Slashdot Mirror

← Back to Stories (view on slashdot.org)

Primers for Semiconductor Physics?

Posted by Cliff on from the quickly-brushing-up-on-those-rusty-skills dept.

mactom asks: "Hello, I am a physics engineer with a background in lasers (non semiconductor), but during the last years I have slipped into a job in semiconductor technology development. We define the manufacturing technology and how the transistors, diodes and other devices are to be designed and manufactured on the silicon level. First I did lithography only but now I am involved in layout and design of devices and in the whole technology development. After all of this, I've discovered that I have some serious gaps in my semiconductor physics understanding! I need some suggestions for books, tutorials or even seminars (in Europe/Germany) about semiconductor physics. Yes, I have some books already, but I always have the feeling that I miss something important when I do some self studying. So, I need some 'semiconductor for dummies' books or seminars.Any suggestions?"

36 comments

  1. GE Transistor Manual by HotNeedleOfInquiry · · Score: 1

    Has been out of print for years but is available on ebay. I don't know about a Deutsch version.

    The first 20-30 pages is a very good introduction to basic semiconductor physics as it was known in the 60's. I first read it when I was in high school.

    --
    "Eve of Destruction", it's not just for old hippies anymore...
  2. Britney's Guide to Semiconductor Physics by 74X0R · · Score: 5, Informative

    Try Britney Spears' illustrated primer, here.

    1. Re:Britney's Guide to Semiconductor Physics by Solder+Fumes · · Score: 1

      Indeed. I always thought my engineering textbooks would have been much more appreciated if they had a hot chick on every page. Every class would have been like that one with the hot teacher!

    2. Re:Britney's Guide to Semiconductor Physics by DrLudicrous · · Score: 3, Informative
      LOL, you beat me to it. And the best part is that a lot of it, if not all of it is correct. Like the opening page, that actually is a correct formula for the density of states in 3 dimensions (different in 1 and 2 D). Honestly, this is probably one of the better sites out there. I love the background for photonic crystals- it's dead on. Shout out to Kathy Kash at Case Western!

      The standard physics text for intro solid state is Kittel. I would avoid this. Try Marder's text, Condensed Matter Physics. It is heady stuff, but if you want to dig deep it is a good source. An old classic for profs and grad students is Ashcroft & Mermin's text (something something solid state something).

  3. Do take a look at.... by Anonymous Coward · · Score: 1, Funny

    Be sure to check out "Semiconductor Production Nanotlithography Praxis and Theory for Idiots." I believe I saw a copy at Walmart for $9.99.

  4. Re:Oooh Look at me, I'm a Physics Engineer... by dhakbar · · Score: 0, Troll

    I think you mean that you make... "lasers."

  5. Re:Oooh Look at me, I'm a Physics Engineer... by RobertB-DC · · Score: 2, Funny

    And I make lasers...

    Friggin' lasers, you insensitive clod!

    --
    Stressed? Me? Of course not. Stress is what a rubber band feels before it breaks, silly.
  6. Try Solid State Electronic Devices by ikeleib · · Score: 3, Informative

    The author's a complete dickhead, but the book is short and gets the point across. The material can by quite boring and the author doesn't exactly spice it up. Anyway, it's by Streetman and Banerjee; amazon it.

    Also there are a few graduate course materials at MIT's courseware that may be helpful.

    1. Re:Try Solid State Electronic Devices by ufnoise · · Score: 2, Informative
      Streetman is a standard text and pretty much covers the whole range of an undergraduate device physics course from an EE perspective. It covers topics including the manufacturing process, quantum mechanics, and the basic analysis of electronic devices. A basic knowledge of calculus is helpful. The standard approach of this books is to:

      1. start with basic concepts in quantum mechanics and materials

      2. find a correspondence to classical physics (F=ma) and electrodynamics (F = q E)

      3. develop a basic intuition of how these devices work by applying classical physics ideas the analysis

      I would start with Streetman's book and look up the references in chapters which are of interest to you.

  7. call me stupid, but by nes11 · · Score: 0

    wouldn't "semiconductor for dummies" be on the same shelf as "brain surgery for dummies" & "rocket science for dummies"??

  8. Mead and Conway... by PaulBu · · Score: 4, Informative

    No, really! That was one of the (http://www.amazon.com/exec/obidos/tg/detail/-/020 1043580/qid=1091570616/sr=1-1/ref=sr_1_1/102-97293 58-6063327?v=glance&s=books)
    books that actually started the whole field, and they had to explain what the whole semi technology was to the "non-initiated" physicists.

    I know how you feel, I've been trying to make a similar transition from superconductor electronics back into the mainstream, and whereever I come I can start from discussing if the place uses classic Mead-Conway colors for their layouts, of if not why not? ;-) And yes, people do relate to that! ;-)

    Another good book which an older friend of mine swears by is the Andy Grove's book on semiconductor processes (more oriented towards fab than design), can not get the link now (maybe it is out of print), but the guy who had started Intel can not be TOO wrong, right?

    Gray and Meyer wrote (http://www.amazon.com/exec/obidos/tg/detail/-/047 1321680/qid=1091571023/sr=1-1/ref=sr_1_1/102-97293 58-6063327?v=glance&s=books)
    this book, my new boss suggested it for me to read (took it off his bookshelf and it takes its well-deserved position on my desk now, I move it around once in a while to indicate the fact that I am checking it out ;-) ), more of analog design with transistors stuff, like how would one build a really good amp and what it takes to design that without computers, SPICE and everything.

    Make your pick! ;-)

    Paul B.

    1. Re:Mead and Conway... by Anonymous Coward · · Score: 0

      Do they also relate to the way you use the word wherever when you actually mean where? Pardon me if you used the word correctly; to imply that you do not, in fact, know where you come from.

  9. a better tactical deployment by nusratt · · Score: 1

    "...physics engineer...lasers...serious gaps in my semiconductor physics understanding..."

    hmmm... is there any possible way that I could interest you in a job with Bin Laden Engineering, Inc.? ;-)

    --signed,
    Wile E. Coyote

  10. you might have to email someone and ask nicely by foog · · Score: 2, Informative

    but see if you can find pdfs of whatever they're using for APh 9 at Caltech these days.

    They whip freshmen through most of what you'd learn better in a junior-level class on semiconductor physics and have them fab their own devices in a simple but effective laboratory. Carver Mead designed the course. Fun class!

  11. You can't make up this stuff by nusratt · · Score: 4, Funny

    Absolutely on the level...
    I went to the Amazon URL provided for Mead & Conway in a prior post.
    Halfway down the page, it says:
    "Customers interested in Introduction to VLSI Systems may also be interested in:

    Free service to meet singles"

    Boy, their artificial intelligence systems are a lot sharper than I imagined!

  12. Un-recommendation by l00sr · · Score: 1

    I can un-recommend a text I had in class, Semiconductor Physics and Devices, by S. Banerjee et. al. Sorry :( OTOH, I can recommend Britney Spears' guide to semiconductor physics for a chuckle, if nothing else.

  13. Re:Oooh Look at me, I'm a Physics Engineer... by Anonymous Coward · · Score: 0

    But what about the sharks!

  14. solid state physics by ppcollins · · Score: 0, Troll

    You first need have a basic understanding of quantum physics and relativity. then look up solid state physics, find some of the Richard Feynman lectures and you should be on your way. Its not easy and takes some hard thinking to understand it all. don't have a name for a book that includes all of solid state physics only "The Elements of Physics" - Grant and Philips, Oxford publications contains a decent amount to get you back on track. P Collins B.Sc

    1. Re:solid state physics by ufnoise · · Score: 1
      Mosts of the standard texts, like Streetman, start out with some basic quantum mechanics and then rapidly work their way to using classical physics analogues. If you are doing device technology, or process integration, I assume your company already understands how these devices work. You are just trying to optimize the behavior through the manufacturing process.

      While quantum mechanics is the fundamental basis, a classical physics approach helps to develop an intuition. This is especially useful when it is the parameters in manufacturing process, not the understanding of quantum physics, that is dominating the device performance.

  15. Semiconductor Optoelectronic Devices by Jazzer_Techie · · Score: 3, Informative

    Author: Pallab Bhattacharya
    Title: Semiconductor Optoelectronic Devices, 2nd ed.
    Publisher: Prentice-Hall, NJ, 1997

    This is a good work. It's a grad textbook that starts simple and then becomes more involved. From the work you describe, I think it would be extremely applicable and helpful.

  16. I recommend Simon Sze by auferstehung · · Score: 3, Informative

    His books contain a wealth of information. Specifically, Physics of Semiconductor Devices, although Semiconductor Devices: Physics and Technology, 2nd Edition may also be useful. Used copies (or the library) won't set you back too much.

    --
    Logic is not Divine.
  17. Sze, Wolf, and Grove by jralls · · Score: 3, Informative

    I made something of a similar journey 15 years or so ago, moving from nuclear engineering (as in reactors) and submarine construction to working in a wafer fab.
    S. M. Sze's "Physics of Semiconductor Devices" is something of a standard which I saw on many bookshelves. It's something of a survey, touching on all of the technologies without really going into depth on any.
    Stanley Wolf's 3-volume set "Silicon Processing in the VLSI Era" is encyclopedic on processing and modelling (and the modelling part goes into the physics in a lot of depth) of CMOS and related processes.
    And my boss used to rave about Grove's book, but I've never seen a copy.
    This is all "vertical" stuff. If your interests are "horizontal" (that is, laying out the circuits rather than building up the devices) then none of it will work for you.

  18. Good book for both CMOS and bipolar device physics by Slack(er)ware · · Score: 4, Informative

    A good book that covers device modeling and design for both CMOS and bipolar VLSI devices is "Fundamentals of Modern VLSI Devices" by Yuan Tar and Tak Ning. It doesn't cover fabrication much but it does a great job of introducing the various electrical effects that modern process integration engineers have to deal with. For a more process-oriented approach there is always Andy Grove's book "Physics and Technology of Semiconductor Devices."

  19. Helpful references. by Christopher+Thomas · · Score: 3, Informative
    I can't point you towards physics references, but I can point you towards texts about modelling and using these devices:

    • Microelectronic Circuits, by Sedra and Smith.
      This is the semiconductor devices bible for electrical and computer engineering (in North America, at least).

    • Analog Integrated Circuit Design, by Johns and Martin.
      Excellent book recapping device behavior and describing analog circuits and the issues that come up when you're trying to integrate elements on a die.

    • Principles of CMOS VLSI Design, by West, Eshraghian, and Smith.
      This also covers layout and circuit issues, though mostly for digital logic design.


    If you want to do digital logic intelligently, or design analog circuits that do their job with precision and effectiveness, you need to go back to school and get a Comp Eng degree. If you want an idea of how engineers use the devices you're trying to optimize, and what factors are important for usability and performance, these books will do.
    1. Re:Helpful references. by jrockway · · Score: 1

      > get a Comp Eng degree

      Isn't that what EE is?

      --
      My other car is first.
    2. Re:Helpful references. by norkakn · · Score: 1

      nah, at least here (University of Michigan) it is about 90% of EE with some computer specific electronics thrown in.

    3. Re:Helpful references. by Christopher+Thomas · · Score: 1
      get a Comp Eng degree

      Isn't that what EE is?

      Depends on the university. About 10-20 years ago, this was true. Right now, around here at least you have EE, CS, and CE as distinct disciplines:

      • EE deals with mostly analog systems, and mostly systems build from discrete components (e.g. power systems, high power RF systems, control for industrial systems, etc.).

      • CS deals with information theory and theory of computation. These are the people who design new programming languages and work on AI and so forth.

      • CE deals with highly-integrated electronics, and the software written to drive it. Focus is more on digital circuits than EE, and more on existing coding techniques for existing or nearly-existing devices than CS.


      At least, that's what it was like when I went through. The flavour of the CE program, and much of its curriculum, will depend on whether it grew out of the EE department, the CS department, or both at the university in question. Likewise, the nature of the EE and CS departments will change depending on how much they offload to CE (EE at a school without a CE department will have a lot more CE material, for instance).

      Your mileage may (still) vary :).
  20. Ashcroft & Mermin by doru · · Score: 2, Informative

    I'm surprised nobody cited Ashcroft and Mermin's "Solid State Physics". If you want to learn about the basics, this is an excellent place to start.

  21. Semiconductor Device Book by Anonymous Coward · · Score: 1, Informative

    I would agree that Sze's "Physics of Semiconductor Devices" is a standard text. However it is rather thick. He also wrote a shorter introductory text which I like very much called "Semiconductor device Physics and Technology". Finally, he wrote/edited a book on ICs called "ULSI Technology" which covers all the manufacturing processes in detail and has loads of references.

    Another introductory text that has more discussion that could be beneficial for self-study is Ben Streetman's "Solid-state Electronic Devices".

    If you are interested in the latest developments in MOSFETs, Professor Walter Hansch has a very good course at the Technische Universitaet Muenchen called "Advanced MOSFETs and novel devices". At the same university there is the "Walter Schottky Institute" that has a seminar on current topics of semiconductor physics (Schottky seminar) - often the topics are concerned with applied physics, such as semiconductor lasers.

  22. My PhD Bookshelf by ofey · · Score: 2, Informative

    I'll just point you in the direction of myAmazon Listmania.

  23. From http://www.avtechpulse.com/faq.html/VII/ by DispassionateObserve · · Score: 1

    • Gerold W. Neudeck, "The PN Junction Diode", 2nd ed., Vol. 2 of the "Modular Series on Solid State Devices", Addison-Wesley. 1989. TK7871.86.N48. This short book has the best diagrams illustrating electron/hole movement in the PN junction that I've seen. Very concise. Highly recommended, if you are interested in diode physics.

      In fact, the entire Modular Series is worth owning. There are books are semiconductor fundamentals, bipolar transistors, FETs, fabrication and manufacturing, "advanced fundamentals", advanced MOSFETs, quantum mechanics, and carrier transport.

    • Ben G. Streetman, "Solid State Electronic Devices", 3rd ed., Prentice Hall, NJ. 1990. TK7871.85.S77
    • A good introduction to semiconductor physics, for electrical engineers.
  24. I worked for Cadence and... by Invisible+Now · · Score: 2, Informative
    I found UC Berkeley's Extension semiconductor coursework very helpfull. Ranges from great Intro to Fab to Discrete Device Design and layout, etc. One or two nights a week for ten weeks with really good instructors...

    Anyway, Good luck. I think this is the single toughest tech skillset out there - particularly Physical Verification of chip design.

    So, if you live in the Bay Area check out UCB Extension...

    --

    "Knowing everything doesn't help..."

  25. CIA arriving at nusratt's house... by Anonymous Coward · · Score: 0

    in 3 ... 2... 1...

  26. Radioshack by scorp1us · · Score: 1

    Oddly enough, they carry the best primers. They have a series of books called "Basic Electronics" "Communication Electronics" and "Digital Electronics"

    All are an easy read, and very clear with good pictures. They make it really easy to understand.

    They are ~8x11 and are thin paperbacks. Absolutely great. I think even kids can understand (assuming they can handle the terminology)

    --
    Slashdot's rate-of-post filter: Preventing you from posting too many great ideas at once.
  27. Link to books: by scorp1us · · Score: 1

    all three

    --
    Slashdot's rate-of-post filter: Preventing you from posting too many great ideas at once.