A Paper Posted Last Month Claims To Have Achieved Superconductivity at Room Temperature, But Other Physicists Say the Data May Be Incorrect (vice.com)

dmoberhaus writes: Last month, two Indian physicists posted a paper to arxiv claiming to have demonstrated superconductivity at room temperature. If this paper is legitimate, it would represent a breakthrough in a problem that has existed for superconductivity for 100 years. Understandably, the paper shook the physics world, but when researchers started digging into the data they noticed something wasn't quite right -- the noise patterns in two independent measurements exactly correlated, which is basically impossible in a random system. The Indian researchers have doubled down on their data, and things only got weirder from there. This is a look inside what could be the biggest drama to happen in physics in nearly a decade.

Comments

That's old news

[Ecuador]

I seem to remember several years ago researchers in Fairbanks, Alaska had already achieved room temperature superconductivity. The trick was to turn of central heating as I recall...

Vice?

[Train0987]

Is Vice really a valid source for news like this?

Re:Vice?

[ceoyoyo]

So you're saying yes, this IS Vice's kind of thing?

Invent a way to verify this

[DickBreath]

If only there was some kind of something-something method by which one scientist could reproduce another scientist's results. Theories could be formed. More experiments tried and reproduced. Etc. Such a thing could be a force that would propel technological advancement forward at an incredible rate.

If someone can invent some kind of scientific method, they should patent it!

Re:Invent a way to verify this

[Anonymous Coward]

You know, it bugs me when people say "well science is not a democracy. Something is either proven or it isn't."

The scientific method is essentially democratic. One person claims to have done something via an experiment, and that doesn't prove anything. You need a whole bunch more people to do the same experiment....to convince them. And their experiments need to be peer-reviewed. To convince even more people. Eventually, when enough people are convinced, your hypothesis is essentially voted into being a theory.

Though there will still be hold-outs within the scientific community and it is always possible some upstart will publish results that totally contradict yours. And the democratic process of truth-establishment will go around again.

And that's how it should be. When truth-establishing processes are dictatorial in nature, you get crazy religious wars. And Hitler.

Re: Invent a way to verify this

[c6gunner]

Fucking magnets. How do they work?

Re: Invent a way to verify this

[c6gunner]

The scientific method is essentially democratic. One person claims to have done something via an experiment, and that doesn't prove anything. You need a whole bunch more people to do the same experiment....to convince them. And their experiments need to be peer-reviewed. To convince even more people. Eventually, when enough people are convinced, your hypothesis is essentially voted into being a theory.

That's not at all how that works. A hypothesis is just an idea. A theory has predictive power. We don't vote to turn a hypothesis into a theory; a hypothesis is just the starting point for an experiment. Based on the results of the experiment you may be able to formulate a theory ... and if that theory is valid, you will be able to predict future results. Popularity and opinion are irrelevant; either your theory predicts future outcomes, or it does not.

Re: Invent a way to verify this

[c6gunner]

Even ICP has a better understanding of how to use magnets than you do, apparently.

Re:No

[Anonymous Coward]

The easy answer is accidental duplication of one input location. If two sensor locations were mis-wired or the collection software had a typo so that one was recorded twice while another was ignored, that would get identical noise in two columns and the appearance of immeasurably fast communication between two locations.

The hard answer is accidental room-temperature superconductivity. It's also the fun answer.

First thought is fraud.

[Anonymous Coward]

Take a look at the green and blue dots in the above graph. They represent the noise measured during two separate experiments run by Thapa and Pandey to test the magnetic susceptibility of their superconducting material.

So they ran the same experiment twice, and got almost identical noise? I don't have to pull any punches here because I'm not publishing a critique.. but that just strikes me as high evidence of fraud.

Entirely speculation here, but I'd guess they ran this experiment once, got the result they wanted, and were incredibly excited. Nobel Prize Time! Then they ran it again, and again, and again.. couldn't replicate the results, but still wanted to publish. So they faked the second set of data, and hoped nobody would notice.

Still it _could_ be something weird... Honestly I hope it is. But realistically this is just fraud, or at best some terrible experimental error.

Re:No legit explanation

[HornWumpus]

The Indian researchers no doubt have doctorates. Hence they are qualified to 'doctor' data. Just SOP.

Anyone notice less scammer calls lately?

[technosaurus]

Turns out they moved on to "science".

Re:No

[johanw]

Occam's razor tells us that cheating indians is the most probable explaination.

Re:"Room temperature"

[iggymanz]

Never mind the silly phrase in the article about "above temperature of liquid water", having superconductor that functioned at temperature a freon or ammonia based cooling system could reach would be earth-shattering, as in ushering in a whole new era of technology. It would be as big as the invention of the electric motor.

Re:That's not even the only problem

[iggymanz]

Not that there is any merit to article's claims but

Silver layer has "proximity effect" on some superconductors, raising the transition temperature.

There is a superconducting alloy with gold, SrAuSi3

Re: No

[HornWumpus]

At this point, they can produce raw data (and a plausible explanation for an honest mistake), admit they duplicated a dataset, or keep quiet.

Only the first will really help them, science is rough on cheaters that get caught while still living.

Cartman: 'I'm sorrry' isn't going anywhere...Maybe inside India, if they are brahmin.

Re:That's not even the only problem

[Sarusa]

Not at all. For example, copper, silver, and gold are not superconducting on their own even at (near) 0K. Their lattices are so tightly packed that even though they're decent conductors they can't generate enough Cooper pairs from free electrons.

Re: To be fair?

[K. S. Kyosuke]

That page could mean any number of things, such as US article authors facing tighter scrutiny, or the US having a larger absolute number of article authors. It is hardly a replacement for a proper study of these incidents.