NOTE: The first big chunk of this entry is a summary of what I've done throughout the whole summer, for those of you looking just for results, scroll down to the RESULTS header.

Well the summer is drawing to an end and my time here at Hunt Utilities Group is winding down, I think it’s time I take a look back on the summer of science that was.

 A Dumb College Kid

When I first arrived at HUG I was a bleary-eyed physics student from the University of Minnesota Duluth who had no idea what the hell he was getting himself into. Having literally no knowledge of cold fusion, LENR, CMNS, whatever you feel like calling it, I had a long ways to go. So, the first half of my internship consisted of totally immersing myself in the literature of the field to at least get some idea of what’s been going on for the last 25 years. When I started to find out about the extremely troubled history of the field and the utter state of scientific exile it stands in today, my reaction was what one would expect from any hopeful grad student in my situation and is best summed up by two words, “Oh shit.” But as I started to read more and more of the theoretical and experimental work that has been done in the field, one thing became very apparent to me: LENR research isn’t being done by whack-jobs and idiots (for the most part). There is a whole raft-load of legitimate scientific papers out there that support the “pseudoscience,” and I spent about six weeks reading as many of them as I could. I started this task with an extremely skeptical frame of mind, and after weeks of reading, my skepticism was, if not eliminated, significantly reduced.

After reading a broad spectrum of papers on virtually every type of experiment in the field, I set out on finding an experiment that I could replicate. This experiment had to be cheap, somewhat easy to perform, and (I hoped) reproducible. Also, it had to be something I could finish within my end-of-the-summer timeframe. In the end, I decided to attempt an electrolysis experiment that was rigged up with a CR-39 particle track detector originally designed by University of Minnesota Professor Richard Oriani. The original plan was to replicate the experiment exactly as done in his 2008 paper, while also mixing in the nickel foil aspect of Aizawa et al.’s work (the nickel foil aspect was eliminated when we discovered how much 6 micron nickel foil cost). I also relied on Kowalski’s work to guide me in the experimental process. The first hurdle to overcome in the process was ordering all of the materials I would need to perform the experiment. While a lengthy process, the ordering and waiting went well, and when everything finally arrived, I was ready to hit the ground running. I thought I would be performing the experiment within the week, but instead, I encountered a whole track’s worth of hurdles to jump over before anything could begin (an experience with which I’m sure most scientists are familiar). The first hurdle came in figuring out the etching process: we had to carefully control the temperature of a 6.5M NaOH solution (super nasty stuff) for six hours. We also had to find the right solution temperature that would be produce tracks on the CR-39 big enough to detect, but small enough to still have some sense of direction.

 CR-39 Overview

Allow me stop here and explain what CR-39 is and how it works. CR-39 is a glass-like polymer that is often used in eyeglasses. When an ionizing radiation particle (such as a proton, electron, or alpha particle) penetrates the material, it leaves (as one could imagine) an ionized damage path in its wake. When the material is then placed in a solution of heated, concentrated sodium hydroxide, the solution eats away at (etches) the ionized damage tracks more quickly than at the rest of the material, the result is a pit anywhere there was an incident particle. Just how quickly the solution eats away at the material (etch rate) is a function of concentration and temperature, and since we stuck with a 6.5M solution, we varied temperature to change the etch rate (a graph of etch rate vs. temp. can be seen here, in Figure 3). Because the solution eats away the exact path that a particle traveled into the detector, the tracks left behind will show evidence of their particle’s incident angle. If a particle hit the CR-39 at a 90^o angle, the resulting track will be a perfect circle; however, if the particle came in with a different angle, the resulting track will be an ellipse whose eccentricity depends on the incident angle. Now, once the solution etches all the way to the bottom of the ionized damage track, etching no longer carries on at an accelerated rate, instead, etching in the track continues isotropically at the same rate as the rest of the material, which means that when etched too far, the elliptical tracks become circular, losing indication of direction, and eliminating a valuable piece of evidence for nuclear emissions. So after some hot plate troubles, we found that etching at 80^oC for six hours gave us the best compromise between etching long enough that tracks are relatively large and not over-etching so that tracks still have direction.

Sitting Motionless for Hours

The second obstacle to overcome was counting the number of tracks on a chip. While counting tracks on a 9cm² piece of material doesn’t sound like a daunting task… it is. The first thing that became clear was that we were going to need a new microscope that allowed us to both take pictures and move the stage a fraction of a millimeter at a time. Once this was ordered, a system for counting was developed: start at the upper left of the chip, move right along the length with the microscope, counting any tracks along the way, then when the other side is reached, move the chip down so that the new field of vision is just below the first, then move left and count the tracks along the way. This zig-zag is repeated all the way down the chip, on both sides to get a total count of tracks on the chip. The problem with this method is that it is often hard to determine whether a track was left by a particle, or by some kind of physical damage to the detector. An example of this can be seen below.

The two tracks on the left are quite light in comparison to many of the larger particle tracks and are often difficult to distinguish from damage tracks.

During the process, the counter has to make up his mind whether or not to count certain tracks, which is, to say the least, an inexact process. This inaccuracy, along with the fact that the procedure takes several hours of sitting motionless in front of a microscope, results in the track counts having very large uncertainties. For example, to get a background reading, we left one chip to sit out in the shop air for about 2 days, etched the chip, then counted it twice to see what kind of deviation we got. On the first count (of just the front of the chip), the result was 1023 tracks, corresponding to 113.67 tracks/cm². On the second counting, the result was 1213 tracks, corresponding to 134.78 tracks/cm². So, our uncertainty in any track count is about 21 tracks/cm², meaning that our background “noise” in the detectors is averaged at 124.26 ± 21 tracks/cm². However, one must keep in mind that this number doesn’t include variation in the actual background radiation, only our variation in counting, so the actual background radiation level has an even higher uncertainty, which, given more time, could have been quantified with a few other background samples. The reason for such a high background level (others in the literature report on the order of 20 tracks/cm²) is two-fold. First, in Minnesota, we have a relatively high concentration of radon in the air, and radon (and its radioactive daughters) emit alpha and beta particles that cause pits on the chip. Second, the detectors (ordered from Track Analysis Systems, Ltd.) were not very well protected in shipment. They came in a sealed bag, but the detectors themselves were uncovered in a large sheet (shown below). Most other CR-39 on the market comes covered in a protective film that can be removed immediately before an experiment, greatly limiting contamination. Unfortunately, without this film, any particle that comes in contact with the detector throughout its entire lifetime (manufacture, shipment, storage, etc.) will show up in etching; also unprotected chips are more prone to physical damage that can appear to be particle tracks during counting. We believe that this lack of protection has a large effect on the background levels for TASL’s CR-39.

The TASL CR-39.

 Electrolysis

Once we had etching and counting down (sort of), we were finally able to start the electrolysis experiment, and, as you probably guessed, there were again a few kinks that had to be ironed out. The electrolysis cell consists of two glass tubes that have an o-ring joint on one end. A CR-39 detector covered in 6µm Mylar is placed between the two tubes (the Mylar is to prevent the electrolyte from chemically damaging the detector). The combination is then clamped together so that the order from top to bottom is: glass tube, o-ring, 6µm Mylar, CR-39, o-ring, glass tube. Both tubes are sealed with rubber stoppers to mildly keep out airborne contaminants and to support the two electrodes: the anode (positively charged), made of 28 gauge (0.3mm) 99.9% Pt wire, and the cathode (negatively charged), made from .25mm 99% Ni wire. The whole setup is shown below. Both electrodes are wound into flat “spiral pancakes” that are positioned about 1cm apart with their spirals’ planes parallel to each other in a 0.2M Li₂SO₄ solution. Since Pt wire is very expensive, only the part of the wire in the solution and a small portion outside of the solution are Pt, the electrode is hooked to the power supply via a copper wire crimped to the Pt. The power supply is to be set at a constant current of either 70mA or 80mA, depending on the experiment. Two of these cells are constructed to be exactly identical, the only difference being that one cell is not hooked up to the power supply and will act as a control. When the experiment is finished, the CR-39 from both cells will be etched and counted, and the difference in tracks between experimental and control will be sought. The idea is that if the phenomenon being seen by Oriani and others is actually an artifact of electrolysis, the tracks seen in the experimental cell will not show up in the control cell.

The experimental and control setup.

The problem encountered on the first electrolysis was a classic “stupid intern” moment: when electrolysis was started (at a constant current of 70mA), the end of the crimp (made of Ni plated Cu) that held the Pt and Cu wires together was slightly in the solution. When I came back about an hour after the experiment began, the solution had turned blue, indicating that copper had been taken out of the crimp and absorbed into the solution. The crimp was quickly moved out of the solution. After discussing the problem with the team, we decided to just let the experiment run the remainder of its planned duration (4 days), while the test was no longer a direct Oriani replication, we thought the results would be worth examining.

The copper contaminated cell after 4 days of electrolysis.

The second electrolysis had a slightly more exciting malfunction. I set the cell up, identical to the previous experiment (except current was set to 80mA instead of 70mA), careful to not have to the crimp in solution, and left it to run overnight. I used a newly wound Ni cathode, but had to use the same Pt anode due to lack of supply. When I came back the next morning, the power supply was showing a very low voltage (it’s usually at ~4.6V, it was at 0.2V), which I determined was caused by the fact that the electrode wires were very slightly touching in the tube outside of the solution. When I turned the rubber stopper in an attempt to separate the wires, there was an arc that ignited the hydrogen in the cell, creating a tiny explosion that blew off the rubber stopper and broke the cathode wire. To reset, a new cathode was wound and the CR-39 was replaced. This experiment was run successfully, but unfortunately it could only be run for three days, as I leave for school soon and need to finish up with etching and counting before my departure.

The third electrolysis attempt, seconds after power was initiated.

Results!

Due to time constraints, I was not able to count all of the control and experimental chips from the two rounds of experiments, however, I was able to get a full count of both the experimental and control detectors from the first electrolysis that ran for 4 days (the one with the Cu intrusion). 

The first chip counted was chip 011, which was placed in the live experimental cell for four days at a constant current of 70mA. About 15 minutes in to the experiment, the copper contamination described above was discovered, and the Cu was pulled out of the solution by slightly raising the anode. The side of the chip that faced the solution (the side with the numbers inscribed on it) was counted to have 1129 tracks on its 9cm² surface area, corresponding to 125.4 ± 21 tracks/cm² (using previously determined uncertainty). The back side of the chip, which faced the empty glass tube on the bottom, was tallied at 974 tracks, or 108.2 ± 21 tracks/cm².

A high local density of tracks near the edge of chip 011.

Chip 012, which sat for the same period of time in the control cell (identical in every way except not hooked to a power supply) was counted next. The side of chip 012 facing the electrolyte was found to contain 1253 tracks, corresponding to 139.2 ± 21 tracks/cm², and the back side had 1164 tracks, or 129.3 ± 21 tracks/cm². 

The heart of the high density area near the etched-in numbers on chip 012.

The conclusion that must be drawn from these two chips is that there was no significant increase in the number of particle tracks produced in the aforementioned electrolysis cell from that produced in a dummy cell. Also, all four track densities lie well within the range of background noise on the detectors, at 124.26 ± 21 tracks/cm² (which, recall, actually has a higher uncertainty than 21 tracks/cm²). 

An interesting fact to note is that, in both chip 011 and 012, the highest local track densities always lay near the edges of the chip, outside of the area subtended by the o-ring, and thus not exposed to electrolysis. On chip 011, there were two of these areas, each relatively small, on chip 012, there was one such area, but it was very large and contained an exceptional amount of tracks.

Another cluster of tracks on the front (electrolysis facing) face of chip 011.

The edge of the enormous cluster on the front face of chip 012. One can note that the edge of the cluster coincides with edge of area subtended by the o-ring.

The fact that these clusters appeared outside of the active area, and that they appeared on both the control and experimental chips lends itself readily to the idea that the clusters were the result of environmental contamination, likely from radon and its daughters. While this experiment was indeed not an exact replication of Oriani’s work, the results, along with the fact that many of Oriani’s clusters (which once seemed to be such strong evidence of something anomalous) also appeared outside of the active electrolysis area of his cell, have lead us to the unfortunate conclusion that any glimmer of a positive result we have seen has been nothing more than an artifact of contamination and has discouraged the rest of the members of the shop to count the remaining chips in my absence. While we can’t rightfully comment on the work done and the results claimed by Oriani, HUG must reluctantly join the ranks who believe this experiment’s positive results to be caused by environmental contamination.

Thanks for your welcoming help Quantum Heat Community! Wish I had better news to share!