Molecular Flow Range Research Articles

Numerical investigation of three turbomolecular pump models in the free molecular flow range

Selecting a suitable model is important for the quick and accurate calculation of the pumping characteristics of a turbomolecular pump (TMP). Three different calculation models (2D, ideal 3D, and real 3D models) in the free molecular flow range were investigated in this study through the Monte Carlo method. Results show that when blade velocity ratio C ≤ 1, the simple ideal 3D model with paralleled blades is a better substitute for the complicated real 3D model compared with the 2D model. When C ≥ 2, maximum compression ratio Kmax of the TMP calculated with the 2D or ideal 3D models tends to saturate as C increases. However, the results computed with the real 3D model increase exponentially as C increases. Further investigation shows that in the rotating reference frame, molecules move toward the tip wall as a result of Coriolis and centrifugal acceleration, resulting in a significantly large Kmax and perfect pumping speed.

Vacuum gauge calibration at the NPL (India) using orifice flow method

This paper describes the primary high vacuum standard established at the National Physical Laboratory (New Delhi) for calibration of gauges in the molecular flow range. The system, based on the orifice flow method, employs a fixed orifice of 8 l\\s argon conductance. A diffusion pump—Ti sublimation pump combination gives base pressure of 10 −8 Pa. The gas from the flow meter enters the calibration sphere through a capillary leak with a flow rate of 10 −10 to 10 −8 mol\\s for reservoir pressures of a few Pa to 10 6 Pa, leading to calibration pressures down to 2×10 −5 Pa. The flowrate is measured by monitoring the time rate of rise of pressure using a spinning rotor gauge. Pressures below 10 −5 Pa are generated by flow division technique, in which the gas is diverted in the lower chamber, leading to the gas being introduced in the calibration chamber at a flow rate reduced by a factor equal to the flow ratio of the system. Results on the determination of flow ratio for nitrogen are presented. Also discussed are the uncertainties associated with the measurement of the flow ratio and the generated pressure.

A numerical study of the performance of a turbomolecular pump

In the free molecular flow range, the pumping performance of a turbomolecular pump has been predicted for various values of the dimensionless blade velocity ratioC by employing the integral method and the test particle Monte-Carlo (MC) method in the calculation of the transmission probability. Also, new approximate method combining the double row solutions, so called double-approximation (DA), is presented here. The calculated values of transmission probability for the single row agree quantitatively with the previous known numerical results. For a six-row pump, the MC method is employed to calculate the overall transmission probability for the entire set of blade rows. When the results of the approximate method combining the single row solutions are compared with those of the MC method atC=0.4, the previously known approximate method overestimates as much as 34% than does the MC method. But, the new DA method gives more accurate results within 10% of relative error with respect to the MC method.

A numerical study of the performance of a turbomolecular pump

In the free molecular flow range, the pumping performance of a turbomolecular pump has been predicted by calculation of the transmission probability which employs the integral method and the test particle Monte-Carlo method. Also, new approximate method combining the double stage solutions, so called double-approximation, is presented here. The calculated values of transmission probability for the single stage agree quantitatively with the previous known numerical results. For a six-stage pump, the Monte-Carlo method is employed to calculate the overall transmission probability for the entire set of blade rows. When the results of the approximate method combining the single stage solutions are compared with those of the Monte-Carlo method at dimensionless blade velocity ratio C=0.4, the previous known approximate method overestimates as much as 34% than does the Monte-Carlo method. But, the new approximate method gives more accurate results, whose relative error is 10% compared to the Monte-Carlo method, than does the previous approximate method.

Conceptual study of the cryocascade for pumping, separation and recycling of ITER torus exhaust

For compliance with the pump requirements of ITER operation a concept of a universal pumping system has been studied. Compact pump trains for installation into pumping ducts are proposed consisting of primary and secondary pumps. The primary pump to be applied in the molecular flow range has a required pump entrance surface of 0.24 m 2 . To maintain divertor pressures, adjustable throttle valves are planned to be installed in front of the primary pumps. Separation of the gas species takes place in the secondary pumps which are operated in the viscous flow range. The impurities and fuel (D-T) with protium and helium are pumped subsequently into three individual chambers which can be isolated from each other during regeneration. For the secondary pump a diameter of 220 mm and a length of less than 1 m are required.

Novel gas target for use in laser harmonic generation

A novel gas target was developed for use in laser-atom harmonic-generation experiments. It operates on the principles of free molecular flow rather than those of fluid flow as does a gas jet. This allows much lower gas pressures to be achieved. The device is a small, cylindrical, double-ended hole through which the focused laser beam passes. Gas enters from the middle of the cylindrical wall and quickly disperses as it emerges out the hole openings. As with the gas jet, the purpose of the gas target is to provide a very narrow and localized distribution of gas for laser interaction. Some of the advantages of the target relative to the jet are that it is easier to systematically align to the laser beam and it can operate at much lower pressures. Results of a Monte Carlo simulation of the gas density and flow rate within the target and experimental measurements of the density profile made by imaging the recombination emission are presented herein. It is found that the gas density decreases rapidly once outside the target hole openings along the symmetry axis. The calculations are in good agreement with experimental measurements. The gas target continues to function without serious deviation at pressures up to ten times above the molecular flow range.

Computer solution to chamber‐gas fractionation during pumpdown

The evacuation process of a gas mixture in the molecular flow range is characterized by the fact that each gas type is separately pumped at its own respective pumping speed. Since the pumping speeds for most high‐vacuum pumps depend on the individual gas type and, further, since the molecular gas‐flow conductance is dependent on the inverse square‐root‐of‐mass behavior, gas fractionation occurs. This merely means that the gases which have higher pumping speed are removed faster leaving the gases with lower speed as the predominant residuals. By far, the most common occurrence of gas fractionation is during the evacuation of almost every high‐vacuum system. While gas fractionation is most predominant in the molecular flow range, it also occurs in the transition range where the gas‐flow behavior is mathematically more complex. The net result of gas fractionation is that the relative composition of the residual gas in the vacuum system steadily changes with time during pumpdown. Since the mathematical aspects of gas fractionation are well established, it seems reasonable that this phenomenon should be simulated in modern computer programs which otherwise simulate the normal vacuum system behavior of gas flow as a function of pressure and of system pressure as a function of time. This article discusses some of the complex interacting problems as well as a final solution which was applied to a specific commercial vacuum‐system‐design computer program, VSD‐II.

Study of the performation of turbomolecular pumps. 1st report. The performance of finite-height blades.

It is important to calculate accurately the pumping performance of the turbomolecular pump for the purpose of optimising the blade design. In this paper, the pumping performance of the finite-height blade with tip clearance is calculated accurately by the Monte Carlo method in the molecular flow range. To simplify this calculation, velocity components of molecules are given by random numbers. The influence of the two parameters, blade height ratio and tip clearance ratio, is investigated in several blades which are selected as optimum combinations of blade pitch ratio and blade angle. Consequently, the influence of these parameters on the maximum compression ratio is notable on the blade which has small blade pitch ratio and small blade angle. Finally, the results calculated by the present method agree well with the experimental results.

Measurement of conductances of models in the molecular and viscous flow regime

Often it is not possible to calculate the conductances of complex shaped objects analytically. Even numerical calculations can be quite difficult. Examples are valves, special kinds of baffles, pumping channels and filtering devices. In order to optimize these objects it is necessary to know the influence of their geometry on the conductance. Besides the statistical calculation of the particle path under conditions of molecular flow using Monte Carlo Methods, measurements on models in the molecular and viscous flow range give a good method to determine the conductances of real objects. A PC-controlled apparatus, able to measure the conductances of objects up to an inner diameter of about 63 mm has been set up. By variation of the gas flow over more than three decades the conductance of the object can be measured in a large pressure range. The results on tubes, slits and valves are presented in this paper.

Use of porous plugs in the transition flow range for the calibration of vacuum gauges

The use of a porous plug for admitting gases in the molecular flow range is well known. In this paper, the authors deal with the transition flow region, for which the pressure of gas leaking through the plug is expressible in terms of a polynomial of nth degree in time unlike the simple exponential relation for the molecular flow region. The conductance of the plug in the transition flow range, determined from the time derivative of this polynomial, varies linearly with pressure. The slopes of the conductance versus pressure curves for helium, nitrogen, and argon when multiplied by respective viscosities are found to be constant for a plug, as expected from theory. The flowmeter thus established has been used for the calibration of vacuum gauges using an orifice flow system both in the constant volume and constant pressure modes.

The pumping character of the magnetic suspension-type turbomolecular pump of high flow rate

The pumping characteristics of the suspension-type turbomolecular pump with a high flow rate, which is operable within a pressure range of 102 Pa or below required for the semiconductor production system (physical vapor deposition, chemical vapor deposition, etc.), are described below. Conventional turbomolecular pumps are available with the operation pressure only up to a molecular flow range of 10−8 to 10−1 Pa. After trial manufacture of various rotor profiles, the rotor which would make the pump operable up to a transition flow range of 102 Pa was reviewed in this experiment. As a result of the experiment, a rotor which could achieve a flow rate of 2900 standard cm3/min (sccm) (6×102 Pa) at a pumping speed of 600 1/s (N2: 10−1 Pa or less) could be obtained. It was also found that the flow rate characteristics differed according to the rotor profile and working pressure range. Also reviewed in this paper is the surface treatment method for protection of the rotor (Al alloy) from reactive gases (Cl, F, and Ga) used in the semiconductor production systems.

5045. Performance of mechanical vacuum pumps in the molecular flow range: M H Hablanian,J Vac Sci Technol,19 (2), 1981, 250–252

5045. Performance of mechanical vacuum pumps in the molecular flow range: M H Hablanian,J Vac Sci Technol,19 (2), 1981, 250–252

Performance of mechanical vacuum pumps in the molecular flow range

An important and remarkable property of oil-sealed rotary vane mechanical vacuum pumps has been demonstrated. Such pumps retain their positive displacement pumping mechanism at partial pressures approaching the ultrahigh vacuum region. Pumping speed measurements for helium, based on partial pressure, have been made at an inlet pressure as low as 10−6 Pa with an ultimate partial pressure of helium near 10−8 Pa. Experiments conducted with a nominal 3 l/s two-stage pump (Varian 200) demonstrate that a constant pumping speed is maintained at very low helium concentrations as long as the presence of helium at the discharge of the pump is minimal, i.e., as long as the maximum compression ratio for helium is not exceeded. Similar results may be expected with other gases except the pump oil vapor or species outgassing from the pump.

A miniaturized 16 I s −1 turbomolecular pump for space application

A miniaturized turbomolecular pump for space application has been developed under a NASA contract. This pump for instance could be used to evacuate a mass-spectrometer on board of a space craft. The most important data of the turbomolecular pump are: pumping speeds for N2: 16 I s−1, He: 20 I s−1; compressions for H2: 95000; cooling by air; weight of the pump: 1450 g; overall length: 19.5 cm; diameter: 7.5 cm; power consumption: approximately 13 W in the molecular flow range.

The Transmission of a Sampling Orifice for Charged and Excited Particles in the Molecular Flow Range

AbstractThe transmission of a circular orifice in the molecular flow range is calculated for the case that the particular state of particles flowing through the orifice is affected by wall collisions. The deactivation of metastable atoms and the neutralisation of charged particles at the orifice tube wall is taken into account.The results give a detailed illustration of the fate of particles passing an extraction orifice, namely the fraction of particles passing without wall collision, the fraction passing after one and after multiple wall reflections as a function of the characteristic dimensions of the orifice and the reflection coefficient.

A survey of flow at low pressures

The theory of internal flows at such low pressures that the Navier-Stokes equations are not valid is reviewed. In the extreme low pressure range (free molecular flow), where the flow rate is independent of pressure, the basic theory is well developed. However, the influence of channel lenght, cross sectional shape, or surface characteristics have not been totally defined. Slip theory, which extends the low pressure range of the Navier-Stokes equation by modifying the wall boundary conditions, is discussed. The theory provides a good basis for experimental data correlation, but is not adequate to extend the viscous equations into the free molecular range. The empirical methods used to provide a smooth transition from slip to free molecular flow are reviewed, and a method of obtaining an equation for the total flow regime is illustrated. The extensive work in physical gas dynamic is discussed. Boltzmann's equation which has been solved numerically for small pressure gradients was found to be valid only in the viscous regime near the free molecular flow range.

A Clean Cryo-Vacuum System with High Pumping Speeds for All Gas Species

Calibration of the Harvard College Observatory spectrometers that are to fly aboard the Apollo telescope mount (ATM) required a completely clean vacuum system having the capability for pumping hydrogen, helium, argon, and other inert gases at high speeds between 10−6 and 10−8 Torr. The spectrometer performance would be degraded if contaminated by exposure to hydrocarbons, electromagnetic radiation, or metallic films. For this reason, a cryogenically pumped system was selected. This paper describes the concept, development, and operation of the vacuum system which was both cryo-condensation and cryosorption pumped in the molecular flow range. Molecular sieve, bonded to liquid helium-cooled panels, provided the clean, fast, and economical pumping device. Data presented show pumping speeds measured, adsorbent capacity of molecular sieve, liquid helium consumption rates, and cryogenic performance of the system. This data permits analysis of the cryo-vacuum pumping technique for other applications.

1237. Scaling laws for experimental determination of pumpdown time in the molecular flow range: G Lewin, J Vac Sci Technol, 5 (3), May–June 1968, 71–74

1237. Scaling laws for experimental determination of pumpdown time in the molecular flow range: G Lewin, J Vac Sci Technol, 5 (3), May–June 1968, 71–74

Gas flow in vacuum system with local temperature variations for the molecular flow range

Gas flow in vacuum system with local temperature variations for the molecular flow range

1238. Gas flow in vacuum systems with local temperature variations for the molecular flow range: G Lewin, J Vac Sci Technol, 5 (3), May–June 1968, 75–78

1238. Gas flow in vacuum systems with local temperature variations for the molecular flow range: G Lewin, J Vac Sci Technol, 5 (3), May–June 1968, 75–78