Page 1 of 1
Internal standard too high
Posted: Thu Oct 23, 2014 6:42 pm
by Allyptedia
Hello
I have a question about the internal standard bromoform. I have to detect 4 different sorts of chloroethanes (111 trichloroethane, 11 dichloroethane, 112 trichloroethane & 12 dichloroethane) in ground water with GC-MS. First, I do a liquid-liquid extraction with pentane and adjust 1 ml internal standard. After detection with the GC-MS, I always have a recovery of 140-150% of the internal standard in comparison with the calibration curve (1 ml of compounds + 50 µl internal standard).
I’ve already changed the injection needle and cleaned the column inlet but nothing helped. Does anyone have an idea why my recovery is too big?
Re: Internal standard too high
Posted: Thu Oct 23, 2014 7:55 pm
by Peter Apps
My first idea is that you are adding too much IS to the sample, but you give no details so it is impossible to tell.
Peter
Re: Internal standard too high
Posted: Sat Oct 25, 2014 5:49 pm
by Allyptedia
I have to detect 4 chloroethanes in ground water samples.
First I do a liquid-liquid extraction. For this I use 50 ml of a ground water sample, 20 ml n-pentane and 1 ml IS (bromorform in water). I shake for 5 minutes. After the extraction I take 1 µl of the upper phase and inject it into a GC-MS.
The callibration line exists out of a 1ml mixture of all my components in pentane and 50 µl IS (bromoform in water).
Properties of the GC-MS:
- DB 5ms column packed with silica
- Carrier gas: helium with a flow of 1 ml/min
- Splitless injection
- Solvent delay is 2,05 minutes
The program of the GC-MS:
- start temp: 30°C and 3 minutes constant
- gradiënt: 1°C/minute till 35°C
- gradient: 25°C/minute till 125°C
Yes maybe I add too much, but how come I've never added too less IS?
Allyptedia
Re: Internal standard too high
Posted: Sat Oct 25, 2014 11:17 pm
by Peter Apps
Having suggested that you added too much IS, I was expecting that your answer would include how much IS you are adding, not how much water.
Are you assuming that because you start with 20 ml of pentane in the extraction and 1 ml of pentane in the standard that 1 ml of IS solution added to the extract is equivalent to 1/20 ml (= 50 ul) added to the standard ?
What is the final volume of pentane from the extract ?
The answer to why you have never added less internal standard is that evaporation never increases the volume of pentane.
Peter
Re: Internal standard too high
Posted: Sun Oct 26, 2014 9:52 am
by dblux_
I have to detect 4 chloroethanes in ground water samples.
First I do a liquid-liquid extraction. For this I use 50 ml of a ground water sample, 20 ml n-pentane and 1 ml IS (bromorform in water)...
Don't waste pentane. Change proportions to 50 mL water sample and 2 mL n-pentane. Your LOQ will be much better.
Properties of the GC-MS:
- DB 5ms column packed with silica
How have you packed this column with silica ?
Re: Internal standard too high
Posted: Sun Oct 26, 2014 5:55 pm
by Allyptedia
In the calibration curve, I use 50 µl IS. For the liquid-liquid extraction, I use 1 ml IS because you have to take the volume of n-pentane I use in account (1 ml has 50 µl IS so 20 ml has 1 ml IS)
The IS for the calibration line has a concentration of 2000 ppm (bromoform in pentane)
The IS for the extraction has also a concentration of 2000 ppm (bromoform in water)
Maybe I can try to take less n-pentane.
Allyptedia
Re: Internal standard too high
Posted: Mon Oct 27, 2014 7:19 am
by Peter Apps
Measure the volume of pentane extract after you have done the extraction.
Peter
Re: Internal standard too high
Posted: Thu Nov 20, 2014 10:19 am
by Allyptedia
Thank you very much for the help. With the correction of the volume after extraction, is my IS correct.
Now I have to change my extraction because I use a lot of pentane. After extraction, my IS is correct but the component 111-TCA is too high.
Calculations spiked water:
Component 111-TCA with a conc of 1000 ppb.
1,1,1 trichloorethane
1,32 g/l = 1,32 kg/l = 1320 g/l = 1320000 mg/l (ppm)
1320000 mg/l . 0,97 = 1280400 mg/l = 1280400000 µg/l
First 1000x dilute in methanol
10 µl 111-TCA in 10 ml (concentration is 1280400 µg/l)
1280400 µg --> 1 l
1280,4 µg --> 1000 µl
(25000 )/(1280,4 ) = 19,53 µl
I spiked 20 µl 111-TCA in 25 ml water. Then I added 50 µl IS and 1 ml of pentane.
After extraction I become a concentration of 8000 ppm of 111-TCA which is 8x too high. Does anybody know why it's too high?
Allyptedia
Re: Internal standard too high
Posted: Thu Nov 20, 2014 11:02 am
by Peter Apps
Thank you very much for the help. With the correction of the volume after extraction, is my IS correct.
Now I have to change my extraction because I use a lot of pentane. After extraction, my IS is correct but the component 111-TCA is too high.
Calculations spiked water:
Component 111-TCA with a conc of 1000 ppb.
1,1,1 trichloorethane
1,32 g/l = 1,32 kg/l = 1320 g/l = 1320000 mg/l (ppm)
1320000 mg/l . 0,97 = 1280400 mg/l = 1280400000 µg/l
First 1000x dilute in methanol
10 µl 111-TCA in 10 ml (concentration is 1280400 µg/l)
1280400 µg --> 1 l
1280,4 µg --> 1000 µl
(25000 )/(1280,4 ) = 19,53 µl - you are dividing a volume by a mass and getting a volume
I spiked 20 µl 111-TCA in 25 ml water. Then I added 50 µl IS and 1 ml of pentane.
After extraction I become a concentration of 8000 ppm of 111-TCA which is 8x too high. Does anybody know why it's too high? check your maths
Allyptedia
Re: Internal standard too high
Posted: Thu Nov 20, 2014 11:54 am
by Allyptedia
Excuse me but I've typed something wrong. It is not 1,32 g/l but g/cm³.
1,1,1 trichloorethane (1000 ppb)
1,32 g/cm³ = 1,32 kg/l = 1320 g/l = 1320000 mg/l (ppm)
1320000 mg/l . 0,97 = 1280400 mg/l = 1280400000 µg/l
First 1000x dilute in methanol
10 µl 111-TCA in 10 ml (concentration is 1280400 µg/l)
1280400 µg --> 1 l
1280,4 µg --> 1000 µl
When I calculate it with C1 . V1 = C2 . V2
C1 = 1000 µg/l C2 = 1280400 µg/l
V1 = 25000 µl V2 = ??
(25000 µl . 1000 µg/l)/(1280400µg/l) = 19,53 µl
I think my calculation is correct, and I don't know why I become a concentration of 8000 µg/l after extraction when I spiked only 1000 µg/l.
Allyptedia
Re: Internal standard too high
Posted: Thu Nov 20, 2014 12:45 pm
by Peter Apps
Have you run a blank - i.e. not spiked ?
Peter