Meteorite Dagger?

Jeff Pringle · 5th August 2006, 02:33 AM

I think it could be done, but it would take some work & money to get to where you can make it ‘definite’.
For example, say I send you a few grams of metal from a blade I just made, 50% meteorite, 50% home-smelted steel. You send that to a lab and have trace element analysis done, you have one data point. It would be a good data point, ‘cause I can tell you which meteorite I used, and you could compare it to the unforged meteorite, there are usually reams of trace element data available on meteorites - now you got two points, which show how the meteorite material is diluted out.
But wait, the blade I did is non-homogenous (as are most keris), you really ought to analyze it more than once, or at least make sure your lab is taking that into account.
Now you check a keris (Datum three? Four?), if it’s old, they weren’t using meteorites from AZ or Argentina, so you have a different element pattern, perhaps – you could study up on element fractionation in the cores of asteroids to figure out which elements are significant, but at this point I’m starting to wonder what non-meteoric kerisses have in them, so you better send a sample of one of them out for testing – there’s another data point (nickel could be a discriminator, but terrestrial nickel is used, too).
Go through that process enough times, you’ll have a great way to graph out any blade, and it will drop into the ‘extra’ or the ‘terrestrial’ field, with an acceptable level of certainty.
But at this time, no one has done that (& published), so we can only speculate – Iridium should be high, that’s how they started thinking the K/T boundary layer was meteorite-related, and that’s got to be more diluted than a keris, it’s geologic in scale

!
Some study of the significant element ratios would be in order before you start, so you can cut lab costs, or just bet on iridium and go for it. I think a couple of different element ratios would be better in the long run, though – you’d be able to sort out which meteorites went into which blade, eventually, and avoid unscrupulous iridium smiths (I don’t really think they exist, yet!).
The best place for asteroid paper research:
http://adsabs.harvard.edu/abstract_service.html
Or, get a planetary scientist to take an interest in the project – but they are really swamped in actual meteorites now, so they might be too busy for this esoteric pursuit.
And if you want a piece of this blade’s metal for testing, look out for the positive Boron anomaly – I did the welding with borax flux

A. G. Maisey · 5th August 2006, 02:58 AM

Thanks Jeff.

I think you just confirmed what everybody else who understood the question had already said. Not in quite the same words perhaps, but the end result is the same.

What it comes down to is that there is no economically viable way to obtain a positive ID on meteoritic content in a keris.

There may be a way to determine if there is meteoritic content, but even that is not certain.

Testing would of course need to be non-destructive.

If the value of the object is $X, and its value might rise to $2X if it can be definitely proven to contain meteoritic material, its not really a proposition to carry out tests that could run to $X to the power of 10.

I reckon we`re just about back where we started as far meteors and keris go:- believe it if you will, its an item of faith, and who can criticise another`s faith?

Jeff Pringle · 5th August 2006, 03:43 AM

Quote:

there is no economically viable way to obtain a positive ID on meteoritic content in a keris

Well, if we can find someone with a large collection of keris, and convince them to drop their current career and get a job as a microprobe tech....

I should point out that I'm just a smith who occasionally ponders on cosmochemistry, I don't necessarily know what I'm talking about; I'm pretty sure the above method would be the way to do it, though.

BSMStar · 5th August 2006, 04:13 AM

Just for clarification (for the Electron Microprobe)...

1. Concentration of meteoritic material in a say a Keris, can be important. As long as it is not present is minute amounts, the trace elements should still be detectable.

2. How homogenous is the mix? If it is not very homogenous, multiple samplings made have to be made.

3. Identification of cosmic origin won’t be a problem. Every known meteorite type has been “fingerprinted” many times over. The traces are established and well known.

4. Sampling… the area of polish will be about a 1/8 inch by 1/8 inch square (not a bad size blemish)… but the whole sample (blade) has to fit in the microprobe… that maybe an issue for large samples without removing a piece of the blade. Not my first choice.

5. Since it was stated that a method could not be advised or suggested, and no one could tell us how to do it… with no rules of engagement for the sample, I just wanted to state that it is possible. But I do not believe that I suggested this would be desirable or easy… just possible. It becomes a soul search at this point; as to how bad do you want to do it and why.

A. G. Maisey · 5th August 2006, 04:58 AM

Quite frankly BSM, I don`t want to do it at all.

I believe I can mostly identify meteoritic material by recognising the origin, method of construction, and probable maker of a blade, and then using the "touch test"---the stuff does feel a little bit different to other materials used as pamor.

I`ve been posing the question all these years because I`ve sometimes thought it might be a good academic exercise to put together a sampling of blades that recognised Javanese experts considered contained meteoritic material, and seeing just how good the indicators that have been used were.

The mix is not at all homogenous. Its layers upon layers upon layers. You could probably test 20 different spots before you struck a square 1/8 inch that actually had meteoritic material in it. On the types of blades that I know contain meteoritic material I could not imagine any keris fancier approving the polish of even one section of 1/8 by 1/8 inch.You probably could ID a section of material that waslikely to contain meteoritic material though. That would probably reduce the number of tests needed.

A normal Javanese keris blade is going to run about 16 inches, including the tang.

But tell me this:- those trace elements that you would be looking for:- are they still going to be there after the material has gone through many, many weld heats?

Jeff Pringle · 5th August 2006, 06:34 AM

Quote:

those trace elements that you would be looking for:- are they still going to be there after the material has gone through many, many weld heats?

I believe so, they track meteorite contamination of the lunar surface by looking at things like PGE (platinum group elements), which are not likely to oxidize away before iron in a welding operation.
Iron meteorites are full of the non-volatile stuff that didn't burn off during the birth of the solar system, and was too heavy to float away from the center of the new-born planets/asteriods - I bet a few minutes at 2300 F would have little effect.

Thanks for that very good synopsis of the ebay auction copy, BSMStar!

I'm curious about what levels of these trace elements are in commercial steel, now...time to do some research...

Jeff Pringle · 5th August 2006, 04:28 PM

Perhaps a better way -
Take a couple months going through the various reference material standards for steels on the web, pulling out the info on the few for which there is trace element data.
Do the same for iron meteorites.
Pick a couple trace elements that aren't typically manipulated in steel production, say Ag, Au, As, Ge, Ta, Zr, Hf, La; the earth steels should plot in one area (or along a single line) when comparing two elements, meteoric metal in one or more areas outside or overlapping - the 'right' pair of elements should allow good discrimination.
You might be able to get the same result by comparing Ni to Co, or P to S; but since those are manipulated elements they might not graph well.
Get a keris tested, it should fall between the two areas, if it's a mix of the two metals - and you've only paid for one lab test!
You'll have to put in some hours with the Excel program's chart function, though

Bonus, you could publish an article in a scientific journal or popular magazine once you're done!
http://www.nist.gov/srm

Jens Nordlunde · 5th August 2006, 04:50 PM

Hi Jeff,

For one who knows something about what you describe, it sounds plausible that the way you describe it will work very well – I don’t know anything about it

. So I will have to ask a question. How can you be sure that the two metals are mixed so well that they will both be represented in the little piece you take out for testing?

5th August 2006, 02:33 AM	#1
Jeff Pringle Member Join Date: Nov 2005 Posts: 189	I think it could be done, but it would take some work & money to get to where you can make it ‘definite’. For example, say I send you a few grams of metal from a blade I just made, 50% meteorite, 50% home-smelted steel. You send that to a lab and have trace element analysis done, you have one data point. It would be a good data point, ‘cause I can tell you which meteorite I used, and you could compare it to the unforged meteorite, there are usually reams of trace element data available on meteorites - now you got two points, which show how the meteorite material is diluted out. But wait, the blade I did is non-homogenous (as are most keris), you really ought to analyze it more than once, or at least make sure your lab is taking that into account. Now you check a keris (Datum three? Four?), if it’s old, they weren’t using meteorites from AZ or Argentina, so you have a different element pattern, perhaps – you could study up on element fractionation in the cores of asteroids to figure out which elements are significant, but at this point I’m starting to wonder what non-meteoric kerisses have in them, so you better send a sample of one of them out for testing – there’s another data point (nickel could be a discriminator, but terrestrial nickel is used, too). Go through that process enough times, you’ll have a great way to graph out any blade, and it will drop into the ‘extra’ or the ‘terrestrial’ field, with an acceptable level of certainty. But at this time, no one has done that (& published), so we can only speculate – Iridium should be high, that’s how they started thinking the K/T boundary layer was meteorite-related, and that’s got to be more diluted than a keris, it’s geologic in scale ! Some study of the significant element ratios would be in order before you start, so you can cut lab costs, or just bet on iridium and go for it. I think a couple of different element ratios would be better in the long run, though – you’d be able to sort out which meteorites went into which blade, eventually, and avoid unscrupulous iridium smiths (I don’t really think they exist, yet!). The best place for asteroid paper research: http://adsabs.harvard.edu/abstract_service.html Or, get a planetary scientist to take an interest in the project – but they are really swamped in actual meteorites now, so they might be too busy for this esoteric pursuit. And if you want a piece of this blade’s metal for testing, look out for the positive Boron anomaly – I did the welding with borax flux Attached Images

5th August 2006, 04:50 PM	#8
Jens Nordlunde Member Join Date: Dec 2004 Location: Europe Posts: 2,718	Hi Jeff, For one who knows something about what you describe, it sounds plausible that the way you describe it will work very well – I don’t know anything about it. So I will have to ask a question. How can you be sure that the two metals are mixed so well that they will both be represented in the little piece you take out for testing?

5th August 2006, 02:58 AM	#2
A. G. Maisey Member Join Date: May 2006 Posts: 7,048	Thanks Jeff. I think you just confirmed what everybody else who understood the question had already said. Not in quite the same words perhaps, but the end result is the same. What it comes down to is that there is no economically viable way to obtain a positive ID on meteoritic content in a keris. There may be a way to determine if there is meteoritic content, but even that is not certain. Testing would of course need to be non-destructive. If the value of the object is $X, and its value might rise to $2X if it can be definitely proven to contain meteoritic material, its not really a proposition to carry out tests that could run to $X to the power of 10. I reckon we`re just about back where we started as far meteors and keris go:- believe it if you will, its an item of faith, and who can criticise another`s faith?

5th August 2006, 04:13 AM	#4
BSMStar Member Join Date: Jan 2005 Location: Kansas City, MO USA Posts: 312	Just for clarification (for the Electron Microprobe)... 1. Concentration of meteoritic material in a say a Keris, can be important. As long as it is not present is minute amounts, the trace elements should still be detectable. 2. How homogenous is the mix? If it is not very homogenous, multiple samplings made have to be made. 3. Identification of cosmic origin won’t be a problem. Every known meteorite type has been “fingerprinted” many times over. The traces are established and well known. 4. Sampling… the area of polish will be about a 1/8 inch by 1/8 inch square (not a bad size blemish)… but the whole sample (blade) has to fit in the microprobe… that maybe an issue for large samples without removing a piece of the blade. Not my first choice. 5. Since it was stated that a method could not be advised or suggested, and no one could tell us how to do it… with no rules of engagement for the sample, I just wanted to state that it is possible. But I do not believe that I suggested this would be desirable or easy… just possible. It becomes a soul search at this point; as to how bad do you want to do it and why.

5th August 2006, 04:58 AM	#5
A. G. Maisey Member Join Date: May 2006 Posts: 7,048	Quite frankly BSM, I don`t want to do it at all. I believe I can mostly identify meteoritic material by recognising the origin, method of construction, and probable maker of a blade, and then using the "touch test"---the stuff does feel a little bit different to other materials used as pamor. I`ve been posing the question all these years because I`ve sometimes thought it might be a good academic exercise to put together a sampling of blades that recognised Javanese experts considered contained meteoritic material, and seeing just how good the indicators that have been used were. The mix is not at all homogenous. Its layers upon layers upon layers. You could probably test 20 different spots before you struck a square 1/8 inch that actually had meteoritic material in it. On the types of blades that I know contain meteoritic material I could not imagine any keris fancier approving the polish of even one section of 1/8 by 1/8 inch.You probably could ID a section of material that waslikely to contain meteoritic material though. That would probably reduce the number of tests needed. A normal Javanese keris blade is going to run about 16 inches, including the tang. But tell me this:- those trace elements that you would be looking for:- are they still going to be there after the material has gone through many, many weld heats?

5th August 2006, 04:28 PM	#7
Jeff Pringle Member Join Date: Nov 2005 Posts: 189	Perhaps a better way - Take a couple months going through the various reference material standards for steels on the web, pulling out the info on the few for which there is trace element data. Do the same for iron meteorites. Pick a couple trace elements that aren't typically manipulated in steel production, say Ag, Au, As, Ge, Ta, Zr, Hf, La; the earth steels should plot in one area (or along a single line) when comparing two elements, meteoric metal in one or more areas outside or overlapping - the 'right' pair of elements should allow good discrimination. You might be able to get the same result by comparing Ni to Co, or P to S; but since those are manipulated elements they might not graph well. Get a keris tested, it should fall between the two areas, if it's a mix of the two metals - and you've only paid for one lab test! You'll have to put in some hours with the Excel program's chart function, though Bonus, you could publish an article in a scientific journal or popular magazine once you're done! http://www.nist.gov/srm