4.1 Testing of the vacuum system
The vacuum system is redesigned to offer an ability to couple with atmospheric ionizations while preserving the ultra-high vacuum inside the MRTOF chamber. We recorded the pressure change in the analyzer using a cold cathode gauge (PKR251, Pfeiffer) with respect to the pumping time at a heating temperature of 100 ℃, which corresponds to about 60 ℃ on the chamber.
The pumping process depicted in Fig. 5 is described by an exponential reduction in pressure, with a slower decreasing trend after the pumping period exceeds 9 hours. For metal vacuum chambers, the desorption of water vapor and adsorbed gases and the outgassing from seals lengthen the pump-down time 53. The process can be shortened by baking out the chambers. After long-time baking, the best pressure in the analyzer reached up to 6.8 × 10–6 Pa for our API-MRTOF-MS, corresponding to a mean free path of 2 km. The result indicates that a baking time of several days is necessary to achieve high vacuum.
Fig. 5. Experimental results of pressure in analyzer as a function of pumping time.
As we reduced the length of the DT between the FLT and the MRTOF analyzer in the design, the analyzer pressure changes must be tested with different cooling gas pressures. We regulated the pressure in the FLT by manually adjusting the intake gas valve. The pressure variation was recorded and compared with the results calculated by the method mentioned in Section 3.
As illustrated in Fig. 6, the FLT outer chamber pressure (Fig. 6A) and the DT chamber pressure (Fig. 6B) increase with the buffer gas pressure, showing a similar tendency with the calculation results. While the experimental results do not change much in comparison with the calculation results, according to the different scales of y-axes, the calculated pressures are about five to six times higher than those in the experiment. In contrast in Fig. 6C, the experimental data differs from the calculated trend. The buffer gas injection process appears to have no effect on the analyzer pressure which is mainly due to the elastomer sealing components (“O-ring”) we used. The diffusion through elastomer seals and their desorption rates caused limitations of the vacuum. These considerations will be used in future vacuum chamber designs to realize an improved vacuum. The differential vacuum system was carefully arranged with a pressure of 6.8 × 10–6 Pa.
Fig. 6. Measured pressures for different buffer-gas pressures in the FLT. Comparison of calculation and experimental results of the pressure outside the FLT (A), the pressure in the drift tube chamber (B), and the pressure in the MRTOF analyzer (C).