GEORGE E. SMITH

I was born in White Plains, New York in 1930. I grew up in several differ-ent states and attended a variety of primary and secondary schools. Upon graduation from high school in 1948, I joined the U.S. Navy where I spent four years as an aerographer’s mate (weatherman), part of it during the Korean War. While stationed in Miami, Florida, I managed to take enough courses at the University of Miami to qualify as a starting sophomore at the University of Pennsylvania in 1952. Since I served during the Korean War, I qualified for the GI Bill, which helped pay tuition. I majored in Physics and graduated with honors in 1955. I married right after graduation. I was accepted into the graduate program at the University of Chicago with a job as teaching assistant and graduated there in 1959 with a Ph.D. in Physics. During that time I also received grants from the National Science Foundation and Bell Telephone Laboratories. Upon graduation, I was offered a job with the Research area of Bell Labs and accepted it. I never bothered to interview at any other institution.

I was assigned to a new department, headed by Bill Boyle, where I started out continuing along the lines of my thesis topic, studying the electronic properties of semimetals. I branched out into other fields as well, including thermoelectric cooling materials and low temperature electronic devices. During the five years in that position I generated many papers and patents. Bill was then promoted to Director of the exploratory semiconductor device development laboratory and he offered me the position as head of a new department entitled Device Concepts. Several bright, imaginative people were assigned to my group and I was given the mandate to hire more of the same. Many fields were pursued including junction lasers, semiconducting ferroelectrics, electroluminescence, transition metal oxides, and the silicon diode array camera tube. In 1969, Bill and I invented the Charge Coupled Device and much of my time was then spent in that field. In addition, my department was renamed the VLSI Device Department where responsibilities covered the physics of devices made with submicron lithography and their use in high performance digital and analog circuits.

In the realm of academic recognition, I have been a member of Pi Mu Epsilon, Phi Beta Kappa, and Sigma Xi. I was made a fellow of the Institute of Electrical and Electronic Engineers, Fellow of the American Physical Society and a member of the National Academy of Engineering. I hold 31 US patents and am the author of over 40 papers. I was founding editor of the IEEE publication “Electron Device Letters”.

My major technical accomplishment, of course, was the inception of the Charge Coupled Device with Willard S. Boyle. We hold the basic patent

89

90

(US 3,858,232) and published the first paper disclosing the device concept accompanied by a paper on its experimental verification. A following inven-tion of the Buried Channel Charge Coupled Device (US patent 3,792,322) significantly improved the performance of the original CCD. These accom-plishments have been recognized in the following awards:

• 1973 Ballantine Medal of the Franklin Institute – “For the invention of the Charge-Coupled Device structure, a conceptually simple semiconductor technology with significant application to image sensing, serial memory and signal processing.”• 1974 Morris N. Liebman Award of the Institute of Electrical and Electronic Engineering – “For the invention of the Charge-Coupled Device and leader- ship in the field of MOS device physics.”• 1986 Progress Medal of the Photographic Society of America – “For the Charge-Coupled Device’s applications to electronic imaging devices.”• 1999 IEEE Device Research Conference Breakthrough – “For pioneering work in the field of Buried-Channel Charge Coupled Devices”.• 1999 Computer and Communications Prize – “For the Invention of the Charge Coupled Device (CCD)”.• 2001 Edwin Land Medal – “For the invention and development of the Charge-Coupled Device, a contribution that has had extraordinary impact on image creation and utilization”.• 2006 Charles Stark Draper Prize – “For the invention of the Charge- Coupled Device (CCD), a light-sensitive component at the heart of digital cameras and other widely used imaging technologies.”

And, of course,

• 2009 Nobel Prize in Physics – “For the invention of an imaging semi- conductor circuit – the CCD sensor.”

Additional honors include:

• 2006 Honored for the invention of the CCD by US Senate Resolution 478• 2006InductionintotheNationalInventorsHallofFame• 2006InductionintotheImagingHallofFame• 2008InductionintotheNewJerseyInventorsHallofFame

Also, I received the: IEEE Electron Devices Society Distinguished Service Award (1997).

On the personal side, I had always wanted to sail and purchased my first boat right after joining Bell Labs and sailed it and subsequent boats on Barnegat Bay,abouthalfwaydownthecoastofNewJersey.Mywifealsoenjoyedsailing

91

but she passed away in 1975. I then commenced to raise children aged 10, 11 and 14 as a single parent. No comment necessary. Two years later, I became partnerswithJanetMurphywhoalsolovedsailingandwehadmanyadven-tures sailing a small (22 foot) cabin boat from Northeast Harbor, Maine to Beaufort, North Carolina. We both decided to retire (she was a teacher) early and sail around the world. To do this, we had a seagoing 31-foot Southern Cross, named Apogee, semi custom built for the task in Bristol, Rhode Island in 1983. After two shakedown trips to Bermuda, we retired in 1986 and start-ed around the world. We did not get back until 2003, although we occasion-ally flew home for a short visit. We now live in our home on a lagoon leading to Barnegat Bay. Apogee is moored to our dock in the back yard and we still enjoy sailing on the bay.

REFERENCES1. W. S. Boyle and G. E. Smith, “Charge Coupled Semiconductor Devices,” Bell Sys. Tech. J.,49,587,(1970).2. G. F. Amelio, M. F. Tompsett, and G. E. Smith, “Experimental Verification of the Charge CoupledDeviceConcept,”BellSys.Tech.J.,49,593,(1970).

92

THE INVENTION AND EARLY HISTORY OF THE CCD

Nobel Lecture, December 8, 2009

by

GEORGE E. SMITH

Bell Laboratories, Murray Hill, NJ, U.S.A. (retired)

To try to indicate how the invention of Charge Coupled Devices in 1969 by W. S. Boyle and myself came about, it is first necessary to describe three important technologies being pursued at the time. Magnetic bubbles, the silicon diode array vidicon for the Picturephone, and MOS semiconductor technology were all contributing factors to the invention of the device.

Figure 1. Magnetic Bubble serial shift register.

Magnetic bubbles were formed by placing a special magnetic material, typi-cally garnet, in a magnetic field so that circular magnetic domains could be formed as shown in Figure 1. A permalloy pattern was deposited on the sur-face which could capture the domain, and when a rotating magnetic field was applied in the plane of the slice, the bubble would be shifted along from one site to the next. With an input device to inject the presence or absence of bubbles representing 0s or 1s and a detection device at the end, one had a serial memory device.

93

Figure 2. The Picturephone set and silicon diode array camera tube.

The second driving technology being worked on, which just happened to be in my department, was the silicon diode array camera tube, shown in Figure 2. This was being developed for the Picturephone, which AT&T − Bell Labs’ parent and the major telephone provider at the time − thought would be a huge market once people discovered they could see as well as hear one another over the existing twisted pair phone lines. The initial imaging device considered was the standard commercial vidicon with an antimony trisulfide coating on glass accessed by an electron beam as the photosensitive element. This device was notorious for its lack of robustness and longevity. Instead, we were working on a tube with a silicon slice with millions of diodes placed on the surface, which were accessed by an electron beam in a similar manner. The silicon was an improvement; however, we strongly hoped to find an all solid state device to eliminate the electron beam.

The third field being vigorously pursued was MOS transistor technology. This was the technology which finally became the basis of all modern inte-grated circuits. The Metal Oxide Semiconductor capacitor which forms the gate of the MOS transistor is shown in Figure 3. Basic studies of this structure were also being carried out in my department.

To understand the atmosphere at the time, one must look at the structure of the Electronics Division at Bell Labs, which was under Vice President Jack Morton. This was split into two divisions, one for semiconductor devices and one for all others. Bill Boyle was Executive Director of the semiconductor part and I was a Department Head under him. Jack Morton was anxious to speed up the development of magnetic bubbles as a major memory

94

technology and there was talk of transferring resources from Bill’s division to the other where the bubble work was being done. For this not to happen, Morton demanded that Bill’s division come up with a semiconductor device to compete with bubbles. To address this demand, on October 17, 1969, Bill and I got together in his office. In a discussion lasting not much more than an hour, the basic structure of the CCD was sketched out on the blackboard, the principles of operation defined, and some preliminary ideas concerning applications were developed. The train of thought evolved as follows.

Figure 3. The basic MOS structure.

First, the semiconductor analogy of the magnetic bubble is needed. The elec-tric dual is a packet of charge. The next problem is how to store this charge in a confined region. The structure which came to mind, of course, was the simple MOS capacitor shown in Figure 3. Charge can be introduced into this depletion region with the amount of charge stored being the magnitude of the signal. To understand this better, a plot of electron energy vs. distance in-to the structure is shown in Figure 4. As charge is introduced into the deple-tion region, the potential at the surface rises until the maximum allowable charge is reached. Any further charge added will flow into the substrate.

95

Figure 4. A plot of electron energy vs. distance through the MOS capacitor.

The last problem was to shift the charge from one site to another, thereby allowing manipulation of the information. This is solved by simply placing the MOS capacitors very close together as shown in Figure 5, one with charge and the second empty. In order to pass the charge from one to the next, one simply applies a more attractive voltage to the second, causing its depletion region to overlap the first and the charge to flow along the surface to the silicon-silicon dioxide interface of the second capacitor.

Figure 5. Two closely spaced MOS capacitors.

96

The original structure using this mechanism to make a shift register is shown in Figure 6. Many MOS capacitors are placed closely together in a row and connected to a three phase voltage source. The top figure shows the storage phase with Va applied to one set of electrodes and a smaller rest voltage Vb applied to the other two. One site has charge and the other has none. The second figure shows the transfer phase where a larger voltage, Vp, is applied to the adjacent plates to transfer charge from one to the next. The last two show resetting the voltages to the initial state with the charge information shifted by one site. This is continued to an output device at the end of the row in order to read the stored information. Many other storage and transfer schemes are possible. The charge can be injected into the device electrically at the beginning of the row making a shift register or supplied by light inci-dent on a structure with empty cells. Then the amount of charge which ac-cumulates by the absorption of photons is determined by the intensity of the light, and the resulting charge pattern can then be read out in shift register fashion after a suitable integration time.

Figure 6. The basic CCD structure.

97

This completed the basic invention. It should be stressed that the basic unit of information in the device was a discrete packet of charge and not the volt-ages and currents of circuit based devices. The CCD is indeed a functional device and not a collection of individual devices connected by wires.

Figure 7. The first CCD device.

Finally, it was decided to go ahead and fabricate a device to show experi-mental feasibility. In less than a week, masks were made and devices were fabricated and tested. This first simple structure is shown in Figure 7. Charge was introduced in the first MOS capacitor by thermal generation and then transferred to the output by applying voltages to the plates where it was de-tected by pushing the charge into the substrate and measuring the substrate current.

98

Figure 8. The first integrated device.

The first device was very crude but charge transfer was successfully demon-strated and this was followed by the first integrated structure, which is shown in Figure 8. This device had a three-phase metallization and diffused input and output successfully demonstrating that it could be operated as a serial memory, the first driving force of the invention. It is no surprise that we tried to use the eight bit CCD as a linear scanning imaging device and the first rather crude image is shown in Figure 9.

Figure 9. Oscilloscope display of the output of the 8-bit device used as an imager.

99

Figure 10. Traps at the Si-SiO2 interface.

Following the initial experiments, it was evident that the main problem with the device was charge transfer inefficiency, the inability to transfer all of the charge from one element to the next. The main reason for this was the trap-ping of charge in traps at the silicon-silicon dioxide interface, see Figure 10. The trapped electrons would emit at a later time causing smearing of the im-age. So Bill and I got together again and invented the buried channel CCD, which placed the stored charge in the interior of the semiconductor where there was relatively little trapping. The structure is shown in Figure 11 where a lightly doped n layer has been added to the original structure. Once the layer has been depleted by transferring the electrons to the output diode, the resulting potential is shown. Electrons in the channel region will now accumulate in the valley created in the interior of the silicon and kept away from the surface traps.

100

Figure 11. The buried channel CCD.

101

Figure 12. An area imaging CCD geometry.

A period of rapid development followed both at Bell Labs and other com-panies. One major activity was to make an area imaging device for video applications. Many different schemes were devised. The one we chose for the Picturephone application is shown schematically in Figure 12. Linear CCDs are formed side by side and split into an imaging and storage area. During a frame time, the image is taken in the upper region and then trans-ferred rapidly to the lower where it is read out a line at a time by the serial readout while the next frame is being taken. The chip that we made for the Picturephone is shown in Figure 13 along with a self-contained experimental camera. Successful testing of the device is shown in Figure 14.

102

Figure 13. The CCD Picturephone imaging chip and a self-contained camera.

Figure 14. W. S. Boyle and G. E. Smith, 1970.

103

Figure 15. An early linear imaging CCD.

Another device that we developed in the early days was a linear scanning device shown in Figure 15 for possible use in facsimile type applications. Many other companies initiated strong development efforts after the initial announcement of the invention was made in early 1970 and the field rapidly became very active. An aerial photo taken with a linear scanning device made by Fairchild is shown in Figure16.

104

Figure 16. An aerial photograph taken with a Fairchild linear CCD.

Table 1. Advantages over photographic film

90% Quantum Efficiency (Film is 5%)Long integration times (at 160°K) 20 electrons/hr/pixel)Large dynamic range105 electrons/pixelRepeatability eliminates systematic errors

Perhaps the most impressive scientific applications were in astronomy, where the impressive advantages over using photographic film listed in Table 1 led to many new discoveries. An early four mega pixel CCD for use in an astro-

105

nomical telescope is shown in Figure 17. A comparison between a state of the art photographic image, Figure 18, compared to an image taken by a CCD, Figure 19 of the same region, exemplifies this. Such improvements led to such things as experimental verification of dark matter in the universe.

Figure 17. An early four 1 Megabit CCD camera for an astronomical telescope.

Figure 18. A film photograph of the sky.

106

Figure 19. A CCD image of the same portion of the sky.

In summary, CCDs were born in the Si-SiO2 revolution and, because of their unique properties, created their own revolution in widespread imaging device applications listed in Table 2.

Table 2. Imaging device applications of CCD.

Digital camerasTV camerasScannersMedical devicesFax machinesBar code readersSatellite surveillanceEarthly surveillanceAstronomy

107

REFERENCES

1. W. S. Boyle and G. E. Smith, “Charge Coupled Semiconductor Devices,” Bell Sys. Tech. J., 49, 587, (1970). 2. G. F. Amelio, M. F. Tompsett, and G. E. Smith, “Experimental Verification of the Charge Coupled Device Concept,” Bell Sys. Tech. J., 49, 593, (1970). 3. W. S. Boyle & G. E. Smith, US Patent 3,792,322 – Buried Channel Charge Coupled Devices, issued February 12, 1974.

Portrait photo of Dr. Smith by photographer Ulla Montan.