[Ifeffit] CN and bond distances in Artemis

[Scott Calvin]

Hi JA,

I agree with Anatoly that remeasuring in 
transmission is a good idea, if you have the beam 
time. But I would also note that the value you 
have for S02 with the correction is within the 
realm of possibility, considering the 
uncertainty: 1.20 - 0.30 = 0.90. And part of the 
reason for starting with a standard in the first 
place is to work out some of the other 
difficulties that may arise; e.g. questions of normalization.

If it is not practical for you to remeasure right 
now, you can try some of the things that can 
reduce the uncertainty in S02: fitting with 
multiple k-weights, changing the k-range being 
fit, fitting multiple coordination shells, etc..

I also just checked a past paper of mine, and see 
I ended up with an S02 for a set of platinum 
samples (transmission) of 1.16 +/- 0.05, 
suggesting that I wasn't doing something quite 
right (probably not sample preparation or 
measurement issues, since most of those tend to 
lower S02). But since that set of samples 
included a standard, and I was really interested 
in the differences between samples (including the 
standard), the fact that I have some systematic 
error that's causing the S02 to come out a tad 
large was acceptable, if a bit troubling. The 
paper is S. Calvin et al., Appl. Phys. Lett. 87, 233102 (2005).

--Scott Calvin
Sarah Lawrence College

P.S. Just as I finished writing this, Matthew 
Marcus' post came through. I do have Pt foil data 
in transmission that I can dig up for you, if 
you'd like. Maybe I'll send it to Matt to put in 
the database; it's pretty good quality, as I 
recall (good data out to 20 inverse angstroms or so).

At 01:31 PM 12/5/2006, Juan Antonio Maciá Agulló wrote:

>Hi all,
>
>I did fits with self absorption corrected data 
>(Fluo and Booth algorithms) and the
>value for S02 is 1.20 (0.30). It seems that I am not very lucky, if I take an
>average value (1.20 and 0.50), it should be ok (0.85) jeje, I am joking.
>
>Then...what can I do now?