Magnesium by Conductivity Detector
Posted: Fri Nov 04, 2011 8:56 pm
We are trying to set up a method for Mg using a conductivity detector and we are having issues with baseline and linearity.
The method was supplied to us from a sister company to perform a cleaning validation. As we will be validating the method in-house we have some room to make changes as necessary.
Supplied method:
Equipment: Ion Chromatgraph (Dionex DX-320)
Detector: Conductivity @35°C
Column: Dionex CS14, 250x4mm, with CS14 guard column 50x4mm
Column Temp: 35°C
Mobile Phase: 10mM methanesulphonic acid (MSA)
Flowrate: 1.2mL/min (to make Mg elute at 8.5min)
Injection volume: 5µL
CSRS Cation Suppressor unit, 4mm
Standard solution is 12mg/L Mg in 0.1M HCl
(20mg MgO dissolved in 100mL 0.1M HCl, diluted 4/100mL in 0.1M HCl)
System suitability solution is 33mg/L Mg and Ca in 0.1M HCl (Resolution NLT 1.5)
System suitability is met.
We are using a Waters 2695 separations unit with a Waters 432 Conductivity detector. We have a Dionex CSRS suppressor and we have determined that the optimum suppressor current to be around 25 – 30mA.
We are injecting 10µL.
We have a stable baseline with a background conductivity of around 8µS.
Injection of solvent (0.1M HCl) gives a smooth baseline with an injection peak at around 3.4 min.
However, when we inject a standard we are getting a dip in the baseline at around the time for Mg which interferes at lower concentrations.
We have tried changing the solvent to 0.01M HCl. This reduces the injection peak but does not change the baseline dip, or improve the linearity.
We have also tried making the Std in mobile phase, this too did not improve the basline dip.
We are also having a problem with linearity. The calibration curve gives a quadratic curve.
Std Conc. (Mg mg/L) Area
2.4 1718
3.0 3109
4.8 9274
6.0 16797
12.1 72240
30.2 382877
Supplied method claims linearity from 2 to 10mg/L with a co-efficient of 0.9996.
Claimed LOD is 0.65mg/L
We see a baseline blip at 2.4mg/L (as above) and can integrate it, but it’s only a peak if you squint hard.
I am going to try making the Std using a different salt of Mg or Ca if we have one, probably Cl.
Any suggestions or advise would be most appreciated, thanks.
The method was supplied to us from a sister company to perform a cleaning validation. As we will be validating the method in-house we have some room to make changes as necessary.
Supplied method:
Equipment: Ion Chromatgraph (Dionex DX-320)
Detector: Conductivity @35°C
Column: Dionex CS14, 250x4mm, with CS14 guard column 50x4mm
Column Temp: 35°C
Mobile Phase: 10mM methanesulphonic acid (MSA)
Flowrate: 1.2mL/min (to make Mg elute at 8.5min)
Injection volume: 5µL
CSRS Cation Suppressor unit, 4mm
Standard solution is 12mg/L Mg in 0.1M HCl
(20mg MgO dissolved in 100mL 0.1M HCl, diluted 4/100mL in 0.1M HCl)
System suitability solution is 33mg/L Mg and Ca in 0.1M HCl (Resolution NLT 1.5)
System suitability is met.
We are using a Waters 2695 separations unit with a Waters 432 Conductivity detector. We have a Dionex CSRS suppressor and we have determined that the optimum suppressor current to be around 25 – 30mA.
We are injecting 10µL.
We have a stable baseline with a background conductivity of around 8µS.
Injection of solvent (0.1M HCl) gives a smooth baseline with an injection peak at around 3.4 min.
However, when we inject a standard we are getting a dip in the baseline at around the time for Mg which interferes at lower concentrations.
We have tried changing the solvent to 0.01M HCl. This reduces the injection peak but does not change the baseline dip, or improve the linearity.
We have also tried making the Std in mobile phase, this too did not improve the basline dip.
We are also having a problem with linearity. The calibration curve gives a quadratic curve.
Std Conc. (Mg mg/L) Area
2.4 1718
3.0 3109
4.8 9274
6.0 16797
12.1 72240
30.2 382877
Supplied method claims linearity from 2 to 10mg/L with a co-efficient of 0.9996.
Claimed LOD is 0.65mg/L
We see a baseline blip at 2.4mg/L (as above) and can integrate it, but it’s only a peak if you squint hard.
I am going to try making the Std using a different salt of Mg or Ca if we have one, probably Cl.
Any suggestions or advise would be most appreciated, thanks.