307) More private messages about Colorado2

Ludwik Kowalski; 8/9/2006
Department of Mathematical Sciences
Montclair State University, Upper Montclair, NJ, 07043

I believe that private correspondence described in unit #285 is worth preserving. It refers to Mizuno-type experiments in which I personally participated. It started as Texas1 project, became Colorado1 project, became my ICCF12 report and will possibly be participation in Paris3 project. Unfortunately, I am still not convinced that excess heat, reported in Paris1 and confirmed in Colorado2, is due to some new Physics. It may be due to a prozaic effect, such as presence tiny droplets, and colloidal particles, in wet steam. Or it can be due to errors in measuring the input energy. Will Paris3 finally convince me that what we called excess heat is really excessive? This remains to be seen.

Why do I spend time on cutting and pasting so many private messages? Because I believe that what we are doing is very important, no matter how the controversy will be resolved. Private messages provide details which might not be apparent when published reports are read. Our correspondence will certainly be appreciated by future historians of science. We are not famous scientists but our messages might become as important as private letters of Galileo, Pasteur, Fermi and Einstein. We are a small part of something significant; future generations will probably study all aspects of our activity with great interest.
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
1) Pierre --> Richard + Ludwik (3/8/06 ?)
Before going by the end of this week with our dewar tests, we have done with Gérard a series of experiments with our traditionnal beaker, just to see if we were able to find again abnormal excess heat.
A long run (30 minutes) at 250 volts: COP between    1.05 to 1.10
2 runs at 300 volts ( 15 and 20 minutes): COP around 1.10 up to 1.20
2 runs at 350 volts (only 12 and 10 min due to cathode burning) COP 1.15 to 1.32
Of course, we made the usual calibration with our ohmic heater ...
We really think with Gérard that we are now able to get abnormal excess heat on demand, within, of course, a particular optimised geometry ( beaker, anode and cathode)...except unknown artefact !
 
I told you that, as in Boulder, we failed in trying to get automatic constant length for the cathode by using gravity. Our small ceramic cup was pierced and even a second one placed on the first one. Happily, the bottom in Nickel stopped the plasma electrolysis. See on the pictures attached. But, however a good result : this cup bottom gives a very good behavior to the electrolyte (no splashing, no waves, no foam...)
 
As asked by Scott, you will find a small movie giving, I hope, plasma colour and sound of our CFR running ( I think here at 300 volts). We will get by the end of the week the Boulder electrolyte analysis...
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
2) Pierre --> Richard + Ludwik
We tested yesterday our dewar by beginning the traditionnal calibration with our ohmic heater. We got the following values:
 
  At 600 watts:   mean value for thermal losses:     40 watts
  At 400 watts:    ---                    ---        40 watts
  At 300 watts:         ---                    ---   29 watts
  At 200 watts          ---                    ---   21 watts
 
Compared to the same tests with our beaker, these values are about one third of the beaker values...
 
Then, we replaced the ohmic heater by our usual electrodes and we made one test at 300 volts during 20 minutes. We have obtained COP values around 1.15 from time 5mn to time 15 mn and even 1.30 at the end of the test ( Pth was about 386 watts at time 20 mn).
 
We tried then several tests at 350 volts. We got good COP (even> 1.3), but we had splashing... The main reason is that, with a beaker, it is very difficult to adjust the length of the active cathode and to see at what depth you really placed the electrodes. A transparent dewar would be a good answer to this problem...
 
But anyway, we think with Gérard that we have found same results with the dewar and with the beaker. We prefer the beaker, easier for the setting up, and also better than the dewar because we can place together the electrodes and the ohmic heater, due to larger diameter for the beaker...
 
Three pictures are joined. I hope by to morrow to have some chemistry analysis...

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

3) Pierre --> Ludwik
Je viens d'envoyer la note ci-dessous à Richard S. et Ludwik K. comme ultime contribution dans le contexte de ma venue à BOULDER. Ludwik comptait faire de nos travaux une note de synthèse. Je ne sais si c'est toujours son intention, à moins qu'il n'attende une duplication par Scott Little. J'ajoute pour répondre à vos questions que le brouillard qui apparaît au dessus du bécher apparaît de la même façon lorsque seule la résistance est opérationnelle. Par ailleurs, je ne comprend pas le débat sur le CO? pouvant s'échapper du bécher. Le CO? potentiel dans les 27g de CO3K2 mis en place dans l'électrolyte représente 8.6 g (le tiers de la masse en carbonate de K). Or, chaque essai consomme environ 250g de H2O et avant de renouveler l'électrolyte, nous faisons des dizaines d'essais. C'est donc pour moi un faux problème. J'attends cependant les mesures chimiques en cours avec Gérard.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
4) Pierre --> Ludwik+Richard+Gerard+ Jean-Francois Fauvarque
Finally, Gérard gives me the following answer (I am not a chemist and I am just trying to translate his note): 1cm3 of electrolyte was mixed with 50 cm3 of distillated water. HCl at 0.05M was used for dosing. You can see the result in the joined curve. First equivalent point for 1.71 cm3 Second equivalent point for 4.91 cm3 that is to say: 4.91-1.71=3.2 cm3
 
The first point seems to correspond to the neutralisation of a strong base: KOH  (0.05x1.71=0.08M) and the second to the neutralisation of the remaining carbonate of potassium ( 0.05x3.2=0.16M). This electrolyte seems to be a little bit concentrated, but the answer is clear : due to the numerous runs we did with, we have not to be worried by the escape of CO2....!

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
5) Ludwik --> Gerard+Pierre+Richard+Jean-Francois
I have no idea what is the meaning of the curve sent to us. How does it help us to make the data more trustworthy? Can this curve be used to show that not much water was lost in the form of tiny droplets?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
6) Pierre --> Ludwik+Richard+Gerard
My previous mail was not explicit enough.....One liter of electrolyte contains 27 g of K2CO3. Inside this carbonate, there is about 8 to 9 g of CO2, which may be released by contact with the very hot plasma. This may be a cause of overestimation of the excess heat. This chemical analysis has shown that our Boulder electrolyte contains always 80% of the initial carbonate content in spite of a lot of runs and hundreds grams of evaporated water. Then it is clear that this very small release of CO2 (20% of 8 g, that is 1.6 g) leads to an insignifiant and negligible correction on all the evaporated water... That's all is said by our chemical analysis ( we will do the same with the initial electrolyte, not used). Are you now OK

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
7) Pierre --> Richard+Ludwik+Gerard
Hi friends, I am to-day at the CNAM Fauvarque lab. and we are preparing with Gérard some experiments Paris2 type to be done on to-morrow friday.  The purpose is to answer the questions raised by Ludwik and to help Pr Fauvarque to prepare his appendix on chemical aspects. We are very interested by the projects of Richard, but be careful Richard with the H2 O2 explosions... We plan to do an experiment similar of the CFR of Jean-Louis Naudin within two weeks, we hope. It seems to us that the thermal losses will be smaller and the global analysis simpler... We will see...

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
8) Ludwik --> CMNS LIST (April 20, 2006):
Colorado2 experiments replicated the COP>1 claim of the Paris-1 team. They showed that by doing the same things we obtain similar results. That was an important step forward. To make Colorado-2 results publishable we must be able to convince honest referees that excess heat cannot be explained in terms of known chemical reactions. That is the main issue. Unfortunately, it has not been addressed by recognized authorities in chemical aspects of high voltage electrolysis. Did Mizuno, or some of his collaborators, publish something along these lines in Japanese?

My understanding was, and still is, that a complete analysis of potentially possible exothermic chemical reactions is now being conducted in Paris. Im I correct? Planning for new setups, ahead of such analysis, does not seem to be productive, at this stage. Writing a paper based on Colorado-2 results also seems to be premature, for the same reason. But we should compare the conductivity of water (from which the electrolyte was made) with conductivity of condensed liquid escaping from the cells (when the COP was measured). In that way we can estimate the percentage of tiny droplets of the electrolyte in the escaping steam. . . .

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
9) Pierre --> Richard+Ludwik+Gerard+jean-Francois (April 22, 2006)
We are also trying to launch a new experiment "Paris 3" with a loop of ordinary water for cooling our experiment and with presently hand measurements giving water flow and inlet and outlet temperatures... I find as a very good idea your choice to take water at 0°C in order to cool the experiment, but of course you must be careful with thermal contributions ( instead of thermal losses...). But I do not understand why your electrolyte is not at 100°C ( or maybe at 94°C in Erie ! ). It is easy to do so with an ohmic heater... and it seems to me that the experiment analysis is simpler when every data are not all varying ( power, temperatures, ...)

You will see in the joined pictures our first experimental set up, here without any heat insulator. Our first tests with the traditionnal ohmic heater show us too large losses of uncondensed steam ( easily measured by our balance), about 40% at 400 watts and even more at 550 watts... Due to the fear of H2-O2 explosion, we managed a too large escaping gases area, in spite of screens located in the center in order to deviate steam on the nylon tubes at the peripherical side... We are now writing our chemical appendix... But, Ludwik, don't be too much impatient.. Gerard and Pr Fauvarque are very busy fort the time being.. I will send you soon the summary !

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
10) Richard --> Pierre + Ludwik+Gerard+Jean-Francois (4/24/2006):
Bonjour Pierre and company, I am also starting up Colorado #3 experiment. I tried a similar set up to what you are showing in your pictures and found that it didn't work very well for me. I ended up putting the condensation coil just above the electrolyte attached to the lid of the cell. This arrangement works well and does a good job of condensing the steam. The lid over the cell only needs a small hole 3.0 cm in the center to let out the accumulated gas. In the pictures you can see that the cathode comes out the hole. I'm able to push the cathode down a little at a time and keep the reaction going for hours. I'm at work right now but I'll send some pictures tonight.

In this new arrangement I am still losing too much water as steam but it's only 50 grams/hour. I'm working on some baffles to force the steam to travel around the coil. I would like to get this down to less than 10 grams/hour. I also need to insulate the lid of the cell. I seem to be loosing a lot of heat to non-evaporated loses and it's almost all though the lid of the cell.

I understand your concerns about thermal contributions. I have thermistors at the inlet and outlet of the coil to measure the input and output temperatures. I'm using a big ice chest right now for my cooling loop. The ice chest gives me several hours of 1-2degC water for the test.

My initial measurements with this new arrangement look good. I can't claim any excess heat. It looks like with just a few more modification I'll be able to run a test for several hours with continuous measurement of the input and output power. I agree it's very important that we have continuous measurements for several hours showing excess heat.

I think this experiment with little loss of water over the duration of the test and continuous excess heat for several hours will be very convincing. Running the test for several hours and taking the before and after electrolyte samples will gives us ideal data for a paper.

Ludwik, do you want to come to Colorado again and run help run experiment #3? I should be ready with all the new setup in about a week.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
11) Richard --> Ludwik (4/25/06):
I’m very encouraged by the results I got last night from my experiment.  The heat exchanger is working nicely with good results.  I’m seeing a COP of about 1.30 for over 90 minutes at 300 volts and 0.95 amps.  I have a constant 3Deg temperature difference in the heat exchanger with a flow of 1.200 liter/minute.  The ohmic heater calibration is showing a heat loss due to conductive and radiated of about 40 watts.  Even if I only use the heat exchanger values I seen a COP = 1.1. Attached is draft of a paper for Colorado #3.  This draft is first a way to make sure that my thought processes for the experiment are correct.  I know I’m a long way from publishing anything.  Your comments are appreciated. 

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
12) Ludwik --> Richard
See green comments in the attached file.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
13) Richard --> Ludwik
Thanks for the comments. Does the experiment sound solid?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
14) Ludwik --> Richard
I can judge what you do on the basis of what you write only. The preliminary description is not sufficient to form an opinion. A flow calorimeter seems to be appropriate when excess energy is generated at the rates exceeding ten watts. How accurately do you measure the flow rate? How much liquid does manage to escape? Suppose that the escaping liquid is 50% droplets. Can this alone explain the COP=1.3, for example? These questions must be addressed; devil is in details.

The most important task, for Colorado-2 and Paris-2 experiments is convincing chemical analysis. I am surprised that Fauvarque, who is presumably a top French electrochemist, did not address that issue in describing results of Paris-1 experiments. I hope this has been addressed in 1997 by Ohmori and Mizuno. Otherwise claims of "unexpected heat" cannot be made. Pierre promised Fauvarque's appendix to me but nothing happens.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
15) Richard --> Ludwik (4/26/06):
Ice water is feed to the condenser at 1.20 liters/minute.   
 
Starting time 8:10   Ending time 9:20                  total time 70 minutes
 
The delta temperature at the heat exchanger is 5 degree a 10K @ 25c thermistor
The delta temperature of the outlet water from the ice bath compared to the returning water was also 5 degree
 
These temperature reading are an average of about 15 random data points over the 70 minutes.
 
Input voltage 355 and input current  1.2 amps   ==  426 watts
This is also a random average of the input voltage and current.  
 
The initial weight of the vessel was 2123 grams and the final weight was 2033 so losses due to evaporation or disassociation of H20  is 90 grams
 
Pin       = 285
Pout     =  Pex + Plosses + Pevap        
Pex     =  5 * 20 *4.178  = 417 
Plosses =  I’m estimating it at about 25 watts. 
Pevap  = 2260 * 90 /( 70*60) = 50 watts
 
Pout     = 50 + 417 + 25
Pout     = 507
 
COP  = 1.18

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
16) Ludwik --> Richard (4/26/06):
Thanks for numerical information. Congratulation for being able to forge ahead so fast. Can you also describe to me your condenser? Where is it located in the 2 liter container? How is it connected to the water pump, etc.?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
17) Richard to Ludwik (4/26/06):
I purchased a 1.5 liter microwave safe plastic container with an air tight lid from the hardware store. I modified the lid to hold a coil of copper tubing 1.5 meters long. There are two pipe fittings on the lid connecting to the coil. Inside these two pipe fittings I have the two thermistors to measure temperature. In the center of the lid is a 3/8 inch hole for letting out steam and H2-02. I also have some baffles below the coil so the steam can not rise straight up but must move in a convoluted way. This prevents any splashing out the hole in the center. I cathode extends out the 3/8" hole so I can just tap it down a little to keep the reaction going. It fits snuggly through the baffles.

I have an ice chest with a submersible pump. I've let the pump fill a 2 liter beaker about 10 times for 30 seconds. The liquid in the beaker was always 0.6 liters. It was right on the 0.6 liter mark. Looks like I can read the beaker to within 0.01 liter easily.

I have plastic hoses going from the ice chest to the beaker lid. These are both covered with pipe insulation. If I fill the ice chest with 20 lbs for ice I can run for about an hour with the inlet coil temperature being at close to 2degC, after that the inlet temperature starts to rise slowly as more of the ice melts. It looks like as longs as the temperature of the water in the ice chest is below about 40C I can still get good data. I'm bidding on several chillers on ebay, just lost the one I really wanted.

I’m really excited about this experimental setup. It really seems to address a lot of the problems with the open beaker. I hope I can get Advance Energy to run the SEM on the electrolye samples and on the cathode.

I would like you to consider helping publish the paper I'm writing and sent you. I should be able to finish up the missing part in the next month. Here is the latest draft I have added some dummy graphs that show what I'm expecting the final data to look like.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
18) Ludwik to Richard + Pierre (4/26/06):
I will start writing a draft of the paper after receiving two essential items:

a) An appendix written by Fauvarque (showing that known chemical reactions cannot possibly be responsible for the COP=1.24 (st. dev. 0.13) that was measured. I take it for granted that he is considered to be a recognized authority in electrochemistry. When should I expect his input. My experience with publishing is a rule: "the longer one waits the more difficult is to write." Sometimes motivation for writing goes away.

b) An estimate of the percentage of tiny droplets, based on some kind of experimental data. If the answer is ~5% then the accuracy with which the number is determined does not have to be excessive. But the open cell approach will be disqualified if the answer turns out to be close to 20%.

I am happy that you are making progress with Colorado-3 and Paris-2 experiments. Does it mean you no longer want to publish the Colorado-2 results in a separate paper? I would go along with your decisions.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
19) Ludwik to Richard:
Sounds like a big step forward. I wish I could assist you again. But not now. After you stop improving the setup you should perform about ten measurments of COP at each voltage. In that way mean values and standard deviations will help to resolve the pending issue. As you know, according to Paris-1 the COP goes up with voltage. But this was not confirmed by Colorado-2 experiments. That contribution alone will be a good reason for publishing an additional paper. On the other hand, considering the problem with droplets, perhaps your new results should be included into the paper we anticipated with Pierre. Just mussing, as you say.

20) Richard --> Ludwik (4/26/06):
I've got a few more things to do and I should be able to start collecting a lot of data for statistical analysis. Like you say we need about 10 runs per voltage level for good stats. This is why I don't understand what Michel is doing or not doing. You've got to have a way to get enough data without taking months.

I have got get some sort of circulating chiller. I keep running out of ice. I'm making a little printed circuit board that will have the 250Mhz AD835 multiplier chip. This will end once and for all the input power measurment controversy. I'm going to change the type of thermistor I'm currently using. I can get a different value that will give 0.1 resolution on the temperature. Repeatable is even better at 0.01.

If I can get AE to do the chemical analysis with their SEM I'll go back and run a couple of test like we did in the Colorado 2 experiment and have them analysis. Apparently they can show a plot with elements as light as carbon. They can't show any kind of consentrations. Hope all is going well with the closing. It's a lot of work to move especailly when your downsizing! Sorry you won't be able to help, but any suggestion on the paper will be appreciated.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
21) Pierre --> Richard+Ludwik (4/27/06):
I will be happy to have your pictures of Colorado#3 and of course to know where you are now in this new experiment... I am going now for two weeks of vacation in the south of France and I will finish these holidays by doing Paris1 test in the lab of JP Biberian (at Marseille) ; that will be Marseille1. Jacques Dufour and maybe Michel Jullian will be there. By the way, what about Scott Little ?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
22) Richard --> Ludwik (4/27/06):
The attache d file is a sample of the type of report Advance Energy can run for me. I'm going to ask them to do a scan of an electrode.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
23) Ludwik --> Richard: (4/27/06):
Information provided in the attached file -- locations of characteristic peaks (in keV) for different elements -- is of no interest to us. This is important before performing the X-ray fluorescence analysis. Once you know which peak (or peaks) to use to identify potassium the information we need is in the heights of peaks. Suppose we are interested in K and we know that it produces a peak at 321 KeV (I am just making it up). First we expose the 1 cc of the original electrolyte to X-rays for 10 seconds. Suppose it produces 10000 counts under that peak. We dilute the electrolyte by the factor of ten and the peak become 1000 counts. We dilute by another factor of ten and this time 1 cc shows a peak close to 100 counts. That is the calibration. It shows that counts under the peak, N, are directly proportional to the molarity, X, of a sample.

X=0.2*N/10000

For example, if N=20000 then X=0.4 M, if N=2000 then X=0.04 M, etc. All this is preparation. Now comes the essential moment. Suppose we take 1 cc of the condensed liquid. Suppose the 321 keV peak has only 2000 counts in 10 seconds. This would mean that the concentration of K ions is about 5 times less than in the original electrolyte. Is it good or bad for us? I think it would mean that about 20% of the lost liquid consists of the droplets of the electrolyte. The COP would be nearly 1.00 if that percentage was taken under consideration. On the other hand, a peak with N=100 would tell us that droplets contributed only ~1% of the lost mass. That is too little to explain the COP>1.2 as an artifact due to droplets.

Scott Little has a setup for that kind of analysis. It took him less that 30 min to demonstrate presence of tungsten in the electrolyte for me. I was planning on asking him to perform the K analysis for us, after you produce a sample of the condensed liquid. My suggestion was to use the cold funnel method. But any other method would be OK. The experiment must be performed in the open cell, as we did when the COP=1.24 was determined. As you know, I prefer to finish Colorado-2 before starting Colorado-3, even when we know that Colorado-3 method is better.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
24) Ludwik --> Scott Little, Pierre and Richard (4/28/06):
Hi Scott:
1) As you know, presence of tiny droplets in the lost liquid is one of the possible explanation of the high COP. One way to show that the COP=1.24 for example, less that ~1% of what is in 1 cc of the original electrolyte. On the other hand ~24% would confirm the suspicion of the illusion. Do you agree?

2) I did talk to a chemistry professor at my university but he said they have no instruments to measure K. This weekend Richard will have samples to be analyzed.

3) Your X-ray fluorescence setup is probably ideal for what we need? Am I correct? If so then perhaps you will be able to analyze Richard's samples.

4) How is your gravitational project? Are you still planning to verify Colorado2 results? Pierre wrote to me that a very convincing appendix, written by Fauvarque, will be sent to me soon. It will presumably convince honest skeptics that excess heat cannot possibly be due to known chemical reactions. Not being a chemist I will simply accept this; Fauvarque is a recognized authority in the field of electrochemistry. I do not want to start writing the draft of our paper before receiving the appendix, and before knowing that the mass of droplets was a negligible fraction of the lost mass. Should I also wait for the Texas2 results? I hope the answer will be positive.

5) Jean-Paul Biberian, if I recall correctly, also has a setup to perform the X-rays fluorescence analysis. He and Pierre are planning on a new plasma electrolysis experiment in France.

6) I am going to be busy this summer with downsizing activities (emptying the house, selling the house, buying an apartment, and moving). But this should not prevent me from writing the draft. In my mind those who worked on Paris1, Texas1, Colorado1 (before the ICCF11), Colorado2, Paris2 and Texas2 experiments will be the authors (listed in alphabetical order). Let me know if you do not like this suggestion.

7) Rising gas prices made me think about your home-based project for the agricultural fuel, Scott. Are you making progress on that front? Most cars in Brasil, I was told, run of fuel made from sugar can.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
25) Scott --> Ludwik + Pierre + Richard (4/29/06):
“2). . . On the other hand ~24% would confirm the suspicion of the illusion. Do you agree?” Yes, that seems correct and obvious. To analyze the K content easily, how about just weighing a small amount of the sample (i.e. 1 gram), evaporating to dryness, and weighing the dry salt? Yes I know there is some concern that the K2CO3 will turn into KOH but that won't create a huge error in the results.

“3). . .If so then perhaps you will be able to analyze Richard's samples.“ Yes, I could probably make a reasonably accurate analysis for elemental K content that way. I need about 15 mL of solution to fill the sample cup properly. Actually if that is a great difficulty, I could probably dilute the samples 10:1 and still attain satisfactory accuracy....so I could use as little as 1.5 mL of sample.

“4) How is your gravitational project?” We've completed construction. Now we are at the hard part: making it work properly. It will be difficult, I'm afraid. “Are you still planning to verify Colorado2 results?” Yes, but I don't know exactly when. The gravity project is out top priority right now and it is not finished.

“Should I also wait for the Texas2 results? . . . “ That depends upon when you want to complete the paper. Perhaps in another month, I'll get to the Texas-2 project.

“6) In my mind those who worked on Paris1, Texas1, Colorado1 (before the ICCF11), Colorado2, Paris2 and Texas2 experiments will be the authors (listed in alphabetical order). Let me know if you do not like this suggestion.” You only need to list me as an author it I actually get around to doing Texas2. If I am still busy with the gravity experiment and you want to get the paper finished, you may just mention the Texas1 experiments without including me as an author.”

“7) Rising gas prices made me think about your home-based project for the agricultural fuel, Scott. Are you making progress on that front?” No. Our process does not make sufficiently pure biodiesel and we do not know how to improve it. I, too, am not a chemist and that makes me uncomfortable with this kind of experimentation. I just feel like I have no clue what to try next.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
26) Richard --> Ludwik (5/5/06):

I'm not sure how to read this chart but it's suppose to show percent of
wet/dry steam at different pressures and temperatures.

http://www.chemicalogic.com/download/mollier_chart_english.pdf

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
27) Ludwik --> Richard: (5/5/06):
My understanding is as follows:

1) Suppose we have a long cylinder with a piston. It contains some water, for example, 1 Lb or 1 kg. Space above water is vapor (air was pumped out and a very small amount of vapor accumulated). The system is said to be in equilibrium, at any given temperature, when the rate of evaporation and the rate of condensation become equal.

2) At room temperature the equilibrium is when nearly 0% of water is vapor (stem) and nearly 100% is liquid (water). Suppose we supply heat but keep the pressure constant. To accomplish this the volume must be increased. We do this by pulling the piston out (as much as necessary to keep the pressure gauge at the same reading). By adding heat (enthalpy) we create more vapor and less liquid. That what the numbers 10%, 20%, etc. are above constant P line.

3) Saturated (100% dry) steam, I guess, is along the 100% line (on the right). Saturated liquid, I guess, is the 0% line (on the left). The distance from the 0% line to the 100% line is the latent heat of evaporation, L, for any chosen P.

4) Instead of plotting P, along the vertical axis, one can plot temperature (P*V=RT for one mole of an ideal gas). That is what is shown in the second figure in:

http://www.spiraxsarco.com/learn/modules/2_15_01.asp

5) I think that steam / liquid percentages (at constant P or T) have nothing do do with what we want to know (% of tiny droplets lost). But the second figure in the above URL reminded me about something I wrote before. Somebody responded that my concern was not valid. But I was not convinced. The figure shows that the latent heat of evaporation (L~2250 j/g) depends on the temperature. It decreases when T becomes higher. This might be responsible for the excess heat we measured. The temperature at which steam is formed (at the surface of the cathode) is very high and not clearly defined. Our system is certainly not at equilibrium. The assumed L=2260 J/g might be wrong. Suppose L=2000 J/g but we are not aware of it. We multiply the mass lost by 2260 and this gives us 10% larger thermal energy than it really is.

I will address this issue at the CMNS list again. Perhaps this time somebody will convince me that using L=2260 is OK.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
27) Pierre --> Ludwik + Richard (5/25/06):
Pr Fauvarque gave us yesterday the draft of the "chemical" appendix. We hope to send you this note by the end of the next week. I met the same kind of difficulties for Marseille 1 as in Boulder for Colorado 2 during the first days. As soon as we change a little bit the design ( for example a new beaker higher than the old one in order to avoid definitively the droplets problem, we have problem with the COP values ( only 1.10 to 1.12 ). And also the wattmeter instability. I hope that Richard will give us soon better news.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
28) Ludwik --> Pierre + Richard (5/26/06):
If taller container gives smaller COP then we must worry about tiny droplets. It is natural to expect less droplets when the container becames taller. In a recent CMNS list posting Michel Jullian wrote: “There was _some_ salted water escaping, by which amount it inflated the COP we don't know yet (among the few excellent qualitative tests performed, . . .) “ What kind of tests were performed and what percentage of droplets was found? I need the answer to this question in order to start drafting our paper. Also how many grams of tungsen is consumed per one kJ of excess heat? I am referring to experiments in which the COP is close to 1.24, as in Colorado2.

Michel also wrote: “Strangely, the addition of a 470µF decoupling capacitor directly across the cell's electrodes in the afternoon yielded sub-unity COP readings (0.8 or so) instead of the over-unity COPs (1.25 or so) which we had obtained in the morning confirming the COPs observed in the previous Fauvarque-ICCF12 type GDPE experiments.” That seems to contradic what Pierre wrote about the COPs in Marseilles -- 1.10 and 1.12. I suppose that the COPs close1.25 were from the original cell, not from the taller cell. I am also puzzled by the effect of the 470µF capacitor on the COP.

How should the last sentence in Pierre’s massage be interpreted? Are we ready to write a paper based on Colorado2 experiment or should we write a paper based on Colorado3 experiment instead?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
29) Pierre to Ludwik + Gerard (5/28/06):
Je ne remets pas en cause nos résultats et notamment ceux de Boulder. Je pense seulement que cette expérience est délicate et que certaines conditions n'étaient pas remplies : lueur jaune-orange du plasma, intensité voisine de 1A à 1A.15, alors que nous avions des valeurs trop élevées, proches de 2A....etc.. Il faut donc reprendre cette expérience calmement et ce ne sont pas les droplets qui doivent modifier les résultats... Je suis bien sûr en faveur d'un papier sur Colorado 2 mais j'attends aussi les résultats de Richard sur Colorado 3 avec beaucoup d'intérêt. J'ai tardé à vous répondre car j'ai du participer à des évènements familiaux, loin de Paris... Je suis de retour et pour gagner du temps , je vous écris en français...

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
30) Ludwik to Richard+Pierre+Gerard (5/29/06):
Pierre wrote in French (see below) to gain time. He is not questioning validity of Colorado2 results. He thinks that "certain conditions were not met," such as color of plasma and the current of ~ 1 A. Our current was ~2A. "Thus we must reconsider these experiments calmly." Pierre does not think that droplets are responsible for the COPs>1. [But that is not enough; we must have data showing that escaping water contained no potassium. That became the central issue when our results were discussed on the CMNS list. My impression was that Gerard was conducting chemical experiments. What did he measure and how can this help us to rule out the effect of droplets? Pierre is in favor of publishing the paper based on Colorado2 but he is eager to see Richard's Colorado3 results.

I am slightly confused by the "we must reconsider these experiments" phrase. What does it imply in practical terms? What should our time-table be? Should we wait or should we try to publish the Colorado2 results as soon as possible? Either way would be OK with me. Let me add some additional observations.

1) What makes the current lower ( ~1 A instead ~2 A at 350V)? I think it is the thickness of the plasma envelope and concentration of ions in it. The resistance goes up when the layer becomes and the current becomes lower. The current is also directly proportional to the concentration of ions. To get lower current we must decrease the ion density. But I do not know how to control the the layer thickness and degree of ionization. Perhaps the diameter of the cathode has something to do with this. The resistance of the layer goes up when the cathode diameter goes down, provided the layer thickness remains the same.

2) Our temperature (near the wall of the cell) was 89 C, that is because the two-litter container was too large for power we used. I am assuming that others used smaller vessels and their temperatures were uniformly equal to the boiling point. One might suspect that differences between currents are due to differences in temperature. But I do not think so. After all, what counts is the temperature of the plasma and not the temperature near the walls.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
31) Ludwik --> to Pierre + Richard (6/2/06):
1) For some reason I had no replies from you. Perhaps my message was lost; That is why I am pasting it below. . . . .

2) Today is Friday; when should I expect Fauvarque's report?

3) What is happening with our attempts to learn about concentrations of potassium in the escaped water and in the electrolyte? My impression was that this work is in progress -- Richard+Scott in the US and Gerard in France. Where do we stand on this?

4) Most of my draft is ready. I will send it to you very soon. My suggestion is to have Richard as the first author for our Colorado2 paper. The rest of us will be listed alphabetically. For the time being I am listing everyone. But I am not sure about George Luce (who works with Scott). And I do not know if Scott will be willing to be the author.

5) Should Fauvarque be the author of the paper or should he only be the author of the appendix? It does not make any difference to me. I will wait for the input from Pierre about this.

6) We also need a good title. Authors: Richard Slaughter, Pierre Paul Clauzon, Jean-Francois Fauvarque, Ludwik Kowalski, Gerard Jean-Michel Lalleve, Scott Little and George Luce

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
32) Pierre --> Ludwik+Richard+Gerard (6/5/06):
I thought that I answered you, Ludwik... We are so much submerged by a lot of e--mails (for instance Michel J.) that I do not know where I am with you. First, for us, we should go ahead with the Colorado 2 paper ! The draft of the chemical appendix you are waiting for is on the desk of Pr Fauvarque since three days at least! But he is a very busy man and last week, he was responsible for an chemical meeting of two days and so, inaccessible for almost all the week ! Today, it is the Monday of Pentecôte, usually a public holliday... and however,  we hope to meet with Gérard in the CNAM Lab.... We will try to definitively give a final answer to Marseille 1 and we hope to get a COP larger than the value of 1.12 we got recently... To my sense, this is only in order to have an other confirmation but this result is not necessary for Colorado 2.

I know that experts in electronics like Michel J. or Jacques Dufour are trying to be sure that the input energy is correctly measured. Jacques Dufour will inspect with his oscillometer our voltage and current signals... We will see, but I am confident with the measurements of Richard and his own expertise... Furthermore, I do not think that droplets or tiny droplets are a real problem. When you get the good operating conditions with only about 1 A and the good plasma color, you do not have droplets at all....Did I answer you , Ludwik?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
33) Pierre to Ludwik ((6/5/06):
The paper joined is just for your personnal information... it is still a draft we have to complete!  If you have questions about this note, please write only to me...Thanks in advance...

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
34) Ludwik --> Pierre (6/5/06):
Thanks for showing Fauvarque’s draft. One thing is clear; I do not understand it. And I do not want to be a coauthor of something I do not understand. Therefore, I suggest that Fauvarque is shown to be the author of the appendix in our paper. It often happens that we have to form opinion about something we are not qualified to understand. When this happens I rely of authoritative statements of experts. In this case we should post the appendix (when Fauvarque agrees) on the list and ask for constructive criticism. Do you agree that this would be the best strategy? Is Fauvarque the member of the CMNS list?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
35) Richard --> Ludwik + Pierre (6/5/06):
I have not been getting any private message from you.  I too think we should go ahead and publish the Colorado report.  I have been busy at work and have not done very much on the experiments.   The calorimetry  is proving to be more difficult than I anticipated.  I’ve had some good results with the ohmic heater showing the correct output power.  I just don’t get the consistency that is needed.    I’ll try once again to get the samples to Scott for checking the amount of K2CO3 in the evaporated water.

36) Ludwik --> Richard + Pierre (6/5/06):
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
You wrote: “I have not been getting any private message from you.” That means that at least one of my recent messages was lost. I was worrying that you might be sick or something else serious. You also wrote: “ I too think we should go ahead and publish the Colorado report.  I have been busy at work and have not done very much on the experiments.   The calorimetry  is proving to be more difficult than I anticipated.  I’ve had some good results with the ohmic heater showing the correct output power.  I just don’t get the consistency that is needed.” 

That is Colorado3. I do not think we should be focusing on that experiment. Except, perhaps at the end, to show preliminary results from "work in progress," if you want. My draft will be about Colorado2 only. You also wrote: “ I’ll try once again to get the samples to Scott for checking the amount of K2CO3 in the evaporated water.”

That is good. I will be waiting for Scott's results, and for the chemical appendix of Fauvarque. He must know how much W is consumed during our typical experiment. Did you give him this information, Pierre? Most of the lost tungsten is probably still pure metal, in the deposit, as in Mizuno experiments. But some of it is probably reacted with something to generate heat. I hope he will show that the total number joules from all possible chemical reactions is no more than 150J in 5 minutes. That would be less than 1% of our mean excess heat. We need the number of joules, and a description of data from which the number was obtained. Also some references, to satisfy chemists. It would take no more than two or three days to incorporate the missing information into my first draft. What each of you think about showing the draft to others on our CMNS list (after discussing it in private)? That would help us to end up with a better manuscript.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
37) Pierre --> Ludwik+Richard+Gerard (6/6/06):
Our W cathode is 15 cm long, 13.2 g heavy, and the cross sction diameter is 2.4 mm large. A typical run of 20 to 30 minutes burns about 1.5 cm of the cathode ( i.e. 1.32 g). A typical abnormal excess energy is about 60,000 joules ( 100 w during 10 minutes).

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
38) Ludwik --> Pierre + Richard + Gerard :(6/6/06):
My suggestion is to collect the residuals (for example, after evaporating water) and to find out how many grams of metallic tungsten it contains. According to Mizuno only a small fraction of lost tungsten participates in chemical reactions. Most of W simply melts away and remains metallic. That means that the amount of tungsten acting as a fuel for chemical heat is actually much less than 13 grams, from each cathode, perhaps less than 0.13 grams. That information will probably be very useful to Fauvarque. How else can he calculate the upper limit of chemical contribution to excess heat? Another suggestion, for the appendix only, is to express things in terms of joules and not in terms of watts. Most people know how to go from joules to eV.

Suppose we report 150 kJ in 25 minutes and that the actually reacted mass of tungsten was 0.13 grams. The 150,000 J translates into 9.4*10^23 eV. The 0.13 grams of tungsten translates into 4.24*10^20 atoms. In other words 9.4*10^23 / 4.24*10^20 = 2217 eV per atom. Such amount of energy per atom is three orders of magnitude higher than what is usually associated with chemistry. Note, however, that the result would be only 22 eV per atom if all the lost tungsten reacted as fuel. That is less dramatic than 2217 eV/atom.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
39) Richard --> Ludwik + Pierre + Richard :(6/6/06):
I have not been getting any private message from you.  I too think we should go ahead and publish the Colorado report.  I have been busy at work and have not done very much on the experiments.   The calorimetry  is proving to be more difficult than I anticipated.  I’ve had some good results with the ohmic heater showing the correct output power.  I just don’t get the consistency that is needed.    I’ll try once again to get the samples to Scott for checking the amount of K2CO3 in the evaporated water.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
40) Ludwik --> Richard (6/6/06):
Below is the message that you probably did not receive. . . .

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Ludwik (in a message posted at the CMNS list) 6/8/06:
. . . My draft describing Colorado2 results will be finished as soon as two missing components are received. I intend to expose it to criticism. And, personally, I would have no objection for not publishing it. Waiting for results from Paris3, Colorado3 and Marseilles1 experiments might indeed be preferable, as suggested by Michel.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
41) Ludwik --> Gerard + Richard (6/9/06):
(1) Is it OK to show, in our paper, that Fauvarque is the author of the appendix? Why is Fauvarque not participating in our discussions? I would very much like to read what he thinks about GDPE. He is the only electrochemist among us, as far as I know. His active participation would give me more confidence. I am hesitant to go ahead without his active participation. (2) Since I am totally ignorant in chemistry I would like this appendix to be discussed on our CMNS list? Can I post it today and ask for comments? (3) What should I write about this appendix in the main text of our paper? That text does not expect readers to know chemistry. Please write a paragraph for me. More questions later.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
42) Ludwik --> Pierre + Richard (6/10/06):
I asked Gerard some preliminary questions this morning. But he did not reply at once. Perhaps you can answer them. Here they are: . . . Meanwhile I had to relearn what "titration" is. This shows how little chemistry I know. I am waiting for your OK about posting the appendix on our CMNS list. I hope others will say that the appendix is well written and should be very convincing to a chemist.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
43) Ludwik --> Jean-Francois (Fauvarque) + Pierre + Richard (6/10/06):
Dear Dr. Fauvarque: I am writing a paper about results of our Colorado2 experiment. The appendix you wrote will be an extremely important part of it. Gerard sent me your appendix yesterday but I am in no position to evaluate it -- I am not a chemist. Our paper would not be worth publishing without convincing arguments that excess heat we measured cannot be due to possible chemical reactions.

Publishing a paper (associated with the controversial field of cold fusion) in a mainstream science paper will not be easy. That is why all possible objections should be anticipated. Our statements should be backed by references and by results of numerical calculations. My first impression was that your appendix is vulnerable to criticism of objective referees. That opinion is probably due to my limited familiarity with chemistry. In the main body of the paper I would like to say something like this: "A possibility of chemical origin of excess heat has been carefully examined by one of us (JFF), as shown in the appendix. The conclusion was that chemical fuels present in the cell could not possibly contribute more than 1% (???) to excess heat measured. We definitely need an upper limit, plus a strong appendix to numerically justify it. The fact that the appendix was produced by a well known electrochemist is essential, as far as i am concerned.

P.S. If the only possible fuel were tungsten, and if most of what was eroded during the electrolysis were found as pure metal at the bottom of the beaker, then the chemical contribution would be much less than 1 %. Here is my recent message to Pierre about this:

On Jun 6, 2006, at 9:09 AM, pierre.clauzon@wanadoo.fr wrote:
Ludwik, our W cathode is 15 cm long, 13.2 g heavy, and the cross sction diameter is 2.4 mm.
A typical run of 20 to 30 minutes burns about 1.5 cm of the cathode ( i.e. 1.32 g).
A typical abnormal excess energy is about 60,000 joules ( 100 w during 10 minutes).

Pierre, my suggestion is to collect the residuals and to find out how many grams of metallic tungsten it contains. According to Mizuno only a small fraction of lost tungsten participates in chemical reactions. Most of W simply melts away and remains metallic. That means that the amount of tungsten acting as a fuel for chemical heat is actually much less than 1.3 grams, in a typical run, perhaps less than 0.013 grams. That information will probably be very useful to Fauvarque. How else can he calculate the upper limit of chemical contribution to excess heat? Another suggestion, for the appendix only, is to express things in terms of joules and not in terms of watts. Most people know how to go from joules to eV.

Suppose we report 150 kJ in 25 minutes and that the actually reacted mass of tungsten was 0.013 grams.
The 150,000 J translates into 9.4*10^23 eV
The 0.013 grams of tungsten translates into 4.24*10^19 atoms.

In other words 9.4*10^23 / 4.24*10^19 = 22200 eV per atom. Such amount of energy per atom is three orders of magnitude higher than what is usually associated with chemistry. Note, however, that the result would be only 222 eV per atom if all the lost tungsten reacted as fuel. That is also impressive, but not as impressive than 22200 eV/atom.

We decided to discuss the draft of the paper among the authors first. Then, after making necessary changes, we want to post it on the restricted CMNS discussion list. There are many highly competent and friendly people on that list; they will probably help us to produce a difficult-to-reject paper. Are you among the CMNS subscribers? I will send you the draft as soon as it is ready. To avoid mistakes made by many, including Fleischmann, our paper should not claim that excess heat is nuclear; we have no evidence for this. Here is the last section of my draft:

Conclusion:
The overall conclusion is that the GDPE thermal-excess energy, discovered in (1), and confirmed by other investigators (9, 10 and 11), seems to be real. Generation of excess heat, at the 0.1 kW level, turned out to be reproducible. The origin of that energy is not clear to us. Can several cells be used in a battery generating excess heat at a much higher rate? Not having an accepted theory one must rely on empirical investigations. Systematic studies of effects of various parameters (sizes, compositions, distances, etc.), on the performance of Mizuno-type cells, seem to be warranted.

Absence of radioactive byproducts is a clear indication that familiar nuclear reactions cannot possibly be responsible for the amount of excess heat measured. A thermonuclear D-D reaction generating excess heat at the rate of 0.1 kW, for three hours (our total electrolysis time), would produce about 4*10^19 neutrons. That would be sufficient to produce radioactive isotopes in many objects near the cell. A Geiger counter we used showed that the level of radioactivity, near the cell, was not higher than natural background.

Selected Referneces:
1) Ohmori and Mizuno "Strong Excess Energy Evolution, New Element Production, and Electromagnetic Wave And/Or Neutron Emission in the Light Water Electrolysis with a Tungsten Cathode." presented at ICCF7,1997.
. . . .
9) Jean-Lois Naudin's web site - “CFR project, a High Temperature Plasma Electrolysis based on the Tadahiko Mizuno work from the Hokkaido University (Japan)” http://jlnlabs.imars.com/cfr/

10) D. Cirillo, A. Dattilo, V. Iorio, “Transmutation of metal to low energy in confined plasma in the water (electrochemical plasma cell),” ,”. in Eleventh International Conference on Condensed Matter Nuclear Science. 2004. Marseille, France. The report can be downloaded from the library at <http://www.lenr-canr.org>

11) Jean-Francois Fauvarque, Pierre Paul Clauzon and Gerard Jean Michelle Lalleve. “Abnormal excess heat observed during Mizuno-type experiments;” 2005. The report can be downloaded from the library at <http://www.lenr-canr.org>
. . . .
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
44) Ludwik --> Pierre + Richard (6/10/06):
That is OK with me. I just sent an email message to JFF, plus a copy to you and to Richard. That would be a good way to start discussing our paper. Is he a CMNS list subscriber?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
45) Pierre -->Ludwik + Richard (6/11/06):
Up to now, JF Fauvarque was not a member of the CMNS list. As I told you, he is a very busy man and I just make for him copies of the most meaningful papers delivered in the CMNS list. We are working together with Gérard, under the leadership of Pr Fauvarque. When I am in the CNAM labs, I use the computer of Gérard for the mails, for example. But, before sending you any mail, I usually ask Gérard for his approval..

46) Richrd --> Ludwik + Pierre (6/11/06):
I think it is a good idea to post Fauvarque’s appendix.  I’m sure he knows we intend to publish the appendix with our paper.  This way we/he can get some additional feedback.  I would add some comments to the paper saying it is not for distribution outside the CMNS group.   I suspect Fauvarque is just very busy, still he did work on the paper.  I think we should send a personal thank you note of appreciation.  Perhaps in this note we could ask how he would like to be credited with the appendix

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
47) Ludwik --> Jean-Francois + Pierre + Richard (6/11/06):

1) Richard, I agree that we should thank Pr. Fauvarque for willing to help us. This morning I looked at the list of his publications, fetched by Google, and I was very impressed. Our claim that the measured excess energy cannot be due to chemical reactions would not be taken seriously by reviewers and editors without a reputable electrochemist. We are lucky to have him. Perhaps he will find time to communicate with us by email during this final stage -- discussing the draft and strategy for publishing (where, what to say, what not to say, etc.). As you know, publishing a paper on a controversial topic is much more difficult than a paper making a contribution to what has already been published.

2) You also received the draft (as an attached file sent by Gerard) of what he wrote. I am reading that appendix again and formulating comments. But before sharing my comments and questions with coauthors I will wait for the reply from JFF. Pierre prefers to discuss the appendix between the coauthors before posting it on the CMNS list. That what we should do for the next two or three days. If nobody comes with suggestions to modify what was written by Jean Francois then I will post the appendix on the CMNS list.

3) Did you send samples of our electrolyte, and of condensed liquid, to Scott? Using his X-ray fluorescence device he promised to determine the percentage of tiny invisible droplets in the escaping liquid. I will be waiting for his input about the droplets. It is also essential. I am no longer counting on Scott’s attempt to replicate our results. My draft cannot be finished unless I know that the percentage of droplets was very small, for example one percent or so. We must be absolutely certain that it not close to 24% -- that would show that the COP=1.24 was an illusion. Was the percentage of droplets measured by Gerard, in Paris2 experiments (immediately after Colorado2)? What was the result and how accurate was it? I would like to say something like --> 2% +/- 0.5%. Anything above ~5% would push us away from the comfortable level of high confidence. (that the COP > 1.0).

4) My notebooks are in boxes that were put into self-storage. That is will probably have some questions to you and to Pierre. For now I want to know the capacity of the 500 V electrolytic capacitor we used at the output of your power supply?

5) I also need information about a test we did not perform. Can one of you perform the following test:

a) Start the experiment as usual (by increasing the voltage and performing the first measurement of COP, the one that should always be rejected).

b) Then turn the voltage off suddenly, disconnect the cables, and measure the voltage between the electrodes. We believe that the voltage will be zero. But this should be checked. It will be useful to describe this test in our paper. If the voltage is nearly zero then we will be able to rule out a possibility of a battery-like effect (storing energy during the preliminary steps and releasing it when the COP is measured). That kind of objection can be made by a reviewer. Our answer, even without the suggested test, would be that stored energy would have to be several times larger than 30 kJ (100 W times 300 seconds case after case.) But reporting that we performed such test will show that such possibility was not ignored.

6) In fact, I would not be surprised to see a small remaining potential between the electrodes. If you observe this then treat the cell as if it were a capacitor and calculate the energy stored. Use a two-trace oscilloscope to see the v(t) and i(t) exponential pulses (slow discharging). That what I would do to find the stored energy. I am 99.9% sure it will be much much less than 30 kJ. Perhaps we will be able to say that stored energy turned out to be 0.01 kJ, or something like this.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
48) Ludwik -->Gerard + Jean + Pierre + Richard (6/12/06):
On Jun 9, 2006, at 8:27 AM, gerard.lalleve@laposte.net wrote:

About now the droplets case, we confirm to you that we
have not at all loss of potassium when the experiment
is operated in a right way. In Marseille also, we used
chemical paper to see if the steam was polluted by
very tiny droplets. The result was again negative.

1) What we need are numerical results. Something like "contribution of tiny droplets to the mass of the escaping wet steam was found to be 2.5% (???)." Or something like "less than 1%," if the titration method is not sensitive enough to produce one number. That would appear in the main text; the Appendix-2 is for the methodology used to establish the result.

2) Likewise, the main text would say something like "contribution of known chemical reactions, to the excess heat measured, could not possibly exceed 0.01%" (???). That statement would be justified in the Appendix-3.

3) Note that the (???) indicates wishful thinking; I have no idea what the actual numbers will be. I already wrote Appendix-1; it is about measuring electric energy. In my opinion three short appendices are better than a single long one.

4) Today I will read the long appendix written by JFF again. My general impression was that it does not provide percentages we need to justify the COP=1.2 claim. I would like to know what each of you think about the long appendix. Do you agree that turning it into two shorter appendices is desirable? If so then we should see and discuss them, as soon as possible.

5) Where do we stand on the X-rays fluorescence test of samples, promised by Scott? Where do we stand on the estimation of the fraction of tungsten, lost by the cathod, that remains metallic? According to Mizuno that fraction (found in the deposited residuals) is very large. What remains metallic did not participate in chemical reactions and did not contribute to chemical excess heat. That should probably be taken under consideration in the Appendix-3.
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

49) Richard --> Ludwik (6/12/06):
I've been very busy at work and just have not had the time and energy to work on CF. I'll try very diligently to get the sample for Scott this week.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
50) Ludwik --> Richard (6/12/06):
1) I understand your dilemma perfectly. I had a similar problem before retirement. Work connected with teaching had to be my first priority, no matter how much I wanted to do something else.

2) I am puzzled by silence from Fauvarque. The appendix he wrote (Pierre said he gave it to them about two weeks ago) was not very helpful to me. I hope it will be changed. Otherwise our paper will be in trouble. Any referee will see that it is not qualitative, as far as what we want to know is concerned. Most numbers refer to intermediate results from titration.

3) Scott would most likely provide reliable numbers for the Appendix-2 (percentage of droplets in wet steam). But rebuttal of the idea of chemical origin of excess heat (Appendix-3) must be produced by an electrochemist familiar with exotic hot-plasma ebvironement.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
51) Ludwik --> Jean + Richard + Gerard + Pierre (6/13/06):
JFF wrote (in our appendix):
1- Acid-Base titration
5 mL of the initial solution is titrated by HCl 1.0 M. The two
successive neutralisations of CO3 2 - at pH 8.5 and of
HCO3- at pH 4.5 are clearly visible. As expected, no other
neutralisation is detected. The titration gives 1.94 M of
K+ ( counter ion), the carbonate being initially under the
form CO3 2 -. . . .

The molarity of 1.94 seems to be impossible. Our electrolyte was 0.2 M. Assuming nearly 100% ionization of K2CO3 in water I would expect the molarity of K+ to be 0.4, or a little less, but not 1.94. Where am I wrong?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
52) Jean --> Ludwik (6/13/06):
I was wrong, it was a writing lapsus, the K+ molarity is 0.388, as expected.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
53) Ludwik --> Jean (6/13/06):
Thanks for the quick reply. Suppose "wet steam" is condensed when a measurement of COP is in progress. If the K+ molarity of condensed liquid turns out to be close to 0.008 then we would be able to say that the contribution of tiny droplets of electrolyte, to the lost mass, is about 2%. The rest is pure vapor. That would be a strong argument that the COP determined in that experiment is not an illusion due to tiny droplets. And our Colorado2 results (mean COP=1.24, st.dev. =0.13) will be taken more seriousely. Do you agree? If so then please ask for the appropriate measurement (or several measurements to determine the precision) to be made.
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

54) Ludwik --> Jean + Gerard + Richard + Pierre (6/13/06):
Dear RS, , PC, JFF, LK and GL:
1) These are initials of authors in my draft, in the order shown. For obvious reason, Richard is the main author, the rest of us are listed in alphabetic order, according to last names. Is this OK with everybody. I still do not know what the status of Scott Little will be. It will depend on his promised contributions.

2) I wander if the message sent nearly 40 hours ago was received by everybody. My expectation was that we will start discussing the appendix, as soon as possible. That was the purpose of comments shown below. What does each of you think about my comments?

3) This morning Jean-Francois asked me to replace the 1.94 M by the 0.388 M, in two or three places. I did this in my file and you should do it in yours. In that way we will be discussing identical texts.
Ludwik

On Jun 12, 2006, at 9:43 AM, Ludwik Kowalski wrote:

1) What we need are numerical results. Something like "contribution of tiny droplets to the mass of the escaping wet steam was found to be 2.5% (???)." Or something like "less than 1%," if the titration method is not sensitive enough to produce a measured value. That would appear in the main text; the Appendix-2 is for the methodology used to establish the result.

2) Likewise, the main text would say something like "contribution of known chemical reactions, to the excess heat measured, could not possibly exceed 0.01%" (???). That statement would be justified in the Appendix-3.

3) Note that the (???) indicates wishful thinking; I have no idea what the actual numbers will be. I already wrote Appendix-1; it is about measuring electric energy. In my opinion three short appendices are better than a single long one.

4) Today I will read the long appendix written by JFF again. My general impression was that it does not provide percentages we need to justify the COP=1.2 claim. I would like to know what each of you thinks about the long appendix. Do you agree that turning it into two shorter appendices is desirable? If so then we should see and discuss them, as soon as possible.

5) Where do we stand on the X-rays fluorescence test of samples, promised by Scott? Where do we stand on the estimation of the fraction of tungsten, lost by the cathod, that remains metallic? According to Mizuno that fraction (found in the deposited residuals) is very large. What remains metallic did not participate in chemical reactions and did not contribute to chemical excess heat. That should probably be taken under consideration in the Appendix-3.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
55) Richard --> Ludwik (6/14/06):
I’m glad you are on top of this.  It’s difficult for me to follow. I’ve got four samples of condensed liquid tonight.  The test when easily but I didn’t get any real high COP.  The highest was 1.20.  I’ll send these to Scott tomorrow.I have three samples of the condensed liquid and one sample of the electrolyte from the beaker.  That will help keep everyone honest.    I’ll try for some higher COP this week.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
56) Ludwik --. Richard (6/14/06):
1) On 4/29/06 Scott wrote to us: "Yes, I could probably make a reasonably accurate analysis for elemental K content that way. I need about 15 mL of solution to fill the sample cup properly. Actually if that is a great difficulty, I could probably dilute the samples 10:1 and still attain satisfactory accuracy....so I could use as little as 1.5 mL of sample."

2) Keep this in mind. My suggestion is to split each of your two sample into two and to send only one half of each to Scott. The others may become very valuable if a serious discrepancy develops between the two results (Scott versus Jean-Francois). In that case we might send the samples to a commercial lab. Many medical and environmental labs are well equipped for the analysis of potassium -- they do this routinely.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
57) Richard --> Ludwik + Pierre + Scott (6/15/06):
Scott, I sent you four samples today.   Each sample is 10 to 15 grams I hope this is enough.   Samples 1 to 3 are of 3 runs that had COP’s > 1.1.  I was not able to get high COP’s but I will try again.   I sent you all of the sample I took.  So you might want to keep part of the sample after you dilute it for further study.  The forth sample is just some of the electrolyte after running the tests. Thanks for the help

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
58) Pierre --> Ludwik + Richard (6/15/06):
Bonjour Ludwik, Je t’écris du laboratoire de Jacques Dufour, beaucoup mieux équipé que notre ancien laboratoire ( enregistrements des mesures, alimentations diverses, etc..). Avec l’accord de nos chefs (Prs Fauvarque et Foos), nous avons lancé l’opération Paris 3 ensemble.
 
Vous êtes entrés en relations, Fauvarque et toi, pour cette annexe chimique. Après en avoir discuté avec lui, nous pouvons dire qu’à la précision des mesures (0.5 à 1%) nous retrouvons dans l’électrolyte tout le potassium initialement présent. C’est aussi une réponse à la question des « tiny  droplets ». Une fois le régime de fonctionnement bien établi, j’estime personnellement qu’il n’y a pas d’entraînement significatif de gouttelettes.(<<1%).
 
Chaque essai de 20 à 30 minutes fait disparaître environ 1.3 g de Tungstène. Une bonne part tombe au fond du bécher (beaker) , peut-être 50%, sous forme de débris métalliques. Nous n’avons pas conservé ces fonds de bécher, mais, là encore, les résultats d’échauffement en terme de réactions chimiques liées au Tungstène sont négligeables comparés aux échauffements anormaux constatés.
 
Tu écris, Ludwik : «  ne laissons pas le mieux être l’ennemi du bien ». J’ajoute dans le même esprit : «  ne coupons pas les cheveux en quatre ! ».  Je crois qu’il est temps de publier nos résultats de Colorado 2. Nous espérons pouvoir remplir tous les blancs que tu laisseras dans ton draft. J’espère que cette expérience simple donnera l’idée à des experts de la fusion froide de confirmer nos dires et d’apporter leurs propres améliorations. PS: I hope, Richard, you read the french... Sorry, if not !

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
59) Ludwik --> Pierre and Richard (6/15/06):
1) In my opinion, as I wrote before, our paper is not worth submitting to a mainstream journal. A good appendix, signed by a reputable electrochemist is essential. That appendix must convince other experts that excess heat we measured cannot be attributed to well known chemical reactions. His words will be taken seriously, my words or your words will not.

2) Pierre, why is Fauvarque not one of those to whom your message was sent? Are there any problems with his willingness to be fully responsible for the thermo-chemical Appendix 3? I wander why he did not comment on what I wrote about his appendix. Are we going to have separate appendices for energy and for tiny droplets? Should I interpret this as an OK for posting his appendix on the CMNS list (after replacing 1.94 m by 0.388 M)? My expectation was, and still is, that each of will write about his appendix, before we show it on the list.

3) Richard, Pierre tells us that Paris3 experiments will be conducted by an enlarged team (people from Fauvarque lab and from Dufour lab at CNAM. Dufour lab is much better equipped for that kind of work).

4) Pierre also gives us a conclusion about the invisible droplets -- their contribution to the lost liquid was found to be between 0.5 and 1 %. That is a great news; I will include that sentence into the main text. The methodology used to establish this result must be described. We can do this in his long appendix or in the Appendix 2, as I suggested. Two or three paragraphs, that any chemist will understand, would probably be sufficient. Fauvarque's appendix does not mention the 1% conclusion; perhaps it is hidden somewhere between the lines.

5) Let us hope that Scott's data will confirm the very low percentage of droplets. Being familiar with his instrument I will add a paragraph or two about the second method to the Appendix 2 (or two Fauvarque's appendix) .

6) No, I am not trying to split a hair, Pierre. We are making an extraordinary claim. Our paper must be nearly nearly perfect in order to have a chance of being published. Please answer as soon as possible; it is frustrating to wait and not being able to finish the draft. I reached the point at which I can do nothing more by myself. I must be satisfied with contributions made by others. And I do not want to share an unfinished draft.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
60) Ludwik --> Pierre and Richard (6/15/06):
In reading Pierre's message more carefully I see that he did not say that droplets were 0.5-1%. He said that the contribution of droplets was much less that 1 %. That is better. The 0.5-1% was their maximum possible error in the statement that the "molarity of the remaining electrolyte did not change." That is a much weaker statement than comparing the 0.2 M of the electrolyte with the molarity of the condensed liquid. If the "much less than 1%" is true that the result should be "much less than 0.002 M." A good referee would certainly ask for a better method of determination of the percentage of droplets. He would also ask if the "much less than 1%" is consistent with the "0.5-1%" accuracy. We should anticipate such questions and answer them in the appendix. Do you agree with this? Does Fauvarque agree with this?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
61) Ludwik --> Pierre and Richard (6/15/06):
1) Pierre; I am terribly confused for the absence of comments about what I wrote to Fauvarque and to you (in two messages). Should I start discussing his appendix on the CMNS list, as we were going to do after discussing it among ourselves?

2) Meanwhile I prepared a message (see below) which will lead to such discussion. I will post it as soon as you and Richard OK it.

MY MESSAGE WILL SAY THIS:

On Jun 14, 2006, at 11:07 PM, Ludwik Kowalski wrote:

List of successful replications of Mizino-type excess heat:

1) J.L. Naudin see his website --> http://jlnlabs.imars.com/cfr/
2) V. D. Cirillo et al. see their ICCF11 report
3) J.F. Fauvarque et al. see their ICCF12 report
4) R. Slaughter et al. Colorado2 experiment (we are working on the report)
5) J.F. Fauvarque et al. -- Paris2 (preliminary results after Colorado2)
6) University Labs in Japan

Experiments in preparation or in progress:
7) R.Slaugther -- Colorado3 (preliminary observations of excess heat)
8) J.P. Biberian -- Marselles1
9) J.F. Fauvarque et al. + another lab at CNAM -- Paris3
10) S. Little -- Texas2

What about your lab? I am not going to list names; many researchers on this list are highly qualified to replicate Mizuno-type experiments. And some are they are likely to have what is needed. Working on nearly identical experiments, and sharing results from our non-patent-motivated work, we are likely to promote the CMNS field faster than working on separate experiments and keeping things secret. Naturally, one must be aware of possible dangers, as described earlier on this thread. . . .

I am slowly realizing that my suggestion would be acceptable only if we had convincing (indisputable ) arguments that the reported excess heat could not be explained by well known chemical reactions. We need a straight forward thermochemical analysis of reactions that are expected to take place in GDPE cells. Potential reactants present are known and chemists, I expected, can calculate excess heats associated with each of them. It would be a matter of adding individual excess heats (some positive others negative, according to existing enthalpy tables and according to amounts of consumed reactants). I assumed that such analysis would show that the net excess heat from known chemical reactions is at least two orders of magnitude smaller than what was actually measured. But I have not seen such arguments, so far.

What I did see, in a Mizuno's paper (see below), was a discussion based on one potential fuel, tungsten. Is it reasonable to assume that the role of other potential fuels can be neglected, in comparison with tungsten? Mizuno seems to be making such assumption. But I am not a chemist and validity of that assumption is not obvious to me. I hope that reputable chemists will provide indisputable evidence that chemical excess heat is too small in comparison with what comes from a Mizuno-type cell containing the K2CO2 electrolyte. By the way, I do not think that reliance on authority of experts is forbidden by the so-called "scientific method." Where would we be if we had to validate everything by ourselves?

Here is how Mizuno et al. addressed the issue (in T. Mizuno, T. Ohmori, K. Azumi, T. Akimoto and A. Takahashi; “Confirmation of heat generation and anomalous element caused by plasma electrolysis in the liquid;” Conference Proceedings Vol. 70, ìICCF8î Società Italiana Di Fisica, Bologna, 2000, p. 75)

1) It is well known that 380 kJ of heat is released when one mole (183.85 grams) of W is consumed to produce H2WO 4.
2) In reality only 0.1 grams of tungsten was lost during a test in which excess heat was measured. That, could produce 0.207 kJ of heat. The amount of excess heat measured, 54.4 kJ, was found to be considerably larger. Thus no more that 0.4% of excess heat measured could be due to chemical consumption of tungsten.
3) This 0.4% fraction, they wrote, is actually an exaggeration because the 0.1 grams of tungsten lost from the cathode was found at the bottom of the cell in the form of pure metallic powder. In other words, tungsten did not react with oxygen, it was simply removed “by hydrogen corrosion as well as heat damage.”
4) Another chemical reaction taking place in the cell, during plasma electrolysis, is “the decomposition of the carbonate in a water solution.” That reaction is said to be endothermic; it removes 274 kJ of heat per mole of K2CO3. Is it true that tungsten is the only well known exothermic reaction that might contribute to excess heat measured in Mizuno-type cells?

By the way, loosing only 0.1 grams of tungsten per experiment is not consistent with what I remember from Colorado2. It is also not consistent with more qualitative results from Paris2 experiments. Referring to these experiments in a private message, Pierre Clauzon wrote: "In each sequence of tests lasting 20 to 30 minutes we loose about 1.3 grams of tungsten. A sizable fraction of it, perhaps 50%, is found as residuals, deposited at the bottom of the beaker. . . . " Note that 1.3 grams is considerably larger than 0.1 grams reported by Mizuno et al. That seems to indicate that experiments in Colorado2 and Paris2 were substantially different from those described by Mizuno. Perhaps the shapes of electrodes have something to do with the difference.

In any case, what conclusion can be drawn from the information provided by Pierre? Following Mizuno et al., I will assume that tungsten is the dominant chemical fuel during the glow discharge electrolysis. I will also assume that excess heat is generated at the average rate of 50 W. That translates into 90,000 joules of excess energy (or . 5.62*10^24 eV) during 30 minutes. The 1.3 grams of tungsten, on the other hand, translates into 2.54*10^22 atoms. If 100% of these atoms reacted chemically, to generate 90 kJ of heat, then the rate is 121 eV per atom. that is already two orders of magnitude higher than what is usually associated with chemical reactions. If Mizuno is correct that tungsten in the residuals is pure metal, and if Pierre is correct that residuals contain 50% of the lost tungsten, then the number reacted atoms becomes two times larger and rate becomes 2*121=242 eV/atom.

Another way of using Pierre's data is to follow Mizuno's approach. The mass of the lost tungsten was 1.3 grams. This is 0.0071 moles. The expected chemical heat, due to reacting tungsten, is 0.0071*380 = 2.69 kJ. That is only 3% of the actually measured excess heat. The result would be 1.5% if only 50% of tungsten reacted, as estimated by Pierre. But what if the actual measurements showed that 90% of the lost tungsten is metallic, at the bottom of the beaker? That the result would be a small fraction of one percent. I hope the amount of metallic tungsten, in the residuals could be determined with the accuracy of 5% or so. Honest referees would appreciate our effort to do the best we can to show that excess heat we measured is indeed abnormal. Publishing a paper on excess heat makes sense only if the abnormality can be established. Abnormality is the central point of our paper and it has to be established as conclusively as possible. We are in no position to argue about nuclear or hydrino's origin of excess heat. That would call for much more information than we were able to gather. Comments and help would be highly appreciated.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
62) Pierre to Ludwik (6/17/06):
I did receive your last e-mails about our absence of comments, Fauvarque and me. I will do my best this w.e. and I will try to get the same from Fauvarque next week. Fauvarque was at the end of this week in Tunisia, for university exams and contracts . He is a quite busy man and please, accept our apology !

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
63) Ludwik --> Pierre + Richard (6/17/06):
1) Thanks for the quick reply, Pierre. OK, I will wait. Did Fauvarque talk with you about my comments?
2) Does he agree with the suggestion to have two shorter appendices, one about chemical excess heat and one about tiny droplets?
3) Did he say that he will calculate for us excess heats of chemical reactions (how many kJ each) on the basis of known amounts of reactants, and on the basis of enthalpy tables used by chemists?
4) I will not start discussing his appendix on the CMNS list. But what do you think about the massage I sent you yesterday. Fauvarque is not mentioned in it; I only quote what you wrote about tungsten. Can I post that message today? It might generate helpful comments.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
64) Richard --> Scott (6/17/06):
Ludwik, said you can do X-ray fluorescence on the samples to find out if there is any K2C03 or just K in the samples.  We are trying to rule out the possibility of droplets or mist being part of the reason for the over unity COP.  Sample 4 is just the electrolyte so it should have 0.2 M K2C03. 

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
65) Ludwik --> Richard (6/17/06):
Just relative heights of potassium peaks (in condensed liquid versus original electrolyte) would probably be enough, provided the geometries are identical. Also an error bar on the ratio. Scott might suggested something else; he knows exactly about the context. (P.S. Actually, this would be true only if the ratio was 0.1 or more. For a much lower ratio, which we hope for, Scott will probably dilute to original electrolyte, for example, by a factor of ten, in order to minimize the error.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
66) Scott --> Richard + Ludwik (6/17/06):
It may be next Friday before I get a chance to go see the XRF system.  It's not at my lab anymore....but it's at the place my daughter works. Meanwhile, I want you to calculate what K concentration in these condensate samples would it take to negate, say, a 1.10 COP.   That value will give me something to "shoot for".   It will tell me whether this is an easy analysis where I have plenty of sensitivity, a borderline analysis where I have to optimize the excitation conditions and use long count times to get the necessary accuracy, or a hopeless analysis where I cannot even detect an important (to this experiment) amount of K.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
67) Ludwik --> Scott + Richard +Jean + Gerard (6/18/06):
Let me try to answer your question. In the draft of his appendix Fauvarque reported the molarity of K+ ions as 0.388 -- determined by the method of the HCL titration. That is only slightly less than the ideal 0.4 M for our totally ionized K2CO3. To prepare the 0.2 M solution of that salt Richard probably used 26 grams per liter or 26 mg/ml. His condensed liquid samples were from experiments that yielded the COP of 1.1.

If such COP were an illusion due to tiny droplets then the concentration of K+ in samples of condensed liquid would be close to 2.6 mg/ml. To negate the illusion, the condensation would have to be below 0.3, hopefully 0.1 or less. Is the 0.1 mg/ml too low for your instrument? My expectation is that Fauvarque will improve the appendix, and that it will be posted on the CMNS list, probably in several days, after we have a chance to discuss it privately.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
68) Scott --> Ludwik (6/17/06):
0.1 mg/mL is about 0.1 mg/gm = 0.0001 gm/gm = 100 ppm. No, that's not too low. I should be able to get a detection limit down below that somewhere, maybe 20 ppm or so. Thanks. I will need to make a more dilute known standard. I'll probably make up a 2500 ppm standard. No problem. I'll get back to you guys later. BTW, XRF measures total K in the sample, regardless of ionization state, etc.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
69) Ludwik --> Richard, Pierre, Gerard, Jean (6/17/06):
Forwarding the above message from Scott.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
70) Richard --> Ludwik (6/20/06):
This is the first real data from Colorado #3. This is about one hour of continuous data.  The voltage and current are normalized using 300volts and 1.333 amps.  This just so all the info is on a single plot.  The sharp increases in current are where I had to push the cathode down because it had burned away.  At about data point 1100 I increased the voltage from 300 volts to 320 because the COP was low.  I’m not sure what to make of point 1280 where the current suddenly drops and the COP jumps 1.35.  You can ponder the graph like I’m doing and let me know what you think.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
71) Ludwik --> Richard (6/20/06):
The fact that you can run for one hour is very impressive. But the main COP is about 1.05. It is too early to comment on this, or to worry about this. I am trying to finish our Colorado2 paper. My suggestion is to include Scott in the list of authors; his XRF result will be a significant contribution. If he refuses then we will simply thank him for help, at the end of the paper. I would prefer him to be a coauthor. But neither you nor Pierre responded to this. Does it mean you have some reservations?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
72) Richard --> Ludwik (6/20/06):
It’s fine to include Scott as a coauthor.  Just give him the chance to decline.  I suspect he will be reluctant to be a co-author without having help take the data.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
73) Ludwik --> Richard (a comment on a picture he sent me, 6/20/06):
COPs are trying to get up when both powers are going down. Is this significant? Perhaps using lower powers is worth trying. What is the nominal diameter of your cathode?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
74) Richard --> Ludwik (6/20/06):
Interesting question! I'm using two 0.040 dia wires twisted together with 1 twist per inch. I had order some wire so I could have longer lengths this was what they sent as a sample. Using two wires got me closer to the original diameter but gave a lot more surface area.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
75) Ludwik to Richard (6/20/06):
I need some information for our Colorado2 paper.

1) About capacitors in parallel with the cell.
a) Was it one or more? How many?
b) Was it 500 V and 470 microF each? If not then how many for each?

2) About the little ac kWh-meter (the one we used for the ohmic heater, when it was also plugged in).
a) What model was it? What manufacturer? Was it from Radio shack?
b) what was the approximate cost?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
76) Richard --> Ludwik (6/20/21):
There are two 3100uF 450Volt spraque capacitors
The little plug in watt meter is a Kill A Watt model P4400 designed for seeing how much power household appliance use.
The meter is about $30.00
http://www.spytown.com/kilwatwatkil.html

The other AC watt meter is a DMMetering DRM75A
http://www.dmmetering.com/product_groups/spd.htm#

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
77) Ludwik --> Pierre + Richard (6/21/06):
Ten days ago I made a suggestion which was forgotten in our subsequent correspondence. I asked about information on a test that was not performed. "Can one of you perform the following test:

a) Start the experiment as usual (by increasing the voltage and performing the first measurement of COP, the one that should always be rejected).

b) Then turn the voltage off suddenly, disconnect the cables, and measure the voltage between the electrodes. We believe that the voltage will be zero. But this should be checked. It will be useful to describe this test in our paper. If the voltage is nearly zero then we will be able to rule out a possibility of a battery-like effect (storing energy during the preliminary steps and releasing it when the COP is measured). That kind of objection can be made by a reviewer. Our answer, even without the suggested test, would be that stored energy would have to be several times larger than 30 kJ (100 W times 300 seconds case after case.) But reporting that we performed the test will show that such possibility was not ignored.

In fact, I would not be surprised to see a small remaining potential between the electrodes. If you observe this then treat the cell as if it were a capacitor and calculate the energy stored. Use a two-trace oscilloscope to see the v(t) and i(t) exponential pulses (slow discharging). That what I would do to find the stored energy. I am 99.9% sure it will be much much less than 30 kJ. Perhaps we will be able to say that stored energy turned out to be 0.01 kJ, or something like this."

What do you think about this suggestion? In my opinion our Colorado2 paper would be more difficult to reject if we say that an attempt to discover a battery-like effect was made but results were negative," or something like this.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
78) Richard --> Ludwik (6/21/06):
Will do.. I’m sure the voltage goes to zero in a very short time but I’ll get some real data.  I’ve done this all the time when I shut down the supply and the voltage always drops immediately. 

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
79) Ludwik --> Richard + Pierre (6/21/06):
On Jun 21, 2006, at 8:20 AM, Richard Slaughter wrote:

1) That is good, Richard. Try to get the energy actually delivered through a battery-like effect, even if it is only 5J, or much less. Numbers are always more convincing than qualitative statements, like "much smaller than ..." If the number is smaller than what can be measured then give the upper possible limit.

2) I do not know why Fauvarque is not replying. Are you giving him all messages as soon as they arrive, Pierre? How is he reacting to my questions and comments? We need to hear from him, the more the better.

3) Other electrochemists on the CMNS list, like McKubre, Oriani, Szpak, Mizuno, Biberian, Zhilov, etc. are also not responding. Why is it so? The same is true for material scientists, like Storms and Dash. I am not comfortable in such unexpected situation.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
80) Ludwik --> Pierre + Richard (6/22/06):
1) Pierre, did you receive the message I sent you yesterday? In that message I asked why Fauvarque is not replying to my comments and questions about the appendix. Now you are not replying. Did my questions create some problems? Were they really so silly that JFF decided to ignore them?

2) According to the appendix only 1 gram, out of 18 grams lost in a typical experiment, was in the form of droplets. That is 5.6%. Can I say, in the main text that due to droplets the mass lost m must be reduced by 5.6 +/- 1 percent? That will lead to the about 5% correction of COP (1.17 instead of 1.24). The 1% error is based on this sentence from the appendix: "The precision of the titration can be estimated to be roughly 1%" Am I interpreting this sentence correctly? If not then what should the bar of errors be for the 5.6% ?

Why am not certain about the 1%? Because a pH value is x, from the 10^x. Was the 1% an error for x or was it an error for pH? A one percent change of x would produce a much larger change in pH. That is why I am not sure that the 5.6 +/- 1 is a correct interpreation of JFF’s sentence. Not being a chemist I need an error on the percentage of droplets, not an error on some intermediate result from which the percentage of droplets was calculated.

3) Scott will probably measure the content of K (in condensed liquid) tomorrow. Will his result confirm the 5.6%? If so then the issue of tiny droplets being responsible for the big part of our excess heat will be put to rest.

4) The only remaining issue will be to convincingly show that corrections to the COP due to chemical heat (positive and negative contributions combined) are also very small, probably much less that 1%. Where do we stand on this issue?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
81) Pierre --> Ludwik (6/23/06)
Bonjour Ludwik, Je retransmets à JFF tous les e-mails le concernant... Mais c'est une véritable étoile filante en ce moment ! Il est en ce moment en Russie. Lundi, j'essaierai de le voir le matin, car dès l'après-midi, il repart pour Grenoble pour plusieurs jours ! Nous sommes en pleine période d'examens en ce moment.

Je t'envoie un peu plus tard mes réponses plus détaillées à tes questions. Mais dès à présent, je peux te dire que lors d'une expérience avec Dufour hier Jeudi, on a mis un papier chimique dans la vapeur pour vérifier s'il y avait des tiny droplets d'électrolyte. Je persiste à te dire que c'est très très faible et que les valeurs que tu tires du papier de JFF sont des bornes très supérieures! Pour ma part, ce problème est inexistant !

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
82) Ludwik --> Richard +Pierre (6/23/06):
1) Richard, Pierre forwards all relevant messages to JFF, who is now in Russia. He will try to see him on Monday morning. On Monday afternoon JFF will leave for a several-days-long visit to Grenoble. It is an exams period in France. Pierre will comment on what I wrote a little later. He wrote: "But I can tell you one thing immediately. The experiment with Dufour [Paris3] started yesterday. We placed a chemical paper into the escaping steam and did not observe any droplets of the electrolyte." Pierre is strongly convinced that the number I extracted from the JFF's appendix are highly exagerated. As far as he is concerned, tiny droplets is not something that we should worry about.

2) But a referee will have right to ask for numbers. What I saw in the appendix was 0.005 M translated into 14 grams of electrolyte. Yes, it was given as an upper limit. But that did not tell me that I should replace 14 grams by 0.14 or 0.014 grams. The upper limit usually refers to what was possible to measure with a given method. Telling the referee that droplets represented much much less than 14 grams of the liquid will not be sufficient. The answers will be "how do you know that it should be 0.14 or 0.014?" or "how do you know it was not 13 grams?"

3) Pierre, what is the realistic limit of detection of tiny droplets when chemical paper is used, as you did yesterday?

4) My suggestion is to design an experiment in which the percentage of the electrolyte in the escaping liquid, for example 1%, is known in advance but not to the person performing the measurements. Measuring the same thing several times (to get the mean value and the standard deviation) would be one way to convince the referee.

5) No, I have no idea how to design an experiment in which the percentage of tiny droplets escaping from the beaker can be controlled. Some kind of a sprayer of pure electrolyte, injecting droplets into steam from pure water, would have to be invented.

6) Yes, I know that most often referees do not ask for numerical data about everything. But we are making an extraordinary claim in a paper that should be "difficult to reject." That is why extraordinary efforts are necessary. Do you agree, Pierre?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
83) Ludwik --> Scott + Richard + Pierre (6/23/06):
1) Scott, I hope other things did not delay you. When you do perform the XFR experiment please try to measure the concentration several times. This will give us not only the mean value but the standard deviation (random error) as well. Also try to estimate the systematic error. We are making an extraordinary claim and referees are expected to ask a lot of unusual questions. In my opinion, your contribution deserves a separate short appendix. Please write if for me.

2) Fauvarque is too busy now to answer questions about the appendix. I am not very happy about this. He will probably give us full attention after the exams in France are over, probably in a week or so.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
84) Richard --> Ludwik (6/23/06):
Hopefully we will be getting numbers from Scott.  His detection is less than 20ppm range.  Don’t you think this is sufficient with added comments about the chemical paper?  We should be able to get a spec on the paper for detectable limits.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
85) Ludwik --> Richard (6/23/06):
Yes, but having a conclusion based on two independant methods would be better. Let us wait and see what will be found by Scott.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
86) Scott --> Ludwik + Richard + Pierre (6/23/06):
I have not found the time to go visit my XRF analyzer yet.  However, I have a nice Mettler balance so I conducted a simple dissolved solids test on the samples you sent.  First I weighed 4 Al weighing cups.  Then I dispensed some of each sample into each cups.  I used a new disposable pipette for each sample to prevent contamination.  Then I weighed the wet cups. Then I gently dried the samples in a 70C oven for 4 hours  This gentle drying should leave the salt as K2CO3+1.5H2O.  I recorded the first set of dry weights.  Then I returned the samples to the 70C oven for another 8 hours.  I recorded a 2nd set of dry weights.  Here are the results:

[I am extracting from Scott’s large table: The mass of solid deposit from each cc of condensed wet steam was 1.68 mg. That was the mean for the samples #1, 2 and 3, after 12 hours in the oven. Individual results were 2.06, 1.15 and 1.84. This leads to the standard deviation of 0.47. For the sample #4 (the original electrolyte) the mass of the solid deposit was 27.4 mg per cc. What does this mean? Let me begin with the original electrolyte. The 27.4 mg/ml translates into 27.4 g/liter. This is in very good agreement with 26.5 grams of K2CO3 that was put into distilled water to make the 0.2 M solution.

Each cc of the solution is essentially 1 gram. That solution was supposed to be from pure steam. Finding 1.68 mg of silid deposit, per one gram of condensed liquid, is significant because 1.68 is 6.13% of 27.4. In other word, our evaporative losses were exaggerated by 6.13 %.This 2.45 times larger than the upper limit in Fauvarque’s appendix. And, according to last Thursday test in Paris3, the upper limit should be changed to a much lower number. What is the new upper limit?]

For the mg/mL column, I assumed that the solution density was 1.000 (same for the molarity calculations)  As you can see, the condensate samples are NOT very free of dissolved solids.  Ludwik calculated that 2.6 mg/mL would "erase" the apparent excess heat and these samples range up to 2 mg/mL. Note that the stock solution came out very close to 0.2M in this analysis.These results make it even more important to go to the XRF analyzer to confirm that this is actually K in the samples.  I'll do that sometime this weekend.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
87) Ludwik --> Scott + Richard + Pierre (6/23/06):
Scott wrote: “These results make it even more important to go to the XRF analyzer to confirm that this is actually K in the samples.  I'll do that sometime this weekend.” Yes, indeed. I suspect that some of the net mass is tungsten. The original solution comes as expected (0.2 M) because it does not contain tungsten. If that suspicion is correct then tungsten peaks will appear in the XRF spectra. Knowing concentration of tungsten might help us to deal with the issue of "chemical contributions to excess heat." According to JFF only ~50% tungsten lost by the cathode is recovered as metallic particles at the bottom of the cell. Where is the rest? You might be able to answer this question, Scott.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
88) Ludwik --> Scott + Richard + Pierre (6/24/06):
1) Yesterday, responding to the above, I speculated that solid deposits (samples 1, 2 and 3) might consist of mostly tungsten rather than potassium. Thinking about this again I realized that this would be equally bad for our excess heat conclusion. It would show that the measured lost mass m, that we used to calculate evaporative losses, was not pure water only. The m was overestimated by a significant fraction.

2) My suggestion is to collect samples of condensed steam in Paris3 experiments (now in progress) and to measure the mass of solid deposits in them, like Scott did. The mass of the solid deposit would be zero for pure steam. Richard's scale is sensitive enough, I think, to measure the mass of solid deposits from condensed steam. Measuring the mass with 10% accuracy would be sufficient to confirm Scott's preliminary results. Is the mystery going to be solved in such disappointing way? That remains to be seen; perhaps some kind of mistake was made by Scott. He is a great experimentalist, but no one is perfect.

3) The XRF analysis will also be an independent confirmation. It might also be very useful in the context "chemical excess heat” corrections.

4) Suppose we become convinced that the COP=1.24 (st.dev. 0.13) was mostly an illusion. What should then be done? In my opinion it will be our obligation to publish the result, for example, at the upcoming conference in Italy. Would you be willing to present the paper there, Pierre? Trying to publish a paper on "unexplained excess energy" in a mainstream journal would be out of question.

5) I hope that other researchers: Mizuno, Naudin, Iori, etc., will also reexamine their results. After all, it would not be difficult for them to do what Richard and Scott did. We should have performed similar analysis before collecting data on the COP. Please share your reflections on that subject.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
89) Scott --> Ludwik + Richard + Pierre (6/24/06):
Agreed [about “tungsten rather than potassium”]. BTW, in one of my Mizuno runs seven years ago I did measure W in the electrolyte and in the swarf (particles on the bottom of the cell) and observed that only 5% of the W lost by the cathode ended up as swarf. 95% of the lost W went into solution. See:

http://www.earthtech.org/experiments/Inc-W/2ndtry/run6.html

[L.K. wrote “but no one is perfect.”] Indeed I am not perfect. The list of my own past measurement errors is a mile long. The dry solids measurement does have a few peculiarities. You saw the difference between the 4 hrs and 12 hrs of oven time. All of the sample were HEAVIER after 12 hours than after 4 hours. That may be due to further chemical reactions in the oven involving the Al weighing cups. Each of the Al cups now has a dark stain on it where the solution dried out....some sort of oxidation caused by the basic solution, I suppose. I don't expect these stains to be very significant. We'll see when I get the K analyses done (looks like Monday is my earliest opportunity to go see the XRF analyzer). [L.K. also wrote about “our obligation to publish the result.”] .But, of course.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
90) Ludwik --> Scott + Richard + Pierre (6/24/06):
Perhaps it would be better to dry samples at room temperature. Doing this under the vacuum (or in a desiccator filled with clean air) would also reduce a chance that something from air reacts with Al and contributes to the net mass difference. Perhaps Al was not the best choice. Just speculating.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
91) Richard --> Ludwik + Scott (6/24/06):
It will be very disappointing if W in the steam turns out to be the cause of the excess heat and of course we should still publish the results. The original Mizuno experiment was a mostly closed cell with little evaporated losses.  My initial Colorado #3 experiment is encouraging but very inconclusive.  Scott you did not see any white deposits in the evaporated test?  At least it did not look like K2C03 deposits?  Scott thanks for all the extra effort!

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
92) Scott --> Richard + Ludwik (6/24/06):
I am afraid it does look like K2CO3, Richard.  Here's the four Al weigh boats in the dried condition. You can see the darkening of the Al but around the outside there is a definite white deposit. [The jpg file attached] The real white one is #4, the stock solution.  The boats are about 4cm in diameter.

BTW, I weighed them again just now after about 48 hours at 70C (I stored them in my oven which is  left on at 70C normally) and they were all within 0.1 mg of the "after 12 hours" weight....except for #4 which was 0.4 mg heavier.  It should be noted that these samples begin gaining weight as soon as you remove them from the oven.  I measured the rate on one of the samples and it was about 0.4 mg/minute.  I suppose that's water vapor attaching itself to the Al surface and to the K2CO3 as things cool off.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
93) Pierre --> Ludwik + Richard (6/24/06):
I will not add any remarks on what I have already said to Ludwik about the tiny droplets. I hope to see to-morrow Pr Fauvarque for that and may-be other points... Do not hesitate, Ludwik, to recall me what are your main concerns.
 
I would like now to draw your attention on the following question : storage and un-storage of energy, due to the transient between two levels of energy. I was not aware of this effect, because of our lack of registered measurements before. Of course, we have to take into account this effect in analysing the results.. or at least to wait a sufficient time (for us about 5 minutes) before taking into account the values obtained. It is easy, Richard, to see what I mean. You have just to make an experiment with only the ohmic heater and you choose various levels of power between 200 and 800 w for example. At least 15 to 20 minutes for each level, and you plot the COP values ( in theory = 1.0). You go of course up and down. What surprises me is that it seems contradictory with the fact that the thermal losses are quite constant on this energy domain...

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
94) Ludwik --> Pierre + Richard (6/24/06):
The only thing I would like to have, as soon as possible is the new upper limit on the percentage of droplets based on the Thursday experiment. It would be desirable to know this before we hear the results of Scott measurement of the concentration of K+. On Friday you wrote that the new upper limit was much much less that 1%.

My main concerns were enumerated in messages about the JFF's appendix. You wrote that all of them were forwarded to him. Can I be certain that he really read them carefully before deleting? I would be happy to send these messages again, if you think that this might help. And here is my comment about your "two levels of energy" idea. Anticipating that a referee might also think about a possibility of some kind of storage we should address the issue in our paper. I think I will write something like this: "The possibility of significant short-time storages and releases of energy can be ruled out the basis of high reproducibility. The mean COP of 1.24 and the standard deviation of 0.13, from 41 tests, would not be possible if such mechanism played a significant role."

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
95) Richard --> Pierre + Ludwik (6/25/06):
You are seeing the same problem that I’m working. What is your heat loss?  I removed all my insulation so that the time lag as the input or output power changed would be less about 2 minutes without the insulation and about 5 minutes with insulation.  I found that from 300 watts to 800 watts the heat loss is directly proportional to the delta temperature of the heat exchanger.  Lowering the Input power below 275 watts you really see the stored heat energy in the electrolyte and it takes a long time to dissipate. I’m looking into ways to greatly reduce the amount of electrolyte used this should improve the response.    Is this a problem with all calorimeter that has a heat storage mechanism?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
96) Pierre --> Richard + Ludwik (6/25/06):
We got a small heat loss of about only 43 watts due, I suppose, to a very good insulator (glass wool). This heat loss seems constant between 250 w to 800 w.I think that we must try to have longer and quite constant power runs by trying to keep the cathode length as constant as possible by motoring it. So this storage effect will no longer be a real problem.You can see our insulated long beaker in the joined picture and the way we choose for diminushing the weigth of the apparatus in order to be able to use our balance ...

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
97) Richare --> Pierre + Ludwik (6/25/06):
This all sounds good.  You are doing the same things that I’m trying.   I making a screw device so I can screw the cathode down into the electrolyte.  The screw will let be control how much cathode is lower into the electrolyte.

98) Pierre --> Ludwik + Richard + Jean + Gerard (6/26/06):
I just saw Pr Fauvarque coming back yesterday from Russia. As you know, he is very busy and will be free only by mid-July. We discussed however your main concerns. I noted the following points: (a) The value of 1% is classical for a chemical titration. Nothing to see with the amount of lost electrolyte in tiny droplets. As you know, my own opinion is that tiny droplets are of a negligible amount in the evaporated water, but we were not able to measure it accurately. What is the point of view of Richard on this point? (b) You have participated to the Boulder experiments (Colorado 2). They have confirmed the Paris 1 conclusions and this was a good second step. We have no more to add at this time in our conclusions than you may have to add for the Boulder experiments case. (c) To answer with a great accuracy your demands needs to do again at least one month of tests. We do not have the material and financial means to do so at this time in our laboratory. I am not very pleased to send you this uncompleted e-mail. But, I hope that the Scott tests will give you the answers you are waiting for...
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

99) Ludwik --> Pierre + Jean + Richard + Gerard (6/26/06):
1) I am also not very pleased. Should the first section of the appendix be eliminated from the draft we received? I counted on two independent determinations of the percentage of droplets. Now we have to rely on Scott only.
2) What about the second issue -- contributions of chemical heat to our COP=1.24? This seems to be a theoretical question based on thermochemistry of dominant chemical reactions. Should I expect this kind of input from Pr. Fauvarque in July? A delay of three weeks is not a big deal, my draft will have to wait. I am convinced that a paper about excess heat must contain statements that what we measured was not due to well known chemical reactions, or to droplets. The manuscript would certainly be rejected without quantitative information about this.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
100) Ludwik --> Richar + Pierre (6/26/06):
Here is a paragraph I just composed for my anticipated draft. The ??? will be replaced by the number provided by Richard. I also believe that the number will be very small, but a paper "difficult to reject' needs numbers. Anything else I can add on this topic?

"Can excess heat measured be atributed to some kind of energy accumulation process? To answer this question we considered a possibility that our cell acted as a recheargeable electric battery. In a control test the cell, after being connected to the power supply for several hours, was suddenly disconnected and discharged through a resistor. The amount of stored energy turned out to be ??? J. That is a small fraction of one percent of the excess heat measured in a typical test. We concluded that the battery-like effect, if any, had a negligible effect on the reported value of COP. Another possibility considered was short-time accumulations and releases of thermal energy. That hypothesis was ruled out the basis of high reproducibility. The mean COP of 1.24, and the standard deviation of 0.13, from 41 tests, would not be possible if such mechanism played a significant role. Furthermore, significant random thermal accumulations and releases would be associated with significant temperature fluctuations during consecutive tests. The actually observed temperature fluctuations, about one degree or so, indicated some thermal instability. It probably contributed to the size of the standard deviation."

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
101) Ludwik --> Richard + Pierre (6/26/06):
I suspect the resistor is too small. Your two parallel capacitors have 6200 microfarads. Discharging them through a 500-ohms resistor should take slightly less than 10 seconds (R*C=3 seconds).
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

102) Scott --> Richard + Ludwik (6/26/06):
I got to the XRF machine today. As usual, the story is not as clear as we would like it to be. So that you can join in the fun, I have attached Excel spreadsheets that contain the raw spectra I obtained today along with some crude calculations (this XRF machine does not have fancy software). The two sheets are K.xls and W.xls for potassium and tungsten analysis, respectively. For K analysis, I used no filter in front of the x-ray tube (so that the softer x-rays would get to the sample) and 10 kV on the anode. Unfortunately, this spectrometer has an air path from the tube to sample to detector so Ar x-rays interfere with the K analysis. In fact, the Ar K-beta is right under the K K-alpha...the most intense K x-ray. So, in fact, I used the Ar K-alpha as a measure of the background under the K peak.

Important: the data show about half as much K in the stock solution than I expected from the 0.2M preparation. And this is based upon a known standard that I prepared today, gravimetrically. More about this later. Despite this setback, there is decent agreement on the ratio of K in the condensate samples to K in the stock solution and the corresponding ratio of % solids in the condensate and % solids in the stock soln.

And then there is the W analysis. As you can see, there are large W peaks in sample #4 which tells me that sample #4 is not virgin stock electrolyte but USED stock electrolyte. At the moment, I do not have any quantitative results for the W but I did calculate the ratio of W in the condensate samples to W in the stock soln. The values match the other two ratios mentioned above pretty well. All three sets of ratios can be found in colored fields in K.xls on the "calculations" worksheet.

I now suspect that the %solids result we got that matched the expected 0.2M so well was a coincidence. The solids I'm weighing must be about half K and half W. Hmmmm..that's a LOT of W dissolved in there....I wonder if that's possible? Maybe there's another component to the solids I'm seeing when I evaporate these samples. And there's still the possibility that I've done something wrong with the K analysis....I'll continue to work on it. If I'm right, though, how did the stock solution lose so much K? Maybe it went out in the mist droplets. Did you guys use this solution for several runs, making up to the same volume each time with pure water? If so, that could explain the reduced K content I am seeing.

A small point that we need to straighten out: Is 70C hot enough to make the K2CO3 anhydrous? It's supposed to be KCO3+1.5H2O when just dried from solution but, of course, there is some point at which it loses all its water and becomes K2CO3. Richard, did you use anhydrous K2CO3 in your preparations? If not, did you calculate so that the solution came out 0.2M K2CO3 and not 0.2M K2CO3+1.5H2O?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
103) Ludwik --> Scott + Richard (6/26/06):
For the time being I will take Scott's numbers on their face values.

1) The mean and standard deviations from three samples (for K) are 0.073 and 0.015. That translates into 7.3% droplets based on K alone.

2) The mean and standard deviations from three samples (for W) are 0.070 and 0.014. That translates into 7.0% droplets based on W alone.

3) Assuming nothing else is present in the samples of electrolyte one would conclude, on the basis of these preliminary results, that the percentage of droplets was 14.3 %. In other words, this seems to be the lower limit. Right? Standard deviations are about 18% of the mean values. In other words the percentage of droplets would be reported as 14% +/- 2.5%.

104) Scott --> Ludwik + Richard (6/26/06):
No, I don't think these two 7's should be added. Instead they just support each other. Each one is an independent measure that says that about 7% of the condensate is due to droplets. And the same goes for my oven-drying tests....they also show around 7% (with considerable variation).

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
105) Richard --> Scott + Ludwik (6/26/06):
I did not use anhydrous K2CO3 but K2C03*1.5H20. I did not understand why my
bottle of K2C03 was different from Pierre until just now when you used the
term anhydrous and it dawn on me...... So it is 26.7 grams of K2C03*1.5H20
in one liter of H20.

The solution was a fresh solution but had been about 10 run adding about 50
grams of fresh water for each run. It took this many runs to get positive
COP's and enough condensate from a single run. [P.S.So sample #4 is the final electrolyte after all the test.]

I would estimate about 4-5 inchs of W 3/32 dia rod used during the tests to
get the condensate.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
106) Scott --> Richard + Ludwik (6/26/06):
At 10:01 PM 6/26/2006, Richard Slaughter wrote:
[ 1) You wrote ”...... So it is 26.7 grams of K2C03*1.5H20 in one liter of H20]. OK, that's good to know....your K content was a bit low to start with, then. You should have used 0.2 * 165.2 = 33 grams of your hydrated salt to get 0.2M K2CO3.
[ 2) You wrote “ many runs to get positive COP's and enough condensate from a single run.”] Very interesting...so, with a steady loss of K via droplets, your final K concentration could easily be significantly lower than your initial K concentration.
[3 ) You wrote: “I would estimate about 4-5 inchs of W 3/32 dia rod.”] OK, I just weighed a new, 6" rod and got 14.46 grams. That's 2.41 grams/inch (excuse us, Ludwik, while we wallow miserably in this American morass of mixed unit systems...:). If 4.5" was used, that's 10.8 grams of W, most of which went into solution according to the seven year old measurements I referenced recently. That's at least comparable to the K content.....sortof.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
107) Ludwik --> Richard + Scott (6/27/06):
My recollection is that you mentioned the COP's close to 1.10 for runs during which samples were collected. Is this correct? At the time I did not consider this to be important (10% of excess heat would also be spectacular if the contribution from droplets were below 1%). But now this fact might become essential. You probably did not use the same vessel and the same geometry as that used when we worked together.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
108) Ludwik --> Pierre + Richard + Scott (6/27/06):
Pierre:
1) Preliminary results from the XFR analysis seem to confirm what Scott reported last Friday. I expect his conclusions to be sent to us today or tomorrow.

2) Meanwhile I would like to share with you the very beginning of my draft. I no longer think that it will become a publishable manuscript very soon. But I do want to send it to the CMNS library. It will be preserved there for posterity. This will be done after the appendix is revised, if necessary, by JFF.

3) Please confirm that nobody who contributed to our work was forgotten. Naturally, I will show you the entire piece when it is finished. We have to wait for JFF to be available again.

4) I really think that what Scott did by simple mass analysis should be done independently in Paris and in Bolder. Too much is at stake, Scott's conclusion must be confirmed before we finalize its validity.

5) Are you planning to go to Italy, Pierre? I suggest that we inform the meeting about the status of our work, especially, if we are no longer as enthusiastic as we were a week or two ago. Would you be willing to do this in the name of all of us, Pierre?

Is the unexplained heat during high voltage electrolysis real?

Richard Slaughter, Pierre Paul Clauzon, Jean Francois Fauvarque, Ludwik Kowalski, Gerard Jean-Michel Lalleve, and Scott Little.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
109) Richard --> Scott + Ludwik (6/26/06):
The COPs were 1.10, 1.15 and 1.20.  You are correct the vessel and geometry were different.  I will work on confirming Scott’s weight measurements.  I think I will let the steam saturate a piece of filter paper and then weight the paper after it dries.  Scott if I sent a piece of filter paper saturated from the steam is that a large enough sample for the XRF machine?  It would be sent wet. Could you soak and/or rinse it to collect the sample?   I can also collect samples of electrolyte after each steam sample.  These results are important that we should run a second test to confirm these results.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
110) Scott --> Richard + Ludwik (6/27/06):
Richard, I need to know the total volume of electrolyte in your cell.  I am working on getting actual W concentration results from the XRF data and I need to see if they make sense with the quantity of W you eroded into solution. . . . [A paper filter] does not sound like an ideal sample, Richard.  A thin layer like that can be analyzed via XRF but with overall reduced sensitivity for the analyte.  I need about 15 mL of liquid to fill up the sample cups we use....so rinsing and collecting the sample from a filter paper would likely result in a significant dilution of the sample.  That would not be good as we are already working at low concentrations.  I need bulk liquid samples, at least 15 mL of each.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
111) Richard --> Scott + Ludwik (6/27/06):
Ok, I was just thinking that I could hang the filter paper above the cell and let if absorb the liquid from the steam. This way I don't have to worry about contamination from the plastic bottles used to collect the electrolyte. I did clean the bottles and interesting I didn't see any signs of electrolyte on the bottles after they dried. There were lots of water droplets on bottles after collected the condensate. The amount of electrolyte I was using was very close to 1 liter. I always started with 1 liter and as the electrolyte evaporated would add more up to the 1.0 or 1.1 liter mark. I know I used only one W rod for all the test collecting the condensate. The 4 inches I mentioned would be the absolute max. The absolute min would be 2.5 inches.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
112) Ludwik --> CMNS list (6/27/06):

1) Due to unforeseen delays, the draft of the paper I was working on could not yet be posted. But what I can do, while waiting for additional input, is to show the beginning of my draft. Details and modified conclusions must wait; they will probably be different from what I anticipated when the introduction shown below was written. Here it is, followed by the list of references. . .

2) Nuances often escape our attention when we quickly read or listen to what others have to say. I was thinking about tiny droplets associated with boiling. My conclusion that such effect, or large systematic errors in measuring electric and thermal energies, cannot lead to COP>1.2 was based on the fact that the measured value of L turned out to be very close to 2260 J/g, in Paris1 experiments. But L was measured with an ohmic heater. Somehow I did not pay attention to this.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
113) Scott --> Richard + Ludwik (6/28/06):
Same spreadsheets attached but this time complete.

1. I used Richard's formula for the stock solution to calculate the original K conc in the stock solution. My XRF data shows that about half the K is now gone from the stock solution....! That is consistent with it leaving via the "steam".

2. I used Richard's estimate of 4" of 3/32 W rod consumed to determine that the stock solution should have been about 1% W if nothing had been lost. My XRF analysis shows the stock solution to be only about half that W concentration....again confirming that W was lost from the stock solution.

Both of these observations at least qualitatively support the XRF results that K and W are found in the condensate samples.

3. I added the COP values that Richard put on the sample jars to the spread sheet. There is not a good correlation between them and the K or W concentration in the condensate samples.

4. I computed the 1 sigma standard deviation in the XRF results. In the case of K, it is about 5% relative of the K conc in the condensate samples. In the case of W, it is larger, nearly 10% of the W concentrations we saw. That could be improved with longer count times but I think it is adequate as is.

Note, the non-zero W readings on the two water samples is a reflection of improper background subtraction, not of actual W contamination in the water. I could do a bit better on the W analysis but the results would be quite similar to what you already have.

In conclusion, it looks like approximately 7% of the condensate volume is due to liquid transferred out of the cell. Does that translate linearly to the COP bottom line?...i.e. that 0.07 must be subtracted from the apparent COP's....i.e. that a COP of 1.10 turns into 1.03 when this is taken into account?

I think that's the way it works but I could be wrong. Ludwik? Richard?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
114) Ludwik --> CMNS list (6/28/06):
I know that Scott, Richard and probably Pierre are busy to verify a tentative conclusion reached so far --> 7% of the lost liquid consisted of the electrolyte droplets. It was based on a simple test performed by Scott. About two weeks ago Richard performed three experiments during which the escaping liquid was condensed outside the cell. The COPs, measured turned out to be 1.2, 1.15 and 1.1. Condensed liquids, plus some amount of the original electrolyte, became samples #1, #2, #3 and #4, respectively. They were sent to Scott Little who has an XRF device able to perform elements analysis.

Note that this approach is very different from that described in JFF's appendix. Titration analysis (Paris2 experiments) was performed on the electrolyte remaining in the cell, not on the re-condensed wet steam. Before performing a sophisticated X-rays analysis, Scott did something much more simple. Here is a message to Richard that I received from Scott last Friday. " I have not found the time to go visit my XRF analyzer yet.  However, I have a nice Mettler balance so I conducted a simple dissolved solids test on the samples you sent.  First I weighed 4 Al weighing cups.  Then I dispensed some of each sample into each cups.  I used a new disposable pipette for each sample to prevent contamination.  Then I weighed the wet cups. Then I gently dried the samples in a 70C oven for 4 hours  This gentle drying should leave the salt as K2CO3+1.5H2O.  I recorded the first set of dry weights.  Then I returned the samples to the 70C oven for another 8 hours.  I recorded a 2nd set of dry weights.  Here are the results: [a table with numbers] . . . These results make it even more important to go to the XRF analyzer to confirm that this is actually K in the samples.  I'll do that sometime this weekend."

Under ideal conditions sample #4 should show that the mass of the deposit is consistent with the molarity of the initial electrolyte. And samples 1, 2 and 3 should have zero mass, corresponding to 100% dry steam escaping from the cell. But that is not what was discovered by Scott. Masses in samples 1, 2 and 3 were not negligible. Comparing them with the mass in the sample 4 one can conclude that ~7% of the escaping wet steam consisted of pure electrolyte. In my opinion, everyone who is going to study excess heat from a GDPE cell (Glow Discharge Plasma Electrolysis) should first perform a similar dry mass test. I also think that those who reported on excess heat from GDPE cells, especially Mizuno, Naudin, Iorio, Richard, Pierre perform similar tests and report the results. It is prudent not to depend on what has been reported by a single researcher, even a very careful one, like Scott. In fact, I would not be surprised to find out that such tests were performed but were not reported.

Scott, can you please describe your X-rays analysis in the form of an appendix for our draft? I just read what you posted this morning and examined your spectra. They seem to confirm the results based on weights. But, as you indicated, absence of correlation between the percentages of droplets and the roughly determined values of COPs is disturbing. If droplets are responsible for the COPs>1 then samples from experiments with larger COPs must contain more electrolyte than samples from experiments in which the COPs were smaller. That was Richard's comment on what you wrote this morning. Only additional experiments can clarify this matter. My suggestion is to focus on this in Paris3 and Marseilles1 experiments. I am also disturbed by conclusions from Texas1 and Colorado1 experiments. In these experiments the COPs were very close to 1.00. Why were tiny droplets emitted in Colorado2 but not in Colorado1 and Texas1? Is answering this question not a good reason to start Texas2 experiments, Scott? I would very much like to assist you again.

115) Ludwik --> CMNS list (6/28/06):
I am looking at Scott's second table again: The mass of solid deposit from each cc of condensed wet steam was 1.68 mg. That was the mean for the samples #1, 2 and 3, after 12 hours in the oven. Individual results were 2.06, 1.15 and 1.84. This leads to the standard deviation of 0.47. For the sample #4 (the original electrolyte) the mass of the solid deposit was 27.4 mg per cc. What does this mean? Let me begin with the original electrolyte. The 27.4 mg/ml translates into 27.4 g/liter. This is in very good agreement with 26.5 grams of K2CO3 that was put into distilled water to make the 0.2 M solution.

Each cc of the solution is essentially 1 gram. That solution was supposed to be from pure steam. Finding 1.68 mg of solid deposit, per one gram of condensed liquid, is significant because 1.68 is 6.13% of 27.4. In other word, our evaporative losses were exaggerated by 6.13 %. This 2.45 times larger than the upper limit in Fauvarque’s appendix. And, according to last Thursday test in Paris3, the upper limit should be changed to a much lower number. What is the new upper limit?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
116) Richard --> CMNS list (6/28/06):
. . . If we POSITIVELY have droplets and/or mist [of electrolyte] then the method of computing excess heat is incorrect and the results of Colorado #2 and Paris #1 are wrong. We still need to validate the presence of electrolyte in the steam. One set of test is not enough and we need test with COP>1.30. . . .

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
117) Richard --> Scott + Ludwik (6/28/06):
I don't think we can just subtract the mass of the electrolyte from our evaporated mass and recalculate the COP. It seems to me that the presence of the electrolye indicates the presence of droplets. The size and mass of those droplets are unknown. I could see the droplets having at least the mass of the electrolyte. If the presence of the electrolyte is confirmed and we decide that this means the presence of droplets then it seems the open cell GDE is not worth more effort. If we decide to abandoned the open cell I do think we should publish something showing the problems we encounter.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
118) Ludwik --> Richard + Scott (6/28/06):
In my mind the term "droplets" stands the "mass lost that is not pure steam." It can be droplets, molecules or tiny crystals. In our context this is not relevant; we just want to be sure that evaporative losses are calculated correctly. In other contexts such details might be of great importance. I like the idea of publishing negative results. In that way what we did will not be totally wasted. But, once again, we need more data. In particular, a correlation between individual COPs and individual "mass corrections" must exist if a large part of COP is an illusion due to the "droplets effect." Data demonstrating the correlation would be a clear demonstration that the droplets effect is real.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
119) Ludwik --> Pierre + Richard (7/1/06):
Pierre, I did not hear from you since Scott wrote that, according to his data, about ~6% of the lost liquid was not dry steam. Are you OK? If Scott is right then Colorado2 and Paris1 conclusions must be modified. Do you agree? But his conclusion should be independently verified. Are you and Richard going to perform "dry-deposits" tests on condensed steam samples? We need many tests to check the reproducibility of results, and to get the standard deviation. Then we would correct the initial COP accordingly. Do you both agree that this is a desirable plan? Can this be done in two weeks, before Fauvarque becomes available? I still hope positive COP, such as 1.10, will be worth publishing. But we need numerical data to succeed.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
120) Pierre --> Ludwik + Richard (7/2/06):
C'est la FRANCE, Ludwik! Pierre  (one of the Colorado 2 team). [That was prompted by what Ludwik mentioned in a messsage posted yesterday at the CMNS list].
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

121) Ludwik --> Pierre + Richard + Scott (7/2/06):
1) I am glad to hear from you Pierre. I was also happy that France won the game. I started worrying that absence of messages from you might mean you are sick, or something like this.

2) If Scott is right (about the ~6% of the escaping mass being something else than pure water) then Colorado2 and Paris1 conclusions must be modified. Do you agree?

3) But his conclusion should be independently verified. Are you and Richard (and Gerard, Jacques, Naudin, etc.) going to perform "dry-deposits" tests on condensed steam samples? We need many tests to check the reproducibility of results, and to get standard deviations under different conditions. My hope is that ~6% will not change significantly with conditions, such voltage or geometry. If this is confirmed then we would correct the COP accordingly. Do you both agree that this is a desirable plan? Can this be done in two weeks, before Fauvarque becomes available? I still hope that positive COP, such as 1.18, will be defendable and worth publishing. But we need numerical data to succeed. Hope is good for motivation, not for justification.

4) And what if experimental data show that our claim of COP>1 is not defendable? I agree with Richard that in that case our paper should also be submitted to a mainstream journal. The experimental part, and theoretical appendix about chemical origin of excess heat , do not have to be changed. But the introduction and conclusions will be totally different.

5) How are you Scott? You were also rather silent recently. Are you OK?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
122) Scott Little --> Ludwik + Richard +Pierre (7/2/06):
Just fine, thank you. The only thing I have done since the K, W, and gravimetric analyses of Richard's 4 samples is to expose the dried residue from sample #4 to 100% relative humidity air by placing it in a sealed box along with an open container of water. The weight gain was about 2.5 times more than expected if K2CO3 was going to K2CO3+1.5H2O so it appears that there are other hydrate ratios involved here and it also appears that my 70C drying oven probably reduced the salt to the anhydrous form: K2CO3.

What this says to me is that we should focus on K concentrations and ignore the state of hydration....except that will give us some trouble when we try to do a real mass balance as Peter Gluck rightfully suggests.

The main problem with the samples that Richard sent me is the fact that the electrolyte was used over and over again and, according to my analyses, became more and more diluted as salty droplets escaped from the cell while Richard added only pure water to make up the lost volume. Thus the condensate samples I have came from runs whose electrolyte concentration was essentially unknown, except that it was somewhere between the original 0.2M and the final ~0.1M.

For a rigorous approach to this problem we need to perform a series of runs in which a sample of the condensate and a sample of the electrolyte are collected for each run.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
123) Pierre --> Ludwik + Jean Francois (7/2/06):
 
Je crois t'avoir dit que nous travaillons maintenant , Gérard et moi, dans le laboratoire de Jacques Dufour dont les moyens sont  plus importants que ceux que nous avions.
Nous avons rencontré cependant les mêmes difficultés que celles rencontrées à Boulder pour reproduire le phénomène. Nous avons dû alors reprendre le même beaker et refaire nos expériences. Tout cela a pris du temps et de plus, nous avons dû regarder le phénomène de stockage-déstockage d'énergie au cours des variations de puissance. Nous étions très préoccupés par ce phénomène qui pouvait remettre en cause tous les résultats acquis sur le CFR... Mes premières conclusions sont que ce phénomène est bien réel, mais que si l'on prend des précautions, par exemple n'enregistrer qu'après au moins 5 minutes de stabilisation et vérifier que les variations de puissance restent modérées pendant l'expérience, il n'y a pas de problème. Le second problème , les "tiny droplets", va être mieux examiné par Jacques dans le cadre du futur Paris 3, mais cela va sans doute prendre du temps...
Pour en revenir à Colorado 2, tu sais que je ne suis pas chimiste et que le temps me manque pour en devenir un. Il faut donc que Jean-François F. te réponde et tu peux compter sur moi pour lui demander de le faire dès la mi-juillet... Ne m'en veux donc pas de ne pas participer aux discussions "chimiques" plus activement... mais je reste cependant persuadé intuitivement que ce problème n'est pas majeur.
Voir dans ton texte ci-dessous quelques réponses en MAJUSCULES.
Amitiés   Pierre
 
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
124) Ludwik --> Richard + Scott +Pierre (7/2/06):
Hi Richard and Scott:

Pierre replied to me in French. I will try to summarize what he wrote after showing what he wrote in French.

= = = = = = = = = = = = = = = = = = = =
Voir dans ton texte ci-dessous quelques réponses en MAJUSCULES.
Below Pierre is replying to my last message by inserting words in CAPITAL LETTERS.

1) I am glad to hear from you Pierre. I was also happy that France won the game. I started worrying that absence of messages from you might mean you are sick, or something like this.

2) If Scott is right (about the ~6% of the escaping mass being something else than pure water) then Colorado2 and Paris1 conclusions must be modified. Do you agree?
OUI, SI CONFIRME (*)

3) But his conclusion should be independently verified. Are you and Richard (and Gerard, Jacques, Naudin, etc.) going to perform "dry-deposits" tests on condensed steam samples? OUI, MAIS PAS DANS UN DELAI RAPIDE (**) We need many tests to check the reproducibility of results, and to get standard deviations under different conditions. My hope is that ~6% will not change significantly with conditions, such voltage or geometry. If this is confirmed then we would correct the COP accordingly. Do you both agree that this is a desirable plan? Can this be done in two weeks, before Fauvarque becomes available? I DO NOT KNOW. (***) I still hope that positive COP, such as 1.18, will be defendable and worth publishing. But we need numerical data to succeed. Hope is good for motivation, not for justification.

4) And what if experimental data show that our claim of COP>1 is not defendable? I agree with Richard AND ME (****) that in that case our paper should also be submitted to a mainstream journal. The experimental part, and theoretical appendix about chemical origin of excess heat , do not have to be changed. But the introduction and conclusions will be totally different.

5) How are you Scott? You were also rather silent recently. Are you OK?

(*) Pierre agrees with need to correct our COP=1.24 to account for Scott results, if confirmed.
(**) They will perform Little-type tests but not without a delay.
(***) He does not know if such tests will be performed during the next two weeks.
(****) The idea of publishing negative results, if it comes to this, was already discussed by us in Denver and Pierre reminds me that we all agreed on this.
= = = = = = = = = = = = = =

Ludwik,
Je crois t'avoir dit que nous travaillons maintenant , Gérard et moi, dans le laboratoire de Jacques Dufour dont les moyens sont  plus importants que ceux que nous avions.
Nous avons rencontré cependant les mêmes difficultés que celles rencontrées à Boulder pour reproduire le phénomène. Nous avons dû alors reprendre le même beaker et refaire nos expériences. Tout cela a pris du temps et de plus, nous avons dû regarder le phénomène de stockage-déstockage d'énergie au cours des variations de puissance. Nous étions très préoccupés par ce phénomène qui pouvait remettre en cause tous les résultats acquis sur le CFR... Mes premières conclusions sont que ce phénomène est bien réel, mais que si l'on prend des précautions, par exemple n'enregistrer qu'après au moins 5 minutes de stabilisation et vérifier que les variations de puissance restent modérées pendant l'expérience, il n'y a pas de problème. Le second problème , les "tiny droplets", va être mieux examiné par Jacques dans le cadre du futur Paris 3, mais cela va sans doute prendre du temps...
Pour en revenir à Colorado 2, tu sais que je ne suis pas chimiste et que le temps me manque pour en devenir un. Il faut donc que Jean-François F. te réponde et tu peux compter sur moi pour lui demander de le faire dès la mi-juillet... Ne m'en veux donc pas de ne pas participer aux discussions "chimiques" plus activement... mais je reste cependant persuadé intuitivement que ce problème n'est pas majeur. Amitiés   Pierre

(a) As you know, Gerard and myself are now working working in Dufour's lab, which is much better equipped.
(b) We also encountered difficulties in reproducing Boulder's results. We had to perform the experiments again in the same beaker. All this took time.
(c) Furthermore, we observed storage and release of energy when consecutive tests are performed at different powers. That was important because the phenomenon could negate our Paris1 conclusions. Fortunately this turned out not to be the case. One must avoid collecting COP data before the temperature of the electrolyte stabilizes. This usually takes more than 5 minutes. Significant electric power changes should be avoided.
(d) The tiny droplets tests will be performed by J. Dufour, in future Paris3 experiments. This will take time.
(e) As you know I am not a chemist. Fauvarque must reply to you and will remind him in two weeks. Now you know why I do not participate in discussing chemical things. Intuition tells me that this is not a major problem.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
125) Richard --> Ludwik + Scott + Pierre (7/4/06):
I'm not convinced that the ~6% is a good correction. We are making the assumption that the K2C03 is coming off as particles or a new compound. Is this a valid assumption? I see three other possibilities 1) It is attached to a droplet of water. 2) The K and/or C03 is being vaporized. 3) The K and/or C03 is part of a chemical reaction. All three of these possibilities have different outcomes. Possibility 1 means we have to account for the droplet size. Possibility 2 means we should be adding the (joules) required to vaporization K2 and/or CO3. Possibility 3 might be ok if the deposit are the same as those of the chemical reaction. Does this make sense?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
126) Pierre --> Richard + Ludwik +Scott (7/5/06):
We got the same results at the beginning of Paris 3, i.e. COP around 1.0. So, we came back to our usual beaker of Paris 1( small size and no insulation), but then we tried 0.1 M electrolyte in order to get higher voltage without too many amperes...(too high power). And that worked and we got again COP between 1.2 to 1.3. The difficulty is to have a very stabilized power level and then to be sure that we have no storage or release of energy during the change in power... We are planning tests to be sure of that... ( we will try to get the same power curve with the ohmic heater and then to see the COP values obtained...)

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
127) Richard --> Pierre + Ludwik +Scott (7/5/06):
The input power needs to be stable over the same time frame as the measurement of the excess heat. Once the beaker is above room temperature it has stored energy. If I just turn off the input power I will have infinite excess heat until the beaker reaches room temperature ie Power_out/zero. As long as the beaker temperature is substantially above room temperature the electrolyte will still evaporate even without input power. In Colorado #2 just watching the current and voltage the input power seemed to be stable over the 5 minute test period. It is possible that this input power was drifting down and we are measuring some of this stored energy as output power. . . .
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
128) Ludwik --> Richar + Pierre + Scott (7/5/06):
Ideally one wants to operate under thermal equilibrium, that is when the dm/dt (rate of evaporation) remain constant. The way to accomplish this, in open beaker during the electrolysis, is to use two power supplies, one feeding the electrolysis and another feeding the ohmic heater, immersed in the electrolyte. I am assuming one can measure the electric energy delivered to the cell by each power supply. One would have to add them to calculate the COP.

Suppose the mean dm/dt is evaluated every 3 seconds and the values are plotted value versus time when an experiment is in progress. To keep the dm/dt constant one simply changes the voltage feeding the ohmic heater. If the dm/dt starts going up the voltage is decreased; if the dm/dt starts going down the voltage is increased. I think that keeping the dm/dt constant, within 2 or 3% would not be too difficult.

The temperature-versus-time plot could be useful below the boiling point. Stirring of the electrolyte is probably worth having in all cases. I thing that the last container in Colorado2 was too big for the power of 300-400 W, especially without stirring. One can probably work without stirring, at such wattages, when the beaker is much smaller, for example, one liter. Under such conditions boiling is sufficiently uniform to get a constant temperature within the cell.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
129) Ludwik --> Scott + Richard + Pierre (7/7/06):

On Jul 7, 2006, at 12:11 AM, Scott Little wrote:

>You, [Ludwik] asked:
>
>>Should I add Texas2 to the list of experiments that will be
>> designed to better address the issue of a possible illusion?
>
> Not just yet, Ludwik. We may eventually get around to Texas2
> but I cannot predict when. There are two problems. 1. The MOND
> experiment remains my highest priority right now and I am having
> a great deal of difficulty with it. 2. Earthtech is MOVING to a new,
> larger facility. It's the first time we have moved in almost 10 years.
> It will be hell. We have to be out of the present location by the end
> of this month!!!!

I hope this will not prevent you from performing the XRF tests on new samples that Richard said will be sent to you soon. These will be from experiments performed as in Colorado2. The initial composition of the electrolyte, before each COP was measured, will be better known this time, I suppose. We should not be surprised, the basis of what you did so far, if the data lead us to a less desirable conclusion. Being able to consistently justify negative results in three different ways (weights, XRF for K and XRF for W) would be very convincing.

Assume we come to a conclusion that the necessary correction of the Richard's COP's leads to the value close to one. We talked about such possibility and agreed that conclusive negative results should be published. A good appendix about the XFR would be needed. Would you also be willing insert a section about your old unsuccessful attempts to replicate Mizuno experiments, Scott? These experiments did not use evaporational calorimetery and this would be important, I think. Another possibility would be to publish two papers at the same time, preferably in the same mainstream journal. One would be about the old Scott's results and another about the Colorado2 results. It is probably not too early to start speculating about these things. But only among four of us, at this time. Right?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
130) Ludwik --> Scott _ Richard + Pierre (7/10/06):
I want you to be the first to see unit 302
http://blake.montclair.edu/~kowalskil/cf/302numbers.html
I will post it tomorrow. Comments, as always, will be appreciated.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
131) Scott --> Ludwik (7/10/06):
Good Ludwik, I like the honest way in which you discuss the situation.  I only wish that Mizuno was an active part of our group.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
132) Ludwik --> Pierre + Richard (7/11/06):
I will respect your request for privacy, Pierre. Should I assume that this refers to the second sentence only, and that the problem of storage and release of energy can be openly discussed on the CMNS list? The issue of storage of thermal energy is important. I wanted to quote you on this in unit #302 but decided not to do so.

I hope that unit #302 answered your question about what should be done to make Colorado2 results publishable. Measuring of the COPs, when 5 to 10 samples of condensed liquid are collected would certainly be desirable. But this is not essential when the purpose is to learn about percentages of the electrolyte in the escaping liquid. Just make sure conditions (geometry and voltage) are as close as possible to what we had. Can this be done in your present laboratory situation? . . .

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
133) Pierre --> Ludwik + Richard + Gerard (7/11/06):
Of course, the problem of storage and release of energy is important and we have to discuss this point openly in CMNS list. But, what I sent to you was preliminary results and reflects only our point of view, Gérard and myself . That's all. Jacques Dufour was absent from the lab these days. . . .  Now, I am wondering about the wet steam. When we make calibrations curves with the ohmic heater, are we concerned with the tiny droplets of electrolyte? Are your corrections, Ludwik, not already taken care of by the calibration curve made with the electrolyte ? I will try to check that point as soon as possible.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
134) Ludwik -->Pierre + Richard + Gerard (7/11/06):
That is a great . . . let us hope for the best. Ludwik
P.S. I do not think that the ohmic heater correction automatically corrects for the mass of tiny droplets. To convince myself I am considering an experiment in a thermos. In that case non-evaporative losses tend to become negligible. But droplets escape in the same way as from a beaker of the same size. How can a nearly negligible correction compensate for a non-negligible effect due to tiny droplets? I am thinking about a case in which the "droplets effect" lowers evaporative losses by nearly 10% while the non-evaporative losses are close to 1% of the supplied energy. Do you agree that this would not be possible, Pierre?

You were probably thinking about a situation in which the percentage of droplets in the wet steam during the electrolysis is the same as during the use of the ohmic heater. This remains to be verified experimentally. Let us suppose that an experiment shows that in both cases wet steam contains 5% of the electrolyte. How would you argue that our non-evaporative correction (such as 50 W determined by the use of the ohmic heater) also corrects for the presence of droplets in the wet steam?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
135) Richard --> Pierre + Ludwik + Gerard (7/12/06):
In one of my early attempts to collect the steam samples I use the ohmic heater as a test and got very clear water sample. When I switch to the plasma the new sample was obviously contaminated (milky white). Just an observation. Is there an easy test to measure the opaqueness of a sample.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
136) Ludwik -->Pierre + Richard + Gerard (7/13/06):

Here is one possibility: Food and drinks are often sold in transparent glass jars or bottles whose walls are flat (not round). Suppose the liquid to be tested for opaqueness is in such container. A beam of light enters the container from the left side and exits from the right side. It falls on a photodiode (from Radio Shack ?). That photodiode, like a thermistor, produces voltage that can be measured; it is proportional to the light intensity. For our purpose opaqueness can be operationally defined as V1/V2, where V1 is the photodiode voltage when the tested liquid is in the jar and V2 is the voltage when the jar is empty.

The experiment should be performed in a darkened room so that most light received by the diode is from the direct beam (not from scattered light, or from other sources in the room). I would place the photodiode into a ~15 cm pipe. A similar pipe would be used on the other side, to collimate the incoming light. A white-light-emitting diode would probably be a sufficient source. If not use a small light bulb, from an old car or a flashlight. The light intensity should match the sensitivity of the photodiode; you might need a dimer.
P.S.
Actually, V1/V2 should be called transparency, not opaqueness. That ratio would change from 0 (not transparent) to 1 (100 % transparent). The V2/V1 could be called opaqueness, if one wanted to use this parameter. Then 1 would stand for the "not opaque at all" and infinity for the 'totally opaque."

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
137) Pierre --> Richard + Ludwik + Gerard (7/13/06):
I think you are right, Pierre... I looked yesterday to the curves we plotted with the ohmic heater calibrations. I found that the slope of this curves, which must give 2260 j / g , give values very close to 2260 j / g ( at + or - 2%). With tiny droplets, these values would have been lower than 2260j/g, I suppose...Your comments, please...

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
138) Richard --> Pierre + Ludwik + Gerard (7/13/06):
I think it would be very difficult to correct the COP value for the misting. At least in a convincing way to skeptics. The original Mizuno experiment was closed cell very similar to Colorado #3 and Paris #3. I think that is the best approach now.
P.S. Pierre I meant to ask any comments from JL Naudin about the misting. Is he going to try and confirm?
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
139) Pierre --> Ludwik + Richard (7/15/06):
From mid-July to the end of August, many labs are in holidays in France... I hope that our contributions will not be too delayed... But any way,  we will look at yours with interest and pleasure. Cheers

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
140) Ludwik --> Pierre + Richard (7/15/06):
Should I still be expecting to hear from JFF in July?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
141) Pierre --> Ludwik + Richard (7/16/06):

I hope so, Ludwik, and I will tell him to do so. But, anyway, do not hesitate to remind him directly of the answers you are waiting for... Next week, with my family, we plan to go to Russia for a cruise on the Volga between Moscow and St-Petersbourg...(12 d)

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
142) Ludwik --> Pierre + Richard (7/16/06):
This cruise is on our agenda, perhaps next year. Meanwhile I would like you to take some pictures for me, at a specific location. As you travel toward Volga river, along the famous canal, about 40 miles north of Moscow, there is a village where I lived during the war. It is called "posiolok Dedenievo." There is a railroad station called "Turist" there. Your guide will know the place. Take pictures on both sides of the canal; both before the dock and after the dock. After Dedenievo there was a village "Medviedki," where a railroad bridge goes over the canal. I remember when this bridge was blown up by the Red Army, in late October 1941. The next railroad station is in a place called "Yachroma." And then a town called "Dimitrovo." Germans were in Yachroma for several days when we were hiding in a cellar, under an officially-closed big church. Then German army was pushed away. That was the beginning of their end. Take a picture of the tower of this church from the boat, if you can. Also the picture of the high hill on the right side of the canal, between the dock and the bridge.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
143) Pierre --> Richard + Ludwik +Gerard + Jean Francois (8/ ? /06):
Back from vacation in Russia ( beautiful cruise from St-Petersburg to Moscow -1800 kms- 4 pictures joined), I went yesterday to our common CNAM laboratory with only Jacques Dufour and his son there. (Gerard, Denis on vacation). Jacques has his own program of experiments using an ice calorimeter, very precise, and his aim is to work on his own theoritical approach to explain cold fusion. Of course, I let him expose his theory in future cold fusion meetings. Jacques has a very long experience on cold fusion experiments. He convinced me to establish the reality of our experimental results by eliminating all the possible artifacts definitively in an irrefutable manner. This is the only way to convince our physicists colleagues, he told me. I think that Ludwik will approve... The tree points we are examining are the following:

1- the measurements of the electric input energy :  we protected all the electric wires with external cladding (armoured? I dont' know the american word) and also used of a special circuit with resistance and capacitors to avoid large electrical impulses. I will ask Jacques to give you more details on that, but I can say that we are now very confident on our measurements.
2- tiny droplets: we have two different approaches with Jacques:
   - Jacques will favor the condensation outside of the beaker of all the steam in order to measure both the water produced and the amount of electrolyte eventually carried on ( and then the thermal energy produced) .
   - Personally, I would prefer to condensate the steam inside the beaker ( as Richard has begun with Colorado 3) and measure the cooling flow and his delta T.
We will choose by September the experimental set-up , but of course, this heavy change will take time to fix...
3- Storage and release of energy, with power changes:
I am not as anxious as Jacques is on this item. We made experiments showing that this could be masterized quite easily. But, the way to avoid this problem is to have longer experiments and to try for that to have cathodes moving very slowly in order to maintain constant their active length and then constant the power. Still there, we have to fix the experimental apparatus...
 
I am confident that we will achieve these goals by the end of this year. Yesterday, we got a COP of 1.3 with our traditional apparatus (analysis by hand to be confirmed). This experiment was to confirm the interest for "armoured" electrical wires... (answer is yes). Richard, where are you on Colorado 3 ?

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
144) Ludwik --> Pierre (8/5/06):
1) I did not hear from Richard since before your trip to Russia.
2) The Paris3 agenda looks promising. I will probably have some comments later.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
145) Ludwik --> Richard + Pierre (8/7/06):
. . . I haven't heard from you since the above question was asked. Is everything OK? By the way, I am also puzzled by Naudin's silence on the CMNS list.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
146) Richard --> Ludwik + Pierre (8/7/06):
I've been very busy at work lately. Haven't had time to work on the experiment much. We had a family reunion last weekend it was a lot of fun. Got to see all 11 of the siblings and a lot of the cousins. Attached is a picture of my brothers and sisters. I enjoy your pictures also. I do plan on working on the experiment again soon.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
147) Ludwik -->Pierre + Richard +Gerard + Jean Francis (8/8/06)
Pierre: Thanks for information about the upcoming Paris3 experiments. To refresh our memories, let me make some comments and suggestions. I hope they will be useful in planing.

1) Shielded-grounded wires should prevent occasional interference effects observed in the U390 meter. My suggestion is to add a kWh meter on the other side of the power supply. That is what was done in Texas1 and Colorado2. At that level the v(t) and i(t) are sinusoidal and a simple instrument can be used to measure electric energy. I thinking about the instrument Richard used for the ohmic heater. The second kWh meter readings will probably be consistently higher than U390. That should not be a surprise; some electric energy is constantly converted into heat inside the power supply. The rate is probable less than 40 W. An ohmic heater of variable R can be used to learn how this number changes (probably not much) with the wattage readings of the U390.

Suppose that the the difference between two kWh meters is the same fore ohmic heater operating at ~300 W and for the plasma electrolysis cell operating at ~300W. That would be a very convincing argument that no significant errors are made in measuring electric energy supplied to the GDPE cell. Sampling
the v(t) and i(t), for example, 100 times per second, and showing that the sum of v(t)*i(t) agrees with the U390, would be another strong indication that no error was made.

2) Another suggestion is to use fresh electrolyte for each long experiment. Also to measure the amount of tungsten lost by the cathode (to calculate eV of excess heat per removed atom). Large eV/atom, for example, 1000 or more, are convincing arguments against chemical origin of excess heat. Chemical reactions never produce more than 10 eV/atom (usually less than 5). Note that tungsten that remains metallic does not react chemically with anything. For that reason only a fraction of the W lost by the cathode should be counted when the eV/atom are calculated. It would be essential to measure that fraction. Mizuno wrote that nearly 100% of the W lost by the cathode was in the form of metallic particles at the bottom of the cell. One way to separate metallic tungsten from the less dense solid components would be to agitate the electrolyte and collect what is deposited quickly at the bottom, like in gold panning. By doing this several times one should be able to separate most of the metallic tungsten. A chemist might suggest a better method.

3) Performing the Little-type test (comparing the mass of residuals per cubic cm in the fresh electrolyte with the mass of residuals per cubic cm in the condensed steam) is highly desirable. That seems to be a good method of determining the percentage of tiny liquid droplets, or colloidal particles, in the escaping steam. Scott found that the mg/cm^3 in two samples were not very different. That is alarming. His conclusion, however, was tentative because fresh electrolyte was not used in each experiment. In my opinion Little-type tests should be performed at the very beginning. Hopefully, they will show that the COP>1 is not an illusion due to tiny droplets or particles. Only then should further investigations be made. I suspect that results, such as COP=1.24, would have to be reduced to account tiny particles or droplets in the escaping steam.

4) I agree that establishing “the reality of our experimental results by eliminating all the possible artifacts definitively in an irrefutable manner” is essential at this stage. We already know that results are reproducible. Now we must convince ourselves, and others, that excess heat is not due to a well known effect. Three to five well done experiments, with the maximum amount of information, would probably be more valuable than a much longer sequence of experiments, with the minimum amount of information (only what is needed to calculate the COP).

5) If the purpose of the experiment is to validate conclusions made in Paris1 and Colorado2 then an open cell (favored by Jacques) is preferable. Pierre's preference -- condensing steam inside the cell -- would be more desirable if one wanted to perform better experiments. That is what Richard is planing to do in Colorado3, I suppose. The most desirable outcome would be to show that Paris3 and Colorado3 (and Texas2 ?) results are not very different.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
148) Richard --> Pierre + Ludwik + Gerard + Jean Francois + Scott. (8/9/06)

Pierre thanks for sharing what is happening in Paris. I've been busy at work and haven't done much. I finally got the Pinnacle power supply working correctly using 220V and single phase instead of the 3 phases it usually takes. It is doing a nice job of regulating the input power.

I'm also seeing the arc suppression that we never saw during Colorado1. It's very interesting I'm able to slowly increase the power from about 100 Watts to about 1500 W. The plasma starts at about 300 Watts it is a bright red if I keep increasing the power the plasm goes from red to bright white. The two attached pictures show input power at about 500 Watts and 1500 Watts. There is little difference in the beaker between the two power ranges.

I'm also seeing the current drop if I find the operating point, just like we saw for Colorado 2. My biggest problem right now is EMI. I just can't collect the data with computer. The USB data acquisition system keeps going crazy. I'm going to work on that this weekend.
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

148) Scott -- >Pierre + Ludwik + Richard + Jean Francois + Gerard (8/9/06):
Thanks for the update, Pierre. You are going in the right direction with your rigorous checking of every part of the experiment. One point I would like to make: Seven years ago we studied the fate of the W lost by the cathode and learned that only a small fraction (~5%) of it ends up as swarf at the bottom of the cell. The rest goes into solution, presumably as H2WO4. Our measurements and rationale for this conclusion are presented in this report:

http://www.earthtech.org/experiments/Inc-W/2ndtry/run6.html

This result is also qualitatively confirmed by my recent XRF analysis of Richard's samples. Each of them contained significant levels of W.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
149) Pierre --> Ludwik + Richard + Gerard + Jean Francios (8/26/06):
[Sending us a diagram of a setup for their Paris3 experiments]

150) Ludwik --> Pierre + Richard * Gerard + Jean-Francois (8/26/06):
Thanks for updating, Pierre. I suppose you will have a small pomp forcing water to circulate through the cooling loop. The mass of the water, passing through the cooling loop during a stable run, must be measured accurately. How accurate is your flow-meter? With the new arrangement the input energy, E, is as before but the output energy, Q, is different. The COP is still defined as Q/E but now Q=Q1 + Q2 + Q3. The first two terms are as before (Q1=evaporative loss, Q2=thermal loss) but Q3 = c*m3*dT, where m3 is the mass of water passing through the loop during a run, dT is the constant temperature difference between your two thermocouples, and c is specific heat of water. Am I interpreting your plans correctly?

I suggest you place at least one thermistor (or traditional thermometer) inside the cell, preferably near the wall. The rate of cooling should allow you to control temperatures below the boiling point. Suppose the COP is significantly larger than 1.0 at 80 C. At that temperature Q1 will be negligible and, as you write, the issue of tiny droplets will disappear. My guess is that the COP will not change suddenly between 95 C and 100 C. The temperature of the electrolyte near the wall is not a measure of local temperatures near the cathode.

As you know, optical thermometers (pyrometers) are routinely used by metallurgist to measure temperatures of glowing objects. Eventually one might try to measure the temperature of plasma with an optical thermometer. How does the COP depend on that temperature? But that is a big project. For the time being the main task should be, as for the Colorado2 results, to convince honest referees that COP>1 is not due to well known chemical reactions. I hope these observations will be useful to you.