How to read LENR experimental results

1st Official Post

    This thread is to review an interesting fairly typical LENR paper
    http://vixra.org/pdf/1309.0070v1.pdf


    Anomalous Exothermic and Endothermic Data Observed by Nano-Ni-Composite Samples
    Akito Takahashi;y, A. Kitamuraz, R. Seto and Y. Fujita
    Technova Inc., 1-1-1 Uchisaiwai-cho, Chiyoda-ku, Tokyo 100-0011, Japan
    Taniike and Y. Furuyama
    Graduate School of Maritime Sciences, Kobe University, Kobe 6580022, Japan
    T. Murota and T. Tahara
    Santoku Corp.


    Oystla summarises its claims as "10X chemical heat output".


    If true that would be a big deal indeed.


    This is not a very complicated paper, but you need to understand what they are doing, and what the numbers mean.


    I should say that although the results here are not impressive and in no way point towards LENR, they are mildly interesting.


    Also, these guys are perfectly reputable scientists doing proper research (though of a very preliminary nature). I don't question their integrity.


    The paper is written in a "house style" suitable for LENR publications. It has two things that are "puff" and would tend to decrease chance of publication elsewhere:
    (1) It headlines figures that are sensational but not in fact helpful because presented out of context.
    (2) It includes some speculation about nuclear mechanisms that is not warranted by the data and does not contribute to the content.


    These guys have some decent equipment, and if they follow up as they say they are going to they will get to the bottom of this anomaly (at least in its current form).


    I'm willing to bet quite a bit that when this happens what they claim is a "must be nuclear" anomaly will become much weaker, and not provide evidence.


    As with all these experiments, the results are sort of in that expected error corridor, in an experiment where methodology could easily be tightened and the effect increased. So, if we do this we expect proof positive of something new. What I predict is that we will not get this. Anomalies will remain, bit the anomalies here will reduce and/or as vanish they tighten up error bounds.


    Before I explain why the impressive-looking figures here are not so great, I'll let oystla say his bit about why they are great!


    When he's done this I'll give my summary of what they have done, how they did it, and what are the issues.

    Quote from Thomas Clarke

    The paper is written in a "house style" suitable for LENR publications. It has two things that are "puff" and would tend to decrease chance of publication elsewhere:
    (1) It headlines figures that are sensational but not in fact helpful because presented out of context.
    (2) It includes some speculation about nuclear mechanisms that is not warranted by the data and does not contribute to the content.


    I 100 percent agree. This paper presents details that will be interesting for someone who is already persuaded that LENR is a thing. For someone who is unpersuaded, the following issues immediately come up:

    • It's unclear what the graphs are showing.
    • Where does the theoretical speculation about lattice displacement, etc., come from? Is that biasing their interpretation of the raw data?
    • What was the control, exactly?
    • Etc.

    There are many other issues that might come under scrutiny as well. All of that said, I thought the paper was interesting.

    Like any debater, Thomas Clarke looks for evidence that support his view that LENR is an illusion supported by bad data and invalid experimental procedures. But Thomas Clarke ignores all evidence that undercuts his debating position. Like Mary Hugo, Thomas Clark is therefore a debater and not a scientist who considered all the data in good faith.

    I will come back with more details on what I think of this paper,


    But the first point I would like to make is the following wrt principlea of Publishing ANY scientific papers:


    On writing a Scientific papers you have to make a few choices and decitions, like;


    Audience: Who will be reading your paper? Usually you will be writing to your peers. Knowing the audience helps to decide what information to include--one would write a very different article for a narrow, highly technical, disciplinary journal vs. one that went out to a broad range of disciplines.


    And this paper in question is no different. It's directed to the peers inside the LENR community. Not to outsiders and "non believers".


    It is published for researchers familiar with the field of research.


    Any sceptics would have to go through the same learnings process as the other scientists that have been convinced after doing LENR research for years.


    Like reading all the referenced papers to understand more of the communication in this specific paper.


    And note, References to the research findings of others are an integral component of any research paper. To understand any Scientific papers, one most likely would have to read more of the referenced papers.


    In this case, the paper have 12 referenced papers. Questions you may have after reading this paper, may well be answerred in one of the referenced ones.


    And that is a usual way of writing papers. You don't necesarily repeat knowledge that your Audience allready posess.


    This paper states in the first sentence of the introduction "Gas-phase hydrogen isotope absorption/adsorption experiments have been performed since 2008 at Kobe University to elucidate the underlying mechanism of anomalously large heat releases [1–5]"


    Allready we are presented the context of this paper. If we want to fully understand the paper, we also would have to read the referenced papers, which the Peers in LENR research allready have done.


    And this is why Thomas never will find a single paper that will convince him of LENR, since all will have som missing info he would want, but which most likely will be found in referenced papers.

    Quote

    Like any debater, Thomas Clarke looks for evidence that support his view that LENR is an illusion supported by bad data and invalid experimental procedures. But Thomas Clarke ignores all evidence that undercuts his debating position. Like Mary Hugo, Thomas Clark is therefore a debater and not a scientist who considered all the data in good faith.


    Axil - if you mean - I've picked a not convincing paper then I'll happily do the same with any other paper.


    If you mean - you could argue on both sides wrt this paper - then you have lost the plot. When the hypothesis (LENR) is extraordinary the null hypothesis (some sort of experimental anomaly) is strongly preferred unless ruled out. Ruling mundane experimental errors and anomalies out is thus the primary job of anyone looking for evidence of LENR.


    Quote

    And this paper in question is no different. It's directed to the peers inside the LENR community. Not to outsiders and "non believers".It is published for researchers familiar with the field of research.Any sceptics would have to go through the same learnings process as the other scientists that have been convinced after doing LENR research for years.


    What you seem to be saying is that heroic LENR researchers have read lots of papers and become convinced LENR is real, therefore this matter is not an issue. That might be true, but it would indicate a strange situation given no killer proof that can convince an objective bystander and gain everyone concerned a Nobel. My instinct, in this situation, would be to put together as compact and bulletproof a case as I could closing all loopholes.


    Quote

    And this is why Thomas never will find a single paper that will convince him of LENR, since all will have som missing info he would want, but which most likely will be found in referenced papers.


    Bulletproof evidence of LENR would be something no researcher could ignore. It would be a big deal. You'd have to be an idiot not to write it up, all loose ends tied up, as a fait accompli. In this case the questions are about things like: "How exactly was the calibration done?". They may have done some similar work in the past, and described calibration, but unless the precise work done on the occasion is available no-one can be sure what was done this time.


    Sometimes, a paper will refer to a precise methodology described in detail in some previous paper, stating any differences and that otherwise the previous method was followed. You'd still want the precise figures, this time.


    Here there is no such care and the authors are obviously not (for this paper) looking to be sure what are the possible errors in their system.


    To be fair, they indicate that they hope to replicate the work with larger quantities of reactant and therefore get data that is clearly beyond error. That is an obvious thing to do.


    Oystla - I believe your ad homs about me here do your side of this debate no service.

    OK - so I'd like to say a bit about how to interpret the eye-catching figures here.


    The abstract headlines an excess power output of 2W/g of Ni


    The paper itself mostly looks at excess energy output measured in eV/Ni atom. There are two types of figures:
    total energy output (over a 700 hour test)
    dynamic energy output in spike (over approximately 5 hours).


    All these figures need decoding by answering some questions:

    • What fraction of total fuel is the Ni?
    • How is energy/Ni atom calculated for the ~ 5 hour spikes?
    • What are the errors in these figures?

    Discussion.
    Obviously, if elements of the fuel other than the Ni can be reactive it makes no sense to consider energy per Ni atom. In this case we have Cu (small qtys, so can ignore) and Zr (large quantities). All these can react with H, so considering only the Ni seems strange.


    In fact, the samples are highly reactive because they consist of nanopowders. There are two consequences of this:
    Chemical reaction rates can be much higher than expected
    Chemical enthalpies can be different from what is expected due to surface effects.


    More details of the samples are given in [7] the paper says:
    H. Sakoh, Y. Miyoshi, A. Taniike, A. Kitamura, A. Takahashi, R. Seto and Y. Fujita, 17th Int. Conf. Condensed Matter Nucl.
    Sci., Daejeon, Korea, Aug. 2012.


    I can only find this as a conference poster here:
    https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/36494/AnomalousExothermicEndotermicData.pdf?sequence=4&isAllowed=y


    This contains no extra information.


    Also relevant is reference [6] from
    http://jcfrs.org/file/jcf12-proceedings.pdf (the first paper, identified with some effort) We will come back to this.


    The samples are "mixed oxide" Zn/Cu/ZrO2. The sample composition is given as both molar fraction, and weight, of Cu, Ni,Zr. It is not entirely clear whether the O is included in the weight. Comparing molar and weight fractions Ni to Zr it is not clear whether O is included in the weight because we don't know the oxidation state of the Ni, but NiO/ZrO2 fits pretty well. In that case the total weight is as given, and the three samples sum to 20g / 10g / 20g. The exact quantity of Ni/Cu/Zr cannot easily be calculated from the molar fraction and total weight (because we don't know the exact composition and therefore fraction of O) however all these errors are quite small.


    It is still frustrating - every loose end like this adds quite unnecessary error and uncertainty to the results.


    The main take-home from this is that since H/D reacts with Zr as well as Ni, and the Ni molar fraction is around 30%, the headline figures are ALL inflated by X3. They remain quite large. More on the exact figures later.


    The other issue relates to the dynamic ad/ab/sorption event figures. The problem with these is clear - hwo do you calculate the number of H/D atoms involved in these phenomena? the authors integrate power spikes over around 5 hours to get energy. They then assume that the H involved in this is that introduced to the sample over this 5 hour period. At least - I think that is what they do. This is not correct, because sudden chemical events causing these spikes can affect any number of atoms in the sample, not just the "pro rata amount". Including all the atoms decreases these values by a factor of 100, so you can see that even a local event involving 10% of the sample could generate these spikes with the real energy per involved atom being 10X lower than the headline figure.


    So as far as extraordinary evidence of LENR goes the dynamic figures - though seemingly higher - are not relevant and apparently much less significant than the average (long-term) figures.


    However in terms of the (chemical) behaviour of these reactive nano-powders the spikes are sort of interesting and may well be responsible for apparent beyond chemical anomalies in other experiments.


    More later.

    Dear Thomas, you accuse 10+ competent scientists of not taking into account any relevant reactions between H/D and Zr in their energy calculations? Please...Well, as I read the paper including [6] and [7] they do actually take All into consideration.


    The excess heat calculation is correlated to Nickel atoms If assuming LENR occurs and it's Nickel atoms that are essential for LENR, call it catalysis of LENR or whatever.


    And for your ad/ab sortion critique:
    You need to read [7] to understand the calculations of H/D Events. This is not new science but fully understood mainstream, nothing magic.


    So the magic is still High excess energy far beyond any possible chemical reactions.


    But anyway: It will be interesting to see what the results will show with the ten times larger reactor.

    And here is another Christmas gift of a LENR paper on Ni-H with large excess heat:


    I think now that Ni-H is a possible better LENR system than the wet Palladium - deuterium system. It's a pitty F&P did not pursue Ni-H, since they actually did include Nickel in their patent from 1989.


    The Authors of the last paper I've read where professor Focardi, Gabbani, Montalbano, Piantelli and Veronesi.
    Paper Title: "Large excess heat production in Ni-H systems"
    Published in the Peer reviewed Italian physics Journal "Nuovo Cimento" in 1998.


    I have not found any criticism (Peer reviewed or not) of this paper. The authors also made a paper in 1994, which was critizied by physcists at CERN. CERN was not able to trigger any excess heat, they saw only excess heat during loading of hydrogen. I've read their paper and it's clear they did not try any trigger mechanism to "turn on" the Ni-H LENR, so they concluded no excess heat other than during Hydrogen absorption in lattice.


    One of the remaining mysteries is what excactly is the trigger mechanism. It's more than pure heat. In the 1998 Focardi et. al paper some trigger mechanisms is mentioned.


    And the Reasons why I think this paper is strong evidence of anomalous heat in Ni-H systems are:


    1. Power input and excess output in the 10's of watts, not milliwatt regions, i.e. Easier to measure outside error margins
    2. Simplicity of their system. No complicated calculations or complicated calibrations required. The calibrations show what temperatures to Expect for certain input heat power, regardless If heat comes from electrical or possible LENR
    3. Two parallell cells to increase confidence of results.
    4. Thermometer registrering total heat, regardless of it's origin (heater or LENR)
    5. Small variations in room temperature would not affect the results, because of the high power regions.
    6. The long test period of excess heat (280 - 320 days), securing accuracy and confidence of results. Indicates longevity of the LENR reactions, as also later Ni-H cells have shown.
    7. Excess heat of 70 watts at less than 100 watts input. Easy to read from calibration curve - far beyond any possible calibration errors.
    8. For cell B a new calibration curve when Nickel is in "excited state" shows clearly higher temperature even for the temperature sensor placed the furthest away from the core.


    There are also similarities with F&P wet cells with Palladium cathodes, that is worth noting:
    - need to load the core material with hydrogen ( Faster than F&P, may be same time span as with CO-deposition of Pd Wet cells)
    - The difference between cell A and B also indicate that this is a surface phenomenon, same as indicated for F&P wet cells with Palladium.


    And how can we scale this up and get more energy? Well, why not try more surface area, i.e. Nickel Powder.....ooops someone is allready onto that one ;)


    And with 900 000 KJ of excess energy you could heat 2,7 m3 of water from 10 degC to 90 degC....some serious amounts of excess energy.


    Paper reference :
    http://www.lenr-canr.org/acrobat/FocardiSlargeexces.pdf

    Quote from oystla

    Paper Title: "Large excess heat production in Ni-H systems"


    Quoting from the end of the paper:


    Quote

    The mean power excess of each cell multiplied by its activity period (278 days for cell A and 319 days for cell B) gives an energy excess of about 900 MJ for cell A and 600 MJ for cell B.


    I wish they had calculated the actual integrated power to get the excess energy (XE) rather than relying on the mean power. Also, I wish they had been able to show that one of the cells was XE<=0 (i.e., was a control).

    Quote

    Thermometer registrering total heat, regardless of it's origin (heater or LENR)


    Thermometers register temperature, not heat, total or otherwise. This sort of misinterpretation causes a lot of difficulty in so-called isoperibolic calorimetry and many other LENR-related endeavors. A temperature at a fixed location may *relate* to heat flow but the relationship is never iron-clad and depends on variables which may not be under adequate control. Want heat measurements? Use heat flux transducers and reliable types of properly calibrated calorimeters.


    I didn't read the paper so the above may not apply to it but it would not surprise me if it did.

    "Thermometers register temperature, not heat, total or otherwise."


    Yes, of course I meant temperature, heat was an inaccurate Word here. Very important to use the right words, especially related to controversial subjects ;)


    "A temperature at a fixed location may *relate* to heat flow but the relationship is never iron-clad and depends on variables which may not be under adequate control."


    Yes, and The most important variables in this case would be room temperature and convective air flows.


    Room temperature where continously monitored. Small variations of convective air flow would not have important impact at these power levels and duration of test.


    Heat calibration curve between power input and temperature will therefore give an accurate relationship between temperature and energy heat flow.

    We are now on the topic of how to evaluate calorimetry. (And I want mainly to consider the paper which is the topic of this thread).


    However I'll say something about the Piantelli distraction here.


    The calorimetry here, as MY points out, is isoperibolic - beloved of LENR researchers. The problem with this is that it is impossible to avoid potential large artifacts under the conditions that LENR researchers typically use it. Specifically, the power is inferred from temperature measurements, and validated by calibration.


    Unfortunately when, as here, the inner sample is heated up to a high temperature there are many possible artifacts. The two temperatures (sample and inner reactor wall) relate to heat flow by the corresponding total thermal resistance.


    For sample we have: convection, radiation, and conduction, both to reactor wall and (mostly conduction) to the outside.
    For inner reactor wall we have: convection and radiation to outer reactor wall (+ some conduction).


    The safety of the measurement relies on the transfer coefficients of all these heat flows staying the same. At the high temperatures used here all materials will change there are many different things that could affect these flows.


    Note that even if the inner/outer wall heat flow is well controlled (and let us suppose that other elements, like conduction, are similarly well controlled) we have an unknown error from the variation in thermal resistance of sample <--> inner wall and sample <-->outside (the authors note that the latter is significant).


    For this type of system to be safer you would want to reduce as much as possible the sample <--> outside thermal resistance. If this could be bounded low enough then all the heat flows sample <--> inner wall <--> outer wall. The inner to outer wall thermal resistance is the one it is easiest to calibrate and control.


    The authors note that the system is bistable with some hysteresis. Once heated past some trigger point it exhibits higher temperatures (for the same heater power) than when cooled down and approaching the same temperatures from below.


    That does not help. Many of the mechanisms that vary the different heat flows are potentially hysteretic.


    So what we have here is a system so complex and difficult to instrument that:
    (1) large errors are possible
    (2) the errors cannot be quantified easily via calibration.


    You'd think, from these LENR experiments, that accurate calorimetry is impossible! But in reality there are perfectly good ways to measure these heat flows, using mass flow calorimetry where it is easy to reduce and control artifacts. F knew this (but did not follow it in his later LENR experiments)


    You will find that the large apparent heat excesses all come from isoperibolic calorimetry in setups like this where the various thermal resistances are very variable.

    Thomas, your last post surprises me, did you think it through ;) ?


    If you have a box, pipe or whatever container and measure the outer wall temperature, the heat flow is governed by the temperature difference between outer wall temperature and surroundings by conduction, convection and radiation to the surroundings.


    It does not matter how the internals look like, number of walls, chambers, heaters etc.


    An even outer wall temperature will therefore have a certain heat flow to the surroundings by conduction,convection and radiation. Therefore a calibration curve will work of outer wall temperature vs internal power.


    And therefore it does not matter If the heat arrives to outer wall from the electrical heater or from the inner core as LENR heat.

    colwyn:


    annealing


    mechanical creep



    But the point is not whether I can think of examples. I'm probably not imaginative enough to think of the right one! LOL - experimental errors are ever this - the thing you never though of...


    The point is that no-one can prove there is NOT such a mechanism. The solution is easy - use better calorimetry. Trying to make isoperibolic calorimetry safe under these extreme conditions - high temperatures, free H - is a mugs game and more to the point you can never prove you have done it because errors cannot be separated from genuine results.

    @Tom,


    Concerning isoperibolic calorimetry as a general topic, beyond the paper by Focardi et al. referenced above, you should bring your analysis of its inherent problems to the Vortex list and defend your thesis there. (This is a trap: you will be an innocent lamb waiting for Jed Rothwell to pull it apart into tiny pieces and methodically deconstruct each one. He is very knowledgeable on the topic. He may even agree with you on a point or two.)


    Eric


    (Curious enough about the topic, I've gone ahead and submitted a question of my own, with a link to your claims.)

    Thomas, I agree up to a point about mass-flow calorimetry.


    As for not being able to disprove (imaginative?) error mechanisms, we could reduce the argument to the absurd: I could posit that invisible aliens came down and zapped in some extra energy with their ray-guns.


    Having clearly stated that, we can roughly judge the potential likelihood (or magnitude of the effect?), and decide whether the burden of proof rests with me, or remains with the original experimentor...

    Quote from colwyn

    Thomas, I agree completely about mass-flow calorimetry. In fact, I'd even say the point is an obvious one.


    As for not being able to disprove (imaginative?) error mechanisms, we could reduce the argument to the absurd: I could posit that invisible…


    The different is that the imagination I posit (creep, annealing, etc) is within normal everyday events, that which you posit (invisible aliens, LENR) is extraordinary and not known ever to exist.


    The point is:
    (1) the burden of poof here is with the scientist who wants to claim an extraordinary mechanism
    (2) the claimed m,echanism (nuclear reaction) has such large fingerprints that however you slice it, if real, it would be easy to prove.


    These two facts together are the context that is lost by LENR researchers because they live in a bubble in which they think LENR is ordinary.


    Think about Lugano. How many people, when challenged, would see the mistake? I know I did not, for a long time... Why should artifacts be easy to work out? If you have done any experiments yourself you will know that once systems get complex the results you get are often surprising due to unanticipated artifacts.


    The things about LENR researchers is that instead of looking hard for these artifacts, like everyone else, they think the artifacts are real. To be fair some LENR researchers go on testing specific experiments and end up discovering the artifact themselves. Others do what will happen in the paper this thread relates to. They scale up the active material by 10X. They find the anomaly does not in increase by 10X. Or they do better calorimetry and finds it vanishes.


    Look at NANORs. They have the magic property that larger ones show a much lower COP than smaller ones. And they are tested by isoperibolic calorimetry...

    Quote

    Having clearly stated that, we can roughly judge the potential likelihood (or magnitude of the effect?), and decide whether the burden of proof rests with me, or remains with the original experimentor...


    I agree this balance of probabilities is the key issue. Not difficult, because the prior probability for LENR is so very low, compared with "artifact".


    If you point out that the precise artifact cannot easily be identified I'd reply by saying the precise LENR theory cannot at all be identified, and that the lack of artifact identification is simply because there are so many possibles, that does not decrease likelihood.

Subscribe to our newsletter

It's sent once a month, you can unsubscribe at anytime!

View archive of previous newsletters

* indicates required

Your email address will be used to send you email newsletters only. See our Privacy Policy for more information.

Our Partners

Supporting researchers for over 20 years
Want to Advertise or Sponsor LENR Forum?
CLICK HERE to contact us.