SPECTROSCOPY PROBLEM
Rainbow Computer
Hello fellow surfers.
I am in the process of constructing a spectrograph to qualitatively and
quantitatively analyse metallic alloys. I am using a DC source to form a
metallic arc, then pass the light through a Littrow spectrograph, which
excites a photosensitive couple that gives an analog reading that will be
converted to digital and given to a PC for later processing and line
matching to an element data base. The details are deep, so before I begin, I
was wondering if any of you out there knew of any technical specifications
for any such instruments in existence todate. From what I've heard, they are
used extensivelly in many areas of the metal industry, to quickly give an
approximate percentage of the contents of many alloys. (for example Carbon
4%, Iron 80%, Chromium 3%, Iridium 2% etc) I would appreciate if anyone has
company names, addresses or phones (internationally) of any such instrument
makers so I can contact them and examine their technical specifications,
before I start building my spectrograph.
If anyone, also, has any information about the line recognition algorithm, I
would appreciate it, since it is an inexact process. (Line width varies with
temperature, pressure, there is line shifting due to the Pole effect at the
arc endpoints, due to the Doppler effect from the EM fields, there is line
reversal from ionization (i.e. sodium at high temperature), elements with
lines close together etc. I will be getting a bunch of digital data which
will be the sensitivity graph of the photosensitive couple. From that point
on, I imagine that I must have some sort of data base with the sensitive
lines of some elements, and do an approximate matching with an error
tolerance of dx. I imagine that the best area for matching will be the UV,
since from the MIT tables for most of the elements, I see that the sensitive
lines are usually in the UV. Also what happens with NON metalic elements,
such as Carbon, that are hard to excite? How many lines for an element,
minimum, do I have to test? If I standardize the spectrograph using a pure
sample, will the intensity of the line for that same metal in an alloy be
directly proportional to its percentage relative to the intensity in the
pure sample? I know, for example that line intensity is proportional to the
content percentage, but how? etc. Any comments will be greatly appreciated.)
Please, send E-mail to the address below, as I do not (yet) have the
capability of connecting to the newsgroups directly, except through my
company's System Administrator.
Thanx a lot in advance for any information.
Sincerely,
John Galidakis, M.S., Athens, Greece.
<rai...@raindrop.ath.forthnet.gr>