Field of Science

Updated #chemnobel odds for tomorrow

So today's physics Nobel Prize was awarded to three physicists who worked on blue LEDs, devices that make light bulbs glow brighter and last longer than normal ones. It's a well-deserved recognition, really one for a physics-based invention rather than for a fundamental discovery (as an aside, it's worth noting how much the invention owes to the chemistry of gallium nitride semiconductors). The last time an invention was recognized was when the prize was given out for CCDs in 2009.

However I think the prize also casts some serious doubt on a lot of predictions that Thomson Reuters and others have been making for tomorrow's chemistry award. It's certainly not unheard of for prizes in two disciplines to be awarded for the same field of study (for instance chemistry and medicine often trail each other in the context of crystal structures of important proteins) and I haven't really done an exhaustive study of how often this has happened before, but I would guess that the odds of it are pretty low.

If nothing else, today's blue LED prize almost certainly means that one of the most popular predictions for tomorrow (and Thomson Reuters's choice) - that for Ching Tang's work on organic LEDs - will have to be tossed out of the window for now. It seems extremely unlikely for two kinds of LEDs to be awarded two different prizes in the same year.

Today's recognition is not just a recognition of materials science but even more so of energy. That might mean that some of the energy-based prize predictions for tomorrow may also be off the mark. Thus, the odds for awarding a prize for Li-ion batteries or for dye injected solar cells may sadly have worsened.

Since both materials and energy are no longer looking that great in my book, I now have a revised set of predictions. In light of today's development I think that the chances for other kinds of chemistry might now be higher. So I am now bumping up my predictions for biochemistry (click chemistry, nuclear receptors, chemical biology - eg. Stuart Schreiber) and also adding a prediction for polymers (especially ATRP) which I missed earlier. Single molecule spectroscopy (Moerner etc.)  and electron transfer in proteins and biomolecules (Gray, Barton etc.) also stand a good chance now.

Chemists, time to break out the pitchforks: A biologist may "steal" "our" Nobel once again...

10 comments:

  1. This is Stuart's year!

    ReplyDelete
  2. Some clarification is in order. Thomson Reuters, when it names new Citation Laureates each year, is not predicting these people (or these prizes) will win that year. Instead, we are using citation analysis to identify researchers 'of Nobel class' and suggest they are probable Nobel Prize winners sometime -- this year or in the future. For example, we named Nakamura a Thomson Reuters Citation Laureate in the first year we started these forecasts on an annual basis, in 2002. Five times we have named individuals Citation Laureates who actually won the Nobel that same year -- but that is wholly by chance. And not all those we name can win because there will always be more researchers of Nobel class (even though it is a very select group) than there are Nobel Prizes to go around. -- David Pendlebury, Thomson Reuters
    P.S. I agree with Ash that after today's prize, OLEDs would be a surprising choice for the Chemistry Nobel tomorrow.

    ReplyDelete
    Replies
    1. Thanks for the clarification David. I am still a little unclear on how Thomson Reuters picks a specific name for a specific year though.

      Delete
  3. Sure, Ash. Our selections are primarily driven by reviewing extremely highly cited papers -- those cited more than 1,000 times and especially those cited 2,000 or more times (our analysis shows that typically at least one of the winners of a Nobel Prize in any of the science fields has one or more papers cited at this level). As you may imagine, such papers are quite exceptional. Since 1960, we've indexed some 55 million items in the Web of Science, of which about 37 million are regular articles (not reviews, letters, abstracts, corrections, etc). Of these, only about 100 have been cited more than 10,000 times, only some 400 more than 5,000 times, about 1,100 cited more than 3,000 times, only 2,500 cited 2,000 or more, and about 10,000 cited 1,000 or more times. Of course, you have to separate these reports (roughly) according to the different Nobel Prizes -- and of course there is overlap in possible assignments. Certain papers lie outside the areas recognized and they are put to the side. We do consider the age of the papers in light of total citations, so that more recently published ones are not expected to have collected as many citations as, say, a 30 year old paper. And there are different citation averages by field and within subfields, but these are not so important at these ultra high frequencies. It's a constant process of weighing (mentally) one report or researcher against another, primarily by quantitative indicators (citations) and secondarily (but importantly since this is not a mechanical process and citations do not speak for themselves) by qualitative considerations (taking into account questions of priority, history of the field and its impact both in terms of new knowledge and practical benefits, other indicators of peer esteem such as national academy membership in home country and receipt of prestigious, especially so-called precursor prizes, and finally instinct). Someone asked a famous jazz musician which player he thought best and he replied that above a certain level it's a matter of taste. I agree. We are looking for Nobel class researchers and then, looking at the history of the Nobel Prize and the types of laureates who have been selected. We are trying to discern the characteristic tastes of the Nobel committees and Academy. In specific, concerning how we choose specific researchers each year, I'll give you an example. Owing to extraordinary levels of citation to the 1995 ATRP paper by Matyjaszewski, and these other considerations, we selected him as one of our 2008 Citation Laureates in Chemistry. Sawamoto's paper of 1995 was cited considerably less (and still is although certainly highly cited -- 2,300 for Sawamoto vs. 3,400 for Matyjaszewski). If I'd made the selection now, with more data, I would choose Sawamoto as a Citation Laureate with Matyjaszewski. But I did not then -- a mistake. And now, in 2014, we have proposed RAFT, published in 1998, as a possible Nobel Prize. We have done so now because citations to the key report of Moad, Rizzardo, and Thang have reached some 2,600. And we noticed it's been awhile for a polymer science prize. A possible choice we have not advanced would be a Nobel recognizing both ATRP and RAFT, including Matyjaszewski, Sawamoto, and Rizzardo. That would again illustrate the cruel consequences of the Rule of 3. By naming Moad, Rizzardo, and Thang Citation Laureates and RAFT separately, we wanted to acknowledge the important research and impact of this team, whatever decisions and groupings are ultimately devised in Stockholm. I'm always as interested as everyone else when the announcements are made and never really have any real notion (except last year in Physics) what prize or people will be chosen. It's a long reply, but I wanted to give you a full answer. -- David Pendlebury, Thomson Reuters

    ReplyDelete
    Replies
    1. Thanks very much for that very detailed reply. I now understand much better the process that goes into the nominations and agree that specific choices for certain years do boil down to taste as much as anything else. Also agree that RAFT and ATRP should still be high on the list.

      Delete
  4. Looks like the Prize was hijacked by three physicists this year! The chemistry Prize just isn't safe from interlopers.

    ReplyDelete
    Replies
    1. We Welcome Interlopers. That's the sign I want hung on top of the Chemistry Academy.

      Delete
    2. I don't think my comment nested properly. That +1 is for the "we welcome interlopers" sign!

      Delete

Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS