Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-L] IBM movie



On 05/01/2013 11:58 AM, Bill Nettles wrote:
The thing that fascinated me was the patterns around the atoms and
collections of atoms. Are those interference patterns of the
E-fields? They also have a wavy appearance. Look at the patterns
around the rectangular platform that the "boy" ends up standing on.

I quote from
http://www.fkf.mpg.de/52586/kk281.pdf
which (in a slightly different context) says
"The interference pattern in the image corresponds to the
oscillations of the local density of states due to the
localized scatterers"

Remember, in the metallic state there are gazillions of electrons
running around. Roughly speaking there is a megamp flowing in
every direction and the only reason you don't notice is that there
is a megamp flowing in the opposite direction also. This is
required by the Pauli exclusion principle; an electron at the
Fermi /energy/ will have a lot of /momentum/ with an associated
de Broglie wavelength. If you mess with that flow (e.g. by
sticking scatterers on the surface) there will be standing wave
ripples aka interference fringes.

Looks like equipotentials to me.

Well, they can't be equipotentials strictly speaking. The only way
to draw equipotentials plural, i.e. equipotental lines plural, is to
plot contours of constant φ=2, φ=4, φ=6, et cetera. That is to say,
you need to look at the /remainder/ modulo something (modulo 2 in my
example). There's nothing in the physics here that will do modular
arithmetic for you.

Besides, the patterns look more like equi-distant spacing rather than
equi-potential spacing. Also you get oscillatory patterns /inside/ a
corral, as we see at e.g.
http://researcher.watson.ibm.com/researcher/files/us-flinte/stm15.jpg
whereas we would expect potential-gradient to be small in the interior
of such a thing. In 3D the potential-gradient would be zero. In 2D
it would not be strictly zero, but it would be less.

Or are these artifacts of the TAFM processing?

Undoubtedly there are /also/ lots of AFM artifacts. It is very hard
to say anything simple about artifacts, because (a) they are artificial
to begin with, and (b) the operators compensate for them as best they
can, so whatever you see is some Nth-order term that escaped compensation.

Also, the problem with movie-making is that they are making a movie,
not making a measurement ... so there is no way of telling to what
extent they post-processed it to reduce the prominence of the electron
surface-state oscillations.