A blue-green glow has been observed on electrodes of high voltage
electrolytic cells and electrolytic rectifiers. [1] [2] Electrodes
to create the blue-green glow can be made from metals like Ni, Zn,
Ti, Al or Zr that form insulating oxides. The glow has been created
in cells containing a weak electrolyte like Pickling Lime (CaO),
Baking soda, sodium metasilicate, or acetic acid, by gradually
increasing the AC voltage through them typically to 200 - 400 V AC.
Zn can only be conditioned to about 76 volts. As the electrodes are
“conditioned” using high voltage AC they begin to act like opposed
diodes and the cell then acts like a capacitor with a low current
bypass resistor. That the electrodes act like diodes can be
confirmed by replacing one of them with a fresh electrode of the same
metal, or with electrodes made of materials which do not form the
insulating layers, like Pb, Mg or C, and gradually increasing the
voltage to avoid an unexpectedly large DC current.
When the electrode is "conditioned" and the glow is formed, the i vs
V curve looks like Fig 1.
0 volts
|
| /|
/ /
----------/ / _______ 0 amp axis
/ /---------/
/ /
|/
Fig 1. - i vs V curve
The i vs T curve looks like Fig. 2.
----
/ \
/ \
------ \
/ \
/..................\..........................
\ /
\ /
------- /
\ /
\ /
----
Fig. 2 - i vs T for blue glow.
These traces indicate there exists a voltage threshold at which
conduction is suddenly increased. This threshold varies depending on
the conditioning of the electrode and choice of metal, but is
typically between 100 and 400 V. The above trace representations
show operation at a peak voltage about twice the breakdown voltage.
THE ELECTROSPARK EFFECT
The electrospark effect, actual sparking or arcing through the oxide
layer, can be brought about following conditioning by pushing the
voltage way above the threshold value. This electrospark regime,
described below, should not be confused with the glow regime, which
operates in a more subtle way and is less obviously visible. The
glow effect is often visible in the background even when a cell is
operating in the electrospark regime.
THE DIODE EFFECT
The "diode effect” can be seen by replacing a conditioned electrode
with a fresh electrode which is not yet conditioned, or a metal that
does not form a similar layer, like lead, magnesium, or carbon. In
one half of the trace there is close to an ordinary linear ohm's law
relation, while in the other half the conditioned electrode's
rectification and breakdown voltage remains evident. For example see
traces at
<http://home.earthlink.net/~lenyr/borax.htm>.
When operating above the breakdown threshold a blue glow can be seen
in a dark room. If the cell current is DC, the blue glow is only
about the anode. The blue glow intensity is proportional to current
flowing above and beyond, i.e. incremental to, that current which
sustains the threshold voltage.
HYPOTHESIS FOR THE GLOW CAUSE
A good working hypothesis for the observations is that during
conditioning an insulating layer, probably primarily an oxide layer,
grows on the surface of the electrodes to a thickness which prevents
molecular ion flow to or from the electrode. This oxide layer passes
electrons however, either through conduction or through tunneling, a
fact which is easily tested by drying the electrodes and measuring
conductivity through the electrodes. The effect of this molecule
filtering yet conductive layer on an anode is to create an interphase
zone in the electrolyte, bounded by the surface of the anode, across
which a large potential drop occurs. This large potential drop
across a short distance thus creates a colossal electrostatic field
gradient at the surface of the anode. This powerful field gradient
is sufficient, by means described below, to free and accelerate
protons to sufficient energy to ionize the electrolyte in the
interphase and cause recombination reactions that cause the glow.
PASSIVATION
This conducting oxide layer is well known, though the reason it
conducts is not conclusive. [3 ] The process of forming this layer,
earlier called “conditioning”, is formally called passivation.
Passivation of an oxide forming electrode occurs when a sufficient
current density is passed for a sufficient time through the electrode
when it is used as an anode. Either DC or AC can be used, provided
the necessary current density is achieved. If insufficient voltage
is used, then no oxide layer forms, and ordinary anode corrosion
results. After a sufficient passivation time interval and current
density, the oxide layer, though initially insulating, begins to be
conductive. As the layer becomes conductive it begins to absorb
incident photons, i.e. becomes black as opposed to reflective, at
least in the case of Ni anodes. [4] When the oxide layer is
conductive, the electrode is protected from corrosion, and is thus
passivated. The oxide film which forms just prior to this is called
a precursor or prepassive film. Passivation time is reduced by
increasing the passivation voltage and thus current density.
THE CATHODE INTERFACE AND THE ANODE INTERPHASE
The anode interphase layer on a passivated anode creates the diode
effect so clearly visible on i vs V traces. The diode effect comes
from the difference in the high mobility of protons through the
cathode (or anode) interphase layer vs the low mobility of the
relatively big negative ions through the anode interphase layer.
The two molecule thick cathode interface is comparatively well
understood and thoroughly described by Bockris. [5] Protons in the
form of hydronium ions (H3O+) are very mobile in water, and easily
electronated at the cathode interphase by electrons tunneling through
the cathode interphase barrier when having a potential on the order
of a few volts. The addition of a conductive oxide barrier at the
cathode merely increases the potential required to achieve this
tunneling by a few volts.
The anode interphase, however, is dramatically changed by a
conductive oxide layer. Here, in order to sustain ordinary
electrolysis reaction, an electron must be removed from an OH-
radical. However, a powerful gradient forms at the anode and may
be sufficient to strip electrons from neutral molecules given a
sufficient availability of OH- radicals in the vicinity of the
anode. In ordinary electrolysis an OH- radical must be adjacent to
the anode surface in order to achieve an effective probability of
electron removal. In fact the glow process can be expected with
just such a reaction:
OH- ---> OH + e-
However, this reaction immediately removes charges from the
electrolyte near the surface of the anode.
The low mobility of the negative molecular ions to the anode is in
part due to the powerful polar nature of water, which forms a two
molecule thick layer about the anode, the cathode, and also around
any molecular ions themselves. This powerfully bound layer prevents
molecular ions like OH- from easily penetrating them. However, this
layer can only account for a couple eV of the 100-400 volt diode
effect breakdown voltage observed. Further the breakdown voltage
does not act like an actual diode breakdown, except in spark mode,
but rather as a range of quickly increasing current/voltage ratio.
What then can account for the large breakdown voltage? Electrons
should be able to tunnel in either direction across the oxide barrier
with the same probability. The answer may lie in part because the
interphase layer on the anode is much thicker than the two molecule
thick interface on the cathode, and thus the cell voltage is divided
across more molecules. The OH- radical is polarized. The O side is
more negative than the H side. It can form chains at the anode.
Fig. 3 is a representation of one example of such a weakly bound
molecular chain.
(anode+)(OH-)(OH2)(OH-)(OH2)(OH-)...
Fig. 3 Representation of anode interphase molecular chain
Multiple water molecules, OH, HOOH, and other polar molecules
would be expected often in place of the OH2 molecules shown in Fig.
3. Water is deliberately designated in Fig. 3 as OH2 in that the
negative O side of the water would be attracted to the hydrogen side
of the OH-. When the potential drop across the first OH- molecule
is sufficient to strip an electron, we have:
OH- ---> OH + e-
In that the OH molecule is polar, we still have a chain as shown in
Fig. 4
(anode+)(OH)(OH2)(OH-)(OH2)(OH-)...
Fig. 4 Anode interphase molecular chain post-discharge
But the front molecule of the chain has to break free of the chain in
order to have another electron exchange. The next OH- radical, now
at least two molecules away from the anode, must force its way to the
anode for another reaction to occur. The diffusion rate for the big
OH- radical is much lower than for a much smaller proton tunneling
its way through water molecules. The creation of the charge free gap
places a large amount of the cell potential drop across that gap.
Given the high field gradient, we might at times see on the anode
surface:
OH- ---> OH + P + 2e-
or possibly
OH ---> O + P + e-
and the free proton , accelerated by the strong local anode field
gradient, zips its way down the chain, causing ionization, some
hydronium formation, but ultimately:
OH- + p --> H2O
a recombination reaction.
In that the glow does not occur from normal low voltage cathode
reactions, i.e.
OH- ---> OH + e-
it appears a substantial percentage of the anode reactions in the
glow regime create free protons.
The diode forming zone in the electrolyte sustains the majority of
the typical 200-400 V drop across a glowing anode cell. Assuming
a 120 V drop across a 4 angstrom thick zone, the field strength is
about 300 billion volts per meter.
The products of disruption caused by energetic protons, various forms
of hydrogen and oxygen, then ultimately recombine to form water or
oxygen, and in the process of recombination emit the characteristic
blue-green glow of this recombination.
ELECTROSPARK REGIME OPERATION
If the voltage is pushed much beyond the breakdown threshold, bright
spots form and in their early stages flash all over the electrode.
The scope traces develop a hair like fringe that represents the
sudden conductivity characteristic of sparks. This is called the
electrospark regime. See:
<http://www.earthtech.org/experiments/sparkly/report.html>
Typically, the sparking sites erode, destroying the electrode, and
sludge forms in the bottom of the cell. In addition, metal oxidation
occurs which throws off calorimetry calculations significantly when
not take into account. Microscopic examination of the electrospark
hot spots show shiny metal interiors to the holes. The sparks must
penetrate through the anode oxide layer. The initial sparkly
effect, which gives the appearance of the motion of sparks around
the electrode, must be due to the fast reformation of the oxide
barrier after a spark. When the spark sites are deeply eroded,
however, the appearance of motion of the sparks stops, and in the
case of aluminum electrodes, leaves tiny static orange glowing hot
spots.
If a DC cell is operated purely in the blue glow regime, it can
operate indefinitely, or at least a very long time. Further, the
current draw of the cell can be much less than an ordinary
electrolytic cell, though the field strength and particle energies at
the anode interphase are much higher than for ordinary electrolytic
cells. The purely blue glow regime therefore is a much better regime
to investigate for free energy creation than the electrospark regime.
Use of AC guarantees the essential anode coating stays in place. Use
of square wave AC, though nonessential, would maximize the peak-to-
peak voltage that can be used without entering the electrospark
regime. Use of regulated DC is desirable if an anode material is
used that forms an oxide that is not consumed by the process and does
not become thick to enough to stop current flow.
POSSIBLE RELEVANCE TO COLD FUSION
In that free hydrogen nuclei and free electrons are undoubtedly
created in the blue glow regime, it is a candidate for both vacuum
energy creation via atomic expansion, and for cold fusion reactions
via electron catalysis, and thus for tritium creation provided D2O
electrolyte is used. Because the input power is low, an over unity
coefficient of power should be much easier to spot as well. If any
energy or nuclear effect is found, the regime can be made far more
energetic by engineering conductive ceramic or semiconducting layers
capable of avoiding the electrospark regime into very high voltages.
If testing for tritium is desired, a cell design maximizing
(electrode area)/(cell volume) is desirable, in order to maximize the
final tritium concentration. Use of a closed cell would be
desirable. The blue glow should be very effective at ensuring
recombination. It may be possible to run a closed cell provided the
anode is located in a horizontal manner at the top of the cell,
having the back of the anode insulated from the electrolyte. The
objective of this is to route any evolved gasses into the blue glow
for recombination. To run a sealed cell, the cell power supply leads
would have to be insulated. Any exposed electrode surface area would
have to be located under the anode recombiner surface. It may be
advantageous to have the anode shaped in a concave dome form, so as
to create a gas trap.
The pure blue glow regime might be investigated by amateurs without
access to D2O by simply using very long run times. Ordinary water
contains some deuterium. If tritium is produced, it should
accumulate in the electrolyte. Electrolytes from very long runs,
months or years, might be affordably tested by amateurs developing
the right arrangements with labs having scintillation counters, or
simply by the use of film.
Unfortunately, in ordinary water, almost no D2O exists. Almost all
the deuterium is in HDO molecules. The probability of two deuterium
atoms in the same close vicinity is unobservably small. If deuterium
is required for a fusion reaction, we would then expect helium-3 to
be the likely product, and there is no convenient test available to
amateurs for helium-3.
The presence of tritium, if produced from ordinary water, would
likely be due to
D + P + e- ---> T + neutrino
and this would be surprising indeed. The electron catalysis [6] for
helium-3 would be:
D + P + e- ---> 3He + e-
The electron catalysis of deuterium to produce tritium looks like:
D + D + e- ---> He* ---> T + P + 2e-
while a dual electron catalysis[7] might look like:
D + D + 2e- ---> He** ---> He + 2e-
where the He** might end up in a heavy nucleus while in a de-
energized state instead of forming helium.
ELECTRON-HOLE ANNIHILATION HYPOTHESIS
The above conjectures are exciting, but not the simplest or most
credible. A more credible hypothesis is that the glow comes from
electron-hole annihilation at the interface between the electrolyte
and the oxide coating.
The oxide layer that is effective in producing the glow becomes
black, i.e. non-reflective, when it becomes effective[4]. A similar
effect was noted by Steiner when making negative resistance layer on
zinc. [8] A reasonable hypothesis then is that the black
semiconductive oxide layer forms the basis for both negative resistance,
the diode effect, and when combined with a transparent cathode like
an electrolyte,
the glow effect. The oxide layer then is an N-type semiconductor,
conducting via holes. When an electron is stripped from an OH- and
then annihilates with a hole, a photon is produced (possibly two,
with at only one visible.)
One problem with this hypothesis is that the glow color seems
relatively constant, despite a range of electrode materials. The
glow should change, depending on the hole energy. However, the
energy to strip the electron from the OH- is vastly greater, so
possibly this accounts for the approximate color constancy.
Another problem is that, for fixed electrode areas, the glow does
not occur until a threshold voltage is reached, and then is a
function of incremental current (actually incremental current
density). If the oxide layer is N-type, then electron-hole
annihilation should be happening even at voltages below the large
threshold voltage which is roughly on the order of a 100 volts. One
possible answer to this is that the oxide layer is simultaneously
both an N-type and P-type conductor. This would also account for the
ability of the anode to function normally as a proton electronator,
i.e. by conducting electrons.
It is possible that both electron-hole annihilation and energetic
free proton effects are happening in the blue glow creating interphase.
SERENDIPITY
Of possible interest is the idea of using electrolytic passivation to
create negative resistance solid state diodes from metals. Zinc can
be passivated in a weak (tablespoon baking soda per 8 oz water)
electrolyte by gradually increasing the AC voltage to about 76
volts. A much better passivation and higher final voltage would be
expected using DC. Other metals require differing voltages, so can
offer stronger negative resistance effects or effects at higher
voltages and power density. Negative resistance devices are useful
for creating oscillators, and may provide an economical way to
oscillate very large amounts of power by use of large area negative
resistance devices created by applying a conductive metal to the
passivated surface of an oxide forming metal.
REFERENCES
[1] Heffner, “AEH and the Blue Glow”, Feb. 2002,
<http://www.mtaonline.net/~hheffner/BlueAEH.pdf>
[2] Steiner, “Borax or Baking Soda Rectifier and the glow”, Oct 2003,
<http://home.earthlink.net/~lenyr/borax.htm>.
[3] J. O’M Bockris and A.K.N. Reddy, Modern Electrochemistry, Plenum
Press,
1977, p 1315 ff.
[4] ibid., p.1319 ff.
[5] ibid., p. 623 ff.
[6] Heffner,”Electron Catalyzed Fusion”, March 2001,
<http://mtaonline.net/~hheffner/EcatFusion.pdf>
[7] Heffner,”Dual Electron Catalyzed Fusion”, Dec. 2004,
<http://mtaonline.net/~hheffner/DualElectronCatFusion.pdf>.
[8] Steiner, “zinc negatice resistance oscillator”,March, 2001,
<http://home.earthlink.net/~lenyr/zincosc.htm>
