The Millennium Semiconductor Survival Guide.

Chapter 5 Discoveries that Connect the Past and the Future

Zhou Xin was a little surprised that he would receive a reply from Hu Zhengming the next morning.

You must know that after 20 years, Hu Zhengming will check his email once a week.

"Could it be that the teacher in his youth read his email once a day?"

"I can't see that the teachers back then worked so hard."

Although Zhou Xin was with Hu Zhengming before, at that age, Hu Zhengming mostly just gave some directional suggestions.

Of course, even so, in 2016, when Hu Zhengming was 69 years old, he still published papers in science that were cited more than a thousand times.

"Professor Hu, hello, I'm Zhou Xin, a sophomore majoring in microelectronics at Yanda University. Although Zhou is old, he wants to restore Zhou Xin."

Cross-border calls are very expensive, and Zhou Xin didn't go to open source after he came.

Just relying on scholarships is still quite stretched.

"Zhou Xin, what a good name.

I thought it was the core of the chip.

I read the email you sent me, and I thought that some professor had a new idea on the MOSFET model.

Your results are good enough to be published in IEEE or even Science. "

The two talked in English. Even though Zhou was an old country, he ordered the restoration, Zhou Xin also spoke in English.

After all, when he went to Berkeley to study for a Ph.D., he specially translated this sentence into English and said it when introducing himself to foreigners.

Zhou was a dynasty of the Hua Kingdom thousands of years ago. Adding this sentence instantly made the Americans who have no history look differently.

This can be regarded as a little trick for Zhou Xin to quickly get acquainted with foreigners abroad.

Talking about history and politics is one of the ways to quickly get closer to another strange man.

"Yes, I am currently a sophomore majoring in microelectronics at Yenching University, and the reason why I sent this email is that I hope to study for your Ph.D.

Because my financial situation is not very good, I still need you to provide a full or half bonus. "

The difference between the full award and the half award in this era is that the half award can get less money, and at the same time, the half award requires work, such as substituting for classes and correcting homework.

The full award will be required by the tutor to work, but you can choose to refuse.

Generally it will not be rejected.

"International phone calls are a bit expensive for me, so I will express my wishes directly."

After Zhou Xin finished speaking, Hu Zhengming paused for a moment and asked:

"I will send you a paper later, and you have five hours to answer.

Reply to me by e-mail after you answer.

If there are [-] points, I will help you arrange everything.

This paper will not be too difficult, it is just the standard of the entrance examination for doctoral students of the Department of Electrical Engineering at Berkeley.

Although it will be a bit difficult for ordinary sophomores.

But your keen intuition and handling of the model shown in the email are not ordinary sophomores.

Even many doctoral students I have supervised are not as capable as you in this area when they graduate. "

Hu Zhengming did not set too high a threshold for Zhou Xin, the difficulty of the doctoral entrance examination.

Of course, for sophomores in Huaguo, no one else could do it except Zhou Xin.

This is not a gap in level, but a gap in all aspects.

Regardless of the textbooks, the level of teachers, the depth of learning, etc., there are thick barriers between the sophomore and doctoral entrance exams.

Not to mention answering in English.

"What if I fail the exam?" Zhou Xin asked rhetorically on the phone.

Hu Zhengming smiled: "As long as you can prove that the email was written by yourself.

Then I will also help you with school transfer and scholarship.

It's just that you need to come to Berkeley to complete the courses that you didn't complete in your undergraduate course. "

As a godfather in the semiconductor industry, he has stayed in Berkeley for more than 20 years and wants to help students get scholarships. It is no exaggeration to describe it as a breeze.

Hu Zhengming admired Zhou Xin very much, not only because of the email, but also because of the frankness shown in the communication and his fluent English.

Even in some modal particles are the same as him.

During Zhou Xin’s stay in American, one of his main communication partners was Hu Zhengming, and his spoken language mainly improved by leaps and bounds during those years in American.

There are of course similarities between the two in terms of oral expression.

Zhou Xin smiled on the other end of the phone: "Okay."

"The MOSFET model can relate Em to all device parameters and bias voltages. Describes its use in explaining and guiding thermal electron scaling. How did you come up with the idea of ​​modeling MOSFET interconnects through the predictive nature of circuit simulations? "

The distance between the two ends of a telephone line spanning thousands of kilometers is not only geographical, but also temporal.

The answer that Zhou Xin sent to Hu Zhengming was Hu Zhengming's own paper in 2000, which was published in the Proceedings of the IEEE Conference on Integrated Circuits in 2000, and ranked eighth among Hu Zhengming's more than [-] papers.

Although the ranking is not very high, it has played a link between the past and the future.

Hu Zhengming's greatest contribution is to develop and optimize the 2D structure of semiconductors into a 3D structure, which is FinFET.

From 1960 until around 2010, the basic planar (2D) MOSFET structure remained unchanged until further increases in transistor density and reductions in device power consumption became impossible.

Zhengming Hu's lab at UC Berkeley saw this as early as 1995.

As the first 3D MOSFET, FinFET transforms a flat and wide transistor structure into a tall and narrow transistor structure.The benefit is better performance in a smaller footprint, as is the advantage of multi-story buildings over single-story buildings in congested cities.

FinFETs are also known as thin-body MOSFETs, a concept that continues to guide the development of new devices.

It stems from the realization that current does not leak through transistors within a few nanometers of the silicon surface because the surface potential there is well controlled by the gate voltage.

FinFETs keep this thin body concept in mind.The body of the device is a vertical silicon fin capped by an oxide insulator and gate metal, leaving no silicon outside of strong gate control. FinFETs reduce leakage currents by orders of magnitude and lower transistor operating voltages.It also points to a path for further improvement: further reductions in thickness.

Current does not leak through transistors within a few nanometers of the silicon surface because the surface potential there is well-controlled by the gate voltage, a concept that was replicated in the laboratory by interconnect modeling for MOSFETs.

It is impossible for Zhou Xin to tell Hu Zhengming that you discovered this by yourself.

However, since Zhou Xin has done intensive reading of Hu Zhengming's most important papers, he has his own analysis of how he thought at that time.

These analyzes and Lao Hu's exchange process 20 years later have also been recognized by the other party.

(End of this chapter)