A Report on the Development of an Inertial Electrostatic Confinement Nuclear Fusion Apparatus
by: Mark C. Langdon
Director, GEMESYS Ltd.,
Waterloo County, Ontario, Canada
November 24, 2006
As a result of the expiration of patents on the Hirsch-Meeks-Farnsworth "fusor" technology, the intellectual property surrounding what is termed "Inertial Electrostatic Confinement" fusion is now in the public domain. Significant research on the science
that surrounds this method of creating nuclear fusion has been carried out by several academic institutions, including the University of Kyoto, Japan, and the Universities of Wisconsin-Madison, and Illinois in the United States of America. As well, there have
been several significant "open-source" efforts to replicate the Hirsch-Meeks-Farnsworth results that have been carried out by independent researchers, due to the elegant simplicity of the associated IEC methodology.
Typically, an IEC "fusor"
consists of a containment grid suspended inside an evacuated, metallic vacuum chamber, which is raised to a high-voltage negative potential within the chamber. The grid operates as a cathode, while the anode is provided by the metallic chamber, which is held
to ground state. Deuterium gas (hydrogen with a proton+neutron nucleus) is introduced into the chamber, at a rate which approximates a controlled leak, and the chamber pressure, applied voltage and current is adjusted to create ionized D2 atoms and circulate
them so that they interact at fusion-reactive energy within the low-pressure regime which is maintained in the chamber. The picture below illustrates my fusor in operation, in the regime which typically generates the most evident neutron flux.
Fusor in operation showing deuterium-fueled fusion occuring within containment grid.
IEC (Inertial Electrostatic Confinement) fusion device which is fueled by deuterium gas is characterized by two possible fusion reactions, which produce neutrons of roughly 2.5 MeV (2.5 million electron volts). The deuterium gas fusion reaction is:
(a) 1D2 + 1D2 ® 2He3 (0.82 MeV) + 0n1 (2.45 MeV)
(b) 1D2 + 1D2 ® 1T3 (1.01 MeV)+ 1p1(3.02
Note that reactions (a) and (b) occur with equal probability.
The neutron field which can be produced by an IEC fusor is small, typically in the range of 10e4 to 10e6 neutrons per second. This reaction rate is
sufficent to allow activation experiments to be carried out, and also allows the fusor to be used as a neutron field-generator. There are efforts underway by several groups to commercialize IEC fusor technology, as the stable nature of deuterium means that
neutron fields can be generated for scanning and analytic purposes, without requiring the presense of any radioactive isotopes.
One immediately obvious benefit is that a device can be configured
to probe for the presence of radio-isotopes, for example in shipping containers, by subjecting the containers to a low-energy neutron field, without also having radioactive material present in the machinery doing the radiation assessment of the container.
The IEC fusor technology can be switched on and off, and when not operating, there is no radiation present. When operating, although the rate of D2-D2 fusion is low, (10e6 neutrons per second is typical for a well-tuned fusor), the resulting weak neutron field
may be sufficient for scanning purposes, if sensitive detection equipment is used. There is also research underway by various groups and individuals into ways to improve the fusion reaction rate in an IEC device. Researchers at University of Kyoto have experimented
with multiple containment grids, using a magnetron to enhance the deuterium ionization rate, and altering the geometry of the confinement field. The University of Wisconsin, Madison IEC device makes use of three hot filaments to enhance deuterium ionization,
and they have experimented with constructing the containment grid from rhenium, and have reached neutron production rates of 1.8 x 10e8, at 180 kilovolts. [Rhenium, "Re" - atomic number 75 - is right next to tungsten, "W" - atomic number 74 - in the periodic
table. Rhenium is used instead of stainless steel to avoid the sputtering problem that occurs with stainless steel.]
My interest in IEC fusion and the Hirsch-Meeks-Farnsworth fusor technology came about as a result of encountering the
open-source consortium which operates a web-site for those doing IEC fusion research outside of traditional academic and institutional settings. I found the existence of the open-source fusor consortium to be absolutely fascinating, as the history of science
is replete with examples of talented, self-educated visionary individuals making significant, world-changing scientific discoveries, outside of the academic mainstream. The most well documented example of course concerns the "thought-experiments" of a Swiss
patent clerk - who developed the most famous equation in history - but there are many, many others in a similar vein. Einstein's revolutionary work , far from being unique, can be seen as the roadmap for how scientific change has to happen. The independent
bicycle mechanics of Kitty-Hawk, the Wright brothers, had no outside funding for their aeronautical research. They not only invented the airplane, they invented the wind-tunnel - the method by which aircraft frames are scientifically assessed. Many people
overlook this. The Wright brothers were superb scientists, who did meticulous research and combined this attention to detail, with the visionary leap of creativity that people must have if they are to create the future.
I was deeply impressed
that serious fusion research and development could still be done outside of the high-cost academic mainstream, and yet the results reported by Richard Hull, and others who have replicated the Hirsch-Meeks-Farnsworth results by designing and developing their
own "fusors", proves this to be the case.
I literally stumbled across the consortium's website while doing some research on the history of the development of radio and television. The development of radio by G. Marconi is probably the
most interesting example of applied scientific research and development in science history. Clerk-Maxwell had developed the theory and equations that described electromagnetic radition, and Hertz had shown that sparks could be essentially transmitted and received.
Marconi began his experiments at his family's farm in Italy, when a very young man. He very quickly established that if he made use of the high voltages supplied by Telsa coils, and used large capacitors, he could increase the transmission and reception range
of the electromagnetic radiation described by Clerk-Maxwell, and demostrated by Hertz.
Marconi carried out his development with little more than coils, capacitors, and various types of antenna configurations, until he could increase the
distances over which his "wireless" device could be effectively used. He had as his starting point the incontrovertable results that Hertz had demonstrated in the laboratory, and the magnificant device that Tesla had created to generate very high voltages.
[A working Tesla coil is an amazingly interesting apparatus, as it is a resonant device - an L-C circuit, which makes use of the charge-discharge characteristics of inductance and capacitance.] Marconi built the first radio devices using an interconnected
configuration of coils and capacitors. He experimented with different combinations, all the while making slight improvements in reception distance. He was routinely ridiculed as a fraud and a charlatan, despite his obvious and demonstrable successes, and his
formal relationship with the British Post Office, which was desperate to defeat the monopoly that the trans-oceanic Cable company had on long-distance communication. Only after his wireless devices were installed on ships, and could be shown to be both effective
and profitable for ship-to-shore communciation, was his technology accepted by skeptics. His early research and development efforts to develop and commercialize radio make the most fascinating reading.
The story of the development of television
is even more outrageous. Like fusion research is at the moment - a promising technology which just does not quite work yet, and yet we all know that at some point it will, and when it does, it will be just wonderful - television existed as an idea for many
years before it was successfully developed. When the technology was finally all pulled together, it proved just as revolutionary and commercially valuable as the most optimistic visionaries forecast. And yet, the key developments came down to a couple of people,
who operated outside of the academic world. A proper description of the development of television would fill several large volumes, and is far beyond the scope of this little paper, but what is most interesting, for the purposes of this note, is that the fellow
who created the most critical piece of technology - the TV camera - was also one of the
inventors of the IEC fusion device - Philo T. Farnsworth. Numerous television schemes were devised that made use of mechanical scanners to create the electronic
image on a screen, but it was V. I. Zworkin, and P.T. Farnsworth, who, in the 1920's, both took out patents on electronic versions of television receivers. And Farnsworth was key, in that he invented the "iconoscope", the basis for an electronic TV camera.
Without cameras, all TV could do was scan films frame by frame, and transmit these pre-scanned film images. With the iconoscope camera, it was possible to create a live image electronically, and transmit that image in real-time. This made television real,
and Farnsworth won his patent fight to establish his prior claim on this technology which he both concieved of, and successfully developed.
Now, it is really interesting in that in the 1950's, Farnsworth
also concieved of the idea of the IEC device, and successfully patented this idea, by 1966. Farnsworth's patent, which is number US 3,386,883, and was filed May 13, 1966, was granted on June 4, 1968. I was able to obtain a copy of this patent, which is titled:
"Method and Apparatus for Producing Nuclear-Fusion Reactions" from the European Patent Office, which has copies of historical USA patents in its database. The patent runs to 35 pages, and provides details on two different fusion-reactor embodiments, as well
as Farnsworth's take on the physics that he expected to be key to the device's operation as a power source.
The other key patent I worked from was filed by a physicist, Robert L. Hirsch, and his research assistant, Gene A. Meeks, who worked
in Farnsworth's lab, just prior to it being shut down by the congolomerate, ITT, in 1969. Hirsch and Meeks filed their patent on April 24, 1968, as serial number 723,825 and they described the version of their fusor which utilized ion-guns to enhance the fusion
reaction rate, and a employed a simple, spherical geometry.
None of the fusor variants developed by Farnsworth's lab were able to develop anything close to net power, as they all consumed more electrical energy than they outputed, either
as heat or as electricity. Farnsworth's final version of the device was designed to produce electricity directly, and was an impressive piece of engineering, even if ultimately unsuccessful for the purpose for which it was designed. I have been unable to determine
to what degree it was evaluated. One key feature of the Farnsworth apparatus, was that his team used tritium, a radioactive isotope of hydrogen (unlike deuterium, which is stable and common). The use of tritium improves the fusion reaction rate, but this meant
the team was forced to work under much more stringent conditions, not the least of which included government monitoring of the lab. This must have both raised the cost of development, and also lessened ITT's interest in continuing to fund what was a rather
high-risk research effort.
Farnsworth's Mark III Fusor
When ITT shut down the Farnsworth
research lab in 1969, Farnsworth relocated to the University of Utah, according to the information I have been able to obtain, but he died in 1971, and was unable to take the Mark III research effort foreward to the point where a detailed evaluation could
be produced. There is no direct evidence that this device ever produced any electric power, despite the impressive claims made in the patent document. But Farnsworth was an exceptional genius, and the Mark III was the embodiment of a tangible research effort
that yielded significant, and verified results. His earlier fusor designs did fusion, but suffered from the problems that plague all fusion efforts, that being simply that power-output, remained stubbornly below the levels of power-input that were needed to
make the fusion reactions occur.
Hirsch-Meeks Research Fusor
The Hirsch-Meeks version of the fusor was smaller, and more oriented as a research tool which could be used to assess the effectiveness
of the ion guns, which were used to accelerate deuterium or tritium ions toward the "poissor", the hot region of trapped ions and electrons, that is evident at the central focal point of the containment grid. A picture of the Hirsch-Meeks device shows Meeks
inspecting the configuration. Both photos are courtesy of Richard Hull, the administrator of the open-source fusor.net website.
of Gene Meeks inspecting the original Hirsch-Meeks Fusor
The spherical chamber was reported as 6 inches in diameter, the anode cage within the chamber was 4 and 1/2 inches diameter, and the containment grid cathode was 1 and 1/2
inches diameter. Gas pressure for optimum operation was reported as being between 10e-3 to 10e-4 torr. The simplicity and elegance of this design makes it the version of IEC technology that is most replicated now. Hirsch and Meeks reported that their device
was successful in producing significant fusion events, but operation was problematic, in that the high power levels - both used by and generated by - the device would cause the cathode containment grid to melt. Nevertheless, it is this variant of the IEC fusor
that has most often been replicated and refined by current researchers.
A schematic of the Hirsch-Meeks variant of this IEC fusor configuration is shown on the next page.
The Hirsch-Meeks "Apparatus for Generating Fusion Reactions"
To the best of my knowledge at the moment, in addition to the various research efforts underway in Japan
and the United States, (Univ. of Kyoto, Univ. of Wisconsin-Madison, Univ. of Illinois) there are 18 non-commercial, "amateur" built devices, which are documented by the fusor.net open-source consortium. My efforts, which first yielded neutron output on September
14, 2006, is the 17th privately developed IEC nuclear fusion apparatus in the world, and I believe, the first in Canada. The most recent fusor to be brought online for private research purposes, has been developed by Thaigo Olson, of Detroit Michigan, a 17
year-old high-school student who is hoping to earn a University scholarship as a result of his efforts. His work has been documented by the local newspaper, the Detroit Free Press and I mention Thiago's efforts, as his work is impressive, and has been followed
by participants in the open-source consortium. We wish him well, and hope that his applications meet with success.
The technology and methodology that is applied to create fusion in an IEC device is both simple and elegant. The deuterium
plasma is confined and concentrated by the containment grid to create a phenomenon know as "star-mode", which seems to correspond to the most optimum ion-circulation regime for neutron production.
In my case, partly I suspect as a result
of my choice of a non-spherical reaction chamber, and also due to my desire to experiment with the topology of the containment grid, I went through three unsuccessful grid designs before I was able to achieve solid evidence of fusion taking place. My fusor's
first successful neutron output was captured by the reaction chamber viewport monitoring camera, and is indicated below:
First successful neutron-producing run of my IEC fusor
This result was documented by the neutron detectors which I acquired from BTI - Bubble Technology Incorporated, a specialist
developer of neutron dosimeters, which are used by personnel in nuclear power plants to monitor their neutron exposure. The BTI detectors, model BDR100's, were configured at the maxium level of sensitivity that BTI makes for ordinary commerical use. These
detectors will indicate 33 bubbles, on average, for a neutron dosage of 1 mrem. As can be seen from the photograph below, taken of the left-side and right-side detectors, there was clear evidence of neutron production, as the left-side detector showed 7 bubbles,
and the right-side detector showed 3 bubbles. This works out to a neutron production rate of roughly 5000 per second, and corresponds to 5/33 or roughly 0.15 mrem dose of neutron radiation directly adjacent the chamber.
Neutron dosimeters showing evidence of neutron production
Of course, at these low levels, there is virtually no chance of any detectable neutron activation occuring, but the presence
of the bubbles proves that the fusor was operating in a fusion-reactive mode. Nuclear fusion of deuterium ions was taking place, albeit at a very low level. These Canadian-made neutron dosimeters are particularly attractive for this type of research, in that
there is no electrical "noise" that can affect them, the way proton counters, He3 detectors, and other related scintilation-based equipment can be plagued with. There is no need, for instance, to take background readings with the BTI bubble detectors, as there
simply is no background that can be detected. These detectors need to be in a fast-neutron field in order to register. They will not even count thermal and epithermal neutrons (neutrons which are moving slowly.) The BDR100s have an energy threshold of 100keV.
Bubble Technology Inc was a spinoff from the Federal Government research establishment at Chalk River, and is still located in Chalk River, near the government research establishment. This allows them to calibrate their dosimeters with high-accuracy isotopic
The GEMESYS Research Fusor
Two photographs of the actual device are provided on the next page. The first shows the fusor partially assembled, which illustrates its non-spherical conical
geometry, while the second shows the completed apparatus, with the fourth version of the containment grid, just prior to successful initial ignition.
Above: The reaction chamber of my fusor, partially assembled, in June of 2006.
Below: The completed apparatus, just prior to first successful fusion ignition.
Some Details on my IEC fusor - Sept. 13th, 2006
Grid "Flashover" Phenomenon
I have observed an interesting phenomenon when operating and configuring
my fusor. The devices are inherently unstable, and when operated manually, it is difficult to balance the flow rate of the deuterium gas, and the appropriate input voltage. Too little pressure, the ionization level falls and this extinguishes the visable reaction,
whereas too much, and the mean-free path of the circulating ions is reduced to the point where they cannot achieve fusion-reactive energy. This makes for the need to maintain a fine balance of input parameters. As vacuum system operation must also be carefully
optimized to the available D2 flow rate, the input voltage must be adjusted so as to avoid exceeding the limits on the power supply so as to avoid tripping load-limiting circuit breakers. It is not surprising that instabilities can occur, I feel.
The following grid "flashover" phenomenon occured in a recent test run, and I captured this as a digital image. "Flashover" seems to be some kind of internal plasma arc, and the factors which trigger it in my device are unclear. But I include it here as
it may represent an exploitable anomoly.
The containment grid "flashover" phenomenon
Grid "flashover" is interesting
and appears dramatic, but the fusor settles down just as quickly as flashover happens, and operation returns to the regime as shown in the first three pictures.
Earlier Unsuccessful Containment Grid Designs
Prior to the Version 4 grid design (as shown in picture above on "flashover"), I used a more sparsely framed containment grid. It did not seem to work at all, but in low-pressure mode, prior to any D2 injections, I caught an image of the following preliminary
Preliminary Test - prior to D2 injection
Although this Version 3 grid did not generate neutrons,
it did have high-permiability, which allowed for the formation of this dramatic plasma jet. The Version 2 grid was smaller, and had only three great-circles. It did not appear to focus the resultant plasma "poissor" well. The "poissor" was Farnsworth's term
for the hot plasma spheroid that appears in the geometric centre of the containment grid.
Poorly-focused Version 2 Grid
The original containment grid design was sprial-shaped, and I had hoped that it might provide some improvements to the typical fusion reaction rates reported by other researchers, perhaps due to improved ion circulation and grid permeability.
This was unfortunately not the case, as the spiral-grid design did not show any evidence of neutron-production, despite the dramatic mother-daughter dual poissor configuration that was evident at higher voltage (> 20Kv) settings. It looked good, but it
did not work. A photo is provided below:
Original Unsuccessful Version 1 Containment Grid
Current Status of Research and Future Plans
At present, further research is ongoing. I have made use of a high DC-voltage, capacitor triggered, arcing (ie. sparking) source within the reaction chamber for the purposes of trying to improve
D2 ion generation, and hence fusion reaction rate. This approach does not at this time appear to enhance fusor operation. So, the ion-generation device will be re-configured. I intend to replace the arcing ion generator with a thermionic nickel-chromium or
tungsten filament, which will be heated by a low-voltage current. An "ion gun" may also be developed, and I will be attempting to obtain some rhenium wire to eventually create a containment grid that is less prone to sputtering the chamber.
I also intend to enhance the neutron monitoring capability with some type of electronic monitoring device, so I can achieve a more precise calibration of the fusor's operating parameters. I hope to use either a proton-scintiliator or a helium-3 detector, and
also install technology that will allow fine metering of the D2 flow into the reaction chamber. The photographic monitoring also needs to be improved, as the CCD camera and associated computer are regularly knocked offline by electromagnetic pulses that are
produced by the running fusor. The cause of this appears to be internal arcing or perhaps grid flashover. A top priority is to obtain a better needle-valve, to allow the D2 insertion rate to be very finely controlled. At present, I have a staged series of
bellows type valves, with an AP-Tech regulator, and this array of plumbing does not lend itself to the fine calibration that is indicated by the physics of the device. I may also need to enhance the vacuum system, though at present, the recently installed
Varian diffusion pump seems to work quite reliably.
Eventually, I hope to try to employ the "laser wakefield" effect, using the hydrogen plasma within the chamber and some form of high-energy laser. This will allow me to assess if there is
any neutron output rate improvement to be found by directing some form of beam energy at the poissor nexus, for example, and dumping a burst of fast electrons down the same pathway. I want to assess if the high-energy electron burst can be used in some way
to enhance the focusing of circulating D2 ions, and hence enhance the fusion reaction rate. A proton is roughly 1840 times heavier than an electron, so attempting to a enhance proton- proton reaction rate by firing a burst of electrons at them is perhaps like
trying to alter the direction of a falling brick by hitting it with a shotgun blast of number 8 lead shot. If you already have the proton-neutron pairs essentially "falling" through the centre of the poissor now, and generating fusion reactions when they impact
each other at sufficient velocity to overcome the coloumb-barrier, then it seems that one only needs to further focus these falling proton-neutron pairs so that the probability of spatial intersection at fusion-reactive speeds is increased. I can only put
so many electrons on the grid, which I have done. But since I have fusion happening (current neutron rate is about 34,000 neutrons/sec), then maybe I can use directed beam energy and high-speed electron bursts, to more tightly focus the system. I hope to try
November 24, 2006