Alternate title: Imminent Death of High-Tc Superconductivity Predicted. Film at 11.
PhysicsWeb has a story about a study of condensed matter papers that has been posted to the arxiv, predicting the imminent death of high-temperature superconductivity:
The new study was carried out by Andreas Barth from the FIZ Karlsruhe and Werner Marx from the Max Planck Institute for Solid-State Research in Stuttgart, who examined the number of papers listed in the INSPEC and Chemical Abstracts Service databases with words like "superconductivity" or "superconductor" used in the title or listed as "keywords".
By plotting these as a function of time, they found that the numbers shot up rapidly in the late 1980s, following the unexpected discovery of high-temperature superconductors by Georg Bednorz and Alex Müller at IBM's Zurich lab in 1986. The numbers in the INSPEC database reached a peak of about 8500 a year in 1990, but have been slowly falling and now stand at about 4400 (figure 1). A simple linear fit to the data reveals that the numbers will reach zero by 2010-2015.
If you ever find yourself thinking that Ph.D. physicists are somehow above statistical silliness, keep this in mind as a counterexample.
Has there been a decrease in the number of papers about superconductivity? Probably (though keyword searching is a lousy way to check). That's usually what happens when a field goes several years without a major new development, either experimentally or theoretically. There are exceptions, and people write books about them, but for the most part, physics is as faddish as anything else-- a splashy discovery leads to a huge burst of publications on a tpic, but in the absence of sustained progress, most people move on to something else.
But honestly, does anyone really think that a simple linear extrapolation to zero is meaningful in any way?
- Log in to post comments
Wow.
I looked at the actual article because I've seen a time or two when someone less-than-seriously extrapolated a current trend to predict the extinction of some area of research, but these guys really seem completely serious to me.
I remember seeing a joke graphic a few years back predicting the date of the "zero-femtosecond laser pulse" from a linear extrapolation. Is this article meant tongue-in-cheek?
Actually, after reading it, they appear serious.
High temp ceramic supercons were postulated and discovered amidst IBM/Zurich managerial threats. Discovery is ever an orphan. YBCO killed discovery and birthed development. Research efforts were increasing better focused and more efficient until program optimization peaked at nothing.
Stanford/Physics' William A. Little proposed decorated polyacetylenes, [-C(Ar)=(Ar)C-]n,
Phys. Rev. A 134 1416 (1964)
wherein "Ar" is a polarizable chromophore/fluophore conjugated to the polyene skeleton. Replacing phonons with excitons for Cooper pairing gives extrapolated Tcs around 23,000 K. That has a decent chance of Tc above 300 K. Nobody could model the structures in 1964 much less make them.
This is 2006. No problem for either. Hydrogens and multiple bonds are omitted from the following stereograms; interior cut of HyperChem-calculated 22-mer displayed.
http://www.mazepath.com/uncleal/benzen1.png
Cheap and easy synthetic model
http://www.mazepath.com/uncleal/pyrene1.png
Better numbers
http://www.mazepath.com/uncleal/decacyc1.png
Much better numbers
http://www.mazepath.com/uncleal/pave1.png
Conjugation varies as cos^2(theta). Prior structures' substituents' pi-planes are almost normal to those of the skeleton and therefore couple by sigma-polarization. Here we have a much decreased twist angle and added intense pi-conjugation.
Iodine fuming gives polymer p-doping. Look for the Meissner effect with a magnet. You have it or you do not.
ArH -> ArCHO --> ArCH(OH)-(C=O)Ar --> O=C(Ar)-(Ar)C=O --> [=C(Ar)-(Ar)C=]n via McMurry coupling
That is Little's [-C(Ar)=(Ar)C-]n with the bracket window shifted by one atom. The model compound starting with $(US)0.16/gram benzil is then a one-step synthesis. No molecular weight control or end-group control, so we do better.
diaryl benzil + Tebbe methylenation --> 2,3-diarylbutadiene
2,3-diarylbutadiene + ADMET (probably Schrock, possibly Grubbs) --> Little + ethylene
O=C(Ar)-(Ar)C=O --> H2C=C(Ar)-(Ar)C=CH2
H2C=C(Ar)-(Ar)C=CH2 --> [=C(Ar)-(Ar)C=]n + H2C=CH2
Also see Ben Zhong Tang for alkyne polymerization,
Acc. Chem. Res. 38(9) 745-754 (2005)
http://home.ust.hk/~tangbenz/ACR0538745.pdf
Do you want room temp supercons? Pull your collective thumbs out of your butts and make them, by the book. Bloody stop whining.
"You want the impossible! I don't believe it."
"That is why you fail"
Uncle Al's weekly number of postings seem to remain flat, while their comprehensibility shows a strong linear trend to zero. This zero point should occur in the next few days, at which point it literally will be all greek and chemistry symbols.
That is why you fail.
Google
"Acyclic Diene Metathesis" 823 hits
ADMET ethylene 15,200 hits
If all you study is what you already have, then you end up with that which you started. Organic synthesis builds its own futures. Science makes stuff, engineering makes things. Deficient stuff cannot be optimized into adequate things (beyond performance bonuses for management).
Learn how to look into a stereogram to see it in 3-D.
Run Uncle Al's post past a decent organic chemist. Is the result derision or a techno-woody?
Soon the philosophers will appear:
"A good forecaster is not smarter than everyone else, he merely has his ignorance better organised."
"Prediction is very difficult, especially if it's about the future." (Niels Bohr)
"An unsophisticated forecaster uses statistics as a drunken man uses lamp-posts - for support rather than for illumination." (Andrew Lang)
and my special favorite:
"Those who have knowledge, don't predict. Those who predict, don't have knowledge." (Lao Tzu, 6th century BC.)
Mark Twain has some fun with linear extrapolation here.
Sorry, link didn't work. go here:
http://math.smith.edu/Local/cicchap1/node12.html