AVS1997 Session VT-WeM: Pumping, Flow Rate, and Outgassing Measurements

Wednesday, October 22, 1997 8:20 AM in Room N

Wednesday Morning

Start	Invited?	Item
8:20 AM		VT-WeM-1 Calculated Adsorption Isotherms of Hydrogen for Technical Surfaces at Various Temperatures¹ G.R. Myneni, C. Dong (Thomas Jefferson National Accelerator Facility) Practical adsorption isotherms of hydrogen for various technical surfaces (metals and metal oxides), at 4.35 K and in the pressure range 10^-11 to 10^-7 Pa, have been reported previously. ¹ It would be very useful for the accelerator community to know the adsorption characteristics of hydrogen on various technical surfaces in the temperature ragne 4.3 - 40 K. On the basis of Polanyi Potential Theory², which states that the adsorption space is independent of temperature and is enclosed by an equipotential surface E_i = kT 1n (P₀/P_i) (where E_i is the adsorption potential, P₀ is the saturation vapor pressure and P_i is the equilibrium pressure at the given temperature), the adsorption isotherms of hydrogen are calculated in the above temperature range and are presented in this paper. These calculations are expected to lead to a better understanding of the behavior of hydrogen in cryogenic ultra high vacuum systems with temperature fluctuations. ¹This work is carried out under D.O.E. Contract No. DE-AC05-84ER40150 ²M.G. Rao, P. Kneisel, and J. Susta, Cryosorption Pumping of H₂ and He with Metals and Metal Oxides at 4.3 K, Cryogenics ICEC Supplement 34, 377 (1994). ³Polanyi, M., Vern Dtsch Phys. Ges. 16 (1914) 1012.
8:40 AM		VT-WeM-2 Experimental Test of the Propagation of a He Pressure Front in a Long, Cryogenically Cooled Tube E. Wallen (CERN, Switzerland) A series of measurements on the propagation speed of a He pressure front in a cryogenically cooled 75.3 m long, 4.3 cm inner diameter stainless steel tube at 1.9 K and 4.25 K and a theoretical model¹ for the phenomenon of limited propagation speed of such a He pressure front are described. The measurements have shown that it took 20 h for a He leak of 5.7·10^-5 Torrl/s to be detected at the other end of the 75.3 m tube at 1.9 K and if the leak was increased to 1.4·10^-4 Torrl/s, still at 1.9 K, the time for the He to be detected at the distant end was reduced to 8.7 h. With the tube at 4.25 K, it took 5.0 h for a He leak of 7.0·10^-5 Torrl/s to be detected at the other end. The He travels through the tube in the form of a pressure front with a steep leading pressure gradient. A model involving the adsorption isotherm for He on stainless steel which is able to predict the time of arrival of the He pressure front at the distant end within 17 % is described. ¹J.P. Hobson and K.M. Welch, J. Vac. Sci. Technol., A11, 1566 (1993)
9:00 AM		VT-WeM-3 Measurements of the Helium Propagation in a 480 m Long Vacuum Tube Cooled at 4.2 K H.C. Hseuh (Brookhaven National Laboratory); E. Wallen (MAX-lab, Sweden) The Relativistic Heavy Ion Collider (RHIC), with two rings of 3.8 km in circumference, uses superconducting magnets operated at ~ 4.2 K to focus the high energy beams. Each sextant of RHIC will have continuous cryostats and cold beam pipes up to 480 m in length. The vacuum in the cold beam pipe will be ≤ 10^-11 Torr for acceptable lifetime of the stored beam. Due to the high vapor pressure of He at 4.2 K even with a small surface coverage, the characteristics of the propagation of a He pressure front due to He leaks will be of importance for beam lifetime and for vacuum monitoring. The travel of the He pressure fronts along the 480 m long, 6.9 cm I.D. stainless steel beam tube has recently been measured during the RHIC first sextant test. The experiment was carried out over a nine-day period by bleeding in a calibrated He leak of 3x10^-5 Torr.l/sec. The helium pressures along this 480 m cold tube were measured at 30 m intervals. The results are summarized and compared with those of a simulation¹ and with those of a recent study.² ^ext J.P. Hobson and K.M. Welch, J. Vac. Sci. Technol., A11, 1566 (1993) E. Wallen, J. Vac. Sci. Technol., A15 (1997) (in press).
9:20 AM		VT-WeM-4 Performance of the RHIC First Sextant Vacuum Systems R. Davis, H.C. Hseuh, D. Pate, L.A. Smart, R.J. Todd, N. Weiss (Brookhaven National Laboratory) The Relativistic Heavy Ion Collider(RHIC), currently under construction at Brookhaven, comprises two interweaving rings ~ 3.8 km in circumference. Superconducting magnets operated at ~ 4.2 K are used to bend and focus the highly relativistic beams. RHIC has three distinct vacuum systems, the ambient (warm bore) beam vacuum, the cold bore beam vacuum inside the superconducting magnets and the insulating vacuum housing for the cold bore and the superconducting magnets. Each RHIC sextant will have continuous cold bore pipes and insulating vacuum cryostats up to 480 m in length. All three vacuum systems have recently been commissioned during RHIC first sextant test. This paper presents the design and layout of the first sextant vacuum systems including: hardware and instrumentation; the construction and commissioning of these three vacuum systems; the methods of locating helium leaks along the 480m insulating vacuum system; the measurement of the helium absorption capacity of the activated charcoal in the insulating vacuum system, and the helium wave propagation inside the cold bore beam tubes.
9:40 AM		VT-WeM-5 Preliminary Design of the NIF Target Area Vacuum Systems J.W. Weed, A.P. Seth, R.W. Wavrik, T.H. Wiseley (Sandia National Laboratories); V.P. Karpenko (Lawrence Livermore National Laboratory) The National Ignition Facility (NIF) is a Department of Energy Inertial Confinement Fusion facility. The goal of the facility is to achieve fusion ignition and modest gain in the laboratory. The $1.1 billion NIF project is responsible for the design and construction of the 192 beam, 1.8MJ laser necessary to meet this goal. The project is a national project with participation by LLNL, LANL, SNL, URLLE, and numerous industrial partners. The NIF target experimental systems includes a hollow, 10m diameter, aluminum (5083 alloy) sphere which is the target chamber. The chamber is required to be evacuated from atmospheric pressure to 5x10^-5 Torr in two hours and 5x10^-6 Torr in 24 hours. The volume of the chamber and attached components is 750 m³. The internal surface area of the chamber and its contained components is on the order of 10⁵ m². Attached diagnostic beam lines and final optics assemblies must also be evacuated by pumping systems independent of the chamber system. The pumping systems must be oil and particulate free to prevent contamination of the target chamber internal components and the target itself. Preliminary design has recently been completed. The pumping systems consists of oil-free roughing, turbomolecular-drag, and cryogenic pumps. Gauging includes total and partial pressure sensors as well as leak detection apparatus. A real-time process control system will operate the various components with automated sequencing, interlocks, alarms, and exception handling features. The hardware, instrumentation, design criteria, and performance goals will be described. This work supported by the United States Department of Energy under Contract DE-AC04-94AL85000.
10:00 AM	Invited	VT-WeM-6 Low-Gas-flow Measurements; Examining Measurement Errors and Their Causes. S.A. Tison (National Institute of Standards & Technology) Gas flow measurement, one of the major process variables in semiconductor processes, is being pushed to lower levels of desired uncertainties. In addition to the desire for more accurate measurement (+/- l%), is the increasing demand for high-accuracies when measuring low-gas flows (> 7 x 10-6 mol/s ,10 sccm). Initial efforts to meet these low uncertainty requirements in gas flow measurement were concentrated on development of improved mass flow controllers (MFCs), which are prevalently used for gas flow measurement and control in semiconductor processing. It was discovered that uncertainties in the calibration of the MFCs in addition to the stability of the MFCs is a significant if not dominant cause of errors made in process flow measurements ¹. To achieve the lowest uncertainties in flow measurement, MFCs are typically calibrated using primary flow standards which are maintain either by the manufacturer, the user, or an independent laboratory. Tests of the primary standards at a number of facilities ² have shown errors in the flow measurement as large as 8%, relative to national flow standards maintained at the National Institute of Standards and Technology (NIST). Errors in flow measurement exceeding their specified uncertainty often result in rejection of the MFCs by the users, or larger process variability and lower process yields. To improve flow measurements requires that the primary flow standards be able to maintain uncertainties which are significantly lower than +/-1% and that their ability to perform at this level be demonstrated. This paper will describe a methodology for improving industrial flow measurements which includes on-site proficiency testing of primary flow standards. A description of the a newly developed laminar flowmeter used to accomplish flow proficiency tests and the test results are presented. ¹S.A. Tison, J. Vac. Sci. Technol. A 14, 2582 (1996) ²D. LeMay, and D. Sheriff, Solid State Tech. 39, 83 (1996)
11:00 AM		VT-WeM-9 Acoustic Flowmeter for the Accurate Metering of Gases for Semiconductor Processing. A.M. Calabrese, S.A. Tison (National Institute of Standards & Technology) Semiconductor production processes such as chemical vapor deposition rely upon accurate control of the flow rate (in the range of 1-1000 standard cm³/min) of several ultra-pure, yet highly toxic and/or reactive gases. Currently, thermal mass flow meters (TMFMs) are widely used in industry for these processes. Possible complications with the use of TMFMs arise because they are usually calibrated with inert gases (heat transfer in TMFMs depends on a complex combination of the gases’ properties), and because the flow must be divided in order to make a measurement due to limited dynamic ranges of the TMFMs. In the present research, an alternative approach to measuring low gas flows has been investigated. An ultrasonic Doppler-shift flowmeter has been developed to measure gas flows. Measurements of the flow-induced phase shift of an acoustic plane wave propagating in an unobstructed tube, combined with the speed of sound of the gas, are used to determine the velocity of the gas. The speed of sound can be measured in the flowmeter via two transducers which are located a known distance from the source. Combination of the mean gas velocity with measurements of the pressure and temperature allow calculation of the mass flow rate. The experimental system uses stainless steel tubes of diameter 0.05 - 0.1 cm, and frequencies near 100 kHz. No division of the flow is necessary, and the meters use materials that are compatible with the relevant process gases. Prototype meters indicate that sensitivities on the order of 1 degree/standard cm³/min are possible in the range of 1-1000 standard cm³/min. Experimental data of two prototype systems using different geometric designs and receivers will be presented.
11:20 AM		VT-WeM-10 Turbomolecular Pump Design for High Pressure Operation. M. Spagnol, R. Cerruti, J.C. Helmer (Varian S.p.A., Italy) The operating range of Turbopumps is being extended to higher pressures as a consequence of many demanding applications. Varian SpA has specialized in the design of Gaede stages for extending the operating pressure of the pump. In this paper, two design aspects of a Gaede stage are identified and analyzed: 1)maximum discharge pressure, and 2)power consumption. A new mathematical model of viscous flow has been developed to describe the pressure distribution in a single Gaede stage. From this a scaling law to optimize the stage design has been derived. Both, model and scaling law have been verified using different channel geometries. A practical limit to high pressure operation is set by the power consumption of the motor. A transition gas flow model of power consumption has been developed and experimentally verified. A program has been developed to optimize the design of a multi-stage Gaede pump. This suggests that a pump with an axial height of only 3 cm can exhaust at pressures up to 100 mbarr.
11:40 AM		VT-WeM-11 A History of the Vacuum Tube Industry R.H. Goehner (ETI Division of the Fredericks Company) The development of the vacuum tube industry was dependent on advances in all the scientific technologies, particularly electronics and vacuum. The history of this development, along with reasons for the continued use of vacuum tubes are discussed. Using significant milestones, the history is traced from the earliest times to the peak of tube manufacture in the 1950s. Among these milestones are the confirmation of Maxwell's equations by Hertz, which lead to radio; a driving force in the development of the vacuum tube. The discovery of the electron in 1897 by J.J. Thompson and the contributions of Edison, Fleming, DeForest, and Armstrong are noted. A typical tube plant at the peak will be described, with an emphasis on the pumping equipment and vacuum porcessing. The effect of foreign competition and the introduction of the transistor on the rapid decline of the US entertainment tube manufacture will aslo be discussed.