Jump to content
Ethan P.

What does a lab do?

Recommended Posts

Hello everyone. I've been pondering this for a while. We all know that we can have our steel sent off to be "analyzed" for both composition and phase. I understand that the phase is determined by microscopy, but what about the alloys? Is it like a sweeping scan (like NMR is for organic chemicals) or are certain reactions necessary to analyze it further (reactions to detect the presence of certain elements and something like spectroscopy to figure out concentration)?

 

I feel like I could do it, with my university's resources, if I only knew how.

 

Thanks for the help,

 

-Ethan

Share this post


Link to post
Share on other sites

I am not sure of the actual process but when i had steel tested it came back with all the alloys including trace elenemts.

Check with the metalurgy dept. at your u. and they should be able to tell and show you what you need to know.

Bob

Edited by Robert Mayo

Share this post


Link to post
Share on other sites

I work at a steelmill and we use spectroscopy to determine most of the alloying concentrations.

Hydrogen levels are tested in a separate machine that measures the H content of the gasses from a sample molten in a small induction furnace.

Same thing with Carbon and Sulphur, the spectrometer is not very good at reading the C and S levels so a sample is melted in another induction furnace and the gasses are analysed to determine C and S content.

Share this post


Link to post
Share on other sites

There are lots of options for analysis. From my experience in the metals industry, C, S, N, O, and H are typically analyzed for using combustion techniques with the dominant equipment manufacturer being Leco - in fact, I've never seen those elements analyzed for reporting by any other method/equipment. For elemental analysis, you're looking at spectrometers. The two most common ones are x-ray fluorescence spectrometers and optical emission spectrometers, sometimes called arc-spark spectrometers. In XRF analysis a polished sample is exposed to x-rays and spectra are generated from the decay of excited electrons. Detectors are used to get the various wavelengths of energy generated, and the amount of an element present determines the amount of energy generated. You run this through complicated programs, that are actually easy to operate, and get the percent of element present in the sample. For optical emission, you again ttake a sample with a polished surface (120 grit finish) and excite an arc on that surface - again you generate spectra by electron excitation/decay and again run it past detectors and through a canned computer program. My general opinion is that XRF is better for exotic alloys and heavier elements - it also looks at a larger surgface area than OE. Optical emission is better for light elements, such as boron, and less exotic alloys.

 

Other options, typically used for very low levels of alloying elements include inductively coupled plasma (ICP) spectrometers and atomic absorption (AA) spectrometers. I've no direct experience with either of those spectrometers.

Share this post


Link to post
Share on other sites

Yup, quite a few options out there... Depends on how accurate you want to be and what your budget is.

 

For spot checks or testing small amounts of material for which it is not worth while doing an expensive in depth analysis, I reckon using spark testing is quite common practice... As far as I know it is some kind of spectroscopy where a small portion of metal is vaporized using al electric arc (I may be totally wrong on how it works though hehe). Last time I checked it cost about $100 per test. It will probably only pick up the metal alloying elements and not carbon or nitrogen content...

 

I know some gold inspectors carry portable testers or something like that to do spot checks for gold and silver purity or correct alloying at jewellery studios, so whatever they use could be relatively simple (I suspect spark testers).

 

It is easy enough to determine carbon content through microscopy provided you can do the heat treatment of the sample properly to get a suitable microstructure. Easy for plain carbon, harder as alloying increases and pearlite gets harder to make.

 

If you want to go minimum budget and just get a ballpark idea for own use you could just do a variety of thermal treatments and see what microstructures they give... Determine carbon content as above (use % area covered in pearlite and cementite multiplied by carbon content of pearlite and cementite for a very slowly cooled sample). With some knowledge of TTTs, CCTs and tempering curves you can match up the curves you get for your steel and compare it to known steels with similar C content. I have used this method a couple of times where they gave us mystery samples in practicals to heat treat.

 

These are the tests that should require the minimum preparation and equipment I can think of. XRD and XRF are also not terribly expensive, but may need special sample prep and you would need training to operate the equipment and interpret the results yourself. With the specific types of XRF and XRD our geology dept. had, we only used these two methods on mineral samples (which were not allowed to contain metal due to issues with platinum crucibles used to melt samples into slags), so I have no idea of their applications in testing metal composition. The metallurgy dept. used "the other" methods :P

Edited by Bertie le Roux

Share this post


Link to post
Share on other sites

OE - set up properly will give good numbers for C & N - maybe not as good as a Leco, but a good ball park number for a knifemaker. In fact, our new x-ray spectrometer, should be able to get us ballpark C and N numbers once we get it totally set-up. Sample prep for either OE or x-ray - get the sample flat in an area large enough to cover the aperture in the spectrometer - most of our OE's a nickel space would be adequate. For the x-rays - a little over 1" in diameter - 27mm). OE - the sample can be larger - must fit inside a cover - about 4" x 6" and you can go pretty thick. X-ray must fit within a cup that has a diameter of about 2" and can be maybe 1 and 1/2 inches tall.

 

OE - you can get by with a good 60 grit finish from a zirconia-alumina belt. X-ray you need to go down to 120 grit from a zirconia-alumina - you can go that far for the OE as well. Use a silicon carbide belt and you'll skew your silicon analysis. OE, especially older ones, have difficulty analyzing wrought iron and pattern welded materials. Also, with OE, you should do at least 2 burns per sample and average the results. If the 2 are very different from each other you'll need to do 3 or more.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...