Why next iPhone screen could be made of SAPPHIRE - and a steal...

Reg mining correspondent talks ores and slags

Top 5 reasons to deploy VMware with Tegile

Man-made sapphire could replace Gorilla Glass as the material of choice for scratch-and-crack-resistant mobile phone screens in the near future, according to a recent speculative piece from MIT Technology Review.

According to the research university's mag:

Manufactured sapphire — a material that’s used as transparent armor on military vehicles — could become cheap enough to replace the glass display covers on mobile phones. That could mean smartphone screens that don’t crack when you drop them and can’t be scratched with keys, or even by a concrete sidewalk.

Having had a little wander around the relevant places and a few chats with people who would know, I'd say that it's actually not just possible but highly likely. But perhaps I should explain with whom I've been having those chats so you can understand why.

My day job for the last year or so has been huntin' slags in the famous Ore Mountains, an entirely wonderful occupation for a middle-aged man, I'm sure you will agree. The area, straddling the border between Germany and the Czech Republic, has been mined since the 12th century and there are all sorts of exciting slag piles from which I might be able to extract my favourite metal, scandium.

In fact, my office is in the factory that used to extract Sc from the local minerals back in the 1950s. I can actually look out my office window at the mountain that provided Sir William Crookes with his Sc back in 1909 as he investigated the various salts.

That doesn't have much relevance for sapphire of course. What does is that the factory has moved on and is now one of the major global producers of that very sapphire that is under discussion. Further, over on the German side, we're working with people who make silicon ingots: they might want some other part of those slag piles after I've extracted my Sc (or possibly before I've done it). This has enabled me to chat with the two sets of people who can inform us as to whether the MIT technology mag's suppositions are reasonable. And it seems that indeed they are.

The basic thing about glass is that it can include all manner of metal oxides. It's actually more of a state of matter* than it is any specific chemical combination and we do often vary the metal oxides in it for different uses.

Sand - but with a few extra bits

Standard plain glass is silica (silicon dioxide). Glass for cathode ray tube screens is usually 25 per cent lead oxide to stop the radiation frying your brain. Camera lenses are 25 per cent or so lanthanum oxide (purportedly to improve visual clarity, though this may be in dispute). Face masks for deep sea divers, meanwhile, are heavily doped with thallium oxide to correct for the problems created by the weird refractive index at great depths. Aluminium is, of course, a metal, while sapphire is an aluminium oxide: it's not quite and wholly true to say that it's just aluminium glass but it's a useful way of thinking about it.

Gorilla Glass is that silica stuff but with extra potassium ions added.

There is a difference in the price of alumina and silica of course: the latter is really just beach sand, and is priced at perhaps $20 or $30 a tonne. Alumina might be $300 a tonne at present. But once we come down to the per kg level that represents the difference between 3 cents and 30 cents for the raw material. Current price differences for screens though are $3 for the Gorilla, $30 and more for the sapphire. The raw materials costs don't explain this price difference: therefore processing costs must.

Sadly, we don't actually produce manufactured sapphire the way we make glass. We make it the way we make silicon ingot** (PDF), the stuff we slice up to make computer chips and solar cells.

We should also note that we've just seen one of the great price reductions of all time as people just get better at slicing up the silicon ingots. It's no great revolution, you understand, just good engineering, and getting better at doing the shit iteration by iteration.

Learning from the solar cell industry

Traditionally, the silicon for solar cells was taken from the offcuts and rejected ingots from computer wafer manufacturing. Then the market took off and the price of Si ingots soared to $450 per kg. There just wasn't enough crud being rejected to feed the new market. So inquiring commercial minds thought about how you might produce Si ingot more cheaply: and they succeeded in doing so. The current market price is around $18 a kg and a fully loaded (ie, including capital costs etc) production cost might be around $25/kg or thereabouts.

The basic method was to break the operation out into its various component parts. You've a raw material cost, of course, and an energy cost, and then the actual manufacturing cost. The first two are pretty much static per kg of material, but that last bit, manufacturing cost, is hugely variable. The thinner (smaller diameter really) the ingot, the more the manufacturing part will cost you per kg of material. So, obviously, to bring costs down you want to be making fatter ingots.*** This is exactly what the Si industry has done. Over recent years, ingots have just been growing ever fatter and production costs have been slumping.

As an aside, this is what bankrupted Californian solar energy start-up Solyndra. It wasn't that it had received gobs of government money ($535m), and it wasn't that is was being undercut by Chinese makers such as Suntech and Yingli. Solyndra's problem was that it had a design which assumed expensive silicon ingot. This meant that the firm found itself concentrating on minimising its use of Si and increasing the efficiency with which it worked. Then Si prices slumped and the design - which minimised Si usage in favour of increased efficiency - was clearly doomed. It wouldn't have worked whoever had been running it and with any amount of government cash props. Before the Si price slump, it wasn't obviously a bad idea, but post facto, it was as dead as a dodo.

The verdict? It's entirely possible

Since I first saw the MIT article, I've been kibbitzing with the various sapphire and silicon guys I meet while huntin' slags and the general view is that, yes, the silicon model is exactly the way the industry will develop. It isn't exactly the same, of course, but it's close enough that we can use it as an analogy.

As the sapphire is made in ever fatter ingots, the price per kg will come down. Currently, at least with the pieces I've seen, it's about the shape and size of the sort of candle you might put on a dinner table. In the coming years we all expect it to get wider and wider, as silicon has done, fattening to the girth of a fat candle carried in the church parade all the way up to the "elephant's tampon girth" of current silicon ingots. We would also expect production costs to come down as they have with silicon: perhaps not a 10x reduction, but no one can see why a 3x or 4x wouldn't be achievable.

And at $10 and under for a screen, most think that sapphire would be competitive with Gorilla Glass at its $3. After all, the sapphire is some three times stronger, three times less likely to crack if dropped, and three times harder to scratch. Of course, these calculations do not take into account the markup Apple may slap on for fanbois who want a sapphire screen...

Amazing the things you can find out when you're off huntin' ores. Smartphones could soon have sapphire screens. Who would have thought it? ®

* This would not be the correct answer to a university materials science, chemistry or physics question. But it's good enough for us here and now.

** Again, this wouldn't get a passing grade in a materials science course. But it is close enough for us here and now.

*** Uni courses etc. But good enough.

Beginner's guide to SSL certificates

More from The Register

next story
PORTAL TO ELSEWHERE scried in small galaxy far, far away
Supermassive black hole dominates titchy star formation
Boffins say they've got Lithium batteries the wrong way around
Surprises at the nano-scale mean our ideas about how they charge could be all wrong
Edge Research Lab to tackle chilly LOHAN's final test flight
Our US allies to probe potential Vulture 2 servo freeze
Europe prepares to INVADE comet: Rosetta landing site chosen
No word yet on whether backup site is labelled 'K'
Cracked it - Vulture 2 power podule fires servos for 4 HOURS
Pixhawk avionics juice issue sorted, onwards to Spaceport America
prev story


Secure remote control for conventional and virtual desktops
Balancing user privacy and privileged access, in accordance with compliance frameworks and legislation. Evaluating any potential remote control choice.
Intelligent flash storage arrays
Tegile Intelligent Storage Arrays with IntelliFlash helps IT boost storage utilization and effciency while delivering unmatched storage savings and performance.
WIN a very cool portable ZX Spectrum
Win a one-off portable Spectrum built by legendary hardware hacker Ben Heck
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?
Beginner's guide to SSL certificates
De-mystify the technology involved and give you the information you need to make the best decision when considering your online security options.