AVS 71 Session VT2-TuM: Measurement, Simulations and Accelerator Vacuum Systems

The increasing complexity of industrial vacuum processes requires broader and deeper knowledge of the vacuum itself. A crucial aspect for increasing quality demands is the necessity of in-situ monitoring and control of pressure and residual gas composition within vacuum processes. A consequence of advanced process control is the reduction of production errors, prevention of failures or major damage in combination with increased operating time. Traditional monitoring devices like hot cathodes or quadrupole mass spectrometers are both only able to measure either pressure or residual gas composition. Therefore, these devices are only conditionally suited for complete process control of vacuum processes. With our novel wide-range vacuum monitor NOVION®, which combines the well-known technology of time-of-flight spectroscopy with our patented ion trap, industrially available pressure and gas analyzation is possible at the same time.

In this talk we present the fundamental principles of the novel vacuum monitor and explain the compact combination of well-known time-of-flight spectroscopy with our own patented ion trap. Within different application cases we discuss advantages and limits of this technology and demonstrate with one single device wide range gas analysis, simultaneous measurement of total and partial pressures, leak detection for Helium and detection of air leaks. With these combined capabilities the novel vacuum monitor is able to quickly capture the complete pressure and gas composition measurement at various stages of the vacuum process chain. In addition, we demonstrate a special signal enhancement method to improve the resolution in the near signal-to-noise range.

NIST is currently constructing a portable robust Fixed Length Optical Cavity (FLOC) pressure standard to be optimized for the calibration of aircraft altimeters, rate of climb indicators, and air speed indicators while also extending the operating pressure range close to 10 MPa.To reduce cost and help simplify the operation of the system we have tested an optical approach that only uses a single laser locked to the reference cavity with a portion of that light being modulated to generate a sideband which is locked to the sample cavity.The tradeoffs and advantages of this technique will be discussed.

Thermal transpiration, also known as the thermomolecular effect, applies when a pressure gauge at temperature T₂ is used to measure the pressure of a gas held at temperature T₁. Examples include gas thermometry (say T₁ = 10 K) and temperature-controlled gauges (say T₂ = 318 K). When the temperature difference is large and the gas mean free path is comparable to the diameter of the tube connecting the two volumes, thermal transpiration can make the pressure ratio P₁/P₂ much less than 1.

Thermal transpiration has been described by physically motivated empirical functions, physics-based numerical models, and a physics-based analytical model. The most common empirical function is that of Takaishi and Sensui (T-S) [1]. Numerical models are rarely used because they rely on details of geometry and surface accommodation that restrict the model’s use to a specific scenario.

There is only one physics-based analytical model, the “dusty gas” model [2], which employs the concept of a gas composed of infinitely heavy “dust” molecules. The dust molecules scatter the ordinary gas molecules, so that the flow in the connecting tube has a viscous component and an opposing rarified-gas component. The dusty gas model was used during the 1960s and 1970s to describe experimental measurements, most notably by Malinauskas and co-workers. Despite that success, it has not been widely used because the model’s core equation requires a numerical solution.

The dusty gas model is superior to the T-S empirical function. The T-S function assumes perfect surface accommodation, while the dusty-gas model does not. Also, the T-S function has three free parameters of obscure meaning, and fitting those parameters to experimental data can hide an error in the data. In contrast, the dusty gas model has only two free parameters with clear physical meaning. The first parameter accounts for imperfect accommodation, and the second accounts for an error in the ratio λ/d, where λ is the mean free path and d is the tube diameter. A re-analysis of literature data found good agreement with the dusty gas model.

1. T. Takaishi, Y. Sensui, Trans. Faraday Soc. 59, 2503-2514 (1963).
2. A.P. Malinauskas, J.W. Gooch, B.K. Annis, R.E. Fuson, J. Chem. Phys. 53, 1317-1324 (1970).