[ATM]

I've got fifteen early 20th century Chinese coins that come in two varieties: copper or brass. I'm trying to tell them apart. That ought to be simple (brass should be yellow), but in these cases I'm not sure - they are maverick. Copper or bronze coins can occasionally turn partly yellowish when exposed to the elements. (The brass versions are the scarce ones, by the way).

Here's what I did: I grinded a tiny part of the edge to see what they were like under the patination. They were all sparkling peach-coloured. I figured they would turn darker through regular oxidation. So I exposed them to hydrogen-peroxide, with minimal effect. Then I rubbed them in my armpit, which had an effect ( ) - but I still can't say that they are clearly one thing or the other.

I refuse to give up (considering that this issue will arise again). So do any of you have a bright idea? 

I should add that there probably were no officially prescribed compositions - or if there were, they probably weren't implemented. I suppose it boils down to testing for the presence of zinc. I'm a non-chemist.

[DemonicAcid]

I should add that there probably were no officially prescribed compositions - or if there were, they probably weren't implemented. I suppose it boils down to testing for the presence of zinc. I'm a non-chemist.

Yup.

[ATM]

Thank you, DemonicAcid. I've heard of a method called electrical conductivity test. By this method, the the composition can be estimated. So, I was thinking: 1) can I do such a test at home (accepting a modest investment in equipment), and 2) are there any alternative (easier or cheaper) methods that could be employed to detect the presence of zinc? I hope somebody can help - I've known for years that I needed to find an answer to this question sooner or later...  (It's a recurring issue).

[Borek]

I've heard of a method called electrical conductivity test. By this method, the the composition can be estimated. So, I was thinking: 1) can I do such a test at home (accepting a modest investment in equipment)

Somehow I doubt it will be possible. I guess it is based on precise measurement of the specific resistance of the alloy, I doubt it can be done without destroying coins. Problem is, you can measure some resistance, but the number you get depends on the coin geometry and size of contact between coins and electrodes used for measuring. Especially contacts will be highly variable leading to irreproducible results.

The only sure way I can think of would be some spectroscopic method - but these are expensive.

[ATM]

Thank you very much for elaborating on this, Borek! I'm not so worried about whether or not the conductivity test can be used, because it has been demonstrated that it can be successfully applied in this field. If there are no simpler alternatives ( ), I'm going to go for that one. I found this info on non-destructive conductivity testing. The test would be used to compare coins where all specifications are the same, except alloy.

I was thinking of something else - perhaps you geeks know the answer  . Suppose the test works and we have a coin that we know consists of an alloy of specified elements. Would the conductivity of the alloy be equal to the weighted average of the conductivity of the elements in the alloy?

[Stepan]

When you measure conductivity be sure that the distance between probes is fixed: I would glue two needle probes into a holder with epoxy. Similar probe and conductivity technique is used to measure moisture content in wood. Be sure that your conductivity meter is sensitive enough. And probe conductivity is sufficiently higher that the one of the coin otherwise you will be measuring conductivity of the needle probe.   

From time to time you will get very low conductivity. This can be due to cracks, cavities, heavy internal corrosion, and nonuniform composition of the metal.

Another test to confirm the composition of the alloy would be a metal density measurement. The density of the coin can be measured with high accuracy. There is 5-8% difference between copper and brass - you will have no problem to detect it.

The both tests are nondestructive, and can be used simultaneously to verify the results.   

[Borek]

When you measure conductivity be sure that the distance between probes is fixed: I would glue two needle probes into a holder with epoxy. Similar probe and conductivity technique is used to measure moisture content in wood.

When using it for wood you can be sure contact surface is large as you are in fact putting electrodes deep into the wood, at least that was the case with the probe I have seen, it required use of a hammer. When applying to coin contact surface will be way too difficult to predict - at least that's what intuition tells me (intuition trained on determination of solution resistances).

[Stepan]

When using it for wood you can be sure contact surface is large as you are in fact putting electrodes deep into the wood, at least that was the case with the probe I have seen, it required use of a hammer. When applying to coin contact surface will be way too difficult to predict - at least that's what intuition tells me (intuition trained on determination of solution resistances).

This days, the wood probe is miniature and is basically two d=1.5 mm nail-like electrodes on the plastic handle. Contact is not the problem. Problem is very low resistance he needs to measure.

[ATM]

Thank you for advice! I have now asked for a copy of the article in which the metal conductivity test (applied to coins) was described. So, hopefully we'll find out exactly how they did.

Regarding the metal density measurement, would it be applicable also in other situations, e.g. where two specified alloys exist, and a particular coin should be attributed to one or another of those alloys? Could I carry out such a test with a device on my own desk?

[Stepan]

Generally speaking you should be able to distinguish two alloys if density difference is significant. Good lab can probably distinguish a 0.1% difference ($1000 investment. For semi-professional user - I would say 2% density difference is doable on instrument budget of $100-500 dollars.

You basically need scales.

[Stepan]

Thank you for advice! I have now asked for a copy of the article in which the metal conductivity test (applied to coins) was described. So, hopefully we'll find out exactly how they did.

Regarding the metal density measurement, would it be applicable also in other situations, e.g. where two specified alloys exist, and a particular coin should be attributed to one or another of those alloys? Could I carry out such a test with a device on my own desk?

Sorry, I did not finish last reply. You need scales with readability of 0.001 mg and preferably with "under-the-scales weighing" capability. I offer similar scales from my web site: http://www.lcsairtest.com/laboratory-equipment/balances.html

The same manufacturer offers professional density measuring kit with scales and all accessories. If there is an interest - let me know.