Headspace Vs Direct injection

[IrishMike]

Hi again

I am trying to put together a method for analysing liquid wastes containg solvents(MeOh, EtOh, Acetone, Chex, ETAC, THF and others) The water content can vary between 0 and ~ 100%

We can do both headsapce and direct injection..

Which would work out best??

Thanks

Mike

[Peter Apps]

Hi Mike

"Best" as judged by what parameters ? - throughput, accuracy and precision, cost, operator skill required, batch size etc etc all ahve to be taken into account.

Peter

[IrishMike]

Hi Peter

I guess best should include all of those but I guess to begin with best in this case refers to ease and reproducibility of calibration...I am leaning towards direct injection as it removes another variable which may affect analysis.

To give you a bit of background I am using combipals so can do either HS or direct...I have got good qualitative results on both however have to get calibrations going....

thanks

Mike

[Peter Apps]

HI Mike

Ok, lets look at some pros and cons:

Headspace pros: the water in the sample will not damage the column as much. Solid muck and dissolved solids in the samples will not damage the hardware.

Headspace cons: peak area for a given concentration of analyte depends on what else is in the sample, so you have to run a calibration for every matrix - a sample with a lot of dissolved solids will give bigger peak areas than one in clean water for example. Repeatability with out of the box headspace hardware is very unlikely to be better than 2 or 3%. MDQ for hydrophilic solvents e.g. ethanol wil be much higher than for hydrophobics

Direct injection (NB I am assuming that you really do mean direct injection and not split splitless) pros: within wide limits the calibration will be the same for different sample matrices. Ultimate repeatability should be below 1 %.

Direct injection cons: for samples with a lot of solvent the column and the detector will be hopelessly overloaded unless you dilute the samples. Suspended solids will block and otherwise destroy the syringe. Any non-volatile materials in the samples will crud up the inlet, which will degrade separation and cause discrimination among analytes. You will need to use a retention gap to protect the column (to some extent) from the water.

So it's swings and roundabouts. If the samples are really cruddy I'd try diluting them 1:10 with saturated salt solution and headpsacing them - they all then end up more or less in saturated salt soln and the calibrations witll be fairly uniform across sample composition.

If the samples are in clean water I would go for direct injection because the calibration for each component will be pretty much unaffected by what other solvents are present.

Peter

[IrishMike]

Thanks Peter

I guess I used the wrong terminology..

We are doing split liquid injections...

Will that alter your suggestions??


Thanks

Mike

[chromatographer1]

Given that your limits of detection are not extremely low, I assume your waste measurement is not critical if the amounts present are below 0.1% v/v .....

I would use a consistent matrix as the balance (DMF or DMAc) and spike this balance with 1% or less of the waste solution. If your waste solution is minimal if should not affect the partition of the volatile solvents to any critical degree. You should be able to achieve linearity and accuracy at a fairly high level. For example, a 10µg (0.01µL) spike of solvent in a 250µL sample of DMAc should be reproducible at ±2% or so. Now if this 10µg was a 1% of a 1000µg sample (1mg or perhaps 1µL) the accuracy by headspace will be 0.02%. And 1µL of sample of sample in a 250µL amount of DMAc will not affect the partition of the solvent from the balance to any serious degree.

So I vote to use headspace. And use direct injection from your HS analyzer with a 5cc/min carrier flow as a minimum.

best wishes,

Rod

[Peter Apps]

Hi Mike

Split injections have the advantage that you can increase the split to keep the peaks in a workable range for concentrated samples, but for aqueous samples it is extremely difficult to get consisitent injections, mainly because a water generates such a large volume of vapour when it hits the hot inlet. This is why genuine direct or on column injection is preferred for liquid injections of aqueous matrices.

What kind of accuracy and/or precision do you need for the data to be useful ?

Since you have the hardware and switching from liquid to headpsace injections is reasonably straightforward why not try both ? Please feed back to us what you find.

Peter

[AICMM]

I would go with Peter's suggestion about dilution (to make the matrix more water like) and headspace. Direct aqueous injections are hard to do, hard to reproduce and hard on the column.

Best regards.

[IrishMike]

Hi all

Thanks for all your suggestions...

I will be trying some of these over the coming days and will let you know how I get on.

Mike

[chromatographer1]

Thanks, Mike.

We are interested in learning which methods you find useful and which have advantages and why.

Good luck in your endeavors.

Rod

[IrishMike]

Hi again

I haven't forgotten your suggestions...I have been working on other things for a while so hope to get this going again this week

I'll keep you updated with any developments.

Mike

[chromatographer1]

IrishMike,

Here is an example of relatively quick chromatography using 0.53mm ID columns and headspace injection of a sample containing 0.5 micrograms of each solvent.



1-Air, 2-Methanol, 3-Diethyl Ether, 4-Ethanol, 5-Acetone, 6-Isopropanol, 7-Dichloromethane, 8-Hexane, 9-Propanol, 10-Ethyl Acetate, 11-Tetrahydrofuran, 12-Chloroform, 13-Benzene, 14-Isobutanol, 15-Trichloroethylene, 16-Dioxane, 17-Methyl Isobutyl Ketone, 18-Toluene, 19-Pyridine, 20-Dimethylacetamide

Xylenes elute before DMAc(#20) without resolution of p/m isomers. MEK coelutes with ethyl acetate under these conditions.

The balance was water w/ 4% DMAc. The measurement was linear over 3 orders of magnitude of concentration of these 18 solvents. This may give you an idea of the possibilities of headspace analysis.

best wishes,

Rodney George
Senior Research and Development Scientist
Gas Separations Research
Supelco
595 North Harrison Road
Bellefonte, PA 16823

814-359-5737 voice
814-359-5459 fax
rodney.george@sial.com

[IrishMike]

Thanks Rod

That looks good to me!

Am I right to assume the DMAC reduces the immisicibility (??) of some of the components??

Mike

[chromatographer1]

The DMAc was added to the water in order to convey the other non-polar solvents to the water. When you are adding 18 solvents to water at trace levels it is easy to have the solvents partition - evaporate out, especially at low levels. This also increased the accuracy of the spiking experiment.

The same anaysis could be made from 100% DMAc but with different partitions of the solvents (Henry's Law) at the same conditions I used for the water solution.

best wishes,

Rod

[shaun78]

If you do decide to go non-headspace route, this will need to become your friend very quickly: http://www.chem.agilent.com/cag/servsup ... s/GCFC.htm

The vapor volume portion of the program will aid you in calculating how much you can safely inject so you can avoid the reproducibility problems everyone has mentioned.