The tech you're reading these words on – you have two Dundee uni boffins to thank for that

Spear and LeComber stumbled on the thin-film-transistor liquid-crystal display 40 years ago

Got Tips? 61 Reg comments

Every time you use a smartphone, glance at your smart watch, fire up a computer, watch TV or endure a PowerPoint presentation, you experience a little bit of Dundee.

The flat-panel technology we use in modern devices wasn't invented by megacorps in Japan or Silicon Valley but by a pair of academics in Scotland's fourth-largest city. And their invention – amorphous silicon thin-film field-effect transistor switches for liquid-crystal displays (better known as TFT-LCD) – has stood the test of time at 40 years and counting.

Given the impact their invention has made on everyday lives, and not just in the developed world, you might have expected Walter Spear and Peter LeComber to enjoy local hero status alongside such celebrated British tech engineers as Tim Berners-Lee, John Logie Baird, Alexander Graham Bell or Charles Babbage. Yet there is no public monument to this pair of Dundee boffins, no banknote – not even so much as a Hollywood biopic starring Benedict Cumberbatch and Tom Hiddleston.

Until recently, that is. The movie option is still on hold but in the meantime, IEEE – the Institute of Electrical and Electronics Engineers, the people who brought you IEEE 802.11 (ie, Wi-Fi) – unveiled a commemorative bronze Milestone plaque at the University of Dundee just after the Easter weekend, dedicated to the invention. It is due to be installed permanently at the doorway to the Harris Building where Spear and LeComber worked within the physics department during the 1970s and 1980s.

Walter Spier L Peter LeComber R photo courtesy University of Dundee Archive Services

Walter Spear (left) and Peter LeComber at their physics department laboratory, photo courtesy of University of Dundee Archive Services

Why are Spear and LeComber so woefully unsung by a nation that usually lacks no such modesty about its contributions to the world? As Edison's contemporaries discovered, often ruinously, it comes down to who successfully registers the patent and noisily claims the credit in public. Everyone loves a showman; studious people wearing lab coats are ignored.

Walter Spear was born in Frankfurt in 1921 but his father, being Jewish, moved the family to London in the nick of time in 1938. He studied physics to PhD in London before becoming a lecturer at University College, Leicester, and eventually taking the position of Harris Chair of Physics at the University of Dundee in 1968. Soon after his arrival in Scotland, he invited Peter LeComber, who he worked alongside when the latter was a junior lecturer in Leicester, to join his new research team in Dundee.

LeComber was born in Ilford, Essex, in 1941 and went on to study physics at Leicester. After completing his PhD, he worked for a few years in research at Purdue University in Indiana before returning to Leicester University as a physics lecturer in 1967. He joined Spear at the Carnegie Laboratory of Physics at the University of Dundee two years later.

What they achieved over the following decade alone was, in the words of Iain Stewart, the university's current dean of Science and Engineering, "absolutely astounding".

Back in the 1960s and 1970s, the ground was more fertile for those in academia wanting to develop the seeds of original ideas. Difficult though it is, try to picture a time when experts were treated with respect and encouragement, not blamed or mocked by politicians holding the purse strings of public investment.

The story begins in 1963, when Labour Party leader Harold Wilson tried to excite voters with talk about igniting a "white hot technological revolution" that would transform post-war British industry and lead to economic progress. When he formed his first Labour government the following year, he put this manifesto pledge into practice, creating a Ministry of Technology and putting Anthony Wedgewood-Benn in charge and latterly the now-notorious (fraudster/death-faker/Czech spy) John Stonehouse MP.

The new ministry took over a number of existing government-funded technical operations, one of which was the Radar Research Establishment in Malvern. On a visit there in 1967, Stonehouse was aghast to learn from RRE's director that the money in royalties it had paid to the Radio Corporation of America (RCA) for licensing its cathode ray tube shadow mask technology in its colour displays exceeded the sum being spent on developing the supersonic passenger jet Concorde.

True or not, the claim prompted Stonehouse to ring up the very next day to approve the launch of a programme to develop an alternative homegrown display system, preferably with a flat screen, to replace shadow mask. RRE's director called in one of his senior managers, a certain Cyril Hilsum CBE, to ask what could be done to produce a flat panel television.

"I said we could make a few LEDs to show people," recalls Hilsum. "He said he didn't think that would be good enough. So we set up a working party and came up with some suggestions, one of which was to work on liquid crystals."

Hilsum is an enduring character in the story of British electronics, what Americans might call a "rock star of seminal technology". He spent most of his career at the Ministry of Defence before becoming director of research at GEC, and subsequently visiting professor in physics at University College London. Now aged 92 but with as much sign of showing it as a Duracell bunny, he has not just seen it all but been an active participant.

By 1970, his display group at RRE was receiving funding from the Ministry of Defence, which wanted a flat display for a portable radar unit, and was also working in the civil sphere because the Department of Technology had morphed into the Department for Trade and Industry.

"To get a flat panel display, you need three things: some way of producing an optical signal from an electrical impulse, a material that can demonstrate this effect, and a way of turning that effect into a picture.

"We had the first in 1971, the twisted nematic effect, invented in Switzerland. In 1972, we had a stable crystal material, the biphenyl – a real breakthrough invented at Hull University with modifications at Malvern. We knew it was good because it was soon being used for digital watches and simple numeric instruments. But we were no nearer the addressing circuit technique which would give us pictures."

This third technology is what Spear and LeComber came up with, somewhat unexpectedly.

Cyril Hilsum CBE, photo copyright Alistair Dabbs

Cyril Hilsum CBE, photo by Alistair Dabbs

To understand the challenge, imagine you want to show four numbers on a mid-1970s digital wristwatch. With seven bar patterns per number, you only need 28 connections; add a few other things on the watch display and you still only come to around 50. But for a TV screen, you'd need millions.

Even if you ran wires running orthogonally behind a display along X rows and Y columns, with half the voltage in one and half in the other, you'd still not operate one pixel element at a time: you'd produce a cross shape. Better would be to have your X and Y lines addressed with transistors to handle a threshold – not like a transistor on a chip but one that covers an entire screen. So, how about using silicon?

"If you try to evaporate and deposit silicon over a screen," notes Hilsum, "what you get is a useless mess. Some people had already tried putting thin film transistors down over a screen but they were made from compounds such as cadmium sulphide (CdS) and cadmium selenide (CdSe). Apart from being poisonous, cadmium isn't stable. We'd have liquid crystal which would last virtually forever, knowing that the thing over it was going to change within a few months."

As for silicon (Si), it is very difficult to make: you have to grow a crystal with atoms very regularly placed to allow electrons to move through the material. But if you just take away the oxygen from sand, which is easier, you get lumps of what's called amorphous silicon (a-Si), which can be evaporated and applied as a thin film over surfaces.

In amorphous silicon, however, electrons tend to sit in the unsaturated bonds that stick out, and they refuse to move. This makes it a very bad conductor; Spear and LeComber's team were researching its properties as an insulator.

Suddenly, almost by accident, the group in Dundee made a breakthrough. They decided to try doping the material as you did with crystalline silicon, while adding hydrogen, to find out exactly why it didn't conduct, only to find that it did.

The performance of electron flow wasn't great, being roughly 1,000 times less efficient than crystalline silicon. It was essentially a really bad transistor. So what Spear and LeComber did was make a p-type conductor and put it together with an n-type conductor to make a simple p-n junction.

"It is one of the simplest circuits you can get in the world," says Ian Underwood, professor of Electronic Displays at the School of Engineering and Electronics, University of Edinburgh. "In each pixel, there is one transistor for one capacitor. The transistor acts as a switch, allows the signal in, and then switches off and isolates it and holds it. This is beautiful."

It's a far cry from passive-matrix LCD displays, which Underwood describes as "the most extreme demultiplexing scenario in the world". It means you can put the drive signal onto the liquid crystal you want and leave it there. "The relative simplicity of the waveforms compared with passive matrix addressing is one of the legacies that has lasted, unlocked by amorphous silicon TFT."

In 1975, they published a paper about their experiments with amorphous silicon. The following year, they wrote a follow-up paper describing how they could control the doping method and make a junction.

They realised they could make devices with it but, being academics rather than businessmen, they knew they needed some advice. The first thing they thought of was making a solar battery, and Spear wrote to the National Research Development Corporation (NRDC), an organisation set up by an earlier Labour government to help universities and government departments get their ideas commercially developed.

NRDC then forwarded the letter to Hilsum.

Hilsum suggested concentrating on thin-film transistors rather than solar batteries, the latter already being a very competitive market sector by that time. And so began several years of an increasingly close partnership between the Dundee team and Hilsum's group, and the birth of modern flat panel displays as we know them today.

In his first letter back to Hilsum, Spear writes that in principle it's very easy to make a thin-film transistor, describing it as being like "turning on a tap". It turned out to be a little more difficult than that: it took another 18 months and Hilsum didn't receive a first sample from Dundee until November 1978. Once Hilsum had confirmed that it switched with enough voltage applied, the project was flying.

The team's key paper, Amorphous Silicon Field Effect Device and Possible Application, was published in March 1979. It includes the amusingly modest statement that TFT "may form a useful switching element in a display panel". Of course, they knew damn well it would: they'd been working with Hilsum for at least six months on just this. But Spear and LeComber were by then aware how competitive things could be in the commercial world and they were playing it quiet while pursuing a patent application – something they did not succeed in achieving.

RCA tried to register the amorphous silicon TFT patent for itself, its application cheekily (or blatantly) making specific reference to Spear and LeComber's work. RCA failed too.

Spear TFT diagram Image courtesy University of Dundee Archive Services

Early diagram of amorphous silicon TFT drawn by Spear, image courtesy of University of Dundee Archive Services

"I was very sorry that we were unable to get a patent but we were able to ensure that their careers benefited considerably from it," says Hilsum. "Both became fellows of the Royal Society. Walter was lauded everywhere [he retired in 1988, succeeded in the Harris chair by LeComber]. It was a great tragedy that Peter died when he did [in 1992 of a heart attack] because he was just rising up in general acclaim and was beginning to get contracts. There is no question that he would have been much in demand as the expert in amorphous silicon."

As University and IEEE representatives drew back the curtain to reveal the bronze Milestone plaque last week, the witnesses who gathered in Dundee to pay their respects included not just Hilsum, Underwood and a veritable army of young and old electronic engineers, but also Peter LeComber's widow and family. For them, recognition comes late but is welcome.

Above all, it's striking to realise how Spear and LeComber's invention remains the dominant flat-panel technology even after all this time.

"How many electronic devices do you know that are still in the same format after 40 years?" challenges Hilsum. "We discovered it wasn't only useful for liquid crystals, it actually worked for the more modern OLEDs that we use today. Amorphous silicon TFT stands proud as the same device that Walter and Peter made in 1978. Three billion mobile phones, 300 million computers, 230 million TVs, 70 million vehicle displays, 10 million projection displays; signage, navigation, medical instruments... all using amorphous silicon TFT.

"It's an amazing and long-lasting legacy." ®

Sponsored: Webcast: Ransomware has gone nuclear


Biting the hand that feeds IT © 1998–2020