Magnetocaloric material based on NdPrFe17 with improved properties

Abstract

The instant invention relates to a magnetocaloric material based on NdPrFe.sub.17 melt-spun ribbons. This material has improved properties when compared with other similar magnetocaloric (MC) materials since it has an enhanced refrigeration capacity in the room temperature range due to its broader magnetic entropy change as function of the temperature curve. This material is useful as magnetic refrigerant as a part of magnetocaloric refrigerators.

Claims

1. A magnetocaloric material comprising: a NdPrFe.sub.17 melt spun ribbon; wherein said magnetocaloric material is a nanocrystallites phase surrounded by an intergranular amorphous phase; wherein the magnetocaloric material is adapted to be used as a magnetic refrigerant.

2. The magnetocaloric material according to claim 1, wherein each element is in stoichiometric proportions.

3. The magnetocaloric material according to claim 1, wherein the magnetocaloric material shows two successive second-order ferromagnetic phase transitions.

4. The material according to claim 3, wherein said transitions are 303 and 332 K.

5. The material according to claim 3, wherein said transitions come from a rhombohedral Th.sub.2Zn.sub.17-type nanocrystallites and a minor amorphous intergranular phase.

6. The material according to claim 1, wherein said magnetocaloric material has a magnetic entropy change curve with a working temperature range T.sub.FWHM of 84 K at .sub.oH=2 T.

7. A method of manufacture a magnetocaloric NdPrFe.sub.17 alloy, according to claim 1 comprising the step of: melt-spinning the alloy to form a ribbon having a two phase microstructure including a nanoscale crystalline phase and an amorphous phase.

8. The method according to claim 7, wherein the melt spinning step is includes a rapid solidification in which the ribbons are form by ejecting a molten metallic alloy onto a rotating copper wheel in Ar atmosphere.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1A shows a graph with experimental (red circles) and calculated (black line) X-ray powder diffraction pattern for as-quenched (aq) NdPrFe.sub.17 alloy ribbons (CuK.sub. radiation). The difference line is depicted at the bottom of the figure. The second series of vertical green bars corresponds to the crystal structure of the impurity bcc a-Fe phase (4 wt. %); the vertical arrow points to its more intense Bragg reflection.

(2) FIG. 1B shows a graph with the Temperature dependence of magnetization under a static magnetic field of 5 mT (red curve) and 5 T (black curve). The vertical arrows point to the magnetic transition of the 2:17 rhombohedral phase and the secondary amorphous phase. Top inset: dM/dT vs. T curve at 5 mT. Bottom inset: low-field M(T) curve between 320 and 400 K.

(3) FIG. 2A shows a typical scanning electron microscope (SEM) micrographs of the ribbons cross-section.

(4) FIGS. 2B-2D show transmission electron microscope (TEM) images of the alloy ribbons collected at different magnifications. In (c), a high resolution TEM image shows the lattice planes of a 2:17 nanograin; the Fourier transform of the square indicated area is shown in the inset. The spots are indexed according to the structure used for Rietveld refinement of FIG. 1A. FIG. 2D shows a high-resolution image shows that nanoparticles are surrounded by a disordered (amorphous) intergranular phase; the corresponding Fourier transform of the square is shown in the inset of the image.

(5) FIG. 3 shows a graph with the temperature dependence of the magnetic entropy change S.sub.M(T) for magnetic field changes of 1.5 and 2.0 T for as-solidified NdPrFe.sub.17 alloy ribbons. For the sake of comparison the curves exhibited by Pr.sub.2Fe.sub.17 bulk alloys are plotted. Inset: normalized temperature dependence of the magnetic entropy change as a function of T/T.sub.C for as-solidified NdPrFe.sub.17 alloy ribbons compared to the curves for bulk Pr.sub.2Fe.sub.17 alloys. The horizontal arrows point to the full-width at half-maximum of the curves. The broadening of the curve observed for the fabricated alloy ribbons is due to the presence of the secondary intergranular amorphous phase.

(6) FIG. 4 shows a graph with the refrigerant capacities RC-1, RC-2, and RC-3 as a function of the magnetic field change for as-solidified NdPrFe.sub.17 alloy ribbons. Inset: field dependence of the temperatures T.sub.hot and T.sub.cold that define T.sub.FWHM (i.e., the full-width at half-maximum of the S.sub.M(T) curve).

(7) FIG. 5 shows a graph with the initial and demagnetization curves in the first quadrant measured at 278 K up to .sub.oH=2 T. Inset: low-field region of the curves.

DETAILED DESCRIPTION OF THE INVENTION

(8) The magnetocaloric material of the invention is made from alloy ribbons of nominal composition NdPrFe.sub.17 in stoichiometric proportions produced by rapid solidification using the melt spinning technique. Samples were produced under a highly pure Ar atmosphere from pure metallic elements (99.9%).

(9) Alloy Constitution

(10) Energy dispersive spectroscopy analyses revealed that the starting chemical composition, namely NdPrFe.sub.17, was well reproduced in the as-quenched (aq) ribbon samples. X-ray diffraction (XRD) analysis [FIG. 1A] shows that the rhombohedral Th.sub.2Zn.sub.17-type crystal structure [space group R-3m with unit cell parameters a=8.553(3) and c=12.543(1) , and cell volume V=794.7(1) .sup.3] is the major phase formed in the as-solidified ribbons. It must be noted that the XRD pattern exhibits low intensity and broad diffraction lines suggesting that the size of crystallites in the samples is small. In comparison with (Pedro Gorria, et al., J. Phys D: Appl. Phys., Vol. 41 (2008) 192003; Pedro Gorria, et al., Acta Materialia, Vol. 57 (2009) 1724-1733; Pablo lvarez, et al., J. Phys.: Condens. Matter Vol. 22 (2010) 216005.), wherein the 2:17 phase is only formed in bulk R.sub.2Fe.sub.17 alloys with R=Nd or Pr, after a long thermal annealing (i.e. several days) at a temperature above 1273 K, it is worthy of mention that in the as-solidified alloy ribbons fabricated the 2:17 phase forms after a one-step rapid solidification process (i.e., as-solidified ribbons were not thermally annealed). This difference is relevant for the purposes of the instant invention, since a one-step process is a competitive advantage towards fabrication costs and energy saving. The low-field temperature dependence of the magnetization M(T), shown in FIG. 1B, reveals that the 2:17 phase of the instant invention shows a Curie temperature of 303 K; in addition, this phase coexists with a secondary magnetic phase having a broad magnetic transition located at 332 K (see top inset of FIG. 1B, which shows the dM/dT vs. temperature curve). Hence, in the obtained alloy ribbons two magnetic phases coexist.

(11) FIG. 2A and its inset show typical low-magnification SEM micrographs of the ribbons cross-section. From these images, we can estimate that the ribbon thickness is around 20 m and also that the ribbon morphology at this length scale consists of different shaped entities with average size of tens of nanometers. High-resolution transmission electron microscopy (HRTEM) observations (FIGS. 2B-2D) show that NdPrFe.sub.17 alloy ribbons are nanostructured. As observed in FIGS. 2B and 2C, the ribbons are composed of nanograins whose size roughly varies between 7 and 15 nm. It must also notice that nanograins are surrounded by an intergranular phase. The fast Fourier transform (FFT) patterns for both, an individual 2:17 nanograin and the intergranular surrounding phase are given in the insets of FIGS. 2C and 2D, respectively. Hence, a two-phase magnetic nanocomposite system is formed in the NdPrFe.sub.17 ribbons (consisting of 2:17 nanoparticles surrounded by an intergranular amorphous phase). The HRTEM images given in FIGS. 2C and 2D provide a more detailed view of the ribbon morphology at the nanometer length scale. As FIG. 2C shows, the dark granular regions observed in FIG. 2B are individual nanocrystalline grains for which well-defined lattice planes can be observed. The selected areas Fourier transform patterns for both regions (shown in the insets of both figures) further confirm the amorphous nature of the intergranular region as well as the crystallinity of the nanograins. The Fourier transform shown in the inset of FIG. 2C shows a NdPrFe.sub.17 crystal in [010] orientation. In contrast, FIG. 2D puts in evidence the atomic disorder of the intergranular phase. Hence, a two-phase magnetic nanocomposite system is formed in the ribbons due to the fast solidification procedure and consists of NdPrFe.sub.17 nanocrystals surrounded by a thin intergranular amorphous phase, as it was previously presumed from the XRD pattern and low-field M(T) analysis.

(12) Magnetocaloric Properties

(13) The magnetocaloric properties of the ribbons produced were evaluated from the magnetic entropy change as a function of the temperature curves, S.sub.M(T). They were obtained by numerical integration of the Maxwell relation

(14) $S_M\left(T,{}_{o}H\right)={}_o{}_0^{{}_o{H}_{\mathrm{ma}x}}{\left[\frac{M\left(T,{}_o{H}^{}\right)}{T}\right]}_{{}_o{H}^{}}d{H}^{}$
from a set of isothermal magnetization curves M(.sub.oH) measured up to a maximum applied magnetic field .sub.oH.sub.max of 2 T. The magnetic field was applied along the major length of the ribbon samples to minimize the demagnetizing field effect. The refrigerant capacity RC, which measures the thermal efficiency of a magnetocaloric material in the energy transfer from cold to hot reservoirs for an ideal thermodynamic cycle, was estimated using the following the three following methods: RC-1=|S.sub.M.sup.peak|T.sub.FWHM, RC-2=.sub.T.sub.cold.sup.T.sup.hot[S.sub.M(T)].sub..sub.o.sub.H dt, and RC-3 by maximizing the product |S.sub.M|T below the S.sub.M(T) curve (usually referred to as the Wood and Potter method) [M. E. Wood and W. H. Potter, Cryogenics, Vol. 25 (1985) 667-683]. In the case of RC-1 and RC-2, T.sub.hot and T.sub.cold are the temperatures that define the temperature interval T.sub.FWHM of the full width at half maximum of the S.sub.M(T) curve (i.e., T.sub.FWHM=T.sub.hotT.sub.cold). The latter defines the working temperature interval of the magnetic material as magnetocaloric refrigerant.

(15) FIG. 3 shows the S.sub.M(T) curves for a magnetic field change of 1.5 and 2.0 T for the fabricated alloys ribbons; for the sake of comparison the curves reported in reference 2 for bulk polycrystalline Pr.sub.2Fe.sub.17 alloys are plotted. Notice that NdPrFe.sub.17 aq ribbons exhibit a lower peak value of the magnetic entropy change S.sub.M.sup.peak but a well broader S.sub.M(T) curve [the normalized S.sub.M(T)/S.sub.M.sup.peak versus T/T.sub.C curves are given in the inset of FIG. 3; the broadening of the magnetic entropy change curve for the nanocomposite ribbons is 78% higher than Pr.sub.2Fe.sub.17 alloys at .sub.oH=2 T]. The S.sub.M(T) the field dependence of T.sub.hot and T.sub.cold is given in inset of FIG. 4; notice that T.sub.FWHM embraces the room temperature interval. Despite the lower |S.sub.M.sup.peak| of the ribbon samples they show a larger refrigerant capacity and working temperature range in comparison with bulk Pr.sub.2Fe.sub.17 alloys (an increase in RC-1 and RC-2 at .sub.oH=2 T of approximately 17 and 12%, respectively, is found). The reversible character of the magnetocaloric effect was confirmed by measuring the first quadrant of the hysteresis loop of ribbon samples at T.sub.cold (i.e., 278 K), which is depicted in FIG. 5. The sample shows an intrinsic coercivity .sub.oH.sub.C of 3 mT, a remanence-to-saturation ratio of 0.2 and a negligible hysteresis loss at this temperature (0.007 J kg.sup.1); given by the area enclosed between the virgin and demagnetization curve in the first quadrant).

(16) Hence, within the operating temperature range T.sub.FWHM, no significant hysteresis losses were measured in agreement with the second-order character of the phase transitions. As a result, these two-phase nanostructured amorphous NdPrFe.sub.17 melt-spun ribbons yield to a reinforcement of the refrigerant capacity of the system owing to the Curie temperature of both phases are close to each other.

(17) The magnetocaloric properties of both materials, i.e., NdPrFe.sub.17 melt-spun ribbons and bulk Pr.sub.2Fe.sub.17 alloys, for magnetic field changes of 1.5 and 2.0 T are compared in Table I. A summary of the magnetocaloric properties of the dual-phase NdPrFe.sub.17 nanocomposite is given in Table II.

(18) TABLE I shows the maximum magnetic entropy change |S.sub.M.sup.peak|, useful working temperature range (T.sub.FWHM=T.sub.coldT.sub.cold), and refrigerant capacities RC-1 and RC-2, for a magnetic field change of 1.5 and 2.0 T for as-solidified NdPrFe.sub.17 alloy ribbons compared to the reported values for bulk Pr.sub.2Fe.sub.17 alloy [Pedro Gorria, et al., Acta Materialia, Vol. 57 (2009) 1724-1733].

(19) TABLE-US-00001 TABLE I T.sub.C .sub.oH |S.sub.M.sup.max| T.sub.cold T.sub.hot T.sub.FWHM RC-1 RC-2 Sample (K) (T) (J kg.sup.1 K.sup.1) (K) (K) (K) (J kg.sup.1) (J kg.sup.1) Aq NdPrFe.sub.17 303 1.5 1.6 280 357 77 126 97 2.0 2.1 278 362 84 175 135 Pr.sub.2Fe.sub.17 bulk 285 1.5 2.6 265 305 40 105 80 2.0 3.2 263 310 47 150 110

(20) TABLE II shows a peak magnetic entropy change |S.sub.M.sup.peak|, RC-1, RC-2, T.sub.FWHM, T.sub.cold, T.sub.cold, RC-3, T.sup.RC-3, and T.sub.hot and T.sub.cold related to RC-3 for as-solidified NdPrFe.sub.17 alloy ribbons.

(21) TABLE-US-00002 TABLE II NdPrFe.sub.17 - as quenched ribbons .sub.oH (T) 0.5 1.0 1.5 2.0 |S.sub.M.sup.peak| (J kg.sup.1 K.sup.1) 0.6 1.1 1.6 2.1 RC-1 (J kg.sup.1) 36 79 126 175 RC-2 (J kg.sup.1) 26 60 97 135 T.sub.FWHM (K) 57 69 77 84 T.sub.hot (K) 344 352 357 362 T.sub.cold (K) 287 283 280 278 RC-3 (J kg.sup.1) 18 41 67 95 T.sup.RC-3 (K) 63 129 132 134 T.sub.hot (K)* 347 372 376 379 T.sub.cold (K)* 284 243 244 245 *related to RC-3.

(22) The magnetocaloric nanocomposite obtained in melt-spun NdPrFe.sub.17 alloy ribbons exhibits two successive second-order ferromagnetic phase transitions that come from the rhombohedral Th.sub.2Zn.sub.17-type nanocrystallites and a minor amorphous intergranular phase, respectively. The dual-magnetic phase character of the system gives rise to a broad magnetic entropy change curve with a well larger working temperature range of 84 K and a higher refrigerant capacity around room temperature if compared with their crystalline bulk counterpart.

(23) It must be noted that T.sub.FWHM at 2 T is superior to other magnetic refrigerants in the room-temperature range including the benchmark MC material Gd (T.sub.FWHM for Gd is typically of approximately 40-45 K).

(24) The use of melt spinning technique avoids the use of a prolonged thermal annealing at high temperatures to produce the 2:17 phase as major phase.

EXAMPLES

(25) Method for Preparing the Magnetocaloric Material

(26) The magneto caloric material of the invention (ribbons), with nominal composition NdPrFe.sub.17, was produced by rapid solidification using a melt spinning system at a linear speed of the copper wheel of 20 ms.sup.1 from bulk pellets previously produced by arc melting. As raw materials, pure metallic elements were used (99.9%). Both the arc melted starting alloys and the melt-spun ribbons were obtained under a highly pure Ar atmosphere.

(27) Characterization Methods

(28) X-ray diffraction (XRD) patterns of finely powdered ribbon samples were collected with a Bruker AXS model D8 Advance X-ray powder diffractometer using CuK.sub.alpha radiation (=1.5418 , 202100; step increment 0.01). The Rietveld analysis of the diffraction data was carried out with the Fullprof suite package. Microstructure and elemental composition were investigated using a Helios FEI Dual beam Helios Nanolab FIB scanning electron microscope (SEM) equipped with and energy dispersive spectroscopy (EDS) system. SEM images were taken on the cross-section of cleaved ribbon samples; the granular microstructure of many ribbons was analysed. The images showing the nanostructure of the samples were collected in a FEI Tecnai high-resolution transmission electron microscope (HRTEM). For TEM examination a tiny amount of finely grounded ribbons were put into a vial with ethanol. The vial was sonicated in an ultrasonic bath for 10 min to form a suspension.

(29) A drop of the upper part of the suspension was applied to a copper grid that was dried in air).

(30) Magnetic measurements were performed by vibrating sample magnetometry in a 9 Tesla Quantum Design PPMS EverCool-I platform. The magnetic field .sub.oH was applied along the ribbon axis (i.e., the rolling direction) to minimize the demagnetizing field effect. The low-field (5 mT) and high-field (5 T) magnetization as a function of temperature, M(T), curves were measured between 100 and 400 K. The magnetic transition temperatures were obtained from the minimum of the dM/dT(T) curve measured under .sub.oH=5 mT. In order to determine the S.sub.M(T) curve from numerical integration of the Maxwell relation

(31) $\left(i.e.,S_M\left(T,{}_{o}H\right)={}_o{}_0^{{}_o{H}_{\mathrm{ma}x}}{\left[\frac{M\left(T,{}_o{H}^{}\right)}{T}\right]}_{{}_o{H}^{}}d{H}^{}\right),$
a set of isothermal magnetization curves, M(.sub.oH), was measured in the temperature range of 200-400 K with a T step of 5 K up to a maximum applied magnetic field of 2 T. With the aim of minimizing the error in the calculation of S.sub.M, the magnetization was measured for a large number of selected values of .sub.oH at each temperature. The values of RC-1, RC-2, and RC-3 were obtained from the criteria stated above (in the section of magnetocaloric properties).