by Dr. Mark O. Kimball, NASA/Goddard Space Flight Center, mark.kimball@nasa.gov

Fifty milliKelvin, now that’s cold! This is the temperature at which the detector array aboard the soft X-ray detector (SXS), destined for the Japanese Astro-H mission, is held when gathering X-rays. This detector is capable of sub-5 eV resolution and is state-of-the-art at the time of this writing [1]. To achieve this amazing feat, the detector array must be cooled to 50mK. This is done by connecting it to the coldest stage of a multi-stage adiabatic demagnetization refrigerator (ADR).

This ADR, developed by the Cryogenics and Fluids Branch at NASA’s Goddard Space Flight Center, is capable of holding the detector array at 50mK for greater than 24 hours with better than 2µK stability over a 10-minute period [2]. After its cooling capacity is expended, the refrigerator may recycle to either a pumped helium bath at 1.2K or a 4.5K Joule-Thomson cooler. These two heat sinks allow a great deal of flexibility and fault tolerance.

The capabilities and performance of the ADR system are covered elsewhere [2]. What has not been described before is the ground-based system used to test both the engineering and flight models of the ADR and detector array. To fully test the ADR, a cryostat that simulates both heat sinks found in the cooling chain of the flight unit is needed. Therefore, both a 1.2 and 4.5 Kelvin interface are required in any simulation of the flight dewar. The higher temperature is achieved using a commercial off-the-shelf cryocooler. The lower temperature may be achieved using a helium-4 bath evacuated until the vapor pressure is below 82 Pa (about 0.6 Torr). However, a helium bath requires periodic refilling during a testing campaign. To perform the refill the bath needs to be warmed to 4.2K, additional liquid helium must be added, and then the bath must be pumped back to operating pressure. To avoid this additional labor, and the associated attention necessary when using sub-atmospheric liquid helium, it was decided early in the program that the helium tank would be simulated using a 2-stage ADR with the same 4.5K cryocooler already required for the higher temperature heat sink mentioned above.

An ADR uses the magneto-caloric effect to cool to low and ultralow temperatures. Simply put, when a paramagnetic substance—typically a pill containing a salt with the proper magnetic characteristics for a given operating temperature range—is subjected to an increasing magnetic field, the individual magnetic moments will align. This alignment lowers the entropy of the paramagnetic material, and energy is liberated in the form of heat. If the pill within the ADR stage is connected to a heat sink at a lower temperature, heat will flow out of the pill while the entropy of the paramagnetic system decreases. After an upper field is reached, the thermal link between the salt pill and the heat sink is opened. A reduction of the magnetic field permeating the salt pill will cool the pill along with anything attached to it—such as a detector array. When the desired operating temperature is reached, one controls the rate of reduction in the magnetic field such that the entropy change in the pill matches the heat coming from the detector array. The end result is a constant ultralow temperature of the array.

To mimic a helium tank with an ADR, one first brings the interface acting as the helium tank to a fixed low temperature, say 1.2K, using an ADR stage. Then one controls the demagnetization rate such that any heat imposed on the tank is taken up by the ADR system. Therefore, the helium tank is held at a constant temperature as long as the ADR has cooling capacity in excess of the heat flowing into the tank.

For this system, we have built an ADR with two stages. Both stages use gadolinium gallium garnet as the paramagnetic material. This choice of material is based upon the ease of procurement and relatively low cost (it is used as an optical or laser crystal). It also has magnetic characteristics so that it works well in the 1.2 to 4.5K range.

This two-stage ADR may be configured in one of two ways. The first uses one stage as an active thermal ballast while the other stage rapidly cycles between the helium tank and cryocooler temperatures. This pulls heat from the helium tank and transfers it to the higher temperature cryocooler. When not transferring heat to the cryocooler, the cycling stage, Stage B, attempts to cool the helium tank below the tank temperature setpoint. However, the ADR stage directly coupled to the helium tank simulator, Stage A, compensates by magnetizing and adding heat to the tank to maintain a constant temperature. When Stage B depletes its cooling capability it decouples from the helium tank via a heat switch. At this point, Stage A must decrease the current in its associated magnet to pull incoming heat from the helium tank. Simultaneously, the cycling stage recycles by lifting its stored heat to the cryocooler. After the recycle finishes, the second stage lowers its temperature below the helium tank’s setpoint and the thermal link between Stage B and the tank is made via the heat switch. Stage A again compensates to keep the temperature constant by increasing the field in its magnet. This cycle continues until the testing campaign is over.

The second configuration is one where both stages cycle as quickly as possible in a manner that is synchronous but out of phase by 180 degrees. Here, one stage is lifting heat to the cryocooler while the other is controlling the temperature of the tank. This configuration trades temperature stability for additional cooling power. This allows the temperature of the tank to be lower than configuration [A] but with larger temperature fluctuations.

A plot of the temperature of the helium tank, Stage A and Stage B during multiple ADR cycles is shown in Figures 3b and 4b. Comparing both plots, one finds a higher temperature stability for configuration [A] in Figure 3 but with a higher overall tank temperature. The opposite is seen for configuration [B] in Figure 4. One could merge the two configurations into a system that combines the positive attributes of both. This would be accomplished by adding a third stage to configuration [B] that would act as an active thermal ballast to smooth out the temperature fluctuations during handoff of control of the helium tank from one of the other two stages. Another method of smoothing the temperature disturbances is to utilize a feed-forward control of both ADR stages, but this is a topic for another time.

So, it is possible to simulate the environment offered by a liquid helium dewar without using a liquid cryogen. The cryostat described here simulates the environment that the instrument developed by NASA for the Astro-H mission will experience when it gazes into outer space. Over the past few years this cryogen-free cryostat was used to test both the engineering and flight models of the Astro-H ADR and detector assembly. At the time of this writing, the flight ADR and detector assembly are being integrated into the spacecraft at the Japanese Aerospace Exploration Agency facility in Tsukuba, Japan.

References

1. F. S. Porter, J. S. Adams, G. V. Brown, J. A. Chervenak, M. P. Chiao, R. Fujimoto, Y. Ishisaki, R. L. Kelley, C. A. Kilbourne, D. McCammon, K. Mitsuda, T. Ohashi, A. E. Szymkowiak, Y. Takei, M. Tashiro, and N. Yamasaki. The detector subsystem for the SXS instrument on the ASTRO-H Observatory. In Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, volume 7732 of Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, page 3, July 2010.
2. P. J. Shirron, M. O. Kimball, B. L. James, D. C. Wegel, R. M. Martinez, R. L. Faulkner, L. Neubauer, and M. Sansebastian. Design and predicted performance of the 3-stage ADR for the Soft-X-ray Spectrometer instrument on Astro-H. Cryogenics, 52(4-6, SI):165-171, Apr-Jun 2012.