Quantitative EDS References From NIST

[Probeman]

Here are several “academic” references to this topic with which [Dale has] been involved:

Newbury, D. E., Swyt, C. R., and Myklebust, R. L., "'Standardless' Quantitative Electron Probe Microanalysis with Energy-Dispersive X-ray Spectrometry: Is It Worth the Risk?", Analytical Chemistry, 67 (1995) 1866-1871.


Newbury, D. E. and Ritchie, N.W.M., “Is Scanning Electron Microscopy/Energy Dispersive Spectrometry (SEM/EDS) Quantitative?”, SCANNING, 35 (2013) 141-168.


Nicholas W. M. Ritchie and Dale E. Newbury, Standardless Analysis – Better But Still Risky, Micros. Microanal 20 (Suppl 3) 2014, 696-697.


Newbury, D. E. and Ritchie, N. W. M., “Review: Performing Elemental Microanalysis with High Accuracy and High Precision by Scanning Electron Microscopy/Silicon Drift Detector Energy Dispersive X-ray Spectrometry (SEM/SDD-EDS)”, J. Materials Science 50 (2015) 493-518.

Electronic version now available:

http://www.springerlink.com/openurl.asp?genre=article&id=doi:10.1007/s10853-014-8685-2

[jared.wesley.singer]

Perhaps this is an appropriate and safe place (away from the EDS bashing) to give props to the Son of Desktop Spectrum Analyzer (DTSA-II).  It is as robust as your standard blocks times patience.   

In my [small] experience, it is not a big time saver over classic EPMA.  The analyst still needs concentration standards, lineshape references, detector calibrations, and drift control.  But, if the laundry list is totally unknown, I'll run 5-10 minute EDS scan on the unknown and DTSA-II will nail it in the hundredths place. 

Jared

[John Donovan]

Hi Jared,
Funny how it takes work to get accurate results with both EDS and WDS!

If you interface your Thermo or Bruker system to Probe for EPMA you can run it in "EDS only" mode. We did this to avoid Bragg defocusing in a magnetic sample as seen here:

Un    4 Ni-Cr (6), trav
TakeOff = 40.0  KiloVolt = 15.0  Beam Current = 30.0  Beam Size =    0
(Magnification (analytical) =  40000),        Beam Mode = Analog  Spot
(Magnification (default) =      400, Magnification (imaging) =    800)
Image Shift (X,Y):                                          .00,   .00
Number of Data Lines: 177             Number of 'Good' Data Lines:  10
First/Last Date-Time: 11/17/2014 06:27:15 PM to 11/18/2014 02:20:28 AM

Average Total Oxygen:         .000     Average Total Weight%:  101.119
Average Calculated Oxygen:    .000     Average Atomic Number:   26.619
Average Excess Oxygen:        .000     Average Atomic Weight:   56.144
Average ZAF Iteration:        3.00     Average Quant Iterate:     2.00

Un    4 Ni-Cr (6), trav, Results in Elemental Weight Percents

SPEC:       Ni      Cr      Co      Fe      Cu      Al       O
TYPE:     SPEC    SPEC    SPEC    SPEC    SPEC    SPEC    SPEC

AVER:     .000    .000    .000    .000    .000    .000    .000
SDEV:     .000    .000    .000    .000    .000    .000    .000
 
ELEM:       Ni      Cr      Co      Fe      Cu      Al       O
BGDS:      EDS     EDS     EDS     EDS     EDS     EDS     EDS
TIME:    60.00   60.00   60.00   60.00   60.00   60.00   60.00
BEAM:    29.79   29.79   29.79   29.79   29.79   29.79   29.79

ELEM:       Ni      Cr      Co      Fe      Cu      Al       O   SUM 
   105  69.521  32.417    .000    .000    .055    .023    .000 102.016
   106  68.030  32.359    .039    .000    .147    .030    .000 100.605
   107  68.267  33.494    .041    .000    .151    .028    .000 101.980
   108  66.965  33.770    .027    .000    .079    .041    .000 100.882
   109  66.563  34.240    .000    .000    .076    .053    .473 101.406
   110  66.658  35.237    .042    .000    .000    .040    .000 101.977
   111  65.309  35.193    .041    .000    .130    .041    .000 100.715
   112  64.648  35.824    .000    .000    .150    .033    .000 100.655
   113  64.175  36.191    .072    .000    .000    .024    .000 100.462
   114  63.680  36.670    .000    .000    .082    .062    .000 100.494

AVER:   66.382  34.539    .026    .000    .087    .037    .047 101.119
SDEV:    1.912   1.523    .025    .000    .058    .013    .150    .657
SERR:     .605    .481    .008    .000    .018    .004    .047
%RSD:     2.88    4.41   96.27     .63   66.16   33.80  316.23
STDS:      528     524     527     526     529     513     912

STKF:   1.0000   .9988   .9977  1.0000   .9974   .9857   .2324
STCT:   380.55  876.47  484.08  597.31  293.04 3610.13  370.11

UNKF:    .6551   .3597   .0002   .0000   .0008   .0002   .0003
UNCT:   249.29  315.61     .12     .00     .24     .68     .44
UNBG:      .00     .00     .00     .00     .00     .00     .00

ZCOR:   1.0133   .9603  1.0558   .9317  1.0604  2.0055  1.6982
KRAW:    .6551   .3601   .0002   .0000   .0008   .0002   .0012
PKBG:      .00     .00     .00     .00     .00     .00     .00


I should mention that this is using standards and constructing k-ratios by ratioing the unk and std EDS net intensities. Pretty cool!

My assistant Julie asked recently if we can perform our quant interference correction on EDS elements and I had to admit, not yet.  But since the interference correction is based on the actual concentrations rather than a simple intensity ratio, it should be totally doable.  Something else to do!

[John Donovan]

My assistant Julie asked recently if we can perform our quant interference correction on EDS elements and I had to admit, not yet.  But since the interference correction is based on the actual concentrations rather than a simple intensity ratio, it should be totally doable.  Something else to do!

One can now perform quantitative interference corrections for WDS elements interfered by EDS elements, WDS by WDS, EDS by WDS and EDS by EDS...

See here for details:

http://probesoftware.com/smf/index.php?topic=482.msg2826#msg2826

[bgarcia]

As a possibly-offensive aside relating to the topic of quantitative EDS, I have a quick question to ask of the forum:

Are there any software programs, outside of the ever-wonderful PFEPMA (for which you need a probe) and DTSA-II, that allow a user to quantify EDS spectra using standards? It sounds like Bruker might have something in their software interface, but not being a customer I can't find too many details on it.

[Probeman]

As a possibly-offensive aside relating to the topic of quantitative EDS, I have a quick question to ask of the forum:

Are there any software programs, outside of the ever-wonderful PFEPMA (for which you need a probe) and DTSA-II, that allow a user to quantify EDS spectra using standards? It sounds like Bruker might have something in their software interface, but not being a customer I can't find too many details on it.

Hi Ben,
It's not so much that other EDS softwares don't have support for standards, they do. But I have several issues with how they have implemented it.

The main thing is they have not made it easy to load ones standard positions and re-acquire the standard spectra (profiles) using automation.  Since 99% of their users don't use standards, they don't have much incentive to fix this.

The method PFE uses is very robust as it automatically acquires EDS spectra along with the WDS data at the same time.  Then it calculates the net intensities necessary and constructs a k-ratio just as one would for WDS.  In addition corrections such as interferences and standard intensity drift are automatically handled.

In fact, several people I know, use PFE even when they just need to acquire EDS quant (without any WDS elements), just for the ease of use in acquiring standard spectra.  But as you said PFE only runs on EPMA instruments at the current time.

A couple of years ago, my suggestion to EDS vendors to eliminate the Standardless button was met with exclamations of: "What?  You expect my customers to buy standards?"   And my answer: "yes, if they want to call it quantitative analysis".

As NIST has said repeatedly, if one is using "standardless" quant, one should not report any numbers at all, simply the words "major", "minor" and "trace" for each element concentration.

Just my 2 cents.
john

[bgarcia]

It's always comes down to incentive, huh? 

So for the Probe implementation, do you need to use reference standards for line shapes (in addition to compositional standards) still, or can you get around that some other way?

[Probeman]

It's always comes down to incentive, huh? 

So for the Probe implementation, do you need to use reference standards for line shapes (in addition to compositional standards) still, or can you get around that some other way?

Hi Ben,
As I tried to say above, in PFE, the standard spectrum acquisition is automatic for both WDS and EDS standards.  The net intensities from the unknown spectra are used for the numerator of the k-ratio and the net intensities from the standard spectra are used for the denominator of the k-ratio.  The peak shape or whatever parameters are utilized separately in the EDS OEM software and therefore PFE allows one to specify an EDS project/fit method with whatever EDS peak fit parameters one wants, from inside PFE, for the Thermo or Bruker system, for best EDS quant results.
john

[bgarcia]

Ok, gotcha.

Hopefully, if the stars align and Congress approves a budget, I'll be calling for an EDS/WDS software hook up in the Spring.

[Nicholas Ritchie]

Another reference that people might find useful:  While it is called "Proposed practices for validating the performance of instruments used for automated inorganic gunshot residue analysis", it really is about validating the performance of an SEM/EDS system.  It is available (for free) from here: https://www.sciencedirect.com/science/article/pii/S2468170920300400?via%3Dihub

[orlandin]

Hi all, I watched an Oxford Instruments webinar on quantitative EDS yesterday (https://analyticalscience.wiley.com/do/10.1002/was.0004000113) and not only was it an excellent summary of the topic and workflow, but I also learned that Dr. Duncan Muir at Cardiff has been publishing standardized EDS data collected and processed in Aztec for years. Not only that, but the analyses are published in the sort of journals that ought to instantly silence any reviewers (or potential lab customers) stuck in the 1990s version of EDS. I highly recommend everyone check out that webinar and these references when you encounter someone skeptical of the accuracy and precision of properly-treated EDS (apologies for the geoscience bias):

Bennett, Emma N., C. Johan Lissenberg, and Katharine V. Cashman. "The significance of plagioclase textures in mid-ocean ridge basalt (Gakkel Ridge, Arctic Ocean)." Contributions to Mineralogy and Petrology 174.6 (2019): 1-22.

Bennett, Emma N., et al. "Deep roots for mid-ocean-ridge volcanoes revealed by plagioclase-hosted melt inclusions." Nature 572.7768 (2019): 235-239.

Buchs, David M., and Samuel AP Oemering. "Long-term non-erosive nature of the south Costa Rican margin supported by arc-derived sediments accreted in the Osa Mélange." Earth and Planetary Science Letters 531 (2020): 115968.

Guice, George L., et al. "Re-evaluating ambiguous age relationships in Archean cratons: Implications for the origin of ultramafic-mafic complexes in the Lewisian Gneiss Complex." Precambrian Research 311 (2018): 136-156.

Lambart, Sarah, et al. "Highly heterogeneous depleted mantle recorded in the lower oceanic crust." Nature Geoscience 12.6 (2019): 482-486.

McDonald, Iain, et al. "Homogenisation of sulphide inclusions within diamonds: A new approach to diamond inclusion geochemistry." Geochimica et Cosmochimica Acta 216 (2017): 335-357.

There may well be more out there that I just haven't heard of or found the right search term for.