mel jacobson on thu 7 sep 00
>Date: Thu, 7 Sep 2000 19:43:07 -0400 (EDT)
>From: Tom Buck
>To: Mel Jacobson
>Subject: Murray's Borate analyses (fwd)
>
>mel:
> mary simmons is having problems (as I am) trying to get the
>following posted on Clayart. It could be local ISP problems.
>please post this if you can. tia. peace. tom.
>
>Tom Buck ) tel: 905-389-2339
>(westend Lake Ontario, province of Ontario, Canada).
>mailing address: 373 East 43rd Street,
> Hamilton ON L8T 3E1 Canada
>
>---------- Forwarded message ----------
>Date: Thu, 24 Aug 2000 20:34:32 -0600
>From: mary simmons
>To: Ceramic Arts Discussion List
>Cc: Tom.Buck@hwcn.org
>Subject: Murrays Borate analyses (long)
>
>Hey Folks-
>
>I've been gone most of the summer, so I am only now getting to answer the
>question some of you asked, which was:
>
>Why is the analysis I posted to the list so different from the one that
>came from Kickwheel Pottery?
>
>I don't know.
>
>But I do know that the Kickwheel analysis reported no LOI. LOI, for those
>of you not in the know, is literally "loss on ignition", which in plain
>English means: all the stuff that burns off in the kiln--such as water,
>CO2, sulfur etc.
>
>I have a material data sheet sent to me by Kickwheel, which states that
>Murray's Borate contains calcite. Calcite is also known as calcium
>carbonate, limestone, whiting....., it's chemical formula is CaCO3. During
>firing, calcium carbonate decomposes to calcium oxide plus carbon dioxide:
>
>CaCO3 + heat = CaO + CO2
>
>CO2 as we all know is a gas, so it just goes away into the kiln
>atmosphere. Same with water, which in clay (and some other minerals)
>exists as (OH)--known as "hydroxl", but is driven off in the kiln
>atmosphere as well.
>
>Therefore, without actually testing for LOI, we know that there SHOULD be
>some, because the material contains calcite.
>
>The problem with NOT reporting the LOI is that the analysis ends up being
>inaccurate--because the original analysis is "normalized" to equal 100%,
>which kicks up the concentration of everything else, by distributing the wt
>% of the LOI over all the other oxides.
>
>There may also be some water in whatever boron mineral was used in Murray's
>borate--naturally occurring borates do.
>
>Now, I am NOT trying to cast aspersions on Kickwheel Pottery, just trying
>to help my fellow potters deal with gerstley borate withdrawals.......I
>believe that the Kickwheel folks are fine upstanding citizens, so I am
>leaning toward picking on the analysis used........I am extremely lucky to
>have access to the high-end analytical equipment of this University
>
>Chemical analyses can be done, using several techniques. I use the famed
>Whole Rock Chemical Analysis via X-Ray Fluorescence, which in a nutshell,
>is a process where a sample of material is bombarded with x-rays, which
>causes electrons to jump out of their shells, and as they move from higher
>to lower energy levels, emit energy (X-radiation). This
>shell-jumping produces a characteristic spectra as the electrons leap out
>of their shells and cascade toward the nucleus--they fluoresce as they do
>this, hence the name of this analytical method. The spectra are resolved
>into wavelengths that are specific to the elements that produced them--ie
>each element will have a unique wavelength. The intensity and position of
>the spectra is recorded by an X-ray counting device, and sent to a computer
>where the intensity and wavelength data is resolved into weight % of major
>element oxides, or as ppm for trace elements.
>
>As you might well imagine, it is unusual for a person to own an X-Ray
>Fluorescence machine, because they are rather pricey--the one here in the
>geology department at the University of New Mexico is about 3 years old and
>was well over $150,000 bucks (American). It was paid for by several
>departments, but lives here in geology (yippppeeeee!)--we do a lot of work
>for those other departments!!!
>
>Moreover, to prepare the samples, which are first ground to 100-mesh
>powder, one needs to have access to a furnace as well as platinum crucibles
>and saucers--which are several thousand dollars EACH! The sample is mixed
>with lithium tetraborate, which is a very strong flux, and then melted in a
>platinum crucible at 1000C and poured into a platinum disc. As it cools,
>it contracts and you can pop the glassy disc out of its platinum
>saucer. We call this a "fused disc". Alternatively, one could mix a
>large amount of sample with a smaller amount of borax, and press the heck
>out of it--enormous pressure fuses the sample material to the flux, to make
>what is called a "pressed powder"
>
>BTW--to get LOI, you heat a weighed sample (with no flux) to 1000C and
>weigh it again--the difference in weight loss gives you the LOI. AND, you
>have to do this in those outrageously expensive platinum crucibles. When
>you have high LOI, such as with GB and GB substitutes, you also have to
>account for the loss of mass in the furnace, where the water, CO2 etc burn
>off. The discs are made with specific weight ratios of sample to flux, so
>if you are getting a 12% loss of mass before the disc ever hits the XRF,
>then the whole analysis has to be adjusted to take this into account.
>
>Major elements (eg SiO2, Al2O3, CaO, K2O, Na2O, MgO, MnO, TiO2, Fe-oxides,
>P2O5) are more accurately analyzed using the fused disc method. Trace
>elements (eg Rb, Sr, Zr, etc) are more accurately analyzed using pressed
>powders.
>
>In order to be certain of one's analysis, it is necessary to run a
>"standard" along with the samples--there are 6 slots on the XRF where you
>can place a sample to be analyzed--one space is for the standard--which is
>sample of a material with a known composition--for the pottery materials I
>am interested in, I generally use a standard that is a granite. So, we
>know the composition of the standard, and with each run, we analyze the
>standard again, to ensure that the machine is giving us good numbers for
>the samples we are testing. If the wt %'s reported for the standard differ
>from what we know it is supposed to be, we know that the other analyses are
>probably also off--in which case, the Senior Scientist of the Universe has
>to make repairs, and we test the sample again. It is also possible to make
>errors in weighing the sample and flux--which would also give spurious
>results.
>
>The major element analysis I reported for Murray's Borate was done on a
>fused disc, with a standard. The wt %'s reported for the standard were
>within 0.4% of it's known composition.
>
>Therefore, I believe in my analysis.
>
>As I always admit, I report the B2O3 wt % to ya'll, as the difference
>between all the major element oxides + LOI and 100--which is why my
>analysis comes out to a perfect 100%. This is not the usual in the real
>world. Our machines are not THAT accurate, to give us perfect
>100%'s. Boron is hard to get on the XRF, because our methods involve a
>flux that contains it as well.
>
>I've been given a few samples of borate minerals (colemanite, ulecite,
>kernite....) and I am going to make standards of these, so that we CAN
>analyze for boron. I do this because I am a geek. In truth, the
>arithmetic method I use to get B2O3 is most likely within 1% of the actual,
>which, persnickety though I am about accuracy, is within acceptable
>tolerances for pottery, I think (we are NOT talking colorants here!!!).
>
>I have not analyzed any of the GB or GB substitutes for trace elements,
>though we know they are there--but in VERY small quantities--we report
>trace elements in parts per million, which are TEENY! I do not know how
>such small concentrations can affect a glaze, but I am inclined to think
>NOT MUCH! However, one of these days, I plane to mix up a glaze recipe
>using pure oxides, and compare that to the same recipe using the usual
>culprits--Custer feldspar, whiting, flint, etc. THEN we'll see just how
>much those trace elements affect things.........
>
>So, to come back to the original question about why my analysis is
>different from the one provided by Kickwheel:
>
>I don't know. Their analysis did not contain all the major element oxides,
>which are indeed there, so I might guess that the analytical method used
>was not XRF--so the analysis was incomplete, but normalized to equal 100.
>
>The XRF analytical machinery is hugely expensive, and I am fortuneate to
>have access to this, at a ridiculously low fee. Commercial outfits have to
>charge a lot more to pay for the equipment and the person who keeps the
>machine running accurately--which is not me!!!!! --THAT requires much more
>understanding of the XRF machine than I have.
>
>I've tried to provide some light as to why these analysis are so different,
>which has to be in the analytical process used. I hope that someone from
>Kickwheel will tell us how their Murray's Borate was analyzed.
>
>Meanwhile, I'd have more faith in my analysis than theirs, if ONLY because
>I reported a rockin' 12.41% LOI and they reported none, though they say
>that calcite is a constituent of Murray's Borate.
>
>cheers, oh my fellow geeks who have made it to the end of this diatribe-
>Mary
>
FROM MINNETONKA, MINNESOTA, USA
http://www.pclink.com/melpots (website)
| |
|
