We use cookies to improve functionality and personalize your content. To ensure compliance with the latest e-Privacy directive, we kindly ask for your consent to set these cookies. Learn more.
General Spectroscopy - How do you verify an ICP measurement ?
I constructed a 3 point calibration curve using 40, 100 and 1000ppm Ca standards where the RSD's were >0.999+.
Not knowing how much Ca the test sample contained, I made 3 dilutions in triplicate which where approx. x50, x200 and x400 dilutions.
The average ICP results for the triplicates for each dilution were
ICP = 392 mg/l x dilution factor of 50 = 19832 mg/l
ICP = 125 mg/l x dilution factor of 200 = 25217 mg/l
ICP = 65 mg/l x dilution factor of 400 = 26093 mg/l
The ICP results for all dilutions fell within the calibration curve. However which result should I take as being representative of the actual concentration of Ca present in the sample and how would I verify this?
Thanks
| Responses | |
|---|---|
|
admin
Total posts: 529
|
I'm also super excited about this thread. I would like to get some more details about it. I hope someone explain it. Thank you so much! |
|
We recommend always checking more than one wavelength. The data from all wavelengths should agree unless there are interferences affecting the results. If possible, can you please provide the data for the calibration curve (i.e., raw intensity and concentration of calibration solutions)?
Another concern is that the sample may not be homogeneous. Please find more information in our Trace Analysis Guide about sampling and strategies for ensuring representative sampling at the following link: https://www.inorganicventures.com/trace-analysis-guide/sampling-and-subsampling. Particularly the “Subsampling” section may be helpful.
Since matrix-matching does not sound feasible in this case, we recommend using the standard additions method of analysis to account for issues with a complex matrix. More information can be found at the following link: https://www.inorganicventures.com/icp-guide/standard-addition-internal-standardization-and-isotope-dilution. Please also see the following link on our website to learn more about method validation and how to ensure that the established method is fit for purpose: https://www.inorganicventures.com/trace-analysis-guide/method-validation. |
|
|
ROMAN HARABURDA
Total posts: 2
|
*Are you using OES or MS? OES *Are you using one or multiple lines/masses? Which ones? just the one 317.933 *What is the limit of linearity of the instrument for Ca at each mass/wavelength used? not sure - however as I was able to create a linear cal line ( in radial view)where the max std was 1000ppm i assuumed that the response for Ca at the wavelength chosen was linear if it fell within the calibration curve. *Has the method been validated using a reference material (similar to sample)? there isnt a reference materal available. The test sample was an aqua regia digest of shredded and milled PCB boards from mobile phones. * What is the RSD for each measurement? at x50 dil RSD = 0.4% (av/3) ; at x200 dil RSD = 0.73% (av/3); at x400 dil RSD= 0.87%. * What is the nature/matrix of the sample? shredded and milled PCB boards from mobile phones * Is the sample homogeneous? the sample has good homeogenity - however even if it wasnt t wouldnt really matter - approx 1g of the sample was digested in triplicate. Whilst there may be some variation between the 1g samples used to make the triplicate solutions, the ICP results for each triplicate for each dilution should show some consistency. * Are there other elements involved in the analysis? yes - lots - e.g. Cu, Ag, Sn, Al, Ni, Fe, Ba where the majors are Fe, Ni, Ca, Al and Sn * Have potential interferences been accounted for? not really - I wouldnt know how to do this. I have assumened the ICP selected the most favourable wavelength. I can go back into the results to see if any wavelenghths are close together which could result in potential interference. * Is the sample matrix matched to the standards? as much as it could be. The sample matrix in terms of analytes is complex and so the only matching done was the acid background. Hopefully this further information helps (?) Many thanks |
|
There are several factors that could be causing the inconsistent Ca results being observed. We recommend using all available wavelengths (or masses for ICP-MS) for Ca and validating your results by analyzing a reference material with similar composition to your sample. You mention that your regression coefficient for the calibration curve was >0.999+, which is good. However, some instrument software packages will draw non-linear calibration curves which could be causing your issue. Linear curves are preferred because they provide a unique concentration for each intensity and limits of linearity are easier to see visually. There are also contamination concerns for Ca and it seems that the working solutions with higher dilution factors are giving higher results which may point to contamination during sample preparation and/or dilution. If you can provide answers to the following questions, this will help us better understand your analysis and we would be happy to provide additional assistance and/or a more specific response:
*Are you using OES or MS? *Are you using one or multiple lines/masses? Which ones? *What is the limit of linearity of the instrument for Ca at each mass/wavelength used? *Has the method been validated using a reference material (similar to sample)? * What is the RSD for each measurement? * What is the nature/matrix of the sample? * Is the sample homogeneous? * Are there other elements involved in the analysis? * Have potential interferences been accounted for? * Is the sample matrix matched to the standards? |