more on SEM

[T3D Edward Gosfield III, MD <gosfield@xxxxxxxxxxxxxxxxx>]

Once again George has provided some excellent comments on SEM, but 
there seem to be some misunderstandings (maybe from before George's 
post was out)

John writes, responding to John Vala and me finding the 'geometric optics'
analogy of SEM with light imaging not very useful--:

>Not even in the limited sense I described in a previous post - that 
>the e-beam leaves the pinhole and is rastered out onto the specimen 
>sort of the reverse of photons going from object to entrance pupil of 
>camera lens?  If not, why not?

I think the optical analogy is apt in describing the perspective of the 
situation. But the field of view (as George points out) is more important in
determining how much tilt is practical. The field of view needed is 
determined by the magnification, or by the specimen size
(sort of the same thing, if you want to see the whole specimen)

>> The way i think of the set up sometimes is like a table top full of
>> ping-pong balls.  I am standing above the table with a powerful 
>> garden hose (no nasty jokes, now)  sweeping it in a raster pattern 
>> across the table top.  As it hits the balls its tiny stream is 
>> reflected off in various directions, which are determined by the 
>> angle of incidence of the stream with the ball's surface. 

>Which brings me to SEMs.  You talk about all these ping pong balls 
>bouncing off surfaces. 

No..No.. the 'stream of water' bounces off the ping pong balls! 
(or actually the primary electrons hit the specimen, and cause secondary
electronss to be emitted).  The angle of incidence of the primary beam on 
the specimen is determined overall by tilt, but also by the 'micro' surface 
structures of the specimen.  

>Do they bounce off a given part of the 
>specimen the same regardless of the tilt of the specimen table?  

No, they don't necessarily.   But the analogy breaks down here, since the
electronic/atomic characteristics of the specimen, the conductivity of the 
specimen and/or its coating (we left out details of specimen prep.) and the 
surface thickness of various features all contribute to determining how many 
secondary electrons are emitted at a particular point on the specimen, AS 
WELL as the simple geometric angle of incidence of the primary beam on a 
given surface feature.

>If not, we might have the equivalent of something between specular and 
>diffuse.  So the questions is, do you notice anything your mind would 
>interpret as a semi-specular surface when looking at SEM stereo pairs?

First of all, as you know, specular means like a mirror, which reflects
light ,and specular points are reflecting light directly back at the lens,
like a mirror.  Obviously this doesn't work quite the same for electrons,
since their wavelength is 
shorter than that of light.  However, there are certain types of surfaces
and surface details which do emit lots of electrons, and look quite bright
in photo images. Remember that this can also be due to charge build up on the
specimen and reasons other than geometric arrangement of the surface feature. 

Although i have seen lots of these 'bright' artifacts (usually due to defects 
in specimen preparation, or the decision not to coat a specimen) the scale of
the bright details is larger than the usual angular extent of specular image
details in light photography.  I have never personally seen retinal rivalry 
due to any thing analogous to specularity, in SEM pairs.  But i photographed 
biological specimens which are pretty soft and squishy (well--until they get 
'fixed' into specimens) and the level of detail in biological specimens is 
orders of magnitude larger than some of the features George and his 
colleagues look at on prepared surfaces of metals, semi-conductors, etc.  

>From: T3D  Dr. George A. Themelis  <DrT-3d@xxxxxxxxxxxxxxxxxxxxxxxxxxx>

>In the SEM a beam of electrons is thrown on your sample.  The beam is
>*scanning* a small area of interest.  Different things are emitted
>from the sample as the beam of incoming electrons interacts with the
>surface.  Things like, secondary electrons, backscattered electrons,
>Auger electrons, x-rays, and others.  Different detectors are placed
>around the sample to pick up those signals.  An image is formed by
>mapping this information on a screen or piece of photographic paper.

The above well describes the essence of the difference--the secondary 
electrons are a 'cloud' which is instantaneously collected and integrated
as the primary beam scans across the specimen. The total electrons detected
provide the signal which results in the 'mapping' of each locus on the 
specimen.  

BUT the secondary electrons are not all (or even predominantly)
directly reflected from a point on the surface of the specimen to the 
collector.  Rather, they are scattered, and subsequently collected 
electronically.  And that is why the optical analogy breaks down 
at this level of detail. 

>Instead of thinking of a beam of electrons coming from the 
>top, hitting the sample and the results of the interactions being detected
>by the detector in the side, you can think of the situation as light being
>emitted from the detector, reflected by the sample and directed to the
>lens at the top to form the image.

But only because the 'mapping' of specimen features by collecting secondary
electrons provides a very good correspondance to what we imagine the surface
would look like at light wavelengths.  In lots of cases, of course, we can 
verify this correspondance by using low mags, so that the specimen 'mapped'
by electrons does indeed look just light a light photograph.

>When it comes to 3d imaging you can use the model of the central
>projection with the very good approximation of orthogonal projection
>in sufficiently high magnifications.

I think this is entirely a function of field of view--we could be 'looking'
down a long cardboard tube, instead of 'looking' through an imaginary lens, 
so does it add anything to our model to use the concept of an optical lens?

>I can send a copy of my unpublished report on how to take view and measure
>stereopairs in the SEM to those who are interested in the subject.  This
>was written in 1990.  

I'd love a copy, George--you have my address (or email me if you need it)

A friend of mine uses SEM routinely to evaluate how well his prep line
is cleaning the surface of polished silicon wafers before they get sent off
to chip manufacturers.  But he hasn't used stereo yet.  Perhaps i can make 
him aware of your work.

>I plan to have a presentation on "3D at LTV" and discuss all potential
>stereo imaging applications at my workplace.  I plan to bring my TDC
>stereo projector and show some of my regular slides too.  Will let you 
>know (and ask for feedback and help!) when I get closer to doing it.

What a great presentation idea.  I suggested 3d slides for anatomy class when
i was in med school, but folks looked at me like i was nuts.  Now, of course, 
there is at least one collection of such slides available, and i'll bet it's 
coming soon to CD.

Ted G


------------------------------