Who Will Win the 2014 Nobel Prize in Physics?

[voko]

ENIAC like computers, abacuses and mechanical Turing machines are all fine and dandy, but without the transistor there would be no Physics Forums, no cell phones that you can stick in your pocket, no Google, no high speed Internet; at least nothing technological, fast, small and light, as we've come to know it today.

That is jumping to conclusions again. Re-read the final sentence of my message that you quoted.

 

[zoki85]

Well, the transistors are not used exclusively in computers, cell phones or similar devices. Far from that. You guys are funny :)

 

[voko]

A more technical paper on the prize: http://www.nobelprize.org/nobel_prizes/physics/laureates/2014/advanced-physicsprize2014_2.pdf

It reminds me of the 2000 prize: http://www.nobelprize.org/nobel_prizes/physics/laureates/2000/advanced-physicsprize2000.pdf

It is interesting that the contribution by Nakamura et al. was mentioned in the 2000 paper. So if nothing else, the Nobel Committee is consistent :)

 

[gravenewworld]

The most hilarious part about this is that the development of LEDs has more to with chemistry than the prize awarded for chemistry, which was for the development of super resolution microscopy. I think physics and chemistry got their nobel prizes mixed up.

 

[Borek]

he most hilarious part about this is that the development of LEDs has more to with chemistry than the prize awarded for chemistry, which was for the development of super resolution microscopy. I think physics and chemistry got their nobel prizes mixed up.

The question "how does the committee define physics and chemistry?" crossed my mind earlier today :)

 

[Medicol]

I was thinking the bee's venom extracted medicine to fight HIV would get the prize but sadly their research has remained silent ever since the report of its "possibility" to be of help for AIDS patients.

 

[dipole]

The transistor only set quantum computing back 50 years! ;)

 

[collinsmark]

That is jumping to conclusions again. Re-read the final sentence of my message that you quoted.

I don't think that is jumping to conclusions. It's quite clear that without the use of transistors, you wouldn't be reading this thread now.
The final sentence that I quoted you in your last message is:

"And even if we hypothesise for a second we do not have semiconductors at all, we cannot really say what other technology we could have developed by now."​

Boldface mine. Hypothesize or speculate?

It's fine to hypothesize here about possible alternatives to the transistor: those based on today's technology and 20th century inventions. But I'd rather not speculate about the results of alternate timelines [sorry if I came across like that]. I'm not speculating what would have or could have happened if the transistor had never been invented [in the timeframe that it was]. Rather I'm commenting about transistor use (or lack thereof) in the here-and-now [and the transistor being invented at any point].

What I can say is that here and now, with today's technology, there is no easy or convenient substitute for the transistor, particularly technology based on 20th century discoveries. Try to build a device that would allow you to read this thread, but does not include any transistors, and you won't get far. And that doesn't even include the network and infrastructure allowing the information to get from Greg's server to the device, or even Greg's server itself.

[Edit: graphene and carbon nanotube based devices might have a promising future, perhaps, but are not a suitable transistor replacement as of today.]

Perhaps hypothetically you could build a slow, clunky thing about the size of room, and that costs more than the GDP of a small country, and that might allow you to slowly load the page, assuming the rest of the network was already in place. But if everybody were required to do that there would be cost overruns, space constraints, and where would the necessary electrical power come from?

Well, the transistors are not used exclusively in computers, cell phones or similar devices. Far from that. You guys are funny :)

Yes, a typical smartphone, for example, contains hundreds of electrical, non-transistor components (ignoring the traces -- i.e., the "wires", so to speak, that connect the components together). Several hundred even. Most are external to the integrated circuits (ICs) and are soldered onto the circuit board, but a handful or two might actually be within the ICs.

Compare that to all the transistors in a smartphone which number in the hundreds of millions. Several hundreds of millions even. [Edit: And considering the memory capacity smartphones have these days, now we're well into the billions.]

(Things get a little tricky if you consider such things as the capacitors involved in the LCD display. These capacitors are not discrete components though, they are actually made of thin film transistors. And those are in addition to the other hundreds of millions [or even billions] of other transistors elsewhere in the phone.)

 

[zoki85]

Yes, a typical smartphone, for example, contains hundreds of electrical, non-transistor components (ignoring the traces -- i.e., the "wires", so to speak, that connect the components together). Several hundred even. Most are external to the integrated circuits (ICs) and are soldered onto the circuit board, but a handful or two might actually be within the ICs.

I was rather thinking about a "little bit" bigger variants of transistors, like power IGBTs ;)
They are widely used in a conjunction with various heavy EE equipment today.

 

[russ_watters]

So, I think this argument over the importance of the transistor sort of proves dipole's point:

The fact of the matter is that the transistor and mini-transistor (IC) ushered-in the modern information age. Yes, if it wasn't invented something else could have been that also could have ushered in the information age and also could have won a Nobel Prize by being just as groundbreaking/revolutionary -- which doesn't change the point that the thing that ushered in the information age was groundbreaking/revolutionary.

But the LED light is not now and likely never will be a groundbreaking/revolutionary technology. We had electric lights for a century and efficient electric lights for decades before LEDs came around and not much changed with LEDs. LEDs are only slightly more efficient than fluorescents, so there is very little potential for them to have much of an impact.

 

[voko]

Hypothesize or speculate?

Call it whatever you want. Note that you are doing this, by saying something about what would happen if t.transistors could not exist; I merely say that we do not really know.

Perhaps hypothetically you could build a slow, clunky thing about the size of room, and that costs more than the GDP of a small country, and that might allow you to slowly load the page, assuming the rest of the network was already in place.

Look what you just did. You are saying that now, in 2014, suddenly without transistors, building that behemoth would be my only option. Let me remind you that such a behemoth would have been the outcome if somebody had tried to achieve the same goal in 1947, using the technology available. 60 years of technological advance is something that you consistently ignore in your conjectures.

 

[nsaspook]

http://spectrum.ieee.org/tech-talk/semiconductors/devices/no-nobel-for-the-father-of-the-led

Given the Nobel Foundation’s statutes (three people at maximum, no posthumous awards), it’s almost inevitable that every year, there will be people who deserve a share of a Nobel Prize that are left out.

Nick Holonyak Jr., the person widely credited with the development of the first visible-light LED, the device that now lights up countless clocks, traffic signals, and other electronic displays, might be one of them. On Tuesday, the Royal Swedish Academy of Sciences awarded this year’s Nobel Prize in Physics to three inventors of the blue light-emitting diode. Holonyak isn’t exactly complaining that he isn’t among them; his objection is that his 1962 invention has never been singled out for recognition by the academy.

“Hell, I'm an old guy now,” Holonyak http://www.miamiherald.com/news/business/technology/article2561243.html with the Associated Press. “But I find this one insulting.”

The old guy seems bitter but I kind of agree and blue leds are the most horrible annoyingly obnoxious led color of them all on devices.

 

[collinsmark]

Call it whatever you want. Note that you are doing this, by saying something about what would happen if t.transistors could not exist; I merely say that we do not really know.

Ahhg, <frustrated>. I am not saying "what would happen" if transistors could not exist, or what would have happened if transistors were not invented circa 1947.

What I am stating, quite matter of factly even, what would not happen without transistors, today, with today's technology (for reasons of practicality, cost, size, power consumption, material resources, etc.): You would not be reading this thread. You would not have access to a computer that could fit in your lap or a smartphone that could fit in your pocket.

Look what you just did. You are saying that now, in 2014, suddenly without transistors, building that behemoth would be my only option.

Yes, that is more like it. That is what I am saying. :)

Let me remind you that such a behemoth would have been the outcome if somebody had tried to achieve the same goal in 1947, using the technology available.

It would still require a behemoth but could be reduced to merely a fraction of the size of what it would otherwise be by replacing the vacuum tubes and electromechanical relays with discretely packaged transistors. (Still a behemoth, but a much smaller behemoth than what would otherwise be required.) This actually was the course of technological advancement in our history, although no one ever attempted to build anything equivalent to the computing power of a modern day smartphone with 1940's technology. What did happen (in addition to computers becoming smaller behemoths) is discretely packaged transistors almost instantaneously changed radios from being pieces of furniture to something that you could hold in your hand and carry around with you.

Since the transistor's invention, engineers, physicists and scientists have been struggling to improve upon the idea. This started by shrinking the transistor size, putting more transistors on the same substrate (integrated circuits), and repeating.

In 1965, Gordon E. Moore published a paper describing a trend: the number of transistors that can fit on a given area of substrate had increased exponentially with time, given the continuous advancements in transistor technology. This eventually became known as "Moore's law." Although the exponential time constant was modified, the exponential quality continued for decades.

But engineers and physicists knew that Moore's law must eventually come to an end. As soon as transistor size hits "the wall" of quantum mechanical limitations, transistor size will shrink no further, at least not without a serious paradigm shift. Engineers and physicists have been searching for such a paradigm shift for decades, in anticipation of Moore's law failure. Today, we are already at the point of diminishing returns. Effort has been put on better utilization of the third dimension (taller gates, stacking transistors on top of one another, etc.) with some limited success, but nothing that will keep Moore's law going as it once was. Better lithography techniques have been accomplished in part by using light sources with higher frequency (makes me think of blue LEDs for some reason), and clever use of optics (liquid lenses for example). [All good stuff, but not enough to maintain Moore's law more than a handful of years or so.]

Before moving on, let me state this clearly: This "wall" seems much, much larger for the technology of vacuum tubes and electromechanical relays. Given what we know today, one cannot squeeze hundreds of millions of electromechanical switches and/or vacuum tubes (and have them work correctly) onto an area the size of your fingernail, and at a manufacturing cost of a few dollars. This version of Moore's law would plateau at the behemoth level.

The paradigm shift of replacing the transistor altogether has not been ignored, however. Quite the contrary. Effort has been made and studies done to explore such things such as purely optical, non-transistor switches; [and separately] properties of carbon nanotubes, to name a couple of examples. Nothing has panned out just yet. But don't think that engineers and physicists have been sitting idly by, twiddling their thumbs this whole time, complacent with the transistor. They have not. Possible alternatives to the transistor have been explored for decades. Maybe with continued research and increased technological advancement, we might move from the transistor to something else one day.

Modern day engineers and physicists have one advantage that scientists circa 1947 did not: transistor based tools, such as computers, and transistor based laboratory instrumentation (something that you can fit on a lab bench rather than filling up an entire building).

60 years of technological advance is something that you consistently ignore in your conjectures.

I am ignoring it because that is speculating about alternative timelines. I don't know what would have happened in the last 60 years if the transistor had not been invented any more than what would have happened if the Nazis had won World War II. Physics Forums is not the place for such speculations.

What I do know is that engineers and physicists have been searching for a transistor replacement for decades, utilizing incredible tools such as the transistor based computer to aid the effort. They have not found a feasible replacement for the transistor just yet, although they keep trying. (Maybe one day). So the real question is: Do you honestly think that taking away their incredible tools would help that effort?

 

[OmCheeto]

But the LED light is not now and likely never will be a groundbreaking/revolutionary technology. We had electric lights for a century and efficient electric lights for decades before LEDs came around and not much changed with LEDs. LEDs are only slightly more efficient than fluorescents, so there is very little potential for them to have much of an impact.

http://www.designrecycleinc.com/led%20comp%20chart.html claims LED's are twice as efficient and last 6 time longer. In the last year, the only replacement bulbs I've purchased are LED.

Boring? I'm sure they don't give out Nobel prizes for being boring. This does not only affect the light bulb above the kitchen table, it's a benefit for society and is influencing the progress of science itself. Because of the blue LED discovery I can live-image biological processes, without overheating the sample.. to just name a personal example :)

Actually, I thought it was kind of boring too. But then I thought about it for a few seconds, and was really glad they won the prize.

I saw somewhere a while back an article about LED lighting for plants. The light made the plants look kind of ugly. It wasn't the following article, but it will do:

http://www.lighting.philips.com/pwc_li/main/shared/assets/images/applications/horticulture/nl/Casestudy_VitroPlus_vitroplus_productiehal_LR_224px.jpg
Case study Vitro Plus, the Netherlands​

Everything is freakin' purple! It's like a bad trip back to the 70's.
But, meh.

So I did some research and found that plants only require certain wavelengths of light, which kind of explained the freakish color.

Light Absorption for Photosynthesis​

Now, I'm no biologist, but I decided that if we stopped wasting energy providing spectral light just to make the plant look pretty, then I could grow basil in my kitchen during the winter months, much more efficiently. And this was when I decided that blue LEDs were most awesome.

 

[russ_watters]

http://www.designrecycleinc.com/led%20comp%20chart.html claims LED's are twice as efficient and last 6 time longer.

According to Lowes.com, they 50,000 hrs is the high end of a range that starts at 18,000 hours and 6-8 watts looks more like 8-10...and CFL life is more like 8000-12,000 hours.

More importantly, "twice" and "6x" aren't as impressive as they sound, given that CFLs were already 5x as efficient and 7x longer lasting as incandescents. So, for example, if you replace an incandescent with a CFL at 3000 hours per year and $.15/kwh, you save $21/yr while if you had replaced it with an LED (assuming 8 watts was accurate) you save $23/yr. Or, if you replace a CFL with an LED you save $2 a year. Which means if you use it enough, you might recoup the 3x higher purchase price in 6-8 years.

However, I do need to look into LEDs for some fixtures in my house since CFLs don't survive in totally enclosed fixtures and I have a bunch. Hopefully LEDs fare better.

 

[zoki85]

According to Lowes.com, they 50,000 hrs is the high end of a range that starts at 18,000 hours and 6-8 watts looks more like 8-10...and CFL life is more like 8000-12,000 hours.

Average CFL life in a typical household is about 3500 hours

 

[dlgoff]

... given that CFLs were already 5x as efficient and 7x longer lasting as incandescents.

bold by me

That's probably true after they've been in service for a few days. Three CFLs from the last package of six I purchased failed in the first 3 days. Just sayin'

 

[zoki85]

bold by me

That's probably true after they've been in service for a few days. Three CFLs from the last package of six I purchased failed in the first 3 days. Just sayin'

Made in China?:D

 

[OmCheeto]

...Three CFLs from the last package of six I purchased failed in the first 3 days. Just sayin'

This also happened to me. I thought it was just a manufacturing fluke, so I didn't mention it.

A quick google search yields some people with similar stories:

The myth of the long lasting CFL light bulb​

April 2014
In my basement I have about 8 ceiling lights that use low energy GE CFL 23 watts bulbs. Over the last 2 years, at least 18 of them have failed. GE claims that each one should last about 8 years before failure.

Premature failure of compact fluorescent lights
January 2013
I have noted that many or even most of the CFL's I've installed in my current house and my previous house have failed in a short amount of time. I've been writing the install date on the body of the lights and have had many fail in less than a year, compared to the rated/advertised 7 to 10 year life.

There are a lot more stories.

 

[nsaspook]

http://spectrum.ieee.org/tech-talk/geek-life/history/rcas-forgotten-work-on-the-blue-led

Nobel Shocker: RCA Had the First Blue LED in 1972

But there’s more to this story. “The background is kind of being swept under the rug,” says Benjamin Gross, a research fellow at the Chemical Heritage Foundation in Philadelphia. “All three of these gentlemen deserve their prize, but there is a prehistory to the LED.” In fact, almost two decades before the Japanese scientists had finished the work that would lead to their Nobel Prize, a young twenty-something materials researcher at RCA named Herbert Paul Maruska had already turned on an LED that glowed blue.

http://online.wsj.com/articles/benj...ghted-the-way-for-a-japanese-nobel-1412900717
http://www.sslighting.net/news/features/maruska_blue_led_history.pdf

 

[russ_watters]

Average CFL life in a typical household is about 3500 hours

That's probably true after they've been in service for a few days. Three CFLs from the last package of six I purchased failed in the first 3 days. Just sayin'

Yes, as I've said I have had such issues as well. Here's hoping the advertised lifespans of LEDs are closer to their realities -- but product lifespans rarely live up to their lab-test condition ratings.

 

[OmCheeto]

http://spectrum.ieee.org/tech-talk/geek-life/history/rcas-forgotten-work-on-the-blue-led

http://online.wsj.com/articles/benj...ghted-the-way-for-a-japanese-nobel-1412900717

Very interesting.

I like the last paragraph of the first article:

Maruska is happy to see his story getting a fresh look again, and there’s no hard feelings on who the Nobel Prize went to. “These three guys really deserve the credit,” he says. “It’s like I say to people: they had been working on the steam engine for 100 years, but they never could make one that really worked, until James Watt showed up. It’s the guy who makes it really work who deserves the Nobel Prize. They certainly deserve it.”
​

Mr. Maruska's Facebook page is open, and he's been talking about this since the prize announcement.

Wait a minute. Mr. Maruska listed his patent number, so it's easy to check his claim:

Gallium nitride metal-semiconductor junction light emitting diode
Patent # 3,819,974

Undoped gallium nitride always occurs highly n-type (n 10 cm) and thus far has not been made conducting p-type. However, a deep acceptor such as zinc has been utilized to compensate the donors and produce insulating gallium nitride crystals. This dopant can be introduced during the growth of the gallium nitride crystal. When the dopant is introduced after initial deposition of undoped material, an i-n junction is formed. In the prior art, red, yellow, green and blue light emitting diodes have been obtained with zinc doped insulating regions forming i-n junctions.

OBJECTS AND SUMMARY OF THE INVENTION It is a general object of the present invention to provide a violet light emitting diode.​

He patented a violet LED.

I cry FOUL!

 

[atyy]

I cry FOUL!

Whose? The Nobel Committee or Maruska?

 

[OmCheeto]

Yes, as I've said I have had such issues as well. Here's hoping the advertised lifespans of LEDs are closer to their realities -- but product lifespans rarely live up to their lab-test condition ratings.

Well, "efficiency" also has economic consequences:

per Wiki:
The Centennial Light is the world's longest-lasting light bulb. ... the bulb is at least 113 years old and has been turned off only a handful of times. ... It is often cited as evidence for the existence of planned obsolescence in later-produced light bulbs.​

You can't very well run a profitable business if your product lasts longer than most humans.

By my calculations, that bulb has been burning for 990,000 hours.

Wow.

 

[OmCheeto]

Whose? The Nobel Committee or Maruska?

Neither.

I call foul on the headline: Nobel Shocker: RCA Had the First Blue LED in 1972

Maruska, et al, patented a violet LED.

Blue is not violet.

Except in certain instances, where you are traveling near the speed of light. Which will, of course, lead to other arguments.

 

[nsaspook]

Looks more blue than violet to me but it was the 70s and violet was a hip color for guys then.
https://scontent-b-mxp.xx.fbcdn.net/hphotos-xap1/v/t1.0-9/1013052_10202593039292688_3207786163505961998_n.jpg?oh=0d616adccda6a0c7364f4d38ddd60ae6&oe=54C5AA2E
https://www.facebook.com/photo.php?...76231271508.2012568.1398003467&type=1&theater

 

[atyy]

Neither.

I call foul on the headline: Nobel Shocker: RCA Had the First Blue LED in 1972

Maruska, et al, patented a violet LED.

Blue is not violet.

Ah, fair point. But I found the history informative. Now I am learning about the history of the steam engine. :)

 

[OmCheeto]

Ah, fair point. But I found the history informative. Now I am learning about the history of the steam engine. :)

On the shoulders of giants.

Did you check out the "REFERENCED BY" section of the patent?

There appear to have been a lot* of people that stood upon Herbert's shoulders. :)

*103, to date.

 

[TumblingDice]

But the LED light is not now and likely never will be a groundbreaking/revolutionary technology. We had electric lights for a century and efficient electric lights for decades before LEDs came around and not much changed with LEDs. LEDs are only slightly more efficient than fluorescents, so there is very little potential for them to have much of an impact.

We have had electric light bulbs for over a century, without much improvement in the incandescent or flourescent areas given the amount of time. When choosing a light for an application, you can select one option for 'temperature'/wavelengths. If it's incandescent you can choose a dual-filament three-way for fixed brightness levels, or go all-out for a dimmer switch. :w

About 18 months ago I learned about Philips Hue LED lamps from an article on Ars Technica. They're only sold by Apple directly, but they have no proprietary ties to Apple. They're expensive at $60/bulb, but I've added on slowly. The original bulbs have an odd trio of amber, magenta and greenish LEDs so they can specifically create everything from standard warm incandescent to brilliant cold white. Of course the three can be used to blend many more vivid color effects. Since then Philips has added "ambient" color lamps and "light strips", and these employ the customary red, green, blue LED combination for blending aesthetic ambient lighting effects.

I've grown my collection to 7 std/original bulbs, 3 ambient color lights, and one light strip. Did I mention the lights are all wireless/Wi-Fi controllable? The "bridge" talks to the lights in a wireless ZigBee protocol, but it takes commands through the network. I can control all lights using my iPad with an iOS app, or the ASUS tablet or GS3 cell phone with Android app when away from home. The cell's NFC feature allows lights to come on when I arrive home. From my front door, to the bedroom, in living areas and on the back patio, all lights can be set for utility or just plain old mood. The apps provide ability to create 'scenes' which are just setting color, levels and grouping lights into your own setups so you can control different lights in various ways as you plan, with a single tap or swipe.

I'm most pleased with how the combination of lighting around the big screen has worked out. I'll have to try the cell camera, or maybe charge up the Canon to see if I can get decent non-flash results that'll do justice to share with y'all.

I could go on about the seeming endless possibilities this technology will open up, and the only barriers to rapid evolutuon/revolution are cost and imagination!

 

[russ_watters]

Did I mention the lights are all wireless/Wi-Fi controllable?

How much extra power are they constantly using?