The world of electronics seems doomed to undergo a change on a scale comparable to that which occurred just over seventy years ago when transistors broke in to end the reign of vacuum tubes. Semiconductor manufacturers are finding it increasingly difficult to continue to improve their manufacturing technology because each step they take brings them closer to the physical limit imposed by silicon. But, fortunately, it seems that we are brushing with the tip of our fingers the solution to this problem.
Sungsik Lee, a professor of electronic engineering at the National University of Pusan, in South Korea, and a former researcher at the University of Cambridge, in the United Kingdom, has published research in which he theoretically describes a new type of device electronic capable of carrying out the inverse function of a transistor.
But what is really interesting is that in his study he argues that these "inverse transistors" will allow us to manufacture simpler, faster-integrated circuits with lower consumption, which is why it is postulated as the technology that could avoid semiconductor stagnation. So much so that in his article, which has been published in the IEEE journal and has been echoed in the MIT magazine and the repository of Cornell University (United States), Lee speaks of a "new paradigm" of the world of The electronic.
What is a reverse transistor and what does this technology promise us?
Before we see what a reverse transistor is, it is good for us to review what a conventional transistor is. Currently, we can find these tiny elements in practically any integrated circuit that we can imagine: microprocessors, power amplifiers, switches, rectifiers, oscillators ... And this in practice means that they reside inside our computers, smartphones, tablets, stereos, televisions, radios, cars, medical equipment and a host of other devices.
Although their precursors are even older, the first transistors as we know them today were invented in 1947 by John Bardeen, William Shockley, and Walter Brattain, three physicists at Bell Labs. A simple way to define a transistor invites us to describe it as a semiconductor electronic device that is capable of responding to an input signal by giving us a specific output. An electronic amplifier, for example, will increase the power, voltage, or current of the signal that we place at its input at its output, resorting, of course, to an external power source.
The first transistors were invented in 1947 by John Bardeen, William Shockley, and Walter Brattain, three physicists at Bell Laboratories.
There are several types of transistors (bipolar, point contact, field-effect, union, single electron, phototransistors, organic electrochemical, etc.), but fortunately, we do not need to delve into them to understand what reverse transistors are, which is what really interests us in this article. It is enough for us to know two more facts about transistors. On the one hand, they are active elements within integrated circuits. And, furthermore, those that have allowed us to reach the level of integration used by current lithographic techniques are field effect (FET).
 |
| This design shows the possible structure of a 'transistor' as proposed by Sungsik Lee in his research. |
Unlike capacitors, coils, or resistors, which are passive elements, transistors are active components of a circuit because they either exert a control function over its behavior, or they introduce a certain gain because they act in a non-functional way linear. This means that the relationship between the applied voltage and the current demanded by the circuit cannot be expressed using a constant value, which introduces a complexity that is not present in linear systems.
However, the behavior of capacitors, resistors, and coils, which, as we have seen, are passive elements in an electrical circuit, is clearly delimited and linear. In addition, they facilitate the connection of the active components and make possible the transfer of the electrical signal by means of storage in magnetic or electrical fields, or by means of the dissipation of electrical energy.
Capacitors, coils, and resistors are passive elements of an electrical circuit, while transistors are active elements.
On the other hand, about field-effect transistors (FETs) we are interested in knowing, without going into complex details, that they use the electric field to let current or not pass through a channel that carries a single type of electric charge. This type of transistors is what has made possible the integrated circuits that we currently use in our digital systems.
During the last decades, many researchers have made an effort to study the characteristics of the passive elements of electrical circuits with the intention of finding out if there are other components with different properties that can replace or complement them. Sungsik Lee, however, has undertaken a similar task, but with the active components. With transistors. And the result of their investigation is the 'transistors'.
 |
| This is the experiment described by Lee in his article to show that the introduction of 'transistors' in an integrated circuit would lead to a significant reduction in both the space it occupies and its consumption. |
The word 'transistor' describes quite precisely what one of these elements does: it takes an input signal and carries current or not at the output. We can imagine it as something like a resistor with variable capacity. In fact, the word 'transistor' comes from the English terms transfer (transfer) and resistor (resistance). What Lee has described is a device with characteristics similar to those of transistors, but, unlike these, capable of taking an input signal and carrying or not carrying voltage at the output. It is somewhat similar to a hypothetical capacitor with a variable energy storage capacity.
The term that Sungsik Lee proposes to identify reverse transistors is 'transistor'
The term proposed by Lee to identify the element he has devised is 'transistor' because its properties, as with the word 'transistor', can be condensed from the English terms transfer (transfer) and capacitor (capacitor). However, we should not think of 'transistors' as active elements designed to replace transistors, but rather as devices designed to coexist with them in the same circuit.
What is the point of using even more elements in our integrated circuits? It seems logical to think that introducing the 'transistors' without eliminating the transistors will increase the complexity, consumption, and size of integrated circuits. However, Lee assures that this is not the case because the introduction of 'transistors' entails the use of a smaller number of transistors. That is, according to this researcher, the key.
To demonstrate his theory, Lee proposes a simple example that illustrates quite clearly the advantages of using 'transceivers'. By combining a single 'transistor' and a single transistor we can make a voltage amplifier, but if we want to obtain it using only transistors we will have to use four of these elements. Just twice as much, which has a clear impact on the complexity, size, and consumption of the circuit.
We have just discussed the advantages of the introduction of 'transistors' in integrated circuits: lower consumption, less complexity, and less space. But there is still another important advantage that Lee also reflects in his research: the circuits that integrate 'transistors' are faster. And all this in practice should have a very clear impact on the devices that we use on a daily basis and inside which integrated circuits reside, such as our computers.
And it is that Sungsik Lee's thesis reflects that the introduction in our electronic devices of "transistor" integrated circuits instead of the traditional «transistor only» circuits should significantly reduce their consumption, their complexity, and their size, but increasing, at the same time, its performance. Hence the "paradigm shift" that this researcher talks about in his article.
What does "Moore's Law" paint in all this?
If we bear in mind everything that we have reviewed up to now and the attractive properties that 'transistors' have on paper are confirmed, it is not difficult to intuit that their introduction into current electronic technology would have a very positive effect. At this point, it comes to us that we have not even painted to recover the definition of Moore's Law that Norberto Mateos, the Manager of Intel Spain, proposed to us during the interview we did at the beginning of last August.
Circuits that combine transistors and 'transistors' will be simpler, faster and consume less, according to research by Sungsik Lee
According to Norberto, “Moore's Law, which was observed in 1965, and therefore is older than Intel itself, has been stated in different ways depending on the moment. During the 90s and the past, we talked about speed. Then we started talking about performance. In the end, what Moore's Law has established is an expectation about the electronics industry in general regarding the amount of advancement, functionalities and technology that we are able to bring to the market, and what motivates users to buy."
If, as Sungsik Lee argues, the combination of transistors and 'transistors' allows us to manufacture microprocessors, which are nothing other than extraordinarily complex integrated circuits, faster, simpler, and with a lower consumption than the current ones, it is Clearly, Moore's Law may be sane for a while. Of course, for this to be possible it is necessary that the reduction in complexity that this innovation promises be confirmed, so that it can have a direct and positive impact on the lithography techniques that we currently use.
The 'transistors' are not ready yet: this is the challenge we have to solve
For the moment, everything that Sungsik Lee proposes in his research is limited to the theoretical field and it is difficult to predict when we will be able to manufacture the first 'transistors'. The reason is quite strong. We know how they should work and what their properties will be. We also know what effect its introduction would have on the integrated circuits we design and manufacture today. But we still don't know how we can get them.
Lee proposes the possibility of manufacturing 'transistors' taking advantage of the well-known "Hall effect", which is a phenomenon that causes the generation of an electric field inside a conductor through which a current circulates when it is subjected to a magnetic field in a direction perpendicular to the movement of the loads. But there is a problem that has not yet been solved: Scientists do not know how to take advantage of this effect in CMOS circuits at the nano-scale.
 |
| In section A of this "Hall effect" diagram, we can see how the electron flow passes through the Hall probe, in the center of the diagram. In sections B and C the electric or magnetic fields are reversed. |
This is the challenge that must be solved so that the technology proposed by Lee in his research comes to fruition. If you want to know what this researcher stands for in more detail and English does not intimidate you, I suggest you take a look at the article he has published, in which he describes in great detail the physical foundations of his innovation. Clearly, there is still a lot to do, but at least having options with a solid foundation allows us to see the future of digital electronics with some optimism.
0 Comments