Path: santra!tut!draken!kth!mcvax!uunet!labrea!rutgers!att!shuxd!attdso!galaxia!rayssdb!rayssd!raybed2!linus!munck
From: munck@linus.UUCP (Robert Munck)
Newsgroups: sci.physics,alt.fusion
Subject: Addendum to "Polarized D-/D+ storage in Pd..." (20 Apr 1989)
Message-ID: <51176@linus.UUCP>
Date: 25 Apr 89 11:49:18 GMT
References: <50901@linus.UUCP>
Reply-To: terry@ctc.contel.com (Terry Bollinger)
Organization: Contel, Fairfax, VA
Lines: 245
Xref: santra sci.physics:6351 alt.fusion:660


(posted for someone without net access)

INTRODUCTION

This is a follow-up to my 20 Apr 1989 letter, in which I suggested that success
of the FAP experiment depends on polarization of palladium into two distinct,
physically separated phases (D+ and D-) of palladium deuteride.  The earlier
letter also made several corollary suggestions on the what the best physical
configurations for FAP reaction systems might be, based on the assumption
that the D+/D- phase separation hypothesis was valid.

In this letter, I discuss the relationship of the proposed D+/D- phases to
the known alpha and beta phases of palladium, describe a few more corollary
suggestions for constructing FAP reaction systems (primarily metallurgical and
chemical in nature), and make a few suggestions as to how a few non-
electrolytic, FAP-like experiments might be performed.  Such experiments might
be useful for eliminating the possibility of chemical battery effects as the
cause of the FAP exothermic reaction.


BACKGROUND

The key hypothesis of my last note (entitled "Polarized D-/D+ storage in Pd
as a factor in the F&P experiment," and dated 20 Apr 1989) was as follows:

Hypothesis -- The critical feature of the F&P reaction is the formation
              of distinct physical regions of D- (palladium deuteride)
              and D+ (mobile deuterons in Pd) under the influence of a
              strong electrical current.  The F&P exothermic reaction
              occurs when conditions permit highly mobile D+ ions to
              recombine or fuse with comparatively immobile D- ions.

Following this hypothesis, two mechanisms for explaining the FAP reaction
were given:  a "chemical battery" interpretation in which a possible energy
density of up to 50.6 kjoules/cm3 of palladium was calculated based on full
separation of D+ and D- phases, and a fusion interpretation in which the
energy provided by D+/D- recombination within a Pd lattice somehow resulted
in fusion of the two deuterons.  The rather interesting physical implications
of a polarized Pd model on building FAP reaction systems were then discussed.


RELATIONSHIP OF D+/D- PHASES TO PALLADIUM ALPHA/BETA PHASES

The alpha and beta phases of palladium, which are mentioned briefly in the
Fleischmann and Pons paper, are two distinct solution phases of hydrogen (or
deuterium) in palladium.  Both phases can exist at room temperature, but the
distinction between them disappears when the palladium is heated to about
300 degrees C.

Although the D+/D- phase argument of the previous letter was constructed in
terms of the net effects of the migration of deuterium ions in palladium under
the influence of a strong current, it is reasonable to surmise that if D+/D-
phases do exist, they could be the same as the alpha and beta solution phases
of hydrogen in palladium.  The Fleischmann and Pons paper seems to suggest
that the beta phase for deuterium is D+ saturated, which would imply that
beta palladium is equivalent to what I have been calling the D+ phase.

Perhaps I am missing some key piece of information, but to me the assertion
that beta Pd is D+ saturated doesn't sound right, particularly when it is then
followed by the Fleischmann and Pons assertion that "...the H+ and D+ in the
lattice behave...possibly as delocalized species...in very shallow potential
wells."  If D+ is delocalized in beta Pd, then that phase should be *more*
metallic in its overall properties -- instead, exactly the opposite is true.
Beta Pd is *less* conductive than alpha Pd, has lower magnetic susceptibility,
and is physically more brittle.  Also, the localization of D to the octahedral
sites between Pd atoms to me sounds closer to the behavior of a conventional
anion than to the behavior of a delocalized "D+ conduction band" deuteron.
The evidence that Fleischmann and Pons quote in their paper for the D+
interpretation ("...hydrogen is in the form of protons...as shown by the
migration in an electric field...") seems rather weak to me, at least as it is
stated in the paper, since it says nothing about the possibility that the H+ or
D+ ions may simply be minority current carriers in a lattice that is otherwise
dominated by D- ions.  All in all, then, I would say that a reasonable working
assumption would be that the beta phase is equivalent to the proposed D- phase,
rather than to the deuteron-rich D+ phase.

(Incidentally, occupation of the beta Pd octahedral sites by D- (vs. D+) ions
would be a significant issue for the fusion debate, since a fair bit of the
public theorizing about the FAP reaction appears to be based on the assumption
that there are *lots* of free deuterons competing for space in the octahedral
sites of beta Pd.  I strongly suspect that D- ions with their nice, fat
electron clouds around them would not be nearly as amenable to density-oriented
fusion arguments as would be naked deuterons.)

In contrast to the poor fit of beta Pd to the proposed D+ phase, the alpha
phase appears to be a fairly good candidate for the D+ phase.  Particularly
interesting is the lack of physical expansion of palladium when the alpha
phase is formed.  (The lattice constant increases by about 4% during the
formation of the beta phase.)  This lack of expansion could be interpreted
(quite loosely!) as an indication that at least in the alpha phase, deuterium
really is predominantly in the form of "delocalized" D+ ions.  Even so, it is
still possible that the alpha phase is not a "true" D+ phase, but rather a
mixture of D+ and D- ions in which D+ and D- are comparable in density.

The bottom line:  There are two known phases of palladium hydride which
correspond reasonably well to the D+ and D- phases postulated in my last
note.  The most likely correspondence of the proposed phases to the known
phases is:

     D+ phase = alpha palladium hydride (?)

     D- phase = beta palladium hydride


METALLURGICAL IMPLICATIONS OF THE D+/D- PHASE MODEL

Moving on to additional experimental implications of the D+/D- polarized phase
model, the metallurgical implications of the model can be summarized quite
succinctly in the following rule:

     The number of grain boundaries crossed by electrical "lines of force"
     in a FAP reactor should be minimized, ideally to zero.

The rational is simply that every grain boundary provides an impediment to the
flow of D+ ions, and thus to the overall level of polarization of the system.
One obvious implication of this rule is simply that FAP palladium components
should be cast directly from molten palladium, and *not* created by powder
metallurgy or extensive machining.

For a cast part, the way in which it is cooled could also have a significant
impact on its effectiveness.  This leads to the following rule-of-thumb:

     During the casting of a FAP palladium component, the thermal gradient
     should be maintained so as to have the same general form as the electrical
     gradient (or its inverse) that the part will be subjected to during use.

The point of this rule, of course, is to try to encourage crystal growth to
follow the same general path that migrating deuterium ions will follow.

For the spherical reactor suggested in the previous note, this would simply
imply that the palladium sphere should be a cast part, and that cooling of
the casting should be as isotropic as possible.


CHEMICAL IMPLICATIONS OF THE D+/D- PHASE MODEL

The D+/D- phase model depends critically on the assumption that two out-of-
equilibrium, physically separated phases of palladium deuteride can be created
and then maintained for extended periods of time (that is, at least for many
seconds, and presumably for much longer).  This leads to some very interesting
observations about the potential chemical fragility of FAP reaction systems,
since both of these out-of-equilibrium phase systems would contain highly
reactive ions (D+ and D-) which might readily react with contaminants.

In particular, it is likely that many contaminants occurring both within the
Pd lattice and on the Pd surface would act as positive catalysts that would
accelerate the return of one or both of the two phase systems to the
equilibrium state.  The D+ phase would probably be particularly sensitive to
such equilibrium acceleration catalysts, since the greater mobility of the D+
ions would allow them to interact quickly with even a very small percentage of
contaminant atoms or molecules.

On the more optimistic side, there might also exist negative catalysts
(inhibitors) whose effect would be to preserve or extend the average life
of D+ or D- ions.  If such inhibitors exist, then doping of the palladium
lattice and/or surface with them could greatly increase the effectiveness of
a FAP reaction system by allowing a larger percentage of the physically
separated D+ and D- ions of the D+ and D- phases to recombine.

Identifying D+/D- phase accelerators and inhibitors would only be possible
through direct experimentation, although a few general observations can be
made.  Elements from groups 1, 2, 6 and 7 would all be worth testing because
of their high oxidation/reduction potentials.  The effects of alkali metals
such as cesium (most active Group 1 element) and lithium (the least active)
could be particularly interesting to observe, due to their chemical similarity
to hydrogen.  Experimentally, the contrast of the Fleischmann and Pons
experiment (in which only lithium was used in the electrolyte) and the Jones
and Palmer experiment (in which they apparently decided to dissolve a kitchen
sink in acid to make their electrolyte;  nothing like adding ten or so more
variables to an already complex problem...) may indicate that lithium acts as
an inhibitor for one or both of the D+/D- phases, while transition elements
(the Jones result) as a whole tend to act as equilibrium accelerators.  In
the case of lithium, it might be worth directly alloying small quantities of
lithium with the palladium to test for volume (vs. surface) effects.

To me it seems a bit remarkable that there has been little or no mention of
the possible effects of oxygen on the palladium electrolysis experiments.
Given that palladium, like platinum, is a good absorber of oxygen, I would
have expected a bit more discussion of the possible effects of good old O2.
In terms of the D+/D- phase model, oxygen could well turn out to be an
important equilibrium inhibitor or accelerator.

In a similar vein, the cathode material that is used to interface with the
palladium may need to be viewed as an integral component of the FAP system.
Selection of a D+ equilibrium accelerator (Copper?  Silver?  Gold?  Is your
electrical clamp keeping you out of the Great Fusion Race?) for the cathode
material might lead to a vanishingly small equilibrium level of D+ ions.
(Sort of like calling people in for dinner and then shooting them at the door.)
 
It should be noted that a contaminant that is an inhibitor for one phase of
a D+/D- system could very well be an accelerator for the other phase; thus, the
possibility of "diode" style reactors in which the cathode/Pd region is doped
with a different inhibitor from the D20/Pd "anode" region could be worth
testing, providing that such inhibitors exist and can be identified.

Overall, the best advice for building FAP reactors without knowing the impact
of contaminants is simply to keep the materials (particularly the palladium
and the electrolyte) as high in purity as possible, and to eliminate
unnecessary variables whenever possible.  Also, an one should keep an eye
on the possibility that seemingly innocuous differences in the environment
of the experiment (the cathode example again) could be drastically altering
the results.


NON-AQUEOUS D+/D- MODEL FUSION EXPERIMENTS

A few last notes on the possibility of fusion in the D+/D- phase polarization
model.  If there is in fact something unique about the recombination of D+
and D- (17.45eV released energy;  not exactly big potatoes by nuclear fusion
standards), then it is at least possible that D+/D- fusion reaction could
occur in non-aqueous environments.  Some possible experiments that come to
mind include:

  1) Direct plasma recombination of D+ and D- at low energies.  (Probably
     the lower the better;  if direct recombination can cause fusion, it
     would have to be accomplished via some extraordinary form of precision,
     rather than brute force.  [Chances for this one would seem low;  also,
     someone surely has tried this before (?).]

  2) Same as (1), but insert a thin palladium plate at the reaction point.
     Titanium or tantalum might also be worth trying -- and would be quite
     a bit less expensive, also.  [I would rate this experiment as having
     fair odds of producing something interesting, provided you can avoid
     vaporizing your palladium plate via mundane plasma heating.  Calorimetric
     measurements could also be quite interesting in such an environment.]

  3) Aim a low-energy D+ plasma directly at a lithium deuteride cleavage face.
     (Or try cesium deuteride if the above configuration is a bit too
     reminiscent of an H-bomb to you.)  [This one might be interesting.]

  4) Same as number (3), but plate the cleavage face with a thin layer of
     palladium, titanium, or tantalum.  [This one might also be interesting.]


                                 Cheers,
                                 Terry Bollinger  (terry@ctc.contel.com)


             "In fact, it has been estimated that [the Soviet Union]
              now provides about 80% of the world's annual supply of
              palladium."

                  -- W. A. E. McBryde
                     The Encyclopedia of the Chemical Elements