2 3 0 D O C U M E N T 9 9 A U G U S T 1 9 2 0
Assumption B. The orientation of each molecule in the crystal powder is practi-
cally completely fixed;
however, an electron ring present within the molecule can execute a quantum
mot[ion] of units +1 or 1. Outside of the magnetic field, both motions occur
equally frequently—but within the magnetic field, not anymore. If the susceptibil-
ity is calculated
here,[18]
[In order not to arrive at an incorrect result, one must set
in the exponent of not the energy, but “Routh’s function,” as was proved in
the dissertation by van Leeuwen.] one again obtains exactly LangevinÊs
formula.[19]
So far, this would be very nice indeed, but for the following problem: above
equilibria furnish good susceptibility. Yet how are these equilibria supposed to set
in? (particularly at those very low temperatures of 2°)—Quite extraordinary colli-
sions are needed in order to make such an adjustment possible.–
Perhaps the adjustment of molecular magnets to rapidly changing magnetic
fields does not, in fact, occur so rapidly either (for gadol. sulfate at 2°). I spoke with
Onnes[20]
about the problem of how such a lag, provided it exists, could be verified
experimentally.–
The reduction in susceptibility with rising frequency seems to be difficult for
him to demonstrate.
The heat development due to the lagging magnetization (I hope I am not mistak-
en that it must cause the powder to heat up) seems to be much more easily demon-
strable to
him.[21]
Because gadol. sulfate at 2° is already almost as strongly magnetizable as iron,
something might be detectable.
Please give me a comment or two, on your part.
In any case, I really do not believe that it is permissible to treat paramagnetism
of solids seriously like a gas. (See
Smekal)–[22]
It works here as with osmotic pres-
sure.
A barmy idea: You know that W. J. de Haas
[23]
has been thinking a lot about how
spontaneous magnetization of tiny crystals in iron could be proved directly.– Well,
I considered whether there could be any use in the fact that in magnetic fields of a
couple hundred gauss, superconductors are not superconductive anymore.—So:
Assuming an Hg-layer of a few molecules thickness is still superconductive
(which would first have to be determined)—and it remains so for very different
base layers,
then one would have to see whether it wouldn’t lose its superconductivity on
iron.
e–KT