He II: From the Lab to an Engineering Fluid

by Dr. John Weisend II, European Spallation Source, CSA Chairman, john.weisend@esss.se

He II (sometimes referred to as superfluid helium) is the second liquid phase of 4He, the most common helium isotope [1]. This phase, first observed in the early 1930s, results from the condensation of some of the helium atoms into the ground quantum state. The existence of He II can only be explained via quantum mechanics and it is a macroscopic example of quantum theory. Under certain conditions, He II can possess zero viscosity and an extremely high heat transfer capability. Despite its unique nature, over the past 40 years He II has moved from a fluid studied for its fundamental physics in laboratories to an engineering fluid used in industrial quantities in some of the most significant scientific instruments of our time.

This article traces the evolution of He II as a cryogenic coolant, discussing both its applications and the research and development activities required to apply He II to large-scale systems. Most of this effort occurred during the 50 years of existence of the CSA and involved many of the institutions and people that also contributed to CSA.

The most important property of He II as a coolant is simply its lower temperature. The He II phase has an upper temperature of roughly 2.2K and exists as a liquid at temperatures approaching 0K. These lower temperatures result in the higher performance of superconductors, in both magnet and radio frequency cavity applications. They also permit infrared detectors to observe objects that radiate at temperatures less than 4K. While the properties of He II, particularly its high effective heat transfer ability, have proven useful in technical systems, the principal motivation for applying He II to large systems is the effect that its lower temperature has on materials and devices.

When describing the engineering applications of He II, it is useful to divide these uses into terrestrial and space applications. The terrestrial applications, mainly driven by particle accelerator and fusion energy projects, generally involve cooling of superconducting magnets or superconducting RF cavities. In terrestrial applications, the main concerns are the development of efficient and reliable refrigeration plants that operate at He II temperatures, the transfer and storage of large amounts of He II and the design of mass-produced cryostats for He II operation. Space applications are typically driven by the needs of infrared astronomy and other measurements made below 2K. In space applications, the systems are smaller but not refillable and the main concerns include development of very low loss He II cryostats as well as fluid management and venting in zero gravity.

An early example of using He II as a coolant comes from the Superconducting Linear Accelerator (SCA), built in the early 1970s at the High Energy Physics Laboratory of Stanford University. This electron accelerator [2, 3], which is also one of the earliest examples of a superconducting linac, operated at 1.9K. Development of the SCA project started in the mid-1960s and included an early realization [4] that operation of such machines at He II temperatures was advantageous. This was due to the improved performance of the superconducting cavities coupled with the excellent heat transfer properties and high heat capacity of the He II. Based on the success of the SCA, similar scale machines were built at a number of institutes in the early- to mid-1970s. These institutes included the Nuclear Research Center in Karlsruhe, Germany [5] and the University of Illinois. The SCA project also developed solutions to the issues of He II refrigeration. A 300 W He II refrigeration plant was developed in cooperation with industry and installed at Stanford as a central refrigeration facility.

One of the first large-scale applications of He II cooling to magnet systems was the Tore Supra Tokamak, built in Cadarache, France, during the 1980s [6-8]. This project further developed designs required for large-scale applications of He II and showed that it could be reliably used to cool magnet systems, setting the groundwork for the Large Hadron Collider (LHC) machine twenty years later.

One of the key requirements of Tore Supra was that the toroidal field superconducting magnets operate with a surface field of 9 T. Such a field was beyond the capability of the standard NbTi superconductor operating at 4.2K but would be possible if the superconductor were operated at 1.8K. However, a bath of He II at 1.8K would have a saturation pressure of 1.6 kPa, significantly below atmospheric pressure. Helium vapor at these pressures has poor dielectric properties and thus does not work well with large magnet systems. The solution used for Tore Supra was a double bath system in which a magnet was immersed in pressurized He II at roughly 1.8K and atmospheric pressure (~ 0.1 MPa) which was in turn cooled by a heat exchanger separating this bath from a saturated bath of He II at 1.6 kPa. This approach solves the problem of low pressure helium vapor near the magnet and is the basis of most
He II cooling of large-scale magnet systems.

The heat transferred into the saturated bath boils off helium vapor that must be pumped away in order to maintain the bath at the correct pressure and temperature. This pumping can be done with room temperature compressors but this has a number of disadvantages, including the need for large diameter pipes operating below atmospheric pressure (and thus prone to leaking) to carry the vapor to the room temperature compressors and the requirement for large low pressure heat exchangers. Tore Supra was one of the first users of cold compressors that operate at cryogenic temperatures. In Tore Supra, a flow of 14 g/s of helium vapor was raised from 1.2 kPa to 80 kPa in two stages between 4.3K and 15K with the remaining pressure rise carried out at room temperature. The initial development and application of reliable, practical cold compressors was an important step towards making large-scale
He II refrigeration practical.

The Tore Supra cryogenic system provided 300 W of cooling at 1.75K. The refrigeration plant was procured, manufactured, installed and commissioned between 1983 and 1987 with first magnet cooling in 1988. Tore Supra’s successful, reliable operation for over 25 years demonstrated that He II can be used at an industrial scale. Figure 1 shows a view of the Tore Supra cold box with a cold compressor.

In the early 1980s, the US DOE held a competition for a new large electron accelerator for nuclear physics research. A design based on a warm linac was chosen. However, very soon after work began at the Newport News VA site, staff from the new laboratory (originally called the Continuous Electron Beam Accelerator Facility [CEBAF], but since renamed Jefferson Lab) proposed that the original warm design be replaced by a superconducting recirculating linac. This change was in part motivated by the demonstrated experience with He II cooling shown by Tore Supra and SCA, along with significant advances in superconducting RF technology carried out at a number of institutions. The new superconducting design was approved in 1986 with delivery of the first beam in 1994.

The CEBAF machine [9, 10] represented a major increase in He II cooling capacity. The cryogenic system at CEBAF provided 4.8 kW of cooling at 2K. The CEBAF system used only cold compressors to pump off the subatmospheric helium vapor and raise its pressure to atmospheric pressure. As part of this effort, CEBAF continued development of cold compressors and paid particular attention to optimizing the control strategies for cold compressors [11].

The CEBAF system has run very reliably from 1994 to the present day and until the advent of the LHC represented the largest single usage of He II as a coolant. The success of CEBAF led the lab to produce the entire superconducting linac and associated cryogenic system for the Spallation Neutron Source (SNS) in Oak Ridge TN. This system [12], which started operation in 2004, provides 2.3 kW at 2K. While the designs of the cryogenic systems of CEBAF and SNS are very similar, there was one important design evolution. In the original CEBAF machine, He at 2.2K and approximately 0.3 MPa was distributed from the Central Helium Liquefier out to the cryomodules. In the SNS design, He was distributed from the Central Helium Liquefier at approximately 4.5K and 0.3 MPa, with all the final cooling to 2K taking place at the individual cryomodules. This distributed cooling approach reduces the heat leak at 2.2K and has additional operational advantages. This approach has now become standard for SRF linac systems that contain cryogenically separate cryomodules. As will be seen, a different approach must be taken in linacs such as XFEL and ILC in which the cryomodules form a continuous cryogenic and vacuum component.

In 2013 [13], the CEBAF machine’s energy was doubled to 12 GeV by the addition of higher performance cryomodules and a second cryogenic plant, bringing the entire He II refrigeration capacity at Jefferson Lab to 9.2 kW. Figure 2 is a view of the recently installed 12 GeV upgrade cold boxes.

The largest current use of He II as a coolant occurs in the LHC at CERN [14, 15]. This project was approved in 1994 and produced first beams in 2008. The motivation for using He II in the LHC was very similar to that of Tore Supra. In order for the machine to reach its desired energy and fit inside the existing LEP tunnel at CERN, the bending magnets were required to reach a nominal field of 8.3 T. An early design choice was to use NbTi superconductor at 1.9K to meet this requirement rather than try to develop large scale production of other, higher field superconductors that operate at 4.2K. This choice necessitated the use of He II.

The He II system cools 1,232 dipole magnets, 392 quadrupole magnets and assorted specialty magnets arranged around a 26.7 km circumference tunnel. Cooling is provided by eight refrigeration plants, each producing up to 2.4 kW of cooling at 1.9K [16]. Like Tore Supra, the subatmospheric pumping was accomplished by a set of cold compressors followed by a final stage of warm compression. CERN funded additional technology development in the area of cold compressors [17]. Figure 3 shows a cutaway view of the LHC magnets installed in the tunnel at CERN.

Even while the LHC was under development, the high energy physics community was interested in a complementary high energy lepton collider. The design envisioned for this machine was two very large linacs (each roughly 12 km in length) accelerating beams to a central interaction point. Two approaches were taken, one, centered at the SLAC National Accelerator Laboratory in the USA, using copper room temperature RF structures, and another, centered at the DESY Laboratory in Germany, using superconducting RF cavities. In the early 2000s, the superconducting approach (known as TESLA [18]) was chosen as the base line design for the future International Linear Collider (ILC). The ILC accelerator and cryogenic design build upon previous experience from SRF linacs as well as a large worldwide R&D program in superconducting RF cavities [19], including extensive work at
DESY.

Ten refrigeration plants, each providing approximately 3.7 kW cooling at 2K as well as additional cooling at 5K and 40K, are planned to be distributed along the length of the ILC [20]. These will cool roughly 2,000 cryomodules containing nearly 20,000 SRF cavities operating at 2K. If built, the ILC will become the largest He II installation in the world. While ILC has yet to be approved for construction, the SRF and cryomodule designs developed for ILC are being used in the XFEL linear accelerator project [21] under construction at DESY. XFEL itself is a substantial He II installation consisting of 120 cryomodules containing roughly 1,000 SRF cavities with a total heat load of about 2.6 kW at 2K.

Due to their size, ILC and LHC share a number of issues. The large numbers of magnets or cavities to be cooled required a different configuration than that used at JLab and SNS. In LHC and ILC, the cryomodules or magnets are tied together into larger continuous cryogenic units. This results in situations where there is two-phase flow of He II. Studies [22-24] were carried out to ensure that this flow would be stable. The large number of cryomodules or superconducting magnets in LHC and ILC requires low heat leak designs that can be economically produced. Significant development and prototyping [25, 26] have resulted in designs that meet these requirements. Figure 4 is a cross sectional view of the third generation ILC cryomodule.

He II continues to be a key feature in large-scale accelerator projects. In addition to the ILC, He II cooling is used in superconducting RF linacs at the European Spallation Source (Lund, Sweden) and the Facility for Rare Isotope Beams (Michigan State University, USA), both in the construction phase, as well as in the recently announced LCLS II linac (SLAC, USA).

Starting in the 1980s, He II became an important coolant in space missions. The first significant use of He II in space occurred in the Infrared Astronomical Satellite (IRAS), a joint mission of the Netherlands, US and UK launched in 1983. IRAS [27] carried out the first sky survey at infrared wavelengths and demonstrated the scientific value of space-based IR astronomy. In order to take measurements at IR wavelengths, IRAS cooled its optics with 600 liters of He II. A similar sized system, the Cosmic Background Explorer (COBE), was launched in 1989. COBE produced the first measurements of the anisotropy in the cosmic background radiation remaining from the Big Bang, a result that has had major implications in our understanding of the early universe. The scientific output of IRAS and COBE has led to a series of space systems cooled by He II from the 1990s to present day. These missions [28] include the Infrared Space Observatory (launched in 1995, 2,300 liters of He II), the Infrared Telescope in Space (launched in 1995, 100 liters of He II), Astro-E2 (launched in 2005, 30 liters of He II), the Spitzer Space Telescope (launched in 2003, 360 liters of He II), Gravity Probe B (launched in 2004, 2,000 liters of He II) and Herschel (launched in 2009, 2,300 liters of He II). NASA, the European Space Agency and Japan have all flown significant missions using He II [29].

While the amount of He II in a single space mission is much smaller than in most terrestrial applications, there are unique challenges in space cryogenics. Since the loss rate of the He II drives the mission lifetime, the development of very low loss helium cryostats for space applications was key to the successful use of He II in space. This problem was made even more challenging by weight and size limits for spacecraft and by the need for the cryostats to withstand launch loads. Innovative solutions were developed. These include Passive Orbital Disconnect Struts [30], which support the cryostat during launch but essentially disconnect in zero g, reducing the heat leak. Examples of low loss space cryostats include Astro-E [31] and Gravity Probe B [32]. Managing and venting He II in zero g was also a challenge and was solved after much research using porous plugs and meshes [33]. Figure 5 shows the He II dewar flown on the Spitzer Space Telescope.

The limitation imposed on mission lifetime by the boiloff of He II in space cryogenics led NASA to consider the possibility of on orbit replenishment of He II from either the space shuttle or the space station. A technology demonstration experiment, Superfluid Helium On Orbit Transfer (SHOOT), was flown on the space shuttle in 1993 [34]. This experiment successfully demonstrated the transfer of He II between two dewars along with a number of other related technologies for handling of He II in space. Fundamental studies of He II have also been carried out in zero g. [35]

Concurrent with the application of He II to larger and larger systems was a global research and development effort carried out in universities, research laboratories and industry on both the fundamentals and technology of He II. This work, which continues to the present day, provided the background data and technical prototypes that enabled the growth of He II as a technical coolant. This effort also provided a trained cadre of people with experience in He II technology who were available to work on the large-scale projects discussed so far. This R&D was funded in part by agencies and institutions (DOE, NSF, NASA, CERN, CEA, etc.) whose projects would later benefit from He II technology. The range of subjects covered were extensive and included fundamental studies on heat transfer [36-38], fluid mechanics [39-41], modeling [42, 43] and flow visualization [44, 45], as well as the development of components such as heat exchangers [46, 47], pumps [48, 49], flow meters [50, 51], porous plugs and fluid management [52, 53], valves [54] and cryostats [55-57]. The important development work on cold compressors and He II two-phase flow were previously discussed above. It’s important to keep in mind that the papers cited above represent only a small fraction of the work conducted over the past 40 plus years.

Research and development in
He II and its applications continues today. For example, at the 2013 Cryogenic Engineering Conference, more than a dozen papers were published on He II fundamentals and applications.

In summary, over the past 50 years He II has transitioned from a laboratory phenomenon to a reliable, well understood cryogenic coolant used both on Earth and in space. The scientific output from machines and instruments using He II cooling has been immense, and future applications promise to be equally productive. This transition has only been possible due to the efforts of many institutions and researchers throughout the world, and the author apologizes in advance to the groups and institutions that space did not permit mentioning.

The development of He II as a technical coolant mirrors the general intellectual excitement and progress that has occurred in cryogenics over the past 50 years [58].

References

1. Van Sciver, S. W. Helium Cryogenics, Springer (2012)
2. McAshan, M. et al. “Demonstration of the Superconducting Electron Accelerator as a High-Intensity High-Resolution Device,” Applied Physics Lett., Vol. 22 No. 11 (1973)
3. Calerco, J.R et al. “Initial Performance of the Stanford Superconducting Recyclotron,” IEEE Trans on Nuclear Science, Vol. Ns-24 No. 3 (1977)
4. Schwettman, H.A. et al. “Low Temperature Aspects of a Cryogenic Accelerator,” IEEE Trans on Nuclear Science, June (1967)
5. Herz, W. et al. “Continuously Cooled Superconducting Linear Accelerator & Particle Detector Models at 4.4 and 2 K,” Adv. Cryo. Engr., Vol. 17 (1972)
6. Claudet, G., Bon Mardion, G., Jager, B., Gistau, G. “Design of the cryogenic system for the Tore Supra Tokamak,” Cryogenics, Vol. 26 (1986)
7. Claudet G. “The Tore Supra He II cryogenic system,” Fusion Engineering and Design 20, 485-490 (1993)
8. Gavil, B. et al. “Experience from the Tore Supra cryogenic system, associated with the statutory ten years tests of the pressure vessels,” Adv. Cryo. Engr. Vol. 45 (2000)
9. Rode, C. “CEBAF Cryogenic System,” Proc. 1995 Particle Accelerator Conference (1995)
10. Rode, C., Arenius, D. et al. “2 K CEBAF Cryogenics,” Adv. Cryo. Engr., Vol. 35 (1990)
11. Wilson, J. D., Ganni, V. et al. “Determination of the Optimal Operating Parameters for the Jefferson Lab’s Cryogenic Cold Compressor System,” Adv. Cryo. Engr., Vol. 49a (2004)
12. Arenius, D., Chronis, W. et al. “Cryogenic System for the Spallation Neutron Source,” Adv. Cryo. Engr., Vol. 49a, (2004)
13. Ganni, V. et al. “Commissioning of the Helium Refrigeration System at JLab for 12 GeV Upgrade,” Adv. Cryo. Engr., Vol. 59B (2014)
14. Lebrun, Ph. “Commissioning and First Operation of the Large Hadron Collider (LHC),” Proc. ICEC 23 (2011)
15. Evans, L. (Editor). The Large Hadron Collider: a marvel of technology, EPFL/CRC Press (2009)
16. Claudet, G., Guyot, Ph. et al. “Specification of 8 2400 W @ 1.8K Refrigeration units for the LHC,” Proc. ICEC 18 (2000)
17. Millet, F., Claudet, S. “Performance Assessment of 35 Cold Hydrodynamic Compressors for the 1.8K Refrigeration Units of the LHC,” Adv. Cryo. Engr., Vol. 51B (2006)
18. Brinkman et al. (eds). Conceptual design of a 500 GeV e+e- with integrated x-ray laser facility, DESY Report 1997-048 (1997)
19. Padamsee, H. Knobloch, J., Hays, T. RF Superconductivity for Accelerators, Wiley (1998)
20. Peterson, T.,J., Geynisman, M., Klebaner, A. “ILC Cryogenic Systems Reference Design,” Adv. Cryo. Engr., Vol. 53B (2008)
21. Botzhko, Y., Lierl, H., Petersen, B. et al. “Requirements for the Cryogenic Supply of the European XFEL Project at DESY,” Adv. Cryo. Engr., Vol. 51B (2006)
22. Rousset, B., Grimaud, A. “Stratified Two-Phase Superfluid Helium Flow: I,” Cryogenics, 37:733 (1997)
23. Xiang, Y., Petersen, B. et al. “Numerical Study of Two-Phase Helium II Stratified Channel Flow with Inclination,” IEEE Trans. On Applied Superconductivity, Vol. 10 (2000)
24. Perraud, S., Rousset, B. et al., “Comparison between Normal and He II Two-phase Flows at High Vapor Velocities,” Adv. Cryo. Engr., Vol 51A (2006)
25. Pagani, C., Barni, D. et al. “Further Improvements of the TTF Cryomodule in View of the TESLA Collider,” Adv. Cryo. Engr., Vol. 45A (2000)
26. Benda, V., Dufay, L. et al. “Measurement and Analysis of Thermal Performance of LHC Prototype Cryostats”, Adv. Cryo. Engr. Vol 41A (1996)
27. “The Infrared Astronomical Satellite (IRAS) Mission,” G. Neugebauer et al., The Astrophysical Journal, 278:L1-L6 (1984)
28. Personal Communication – P. Shirron NASA-GSFC
29. Linder, M. et al. “Cryogenics in Space-A review of missions and technologies,” ESA Bulletin 107 (2001)
30. Spradley, I. E. and Parmley, R. T. “Design and Test of a Modified Passive Orbital Disconnect Strut (PODS-IV),” Adv. Cryo. Engr., Vol. 33 (1988)
31. Parmley, R.T. “Unique Cryogenic Features of Gravity Probe B (GP-B),” Adv. Cryo. Engr., Vol. 33 (1988)
32. Breon, S., Branch, H.D., et al. “Design of the XRS Helium Insert,” Adv. Cryo. Engr., Vol. 41B (1996)
33. Schotte, U. “He II Phase Separation with Slits and Porous Plugs for Space Cryogenics,” Cryogenics, 24:10 (1984)
34. DiPirro, M., Shirron, P. and Tuttle, J. “On-orbit superfluid transfer: preliminary results from the SHOOT flight demonstration,” Cryogenics, 34:5, (1994)
35. Yang., L.C. and Mason, P.V. “Superfluid Helium Flow in Zero Gravity,” Cryogenics 20:2 (1980)
36. Seyfert, P. “Practical Results on Heat Transfer to Superfluid Helium,” Stability of Superconductors, IIR Commission A 1/2 (1981).
37. Kashani, A., Van Sciver, S.W. “Kapitza Conductance of Technical Copper with Several Different Surface Preparations,” Cryogenics 25,238 (1985)
38. Takada, S., Murakami, M., Kimura, N. “Heat Transfer Characteristics of Four Film Boiling Modes around a Horizontal Cylindrical Heater in He II,” Adv. Cryo. Engr., Vol. 55B (2010)
39. Walstrom, P.L., Weisend II, J.G. et al. “Turbulent Flow Pressure Drop in Various He II Transfer System Components,” Cryogenics, Vol. 28, 101 (1988)
40. Rousset, B., Claudet, G., et al. “Pressure Drop and Transient Heat Transfer in Forced Flow Single Phase He II at High Reynolds Numbers,” Cryogenics, Vol. 34 217 (1994).
41. Nishigaki, K., Takeda, M. et al. “Observation of Cavitation in Superfluid Helium,” Adv. Cryo. Engr., Vol. 47A (2002)
42. Snyder, H.A., Mord, A.J. “Modeling Superfluid Phase Separator Systems,” Adv. Cryo. Engr., Vol. 43B (1998).
43. Darve, C., Bottura, L. et al. “A Method for Numerical Simulation of Superfluid Helium,” Adv. Cryo. Engr., Vol. 57A (2012).
44. Yokota, N., Murakami, M., et al. “Superfluid Turbulence Properties of Thermal Counterflow Jet measured with PIV,” Adv. Cryo. Engr. Vol. 55A (2012).
45. Zhang, T. and Van Sciver, S.W. “Use of the particle image velocimetry technique to study the propagation of second sound shock in superfluid helium,” Physics of Fluids, 16 (12), 99-102 (2004)
46. Bertinelli, F., Favre, G. “Design and Fabrication of Superfluid Helium Heat Transfer Tubes for the LHC Superconducting Magnets,” Proc. EPAC 2004 (2004)
47. Weisend II, J. G., Huang, Y. and S.W. Van Sciver. “Testing of Two Heat Exchangers in He II,” Cryogenic Processes and Equipment 1989, ASME 141 (1989)
48. Jahromi, A.E., Miller, F. et al. “Modeling and Development of a Superfluid Magnetic Pump with No Moving Parts,” Adv. Cryo. Engr., Vol. 57B (2012)
49. Kittel, P. “Losses in Fountain Effect Pumps,” Proc. 11th ICEC (1986)
50. Van Sciver, S.W., Holmes, D.S. et al. “He II Flow Metering,” Cryogenics, 31:75 (1991).
51. De Jonge, T., Patten, T. et al. “Development of a Mass Flow Meter Based on Coriolis Acceleration for Liquid, Supercritical and Superfluid Helium,” Proc. ICEC 19 (2003).
52. Yuan, S.W. K., Frank, D.J. “Integrated Fountain Effect Pump Device for Fluid Management at Low Gravity,” Cryogenics 28:115 (1988)
53. Yamaguchi, H., Ezoe, Y. et al. “Developments of Porous Plug Phase Separator and Superfluid Film Flow Suppression System for Soft X-Ray Spectrometer (SXS) on Board ASTRO-II,” Proc. ICEC 23 (2011)
54. Mills, G. L., “Design and Development of a Leak Tight He II Valve with Low Thermal Impact,” Adv. Cryo. Engr., Vol. 37B (1991)
55. Urbach, A.R., Vorreiter, J. , Mason, P. “Design of a Superfluid Helium Dewar for the IRAS Telescope,” Proc. ICEC 7 (1978)
56. Weisend II, J.G., Pagani, C., et al. “The TESLA Test Facility (TTF) Cryomodule: A Summary of Work to Date,” Adv. Cryo. Engr., Vol. 45A (2000)
57. W.J. Schneider et al. “Design of the Spallation Source (SNS) Cryomodule,” PAC2001 (2001)
58. Timmerhaus, K.D., Reed, R. Cryogenic Engineering: Fifty Years of Progress, Springer (2007)