James R. Biard

James R. Biard
Inventor of the LED (light-emitting diode)
Born	(1931-05-20) May 20, 1931 Paris, Texas
Residence	United States
Nationality	American
Fields	Electrical engineering
Alma mater	Texas A&M University in College Station, TX; BS 1954, MS 1956, PhD 1957

Dr. James R. "Bob" Biard (born May 20, 1931) is an American electrical engineer/inventor who holds 70 U.S. and 6 foreign patents;^[1] including the first U.S. patent for the light-emitting diode (LED),^[2][3] the optical isolator,^[4] the Schottky transistor,^[5] and Metal Oxide Semiconductor Read Only Memory (MOS ROM).^[6] Dr. Biard currently works as a Senior Scientist for the Advanced Optical Components Division of Finisar in Allen, TX. He has also been on the staff of Texas A&M University as an Adjunct Professor of Electrical Engineering since 1980.

The IR Light-Emitting Diode

In the fall of 1961, while working at Texas Instruments Inc., Dr. Biard and Gary Pittman found that gallium arsenide (GaAs) emitted infrared light when electric current was applied. On Aug. 8th, 1962, Biard and Pittman filed a patent based on their findings for the "Semiconductor Radiant Diode" (U.S. Patent US3293513). After establishing the priority of their work based on engineering notebooks predating submissions from G.E. Labs, RCA Research Labs, IBM Research Labs, Bell Labs, and Lincoln Labs at MIT, the U.S. patent office issued the two inventors the first patent for the infrared (IR) light-emitting diode.^[7] T.I. immediately began a project afterward to manufacture infrared diodes and announced the first commercial product, the SNX-100, in October 1962. T.I. gave Biard and Pittman $1.00 each for their patent.

Infrared LEDs continue to be used today as transmitters in fiber optic data communication systems. They are also used in the remote control units of many commercial products including televisions, DVD players, and other domestic appliances.

Visible spectrum LEDs that could be used for watch displays were later developed by Nick Holonyak at G.E. in 1962.

Life and career

Bob grew up and attended school in Paris, TX where his father, James Christopher "Jimmy" Biard, worked as a farmer and a Dr. Pepper route salesman for the local Dr. Pepper company. His father eventually became manager of the local 7-Up company and ended up buying it from the former owner. His father also sold used cars, worked as a master plumber at Camp Maxey (an army camp north of Paris) during and after WW-II, and did plumbing work for homes and businesses in the Paris area. Later in life Jimmy became chief deputy sheriff in Lamar County, TX. Bob’s mother, Mary Ruth Biard (née Bills), worked as a retail sales person at the Collegiate Shop in downtown Paris. She also sang in quartets at weddings and funerals. While in high school, Bob worked for his father and an off duty fireman, who was also a plumber, during the summer as a plumber's assistant.

After receiving an associate degree from Paris Junior College in 1951, Bob transferred to Texas A&M University in College Station, TX where he received a B.S. in Electrical Engineering (1954), an M.S. in Electrical Engineering (1956), and a Ph.D. in Electrical Engineering (1957). While a student at Texas A&M, he met his wife Amelia Ruth Clark. They married on May 23, 1952 and later moved to Richardson, TX. Together they have 3 children (Jimmy, Jan, and Becky), 10 grandchildren, and 6 great-grandchildren. 

On June 3, 1957, three days after graduation, Dr. Biard was hired, along with his former Texas A&M professor Walter T. Matzen, by Texas Instruments Inc. in Dallas, TX as a molecular engineer. Here his work focused on the development of transistor circuits, microwave and optoelectronic components, avalanche photodiodes, silicon MOS Technology, high frequency transmission lines, solid state devices, and compound semiconductor materials technology.

In the summer of 1958, Texas Instruments hired Jack Kilby (the inventor of the integrated circuit). According to Biard, “At the time we were new, so we had to work while the others were on vacation.” Kilby held more than 60 U.S. patents, including two with Dr. Biard (the first optical isolator - US3304431 and an electro-optical transistor switching device - US3413480). Biard later stated, “I had the pleasure of being the co-inventor on two of his 60 patents. It was an honor to have my name with his."

Diagram of the tunnel diode constructed on a zinc diffused area of gallium arsenide semi-insulating substrate

In the fall of 1961, Dr. Biard and Gary Pittman worked on GaAs tunnel diodes and GaAs varactor diodes for X-band radar receivers in the Semiconductor Research and Development Laboratory (SRDL) at Texas Instruments. While investigating the valley current region of a tunnel diode constructed on a zinc diffused area of gallium arsenide semi-insulating substrate, they discovered a significant drop in resistance between the two Ohmic side contacts on the semi-insulating substrate when the diode was operated in forward bias. This photoconductive response in the semi-insulating substrate material was a result of photon emission in the forward biased tunnel diode. The photons were infrared, which could not be seen by the human eye. Using an infrared microscope recently brought in from Japan, they discovered all of the GaAs varactor diodes and tunnel diodes being manufactured at the time emitted light when operated under forward bias. On Aug. 8th, 1962, Biard and Pittman filed a patent based on their findings, which described a zinc diffused p-n junction LED with a spaced cathode contact to allow for efficient emission of infrared light under forward bias. After establishing the priority of their work, the U.S. patent office issued the two inventors the patent for the infrared (IR) light-emitting diode (U.S. Patent US3293513), the first modern LED. Most other organized research seeking LEDs at the time used II-VI semiconductors like cadmium sulfide (CdS) and cadmium telluride (CdTe). Biard and Pittman's patent used gallium arsenide (GaAs), a III/V semiconductor. Based on their findings, T.I. immediately began a project to manufacture infrared diodes and announced the first commercial product in October 1962, the SNX-100. This was the first LED on the market; $130 each. The SNX-100 used the zinc diffusion and gold-zinc P-type contact from the varactor diode and the tin alloy from the tunnel diode for the N-type Ohmic contact. The N-type contact was achieved by plating molybdenum wires with tin and alloying them into the N-type surface of the die using a strip heater. The IBM 059 Card Verifier, which was the Verifier companion to the IBM 029 Card Punch (announced October 14, 1964), was the first device to use infrared LEDs. The LEDs replaced tungsten bulbs that controlled punched card readers. Infrared light was sent through the holes or blocked by the card, which not only significantly reduced the size and power required, but also improved the reliability.

In Dec. of 2013, during a recollection of the patent, Dr. Biard stated the following:

The tunnel diode we made on the semi-insulating substrate had an N+ GaAs region under the tin dot that was formed by heating the sample to a temperature above the tin/GaAs eutectic temperature. In that process some of the P-type GaAs was dissolved into the liquid at the interface between the tin dot and the GaAs. As the sample is cooled back to the eutectic temperature, that GaAs grows back out of the liquid onto the P-type GaAs as a single crystal deposit. However, the regrown GaAs is then very heavily doped with tin, which acts as a donor impurity. This makes the regrowth region N++ due to the high tin doping level that results from the heating above the eutectic temperature. The material that was dissolved was heavily doped P-type, but the tin concentration was so high that it converted the regrown layer to N++. To make a tunnel diode with its unique current-voltage characteristic requires that the semiconductor be heavily doped on the N-type side and the P-type side of the junction. A tunnel diode is so heavily doped that it has a breakdown voltage of zero volts. The current rises rapidly for small forward voltage and reverse voltage. As the forward voltage is increased, the current peaks and falls off as the forward voltage is increased further. As the forward bias is increased even more, the current begins to increase exponentially due to carrier injection across the PN junction. The PN junction tunnel diode we made had a heavier doping concentration on the N side than on the P side. Thus you could say it was an N++P+ junction. Both sides had to be very heavily doped to form a tunnel diode. The Ohmic contact to the N++ regrowth layer was a tunneling Schottky diode between the tin metal in the dot and the N++ regrowth region. The gold-zinc metal layer on the P+ GaAs was heated to a temperature at which the zinc diffused from the gold-zinc metal deposit into the P+ GaAs. This made the GaAs P++ in a thin layer at that surface next to the metal layer. The Ohmic contact to the P+ GaAs was formed by a tunneling Schottky diode formed between the P++ zinc diffused layer and the gold metal. The result of all of this is an Ohmic contact to the N++ regrowth layer, an Ohmic contact to the P++ zinc diffused layer, and a tunnel diode junction between the N++ regrowth and the P+ GaAs layer. It was not important that the junction formed was a tunnel diode. The light was emitted under the condition of high forward bias where minority carriers were injected across the N++P+ junction. Since the N-type layer was more heavily doped than the P-type layer, the major carrier injection was minority carrier electrons into the P+ material. When those minority carrier electrons recombined with the majority carrier holes, the photons were emitted. At the same time we also saw light emission from varactor diodes, which were made by zinc diffusion into lightly doped N-type GaAs. Those devices were P++N- diodes in which the forward bias caused minority carrier holes to be injected into the lighter doped N-type GaAs. Photons were emitted when the minority carrier holes recombined with the majority carrier electrons in the lightly doped N-type material. The varactor diodes had a reasonable reverse breakdown voltage (10-15V) and did not show the negative resistance characteristic of the tunnel diode. The observation that light was emitted from both the tunnel diode and the varactor diode, when viewed with the IR microscope, proved that it was a minority carrier recombination effect in the direct bandgap semiconductor (GaAs). The test structure that we built with the tin dot on P-type GaAs on the semi-insulating GaAs substrate that demonstrated the photoconductive effect in the semi-insulating material just happened to be a tunnel diode. We picked that structure because it was easy to fabricate. We would have gotten the same result if we had deposited an N-type layer on the semi-insulating substrate and diffused zinc into it to form the PN junction.

On Nov. 29th, 1963 Dr. Biard, Gary Pittman, Edward L. Bonin, and Jack Kilby filed a patent titled "Photosensitive Transistor Chopper Using Light Emissive Diode" (U.S. Patent US3304431); the first optical isolator. Within the patent they described a phototransistor chopper consisting of an LED optically coupled with a dual emitter, photosensitive, silicon transistor. The arrangement provided a switching function in which the switch was completely electrically isolated from the LED that drove it. The transistor operated in response to light emitted from the LED when forward current bias was generated across the junction of the diode. When emitted light struck the surface of the transistor, it was absorbed in the regions of both the emitter-base and base-collector junctions causing the transistor to conduct. This photoconductive transistor could be rapidly turned on and off by intensity modulating the LED at a very high frequency using high frequency alternating voltage. Prior to their invention, complete electrical isolation of the switch element in a chopper from the driving source for opening and closing the switch element was not possible, even through use of isolation transformers. Using such transformers, which were bulky and expensive for use in miniaturized circuits, to separate the driving source and the switch element resulted in magnetic pick-up and spike feed-through due to the transformer winding capacitance. Optical isolators were ideal because they're very small and can be mounted to a circuit board. In addition, they offer protection against excessively high voltages, reduce noise levels, and make measurements more accurate. Optical isolators are found in all sorts of electronic circuits in which electrical isolation is important.

In the mid-1960s, Dr. Biard was placed in charge of both the Optoelectronic branch and the MOS branch in the Semiconductor Research and Development Lab (SRDL) at Texas Instruments. In 1964, TI's Opto branch developed a monolithic visible LED element consisting of a 3x5 array of red LEDs capable of displaying the numbers 0-9. The device was lacking a means of driving the array, so Dr. Biard and Bob Crawford (from the MOS branch) designed a P-channel MOS circuit using binary coded decimal inputs to turn on the appropriate 15 output lines. The MOS circuit worked on the first pass and was implemented into a simulated cockpit altimeter display shown in the TI booth that year at the New York IEEE show and convention. Biard and Crawford filed a patent for their device (U.S. Patent US3541543) on July 25, 1966, referred to as the "Binary Decoder", unaware at the time that it was an MOS ROM. Within the patent they described how to fabricate metal gate ROMs of small size to permanently program an MOS logic array upon manufacture by the gate level mask or moat mask. In the 1980s, while employed for Honeywell, Dr. Biard testified before the International Trade Commission in Washington D.C., per the request of Texas Instruments, in a royalties suit regarding a Japanese firm's use of their Binary Decoder patent in their own MOS ROM circuits. The judge in the lawsuit determined that the Japanese firm did not violate TI's patent rights because they had made enough changes in the implementation of their circuits. Today MOS ROM devices mounted on circuit boards are the most common example of nonvolatile memory used to provide the storage of fixed programs in digital equipment such as minicomputers and microprocessor systems (for ex. most calculators employ MOS ROM to store a program consisting of a large number of instruction words).

In 1964, Dr. Biard designed linear amplifiers to work with photodetectors and used hot carrier diodes to keep the transistors from going into saturation. The engineer in the next office was working on developing Diode Transistor Logic (DTL) ICs and was also having saturation problems. On Dec. 31st, 1964, Dr. Biard filed a patent for the Schottky transistor (U.S. Patent US3463975), a.k.a. the Schottky-clamped transistor, which consisted of a transistor and an internal metal-semiconductor Schottky-barrier diode. The patent was filed based on Schottky Clamped DTL monolithic integrated logic circuits using aluminum-silicon Schottky diodes across the collector-base junctions of the transistors and in the input to adjust the logic levels. The diode prevented the transistor from saturating by minimizing the forward bias on the collector-base transistor junction, thus reducing the minority carrier injection to a negligible amount. The Schottky diode could be integrated on the same die, it had a compact layout, it had no minority carrier charge storage, and it was faster than a conventional junction diode. Dr. Biard's patent was filed before Transistor-Transistor Logic circuits (TTL) had been invented, yet it was written broadly enough to cover the Schottky clamped TTL ICs using platinum silicide Schottky diodes, which were much more predictable and manufacturable than the aluminum Schottky diodes he originally used. His patent ultimately improved the switching speed of saturated logic designs, such as the Schottky-TTL, at a low cost. In 1985, Dr. Biard received the Patrick E. Haggerty Innovation Award for this patent.

In the 1960s, during the ongoing development of integrated circuit related technologies, avalanche photodiodes were afflicted by a relatively high bulk leakage current, which was amplified by the avalanche gain. The leakage current resulted from holes and electrons thermally generated in the device. This leakage current restricted the photodiode's use, unless a cooling apparatus was used conjunctively. On Feb. 15th, 1968 Dr. Biard filed a patent titled "Low Bulk Leakage Current Avalanche Photodiode" (U.S. Patent US3534231), which presented the design of an avalanche photodiode to reduce the bulk leakage currents without having to be cooled. The design consisted of three semiconductor layers, located one on the other, with a barrier layer below the photosensitive junction in the form of a reverse biased second junction. The first two layers constituted the photosensitive junction and the third layer constituted a highly doped semiconductor back region present at a distance from the photosensitive junction smaller than a diffusion length of the thermally generated carriers.

In 1969, Dr. Biard left Texas Instruments to join Spectronics, Inc., when the company was founded, as Vice President of Research. While at Spectronics, Dr. Biard worked on the development of optical couplers used in a data bus developed for airborne avionics systems. He also worked on integrated circuits consisting of an LED driver and pin diode receiver used for digital fiber optic communications. Spectronics, Inc. was acquired by Honeywell in 1978. From 1978 to 1987, Dr. Biard worked as Chief Scientist of the Honeywell Optoelectronics Division in Richardson, TX. Dr. Biard started their MICROSWITCH IC & Sensor Design Center and served as a member of the Components Group Sensor Planning Team. He was also the Components Group representative on the Honeywell Technology Board (HTB). The HTB was concerned with the development and transfer of technology throughout the Honeywell corporate structure. Dr. Biard's product development responsibilities included optoelectronic components (light emitting diodes and photodetectors), fiber optic components, transmitter & receiver modules, silicon Hall effect sensors, and pressure sensors.

In 1987, Dr. Biard became Chief Scientist of the Honeywell MICRO SWITCH Division. He then retired in Dec. 1998 only to be hired back on as a consultant. As a consultant, he became part of a team developing Vertical Cavity Surface Emitting Lasers (VCSELs). He was also involved in the interface between the VCSEL team, universities, and the Honeywell Corporate Research Laboratory.

In 2006, Honeywell sold the VCSEL group to the Finisar Corporation, which hired Dr. Biard on half time as a consultant Senior Scientist for the Advanced Optical Components Division in Allen, TX. While working for Finisar, Dr. Biard has been issued a total of 28 engineering patents related to VCSELs, photodiodes for high-speed fiber optic data transmission, and semiconductor devices.

Bob is also an avid harmonica player. He's performed locally in the Dallas area at banquets, schools, churches, hospitals, retirement homes, and performance halls. His renditions of classic songs are done with several harmonicas and a musical saw.^[8] 

Publications

In the course of his technical career, Dr. Biard has published more than two dozen technical papers and made about the same number of unpublished presentations at major technical conferences.  He also developed a one-week seminar on Fiber Optic Data Transmission that he's presented on five occasions. Some of his key papers include: 

W. T. Matzen and J. R. Biard, "Differential Amplifier Features D-C Stability", Electronics; January 16, 1959.

J. R. Biard, E. L. Bonin, W. N. Carr, and G. E. Pittman, "GaAs Infrared Source", PGED Electron Device Conference, Washington, D.C.; October, 1962.

J. R. Biard, "Low frequency Reactance Amplifier", IEEE Proc., Vol. 51, No. 2, pp. 298–303; February, 1963.

W. N. Carr and J. R. Biard, "Common Occurrence of Artifacts or 'Ghost' Peaks in Semiconductor Injection Electroluminescence Spectra", Journal of Applied Physics, Vol. 35, No. 9, pp. 2776–2777; September, 1964.

J. R. Biard, "Degradation of Quantum Efficiency in GaAs Light Emitters", GaAs: 1966 Symposium Proceedings, (Reading England), Institute of Physics and Physical Society, pp. 113–117; September, 1966.

J. R. Biard and L. L. Stewart, "Optoelectronic Data Bus", IEEE Electromagnetic Compatibility Symposium Rec., IEEE 74CH0803-7 EMC; October, 1973.

Ralph H. Johnson, Brian W. Johnson, and J. R. Biard, "Unified Physical DC and AC MESFET Model for Circuit Simulation and Device Modeling", IEEE Electron Devices Transactions; September, 1987. 

A. Peczalski, G. Lee, J. R. Biard, et al., "A 6 K GaAs gate array with fully functional LSI personalization", Honeywell Syst. & Res. Center, Page(s): 581 - 590; April, 1988.

Ananth Ramaswamy, Jan P. van der Ziel, J. R. Biard, Ralph Johnson, and Jim A. Tatum, “Electrical Characteristics of Proton-Implanted Vertical-Cavity Surface-Emitting Lasers", IEEE Journal of Quantum Electronics, Vol. 34, No. 11; November, 1998. 

Awards and honors

In 1969, Dr. Biard was elected as a Life Fellow of IEEE cited for "outstanding contributions in the field of optoelectronics".

In 1985, he received the Patrick E. Haggerty Innovation Award for his contribution to the design and development of Schottky Logic.

In 1986, he was recognized as a Distinguished Alumnus of Texas A&M University.

In 1989, he received the Honeywell Lund Award.

In 1991, he was elected to membership in the National Academy of Engineering.

In May 2013, he was awarded the degree of Doctor of Science, honoris causa, from Southern Methodist University.

In September 2013, he received the “Distinguished Graduate Award” from Paris High School in Paris, TX.

References