simply take a few seconds to let us all know the error in my ways please.
- upandatom 8y agoPure is not 100%. The article talks about 99.XXX percentages depending on application.
- qume 8y agoThe article isn't the canonical source of the meaning of the word pure. And yes, the word pure _literally_ means 100% My point was that the author has misused the word pure. Which remains my point as your argument has failed to add anything valid to change this.
- acqq 8y ago> the word pure _literally_ means 100% Only to you, if you observe it out of the context. The practice is completely different, and the OP article wasn't incorrect. http://www.thequartzcorp.com/en/applications/silicon-metal.html http://www.thequartzcorp.com/en/applications/silicon-metal.h... "For silicon metal to be used in solar and semiconductor applications it must be extremely pure. For solar applications, the metal must be 99.9999% Si, while semiconductor applications require 99.999999999% purity. No quartz product can be purified above 99.999% SiO2, which means silicon metal must go through successive purification processes before it can be used." As you see, the common use in the industry of "pure" is somewhat as "went through the process of the purification," therefore if it is purified more than what is commonly referred to as "pure" then it's "extremely pure" or can even be called "ultra pure" (edit: note, in the sense of "purer than what is typically offered as pure" and not in the sense of "more than what anyway doesn't exist"). In practice, anything "completely" purified, so that "not even single atom of the impurity remains" simply doesn't exist. There's no absolute, only different achieved levels of what is commonly called "pure": See e.g. "Purest silicon sphere ever created": https://www.chemistryworld.com/news/purest-silicon-sphere-ever-created-will-redefine-the-kilogram/3007588.article https://www.chemistryworld.com/news/purest-silicon-sphere-ev... The common usage, that you object to, reflects the reality.
- qume 8y agoThank you, excellent points amd compelling argument. My point however remains - 99.999999% pure makes 100.000% sense. Ultra pure does not.
- tabletopneedle 8y agoFor example, cleanroom has classification that implies some amount of dust particles/impurities in the air. The technicians don't talk about the number of dust particles allowed, they just consider class 5 suitable for some applications whereas others might need class 4. Class 4 cleanroom makes sense, as does "Ultra pure" as long as industry sees it as a standard. Another example: UHD makes as little sense as FHD in terms of display resolution. I mean, Full must mean it's at maximum possible amount of high definition? Well we just know it stands for 1920x1080, and we know UHD or 4K is larger.
- gh02t 8y agoIn this context "pure" is meaningless (truly 100% pure silicon is impossible to obtain in quantity), whereas "ultra pure" is a technical term that refers to a specific level of purity. The meanings are context dependent. It isn't haphazard, it's actually hyper specific and the terms have clear definitions in context. And in this case they are defined in the article.
- qume 8y agoI didn't realise the word pure was context sensitive. OED doesn't either, perhaps you could contact them and help them update their definition: https://en.oxforddictionaries.com/definition/pure https://en.oxforddictionaries.com/definition/pure
- close04 8y agoA pure metal is never 100% simply because that's practically impossible. Yet the term is perfectly correct when used in the right context. Now that you know it's context dependent you can maybe stop insisting it's wrong. Pure or ultrapure simply mean different and very clearly defined levels of purity for every field they are used in. Including some that stretch your concept of pure, like pure quantum states. Please don't try to redefine it. You're going into deeper and deeper mud while trying to walk out clean from this.
- windows_tips 8y ago>1 Not mixed or adulterated with any other substance or material. You could argue the definitions of mixed or adulterated if you'd like. >Pure or ultrapure simply mean different and very clearly defined levels of purity for every field they are used in. Not so clear in the case of dictionaries. >Please don't try to redefine it. Maybe fields that use such highly context-dependent definitions of common words should take a look at their wider usages and determine if an update of their nomenclature is appropriate.
- close04 8y ago
- sanxiyn 8y agoYour argument is valid but uninteresting, hence my downvote. You should mail WIRED for such corrections instead of posting comments to HN. WIRED won't read your HN comments.
- qume 8y agoI apologize if you found my post uninteresting. You must downvote a lot.
- jsjohnst 8y agoI would argue it isn’t even valid. Words have meaning in context, not isolation. There were lots of examples of words misused from their most exact dictionary definition given in this thread, which I’ll leave as an exercise to scroll up if you want to read. This was instead a pointless pedantic rant by a few folks.
- dang 8y agoPlease don't do flamewars on HN and for heaven's sake not nitpicky pedantic ones. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
- robertAngst 8y ago>That quartz is then blasted in a powerful electric furnace, creating a chemical reaction that separates out much of the oxygen. That leaves you with what is called silicon metal, which is about 99 percent pure silicon. But that’s not nearly good enough for high‑tech uses. Silicon for solar panels has to be 99.999999 percent pure—six 9s after the decimal. Computer chips are even more demanding. Their silicon needs to be 99.99999999999 percent pure—eleven 9s. From the article. Pure/purity.
- agumonkey 8y agorecently learned that silicon solar panels used to source their material from 'bad' computer chips silicon :)
- thaumasiotes 8y ago>> Silicon for solar panels has to be 99.999999 percent pure—six 9s after the decimal. Computer chips are even more demanding. Their silicon needs to be 99.99999999999 percent pure—eleven 9s. Yuck. Their style of reporting numbers isn't even self-consistent; an impurity rate of 1e-8 is "six 9s after the decimal", but 1e-13 is "eleven 9s". Obviously, they're wrong about where the decimal point in a percentage is in the first place, but if they're going to invent a distinction between "9s after the decimal place" and "9s", you'd think they could try to get that right.
- Jtsummers 8y agoThe “eleven 9s” is written with the earlier sentence still in context. “After the decimal” can be elided in situations like that because the literate reader knows what they’re referring to.
- thaumasiotes 8y agoWhat does "5 nines of reliability" mean to you? "Nines after the decimal" was nonsense to begin with. But no, it doesn't make sense to assume, when someone uses established terminology, that what they really meant was to refer back elliptically to their own different idiosyncratic terminology.
- robin_reala 8y agoHow are 5 year non-compete clauses remotely legal?
- tabletopneedle 8y agoThey're not. The message they convey is "we do what we please because even if we lose a case every now and then, we still win more in the long run. We own this business". Until the fines for such legal harassment are so severe losing actually damages the business, this will continue until our race is wiped by the climate change.
- close04 8y agoThey are legal because the actual wording of the agreement has to be as narrow as possible. So you would be barred from competing in the same industry, in a certain geographical region, etc. It all has to be very specific. Otherwise it just gets dismissed in a court. If I sell you my self driving tech company and tomorrow I'm founding a new one, right next door, ideally even stealing all the former employees back, that's competing. If I move across the ocean or start building self sailing yachts that's probably good. Basically the limitations can only go as far as needed to protect the legitimate interests of the parties. There's of course some grey area where those interests overlap but I think overall 5 years isn't necessarily exaggerated.
- arminiusreturns 8y agoThe thing most companies don't realize is that non competes are extremely hard to enforce. Longest I had was 2 years, and that was just as a sysadmin at a genetics company. 5 does seem excessive and likely unenforceable.
- DoctorOetker 8y agoThe comment from the security guard paints a big terrorist target on the facility!