Testing the Dilution Factor using an ICP-OES Chemical Standard

We are running a series of dilution studies on our ICP-OES 5110 (Intelliquant Software Version 7.6).

We want to determine the dilution factor that offers the most reliable results for each one of the samples from our plant.

One dilution study was done with Tin (Sn) Standard from Ricca Chemical Co. 1000.0 ppm +/- 7.2 ppm (Matrix: 1% HNO3, 1% HF), used as our testing probe.

Testing procedures. Five dilution samples were prepared (digital pipette used for precise volume) and tested, using the Tin Standard.

Dilution Factor 1: 12 mL of Tin Standard + 0 mL of 1% HNO3

Dilution Factor 2: 6 mL of Tin Standard + 6 mL of 1% HNO3

Dilution Factor 3: 4 mL of Tin Standard + 8 mL of 1% HNO3

Dilution Factor 4: 3 mL of Tin Standard + 9 mL of 1% HNO3

Dilution Factor 5: 3 mL of Tin Standard + 12 mL of 1% HNO3

ICP-OES configuration: Radial View, RF Power 1.2 kW, all other parameters set to standard. Tin wavelength 189.925 nm.

Results:

Sn RADIAL 189.925 189.925
Dilution Factor Analyte 1% HNO3 Concentration Intensity
(mL) (mL) (ppm) (ct/s)
blank 0.00 11.12
Sn standard 1000.00 222599.22
none 1 12 0 993.50 221152.45
1:1 2 6 6 522.34 116276.68
1:2 3 4 8 364.89 81230.91
1:3 4 3 9 275.68 61374.56
1:4 5 3 12 224.78 50045.01

What should it be the relationship between the Intensity (Concentration) vs Dilution Factor?

We have found a power relationship between them. Shouldn't it be linear?

Appreciate any ideas or comments. Thanks.

Parents
  • The issue I believe is the density of the solutions is changing as you dilute, and you prepared by volume which does not account for the density. You should prepare by weight.

  • You should use 5 points calibration from std 200ppm to std 1000ppm vs their intensities, instead of only 1 point calibration of Std 1000pm vs its intensities. See the excel picture I attached. Using 1 point cal, you just force the cal curve to zero at 0 concentration which is not correct because at 0 concentration, there are still signals in the instrument.

    When using 5 points calibration, the %bias to their targets is only -3 to 2%. 5 points cal gives you the nature of the instrument bias with the natural intercept and slope. You can see the %bias of the cal'ed concentration of you stand stock based on the 5 points cal is only -0.28%.

Reply
  • You should use 5 points calibration from std 200ppm to std 1000ppm vs their intensities, instead of only 1 point calibration of Std 1000pm vs its intensities. See the excel picture I attached. Using 1 point cal, you just force the cal curve to zero at 0 concentration which is not correct because at 0 concentration, there are still signals in the instrument.

    When using 5 points calibration, the %bias to their targets is only -3 to 2%. 5 points cal gives you the nature of the instrument bias with the natural intercept and slope. You can see the %bias of the cal'ed concentration of you stand stock based on the 5 points cal is only -0.28%.

Children
Was this helpful?