For the benefit of the general public I am going to try and keep this very simple so please excuse any overly reduced concepts or calculations. It would be useful if some reading was done on the relevant subjects and so I provide some links.
Here are some useful links for introductory reading;
Secondary Electron Emission
Electron Tunnelling Calculator
Here are some values to use in the electron tunnelling calculator
Particle energy = 0.05eV
Electron mass (kg) = 9.11×10−31
Barrier Height = 0.1eV
Thickness (m) = 2 x 10-9
Note; for room temperature we will use 300K, at this temp the average electron particle energy is approx 0.026eV, for the purpose of calculating a tail energy electron tunnelling probability using the above equation link we will use a value of 0.05eV, this particle energy number relates to the the population of electrons of a room temperature emitter having an effective temperature of approx 600K or 2x the average temperature. The population of such electrons as a percentage of the total available free electrons is a few percent (the actual number will be calculated later in this article) but for now let’s say it is 1%.
Using these numbers in the calculator you will see that we get an electron tunnelling probability of approx 1:100 and therefore a predicted forward tunnel current of 1:100 x 1% = 0.0001 of the available electron population.
The population of quasi free electrons on the surface of a typical metal can be estimated roughly by the total number of atoms of a single layer, this is approx 1E12/cm2, and by assuming one quasi free electron per atom, now as the electrons at the surface are oscillating at a mean frequency of approx 3E13Hz (10um wavelength) the total population of electrons expressed as though it is an electrical current is 3E25, as an Ampere is 6.24E18 electrons then we get a gross value of 5,000,000A/cm2. If we use a lesser figure to take into account only electrons near normal to the surface we arrive at the accepted number of 1,000,000A/cm2.
So for now we shall see that the theoretical currents of thermally excited electrons tunnelling across nm sized gaps are of the order of 100A/cm2 for a 1:100 probability, and if the gap is decreased or the barrier height reduced (or both) then the current increases to more than 10,000A/cm2. Of course this tunnelling current is a one way measure, at this point we have not calculated the reciprocal currents as may flow, nonetheless these figures are in total agreement with present day tunnel diodes where 460kA/cm2 currents have been recorded with only modest DC bias (<100mV).
The available forward current in a normal unbiased MIM tunnel diode (MIMTD) are of course completely in balance with the reverse current giving a net zero current flow. Quenco operates by suppressing the reverse current to create a net positive current flow. This is of course the claim made that will be publicly proved in due course, but for now we shall see how it works in theory and how it works in a macro device (already proved).
That a MIMTD does not produce power by converting ambient heat we can take as an assumed fact for otherwise it would have been detected, but the mechanism by which this balance is achieved is important in understanding how the Quenco works. We see that if we have a low work function emitter that forms a low barrier height to the insulator it in the first instance allows high forward tunnelling to occur, this tunnelling however quickly becomes balanced by a reverse tunnelling current. The mechanism is simple enough, the electrons captured by the collector raise the electrical potential of that electrode (often referred to as the chemical potential) and as it rises then the quasi free electrons have a reduced barrier height so the probability of tunnelling back the emitter increases. One can simply deduce by the absence of power production in unbiased MIMTD that the potential increase at the collector never exceeds the potential barriers in the external circuit (the wires / circuit that go from the collector to the emitter via a load). The external potential barrier(s) of the return circuit is primarily composed of dissimilar metals that form a barrier known as a contact potential difference or cpd, in trying to determine the work function of metals one method is the cpd, there is also a thermionic and photoelectric measurement. If we consider two different metals in contact we can imagine that the free electron binding energy of one being greater than the other tends to prevent electrons going from the high work function metal to the lower work function metal, going against this tendency requires a bias and as it happens this is equal to the MIMTD potential at equilibrium, in other words nature provides a perfect balance even though the mechanisms involved are complex.
Contact Potential Difference http://www.kayelaby.npl.co.uk/atomic_and_nuclear_physics/4_3/4_3.html
So we see that what we need to produce power is an imbalance and this is precisely what the Maxwell Demon debate is all about, is there a mechanism that can favour the movement of energetic particles in a particular direction? There are many who have worked on this idea and in my opinion Professor X.Y. Fu of China demonstrated such about 15 years ago but he has been either ridiculed or ignored, also working in this field are Prof Daniel Sheehan of USC and Dr Germano D’Abramo, and many others.
Professor XY Fu’s paper http://arxiv.org/abs/physics/0311104
To take this brief course out of the purely theoretical we introduce the already many many times reproduced (including independently) and tested (so the claim of proved) macro version for analysis. The next two diagrams show the conventional vacuum diode that is in assumed perfect balance (no output) and the device previously referred to as the sebithenco (self biased thermionic energy convertor). Note the only difference between the devices is the additional grid that is shorted to the Anode (collector).
The metal oxide coating of the emitter is the common type found in radio tubes, a better (room temperature) emitter would be the Ag.O.Cs typically found in old PM tube
First we shall carefully consider the above device without the Grid, in this device at time zero and irrespective of temperature (though this device is best suited for 1000K) energetic electrons escape the emitter and some will cross the vacuum gap and collide with the collector, some will bounce off the collector and some will be absorbed. Of the ones that bounce back some will do so elastically, some with lost kinetic energy and some with more energy (so called super-elastic), the amount of each species is dependent on the chemical potential of the collector, the incident angle of the collision and the incident energy. Furthermore there exists a yield of secondary electrons, this yield rises with chemical potential, incident angle and collision energy, typically the yield of secondary electron emissions in a vacuum tube device when used as a power amplifier or rectifier is not of much interest as it is a small percentage of the beam current. When vacuum tubes were first employed it was noted that in small signal amplifiers there could be a high level of noise attributed to secondary emissions and as a solution designers added a wire grid in close proximity to the Anode, the purpose of this Grid was to suppress secondary electron emissions and it is in fact called the Suppressor grid, or otherwise G3.
Still with the first device, we understand that at some point the number of electrons captured by the Anode must be equal to the number emitted otherwise there would be a nett flow of electrons. This obvious fact leads us to an understanding that thermionic emissions and secondary electron emissions must form a balanced equation, as with all such balanced equations we see that there has to be a parameter that provides a regulating effect. Clearly the self balancing effects are that as the chemical potential rises in the Collector so too must the emission of thermionic electrons, however the production of secondary electrons, while increased by rising chemical potential is at the same time reduced when collision energies drop. The fact is that the collision energy of an electron from the Emitter to the Collector is reduced as the Collector becomes negatively charged, this is a simple matter of electric fields causing a ballistic electron to do work as it approaches the Collector. Of course this change in secondary electron emission yield is also happening at the emitter for electrons colliding with it from the Collector. It matters not in explaining the Quenco or the Sebithenco to reduce these counterbalancing effects with any great precision as the starting point is that they do in fact balance perfectly (or at least as perfect as would likely have been casually observed). The issue of perfection of balance was dealt with in a series of papers by Germano D’Abramo, we have had many conversations on the subject over the years, we disagree on some things but nonetheless he is a noted international mathematician who finds the issue worth discussion. The gist of his paper, which at one revision stage acknowledged my contribution, is that he argues even a simple vacuum diode will show a tiny imbalance, however it should be noted that he talks of room sized devices.
Paper by Dr Germano D’Abramo on ambient heat to power using a simple structure http://arxiv.org/pdf/0912.4818.pdf
We now consider the effect of the Suppressor grid as used in the Sebithenco and the analogous mid layer of the Quenco. The actual current flowing from the emitter (cathode) to the collector (anode) in an unbiased vacuum diode, Sebithenco or Quenco is not a constant at a constant temperature but rather a function of the potential of the collector, clearly as the collector becomes more negative it will choke off the population of electrons having sufficient energy to reach it, thus when the device(s) are operated in an open circuit we will see that the voltage will plateau, not particularly surprising except that the level of this voltage is a lot higher than most would expect (if they expected anything other than zero). The full appreciation of the dynamics of this class of device necessitates us to understand both the short circuit device and the open circuit one, this essentially allows us to estimate the governing equations. The ability to provide a comprehensive and proved mathematical proof is as yet impossible for the surface physics, sub surface physics and interfacial considerations are just too complex and as yet poorly understood by science. I point the interested reader to the fact that most attempts in this area are about arguing which model better fits empirical data and that there are a number of monte carlo simulations that can be Googled. Nonetheless I will attempt to explain at the lay level how clearly the Quenco and Sebithinco can be theoretically “proved”.
When used in classic radio applications the suppressor grid is biased so as to be negatively charged in absolute terms and so being a negatively charged coulombic barrier, this results in the secondary electron emissions being significantly contained within the interval of the anode and the suppressor grid, the electrons leaving the anode are subject to an electrostatic repulsive force tending to deflect them back towards the anode (which is of course positively charged). The electrons that created the secondary electron emissions have a tendency to be normal to the anode (impacting vertically) whereas the secondary emission electrons have a wide range of emission angles and in fact tend to not be represented in the normal to the plane vector since that is the direction of the incoming impacting electron. In any case the simple fact that incoming electrons are essentially vertical and the secondaries are not means that the charged suppressor grid will tend to frustrate the back flow current. From the last paragraph it was noted that the perfect balance of a thermionic diode immersed in an isothermal bath (ie just sitting in an ambient environment) involved components of secondary electron emission currents, so any influence on the secondary electron emission current needs to be balanced lest there be a macro and continual imbalance in the device that would, for want of a better expression, violate Kelvin’s interpretation of the Second Law of Thermodynamics.
Second Law of Thermodynamics http://en.wikipedia.org/wiki/Second_law_of_thermodynamics
It should be noted that the suppressor grid is proximal to the anode in radio tubes and that in the Sebithenco and the Quenco the same geometry exists. This asymmetry means that the suppression of secondary electron emission is greater at the anode than the cathode even when the device is otherwise in equilibrium, this is perhaps not obvious and to understand this effect we must first imagine that the suppressor is uncharged. When the classic vacuum diode is immersed in an isothermal bath the nett current is zero, similarly when the Sebithenco is immersed in an isothermal bath with the grid unbiased it will have a nett zero current. If we bias the Sebithenco grid with a negative charge we know there is an effect on the secondary electron emission current that was just prior in balance, significant as this is it is not the only shift in the parameters needed for perfect balance, the proximal negative charge of the grid also means that collector’s chemical potential is also reduced. Since the quasi free electrons in the metal are mobile then the presence of an external electric field (negative) will push them inward (away from the charged grid), when an electron is pushed inwards it of course means that it has to have greater kinetic energy in order to thermionically escape the metal, so the charge on the grid means that reverse direction secondary electron emission current is reduced, and so too is the reverse direction thermionic current. Further the depression of the surface charge of a metal surface also means that there is a partial shielding effect to thermionic emission simply as the atoms of the metal become a physical barrier, whilst this effect is not large it is not insignificant when we talk of perfect balance. Now if the device is to remain in perfect balance we will need to equally effect the forward currents (thermionic and secondary), that they are not so equally effected is the essence of the operation of the Sebithenco and Quenco.
We will now examine the simultaneous change to the forward and reverse currents in the presence of a biased grid.
With reference to what most say is the impossibility of the Sebithenco or the Quenco as a Maxwellian Demon I think it is worth noting that the most persistent argument is that a demon would need to expend energy in sorting hot from cold, fast from slow. It is clearly not the case in either the Sebithenco or Quenco, the sorting is of energy by electrostatics where the sorted particles act only against static charge. So for those that wish to deal with this other than by evidence (replicated working Sebithenco’s) the argument would need to be different, and I say I know of no sensible argument against a working demon if the said demon does not require energy to be expended on its function. Even arguments involving information theory propose that there is an energy cost in observation that is greater than the benefit of the particle energy partition, though it is oft expressed as an increase in entropy caused by lost information, incredulously one widely accepted argument is that the Demon runs out of paper upon which to write down observations and so must reuse the paper for new entries, I must say it is from such tripe and silly arguments that I always felt there was a reason and justification to doubt Lord Kelvin and search for a viable Demon, so here I am………….. and it only took me 34 years.
As a background to the development of the Quenco concept the above conceptual device was considered, it simply shows thermionic electrons emitted from a room temperature emitter. The emitted electrons are subject to an electrostatic repulsive force resulting in the interaction with the fixed charge (shown as purple spheres). The effect is that only the most energetic electrons (hot electrons) are able to reach the upper plateau where they have lost their kinetic energy and are hence then ultra cold, here subject to the electrostatic forces they are essentially pushed into the gold collector. It should be noted that the percentage of electrons that can make this journey can be adjusted by the charge of the purple spheres and by the geometry. Accordingly it can be seen that we can raise the potential energy accumulated at the upper gold plateau and to thus have a calculated emf (voltage) greater than the counter force existing by virtue of the contact potentials in the return circuit. In simple terms the device suggests we can violate the Lord Kelvin interpretation of the 2nd Law. This idea was never itself built as a real device but it led to the testable device shown in the $25,000 challenge.
Note from the previous description that the production of secondary electrons is suppressed by the presence of a negative external proximal charge and that the work function of Gold is so high that the emission of a primary thermionic electron is almost zero at room temperature unless the chemical potential can rise by many volts, however with the charged purple spheres the chemical potential will never get so high. Relevant to this idea is the drawing of the gold leaf electroscope below, from this you can see the effect of an external charge on a conductor.
Furthermore even if the chemical potential could rise so significantly as to have an emission of thermionic electrons from the collector the emission would in the main part simply be reflected back by the overhead electrostatic barrier (purple spheres).
For those wishing to test this idea mathematically assume the charge on the purple reflector is equal a piece of silicon with electrons trapped at a spacing of 1E-9m and the distance to the upper Gold collector is 1mm. Use a vertical dimension of at least 5mm from the emitter to the upper ledge.
In any case this led to the idea below of arranging the geometry such that the charged region was a mid layer, not as easy to understand and it is not as effective but it allows a more compact device where we can take advantage of quantum tunnelling.
Migration of charge in a conductor under the influence of a charged region