History of Semiconductor Engineering

Mainly for Geeks?

In recent years a number of good accounts of the early history and development of the semiconductor industry have appeared, notably ?Crystal Fire? [1] and ?The Man Behind the Microchip? [2]. These books give excellent accounts of the development of the transistor and integrated circuit respectively without going into too much technical detail.

Lojek?s book sets out with two objectives: to provide a detailed account of the technical development of semiconductor manufacturing techniques and to record the contributions made by a host of engineers whom Lojek clearly feels were overlooked or under-represented in previous accounts of the industry.

History of Semiconductor Engineering is an extremely well researched book. Lojek has clearly amassed a huge archive of documents and photographs, many of which have not been previously published and he uses this material to good effect. His account of the development of the transistor from its origins in Bell Labs through Shockley Semiconductors and Fairchild emphasises the huge task faced by the engineers involved. Not only did they have to gain understanding of the workings of this new device but also had to invent, often from scratch, the manufacturing equipment and techniques required to bring the devices to reality in a form in which they were reproducible and reliable enough to perform real-world tasks.

The book is interspersed with images from lab notebooks and process flow diagrams which provide the reader with a clear thread of how the semiconductor industry developed from rather crude beginnings to a technology unsurpassed at manufacturing huge numbers of highly complex products at ever-decreasing cost. His account of the development of the Fairchild planar process, arguably the most important technical advance in the history of the industry is particularly fascinating.

In parallel with the accounts of technical developments Lojek records the activities of numerous players, both at engineering and management level such that the reader gains a flavour of the frenetic atmosphere in these fledgling companies. The two themes, technical and human come together most successfully in the complete chapter Lojek devotes to Robert Widlar, the brilliant maverick circuit designer who pioneered new techniques required to successfully implement linear integrated circuits, first at Fairchild and latterly at National Semiconductor. Lojek here gives a very clear insight into how Widlar?s creative genius overcame the severe limitations imposed by silicon processes in terms of the restricted range and relatively poor performance of passive electronic components that can be implemented on silicon.

Lojek frequently seeks to elevate the contributions of lesser-known engineers in the industry. This is a laudable objective but sometimes he seems to go over the top, as in a lengthy passage in which he seeks to downgrade the significance of the integrated circuit patent awarded to Robert Noyce at Fairchild in 1959 by providing a lengthy list of engineers whose previous efforts had provided vital grounding to the invention. This is not entirely convincing: many brilliant inventions can with hindsight be seen to have been natural developments from work that went before them but they nevertheless needed an inspired individual to see the possibility that hadn?t been spotted by others.

Unfortunately, despite its fascinating content, History of Semiconductor Engineering is not an easy book to read. Lojek?s first language is not English and this is evident in the stilted style of writing throughout the book. The material is often not well organised with abrupt jumps in chronology or subject matter that frequently leave the reader adrift. At one point (p 37) an unfortunate juxtaposition of sentences lead this reader to the erroneous conclusion that Jack Morton, a former manager of Bell Labs, was murdered by members of the American feminist movement ? only a careful rereading of the paragraph showed this not to be the case. Much of the blame for this must be laid at the door of the publisher: the book appears to have undergone no editing whatsoever. It is hard to understand how a long-established, reputable publisher like Springer could have failed one of its authors so badly.

The lack of editing and high technical content of History of Semiconductor Engineering probably make this book less well suited to the general reader than other accounts of the industry. However, for those willing to put in the effort, it offers insights not available elsewhere.

1. Crystal fire: The Birth of the Information Age ? Riordan and Hoddeson 1997
2. The Man Behind the Microchip: Robert Noyce and the Invention of Silicon Valley ? Berlin 2006