Alternates to Helium Leak Testing: Hydrogen as a Tracer Gas
Our previous article dealt with alternates to helium leak testing, which included the use of Argon as a tracer gas. The use of hydrogen as a tracer gas has also generated much interest due to helium supply issues. Two hydrogen sensor technologies are currently in use in production leak testing applications. The first is the same magnetic sector mass spectrometer technology that is used in helium leak detectors. In most commercially available helium mass spectrometer leak detectors the option is available to tune the mass spectrometer to detect hydrogen. The second detector that has gained some popularity is a chemical sensor that is specific to hydrogen gas and is typically mounted on the end of a handheld probe for sniffing applications.
Why Consider Hydrogen?
In considering alternate tracer gases let’s revisit the following list of characteristics to consider in selecting a tracer gas for leak testing:
- Natural Ambient Concentration
- Environmental Concerns
- Atomic Mass
- Gas Viscosity
- Detectability (sensors or instruments available, and detection limits)
- Suitability of Sensor (rugged for production environments)
- Applicable Test Methods (influenced by sensors)
Both low cost and availability make hydrogen a qualified tracer gas candidate. Because of safety issues, those using hydrogen as a tracer gas are using a mixture of 5% hydrogen in 95% nitrogen, called Forming Gas. This mixture is considered safe and non-flammable. However, this creates a dilemma when comparing using forming gas to other tracer gases in 100% concentration. Simply stated, when using this diluted gas mixture the concentration of gas escaping the leak will be only 5% of what it would be if the concentration was 100%. Thus, the reject limit for the specific application would be reduced by the same amount. For example, a reject limit of 1 x 10-5 atmcc/sec for 100% tracer gas would be 5 x 10-7 atmcc/sec for the same gas at 5% concentration.
Hydrogen gas has both a lower atomic mass and gas viscosity compared to helium giving it a slight advantage over helium. Hydrogen also has a lower natural concentration (0.5 ppm) in ambient air compared to helium (5 ppm). However, there are many sources of hydrogen that need to be considered that would increase the natural hydrogen concentration and have a significant impact on the sensitivity of an instrument. This can be a challenging problem in a mass spectrometer leak detector where background levels of hydrogen emitting from hot metals and residual water vapor can cause background levels in the mass spectrometer to be as much as 100,000 times higher compared to helium. This, perhaps, is the single biggest factor that limits the use of a mass spectrometer as a sensor in hydrogen applications.
Finally, the design of the sensors and how the sensors can be implemented into a production leak testing application have a significant impact on how hydrogen can be used as a tracer gas.
Test methods using Hydrogen
Manual sniffing is the most common method where hydrogen has been recently implemented as a tracer gas. Inficon’s Sensistor leak detector claims a leak rate sensitivity of 1 x 10-7 atmcc/sec with 5% hydrogen tracer gas. This is close to the sensitivity of what is claimed for a helium mass spectrometer leak detector in sniffing mode. Real life use, however, shows that a more practical leak rate limit for this instrument is closer to 1 x 10-5 atmcc/sec.
Helium mass spectrometer leak detectors tuned to hydrogen can also be used in sniffing applications. However, they suffer from the hydrogen background limitations described above. These hydrogen background levels may vary significantly over a range of different manufacturers making it difficult to state the true sensitivity in sniffing mode.
The chamber accumulation method is one way to perform more automated and global testing with hydrogen. In this method a sniffer probe is used to monitor rising concentrations of the leaking tracer gas that is collected in a surrounding chamber. Proper integration of the hydrogen sensor or sniffer is critical to the success of this method. Nevertheless, the accumulation method suffers from lack of sensitivity, particularly for larger test objects and when fast cycle times are required. Typical sensitivities are usually no better than 1 x 10-3 or 1 x 10-4 atmcc/sec. Hydrogen accumulation is fairly close in leak rate sensitivity compared to helium accumulation. However, when compared to the more common helium hard vacuum technique, the accumulation method is much less sensitive, less repeatable, and always requires longer test times.
The traditional helium hard vacuum method can also be adapted to hydrogen tracer gas for a typical helium mass spectrometer leak detector equipped with the option. However, as mentioned above, in traditional mass spectrometers the hydrogen background can be very high. For this reason it is uncommon to implement hydrogen tracer gas leak testing in the hard vacuum chamber mode. Typical sensitivities are no better than 1 x 10-5 atmcc/sec.
While hydrogen appears to have some of the ideal properties of a tracer gas, particularly compared to other alternate choices such as argon, there are significant challenges when implementing hydrogen. The use of hydrogen has seen modest success in sniffing applications, but due to sensor and hydrogen background issues, it is not an equivalent replacement for helium in the chamber test methods that are commonly used in production applications.