Hi all, in this long and interesting discussion I think that Prof. Leinenweber has indicated where it is one of the problems in quantitative phase analysis. Regards,
-- ******************************** Jose M. Amigo http://www.uv.es/~amigo/ ******************************** > Interesting thought and probably worth following up, although it sounds like a meticulous and possibly time-consuming study for someone or other - any volunteers? :-) > Following Deane Smith's methodology you should be able to calculate the number of diffracting crystallites per phase for particular crystallite sizes and fraction (assuming some sampling volume - I vaguely remember his being 20mm^3 or something). I suppose that whether you're using a PSD or 2D detectors as opposed to point detectors and the geometry would also have to be taken into account. > > In our cement experiment the % difference between 0.1% and 2% periclase is quite large to say the least. In terms of practical versus theory, we did find that up to a point, the 'real' standard deviation from repeated experiments with micronized samples can be smaller than the esds calculated by the Rietveld program - a happy note for repeatability! On a less happy note that doesn't mean that something like severe microabsorption doesn't come along and mess up your accuracy. > > Pam > > -----Original Message----- > From: Kurt Leinenweber [mailto:[EMAIL PROTECTED] > Sent: May 30, 2008 11:11 AM > To: Whitfield, Pamela; [EMAIL PROTECTED] > Cc: [email protected] > Subject: RE: RES: ADS > > Hi all, > > This is an interesting discussion of quantitative phase analysis. It sounds to me from what Pam is saying that we can calculate an esd for each phase if we know its particle size, and that the relative error would be larger for phases in smaller abundance. If this is correct, then you could also repeat the physical experiment several times and get a variation of similar magnitude to the one that is calculated. Is this what happens? Pamela refers to the errors as being "worse" for minor phases, but they are what they are, and why not estimate them? > > - Kurt > > -----Original Message----- > From: Whitfield, Pamela [mailto:[EMAIL PROTECTED] > Sent: Friday, May 30, 2008 7:30 AM > To: [EMAIL PROTECTED] > Cc: [email protected] > Subject: RE: RES: ADS > > Hi Rheinhard > > The comments weren't aimed at you in particular! We've been banging our heads against a brick wall about this stuff for some time, but not many in the cement community seem to be listening. > > Anyway - there is a fundamental statistics problem with quantifying minor phases (of any symmetry) as the stats get worse with fewer diffracting crystallites. > The numbers that Deane Smith produced in terms of reproducibility of the quartz reflection dealt with a single phase system. It doesn't take much to imagine (or at least I do) that things rapidly get worse in a multi-phase system where theoretically every phase should consist of a nice random powder, reducing the required average particle size even further. It should follow then that the errors obtained with minor phases will be worse than major phases where the stats are better even without any other effects. Maybe my logic is flawed but I don't think so! Having said that it has been demonstrated that the situation isn't quite as bad practically as it is theoretically. > > Pam > > -----Original Message----- > From: Reinhard Kleeberg [mailto:[EMAIL PROTECTED] > Sent: May 30, 2008 10:10 AM > To: Whitfield, Pamela > Cc: [email protected] > Subject: Re: RES: ADS > > > Hi Pam, > maybe a misunderstanding: I did not try to discuss the amount of > "amorphous" material in cements. I'm aware of the problem, but a > careful Rietveld analysis with an internal standard should be able to > determine at least the magnitude of the amorphous part. Even if not > (e.g. if profile errors or structural uncertainties of the clinker > phases do prohibit a reliable quantification of the amorphous), a > systematic error of 7 instead of 5 wt% for a highly symmetric phase with > well resolved peaks like periclase is neither acceptable nor > unavoidable. And in my opinion a careful sample prep, measurement and > Rietveld analysis (maybe with internal standard) should be able to > perform better. I only tried to relativise Lubo's general denial of the > potential of quantitative phase analysis to reach an accuracy of about 1 > wt%. Maybe I'm simply more optimistic than Lubo ;-) > Reinhard > > Whitfield, Pamela schrieb: > > > Hi Reinhard > > > > Saying the words amorphous content and cement in the same breath is > > heresy in many circles! Personally I think that it's a case of 'see > > no evil, hear no evil'! > > Anyway, many moons ago we did yet another study on the amorphous > > content in clinkers/cements and they can vary alot - we found up to > > 20wt% in some. The amount probably depends on the kiln conditions, > > alkali metal content and quenching rate. > > Anyway don't take my word for it, Lerch and Brownmiller did a lovely > > bit of work (sadly forgotten) in 1937 using quenching in mercury (?!) > > and calorimetry yielding amorphous contents in clinker between > > 10-28%. The Ruland method has also been used to show significant > > amorphous contents in the various components of cement. > > > > In my opinion the cement companies are probably more interested in > > consistency rather than accuracy to track changes in their process. > > Case in point - strictly speaking the XRF samples should be micronized > > (or fused) as well as the XRD ones - the XRF errors from the matrix > > effects of a 30micron cement in a pressed pellet are not exactly ideal. > > > > I would agree that it's worth trying to push the boundaries of QPA. > > All is not lost for cements though - I saw the results of a test that > > Mati Raudsepp did with a very messy synthetic mineral mixture that > > looked completely ridiculous with the number of phases, but the result > > matched the weighing figures quite nicely. > > > > Pam > > > > ------------------------------------------------------------------ ---- > > -- > > *From:* Reinhard Kleeberg [mailto:[EMAIL PROTECTED] > > *Sent:* Fri 30/05/2008 5:36 AM > > *To:* [email protected] > > *Subject:* Re: RES: ADS > > > > Hi Lubo, > > the difference between 6 and 7 % MgO is bigger than 15 % relatively, > > thus the error by wrong re-scaling (e.g. if 10 % "amorphous" are > > missing) is smaller than the error discussed. And, if the "true" > > content of periclase should be about 5 %, than a measured value of 7 % > > is worth to be discussed to be a real systematic error. This is an > > over-estimation by 40 %. In the very past (30 years and more ago) > > people who applied single line methods carefully have been able to > > determine "simple" phases like free lime, quartz, anatase in low > > concentrations very accurate (of course after optimizing their > > technique), and I believe that such methods are still in use in > > routine product control. So I feel it is worth to think about why a > > Rietveld refinement should perform worse. I agree with you that the > > cement business is full of dubious advertisement for machines and > > methods for accurate QPA. But this is not a reason to finish thinking > > about how to improve QPA methods. Best > > Reinhard > > > > Lubomir Smrcok schrieb: > > > > > Hi, > > > I really wonder why do you bother about 1% difference when the error > > > of the method (XRD, quantitative phase analysis) could reach 10% > > > (absolute) ... > > > People in the discussions appearing here seem to forget about two > > things: > > > > > > i) "quantitative" phase analysis done by rietveld method is always > > > re-scaled to 100%, i.e. the accuracy of % attributed to the > > > individual phase directly depend on %'s attributed to the another > > > phases present in the mixture (i.e. when you forget or ignore one > > > phase having 10% share, those %'s are redistributed among the other > > > phases); > > > > > > ii) this method completely ignores any amorphous parts (i.e. those > > > not providing Bragg peaks, to be more exact) - ignores, because it > > > cannot do anything else about it. Obviously, when you compare XRF > > > and XRD you can find yourself in troubles as XRF dos not bother > > > about the state of the compounds. > > > > > > Finally, my personal advice. If someone tells you you can analyze > > > clinkers or even cements by XRD accurately, do not trust him. He, > > > most probably, thinks of selling you a machine you can hardly use > > > for this purpose :-) Bad luck if you have already bought it ... > > > > > > Best, > > > Lubo > > > > > > > > > > > > On Thu, 29 May 2008, [iso-8859-1] José Carlos Cordeiro wrote: > > > > > >> My samples are cements (fine samples with 30 microns) and don't > > >> need micronize its. The problem in my refinements is that periclase > > >> quantified with fixed slit is bigger than MgO by Fluorescence, and > > >> its impossible. Ex: > > >> > > >> > > >> > > >> Periclase with fixed slit: 7,14% > > >> > > >> MgO by XRF = 6,84% > > >> > > >> > > >> > > >> We know that part of MgO (1,5-2,0%) go to the structure of C3S an > > >> C2S phases, so periclase expected in this sample is 4,8% - 5,3%. If > > >> we consider this, the errors of refinement is not very slim. > > >> > > >> > > >> > > >> Sorry for my bad English!!! > > >> > > >> > > >> > > >> regards > > >> > > >> > > >> > > >> =========================================== > > >> > > >> José Carlos Cordeiro > > >> > > >> Diretoria Técnica - Laboratório Central > > >> > > >> Centro Técnico - Curitiba > > >> > > >> Rodovia PR-092, 1303, Abranches, 82130-570 Curitiba-PR > > >> > > >> Votorantim Cimentos > > >> > > >> [EMAIL PROTECTED] > > >> <mailto:[EMAIL PROTECTED]> > > >> > > >> www.votorantimcimentos.com.br > > >> > > >> Fone: + 55 041 3355-1380 > > >> > > >> Fax: + 55 041 3355-1358 > > >> > > >> =========================================== > > >> > > >> -----Mensagem original----- > > >> De: Whitfield, Pamela [mailto:[EMAIL PROTECTED] > > >> Enviada em: quinta-feira, 29 de maio de 2008 17:40 > > >> Para: José Carlos Cordeiro; [email protected] > > >> Cc: Mitchell, Lyndon > > >> Assunto: RE: ADS > > >> > > >> > > >> > > >> Hi Jose > > >> > > >> > > >> > > >> Errors in the region of 1wt% aren't unheard of in complex mixtures > > >> (and the liklihood of the second decimal place being meaningful is > > >> very slim). However there are other possible explanations and one > > >> in particular comes to mind. > > >> > > >> Did you micronize your sample? If yes then feel free to ignore the > > >> following! If not then it might be worth thinking about. > > >> > > >> > > >> > > >> We did a systematic study on the effect of particle statistics on > > >> the repeatability of quantitative analysis on one of the NIST > > >> cements last year (presented at the ICCC in Montreal - should > > >> really go somewhere more accessible). We compared different > > >> divergence slits, micronized and unmicronized samples, spun and > > >> static, and remounted/repeated the experiments 3 times each for > > >> real statistics as opposed to esds. > > >> > > >> > > >> > > >> Funnily enough the phase that showed the biggest anomaly in one of > > >> the unground sample was the periclase - obviously a big (by XRD > > >> standards) lump fell into the sample from the kiln that made up the > > >> NIST batch and made it through the mill at the plant. Without > > >> grinding it we got 0.1, 0.1 and 2wt% one series of 3, versus 0.6, > > >> 0.6 and 0.6 wt% (this is from memory so don't don't quote me!) for > > >> the micronized sample. > > >> > > >> This was the result that we should have got theoretically but it > > >> was nice to see it work in practice, and the periclase did such a > > >> nice job for us by sticking out like a sore thumb even without a > > >> Rietveld analysis! BTW there was no statistically significant > > >> difference (95% > > >> limits) in the results of the C3S and C2S phases between ground and > > >> unground samples - mostly because the errors in the unground samples > > >> were so large! The minor phases were another matter. In the > > >> micronized samples the choice of divergence slit size made little > > >> difference, but with the unground samples they made a significant > > >> difference. > > >> > > >> > > >> > > >> To cut a long story short it all comes down to particle statistics > > >> - with an ideal powder the choice of fixed versus variable slits > > >> (or even different fixed slits) should make no difference (assuming > > >> that the Rietveld software can handle the difference in profiles). > > >> Moral of the story - sample prep, sample prep and more sample prep! > > >> > > >> > > >> > > >> Pam > > >> > > >> -----Original Message----- > > >> From: José Carlos Cordeiro [mailto:[EMAIL PROTECTED] > > >> Sent: May 29, 2008 3:53 PM > > >> To: [email protected] > > >> Subject: ADS > > >> > > >> I have a Programmable Divergence Slit (incident beam) in my > > >> difratctometer, that can work in automatic mode (irradiated length) > > >> or fixed mode (0.5 degree or 1.0 degree etc). What is the best > > >> collect mode for Rietveld Refinament?? > > >> > > >> > > >> > > >> I did refinement two difractograms collected in booth > > >> configuration (fixed 0.5 degree and automatic mode), the results in > > >> the fase "Periclase" (cement fase) changed, see below: > > >> > > >> > > >> > > >> Periclase with 0.5 degree = 6.08% > > >> > > >> Periclase wich automatic mode = 7.14% > > >> > > >> > > >> > > >> Note.: The refinement of the difractograms collected with > > >> automatic mode went converted to fixed mode by software. > > >> > > >> > > >> > > >> > > >> > > >> > > >> > > >> What do you think about this? > > >> > > >> > > >> > > >> regards > > >> > > >> > > >> > > >> =========================================== > > >> > > >> José Carlos Cordeiro > > >> > > >> Diretoria Técnica - Laboratório Central > > >> > > >> Centro Técnico - Curitiba > > >> > > >> Rodovia PR-092, 1303, Abranches, 82130-570 Curitiba-PR > > >> > > >> Votorantim Cimentos > > >> > > >> [EMAIL PROTECTED] > > <mailto:[EMAIL PROTECTED]> > > >> > > >> www.votorantimcimentos.com.br > > >> > > >> Fone: + 55 041 3355-1380 > > >> > > >> Fax: + 55 041 3355-1358 > > >> > > >> =========================================== > > >> > > >> > > >> > > >> > > > > > > > > >
