Cryogenic Test Stand for Characterization of Magnetocaloric Materials

Alternative cryogenic refrigeration methods are needed to improve the low efficiency of traditional gas cycles cryocoolers. Magnetic refrigeration is an old known technique used to reach below Kelvin temperatures, though there is rising interest in using this method at liquid helium and liquid hydrogen temperatures due to its high Carnot efficiency. However, further experimentation is needed to fully understand the heat transfer dynamics of magnetocaloric materials at cryogenic temperatures. A test stand has been designed to evaluate the heat transfer coefficients of these materials using the single-blow transient test technique. The system consists of a hermetic helium gas circuit, which is cooled down to cryogenic temperatures and is forced to flow through a packed bed of magnetocaloric material. In order to test the heat transfer properties of the magnetocaloric materials at different magnetic fields the test stand has a superconducting solenoid capable of providing up to 4 T. The design and validation of the test stand for the characterization of magnetocaloric materials is described in the present paper.

Cryocoolers have undergone major advances in the last 20 years, improving their heat rejection capacity, and reliability.However, the cost and Carnot efficiency of these devices are still limiting factors [3] for the development of breakthrough superconducting applications.Therefore, alternative refrigeration methods are being continuously explored.Magnetic refrigeration is a promising alternative, since the reversible nature of the undergone process delivers high Carnot efficiencies [4].The magnetocaloric effect (or MCE) [5], is the basis of magnetic refrigeration.
For the modelling of a magnetic refrigerator, it is essential to establish the heat transfer and fluid dynamics mechanisms.Typically, correlations used for ambient temperature processes are used [6], which have provided good results for temperatures as low as 70 K [7].However, there is a limited literature on the validation of heat transfer and fluid dynamics correlations at liquid helium temperature.The most comprehensive work is found in [8], but the conclusions are not definitive since large fluctuation in the measured temperatures were observed.
This article presents the design of a test station for the characterization and heat transfer analysis of magnetocaloric materials down to liquid Helium (LHe) temperature based on the single-blow technique.The system has a superconducting magnet capable of providing up to 4 T.

II. EXPERIMENTAL APPARATUS
The single-blow technique is a transient method which allows to measure the heat transfer coefficient of a heat exchanger.This procedure requires a constant flow rate through the test bed and a step-wise fluid flow inlet temperature change.
Fig. 1 shows the schematic of the experimental system.It is a closed system which uses a pump to circulate the helium through the circuit.The pressure and mass flow of the system are controlled with two mass flow controllers (MFC), MFC1 and MFC2 in Fig. 1.
The helium gas goes inside a cryostat and is cooled down in a double counterflow heat exchanger, rejecting heat to the outgoing helium gas, and to the evaporated helium from the liquid helium bath where the magnet is submerged.A heater controls the temperature at the inlet of the regenerator.

A. Magnetocaloric Material (MCM)
Inside the test bed there is the magnetocaloric material, crushed into particles of small diameter (less than 1 mm.),  positioned in a packed bed as depicted in Fig. 2. Two MCMs have been selected: Gadolinium Gallium Garnet, GGG or Gd 3 Ga 5 O 12 , and Erbium Aluminum (II), ErAl 2 .
The selection of the MCM obeys to the following rules [9]: the selection of a suitable Curie temperature, intensity of the magnetocaloric effect, high electrical resistivity (to prevent eddy currents), good manufacturing, and corrosion properties and low price.
GGG, is an antiferromagnetic material, which has been the most employed magnetic refrigerant for cryogenic applications, as its low Néel temperature (2 K), gives this material great versatility for LHe applications.Fig. 3 shows the significant entropy change near LHe boiling temperature.
On the other hand, ErAl 2 was selected since there are additional applications at slightly higher temperatures, like hydrogen liquefaction, that have sparked interest in recent years [10].ErAl 2 is a ferromagnetic material, with a Curie temperature of 13.4 K.A ferromagnetic transition can be slightly appreciated in Fig. 4. Its optimum refrigerant capacity is theoretically higher than any other type of material.
The MCE properties of the magnetocaloric materials, as well as the quantity of material to be used in the tests, provides the boundary conditions for the rest of the components of the test   stand: the magnetic source and heat transfer system, which are summarized in Table I.

B. Magnetic Source
For a full evaluation of the material properties a magnetic field of 4 T was established, the best option is to use a superconducting magnet.
The magnet used in this test station, shown in Fig. 5, was developed for the CRISA project.During CRISA the power source of the antimatter spectrometer (AMS) experiment, a Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.particle detector for the international space station (ISS), was tested [11].Its characteristics are summarized in Table II.

C. Heat Transfer Circuit
Helium gas flowing through the regenerator will act as the heat transfer medium.The helium is pumped from ambient temperature, which provides a more precise control of helium mass flow and pressure.The equipment used is summarized in Table III.
Gas helium will also be responsible for cooling down the magnetocaloric material down to near its Curie (or Néel) temperature, where the MCE effect is more significant.For this purpose, a cryogenic heat exchanger was developed to recover the enthalpy of the evaporated Helium from the liquid bath which is cooling the superconducting magnet.
This cryogenic heat exchanger is similar to a Giauque-Hampson heat exchanger [12], although in this case there is an additional heat recovery stage.The helium that exits the packed Fig. 6.Cryogenic heat exchanger for MC test stand (left), detailed view of the cross section of the heat exchanger, showing the heat transfer (Q in ) from the high pressure helium gas (HP GHe) to the evaporated helium gas, and the heat transfer (Q out ) to the low pressure helium gas (LP GHe) (right).bed, flows inside the incoming helium circuit to increase the test station efficiency.The schematic of the heat exchanger is shown in Fig. 6.To control the temperature at the inlet of the regenerator a heater is placed immediately at the exit of the heat exchanger.

A. Magnetocaloric Material: Preparation and Validation
For the construction of a packed bed regenerator, the procedure followed described in [6], [13] was adopted: 1) Pre-crush the magnetocaloric material crystals to the required size.To achieve the chosen size the crushed crystals are passed through a series of sieves of unique dimensions.Fig. 7 shows a sample of crushed ErAl 2 .2) Fill the regenerator with the crushed crystals and apply certain packing pressure to achieve the chosen porosity, defined as the relation between the void volume and the total volume of the packed bed.

TABLE IV TYPICAL REGENERATOR VALUES
Ideally, the porosity of the regenerator should be minimized and therefore the packing pressure should be as high as possible.However, if the packing pressure is too high certain regions of the regenerator can become occluded, impeding the flow and isolating that region of the regenerator.Fig. 8 shows theoretical and experimental pressure drop for different porosities, it can be observed how with porosities greater than 0.45, the regenerator becomes occluded, as there is a divergence between the theoretical, given by Ergun correlation [14], and experimental pressure drop.
One more parameter is needed for the characterization of the regenerator, the shape factor (φ).The shape factor describes the 'irregularity' of the particles, it is defined as the ratio of the specific area of a spherical particle to the specific area of the particle in question.The geometric parameters and values used in the analysis are presented in Table IV.

B. Superconducting Magnet Tests
The superconducting magnet was manufactured in the early 2000s, therefore, there was certain probability that its superconducting properties have been degraded.Consequently, the magnet was tested alone at ambient temperature, with electrical and magnetic measurements.And at cryogenic temperatures to test the instrumentation, quench protection system, and the power source.
Fig. 9 shows the schematic adopted for the cryogenic tests, where R c is the ambient cables resistance, R cl the current lead resistance, L sc the inductance of the superconducting magnet,  R p is the dump resistor for quench protection, and D 1 is the protection diode, defined by its nominal current I c , and the forward voltage V f .Fig. 10 shows retraining from the magnet before achieving its nominal current, as it quenched during the first test at 312 A, probably due to the long time out of operation.Detection and quench system behaved as expected.

C. Heat Transfer Circuit
To cool down the magnetocaloric material down to its operating temperature we use the double counterflow heat exchanger, Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.being the heat flux, Q, in each direction: where h is the heat transfer coefficient, T f,annular the hot fluid temperature, flowing through the annulus, and T f,(int.,ext) the cold fluid temperature, which could be the returning gas, or the evaporated helium gas.To model the heat transfer, and define the heat exchanger main parameters, the correlations shown in Table V are used.
Based on these correlations the heat exchanger was designed.Table VI shows the main dimensions.

IV. CONCLUSION
Magnetic refrigeration is considered to have the potential to increase the efficiency of current state of the art cryocoolers increasing their competitiveness.In this paper we presented the design and validation of a test stand to understand and characterize the thermal behavior of magnetocaloric materials at cryogenic temperatures, which will allow the development of optimized magnetic refrigerators.The system components are currently being manufactured, and the first results are expected to be obtained in the following months.

Fig. 1
Fig. 1 Schematic of apparatus and data acquisition system.The portion enclosed by dashed lines are inside a cryostat, which is shown at the right side of the figure.

Fig. 3 .
Fig. 3. Entropy of GGG with temperature and magnetic field.

TABLE I RANGE
OF VALUES FOR TEST STAND COMPONENT VARIABLES

TABLE II MAIN
ELECTROMAGNETIC PROPERTIES OF CRISA MAGNET

TABLE III HELIUM
CIRCUIT PUMPING COMPONENTS

TABLE V CORRELATIONS
FOR EXTERNAL AND INTERNAL FORCED CONVECTION TABLE VI THERMAL DESIGN VALUES FOR THE COUNTERFLOW HEAT EXCHANGER