From: Grant¹ 18/10/99 23:05:48 Subject: Dr. Ed G post id: 934 Found this article on the web "Researchers will reveal some exotic technologies of the not-so-distant future at the upcoming International Electron Devices Meeting in Washington, D.C. This year's highlights include TRW's high-speed 69-GHz transistor, which uses an expensive semiconductor material called indium phosphide instead of silicon. Intel plans to show a 16Mb memory chip running at 1.06 GHz and using technology that may one day push the clock frequencies of microprocessors beyond 1 GHz. Seiko Epson will present an active-matrix flat-panel display built from a light-emitting plastic called phenylene vinylene. And researchers from Philips Semiconductors will demonstrate a new CCD image sensor that packs in 6 million pixels-- 4 million pixels more than today's high-end digital cameras. Much of the discussion at the December conference will focus on the challenges researchers face in trying to make smaller and smaller transistors. The industry may be approaching the point where some of these problems will become insoluble." What are your comments on the last sentence? From: Dr. Ed G (Avatar) 19/10/99 3:21:31 Subject: re: Dr. Ed G post id: 946 Well, it's true that we're going to hit a limit eventually, the question is whether that limit will be physical or technological. Personally I don't think we're even close to the technological limit yet. In terms of mass production of the top of my head I think the standard for minimum line width is around 0.4 micron (400 nm) for deep UV photolithography. However, in the next few years I reckon we'll see a move from UV lithography to X-ray lithography which should be able to achieve line widths down to 25 nm fairly readily, so I don't think we've reached the technological limit quite yet (though we very well might within the next 10 years). In terms of physical limits we may well be approaching them. In a field effect transistor (FET), the basic microelectronic device from which all the logic circuitry in modern computers are built, there is a limit to how small you can make the "channel" (the bit between the "drain" and the "source" who's electrical conductivity is controlled by the voltage applied to the "gate") beyond which you cannot achieve effective switching. IBM have just recently made an FET with the smallest channel ever, but I can't remember the dimensions or its characteristics (but basically, the smaller the device, the faster you can switch it). Furthermore, if things get too fast we may reach a point were the clock speed gives an EM wavelength equivalent to the lengths of interconnects between IC, at which point they will act like microwave antennae and all the energy will be radiated into the cosmos with none left to transmit information. (though we're not there yet, this could start to become a problem when we start to push into the 10's of GHz range) However, the people involved in ultra-microelectronics (we really should start refering to it as nanoelectronics soon) are a bunch of pretty clever cookies, and will probably come up with something to continue to push the envelope of technology further and further along, for a little while yet. People have been talking about our imminent approach to the absolute physical limit of technology for at least a decade now, and we still haven't hit it. Although, to abuse a well worn metaphor still further, it is usually difficult to see what might lie much beyond the frontier of human technology/knowledge. I give it about another decade still before something BIG has to happen. Soupie twist, Ed G. From: Dan B. 19/10/99 9:00:41 Subject: re: Dr. Ed G post id: 953 Those plastic screens have been around for a bit. I remeber when the original B&W (breakthrough!) version was shown in early '98. The colour one was a couple of years off then so it makes sense that it's out now. I wonder if theyre still the small 10x7 cm jobbies? Dan. From: Grant¹ 19/10/99 19:58:15 Subject: re: Dr. Ed G post id: 1074 Thanks. I was reading the posts on a benchmarking site forum & inbetween the flames there were some interesting comments on the movement from aluminium to the use of copper in ICs & why it's only now that companies are starting to move in that direction- apparently the increased costs are due to the increased number of steps in the manuafacturing process. From: Dr. Ed G (Avatar) 19/10/99 20:06:14 Subject: re: Dr. Ed G post id: 1080 I believe copper can act as a contaminant causing carrier traps which degrades performance, so you have to take special precautions that you intergrate it into the design so that it can't, and this is where the extra steps come in. Soupie twist, Ed G. From: Cam 22/10/99 18:36:05 Subject: re: Dr. Ed G post id: 1822 those gates you talked about..... in a computer, are they similar to the sodium/potassium gates in the brain? I heard (don't remember where) that THIS is the limit to which we can push current IC tech. i am not using my brain properly with this post, so i will stop before i look too silly. From: Dr. Ed G (Avatar) 22/10/99 21:11:56 Subject: re: Dr. Ed G post id: 1846 Hmmm... don't know enough about nerve function to say. They might be analogous, but the actual physical mechanism, and the limits imposed are unlikely to be related. In a field effect transistor (FET) you use the electric field applied to a 'gate', which is insulated from the device by an oxide, to control the electronic energy bands, and therefore the concentration of charge carriers in a semiconducting 'channel' between a 'source' and a 'drain' (the name for the two contacts on either side of the channel). This controls the flow of current between the source and drain, and thereby allows you to switch the device between high and low currents with a relatively small gate voltage. If you connect a number of these switch FET's together, you can make a logic element (like AND, OR, NAND, NOR, and NOT), and if you connect a number of these logic elements you make a computer. The problem with size comes when make the channel too small that the source and the drain are so close together that current flow is no longer effectively controlled by the gate, and so you can no longer switch the device. I'm not sure how close we are to this limit in commercially produced microelectronics, though (I suspect we're still some way off this limit, yet). Soupie twist, Ed G. This forum is un-moderated. The views and opinions expressed are those of the individual poster and not the ABC. The ABC reserves the right to remove offensive or inappropriate messages.