# Analysis of Atomic Absorption Spectroscopy results

1. Apr 8, 2016

### Organoid

1. The problem statement, all variables and given/known data
Hello,

This week I ran an AA Spectroscopy experiemnt in my university, in which we had to analyze the percentage of Copper in a metal coin, and the containment of Iron in morning cereal.

We used 2 standard solutions of 1 and 10 ppm (containing both Copper and Iron), and exprimented a few diluted solutions of the extracted Copper and Iron from the substances.

The results we received were an average value and standard deviation of the 3 tests run by the machine.

2. Relevant equations
My question is: what do the concentration results that the machine provides us with mean? i.e. if 1 and 10 ppm standards had an average of 0.525 and 1.426 respectively, and the diluted solutions (x10, x100, x1000) had an average of 10.72, 4.094 and 2.644 respectively, how can I analyze these results to calculate the original conc. in the original substance?

3. The attempt at a solution
I've generated a standard curve using the standards' results and reveied a linear-fit equation of the graph with a matching slope.
I know that the equation connecting those variables is AA = K x C, (where AA=absorbance, k is the trend line's slope and C is the conc.) but when I tried to apply my results of the diluted solutions to that equation, and by multiplying by the dilution factor, they don't make sense. Each solution gives different concentration of the original solution.

Am I analyzing my results using the standard curve wrong?

I'd appretiate any help, thank you.

2. Apr 8, 2016

### drvrm

i think you try to relate concentration of the material with dilution so that the AA value can be correlated -perhaps you know k value from the graphs.

3. Apr 8, 2016

### Organoid

I do know the k value from the graphs, but I actually don't have any correlation between the results - if I calculate backwards to get to the initial conc. I get totally different results from each dilution. it doesn't make sense

4. Apr 8, 2016

### Staff: Mentor

If these numbers are correct, then you definitely don't have a linear relationship between AA and C.

I guess you can't redo the experiment, but it is not good at all that the AA of the standard are not in the same range as the AA of the sample.

5. Apr 8, 2016

### Organoid

I think that we couldn't predict such high conc. of the original substances. Either way, I could use extrapolation to calculate the initial conc., but if these are the results we recieved, how can I still utilize them?

6. Apr 8, 2016

7. Apr 8, 2016

### Organoid

I'm already familiar with this procedure, I wish I only had tested one solution but I had a series of dilutions so it makes it much more complicated since there is actually no correlation between the absorbance and the dilute. By my last phrase I mean that if I dilute by x10, I don't get a linear fit between the absorbance of the dilution and the original solution I used to make the x10 dilution for.