It is quite unlikely that you will ever use an absolute method for gas detection. Rather, you will employ any one of dozens of “relative” [or “reference,” but not necessarily EPA Reference] methods—that is, methods that use a sensor to produce some electrical output that must be calibrated against a known standard. Then, its display can be directly read out in units of concentration, usually parts-per-million (ppm).
Even though proper calibration is 90% of successful gas detection, it is a subject that has been neglected—often purposely—by far too many gas detection instrument manufacturers. There’s a good reason for this, of course: Proper calibration can often be difficult and expensive. But, we’re getting a bit ahead of ourselves.
Gas Blends in Cylinders
Early occupational health toxic gas detection focused on carbon monoxide (CO) and hydrogen sulfide (H2S). The calibration standards were supplied as gas blends in cylinders, and in the case of CO, at least, things worked out pretty well. This is because CO is not very reactive, and, within reason, maintains a stable concentration in the cylinder, as the pressure drops with use.
On the other hand, H2S is very reactive, and the original simplistic techniques used to create the cylinder gas blends could not provide a stable product. The problems observed with H2S blends were soon seen in blends for many other toxics. To make matters worse, improper analogies were drawn between experiences in combustible gas detection and toxic gas detection, establishing a false sense of security about poorly prepared gas blends.
In fact, other than the obvious point that both combustible and toxic gas detection get involved with detecting gases, the two fields of endeavor could not be more different.
The combustible gases of interest are nearly all stable (unless they are ignited by some external source), while nearly all toxic gases are unstable, and in many cases are extremely reactive. Most importantly, though, combustible gas detection is done in percent level concentrations, while toxic gas detection is done in parts-per-million, and even parts-per-billion concentrations—10,000 and 10 million times lower, respectively!
Fortunately, calibration gas blending technology has improved, encompassing specialized techniques for passivating the cylinders, as well as logging experience to determine how long a blend must age to become stable, and how long stability can be guaranteed. Much of the technological development has been done with aluminum cylinders, since this material seems to be less prone to wall effects and unwanted chemical reactions than steel.
Interscan can recommend good gas blend suppliers, but no matter what company you choose, the following points are important:
- Order the blend so that the concentration is about 50% of the instrument’s measuring range.
- Ensure that the blend’s analysis is ± 2% accurate (or better).
- Insist on NIST-traceability.
- Obtain a written guarantee as to how long the blend will be stable.
- Since most of the cost of the blend is in the analysis labor, order the largest cylinder you can use. Stay away from disposable cylinders, which just become a solid waste problem. After all, we ARE in the environmental business!
- Before you order, ask for references for the exact blend, or one that is similar, and check them.
Some material courtesy of VICI Metronics
Certain toxic gases are not well-suited to being stored in cylinders, and cylinder blends are cumbersome to re-standardize, in that a separate (usually wet chemical) analytical method is required. In addition, some instrument users need a source for several different calibration standards. These situations call for permeation devices.
Permeation devices are small, inert capsules containing a pure chemical compound in a two-phase equilibrium between its gas phase and its liquid or solid phase. At a constant temperature, the device emits the compound through its permeable portion at a constant rate. This rate can always be determined via differential weighing at constant temperature. Permeation devices are typically inserted into a carrier flow to generate test atmospheres for calibrating gas analyzer systems.
These devices are discussed in some detail in other Knowledge Base articles (here and here). Typical applications for Interscan analyzers include calibration for bromine, chlorine, formaldehyde, hydrazine, hydrogen bromide, and hydrogen chloride. Many Interscan customers who do not wish to perform their own permeation device calibration—although it is recommended to calibrate on site if at all feasible—can take advantage of our SENSOR EXPRESS® program.
Note that having calibration facilities on site provides the best possible answer to the question “How do I know that this monitoring system actually works?” You can challenge the system with a known concentration of gas at any time.
On-Demand Calibration Gas Generators
In certain cases, an electrolytic cell can be arranged, whereby a liquid containing the generating source is subjected to an electrical current. As such, a known amount of gas is generated on demand, and can be blended with air to create a calibration standard. This technique is available for several substances, and by design can be quite economical.
As you can imagine, if your measurement range is in the low ppm (or even more sensitive), accurately zeroing the instrument is of vital importance. Consider that it is not a trivial matter to remove contaminants such as carbon monoxide from air below tenths of a ppm.
Zero air can be obtained from the same vendors who manufacture gas blends. We would make the following recommendations:
- Tell your supplier your target gas and measuring range, and have him suggest the proper zero gas for your application.
- Ask for a written analysis of the zero gas. Ideally, there will be specific information and not just a series of “less thans.”
- As we noted for your calibration gas, before you order, ask for references for applications as close as possible to your own, and check them.
There are compounds that will present challenges. Hydrazine, for example, done correctly, requires an expensive and elaborate set-up, and a skilled operator. Chlorine dioxide is unstable, and although in situ calibration methods are available, great care is required to produce accurate results. An updated and promising in situ method utilizes an electrochemical gas generator.
Other hard cases include hydrogen peroxide and peracetic acid, in that commercial standards simply do not exist. Rather, standards have to be generated in a lab environment, back-titrated against wet chemistry.
How Frequently Should You Calibrate?
In general, the lower your measuring range, and the greater accuracy you desire, then the more frequently you should calibrate. Calibration monthly is a good median recommendation, and bi-monthly is even better. When we say “calibration,” we mean a good patient effort, that allows for sensor and instrument stabilization, to get a good, solid, reproducible reading. So-called bump tests, that challenge the instrument with some unknown, but high concentration of gas prove little, and can often be misleading. For the most part, these are NOT recommended.
In certain cases, less frequent calibration will still afford satisfactory results. Feel free to discuss this at any time with our service department.
The bad news is that calibration for some chemicals can be difficult, yet it is essential for proper gas detection. The good news is that we are here to help.
Please contact our service department with any calibration inquiries.