De: City of Light The Story of Fiber Optics




Скачать 87.64 Kb.
НазваниеDe: City of Light The Story of Fiber Optics
страница1/2
Дата конвертации11.02.2013
Размер87.64 Kb.
ТипДокументы
  1   2
De: City of Light - The Story of Fiber Optics

Jeff Hecht

Oxford University Press 1999

THE SLOAN TECHNOLOGY SERIES

Capítulo 12 pp.147-159






(figuras no final)

p.147


Recipes for Grains of Salt The Semiconductor Laser (1962 to 1977)


Jack Tiliman said "Did you know the semiconductor laser has just been invented? . . . I think we should get some and see if they're any use

at all for communications." So I went over [in 1963 to the joint Services Electronics Research laboratory in E@aldock, north of London] for a month, and brought some of these things back with me. They only operated at liquid nitrogen temperature, only operated in pulses of about 1 0

microseconds, and it took 1 00 am- peres in each pulse to drive the bioody things. We con- cluded that they weren't very promising. -David Newman, British Post Office Research laboratories' s Corning all too keenly recognized, clear optical fibers were not the only Abuilding block vital for fiber-optic communications. A second crucial ingredient was a matching light source. From Bell Labs to the British Post Office, developers of optical communication systems thought the ideal light source would be a semiconductor laser. By a curious coincidence, the spring and summer of 1970 also marked symbolic breakthroughs on the semiconductor laser frontier, although as with fibers, practical devices took longer. The allures of semiconductor lasers were considerable. One was the sheer magic of solid-state devices. Semiconductor technology was hot in the 1960s;


p.148


transistor electronics were pushing vacuum tubes into the grave of obsoles- cence. Semiconductor lasers generated light easily, with the intensity con- trolled by the electrical currents passing through them. Tiny as grains of salt, they matched the size of optical fibers, but the potential for compact, solid- state laser transmitters attracted even the developers of hollow optical wave- guides. Unfortunately, early semiconductor lasers were as useless for practical communications as the optical fibers of the early 1960s. The lasers operated only at the cryogenic temperature of liquid nitrogen, -321'F or -196'C. They burned out quickly and unpredictably. Worse yet, the warmer the laser got, the more current you needed to make it emit light, and the higher the current, the faster the laser burned out. It didn't look good, and after an early burst of energy, progress stalled in the mid-1960s. Developers needed better recipes for their grains of salt. A Curious Class of Mciterials Both the allure and the problems of semiconductor lasers came from the nature of semiconductors. From an electrical standpoint, materials fall into three classes-conductors, semiconductors, and insulators. The essential dif- ference among them is the amount of energy needed to free an electron from the bonds that link atoms in the material. In conductors, notably metals, very little energy is nepded, so electrons flow freely through copper wires. In glass and other insulators, the electrons are bound so tightly between atoms that essentially none of them have enough energy to escape. Semiconductors fall between the two extremes, because a few electrons do escape from atomic bonds to conduct electricity in the crystal. The bonded electrons occupy a "valence band," where they form bonds between atoms in the crystal. Electrons that get enough energy to escape those atomic bonds fall into another energy state called the "conduction band." No energy states exist between the valence and conduction bands, so electrons have to get enough energy to jump this "band gap" before they can carry a current. In a pure semiconductor, only a few electrons have enough energy to reach the conduction band. Those that escape the valence band leave behind vacancies called "holes," which effectively have a positive charge equal to the negative charge of the electron. Other valence-band electrons can move to fill the hole, leaving another hole behind, so essentially the holes move as well as the electrons in the conduction band. For practical purposes, you can think of holes as positive charge carriers and electrons as negative charge carriers, although it really isn't that simple. One way that engineers can adjust the number of electrons and holes in a semiconductor is to add impurities to replace some atoms in the crystal lattice. If the impurity has one more outer electron than the atom it replaces, that extra electron is free to roam through the crystal, forming an n-type (for transistor electronics were pushing vacuum tubes into the grave of obsoles- cence. Semiconductor lasers generated light easily, with the intensity con- trolled by the electrical currents passing through them. Tiny as grains of salt, they matched the size of optical fibers, but the potential for compact, solid- state laser transmitters attracted even the developers of hollow optical wave- guides. Unfortunately, early semiconductor lasers were as useless for practical communications as the optical fibers of the early 1960s. The lasers operated only at the cryogenic temperature of liquid nitrogen, -321'F or -196'C. They burned out quickly and unpredictably. Worse yet, the warmer the laser got, the more current you needed to make it emit light, and the higher the current, the faster the laser burned out. It didn't look good, and after an early burst of energy, progress stalled in the mid-1960s. Developers needed better recipes for their grains of salt. A Curious Class of Materials Both the allure and the problems of semiconductor lasers came from the nature of semiconductors. From an electrical standpoint, materials fall into three classes-conductors, semiconductors, and insulators. The essential dif- ference among them is the amount of energy needed to free an electron from the bonds that link atoms in the material. In conductors, notably metals, very little energy is needed, so electrons flow freely through copper wires. In glass and other insulators, the electrons are bound so tightly between atoms that essentially none of them have enough energy to escape. Semiconductors fall between the two extremes, because a few electrons do escape from atomic bonds to conduct electricity in the crystal. The bonded electrons occupy a "valence band," where they form bonds between atoms in the crystal. Electrons that get enough energy to escape those atomic bonds fall into another energy state called the "conduction band." No energy states exist between the valence and conduction bands, so electrons have to get enough energy to jump this "band gap" before they can carry a current. In a pure semiconductor, only a few electrons have enough energy to reach the conduction band. Those that escape the valence band leave behind vacancies called "holes," which effectively have a positive charge equal to the negative charge of the electron. Other valence-band electrons can move to fill the hole, leaving another hole behind, so essentially the holes move as well as the electrons in the conduction band. For practical purposes, you can think of holes as positive charge carriers and electrons as negative charge carriers, although it really isn't that simple. One way that engineers can adjust the number of electrons and holes in a semiconductor is to add impurities to replace some atoms in the crystal lattice. If the impurity has one more outer electron than the atom it replaces, that extra electron is free to roam through the crystal, forming an n-type (for


p.149


- - ---- -~-- -- - semiconductor laser. He worked through the numbers on the train ride home, negative carrier) semiconductor. An impurity with One less outer electron creates a hole, forming a semiconductor of thd p-type (for positive carrier, because the absence of an electron leaves a positive charge on the atomic nucleus). Adding such dopants releases more carriers in the crystal, making better electronic devices. Put layers of n-and p-type semiconductor together, and you have a simple two-terminal device called a diode .2 Things get interesting when you apply a voltage across a diode. A positive electrical voltage attracts the negative electrons, while a negative voltage attracts the positive holes. Apply a positive voltage to the n-type material and a negative voltage to the p-type material, and the electrons and holes move to the terminals and stay there, so no current flows through the junc- tion between the two materials. However, a current will flow if you switch the voitages, because electrons in the n material move toward the positive terminal on the p material, and holes move in the opposite direction. Thus, a semiconductor diode conducts electricity in only one direct!0n3 (figure 12-1). The real action happens at the junction between n and p material, where electrons from the n material combine with holes from the p material. As the electron drops into the hole, it releases its extra energy by a process called recombination. In silicon, the energy typically is released as heat. However, in gallium arsenide, indium phosphide, and certain other semiconductor crystals, some energy is released as light. This is the basis of a light-emitting diode or LED. Initially, the process was quite inefficient. The first LEDs turned only about 0.01 percent of the input electrical energy into visible or infrared light.' Birth-of the Semiconductor Laser When the laser hit the headlines, a few physicists thought of making lasers from light-emitting semiconductors. However, nobody took the idea too seriously until 1962, when Robert Rediker, Tom Quist, and Robert Keyes changed the way they were processing gallium arsenide at the MIT Lincoln Laboratory. LEDs made from the new material generated light surprisingly efficiently. When Keyes announced their results at a July 1962 meeting in New Hampshire, an astounded member of the audience stood up to say that Keyes's statement violated the second law of thermodynamics-a cardinal principle of modern physics which holds that the degree of disorder or "en- tropy" always increases. Keyes returned to the microphone and deadpanned, "I'm sorry."-' Bob Hall, a semiconductor expert from the General Electric Research Lab- oratory, immediately figured out what the MIT group had done. After every- one stopped laughing, he stood up and explained why the results were pos- sible. He also realized that high efficiency opened the door to making a semiconductor laser. He worked through the numbers on the train ride home,


p.150


Figure 12-1: The first semiconductor laser was a simple cube of gallium arsenide (GaAs), half p-type and half n-type.


and back in Scheneetady he rounded up a crew of other scientists to try some expe ntS.6 In short order, they made gallium arsenide chips with edges rime polished to act as the mirrors needed to generate a laser beam. When they fired powe;ful electrical pulses through the chips, they emitted light at a nar- row range of wavelengths, a sign of laser action. The whole process took just two and a half months .7 It was a scientific tour de force. Hall was good, but three other labs were hot on his heels. A team from IBM came in a close second with a slightly different variation, although no one from the lab had attended the New Hampshire meeting." Nick Holonyak, Jr., returned from the New Hampshire meeting to try making lasers from gallium arsenide phosphide at GE's Syracuse lab but didn't succeed until he tried Hall's approach of polishing the crystals.9 The Lincoln Lab group, dis- tracted by other projects, wound up a close but frustrating fourth."' That remarkably close finish-achieved in the shadow of the Cuban missile crisis-seemed to herald good things for semiconductor lasers, but there the technology stalled. The developers of communication systems wanted lasers that generated steady beams and operated for long periods at room temper- ature. The lasers made by Hall and the others only fired intermittent pulses and didn't last long even when cooled to liquid nitrogen temperature, - 32 I'F or -196'C. Progress from there was painfully slow. By the end of 1964, the best semiconductor laser could fire a single pulse at room temperature when 25 amperes-more current than a standard household refrigerator draws- flowed through an area of 0.02 square millimeter for 50 billionths of a second. Then it had to cool before firing again." and back in Scheneetady he rounded up a crew of other scientists to try some r' In short order, they made gallium arsenide chips with edges expe iinents .6 polished to act as the mirrors needed to generate a laser beam. When they fired power*ful electrical pulses through the chips, they emitted light at a nar- row range of wavelengths, a sign of laser action. The whole process took just two and a half months .7 It was a scientific tour de force. Hall was good, but three other labs were hot on his heels. A team from IBM came in a close second with a slightly different variation, although no one from the lab had attended the New Hampshire meeting," Nick Holonyak, Jr., returned from the New Hampshire meeting to try making lasers from gallium arsenide phosphide at GE's Syracuse lab but didn't succeed until he tried Hall's approach of polishing the crystals.9 The Lincoln Lab group, dis- tracted by other projects, wound up a close but frustrating fourth.10 That remarkably close finish-achieved in the shadow of the Cuban missile crisis-seemed to herald good things for semiconductor lasers, but there the technology stalled. The developers of communication systems wanted lasers that generated steady beams and operated for long periods at room temper- ature. The lasers made by Hall and the others only fired intermittent pulses and didn't last long even when cooled to liquid nitrogen temperature, - 32 I'F or -196'C. Progress from there was painfully slow. By the end of 1964, the best semiconductor laser could fire a single pulse at room temperature when 25 amperes-more current than a standard household refrigerator draws- flowed through an area of 0.02 square mffiiineter for 50 billionths of a second. Then it had to cool before firing again.11


p.151


Endless problems frustrated semiconductor laser developers. The devices didn't last long. Some died suddenly; others degraded gradually, emitting less and less light. The beams were messy, not pencil thin like those of gas lasers but spreading rapidly and unevenly, a fuzzy blur instead of the pinpoint of light wanted for communications. No one was sure what caused the problems. Was the design inadequate? Were requirements for crystal quality impossibly highp Or was the whole problem simply insoluble, dooming semiconductor lasers to the oblivion of interesting but impractical devicesp Many researchers interested in civilian conununications bailed out, but not the military. Compact semiconductor diode lasers looked promising for use in portable systems with purposes ranging from battlefield communications to measuring the distance of an air-to-air missile from its target. Pulsed lasers could do many of those jobs, although room-temperature operation was vital. Besides, the Pentagon was flush with money. As often happens, ideas that would help solve those problems had already been suggested but had not been tested. The first diode lasers were made entirely of one material, gallium arsenide, with different dopants. In 1963, Herbert Kroemer, an engineer at the Varian Central Research Laboratory in Palo Alto@ suggested adding layers of different compositions So did one of the Soviet Union's top semiconductor researchers, 1 3Zhores Alferov of the loffe Physico-Technical Institute in Leningrad (now St. Petersburg), but the Soviet ,government classified his patent disclosure."' Their basic idea was the same, to trap electrons at the junction so they could recombine more efficiently with holes. They hoped this would make a more efficient laser, able to operate at warmer temperatures. They realized they could do this by adjusting the semi- conductor composition, which affects the band-gap energy electrons need to be in the conduction band. Substituting aluminum for some gallium atoms in gallium arsenide increases the energy requirement, so electrons in gallium arsenide lack enough energy to move into gallium aluminum arsenide. Place a layer of gallium aluminum arsenide next to a p-n junction in gallium ar- senide, forming what specialists call a "heterojunction," and you've trapped electrons on the gallium arsenide side of the junction, increasing the odds they will recombine with holes and emit light. Unfortunately, no one knew how to make heterojunctions in 1963. Chang- ing semiconductor composition also affects the spacing between atoms in the crystalline lattice, and mismatches cause fatal flaws in devices. The trick was to find a combination of materials where the lattice difference was small. Gallium arsenide was the logical place to start. The question was what to add. Alferov's group considered two possibilities: replacing some gallium with aluminum, or replacing some arsenic with phosphorus. Adding aluminum hardly changes the lattice constant, but aluminum arsenide decomposes in moist air, so the Russians doubted gallium aluminum arsenide would be sta- ble. Instead, they added phosphorus, but that changed the lattice constant so Endless problems frustrated semiconductor laser developers. The devices didn't last long. Some died suddenly; others degraded gradually, emitting less and less light. The beams were messy, not pencil thin like those of gas lasers but spreading rapidly and unevenly, a fuzzy blur instead of the pinpoint of light wanted for communications. No one was sure what caused the problems. Was the design inadequate? Were requirements for crystal quality impossibly highp Or was the whole problem simply insoluble, dooming semiconductor lasers to the oblivion of interesting but impractical devicesp Many researchers interested in civilian communications bailed out, but not the military. Compact semiconductor diode lasers looked promising for use in portable systems with purposes ranging from battlefield communications to measuring the distance of an air-to-air missile from its target, Pulsed lasers could do many of those jobs, although room-temperature operation was vital. Besides, the Pentagon was flush with money. As often happens, ideas that would help solve those problems had already been suggested but had not been tested. The first diode lasers were made entirely of one material, gallium arsenide, with different dopants. In 1963, Herbert Kroemer, an engineer at the Varian Central Research Laboratory in Palo Alto@ suggested adding layers of different compositions So did one of the Soviet Union's top semiconductor researchers, 1 3Zhores Alferov of the loffe Physico-Technical Institute in Leningrad (now St. Petersburg), but the Soviet ,government classified his patent disclosure."' Their basic idea was the same, to trap electrons at the junction so they could recombine more efficiently with holes. They hoped this would make a more efficient laser, able to operate at warmer temperatures. They realized they could do this by adjusting the semi- conductor composition, which affects the band-gap energy electrons need to be in the conduction band. Substituting aluminum for some gallium atoms in gallium arsenide increases the energy requirement, so electrons in gallium arsenide lack enough energy to move into gallium aluminum arsenide. Place a layer of gallium aluminum arsenide next to a p-n junction in gallium ar- senide, forming what specialists call a "heterojunction," and you've trapped electrons on the gallium arsenide side of the junction, increasing the odds they will recombine with holes and emit light. Unfortunately, no one knew how to make heterojunctions in 19 6 3. Chang- ing semiconductor composition also affects the spacing between atoms in the crystalline lattice, and mismatches cause fatal flaws in devices. The trick was to find a combination of materials where the lattice difference was small. Gallium arsenide was the logical place to start. The question was what to add. Alferov's group considered two possibilities: replacing some gallium with aluminum, or replacing some arsenic with phosphorus. Adding aluminum hardly changes the lattice constant, but aluminum arsenide decomposes in moist air, so the Russians doubted gallium aluminum arsenide would be sta- ble. Instead, they added phosphorus, but that changed the lattice constant so
  1   2

Добавить в свой блог или на сайт

Похожие:

De: City of Light The Story of Fiber Optics iconHistory & Chronology of Fiber Optics

De: City of Light The Story of Fiber Optics iconChapter 15. Laser and fiber optics

De: City of Light The Story of Fiber Optics iconSeattle City Light Integrated Resource Plan

De: City of Light The Story of Fiber Optics iconThis story is a work of fiction. Other than factual locations like the city of New Orleans and the former Cleveland Joseph Peete Projects, etc., names

De: City of Light The Story of Fiber Optics iconThis response from the swgmat fiber Subgroup is not all inclusive of all the relevant literature in the field of Forensic Fiber Examinations

De: City of Light The Story of Fiber Optics iconСвет … Ньютон отдавал предпочтение корпускулярной теории света
Спеллиусом and Descartes [3, 4]; diffraction and an interference of light were opened by Гримальди (the publication 1665), double...

De: City of Light The Story of Fiber Optics iconВлияние стеклотканевых прослоек на прочность клееных деревянных балок
Тhere were investigated the possibilities of increasing the strength of wooden building structures using a fiber dispersion of reinforcement....

De: City of Light The Story of Fiber Optics iconL'icona 2012 2-й Международный симпозиум
В рамках международной научно-технической конференции «Laser Optics`12» («Оптика лазеров-2012») в период 26-27 июня 2012 г в Санкт-Петербурге...

De: City of Light The Story of Fiber Optics iconThe light is dim as you enter the room after receiving no answer from repeated knocking. The lack of light is due to aluminum foil and newsprint taped to the

De: City of Light The Story of Fiber Optics iconThe laser is a device that a beam of light that is both scientifically and practically of great use because it is coherent light. The beam is produced by a


Разместите кнопку на своём сайте:
lib.convdocs.org


База данных защищена авторским правом ©lib.convdocs.org 2012
обратиться к администрации
lib.convdocs.org
Главная страница