Path: santra!kth!mcvax!uunet!lll-winken!csd4.milw.wisc.edu!bionet!apple!sun-barr!rutgers!joyce!mordor!voyager!lip
From: lip@voyager..ARPA (Loren I. Petrich)
Newsgroups: sci.physics.fusion
Subject: Problems with Fusion Reactions and Pathological Science (long)
Message-ID: <24588@mordor.s1.gov>
Date: 6 Jul 89 22:34:08 GMT
Sender: news@mordor.s1.gov
Reply-To: lip@s1-voyager.UUCP ()
Organization: Supercomputer R&D Project, LLNL
Lines: 161


	I have done some somewhat amateur nuclear-physics research on
the possibility of cold fusion. Given the raw materials, p, D, T, and
He3; there are the following possible reactions:

	p + D -> He3 + gamma	 5.40
	p + T -> He4 + gamma	19.81
	D + D -> He4 + gamma	23.65
	D + D -> T + p		 3.84
	D + D -> He3 + n	 3.08
	D + T -> He4 + n	17.49
	D + He3 -> He4 + p	18.26

The energies released are in the rightmost column, and are in MeV.

	Reaction rates? These should be unaffected by the chemical
environment, since that's all long-range, low-energy.

	The reactions involving release of a photon should be
suppressed by at least a factor of alpha (the fine structure constant)
compared to the other reactions; there is no possibility of getting D
+ D -> He4 + gamma as the primary reaction. It has been suggested that
internal conversion is so strong that it makes this reaction take
place at 10^9 the rate of neutron-releasing reactions; thus explaining
Pons and Fleischmann's "results." Wasn't it Hagelstein?

	I have attempted to estimate the branching ratio for internal
conversion, and I have found it to be extremely small. The ratios
(internal conversion/direct) are (approximately):

b(electric-l) = l/(l+1)*alpha^4*Z^3*(2me/E)^(l+5/2)
b(magnetic-l) = alpha^4*Z^3*(2me/E)^(l+3/2)

with released energy E much less than electron mass me. "l" is the
multipole order. I haven't been able to find any similar formulas for
the case in which electrons are relativistic, but some phase-space
arguments lead to the formulas

b(electric-l) = l/(l+1)*alpha^4*Z^3*(2me/E)^(l+3)
b(magnetic-l) = alpha^4*Z^3*(2me/E)^(l+2)

The exponents I am not too sure about; they may be (l+2) and (l+1)
respectively.

Here is a table of estimated branching ratios:

	p + D -> He3 + gamma	 (5.40)		8e-10	2e-10	3e-11
	p + T -> He4 + gamma	(19.81)		6e-11	3e-12	2e-13
	D + D -> He4 + gamma	(23.65)		4e-11	2e-12	8e-14

using exponents 2, 3, and 4 respectively in the last three columns. It
is rather evident that internal conversion cannot possibly be a
significant energy sink for the reaction.


	And is it possible to have a reaction that produces no
neutrons? It seems not, from considerations of the symmetries that
govern the strong interactions. Strong interactions respect not only
the familiar kinematic symmetries (energy, momentum, and angular
momentum), but also isospin symmetry. The latter symmetry is broken by
the electromagnetic field; but EM is weaker by a factor of (alpha),
and is insignificant in the reaction dynamics.

	What are the "regular" spin and isospin quantum numbers? Here
they are (spin, isospin j and jz):

	n	1/2	1/2,-1/2
	p	1/2	1/2,+1/2
	D	1	0,0
	T	1/2	1/2,-1/2
	He3	1/2	1/2,+1/2
	He4	0	0,0

The most important reactions where the main hydrogen isotope is D are
D + D -> T + p and D + D -> He3 + n. It is easy to transform the first
reaction into the second by flipping all the isospins. The
nuclear-reaction rate will be the same.

	However, EM effects are important in what happens afterwards;
a T and a p repel each other, which a He3 and a n do not. This
repulsion suppresses the reaction slightly. However, it has a slightly
larger phase space, due to its higher released energy. I don't know
how these two effects balance out.

	Of the other reactions, D + T will produce neutrons.

	Thus, neutrons are an inevitable byproduct of fusion
reactions, and their presence at appropriate energies should be a good
index, superior to simple heat production.

	Effects that would invalidate these conclusions, (1)
suppressing neutrons and (2) enhancing internal conversion, would be
very important discoveries -- if they existed. But the supposed
"evidence" is ambiguous at best, and the theory behind these
conclusions seems very well-established.

	I am told that a Caltech group had obtained excess heat	with a
P and F -type cell, but that when they put in a stirrer, the "excess
heat" disappeared. The "excess heat" was pure mis-accounting; only
part of the water got hot. And this may well be the answer in most, if
not all, the other cases.

	I wonder about the statistical significance of the neutron
observations; if they are on the borderline of sensitivity; then the
"observations" may be due to instrument or environmental fluctuations.
It is interesting that the Frascati group got an upper limit less than
some other groups' positive results.

	It is certainly good that P and F are letting others analyze
pieces of their electrodes; I think it means that they are moving away
from acting in a Paul-Kammerer-like fashion.

	As to Irving Langmuir's "pathological science;" I still
haven't found a reference to his early 1950's(?) paper, but I remember
seeing some of its conclusions:

	Pathological science features far-reaching and dramatic
claims.

	Pathological science features reliance on borderline and
hard-to-detect effects.

	Langmuir gave as examples N-rays, the Allison effect,
mitogenetic rays, and the canals of Mars. N-rays were discovered by
some researchers in Nancy, France not too long after X-rays. They were
emitted by a variety of objects, but never from wood. N-rays could
only be seen in near-total darkness. These rays were observed by
French scientists, but could not be observed by British or German
ones. The British physicist R.W. Wood went to the lab of a leading
N-ray researcher, Blondlot, and got suspicious. Blondlot was taking
spectra of N-rays, and Wood replaced the metal file he was using with
a wooden ruler. Blondlot continued observing N-rays. The ultimate
result: N-rays existed only in the imaginations of those who thought
they saw them.

	The "canals" of Mars were seen by some astronomers around the
turn of the century, but not by others. In the 1880's, Schiaperelli
had observed "canali" on the planet, which translates into "channels"
(anyone who knows Italian: correct me if I am wrong), with no
assumptions about origins. In the 1920's Percival Lowell publicized
the idea that they were canals constructed by Martians. Other
astronomers at the time could not see them (see Shklovskii and Sagan's
_Intelligent Life in the Universe_ for more details), and the
consensus grew up that the canals were probably optical illusions
caused by borderline perceptions. The issue was finally resolved well
after Langmuir's paper; the whole planet has been photographed at high
resolution by several spacecraft, and there are at most one or two
features that can be matched with the "canals."

	I wonder if cold fusion is going to turn out as real as
N-rays or the canals of Mars.
						        ^    
Loren Petrich, the Master Blaster		     \  ^  /
	loren@moonzappa.llnl.gov		      \ ^ /
One may need to route through any of:		       \^/
	sunlight.llnl.gov			<<<<<<<<+>>>>>>>>
	lll-lcc.llnl.gov			       /v\
	lll-crg.llnl.gov			      / v \
	star.stanford.edu			     /  v  \
						        v    
"I'm just a spud boy looking for that real tomato" -- Devo