The Age of Information
Although there have been many technological advancements that have contributed to the development of the modern computing world, few have been as noticeably influential as the transistors. They are the brain/processing power behind everyday items from smartphones, to smoke alarms. As Michael Riordan eloquently put in his article,
"The Incredible Shrinking Transistor" the modern transistors is
"Little more than an abstract physical principle imprinted innumerable times on narrow slivers of silicon-millions of microscopic ripples on a shimmering crystal sea."
It seems physically impossible that the iPhone 14 has 15.8 billion transistors, each with a diameter of five nanometers, contained within its six inch by three inch frame. The smallest modern commercial transistors are two nanometers and over 18000 times smaller than the first transistor invented over 7 decades ago. That would be like comparing the length of three cruise ships lined up to a cricket. The transistor dimensions have halved in size every two years since their creation; a phenomenon we refer to as Moore’s law, after Gordon Moore pointed out the trend in 1965. The resulting outcome of this trend is a world where one can experience the advancements in processing power in the palm of one's hand.
But how did this technological renaissance come to pass?
The Problem
In the 1930's, American Telephone and Telegraph AT&T was having trouble expanding their business due to limitations with triodes. At the time, the triode, which consisted of three electrodes embedded within a vacuum tube, was the best technology available for amplifying electrical signals long distance. However, they were prone to overheat, and used too much power. A better device was needed for AT&T's continued expansion of telephone lines across the country. The solution was found within a group of materials that all shared a unique property of being neither a conductor like copper, or an insulator like Styrofoam, but somewhere in-between. This unique property, allows these semiconductors to modify their conductivity by applying a charge.
The transistor was developed in 1947 at Bell Laboratories, which was the Research and Development arm of AT&T. The device used to test the first transistor was about the size of a light bulb. It was created with a germanium crystal sitting on a power source, and a small plastic triangle wrapped in gold lightly touching the germanium surface. Using a thin razor, a slit was cut in the middle of the triangle to create a gap of at most .002 inches. Then the triangle was balanced over the germanium, so that the two gold points are in contact with the germanium surface.
How it Works
When the current being fed in through the first contact passes a certain voltage threshold. This causes a stronger output current to exit through the second gold contact. At the subatomic level, one can start to understand how the properties of the semiconductor are what gives this effect. Semiconductors have the ability to collect and release electrons from their outer-most ring. When an electrical charge exceeds the threshold, the electrons are actually knocking into those free floating electrons in the germanium's outer-most ring. This causes them to break away from the germanium and exit through the second gold contact producing a stronger current. However, if the current passed through the first contact doesn't exceed the voltage threshold, then there will be no output current exiting the second gold contact. This is because germanium also has the ability to loosely hold onto electrons in it's outermost ring, preventing them from exiting through the second contact. This less than a hair-length gap on the germanium between the two gold points is effectively a switch.
Shrinking the Transistor
The science team at Bell Labs had finally found a way to amplify electrical signals more reliably and with less power use than with triodes. This allowed for the expansion of AT&T's phone lines across the country, which was the original goal for developing the transistor. However, the true potential for the transistor in its use for transferring and storing information was yet to be realized.
Several early computers were already around by the 1930's. Giant hulking monstrosities packed full of vacuum tubes and triodes. Around the time the transistor was invented, computers like the UNIVAC, were already being commercially sold. However, only businesses and government entities purchased the UNIVAC. It was probably difficult for any other interested buyers to find a place to put an eight-ton, 14-foot long, glorified calculator.
Nonetheless, humanity was content with their vacuum tube computers. After all, they had helped us decipher the NAZI's code and win WWII. Why fiddle with that little gold wrapped triangle thingy on a crystal, when one can simply go buy some vacuum tubes if they want to make a computer. But some saw the true potential of the little gold wrapped triangle thingy. Geoffrey Dummer believed we could improve upon the transistor by effectively having multiple transistors in one unit. He suggested that instead of attempting to make the connections with wires, it could be possible to embed electronic components in a semiconducting material. Dummer theorized that by creating tiny spaces in the layers of the semiconducting material, several transistors could be on one integrated circuit.
It soon became clear that shrinking computer components was feasible as well as a lucrative venture. The US Department of Defense saw it as crucial that we find ways to miniaturize electronic components to help in the ongoing Cold War with USSR. Millions of dollars in grants were invested by the Department of Defense to spur the research and development process. It wasn't until the late 1950's when an integrated circuit housing a few transistors in one unit was developed, that humanity began to see the true potential for the transistor.
The number of transistors that could fit in an integrated circuit continued to increase, from hundreds of transistors in the mid 1960's, to thousands of transistors by the early 1970's.
Computers finally became more mainstream in our society in the mid 1970's. The components required to make mid to large scale integrated circuits were finally cheap enough to mass produce, and soon everyone was seeing the benefits of increased processing power. We were able to send humans to the moon, we converted to an automated modern banking system, and the first personal computers were starting to hit the market.
We continued to develop newer and more effective techniques for shrinking the dimensions of the transistor. By the mid to late 1980's companies were mass producing computer chips with millions of transistors.
The Limit of Transistors
As humanity entered the next millennia, it seemed as though we were fast approaching the limits of physics itself in our pursuit to have more processing power. As transistors shrink to single digit nanometers, electrical charges from transistors bleed over to transistors within close proximity.
Even given these limitations, the development of new techniques as well as new materials has lead a team in China to create an even smaller transistor. At just one third a nanometer it is literally a single layer of carbon atoms. It seems hard to fathom going smaller than a carbon atom's width, but chances are, there is already a team of scientists figuring out the next path forward.
Sources:
Transistorized!
What is a Transistor
History of computers: A brief timeline
How to squeeze billions of transistors onto a computer chip
Top comments (0)