I frequently need to discriminate between 304 and 316 stainless steel which cannot be done by eye.

304 SS has 18% chromium and 8% nickel

316 SS has 16% chromium, 10% nickel and 2% molybdenum

How can I do a chemical assay to discriminate between them and ideally identify their precise composition to 1% or better accuracy? I assume the basic procedure is to create a solution, then do a titration of some kind.

If the first step in your answer is... "dissolve the sample in acid," be aware that stainless steel is pretty resistant to most acids and I need a procedure that will work within 24 hours.

Also, please no answers like "use an xrf machine" or some surface detection technology. I need a chemical process that assays the whole sample, not just its surface.

  • 1
    $\begingroup$ Why is your supplier not marking them prominently and correctly? Note that 316 is not magnetic, although 304 is not guaranteed to be magnetic. As for assaying the 'whole sample' - how do you plan to do that? $\endgroup$ – Jon Custer May 7 '18 at 15:38
  • $\begingroup$ On what order of magnitude size is that "whole sample" piece? 1g? 50g? $\endgroup$ – Karl May 7 '18 at 19:27
  • $\begingroup$ 316 is much more likely to show ferromagnetism than 304. Cold worked 316 will be about half as magnetic as carbon steel . $\endgroup$ – blacksmith37 May 7 '18 at 19:35
  • $\begingroup$ Why do you think a blank surface of your material would have a different composition than the interior? Saw off a piece and analyse the cut if you think that would be necessary. $\endgroup$ – Karl May 8 '18 at 7:03

Digest a sample of the metal in aqua regia, I have found that the gauze in a kitchen sieve can be digested in a very short time with a mixture of 16 M nitric acid and concentrated hydrochloric acid.

Next you should dilute it with water and filter it to remove any particles.

I would use either flame AAS or ICPOES to measure the metals after dilution to a suitable concentration in plastic ware. I would want to measure the Fe, Ni, Cr and Mo in the sample. The Fe to Ni and Fe to Cr ratio may help you distingish between a lot of grades of stainless steel.

One alternative for iron is to use 2,2'-bipy and hydroxylamine to form the deep red iron(II) tris bipy complex.

This can be measured with a UV / vis machine in a 1 cm cuvettes, you will have to make a series of solutions in volumetric flasks. You make 0, 1, 2, 5, 10 and 20 ppm iron by adding a 1000 ppm standard to the flasks, then add a solution of 2,2'-bipy in water, the hydroxylamine and then dilute to the line. Mix them well and then measure the red colour with the UV / vis machine.

I am sure that there are colourmetric methods for Mo and Cr but sadly I do not know them.

  • $\begingroup$ I am looking for a chemical solution. I do not want to buy a big complicated machine. $\endgroup$ – Shaka Boom May 8 '18 at 0:36

I did some more searching around and found that there is a standard spot test for molybdenum in ASTM publication 550 that can be used on stainless steels. It is not ideal in that it does not determine the proportions of chromium and nickel, but just tells whether molybdenum is present:

a) Place equal drops of concentrated hydrochloric acid and concentrated nitric acid on the metal surface. Allow 3 to 4 minutes for reaction and transfer 2 drops to a spot plate.

b) Add 2 drops of 10% potassium thiocyanate.

c) Add 25% sodium thiosulfate drop-wise stirring until the red color begins to disappear. A violet color will appear if molybdenum is present.


Asking this question, I assume you are not a chemist, and neither have a lab or a chemist available in the place you want to do this, right?

Buy yourself a handheld XRF analyser. They look like a toy pistol, come nicely coloured and with touchscreen displays, and there are a dozen or so companies building them. Google gives you many hits, although no prizes. I'll guess you can get one in the 2-3000 € region? Definitely cheaper than hiring a chemist, or starting to fiddle around with acids and other stuff yourself, if you need to do it on a daily basis, and professionally.

(The machines on wikipedia https://en.wikipedia.org/wiki/X-ray_fluorescence look very intimidating, google has much nicer pics https://www.google.com/search?q=handheld+xrf . ;-)

  • 1
    $\begingroup$ Sorry, didn't see the last line in your question that you don't want XRF. ;-) Still, instead of dissolving a sizeable sample pice, can't you just file it to a fine powder and do XRF on it then? Or just file off anything that might be inhomogeneous surface before holding the XRF on it? $\endgroup$ – Karl May 7 '18 at 19:22
  • $\begingroup$ I want to analyze the sample, not scrapings off the surface of the sample. I thought I made that clear in my question. I want a chemical assay solution, not xrf/spectroscopic solutions. $\endgroup$ – Shaka Boom May 8 '18 at 0:34
  • $\begingroup$ @ShakaBoom I noticed. Hence my above comment. :-/ $\endgroup$ – Karl May 8 '18 at 6:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.