of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  [MUSIC]Hi there and welcometo week two ofNanotechnology:The Basics.My name is Dan Mittleman.And I'm going to be delivering a series oflectures this week on nanoelectronics,which is the intersectionof Nanotechnology with the electronicsindustry.By electronics, what I'm referring to are,the electroniccomponents that are inside many of yourconsumer devices today.They make your TV run, they makeyour cellphone run, they make yourcomputer run.All of those components that haveobviously gotten better andmore powerful and smaller as the yearshave gone by.Which is why our computers become obsoleteevery threeyears and we have to buy a new one.What we will talk about in this lecture isthe trend of how that's happened and whythat's happened and a little bit aboutwhat wemight expect will come next, so let's getstarted.To put things in context, we can take alookat the, these images here, which show acellphone and acomputer from about 30 years ago and theywere obviouslymuch less functional and also much largerthan they are today.Nowadays, we understand that electroniccomponents have begotten a lot smaller anda lot faster and a lot more powerful.In your cell phone, if you have asmartphone, then the phone inyour pocket has more computing power thanthe entire world had in 1950.That's a fairly extraordinary advance inreally, just about one lifetime.So, to understand what's going on here,let's firstlook inside a cell phone and see what wesee.What we find is that there are a number ofelectronics circuitscomponents, these little black boxes thatI'm circlinghere are electronic circuits and you can'tsee them.They're packaged inside of thermalpackaging  and inside there you have Semiconductors.The semiconductor in this case is silicon,silicon isthe material that makes all of thesedevices work.Silicon can be formed into transistors anddiodes and other components and theseneed to be all connected by wires and wecan make little logic circuits.And the smaller you can make thetransistorsand the wires, the more of them you canpack into a certain space and therefore,the morecomputing power you can get in a certainspace.Those these electronic devices are insideeverything you use these days,practically.Their shrinking size has been,what hasrevolutionized hasbrought us what we call now thecommunications industry.The context that we can use to understandthis is aparadigm known as Moore's law, which isnot really a law,there's no law of science or nature thatsays this has to happen.But Gordan Moore, which was one of thefoundersof the company Intel noticed in the mid70's,that the number of transistors that onecan puton a single chip, was doublingapproximately every two years.And he observed that if this trend were tocontinue, we would very rapidlyreach a regime, where we had so manycomputers on a chip that the computingpower on a single chip, would beextraordinary, by the standards of theday.Well, he was absolutely right, as thisgraph shows you, here's the first,this is the 286 computer, here is the 486,here is the first pentiumright there and so on it goes, all the wayup to, thisgraph ends in 2008, but it has continuedsince then, even up until today.So, the number of transistors on a chipdoubles every 2 years andtherefore, the computing power scalesexponentially with time.And this trend is why computers are sopowerful, somuch more so than they were even just fiveyears ago.  There have been many occasions on whichprognosticators, including GordonMoore himself, have predicted the end ofthis law havesaid, no, it can't go on forever, it'sgoing tostop in a certain year and yet it hasn't,it continues.I think it's safe to say that it will endeventually,but there's a lot of discussion about whenthat will happen andthat's an interesting question, whichreally will be answered byNano-technologists.Because, as we will see, the devices thatare being built, at in the modern times,ever since about here, are all devicesthat arebased, that have nano- technology in themthe materials are nano sized.So here's another version of Moore's lawand itturns out there are many versions ofMoore's law.This onedoesn't show the number of transistors ona chip, versus here.This one shows the size of thosetransistors, the feature size, asa function of year and this is also ishonoured exponentially favorable trend.In other words they gets smaller andsmaller as time goeson the doubling, the having time may notbe two years.In this case it's actually a little fasterbutyou can see that there is a an exponentialscaling.And about the year 2001 we passed thepoint at which transistorsizes be, where, now below 100nano-meters, this data point here, thisdata point here is 35 nano-meters, thatwas in about 2009 or 2010.Right now, in 2013, we're down here wherethe size is about 22 nano-meters and it'savery interesting question to ask, whathappens when wetry to extrapolate this line and predictthe future?Another version of Moore's law is the costof a transistor, transistors get cheaperas you make them smaller.And so here, we see another exponentialscaling and nowadays, the cost ofa transistor is point oh, oh, oh, lots ofzeros one dollar per transistor.  There are lots of zeros in thereand that's why we can afford personalcomputers.So, really, almost anything associatedwithcomputation or the semi-conductor industryis onsome sort of exponential trend and hasbeen for several decades and that's reallywhat has driven that industry forward.That industry has taken to using Moore'slaw as a method for predicting, for makingmarket forecasts and so the fact thatthingscontinue to fall along the Moore's lawline.Is not really surprising, it's, it's beenpredicted and it's what they'reaiming for, so it's not surprising thatthey get what they're aiming for.There are also some unfavourable versionsof Moore'sLaw, for example, here's a plot of thepowerdensity on an integrated circuit, as afunction of yearand this is actually a predicting that wasmade in2001, by a vice president at Intel and hepredictedthat if power scaling continued around theMoore's Law trend.That very quickly we would get to a point,where thepower density inside the chips was as hotas the surfaceof the sun and of course this would bebad, becausethen the chip would melt Prediction hasnot come true, because ofmulti core processors, the transition fromsingle core processors, to multi coreprocessors.And also because of the development ofcleverways, to do power management inside thesechips.But it is true that the power scaling isgoing up andso eventually that becomes an issue thatone has to deal with.Here's another unfavourable version ofMoore'slaw this shows the cost of buildinga factory where you can manufactureintegrated circuits as a function of time.And that also goes up exponentially thesedays, if Intel or one of those companies,wants to build a factory to, tofabricate chips, that costs them several
Similar documents
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks