Re: [Ifeffit] Parameters for parent function in wavelet transform analysis

[Scott Calvin]

I think I finally have the answer!

The short version is that only the product of eta and sigma affects analysis 
using Morlet wavelets. It makes absolutely no difference whether an analysis 
uses sigma = 0.5 and eta = 10.0 or if it uses sigma = 1.0 and eta = 5.0.

Formally, though, the wavelet transforms can take different values in the two 
cases, because the most commonly used normalization depends on eta and sigma 
separately, and not just their product. But since the same normalization will 
be used on two different measurements (different samples, different locations 
on a sample, or a sample at different times), or on a sample and a theoretical 
model, it has no practical impact. The scale on a contour plot of the wavelet 
transform will change, but that scale isn’t even included in some publications, 
because it’s essentially arbitrary. That’s especially true since there are 
different normalization schemes out there: 
https://cp.copernicus.org/preprints/cp-2019-105/cp-2019-105-supplement.pdf 
<https://cp.copernicus.org/preprints/cp-2019-105/cp-2019-105-supplement.pdf>.

I arrived at these conclusions by examining the mathematics. Can anyone who has 
actually worked with Morlet wavelets confirm that’s the behavior? I could 
imagine that some piece of software might implement wavelets in such a way that 
different combinations of sigma and eta that have the same product might 
nevertheless yield very slightly different results just because of 
computational effects (e.g. differences in interpolation and such).

Best,

Scott Calvin
Lehman College of the City University of New York

> On Aug 6, 2022, at 7:29 AM, Xinyu Luo  wrote:
> 
> Hello Scott,
> 
>   Thanks! I actually have the same question. I'm a graduate student. I 
> learned the wavelet transform from the website of ESRF (and I learned XAFS 
> from your book!). I can share my opinions on this question.
> 
>   There is a mother wavelet called cauchy [Muñoz, M.et al, Am. Mineral. 2003, 
> 88 (4), 694], which seems to be pretty nice because only one parameter used 
> to manipulate the resolution---n. And that's where I started. After reading 
> more literatures and I was taught that wavelet analysis can be more powerful 
> if the mother function looks more like to the path of interested. So 
> probably, the two parameters of Morlet can be used to ,mimic the path of 
> interest generated by FEFF. I wrote an email before to ask about how to 
> realize the steps described in a literature[Funke, H. et al, J. Synchrotron 
> Radiat. 2007, 14 (5), 426]. You can see that they customized a mother 
> function on the base of FEFF to surpass the resolution of Morlet. If I 
> understand it correctly, the uncertainty limit, though which is a law in 
> physics, can't be compared on different Mother functions. And go back to your 
> question, perhaps you can an optimized choice of sigma and eta with a fixed 
> (sigma)(eta). BUT I DON'T KNOW HOW TO GET THAT. It's clumsy to try a bunch of 
> value and I hope someone can teach me on how to choose a starting value of 
> sigma, or eta.
> 
> 
> 
> Best
> 
> Xinyu
> 
> 
> 
> -Original Messages-
> From:"Scott Calvin" 
> Sent Time:2022-08-06 01:41:17 (Saturday)
> To: ifeffit@millenia.cars.aps.anl.gov
> Cc: 
> Subject: [Ifeffit] Parameters for parent function in wavelet transform 
> analysis
> 
> Hi all,
> 
> I’ve never actually tried wavelet transform analysis before, so I’m trying to 
> understand it better, but there’s one aspect that’s puzzling me:
> 
> The Morlet wavelet has two parameters, eta and sigma. Eta controls how fast 
> it wiggles, and sigma how quickly the function dies out. (Essentially, it’s 
> like the basis function for a Fourier transform multiplied by a Gaussian 
> envelope.)
> 
> In wavelet analysis we pick a parent function, which is often a Morlet 
> wavelet with particular values of eta and sigma that we have chosen. 
> 
> We then create a set of child functions from the parent function by shifting 
> and dilating the function in k-space.
> 
> Each child function is then used as a basis function for a transform of 
> chi(k) calculated by integrating over a range of k-values, much like what is 
> done to calculate a Fourier transform. Since the Morlet wavelet is localized 
> in k-space, and the child functions are shifted to focus on different regions 
> of k-space, our wavelet transform produces plots which are a function of k. 
> But since dilating the parent function by different amounts yields child 
> functions with different frequencies, the plots are also a function of R. 
> Therefore the result is a two-dimensional (k and R) contour plot.
> 
> So far, so good. 
> 
> But what I’m wondering about is the effect of the initial choice of e

[Ifeffit] Parameters for parent function in wavelet transform analysis

[Scott Calvin]

Hi all,

I’ve never actually tried wavelet transform analysis before, so I’m trying to 
understand it better, but there’s one aspect that’s puzzling me:

The Morlet wavelet has two parameters, eta and sigma. Eta controls how fast it 
wiggles, and sigma how quickly the function dies out. (Essentially, it’s like 
the basis function for a Fourier transform multiplied by a Gaussian envelope.)

In wavelet analysis we pick a parent function, which is often a Morlet wavelet 
with particular values of eta and sigma that we have chosen. 

We then create a set of child functions from the parent function by shifting 
and dilating the function in k-space.

Each child function is then used as a basis function for a transform of chi(k) 
calculated by integrating over a range of k-values, much like what is done to 
calculate a Fourier transform. Since the Morlet wavelet is localized in 
k-space, and the child functions are shifted to focus on different regions of 
k-space, our wavelet transform produces plots which are a function of k. But 
since dilating the parent function by different amounts yields child functions 
with different frequencies, the plots are also a function of R. Therefore the 
result is a two-dimensional (k and R) contour plot.

So far, so good. 

But what I’m wondering about is the effect of the initial choice of eta and 
sigma for the parent function.

The dimensionless product of eta and sigma has a clear effect. If (eta)(sigma) 
is small, the parent function will not have very many oscillations of 
significant size; it it’s large, it will have a lot of oscillations. So if we 
want good resolution in R (and thus relatively poor resolution in k), we want a 
big value of (eta)(sigma). In the limit as (eta)(sigma) becomes arbitrarily 
large, we recover the Fourier transform. If we want good resolution in k, we 
use a small value for (eta)(sigma). In the limit as (eta)(sigma) becomes 
arbitrarily small we recover chi(k).

That still makes sense to me!

But what difference do sigma and eta make individually? In other words, how is 
sigma = 0.5 and eta = 10.0 different from sigma = 1.0 and eta = 5.0? Since we 
end up creating child functions that dilate the parent function anyway, I can’t 
see that it should have any effect at all. 

And if those two parameters don’t make an independent effect, why do we pretend 
there are two independent-looking parameters? 

I’ve seen at least one early paper that suggests choosing eta so that it’s 
close to the path length you’re most interested in probing, but I can’t see how 
that actually makes any difference, given the dilation.

I expect there’s a good chance I’m missing something obvious, or have a 
fundamental misunderstanding of part of the process, and look forward to 
learning more!

Best,

Scott Calvin
Lehman College of the City University of New York

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] EXAFS Fitting with 2 CIF files

[Scott Calvin]

Hi Zihao,

Do it by folding the molar ratio in to the amp parameter for each CIF. Of 
course S02 is part of that parameter, too.

So the amp for PtO2 might be parameterized as S02 * fracoxide and the amp for 
Pt3Fe as S02 * (1-fracoxide)

You could either assume the S02 is the same for the two phases (as I implicitly 
did by writing it with only one S02 variable), or you could use a standard for 
each (from a sample prepared and measured in a similar way) to fix the S02 for 
each phase.

Best,
Scott Calvin
Lehman College of the City University of New York

> On Sep 4, 2021, at 4:02 PM, Zihao Yan  wrote:
> 
> Dear Ifeffit members:
> 
> I have a sample which is a mixture of PtFe and PtO2. In order to do the EXAFS 
> fitting, I imported 2 CIF files on Artemis and generated the following 3 
> paths:
> From PtO2: Pt-O @ 2.07 nm
> From Pt3Fe: Pt-Pt @ 2.736 nm, Pt-Fe @ 2.736 nm
> Then, I used these 3 bonds to do the fitting as in the online tutorial (which 
> used 1 CIF file). The fitting result is good in terms of R^2 but it is not 
> scientifically correct.
> 
> I think I didn't set up a parameter which determines the ratio between 
> CIF(PtO2) and CIF(Pt3Fe). I mean there should be a parameter that regulates 
> how much each CIF contributes to the fitting or the molar ratio between PtO2 
> and Pt3Fe in my system.
> 
> Could someone please give me some suggestions on how to do EXAFS fitting with 
> 2 CIF files? Is there anywhere I can set up the molar ratio between PtO2 and 
> Pt3Fe?
> 
> Yours sincerely,
> Zihao Yan
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] EXAFS fitting - noisy data

[Scott Calvin]

Hi Yang,

There are many reasons why you would prefer your first-shell fit to have your 
fit include high-k data, when feasible. I’ll name one, just to give you the 
idea:

Suppose you have a substance where you have a good estimate of S02 
(transferability from a similar substance measured under the same conditions), 
but you don’t know the first-shell coordination number or sigma^2. Coordination 
number and sigma^2 are correlated, since they both primarily affect the 
amplitude of chi(k), but they are not completely correlated, as the 
coordination number has the same effect across the entire k-range, wherease 
sigma^2 has a much greater effect at greater k (in the EXAFS equation, it’s 
weighted by k^2). So extending the fit up to higher k reduces the correlation 
between coordination number and sigma^2, which can be very important to 
understanding the structure of your material.

Of course, there are other ways to reduce that correlation, such as by 
measuring a series of spectra at different temperatures. But it does provide a 
good example of why high-k information is useful when you can get it.

Best,

Scott Calvin
Lehman College of the City University of New York


> On Aug 30, 2021, at 4:32 PM, Hu, Yang (HIU)  wrote:
> 
> 3)For fitting the 1st and 2nd shells, I still lack of clear understanding 
> how the high-k portion can influence. If I measure a set of samples, and one 
> of them has much noisy data so a shorter k-range is picked up for background 
> subtraction and FT. In this case, can I still consider the change or 
> evolution in the fitted parameters “systematic”?  
> 

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] smoothing XAS data

[Scott Calvin]

Not only can smoothing XANES data attenuate sharp features, it can also shift 
them, since many XANES features, particularly in the “pre-edge” region, are 
significantly asymmetric about their peak. That can make analysis quite 
confusing, since, e.g., the size of the boxcar ends up affecting the position 
of the peak!

Best,
Scott Calvin
Lehman College of the City University of New York

> On Jan 22, 2020, at 8:44 PM, Carlo Segre  wrote:
> 
> I generally don't like to smooth XANES data since smoothing does tend to 
> attenuate sharp features such as the ones which exist near the absorption 
> edge.

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Two analysts, one spectrum, similar outcome?

[Scott Calvin]

Yes, I did.

I set up a double-blind experiment with mixtures of various iron standards
and an “unknown” iron-containing compound. This was years ago, so I may
have a few details wrong, but it will get the gist:

 I used undergraduates to make the mixtures with random amounts of random
selections of the standards. An undergraduate also ordered the “unknown”
compound.

We then measured the spectra, along with the spectra of the pure standards
(but not the pure “unknown”). The spectra were measured at somewhat
different temperatures so that simple linear combination analysis would be
of limited use. Three of us then attempted analysis independently: myself,
one of my most advanced undergraduates (who did not participate in the
preparation of the samples), and a high school student with roughly two
weeks training in analysis. We attempted to find what compounds were
present, how much of each, and, in the case of the unknown compound, as
much as we could suss out structurally.

The results were that we all did OK, but my analyses were the most
accurate, the advanced undergraduate less so, and the high school student
the least (but still generally in the ballpark).

It was satisfying to know that different analysts could find similar
results, that those results reflected reality, and that those with greater
expertise did achieve more accurate results.

If I recall correctly, I didn’t publish because the undergraduate who
prepared the pure standards did a lousy job (pinholes, etc.), which
distorted our measurement of the standards enough to make it more difficult
to evaluate our analyses. Still, it showed things work in principle.

The data, including a remeasured set of standards, is still available as
the EXAFS Divination Set <http://xafs.org/EXAFS_Divination_Set>.

Best,

Scott Calvin
Lehman College of the City University of New York

On Tue, Aug 13, 2019 at 9:41 PM Mike Massey  wrote:

> Hi Everyone,
>
>
> I'm curious, has anyone ever tried turning two analysts loose on the same
> unknown EXAFS spectrum to see if their fits come out with similar
> conclusions? If you have tried it, how did it work out? Were the
> conclusions indeed similar? If not, why not, and what did you end up doing
> about it?
>
> I was talking with a colleague today about our plans for data analysis,
> and we settled on this approach (since there are two interested parties
> willing to try to fit a series of unknown EXAFS datasets).
>
> The hope is, of course, that the two analysts will independently reach
> similar conclusions with similar fits and structural models, but to my mind
> that outcome is by no means guaranteed. Given the (presumably) wide
> variation in fitting customs and procedures, I can envision a scenario in
> which there are major differences.
>
> This got me wondering, "Has anyone tried this?" So I thought I'd ask.
>
>
> Your thoughts and experiences would be welcome. Thanks!
>
>
>
> Mike Massey
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit
>
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Advice on using Athena for XANES LCF on spectra with a small edge step?

[Scott Calvin]

Hi Mike,

Is there a reason the data stops at 2200 eV—e.g., is there another edge just 
above that?

—Scott

> On Jan 15, 2019, at 8:02 PM, Mike Massey  wrote:
> 
> Thanks Riti and Carlo.
> 
> Riti: will give it a go, thanks! I think the data are salvageable, it's just 
> a question of how...
> 
> Carlo: Interesting point. In terms of how I'm fitting it, I just use the LCF 
> window to do a two-component fit. Nothing super fancy.
> 
> But what you're suggesting is that maybe I have a misunderstanding in the way 
> I'm normalizing my data. That's entirely possible. I generally try to get a 
> nice flat background by following the "flat part" of the spectra but in cases 
> like this the pre- and post-edge lines end up not parallel. I can force it to 
> be parallel, and that helps the fit to behave better, but ends up creating 
> other normalization issues.
> 
> 
> Much obliged for the responses so far.
> 
> 
> 
> Mike
> 
> 
> 
> 
>> On Jan 15, 2019, at 4:42 PM, Carlo Segre  wrote:
>> 
>> 
>> Hi Mike:
>> 
>> How are you trying to fit it?  It looks to me like the post edge line is not 
>> in the right place.  You need to tune it by hand when you have such a short 
>> energy range after the edge since the normalization is critically dependent 
>> on how you set this line.
>> 
>> carlo
>> 
>>> On Tue, 15 Jan 2019, Mike Massey wrote:
>>> 
>>> Hi All,
>>> 
>>> 
>>> Does anyone have advice for LCF of spectra with a small edge step?
>>> 
>>> I've recently found myself with more and more issues fitting spectra like 
>>> the one shown in the attached photo, where the spectrum is of good quality, 
>>> but the fits end up not working out due to normalization (as far as I can 
>>> tell).
>>> 
>>> Sadly, I have so far ended up just throwing these fits out, but I'm hoping 
>>> there is a better solution out there that someone might be able to 
>>> recommend.
>>> 
>>> One solution is "get more of the element of interest," as a beamline 
>>> scientist recently told me. I'm just reluctant to really load it up in the 
>>> data collection, because at 2 keV I get nervous about artifacts like 
>>> self-absorption.
>>> 
>>> Is there a better way? What am I missing?
>>> 
>>> Thanks for your thoughts,
>>> 
>>> 
>>> 
>>> Mike
>> 
>> -- 
>> Carlo U. Segre -- Duchossois Leadership Professor of Physics
>> Interim Chair, Department of Chemistry
>> Director, Center for Synchrotron Radiation Research and Instrumentation
>> Illinois Institute of Technology
>> Voice: 312.567.3498Fax: 312.567.3494
>> se...@iit.edu   http://phys.iit.edu/~segre   se...@debian.org
>> ___
>> Ifeffit mailing list
>> Ifeffit@millenia.cars.aps.anl.gov
>> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] the meaing of "beta=125.264" in path interpretation

[Scott Calvin]

You’ve figured it correctly—it’s the scattering angle.

Best,
Scott Calvin
Lehman College

> On Oct 19, 2018, at 10:22 AM, woschangqi...@126.com wrote:
> 
> Dear,
> 
> Firstly, i'm very grateful for the useful responses from Bruce Ravel and Matt 
> Newville on my previous question.
> 
> Now, I have another question about path. The screenshot below is the 
> interpretation of a path in α-Fe. I was wondering what does "beta" means. Is 
> that scattering angle? I tentatively construct a triangle according to the 
> rleg values, and find that the three angles are consistent with the beta 
> values. What does this mean? I searched the Artemis user guild and the 
> mailing list, but could not find the answers. Do you have any ideas?
> 
> Thanks in advance
> 
> 
> 
> 
> 
>>> Qiang Chang
>>> Institute of Coal Chemistry, CAS
>>> Leyuan South II Street 1, Yanqi Development Zone
>>> Huairou District, Beijing, POB. 101407
>>> P.R. China
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov <mailto:Ifeffit@millenia.cars.aps.anl.gov>
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit>
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/options/ifeffit>
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Constraining values in Artemis fit

[Scott Calvin]

Hi Andrew,

I’d add to this that it’s important to think about what it is that you are 
trying to figure out. At least with the systems that I’ve worked with, it is 
rare for the answer to that being “I would like to know the coordination number 
for this path.” Instead, it’s usually something like “I would like to know what 
fraction of the manganese ions are in sites octahedrally coordinated to oxygen 
as opposed to tetrahedrally coordinated,” or “I would like to know which of 
several possible ligands are attached to my central atom, and in what 
proportions.”

Those questions can yield to clues aside from the coordination number alone. 
For example, the metal-oxygen bond lengths are significantly different for 
octahedral coordination than for tetrahedral coordination. If you’re trying to 
distinguish between ligands, sometimes you can include paths beyond the 
nearest-neighbor, and so on.

I’ve often seen people who are first trying to learn EXAFS analysis think that 
the way to analyze their system is to first see what information they can get 
from a single nearest-neighbor path from a single sample under a single set of 
conditions. If they can’t get a precise result from that, they figure anything 
more complicated will yield even fuzzier results. But if you have some 
knowledge about your system, or about possible alternative models of your 
system, then that’s not the case. Comparing samples in some way, or employing 
more scattering paths, or measuring multiple edges and performing a combined 
fit, can reduce the uncertainties in fitted parameters.

In your example of coordination number and sigma^2, one possibility is to 
measure the same sample at different temperatures. Of course, if you expect the 
sample might undergo a phase change or otherwise change structure with 
temperature, that won’t work very well. But if you expect the structure to be 
fairly stable over a range of temperatures, then the coordination number might 
be the same for all measurements, while the sigma^2 varies. A multiple-spectrum 
simultaneous fit can then help.

Of course, every system, and every problem, is different. But I do think it’s 
important to think about your system and about what the scientific problem is 
that you’re trying to solve.

Best,
Scott Calvin
Lehman College of the City University of New York

> On Aug 23, 2018, at 12:13 PM, Bruce Ravel  wrote:
> 
> On 08/23/2018 10:55 AM, Thomas, Andrew (AGW) wrote:
>> Hello,
>> Does anyone have any suggestions for breaking the correlation between the 
>> coordination number and sigma^2? Specifically, is there a way to set a 
>> maximum or minimum value for parameters in a fit?
> 
> Matt is, of course, correct when he says that this correlation is inherent to 
> EXAFS analysis and cannot be "broken".  I would like to suggest a broader way 
> of thinking about this sort of issue.
> 
> When you have a very well behaved analysis problem -- that is, something like 
> my teaching example of FeS2, something for which the structure is pretty well 
> known -- then your EXAFS analysis will yield defensible values for CN and 
> sigma^2 without too much effort.  They will still be correlated, but sensible 
> numbers will tend to just fall out.
> 
> In a harder problem -- y'know something you are doing actual research on -- 
> you often run into the situation where these correlations preclude a "fall 
> right out" analysis.  It is tempting to assert that the CN must be SOMETHING 
> and the ss must be SOMETHING.  That's true in a sense, but you have made a 
> real measurement and are doing a real analysis with real measurement 
> uncertainties.  And you have to respect that.
> 
> When you are running into trouble in your EXAFS analysis -- big 
> uncertainties, indefensible values, that sort of thing -- that is usually the 
> program trying to tell you something about your analysis. Usually, that would 
> mean that you want to know something that the data do not support (or do not 
> support beyond some level of precision or accuracy).  Or it might mean that 
> your fitting model is not realized in the data -- that is, your model is 
> missing some important feature and the bad fit results are the result of 
> missing that important feature.
> 
> To summarize, I want to encourage you not to assert that you need to find a 
> way to break the correlations.  i want to encourage you think about what the 
> wonky fit results are trying to tell you about your data or your fitting 
> model.
> 
> 
> That was kind of rambling, I admit.  Hopefully it was helpful nonetheless
> 
> B
> 
> 
> -- 
> Bruce Ravel   bra...@bnl.gov
> 
> National Institute of Standards and Technology
> Synchrotron Science Group at NSLS-II
> Lead Beamline Scientist, 06BM (BMM)
&

Re: [Ifeffit] Fitting Problems in Artemis version 0.9.26

[Scott Calvin]

Dear Huan,

Neither of those sound like a flaw in Artemis to me. 

1) Fitting with a range that goes below Rbkg means fitting to a part of the 
data’s Fourier transform that has been suppressed by background subtraction, 
even though the theory’s Fourier transform has not been similarly suppressed. 
That will yield a misleading fit. If the newer version of Artemis does a check 
to prevent this, that’s actually an improvement.

2) This could always happen—you may just not happen to have seen it before. 
Using abs() doesn’t make the variable you are fitting positive; it makes the 
result which is used in the fit (the “path parameter”) positive. So if, for 
example, Artemis reports ss1 = -0.002, that means it used +0.002 in the fit.

Best,
Scott Calvin

> On Jun 28, 2018, at 10:03 PM, LIU HUAN  wrote:
> 
> Dear Dr. Ravel,
> 
> I want to ask you that when I use the new version "Artemis 0.9.26" I have two 
> problems. (1) The fitting Range was reported to be a error that the Rmin 
> should be larger than the Rkbg value. It makes that the fitting range cannot 
> be smaller than 0.9 (the Rkbg value is 0.9). (2) When Fitting the 
> Debye–Waller factor  (ss) and coordination number (N), although I use "abs()" 
> to make it to be positive, but the fitting results still be negative. But I 
> did not meet these problems in the old version in Artemis 0.8.012. I am 
> wondering that whether could you tell me how to solve these problems. Thank 
> you so much.
> 
> Sincerely,
> Huan
> 
> Huan Liu
> School of  Earth Sciences and Engineering 
> Nanjing University 
> 163 Xianlin Road, Nanjing, Jiangsu, 210023 
> P. R. China
> 
> Tel  86-15861809073
> Email: huanliu_ea...@hotmail.com <mailto:huanliu_ea...@hotmail.com>
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov <mailto:Ifeffit@millenia.cars.aps.anl.gov>
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit>
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/options/ifeffit>
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Amplitude and sigma square

[Scott Calvin]

Hi all,

> On May 29, 2018, at 11:35 AM, Carlo Segre  wrote:
> 
> Finally, the answer to your second question is that you do NOT need to fit 
> longer paths if all you want to know is the first shell information.

As a rule of thumb, of course, this is true. But there are two situations I 
think are worth noting, even though I think neither applies to Abdul’s 
particular example.

One is with a well-characterized system for which highly accurate first-shell 
information is desired; e.g. bond lengths within a hundredth of an angstrom or 
so. Outer shells can contribute some in the Fourier transform in the range 
being used to fit the first shell. The effect is likely to be modest, and for 
some purposes can be considered negligible, but when very high accuracy is 
desired it may be relevant.

The other is when the long-range structure is roughly known, but distortions 
relative to the model structure are expected. For example, substitutional 
doping might leave a crystal with the same overall structure, but with both 
long-range differences (e.g. Vegard’s law type stuff) and local distortions. In 
these cases, including paths beyond the first shell constrained in a physically 
defensible manner can help reduce the correlations between fitted parameters 
and lead to smaller uncertainties in the fitted parameters. For example, the 
mixed-metal ferrite systems I’ve worked with were like this.

Best,

Scott Calvin
Lehman College of the City University of New York

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] boundary issue in artemis

[Scott Calvin]

Hi Marcus,

From the graph, it looks likely that one of your sigma2’s is negative, causing 
the fit to blow up at high k. 

If so, then the apparently good fit below k=9 is actually a “false fit,” 
meaning that it looks good on a graph, but is nonphysical. Unfortunately, that 
means you have more work to do in adjusting the model; you shouldn't just 
narrow the fitting window and ignore the misfit.

Best,
Scott Calvin
Lehman College of the City University of New York

> On Jan 17, 2017, at 4:21 AM, mfehse  wrote:
> 
> Dear all,
> 
> im fitting Ti next neighbour environment and I'm facing the problem that 
> above a k value of 9 the fit deviates completely and does not follow the data 
> any more. I dont really understand what is happening, so im forced to limit 
> my fitting window in order to avoid this effect.
> i have attached a plot in k space so you can better understand what im 
> talking about.
> 
> any hints would be 
> appreciated___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] about sigma2 for exafs fitting

[Scott Calvin]

Hi Shaofeng,

Thanks for providing a thorough description of what you did in the Word file—it 
makes it much easier to comment!

I wouldn’t say that sigma2 is not important, in this case. I would say, 
instead, that your fits have not yet successfully distinguished between the 
different proposed models, although some of those models yields significantly 
different values for sigma2.

Here’s my summary of what I’m seeing in your file and from your description:

The Hamilton test doesn’t indicate the differences between fits are 
statistically significant.

The Ba3(AsO4)2 model yields E0’s a bit smaller than the corresponding standard 
(although with overlapping error bars) and sigma2’s significantly larger. The 
larger sigma2 could just indicate it’s somewhat more disordered. I certainly 
can’t reject this model on the basis of the information you’ve provided, but it 
seems a little less likely than the others.

The BaHAsO4*H2O model yields parameters almost identical to the corresponding 
standard. Yay! The results are therefore consistent with the sample being 
BaHAsO4*H2O.

The C2 model again yields parameters that are a bit different from the fits to 
the standards. The sigma2’s are, as you mentioned, small. The E0’s are also 
large compared to the standard, although the error bars overlap.

This raises, for me, a question: did you run a fit with the BaHAsO4*H2O 
standard using the C2 model? I know that’s not the structure you expect for the 
standard, but it would be very nice to know if using that model with the 
standard also yields small sigma2’s, big E0’s, and smaller R-factors. If that 
happens, then you’d gain confidence that you had BaHAsO4*H2O, but you also 
wouldn’t have any evidence that the C2 model was a better description of your 
samples than just the regular BaHAsO4*H2O structure. If, on the other hand, the 
C2 model doesn’t yield a similar set of results for the standard as it did for 
the sample, then you’ve got some more investigating to do.

Does that make sense?




On Jan 8, 2017, at 6:53 AM, Shaofeng Wang 
mailto:wangshaof...@iae.ac.cn>> wrote:

Dear Scott and demter users,

I also tried to use BaHAsO4 crystallographic data as the initial model for 
fitting of the EXAFS data. Please see the attached file. It was found the 
sigma2 was around 0.0012-0.0014. The Hamilton tests also show the improvement 
was not significant (> >0.05). But thease results could be considered to be 
better than using DFT optimized Configuration C2 as the initial model. This 
reuslt may imply the sigma2 is strongly denpendent on the initial model. Does 
this mean the sigma2 is not so important to determine the fitting quality?

Regards,

Shaofeng

--
Shaofeng Wang, Ph.D of Geochemistry
Environmental Molecular Science Group
Institute of Applied Ecology, Chinese Academy of Sciences
Shenyang, 110016, China
wangshaof...@iae.ac.cn<mailto:wangshaof...@iae.ac.cn>
www.iae.cas.cn


From: Scott Calvin<mailto:scal...@sarahlawrence.edu>
Sent: Friday, January 06, 2017 3:34 AM
To: XAFS Analysis using Ifeffit<mailto:ifeffit@millenia.cars.aps.anl.gov>
Cc: Shaofeng Wang<mailto:wangshaof...@iae.ac.cn>
Subject: Re: about sigma2 for exafs fitting

Hi Shaofeng,

The parameter b does not have to do solely with the difference in N_var between 
the two fits! As I say in the book,

“Changes in the theoretical standards used by the model can often be accounted 
for as if they were changes in parameters. For example, if fit A assumed that 
the nearest neighbors were oxygen and included coordination number, bond 
length, and MSRD for those oxygen atoms as free parameters, while fit B assumed 
the nearest neighbors were sulfur and used the same three parameters, then a 
Hamilton test could be applied with the assumption that those three parameters 
were “changed”—instead of applying to oxygen, they now apply to sulfur. If E0 
and S02 were also free parameters, they would not be considered to be changed, 
as they are primarily a property of the absorber, not the scatterer.”

In your case, it looks to me like two or three parameters are changed between 
models (I count four free parameters in your table, one of which is E0, so I’m 
not sure why you say there are only three…is there another constraint?) Suppose 
there are two free parameters changed. b = 2/2 = 1.

x is the ratio of R-factors between the fits.

Using the values you list below, then, gives a result of 0.64, which indicates 
such variation is quite likely to have taken place by chance.

You chose the worst case of your three, though; the pH 3 case. The pH 10 case 
shows the most improvement in the R-factor with x = 0.007/0.012 = 0.58. The 
probability of getting that much improvement by chance drops to 0.30—less 
likely, but still pretty close to a coin flip.

Even the fact that you got all three to show improvement doesn’t quite reach 
the 95% confidence level, although it’s becoming suggestive.

W

Re: [Ifeffit] Ifeffit Digest, Vol 167, Issue 11

[Scott Calvin]

Hi Raj,

The kind of thing I gather you did for #1 is almost always a bad idea. Guessing 
parameters in one fit and then setting them to the results  for that fit in 
subsequent fits without further justification is asserting a certainty about 
those parameters that you do not possess. Occasionally it might be useful in an 
intermediate fit as a kind of diagnostic, but never in fits from which you want 
to extract information.* Note that the fit in which you set S02 is a different 
kind of constraint; rather than using the result of a previous fit, you used an 
a priori plausible value. While that can introduce its own biases, it at least 
is based on external information (a “reasonable" value for S02) rather than 
trying to do the fit in pieces (I’ll run a fit, find values for guessed 
parameters, and then set those to the first results for the next fit).

Fundamentally, you just don’t have a good fit yet. Likely, something about the 
system is a bit different than you envisioned it. But that’s part of the point 
of using EXAFS, right? For fits to tell you that, and then you to use your 
knowledge about the system to try out plausible modifications to the model 
until you find something that works well, hopefully also supporting it with 
other lines of evidence? 

That’s not something we’re going to be able to do for you on this list; it’s 
the heart of the investigation you’re undertaking. 

Best,

Scott Calvin
Lehman College of the City University of New York

*I say “never,” but I can think of one case where it might be justified, and 
wouldn’t be surprised if there are others. If you were trying to decide between 
a compound being 4-coordinated or 6-coordinated, and had some reason for 
thinking it was purely one or the other, then a fitted coordination number of, 
for example, 6.3 +/- 0.7 could reasonably be set to 6 in future fits. But 
that’s not the kind of circumstance you have.


> On Jan 7, 2017, at 9:17 AM, Raj kumar  wrote:
> 
> Hi Scott,
> 
> Thanks for your comments. Please see below for my answers.
> 
> 1. Initially, i have started the fitting using single Eo but couldn't achieve 
> satisfactory fit. Hence, moved to use two Eo's, one for first two shells and 
> another for rest other shells, for fitting the curve. With this strategy, I 
> have varied all parameters, including Eo's, mentioned in the previous table. 
> As a result, i have identified appropriate Eo's for respective paths. 
> Subsequently, evaluated all other parameters keeping both Eo's fixed. The 
> result remains almost same irrespective of varying or fixing Eo's.  
> 
> 2. YbVO4 belongs to zircon structure. In this kind of structure, only the 
> position of oxygen varies and leaving both metal atoms unmoved. The atomic 
> config, I have used (YbVO4-Demeter: @ standard zircon position) for modelling 
> is attached along with the COD database file. Both files having different 
> oxygen positions and hence different splitting distances for first oxygen 
> shell. Further, i have compared first two oxygen paths, obtained from COD 
> file, with experimental signal. It seems, the second oxygen distance might be 
> exaggerated. Again, please see attachment for the confirmation. 
> 
> 
> 
> 
> Regards,
> Raj
> 
> On 7 January 2017 at 01:06,  <mailto:ifeffit-requ...@millenia.cars.aps.anl.gov>> wrote:
> Send Ifeffit mailing list submissions to
> ifeffit@millenia.cars.aps.anl.gov 
> <mailto:ifeffit@millenia.cars.aps.anl.gov>
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit>
> or, via email, send a message with subject or body 'help' to
> ifeffit-requ...@millenia.cars.aps.anl.gov 
> <mailto:ifeffit-requ...@millenia.cars.aps.anl.gov>
> 
> You can reach the person managing the list at
> ifeffit-ow...@millenia.cars.aps.anl.gov 
> <mailto:ifeffit-ow...@millenia.cars.aps.anl.gov>
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Ifeffit digest..."
> 
> 
> Today's Topics:
> 
>1. Re: error when fitting in Artemis (George Sterbinsky)
>2. Re: Problem with large so2: Updated (Scott Calvin)
> 
> 
> --
> 
> Message: 1
> Date: Fri, 6 Jan 2017 16:32:45 -0600
> From: George Sterbinsky  <mailto:georgesterbin...@u.northwestern.edu>>
> To: XAFS Analysis using Ifeffit  <mailto:ifeffit@millenia.cars.aps.anl.gov>>
> Subject: Re: [Ifeffit] error when fitting in Artemis
> Message-ID:
>  <mailto:pyokfkivpjffeqmoala95wjk%2b%2byen%2bhai70v3kh...@mail.gmai

Re: [Ifeffit] Problem with large so2: Updated

[Scott Calvin]

Hi Raj,

I don’t have an easy way to read Artemis files on my computer at the moment, so 
I haven’t looked directly at those. But looking at your Word document, I’m 
puzzled by a couple of things:


  *   You have different E0’s for the nearby oxygens from more distant shells. 
That can be defensible in some cases. But the strange thing is that the values 
are identical in different fits, as if you’re setting them to some value. How 
did you choose those values?
  *   I downloaded a YbVO4 structure from the Crystallographic Open Database 
and ran it in WebAtoms. While I wouldn’t expect it to generate identical values 
to those from your model, it does show much greater splitting between the 
nearest-neighbor oxygens (at 2.27 A) and the next-nearest (2.56 A). Your model 
has 2.28 A and 2.39 A. That’s quite different. Are you sure you entered the 
starting structure right?

Best,

Scott Calvin
Lehman College of the City University of New York

On Jan 5, 2017, at 3:12 PM, Raj kumar 
mailto:rajr...@gmail.com>> wrote:

Dear all,

Earlier, I have posted few questions on large so2 value and got valuable 
suggestions from Scott and Bruce. Here i am presenting my summary of earlier 
emails and the present views. Could you please tell me whether am i going in 
right direction.

I have been trying to fit EXAFS signal of YbVO4 both in bulk and nanoparticles 
form. Here, I would like to estimate so2 from bulk (reference) and subsequently 
use the number, as normalization parameter, to calculate the co-ordination 
number in all nanoparticles. Nanoparticles were synthesized by solution route. 
First, I have modelled my reference (bulk YbVO4) with two so2, N and R as free 
parameters. So2-1 is for first two shell and so2-2 is for rest other shells. If 
I use single so2, the value is obtained to be around 0.9 with unsatisfactory 
fit whereas second so2-2 incorporation leads to satisfied fit with large 
uncertainty on the obtained number i.e., (2.04 +/- 0.9). Using the same 
strategy, I could successfully fit the nanoparticles again with large so2-2 
(1.79). From literature and IFEFFIT earlier discussion, i understood that so2 
should be between 0.8-1.2 value.  Further, so2-2 is found to well correlated 
with ss1 and delc (a parameter to define the movement of atoms in c direction). 
Among two, the former parameter (ss1) strongly correlate with so2 and take 
value of 0.8 and 1.9  for fixed and varying ss1, respectively, during the fit. 
This large value of  so2 cannot be controlled with the used model. Also, I have 
tried to fit the curve with fixed so2 i.e., below 1 but leads to unsatisfactory 
fit. For your info, i have tabulated various strategy and the resultant 
parameters. After considering Scott suggestions (and previous posts), i have 
concluded the following: Since, so2 is used as a normalization parameter and 
with the estimated uncertainty, i would like to consider the obtained number as 
reasonable value to normalize the unknown to calculate the CN in other systems. 
Is this strategy good to calculate the CN in unknown? and later to describe the 
trend on obtained parameters?

Regards,
Raj
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] about sigma2 for exafs fitting

[Scott Calvin]

Hi Shaofeng,

The parameter b does not have to do solely with the difference in N_var between 
the two fits! As I say in the book,

“Changes in the theoretical standards used by the model can often be accounted 
for as if they were changes in parameters. For example, if fit A assumed that 
the nearest neighbors were oxygen and included coordination number, bond 
length, and MSRD for those oxygen atoms as free parameters, while fit B assumed 
the nearest neighbors were sulfur and used the same three parameters, then a 
Hamilton test could be applied with the assumption that those three parameters 
were “changed”—instead of applying to oxygen, they now apply to sulfur. If E0 
and S02 were also free parameters, they would not be considered to be changed, 
as they are primarily a property of the absorber, not the scatterer.”

In your case, it looks to me like two or three parameters are changed between 
models (I count four free parameters in your table, one of which is E0, so I’m 
not sure why you say there are only three…is there another constraint?) Suppose 
there are two free parameters changed. b = 2/2 = 1.

x is the ratio of R-factors between the fits.

Using the values you list below, then, gives a result of 0.64, which indicates 
such variation is quite likely to have taken place by chance.

You chose the worst case of your three, though; the pH 3 case. The pH 10 case 
shows the most improvement in the R-factor with x = 0.007/0.012 = 0.58. The 
probability of getting that much improvement by chance drops to 0.30—less 
likely, but still pretty close to a coin flip.

Even the fact that you got all three to show improvement doesn’t quite reach 
the 95% confidence level, although it’s becoming suggestive.

Why, then, are fits that seem to have much lower R-factors not attaining 
statistical significance? The main culprit is that you have a relatively small 
number of independent points for studying this kind of system.

So where’s that leave you?

I’ll stand by the results of the Hamilton test…there’s not quite enough data 
here, on its own, for me to decide in favor of your second model. If you have 
other evidence pointing in that direction, these results could be used as 
supporting evidence. But with the combination of the closeness of fit not 
improving to a statistically significant degree and the sigma2 taking on an 
unusual (but not impossible) value, there needs to be more to the argument, in 
my opinion.

Best,
Scott Calvin
Lehman College of the City University of New York

On Jan 5, 2017, at 12:19 AM, Shaofeng Wang 
mailto:wangshaof...@iae.ac.cn>> wrote:

Dear Scott,

I have learned the Hamilton test. However, this method seems not suitable to 
distinguish our results because the Nvar are same for the two fittings (3 when 
S02 was fixed). So the b value should be zero and the calculator on the website 
 http://www.danielsoper.com/statcalc/calculator.aspx?id=37 can not carry on.

In addition, I am not sure the value x. Is it the ratio of R-factors between 
two fits? I calculated the results using a=2.18 (Nidp and variables (Nvar) were 
7.35 and 3), b=0.1, and x=0.818 (the ratio of two R-factors) and got the 
Regularized lower incomplete beta function of 0.0807. Does it mean 
something?

Cheers,

Shaofeng

--
Shaofeng Wang, Ph.D of Geochemistry
Environmental Molecular Science Group
Institute of Applied Ecology, Chinese Academy of Sciences
Shenyang, 110016, China
wangshaof...@iae.ac.cn<mailto:wangshaof...@iae.ac.cn>
www.iae.cas.cn


From: Scott Calvin<mailto:scal...@sarahlawrence.edu>
Sent: Thursday, January 05, 2017 11:50 AM
To: XAFS Analysis using Ifeffit<mailto:ifeffit@millenia.cars.aps.anl.gov>
Cc: Shaofeng Wang<mailto:wangshaof...@iae.ac.cn>
Subject: Fwd: about sigma2 for exafs fitting

Shaofeng has given me permission to repost her question here on the ifeffit 
mailing list. It is quoted below my response.

Dear Shaofeng,

As Bruce and I said before, a sigma2 of 0.0007 A^2 is not impossible, although 
it indicates less disorder than is typically present. Your attached table does 
seem to show some improvement by using the model from the hydrogen-containing 
structure as compared to the arsenate. A more rigorous test for statistical 
improvement can be conducted using the Hamilton test (you mentioned you’ve 
consulted XAFS for Everyone; full details of the Hamilton test are given there).

It’s also encouraging that the uncertainties on your sigma2 determinations 
using the hydrogen-containing model are quite small; it appears that the fit is 
not getting confused by correlations even though it’s fitting both coordination 
number and sigma2, as that would generally also cause high uncertainties in the 
correlated parameters.

Is such a stiff sigma2 reasonable in this case? I have no idea. I just don’t 
know enough about this particular system; perhaps someone else on the list 
does. Oh, and one other question—

[Ifeffit] Fwd: about sigma2 for exafs fitting

[Scott Calvin]

Shaofeng has given me permission to repost her question here on the ifeffit 
mailing list. It is quoted below my response.

Dear Shaofeng,

As Bruce and I said before, a sigma2 of 0.0007 A^2 is not impossible, although 
it indicates less disorder than is typically present. Your attached table does 
seem to show some improvement by using the model from the hydrogen-containing 
structure as compared to the arsenate. A more rigorous test for statistical 
improvement can be conducted using the Hamilton test (you mentioned you’ve 
consulted XAFS for Everyone; full details of the Hamilton test are given there).

It’s also encouraging that the uncertainties on your sigma2 determinations 
using the hydrogen-containing model are quite small; it appears that the fit is 
not getting confused by correlations even though it’s fitting both coordination 
number and sigma2, as that would generally also cause high uncertainties in the 
correlated parameters.

Is such a stiff sigma2 reasonable in this case? I have no idea. I just don’t 
know enough about this particular system; perhaps someone else on the list 
does. Oh, and one other question—was the data collected at room temperature? If 
it were collected at cryogenic temperatures, that would tend to reduce thermal 
disorder and thus lower sigma2’s.

Even if no one on the list has insight in to this particular system, anyone 
have good published examples of room-temperature systems with sigma2’s < 0.001? 
It might help Shaofeng with her referee…

Best,

Scott Calvin
Lehman College of the City University of New York

P. S. I certainly hope no reviewer is using the “typical values” I provide for 
parameters in XAFS for Everyone as rigid criteria for rejecting results! It is 
most certainly not the way I use them in the book. Some systems actually are 
atypical!

Begin forwarded message:

From: Shaofeng Wang mailto:wangshaof...@iae.ac.cn>>
Subject: about sigma2 for exafs fitting
Date: January 4, 2017 at 8:31:33 PM CST
To: mailto:scal...@sarahlawrence.edu>>
Reply-To: Shaofeng Wang mailto:wangshaof...@iae.ac.cn>>

Dear Dr. Calvin,

I am a research from China. I know you are an expert on XAFS analysis. So I 
write this letter to you for some xafs analysisi problem.

Recently, I am attempting to study the incorporation of arsenate into the 
barite structure. To investigate the species of arsenate in barite, I fitted 
the exafs data using Ba3(AsO4)2 and a dft optimized structure with HAsO4 
incorporated in barite supercell (configuration C2), respectively and got some 
results. Please see the attached table. Smaller reduced chi2 and R-factor were 
obtained by using configuration C2 as the initial model. Does this mean the 
species of arsenate in barite is more likely to be HAsO42- instead of AsO43- ? 
We also have other evidences including XANES and vibrational spectroscopy to 
support this conclusion. However, using C2 as the initial model we obtained 
samll sigma2 (~0.0007). This seems too small and out of the normal range (0.002 
每 0.03) as you mentioned in your publications (XAFS for everyone). So, my 
question is if our data are reasonable. If yes, could you provid some 
references to support?

By the way, I asked simialr question on the ifeffit forum and you gave me some 
answer. However, the reviewer was not convinced and he insited on that sigma2 
must be in the range of 0.002-0.03. How can I response this question?

Any help is very appreciated.

Best regards,

Shaofeng



--
Shaofeng Wang, Ph.D of Geochemistry
Environmental Molecular Science Group
Institute of Applied Ecology, Chinese Academy of Sciences
Shenyang, 110016, China
wangshaof...@iae.ac.cn<mailto:wangshaof...@iae.ac.cn>
www.iae.cas.cn<http://www.iae.cas.cn/>

BEGIN:VCARD
VERSION:2.1
N:;Shaofeng
FN:Shaofeng
REV:20170105T023133Z
END:VCARD


EXAFS fitting parameters.docx
Description: EXAFS fitting parameters.docx
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Problem with large so2

[Scott Calvin]

Hi Raj,

I haven’t looked at your files, but the kinds of results you are describing 
often indicate the model you are using beyond the first shell is incorrect. It 
may be that it has the wrong elements (e.g., heavier elements are present than 
what you modelled) or that the model includes split distances when in reality 
the compound is more symmetric. 

In short, no, the high second S02 does not make sense (although it would be 
good to know the uncertainties). 

Best,
Scott Calvin
Lehman College of the City University of New York

> On Jan 3, 2017, at 9:38 AM, Raj kumar  wrote:
> 
> Dear All,
> 
> 
> 
> I have been trying to fit EXAFS signal of YbVO4 both in bulk and 
> nanoparticles form. Here, I would like to estimate so2 from bulk (reference) 
> and subsequently use the number, as normalization parameter, to calculate the 
> co-ordination number in all nanoparticles. Nanoparticles were synthesized by 
> solution route. From our earlier understanding on the synthesis, in some 
> cases, Yb may or may not completely involve in reaction. Therefore, I except 
> Yb in two environments, one surrounded by mere oxygen (in precise 
> oxygen-Hydrogen; due to disorder one may expect a single peak) and another 
> one in YbVO4 environment. With this info and for reasonable fit, I have used 
> two so2 as free parameters for N: one for first shell (oxygen at two 
> distances) and another for rest other shells.  If I use single so2, I obtain 
> the value of around 0.92 with unsatisfactory fit whereas the fit seems 
> plausible with two so2 i.e., 0.75 and 2.2, respectively. Using the same 
> strategy, I could successfully fit the nanoparticles again with large second 
> so2 (1.79). From literature and IFEFFIT earlier discussion, i understood that 
> so2 should be between 0.8-1.2 value. Here, I would like to know, does this 
> huge second so2 value makes sense? is the obtained so2 value can be used to 
> calculated coordination number for unknown? Further, i have attached my 
> worked files with this message.
> 
> 
> 
> Regards,
> 
> Raj
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Time reversibility of multiple scattering paths

[Scott Calvin]

Hi David,

The TS path in this example, when reversed, is the identical path. A 
straightforward enumeration of paths would only list it once.

The DS paths would be listed twice in the enumeration: M — M(1) — M(4) — M and 
M — M(4) — M(1) — M. Those two paths have identical scattering characteristics, 
and thus could be replaced by a single path with degeneracy two.

In other words, to find degeneracy, list every possible sequence of scattering. 
Paths that have identical scattering properties may be grouped together and 
replaced by a single representative with degeneracy equal to the number of 
instances. If you think about it that way, the answer to your question becomes 
evident.

Best,

Scott Calvin
Lehman College

> On Nov 18, 2016, at 6:03 AM, Inwood D.  wrote:
> 
> Hi All,
>  
> I’ve just been reading the paper with the following reference:
>  
> Nashner, M. S.; Frenkel, A. I.; Adler, D. L.; Shapley, J. R.; Nuzzo, R. G. 
> Structural Characterization of Carbon-Supported Platinum - Ruthenium 
> Nanoparticles from the Molecular Cluster. J. Am. Chem. Soc. 1997, 7863, 
> 7760–7771.
>  
> In particular I was interested about the data analysis section on page 7763 
> (footnote 32) where the authors talk about time reversibility of multiple 
> scattering paths. It is mentioned that the collinear DS path, in the fcc 
> structure, (M – M(1) – M(4) – M) is time reversed, so has twice the 
> degeneracy, but that the collinear TS path (M – M(1) – M(4) – M(1) – M) is 
> not time reversed.
>  
> Why is the DS path time reversible, but not the TS path despite the apparent 
> ‘symmetry’ of the TS path? What makes a scattering path time reversible? Does 
> time reversibility only apply to collinear paths or can it also apply to 
> triangular paths as well?
>  
> Thanks for your help,
>  
> David
>  
> David W Inwood
> PhD Researcher
> Chemistry
> University of Southampton
> Southampton, UK
> SO17 1BJ
>  
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov <mailto:Ifeffit@millenia.cars.aps.anl.gov>
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit>
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/options/ifeffit>
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Converting PDF files to CIF files for usage in Artemis

[Scott Calvin]

Hi Ed,

In case it’s not clear, ATOMS is built in to Artemis. (Long, long ago ATOMS was 
a separate program, so many people still refer to that functionality of Artemis 
by that name.)

—Scott Calvin
Sarah Lawrence College

> On Aug 9, 2016, at 12:40 PM, Matthew Marcus  wrote:
> 
> There are two different input files which take the extension .inp.  One is 
> the feff.inp file, which is the input to FEFF and contains a list of
> coordinates (Cartesian, not crystallographic) of all atoms surrounding an 
> absorpber atom out to a certain distance.  ATOMS produces this.  The other
> is an input file for ATOMS, which has a subset of the information found in a 
> CIF file.  ATOMS can save this out as well once you fill in the form, and you
> can thus re-use it if, for instance, you want to re-run ATOMS to produce a 
> feff.inp file for a different site as absorber.  ATOMS is the only program
> I know that produces FEFF files.
>   mam
> 
> On 8/9/2016 9:26 AM, Ed Han wrote:
>> I see. Thank you for the insight!
>> 
>> Is it possible to create the same .inp or .cif file in FEFF6 (or any of the 
>> FEFF versions) as well?
>> 
>> Or is the ATOMS program the most recommended software to perform this task?
>> 
>> 
>> Regards,
>> Ed Han
>> 
>> On Mon, Aug 8, 2016 at 5:12 PM, Carlo Segre > <mailto:se...@iit.edu>> wrote:
>> 
>> 
>>Hi Ed:
>> 
>>You do not need to start with a CIF file at all.  All you need is the 
>> space group, the lattice constants and the atomic positions.  You can then 
>> fillin the blanks for the Atoms program yourself.
>> 
>>Carlo
>> 
>> 
>>On Mon, 8 Aug 2016, Ed Han wrote:
>> 
>>Hello All,
>> 
>>I apologize in advance if this is not the correct method to reply to 
>> the
>>thread.
>> 
>>Thank you for your replies and suggestions. I will look further into
>>determining the atomic structure of the Hf-Al intermetallics through 
>> their
>>papers.
>> 
>>Once I obtain this information, how do I go about creating the actual 
>> CIF
>>file? Can I just open up any arbitrary CIF file and then change the
>>parameters and name to match that of the Hf-Al systems?
>> 
>>Lastly, going through some CIF files from crystallography.net 
>> <http://crystallography.net>, I notice
>>that some of the 3D models from the CIF has a different stoichiometry 
>> from
>>the designated crystal. For example, I have attached a CIF file of 
>> SiO2
>>that I have found; however, the 3D representation only has 1 Silicon 
>> atom
>>and 1 Oxygen atom. Do you have any insight on this disparity?
>> 
>>Regards,
>>Ed Han
>> 
>> 
>>--
>>Carlo U. Segre -- Duchossois Leadership Professor of Physics
>>Interim Chair, Department of Chemistry
>>Director, Center for Synchrotron Radiation Research and Instrumentation
>>Illinois Institute of Technology
>>Voice: 312.567.3498 Fax: 312.567.3494 
>> 
>>se...@iit.edu <mailto:se...@iit.edu>   http://phys.iit.edu/~segre   
>> se...@debian.org <mailto:se...@debian.org>
>> 
>>___
>>Ifeffit mailing list
>>Ifeffit@millenia.cars.aps.anl.gov 
>> <mailto:Ifeffit@millenia.cars.aps.anl.gov>
>>http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit 
>> <http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit>
>>Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit 
>> <http://millenia.cars.aps.anl.gov/mailman/options/ifeffit>
>> 
>> 
>> 
>> 
>> ___
>> Ifeffit mailing list
>> Ifeffit@millenia.cars.aps.anl.gov
>> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit
>> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Questions about fitting of EXAFS

[Scott Calvin]

Just to clarify: the shape of the FT may certainly change when you change kmax. 
But if the data and the model change in the same way, then the fit is 
insensitive to the choice of kmax, in the sense that the fitted parameters 
remain stable to within their reported uncertainties.

—Scott Calvin
Sarah Lawrence College

On Aug 4, 2016, at 2:50 PM, Fuxiang Zhang 
mailto:zhan...@umich.edu>> wrote:

The Kmax effect on the shape of magnitude of FTs of one of my samples may come 
from the low signal/noise ratio.

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Questions about fitting of EXAFS

[Scott Calvin]

I’ll add to Bruce’s comment that if the choice of kmax has a great effect on 
the parameters found by the fit, that’s an indication of a problem! Good fits 
should be stable to modest changes in kmax (e.g., an inverse angstrom or two 
one way or another). The statistics may suggest one kmax or another is somewhat 
better, but the fitted parameters should not be drifting outside of the ranges 
defined by their uncertainties. If they are, you have an unstable fit. In that 
case, there are several possibilities: perhaps you are including an artifact in 
your k-range (e.g. the beginning of another edge) or a lot of data dominated by 
noise. Or perhaps your model is having trouble distinguishing between two 
fitting minima, and is flipping back and forth between them (this is more 
likely if your fitting model is fairly complicated, with many free parameters). 
In any case, if you’re fit is highly sensitive to kmax you should investigate 
to try to determine why.

—Scott Calvin
Sarah Lawrence College

> On Aug 4, 2016, at 11:07 AM, Bruce Ravel  wrote:
> 
>> 
>> 2. The cutting range of Kmax (FT transform parameters) has great effect
>> on FTs of EXAFS, how do I know to use the best value of Kmax;
> 
> If you have measured data with signal well above the level of noise, why 
> would you choose to use less data?
> 
> Similarly, if, at some point in your data, the signal becomes dominated by 
> noise -- either statistical or systematic -- why would you include it in the 
> analysis?


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Applicable scope of delr=alpha*reff

[Scott Calvin]

If you think it’s plausible that the crystal is expanded or contracted 
isotropically and homogeneously, then it’s a model worth trying. It’s usually 
my first model for fitting just about any crystalline material. Often, the 
resulting fits are not good, and then I have to try other models. But 
sometimes, particularly when the differences from my model structure are small, 
it works fairly well.

In terms of the criteria I described 
here<https://www-ssrl.slac.stanford.edu/conferences/workshops/srxas2011/presentations/calvin_2011.pdf>
 and elsewhere, it has the advantage of “simplicity” under criterion 4. Whether 
or not it produces a good fit under the other criteria is, of course, something 
that you have to find out for each system.

—Scott Calvin
Sarah Lawrence College

On Aug 2, 2016, at 6:27 AM, woschangqiang 
mailto:woschangqi...@126.com>> wrote:

Hi, everyone.
delr=alpha*reff is very powerful in the fit of cubic system. And I want to know 
whether this function can be used in other crystal system. Thank for your 
patient answer.



___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Multiple scattering path parameters

[Scott Calvin]

Hi Neil,

For some kinds of paths, the answer is unambiguously “yes.” For example, 
consider the path A -> B -> A -> B -> A, where A is the absorbing atom and B is 
another atom; i.e., the path has four legs and 180 degree angles between each 
leg. In that case deltaR_MS = deltaR_SS, and sigma2_MS = 4*sigma2_SS.

It’s also close to exact to assert that if a path is perfectly focussed (I 
think that’s what you’re calling “forward through”), that the deltaR is the 
same as the deltaR for single-scattering off the further angle, and likewise 
with sigma2. That’s not quite perfect, but its very close unless the disorder 
is very high.

But for most cases, the constraints imply a model of how the atoms are moving 
relative to each other. Is the motion independent? Completely correlated? 
Isotropic? In something like a monatomic metal, you can make a pretty good 
argument as to what the relative motion should be like. In more complicated 
materials such as oxides, however, it becomes more difficult (although 
qualitative educated guesses can certainly be made).

The good news is that in many of the cases where a physical model is not 
obvious, the multiple-scattering paths that are difficult to constrain (e.g. 
triangles) make relatively small contributions to the signal, and may be 
numerous as well. In that kind of case, I’ve done some work that suggests that 
the precise constraint scheme employed does not make much difference to the 
fit, but leaving those paths altogether does degrade the fit. In other words, 
any reasonable constraint scheme is better than the “constraint” of leaving off 
the triangle paths altogether.

That’s a little scary, I know. But the thing to do is to try a few different 
constraint schemes and just see if it changes your fits much. If it doesn’t, 
and it often doesn’t, then you don’t have to worry! You can also try to leave 
the smaller MS paths off altogether, and see if that makes a difference.

—Scott Calvin
Sarah Lawrence College

> On Jul 28, 2016, at 5:52 PM, Neil M Schweitzer 
>  wrote:
> 
> Are there standard ways of relating the parameters (deltaR and sigma^2) of 
> multiple scattering paths to the parameters of the single scattering paths 
> based on the type of MS path? For example, would deltaR_MS = deltaR_SS*2 if 
> the MS was a “forward through” geometry? What about other geometries and 
> paths that involve different types of atoms?
>  
> Neil
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov <mailto:Ifeffit@millenia.cars.aps.anl.gov>
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit>
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/options/ifeffit>
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] determining reasonable fitting parameters

[Scott Calvin]

This is also related to whether it is appropriate to apply a McMaster 
correction to sigma2 after fitting, right? (Or “during,” but since it’s just an 
offset, it amounts to the same thing.) In Bruce’s method, a McMaster correction 
is appropriate, although often it’s small in comparison with the reported 
uncertainties anyway. In Matthew’s method, there is no need for a McMaster 
correction. Essentially McMaster attempts to correct for using a constant 
normalization instead of the expected smooth decrease in the normalization 
factor after the edge. Or am I remembering the meaning of that correction wrong?

—Scott Calvin
Sarah Lawrence College

> On Jul 27, 2016, at 5:26 PM, Matthew Marcus  wrote:
> 
> I think that if the wiggliness in the spline is enough to cause trouble by 
> the mulriplicative effect, then it's probably eating up the wiggles by 
> subtraction, anyway.
> However, I can see that you wouldn't want the bother and potential for 
> trouble involved in having a second curve for the post-edge division.
>   mam
> 
> On 7/27/2016 2:08 PM, Bruce Ravel wrote:
>> On 07/27/2016 04:56 PM, Matthew Marcus wrote:
>>> Oh, so it assumes that chi(k) = (mu(E(k))-spline(k))/const.?  Mine does
>>> (mu(E(k))-spline(k))/spline(k), which I guess has its own problems
>> 
>> Exactly.  Both solutions have problems.  The potential for trouble with 
>> "/const" seems less severe to me and easier to support (in the sense of 
>> supporting the use of a software package).  So that's why Athena does what 
>> she does.
>> 
>> B
>> 
>> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] about sigma square

[Scott Calvin]

Hi Bruce,

A side comment on a topic you’ve raised before. You said to Shaofeng that:

On Jul 26, 2016, at 8:39 AM, Bruce Ravel 
mailto:bra...@bnl.gov>> wrote:

 It also means that the uncertainty is
such that you can support your conclusion.  While the red line might
overplot the blue with sigma^2=0.0003+0.001, that may be a troubling
result because sigma^2 is not positive definite!

I’m not convinced that it should be disturbing at all if a fit for sigma2 
yields a result that is not positive definite.

Suppose, for a moment, that the true sigma2 for a scattering path is 0.0003 Å2, 
and that data is being analyzed up to k = 9 Å-1. The EXAFS equation tells us 
that the effect of sigma2 on chi(k) is quite modest in that case, and is also 
relatively insensitive to the precise value of sigma2. According to the EXAFS 
equation, at the top of the data range, where it’s effect is greatest, the 
sigma2 factor, e^(-2k^2 sigma2), is multiplying the amplitude by 0.95. Suppose 
further that the uncertainty was +/- 0.0005 Å2. That would imply the sigma2 
factor was at the top of the range was as small as 0.88 or as large as 1.03. It 
doesn’t seem terribly different to me than a fit which yields an S02 of 0.95 
+/- 0.08. Or, in a case where coordination number is expected to be either 4, 
6, or a mixture of the two, N = 5.7 +/- 0.5. The latter result is not generally 
considered troubling, even though the range implied by the uncertainty overlaps 
with values (N > 6) that might be considered wildly implausible for the system 
being studied.

The comparison to coordination number or S02 is not perfect, because sigma2 is 
sensitive to the difference between the amplitude of chi(k) at low k and at 
high k, whereas S02 or N correspond to a uniform suppression. Still, in a case 
when the true value of sigma2 was 0.0003 Å2, the difference between the chi(k) 
amplitude at the bottom of the k-range and the top is quite modest, and might 
reasonably be statistically indistinguishable from no difference at all, 
particularly if it is for one path in a multi-path fit, if the data is somewhat 
noisy, or if the k-range is small.

To borrow a similar example from another specialty in physics, for quite some 
time, measurements of the square of the electron neutrino mass often yielded 
results that were not positive definite. This was taken neither as evidence 
that the neutrino mass was imaginary, nor that the data was bad.

I worry that sending the message that it is troubling to get a result for 
sigma2 that is not positive definite can lead to beginners rejecting such fits, 
thus introducing a bias toward larger values of sigma2. Such a user might 
prefer a model that generates sigma2 = 0.0020 Å2 +/- 0.0018 to one that gave 
0.0003 Å2 +/- 0.0005, for example. To me, all else being equal, the second 
result is better, because it is more precise.

Of course, a result that was negative definite, such as sigma2 = -0.0009 Å2 +/- 
0.0004, would indeed be troubling, and good evidence that something about the 
model or the data was problematic.

—Scott Calvin
Sarah Lawrence College
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] determining reasonable fitting parameters

[Scott Calvin]

Hi Neil,

This<https://www-ssrl.slac.stanford.edu/conferences/workshops/srxas2011/presentations/calvin_2011.pdf>
 might help. It’s an abbreviated version of the discussion of criteria for 
judging fit quality that I later expanded on in my textbook. Northwestern 
doesn’t seem to have a copy of the full book, but you can certainly get it via 
inter-library loan.

—Scott Calvin
Sarah Lawrence College

On Jul 20, 2016, at 6:59 PM, Neil M Schweitzer 
mailto:neil.schweit...@northwestern.edu>> 
wrote:

Being somewhere between a novice and an advanced user of XAS, I have two 
“philosophical” questions about EXAFS fitting (I hope this is the correct forum 
to ask them!). I also have a bug to report. I have included a project file of a 
CeO2 powder for those interested.

1)  Generally, I have developed my own methodology for EXAFS fitting which 
I think generates ok fits, at least from the standpoint that the red line seems 
to follow the blue line pretty closely. My question relates to interpreting the 
statistics of the fit. Generally, I am studying catalyst (i.e. very small metal 
or oxide nanoclusters supported on a different oxide phase). Due to the size of 
the clusters of interest, I wouldn’t expect them to resemble a bulk crystal, 
and therefore it is difficult for me to reduce the number of independent 
variables using know crystal structures. For example in this reference 
material, I am giving every single scattering path its own delR and ss, and 
using a single delE and SO2 for all the paths since they come from the same 
feff calculation. Once I am through with my own methodology, I can usually get 
a close fit even while making every variable a guess (refer to fit 17 in the 
project file). Admittedly, the reduced chi is rather large (5262) and the error 
of each of the DWF’s is the same magnitude as the value itself, which is 
troubling. However, if I use the best fit values of fit 17 and set all of the 
values except for the 3 DWF’s and generate a new fit, all of the statistics 
significantly improve even though the values of the DWF’s did not change at all 
(Fit 19). That is, merely setting the other variables improved the fit (the 
reduced chi square is now 2951) and improved the error of the guess variables 
(the errors of the DWF’s are cut close to half). Why is the fit better since 
none of the values actually changed, and why do the errors improve? How should 
I be reporting the statistics (i.e. the errors) in a publication? Should I 
report the fit 17 errors or the fit 19 errors for the DWF’s?
2)  In terms of the DWF’s in general, what value is considered too high? I 
know the DWF’s have a component that relates to temperature induced disorder in 
the scattering shell and a component that relates to physical disorder in the 
scattering shell, but what value would be considered too big for a sample 
measured at room temperature. I have seen values as high as 0.03 and 0.04 in 
presentations (sorry, no references) but these seems too large to me. At some 
point, if the sample is disordered enough, it seems like EXAFS is no longer an 
appropriate characterization tool to use. What value of DWF would that 
represent (for a sample measured at room temperature)?
3)  I found a bug in the log files in the history window. I am running 
Demeter 0.9.21 on Windows 7. When I generate a new fit, if I change the k-range 
of the fit, the k-range of previous fits in the history window will also change 
to the current range (e.g. fit 19 was run from 2.1-9.3, as shown in the data 
window. Fits 14-16 were run in smaller ranges). What’s even stranger, is if I 
go to a very old fit (i.e. fit 1 or 2 in the project file) and then go back to 
the new fit, the log file will report the k-range from the earlier fit, not the 
current fit. Obviously, this is making it very difficult to keep track of fits 
generated with different k-ranges. I have not tested to see if the R-range 
behaves similarly.

Neil
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] How to input atoms in solid solution for Artemis

[Scott Calvin]

One additional thought:

Distinguishing Ni-Ni and Ni-Fe by creating two different feff paths and fitting 
the spectrum as a mixture of the two is very difficult, to put it mildly, as 
both Anatoly and Matt indicated.

But the bond length determined by your fit may be able to give you a clue, 
particularly when compared to appropriate standards. Depending on the details 
of your structure, the average bond length may depend on the fraction of sites 
occupied by iron to a degree which is measurable by EXAFS. Even if you simply 
use a nickel-nickel scattering path in feff to model your structure, 
differences in bond length could be indicative of changes in site occupancy.

Ideally, this could be done by creating a series of standards with known 
iron-nickel ratios, fitting the bond length for each, and then comparing a fit 
of a sample with a similar structure but unknown iron-nickel ratio to the 
results for the standards.

—Scott Calvin
Sarah Lawrence College

> On Jun 23, 2016, at 3:41 PM, Fuxiang Zhang  wrote:
> 
> Hi, Matt and Anayoly,
> Thank you very much for the help.
> 
> Fuxiang
> 
> On Thu, Jun 23, 2016 at 1:51 PM, Anatoly I Frenkel  <mailto:afren...@yu.edu>> wrote:
> Whatever you decide to do, remember that Ni Ni and Ni Fe pairs look very 
> similar by EXAFS, so it does not matter how you end up doing it.
> You can as well use pure Ni but you may want to consider both fcc and bcc 
> structure for making the model because the alloy may end up to be either that 
> or that, especially if it is a nanoalloy.
> 
> Anayoly
> 
> 
> 
> Sent from my Verizon Wireless 4G LTE Tablet
> 
> 
>  Original message 
> From: Fuxiang Zhang mailto:zhan...@umich.edu>> 
> Date: 6/23/2016 12:51 PM (GMT-05:00) 
> To: ifeffit@millenia.cars.aps.anl.gov 
> <mailto:ifeffit@millenia.cars.aps.anl.gov> 
> Subject: [Ifeffit] How to input atoms in solid solution for Artemis 
> 
> Hi, All,
> I am new in using EXAFS techniques. Just wondering how to input atomic 
> structure of NiFe solid solution to Artemis. It has a simple Ni-structure, 
> but both Ni and Fe occupy the same site. If I use partial occupancy, Artemis 
> program does not allow. Anyone can give me an ideal?
> Also the coordination number N cannot be refined in this simple system (12), 
> is that right? In fact I just want to get the N through the fitting of EXAFS 
> because of the partial occupancy.
> Thank you very much
> 
> Fuxiang Zhang 
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov <mailto:Ifeffit@millenia.cars.aps.anl.gov>
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit>
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit 
> <http://millenia.cars.aps.anl.gov/mailman/options/ifeffit>
> 
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] doubt on number of variables

[Scott Calvin]

Oops, accidentally hit send before I meant to!

You might also find the information in this presentation helpful, as it 
addresses some of the kinds of questions you are asking: 
https://www-ssrl.slac.stanford.edu/conferences/workshops/srxas2011/presentations/calvin_2011.pdf

—Scott Calvin
Sarah Lawrence College

On May 26, 2016, at 10:23 AM, Scott Calvin 
mailto:scal...@slc.edu>> wrote:

Hi Jesus,

Yes, that procedure is fine. As long as the final, reported, fit guesses 
everything which is varied, whatever you do with earlier fits amounts to 
exploration and diagnostics.


On May 26, 2016, at 10:02 AM, Jesús Eduardo Vega Castillo 
mailto:jeve...@gmail.com>> wrote:

Thanks Bruce for your answer,

Just for clarifying, this is not what I have been doing. I just was asked to 
add new independent variables (individual DW factors for each path) while I am 
already at the limit of independent points.

I have been using a number of parameter always lower than the number of 
independent points. I have also been managing high correlations by not varying 
two strongly correlated parameters at the time. But at the end I do a final fit 
setting to guess all the parameters, no matter the correlations, in order to 
obtain a "true" reduced Chi2 to report which includes all parameters. Do you 
consider this procedure right?








2016-05-26 10:36 GMT-03:00 Bruce Ravel mailto:bra...@bnl.gov>>:
On 05/26/2016 09:17 AM, Jesús Eduardo Vega Castillo wrote:
As far as I understand the number of variables refers to those
parameters which have been "guessed" in the fit and it is limited by the
number of independent points.

Correct.

If I have a group of variables but only set to guess a few of them while
fixing the others, then fix some of the obtained values and guess other
variables; I might include a large number of parameters in the resulting
fit (even more than the independent points) without increasing the
Number of variables of each individual fit. This also would keep reduced
chi2 from increasing too much.

Would this be cheating?

The answer, almost certainly, is "yes".

The total number of parameters included is always restricted by the
Number of independent points?

I'm going with this one ...

or
Can I include all the parameters I want as long as I set to guess no
more that the number of independent points at the time?

... and I really don't like this one.


The thing you seem not to be considering is that parameters have correlations.  
The correlations between parameters are an important part of the assessment of 
uncertainty.

If you artificially suppress correlations between parameters in the way that 
you are describing, it is unlikely that the fitted results and their 
uncertainties would be defensible.

While it is possible to probe correlations between parameters in a defensible 
manner by selectively setting and guessing parameters in a lengthy sequence of 
fits, your brief description does not give me confidence that that is what you 
are doing.

The trick to EXAFS analysis (and any non-linear minimization analysis) is to 
defensibly get what information you can from the data.  Rarely do we get 
everything we want.  In those situations where we aspire to  more than our data 
can provide, it's important to remember that something is better than nothing.

B

--
 Bruce Ravel   
bra...@bnl.gov<mailto:bra...@bnl.gov>

 National Institute of Standards and Technology
 Synchrotron Science Group at NSLS-II
 Building 535A
 Upton NY, 11973

 Homepage:http://bruceravel.github.io/home/
 Software:https://github.com/bruceravel
 Demeter: http://bruceravel.github.io/demeter/
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] doubt on number of variables

[Scott Calvin]

Hi Jesus,

Yes, that procedure is fine. As long as the final, reported, fit guesses 
everything which is varied, whatever you do with earlier fits amounts to 
exploration and diagnostics.


On May 26, 2016, at 10:02 AM, Jesús Eduardo Vega Castillo 
mailto:jeve...@gmail.com>> wrote:

Thanks Bruce for your answer,

Just for clarifying, this is not what I have been doing. I just was asked to 
add new independent variables (individual DW factors for each path) while I am 
already at the limit of independent points.

I have been using a number of parameter always lower than the number of 
independent points. I have also been managing high correlations by not varying 
two strongly correlated parameters at the time. But at the end I do a final fit 
setting to guess all the parameters, no matter the correlations, in order to 
obtain a "true" reduced Chi2 to report which includes all parameters. Do you 
consider this procedure right?








2016-05-26 10:36 GMT-03:00 Bruce Ravel mailto:bra...@bnl.gov>>:
On 05/26/2016 09:17 AM, Jesús Eduardo Vega Castillo wrote:
As far as I understand the number of variables refers to those
parameters which have been "guessed" in the fit and it is limited by the
number of independent points.

Correct.

If I have a group of variables but only set to guess a few of them while
fixing the others, then fix some of the obtained values and guess other
variables; I might include a large number of parameters in the resulting
fit (even more than the independent points) without increasing the
Number of variables of each individual fit. This also would keep reduced
chi2 from increasing too much.

Would this be cheating?

The answer, almost certainly, is "yes".

The total number of parameters included is always restricted by the
Number of independent points?

I'm going with this one ...

or
Can I include all the parameters I want as long as I set to guess no
more that the number of independent points at the time?

... and I really don't like this one.


The thing you seem not to be considering is that parameters have correlations.  
The correlations between parameters are an important part of the assessment of 
uncertainty.

If you artificially suppress correlations between parameters in the way that 
you are describing, it is unlikely that the fitted results and their 
uncertainties would be defensible.

While it is possible to probe correlations between parameters in a defensible 
manner by selectively setting and guessing parameters in a lengthy sequence of 
fits, your brief description does not give me confidence that that is what you 
are doing.

The trick to EXAFS analysis (and any non-linear minimization analysis) is to 
defensibly get what information you can from the data.  Rarely do we get 
everything we want.  In those situations where we aspire to  more than our data 
can provide, it's important to remember that something is better than nothing.

B

--
 Bruce Ravel   
bra...@bnl.gov

 National Institute of Standards and Technology
 Synchrotron Science Group at NSLS-II
 Building 535A
 Upton NY, 11973

 Homepage:http://bruceravel.github.io/home/
 Software:https://github.com/bruceravel
 Demeter: http://bruceravel.github.io/demeter/
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Absorption length

[Scott Calvin]

Rob Scarrow did a particularly good job with this topic: 
http://xafs.org/Workshops/NSLS2002?action=AttachFile&do=view&target=Scarrow.pdf

—Scott

On Mar 9, 2016, at 7:54 PM, 이국승(에너지환경소재연구팀) 
mailto:lks3...@postech.ac.kr>> wrote:

Dear members,
I have seen many times the absorption length in materials for introduction of 
XAFS.
The absorption length equals 1/mu, which is the thickness that I0 is attenuated 
to 1/e.
But I do not understand why the absorption length is important.
More importantly, I do not understand why the particle size in samples should 
be smaller than the absorption length.
Could any one explain the absorption length including the question above and 
the relation between the absorption length and edge step calculations?
Thanks in advance.

Kug-Seung Lee


Kug-Seung Lee, Ph.D.
Staff Scientist
Manager, 8C Nano Probe XAFS Beamline
Pohang Accelerator Laboratory
Pohang 790-784, Republic of Korea

Office: +82-54-279-1522
Fax: +82-54-279-1599
C. P.: +82-10-6419-7801
Researcher ID: http://www.researcherid.com/rid/D-7088-2011

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Collected k-space and fitting range

[Scott Calvin]

Hi Terry,

The only possible difference would be due to differences in background 
subtraction. In practice, it’s a good idea to collect data ~1 inverse angstrom 
above the range you will analyze to make sure the background behaves properly 
toward the top of the fitted range. But beyond that it shouldn’t make much 
difference.

—Scott Calvin
Sarah Lawrence College

On Mar 3, 2016, at 8:33 PM, Taehoon Kim 
mailto:oxonh...@outlook.com>> wrote:

Hello,

I am a bit confused about the range of k-space, which may affect the quality of 
EXAFS fitting.
I would like to understand whether there is a difference between the fitting 
using an identical fitting range of k-space obtained from different range of 
k-space collected.

For example, is there a notable difference between a fitting of k = 2.3-11 
which was deduced from a spectra data collected up to k=12 and a spectra data 
collected up to k=13 ?

Kind regards
Terry
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] XANES pre-edge vs. edge features

[Scott Calvin]

Hi Aditya,

Matt gave a good explanation of what’s responsible for the various features, 
which was part of your question. Another aspect of your question has to do with 
the way we use terminology.

(Bruce’s answer, and your response, came in while I was writing this. So it may 
be somewhat superfluous, but since it’s already mostly written, I figure I’ll 
pass it along.)

As Matt said, “edge” is a poorly defined term. One meaning of it is that it’s 
the big rise in absorption that occurs at the start of a XAFS spectrum. That, 
for example, is the meaning that’s in use when we try to quantify an “edge 
step” or an “edge jump” so that we can normalize the data. In that sense, if a 
feature occurs before most of that rise, it can be called “pre-edge.”

But another use of the word “edge” is as a synonym for “E0,” which is itself a 
concept that is not unambiguously defined. One definition of E0 would be “the 
energy origin needed to make the EXAFS equation accurate.” But the EXAFS 
equation is itself an approximation which is only useful at energies starting 
many eV above E0. So there’s no theoretical reason that the E0 used in EXAFS 
has to correspond to an energy with a simple theoretical meaning or to any 
particular feature in the spectrum. The best we can say is that it should be 
within, or sometimes a bit above, the rising portion of the absorption spectrum.

But when we’re trying to align and process data we need some working definition 
of E0, both for XANES analysis and to get a preliminary chi(k) for EXAFS 
analysis. So we come up with other definitions, such as “the maximum of the 
first significant peak in the first derivative spectrum.” As Bruce said, the 
exact definition used is not important, but it is important that the same 
definition be used with all spectra being compared.

Depending on the definition chosen, the preliminary value of E0 may very well 
be chosen well below the bulk of the rise in the spectrum. If we call the 
energy of E0 the “edge,” but also call features that come before the big rise 
"pre-edge features,” we end up with the confusing terminology that the edge 
energy may well be below some of the pre-edge features!

That terminology is unfortunate for people new to the field, but rarely causes 
any actual ambiguity.

—Scott Calvin
Sarah Lawrence College


On Jan 25, 2016, at 9:40 AM, Matt Newville 
mailto:newvi...@cars.uchicago.edu>> wrote:

Aditya,

The distinction between "edge" and "pre-edge" is not very clear, either when 
looking at a single spectrum or even conceptually.

In broad terms, the main edge is at the energy where the unoccupied electron 
levels start - the Fermi energy.   For 1s levels, the transition is to p levels 
(and for Fe K edge, the 4p level).  So, the main edge is at the energy of the 
empty 4p levels.   This the transition as being to an atomic level.  In a solid 
(or liquid), the energy levels above the Fermi level are highly delocalized and 
spread over many (if not all) atoms in the systems.  Once you get much above 
the main edge, it's not very easy to assign transitions to identifiable atomic 
transitions, or even assign a good quantum number to them.

Pre-edge features are generally considered to be unoccupied atomic levels (that 
is, still assignable to a particular atom, or at least almost so)  below the 
main edge.   For the transition metal K edges (such as Fe), the main edge is 1s 
-> 4p.  But Fe has many unoccupied 3d levels.   For a K edge to get to 
transition to these levels, either you need a quadrupole transition  (unlikely, 
but not impossible), or (more likely) for bonding/anti-bonding with ligands 
(typically oxygen) to mix their p-orbitals with the metal d-orbitals.This 
hybridization is often called a ligand field or crystal field.It often 
gives very identifiable (and at very predictable energies) peaks below the main 
edge.Two and sometimes even three peaks can be seen and assigned with 
ligand field terminology.There's sort of a whole industry built up around 
these peaks for transition metal oxides.

These peaks can "leak" into the main edge, and in some cases (say, Cu1+) the 
classification of "sharp features at the edge" is not very clear.  For Fe 
metal, it's pretty clear that the main edge (derivative at 7110.75 eV, a small 
peak on the main edge around 7112.5 eV) is the 4p level, and the rest of the 
features are actually explainable as EXAFS.

Hopefully, someone will correct anything I got wrong!

--Matt



On Mon, Jan 25, 2016 at 8:13 AM, Aditya Shivprasad 
mailto:aps...@psu.edu>> wrote:
Dear list,

I was looking at the XANES standard for Fe foil from Hephaestus and I noticed 
that there was a small, curved feature at the edge (7112 eV), another 
inflection point at 7116.4 eV, and then the edge step at around 7131 eV. My 
question is: why is the feature at 7112 eV considered as the edge and not as a 
pre-edge f

Re: [Ifeffit] Regarding calculations for scattering amplitude

[Scott Calvin]

Hi Pushkar,

It sounds to me like you have the right idea.

Using this procedure with free nanoparticles tends to give sizes that correlate 
to TEM sizes, but are about ten times smaller, just as you found. This is shown 
in more detail in this later paper of mine:


“Comparison of extended X-ray absorption fine structure and Scherrer analysis 
of x-ray diffraction as methods for determining mean sizes of polydisperse 
nanoparticles,” S. Calvin, S. X. Luo, C. C. Broadbridge, J. K. McGuinness, E. 
Anderson, A. Lehman, K. H. Wee, S. A. Morrison, and L. K. Kurihara, Appl. Phys. 
Lett. 87, 233102 (2005).

I’m still not completely convinced as to why the EXAFS size for free 
nanoparticles is often so much smaller than the TEM and XRD sizes. One 
possibility, raised by Anatoly Frenkel, is that there are often lots of 
molecular-scale fragments that TEM and XRD don’t pick up, but EXAFS does. 
Another possibility, explored in some of my papers, is that the size 
distribution is broad enough so that, for XRD in particular, the signal is 
dominated by the largest crystallites. Another possibility, not well explored, 
is that the amplitude reduction factor S02 has a weak dependence on R, or that 
there’s some other R-dependent effect (e.g. a surface effect) that is 
systematically reducing the sizes given by EXAFS.

Overall, then, my advice is that EXAFS-determined sizes can be compared to 
other EXAFS-determined sizes for similar systems. For example, if you have two 
sets of nanoparticles of the same composition, EXAFS can be used to get the 
relative size of those two samples. EXAFS-determined sizes, however, can not be 
compared directly to TEM and XRD determined sizes (or, for that matter, to 
other methods), particularly for polydispersed samples,in part because each 
technique places different weight on the low and high end of the actual size 
distribution.

—Scott Calvin
Sarah Lawrence College

On Jan 21, 2016, at 4:01 AM, pushkar shejwalkar 
mailto:pshejwalkar2...@gmail.com>> wrote:

Dear All,
   I sent a related question in the month of Dec but it seems that the question 
did not appeared on list because of mailbox being full, as we all have received 
an email. Since mailing list is started again I am sending this question once 
more.
My question is actually regarding finding out the nanoparticle size. In this 
respect, I am referring to Harris's work in which by using a simple equation we 
can roughly estimate the nanoparticle size

J. Appl. Phys., Vol. 94, No. 1, 1 July 2003

 Now the question is that in this paper and another paper I found online as 
well, it refers to Nnano and Nbulk. I am guessing that this is basically 
coordination number of nanoparticle and bulk. But my first question is how do I 
calculate it?? e.g. I am working on Pd K-edge. So should I fit Pd foil with Pd 
crystal data and get the CN? but that would be 12 anyways. So would Nbulk be 12 
in that case? and what will be Nnano then?

In another context, CN is proportional to amplitude, so in that case is it 
directly proportional? as in, can I simply replaced in formula, Nbulk/Nnano 
with Amplitudebluk/amplitudenano?

Third thing is that, if above is true then how should/can I calculated the 
amplitude? isnt that amplitude is always calculated by Feff? if so could 
someone help me with the procedure to do so? How will I know amplitude for 
nanoparticles, because I cannot have FEFF for nanoparticles, right?

I went through lot of material, however, so far my calculations give me a 
solution that is vaguely correlated to TEM images. The TEM value is about 2nm 
(rouhgly 20 angstrom). my current values I am getting by solving the equation 
by my current understanding is only 2Ao. about 10 times less. What I am doing 
is simply plotting the crude data in athena to get x(R) Vs radial distance 
graph, calculate A-3 for nanoparticles and bulk and using this as amplitudes. 
It is possible that I may be doing something totally irrelevant, so please help 
me in this aspect.

Thank you

Sincerely

Pushkar

--
Best Regards,
Pushkar Shejwalkar.
Post-doctoral -Researcher,
Tokyo Engineering University,
Tokyo-to
Japan

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] Most probably a bug or something.

[Scott Calvin]

Hi Pushkar,

Bruce Ravel replied to your question on November 30: 
http://www.mail-archive.com/ifeffit@millenia.cars.aps.anl.gov/msg05367.html

—Scott Calvin
Sarah Lawrence College


On Dec 2, 2015, at 7:40 PM, pushkar shejwalkar 
mailto:pshejwalkar2...@gmail.com>> wrote:

Dear All,
   I have sent the following email about 4-5 days before but I havent heard 
anything back as yet. I was wondering as to if this is not something to worry 
about we can continue to use it? Or did I not posted the question correctly or 
to right forum?
Best
Pushkar
-- Forwarded message --
From: pushkar shejwalkar 
mailto:pshejwalkar2...@gmail.com>>
Date: Mon, Nov 30, 2015 at 12:24 PM
Subject: Most probably a bug or something.
To: XAFS Analysis using Ifeffit 
mailto:ifeffit@millenia.cars.aps.anl.gov>>


Dear All,
I have recently found this particular problem with plotting my graphs in k 
fitting. This is a problem with athena as well as artemis. When I plot the 
graphs in R or E or q there is no problem, but when I do so in k suddenly these 
four black circles show up on the left side of the graph. Interestingly this 
has appeared suddenly just now and after this (probably i can recollect 
exactly) I could not maximize my plotting window at all. although if I drag the 
corners it can be resized and these four circles then appears relatively 
smaller (basically it does not exapnds with the window. Anybody else have this 
problem? is there a quick fix? I am not sure if this is a bug or not so I am 
sending this here.
my operating system is Windows 10, Athena and artemis 0.9.22 8GB RAM. (Please 
let me konw if anything else is needed)
Pictures of these problem is attached below. I am attaching R-space just to 
show that it doesnt happen for the R-space. First picture is Athena, second is 
from Artemis fitting.



--
Best Regards,
Pushkar Shejwalkar.
Post-doctoral -Researcher,
Tokyo Engineering University,
Tokyo-to
Japan



--
Best Regards,
Pushkar Shejwalkar.
Post-doctoral -Researcher,
Tokyo Engineering University,
Tokyo-to
Japan

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Re: [Ifeffit] a small question

[Scott Calvin]

Dear Shaofeng,

You really do need to fit to know for sure. Aside from changes in bond length, 
there are many other effects that can shift a peak.

—Scott Calvin
Sarah Lawrence College

On Nov 4, 2015, at 7:29 AM, Shaofeng Wang 
mailto:wangshaof...@iae.ac.cn>> wrote:

Dear all experts,

I am processing two As K edge EXAFS data  shown as following figure. The first 
shell is As-O. The difference of the As-O peak btween sample As1 and As2 is 
around 0.03A. Does this represent that the average As-O bond length of sample 
As1 should be around 0.03 A longer than that of sample As2? Or this depends on 
fitting results ?

Any help is very appreciated.




Cheers,


Shaofeng
--
Shaofeng Wang, Ph.D of Geochemistry
Environmental Molecular Science Group
Institute of Applied Ecology, Chinese Academy of Sciences
Shenyang, 110016, China
wangshaof...@iae.ac.cn<mailto:wangshaof...@iae.ac.cn>
www.iae.cas.cn<http://www.iae.cas.cn>
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] La2Zr2O7 EXAFS fitting

[Scott Calvin]

Hi Herve,

I haven’t looked at your project file, but the 1.80 A peak is not that big a 
surprise, as peaks in the Fourier transform typically show up 0.3 to 0.5 
angstroms lower than the corresponding half-path length, due to the effects of 
the potential of the absorbing and scattering atoms. In other words, the 
Fourier transform is not a radial distribution function, although it’s 
correlated to one.

The interaction between the two similar path lengths further complicates peak 
location, since it can establish a “beating” between the two signals.

Under the circumstances, a peak at 1.80 angstrom is not remarkable.

The peak at 1.28 angstrom, on the other hand, is well in to the background 
region for this compound, and does not need to correspond to a path.

The fact that the Fourier transform is not a radial distribution function is 
why some sort of comparison to a standard is necessary. For Ifeffit users, that 
typically means fitting to a so-called “theoretical standard,” that is, a 
computed spectrum, using Artemis. It could also mean comparison to a 
well-chosen series of measured materials (“empirical standards”). But it is not 
possible to just look at the Fourier transform and read off bond lengths.

—Scott Calvin
Sarah Lawrence College

On Aug 14, 2015, at 4:41 AM, herve muguerra 
mailto:herve.mugue...@neel.cnrs.fr>> wrote:

Dear All,

First of all, I am still very new in EXAFS analysis, so please excuse me
for any misunderstandings.

My sample is a crystalized La2Zr2O7 thin layer (thickness = 100 nm) on a
NiW substrate and we have realized EXAFS measurements by grazing angle
under primary vacuum at Lathanum L3 edge (ESRF, Grenoble, France).

My objectives are to establish the local distortions due to the
different constraints in the oxide thin layer.

The La2zr2O7 crystalized in a pyrochlore structure (SG = Fd-3m ; a =
10.78A).
First scattering path
La-O2.12.3352A
La-O1.12.6453A

But I have a problem: I observe the first peaks at 1.28A and 1.80A on
the EXAFS spectra as a function of R.

How do you explain this feature?

Can you point me in the right direction for these issues?

I attached the project file La_80_LZO_1000.prj of my sample:
LAHM045 to LAHM50 = 6 different scans
La_80_LZO_1000 = Data merge

Thanks

Best regards

Hervé Muguerra
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Measuring particle size by coordination number

[Scott Calvin]

Oops—too early in the morning for me! I meant "The formula works poorly when it 
is only applied with one path”

—Scott Calvin
Sarah Lawrence College

On Aug 3, 2015, at 7:11 AM, Scott Calvin 
mailto:scal...@sarahlawrence.edu>> wrote:

Hi Abhijeet,

The formula looks poorly when it is only applied with one path.

The strength of this approach, and of Anatoly Frenkel’s alternative strategy 
for finding particle sizes, is that it compares coordination numbers for 
different shells, helping to break the correlations between N, S02, and sigma2.

Even then, it tends to report sizes that are small compared to those provided 
by other techniques, but compares sizes between different samples quite well.

—Scott Calvin
Sarah Lawrence College

On Aug 3, 2015, at 4:41 AM, Abhijeet Gaur 
mailto:abhijeetga...@gmail.com>> wrote:

Hi all,

 Sorry, I may be asking a question already asked in the mailing 
list, but I am unable to get it from the earlier discussions.

I am reading one of the research article ´´ Determination of 
crystallite size in a magnetic nanocomposite  using extended x-ray absorption 
fine structure´´ .J. Appl. Phys,. vol 94, page 778, 2003

   In this paper there is a formula N_nano = 
[1-3/4(r/R)+1/16(r/R)^3]N_bulk, which you said can be used in the Artemis for 
extracting the  particle size.

   I am trying to use this formula for determination of size of Ni 
particles in an Ni / Al2O3 catalyst. From TEM the size determined is around 4 
nm.

   So, I am trying to fit the theoretical model of Ni metal to the 
sample data in the Artemis

   I defined a guessed parameter R and then put the whole formula 
for S02 = amp*(1-(0.75*(Reff/R))+(0.063*(Reff*Reff*Reff/(R*R*R.

   The value of R I am getting is 0.81 (0.05) which is quite small.

   Is there something I am missing in the formula or I have to 
define the N seprately for sample cooridnation number.

  Currently, I am using only the first path in the fitting.



With thanks


Abhijeet Gaur


--
Dr. Abhijeet Gaur
Post doctoral fellow in the  Institute for Chemical Technology and Polymer 
Chemistry
Karlsruher Institut für Technologie (KIT), Karlsruhe Germany.
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Measuring particle size by coordination number

[Scott Calvin]

Hi Abhijeet,

The formula looks poorly when it is only applied with one path.

The strength of this approach, and of Anatoly Frenkel’s alternative strategy 
for finding particle sizes, is that it compares coordination numbers for 
different shells, helping to break the correlations between N, S02, and sigma2.

Even then, it tends to report sizes that are small compared to those provided 
by other techniques, but compares sizes between different samples quite well.

—Scott Calvin
Sarah Lawrence College

On Aug 3, 2015, at 4:41 AM, Abhijeet Gaur 
mailto:abhijeetga...@gmail.com>> wrote:

Hi all,

 Sorry, I may be asking a question already asked in the mailing 
list, but I am unable to get it from the earlier discussions.

I am reading one of the research article ´´ Determination of 
crystallite size in a magnetic nanocomposite  using extended x-ray absorption 
fine structure´´ .J. Appl. Phys,. vol 94, page 778, 2003

   In this paper there is a formula N_nano = 
[1-3/4(r/R)+1/16(r/R)^3]N_bulk, which you said can be used in the Artemis for 
extracting the  particle size.

   I am trying to use this formula for determination of size of Ni 
particles in an Ni / Al2O3 catalyst. From TEM the size determined is around 4 
nm.

   So, I am trying to fit the theoretical model of Ni metal to the 
sample data in the Artemis

   I defined a guessed parameter R and then put the whole formula 
for S02 = amp*(1-(0.75*(Reff/R))+(0.063*(Reff*Reff*Reff/(R*R*R.

   The value of R I am getting is 0.81 (0.05) which is quite small.

   Is there something I am missing in the formula or I have to 
define the N seprately for sample cooridnation number.

  Currently, I am using only the first path in the fitting.



With thanks


Abhijeet Gaur


--
Dr. Abhijeet Gaur
Post doctoral fellow in the  Institute for Chemical Technology and Polymer 
Chemistry
Karlsruher Institut für Technologie (KIT), Karlsruhe Germany.
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Plotting chi(k) linear combination fit in energy

[Scott Calvin]

Hi Ellen,

Either I’m not understanding what you’re trying to do, or what you’re trying to 
do doesn’t make sense.

XANES is generally plotted as mu(E) or norm(E). To get from norm(E) to chi(k), 
two things must be done: a smooth background must be subtracted from norm(E), 
and E has to be converted to k. The conversion from E to k is reversible, but 
the background subtraction is not, since chi(k) is meant to be background 
independent. You therefore cannot, in principle, plot chi(k) data as norm(E).

—Scott Calvin
Sarah Lawrence College

On Jul 20, 2015, at 7:48 PM, Ellen Daugherty 
mailto:ellen.daughe...@colostate.edu>> wrote:

Hello,

I was unable to find an example of this question in the archive, but please let 
me know if it has already been addressed. I would like to plot the results of 
linear combination fits performed in chi(k) in energy space to compare them to 
XANES fits. When I try to plot the saved fit in energy, I get a message saying: 
"chi data cannot be plotted in energy." Is there any way to do this other than 
to use Excel to scale the weights of the standards? This would be exhausting to 
perform for combinatorial fits. I am using Athena in Demeter 0.9.20 on Windows 
7. I have attached a sample and a few standards for aid in addressing the issue.

Thank you!

Ellen

--
Ph.D. Candidate
Research Assistant, Borch Lab<http://borch.agsci.colostate.edu/>
Department of Chemistry
Colorado State University
http://borch.agsci.colostate.edu/group-members/ellen-daugherty/
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Mn3O4 EXAFS fitting

[Scott Calvin]

Thanks, Kug-Seung. If you have the book, a spinel is one of the case studies in 
the last chapter! If you flip to section 16.4, that might provide additional 
help.

—Scott Calvin
Sarah Lawrence College

On Jun 1, 2015, at 9:43 PM, 이국승 
mailto:lks3...@postech.ac.kr>> wrote:

Dear Scott,



In addition to practical explanation from Fred, you give theoretical details 
and a good example.
I am reading your book 'XAFS for everyone' from time to time. It helps a lot.
Thank you.



Kug-Seung Lee




___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Create Atoms input with just coordinates

[Scott Calvin]

Hi Maryline,

Use a = b = c = 10, with alpha = beta = gamma = 90. Now enter the coordinates 
from your simulation in units of angstroms/10 (which happens to be nm).

—Scott Calvin
Sarah Lawrence College

On Jun 1, 2015, at 10:43 AM, M Ferrier 
mailto:mferr...@hotmail.com>> wrote:

Hi,

I am trying to create an Atoms input file for actinium. Since not a lot is 
known about this element we run some DT calculations. One molecule of Ac 
acetate in a box surrounded by water.
I have the coordinate of each elements. I was told the symmetry was P1 (I guess 
for the box) but unfortunately their is no symmetry in liquid and thus we 
cannot input the cell parameters (distances a, b, c and angles alpha, beta, 
gamma). I was wondering if their were a way to make Atoms run to obtain the 
Feff calculations. Or we should use a different model with a known solid?

Thank you for your help

Maryline Ferrier
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Mn3O4 EXAFS fitting

[Scott Calvin]

Hi Kug-Seung,

Fred has explained the basic issues with spinels well. For an example of how to 
handle them, you can see this article of mine:

http://journals.aps.org/prb/abstract/10.1103/PhysRevB.66.224405

If you don’t have access to the article, let me know, and I can email you a 
copy.

—Scott Calvin
Sarah Lawrence College

> On Jun 1, 2015, at 5:07 AM, fred.mosselm...@diamond.ac.uk wrote:
> 
> Dear Kug-Seung 
> 
> Your main initial problem is you didn't understand the structure of Mn3O4. It 
> is a spinel with two Mn sites one octahedral and one tetrahedral. You need to 
> consider both sites to have a chance of fitting, and set multiplication 
> factors on the relative amps in the ratio. It has 1/3 tetrahedral sites and 
> 2/3 octahedral. You need to do the feff calculation twice with each site as 
> the core atom.
> In the attached project I have done this. You then end up with a lot of paths 
> so to try to restrict the no of variables I have made several of sigma 
> squared to be the same.  I am sure this model could be improved upon but  I 
> wanted to show you how to get going in fitting the structure. In theory as 
> the Mn tetrahedral site and octahedral sites are different oxidations states 
> they might (should) have different E0s as well but with a limited data range 
> this gets hard to test properly.
> 
> Best wishes
> Fred Mosselmans
> 
> 
> Message: 1
> Date: Mon, 01 Jun 2015 15:59:40 +0900
> From: ??? 
> To: ifeffit@millenia.cars.aps.anl.gov
> Subject: [Ifeffit] Mn3O4 EXAFS fitting
> Message-ID: <1433141980110.23146.post...@postech.ac.kr>
> Content-Type: text/plain; charset="us-ascii"
> 
> An HTML attachment was scrubbed...
> URL: 
> <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20150601/4ec8a237/attachment.htm>
> -- next part --
> A non-text attachment was scrubbed...
> Name: Mn3O4.fpj
> Type: application/octet-stream
> Size: 1569403 bytes
> Desc: not available
> URL: 
> <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20150601/4ec8a237/attachment.obj>
> -- next part --
> A non-text attachment was scrubbed...
> Name: Mn3O4.prj
> Type: application/octet-stream
> Size: 27599 bytes
> Desc: not available
> URL: 
> <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20150601/4ec8a237/attachment-0001.obj>
> 
> --
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> 
> 
> End of Ifeffit Digest, Vol 148, Issue 1
> ***
> 
> -- 
> This e-mail and any attachments may contain confidential, copyright and or 
> privileged material, and are for the use of the intended addressee only. If 
> you are not the intended addressee or an authorised recipient of the 
> addressee please notify us of receipt by returning the e-mail and do not use, 
> copy, retain, distribute or disclose the information in or attached to the 
> e-mail.
> Any opinions expressed within this e-mail are those of the individual and not 
> necessarily of Diamond Light Source Ltd. 
> Diamond Light Source Ltd. cannot guarantee that this e-mail or any 
> attachments are free from viruses and we cannot accept liability for any 
> damage which you may sustain as a result of software viruses which may be 
> transmitted in or with the message.
> Diamond Light Source Limited (company no. 4375679). Registered in England and 
> Wales with its registered office at Diamond House, Harwell Science and 
> Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Athena reporting incorrect element on import

[Scott Calvin]

Following up on Anna's suggestion, note that when you let Athena normalize, you 
can instruct it to fix the edge step at 1. In effect, that tells it to not 
actually renormalize (soft x-ray edges can be tricky to normalize, and you may 
like what the Australian synchrotron has already done), but it does let it do 
its interpolations to get the data in the format it needs.

--Scott Calvin
Sarah Lawrence College

On Apr 20, 2015, at 2:48 AM, Zimina, Anna (IKFT) 
mailto:anna.zim...@kit.edu>> wrote:

Hi, Ken,

try to load your data as raw data and let the Athena normalized your „already 
normalized“ data again. As far as I understand, the normalized data for Athena 
means special format of the data (equidistance E step?) which you do not get 
from your treated data.

Hope it helps a bit,

Anna.

Von: 
ifeffit-boun...@millenia.cars.aps.anl.gov<mailto:ifeffit-boun...@millenia.cars.aps.anl.gov>
 [mailto:ifeffit-boun...@millenia.cars.aps.anl.gov] Im Auftrag von Kenneth 
Latham
Gesendet: Montag, 20. April 2015 07:40
An: ifeffit@millenia.cars.aps.anl.gov<mailto:ifeffit@millenia.cars.aps.anl.gov>
Betreff: [Ifeffit] Athena reporting incorrect element on import

Hi,

I am having a number of issues with importing my data into Athema, the data 
that I collected was on the carbon K edge. As there is an issue with recording 
carbon (as it builds up on the mirrors), the synchrotron that I performed my 
measurments at (Australian Synchrotron) has developed software to perform a 
double normalization to remove these effects.

The problem comes when importing the now normalized data into Athena. As it 
doesn't require normalization again, I select the norm(E) data type and the 
correct column for energy and response which produces the correct plot of the 
data. When I click OK it attempts to produce the other plots, sometimes it 
manages to make a k-space and R-space but q-space is never displayed and other 
times it only manages the energy plot. Additionally, the word element is lit up 
red and it has selected beryllium as the element on the L2 edge. Changing this 
to carbon does not seem to fix the problem as it reverts back to beryllium as 
soon as I leave the main page.

Finally, when I go to peak fitting, any value added for peaks informs me that 
it is out of range.

I have attached some normalized data.

Any help with this would be greatly appreciated.

Regards,
Ken Latham
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] amplitude parameter S02 larger than 1

[Scott Calvin]

Hi Anatoly,

The method Ifeffit uses to compute uncertainties in fitted parameters is 
independent of noise in the data because it, in essence, assumes the fit is 
statistically good and rescales accordingly. This means that the estimated 
uncertainties really aren't dependable for fits that are known to be bad (e.g. 
have a huge R-factor, unrealistic fitted parameters, etc.), but since those 
fits aren't generally the published ones, that's OK. 

Secondly, the high-R amplitude will not be essentially zero with 
theoretically-generated data, even if you don't add noise, because the effect 
of having a finite chi(k) range will create some ringing even at high R.

Frankly, the default method by which Ifeffit (and Larch? I haven't used Larch) 
estimates the noise in the data is pretty iffy, although there's not really a 
good alternative. The user can override it with a value of their own, but as 
you know, epsilon is a notoriously squirrelly concept in EXAFS fitting. The 
really nice thing about the Ifeffit algorithm is that it makes the choice of 
epsilon irrelevant for the reported uncertainties. 

What it is NOT irrelevant for is the chi-square. For this reason, I personally 
ignore the magnitude of the chi-square reported by Artemis, but pay close 
attention to differences in chi square (actually, reduced chi square) for 
different fits on the same data.  

--Scott Calvin
Sarah Lawrence College

> On Mar 23, 2015, at 10:18 PM, Anatoly I Frenkel  wrote:
> 
> Hi Yuanyun and Bruce,
> 
> Your test is a very nice way to check the effect of the muffin tin radius on 
> the fit value of the amplitude fator.
> Although I hadn't run your test myself I would have expected that the error 
> bars produced by the fit of a theory to "theoretically generated data" would 
> have been huge because you haven't added noise to the "data". Or have you? It 
> is not clear why the uncertainties reported by Yuanyun are so small, given 
> that the analysis method uses the high r signal in the data as a measure of 
> noise. Since the data had no noise the high r amplitude should be essentually 
> zero and that would renormalize error bars to huge values.
> 
> I may have misunderstood what was done in this test, and, perhaps, the noise 
> was added automatically as one of the option that Yuanyun chose. Or there was 
> some other reason why errors came out to be small? Thanks for your comments,
> 
> Anatoly
> 
> 
> From: ifeffit-boun...@millenia.cars.aps.anl.gov 
> [ifeffit-boun...@millenia.cars.aps.anl.gov] on behalf of 
> huyan...@physics.utoronto.ca [huyan...@physics.utoronto.ca]
> Sent: Monday, March 23, 2015 9:46 PM
> To: XAFS Analysis using Ifeffit
> Subject: Re: [Ifeffit] amplitude parameter S02 larger than 1
> 
> Hi Bruce,
> 
> Thanks for the comments.
> 
> Just to make an end to this topic so far: the samples which give large
> amplitude are suspected to be more complicated than the normal
> skutterudites I showned to you last fall. Normal skutterudites data
> actually didn't cause me worried because its S02 is fitted in a
> typical range. In addition, there is one example on the paper PRB 86,
> 174106(2012) which has >3 angstrom nearest distance and the authors
> reports very normal s02 values. Therefore, I will keep tracking down
> why the amplitude is fitted to large and how to relate this amplitude
> to what we found here for my specific case.
> 
> Thank you everyone.
> 
> Best,
> Yanyun
> 
> Anyway, this test result gave me some hints to tracking down why there is
> Quoting Bruce Ravel :
> 
>> On 03/23/2015 04:10 PM, huyan...@physics.utoronto.ca wrote:
>>> 
>>> This is to follow up the *test* experiment you suggested. Attached are
>>> three Artemis files.
>>> 
>>> I chose the Fe (im-3m) structure to do the test. The normal crystal
>>> structure has its first-shell distance at 2.48 angstrom. A large
>>> structure was created so that the first-shell distance reaches 3.11
>>> angstrom.
>>> 
>>> Both normal and large structure are calculated in JFEFF in the same
>>> manner (i.e., same path number, same calculation procedure, same
>>> sigma2=0.0045 for all paths) to generate the calculated chi data.
>>> 
>>> File #1 attached was exactly following the procedure you mentioned. The
>>> quick first shell fits very well, four parameters except S02 give normal
>>> results. The amplitude parameter S02 is very large (2.2+/-0.14).
>>> 
>>> I extended the test in #2 and #3 for comparison. In #2, calculated data
>>> based on large structure is presented to fit to the same large
>>> structure.

Re: [Ifeffit] Breaking down correlationships between parameters

[Scott Calvin]

Hi Jatin,

The key is that S02 should be the same for all paths.

For example:

Suppose you are very confident path 1 has a coordination number of 6, because 
of prior knowledge you have about the system. Paths 2, 3, and 4 have unknown 
coordination numbers, however.

N1 (i.e. the degeneracy of path 1) you set to 6.

You guess S02, N2, N3, and N4. Of course, the way Artemis and Ifeffit implement 
that, you're writing N1*S02 for the S02 field of Path 1, N2*S02 for the S02 
field of Path 2, etc., and then setting N1 = 6 and guessing S02, N2, N3, and N4.

There is now enough information for S02, N2, N3, and N4 to each be fitted 
without 100% correlation.

—Scott Calvin
Sarah Lawrence College


> On Mar 23, 2015, at 8:54 AM, Rana, Jatinkumar Kantilal 
>  wrote:
>
> Hi Chris,
>
> The term N*S02 is fitted for each path of the FEFF calculation. So my 
> question is, even if we know N with a great certainty for some path, how can 
> we vary both N and S02 for other paths ? or Did I understand it wrong ?
>
> Best regards,
> Jatin
>
> -Original Message-
> From: ifeffit-boun...@millenia.cars.aps.anl.gov 
> [mailto:ifeffit-boun...@millenia.cars.aps.anl.gov] On Behalf Of 
> ifeffit-requ...@millenia.cars.aps.anl.gov
> Sent: 23 March, 2015 12:16
> To: ifeffit@millenia.cars.aps.anl.gov
> Subject: Ifeffit Digest, Vol 145, Issue 42
>
> Send Ifeffit mailing list submissions to
>ifeffit@millenia.cars.aps.anl.gov
>
> To subscribe or unsubscribe via the World Wide Web, visit
>http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> or, via email, send a message with subject or body 'help' to
>ifeffit-requ...@millenia.cars.aps.anl.gov
>
> You can reach the person managing the list at
>ifeffit-ow...@millenia.cars.aps.anl.gov
>
> When replying, please edit your Subject line so it is more specific than "Re: 
> Contents of Ifeffit digest..."
>
>
> Today's Topics:
>
>   1. Re: Breaking down correlationships between parameters
>  (Rana, Jatinkumar Kantilal)
>   2. Re: Breaking down correlationships between parameters
>  (Chris Patridge)
>
>
> --
>
> Message: 1
> Date: Mon, 23 Mar 2015 11:00:19 +
> From: "Rana, Jatinkumar Kantilal"
>
> To: "ifeffit@millenia.cars.aps.anl.gov"
>
> Subject: Re: [Ifeffit] Breaking down correlationships between
>parameters
> Message-ID: 
> Content-Type: text/plain; charset="utf-8"
>
> Hi Scott,
>
> Thank you for your comments. Can you please elaborate a little bit more on 
> this "In cases like that, both N for all paths but one and S02 can be fit 
> without 100% correlation."
>
> Best regards,
> Jatin
>
> -Original Message-
> From: ifeffit-boun...@millenia.cars.aps.anl.gov 
> [mailto:ifeffit-boun...@millenia.cars.aps.anl.gov] On Behalf Of 
> ifeffit-requ...@millenia.cars.aps.anl.gov
> Sent: 23 March, 2015 10:16
> To: ifeffit@millenia.cars.aps.anl.gov
> Subject: Ifeffit Digest, Vol 145, Issue 41
>
> Send Ifeffit mailing list submissions to
>ifeffit@millenia.cars.aps.anl.gov
>
> To subscribe or unsubscribe via the World Wide Web, visit
>http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> or, via email, send a message with subject or body 'help' to
>ifeffit-requ...@millenia.cars.aps.anl.gov
>
> You can reach the person managing the list at
>ifeffit-ow...@millenia.cars.aps.anl.gov
>
> When replying, please edit your Subject line so it is more specific than "Re: 
> Contents of Ifeffit digest..."
>
>
> Today's Topics:
>
>   1. Re: Breaking down correlationships between parameters
>  (Scott Calvin)
>   2. Re: Breaking down correlationships between parameters
>  (Matt Newville)
>   3. Re: Breaking down correlationships between parameters
>  (Rana, Jatinkumar Kantilal)
>
>
> --
>
> Message: 1
> Date: Sun, 22 Mar 2015 13:44:28 -0400
> From: Scott Calvin 
> To: XAFS Analysis using Ifeffit 
> Subject: Re: [Ifeffit] Breaking down correlationships between
>parameters
> Message-ID: 
> Content-Type: text/plain; charset="utf-8"
>
> One side-comment from me:
>
> On Mar 22, 2015, at 12:52 PM, Matt Newville 
> mailto:newvi...@cars.uchicago.edu>> wrote:
>
> N and S02 are always 100% correlated (mathematically, not merely by the 
> finite k range).
>
> Matt is saying that N and S02 are always 100% correlated for a single path. 
> But in some situations you might know N for one path but no

Re: [Ifeffit] Breaking down correlationships between parameters

[Scott Calvin]

One side-comment from me:

On Mar 22, 2015, at 12:52 PM, Matt Newville 
mailto:newvi...@cars.uchicago.edu>> wrote:

N and S02 are always 100% correlated (mathematically, not merely by the finite 
k range).

Matt is saying that N and S02 are always 100% correlated for a single path. But 
in some situations you might know N for one path but not others. For example, 
you might know that the absorbing atom is octahedrally coordinated to oxygen 
but not be as certain as to next-nearest neighbors, or that there are copper 
atoms on the corners of a simple cubic lattice with a mixture of atoms at other 
positions. In cases like that, both N for all paths but one and S02 can be fit 
without 100% correlation.

The degeneracy of multiple-scattering paths can often be constrained in terms 
of the coordination numbers for direct-scattering paths, which can further 
reduce (not “break”) the correlation.

In terms of the main question, I agree with Matt: I don’t think there’s much 
point in using the line-crossing technique nowadays; fitting using multiple 
k-weights simultaneously accomplishes the same thing but is a bit easier to 
interpret statistically.

—Scott Calvin
Sarah Lawrence College
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] amplitude parameter S02 larger than 1

[Scott Calvin]

Hi Yanyun,

Good. So here's the procedure for a Hamilton test.

We're comparing the fit with S02 guessed to the one with S02 set to 0.90, 
because that was your a priori best guess at S02.

I take the ratio of the first R-factor to the second. You didn't actually say 
the R-factor for the fit with S02 guessed, but it's clearly around 0.0055 based 
on the other information you gave. The R-factor for the 0.90 fit is 0.021. So 
the ratio is 0.0055/0.021 = 0.26, which we'll call x.

For the first fit the degrees of freedom is 31.2 - 24 = 8.2. Take half of that 
and call that a. So a is 4.1.

The first fit guesses 1 parameter that the second one doesn't. Take half of 1 
and call that b. So b is 0.5.

Find a regularized lower incomplete beta function calculator, like this one: 
http://www.danielsoper.com/statcalc3/calc.aspx?id=37

Enter x, a, and b.

The result is 0.001. This means that there is a 0.1% chance that the fits are 
actually consistent, and that the difference is just due to noise in the data. 

So in this case, we can't just explain away the high S02 as insignificant. 

Of course, you could pretty much eyeball that once you gave me the 
uncertainties; since your fit said 1.45 +/- 0.14, that's likely to be quite 
incompatible with S02 = 0.9. Still, it's nice to put that on a firmer 
statistical basis, and I've personally found the Hamilton test quite helpful 
for answering "do I need to worry about [X]?" type questions.

But in your case, you do need to worry about it. This discussion has generated 
several suggestions; hopefully one of them is a good lead!

--Scott Calvin
Sarah Lawrence College

> On Mar 20, 2015, at 4:30 PM, huyan...@physics.utoronto.ca wrote:
> 
> Hi Scott,
> 
> In all situations, 31.2 independent data points and 24 variables were
> used. In the case of setting S02 to a value, 23 variables were used.
> 
> Let me know if there is any other info needed.
> 
> Best,
> Yanyun
> 
> 
> Quoting Scott Calvin :
> 
>> Hi Yanyun,
>> 
>> To actually do a Hamilton test, the one other thing I need to know
>> the number of degrees of freedom in the fit...if you provide that,
>> I'll walk you through how to actually do a Hamilton test--it's not
>> that bad, with the aid of an online calculator, and I think it might
>> be instructive for some of the other people reading this list who
>> are trying to learn EXAFS.
>> 
>> --Scott Calvin
>> Sarah Lawrence College
>> 
>> 
>>> On Mar 20, 2015, at 3:46 PM, huyan...@physics.utoronto.ca wrote:
>>> 
>>> Hi Scott,
>>> 
>>> Thank you so much for giving me your thought again. It is very helpful
>>> to know how you and other XAFS experts deal with unusual situations.
>>> 
>>> The floating S02 is fitted to be 1.45+/-0.14, this just means the fit
>>> doesn't like the idea of an S02 in a typical range. Instead of setting
>>> S02 to 0.9, I have to figure out why it happens and what it might
>>> indicate.
>>> 
>>> I guess a Hamilton test is done by adjusting one parameter (i.e., S02)
>>> while keeping other conditions and model the same. Is that right?  So
>>> I record this test as following:
>>> 
>>> 1) Floating S02: S02 fits to 1.45+/-0.14, R=0.0055, reduced
>>> chi^2=17.86, Percentage=0.53+/-0.04
>>> 2) Set S02=0.7, R=0.044, reduced chi^2=120.6, percentage=0.81+/-0.2
>>> 3) set S02=0.8, R=0.030, reduced chi^2=86.10, percentage=0.77+/-0.07
>>> 3) set S02=0.9, R=0.021, reduced chi^2=60.16, percentage=0.72+/-0.06
>>> 4) set S02=1.0, R=0.017, reduced chi^2=49.5, percentage=0.67+/-0.05
>>> 5) set S02=1.1, R=0.012, reduced chi^2=35.1, percentage=0.62+/-0.03
>>> 6) set S02=1.2, R=0.009, reduced chi^2=24.9, percentage=0.59+/-0.02
>>> 7) set S02=1.3, R=0.007, reduced chi^2=18.9, percentage=0.57+/-0.02
>>> 8) set S02=1.4, R=0.0057, reduced chi^2=16.1, percentage=0.55+/-0.02
>>> 9) Floating S02 to be 1.45+/-0.14
>>> 10) set S02=1.6, R=0.006, reduced chi^2=17.8, percentage=0.53+/- 0.02
>>> 11) set S02=2.0, R=0.044, reduced chi^2=120.7, percentage=0.37+/-0.06.
>>> 
>>> Therefore, I will say S02 falling in the range 1.2~1.6 gives
>>> statistically improved fit, but S02=0.9 is not terrible as well. I
>>> agree with you that I could always be confident to say the percentage
>>> is 0.64+/-0.15, but I do want to shrink down the uncertainty and think
>>> about other possibilities that could cause a large S02.
>>> 
>>> I did double-check the data-reduction and normalization process. I
>>> don't think I can improve anything in this step. By the way

Re: [Ifeffit] amplitude parameter S02 larger than 1

[Scott Calvin]

Hi Yanyun,

To actually do a Hamilton test, the one other thing I need to know the number 
of degrees of freedom in the fit...if you provide that, I'll walk you through 
how to actually do a Hamilton test--it's not that bad, with the aid of an 
online calculator, and I think it might be instructive for some of the other 
people reading this list who are trying to learn EXAFS.

--Scott Calvin
Sarah Lawrence College


> On Mar 20, 2015, at 3:46 PM, huyan...@physics.utoronto.ca wrote:
> 
> Hi Scott,
> 
> Thank you so much for giving me your thought again. It is very helpful
> to know how you and other XAFS experts deal with unusual situations.
> 
> The floating S02 is fitted to be 1.45+/-0.14, this just means the fit
> doesn't like the idea of an S02 in a typical range. Instead of setting
> S02 to 0.9, I have to figure out why it happens and what it might
> indicate.
> 
> I guess a Hamilton test is done by adjusting one parameter (i.e., S02)
> while keeping other conditions and model the same. Is that right?  So
> I record this test as following:
> 
> 1) Floating S02: S02 fits to 1.45+/-0.14, R=0.0055, reduced
> chi^2=17.86, Percentage=0.53+/-0.04
> 2) Set S02=0.7, R=0.044, reduced chi^2=120.6, percentage=0.81+/-0.2
> 3) set S02=0.8, R=0.030, reduced chi^2=86.10, percentage=0.77+/-0.07
> 3) set S02=0.9, R=0.021, reduced chi^2=60.16, percentage=0.72+/-0.06
> 4) set S02=1.0, R=0.017, reduced chi^2=49.5, percentage=0.67+/-0.05
> 5) set S02=1.1, R=0.012, reduced chi^2=35.1, percentage=0.62+/-0.03
> 6) set S02=1.2, R=0.009, reduced chi^2=24.9, percentage=0.59+/-0.02
> 7) set S02=1.3, R=0.007, reduced chi^2=18.9, percentage=0.57+/-0.02
> 8) set S02=1.4, R=0.0057, reduced chi^2=16.1, percentage=0.55+/-0.02
> 9) Floating S02 to be 1.45+/-0.14
> 10) set S02=1.6, R=0.006, reduced chi^2=17.8, percentage=0.53+/- 0.02
> 11) set S02=2.0, R=0.044, reduced chi^2=120.7, percentage=0.37+/-0.06.
> 
> Therefore, I will say S02 falling in the range 1.2~1.6 gives
> statistically improved fit, but S02=0.9 is not terrible as well. I
> agree with you that I could always be confident to say the percentage
> is 0.64+/-0.15, but I do want to shrink down the uncertainty and think
> about other possibilities that could cause a large S02.
> 
> I did double-check the data-reduction and normalization process. I
> don't think I can improve anything in this step. By the way, I have a
> series of similar samples and their fittings all shows floating S02
> larger than one based on the same two-sites model.
> 
> Best,
> Yanyun
> 
> 
> 
> 
> Quoting Scott Calvin :
> 
>> Hi Yanyun,
>> 
>> Lots of comments coming in now, so I’m editing this as I write it!
>> 
>> One possibility for why you're getting a high best-fit S02 is that
>> the fit doesn't care all that much about what the value of S02; i.e.
>> there is broad range of S02's compatible with describing the fit as
>> "good." That should be reflected in the uncertainty that Artemis
>> reports. If S02 is 1.50 +/- 0.48, for example, that means the fit
>> isn't all that "sure" what S02 should be. That would mean we could
>> just shrug our shoulders and move on, except that it correlates with
>> a parameter you are interested in (in this case, site occupancy). So
>> in such a case, I think you can cautiously fall back on what might
>> be called a "Bayesian prior"; i.e., the belief that the S02 should
>> be "around" 0.9, and set the S02 to 0.9. (Or perhaps restrain S02 to
>> 0.9; then you're really doing something a bit more like the notion
>> of a Bayesian prior.)
>> 
>> On the other hand, if the S02 is, say, 1.50 +/- 0.07, then the fit
>> really doesn’t like the idea of an S02 in the typical range. An S02
>> that high, with that small an uncertainty, suggests to me that
>> something is wrong—although it could be as simple as a normalization
>> issue during data reduction. In that case, I’d be more skeptical of
>> just setting S02 to 0.90 and going with that result; the fit is
>> trying to tell you something, and it’s important to track down what
>> that something is.
>> 
>> Of course, once in a while, a fit will find a local minimum, while
>> there’s another good local minimum around a more realistic value.
>> That would be reflected by a fit that gave similarly good
>> quantitative measures of fit quality (e.g. R-factors) when S02 is
>> fit (and yields 1.50 +/- 0.07) as when its forced to 0.90. That’s
>> somewhat unusual, however, particularly with a global parameter like
>> S02.
>> 
>> A good way to defend setting S02 to 0.90 is to use the Hamilton test

Re: [Ifeffit] amplitude parameter S02 larger than 1

[Scott Calvin]

Hi again,

I'm not sure about your structure overall, but including more distant 
scattering shells can help reduce that particular correlation, if the data 
supports it and if you enough confidence in what the lattice looks like to 
introduce more distant shells in to your model without having to fit lots of 
new free parameters.

--Scott Calvin
Sarah Lawrence College

> On Mar 20, 2015, at 1:44 PM, huyan...@physics.utoronto.ca wrote:
> 
> Hi Josh and Anatoly,
> 
> Thank you for bringing up this point.
> 
> I agree that P would be 100% correlated with S02 if we assume both
> sites have the same first-shell distance. In my fit, the correlation
> between the percentage parameter P and S02 is -0.73.
> 
> Do you have any good idea on how to break this kind of correlation, or
> alternative way to find the 'true S02' and 'true P'?
> 
> Best,
> Yanyun
> 
> 
> Quoting Joshua Kas :
> 
>> Anatoly makes a very good point, and his example can also be used to show
>> that there could be very large correlation between s02 and your percentage
>> parameter. For example, if you assume that the first shell distances are
>> the same for the two sites, and take Anatoly's example of 4 and 6 nearest
>> neighbors, then your parameter P is 100 percent correlated with s02, i.e.,
>> all it does is change the coordination smoothly from 4 to 6. In a real
>> system this probably won't happen, but the correlation could still be
>> large. Have you looked at it?
>> 
>> Josh
>> 
>> On Fri, Mar 20, 2015 at 10:00 AM, >> wrote:
>> 
>>> Send Ifeffit mailing list submissions to
>>>ifeffit@millenia.cars.aps.anl.gov
>>> 
>>> To subscribe or unsubscribe via the World Wide Web, visit
>>>http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>>> or, via email, send a message with subject or body 'help' to
>>>ifeffit-requ...@millenia.cars.aps.anl.gov
>>> 
>>> You can reach the person managing the list at
>>>ifeffit-ow...@millenia.cars.aps.anl.gov
>>> 
>>> When replying, please edit your Subject line so it is more specific
>>> than "Re: Contents of Ifeffit digest..."
>>> 
>>> 
>>> Today's Topics:
>>> 
>>>   1. Re: amplitude parameter S02 larger than 1 (Anatoly I Frenkel)
>>> 
>>> 
>>> --
>>> 
>>> Message: 1
>>> Date: Fri, 20 Mar 2015 16:57:02 +
>>> From: Anatoly I Frenkel 
>>> To: XAFS Analysis using Ifeffit 
>>> Subject: Re: [Ifeffit] amplitude parameter S02 larger than 1
>>> Message-ID:
>>><
>>> 545f89325f51764c902d37b42c9f4d530148fda...@yuwexcpm11.yuad.uds.yu.edu>
>>> 
>>> Content-Type: text/plain; charset="us-ascii"
>>> 
>>> One possible scenario: If one site has 6 nearest neighbors and the other -
>>> 4, and if you choose the site with 4 neighbors to construct FEFF to model
>>> your EXAFS data; and if you set your degeneracy equal to 4 and make your
>>> amplitude factor as S02*x of one site + S02*(1-x) of another site, then
>>> your S02 will come out larger than it should be because it will compensate
>>> for the fact that you underestimate the degeneracy of the 6-coordinated
>>> site.
>>> 
>>> Anatoly
>>> 
>>> 
>>> From: ifeffit-boun...@millenia.cars.aps.anl.gov [
>>> ifeffit-boun...@millenia.cars.aps.anl.gov] on behalf of
>>> huyan...@physics.utoronto.ca [huyan...@physics.utoronto.ca]
>>> Sent: Friday, March 20, 2015 12:04 PM
>>> To: XAFS Analysis using Ifeffit
>>> Subject: Re: [Ifeffit] amplitude parameter S02 larger than 1
>>> 
>>> Hi Chris,
>>> 
>>> Thank you for your suggestion. But I don't have a standard. Also, I
>>> wonder whether a multiple-site situation could be different from the
>>> normal one-site case with respect to S02.
>>> 
>>> Best,
>>> Yanyun
>>> Quoting Chris Patridge :
>>> 
>>>> Hi,
>>>> 
>>>> One thing that could be considered is transferring the SO2 factor
>>>> from a reliable source such as a standard and then use that value in
>>>> the fit. Chemical transferability of SO2 to similar systems is often
>>>> acceptable. You could also try constraining the value in the fit as
>>>> well.  SO2 and Debye are also correlated so this ma

Re: [Ifeffit] amplitude parameter S02 larger than 1

[Scott Calvin]

Hi Yanyun,

Lots of comments coming in now, so I’m editing this as I write it!

One possibility for why you're getting a high best-fit S02 is that the fit 
doesn't care all that much about what the value of S02; i.e. there is broad 
range of S02's compatible with describing the fit as "good." That should be 
reflected in the uncertainty that Artemis reports. If S02 is 1.50 +/- 0.48, for 
example, that means the fit isn't all that "sure" what S02 should be. That 
would mean we could just shrug our shoulders and move on, except that it 
correlates with a parameter you are interested in (in this case, site 
occupancy). So in such a case, I think you can cautiously fall back on what 
might be called a "Bayesian prior"; i.e., the belief that the S02 should be 
"around" 0.9, and set the S02 to 0.9. (Or perhaps restrain S02 to 0.9; then 
you're really doing something a bit more like the notion of a Bayesian prior.)

On the other hand, if the S02 is, say, 1.50 +/- 0.07, then the fit really 
doesn’t like the idea of an S02 in the typical range. An S02 that high, with 
that small an uncertainty, suggests to me that something is wrong—although it 
could be as simple as a normalization issue during data reduction. In that 
case, I’d be more skeptical of just setting S02 to 0.90 and going with that 
result; the fit is trying to tell you something, and it’s important to track 
down what that something is.

Of course, once in a while, a fit will find a local minimum, while there’s 
another good local minimum around a more realistic value. That would be 
reflected by a fit that gave similarly good quantitative measures of fit 
quality (e.g. R-factors) when S02 is fit (and yields 1.50 +/- 0.07) as when its 
forced to 0.90. That’s somewhat unusual, however, particularly with a global 
parameter like S02.

A good way to defend setting S02 to 0.90 is to use the Hamilton test to see if 
floating S02 yields a statistically significant improvement over forcing it to 
0.90. If not, using your prior best estimate for S02 is reasonable.

If you did that, though, I’d think that it would be good to mention what 
happened in any eventual publication of presentation; it might provide an 
important clue to someone who follows up with this or a similar system. It 
would also be good to increase your reported uncertainty for site occupancy 
(and indicate in the text what you’ve done). I now see that your site 
occupancies are 0.53 +/- 0.04 for the floated S02, and 0.72 +/-0.06 for the S02 
= 0.90. That’s not so bad, really. It means that you’re pretty confident that 
the site occupancy is 0.64 +/- 0.15, which isn’t an absurdly large uncertainty 
as these things go.

To be concrete, if all the Hamilton test does not show statistically 
significant improvement by floating S02, then I might write something like this 
in any eventual paper: “The site occupancy was highly correlated with S02 in 
our fits, making it difficult to determine the site occupancy with high 
precision. If S02 is constrained to 0.90, a plausible value for element [X] 
[ref], then the site occupancy is 0.53 +/- 0.04. If constrained to 1.0, the 
site occupancy is [whatever it comes out to be] To reflect the increased 
uncertainty associated with the unknown value for S02, we are adopting a value 
of 0.53 +/- [enough uncertainty to cover the results found for S02 = 1.0].

Of course, if you do that, I’d also suggest tracking down as many other 
possibilities for why your fit is showing high values of S02 as you can; e.g., 
double-check your normalization during data reduction.

If, on the other hand, the Hamilton test does show the floated S02 is yielding 
a statistically significant improvement, I think you have a bigger issue. 
Looking at, e.g., whether you may have constrained coordination numbers 
incorrectly becomes more critical.

—Scott Calvin
Sarah Lawrence College



On Mar 20, 2015, at 12:48 PM, 
huyan...@physics.utoronto.ca<mailto:huyan...@physics.utoronto.ca> wrote:

Hi Scott,

Thank you. Our group has one copy of your book, I'll read it again
after my colleague return it to shelf. I still want to continue our
discussion here:

If we treat S02 as an empirically observed parameter, can I just set
S02=0.9 or 1.45 and let other parameters to explain the k- and R-
dependence? Because S02 is not a simplistic parameter which may
include both theory and experimental effects, I feel that S02 is not
necessarily to be smaller than 1, although I admit S02 smaller than 1
is more defensible as it represents some limitations both in theory
model and experiment, but I have a series of similar sample and all
their S02 will be automatically be fitted to 1.45~1.55, not smaller
than 1. Could this indicate something?

I actually found in my system, when I set S02=0.9 (instead of letting
it fit to 1.45), other parameter will definitely change but the
fitting is not terrible, it is still a close fit but im

Re: [Ifeffit] amplitude parameter S02 larger than 1

[Scott Calvin]

Hi Yanyun,

I am hesitant to promote a commercial project from which I directly profit on 
this list, but it seems to me you are asking a bigger set of questions than can 
comfortably and sufficiently be answered in this format, and they are questions 
which have been answered in detail elsewhere.

In my book XAFS for Everyone, I have four pages devoted solely to S02, along 
with related information elsewhere in the book.

Since you have a University of Toronto address, I am guessing you have access 
to their library. If you don't wish to purchase the book, you can request it 
via interlibrary loan, at no cost to you or your institution. 

In the mean time, a quote from the book that may be useful in thinking about 
S02:

"Alternatively, one can treat So2 as a phenomenological parameter that accounts 
for any amplitude suppression independent of k and R, regardless of physical 
cause (Krappe and 
Rossner 2004). Under this view, So2 does not have any particular physical 
meaning, and the k or R dependence of intrinsic losses can be assigned to other 
parameters."

That's the way I usually think about it--as not having a single physical 
meaning, but rather as being an empirically observed correction factor relative 
to simplistic theories which is indicative both of experimental effects and 
limitations in the theoretical model.

Hope that helps...

--Scott Calvin
Sarah Lawrence College

> On Mar 19, 2015, at 6:32 PM, huyan...@physics.utoronto.ca wrote:
> 
> Hi all,
> 
> I know this question has been asked for many times. S02 is expected to  
> be around, but smaller than 1,  a fact that has been explained, such  
> as in the following previous emails, in our mailing list.
> 
> http://www.mail-archive.com/ifeffit%40millenia.cars.aps.anl.gov/msg02237.html
> http://millenia.cars.aps.anl.gov/pipermail/ifeffit/2003-February/000230.html
> 
> However, I am continually get S02 value larger than 1 for a series of  
> similar samples when I fit data in Artemis. I think my fit is very  
> good, because my suspected model(based on other technique) could be  
> verified in XAFS analysis (i.e., defensible in physics), the  
> statistics is good ( R=0.01, reduced chi-square=31.4, fit-range:1.5~6  
> Angstrom, k-range: 3~14 angstrom-1) and all the parameters such as the  
> bond length, sigma2 are physically reasonable. The only thing makes me  
> uncomfortable is that parameter S02 keeps between 1.45 to 1.55 during  
> the fitting.
> 
> In my system, the absorber atom occupies two crystallographic sites.  
> So I built a model with paths generated from two FEFF calculations.  
> For paths generated from the 1st and 2nd FEFF calculation, the  
> amplitude parameters are set to be S02*P% and S02*(1-P%) respectively,  
> where P% is the first site occupancy percentage. Both S02 and P are  
> free parameters during the fit, and P is an important conclusion I  
> want to extract from XAFS fitting.
> 
> However, the fit result gives me S02=1.45 ~ 1.55 and P=0.51 ~ 0.56 all  
> the time (i.e., for each path the 'total amplitude' S02*P% or  
> S02*(1-P%) are about 0.7~0.8, smaller than 1). It looks to me that I  
> got a 'perfect' fit but I am not sure if S02 larger than one is  
> defensible. So I have to ask:
> 
> 1) Is my current fit with S02 larger than one reasonable? If not, what  
> could be suggested to get around it?
> 
> 2) What's the meaning of S02? It is interpreted in physics that it is  
> a reduced electron excitation parameter, but is it possible that S02  
> will be affected by any experimental condition?
> 
> 3) Can anyone share whether you had the multiple site system that gets  
> S02 larger than one?
> 
> Looking forward to your help.
> 
> Best,
> Yanyun
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] A small query on constricting mean square displacement in multi-edge EXAFS refinement

[Scott Calvin]

Yes.

The only minor adjustment I can think of to that is something called the 
“McMaster correction,” which adjusts for an oversimplification in the usual 
EXAFS method of normalization, and thus can be different for the two edges. But 
that correction is generally quite small, and the difference in it between two 
edges is even smaller.

—Scott Calvin
Sarah Lawrence College

> On Dec 3, 2014, at 7:54 AM, Badari Rao  wrote:
> 
> Dear All,
> 
> Suppose I have a compound, say AB. I have the EXAFS data from both A-edge as 
> well as B-edge and I am simultaneously refining these data.
> Would it be correct to force the mean square displacement of the A-B path in 
> both the edges to be the same?
> 
> -- Regards,
> Badari Narayana. A. R.
> Ph.D Student, Nanoscience and Nanotechnology,
> c/o Prof Rajeev Ranjan,
> Electroceramics Lab,
> Department of Materials Engineering,
> Indian Institute of Science.
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] multiple absorber atoms

[Scott Calvin]

Hi Jesus,

Just make a path list for each crystallographic site, and then weight S02 for 
the paths by the fraction of absorbing atoms in the associated site.

--Scott Calvin
Sarah Lawrence College


On Oct 20, 2014, at 10:00 AM, Jesús Eduardo Vega Castillo  
wrote:

> Hello,
> 
> This is my first message to the list so I expect to do it right.
> 
> I need to generate a path list for EXAFS fitting within ifeffit, based on a 
> crystalline structure that has two different crystallographic positions for 
> the absorber atom (edge K of Mo). Is there any way to do this? 
> 
> I would appreciate any information you could give me.
> 
> Thanks in advance
> 
> Jesus
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Inconsistency of the amplitude reduction factor

[Scott Calvin]

This is an excellent point, and I'd like to elaborate yet further. :)

How does one tell whether transferability is justified or not, based on the 
experimental data?

Perform fits with the S02 constrained to some reasonable value (perhaps the 
average of your results). Compare the reduced chi square reported by the fit 
for the case where S02 was guessed to the one where it was constrained. If the 
reduced chi square is significantly higher for the constrained case, then 
transferability is not justified; i.e., the data is fit better, in a 
statistical sense, when S02's are fit separately for each spectrum. If it's 
lower or about the same, then transferability is justified. (But keep in mind 
that an apparent lack of transferability might be due to inconsistent 
determinations of the edge jumps.)

How much higher is "significantly" higher? There's a statistical test called 
the Hamilton test that can help you decide that, if you really want to be 
careful about things. But looking at the reduced chi square is a good first 
step.

Note that what I'm saying is closely related to Bruce's other comment about 
including error bars on the reported values. If the reported ranges including 
error bars all overlap, it's also likely that you'll find that constraining all 
the fits to have the same value of S02 will not increase the reduced chi 
squares significantly.

So what if transferability is justified by the data (i.e. compatible with it)? 
Does that mean you should do it? 

At that point the decision which to present as the "primary" fit becomes to 
some extent aesthetic. The statistics are then telling you that the data is 
consistent with transferability, but of course there may be some "actual" 
variations due, perhaps, to the cause Bruce suggested. Would you prefer to 
explain in a paper why you are constraining the S02's to be the same, or why 
you have different values for them? In either case, if you had performed fits 
both ways, you could look at the values generated by both sets of fits and see 
if they were consistent; that is, if the ranges specified by their error bars 
overlap. If that were the case, you could say so in your paper, and then 
whether or not you constrained the S02s to be the same becomes mostly a 
non-issue for readers. And if the ranges didn't overlap, then you should report 
that too, suggesting that there is some additional uncertainty based on your 
results based on the question of S02.

 --Scott Calvin
Sarah Lawrence College


On Sep 18, 2014, at 9:04 AM, Bruce Ravel  wrote:

> On 09/18/2014 07:23 AM, Scott Calvin wrote:
>> Hi Hoon,
>> 
>> Using a reference value is not always a good idea, because experimental
>> effects can play a role.
> 
> I want to elaborate a bit on this point in Scott's post, particularly 
> given the nature of your sample.
> 
> The materials in a battery, as they charge or discharge, can undergo 
> changes in morphology.  Morphology can have an impact on the measured 
> amplitude.  More specifically, inhomogeneity in the sample -- pinholes, 
> for instance -- have a known effect on the amplitude.  See, for example:
> 
>   http://dx.doi.org/10.1103/PhysRevB.23.3781
>   http://dx.doi.org/10.1016/0167-5087(83)90730-5
>   http://gbxafs.iit.edu/training/thickness_effects.pdf
> 
> My point is that you may need to question the assumption that S02 even 
> should be transferable in your measurement.
> 
> B
> 
> 
>> BUT, S02 should not change during charge-discharge on a single sample,
>> or a series of samples prepared and measured similarly. Instead, it's
>> likely something correlated with S02 in the fit is changing, and so the
>> fitting routine is getting a bit confused and attributing part of the
>> change to S02. (That's not a knock on the fitting routine; it doesn't
>> know any better unless you tell it!)
>> 
>> I think the best recommendation is to do a simultaneous fit on multiple
>> spectra, constraining the S02 to be the same for each. So you're still
>> fitting S02, but forcing all the spectra to use the same value.
>> 
>> Second best is to fit one spectrum and allow S02 to vary, and then
>> constrain all the other fits to use that value.
>> 
>> --Scott Calvin
>> Sarah Lawrence College
>> 
>> On Sep 18, 2014, at 7:06 AM, HOON Kim > <mailto:scie...@live.co.kr>> wrote:
>> 
>>> Hello,
>>> 
>>> I am a bit confused about the amplitude reduction factor (S0^2), in a
>>> sense that whether this factor must be determined by fitting or
>>> constrained by a reference value for a specific element. I'm dealing
>>> with a cathode composite (for lithium-ion battery) comprised of two
>>&g

Re: [Ifeffit] Inconsistency of the amplitude reduction factor

[Scott Calvin]

By the way, you should also look closely at how well the edge jump is being 
determined by Athena (or whatever software you use for the data reduction). If 
there's a white line or pre-edge feature that's changing, it's possible the 
default behavior of Athena is not determining the edge jump in a consistent 
manner. That would directly lead to an apparent change in S02. The solution in 
that case is to adjust the parameters in Athena to assure the edge jump is 
determined in a consistent fashion.

--Scott Calvin
Sarah Lawrence College

On Sep 18, 2014, at 8:20 AM, HOON Kim 
mailto:scie...@live.co.kr>> wrote:

Dear Scott,

Thank you for your advice.

Q) I assume that your second solution allow different S02 values for each 
spectrum?

I'd like to be more specific about my obtained results,

At reference state (that means without any applied current), the amplitude 
reduction factor was 0.77
At 20% charge, it was 0.67
40% charge = 0.66
60% charge = 0.63
80% charge = 0.68
full charge = 0.67

Most of the fitting parameters were reasonable and the result well corroborates 
with previous study (papers) from other group (but I'm not sure about the value 
of the amplitude reduction factor as there were little information about it).

Kind Regards,
HOON


From: scal...@sarahlawrence.edu<mailto:scal...@sarahlawrence.edu>
To: ifeffit@millenia.cars.aps.anl.gov<mailto:ifeffit@millenia.cars.aps.anl.gov>
Date: Thu, 18 Sep 2014 07:23:53 -0400
Subject: Re: [Ifeffit] Inconsistency of the amplitude reduction factor

Hi Hoon,

Using a reference value is not always a good idea, because experimental effects 
can play a role.

BUT, S02 should not change during charge-discharge on a single sample, or a 
series of samples prepared and measured similarly. Instead, it's likely 
something correlated with S02 in the fit is changing, and so the fitting 
routine is getting a bit confused and attributing part of the change to S02. 
(That's not a knock on the fitting routine; it doesn't know any better unless 
you tell it!)

I think the best recommendation is to do a simultaneous fit on multiple 
spectra, constraining the S02 to be the same for each. So you're still fitting 
S02, but forcing all the spectra to use the same value.

Second best is to fit one spectrum and allow S02 to vary, and then constrain 
all the other fits to use that value.

--Scott Calvin
Sarah Lawrence College

On Sep 18, 2014, at 7:06 AM, HOON Kim 
mailto:scie...@live.co.kr>> wrote:

Hello,

I am a bit confused about the amplitude reduction factor (S0^2), in a sense 
that whether this factor must be determined by fitting or constrained by a 
reference value for a specific element. I'm dealing with a cathode composite 
(for lithium-ion battery) comprised of two crystal phases. During 
charge-discharge, the amplitude reduction factor changes and at a certain state 
of charge (SOC) it changes a lot  such as from 0.77 to 0.67. My understanding 
is that it may reflect the phase transition of the material into the amplitude 
reduction factor. But, I'm not sure ... I need advice on this.

Thank you !

Kind Regards,
HOON
___

___ Ifeffit mailing list 
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Inconsistency of the amplitude reduction factor

[Scott Calvin]

On Sep 18, 2014, at 8:20 AM, HOON Kim 
mailto:scie...@live.co.kr>> wrote:

Dear Scott,

Thank you for your advice.

Q) I assume that your second solution allow different S02 values for each 
spectrum?

No. My second solution is to fit one of them, say the reference at 0.77, and 
then force all of the other fits to use the value 0.77.

--Scott Calvin
Sarah Lawrence College


I'd like to be more specific about my obtained results,

At reference state (that means without any applied current), the amplitude 
reduction factor was 0.77
At 20% charge, it was 0.67
40% charge = 0.66
60% charge = 0.63
80% charge = 0.68
full charge = 0.67

Most of the fitting parameters were reasonable and the result well corroborates 
with previous study (papers) from other group (but I'm not sure about the value 
of the amplitude reduction factor as there were little information about it).

Kind Regards,
HOON


From: scal...@sarahlawrence.edu<mailto:scal...@sarahlawrence.edu>
To: ifeffit@millenia.cars.aps.anl.gov<mailto:ifeffit@millenia.cars.aps.anl.gov>
Date: Thu, 18 Sep 2014 07:23:53 -0400
Subject: Re: [Ifeffit] Inconsistency of the amplitude reduction factor

Hi Hoon,

Using a reference value is not always a good idea, because experimental effects 
can play a role.

BUT, S02 should not change during charge-discharge on a single sample, or a 
series of samples prepared and measured similarly. Instead, it's likely 
something correlated with S02 in the fit is changing, and so the fitting 
routine is getting a bit confused and attributing part of the change to S02. 
(That's not a knock on the fitting routine; it doesn't know any better unless 
you tell it!)

I think the best recommendation is to do a simultaneous fit on multiple 
spectra, constraining the S02 to be the same for each. So you're still fitting 
S02, but forcing all the spectra to use the same value.

Second best is to fit one spectrum and allow S02 to vary, and then constrain 
all the other fits to use that value.

--Scott Calvin
Sarah Lawrence College

On Sep 18, 2014, at 7:06 AM, HOON Kim 
mailto:scie...@live.co.kr>> wrote:

Hello,

I am a bit confused about the amplitude reduction factor (S0^2), in a sense 
that whether this factor must be determined by fitting or constrained by a 
reference value for a specific element. I'm dealing with a cathode composite 
(for lithium-ion battery) comprised of two crystal phases. During 
charge-discharge, the amplitude reduction factor changes and at a certain state 
of charge (SOC) it changes a lot  such as from 0.77 to 0.67. My understanding 
is that it may reflect the phase transition of the material into the amplitude 
reduction factor. But, I'm not sure ... I need advice on this.

Thank you !

Kind Regards,
HOON
___

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Inconsistency of the amplitude reduction factor

[Scott Calvin]

Hi Hoon,

Using a reference value is not always a good idea, because experimental effects 
can play a role.

BUT, S02 should not change during charge-discharge on a single sample, or a 
series of samples prepared and measured similarly. Instead, it's likely 
something correlated with S02 in the fit is changing, and so the fitting 
routine is getting a bit confused and attributing part of the change to S02. 
(That's not a knock on the fitting routine; it doesn't know any better unless 
you tell it!)

I think the best recommendation is to do a simultaneous fit on multiple 
spectra, constraining the S02 to be the same for each. So you're still fitting 
S02, but forcing all the spectra to use the same value.

Second best is to fit one spectrum and allow S02 to vary, and then constrain 
all the other fits to use that value.

--Scott Calvin
Sarah Lawrence College

On Sep 18, 2014, at 7:06 AM, HOON Kim 
mailto:scie...@live.co.kr>> wrote:

Hello,

I am a bit confused about the amplitude reduction factor (S0^2), in a sense 
that whether this factor must be determined by fitting or constrained by a 
reference value for a specific element. I'm dealing with a cathode composite 
(for lithium-ion battery) comprised of two crystal phases. During 
charge-discharge, the amplitude reduction factor changes and at a certain state 
of charge (SOC) it changes a lot  such as from 0.77 to 0.67. My understanding 
is that it may reflect the phase transition of the material into the amplitude 
reduction factor. But, I'm not sure ... I need advice on this.

Thank you !

Kind Regards,
HOON
___
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Questions for Artemis fit

[Scott Calvin]

Hi Yanyun,

The shift observed in R-space is due to the effect of the potential of the 
absorbing and scattering atoms. 

You are asking a series of questions that, while thoughtful, are covered in 
basic tutorials and XAFS textbooks. I suggest you search for introductions to 
XAFS analysis, many of which are freely available and which come in many forms 
(videos, tutorial files, wikis, PowerPoints, chapters/books, etc.) and work 
through one or more of them.

Beginners are encouraged to use this list, but beginners and advanced users 
alike are also urged to see if the answer to their questions is readily and 
clearly available elsewhere first.

--Scott Calvin
Sarah Lawrence College

On Aug 20, 2014, at 1:11 AM, huyan...@physics.utoronto.ca wrote:

> Hi all,
> 
> Does Feff take into account of the radius of the each atom when it  
> calculates atom-pair distance? I am wondering if the calculated  
> atom-pair distance (Reff) in each SS path has already excluded the  
> radius size of absorber and scatterer atoms? In the R-space spectra, I  
> always have a big peak at R smaller than any of the calculated path  
> distances, in other words, I feel there is no possible path  
> contributing to this peak.
> 
> My goal is trying to understand where the filler atom is in a unit  
> cell, so if the above question is 'yes', the true position of the  
> filler atoms is very likely to be far away from the site I input in  
> the ATOM file ?
> 
> Best regards,
> Yanyun
> 
> 
> Quoting Scott Calvin :
> 
>> Hi Yanyun,
>> 
>> This topic has been dealt with multiple times on the mailing list.  
>> Please search the archive:
>> 
>> http://www.mail-archive.com/ifeffit@millenia.cars.aps.anl.gov/
>> 
>> For example, searching for "sites" will give you several discussions  
>> of the topic.
>> 
>> --Scott Calvin
>> Sarah Lawrence College
>> 
>> On Aug 19, 2014, at 2:49 PM,   
>>  wrote:
>> 
>>> Hi all,
>>> 
>>> How would you deal with the case that the absorber atoms are occupying
>>> two different crystallography sites? Should I still use/guess the same
>>> 'S02' for paths calculated from two sites?
>>> 
>>> Best,
>>> Yanyun
>>> 
>>> 
>>> Quoting huyan...@physics.utoronto.ca:
>>> 
>>>> Hi Scott,
>>>> 
>>>> Thank you for giving me detailed examples. I will go with your
>>>> suggestion and try to reduce free parameters.
>>>> 
>>>> Best,
>>>> Yanyun
>>>> Quoting Scott Calvin :
>>>> 
>>>>> Hi Yanyun,
>>>>> 
>>>>> On Aug 19, 2014, at 12:05 PM, huyan...@physics.utoronto.ca wrote:
>>>>>> 
>>>>>> But my question is, for each path there are a set of Path Parameters,
>>>>>> does that mean we have to fit this set of path parameters independent
>>>>>> from those used in other paths?  For instance, if I am going to
>>>>>> include 30 paths in my fitting, do I need to guess about 150
>>>>>> (30*5=150) parameters (except guess the same 'enot' for all paths)?
>>>>> 
>>>>> No. One of the great principles of Ifeffit (and thus Artemis) is
>>>>> that the path parameters do not have to be the same as the fit
>>>>> parameters.
>>>>> 
>>>>> As a simple example, it's good to start with the model that all
>>>>> paths have the same value for E0. So you can guess a single
>>>>> parameter and then use that parameter for the E0 for all paths.
>>>>> 
>>>>> As another simple example, a cubic crystal might be modelled as
>>>>> having a uniform thermal (or Vegard's law) expansion. Then there
>>>>> could be a single guessed parameter indicating the fractional
>>>>> expansion (called, for example, alpha) while the delr for each path
>>>>> could then be entered as alpha*reff.
>>>>> 
>>>>> Much more complicated constraint schemes are possible, but for fits
>>>>> with large number of paths, the number of guessed parameters is
>>>>> almost always much much lower than the number of path parameters.
>>>>> 
>>>>>> 
>>>>>> As shown on the fit Log file, the Correlation value between two fit
>>>>>> parameters is the bigger the better, or the smaller the better?
>>>>> 
>>>>> Not necessarily. Correlations are provided to hel

Re: [Ifeffit] Questions for Artemis fit

[Scott Calvin]

Hi Yanyun,

This topic has been dealt with multiple times on the mailing list. Please 
search the archive:

http://www.mail-archive.com/ifeffit@millenia.cars.aps.anl.gov/

For example, searching for "sites" will give you several discussions of the 
topic.

--Scott Calvin
Sarah Lawrence College

On Aug 19, 2014, at 2:49 PM,  
 wrote:

> Hi all,
> 
> How would you deal with the case that the absorber atoms are occupying  
> two different crystallography sites? Should I still use/guess the same  
> 'S02' for paths calculated from two sites?
> 
> Best,
> Yanyun
> 
> 
> Quoting huyan...@physics.utoronto.ca:
> 
>> Hi Scott,
>> 
>> Thank you for giving me detailed examples. I will go with your  
>> suggestion and try to reduce free parameters.
>> 
>> Best,
>> Yanyun
>> Quoting Scott Calvin :
>> 
>>> Hi Yanyun,
>>> 
>>> On Aug 19, 2014, at 12:05 PM, huyan...@physics.utoronto.ca wrote:
>>>> 
>>>> But my question is, for each path there are a set of Path Parameters,
>>>> does that mean we have to fit this set of path parameters independent
>>>> from those used in other paths?  For instance, if I am going to
>>>> include 30 paths in my fitting, do I need to guess about 150
>>>> (30*5=150) parameters (except guess the same 'enot' for all paths)?
>>> 
>>> No. One of the great principles of Ifeffit (and thus Artemis) is  
>>> that the path parameters do not have to be the same as the fit  
>>> parameters.
>>> 
>>> As a simple example, it's good to start with the model that all  
>>> paths have the same value for E0. So you can guess a single  
>>> parameter and then use that parameter for the E0 for all paths.
>>> 
>>> As another simple example, a cubic crystal might be modelled as  
>>> having a uniform thermal (or Vegard's law) expansion. Then there  
>>> could be a single guessed parameter indicating the fractional  
>>> expansion (called, for example, alpha) while the delr for each path  
>>> could then be entered as alpha*reff.
>>> 
>>> Much more complicated constraint schemes are possible, but for fits  
>>> with large number of paths, the number of guessed parameters is  
>>> almost always much much lower than the number of path parameters.
>>> 
>>>> 
>>>> As shown on the fit Log file, the Correlation value between two fit
>>>> parameters is the bigger the better, or the smaller the better?
>>> 
>>> Not necessarily. Correlations are provided to help understand the  
>>> relationship between guessed parameters, but there's nothing  
>>> inherently wrong with a high correlation...the uncertainty  
>>> associated with that relationship is already represented in the  
>>> uncertainties reported with the fit. In other words, it is more  
>>> direct to focus on getting the uncertainties to be lower, rather  
>>> than the correlations.
>>> 
>>> --Scott Calvin
>>> Sarah Lawrence College
>>> 
>>> 
>>> ___
>>> Ifeffit mailing list
>>> Ifeffit@millenia.cars.aps.anl.gov
>>> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>> 
>> 
>> 
>> ___
>> Ifeffit mailing list
>> Ifeffit@millenia.cars.aps.anl.gov
>> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> 
> 
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Questions for Artemis fit

[Scott Calvin]

Hi Yanyun,

On Aug 19, 2014, at 12:05 PM, huyan...@physics.utoronto.ca wrote:
> 
> But my question is, for each path there are a set of Path Parameters,  
> does that mean we have to fit this set of path parameters independent  
> from those used in other paths?  For instance, if I am going to  
> include 30 paths in my fitting, do I need to guess about 150  
> (30*5=150) parameters (except guess the same 'enot' for all paths)?

No. One of the great principles of Ifeffit (and thus Artemis) is that the path 
parameters do not have to be the same as the fit parameters.

As a simple example, it's good to start with the model that all paths have the 
same value for E0. So you can guess a single parameter and then use that 
parameter for the E0 for all paths.

As another simple example, a cubic crystal might be modelled as having a 
uniform thermal (or Vegard's law) expansion. Then there could be a single 
guessed parameter indicating the fractional expansion (called, for example, 
alpha) while the delr for each path could then be entered as alpha*reff.

Much more complicated constraint schemes are possible, but for fits with large 
number of paths, the number of guessed parameters is almost always much much 
lower than the number of path parameters.

> 
> As shown on the fit Log file, the Correlation value between two fit  
> parameters is the bigger the better, or the smaller the better?

Not necessarily. Correlations are provided to help understand the relationship 
between guessed parameters, but there's nothing inherently wrong with a high 
correlation...the uncertainty associated with that relationship is already 
represented in the uncertainties reported with the fit. In other words, it is 
more direct to focus on getting the uncertainties to be lower, rather than the 
correlations.

--Scott Calvin
Sarah Lawrence College


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Can the degeneracy (N) in Artemis be a variable?

[Scott Calvin]

I can't come up with a properly humble non-self-promoting way to say this, but 
for common fitting issues like this I'd cite my textbook. In my experience, I 
found it difficult to find good discussions of this kind of thing in the 
peer-reviewed literature, because they're either considered common knowledge or 
technical detail.

S. Calvin, XAFS for Everyone (CRC Press, 2013). 

The degeneracy S02 correlation issue is most directly discussed in Section 
10.1.2.

--Scott Calvin
Sarah Lawrence College

On Jul 28, 2014, at 9:51 AM, Bruce Ravel  wrote:

> 
> 
> Peng's question is so common on our mailing list that I would like to
> write a section of the Artemis document addressing it.
> 
> At heart, there are two issues.  The more important is that N is very
> highly correlated with S02 (and other things!).  The operational issue
> is the choice I made about only allowing numbers in the N box in
> Artemis.
> 
> I would like some suggestions for literature references on the more
> important part of this question.  Specifically, I'd like to know what
> references all of you cite on the topics of this high correlation and
> of solving the problem for real measurements.
> 
> With this bounty, I'll write up a section of the document which we can
> use when this question comes up in the future.
> 
> Thanks,
> B
> 
> 
> On 07/26/2014 12:56 PM, Peng Liu wrote:
>> Hi all,
>> 
>> When I model a standard, I know the degeneracy through the other data.
>> But for spectra of unknown material, I would like to set degeneracy as a
>> variable like the amplitude, deltE0, e.g.. However, I can assign a
>> letter to N and define it. Do you have any idea how I should do it?
>> 
>> Thanks for your reply,
>> 
>> Peng
>> 


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] from Rui Si at SSRF: help needed

[Scott Calvin]

Hi Rui,

Since XRD indicates your material is phase pure, I can think of three likely 
contributors as to why the 3 angstrom peak is "missing."

First is that you never actually plotted the model. You plotted the magnitude 
of individual paths from the model, but you did not sum the paths and plot the 
magnitude of the result. Your paths 2, 3, and 4 all show considerable amplitude 
around 3 angstroms, but the magnitude plot does not show phase. It is quite 
possible for those paths to add out of phase, and thus partially cancel. You 
should try summing the paths and plotting the result.

Second, Co3O4 is a spinel, and has two different absorbing sites for cobalt 
(the tetraehdrally and octahedrally coordinated ones). It very much looks to me 
like you only did a feff calculation for one of those sites. You need to 
calculate for both sites. Since there are two octahedral sites for every 
tetrahedral one, you need to weight the octahedral sites twice as heavily when 
you do the sum (you can use the S02 path parameter to affect weighting).

Finally, you didn't include any multiple-scattering paths. Multiple-scattering 
paths are significant contributors to spinel spectra in that region of the 
Fourier transform.

--Scott Calvin
Sarah Lawrence College


On Jul 15, 2014, at 5:25 AM, Bruce Ravel  wrote:

> 
> 
>(1) Why the second peak (ca. 3 Å) in experimental data is missing in
>the Co3O4 model?
> 


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Multiple phases in samples

[Scott Calvin]

Hi Dien Li,

Create a FEFF calculation and for each phase.

Weight the SO2 of the paths for each calculation by the proportion of that 
phase that is present; that is often done with a guessed parameter.

For example,

Set SO2_U = 0.85
Guess frac_goethite = 0.3
Guess frac_organic = 0.3
Def frac_uraninite = 1 - frac_goethtite - frac_organic

Then, for the paths associated with U on goethite, enter in the SO2 field the 
expression SO2_U * frac_goethite, and similarly for the other phases.

While that's how you do it in theory, these kinds of fits often end up 
unstable; Ifeffit may find fits that work well statistically but are utter 
nonsense; e.g., they might have negative fractions. The more you can constrain 
or restrain the fit, the less likely you are to have that happen.

In fact, if you have more than one sample in this class, you might want to try 
PCA first, including target transforms, to start getting a sense of what's 
present, and then switch to Artemis once the possibilities have been narrowed 
down.

--Scott Calvin
Sarah Lawrence College

On Jul 7, 2014, at 1:20 PM, dien...@srs.gov<mailto:dien...@srs.gov> wrote:

Hi, All

"Assuming" there are multiple phases in one sample, saying, U on goethite, U on 
organic matter, and uraninite, how to fit the EXAFS data using Artemis? Thanks.



Dien Li, Ph.D.
Environmental Sciences and Biotechnology
Savannah River National Laboratory
Aiken, South Carolina, 29808
Tel: 803-725-7520
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Demeter-devel MacPorts Athena issues & work-around

[Scott Calvin]

Hi all,

This is probably as good a time as any to mention a miscommunication I've seen 
come up a couple of times on the list that may have caused unnecessary 
friction. It's a good time because it =isn't= what transpired in the current 
thread (which ended happily for all involved), so it hopefully won't get bogged 
down in particulars.

Sometimes I've seen someone post to the list that they've been having an 
intermittent, difficult to diagnose, or otherwise mysterious problem, asking if 
anyone else had seen something similar. That makes for a lousy bug report, but 
is, in my opinion, a reasonable and rational use of a list like this. 

As a hypothetical, something like "is the latest distributed version of 
Hephaestus working for other people using MacPorts?" is a useful diagnostic 
question, and a terrible bug report. It does not require a response of any kind 
from Bruce, Matt, etc., and they should not feel compelled to give one. "Works 
for me" or "No, Hephaestus crashes" would be useful responses from other 
members of the community. 

To put it another way, I think this list straddles the boundary between being 
part of the reporting system for software and a community board. It's important 
to recognize, I think, when each function is being utilized.

That's not to say that there aren't lousy bug reports that get posted 
here--there are. When someone says "Athena crashes every time I open it--please 
fix this bug," they are not being useful to anyone. But I am saying that not 
every report of problems is actually a bug report.

That said, I think this is a good list. Everyone who contributes to it is a 
volunteer, and there are interesting and helpful discussions that take place at 
a wide range of levels.

--Scott Calvin
Sarah Lawrence College

P.S. If we're looking for ways to improve the list in the future, I'd suggest 
considering the possibility of a more modern forum format. I am sure there are 
open-source platforms for such a thing. It would be wonderful to have an upvote 
type feature (Facebook's "like" is the most familiar example), more 
user-friendly threading, and the ability to delete and edit posts. 



___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] How to distinguish whether the coordination element is heavy or light

[Scott Calvin]

Hi Zhanfei,

A few more words to add to Bruce's answer on the +/- 5 "rule."

It refers to cases where the bond length is treated as a free parameter. 
There's a gradual change of phase shift due to scattering with Z which looks a 
bit like a change in bond length. So if you're fitting bond length, the fit can 
compensate for an incorrect choice of scatterer by also returning an incorrect 
bond length. If you don't know what the bond length should be and your 
comparing two fits with scattering atoms of similar Z, then you don't know 
which reported bond length is incorrect and thus which fit is more valid.

But some people (Joe Woicik comes to mind) have pointed out that sometimes you 
do know what bond length would correspond to which Z. As an example, you might 
be trying to determine which of two possible known structures is found in a 
nanoparticulate sample. Each candidate structure might have bond lengths that 
are well known. In that case, bond length would not be a free parameter, but 
would be fixed to the appropriate value for each fit. In a case like that, a 
change in Z of 1 is often distinguishable; i.e., there is no +/- 5 rule.

--Scott Calvin
Sarah Lawrence College

On Jul 4, 2014, at 9:06 AM, Bruce Ravel  wrote:

> On 07/04/2014 03:51 AM, ZHAN Fei wrote:
>> In the picture you recommended,it says "The variations in functional
>> form allow Z to be determined (±5or so) from analysis of the
>> EXAFS".But I don't find any publication use it.
> 
> Hi Zhanfei,
> 
> The figure I referred you to was a plot of the scattering amplitude (the 
> bottom panel had the phase shift) for three different scattering 
> elements.  I don't know what's plotted in the figure you attached 
> because there are no labels on the axes.  In any case, the scattering 
> amplitude is the F(k) term in the EXAFS equation.  As such, it's 
> relationship to chi(R) or chi(q) is subtle.  And, of course, in real 
> data, all the different scattering paths interfere with one another.
> 
> As for the Z+/-5 rule, I don't know who first stated that.  Perhaps Teo 
> and Lee...?
> 
> In any case, it is easy to test.  Measure, say, a NiO standard.  Try 
> replacing O in the feff.inp by N or F and do the analysis.  Try again 
> with C or Ne.  Try again with B or Na.  You will find that the fit using 
> F is basically indistinguishable of the fit with O.  Ne will be a bit 
> worse, but not much.  Na a bit worse, but not much.  Eventually, you 
> will get far enough away from O that you can clearly see the difference 
> in the fit.
> 
> Some years ago, I tried to work on a FeGa alloy.  Although 5 apart, I 
> could not distinguish Fe scatterers from Ga scatterers well enough to 
> say anything about how the dopant was distributed in the lattice.


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] to use scattering path in different model or to construct a new model which is the combination of different simple compound

[Scott Calvin]

Hi Zhanfei,

On Jul 1, 2014, at 10:43 AM, ZHAN Fei 
mailto:zhan...@ihep.ac.cn>> wrote:

2 I have confusion of the relationship pf N between single path and multiple 
path,for example I think U-Oe-C U-Oe-Pshould be 1 2 respectively,instead 2 4 in 
paper.

I didn't look at the paper, but multiple scattering paths usually have even 
degeneracies because it counts separately when you go around the path in the 
opposite direction. Thus, if U is the absorber, U-Oe-C and U-C-Oe are separate 
but degenerate possibilities. So even if there is only one of each atom, the 
degeneracy would be 2.

--Scott Calvin
Sarah Lawrence College
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Use of the 3rd and 4th Parameter

[Scott Calvin]

Hi Michael,

It is best to start by guessing the third cumulant only, leaving the fourth 
cumulant set to zero. A survey of the published literature shows very few 
examples where the fourth cumulant has been fit--by that point the cumulant 
expansion itself is starting to become iffy, which is not the case when you 
just have "some" disorder.

Guess the third cumulant as zero, and then observe the best-fit value and its 
uncertainty. A typical result might be something like 0.0005 +/- 0.0002 
angstroms cubed. That result would suggest that the third cumulant was nonzero, 
as the bestfit value is outside the reported uncertainty. Comparing the reduced 
chi-square parameter with the third cumulant set to zero and guessed will give 
you a sense of whether it resulted in a statistical improvement to the fit; 
using the Hamilton test would be even better.

The third cumulant can be positive, negative, or zero.

To get a sense of how large the magnitude can be, think about the units: 
angstroms cubed. Very roughly, we usually expect delr to be less than 0.1 
angstroms, sigma2 to be less than 0.01 angstroms, and the third cumulant to be 
less than 0.001 angstroms. If we're getting numbers bigger than that, it 
suggests that a different starting model might be called for, rather than 
trying to use a cumulant expansion. You certainly should never guess the third 
cumulant at 0.5--that's way too big!

--Scott Calvin
Sarah Lawrence College


On Jun 16, 2014, at 11:15 AM, Michael Kömm  wrote:

> Hello,
> 
> I'm trying to make an EXAFS analysis on the Structure of Tb in a 
> borogermanate glass System. The reference is an .cif of 
> Terbium-Gallium-Garnet (TGG).
> 
> I expect some disorder caused to working in glass. Therefore I would 
> like to use the 3rd and/or the 4th parameter. I tried to set a guess 
> parameter to evaluate those parameters. Could you tell me a reasonable 
> value for these parameters? The 3rd only varies a little bit around its 
> guessed value (+- 0.01) and if I try to do the same with the 4th 
> parameter no fit at all takes place. The errors in the GDS-window are 
> exactly 0.0 with no fit taking place (enot = 0, amp = 0 (guess amp = 
> 1) delr = 0 and ss = 0.003) and the R-factor of the fit is way over the 
> top (>10^20).
> 
> I've chosen a value of 0.5 for the 3rd and the 4th parameter just to see 
> if they're working and I was only doing a quick first shell fit with the 
> surrounding oxygen-shell.
> 
> Please can somebody tell me how to deal with this problems.
> 
> Thanks a lot for your time and efford.
> 
> Sincerly yours,
> 
> Michael Kömm
> 
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Fitting EXAFS simulation to experimental

[Scott Calvin]

Dear Carl,

The short answer is yes, but it takes a little bit of effort.

One method is to actually figure out how stretching the c-axis impacts the 
lengths of each path geometrically, and then use the functions determine to 
express the delr's of each path in terms of the change in c. This method is 
rigorously correct, but requires digging in to the geometry, and entering 
somewhat lengthy expressions for many of the scattering paths.

A related method is to in effect use a first-order approximation of those 
changes. To do that, create two atoms files with slightly different values of 
c. Generate the corresponding paths, and see how the reff of those paths 
changes. For example, changing c by 0.01 might result in a certain path having 
a reff change by 0.014. Use the ratio of the change in reff to the change in c 
to scale the delr of those paths with respect to a guessed parameter delc. In 
the example I just provided the delr of the path in question would be chosen to 
be 1.4*delc.

The second method is, of course, approximate, but for small changes in a 
parameter such as c can work pretty well, and can be easier to implement than 
the first.

--Scott Calvin
Sarah Lawrence College


On May 29, 2014, at 12:33 PM, Carl Brozek  wrote:

> Dear All,
> 
> Thank you for your responses so far and I apologize for not being clearer.
> 
> Here is the problem at hand: I have a material that is best thought of as 
> organic, graphene-like 2-D sheets connected by infinite chains of Fe-S. Based 
> on PXRD and other techniques, we know the structure of the AB plane (the 2-D 
> organic part), but we aren't able to make sense of what is going on with the 
> Fe-S chain. We can simulate the PXRD based on proposed .cif files, but it's 
> off most likely because of the c direction (Fe-S chain).
> 
> We collected Fe K edge XAS to tease out Fe-S bond angles and distances, 
> hoping we could feed this into our simulation for the PXRD to improve the 
> fit. 
> 
> Modifying the ATOMS input by hand, by constructing different reasonable 
> geometries, was improving the simulation, but using a error minimization 
> fitting program would be best.
> 
> So, is it possible to vary specific angles and distances between atoms along 
> the c-direction only in an error minimization program based on our 
> experimental data?
> 
> I apologize if you have already responded to this reformulated question.
> 
> best,
> Carl 


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Fitting EXAFS simulation to experimental

[Scott Calvin]

Hi Carl,

I'm not sure what level you were asking your question from. Bruce provided an 
answer to one interpretation of the question.

If the question is, however, "has anyone written an automated script that 
allows the user to choose parameters given to ATOMS as fitting parameters for 
all space groups," then my answer is that I'm not aware of that having been 
done.

For example, it is very easy to make the lattice parameter for a cubic space 
group be a fitting parameter, as changes to it will result in a uniform 
fractional change to the length of all paths, and the reff keyword makes it 
easy to implement a uniform fractional change. Plenty of people have computed 
the relationship between ATOMS parameters and path parameters for specific 
geometries (including more complicated ones), and used that for fitting.

But in each case that requires some effort to think about the geometry 
involved. That's different from having a script that allows you to designate a 
shift in, say, the x-coordinate in fractional coordinates of one kind of atom 
in the input file and automatically have that result in the generation of the 
correct function of that fitting parameter in each of the paths used in FEFF, 
including multiple-scattering paths. And that's the thing that I don't think 
currently exists.

--Scott Calvin
Sarah Lawrence College
 
On May 29, 2014, at 9:04 AM, Bruce Ravel  wrote:

> On 05/28/2014 06:58 PM, Carl Brozek wrote:
>> Dear listhost,
>> 
>> I'm using Artemis as a front end for ATOMS and FEFF and I would like to
>> build a simulated EXAFS spectrum based on experimental data I have. I
>> know how to generate a simulation based on a .cif or .xyz file, but is
>> it possible to vary parameters in these input geometries to best match
>> the experimental?
>> 
>> Has someone implemented this in artemis or matlab?
>> 
>> thanks,
> 
> 
> Carl,
> 
> In Artemis, this is called a VPath.
> 
>   http://bruceravel.github.io/demeter/artug/plot/vpaths.html
> 
> You will need to import data of some sort because a VPath is created
> from paths that have been associated with a data set.
> 
> Run Feff.  Drag and drop any number of paths from the Path tab in the
> Feff window onto your data Window.  Mark the paths paths you want to
> include in your VPath by clicking their little check buttons.  From
> the Actions menu, select "Make VPath from marked".  This will prompt
> you for a name for the VPath and insert the VPath into the plotting
> list, as described in the page whose URL is given above.
> 
> As for varying structural parameters, that is what we call a "fit".
> You can parameterize the various Delta R parameters to represent
> changes in lattice positions.  There are plenty of examples of this
> using Ifeffit -- Daniel Haskel's papers on cuprate superconductors, my
> own papers on titanate perovskites, lots of work buy Anatoly Frenkel
> on various catalyst materials, Sam Webb's work on manganites, just to
> name a few.
> 
> B
> 
> 
> 
> -- 
>  Bruce Ravel   bra...@bnl.gov
> 
>  National Institute of Standards and Technology
>  Synchrotron Science Group at NSLS --- Beamlines U7A, X24A, X23A2
>  Building 535A
>  Upton NY, 11973
> 
>  Homepage:http://xafs.org/BruceRavel
>  Software:https://github.com/bruceravel
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] EXAFS simulations - peak too high

[Scott Calvin]

Hi Lisa,

On Feb 3, 2014, at 4:24 PM, Lisa Bovenkamp  wrote:
> 
> Dear Scott,
> That was the answer I was looking for. 
> I believe the paths are reinforcing each other.
> This is now obvious from the Im chi(R).
> Thanks for the presentation of yours.
> 
> Now one question remains: 
> Do I discard/disregard this site from my calculations (in the averaging 
> process) or not?

No; the calculation is telling you must include it, because it is particularly 
important!

Presumably your "averaging process" involves a weighted average of chi(k), or 
maybe chi(R) (it doesn't matter which). It is invalid to average the magnitudes 
of chi(R). 

--Scott Calvin
Sarah Lawrence College



___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Absorption-coefficient data repository?

[Scott Calvin]

Hi Damon,

Hephaestus will calculate that for you. (There are other options, too, but that 
may be the simplest.)

--Scott Calvin
Sarah Lawrence College

On Jan 27, 2014, at 1:30 PM, Damon Turney  wrote:

> Dear XAFS Community,
> 
> I'm looking for values of absorption coefficient (or mass absorption
> coefficient) of ZnO in the XANES region of the Zn edge, but all the
> publications give XANES plots in "arbitrary units", so I have no way
> of knowing the absorption coefficient for ZnO.
> 
> Does anybody know a data repository where absorption coefficients are
> collected?
> 
> Thanks,
> Damon
> 
> 
> 
> -- 
> Damon Turney PhD
> Energy Institute, Department of Chemical Engineering
> City University of New York
> New York, NY 10031
> dtur...@che.ccny.cuny.edu
> office: (212) 650-5470
> cell: (805) 571-3764
> fax: (805) 823-4593
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] correlating number of nearest neighborst to the particle size

[Scott Calvin]

Hi Jatin,

There are two strategies that have been used to do this that you can find in 
the literature, both of which are by members who frequent this list (myself and 
Anatoly Frenkel).

I have published articles outlining a strategy where you assume the shape of 
the particles. Most often in my case that has been spheres, but it could be 
plates or hemispheres or whatever. Then you derive how the coordination number 
depends on dimensions for that shape (as I said, I an others have done that for 
spheres), and make those dimensions free parameters in an Ifeffit (e.g. 
Artemis) fit.

Anatoly has published articles outlining a strategy where you treat the 
coordination numbers themselves as free parameters, and then check for 
consistency with various shapes.

I think Anatoly's method is better when the nanoparticles are very consistent; 
i.e. they have a narrow size distribution and consistent shapes, and when data 
quality is good. I think my method may work better when the nanoparticles are 
"messier" or when the data is more limited, if only because it requires fewer 
free parameters.

A typical reference for Anatoly's method is

A View from the Inside:  Complexity in the Atomic Scale Ordering of Supported 
Metal Nanoparticles
Anatoly I. Frenkel,*,†,‡, Charles W. Hills,§ and, and Ralph G. Nuzzo*,†,§
The Journal of Physical Chemistry B 2001 105 (51), 12689-12703


A typical reference for my method is

“Estimating crystallite size in polydispersed samples using EXAFS,” S. Calvin, 
C. J. Riedel,* E. E. Carpenter, S. A. Morrison, R. M. Stroud, and V. G. Harris, 
Physica Scripta T115, 744 (2005).

although that can be a little hard to find.

Another reference which is more widely available is

“Comparison of extended X-ray absorption fine structure and Scherrer analysis 
of x-ray diffraction as methods for determining mean sizes of polydisperse 
nanoparticles,” S. Calvin, S. X. Luo, C. C. Broadbridge, J. K. McGuinness, E. 
Anderson, A. Lehman, K. H. Wee, S. A. Morrison, and L. K. Kurihara, Appl. Phys. 
Lett. 87, 233102 (2005).

--Scott Calvin
Sarah Lawrence College

On Jan 9, 2014, at 12:55 PM, Rana, Jatinkumar Kantilal 
 wrote:

> Hi All,
>
> I would like to estimate the size of Cu nanoparticles grown in my sample by 
> EXAFS fitting. I can refine the number of nearest neighbors in these Cu 
> nanoparticles up to 4 shells (say NN1, NN2, NN3 and NN4), by constraining S02 
> to the value obtained by fitting the data of Cu reference foil. But, how do I 
> correlate the refined values of nearest neighbors to the actual size of Cu 
> nanoparticles ? Any help in this regard would be much appreciated.
>
> Regards,
> Jatin
>
> -Original Message-
> From: ifeffit-boun...@millenia.cars.aps.anl.gov 
> [mailto:ifeffit-boun...@millenia.cars.aps.anl.gov] On Behalf Of 
> ifeffit-requ...@millenia.cars.aps.anl.gov
> Sent: 09 January, 2014 17:51
> To: ifeffit@millenia.cars.aps.anl.gov
> Subject: Ifeffit Digest, Vol 131, Issue 8
>
> Send Ifeffit mailing list submissions to
>ifeffit@millenia.cars.aps.anl.gov
>
> To subscribe or unsubscribe via the World Wide Web, visit
>http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> or, via email, send a message with subject or body 'help' to
>ifeffit-requ...@millenia.cars.aps.anl.gov
>
> You can reach the person managing the list at
>ifeffit-ow...@millenia.cars.aps.anl.gov
>
> When replying, please edit your Subject line so it is more specific than "Re: 
> Contents of Ifeffit digest..."
>
>
> Today's Topics:
>
>   1. Query re: Athena and Artemis user guides (Leon Clarke)
>   2. Re: Query re: Athena and Artemis user guides (Bruce Ravel)
>   3. Re: Query re: Athena and Artemis user guides (Kathy Dardenne)
>   4. Re: Demeter 0.9.19 (Schima, Frank)
>
>
> --
>
> Message: 1
> Date: Thu, 9 Jan 2014 15:56:29 +
> From: Leon Clarke 
> To: "ifeffit@millenia.cars.aps.anl.gov"
>
> Subject: [Ifeffit] Query re: Athena and Artemis user guides
> Message-ID: <83CF3E8AA2102C4598B316168B472EFE98BEEECD@drmb1>
> Content-Type: text/plain; charset="us-ascii"
>
> Dear Dr Ravel and Demeter users,
>
>
>
> I recollect that the Athena and Artemis user guides used to be available as 
> PDF files (which I found useful). Am I correct that the latest guides are now 
> only accessible as internet pages, as I can't seem to find a download link?
>
>
>
> Best regards, Leon
>
>
> -
> Dr Leon J. Clarke
> Lecturer in Environmental Analytical Chemistry Division of Chemistry and 
> Environmental Sciences School of Science and the Environment Faculty of 
> Science and Engineering Manchester Metropolitan Univers

Re: [Ifeffit] Itinerant vs localized XAFS

[Scott Calvin]

Dear Maurits,

My dissertation looked at the possibility that the kind of change you describe 
is accompanied by a discontinuity in the third cumulant parameter for (in this 
case) the nearest-neighbor bond. My evidence was not sufficiently convincing to 
allow for publication elsewhere, but it was suggestive.

Of course, the third cumulant tends to increase with temperature anyway. One 
way around this is to measure just above and below the putative transition; a 
jump in third cumulant (rather than a modest change) would be in accord with 
what I described in my dissertation. Another way around this is to use one of 
the theoretical models for how third cumulant should change with temperature in 
such a system, and look to see if the change is significantly bigger than that.

Changes in the third cumulant correlate in fitting with changes in lattice 
parameter, so your strategy of using diffraction at the measured temperatures 
to fix the inter-atomic distances is a good idea. Also, interatomic distances 
and the third cumulant are weighted differently by k in the EXAFS equation, so 
if you can collect and analyze data over a wide k-range, you've got a better 
chance of teasing apart the two parameters.

--Scott Calvin
Sarah Lawrence College

On Dec 9, 2013, at 10:02 AM, Maurits Boeije 
mailto:m.f.j.boe...@tudelft.nl>> wrote:

Dear XAFS specialists,

We’ve performed an experiment at the ESRF in Grenoble to find experimental 
evidence of an electronic change in our system. We suspect, from ab initio 
calculations, that there is a change in how electrons are allocated in our 
Mn/Fe system. Calculations show that some electrons change from being itinerant 
at high temperatures, to being localized at low temperatures.
We performed an XAFS experiment and hope to see a difference between the low 
and high temperatures measurements. There is always difference because the cell 
parameters change as function of temperature, and I used your Artemis program 
to investigate the influence of the cell parameters on the EXAFS signal. I can 
get a decent fit using the crystallographic data we have, but I’m not sure if 
I’m not throwing away any evidence of our electronic change. I can interpret a 
fitted interatomic distance of 2.1 A (compared to 2.0 A we got from 
diffraction) as being the mean interatomic distance in our sample, but I could 
also interpret it as a 0.1 A difference because of our suspected electronic 
change.
The challenge now lies in seeing an electronic difference despite the 
crystallographic change. What we would ideally do, is to keep the (known) 
atomic positions and cell parameters fixed, fit the EXAFS parameters and see if 
we can predict either the high or low temperature spectrum. Do you think this 
is possible? Are there any parameters which are calculated by IFEFFIT who are 
not precise enough for such a conclusion? It could be that the effect we are 
looking for is so small that they are nullified by the assumptions in the EXAFS 
equation.

I would appreciate any insight you could have in this particular problem.

With kind regards,

Drs. MFJ Boeije
Fundamental Aspects of Materials and Energy

TU Delft
T +31 (0)15 27 83793

No trees were killed to send this message, but a large number of electrons were 
terribly inconvenienced

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] fitting a specific k range

[Scott Calvin]

Hi Matt,

Yes--we clearly disagree. 

The word "complaining" has come up a couple of times. I didn't think of it that 
way, although clearly it's coming off that way. I said it was my least favorite 
warning, and that usually one of the first things I tell students to do is to 
change the preferences to turn it off. That isn't a complaint--it's a 
description of what I do. I think descriptions of the actual behavior of users, 
whatever their level of expertise, is useful to developers and community. If 
that's counterproductive, wouldn't it also be counterproductive to say "I never 
use the button for corrected Fourier transform, because I think it encourages 
confusion" or "I don't see what the benefit is to fitting in k or q space, and 
so I don't do it?" 

> Having a warning about this is vastly preferable to not having such a warning.


To be clear, my view is that false positives can indeed be a significant 
problem--the medical community, for example, knows this very well. I've 
repeatedly witnessed people starting out with Artemis who fall in to a pattern 
of ignoring all warnings because of this particular false positive.

It seems to me that a user has to know a bit more to generate the true positive 
in this case than the false one. They have to add free parameters for paths 
outside the fitting range--Artemis doesn't do that for them. And then they get 
another strong sign that something has gone wrong because, as you say, the 
error bars blow up. So, unlike the consequences that can come from treating the 
false positive as a true one (removing a path that actually contributes 
significantly to the fitting range), in the case of a true positive it will at 
least be clear something is wrong. The message helps diagnose the problem, but 
the lack of the message wouldn't cause someone to think everything is 
hunky-dory when it's not (unless they don't even understand the concept and 
importance of error bars) .

So yes, it's closer to my view that the rule is always wrong and not worth 
fixing, although of course that's a bit oversimplified. For certain kinds of 
fitting strategies, it can be useful, and it's good that the software offers it 
as an option. But I don't think it should be the default option, as I think it 
confuses or misleads novices more often than it helps them. So I would suggest 
that the out of the box behavior for the warning to be off.

That's not an attack on the software in general, or on the warning feature it 
includes. It is my opinion, and it seems to me that this mailing list is an 
appropriate place to express that kind of opinion.

--Scott Calvin
Sarah Lawrence College

On Dec 6, 2013, at 12:29 AM, Matt Newville  wrote:

> Hi Scott,
> 
> On Thu, Dec 5, 2013 at 4:56 PM, Scott Calvin  
> wrote:
>> I love the warnings Artemis gives! They're not just for novices--they often 
>> catch when I've made a dumb mistake somewhere. I praise them, defend them, 
>> and generally think Bruce has done a wonderful thing by having them.
>> 
>> The out of range default warning, however, I find counterproductive and 
>> confusing to novices. There are two reasons for my opinion (and it is, of 
>> course, my opinion--as Bruce points out, if others differ, they can set the 
>> defaults differently.) One has to do with the kind of example I mentioned 
>> earlier. Here's the other reason:
>> 
>> The default behavior is to warn if there is a path beyond 1.1 times the top 
>> of the range, correct? (It's kind of a pain for me right now to fire up the 
>> most recent version of Artemis, so I can't actually easily confirm that at 
>> the moment.) The default R-max is, if I recall correctly, 3.0 angstroms. 
>> Thus, by default, a warning is generated if there is a path above 3.3 
>> angstroms in the list.
>> 
>> But, as we know, the path list uses distances which are half path lengths, 
>> while the Fourier transform range is in terms of the conjugate variable to 
>> k. For edges around the third row of the periodic table, the peaks 
>> corresponding to a path tend to show up about 0.3 to 0.5 angstroms lower in 
>> the Fourier transform than their half path-length. And, of course, that's 
>> just the peak--the path has significant amplitude a bit below (and above) 
>> that.
>> 
>> So, a novice user fires up Artemis, imports her data, and uses atoms to 
>> generate a feff file. Because she's appropriately thoughtful about what 
>> she's doing, she looks at what the unfitted paths of the FEFF calculation 
>> look like. She sees the fitting range goes up to 3 angstroms, and then 
>> selects all the paths that contribute significant amplitude to 

Re: [Ifeffit] fitting a specific k range

[Scott Calvin]

I love the warnings Artemis gives! They're not just for novices--they often 
catch when I've made a dumb mistake somewhere. I praise them, defend them, and 
generally think Bruce has done a wonderful thing by having them.

The out of range default warning, however, I find counterproductive and 
confusing to novices. There are two reasons for my opinion (and it is, of 
course, my opinion--as Bruce points out, if others differ, they can set the 
defaults differently.) One has to do with the kind of example I mentioned 
earlier. Here's the other reason:

The default behavior is to warn if there is a path beyond 1.1 times the top of 
the range, correct? (It's kind of a pain for me right now to fire up the most 
recent version of Artemis, so I can't actually easily confirm that at the 
moment.) The default R-max is, if I recall correctly, 3.0 angstroms. Thus, by 
default, a warning is generated if there is a path above 3.3 angstroms in the 
list.

But, as we know, the path list uses distances which are half path lengths, 
while the Fourier transform range is in terms of the conjugate variable to k. 
For edges around the third row of the periodic table, the peaks corresponding 
to a path tend to show up about 0.3 to 0.5 angstroms lower in the Fourier 
transform than their half path-length. And, of course, that's just the 
peak--the path has significant amplitude a bit below (and above) that. 

So, a novice user fires up Artemis, imports her data, and uses atoms to 
generate a feff file. Because she's appropriately thoughtful about what she's 
doing, she looks at what the unfitted paths of the FEFF calculation look like. 
She sees the fitting range goes up to 3 angstroms, and then selects all the 
paths that contribute significant amplitude to that range. That might include a 
path with a half path length of 3.4 angstroms. She then runs a fit--and Artemis 
gives her a warning that something may be wrong.

At that point, she could stick to her guns and tell the fit to continue. Since 
that's going to happen with pretty much every fit she runs, it becomes very 
tempting not to read the warning each time, but just dismiss it. And at that 
point, if there's a highly useful warning, she'll miss it.

Or, she could decide that she's the novice, and what she's doing isn't that 
unusual, so maybe she shouldn't be including that path at 3.4 angstroms, and 
take it out. She is now getting distorted results, because she's leaving out a 
path that has significant amplitude in the region.

I think, not just from personal preference, but also from a consideration of 
what is best for people learning to use the software, that the warning is set 
too conservatively. And I'm not even clear what misstep it's trying to prevent.

--Scott Calvin
Sarah Lawrence College

On Dec 5, 2013, at 2:32 PM,  
 wrote:

> 
> 
> What I find reassuring about Bruce's warnings is they allow you to think 
> properly about what you are doing. Fortunately, unlike Chris, I am not of a 
> size where I have to worry about falling down rabbit holes, unless American 
> rabbits are much bigger than English ones.
> I would commend Matt's original question as it opened a debate with 3 of the 
> mainstays of the list. That may not have been the intention but it makes 
> interesting reading for us mailing list voyeurs.
> 
> Cheers
> Fred
> Sent from Samsung tablet


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] fitting a specific k range

[Scott Calvin]

First, a disclaimer--I haven't looked at the data Matt sent (it's a busy time 
of year!), but I disagree in a general sense with my reading of what Bruce 
wrote (perhaps I am reading it wrong). In particular, I disagree with this 
statement:

> Given that you are fitting in q-space, it is completely unreasonable (from a 
> numerical perspective) to expect that the fit could possibly reproduce a 
> feature that you have (intentionally or otherwise) filtered out of the data.
> 
> To say that another way, given how you constructed the fit, you got a good 
> fit.  You made the fit in a way that it cannot possibly reproduce the feature 
> you are asking about, thus your question is ill-posed.

To explain why, suppose I am fitting a standard--for the sake of a simple 
example, suppose it's copper. I include in my model paths out to 7 angstroms, 
including multiple-scattering paths, and use a Debye model for the MSRDs. It's 
possible to do a pretty good job in that kind of fit with just a few free 
parameters: S02, E0, an isotropic lattice expansion, and a Debye temperature. 
Now, suppose I perform the fit from 1 to 3.5 angstroms. Usually, the fit will 
do a pretty good job reproducing features well above 3.5 angstroms, because 
they're in the model (the paths are included) even though they're not in the 
fitting range. That's true for features in k-space that are caused by high-R 
paths too, of course.

In fact, that kind of fit is particularly valuable to me, because it strongly 
suggests I've got the model right--I've essentially hid the high-R data from 
the numerical routines, so if it fits that region well anyway, it's probably 
because the model itself is a good one.

While I defend that principle as a very important one, I'm not claiming it 
applies in Matt's case--it probably doesn't. To reproduce features at high-R, 
it IS necessary to have a model (i.e. paths) that cover the high-R 
contribution, and it sounds like Matt does not have those paths included in his 
model. But it's not necessary to FIT up to high R.

--Scott Calvin
Sarah Lawrence College

P.S. My least favorite warning that Artemis provides is the one that tells you 
that you're including paths outside the fitting region, for just this reason. 
It's easy enough to change the preferences so that Artemis doesn't give that 
warning, and it's usually one of the first things I have my students do when 
they're first learning to fit.

On Dec 5, 2013, at 10:11 AM, Bruce Ravel  wrote:

> 
> Matt,
> 
> At the risk of coming off sounding a bit mean, I don't think you are
> asking a very well-posed question.
> 
> Examining the history of this project, I see that you are fitting in q
> space.  Like Matt, this is not my favorite choice, but there is
> nothing horribly wrong about it, so long as you understand what you
> are doing.
> 
> What is problematic is your expectation that, in doing so, you should
> be better able to fit a particular feature in k space.  If you examine
> the data in k and q space, you will see that the act of Fourier
> filtering the data (i.e. plotting in q space) has the effect of
> suppressing the wiggle at 5 inv. Ang. that you are asking about.
> Given that you are fitting in q-space, it is completely unreasonable
> (from a numerical perspective) to expect that the fit could possibly
> reproduce a feature that you have (intentionally or otherwise)
> filtered out of the data.
> 
> To say that another way, given how you constructed the fit, you got a
> good fit.  You made the fit in a way that it cannot possibly reproduce
> the feature you are asking about, thus your question is ill-posed.
> 
> I think the deeper problem is that you don't have a deep grasp of what
> happens in Fourier analysis.  So let's talk about that a bit.
> 
> When you do the transform from k to R-space, you are representing the
> frequency spectrum contained in the original data.  Slow wiggling
> features in the original data give rise to the low-R
> (i.e. low-frequency) features in the chi(R) data.  Fast wiggling
> features in chi(k) give rise to high-R features in chi(R).  Your
> wiggle at 5 inv. Ang. looks to my eye like a pretty high frequency
> feature.
> 
> When you do the backwards transform from k to R with a restricted R
> range (in your case, from 1 to 3.5), you are filtering frequencies out
> of the data.  The chi(q) data only contains those frequencies from the
> original chi(k) spectrum that fall in your R range.
> 
> What I am suggesting is that the wiggle in question is due to Fourier
> components beyond 3.5 Ang in chi(R).
> 
> So, how would you reproduce that feature in chi(k)?  That's simple --
> fit features in the data beyond 3.5 in chi(R).  That is, do a

Re: [Ifeffit] calculating particle size from coordination numbers

[Scott Calvin]

Hi George,

Do you have a rough sense of the shape of your nanoparticles?

I've consistently found (and I think Anatoly Frenkel has as well) that these 
kinds of methods tend to underestimate the size of nanoparticles, and in fact 
give a "size" for bulk metals which suggests they're actually large-ish 
nanoparticles.

I've never been able to nail down why that is. The hypothesis that I used to 
favor is that EXAFS is emphasizing the small end of the size distribution, and 
so there's actually a low-end tail that is being measured. More recently I've 
started to wonder if it's related to the assumption that S02 is independent of 
R and k within the EXAFS region. It's been known since the 80's that this is 
not exactly true, and it seems possible to me that there's a slight systematic 
trend with R that gets confounded with the particle size effect. As I say, 
that's just a guess, but in any case the underestimate of particle size comes 
up quite often--I've discussed it in a number of papers.

Therefore, if you do have a sense of the shape of your nanoparticles (e.g. 
roughly spherical, or roughly raft-like), I'd suggest constructing a model that 
describes the morphology with a small number of free parameters. For a 
sphere-like shape, for instance, the only free parameter needed is the radius 
(and S02). Then calculate the effect on the coordination number as a function 
of those free parameters. Next, fit the copper foil using the model, and see 
what it finds for the free parameters-. Finally, apply to your nanoparticulate 
samples, setting S02 to the value you found for the foil using the model, and 
compare the results for the other free parameters, such as radius.

I suggest this procedure for several reasons:

*By using a relatively small number of free parameters, it reduces the ability 
of the fit to compensate for a host of other systematic errors by tuning each 
individual coordination number up and down.

*By comparing to the fit to the foil, and using the S02 for that fit, some 
kinds of systematic errors are more likely to apply equally to each fit, thus 
making the procedure useful for at least judging relative sizes.

*The procedure is relatively insensitive to the details of the shape of the 
nanoparticles; you don't have to get much further than "raft-like" or 
"sphere-like" to have it produce useful results.

Hopefully that helps!

--Scott Calvin
Sarah Lawrence College

On Oct 29, 2013, at 1:23 PM, Georges Siddiqi 
mailto:gsidd...@gmail.com>> wrote:

Hi All,

I'm trying to use EXAFS data I have of Cu nanoparticles dispersed on SiO2 and 
ZnO to calculate their particle size.

Searching the archives, i've come by these topics and papers, which have helped 
immensely, but I'm still a little bit in the dark about the specifics of the 
fits:
http://www.mail-archive.com/ifeffit@millenia.cars.aps.anl.gov/msg02994.html
http://www.mail-archive.com/ifeffit@millenia.cars.aps.anl.gov/msg03430.html

http://pubs.acs.org/doi/full/10.1021/jp012769j
http://pubs.rsc.org/en/Content/ArticleLanding/1999/CP/a904654b#!divAbstract

So these discussions and papers make it clear that CNs for the 2nd or 3rd 
coordination shells will give a more accurate picture of the particle size (and 
even shape).  Also, just comparing SO2s between a foil and the sample isnt the 
most accurate.

My initial thought was to get a good fitting model on Cu foil, obtain 
coordination numbers which are very close to 1, then apply that same fitting 
model to the Cu nanoparticles (for the Cu-SiO2 species we have TEM data I can 
again calibrate myself with)

However, my specific problem is getting reasonable coordination numbers for the 
Cu foil.
- I can set the SO2 to 1 and get a very good fit for the data, but this is not 
useful for the nanoparticles
- I can give all paths the same SO2 variable, and get a fitted value close to 
1, but this isnt useful for comparing first and second shell CNs, as the ratio 
will always be the same.
- And finally, when I try to give each scattering shell its own independent CN, 
I again get a good fit, but while the first shell SO2 is close to 1, the 2nd 
shell value is closer to 1.5

I've been playing with which variables are the same between paths, etc, and I 
can get more reasonable numbers, but the problem then becomes that I cant 
easily justify my fitting model.

Attached is my Artemis file for Cu foil.

Any help or suggestions with how to proceed (specifically with how to group 
variables) would be much appreciated.  I apologize in advance if i'm asking 
about a topic that's been covered extensively.

thanks,
Georges Siddiqui
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Artemis Software Issue

[Scott Calvin]

On Oct 25, 2013, at 1:24 PM, Bruce Ravel  wrote:

> 
> Honestly, sometimes I wonder why I even bother with all this
> 
> B
> 

You bother, I would guess, because of the enormous amount of sound science that 
has been generated all around the world with the help of this software and its 
predecessors, in a way that strives for egalitarian accessibility with low 
barriers to entry.

--Scott Calvin
Sarah Lawrence College
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

[Ifeffit] DAthena crashes when reconstructing data from PCA

[Scott Calvin]

Hi Bruce,

Athena is crashing every time I try to "Reconstruct Data" in the PCA dialog. It 
doesn't seem to matter what data I use, or what space I'm working in (mu or 
chi), or how many components I use in the reconstruction.

I'm using Demeter 0.9.17 with Ifeffit 1.2.11d, under Windows 8 running in 
Parallels 8 running under OS 10.8.4 on a MacBook Pro.

The log file for the latest crash is below. If you'd like a project file too, 
let me know, but it happens, e.g., with the gold cyanobacteria project file, 
and with every data set I've tried.

--Scott Calvin
Sarah Lawrence College



dathena.log
Description: dathena.log
___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] L-edge soft X-ray data in Artemis and Athena

[Scott Calvin]

Hi all,

Can the EXAFS of the L1 edge of 3rd row transition elements be analyzed? I've 
never tried it, and the discussion so far has been for the L2 and L3 edges. I 
know there would be some interference from the EXAFS of the lower edges, but 
that's just a technical problem, and can be corrected for.

--Scott Calvin
Sarah Lawrence College


On Apr 11, 2013, at 10:03 AM, Dr. Dariusz A. Zając  wrote:

> Hi,
> 
> in Athena you can still normalize data, plot together, etc, but due to that 
> lines L2 and L3 are very closed you will observe no EXAFS oscillations.
> In opposite I  use Athena and Artemis to analyse EXAFS on e.g. O K edge
> L edges for EXAFS are useful for mainly 5d and actined elements...
> 
> regards
> kicaj
> 
> W dniu 13-04-11 13:46, Shoaib Muhammad pisze:
>> Hi,
>>  
>> I am a research student and relatively new to XAFS. I have soft X-Ray L-edge 
>> data of Mn and Ni. Is it possible to process this data in Athena and then 
>> fit in Artemis like usual K-edge data?
>>  
>> Thanks
>> -- 
>> Shoaib

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] plotting

[Scott Calvin]

Hi Eva,

Personally, I'm pretty happy with Pro Fit. It's a commercial product (like 
Origin) but is much cheaper. It's also got the stupidest name imaginable, since 
Google will assume you misspelled "profit." With that in mind the website is 
http://www.quansoft.com . If you're willing to spend more, Igor Pro is a good 
product.

--Scott Calvin
Sarah Lawrence College

P.S. Disclosure: I have no financial relationship with either software producer.

On Mar 26, 2013, at 4:50 PM, Eva Vanamee  wrote:

> Thanks everyone for the helpful comments! I'll go with exporting to Origin.
> A final question. I am now a Mac user, is there a good alternative to Origin?
> 
> Best,
> 
> - Eva
> 


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] generating high res figures

[Scott Calvin]

Hi Eva,

Demeter is not intended to be used to produce publication-quality figures. It's 
best to export the data, fits, etc., and then use a proper graphing program to 
generate figures. Even a spreadsheet will do in a pinch, although most of us 
use dedicated scientific graphing software.

--Scott Calvin
Sarah Lawrence College

On Mar 26, 2013, at 1:13 PM, Eva Vanamee  wrote:

> Hi,
> 
> I'm in the happy stages of making figures. 
> I'd like to know how to change figure legends, make high resolution figures 
> from a fit, etc.
> 
> For instance:
> I saved the session file. Can I go back to a particular fit to make a figure 
> or do I need to rerun it?
> Can I change the figure legends? I used sample 1 and I want to give it a 
> different name
> Can I use special characters (e.g. Greek letters)?
> Can I adjust the resolution for different formats (talk vs. publication)?
> 
> I am running Demeter 0.9.9.9 under XP. 
> 
> Thanks in advance for the help!
> 
> - Eva
> 
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] k-range question & R-factor

[Scott Calvin]

Hi Chris,

I don't see a reason to think that data is a glitch. For one thing, it's not 
consistent across datasets. The features also look smooth, and not so 
glitch-like. The spike around 8.2 inverse angstroms in some of the datasets 
looks a bit more like a glitch, but it's fairly modest and narrow enough not to 
mess you up too much.

The spacing of those features look OK--there's a double feature in some of the 
datasets around 6-7 inverse angstroms; it's plausible there would be another 
reature like that above it. In fact, I can make an argument that there's some 
kind of beating going on that gives a shoulder at 3.5-5, a double peak at 5-7, 
and two peaks at 7-8 inverse angstroms.

So I would recommend including that data and seeing what it does to your fits. 
If that range is garbage, your fits will probably reject it.


As for your second question, R-factors are always a kind of average across the 
data, by definition. So "total" mismatch doesn't really make sense. Off-hand, 
though, I don't recall how ifeffit weights the data for the purposes of 
calculating R-factors for multiple datasets, and that may be your question.

--Scott Calvin
Sarah Lawrence College

On Jan 15, 2013, at 9:21 AM, Christopher Patridge wrote:

> Hello Users,
> 
> I was looking for an opinion about the chi(k) signal in a set of data I 
> am analyzing.  Brief background, this is a set of in-situ XAS data 
> collected at the Fe K edge from a working electrochemical cell at a 
> range of potentials during charge; I did not collect the data. I suspect 
> the feature at ~ 8 angstroms-1, although present in all the spectra is 
> noise or glitch and wondered if I am being overly cautious?
> 
> My conservative range ( k = 2-7 and R = 1-2) really constrains the model 
> Nidp = 3.31.  Luckily, multiple datasets ( 8 ) to the rescue to give me 
> some flexibility.  In a multiple dataset fitting, is the R-factor of the 
> whole set just the average or total mismatch across all the datasets or 
> it calculated another way?
> 
> Working towards happiness,
> 
> Chris Patridge
> 
> -- 
> 
> Christopher J. Patridge, PhD
> NRC Post Doctoral Research Associate
> Naval Research Laboratory
> Washington, DC 20375
> Cell: 315-529-0501
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] S02 question

[Scott Calvin]

Hi Eugenio,

It has generally been found that S02 is transferrable, even from metals to 
oxides!

In practice, the bigger effect is generally sample and measurement issues. For 
example, if you are measuring powders in transmission and the grain size is 
larger than a small fraction of the absorption length, then the EXAFS will be 
suppressed and mimic a smaller S02. If your standard is then a metal foil, the 
effective S02 for your powder will be smaller than the foil, not because of an 
intrinsic difference in absorption, but because of the distortion due to grain 
size. 

--Scott Calvin
Sarah Lawrence College

On Jan 11, 2013, at 4:23 AM, Eugenio Otal wrote:

> Hi all,
> I have a question about S02 measurement in samples for CN calculation.
> By example, If I have an oxynitide, LaTiO2N, Can I use TiO2 as
> standard for S02 even the Ti in LaTiO2N is coordinated to O and N? In
> a more extreme case, is possible to use metals? Which are the
> recommendations and limits?
> Regards, euG


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Using this mailing list effectively

[Scott Calvin]

On Jan 6, 2013, at 5:54 PM, Zhaomo Tian wrote:

Ps. I am always curious why people think me as Mr.by just seeing my name, 
actually I am Ms., but just call me Zhaomo is ok.

I think I can answer that, Zhaomo.

In Romance languages (Spanish, French, Italian, Portuguese, etc.) given names 
that end in "o" are usually masculine, just as those that end in "a" are 
usually feminine. Since most English speakers are more familiar with Romance 
names than east Asian ones, "Zhaomo" subconsciously looks male to many of us.

--Scott Calvin
Sarah Lawrence College

___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] background subtraction process in Athena

[Scott Calvin]

Many of your questions are addressed in the documentation for Athena:

http://cars9.uchicago.edu/%7Eravel/software/doc/Athena/html/

--Scott Calvin
Sarah Lawrence College

On Jan 2, 2013, at 7:14 PM, Zhaomo Tian wrote:

Dear all,

I feel confused about the backgound subtraction process in Athena, I think the 
backgound subtraction is actually a complicated process containing some 
algorithms , but it seems the Athena program just let us do backgound 
subtraction by applying some easy work such as energy calibration and by 
changing some values of parameters such as pre-edge range or spline range...Is 
that true?

And I ususally just use the default values of parameters in the left panel in 
Athena,are those parameter with different values mattering a lot to the quality 
of backgound subtraction? I do not know how to change parameter values in order 
to get a better backgound subtraction.How can I compare between different 
backgound subtractions? ?

Thanks so much!



Tian Zhaomo
M.S. candidate





Lab. For Photosynthesis Materials and Devices
Department of Materials Science and Engineeing,POSTECH
san 31, Hyoja-Dong, Nam-Gu,
Pohang, 790-784, Republic of Korea



office: +82-54-279-2827
mobile: +82-10-7747-3790
e-mail: zhaomo1...@postech.ac.kr<mailto:zhaomo1...@postech.ac.kr>




 ___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] copper 1st shell coordination number

[Scott Calvin]

You're thinking about this the right way, Zhaomo Tian.

The measurement should come primarily from the bulk portion--even electron 
yield, which is most surface sensitive, will probe deep enough for that, and 
fluorescence or transmission certainly would.

Are you getting a reasonable value for sigma2? If you have a negative sigma2 
and a coordination number that is too small, that suggests some sort of data 
problem. If, on the other hand, you're getting a more typical value for sigma2 
(in the 0.001 to 0.01 angstrom^2) range, then that suggests you're seeing 
something structural in the sample. Could there be vacancies? Some sort of 
non-close-packed structure? A modest degree of oxidation or other impurities?

--Scott Calvin
Sarah Lawrence College

On Jan 1, 2013, at 4:57 PM, Zhaomo Tian wrote:

Dear all!

Really thanks for your kind and helpful reply to my previous questions about 
confusion between CN and degeneracy number.I understand it now.

Now another problem happened is that when I tried to fit the first shell of Cu, 
the CN would never be 12(initial guessed value), but turned out to be some 
value around 8. At first, I thought something might be wrong such as choosing 
bad value of S02, but after I set different values for S02 in a certain range, 
the CN won't change too much and remains around 7~8. Then I suddenly realized 
that the guessed CN of Cu first shell should be closed to 8 rather than 12, I 
do not know if it is correct so I am asking if it is possible. The reason is : 
actually, my copper data is not from copper foil but from a copper thin 
film(thickness around 300nm deposited on silica substrate), the EXAFS data I 
wanted is also the surface copper atom, For surface copper atom, there is only 
8 nearest neighbors(the upper 4 nearest neighbors do not exist). So I think if 
the EXAFS signal came mainly from the surface copper atom, then the ideal CN 
for 1st shell is 8, but since there is only one single! to!
 p atomic layer and the rest are all bulk, I think the major EXAFS signal 
should come from bulk, in this case the CN for 1st shell should be 12.

Maybe I am thinking in a wrong way, I am waiting to be corrected.

Thanks so much!
 ___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] The degeneracy and the coordination number

[Scott Calvin]

Dear Zhaomo Tian,

This is an idiosyncrasy of Artemis/Ifeffit.

In the EXAFS equation, coordination number only appears multiplied by S02. 
Therefore it makes no sense to fit both S02 and N for a single path, as they 
would be 100% correlated.

Artemis' solution to this is to only allow you to fit S02, requiring you to 
interpret whether you are doing that to fit the actual value of S02, or to fit 
the coordination number.

To fit coordination number for the first shell, therefore, you need to know the 
value of S02.

Hopefully, examples will make this clearer. Suppose you know S02 = 0.90 (from, 
for example, fitting a standard).

One way to fit the first-shell coordination number would be to Set a parameter 
amp = 0.90 and guess a parameter N1 = 12. Then enter amp*N1 in the S02 box for 
path 1 in Artemis, and 1 in the N box.

Another way would be to Set amp = 0.90 and N1 = 12, and then enter amp*N1/12 in 
the S02 box and 12 in the N box.

There are, of course, other strategies that are also mathematically equivalent.

The key is that you must never put a variable in the N box, instead allowing 
the S02 box to do all the work of fitting.

--Scott Calvin
Sarah Lawrence College

On Dec 26, 2012, at 3:23 AM, Zhaomo Tian wrote:

Dear all,

I am confused about the degeneracy and the coordination number used in the 
Artemis program.
I want to get the coordination number of 1st shell of my copper sample, so I 
set a guess number N1=12(initial guess) as the coordination number which will 
vary during the fitting iteration, but when I am looking at the 1st path's 
parameters, there is another N appear which may stand for the degeneracy.The 
1st path of copper model is single scattering, so the degeneracy N equals 
coordination number . But when I set N=12, the guessed N1 will have a value 
which is closed to 1, only when I set N=1, the guessed N1 will have a value 
which is closed to 12. It seems that N and N1 are reciprocals. But I think N 
actually equals N1 in the single scattering case.

So I am really confused, If I want to get coordination number of a shell in a 
fitting, how can I deal with floated coordination number and its path 
degeneracy N?

Thanks so much.
 ___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov<mailto:Ifeffit@millenia.cars.aps.anl.gov>
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Help: some operation using Athena and Artemis

[Scott Calvin]

A comment on this thread, started by Bingjie:

On Nov 26, 2012, at 11:01 AM, Bruce Ravel wrote:

2, In Artemis, since I was told that each shell in R-space can be fitted
separatly, is it OK to fit each shell in the R-range respectively, and then
join up the fitted curve and just ignore the not match part in other
R-ranges? And can a shell be comprised of several different paths (like
Fe-O and Fe-S)?

There is, currently, no way to exclude an interior region from a fit
in Artemis (i.e. there is no way to fit from 1 to 2 and from 2.5 to
3.5, but to exclude the region from 2 to 2.5).  I actually think
that's a bad idea and doubt that I would ever implement such a thing.


Sure there is. Just make a copy of the data set and treat it as a multiple 
dataset fit, with the two epsilons forced to the same value.

I've considered using that trick in the past, but never done so in earnest, in 
part because of the very important considerations Bruce describes.

So why did I consider it? To get around a really messy forest of glitches in 
data. While a single glitch can be removed, replacing a forest of glitches with 
zeroes is creating data that isn't there. Better to somehow exclude the data 
without fitting it, working only with the non-glitchy data above and below.

As I've said, I never got quite desperate enough to do that on a real fit, 
though--I always found some better way out of the problem.

But I agree strongly with Bruce that it's problematic to use this trick to 
avoid data, as opposed to glitches.

The reason I don't like this idea is that, in general, shells in the
EXAFS signal are not completely isolated.  Because of the limited data
range and other reasons, the signals form the various shells have long
tails and overlap considerably.  That overlap is a very important part
of the problem.  Getting the small parts of the chi(R) spectrum right
is just as important as getting the large parts right.

Excluding an interior region could artificially remove this overlap
from the evaluation of the fit.  That would have a very serious impact
on the statistical quality of the analysis.  I also worry that an
interior exclusion would be used in an attempt to minimize the impact
of multiple scattering paths on the analysis.  That, too, would be a
statistics disaster.

In short, you should embrace the correlations between the parts of
the problem.


--Scott Calvin
Sarah Lawrence College


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Linear combination fitting without normalization

[Scott Calvin]

Yes, Jan.

The workaround is to import your pre-normlaized data into Athena, and then fix 
the edge step to 1. I think you might also need to choose the "additional 
parameters" and uncheck "flatten normalized data" option.

With the edge step fixed and no flattening, normalization should leave your 
data unchanged, allowing you to perform LCF on your pre-processed data without 
a second round of normalization.

--Scott Calvin
Sarah Lawrence College

On Oct 26, 2012, at 5:15 AM, Stummelfinger wrote:

Hi there,
Is there a way to do Linear combination fitting in Athena with data which were 
already proceeded (background corrected and normalized)? I don’t want to do a 
double background correction /normalization on these data. However the Linear 
Combination Fitting dialog gives only the option “Norm(E)”.
Is there a work around?
Thanks a lot,
Cheers, Jan


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] determining S02 from ZnO vs. Zn foil

[Scott Calvin]

Hi Jeremy,

What you describe matches very well with my experience.

It is difficult to prepare powder transmission samples well enough so that 
there is no attenuation of the EXAFS due to particle size. Not impossible, but 
difficult. Much of the published work I've seen on first-row transition metal K 
edges settles for a little bit of attenuation; if they're fitting S02 and are 
primarily interested in phase variables such as bond length, there's little 
harm done. (Of course, you're looking for an amplitude variable, i.e., 
coordination number!)

This means powder standards often show a somewhat lower fitted value for S02 
than foil standards.

As to which you should use to constrain your samples, it depends on the form of 
the samples. You say they're "glass," and are implying that they are measured 
in transmission. If "glass" means literally a piece of macroscopic glass, and 
the ZnO is distributed homogeneously through it, then you should use the foil 
S02. If it's a glass microscopically, but are measuring it as a powder, the ZnO 
S02 is better. Either way, I'd take into account the additional uncertainty in 
S02 when reporting the uncertainties in your fitted coordination numbers.

--Scott Calvin
Sarah Lawrence College

On Oct 9, 2012, at 3:21 PM, Jeremy Thorbahn wrote:

> 
> [note: when putting together this post I discovered a possible bug regarding 
> opening my project files on different PCs. I will submit that issue 
> separately] 
> 
> 
> Hi all:
> 
> I am writing to probe the community's knowledge regarding my current problem 
> in measuring S02. 
> I have data on a series of glass samples containing ZnO. I also measured a 
> ZnO crystalline powder, and a Zn foil was measured between detectors 2 and 3 
> (transmission mode) simultaneously with every run. 
> 
> My goal is to determine coordination numbers around Zn in the glass samples, 
> which requires an accurate measurement of S02 for zinc. 
> 
> My main question is if there is any reason why the Zn foil would provide a 
> more accurate result than the ZnO powder, or vice versa. I see no reason why 
> it should be different, especially since this method depends entirely on the 
> 'chemical transferability' of S02 in the first place. Nevertheless, I obtain 
> different results from each, with S02 ~ 0.75 from the ZnO and ~0.88 from the 
> foil. 
> 
> Statistically, the Zn foil dataset is "better" since it is a merge of all of 
> my scans while there are only two scans of ZnO. 
> However, part of me wants to trust the ZnO data more, for somewhat vague 
> reasons such as that it is a "more similar" coordination environment to what 
> I would expect in my samples, and that it was measured between the same two 
> detectors, etc. 
> 
> If anyone has any experience that would indicate why one should be better 
> than the other, I would love to hear about it. 
> I will attach Artemis project files for each if you'd like to take a look at 
> my fits. It is entirely possible that the discrepancy is due to an error in 
> data processing or analysis on my part; I am by no means an expert. I hope 
> the project files will not clog up anyone's mailbox, at a couple of megabytes 
> each.
> 
> 
> Thanks,
> Jeremy
> 
> 
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Importing Multiple Data Sets into Athena

[Scott Calvin]

Hi Dan,

Could you attach the data files you are trying to import? We can see if the 
problem is reproducible on our systems.

--Scott Calvin
Sarah Lawrence College

On Sep 5, 2012, at 10:50 AM, Daniel Whittaker wrote:

Hello all,

I am having a small issue when importing multiple data sets into Athena all of 
which have a reference channel.

The issue i am experiencing is that when importing multiple data sets together 
they do not all import correctly. The first data set (of all selected) imports 
fine (The data is Np and the reference Zr foil) - it imports the reference file 
which can be aligned to the calibration Zr foil file. However, the remainder of 
the files do not work at all, the reference channel does not import correctly 
at all. I have tried (best as i can tell) all of the pre-process options to no 
avail. When i import them individually it works although I have to tell it what 
element the ref channel is and the edge and E0 (couldn't the edge and E0 be 
linked such that selection of one filled the other?) but this isn't too much of 
an issue.
It is by no means a massive problem but i thought i would ask if there is a way 
around it or if it is a real bug?
All other multiple data processing - such as "align marked groups" - works fine.
Regards,
Dan Whittaker



___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Ratio of peak amplitudes in R-space

[Scott Calvin]

Hi Jatin,

It's really best to model such a system using Ifeffit or similar software. 
There's a lot that goes into peak height. For one thing, I'm assuming you're 
looking at the peak height in the magnitude of the Fourier transform. But the 
Fourier transform of the individual scattering paths is complex-valued, not 
positive-definite. Thus, two paths contributing at some value of R can 
sometimes result in less amplitude than either path alone. Depending on data 
ranges and windowing, there's likely to be some contribution from the path that 
you're thinking of as corresponding to the first peak at the R of the second, 
and vice-versa, and that contribution is not additive when viewed in the 
magnitude.

In addition, for the peak at higher-R there may be a significant contribution 
from multiple scattering.

It should also be noted that a single peak may be due to unresolved splitting. 
For example, a nearest-neighbor octahedron might be distorted, with different 
axial and radial distances. In a case like that, a change in the amount of 
distortion can have a dramatic effect on the height of the peak.

Having said that, you can make qualitative comparisons in a series like the one 
you describe. If you have a suspicion of what might be happening (change in 
coordination, change in disorder, change in splitting, etc.), then you can see 
it progress by just inspecting the graphs in the way you suggest, and could 
probably identify, e.g., a phase change. But for quantitative determinations, 
you really need to model the system.

--Scott Calvin
Sarah Lawrence College

On Sep 5, 2012, at 8:01 AM, Rana, Jatinkumar Kantilal wrote:

Dear Ifeffit community,

I was wondering If I can make a qualitative comparison about the amplitude 
ratio of first two shells for various states of same sample ?

For example, I have a sample which is measured at various states, say A, B, C, 
D and E during any physical or chemical process.

I get systematic change in the ratio of first two peaks at each state. This 
could be interpreted either as change in disorder or as change in number of 
atoms associated with first two shells as a result of any physical or chemical 
change material has undergone.

As we know that amplitude of peak decreases with increase in distance from 
central absorber.

Can I compare the amplitude ratio of first two peaks for various states of same 
sample ??

Many thanks in advance !!!

With best regards,
Jatin



___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] How to fit "N" number of Neigbors in (D)Artemis?

[Scott Calvin]

Just to fill out some details about how this somewhat quirky aspect of 
Ifeffit/Artemis can be used:

Suppose you want to find the degeneracy of nearest-neighbors, represented by 
path 1, and next-nearest neighbors, represented by path 2. To fit the 
degeneracy of one of the paths,

1. Set "N" to 1 for both paths.

2. Create parameters named, say, "amp," "N1," and "N2." 

3. Fill in the N box for both the first and second path with "1" (not "N1" and 
"N2"!)

4. Fill in the S02 box for path 1 with "amp*N1" and for path 2 with "amp*N2."

5. EXAFS does not contain enough information to guess all three parameters; one 
of them must be set. Often, that is amp, perhaps because it is obtained from a 
standard. But maybe it's N1: you might know the near-neighbor coordination, but 
not the next-nearest neighbors. In any case, you need to set at least one of 
the three parameters to a known (or suspected) value.

6. All your named parameters now mean what you think they should mean in the 
log files: e.g., "N2" is the coordination number for the next-nearest 
neighbors. But the "S02" path parameter in the log files is amp*N1 for the 
first path and amp*N2 for the second path. That's kind of confusing, but 
because you've designed your guess/def/set parameters effectively, it's not a 
problem!

Hope that helps...

--Scott Calvin
Sarah Lawrence College

On Aug 20, 2012, at 9:24 AM, Huaming Guo wrote:

> 
> Thanks. In this case, is the number of neibors equal to N * SO2 in the model 
> result?  
> 
> Regards 
> 
> HM 
> 
> Bruce Ravel 写: 
> 
> > On Monday, August 20, 2012 09:02:28 PM Huaming Guo wrote: 
> >> I am using the (D)Artemis for EXAFS fitting. However, the "N" is set to be 
> >> a 
> >> constant in the (D)Artemis. We can not set the variable for this 
> >> parameter. 
> >> How can I do that if I model non crystalline samples using (D)Artemis? 
> >> Thanks a lot.  
> >  
> > In Artemis, N is a pure number, but S02 can take a math expression. 
> >  
> > In the EXAFS equation, N and S02 are multiplied together. 
> >  
> > Thus, you might consider parameterizing the amplitude by setting N to 
> > 1 and putting all of the parameterization in the math expression for 
> > S02. 
> >  
> > B 
> >  
> >  
> > --  
> >  
> >  Bruce Ravel   bra...@bnl.gov 
> >  
> >  National Institute of Standards and Technology 
> >  Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 
> >  Building 535A 
> >  Upton NY, 11973 
> >  
> >  Homepage:http://xafs.org/BruceRavel 
> >  Software:https://github.com/bruceravel 
> > ___ 
> > Ifeffit mailing list 
> > Ifeffit@millenia.cars.aps.anl.gov 
> > http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit 
> 
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Ifeffit Digest, Vol 114, Issue 15

[Scott Calvin]

Hi Marc,

After you import it into Artemis, just mark the moly atom as core, and make 
sure K edge is selected. Since it's a crystal structure, the ATOMS routine in 
Artemis will figure out the rest correctly.

--Scott Calvin
Sarah Lawrence College

On Aug 15, 2012, at 6:05 PM, Marc Porosoff wrote:

Thanks for the help, Scott.

I had a look at that file and it uses carbon as the central scattering atom and 
not Mo. Can I simply change the central atom to Mo or do I need to make other 
modifications to the file? The data I collected was at the Mo K-edge, 20 keV.


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] Mo2C

[Scott Calvin]

Hi Marc,

The Crystallography open database has a CIF file for Mo2C, which can then be 
read into Artemis:

http://www.crystallography.net

--Scott Calvin
Sarah Lawrence College

On Aug 14, 2012, at 10:30 AM, Marc Porosoff wrote:

> Hi everyone,
> 
> Does anyone have any experience modeling Mo2C in Artemis? I can't find any 
> existing INP files for Mo2C. I was wondering if anyone had any experience 
> with Mo2C or knowledge or where I can find some relevant information.
> 
> Thanks!
> Marc


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] question for ifeffit mailing list - ATHENA NORMALIZATION

[Scott Calvin]

Hi Marie,

I think you can try linear combination fitting, but you'll have to build in the 
uncertainty in normalization in to your own estimates of uncertainty. 

If you don't check the box that says "force weights to sum to 1," then you can 
allow for normalization errors in your sample.

If your standards also have only short energy ranges, then there's nothing you 
can do about that, though. As I said, you'll just have to build them in to your 
uncertainty. If you are unsure of the edge jump of a standard to 20%, then the 
contribution of that standard to the linear combination fit is uncertain by 
20%. 

Uncertainty in normalization is one of the leading contributions to the 
uncertainty that should be associated with linear combination fitting, even 
when the energy ranges are sufficient. In a case like yours, they're bigger 
than they could have been, but it doesn't mean you can't use linear combination 
analysis at all.

--Scott Calvin
Sarah Lawrence College

On Aug 8, 2012, at 1:17 PM, Marie Zwetsloot wrote:

> Hi Scott Calvin,
> 
> Thanks for your help. Yes, this is as far as my pre- and post-edge range go. 
> I realized I should have made them longer; it was my first time doing this 
> and wasnt aware that i should lengthen my pre and post-edge for later on 
> analysis. This will be good lesson for the future.
> 
> So you would not recommend doing linear combination fitting? I was planning 
> on trying it out.. But I wouldnt want to do it if with my pre- and post-edge 
> range, I am bound to derive wrong conclusions from the data. 
> 
> Best,
> Marie
> ___
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] question for ifeffit mailing list - ATHENA NORMALIZATION

[Scott Calvin]

Hi Marie,

Is that as far as your data goes? If so, the lesson for the future is to 
collect a longer pre-edge and post-edge region, if you don't plan to analyze 
them. If time is an issue, you can collect every ten eV or so in those extended 
regions--that's plenty to get a background, and doesn't take much time.

In any case, it's hard to tell from just XANES data how they should be 
normalized. Maybe what you show is fine, and maybe it's off by a bit--it's too 
hard to tell the trend of the background.

If you're not doing linear combination analysis or PCA, but are just comparing 
one spectrum to another, looking for changes, and interpreting them 
qualitatively, I think what you have is fine--in that cases, what's important 
is that the spectra be normalized similarly, and that does appear to be what 
you've done.

--Scott Calvin
Sarah Lawrence College

On Aug 7, 2012, at 4:24 PM, Marie Zwetsloot wrote:

Dear Ifeffit community,

Thanks for your questions. I am very new to this field (M.Sc. student in soil 
science), so I really appreciate that you are taking the time to give me some 
advice.
I have attached a picture of some of the normalized spectra I am working with 
as well as mu(E) spectra (I am sorry for attaching the picture, for some reason 
my email did not allow me to copy it into this message). I am doing Xanes 
analysis on phosphorus in bone char. My normalization range is from 15-60 eV 
and pre-edge range is from -13.5 to -6.75.
Maybe my normalized spectra looked actually fine; they are not that far above 1 
(more or less around 0.3 above). I just thought they looked a little 
strange/skewed.

Let me know what you think.

Thanks,
Marie



___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

Re: [Ifeffit] [IFEFFIT] question about chemically unordered systems

[Scott Calvin]

Hi Falk,

These pages may help:

http://cars.uchicago.edu/ifeffit/FAQ/FeffitModeling?highlight=%28doping%29#How_do_I_handle_doped_materials.3F_Why_doesn.27t_Atoms_handle_doped_materials.3F

http://cars.uchicago.edu/ifeffit/Doped

http://cars9.uchicago.edu/~ravel/software/doc/Atoms/Atoms/node4.html#SECTION0044

--Scott Calvin
Sarah Lawrence College

On Aug 7, 2012, at 10:52 AM, Meutzner, Falk wrote:

> Dear Matt, dear community
> 
> thank you for your fast reply!
> I am not sure, if I understand you correctly. I tried what you suggested, but 
> the result is just the same. I will try to give some more insight:
> 
> Our binary system is bcc (pm-3m), there is one atom on the position (0,0,0) 
> and another one on the position of (0.5,0.5,0.5).
> After creation of the material, we have a chemically disordered state. Any of 
> the two atoms can occupy any of the aforementioned positions.
> After heat treatment however, the material chemically orders itself and 
> becomes CsCl structure. In theory, atom A will always be on the corner at 
> (0,0,0), while atom B will always be in the centre of the cell at 
> (0.5,0.5,0.5).
> 
> Our problem is, we would like to distinguish between these two phases, 
> because 
> we possibly have mixtures of the ordered and disordered states.
> 
> We already achieved the ordered state with atoms, because they are unique 
> positions (Atom A at (0,0,0), atom B at (0.5,0.5,0.5). Unfortunately we are 
> unsure how to model the disordered state. We would need the possibility to 
> tell the programme, that the observed atom will always be A at (0,0,0), while 
> the other atom has a possibility of 50% to be A and 50% to be B.
> 
> For the programme it seems not to matter which of the positions our observed 
> atom is at. Therefore the calculated spectra are the same.
> 
> I hope I was able to give more insight into our problem.
> Thank you! Best Regards
> Falk


___
Ifeffit mailing list
Ifeffit@millenia.cars.aps.anl.gov
http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit