SS304 / SS316 - how to check on them ? 4

[azertyuiop]

Dear collegues ,

I want to check if the contractor is supplying my company an installation with the much more expensive SS316 materials that was demanded , and make sure he doesn't supply us cheaper SS304 materials instead (pharma applications).

How to check on this without damaging the delivered equipment/piping networks?

I have already used magnets to check if we are dealing here with stainless steel (the magnet doesn't clinch = SS) , but how to differentiate between the SS304 and SS316 variety ? Some tubing has SS316L inkinscriptions , but most (polished) parts are not tagged.

Expert advise needed.

regards.

 

[Metalguy]

There are non-destructive methods you can use. If the amount of material is large, it might pay to find a met. lab. that can take a portable "metal analyzer" right to your equipment and find out what you have. Another way is to use a chemical kit offered by a few companies (I think Koslov-? is one).

 

[kenvlach]

Portable X-ray fluorescence units are nice but $$$ (if you have to ask, you can't afford). Can give total analysis of composition (except very light elements like B & C).

Metalguy is referring to Stainless Steel ID Kit 1542C from Koslow Scientific. Nice electrochemical spot test.

http://www.koslow.com/1542c.html

Chemical spot tests are just one of many methods described in ASTM E1476-97 Standard Guide for Metals Identification, Grade Verification, and Sorting. Available for purchaase at

http://www.astm.org/cgi-bin/SoftCart.exe/DATABASE.CART/REDLINE_PAGES/E1476.htm?L+mystore+mcqk8236

Qualitative & quantitative inorganic chemical analysis books also give methods for detecting or measuring molybdenum, but it's some work to get a sample into solution (aqua regia works). Then an indicator such as potassium xanthate (or potassium ethyl xanthogenate) is added, which turns red-purple in the presence of slightly acidified molybdate. Sensitivity is 0.04 microgram Mo, so don't need a big sample. -- Vogel's Qualitative Inorganic Ananlysis, 6th Edn., p. 252-252 (1987).

Hanna Instruments sells a wastewater tester for Mo that uses mercaptoacetic acid to form a colored complex for quantitative analysis by photocell. Detection limit is 0.1 mg/L, so you wouldn't need a big sample; sand off a mg and dissolve within a 100 mL Teflon beaker. Starting to sound like work, though.

 

[usjbh]

I've used the Koslow test kit 25 years ago to distinguish between 304 & 316 and identify the Mo. I found it to be not as difinitive as I would like, particularly when objective evidence is needed should the supplier be using 304 ss. I prefer sending a very small sample of the material to an outside analytical lab for Light Element Energy Dispersive X-Ray Spectroscopy and getting a formal report & interpretation. This report can then be retained as objective evidence in your project folder for reference should any problems arrise.

 

[sensorman]

You should be requiring your supplier to provide copies of the material certification reports from the mill.
We require all of our vendors to provide us this information when they ship material to us.

 

[mcguire]

You need to be fairly accurate. Today's 304 often has residual molybdenum of 0.5% or more. Enough to give a false positive, but not enough to make it more corrosion resistant.

 

[NickE]

When I was in SEM classes we were told taht EDS analysis was not generally considered a valid method for bulk chemical analysis. XRF should be used instead. There is likely a lab near you with a stationary machine. Since it sounds like this is a critical requirement I would D*** the expense and cut a tab off and send it out. (Bodycote comes to mind.) (oops missed your non-destructive need)

(If this is going to be a common test then you might look into the portable analyzers that Metalguy and Kenvlach mention)

just as a mention welded and/or worked SS will stick to a magnet.



Nick
I love materials science!

 

[usjbh]

In response to NickE's reply I had to come back to defend my EDS recommendation vs. XRF. We use SEM EDS all the time because our samples are usually unknown particles trapped in 2 micron filters that we are trying to I.D. XRF, which is typically used in foundry environments, requires a large sample due to the nature of the beam penetration. The energy dispersive methods only require a small particle of a sample, something you can easily remove without destroying the host. Because all these techniques suffer from limitataions of one sort or another, EDS being peak overlap between sulphur & moly and a 0.05% sensitivity limitation, you have to determine if the application is appropriate. In this case differences between 304 & 316 can easily be determined with the tiniest of sample.
Regarding magnetic properties of 316 or 304, my experience is that it is highly dependent of the % ferrite formation in the alloy but even under heavy cold worked condition, 300 series alloys wiill only exhibit light magnetic properties whereas martinsite alloys exhibit strong magnetic properties and you can easily sense the diffrences with a small hand held wall magnetic in a pinch.

You're right, materials science and art are fun and interesting.

 

[unclesyd]

Being a Pharma Co. you could use a simple spot test we used to differentizte between 304S/S and 316S/S.
In a dropper bottle make a 50/50, V/V, water/hydrochloric acid and saturate, 5 minutes of bubbling, with sulphur dioxide gas (SO2).
Take a known piece of 304 and 317 rub with 180 grit paper and put a drop of solution on each, wait one 1 minute and rinse. By looking at the spot in good light you see the difference. The 316S/S will be purple and the 304 grey to black.
Keep the solution saturated and renew each day.

 

[NickE]

In response to usjbh.

The interaction volume (beam-sample) of EDS analysis is the critical thing. I'm not debating the validity of EDS chem. but the sample size is so very small that for chemical analysis of things that are (nnn) times greater in size than the interaction volume XRF is indicated. (From my experience and schooling) not that SEM-
EDS wont work. just that if this is mission critical I would recommend that the (beam-sample) interaction volume be on the same order of scale as the item.

From my GEM the diffeneces in chemistry btw. 304l/316l are:
(Nominal WT%)
304 --> 2% greater Cr
2% less Ni
no Mo
316 --> 2% less Cr
2% more Ni
2.5% Mo

The chemistry of the two is vey close (IMHO) and it would not be a good thing if the EDS sample area was a slightly segregated one. I could easily belive that an analysis would then be off by enough to produce a false positive (or negative).





Nick
I love materials science!

 

[stanier]

You could ask for certificates from the original material supplier.

 

[quadswift]

Here we go again!
There have been several questions about PMI and certification / traceability.

As I have said before, why in this day and age of 'Q/A, ISO 9000 etc' are we querying the supply and the customer is looking at spending money to ensure he got what he asked for?

You are entitled to get what you ask for / specify even by the sale of goods act. 'Quality' is frequently said to be 'meeting the requirements'. If the requirements are not clear ....then!

To get the level of verification you appear to be looking for you need to specify the certification / traceability levels in your purchase orders.
Did you specify certification and traceability levels? If not then you have 'got what you asked for'.

If you are so concerened then you should fix the problem at source and:
Specify the levels required in your purchase orders.
e.g. That all materials are certified, identifiable and traceable back to the certification, heat and batch numbers. With EN10204 3.1.b certification, verified true copies.

You could go on to ask for the testing and material identification transfer to be witnessed by an independant third party, nominated by you, if required (with 3.1.c inspection certificate).
You also need to address how you can ensure that the condition of the parts as supplied permit some form of marking of identity (you are talking of 'polished parts'). Vibro etching has been used in some cases. The other option for these is to have some form of inspection representation at the suppliers.

This will be reflected in the price of the product of course.

Then you can be (reasonably - or am I just an old cynic?) sure that you get what you specify.

Quadswift

 

[azertyuiop]

quadswift ,

you are right on spot concerning the fact that I am earning what I should earn , except that you are firing at the wrong guy. I am new in this company , and new in the heavily regulated pharma sector , replacing another engineer who got fired for incompetence , dixit his collegues. I only hope not to get the same treatment later on , for which I posted this intel request.

My trust in the QA/QC teams is very relative , to say the least, given the number of meetings and coffee cups the guys are swallowing. I have been submitted a mountain of certification papers and check-up's stating that everything is supplied according to the spec's. But having worked in exotic locations , where rubber stamping of certificates by client personnel could be obtained by an appropriate greasing with green coloured monies or sufficient supply of extremely accomodating and attractive women , I have grown a relative distrust to what paper can certify.

I even can obtain a paper certifying that my mother-in-law is a saint , where in effect she is a real bitch. Just don't tell her . . .

Therefore my request for simple physical methods to random check some spots on the overal installation , before approving the commissioning phase of the ongoing project.

I hereby also wants to thank all the other respondents for their very valuable inputs. You are never too old to learn a few new tricks to beat the conman.

 

[usjbh]

It is interesting how a request for a simple test to determine the difference between 304 & 316 evolved into an ethics discussion. Ethics aside, I came across an old Carpenter Material Data Guide from the early 70's that listed methods for identifying stainless steel in the back of the catalog, probably worth its weight in GOLD today.

MOLYBDENUM SPOT TEST:
Three solutions are required for this test:
Solution #1 = Add 1 ml of the following mixture: 50 ml H2O + 10 ml H2SO4.
Solution #2 = Add 1 ml of the following mixture: 5 gm sodium thocyanate + 95 ml H2O.
Solution #3 = Add 3 ml of the following mixture: 100 ml H20 + 25 gm stannous chloride + 35 ml HCL.

PROCEDURE: Place one drop of Solution #1 on a newly ground spot on the specimen. Allow it to react for 30 seconds and then add one drop of Solution #2. This mixture is allowed to react for an additional 30 seconds and then is absorbed with white blotting or filter paper. Place one drop of Solution #3 on the moist spot of the paper. The formation of red or pink coloration in the spot denotes the presence of molybdenum.

Of course this is not definitive for 304 but will certainly detect the presence of the moly that should not be there with 316.

It appears that a legacy FAQ for these old Carpenter processes that were originally created to help identify various types of Stainless Steels quickly without undertaking a chemical analysis may be worth replicating for the Forum. I recalled the Carpenter salesman, a decade later, suggesting that I keep the old catalog because they no longer provide printed process information in their new ones.

 

[unclesyd]

No matter which spot test you use always use a standard coupon with a fresh surface. Always test on a fresh surface.

 

[moltenmetal]

Again I caution against the use of chemical spot tests. Most 304 these days contains a fair bit of mixed stainless scrap, and therefore contains a little molybdenum- not enough to meet the 316SS spec, and not enough to provide the necessary additional corrosion resistance, but enough to possibly show up in a chemical colourimetric spot test.

If you really care and a single piece of misplaced 304 will cause a disaster, get a certified inspection company with a handheld XRF machine to come in and shoot every piece of pipe, every fitting and all welds which have been made with filler metal- this will tell you accurately the composition of your alloy as far as the metals go, but will not tell you if you have a carbon content problem. All other approaches require that you trust the fabricator, the fabricator's supplier, and the mill. Given that you claim to be replacing an incompetent, you may be forced into this approach. But I'd never select 316 in a service where a single piece of 304 in the piping system would cause a disaster- I'd go up at least one grade to be safe.

For future projects, choose a designer/fabricator you can trust, pay them adequately for the job so they're not forced to take shortcuts, insist on traceability and mill certs for all materials and review them in process, make sure they have an externally-audited QA program, and if you are still suspicious, shoot maybe 5% of the finished product with the XRF machine- and tell the fabricator up-front that you'll be doing it and he'll be responsible for any re-work (materials and labour). A good fabricator won't shy away from the challenge to get things right.

There was a time when both our customers and our company just wouldn't accept fittings from China (though we would from Taiwan) because we couldn't trust that the mill certs were valid and accurate. At the rate things are going, I doubt you'll have much choice BUT to accept materials from China unless you're willing to pay a very significant cost and possibly schedule premium.

 

[quadswift]

USJBH's comment reminds me.
I remember in the 60's, watching a guy in a steelworks who was sorting some billets that had got mixed when some rail wagons had shed there loads in a shunting mis-hap in the marshalling yard.

He had a small grinder and when he touched it on the billets he could segregate the materials by the type / colur and shape of sparks it gave off.
Would that be any use in this case?
Regards,
Quadswift

 

[unclesyd]

moltenmetal,
I've got to get on you a little concerning 304/316 S/S as a material of construction. We had a process that converted Adipic Acid(HOAd) to Adiponitrile(ADN) where the choice of 304 vs 316 was an absolute. HOAd was made from HNO3 oxidation of Cyclohexanol. 304L was the only material of construction for this conversion. All purification equipment after the HNO3 was removed could be either 304/304L. Once the material was dried it was conveyed to a melt tank which absolutely had to be 316 S/S. The corrosion rate of 304 S/S at melting point of the HOAd(162°C) changed from <.001 IPY to greater than >2000 IPY. The material of construction, mostly jacketed piping, for molten HOAd had to be 316 S/S until we reached 192°C, corrosion of the 316 S/S started. At this point we used Titanium until 222°C. Again another change to Hastalloy C until 238°C. Above 238°C the only material that was usable was pure Copper or Everdure 1010 (Silicon Bronze).
I can’t describe the process from this point on as you would come after me with a code book.

We used the spot test I described in my post to check every piece of S/S, even if marked, that went into molten HOAd service until the process was abandoned after 25 years.

 

[moltenmetal]

Unclesyd:

We agree there's a reason to know the difference between 304/L and 316/L. They are different alloys with different corrosion resistance to a wide variety of corrosive environments. There are cases where one will work and the other won't, certainly.

All I was saying was that if that 2.5% molybdenum is so important to you that one piece out of place can lead to a rapid failure, you have two choices:
1) to XRF every single piece before it goes in (because these days, the spot tests are anything but reliable, and mill certs are far from 100% even in the best of circumstances)- and then to XRF any manual welds for filler metal composition too, or
2) to go up a grade to, say, a 6 moly stainless, or a duplex grade (or whatever else will work, depending on your particular corrosive environment conditions). That will reduce the risk of an errant piece of 304 getting in there and wreaking havoc.

What a wild ride that plant must have been! Corrosion turning on like a light switch in such narrow temperature bands- it's remarkable that you were able to find materials which worked, period! A testament to your skill and experience!

 

[unclesyd]

azertyuiop,
How much piping and fittings are you talking about?

moltenmetal,

Can't take all the credit as there were 6 people in our group. We had the tools and we were able to use them with pretty good results over the years. The reason for the our shifts in direction was guided by trying to be cost effective. At the time (1960's) Titanium was around $25.00/# and C-276 was very limited in sizes and also very, very expensive. We didn’t have the alloys available today and we found out that additional Moly wasn’t very helpful.

A little anecdotal but this molten HOAd was fed by spinning Cu disks into the walls of a vaporizer made from Everdure 1010 (4' dia 7' tan/tan 1&quot; wall) with electricaly heated walls operating at 700°F. Superheated NH3 from an Inconel superheater at 900°F was feed into the bottom of the vaporizer and the resulting mixture was reacted over a catalyst at 800°F. The converter was heated with a direct gas fired heater. The outlet of the converter was piped to a condenser in Cu pipe with the condenser channel section and tubesheet along with 6&quot; of the 316 S/S tubes protected by Cu.
We had 17 of these units.