: RE PIT DEPTHS

[Goldsmith, Noel]

Michael Shaffer asked about measuring pit depths.
In particular laser ablated pits.
I have looked at this (laser pits) a bit and can say that what you do and
how you do it all depends on the pit aspect ratio, and the actual size of
the pit.
If pits are narrow and deep, then getting an image of the bottom is very
difficult. In the case of optical microscopy, the fraction of light which
comes back out of the pit is vanishing small, and the human eye may find it
impossible to detect, the brightness of the surface around the pit
completely overpowers the tiny signal from the pit bottom. If you have a
digital camera, which allows control of contrast, brightness and gain, then
it is possible to pull the weak signal from the pit up from the darkness,
and image it well enough to fix the focus, which is what you need for a Z.
This may take a bit of time, but a users skills will improve with practice.
But, assuming you can see the bottom, and the undamaged surface around the
pit, and you can measure Z as well as X and Y to the precision you need, it
is all straightforward.
If you use a confocal microscope it will be possibly even better because of
the confocal properties of only having any light when the object is
focussed.
If you use an SEM, then getting electrons down a narrow pit is also
difficult, but as the NA may be quite low and you can control all of the
imaging conditions pretty well, it should be reasonably good. Of course as
the NA is reduced, the depth of focus increases, determining when you are in
focus for Z measurement becomes more difficult.
Using Stereo, ie two images of object at each place, and measuring the
parallax to determine the depth will also work, but if the same places in
bottom of the pit are not visible in each image, it will be no good.

As Michael suggests, three points on the pit surroundings will define a
plane which may be taken to represent the top, and a point (or more) will
then define the base and a bit of trigonometry and you are done.
Our systems for measurement rely upon a Plug-in for ImageJ which is closely
integrated with the stage on the microscope (LUDL electronic products) and
the camera (Diagnostic Instruments Spot (various models).
The plug-in displays a panel which shows the X, Y and Z coordinates of the
specimen (at the point in the field of view indicated by the cross hair in
the camera and the microscope). This point is set at zero when the operator
wishes, by pushing the appropriated button on the stage joystick. Then the
specimen is moved using the stage joystick and focus knob on the joystick,
to a desired point for measurement and the appropriate button on the stage
joystick is pressed, recording the X, Y and Z coordinates in a table of
values on the screen, the the next point is found and the button again
pressed. After all desired points are measured, in this case four, the
process would be repeated for each pit.
The data collection plug-in could be easily modified to produce new columns
of derivative data, in the current form it produces a column which contains
the radial distance from the zero point, calculated from the sum
(sqrt(X^2+Y^2).
Or.
The data file would then be saved and read into some data processing
program, such as excel or Igor-Pro or Origin or Kaleidagraph or whatever you
have or are familiar with and which will do the job. And then you could
apply the trig.
It seems to me that the difficulty you might have is concerned with getting
the microscope linked into imageJ, not making a plug-in or even a macro do
the sums you need.
The path to lots of data, when you need the user to be a critical element
providing judgment in the data gathering process, is to reduce the operators
involvement to the act of deciding the points to be measured, and then
having them perform a simple act, such as pressing a button. The reading,
recording and saving of data is done by the computer.

HTH
Noel Goldsmith


IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.