Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

The Difference Between an Alloy and an Amalgam - It's Thermoelectric!

7 views
Skip to first unread message

Keith P Walsh

unread,
Jan 28, 2009, 4:34:55 AM1/28/09
to

The following statement was made by Professor Xu Wang of the
University of Akron in response to a question regarding the
relationship between the thermoelectric properties and the
electromagnetic behavior of metal amalgam dental fillings.

"Theoretically electromagnetic field will be generated by thermal
gradient. But the thermoelectric coupling parameter in most metals is
very low (0.001-0.01). So I don't think the induced electromagnetic
field is significant enough to influence the neurological tissue
nearby."

Amalgams, including dental amalgams, are not like most other metals in
at least one crucial respect; they have a much greater degree of
material inhomogeneity.

This is true when compared either with pure metals, such as copper,
silver, etc., or with true alloys such as brass.

The explanation for the difference in the material homogeneities of
amalgams and true alloys lies in the difference between the methods by
which the two types of material are formed.

When a true alloy is formed, the component metals are mixed together
at a temperature which is greater than the melting point of all of
them. Then, after having been mixed thoroughly in its fully liquid
state, the mixture is allowed to solidify by cooling at a controlled
rate.

By contrast, in an amalgamation process, bits of solid metal, which
may themselves be of either pure metal or an alloy, are mixed together
with a liquid metal at a temperature which is BELOW THE MELTING POINT
of the solid component(s). (And in the case of dental amalgam, where
mercury is used as the liquid metal amalgamating agent, this process
is normally performed at room temperature.)

In the setting process of such an amalgam, the liquid mercury becomes
part of the solid material not as a result of any subsequent reduction
in temperature, but by joining in solid solution with the outer layer
of the solid particles of metal with which it was mixed. But of
course, not all of the volume of the solid particles is involved in
this process and, as a result, the microstructure of the resulting
solid amalgam is as depicted in the schematic diagram at:

http://book.boot.users.btopenworld.com/setting.htm

In this diagram the lumps of "unreacted alloy" (denoted "gamma") are
the cores of the original grains of solid silver-tin alloy which have
not mixed with any of the mercury during the amalgamation process.

At this scale it is not possible to show the spatial relationship
between the atoms of silver and the atoms of tin in these alloy
"cores". The alloy has too great a degree of homogeneity for this to
be done.

However, the relative inhomogeneity of the "amalgam" is clearly
depicted by the sizes of the unreacted alloy cores. (These being held
together by a solid matrix of a dissimilar mixture of metals (denoted
"gamma-1") which does have mercury in it, and which may be presumed
therefore to have dissimilar physical properies.)

Now, I have it on good authority from Professor David B Mahler of The
Oregon Health & Science University School of Dentistry that the median
size of the "unreacted" grains of original solid alloy in dental
amalgams is in the order of 30 microns. Scientists with experience of
electrical phenomena at the nano scale might provide some testimony to
the significance of this figure.

The question which remains unanswered is this; whilst it may be
appropriate to quote a "coupling parameter" for the local
electromagnetic effect arising purely from temperature difference in a
homogeneous metallic material, such as a pure metal or an alloy of
metals (for example the type of alloy which may be mixed with liquid
mercury to form an amalgam), is it not the case that the dominant (and
potentially much larger) electromagnetic effect arising as a result of
temperature differences in a more inhomogeneous mixture of dissimilar
metals, such as an amalgam, is more likely to be that caused by the
establishment of thermoelectric eddy currents which would be necessary
for maintaining physical equilibrium against temperature gradient in
such an inhomogeneous medium?

Can anyone think of any experimental procedure which might be employed
in order to demonstrate the case either way?

And is it possible that Professor Wang was failing to take into
account the degree of inhomogeneity of metal amalgams, which is much
greater than "most metals", when estimating the size of the
electromegnetic disturbance produced by thermoelectric effects in
dental amalgams?

Professor Wang's home page is at:

http://www.ecgf.uakron.edu/~pan/wang/

Keith Walsh

PS, Any dentists out there who don't know anything about the
thermoelectric behavior of dental amalgams can always quote professor
Wang's "coupling parameter" statement if pressed. But I would strongly
advise you to make sure that he knows what he's talking about first.

Paul O

unread,
Jan 28, 2009, 12:46:46 PM1/28/09
to
Keith P Walsh wrote, On 1/28/2009 4:34 AM:
> <snip material we've seen a hundred times before>

>
> Can anyone think of any experimental procedure which might be employed
> in order to demonstrate the case either way?
>
> <snip>
>
Keith,
Yes, I can think of an experimental procedure. Several people on these
newsgroups (myself included) have described to you exactly how to
measure the thermoelectric properties of metal amalgam dental fillings.
But it has been a complete waste of time. In the last 15+ years that
I've been reading your posts, you have never once made even the
slightest effort to test your hypothesis.

Look, this is not that difficult. I will help you if you are willing
purchase some materials, borrow or rent a good digital volt meter,
construct a test apparatus, and then run your experiment. Otherwise,
feel free to continue shouting in the wilderness.

--

Paul D Oosterhout
I work for SAIC (but I don't speak for SAIC)

ji...@specsol.spam.sux.com

unread,
Jan 28, 2009, 1:15:01 PM1/28/09
to

Hell, a couple of years ago I found what he needed on e-bay for well
under $100 and offered to buy it for him if he made and published
the measurements.


--
Jim Pennino

Remove .spam.sux to reply.

Keith P Walsh

unread,
Jan 30, 2009, 3:41:38 AM1/30/09
to


I am proposing the hypothesis that the degree of electromagnetic
disturbance caused by the application of a temperature gradient to an
amalgam should be greater than that caused by the application of a
temperature gradient to either a pure metal or a true alloy.

The justification for this proposal is based on the fact that an
amalgam has a much greater degree of material inhomogeneity than
either a pure metal or an alloy, and it should therefore be expected
to generate internal thermoelectric eddy currents to a much greater
extent when subjected to thermal gradients.

Now, Professor Wang of the University of Akron has asserted that the
relevant "coupling parameter", which gives an indication of the size
of the electromagnetic disturbance generated by a material with
respect to the size of any temperature differential applied to it,
would be too low in "most metals" to suggest that any local
electromagnetic effect produced could have an influence on
neurological tissue in the direct vicinity of the material.

However, as I have shown (and I note that in your own contribution you
have not registered any disagreement with this point) amalgams
(including dental amalgams) are different from "most metals" in the
crucial respect that they have a much greater degree of material
inhomogeneity.

So, the question is, how can we be sure that amalgams (and most
importantly dental amalgams) come within the "safe" range defined by
Professor Wang with regard to electromagnetic activity?

Well there's only one scientific way to do that and that is to carry
out experimental investigations to see if the electromagnetic
disturbances generated by amalgams as a result of their thermoelectric
behavior can be measured.

And if they can be measured then it would also be necessary to carry
out further experimental investigations to determine whether or not
the measured disturbances are able to influence the function of any
neurological tissue nearby.

According to the established principles of scientific understanding,
without any of the experimental investigations here described having
been carried out, it is not possible for any of us (including
Professor Wang) to conclude whether or not the natural electromagnetic
behavior of an amalgam dental filling is capable of dissipating
electrical energy through the nerves in people's heads.

And it is therefore not possile to declare amalgam fillings "safe" in
this respect.

I do not advocate the use of metal amalgams in restorative dentistry.

Professional bodies such as the American Dental Association, the
British Dental Association, the US FDA, etc., they do that. And
therefore the responsibility for carrying out the experimental
procedures to demonstrate whether or not such materials are "safe" for
this purpose lies with them, not with me.

Now I appreciate that you have succeeded in convincing yourself that
this responsibility does lie with me.

But you are wrong, it doesn't.

Keith P Walsh

Paul O

unread,
Jan 30, 2009, 11:27:30 AM1/30/09
to
Mr Walsh,
Why would you trust the results from any experiments carried out by the
ADA, the BDA, or the FDA? Do the experiment yourself and then you will
know for sure whether your hypothesis is correct.

If your experiment has any sort of positive result, then other
researchers will be come interested in the phenomenon and begin their
own investigations. Then the ADA and the BDA will be forced to take notice!

You could be a hero! All you have to do is a little work.

Best of luck
Paul O.

Keith P Walsh

unread,
Jan 31, 2009, 3:30:24 AM1/31/09
to
On 30 Jan, 16:27, Paul O <first.d.l...@company.com> wrote:

>
> You could be a hero! All you have to do is a little work.
>

I'll bear it in mind.

In the meantime can I take it that you agree with my description of
the difference between an alloy and an amalgam?

Keith P Walsh

Keith P Walsh

unread,
Feb 2, 2009, 9:32:46 AM2/2/09
to
On 28 Jan, 19:41, alchazz <no.s...@wanted.com> wrote:

>
> I think that by definition, an amalgam has to have mercury in it.
>

That's a common misconception.

That "an amalgam" has come to imply mercury, particularly in the
dental profession, is because mercury amalgams are the most common.

But mercury amalgams are NOT the only type of amalgam, as you can see
by extracts from the book, "Principles of Soldering", by Giles
Humpston and David M Jacobson, published by ASM International (April
2004) and endorsed by The Materials Information Society.

The parts which are of interest to us are:

Page 115 - Section 5.4.1 - Amalgams Based on Mercury

Page 116 - Section 5.4.2 - Amalgams Based on Gallium

"Gallium melts at 29 deg C and is therefore a potential base for
formulating very low-process-temperature amalgams without the toxic
hazard associated with mercury"

(Note, there's no mercury in these amalgams.)

Page 117 - Section 5.4.3 - Amalgams Based on Indium

"Indium is another liquid metal that can be considered as a base for
amalgam systems. "

(There's no mercury in these amalgams either.)

You can find details of this book at:

http://books.google.com/books?id=cQ6khQScBF4C&pg=PA103&dq=%22gallium+amalgam%22&sig=Um_AAj4cKineMZPCpE4sXOWq3Uw#PPP1,M1

When I first quoted these pages (see message in thread "Lessons on
Alloys, Amalgams, and Pleonasms (2)" of 1 September 2007)
they were available for reading at the above URL.

Unfortunately it looks like since then they've been made unavailable!
(I tell ya, if I were a conspiracy theorist ... ).

Anyway the quotes are genuine.

So come on Al. Please don't simply ignore this post just because
you've been proved wrong.

Get typing and give us your opinion now (and don't be shy about cross-
posting to the other groups).

Keith P Walsh

Paul O

unread,
Feb 2, 2009, 9:45:51 AM2/2/09
to
Yes I do. But It do not agree that you will find any significant
thermoelectric effect across an thermal gradient in an amalgam. Do the
experiment and find out for yourself.

Keith P Walsh

unread,
Feb 2, 2009, 10:27:38 AM2/2/09
to
On 28 Jan, 19:41, alchazz <no.s...@wanted.com> wrote:

> I think that by definition, an amalgam has to have mercury in it.
>

> Al

Sorry to labor the point Al, but I found my other favorite reference
for gallium amalgams (i.e. where the liquid metal used to formulate
the amalgam is gallium and not mercury - so there's no mercury in
'em).

This is from the Institute of Physics document service website at:

http://www.priorartdatabase.com/IPCOM/000137544/

"Hermetic Sealing of Electronic Hardware Using Gallium Amalgams

In this proposed invention, the gallium amalgam is mixed in a ball
mill, but the mixing action is minimized to limit the abrasion of the
metal particles so as not to have the liquid gallium wet the metal
particles. Since the metal particles are not wetted by the liquid
gallium, they will not dissolve in the liquid gallium. The pot life is
thus much extended. The liquid gallium, laden with metal particles, is
placed on the surfaces to be joined. The faying surfaces are pressed
together and then vibrated with respect to each other. The compressive
stresses and relative motion between the surfaces cause the metal
particles to abrade against each other and the faying surfaces. The
particles and the faying surfaces become clean and thus wettable by
the liquid gallium. The amalgamation process begins solidifying the
amalgam and joining the two surfaces. When the amalgamation process is
complete, the amalgam becomes solid and holds the faying surfaces
together.

The amalgam can be a ternary alloy having 5-35% Cu and 15-30% Ni in
liquid gallium. The nickel and copper particle size has to be less
than about 0.05 mm so that the faying surfaces are not kept
excessively apart by the particles. The relative motion between the
faying surfaces to initiate the amalgamation action can be achieved
by..."

So you see there really are amalgams with no mercury in them - but it
is still correct to call them amalgams because of the nature of the
process by which they are formed.

But like I said, yours is a common misconception, so there's no need
to feel stupid.

On the other hand, if you begin to wonder whether the idea that
"there's nothing wrong metal amalgam dental fillings" might also be a
common misconception let me know and I'll give you some pointers on
how you might find out.

Keith P Walsh

Olin Perry Norton

unread,
Feb 2, 2009, 4:12:41 PM2/2/09
to

It seems to me that we go for several months without
hearing from Mr. Walsh, and then he resurfaces with
a post of basically the same stuff he posted the previous
time around.

Has anyone determined the period of the Walsh function?
Is it correlated with the phase of the moon, perhaps?

At any rate, during a previous iteration, Mr. Walsh
gave us a reference to a paper on the eddy currents induced
inside a metallic conductor which contains an inclusion of
a different metal and is subjected to a temperature gradient.

Let's see, when was that ... ah yes, it was back in October 2006.

In the paper Mr. Walsh referred us to then, the authors considered
a copper-constantan system. If one had a block of copper
with a constantan inclusion, and if that block were subjected to
a temperature gradient, then there would be a steady, DC eddy
current flowing inside the block of metal, due to the thermoelectric
effect and the fact that one end of the inclusion would be at a
different temperature than the other end due to the temperature
gradient.

This would presumably occur inside any conductor which was not
of homogeneous composition (i.e., a dental amalgam) and was
subjected to a temperature gradient.

I pointed out at the time that the external manifestation of
this eddy current was a very, very small (pT, that's
right -- PICO-Tesla -- PICO, as in 10^-12)
steady magnetic field. This result was stated in the very
paper Mr. Walsh referred us to, but somehow he seemed to
have missed it in his reading. This is an extremely
tiny magnetic field, many orders of magnitude less than
the earth's magnetic field, and it is hard to see how it would
have any effect whatsoever on the human body.

In fact, the authors of Mr. Walsh's reference stated that
"only the most sensitive magnetometers have a significant
chance to detect the expected magnetic field values at a
proper signal to noise ratio." This magnetic field is so
tiny that you would need a SQUID detector to have any chance
of even measuring it.

So, how could this extremely weak DC magnetic field, many
orders of magnitude less than the already existing DC
magnetic field of the earth, possibly have any effect on the
human body?

I seem to recall that Mr. Walsh ignored this question back in
2006.

Olin Perry Norton

Androcles

unread,
Feb 2, 2009, 6:58:06 PM2/2/09
to

"Olin Perry Norton" <mylas...@icet.msstate.invalid> wrote in message
news:gm7nk8$eol$1...@nntp.msstate.edu...


By analogy, it would be hard for you to "see" how a pebble could
topple a boy on a skateboard (any effect whatsoever) when a boulder
or brick wall would stop him dead. Perhaps you do not want to see.

Keith P Walsh

unread,
Feb 3, 2009, 4:50:57 AM2/3/09
to
On 2 Feb, 21:12, Olin Perry Norton <mylastn...@icet.msstate.invalid>
wrote:

Mr Norton.

Thank you for your contribution.

In reply, first of all, can I take it that you agree with my
description of the difference between an amalgam and an alloy (which
is not dependent upon whether or not the thermoelectric eddy currents
generated by the amalgams are detectable by human neurological
tissue.)? And also, for the benefit of my friend alchazz, that
amalgams may be formed using liquid gallium as the amalgamating agent
rather than liquid mercury, with the consequence that there is no
mercury in the resulting gallium amalgams?


>
>     So, how could this extremely weak DC magnetic field, many
>     orders of magnitude less than the already existing DC
>     magnetic field of the earth, possibly have any effect on the
>     human body?
>
>     I seem to recall that Mr. Walsh ignored this question back in
>     2006.
>


OK then, I'll address it now.

There is a basic inconsistency in the argument that you have presented
here (OK I know, you've really only presented a question).

The inconsistency which you imply in your question is that, although
the SQUID detector is sensitive enough to detect a very weak local
electromagnetic effect in the presence of the Earth's magnetic field,
which may be considered much stronger, human neurological tissue is
not sensitive enough to do the same.

I do not believe that you actually know whether or not this apparent
implication is correct. At least, I do not believe that you are able
to provide any direct scientific (i.e. experimental) evidence in
support of any such assertion.

And there in lies the rub. It appears that, in this "science obsessed"
world of ours, no-one has ever bothered to even try to find out (by
experiment) whether or not human neurological tissue is capable of
detecting the electromagnetic disturbances generated by the lumps of
inhomogeneous mixtures of metals that dentists routinely place in
children's teeth.

So a word of warning, if you want to set yourself up as some sort of
expert on the electromagnetic behavior of amalgam dental fillings, you
might first wish to consider why there doesn't appear to be anyone in
that role ahead of you (just think, with a little work you could be a
hero! And I'm sure that the American Dental Association would be most
generous in its gratitude for your expertise - but perhaps only if you
are able to tell them what they want to hear.)

My answer to your question is this. Perhaps human neurological tissue
detects weak local electromagnetic disturbances, even in the presence
of much stronger magnetic fields, in a similar way to that in which
the SQUID detector does. In the absence of any experimantal evidence
to the contrary why shouldn't it? Perhaps that's a question which you
might like to address (although I understand perfectly well that in
the absence of any relevant experimental evidence at all you'd only be
able to guess your answer anyway).

Electric, magnetic and electromagnetic fields are all manifestations
of the fundamental energy that makes up the Universe. They are not
characterised by field strength alone. Their nature is much more
complex than that. In the modern technique of nuclear resonance
imaging, the strength of the signal is largely irrelevant with regard
to the degree by which it is attenuated by the target. It's the
frquency/wavelength of the signal that matters most crucially, and if
it doesn't match the resonant frequency/wavelength of the target then
the attenuation doesn't happen, no matter how strong the signal is.

That's just an example to establish the principle that in certain
circumstances a weak signal may have a significant effect, whereas a
much stronger signal which has different characteristics might have no
effect at all in otherwise identical circumstances (a principle which
I believe that "Androcles" may have been alluding to in his response
to your contribution - thank's for the prompt Androcles).

If you would prefer to believe that there isn't any analogous
principle which might apply in our topic of interest well fine, go
ahead, but as I say, in the absence of any authoritative scientific
evidence on the matter, you're still only guessing.

There was another thread concerning this subject which I left
uncompleted a few years ago. I'd asked how one would know that the
process by which the local electromagnetic field generated by
thermoelectric eddy currents is not reversible, and I was presented
with the answer, "Elementry (sic) physics" - I'll just go and find it
(isn't Google Groups wonderful. In centuries to come they'll look back
on these arguments and laugh - or cry.)

Found it. Jim Pennino, 24 October 2006, in thread, "Electromagnetic
Effects in Inhomogeneous Materials", the relevant part of which went
like this:

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+
> How does this assertion have any bearing of the question as to whether
> or not this same effect is reversible?
> Are you saying that it would not be possible to generate a "magnetic"
> field of the appropriate characteristics via an alternative external
> source?


A static magnetic field will not generate a current.


> And if so, how do you know that it wouldn't?


Elementry physics.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++


I'll start this one back up again now as well.

I agree that the influence of a static magnetic field is not able to
drive an electric current in a homogeneous electrical conductor. That
much is "elementary physics". However, having said that, at the moment
of establishing the new local magnetic field in the vicinity of the
conductor the charge carriers (electrons) in the conductor will be
disturbed momentarily, and may adopt a new static arrangement in order
to maintain equilibrium with the newly established local field,
although no further movement of the electrons will take place once
this static equilibrium condition is reached.

This situation is analogous to the effect of the application of a
temperature differential to a homogeneos electrical conductor, in that
on the establishment of the temperature differential the charge
carriers in it must adjust their position (thereby creating a
momentary "current" - i.e. displacement of charge) in order to
maintain equilibrium with the new temperature gradient. And that on
reaching the new equilibrium position, no further displacement of of
charge (current) can be induced or maintained by the static
temperature differential.

However, where inhomogeneous conductors are involved, such as a
circuit comprising elements of, say, two dissimilar metals, then a
static temperature differential applied between the contact points of
the two dissimilar metals DOES induce an electric current in the
circuit, and this current continues to flow for as long as the static
temperature differential is maintained. And of course the equivalent
of this phenomenon when applied to an inhomogeneous material such as a
dental amalgam is that the static temperature differential should be
expected to induce and sustain "thermeoelectric" eddy currents around
the inclusions of dissimilar metal which are internal to the material
itself.

I don't know whether this stuff is included under the description
"elementary physics" or not. Though it does appear that dental
students in dental schools are never taught it. Nevertheless, it makes
me wonder whether or not the reverse analogy is also true for static
magnetic fields when applied to inhomogeneous mixtures of metals, and
that these too might be able to induce eddy currents in dental
amalgams.

Food for thought.

If only we had some experimental evidence which might settle the issue
either way.

Whatever the case, I'd advise anyone who might be feeling the urge to
step into the role of "expert on the electromagnetic behavior of
dental amalgams" at this point to be very careful.

Because I for one can feel the rumblings of a next question which
might be on the lines of, "and are you presuming that static
electromagnetic fields are the only kind that metal dental
restorations are subjected to?".

"Elementry physics" my eye!

Keith P Walsh


Mark & Steven Bornfeld

unread,
Feb 3, 2009, 9:33:20 AM2/3/09
to
Keith P Walsh wrote:
> On 2 Feb, 21:12, Olin Perry Norton <mylastn...@icet.msstate.invalid>
> wrote:
>
> Mr Norton.
>
> Thank you for your contribution.
>
> In reply, first of all, can I take it that you agree with my
> description of the difference between an amalgam and an alloy (which
> is not dependent upon whether or not the thermoelectric eddy currents
> generated by the amalgams are detectable by human neurological
> tissue.)? And also, for the benefit of my friend alchazz, that
> amalgams may be formed using liquid gallium as the amalgamating agent
> rather than liquid mercury, with the consequence that there is no
> mercury in the resulting gallium amalgams?


I will accept your use of the term "gallium amalgam" though it is
seldom used and hence can be misinterpreted.
In fact, dental restoratives have been made using gallium as a major
component; unfortunately their performance has been disappointing:

http://www.cda-adc.ca/jcda/vol-64/issue-9/645.html

Steve

--
Mark & Steven Bornfeld DDS
http://www.dentaltwins.com
Brooklyn, NY
718-258-5001

Keith P Walsh

unread,
Feb 4, 2009, 3:47:33 AM2/4/09
to
On 3 Feb, 14:33, Mark & Steven Bornfeld <bornfeldm...@dentaltwins.com>
wrote:

>
>         I will accept your use of the term "gallium amalgam" though it is
> seldom used and hence can be misinterpreted.
>         In fact, dental restoratives have been made using gallium as a major
> component; unfortunately their performance has been disappointing:
>
> http://www.cda-adc.ca/jcda/vol-64/issue-9/645.html
>
> Steve
>

Dear Steve,

I think that the main reason why the term "gallium amalgam" might be
misinterpreted is because of the widespread misconception that, by
definition, an "amalgam" must have mercury in it.

This also explains why the the types of gallium amalgams that we are
talking about here, which do not have any mercury in them, are
sometimes erroneously referred to as "gallium alloys", e.g.:

http://www.cda-adc.ca/jcda/vol-64/issue-9/645.html

Of course, it may be argued that, whilst establishing a consistent
nomenclature is important in any scientific discipline (including that
of dental materials science), the names that we actually give to
physical entities is perhaps of secondary importance to the correct
identification of their physical properties and behaviors.

However, I do think that in this instance it might be helpful, as well
as perfectly correct, for us to establish that the true origin of the
word "amalgam", as applied to certain types of mixtures of metals, is
NOT dependent upon there being mercury included in the mix, but is in
fact defined by the general process by which they are formed; i.e.
that of mixing a quantity of a liquid metal "amalgamating agent"
together with solid particles of another metal at a temperature which
is below the melting point of the solid metal, and then allowing the
mixture to harden by a process which might reasonably be termed
"amalgamation".

The metallic element gallium melts at around 30 degrees celsius. It
can therefore be used in amalgamation processes such as this to form
amalgams by mixing it together with bits of solid metals or alloys at
a temperature slightly above that of room temperature (say, below 40
degrees celsius). The type of material that you get from such a
process could not be accurately described as an alloy (or what I
sometimes call a "true alloy) because a significant part of its volume
would remain as the "unreacted", or "undissolved" cores of the
original grains of solid alloy used in the mix, and these would not
have any gallium in them at all.

So it wouldn't be "an alloy". It would be "bits of an alloy all held
together in a matrix of a dissimilar metallic composition formed by
the liquid gallium in dissolving the outer layers only of the original
solid particles". Or, more generally, an inhomogeneous mixture of
dissimilar mixtures of metals.

It is from the requirement to distinguish materials of this type that
the term "amalgam" originates.

Primarily such materials are distinguished in the physical sense by
their much greater degree of material inhomogeneity, as compared with
that of true alloys, and this distinction comes from the complementary
fact that true alloys have a much greater degree of material
homogeneity due to their being formed by mixing together the
constituent metals at a sufficiently high temperature such that they
are ALL in the liquid state.

By this definition, therefore, a gallium amalgam may be considered no
more a true alloy than a mercury amalgam is.

I am already aware that attempts are been made to formulate suitable
gallium amalgams for use as restorative dental materials.

However, as far as I can tell the principal motivation for this is to
find a viable alternative to the use of mercury amalgams.

And, again to the best of my knowledge, the reason for this is concern
for the possible toxic effects that the mercury in mercury amalgams
may have on the human body.

Since I know that you have read, and often responded to, many of my
enquiries to sci.med.dentistry over the years, I presume that you will
be aware that my own particular interest in this topic is not the
toxic properties of mercury, but rather the electrical behavior of
amalgams. And in fact my interest in the subject dates back to the
early 1990s, beginning with a series of letters which I wrote towards
the end of 1992 enquiring about the electrical properties of dental
amalgams. You can read them at:


http://book.boot.users.btopenworld.com/intro.htm


As a result of the response to these letters, and of my own personal
experiences both before and after they were written, I have come to
believe that the electrical potentials generated by metal amalgam
dental fillings are able to dissipate electrical energy through the
nerves in people's heads and, in so doing, make them unhappy.

And in extreme though not unusual cases they are also able to cause
permanent neurological injury which cannot be repaired simply by the
removal of the fillings.

Scientific evidence for the existence of amalgam potentials is well
documented.

And from my knowledge of the electrical behavior of mixtures of
metals, I have no reason to expect that the electrical potentials
generated by dental restorations constructed using gallium amalgams,
rather than mercury amalgams, will be any less significant in
magnitude.

I remember reading a comment some years ago that if dental (mercury)
amalgams were being invented right now their use would never be
approved by the US Food and Drug Administration (the implication being
that the implied approval that their use has now is only due to the
application of the somewhat unscientific principle that "we've been
using them for so long that there can't possibly be anything wrong
with them!").

My hope on the advent of gallium amalgams is that someone, perhaps the
FDA, or the ADA, or an erudite young scientist in a university
materials science lab somewhere, will recognise that the interaction
of the electrical (not only galvanic, but also thermoelectric and
electromagnetic) behaviors of these new materials with the human body
ought to be investigated thoroughly first before they can be approved
for this use. Something which has still never been done with mercury
amalgams.

Keith P Walsh

Paul O

unread,
Feb 4, 2009, 10:44:54 AM2/4/09
to
Keith P Walsh wrote, On 2/4/2009 3:47 AM:
> On 3 Feb, 14:33, Mark & Steven Bornfeld <bornfeldm...@dentaltwins.com>
> wrote:
>
>
>> I will accept your use of the term "gallium amalgam" though it is
>> seldom used and hence can be misinterpreted.
>> In fact, dental restoratives have been made using gallium as a major
>> component; unfortunately their performance has been disappointing:
>>
>> http://www.cda-adc.ca/jcda/vol-64/issue-9/645.html
>>
>> Steve
>>
>>
>
> Dear Steve,
>
> <snip dissussion about the definition of an amalgam>
>

> http://book.boot.users.btopenworld.com/intro.htm
>
>
> As a result of the response to these letters, and of my own personal
> experiences both before and after they were written, I have come to
> believe that the electrical potentials generated by metal amalgam
> dental fillings are able to dissipate electrical energy through the
> nerves in people's heads and, in so doing, make them unhappy.
>
> And in extreme though not unusual cases they are also able to cause
> permanent neurological injury which cannot be repaired simply by the
> removal of the fillings.
>
> Scientific evidence for the existence of amalgam potentials is well
> documented.
>
> And from my knowledge of the electrical behavior of mixtures of
> metals, I have no reason to expect that the electrical potentials
> generated by dental restorations constructed using gallium amalgams,
> rather than mercury amalgams, will be any less significant in
> magnitude.
>
>

Mr Wash
Your knowlege of the electrical behavior of mixtures of metal is purely
speculative. You have never attempted to make even the most basic
measurements to back up your claims. Purchase some amalgam materials,
fabricate some test samples, assemble a test fixture, rent (or borrow) a
good digital volt meter and take some measurements. Then you will KNOW
the thermal electrical behavior of dental amalgams.

> I remember reading a comment some years ago that if dental (mercury)
> amalgams were being invented right now their use would never be
> approved by the US Food and Drug Administration (the implication being
> that the implied approval that their use has now is only due to the
> application of the somewhat unscientific principle that "we've been
> using them for so long that there can't possibly be anything wrong
> with them!").
>
> My hope on the advent of gallium amalgams is that someone, perhaps the
> FDA, or the ADA, or an erudite young scientist in a university
> materials science lab somewhere, will recognise that the interaction
> of the electrical (not only galvanic, but also thermoelectric and
> electromagnetic) behaviors of these new materials with the human body
> ought to be investigated thoroughly first before they can be approved
> for this use. Something which has still never been done with mercury
> amalgams.
>
> Keith P Walsh
>

There you go again, trying to get someone else to do your work for you.
Do your own work, publish your results, and then bask in praise that
will befall unto you.

Best of luck,

Mark & Steven Bornfeld

unread,
Feb 4, 2009, 11:41:41 AM2/4/09
to
Keith P Walsh wrote:
> On 3 Feb, 14:33, Mark & Steven Bornfeld <bornfeldm...@dentaltwins.com>
> wrote:
>
>> I will accept your use of the term "gallium amalgam" though it is
>> seldom used and hence can be misinterpreted.
>> In fact, dental restoratives have been made using gallium as a major
>> component; unfortunately their performance has been disappointing:
>>
>> http://www.cda-adc.ca/jcda/vol-64/issue-9/645.html
>>
>> Steve
>>
>
> Dear Steve,
>
> I think that the main reason why the term "gallium amalgam" might be
> misinterpreted is because of the widespread misconception that, by
> definition, an "amalgam" must have mercury in it.
>


Sorry Keith. I'm not interested in the semantics of what your
definition of amalgam vs. alloy is. I'm a clinical dentist, and I want
my materials to work on planet earth.

Carry on,

Keith P Walsh

unread,
Feb 5, 2009, 5:02:27 AM2/5/09
to
On 4 Feb, 15:44, Paul O <first.d.l...@company.com> wrote:

>
> Mr Wash
> Your knowlege of the electrical behavior of mixtures of metal is purely
> speculative. You have never attempted to make even the most basic
> measurements to back up your claims. Purchase some amalgam materials,
> fabricate some test samples, assemble a test fixture, rent (or borrow) a
> good digital volt meter and take some measurements. Then you will KNOW
> the thermal electrical behavior of dental amalgams.
>

The measurements which can be made with a good digital voltmeter have
already been demonstrated, see,

http://book.boot.users.btopenworld.com/potentials.htm

(These were taken by Dr Jack Levenson, hero of the mercury-free
dentistry movement in the UK, but sadly now deceased.)

The potentials measured here showed up even though the fillings were
not in contact with any saliva - which shouldn't be surprising since
it was first demonstrated that this is the case way back in 1952, see:

http://jdr.sagepub.com/cgi/reprint/31/2/205

Experimental procedures to confirm whether or not these electrical
potentials do in fact excite neurological synapses in the vicinity of
dental fillings would have to positively detect the occurrence, or the
lack of occurrence, of such excitation in the nerve fibers.

I believe that instruments do exist which are sensitive enough to be
able do this. But a "good" digital voltmeter costing less than $100 is
not one of them.

I'm sure I've explained all this to you before. (Perhaps it's you who
is habitually failing to retain the significant parts of the argument
here.)

Keith P Walsh


Paul O

unread,
Feb 5, 2009, 11:13:13 AM2/5/09
to
Keith,
Stop being such a cheapskate! If you really felt strongly about the
health of children you wouldn't let a few measly dollars stand in the
way of this important research. Now take a crowbar, open up your wallet,
an get that DVM.

Besides, you don't need to purchase a DVM they are readily available for
lease at any test equipment rental store. See:
<http://www.testequipmenthq.com/>
<http://www.livingston.co.uk/>
<http://www.etestdirect.com/>
Do a web search, you will be able to find dozens more.

Another way to go is to purchase a used DVM. You will not need the
latest and greatest equipment for your experiment. Just a good, reliable
meter that can measure in the microvolt range.

Good luck. We are all waiting to hear about your results.

Keith P Walsh

unread,
Feb 6, 2009, 2:27:19 AM2/6/09
to
On 4 Feb, 16:41, Mark & Steven Bornfeld <bornfeldm...@dentaltwins.com>
wrote:

>


>         Sorry Keith.  I'm not interested in the semantics of what your
> definition of amalgam vs. alloy is.  I'm a clinical dentist, and I want
> my materials to work on planet earth.
>

It has been demonstrated on planet earth that (mercury) amalgam dental
fillings generate electrical potentials with magnitudes of up to 350
millivolts. See:

http://book.boot.users.btopenworld.com/dutch.htm

It has also been demonstrated that these potentials can be detected
even when the amalgams are not in contact with any saliva in the
mouth. See:

http://jdr.sagepub.com/cgi/reprint/31/2/205

And another research study carried out in Poland (which I believe is
also on planet earth) has even demonstrated experimentally that the
electrical potentials generated by amalgam fillings are quickly
regenerated after being discharged. See:

http://www.ncbi.nlm.nih.gov/pubmed/2103035?dopt=Abstract

Do you think that it should be possible to carry out (on planet earth
of course) research studies to compare the electrical behavior of
amalgams made with liquid gallium (gallium amalgams) with that of
mercury amalgams?

Keith P Walsh

Keith P Walsh

unread,
Feb 6, 2009, 3:49:49 AM2/6/09
to
On 5 Feb, 16:13, Paul O <first.d.l...@company.com> wrote:
>
> Keith,
> Stop being such a cheapskate! If you really felt strongly about the
> health of children you wouldn't let a few measly dollars stand in the
> way of this important research. Now take a crowbar, open up your wallet,
> an get that DVM.
>

You're beginning to sound desperate Paul.

The reason why I keep returning to these newsgroups with the same old
questions is because those questions have never been answered.

I recognise that people like you and Jim Pennino are able to convince
yourselves that you have answered my questions, but you are mistaken.
All you ever do is simply make up excuses for ignoring them because
you don't know what the answers are.

Olin Perry Norton even appeared to have convinced himself that he had
conclusively dealt with my questions by asking me one of his own which
I couldn't answer!

How obtuse is that!?

Some of the earliest questions I posted to sci.med.dentistry were
concerned with the physical properties of dental amalgams, and why it
appears that properties such as their electrical conductivity and
magnetic susceptibility had ever been measured. And those questions
are still not resolved today.

I still bother to check. (I doubt very much whether any of you guys
ever do.) And on a recent search I found a paper with the following
title:

"Magnetic susceptibility and electrical conductivity of metallic
dental materials and their impact on MR imaging artifacts"

- the summary of which can be seen at:

http://www.demajournal.com/article/S0109-5641(07)00199-6/abstract

I was sufficiently interested to pay the online fee to buy the
complete article (see, I'm not such a cheapskate after all!), and do
you know what I discovered? I discovered that the "amalgams" which the
paper claims to have tested were not amalgams at all. They were simply
samples of the various ALLOYS which are supplied to dentists for
mixing with liquid mercury to form amalgams.

As it happens, two of those alloys do have some mercury in them -
Starfill NG2 (3% Hg) and Ana 2000 Duet (2% Hg). However, despite their
mercury content, these materials are NOT amalgams. They are ALLOYS.
Dentists will appreciate perfectly well that you can't make amalgams
with those proportions of mercury. These materials would have been
formed by mixing ALL of the constituents (Ag 70%, CU 15%, Hg 3%, Sn
12%; and; Ag 43%, Cu 25.4%, Hg 2%, Sn 29.6%, respectively) in their
molten states, i.e. at a temperature well above room temperature, and
then allowed to cool at controlled rate.

The other "amalgams" tested, i.e. those with 0% mercury content, were
not amalgams either, they too were ALLOYS (Safargam Plus, Safargam
NG2, Safargam Special and Ana 70 Duett).

These alloys are purposely formulated for use in amalgams. It may make
sense therefore to refer to them by the term "amalgam alloys", in
which term the word "amalgam" serves as an adjective.

But they are NOT amalgams. And in the physical world, regardless of
what anyone calls them, they do not accurately represent the physical
composition of any material that would be found in anyone's mouth in
the form of an amalgam dental filling.

The discovery of this paper therefore does not answer any of my
questions regarding the physical properties of amalgam fillings.

On the contrary, it raises more questions.

Such as, if someone can get a Phd measuring the electrical and
electromagnetic properties of the alloys used for mixing amalgams, why
hasn't anyone bohered to do the same with samples of actual amalgams
(i.e. after the alloys have been turned into powder and mixed with
liquid mercury to form the actual amalgams which are placed in peopl's
teeth.)?

The same paper also presents the results of measurements for a range
of other metal alloys, those used for crowns, bridges, braces, etc,.

So we have the properties of electrical conductivity and magnetic
susceptibility for just about every type of metallic material that
dentists place in people's mouths, and for the ALLOYS used for mixing
amalgams, but NOT for the mixed amalgams themselves.

And with regard to the actual amalgams themselves, i.e. as they appear
in people's teeth, we are still ignorant.

How do you explain that?

And don't tell me that it's up to me to do the measurements myself.

That's just the same old excuse that you always use simply to avoid
the question.

Keith P Walsh

Steven Bornfeld

unread,
Feb 6, 2009, 9:27:05 AM2/6/09
to
Keith P Walsh wrote:
> On 4 Feb, 16:41, Mark & Steven Bornfeld <bornfeldm...@dentaltwins.com>
> wrote:
>
>> Sorry Keith. I'm not interested in the semantics of what your
>> definition of amalgam vs. alloy is. I'm a clinical dentist, and I want
>> my materials to work on planet earth.
>>
>
> It has been demonstrated on planet earth that (mercury) amalgam dental
> fillings generate electrical potentials with magnitudes of up to 350
> millivolts. See:
>
> http://book.boot.users.btopenworld.com/dutch.htm
>


(snip--sorry).

I'm sure it's possible. I've watched for years as the denizens of
sci.materials and perhaps sci.physics have encouraged you to conduct
your tests.
You can try it with plutonium, for all I care. I'm not using gallium,
and I'm neither a physicist nor a metallurgist. And I'm not going to
use gallium "alloys" in my practice.
You should know though that I'm the type of guy who tests 9-volt
batteries by touching them to my tongue. Maybe that's my problem right
there.

Steve

Keith P Walsh

unread,
Feb 7, 2009, 3:59:48 AM2/7/09
to
On 6 Feb, 14:27, Steven Bornfeld <dentaltwinm...@earthlink.net> wrote:

>         You can try it with plutonium, for all I care.  I'm not using gallium,
> and I'm neither a physicist nor a metallurgist.  And I'm not going to
> use gallium "alloys" in my practice.

How about mercury amalgams?

Do you use them?

(And remember, it's NOT a pleonasm.)

Keith P Walsh

Mark & Steven Bornfeld

unread,
Feb 7, 2009, 10:06:19 AM2/7/09
to


Take it up with Webster:

http://www.merriam-webster.com/dictionary/amalgam


>
> Keith P Walsh

Keith P Walsh

unread,
Feb 10, 2009, 6:06:24 AM2/10/09
to
On 7 Feb, 15:06, Mark & Steven Bornfeld <bornfeldm...@dentaltwins.com>
wrote:

>

Like I said, it's a common misconception.

Autymn D. C.

unread,
Feb 28, 2009, 7:14:39 PM2/28/09
to
On Feb 3, 1:50 am, Keith P Walsh <keith.p.wa...@btinternet.com> wrote:
> In reply, first of all, can I take it that you agree with my
> description of the difference between an amalgam and an alloy (which
> is not dependent upon whether or not the thermoelectric eddy currents

whether or not = whether or not whether

> I do not believe that you actually know whether or not this apparent
> implication is correct. At least, I do not believe that you are able
> to provide any direct scientific (i.e. experimental) evidence in
> support of any such assertion.
>
> And there in lies the rub. It appears that, in this "science obsessed"
> world of ours, no-one has ever bothered to even try to find out (by
> experiment) whether or not human neurological tissue is capable of
> detecting the electromagnetic disturbances generated by the lumps of
> inhomogeneous mixtures of metals that dentists routinely place in
> children's teeth.

Is human tissva a superconductor--in the relevant band? That is, not
in the picovolts-microvolts of the nuclear radio and aqveose ionic
background, but in the millivolts of lone metals, which the human body
is not?

> Electric, magnetic and electromagnetic fields are all manifestations
> of the fundamental energy that makes up the Universe. They are not
> characterised by field strength alone. Their nature is much more
> complex than that. In the modern technique of nuclear resonance
> imaging, the strength of the signal is largely irrelevant with regard
> to the degree by which it is attenuated by the target. It's the
> frquency/wavelength of the signal that matters most crucially, and if
> it doesn't match the resonant frequency/wavelength of the target then
> the attenuation doesn't happen, no matter how strong the signal is.

Yes. And reception doesn't happen either.

> That's just an example to establish the principle that in certain
> circumstances a weak signal may have a significant effect, whereas a
> much stronger signal which has different characteristics might have no
> effect at all in otherwise identical circumstances (a principle which
> I believe that "Androcles" may have been alluding to in his response
> to your contribution - thank's for the prompt Androcles).

That's a cretinose argument for thome who can't add. If there's a
damping mekanism for stronger signals, the output still must show up
somewhere.

> I agree that the influence of a static magnetic field is not able to
> drive an electric current in a homogeneous electrical conductor. That

A static magnetic field is itself born of elèctric currend, and would
need to feed another body to make another, and no longer stay static.

> much is "elementary physics". However, having said that, at the moment
> of establishing the new local magnetic field in the vicinity of the
> conductor the charge carriers (electrons) in the conductor will be
> disturbed momentarily, and may adopt a new static arrangement in order
> to maintain equilibrium with the newly established local field,
> although no further movement of the electrons will take place once
> this static equilibrium condition is reached.

Which I like to call IC (AC + DC), intermittent currend, with one sign
and on one side of the baseline.

[snip]

> Food for thought.

It was pap.

[snip]

-AutOn Feb 3, 1:50 am, Keith P Walsh <keith.p.wa...@btinternet.com>
wrote:


> In reply, first of all, can I take it that you agree with my
> description of the difference between an amalgam and an alloy (which
> is not dependent upon whether or not the thermoelectric eddy currents

whether or not = whether or not whether

> I do not believe that you actually know whether or not this apparent
> implication is correct. At least, I do not believe that you are able
> to provide any direct scientific (i.e. experimental) evidence in
> support of any such assertion.
>
> And there in lies the rub. It appears that, in this "science obsessed"
> world of ours, no-one has ever bothered to even try to find out (by
> experiment) whether or not human neurological tissue is capable of
> detecting the electromagnetic disturbances generated by the lumps of
> inhomogeneous mixtures of metals that dentists routinely place in
> children's teeth.

Is human tissva a superconductor--in the relevant band? That is, not
in the picovolts-microvolts of the nuclear radio and aqveose ionic
background, but in the millivolts of lone metals, which the human body
is not?

> Electric, magnetic and electromagnetic fields are all manifestations
> of the fundamental energy that makes up the Universe. They are not
> characterised by field strength alone. Their nature is much more
> complex than that. In the modern technique of nuclear resonance
> imaging, the strength of the signal is largely irrelevant with regard
> to the degree by which it is attenuated by the target. It's the
> frquency/wavelength of the signal that matters most crucially, and if
> it doesn't match the resonant frequency/wavelength of the target then
> the attenuation doesn't happen, no matter how strong the signal is.

Yes. And reception doesn't happen either.

> That's just an example to establish the principle that in certain
> circumstances a weak signal may have a significant effect, whereas a
> much stronger signal which has different characteristics might have no
> effect at all in otherwise identical circumstances (a principle which
> I believe that "Androcles" may have been alluding to in his response
> to your contribution - thank's for the prompt Androcles).

That's a cretinose argument for thome who can't add. If there's a
damping mekanism for stronger signals, the output still must show up
somewhere.

> I agree that the influence of a static magnetic field is not able to
> drive an electric current in a homogeneous electrical conductor. That

A static magnetic field is itself born of elèctric currend, and would
need to feed another body to make another, and no longer stay static.

> much is "elementary physics". However, having said that, at the moment
> of establishing the new local magnetic field in the vicinity of the
> conductor the charge carriers (electrons) in the conductor will be
> disturbed momentarily, and may adopt a new static arrangement in order
> to maintain equilibrium with the newly established local field,
> although no further movement of the electrons will take place once
> this static equilibrium condition is reached.

Which I like to call IC (AC + DC), intermittent currend, with one sign
and on one side of the baseline.

[snip]

> Food for thought.

It was pap.

[snip]

-Aut

Autymn D. C.

unread,
Feb 28, 2009, 7:33:21 PM2/28/09
to
On Feb 4, 12:47 am, Keith P Walsh <keith.p.wa...@btinternet.com>
wrote:

> degrees celsius). The type of material that you get from such a
> process could not be accurately described as an alloy (or what I
> sometimes call a "true alloy) because a significant part of its volume
> would remain as the "unreacted", or "undissolved" cores of the
> original grains of solid alloy used in the mix, and these would not
> have any gallium in them at all.

It was not dereacted or dedissolved.

> So it wouldn't be "an alloy". It would be "bits of an alloy all held
> together in a matrix of a dissimilar metallic composition formed by
> the liquid gallium in dissolving the outer layers only of the original
> solid particles". Or, more generally, an inhomogeneous mixture of
> dissimilar mixtures of metals.

a composite

> As a result of the response to these letters, and of my own personal
> experiences both before and after they were written, I have come to
> believe that the electrical potentials generated by metal amalgam
> dental fillings are able to dissipate electrical energy through the
> nerves in people's heads and, in so doing, make them unhappy.

More unhappy than, say, piezoelèctric chewing effects within the CEJ?

-Aut

0 new messages