Very strange question for the experts: 1

[v20102]

Hello all,

Thank you in advance for any information.

I have a strange question and found this forum, so I am hoping that you can help me.
I work with museums and collectors in the US and Europe in the authentication of military and other artifacts (swords, helmets, etc.).

We have an issue with items made of steel and other metallic materials.
The issue is this: many museums, collectors, etc. are looking for a way to authenticate artifacts based on their metallic composition. This is because of a large number of post-1945 reproductions flooding the market.

For many of these items (European swords for example) we know the source of the steel (Solingen Germany).

Would it be possible to use (or develop) a non-distructive test that could be used on artifacts "in question" to compare their metallic composition to that of known period Solingen steel?
I.E. a "fingerprint" of the steel composition.

If not, is there some other way to make this determination? (assuming that the same forging methods were used in both pre and post 1945 periods)?

Anything at all would be helpful!

Thank you in advance!
John

 

[eyec]

not sure but you may want to look at an alloy analyzer. it will give you the material properties in percentages of elements present.

 

[metengr]

My advice would be to contact several material/metallurgical testing labs that have the capability to perform positive material identification (PMI), which requires the use of a portable alloy analyzer and inquire about your project. The x-ray fluorescence units will determine qualitative amounts of alloy elements present. This information must be compared to a known steel artifact composition to determine a comparative match. Be aware, there are significant limitations with the alloy analyzer, for example no element below sulfur can be detected (e.g., carbon content).

 

[enmax]

Would a check for radioactivity help here? Some years ago I recall some manufacturers of gamma radiation monitors seeking pre-nuclear age steel since all steel since the start of atmospheric testing contained radioactive contaminants which interfered with the calibration of the monitors. I would think the contaminants must be at a very low level so the checking instrument would need to be very sensitive.

http://uk.linkedin.com/pub/terry-roberts/28/a9a/a10

 

[v20102]

Thank you all!

I have looked into alloy analyzers and they seem ideal. I have access to several museums who could provide unquestioned original pieces for testing in order to create a database for comparison.

My concerns are only:
1) Is the analyzer accurate enough to detect minute differences in steel "source" composition?
2) Many of these artifacts date to the early 20th century. Was steel at this time refined to such an extent that there may be not enough other allows present for detection or accurate comparison?

Emax- You read my mind. I have heard this as well. In fact, I posted this exact question in the nuclear engineering section of this very website… no responses yet.
If you now the tools/methods by which such measurement of such radioactive contaminants could be measured, I would be very interested.
Also, is this actually a fact? My sourse was merely here-say.

Thank you all!!!!!
John

 

[v20102]

PS: Sorry for the 2 spelling errors (allows/alloys and now/know), typing too quickly.
I also cannot seem to find a way to edit my post and thus stem my embarrassment.

 

[blacksmith37]

You will need to "fingerprint" a few known samples from Soligren (and other known sources). I think you will find the old steels will use higher Mn then modern steels. Also the Europeans were more likely to use V alloy. The trace contaminates from the ore are likely to help. Modern steels can be exceptionally low in S and P; Also the ratio of S to P may help. Al "fine grain practise" is likely to be a marker.
Sounds fun, you are likely to need an experienced (old) metallurgist.

 

[MikeHalloran]

About editing: You can't.

However, if you Red Flag your message and leave a clearly worded request in the dialog box that appears, the site's magic bunnies will make reasonable edits.



Mike Halloran
Pembroke Pines, FL, USA

 

[IRstuff]

But, even if you COULD determine age based on composition, that would merely push the counterfeiters into the next level of complexity, much life they've done with counterfeit documents. With documents, "washing" a worthless document from a particular era and putting age-specifc and composition inks for a more expensive document is par for the course.

Likewise, one could imagine that anyone sophisticated enough to counterfeit the look and feel of a turn-of-the-century metallit item would simply push on to duplicating the composition and heat cycles to fool the analysis.

TTFN
faq731-376

 

[v20102]

IR Stuff - Thank you for your post and, good point. I have been thinking about this as well.

One key to success would be maintaining a "trade secret" database of steel composition from known original pieces. Without this information, the "unscrupulous" would need to invest in this testing equipment themselves, gather their own independent data, and then commission a steel manufacturer to duplicate the composition. Sounds expensive...

I believe that currently, this additional expense would make new pieces (able to fool the proposed analysis) too costly to justify.

I do concede however that should the values of these pieces continue to rise, in another 20 years values may reach a point that would justify the "unscrupulous" taking these additional steps…

In any case, at least this method could put us "1 step ahead"… for now.

 

[v20102]

One other thought…

Many of these pieces are made of several parts, which in turn are made from different metals (blades from steel, but the crossguards, pommels, etc. are made from other alloy mixtures.

If the alloy analyzer could accurately measure the composition of these pieces as well, then the "unscrupulous" would have a much bigger challenge in creating a fake which could beat the analysis.
Not only would they need to get the blade metal right, but also the other fittings.

This might buy us another 10 or 20 years.

 

[EdStainless]

V
I have been associated with some archaeological metallurgy work.
With good modern hand held x-ray fluorescence analyzers you should be able to do most of what you have in mind.
You will need to do surface prep in order to get good readings. But we are talking an area of less than 1/2".
You don't care how fast it is so you can allow long analysis times and get more accurate readings on trace elements.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[SnTMan]

v20102, interesting post.

Regards,

Mike

 

[v20102]

EdStainless - Thank you for your post!

I was unaware that any type of preparation is necessary?
What would this involve?

In many cases, people do not want these pieces to be handled unnecessarily, so I am hoping for as little "touchie feelie" as possible.

I have contacted a few manufacturers and hope to hear something soon. They too however are probably puzzling over my intended application.

All the best,
John

 

[v20102]

Also, how long a time period are you referring to with "long analysis times"?
Although, this should not be a serious issue.

 

[metengr]

v20102;
Please do not take the following advice as criticism. As I stated earlier and will repeat it again, you need to have this work performed by a competent metallurgical lab. Unless you know the limitations of the machine and fully understand preparation requirements for a clean surface, you will obtain erroneous results.

 

[Wrenchbender]

Of course there are exceptions, but here are some generalizations that might help in screening for counterfeits or in building a database of ferrous materials.

Impurity levels:
Steel >100 years old will contain more impurities such as P and S due to 'dirty' processing (such as Bessemer or open-hearth methods), which are not as 'clean' as today's (electric arc or induction furnace melting).

Good points above in post # 22Jan12, 15:42. Knowledge of the authentic steelmill's melting practice is desirable - what alloying elements (if any) were typically used; and did they 'kill' the melt with a few shovels full of sand (Si) or use Al instead. All these would be easily detectable.

Due to the thermochemistry of steelmaking, certain impurity elements such as Cu and Sn (I believe), cannot be removed economically. As scrap is recycled, these levels have risen over the last ~75 years.

Radioactivity:
Not sure, but I do know that today's domestic steel production uses a huge amount of Fe scrap which itself may be pre-1945. Also radiocarbon dating would only set the age of the carbon added to the melt. (e.g., millions of years old.) Still feel it should be pursued.

Non-ferrous metals:
JOM, a publication of the TMS,

http://www.tms.org,

once had a good article concerning ancient metals. I don't recall the detail but you might check there.

 

[EdStainless]

A spot on the surface would need to be cleaned, or else what ever was applied (or happened to end up) to the surface will be included in the results. Fine grit sandpaper is the most common method.

Long time for the new machines is a minute or two. The new machines actually contain miniature x-ray tubes and they can be tuned to optimize the results for various elements.

If you were in the neighborhood I would have you drop by and show you how this would work.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[swall]

Following up on what Wrenchbender said, today's steels, especially electric furnace melted steels, will have higherlevels of residual elements like chrome, nickel, vanadium, etc. than the older steels.

 

[mrfailure]

Following up on EdStainless' comment: Specifically, surface oxides alter the results; this is why such a technique requires surface preparation. This is fine when working on a part in the field, but you should be concerned regarding how this would mar the surface.

I hate to be a naysayer but I have little faith in using PMI for this purpose. It can work if the alloy composition is so different from the original alloy that you would see a definite difference. However, some problems include:

-More recent alloys may be similar or not have a meaningful difference in composition.

-PMI is a semi-quantitative technique. It is good for identifying alloys, but cannot be used, for example, to determine if the alloy is in spec

-PMI cannot identify carbon, sulfur, or phosphorus contents.

Aaron Tanzer

http://www.lehightesting.com