| Artikelinhalt |
- From freidus@wwnet.net Thu May 25 19:22:01 2000
Return-Path: <freidus@wwnet.net>
Received: (qmail 32061 invoked from network); 26 May 2000 02:22:00 -0000
Received: from unknown (10.1.10.26) by m4.onelist.org with QMQP; 26 May 2000 02:22:00 -0000
Received: from unknown (HELO ziggy.wwnet.net) (209.142.193.4) by mta1 with SMTP; 26 May 2000 02:22:00 -0000
Received: (qmail 11920 invoked from network); 26 May 2000 01:23:00 -0000
Received: from max5-179.pontiac.usabestnet.net (HELO ?209.142.225.179?) (209.142.225.179) by ziggy.wwnet.net with SMTP; 26 May 2000 01:23:00 -0000
X-Sender: freidus@mail.wwnet.net
Message-Id: <v04003a01b5538976feb2@[209.142.225.179]>
In-Reply-To: <17.610b494.265ef61b@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 25 May 2000 22:20:31 -0400
To: colonial-coins@egroups.com
Subject: XRF
From: Dan Freidus <freidus@wwnet.net>
I did some neutron activation analysis (NAA) on Higleys but I haven't done
any XRF.
Here's my impression of the major difference between the usefulness of XRF
and NAA. If anyone has any information or references that would help
further our decisions about which method to use (if any), that owuld be
great. I'm also working off the top of my head right now, so I won't
include references. I also don't own a copy of Volume 2 of the RNS
metallurgy series, so I do need to either buy a copy or stroll over to the
library soon and take it out.
XRF is a technique to see what elements are on the surface of a coin. I
believe that I have seen an esitmate of the thickness of the layer that
contributes to the signal. It was measured in microns. Anyone who has
ever cleaned a coin, whether with a thorn or any other tool, knows that it
is not unusual for coins to have a layer with significantly different
composition than the bulk of the coin. Even if there is nothing that can
be easily removed, the patina is the result of chemical reactions,
therefore the elemental composition of the surface layer has had some
changes. TO determine whether XRF is useful on uncleaned coins with
typical patination or those with significant encrustation, or dirty
surfaces, we need a study to determine how much these factors affect the
results. What would be needed are repeated measures of the same coin; as
found, then cleaned down to bare metal. Skip, if you have an X-Ray
fluorometer, I'd be glad to donate some coins for destructive testing
[*not* Higleys :) ].
NAA requires putting a sample in the path of a neutron source and then
reading the strenth and wavelength of radiation that is emitted from the
sample. THis can be done with entire coins (and there are soem good
reasons for doing this). However, the study I did using this technique was
done with Don Gordus at Unviersity of Michigan (we're lucky to have a
reactor on campus with an endowment, a holdover from "peaceful atom" days).
Don prefers to use samples of a few micrograms scraped from the edge of a
coin, which does require cleaning the sample area down to bare metal,
though that area can be a square millimeter on the edge. I think this
preference comes form having done most of his work in the past with silver
and gold coins. Certain elements in these alloys remain radiioactive
longer than the elements in copper and bronze coins. I once calculated
that if one did NAA on a few hundred coins form their collection they
would then need a federal license to store this now radioactive material.
Can XRF be done on microgram samples (Don collects them by scraping the
clean area with a sandblasted quartz rod and then using the rod as the
sample, or sample holder.
NAA on entire coins should be pretty good because the surface contributes a
small amount to the entire signal. NAA on scraped samples elimanates the
corrosion entirely. XRF ends up looking at the corroded layer exclusively
(i.e. the patina). How significant of an issue is that? If Skip or John
have an answer or anyone has a reference, great. Otherwise, I hesitate to
draw much inference from XRF data.
That said, I think all the various data sources need to be used, with the
constraints on their usefulness given full consideration. If the
interesting quesions had simple answers, we'd probably already have them.
I think we need to start with coins of known origin (at least down to
continent), e.g. state coppers and Regal halfpennies. Then see what
characteristics differ between these populations. Then look at the
unknowns and see if they neatly fall into one or another category. Some
probably will, I assume a large number will remain questionable. But that
is my vision of what progress would look like. It's probably a constraint
of my background as a scientist trying to do history or archaeology. Sure
artistic style could be part of the data, buyt my gut feeling is that there
were so many engravers that there was probably big overlap in the skill
level of engravers on either side of the pond. Was the average skill level
higher on the Eastern shore? Probably? But were there any decent
engravers on the Western shore? Yes, the state coppers show us that.
I agree with Mike in thinking that there are a lot of types of data that
could, and should, be used. I'm sorry that this turned out to be the
length of one of my columns (hmmm, maybe I just saved myself some time next
week...), but I think it's important for us to not latch onto one detail to
the exclusion of other useful factors.
Dan
|
|---|
| Quell-URL |
|
|---|
| Veröffentlichungsdatum |
|
|---|
| Volumen |
|
|---|
