Improving The Performance Of Silicon Chips Over The Past 50 Years Has Followed A Special Rule Called Moore’s Law; But What Is Another Way To Improve Silicon Chips?
The advent of the semiconductor silicon chip revolutionized the electronic world and the computerization of life, marking the twenty-first century’s first years. Integrated circuits, or ICs, form the basis of the entire digital world around us, controlling our systems and allowing us to access and share information in the blink of an eye.
The development of silicon transistors since the prototype in 1947 has been very rapid. The total number of transistors on a primitive integrated circuit chip has increased from a few thousand to more than two billion. According to Moore’s Law, which is still valid after 50 years, the density of transistors will double in both years.
However, these components face a significant challenge:
the last integrated circuit produced has a 5-nanometer architecture that measures between a blood cell (7,500 nanometers) and a diode string (2.5 nanometers). Also, the size of single silicon atoms (about 0.2 nm) in integrated circuits with a width of one bit will create a problematic physical constraint.
On the other hand, it will destroy the stability of its behavior and performance and make it difficult to control. So what is the solution? Is reducing the size of silicon chips the only way to improve their performance?
It should note that the inability to reduce the size of the ICs will interrupt the growth process of silicon components. It is inevitable to reconsider the way devices are manufactured or even offer alternatives to silicon. In this article, we will explore other ways to improve the performance of silicon chips besides reducing their size.
Problems with the use of silicon in the chip; Speed, heat, and light
To understand the challenge ahead, we must first understand why silicon became the best raw material for electronic components. In addition to the advantages such as abundant access to this material and its easy extraction and good physical properties, the presence of stable native oxide makes it a good insulator. It has all the necessary properties for use in the chip. Of course, this article also has its drawbacks.
For example, one of the significant benefits of increasing the composition of more transistors on a chip is that it can speed up data processing in an integrated circuit; However, this increase in velocity is primarily related to the degree to which electrons can move and move inside a semiconductor material, called electron mobility.
Although the electrons in silicon are fully mobile, they are much less mobile than other semiconductor materials such as gallium arsenide, indium arsenide, and indium antimonide. So the first problem with using silicon is the low excitability of the electrons.
However, not all the helpful conductivity properties of semiconductors are limited to the mobility of electrons, and the movement of so-called electron holes also affects the conductivity and speed of information transfer. The electron-hole is the remaining gaps in the network of electrons that, as the electrons move, also move around the nucleus, increasing the conductivity.
Today’s ICs use the CMOS technique, or complementary oxide-metal semiconductor, and use a pair of transistors, one of which uses the electrons to move and the other the electron holes. Still, the drift of electron holes in silicon is so poor that it stands in the way of higher performance, forcing manufacturers to use germanium alongside silicon for several years to speed up drift.
Another problem with silicon is the strange reduction in its performance at high temperatures.
Modern ICs, with billions of transistors, generate considerable heat, much effort being expended by cooling them with fans and heat sinks (for example, desktop computer processors that are equipped with fans).
In contrast, semiconductors such as gallium nitride (GaN) and silicon carbide or silicon carbide (SiC) perform better at high temperatures and can operate faster; Therefore, in high-consumption electronic amplifiers, they have been replaced by silicon to reduce their performance.
The last major problem with silicon is its poor light transmission. While lasers, electrodes, and other photonic devices are less complex, they use silicon replacement semiconductor compounds. As a result, this led to two distinct industries: silicon in the electronic structure and the semiconductor composition in the photonic form.
This situation continued for years, But in recent years, a lot of pressure has been put on a chip to combine electronics and photons, which has become a significant problem for manufacturers.
Change the silicon instead of changing the size.
Manufacturers tried different ways to improve the performance of silicon chips instead of reducing the nanometer size of the chip architecture according to Moore’s law and finally decided to change the way they use silicon. In this regard, various materials for combination with silicon to improve performance were tested, and among them, finally, three selected materials were.
The first material was to eliminate the weakness of silicon in the drift of electron holes. A small amount of germanium has already been added to the chips to improve this problem, But increasing the amount or even using all-germanium transistors can improve the situation.
Germanium was the first material used in semiconductors; Therefore, reusing it is precisely a return to the future. However, rebuilding the large germanium semiconductor industry and removing silicon will be very difficult and costly for manufacturers.
The second material relates to the subject of metal oxides.
Silicon dioxide has been used in semiconductors for years, But as the silicon dioxide layer shrunk, it became so thin that it lost its insulating properties, making transistors almost unreliable.
After much research, the rare substance hafnium dioxide (HfO2) began to be used as alternative insulation. Of course, despite the move towards using this rare material, research is still ongoing on alternatives with better insulation properties.
Next, and to solve the problem of using silicon, the most interesting may be composite semiconductors III-V, especially samples containing indium, such as indium arsenide and indium antimonide. These semiconductors offer electron excitability 50 times greater than silicon and, when combined with germanium-enriched transistors, will be able to increase speed dramatically.
However, not everything is going as smoothly as it seems. Silicon, germanium, oxides, and III-V materials all have crystalline structures whose properties depend on crystalline integrity.
Therefore, they can not be easily combined with silicone and expect the best performance from them. Thus, solving the problem of crystal lattice mismatch became the main challenge of this technology. Various ways were examined, which eventually led to better use of silicon.
Types of silicone
Despite silicon’s limitations, silicon electronics have proven compatible, reliable, and cost-effective in the mass-market device market. Thus, contrary to the notion of the “end of the silicon age” or the unrealistic promises of alternative materials, silicon has proven to be the best choice and is strongly supported by the global semiconductor industry and will not be abandoned at least for the rest of our lives.
Instead, advances in electronics will lead to improved performance of silicon through integration with other materials. Companies such as IBM, Intel, and university labs worldwide have spent time and effort meeting this challenge. Research results promise a hybrid approach that combines materials III-V, silicon, and germanium can achieve in minutes.
Conquer the market year.
Compound semiconductors have already shown their power in essential areas such as lasers, lamps, LED displays, and solar panels, in which silicon has nothing to say. But, in the future, we will need more advanced combinations because electronic devices are getting smaller and less energy-consuming every moment. In addition, high-consumption electronic devices are exacerbating the need for advanced compounds due to their properties beyond silicon capacity.
The future of the electronics world is very bright. Although much of it will still be owed to silicon, this time, silicon, in combination with other materials, will have different types to provide more improved properties.
What do you think about the future of chips?