AVS2012 Session VT-TuP: Vacuum Technology Poster Session and Student-built Vacuum System Poster Competition

T. Fujimoto and M. Usami, ASME Trans. J. Fluids Eng. 106, 367 (1984).

R. Johnsen and B. K. Chatterjee J. Vac. Sci. Technol. A 29, 011002-1 (2011)

Facility for Rare Isotope Beams (FRIB), Michigan State University, East Lansing, MI 48824 USA

The Facility for Rare Isotope Beams (FRIB), a new national user facility for nuclear science funded by the U.S. Department of Energy - Office of Science and operated by Michigan State University (MSU), is currently being designed and established to provide intense beams of rare isotopes. It will enable scientists to make discoveries about the properties of these rare isotopes in order to better understand the physics of nuclei, nuclear astrophysics, fundamental interactions, and applications for society. The FRIB driver linac accelerates ions as heavy as 238U to energies beyond 200 MeV/u at beam powers up to 400 kW. Machine maintainability requires an average uncontrolled beam loss below 1 W/m. Baseline vacuum levels have been established to support this requirement and pumping systems have been defined. Vacuum levels in high loss regions such as beam stripping and collimation areas have also been evaluated and pumping systems defined. Value engineering has been applied to minimize the number of different vacuum pump and beam box types and sizes where possible. The pumpdown and operating vacuum of large, complex Target System chambers has been analyzed and pumping systems defined. This talk reports on the status of the FRIB vacuum system design.

This material is based upon work supported by the U.S. Department of Energy Office of Science under Cooperative Agreement DE-SC0000661

è Pohang Light Source’s upgraded machine (PLS-II) has been on commissioning stage since the installation completed in June 2011. PLS-II storage ring vacuum system is designed to maintain a base pressure in the low 10^-9 Torr region for the beam-gas scattering lifetime in excess of 20 hours, and the vacuum components, especially photon absorbers, are designed to endure 3 GeV, 400 mA beam operation. Several issues regarding the vacuum system will be presented, such as BPM stability from mechanical or electrical point of view, vacuum chamber displacement, photon absorbers, and vacuum related events during the commissioning.

Thermal management within vacuum chambers is sometimes a concern for temperature dependent vacuum processes. We describe a phenomenon in which using a magnetically levitated turbo molecular pump causes internal heating of a chamber. We measured the heat load based on a thermocouple inside the vacuum chamber, exposed line of sight to the magnetically levitated turbo pump. The measurements showed a temperature change of more than 10C based on the heat load from the pump. We also measured the effect of any heating due to gas flows and found that the dominant effect was not related to gas flows and was likely simply radiative heat transfer from the pump. The source of the heating of the pump is likely Ohmic heating due to eddy currents in the turbo pump.

This paper introduces unique statistical features extracted from the measured state variables of dry vacuum pumps in the semiconductor processes. They were found to have three distinctive means and overlapped distributions, not a single normal distribution. More specifically, two distinctive distributions near the upper and lower asymptotic bounds are obviously observed from the gas-loaded states of the vacuum pump and the third one from the idle states. These observations have provided new motivations of not only separating the pump operation state into the gas-loaded and idle states but also modeling the upper and lower bounds as a separated distribution. A linear adaptive parametric model (APM) is proposed such that their linear trends of each state variable are shown to be mapped onto their model parameters. Those estimated model parameters are used to construct the batch data obtained after each process. The APM-based batches are also exploited to construct the batches under the normal operating conditions (NOC) such that the major eigenvectors of the NOC batches are used to diagnose the current process batch data. It should be noted that the APM-based batch provides a dramatic reduction of memory usage and computation time (for example, 1~2 % memory usage and 10 times faster computation time) in comparison to the conventional dynamic-time wrapping methods. The feasibility of the proposed APM for the predictive maintenance of dry vacuum pumps is demonstrated to be successful by illustrating test results obtained from the six dry vacuum pumps.

This presentation addresses measurements related to energy, utilities and materials usage on dry vacuum pumps. A particular course of action was taken into account with regard to the measurement into equivalent energy suggested by SEMI Document 4399.

Recent SEMATECH and SEMI studies showed that 50 ~ 60 % of equipment power is used for vacuum pumps. Currently vacuum pump suppliers have responded by reducing power consumption and cooling water flow requirement in energy consumption at the component level. Actual process studies showed that for some processes, the energy consumption level did not change significantly during idle and processing operation modes. However, specified studies in experimental scale to characterize the energy consumption pattern have not been reported yet.

We have performed a simulation study to characterize energy consumption pattern in the idle and process modes.

The pressure range of about 0.1 to 50 mbar for 7 minutes was assigned to the simulated process mode, meanwhile the pressure of <0.1 mbar for 3 minutes to the idle mode. The integrated characteristics evaluation system for dry vacuum pumps has been utilized to gather the dry pump characteristics data for the simulation. Roots, claw, classical screw, and multi-stage type vacuum pumps supplied from the manufacturers have been evaluated using the evaluation system in terms of ultimate pressure, pumping speed, power consumption, vibration, sound power as well as nitrogen purge, cooling water rate from the single pump monitoring system in time-synchronized mode. This study includes the application of the SEMI S23-0705 standards – A Guide for Conservation of Energy, Utilities and Materials Used by Semiconductor Manufacturing Equipment.

The estimated power consumption per pump per year was ranged from 10 to 30 MWh and 15 to 50 MWh for 600 ~ 1200 m³/h dry pumps in idle and processing modes, respectively. The utility energy consumption was also ranged from 5 to 10 MWh and 10 to 30 MWh, respectively. More specific energy consumption patterns with respect to the pressure are also presented.

In this work we suggest that the correlation mechanism dependant on the actual process lines should be carefully analyzed and furthermore understood, for example, the relationship between cooling water flow rate and temperature variation during processes. Simple characteristic modeling of energy consumption patterns dependent on dry and roots pump types are also discussed.

Acknowledgements: Results are partially attributed to two national projects (Contract Nos. 10031858 and 10031836) sponsored by the Korean Ministry of Knowledge Economy.

Advanced industrial processes such as semiconductor and display manufacturing continuously requires the precise measurement and control of the low mass flow of gases. The requirements include very low mass flows of less than 1sccm or 2 x 10^-2 mbar∙l/s. However, the lower limit of traceable mass flow ranges are not well defined even in the sophisticated NMIs (National Measurement Institute). Since the primary standard for mass flow (kg/s) must provide a method for deriving mass flow directly from its base units, almost all mass flow systems heavily rely on the gravimetric method. Current technical measurement limit of mass flow rate with the gravimetric method is 2 x 10^-8 kg/s (10^-2 ~ 10^-3 mbar∙l/s).

Ensuring mass flow traceability ability in the range of greater than 10² mbar ∙ l/s with the Korea Research Institute of Standards and Science (KRISS) standard system, an attempt to trace the lower mass flows down to 10^-4 mbar ∙ l/s has been undertaken with constant volume flow meters (CVFM), sonic nozzle systems, and orifice method. In this work we briefly demonstrate the systematic attempt for the completion of the traceability chain from 10⁴ to 10^-4 mbar ∙ l/s.

Meanwhile the most effective way of ensuring completion of the traceability chain is, to say, characteristics monitoring of the mechanical rotary machines such as vacuum pumps used in advanced industry since these pumps have normally clean and smooth pumping capabilities during their stable operation.

Utilizing the combined mass flow systems mentioned above, the measurement of pumping speed has been performed with the throughput and orifice methods dependent on the mass flow regions. However, in the HV range of the molecular flow region, the high uncertainties of the gauges, mass flow rates, and conductance are too critical to precisely accumulate reliable data. In order to solve the uncertainty problems of pumping speeds in the HV range, we introduced an SRG with 1 % accuracy and CVFMs to measure the finite mass flow rates down to 10^-3 mbar ∙ l/s with 3 % uncertainty for the throughput method. In this way we have performed the measurement of pumping speed down to less than 10^-6 mbar with an uncertainty of 6 %. In this article we suggest that the CVFM has an ability to measure the conductance of the orifice experimentally with flowing the known mass through the orifice chambers, so that we may overcome the discontinuity problem encountering during introducing two measurement methods in one pumping speed evaluation sequence.

We present the results of an extensive study of the outgassing rates of four nominally identical stainless steel vacuum chambers. Chambers were either heat treated (semi-vacuum) at 400°C, coated with TiN, coated with SiO₂ (SilcoGuard™), or heat treated then coated with SiO₂. The outgassing rates were measured at a variety of room temperatures to determine the temperature dependence and compare to theoretical models of diffusion- or recombination-limited outgassing. In addition, outgassing rates were evaluated following venting to air and re-baking at 150°C and 250°C to simulate the impact of putting a chamber into service.

The need addressed here is the problem of pumping out internal volumes in vacuum systems that may be blocked by construction methods. Vented screws are a common approach to reducing this problem. This discussion will focus on evaluating the magnitude of the problem and various solutions. Calculations on conductance for vented and unvented screws and some accessories will be given. These conductance values will be used to model different common configurations that have different amounts of trapped volumes.

The conclusions are clear; vented screws and venting accessories are essential.

The direct measurements on un-vented screws were flawed because the screws tended to seal to varying degrees by the head of the screw. A use of a vented washer helped but the leak rate was hard to measure because of variability. Direct comparison to calculations is difficult because the tolerances in screw fasteners. There are large differences in conductance given the dimensions of the clearance between the screw and the mating fixtures. Calculating the ranges of conductance that can be expected for unvented screws, venting channels and internal volumes were done for different applications in vacuum systems. Ranges in excess of a factor >15 were encountered for a single divinition of a particular fastener. Some comparisons with data shed insight to the nature of the task.

The result is a "set of ground rules for design" based on the quantitative information derived. Examples are given for different designs with different parameters in terms of vacuum performance.