You mentioned in your discussion about
pseudo potentials and your work as a

graduate student in calculating pseudo
potentials.

That word did not exist when I started.

I started when was it 1954?

Yes.

What did exist was the orthogonalized
plane wave method for calculating

electronic structure.

That had only once been used as a test
at about the time, about 1950, 60, and

then by around

No, 1940, around 1950.

And then used in 1950 for the first time
and properly used for the electronic

structure of silicon.

Because silicon, after World War II, was
the important whole idea of

semiconductors and so on, was the new
science.

So I, for my calculation on aluminium
electronic structure, I used the

orthogonalized plane wave method.

Now, that gradually evolved by different
steps until middle of the 1960s, the

main ideas.

Then there were two extra

bits that then improved.

One was, I forget the man's name, coming
from the United States as a postdoc to

us, the effective mass of the electrons
and the fact that there was a charge

deficit in the core of the atom, and
that had to be taken into account.

And then another

Eight years later after that, up till
that time, people had just calculated

the pseudoprotential when they needed
it.

But then the people at Bell Labs, this
is now the end of the 1970s, suddenly

said, we'll just calculate all the
elements in the periodic table.

And they made an extra innovation.

Instead of using plane waves, they used
the circle Bessel functions.

If you take a plane wave and decompose
it in angular momentum, you get the

circle Bessel functions.

And so they worked directly in real
space with the circle Bessel functions

that hold all those 92 elements.

It was done like that.

But

A few years ago, I had a discussion with
a friend, a previous TCM member, Dennis

Weir, in Dublin, and we were going to
write...

a sort of history and a proper history
of pseudo potentials.

Because nobody, I mean, they just exist
as separate papers and so on.

And it's a nice science story of how
science developed.

But then unfortunately, his health
deteriorated.

He was an ex-CCM member, but from early
days.

And so that paper has never gotten
written.

But there's an outline of it somewhere
on my computer.

And maybe you and I should write it.

It's something, I think it's not first
priority, but I think it would be a nice

thing to do.

So from the orthogonalized plane wave
method of calculating, which was a band

structure method, but one could,
nowadays, it was obvious that you can

write it in pseudo potential terms.

It was not so obvious then, but
gradually, step by step, it became.

And suddenly, in the middle of the
1980s,

60s, by putting two and two together
from a visitor and so on, suddenly it

became easy to generate a pseudo
potential for any element.

And Alex Animalu was a graduate student
at that time, and instead of just

publishing one pseudo potential, I said,
we can now do all the elements.

I didn't want to do 92 elements, but we
did 25 or something elements of the main

ones that in semiconductor and the kind
of metals.

Not the transition metals, but the SP
bonded elements that were widely used in

semiconductors and atomic physics.

So we suddenly, it was a paper with 25
elements.

And that was, it was one of the top
cited papers even until the 1980s.

I remember coming across a reference to
it in the 1980s.

It was one of the top cited papers at
the time in the late 1960s.

So that's something that's, if we have
time and we have the brain, then to do it.
68
00:06:45,138 --> 00:06:48,8
Not the first priority, but the second
priority.

Yes.

Alright, just stop there, finish that
subway.