You are here - Home ::

Getting a vacuum

The typical mass detector ionisation systems used in LC-MS work at atmospheric pressure, but ions do not travel well through gasses. Unnecessary collisions with gas molecules cause at best fragmentation, and at worst total loss of signal. Therefore a good vacuum is essential.

As an example, the typical distance between the ion source and the actual detector in a single quadrupole machine may be about 0.5m. A pressure of about 5*10^-5 Torr gives an ion a mean free path (distance before expecting a collision) of 1m. This is therefore an appropriate sort of pressure. Unfortunately it is better than can be achieved by an average laboratory vacuum pump.

The other problem is how to interface an area at normal pressure with an area at very high vacuum while still letting the sample in. The two methods are:

A long thin hole!
A series of shorter holes!

Our HP-Agilent machine, for example, uses a combination of both. Initially the sample is drawn into the detector at right angles to the nebuliser spray (for details of this see ionisation), into a glass capillary about 15cm long. This is metal plated at the ends so voltages can be applied to accelerate ions into the machine. Any solvent droplets that haven't been dried properly, or general uncharged gunk, tends to carry on the way it was sprayed, and is diverted to waste rather than contaminating the detector.

After the capillary are two "skimmers". These are conical plates with a small hole in the middle. The gaps between them are connected to ever better stages of the vacuum system, so that drying gas and vaporised solvent are drawn away sideways, while charged ions carry on the way they were going, through the little hole. Of course voltages are applied to ensure that the electric field continues to accelerate the ions into the machine in an appropriately controlled manner...

The next stage is to find the mass of the ions.