The Birth of Semiconductor 2.0

Roland Piquepaille writes "According to several articles in the press, an Austrian company has opened a new chip printing factory. But there is a twist. The chips produced by this factory, dubbed Semiconductor 2.0 by the company, will be organic semiconductors, and will be produced by inkjet printers. According to the company, the new factory will be able to produce 40,000 square meters of semiconductors per year, mainly for the biotech, clean tech, and defense industries."

Comments

Hardly Semiconductor 2.0

[26199]

The feature size (10-100 micrometres) is 100-2000 times what you'll get from a modern silicon fab plant (50-100 nanometres). Call it 1000 times for the sake of argument. So for every organic transistor you could instead have 1,000,000 in silicon. It's not exactly going to be a revolution in processing power. (In fact it puts us back in the early 60s).

The market is apparently cheap, disposable logic. From the sound of it, the fab plants are about 100 times cheaper for the same chip-area/year output. That means each transistor will be up to 10,000 times more expensive; it's going to have to be very simple logic to be cost-effective.

The process sounds like it could be well suited to doing small runs, so I suppose that's something.

Ah well. I will go on record as saying that this is not hugely exciting. When in 50 years' time organic semiconductors have taken over you can all mock me as appropriate :)

Re:Hardly Semiconductor 2.0

[MITEgghead]

There are actually cases when you want one cheap big transistor as opposed to a million tiny ones. For instance, if you have a display, also made of organics, with pixels around a millimeter then you can manufacture driving circuits on the same substrate even directly below the pixel. This keeps costs low and allows new flexible substrates to be used for mobile applications. Other applications from the article are biotech and military where the transistors would be used as part of real-time biological/chemical detectors which need relatively large areas to pick up enough of the substance to be detected.

The bottom line is that no one thinks these things are going to be doing heavy-duty logic. But they can certainly do easy logic cheaply and in novel applications.

Re:Hardly Semiconductor 2.0

[MobyDisk]

When I was taking Computer Engineering at University of Maryland, 3 students were allowed to have their chip designs fabricated. We used a 250nm feature size, which was fine for what we were doing. Perhaps something like this could eventually be good enough that every student could have their chip fabricated. The larger feature size would be sufficient for a learning excercise. 10 micrometers does seem kinda big, but if the wafer is larger, it might still be good enough for academics.

Re:Hardly Semiconductor 2.0

[QuantumG]

If you RTA or if, say, the people who modded you up to +5 had RTA, you'd see why feature size is terribly unimportant.

Fuckin' Slashdot.

Re:Hardly Semiconductor 2.0

[Original Replica]

the idea of degradable logic is pretty enticing for the military.

This isn't news, our military has been using degraded logic for years.

injet

[mastershake_phd]

Inkjet printers? Man they are going to be expensive.

Of course, the REAL money . . .

[StefanJ]

. . . will go to the folks who supply the ink cartridges.

Seriously, this is good news. Cheap, low performance electronics could play a big role in "leapfrogging" in the developing world. Going straight from low-tech to whoa!-tech, leaving out the capital and infrastructure intensive middle.

Re:Of course, the REAL money . . .

[veganboyjosh]

not only developing world. i was thinking more along the lines of cheap-o electronics inside cereal boxes, happy meals, etc...a lot of overlap, tho, i'm guessing there is.

Re:Of course, the REAL money . . .

[Intron]

And what would cheap electronics printed right onto packaging do?

*** ADVERTISING ***

This may be an advance, but I'm sensing a mixed blessing.

The birth of 2.0 version 3

[spun]

2.0 is so 1.0 these days, isn't it time to move on to 3.0 yet?

Re:The birth of 2.0 version 3

[Threni]

3.0 is the new black.

Imagine a beowolf cluster...

[vrmlguy]

...that you can wear!!!

Organic hmm delicious

[aiwarrior]

Now with OLED's i can order electronic food, display the picture in my OLED display and eat it.
Simple.

Semiconductor 2.0

[Anonymous Coward]

FINALLY, I'll have something to run my web2.0 on.

Re:Roland Piquepaille again!

[$RANDOMLUSER]

Oh wait. Never mind.

Exactly. I think he deserves a little credit for turning over a new leaf. He's not the click-whore of old, and like him or not, he always did have a talent for finding stories that appealed to the /. crowd. I used to hate him, but these days, I have more tolerance for him than I do for Zonk.

Semiconductor 2.0 is quite appropriate

[mangu]

It's not exactly going to be a revolution in processing power. (In fact it puts us back in the early 60s).

Well, considering that silicon semiconductors must be somewhere around "Semiconductor 10.0" or so by now, calling something that reminds us of the early 60s "2.0" seems all right to me...

Re:Leapfrogging the developing world

[PopeRatzo]

Why is it that westerners think that the developing world has to start way back in the 1960

Have you ever been to a "developing" nation? Many of them are "starting" way before "1963". The technological revolution in Europe and the West came after most of the residents of those regions were reasonably well-fed, healthy, had some plumbing, indoor refrigeration for food, etc. That is not the case in much of the "developing" world. I was in Tanzania a few years ago, and traveled through some of the countries along the Western coast of Southern Africa a few years before that. I've been to Sri Lanka, Pakistan, and have seen much of Central America. They wish they had 1963 technology in many of those places.

It's not about the level of technology, it's about how widespread it is among the population. Believe me when I tell you that just because the friends of the Prime Minister all have Mercedes, iPods, satellite phones and Playstations does not mean that their country has technology.

And just because the gluttonous creeps who run global corporations decide to set up factories or call centers or even code farms CERTAINLY doesn't mean that those countries have any technology. It just means that there's someone so despicable that they'd suck what little wealth is in those countries for themselves and their shareholders while taking advantage of the fact that the life expectancy is 49 so they won't have to worry about paying retirement benefits.

I wish I could write vividly enough to express how disgusting the exploitation of the "developing world" by global corporations is and how badly it's misusing the people in those places. And remember, once they can drive down the middle class here in the US so that Americans will accept those wages, they'll deign to come back here, but only after we promise never to utter the word "union" or "collective bargaining" again.

Strangely, what computers are in those countries mostly seem to be running Windows. Hmmm.

The equivalent of one silicon wafer a year? Wow!

[nick_davison]

According to the company, the new factory will be able to produce 40,000 square meters of semiconductors per year. Given that the whole goal of semiconductors has been to make them smaller and smaller, boasting about how much area you can use up isn't necessarily a good thing.

It's worse when you consider this tech is roughly micrometers to silicon's nanometers. 10^3x10^3 means you're looking at a millionth the area utilization of silicon. Divide 40,000 by a million and you're looking at the equivalent of 0.04 square meters of silicon or roughly that of a single 12" wafer. A whole factory to produce the equivalent of one silicon wafer a year? Not such a great boast.

Yeah, I'm sure I've got meters squared and square meters confused, messed up an area calculation or somesuch... But you get the idea.

Funny that square. . .

[treeves]

meters are used as the measure of production, rather than #transistors or some other more meaningful measure. Hiding the fact that 1 sq cm only has 1000 transistors instead of 1,000,000?

Re:Roland Piquepaille again!

[bill_mcgonigle]

All of which can be found days earlier by having RSS feeds to ArsTechnica, The New Scientist, Computer World, The National Academies, and the AP Science and Technology wires. What's his value add again?

Um, you just proved your counterpoint - by us not having to read all the RSS feeds in existence. Cripes, I have 4762 unread items in NewsFire and I have none of those feeds you mentioned.

"Would you like to print your new graphics card?"

[7Prime]

I can just imagine 20 years from now:

"Congradulations! You have just downloaded a new video card. Print video card? [Yes] [Cancel]"