by John J. Manura,
Scientific Instrument Services
Electron multiplier life is dramatically effected by its operating environment. Vacuum level, pressure surges, solvent plugs and air plugs all can drastically reduce multiplier life. In our labs we have two H.P 5971 instruments, in which we have increased multiplier life to greater than 2 years. This compared to the normal operation by most users of 9 months to 1 year for electron multiplier life. We have accomplished this increased life by two means.
First, we replaced the original small rotary vacuum pumps on the instrument with larger Alcatel pumps. On one instrument we installed an Alcatel Model UM2010 (200 l/min) and on the other instrument we installed an Alcatel Model UM2015 (300 l/min). The original rough pumps supplied with the mass spectrometer had a pumping capacity of 90 l/min or less. We also replaced the vacuum pump oil with a new lower vapor pressure oil (Inland 45). Both of these changes result in better vacuum and quicker pumping out of the mass spectrometer. Since the MSD is not differentially pumped, every improvement in vacuum performance helps increase multiplier life and also filament life.
The second thing we do differently, is that nearly all of our samples are analyzed using our thermal desorption system. Therefore no solvent slugs are introduced into the mass spectrometer. If we need to inject a sample with a solvent front, we first inject it onto a Tenax® TA trap, then purge off the solvent front before desorbing the sample into the mass spectrometer. In many cases this is not practical, but it points out that solvent slugs and large sample pressure surges can reduce both multiplier as well as filament life. Solvents slugs should be kept to a minimum, and the mass spectrometer (and therefore the multiplier) should not be turned on until the solvent slug has pumped out of the system. Likewise sample overloading and the introduction of air can also reduce multiplier life.
As a result of these improvements in the vacuum system and the injection techniques that we use in our lab, we have never had a multiplier last less than 2 years. In addition we rarely burn out a filament. In most cases we need to shut down the mass spectrometer to clean the source or change columns before the filament burns out.
Yttria coated filament at start
Yttria coated filament after 16,000 cycles