20 years of Photovoltaic in China: a record of Rivers and lakes (4)

In 1951, thanks to a large number of defense electronics contracts from the United States Army and Navy during World War II (after the United States participated in World War II, national industrial production was mobilized to produce one light aircraft carrier a week). Texas Instruments was born. In 1952, Texas Instruments bought the patent certificate for the production of transistors, began to produce and research transistors, and soon successfully produced the world's first commercial silicon transistor.

To paraphrase the idiom of reviewing the industry every year: "2020 is an extraordinary year" for the polysilicon-photovoltaic industry. Unwittingly, in retrospect, I found that I had been writing about this "extraordinary" for 10 years. From full-time photovoltaic journalists to employees of engineering companies investing in polysilicon and photovoltaic industries, the polysilicon-photovoltaic industry has always affected my thoughts over the past 10 years. This "extraordinary" is by no means a simple habit of writing and copying, but from the beginning of this century to the present, China's polysilicon-photovoltaic industry has experienced changes and ups and downs in a short period of 20 years. It has condensed the development trajectory of other industries for decades or even hundreds of years.

"when we go forward, we must not forget the road we have traveled; no matter how far we go, no matter how glorious the future is, we must not forget the past we have passed, nor can we forget why we started." At the key point at the end of the 13th five-year Plan and the beginning of the 14th five-year Plan, standing at the intersection of "two hundred years", I look back at the development of the industry, hoping to inherit the history with a pen between the vicissitudes and successes, and save up the strength to move forward steadily in the next section of the road.

As a souvenir, as a wish.

First, the origin: the birth of photovoltaic.

(19th century to 1960s)

5. Texas instrument

In 1951, thanks to a large number of defense electronics contracts from the United States Army and Navy during World War II (after the United States participated in World War II, national industrial production was mobilized to produce one light aircraft carrier a week). Texas Instruments was born. In 1952, Texas Instruments bought the patent certificate for the production of transistors, began to produce and research transistors, and soon successfully produced the world's first commercial silicon transistor.

The above-mentioned Tier, after leaving Bell Labs, joined Texas Instruments and became a "tough player" in the development of the industry. The biggest technical obstacle to the development of PN junction germanium transistors is the lack of enough pure and uniform semiconductor "germanium" materials, which greatly restricts the development of junction transistors. Therefore, Thiel put forward the idea of semiconductor single crystal: first to produce a large "germanium single crystal", it is possible to doping P or N junctions. But his proposal was not supported. In a mockery, he and his lab technicians built a crude device capable of vigilant growth, and used Tchaikovsky's CZ method to try to grow monocrystalline silicon, suspending a small "seed"-germanium crystal in a molten germanium crucible-and slowly "pulling it out" to form a long, narrow germanium single crystal. After the successful manufacture of germanium single crystal, the development of PN junction transistor is like taking things out of the bag. Chemist Morgan Sparks took the lead in using Gordon Ktier's method for the first time. He dripped tiny impurities into the molten germanium during crystal growth, and finally formed a very simple Pmurn junction, which is also the first PN junction in the global semiconductor industry.

Thiel then put the implementation into the NPN junction transistor mentioned above. In April 1950, in Bell Labs, the two men added "two continuous balls" to the molten germanium, one with P-shaped impurities and the other with N-shaped impurities, resulting in a thin base layer-NPN structure in the inner layer. In 1951, the first batch of PN junction transistors M1752 in Bell Laboratories began to be commercialized.

On Feb. 14, 1954, Gordon Ktier of Texas Instruments also independently successfully developed the world's first commercial silicon transistor, and on May 10th of that year, he presented his research findings-"recent development of silicon and germanium materials and equipment" (SomeRecent Developments in Silicon and Germanium Materials and Devices) at the IRE National Avionics Conference held in Dayton, Ohio. : IRE (, the predecessor of today's IEEE (American Society of Electrical and Electronic Engineers), is an international association of electronic technology and information science engineers and one of the largest professional and technical organizations in the world, with more than 300,000 members from more than 170 countries. IEEE was formed on January 1, 1963 by the merger of the American Society of Radio Engineers (IRE, founded in 1912) and the American Society of Electrical Engineers (AIEE, founded in 1884). Professor Terman of Stanford University and his student Hewlett, one of the founders of Hewlett-Packard, are both chairmen of IRE. We are familiar with the standards developed by the IEEE 802 committee, including the IEEE 802.11F wireless LAN specification. The essay would like to add that many people mispronounce IEEE and habitually pronounce it as "Eye-triple-E", but the correct way to pronounce it should be "Imure".

This IRE conference in 1954 is enough to go down in the history of the global semiconductor "silicon cycle". In May of that year, Thiel represented Texas Instruments at the National Conference of Radio Engineers in Ohio. In the previous few hours' report, various manufacturers complained that the production of silicon transistors was facing many technical problems, and many people even pessimistically thought that the research of silicon transistors might come to a dead end.

Finally, it was Texas Instruments (Ti) 's turn to speak. Thiel first echoed, "germanium transistors are so unstable, and the development and production of silicon transistors are so difficult." but, Thiel finally turned, "I can tell you responsibly that Ti has solved the problem of mass production of commercial silicon transistors!"

Thiel's voice immediately shocked the audience, and the delegates in the audience immediately became agitated, with someone shouting, "are you saying that Ti has already produced commercial-grade silicon transistors?"

"Yes, we already have three models of silicon transistors on the production line." Thiel carelessly took a transistor out of his pocket. "I happen to have one with me today. I can show you."

Thiel took out a modified tape recorder, first connected to a germanium transistor amplifier, after playing a piece of music, dipped the germanium transistor into a cup of hot oil, and the music stopped abruptly. Thiel then replaced the germanium transistor with Ti's silicon transistor, which was also immersed in hot oil, and the music continued to play. There was thunderous applause from the audience, and the whole meeting was boiling up.

The wonderful release of the world's first commercial silicon transistor shocked the entire semiconductor industry, and made Texas Instruments, a "newcomer" in the semiconductor field, a star and leader in the industry.

In May of that year, Texas Instruments quickly turned Gordon Ktier's research results into productivity. The first batch of silicon transistors completed production and trial production, and were soon successfully applied to the US military department, and it was almost completely monopolized. From 1955 to 1958, Texas Instruments received generous "technology dividend" returns. Texas Instruments easily won the "first bucket of gold" in the field of semiconductors. It was not until the founding of Silicon Valley that Texas Instruments ushered in its biggest rival in the semiconductor industry, the leader of the Fairy Child system, the American Fairchild Semiconductor Company.

In the following three years, Texas Instruments achieved a monopoly in the American market, and the time window of the "silicon" transistor market enabled Texas Instruments to make the biggest profit since its inception, which is the charm of the technology dividend. During this period, Thomas Watson Jr., the second-generation head of IBM, bought 200 transistor radios and gave them to senior IBM executives, telling them: "if this company's products are good at radios, then they should also be good at our computers." Soon, IBM became one of Texas Instruments' biggest "big customers" since its inception, using "silicon" transistors to replace vacuum tubes in IBM mainframe computers.

Tannibarn, mentioned in the previous chapter, failed to continue his research in January 1954 and received attention here at Texas Instruments: thanks further to DuPont's high-purity silicon, a silicon transistor, a N-P-N transistor with growth junction technology, was successfully developed on April 14, 1954.

In 1956, Texas Instruments' sales reached $45.7 million, an increase of 59% over the previous year. Although the annual sales of 12.5 million "silicon" transistors are negligible compared with the annual sales of billions of vacuum tubes, it is clear that this is a growing business, and it is a visible "blue ocean". Under the guidance and stimulation of Texas Instruments'"silicon" transistors, from 1957 to 1958, the market demand doubled, and the world market demand for transistors increased from 12 million to 25 million. The sales value of the whole industry was about 228 million US dollars in 1958 and more than 400 million US dollars in 1960. Transistors are replacing vacuum tubes in almost every field and every aspect. And what about Texas Instruments itself? By 1956, the company's sales had reached $45.7 million, an increase of 59% over the previous year. Although the annual sales of 12.5 million transistors are negligible compared with the annual sales of billions of vacuum tubes, it is clear that this is a fast-growing business. In the following three years, Texas Instruments continued to grow at an alarming rate, with annual sales growing by an average of 61% and annual profits by an average of 73%. By 1959, the company reported sales of $193 million, more than double that of the previous year, and annual profit increased by 136% to $14 million.

In January 1959, an article in Forbes magazine enthusiastically introduced Texas Instruments: "since entering the electronics industry, it can now be called the number one manufacturer of transistors in the country. This small electronic device can do vacuum tubes, forming the fastest growing field in the United States."

All kinds of mainstream financial media in the United States continue to exclaim: "since 1950, Texas Instruments' sales have increased 12 times, in 1958 it is expected to reach 90 million US dollars, and its profits have increased more than 10 times. Last year, it reached about 5.2 million US dollars."

All kinds of Wall Street commentators are after it: "as seen in the stock market, Texas Instruments' stock has soared from $73 a share to nearly $250 in 12 months."

Analysts at major investment banks have been unable to express their admiration: "over the years, Haggerty has captured and won the attention of Wall Street securities analysts." what fascinates analysts most about Haggerty is that you only need to repeat a little simple arithmetic. Numbers have never had such a wonderful magic since Pythagoras. "

In November 1961, Fortune magazine summed up the successful experience of Texas Instruments: "Silicon transistors are the turning point in the development of Texas Instruments (Ti). The sales of Ti are almost vertical, and the company suddenly becomes one of the big companies."