In the post-Moore era, the research and development of advanced processes has become increasingly difficult. Even the world's top wafer foundry giants like TSMC and Samsung have frequently reported news of R&D obstacles in the research and development of 3nm processes.

Text/Youyu Review/Zhang Ziyang Correction/Zhiqiu

In the post-Moore era, the research and development of advanced processes has become increasingly difficult. Even the world's top wafer foundry giants such as TSMC and Samsung , there are frequent news of R&D obstacles in the research and development of 3nm processes.

If nothing unexpected happens, 1nm will become a watershed and other semiconductor technologies are required to achieve it. Therefore, in order to maintain its technological leadership, TSMC and Samsung are both developing new technologies.

As we all know, TSMC is the most advanced in the global foundry field, and its profitability and R&D capabilities are very strong. Many people believe that TSMC will always remain ahead.

However, just recently, good news came from Samsung that it joined forces with IBM to break through the 1nm technical bottleneck.

December 14 EETOP news, at the recent IEDM 2021 International Electronic Components Conference, Samsung and IBM jointly announced a chip design technology called "Vertical Transistor (VTFET).

According to IBM and Samsung, the working principle of vertical transmission effect transistors is that stacks transistors vertically to increase the number of transistors stacking density, while remembering the vertical flow of current to improve the efficiency of power use.

Most importantly, IBM and Samsung said that can break through the current dilemma in 1nm process design with the help of vertical transmission effect transistor technology.

You should know that the number of transistors that can be accommodated per unit area of ​​a wafer is now approaching the physical limit of the mainstream semiconductor material, silicon, and the chip efficiency can no longer be significantly improved year by year. The semiconductor industry urgently needs new technologies. Therefore, IBM and Samsung's claim that the new technology they developed can break through the 1nm technology bottleneck really shocked the industry.

Given that IBM previously released the world's first 2nm chip manufacturing technology, its strength has been proven, the vertical transmission effect transistor technology jointly developed by IBM and Samsung is also worth looking forward to.

However, now that Samsung and IBM have joined forces, does this mean that TSMC will be overtaken by Samsung in the future? The author believes that there is no answer to this question yet.

If Samsung can enter the mass production stage of vertical transmission effect transistor technology in the short term, this will indeed be a heavy blow to TSMC.

But it should be noted that neither Samsung nor IBM disclosed when this technology will be applied to the actual product . Considering that IBM's 2nm process is still in the laboratory stage, it may take a long time for the commercial use of vertical transmission effect transistor technology, and during this period there must be many technical problems to be overcome.

In addition, TSMC is not waiting to die and is actively looking for ways to break Moore's law . However, unlike IBM and Samsung, TSMC seems to be targeting new materials.

In May this year, TSMC announced that it jointly conducted research with Taiwan University and MIT and found that two-dimensional material combined with semi-metal bismuth (Bi) can reach extremely low resistance, close to the limit of quantum . This also means that by using the bismuth metal properties, the production limit of 1nm process can be broken and the process technology can be lower than 1 nanometer.

Moreover, according to Professor Wu Zhiyi of the Department of Electrical and Optoelectronics of National Taiwan University, it was found that the effectiveness of two-dimensional material transistors is comparable to that of current silicon-based semiconductors.

At present, as Moore's Law approaches its limit, many semiconductor companies around the world are working hard to develop, hoping to take the lead in winning the commanding heights of the process technology layout. The author believes that it is impossible to draw a conclusion for the time being, which will win between Samsung and TSMC.