What's up in this SPAWAR video?

Here is a video presentation of infrared recordings of anomalous heating in a deuterium palladium cell: ( youtube video) (see also this presentation if you want more detail, and have time). There are two papers referenced in the link, which give more detail. You can see with your own eyes sporadic localized bursts of energy, which show up as white flashes in the camera (I would like to encourage people to read the linked papers, which include thermal photos with a different color map, and also include piezoelectric detection of bursts)

Excluding cold fusion, what could possibly be causing this? Please try to account for the nature of the bursts--- the small radius, the qualitative amount of energy released, etc.

Just to be clear, I think the answer is nothing, that it is cold fusion, but I would like to see what explanations people come up with. I will also point out that this research group has been politically shut down recently.

-

Without knowing anything about the experiment nor the camera, I would suggest that what is shown in the video is a combination of shot noise and aliasing due to a poor choice of gradient mapping. Note that the gradient bar at the bottom of the frame jumps from a fairly deep red (actually darker than precedent tones) to pure white in one increment - this may result in small fluctuations in temperature appearing much more visually significant than they are.

I had a go at my own gradient maps (in Photoshop) to show off this effect - I tried to go for something similar to their gradient mapping, which for all I know is hardwired into the camera:

Something to consider. I read an article recently about injudicious gradient mapping overemphasising tiny differences in geospatial and medical datasets (and also obscuring important distinctions in other scenarios). It's a good read and I'll link it when I find it again.

-
Shot noise seems like a reasonable explanation to me. All I see there is a grainy image. Also, I must say that the quality of the entire presentation seems quite low. No mention of any control experiments or characterization of their detectors, no quantitative analysis of these "hot spots", no correlative measurements between their camera/piezo detector. It is very unconvincing to me. – user2963 Aug 4 '12 at 12:25
@RonMaimon I have a ton of respect for your physics knowledge and intelligence, but do you mind if I ask what experience you have with experimental science? This seems easily explainable as measurement noise to me. Their imaging system is low resolution and has a lot of variance, which is visible in the sections of the line profile which lie outside the electrode - the level of variance does not look substantially smaller than the center. – user2963 Aug 5 '12 at 22:14
Also, take a look at this screenshot of the video, which I believe show the limitations of their imaging detector: imgur.com/FuTkX – user2963 Aug 5 '12 at 22:14
Can you give a physical explanation for the anisotropy of the white segments? – user2963 Aug 5 '12 at 22:15
Finally, I'd like to reiterate that I think this is very poor science, even if the effect is real: these kinds of doubts could easily by reduced by showing some basic control experiments, such as imaging with the current off, or using a resistive heater to warm the electrode without the reaction and comparing. – user2963 Aug 5 '12 at 22:16

I'm an experimental electrochemist. The problem with experiments such as those mentioned above is that they lack the necessary details to reproduce it, so that we can verify it or improve upon it.

In the first video, a paper linked is here: http://www.lenr-canr.org/acrobat/SzpakSpolarizedd.pdf

They vaguely mention a "negatively polarized Pd/D$_2$O system". Then they refer us to Figure 6 for more "complete" experimental details. Let's ignore all the piezo stuff. We have a potentiostat (what kind), with electrodes (unspecified counterelectrode and a vague Pd/D film working electrode without mention of preparative conditions), in an undefined electrodeposition solution with unknown concentrations or purities of reagents, and without any mention of the voltage/settings applied on the potentiostat during the course of the experiment. Moreover, they do not tell us about the thermal scanner, even the make or model. There is a mention to other journals but a quick glance at the few I could easily access online did not illuminate any of these details better. Was the container even cleaned for trace impurities?

In an excellent paper you will find all these experimental details and more. In a field which is highly contested among leaders in electrochemistry you are already held to a higher bar of scrutiny and as such greater care must be taken in reporting your results for any serious scholarly interest. Lacking these important details makes the entire paper worthless. If this were an undergrad's first lab report given me to then I wouldn't even award a D-. This is especially true when people in cold fusion report that it is very sensitive to experimental details/preparation.

This "paper" also suffers because it does not address why this could not be due to joule heating, a more likely scenario but cannot be judged without knowing the voltages or currents involved.

If you can find a paper (with the PDF) that has a complete experimental then e-mail it to me and I'll take a closer look. I would honestly really love for cold fusion to be true and to see if it could work, but the work I've seen doesn't even hold a candle to the care you'll find in a Letter published in the Journal of the American Chemical Society and therefore I cannot spend my time trying to reproduce these results.

-
This is a nice honest answer. I'll try to dig up the review and link it here. I think you are not fair to these folks, they have published many of their results in journals, and the reagents and concentrations for the experiment are widely available in the CF field, in the SPAWAR published papers. The field is neglected, and the papers tend to look bad, but that's not a way to evaluate experimental results--- you need to look at the data, even if it looks crappy, and see if there is an anomalous effect. – Ron Maimon Jan 22 '13 at 7:08