I am getting closer and closer to have enough material to submit an article to PRL. Gradually I am getting better at controlling parameters and making the effect easier to replicate.

Replication Steps

I am able to replicate neutron emission by following these steps:

1. I thoroughly degas oil. I pull vacuum on oil in the reactor while heating using an electrical heat strip wrapped around the reactor. In the same time I cavitate the oil with the duty cycle of 1 second; bubbles need time to rise to the surface; continuous cavitation traps bubbles due to Bjerkens forces and thus makes degassing less efficient. I stop degassing when I read ~1E-3 Torr in the reactor, which may take half a day.
2. I introduce ~1 ml of benzene-d6 into the reactor. Because I am out of deuterium gas I decided to try mixing a small quantity of fully deuterated benzene (aka benzene-d6) with the oil. I have a 90-degree KF16 valve, which forms an upward facing inlet. I deposit ~1 mL of benzene-d6 into the valve using a pipette and top off the inlet with oil to remove air. Then I seal the valve opening with a KF16 blank and open the valve to admit the deposited fluids into the reactor. The pressure spikes to 6 Torr as the benzene-d6 evaporates and (along with some of the residual air) fills up the reactor headspace.
3. I turn on the acoustic drive. My Branson 550 Sonifier is operating at 100% power with 0.02 / 0.01 second on/off duty cycle.
4. I register neutron flux. My neutron counts go up to 10-20 CPM almost immediately, and I am able to sustain the emission for 5-10 minutes. Eventually some critical parameter(s) change and the neutron production quits. I am yet to determine how to sustain the neutron flux indefinitely.

When the neutron flux stops I can repeat the steps 1-4 and get it working again. So far it works every other time, which gives me enough repeatability to make progress.

Neutron and Gamma Counts

I am gradually gaining confidence in my neutron counts and I must say that the data looks better and better every time. Here is what I have so far, using my bank of 6x LND 251106 ³He-tubes.

1. Background counts average 0.5 CPM and never exceed 4 CPM. I have collected 139 4-minute background samples overnight – Fig. 1.
2. Reactor neutron flux ranges from 10 to 20 CPM, which is 20-40 times in excess of the background.
3. Neutron events are coincident with the acoustic drive – Fig. 2.
4. I have ruled out vibration as a possible source of a systematic error. I have folded two paper towels to break mechanical contact between the detector bank and the reactor; the bank was sitting on a crumpled paper towel and another crumpled paper towel was between the reactor side and the detector bank. I got almost exact same counts, but now the detector signal was very clean and free of EM-noise, which the bank must have been picking up from the reactor surface – Fig. 3. Now I could not tell if neutrons were coincident with cavitation, but the shape of the neutron spectrum improved significantly and began to approximate the thermal neutron spectrum – Fig. 4. Given long enough counting time I should see a good thermal neutron peak on the spectrum. This means that my neutron detector bank is operating as expected and actually counting neutrons.
5. I see gamma events. Detection of x-rays or gammas is critical to corroborate thermonuclear fusion. The -2.5 mV discriminator on my neutron detector bank effectively rejects gamma events, which I nevertheless can see when I view the raw signal – Fig. 5. And there are a lot of gamma events! And these gamma events are not present when I capture background (i.e. when the cavitation is off).
6. I think I see a ~100 keV line on the gamma spectrum – Fig. 6. I used a GAMMA-PRO 3″ system fitted with a 3″ x 3″ NaI(Tl) detector to capture a gamma spectrum in the 10 to 1000 keV range when the cavitation was on and when the cavitation was off (background). The counts with the cavitation on were ~1% higher than background, the result was highly statistically significant (P = 0.012) – Fig 7. The increase in gamma counts corroborates the increase in gamma events visible on the neutron detector trace.
7. Acoustic trace (captured using the PCB 113B22 transducer) is full of 1,000 psi peaks – Fig. 8.

Summary of Results

My neutron and gamma counts corroborate each other and begin to paint a picture consistent with a nuclear reactions. I ruled out detector systematics and I definitely get a 20-40 above-background neutron flux coincident with cavitation. The neutron flux seems to be accompanied with a gamma emission, which may have a peak at 100 keV.

When there is no cavitation or when there is no deuterium present – neutron and gamma counts remain at background level.

Next Steps

1. I need to add a fixture to my reactor (like a vacuum grade syringe) that would introduce a known volume of benzene-d6 (or any other fluid) into the reactor oil. This will help with reputability as well as reduce contamination with error.
2. I need to capture better gamma and x-ray spectra with improved signal to background ratio so I could better discern peaks that may be present in the differential spectrum. I am thinking about using a CapeSym SrI₂ 14×25 mm gamma detector (10 keV to 2 MeV) and a CapeSym SrI₂ 14×1 mm beryllium window x-ray detector positioned inside the reactor just above the surface of the oil to capture better spectra. Or maybe I will reinstall the glass viewport, position the detectors in front of it, and shield the detectors on all sides with lead bricks to reduce the background counts.
3. I need to start capturing acoustic trace concurrently with gamma and neutron traces to see if I can spot a correlation between neutron events, gamma events, and acoustic signal.
4. I need to look for trace amounts of tritium in the reactor headspace using my Femto-Tech PTM-1812 tritium detector.
5. I should get a BubbleTech thermal neutron detector vial, wrap it in paper towel to eliminate vibration and use it as a secondary means of confirmation of my neutron flux to meet the publication standard.

It would be really nice to workout the degassing and the deuterium introduction procedure to where I could obtain the neutron flux every time and sustain for longer than just a few minutes. Back in September of 2021 by chance I got a flux in excess of 1,500 CPM that I could replicate for several days. But that was with tritium. It would be nice to build the flux up to the same level so I could activate an indium foil and look for the activation gamma spectrum using my HPGe detector.