Calibrating microlitre syringes (liquid < 10 ul)

[MikeD]

I am interested in canvassing opinion on the subject of calibrating low volume microlitre syringes (liquid capacity ≤10 µl). Why would anyone want to do this? Well, because when we introduce a calibration sample say 0.01 – 10 µg to the chromatograph by thermal desorption, as opposed to comparing two solutions, we are obliged somehow to deliver an absolute mass to the GC inlet. Liquid phase spiking is one method of loading the intermediate sorbent trap, but the point is no dilution solvent is involved in taking the real sample. There is a currently a debate in a laboratory intercomparison. Your options are:

A. Ignore microlitre syringe – establish traceability by gas phase spiking if you can.
B. Don't calibrate - trust the syringe manufacturer markings or certificate.
C. Follow any ISO/ASTM/EPA etc. methods – if there are any.
D. Weigh the syringe before and after spiking the (cool) intermediate sorbent trap
E. Weigh liquid into a flask sealed with a septum.
F. Other

A few restrictions before opting C-F.

Mercury as the calibrant liquid is not allowed.
The syringe calibrant liquid must be the same as the dilution solvent for spiking the sorbent trap.
The dilution solvent cannot be semivolatile. It must be capable of being purged from the sorbent trap before analysis.
The uncertainty of the mean delivered volume must be not more than 2% (95% confidence.
The repeatability of the delivered volume must be not more than 1% (95% confidence).

[Bruce Hamilton]

My first question would be:-
What are you trying to achieve. Calibration of the syringe, or calibration of your use of the syringe?

I'd want to calibrate the syringe under standard conditions ( eg temperature = 20C / test fluid=water / humidity 50% etc ) to ensure the syringe met manufacturer's specifications. Then, if necessary, I would calibrate the syringe's performance under actual operating conditions, using whatever fluids etc are relevant.

I would expect the initial acceptance calibration of a syringe to be conducted in accordance with the recommendations of the manufacturer for the type of syringe, as a plunger-in-needle syringe may have a different protocol to a plunger in barrel type. I suspect paying for a calibrated syringes would be a far cheaper option.

Most glass syringe manufacturers offer calibration certificates, and claim their methods are far more accurate than any of the traditional methods ( measuring dispensed liquid mass, dispensing into a solution and measuring the change in solution property ( eg UV absorption, fluoresence, radioactivity, etc. ).

If you want to calibrate the syringe to your application, then there could be significant additional variables. I would also start with a couple of new manufacturer-calibrated syringes, keep one as a reference, and the other as the working syringe.

I've used the mass of water dispensed to check the calibration of 10 uL syringes used for dispensing solutions or gases. It's a tedious process, as the syringes have to thoroughly cleaned, allowed to come to thermal equilibrium, etc. but at least that data relates to a reasonably-common calibration fluid. From memory it was based on an ASTM? method for microliter volumetric dispensers. It was accepted by client auditors for routine confirmation of GC syringe status, but rules could have changed.

Please keep having fun.

Bruce Hamilton

[MikeD]

Bruce

Good point about calibration of the use of the syringe. That is what is required and my comments are about the plunger in barrel type.

A positive stop adapter must be used for good repeatability and its exact position doesn't matter. You can buy a certificate but it means nothing other than proving engineering quality. The manufacturer's certificate just certifies the dimensions of the mandrel that made the syringe. It's an academic exercise to calibrate these syringes with mercury and find not surprisingly that they are spot on to better than 0.2 %. Manufacturers may claim better accuracy than "traditional" methods, but in private they don't . They know the user application is critical.

In our case the carrier solvent is normally methanol, so methanol must be used for calibration. If the spiking process involves injection through a (cool)septum about 0.1 ul methanol is lost from the needle as it penetrates, therefore the calibration must involve septum penetration.
With care the method of injecting into a preweighed 5 ml septum-sealed flask has the fewest traceability steps for our application.

[WK]

Mike D,
I wouldn't like having to calibrate syringes with mercury!
WK

[Bruce Hamilton]

A positive stop adapter must be used for good repeatability and its exact position doesn't matter. You can buy a certificate but it means nothing other than proving engineering quality. The manufacturer's certificate just certifies the dimensions of the mandrel that made the syringe.

I suspect your syringe supplier may only be providing Certificates of Conformance ( which should accompany every volumetric glass syringe ), and my understanding is that the Certificate of Calibration is specificed at time of purchase, and is testing of the actual syringe you purchase. But I could be wrong...

http://www.sge.com/uploads/vu/dB/vudBTC ... 0061-S.pdf

I'd be very wary of calibrating with methanol, the volatility issues are pretty serious. I'd probably investigate first, by sticking one septum blocking the syringe needle end, weigh, inject into weighed septum flask ( through both septa ), pull needle back into top septa and reweigh both flask and syringe.

With that information about precision and I'd either continue your process, or commit suicide.

Good luck, and please keep having fun,

Bruce Hamilton

[Peter Apps]

Hi Bruce

I have been looking at the errors that come from solvent evaporation when using microsyringes, and as you say they can be serious if you do not take the appropriate precautions.

I have been using septa to cap the syringe needle, but not the paticular technique of injecting through both septa and pulling back into the needle seal. Is this technique written up anywhere ?

Thanks Peter

[MikeD]

Bruce

SGE are indeed the makers in question and excellent products they are. Yes they would calibrate your individual syringe on request, but it is just another certification of dimensions, not the dispensed volume with your chosen liquid using it the way you do. The traceability of dispensed volume to national standards is not there, which is to a degree admitted by the qualifying statement on the certificate. It wouldn't be far out - maybe 4-5% high but enough to degrade your scores in a PT scheme for thermal desorption.

Yes methanol is a tough choice and I wish it were otherwise. It has to be purgeable from the traps at ambient temperature leaving the analyte behind. Bromoform is good but it's not purgeable and the viscosity is different, so again the traceability is dubious.

Measuring VOCs in air by thermal desorption is a specialised area and ease of calibration is its weakness. Nevertheless my favoured method is the one used by most collaborating labs in certifying the European reference materials.

Regards and not suicidal
Mike

[Bruce Hamilton]

I have been using septa to cap the syringe needle, but not the paticular technique of injecting through both septa and pulling back into the needle seal. Is this technique written up anywhere ?

Doubt it. I used that technique when I was checking gasoline injection volume repeatability from cooled ( 4C ) syringes. I was able to account for all the gasoline, and the flask and the syringe data were consistent.
I think I arrived at the technique by trial and error.

I used SGE TCS ( 3 layer - but quite firm PTFE layer ) septa, with the PTFE facing inwards on the flask, and facing towards the flask on the syringe needle. The septum on the flask was 11 or 12 mm diameter, and the syringe septum was about 6 or 7mm diameter so it sat snugly inside the flask's septum retainer.

Please keep having fun,

Bruce Hamilton

[Bruce Hamilton]

SGE are indeed the makers in question and excellent products they are. .... Nevertheless my favoured method is the one used by most collaborating labs in certifying the European reference materials.

I think I would be talking to SGE technical support in Oz, probalby by email unless your boss wants to send you there.

They should be able to advise on both their technique and traceability, and may offer detailed calibration instructions, or point to suppliers of such services.

Please keep having fun, and stay non-suicidal,

Bruce Hamilton