How to create semiconductor lasers. Part I

A laser is a potent laxative rubber-like resin, a laser-root derived from a plant.
Laser-root is a plant of the Lazapitium genus of the carrot family (umbrella).

(Webster’s Dictionary of 1939)


Laserpitlum latifolium (Lazarpitium broadleaf).

It all started with radio

Soon after the first demonstrations of Popov and Marconi in the years 1895-1896, the idea of ​​a detector receiver (the one with a single diode) appeared. Just a few years before, semiconductor diodes were invented. Then they were called crystal detectors - neither the concept of "semiconductor", nor the word "diode" have yet been invented. Moreover, no one understood why the crystal detector basically works - however, it was clear that the matter was in the special crystal magic contact between the crystal and the metal wire.
')

Crystal detector . The metal wire touches the semiconductor (on a round stand), forming a Schottky barrier. Because of the protruding wire, such a detector was called “cat's-whisker,” that is, “cat's whisker.”

Henry John Round worked on such diodes in the Marconi laboratory. In 1907, he noted that if voltage was applied to silicon carbide-based detectors, some of them began to glow. The glow is usually yellowish, but it can be green and even blue. In the years of the formation of the radio, everyone was a little ill at ease, therefore the Round was limited to a note in the magazine.

By the way, after more than a hundred years, craftsmen repeat this experience . It turns out very cool.

Quite often, the same new phenomenon is independently observed by different teams. It happened with the silicon carbide glow. The second time - in 20 years - Oleg Losev from the Nizhny Novgorod radio laboratory noticed him, examining the crystal detector under a microscope. Unlike Round, Losev tried to study the glow in more detail and try to understand its nature. Looking ahead, I would say that it was not possible to figure out the physics of luminescence - the band theory did not exist in principle, and quantum mechanics was only being created. Nevertheless, Losev demonstrated an unusually consistent and detailed scientific approach.


Henry Round (left), Oleg Losev .

First of all, it was necessary to understand the conditions under which the glow occurs. Since the main parameter in the diode operation is voltage, Losev measured the threshold voltage at which the detector began to glow. It turned out that this can occur with direct and reverse polarity of the applied voltage. (Today we know that the first is the injection luminescence mode, in which all modern LEDs and lasers work; the second is observed before irreversible diode breakdown and is called prebreakdown luminescence ).


CVC of Losev detector. The arrow indicates the beginning of the glow at about 8 V. [1]

The next question is the nature of the glow. What if a thin contact glows due to high temperature, like an incandescent bulb? Losev dropped a drop of benzene on contact and observed its evaporation; benzene evaporated very slowly. So it's not about temperature. Losev suggested that he was dealing with an “inverse photoelectric effect”: an electron, having accelerated in an electric field, flies into the contact area and is inhibited, and its energy goes to the generation of light. Perhaps it was the most logical explanation at the time: the effect was probably quantum, but there were still 10 years left to the band theory.

What else can be measured at the light source? Of course the spectrum! It turned out that it is clearly non-thermal and depends on the applied voltage. Nothing more could be said: there was no available theory. According to loffe, Losev wrote about this to Einstein, but he did not wait for an answer.

Finally, Losev showed an amazing response speed. He managed to modulate the radiation of a diode with a frequency of up to 78.5 kilohertz - his equipment simply did not allow it above. Losev made a far-reaching conclusion about the possibility of applying the effect to high-speed transmission of information, and also wrote a patent for a fast "light relay".

After the war

A quantum theory of solids has finally been developed. At the beginning of the 40s, the first pn junction was created at Bell Labs, and by 1948 the first transistor was created. Semiconductor physics is becoming more relevant than ever. Kurt Legovets, who recently emigrated to the USA from Germany, repeats Losev's experiments on higher-quality silicon carbide samples. In general, confirming Losev's results (measuring the same IV characteristics and emission spectra of diodes at different temperatures), Legovets and his colleagues point out the shortcomings of his physical model. Instead, they show why the pn junction emits light if a voltage is applied to it.


From the work of Legovec . The pn junction transmits a current, the electrons recombine with the holes, emitting light.

A year later, John Heinz from Bell Labs manufactures LEDs based on silicon and germanium, and soon clearly confirms the findings of Legovtse. True, the efficiency of LEDs is extremely low. The reason for all is the indirect location of Si and Ge (I mentioned this in the article about blue diodes ).

At the same time, other semiconductors, gallium arsenide (GaAs), gallium phosphide (GaP) and their solid solutions (GaAsP), which are direct-gap, begin to be investigated. The first LEDs based on them were demonstrated in 1962: an infrared diode on GaAs - Jacques Pankov from RCA; GaAsP red light diode - by Nick Holonyak of General Electrics.


From left to right: Kurt Legovec , Jacques Pankow , Nick Holonyak . Photos of John Heinz could not be found.

Lasers

The world entered a new era in 1954, when the first generators of coherent microwave radiation in ammonia — masers — were created. After 10 years, Basov, Prokhorov and Townes will receive the Nobel Prize for this, and shortly before that, in 1960, Meiman will demonstrate the first optical laser on a ruby ​​crystal. The semiconductor revolution will be followed by a laser revolution.

In classical lasers, we deal with energy levels in atoms or ions. You need at least two levels: first, we “throw” electrons onto the upper one, after which they return to the lower one, generating a laser pulse. But what if we use the valence band and the conduction band of a semiconductor as these two levels? This idea comes to mind to the pioneer of laser physics, Nikolai Basov, along with Oleg Krokhin and Yuri Popov from FIAN.



The most pleasant thing was that from the LED to the laser diode there was only one step left - the creation of an external resonator from two mirrors around the pn junction. In reality, everything turned out to be even simpler: instead of mirrors, the polished surface of the crystal could be used, since the internal reflection from the surface of the semiconductor is quite large.

For this reason, the first semiconductor GaAs laser was created just a few months after the first LED. The author of the work was Robert Hall from the same General Electric.

In the same 1962, Nick Holonyak, already known to us, made a GaAsP laser. With mirrors, he did even more cunning. The fact is that high-quality crystals break very easily along the crystal axes, and the cleavage surface turns out to be very smooth (ideally, almost atomically smooth). Holonyak simply chipped the edges of the crystal from two sides and thus turned it into a laser.


From left to right: Nikolai Basov , Oleg Krokhin , Yuri Popov , Robert Hall .

Finally, in the same year, Basov, Krokhin, and Popov made a GaAs laser in FIAN. Thus, the year 1962 was truly a breakthrough for optoelectronics. True, all pioneering work united one big problem - lasers worked only in a pulsed mode, only at liquid nitrogen temperature, did not differ much in efficiency and quickly went out of order. Some scientists believed that the creation of a continuous semiconductor laser is impossible in principle.

(to be continued)

Literature
[1] N. Zheludev, a 100-Year History, Nat. Photonics 1, 189 (2007).
[2] Lecture of Zh. I. Alferov in the television program “Academia”, parts one and two .
[3] Nobel lectures in physics - 2000 (translated UFN).
[4] Additional information about the 2014 Nobel Prize in Physics.
[5] Charles N.V., Kirichenko N.A. The initial chapters of quantum mechanics. - M .: FIZMATLIT, 2004.
KDPV from here .