In this article we will go over the different terms as they relate to the overall accuracy of a gas mixture.
Firstly, understand that “accuracy” is a general term which is actually made up of two components: Preparation Tolerance and Measurement Uncertainty
Preparation Tolerance or Blend Tolerance
This describes how closely the manufacturer can match the customer’s requested values, based on their manufacturing technique. Think of this like throwing a dart at a dart board. You’re aiming for the centre, but you may not always hit it. But you know where you did hit, because you can check. Some manufacturing methods make it easier to get close to the centre compared to others – a good example is volumetric manufacture vs gravimetric manufacture.
A quality calibration gas’ preparation tolerance will not have a negative effect on an accurate instrument calibration, as the values will be close to the requested values. As long as you know what the actual value of each component is then you can then set the instrument to accept the certified prepared values. If the value falls far from the requested value, you are dealing with a poor quality product and the calibrated instrument’s accuracy and performance can be affected.
Measurement Uncertainty or Analytical Accuracy
These terms describe the accuracy of the tools you used to measure the result you think you obtained. In our dart board analogy, let’s say you used a ruler to measure the distance from the centre of the dart board to where the dart landed. You determine it is 12mm based on what the ruler says – but how accurate is the ruler? Possibly the ruler is +/- 0.5mm from the true value, so the dart may actually be 11.5 to 12.5mm from the centre. This is your measurement uncertainty.
But this does not tell the entire story. Sometimes you need to dig a little deeper when you see these terms. You need to ask what type of measurement is being made. If you are comparing two gas mixtures, you need to make sure you’re comparing the same type of measurement.
There are two types of measurements made when preparing gas mixtures. The first is measuring what goes into the cylinder (volumetric or gravimetric measurement), and the second is measuring what comes out of the cylinder (measurement based on analysis). The first type is relatively simple, as you need only to have;
- In the case of volumetric preparation: a calibrated pressure or flow device, OR
- In the case of gravimetric preparation: a calibrated balance.
The second type, analysis-based measurement, is much more involved. It requires not only an expensive gas analyser, but also several other reference gas mixtures. The type of analyser used, operator skill, quality of reference gas mixtures, and accreditation of the whole process can each have an effect on the overall quality of the gas mixture. Analytical results could be referred to as either measurement uncertainty, or analytical accuracy. If your gas mixture certificate doesn’t give specific information as to the methods used, ask the manufacturer for all the relevant information, or you won’t be able to make an accurate comparison.
The best quality gas mixtures will be manufactured gravimetrically, and then analysed with an accredited process, all clearly outlined on the cylinder certificate.
“Accuracy” in Non-Refillable Cylinders
Mixtures in NRCs are offered stating only “accuracy,” which is a composite figure based on the product of the two components above: preparation tolerance and measurement uncertainty. Because all NRCs are trans-filled (that is, they are decanted from a larger cylinder), they don’t have any direct measurements made against them.
The best total “accuracy” available in an NRC is +/- 2%, it really can’t get any better than that without doing a direct measurement. Furthermore, preparation tolerance or measurement uncertainly can’t be provided individually. That can only be done where the cylinder being certified is the cylinder in which the mixture was prepared (and not trans-filled).
“Accuracy” in High-Pressure Cylinders
Accuracy in high pressure or refillable cylinders is much more clearly defined. Often there can be several classes available, depending on the mixture type.
Preparation tolerance varies according to component amounts, but in terms of measurement uncertainty, there are generally four classes:
Certified Gravimetric Standards
- Gravimetric manufacture
- Measurement Uncertainty +/- 2.0% relative or better
These standards are everyday working standards. With reliable gravimetric measurements, these are used in many different applications from calibration of safety devices to laboratory analysers.
Reference Gravimetric Standards
- Gravimetric manufacture
- Measurement Uncertainty +/- 1.0% relative or better
Reference Gravimetric Standards provide another level of gravimetric accuracy using high-precision balances.
- Analytical measurement, unaccredited (uncertainties vary)
- Measurement Uncertainties vary
Where the application demands analytical data on reference materials, analytical standards are used. These standards are produced gravimetrically but also provide the user with analytical information about each specific gas mixture.
Accredited Analytical Standards
- Analytical measurement, accredited (ISO17025)
- Measurement Uncertainties vary
- Secondary Reference Gas Mixture
- Primary Reference Gas Mixture
Accredited analytical standards provide the lowest uncertainty and highest quality measurement data, authenticated by ISO17025 accreditation. Available in either Primary Reference Gas Mixtures – (true primary reference materials, measured relative to Primary Standards held at national metrology institutes), or Secondary Reference Gas Mixtures (measured relative to Primary Reference Gas Mixtures).
Determining which level of “accuracy” you require is dependent upon your application, measurement criteria, and what your objectives are in your measurement plan. With so many variation’s of gas mixtures available it is best to discuss your requirements with a specialist in calibration gas mixtures to you ensure get the accuracy you require.