Troubles with calibration

[natali.g17]

Hi,

I'm having some trouble with calibration curves for some of the anions (especially bromides). There are 6 calibration levels for fluorides, chlorides, nitrates, sulfates and phosphates: 0.1 mg/l, 1 mg/l, 2.5 mg/l, 5 mg/l, 7.5 mg/l and mg/l. Nitrites and bromides have one more: 0.05 mg/l. Fluorides always seem to have the best linearity (>0.999) while bromides always seem to be problematic. Namely, their linearity meets the criteria (>0.995) but concentractions below 5 mg/l are below the curve and when I spike at those concentrations recoveries are too small (around 70%, requirements are 80-110%), also two highest concentrations are above the curve. So I'm not sure what exactly is the problem, are lower levels too low or higher levels too high. I tried splitting the calibration in two (for lower and higher levels) it kinda worked for lower one but there is still issue with the higher one so I suppose it's 7.5 mg/l and 10 mg/l that are too big. I thought maybe there is cross-contamination but blanks afterwards are fine. If anyone has an idea how to improve this I would be very grateful.

[eja85]

Could you provide more detail about your instrument? What column are you using? What is your eluent?

[anionman]

There is a limit to how wide your calibration can be. I use an as-18 column and have to calibrate fluoride from 0.1 - 4.0 mg/L. I tried 0.1 - 8.0 mg/L but that would not hold for more than a week. I just have to dilute all fluorides > 4 mg/L.

I have never calibrated Bromide lower than 1 mg/L. It would be a problem due to coelution with carbonate and sulfate.

[peterbopei85]

1.Nonlinear detector response: For certain ions, the detector's response characteristics can change in the low and high concentration regions, which can easily lead to deviations from linearity.
2.Matrix effect: Other components in the sample may interfere with the detection of bromide. At low concentrations, the impact of this interference on the measurement results is more significant.
3.Problems with standard solutions: If high-concentration standard solutions are stored for too long, they may volatilize or become contaminated, leading to changes in concentration.
4.Injection system residue: Although the blank test results are normal, there may still be residue in the injection system, affecting the determination of high-concentration samples.

[Nadine Seifert, MIHQ]

What column and eluent are you using? That linearity of bromide is bad is quite strange, but perhaps it could be related to a system peak close by?