Specialty gas calibration standards and the road to getting it right

Recently, I was reflecting on my experiences while at Pittcon several weeks ago. For folks in the analytical field, that show remains as one of the best for renewing old acquaintances. More importantly, however, it offers an opportunity to view all the latest technologies.

I’ve attended that show many times over the years, and during those visits I often liked to look at not only the instrumentation itself, but more so at the applications and technologies that support that instrumentation.

Having been involved with the analysis of specialty gases throughout my career, I always look for new instruments and technologies that will enhance the process. Two areas in particular that I always look at relate to the operation of instrumentation. The first is the sampling methods employed during analysis and second is the calibration of the instrument.

Proper Introduction Pays

You can spend a lot of money on stateof- the-art instrumentation and still not achieve accurate results if you fail to consistently introduce the sample to the instrument properly. Recent upgrades in sampling systems and automation have helped a great deal in this area. Also, new technologies permit an analyst to more easily conduct a reproducibility test to confirm whether the sampling procedure is working satisfactorily.

But what about calibration of the instrument? As with many other questions that characterize the specialty gas industry, the answer to this question can be elusive. In this case, calibration depends on what you are trying to analyze and the accuracy of the results you are expecting.

For instance, during my tour at Pittcon, I attended a symposium that addressed the issue of calibration standards, which was directed primarily to the auto industry. Representatives from NIST (National Institute of Standards and Technology) and EPA (Environmental Protection Agency) presented an update on activities covering standards for environmental emissions testing. They presented a nice update on standards they have developed and tracked stability over the last several years.

It was explained that the data and standards discussed are used primarily to produce SRMs (Standard Reference Materials) for the emissions testing community. This, however, covers only one market sector of the overall special gases requirements. In addition, SRMs are expensive and in some cases difficult to obtain.

For those laboratories that need to calibrate instruments for a full range of applications, the question is: What standards are commercially available and what kind of analytical accuracies can they expect? To help answer this question, the rest of this article is dedicated to providing a general guideline of standards that are commercially available from both specialty gas companies and NIST. Be prepared to find that the standards offered vary from company to company, despite the fact that designations given to the core standards produced are typically the same.

The key to understanding exactly what you are getting is to understand the listed blend tolerance, and more importantly the listed analytical accuracy. Table I lists a summary of the standards discussed in this article. Again, this table represents a composite of what is available. Obviously, what each company offers will vary slightly by name, blend, and analytical tolerance. You also need to understand what their guarantees are based on.

 

NIST- SRM

NIST is responsible for “ensuring accurate and compatible measurements through the development, certification, and distribution of Standard Reference Materials.” Standards produced by this body are the top of the line. While the accuracy associated with NIST mixtures is probably, in most cases, around plus or minus 0.1 to 0.2 percent relative, the listed accuracy is plus or minus one percent relative to the mixture you receive. Using an SRM to standardize your equipment is by federal regulations the best you can do. Unfortunately, these standards are available for just a small segment of the overall special gas laboratory.

The only standards that exist today are those that support continuous emission monitoring. SRM’s are also very expensive, ranging in cost from $2,000 to $3,800 for a small cylinder. Despite their high cost, even these are difficult to obtain in some cases.

EPA Protocols

EPA standards are used primarily to calibrate continuous emission monitoring systems. They are NIST traceable with a plus or minus one percent accuracy. They are produced in accordance with the latest EPA specifications found in document 600/R-97/121 Rev. 9/97. These standards may be considered as the next best thing to having an SRM to use on a daily basis.

As part of the protocol, the producer of an EPA protocol mixture must allow the mixture to sit for four days before performing the first set of triplicate analysis, then let the mixture sit for another seven days before performing a second set of triplicate analysis. This should be kept in mind when ordering an EPA protocol mixture because it will affect your leadtime. Also, keep in mind that a supplier can make an EPA protocol mixture for you only if that supplier has the corresponding SRM’s on hand or available.

Up to this point, this article has discussed standards used primarily in continuous emissions monitoring; these have regulatory documentation to back them up. But what about the calibration gases required for other specialty gas applications? In these areas, it is not as clear-cut as to what is needed to determine what standard you need and what you are getting.

No single definition can be applied throughout the special gas industry when it comes to calibration standards. However, this article describes two commonly named types of standards used throughout the industry. They are Primary Standards and Certified Standards. Table I shows some common features found within both these standards.

Primary Standards

You also may have heard these standards referred to as “Master Standards” or “Gold Standards.” Whatever the name, this set of standards to which I refer are high accuracy mixtures that are blended gravimetrically on electronic, high precision balances. There are no SRM’s with which these mixtures can be compared. NIST traceability, however, can be achieved through traceability of the NIST weights that are used to measure gravimetrically each component of the mixture. Mixtures produced using this technique are extremely accurate, but they are also subject to errors, which can be introduced by impurities in the raw material used, by poor filling techniques, and other factors. For this reason, the producer of the standards mixture should not only use the weight calculations but also run additional quality control tests to ensure that the mixture accuracy has not been compromised.

When reviewing Table I under the Primary Standards section, you will notice that at the low and high concentration ranges the accuracies shown are in units absolute instead of percent relative. The reason for this is that the error for these mixtures is due to the error associated with the weight measurement itself. The error is an absolute value that does not change. Thus, as more weight is added, the total error is decreased. Typically, the absolute error associated with making a mixture using high precision scale and NIST weights is plus or minus 0.1 gram.

Certified Standards

For analyses that do not require the high accuracy achieved with a primary standard, a certified standard is recommended. Certified standards typically have a two percent relative accuracy. Some of these mixtures can have a low ppm five percent relative, which is probably realistic. While certified standards do not provide the blend and analytical tolerance other standards do, they are usually much cheaper than other types of standards discussed in this article. In some cases, a certified standard is the only standard that can be produced.

Conclusion

The importance of selecting the right type of standard for any given task can be easily overlooked. In reality, you can have the best analytical equipment money can buy, yet produce nothing but bogus results if the standard you are using is off. One way to make sure any new standard you receive is correct is by running it directly against the standard it is replacing. If they do not correlate within the certified accuracy, then further investigation is warranted...

Mark Conroy is Special Gas & Quality Manager, GTS, Inc. He is a chemist who has been in the specialty gas business for more than 20 years. He can be reached at: Tel: 215 736 5203 or email: mark. conroy@gtsgas.com

Specialty Gas Report SECOND QUARTER 2007 //