January 26, 1954

Professor Linus Pauling

Department of Chemistry

California Institute of Technology

Pasadena 4, California

Dear Professor Pauling:

Thank you very much for the draft of the enneamolybdomanganate ion paper. I am glad to see that off the docket. Is anyone in particular planning to continue the work and refine the structure?

Thank you also for getting after Tucker. Tucker had sent a previous draft of his β-Uranium paper to me, and it was much worse. I sent him a bill of complaints, and when his paper came back from the referees for other changes he made the ones I had insisted on, but the result, as you saw, wasn't too happy.

Proof for the paper on Brillouin zones came today and I'll get it off right away.

Today I am sending the draft of the sigma phase paper back to Gunnar. I am in accord with virtually all of the changes and deletions that you and he made, and have made some more of my own. The paper is now by no means short, but is perhaps not over-long. On the seven pages that you suggested be omitted, I agree that there was much too much there considering the paucity of conclusions that could be reached. The suggestion of mine that the space between the two zones is filled with electrons was made rather half-heartedly, and with the half-expectation that you and Gunnar would not like it and with the half-hope that you and he would come up with something better in the way of a Brillouin-zone interpretation. I am now just about reconciled to the conclusion that that big, wonderful Brillouin zone is only about 5/6 filled. It is, however, something of a disappointing contrast to the alloys which so beautifully show filled zones.

My main purpose in writing this letter is to ask you about one of the deletions you made. This was in connection with the fact that the Curie temperatures of sigma phases seem to be very sensitive to (say) the iron-chromium ratio, but not affected at all by the replacement of iron by cobalt or of chromium by molybdenum. I cited this as partial support for the supposition that the ordering of atoms in Sigma FeCr might be the same as in FeMo, on the grounds that the Curie temperature ought to be to some extent a function of the arrangement of the atoms that carry the magnetic moment. Your notation beside the deletion was, "Not required by my theory". By which you mean, I take it, that the uncoupled electrons at the top of the band do not care where the atoms with magnetic moments are.

However, in your paper in the Trans. Far. Soc., in connection with cobalt and nickel, you point out that an improvement in the treatment can be made by assuming that the wave functions are perturbed so that (in the case of nickel) the unpaired electrons spend more of their time near the atoms with J = 1 than near those with J = 0. Presumably, as you state, this would affect the energy and improve the agreement with the observed Curie temperature. However, if it can affect the Curie temperature in the case of cobalt and nickel, could it not also affect the Curie temperature in the case of the sigma phase, if the positions of atoms with magnetic moments were changed? Then, if the Curie temperature is constant, cannot one take this as some indication that the positions of the atoms with magnetic moments are constant, other things being the same?

I do not regard this as a point worth holding up the sigma phase paper over, though, if my reasoning above is correct, my personal preference is that the deletion be restored. However, I would appreciate very much hearing your views on this.

By the way, if you have any spare reprints of your article on ferromagnetism, I would appreciate it very much if you would send me one.

With best regards,

Sincerely yours,

Dave

David P. Shoemaker

DPS:fd

cc: Professor Gunnar Bergman