Published in Nature Scientific Reports, Bin-Juine Huang et. al paper entitled “Water can trigger nuclear reaction to produce energy and isotope gases” has drawn a lot of attention. I have been asked many times to comment on it. So, hereby offer an honest ‘peer review’ of this work (for whatever it is worth). Please keep in mind that this review comes from someone who is not prejudiced against LENR in principle.

The Goal of Experimental Physics

I do believe that we must hold each other to the highest standards and solicit critical opinions from real experts whose expertise narrowly applies to particular kinds of measurements used to derive experimental data. It is foolish to believe that one can simply pick up an instrument and start measuring regardless of how user-friendly the instrument appears to be. The devil is always in the detail. And the biggest ‘detail’ is the search for and elimination of sources of systematic errors. The analysis and elimination of sources of systematic errors is the biggest problem of experimental physics. LENR research in particular way too often suffers from the following shortcomings:

1. We are not critical enough of our own experimental data;
2. We are too eager to interpret our data as supportive of our preferred conclusions;
3. We completely overlook alternative explanations, and
4. We most often do not spend any time looking for reasons why the instrument produced the ‘spectacular’ or ‘interesting’ readings that it did.

Unfortunately, more often than not instruments produce data that must be interpreted and interpreted correctly, because when taken at face value it could mean anything. This is where the lack of expertise in a certain kind of measurement really dooms an experiment: too many experimentalists believe that they can pick up an instrument and start measuring without spending years on learning and mastering relevant measurement techniques. I would go so far as to say that the real objective of an experiment is not to capture an ‘interesting’ result, but to spend hours, days, weeks and months on proving these result to be wrong. Yes, you read that right. The effort must be on proving the results to be wrong. And only when we have exhausted all possibilities and failed to prove our results to be wrong – only then we can dare to conclude that maybe, just maybe our results are indeed genuine.

This is quite contrary to what most LENR experimentalists actually do. They tend to attain an ‘interesting’ result and run with it claiming that it supports their theory. The real objective of a real honest experimentalist is to prove the ‘interesting’ measurement wrong. It is not a self-defeating task – far from it. This is how we establish the truth: by eliminating all possible alternatives, whatever remains however improbable must be true. Too bad very few LENR researchers actually follow this practice. Mostly because those who do have nothing to report…

It is only when we try, try and fail to find mundane explanation of our ‘interesting’ results, only then we can speak of discovery. And our peers must see this effort in all minute detail in order to be convinced. The burden of proof is on us! We must do the work and show how we did it. Too bad almost nobody does through the trouble… And this is why I daresay LENR did not make any progress in 35 years.

For those, who want to learn more I refer to this blog, which dives into some calorimetry issues and provides pertinent references.

Criticism of the Results

The paper reports on “the discovery that water can trigger a peculiar nuclear reaction and produce energy”. But does it give us convincing evidence? I am afraid the answer is NO and here is why.

1. COP

The paper claims that the coefficient of performance (COP) is greater than 1.05 yet makes no effort to describe the calorimetry procedure. How was the temperature measured? How was the calorimetry setup calibrated? What are possible sources of systematic errors? None of this got any attention in the paper yet these are the most important details to analyze.

Flow calorimetry is notoriously difficult and notoriously prone to errors. The prevailing opinion in the field is that conclusively proving differences in heat production / consumption on the order of a few percent is all but impossible. Granted, I am not an expert on flow calorimetry, so I cannot speak about calorimetric issues with authority; this is where a real expert is needed. One must speak to someone who has devoted their life’s work to flow calorimetry in order to receive a useful feedback.

Nevertheless, I have done some work in basic water calorimetry and I can tell you the following: it is universal understood that power input into the system and power output of a system is nearly impossible to establish by using thermocouples alone because the medium (e.g. water) flow is not uniform, flow meters are not precise, and thermocouples may not be properly calibrated. This is cold fusion 101.

The paper, unfortunately, lacks sufficiently detailed discussion focused on the employed flow calorimetry technique. There is no analysis or discussion of possible sources of systematic errors. Calibration technique is not even mentioned.

So, when I read the paper all I see is a non-substantiated claim. I do not dispute the fact that their thermocouples may have measured an increase in temperature of the out-flowing water with respect to the in-flowing water – not at all! But the authors did not bother to look into the alternative explanation for those readings (in other words they did not mention any effort to look into why those readings might have been in error).

The Presence of ²²Ne

I have to give the authors credit that their discussion pertaining to the presence of ²²Ne is actually more substantial and appears to comprise the bulk of the paper. But here in my opinion the authors made a critical mistake by thinking that a mass spectrometer (such as a residual gas analyzer or RGA) can prove a presence of certain atomic species. All mass spectrometer does is shows a presence of particles with a certain mass to charge ratio forming a mass spec (or RGA) spectrum. There is no species information present in this spectrum. Granted, the species information can be deduced, but deduced does not mean proven. In order to prove that the observed m/z = 22 peak is indeed neon they should have done optical spectroscopy! Optical spectroscopy is universally understood to be the gold standard for establishing elemental composition of gaseous species. Unlike mass spectroscopy, optical spectroscopy is a conclusive technique.

Hence I find the claim that the authors established the presence of ²²Ne in their gases as a mere conjuncture (as opposed to a proof).

But what about their data you may ask? The authors correctly point that ²²Ne+ and CO₂++ produce the same m/z = 22 peak on RGA spectrum, and they their residual gas m/z spectrum on Fig. 5, which I duplicate below.

But where is the RGA mass spectrum showing the m/z = 22? Somehow mass spectrum plots illustrating the purported ²²Ne+ peak are entirely omitted from the paper.

To me this omission raises a pointed question: is the purported m/z = 22 peak quire small and sitting right at the edge of the instrument’s detection limit? Unfortunately, this is a very common problem in LENR, taking too much stock in readings at the edge of the measurement range. But there is no plot for us to see, so who knows. Regardless, what the authors should have done instead – they should have worked to boost the magnitude of this peak until it was sufficient enough for optical spectroscopy to produce distinct neon lines.

Overall I find the CO₂++ discussion of little relevance since this is the wrong technique to begin with. What the authors did is the opposite of what they should have done: their ²²Ne discussion centers around why they think it is neon as opposed to why this cannot be something else (see the first section of this post on the Goal of Experimental Physics). Although they considered the possibility of the peak being CO₂++, but in my mind even here they have not given enough details to convince the reader that they have really eliminated this possibility. But they did put in some effort.

PS Actually they could have lowered the RGA ionization potential below neon ionization energy to see if the purported ²²Ne peak would disappear. Lowering ionization potential creates fewer ionic species thus allowing one to clean up a RGA spectrum by eliminating gases that would not ionize at the given ionization potential.

Other Claims

“Possible nuclear transmutation was found by SEM/EDX inspection of ruptured copper pipe samples (C increases 200–500%, O 300–600%, Fe 400%, and new elements P, S, Ca appears).” This sentence appears out of the blue and amounts to a wild unscientific speculation. One sentence, really? And we are talking about common elements, which are present everywhere. I have done a lot of work with SEM/EDS an I can tell you that you find these elements on every sample. How can you possibly show the increase with respect to a control? How can you exclude contamination? Migration? Deposition? I am not saying that this is impossible (even though I have no clue how you can do it), but the authors did not bother to provide any detail and simply expect us to believe their claim.

Claims pertaining to the formation of ²H (deuterium) and ¹⁷O are equally baffling. On one hand deuterium is always present in water therefore water m/z spectrum contains numerous peaks corresponding to all possible combinations of hydrogen-1,2 and oxygen-16,18 ionic and molecular species (including molecular fragments such as OH). I am not saying it is impossible to establish the presence of the ¹⁷O in residual gas using the RGA technique – with enough work and care, perhaps – but the authors did not spend enough time describing their technique, providing spectra, providing controls and otherwise convincing us that they have really done the work. Instead they jump into ‘isotopic ratios’ as if they are referring to some common run-of-the-mill technique (which might have been the case have they been talking about detecting heavy metal isotopes in water-based solutions using ICP-MS, but this is not what they are talking about).

Summary

I must admit, I am rather taken aback by this paper as it has all the flaws that have been plaguing the LENR research for the past 35 years. We must do better! We really must if we want others to take us seriously. I repeat, we must hold each other to higher standards and solicit critical opinions from real experts whose expertise narrowly applies to particular kinds of measurements used to derive experimental data. Then and only then we can make progress.