The Incredible Accelerating Technology The year was 1947. On December 23, the first working solid state amplifier was built. We now had transistors. Here is the strange part, prior to that, there had been a problem with condensation forming on the transistor and preventing it from working. Walter Brattain, who was working on the experiment, in frustration, put the entire apparatus under water and believe it or not, that cured the problem. The first transistors were huge devices compared to the ones on a computer central processing unit. Today there are actually chips that have a billion transistors and the entire chip is smaller than the first transistor. Can you imagine it, we have shrunk this thing to one billionth of its original size? We did this in only 60 years. There is an enemy to this efficiency however and that is heat. Moore's Law states that computer power will double every 18 months and since the invention of the computer central processing chip, it has been right on schedule. There has been much talk of using new materials to make computer chips. Right now, silicon is the most popular and cost effective, but it has its limits. It was fine when chips were larger, but as they grow smaller, we are going to hit a limit to how small they can be made when using this material. At most, it is said that silicon will reach its size limit in about 20 years. Once the chips get small enough, they just won't function. Scientists feel that because of the increases in power of the chips, that they might reach metal melting temperatures by the end of the decade. What is the sense of creating very small chips if we need cooling systems that will fill the inside of a computer case? Then too, a cooling system failure might be the cause of fires and such. How would we use our laptops under these circumstances or would they just have to use different, slower chips to avoid the heat problems? There are many companies working on solutions to this problem. IBM is experimenting with carbon nanotubes. They are weaving them one atom at a time into transistors. One hundred thousand carbon nanotubes lined up side by side are only the width of a human hair. It is felt that using them in chip making will allow great increases in the number of transistors on a chip, lower the heat and make the process of manufacturing computer chips much cheaper. This is necessary because today it costs over two billion to construct a fabrication plant and it is predicted that the price will rise to ten billion by the next decade. Chips will get smaller and this is great for some devices, but for other things we may have already gotten to a size as small as we would want. Remember that not everything is dependent on the size of its chip. An example of this might be a transistor radio. We wouldn't want to make it so small that we couldn't comfortably hold it in our hand to see the stations, same with a MP3 player. Would you want an electronic watch that you would need a magnifying glass to see? What about things like refrigerators? You can't make these things too small or you wouldn't be able to fit enough food in them. When you have something that has to be at least a certain size, it won't matter if you have a chip in it that is microscopic, or an inch across. What this may mean is that older chips will be able to be used and they will be much cheaper. This will keep the cost of the device down. This can be seen right now, as a matter of fact. I think that everyone is familiar with greeting cards that play a song when opened. This would have been impossible to do before the dawn of the computer age. Look at the prices of electronic watches and calculators. It is possible to buy these things for one dollar or less. Remember the watch has a chip in it and a battery. Can you imagine how cheap they are to make? Nanotubes may have other advantages in chips. They are over 10 times stronger than steel and radiation resistant. Everyone knows that nanotubes will improve computer chips, but no one can figure out how to manufacture them yet, while keeping the cost at the same level as today's chips. This is a hurdle that will no doubt be surmounted as time goes by. Making things harder for the scientists is the fact that there are all different types of nanotubes. So why do current chips get hot? This is how things work, as the circuits get narrower because of shrinking chip size, the power has to be boosted by increasing voltage so the same amount of current gets through. The more you do this the more the chips leak. The reason for the leaks are the chips have fewer atoms for insulation to keep electrons in place. Even with today's technology we have been able to create some chips that are the size of a grain of sand. New tracking chips called RFID tags are now available and they measure only 0.05 x 0.05 millimeters. One of these babies could be placed on your person and you would never know it. It could even be put into your food and you would ingest it. It has been suggested that these things could be sprinkled on an area where a protest is going to take place and unsuspecting protesters would walk through not realizing that some of these things have stuck to them. Have we decided that nanotubes are the way to go, to shrink computer chip size? Yes? Don't answer too soon, there is another material that looks like a player in the race to shrink and that is Graphene. Graphene is the thinnest material known to man. There is an ongoing program at the University of Manchester to perfect this material and use it in chip manufacture. So how small could you make chips using this material? This is going to sound crazy, but the new transistors would be only one atom thick and 50 atoms or less wide. You just can't get much smaller than that on the thickness unless science finds smaller units of material than the atom, in the future. Graphene is a type of carbon that is only one atom thick. The team at Manchester has already created fully functional Graphene transistors. These transistors can still function at one tenth of the size limit for silicon transistors. So how small can we go? It seems that one atom thick and about 50 atoms wide is the limit for now. Can you imagine, we could have a small cup that contained billions of these devices. If we could manufacture them cheap enough, computers may become sub $100 devices. If we have enough of these tiny chips, might it be possible to outdo a quantum computer just by sheer force of computing power using millions of chips? I can't wait to see what happens.
|
