This website contains other cold fusion items.
Click to see the list of links

326) Online logbook of a PACA experiment

Ludwik Kowalski; 4/9/2007
Department of Mathematical Sciences
Montclair State University, Upper Montclair, NJ, 07043

This unit is the continuation of unit #320; please read unit #320 first. As you know, Oriani and Fisher wrote that the secret of reproducibility, in working with PACA detectors, is in a bit of hardware taken from a successful experiment. Oriani had eight consecutive successes because his cell was preconditioned to reproduce the effect. A success leads another success, he believes. He suggested that I incorporate the O-rings, and the mylar, that were used in his most successful experiment into my setup. These pieces of hardware were sent to me today (1/9/07). This unit will be the electronic logbook for the experiment that I will start very soon. I want it to be open to all who are interested. Hopefully, people will look into my reporting each day and make useful suggestions or critical comments. The experiment will probably last several days; this will be followed by the analysis of CR-39 chips, and possibly by additional control experiments. Entries of this unit will be numbered sequentially.

1) Anticipated protocol:
The goal is to replicate Oriani’s experiment I will follow his procedure. The electrolyte will be Li2SO4 in light water (concentattion 20 grams per 100 cc). The cathode will Ni (wire whose diameter is 0.25 mm) and the anode will be Pt, more or less as in Figure 2 of unit #320. The exposed length of the cathode will be 2 cm and the constant current will be 25 mA, as recommended by Oriani today. I will have two cells in series; one made of glass (a replica of Oriani’s cell) and another made from a polyethylene bottle (as described in the appendix -- see below).

Important insertion (5/1/07):
A serious mistake was made in the experiment described in this unit. The concentration of the Li2SO4 electrolyte in water was 9 times larger than that used by Oriani (it should have been 2.2 grams per 100 cc of H2O). Is this the reason why I was not able to replicate his results? Yes, that is a possibility. The mistake was discovered today, as I was examining my notes critically. The experiment with correct concentration, to be performed this week, will be described in unit #327. The CR-39 chip used in the new experiment will by about 3 by 3 cm.

2) Anticipated control experiments:
The most obvious control experiment, testing the radioactivity of the electrolyte (both as powder and after dissolving it in discilled water) are not necessary because I have already done this one and a half years ago. Alpha radioactivity from glass, from nickel and from platinum were also shown to be negligible. But I will perform these tests again if a significant excess of tracks is found (in comparison with background). I will also check for alpha radioactivity of mylar. Oriani urges me to begin the experiment as soon as possible because the O-ring and mylar might be loosing what is responsible for favorable conditions.

3) About statistical levels of confidence.
Suppose a Geiger counter is used to measure radioactivity from a weak source. One measurement, lasting two hours, gives N1=100 counts when the source is near the detector. Another measurement, also lasting two hours, gives N2=49 counts, when the source is removed. Is the difference 100-49=51 significant? And if so then what level of confidence should be assigned to this significance? That is well known problem. Knowing that distributions of results (if measurements were repeated many many times) are essentially Gaussian we estimate standard deviations as square roots of the numbers of counts. Thus S1=10 and S2=7. To simplify the problem slightly we use the mean standard deviation, 8.5. The difference of counts, 51, is six time larger than standard deviation. That corresponds to a very high level of confidence. The well known rule is that, for example, the level of confidence is about 99.7% when (N1-N2)/S is equal to 3. According to statistical tables, the level of confidence becomes higher than 99.999% when the ratio is 6.

But that kind of analysis would be questionable, for example, if the electronic threshold of detection were not very stable, or if the high voltage fluctuated during each experiment. Under such conditions the S=sqr(N) would be strongly underestimated. It is conceivable that by repeating each measurement many times one could find that S1=60 and S2=50. Now the mean S is nearly the same as the difference between the mean values of N1 and N2. The level of confidence, in such case, becomes 68%.

A well designed Geiger counter is usually more stables than in the above illustration. But the illustration has a pedagogical value. Suppose that a measurement lasting four days yielded N1=100 tracks in the post-electrolysis chip and only N2=49 tracks in the control chip. I am assuming that chips were etched identically and that their areas are identical. How to estimate standard deviations. This if far from being obvious because of the “count or not to count dilemma. I know from experience that in addition to tracks of nuclear particles the surface of an etched CR-39 chip displays other defects, some of them do not resemble tracks while others resemble them. It is very difficult to be consistent about what to count and what not to count. This is equivalent to a Geiger counter fluctuating threshold of detection.

Ideally, the values of N1 and N2 should be determined by several, for example ten, equally trained people. This would allow one to calculate mean values of N and corresponding standard deviation. The alternative, which I will use, is to count how many times, N’, the “to count or not to count” hesitations were encountered. A typical example might be N1=100 with N’=60 and N2=49 with N2’=20. In that example I would assume, arbitrarily, that S1=60 and S2=20. Note that the mean S’=40 is not very different from the N1-N2=51. Also note that N1’ and N2’ would probably not be very different if the N1 were much larger, for example, 300. In that case the ratio of (N1-N2)/S’ would be 251/40=6.3 and the level of confidence (believing that the effect is real) would be very high. All this simply illustrates the obvious, the value of S must be estimated and the (N1-N2)/S must be large to be certain that the effect is real.

I know from experience that S is often not very different from N2/2. Taking this for granted, one has the following table for the levels of confidence:

99.7% confidence when N1/N2 =2.5
95% confidence when N1/N2 =2.0
68% confidence when N1/N2 =1.5
0% confidence when N1/N2 =1.0

In other words, under specified assumption (to be verified each time) the number of tracks in the post-electrolysis chip must at least twice as large as on the control chip. Otherwise the level of confidence (in claiming that the effect is real) becomes questionable. In a much more favorable case, when the “count or not to count” dilema is faced in only 10% of counting for N2 (instead of 50%, as above) one has:

99.7% confidence when N1/N2 =1.3
95% confidence when N1/N2 =1.2
68% confidence when N1/N2 =1.1
0% confidence when N1/N2 =1.0

The above is true when only the “count or not to count” uncertainty is responsible for random fluctuations of experimental results. And only when results are qualitatively reproducible. Random fluctuations contribute to limited precision while systematic errors, for example personal bias in dealing with the “count or not to count” dilemma, contribute to limited accuracy. It is very difficult not to be biased when one knows what is at stake. The way to avoid personal bias is to have an assistant who does not know what is better, too many tracks or not enough tracks. Fortunately, I am trying to replicate experiments in which N1 was always at least ten times larger than N2. Therefore I do not have to worry about gray areas in which statistics becomes intolerably sadistic.

4) Experimental setup
- - - - -
The geometry of cell #1 is essentially the same as shown in Figure 2 of unit #320. But geometry of the cell #2 is slightly different. The bottom of the cell is the bottle cap. Inside that cup is one CR-39 chip (area 2 cm2); it is protected from the electrolyte by a mylar film. As in the cell #1, the anode-cathode system, shown below, is rigid. It stands on the nickel cathode (a foot), supported by CR-39, below the mylar film. A large copper ring fits loosely inside the bottle. The nominal thickness of my mylar, 6 microns, corresponds to the surface density of 0.84 g/cm2 (based on the mylar’s density of 1.4 g/cm3). This is small in comparison with ranges of alpha particles of 2 and 6 MeV (1.10 mg/cm2 and 5.14 mg/cm2, respectively, according to “Charged Particles tracks in Polymers” by R.P. Henke and E.V. Benton, 1966.)
- - - - - - - - - - - -

Figure 1
The rigid cathode-anode setup for the cell #2. The end of the nickel wire, near the upper right corner, is going to be the foot, standing on mylar above the CR-39 chip. Only 2 cm of that wire is outside the heat shrink. The other end of the cathode wire will comes out of the heat shrink above the electrolyte. A thick copper wires also inside the heat shrink. But it is not in contact with the electrolyte. The purpose of this wire is to provide rigidity, in the form of the large circle, inside the bottle, The spiral platinum anode is mounted on the heat shrink column; its distance from the cathode (foot) is 22 mm. Diameters of the Ni and Pt electrodes are 0.25 mm.
5) It is 4/12/07; I was ready but . . .
The two O-rings that Oriani sent me arrived today. So I started preparing cell #2 (made from the bottle). I am using a new mylar film. Here are labels of my CR-39 chips:

1) Chip below the mylar film (the cathode wire is on top of it) -- 8810900
2) Control chip (blank) in the next room -- 8810888
3) Chip outside the cap -- 8810896
4) Chip outside the cap (below it) -- 8810901
5) Chip outside (on the conical part of the bottle -- 8810891
6) Chip outside (about 60 mm above the cathode)

The cell has been positioned and connected to the cirquit. But it is still empty. Then I started preparing the cell #1 (replica of Oriani’s cell). The mylar film has dark spots on it, probably some kind of dirt from the experiment in which it was used. At that moment I decided not to start the experiment today. I took another CR-39, chip 8810899, and wrapped the mylar around it. It would be a big mistake not to check for alpha radioactivity on mylar foil. I know that Oriani already did this; but it is a good idea to make another check. I will start the experiment tomorrow, without etching this last chip. It will etched with all other chips, probably in one week or so. It will exposed to Oriani’s mylar for only 12 hours. The mylar in cell #2 is not going to be tested for alpha radioactivity before the experiment. I can do it later if a lot of tracks are found.

6) Observations during the electrolysis:
a) The CR-39 that was exposed to Oriani’s mylar was put into another room (where the balnk chip is kept). Am-241 was used to irradiate a chip (half of it through one layer of mylar). The cell #1 has only two chips, one about 2 mm below mylar and another on the outside glasss wall. The center of that chip is at the level of the anode.

b) The electrolysis stareted, with two cells in series, at 13:45 of 4/13/07. With the current of 25 mA the potential differences were 4.49 V and 4.17 V, on cells #1 and #2 respectively. Bubbling of hydrogen and oxygen is not very intensed.

c) One hour later. All is about the same. Below is volt-amp lines for my two cells. They were taken at about 2:00 p.m. I suppose that the common intercept, at about 3 volts has something to do with work functions; two cells are chemically identical. Small differences in slopes are are probably due to differences in geometry inside the Li2SO4 electrolyte.

- - - - - - - - - - - - - - - - - -

Figure 2
The volt-amp lines for my two cells.
- - - - - - - - - - - - - - - - - -

d) 4/14/07: I forgot to mention that no electronic current stabilization is used in this experiment. But everything is very stable. Loss of water has been small so far. Here are the values from my notebook:

4/13/07 --18:00 -- 25 mA, -- 4.44 V and 4.08 V
4/13/07 --21:00 -- 25 mA, -- 4.42 V and 3.98 V
4/14/07 - - 5:30 -- 25 mA, -- 4.49 V and 4.05 V
4/14/07 --10:30 -- 25 mA, -- 4.43 V and 4.02 V
4/14/07 --12:00 -- 25 mA, - 4.38 V and 4.05 V . . . . .adding 1 cc of water to each cell
4/14/07 --12:30 -- 25 mA, - 4.43 V and 4.02 V

My milliammeter is analog and fluctuations of up to one or two mA might remain unnoticed. The current 25 mA is used because it was Oriani’s last recommendation.

e) An observation: Tiny bubbles continuously rise up, like a cigarette smoke, toward the surface. Occasionally large bubbles appear. I have seen this many times before. But my cell #2 is very wide and I discovered something peculiar. Big bubbles, above the spiral Pt anode suddenly, yes very suddenly, stop rising and turn toward the wall of the bottle, or toward the heat shrink tube. It seems that they are attracted toward the plastic maescape from the electrolyte at once, like tiny bubbles?
terial. After traversing at least 1 cm of the electrolyte (sometimes 2 or 3 cm), the bubbles stick to the surface and remain there for long time, for example an hour or more. Their sizes are between 1 mm and 3 mm. At any given time I can see at least 20 or 30 bubbles. Why large bubbles are formed during the electrolysis? Why do they not
g) It is 8:00, Monday, 4/16/07. No significant changes in the current and voltages. The cell #2 needs more water to compensate for more evaporation from its much larger surface, in comparison with the cell #1. Losses due to electrolysis are probably identical because the same current flows through two cells.

h) It is 23:00, all as before. My notebook has many nearly identical lines, like those on the first day.

i) It is 10:20, Tuesday, 4/17/07. Oriani suggested to run the experiment for four days. That means I have about 3 more hours. But I am tempted to add another day, at much larger current. This will not destroy tracks that are already on The chips. The only disadvantage will be loss of information about when the tracks were formed. But that does not matter. At this stage I just want to see a lot of tracks with my own eyes. I would not mind performing several additional experiments, one for each current, if this two-in-one experiment produces the same result as Oriani’s experiments.

j) Tuesday, 4/17/07; last line before going to the higher current.

4/17/07 --9:30 -- 25 mA, - 4.42 V and 3.98 V (total duration nearly 4 days, 92 hrs)

k) Increasing the total voltage to get higher current
4/17/07 -- 9:44 -- 150 mA -- 10.03 V and 8.17 V
4/17/07 -- 10:12 -- 150 mA -- 9.34 V and 7.86 V
4/17/07 -- 10:30 -- 150 mA -- 9.50 V and 7.79 V
4/17/07 -- 16:40 -- 168 mA -- 9.60 V and 7.93 V
4/17/07 -- 17:00 -- 150 mA -- 9.36 V and 7.35 V
4/17/07 -- 17:45 -- 150 mA -- 9.86 V and 7.42 V (Stop electrolysis. 92+8 = 100 hrs)

Note that 8 hrs at 150 mA --> 48 hrs at 25 mA (same number of coulombs).

7) Dismounting the cells.
a) The CR-39 below mylar were found to be dry below each cell. That is VERY GOOD.
b) In cell #1 (glass) only a small part of cathode (perhaps 3 mm) was in contact with mylar because the foot was nos not horizontal.
c) The cathode in cell #2 was nearly horizontal. That means the contact between the cathode and the mylar was probably on about 10 mm of the wire.
d) Visual examination of chips (before etching) did not reveal anything abnormal.

- - - - - - - - - - - - - - - - - - - - - - -

Figure 3
Cell #1. Two O-rings and the 6 micron mylar film (received from Oriani) are located between two parts made from glass. Cell #2 is not shown because it is just a polyethylene storage bottle with a removed lower part, as described above. The coin leaning on the cell has the diameter of 25mm. Let me add that glass components of Oriani cell can be ordered from: The inner diameter of tubes are close to 16 mm.
- - - - - - - - - - - - - - - - - - - - - - -

Figure 4
The rigid cathode-anode setups for two cells. The setup from the cell #1 is on the left (vertical) while the setup from the cell #2 is on the right (horizontal). Spiral anodes can also be seen.

8) Etching:
(NaOH at 6.25 N, temp ~73 C) 6 hours.

9) Results:
a) Executive summary: Nothing above background.

b) Detailed results of counting:
Chip exposed to Am-241 -- numerous tracks, as usual
Chip 8810-0888 (blank, 1.8 cm^2) 9 tracks on the labeled side and 4 on the other side.
Chip 8810-0899 (15 hrs exposure to Oriani's mylar, 1.8 cm^2) 5 tracks on the labeled side and 4 on the other side.

Chip 8810-0888 (PACA chip, 0.9 cm^2, below mylar cell #1) 4 tracks on the labeled side and 7 on the other side.
Chip 8810-0900 (PACA chip, 1.8 cm^2) below mylar cell #2) 11 tracks on the labeled side and 8 on the other side.

Chip 8810-0898 (1.8 cm^2, chip outside cell #1) 5 tracks on the labeled side and 4 on the other side.
Chip 8810-0891 (1.8 cm^2, chip outside cell #2, lower) 4 tracks on the labeled side and 14 on the other side.
Chip 8810-0885 (1.8 cm^2, chip outside cell #2, higher) 8 tracks on the labeled side and 8 on the other side.

10) Discussion:
I do not know why my three consecutive attempts to replicate Oriani's results were not successful. Something essential must be present in his cell but in my cells. How else can this be interpreted? Will other TGP researchers, still conducting experiments, be more successful? That remains to be seen. For some reason the CMNS list for cold fusion researchers is silent about TGP experiments performed in March and April.

11) Inserted on 4/19/07:
From a private telephone conversation with Oriani I heard about another spectacular result. He is now investigating the effect of another parameter. Meanwhile he asked me to put the CR-39 between the O-ring for several days. I started this experiment today.

12) Appendix :
The easiest way to describe my polyethylene cell is to reproduce a message that I posted several days ago, on a private list for CMNS researchers. Here it is:


On Apr 5, 2007, at 11:06 AM, Ludwik Kowalski wrote:

2) In a private message to me a TGP researcher wrote: "Ludwik, Thanks for suggesting this very low-tech way to make a cell with Mylar protecting the CR-39 in the screw cap of a vitamin bottle. Upon having a repeatable result, this could be a simple setup for others to duplicate without machine tools or laboratory glassware. Have you leak-checked it yet?" Well, the test performed yesterday turned out to be negative. Water did manage to leak (during the night) into the space that was expected to remain dry. Perhaps the cap was not screwed strong enough. I will try again, also with water (and with dry tissue instead of CR-39). Please join me in the effort to offer a simple setup for students and teachers. Weeks ago Scott reported on using the epoxy glue. That is worth trying.

Here is my original description:
a) Take a small plastic bottle with a screw-top lid. I am experimenting with a bottle in which vitamins are sold in the USA.
b) Using a knife, remove the bottom from the bottle (the created opening will become the top of your cell).
c) Place the CR-39 detector into the lid.
d) Cover the detector with the six-micron-thin mylar foil (~ 5 by 5 cm) and screw the bottle into the covered lid.
e) Gently pour the electrolyte into the bottle. The electrolyte will not be in contact with CR-39.
f) Mount the electrodes (more or less as in Figure 2 mentioned above) and start the experiment.

I mentioned the 2-HDPE label on the bottle. Francis Tanzila commented: "2-HDPE is a grade of high density polyethylene and the white pigment is probably TiO2. These should be fine with Li2SO4 but I would look for an un-pigmented plastic bottle just to be safe."

The second test (a better bottle) turned out to be positive. Here are details, for those who might be interested:

a) The vitamin bottle cap used in the first test was wider than the cup on the new bottle -- inner diameter 45 mm versus 29 mm.
b) The first cap takes only about 150 degrees to either screw or unscrew it. The second cap takes slightly more than 360 degrees.
c) The second cap has an O-ring-looking plastic rim at the bottom, to seal the bottle very well. That built-in plastic O-ring was probably the most important contributor to my success.

The material is "high density polyethylene" but without pigment. It is a standard 500 cc bottle used by chemists; the inner diameter 70 mm. Such storage bottles can be purchased from any supplier of chemical hardware. I used the same mylar sheet that failed in the vitamin bottle test. First I had a column of water of 5 cm. I emptied the cell after 6 hours and found that the mylar-protected tissue was dried. Then, using the same mylar sheet, I increased the height of the water column to 9 cm. This test was also positive, after another 6 hours. The 500 cc bottle is probably too large for our purpose. But that what they use in our chemistry lab.

This website contains other cold fusion items.
Click to see the list of links