Design of efficient magnetocaloric materials for energy conversion

The magnetocaloric effect (MCE) is a magneto-thermodynamic phenomenon in which a temperature change of a material is caused by exposing the material to a changing magnetic field under adiabatic conditions. There are two main applications based on the MCE. One application is magnetic refrigeration, which can expel heat in a magnetic field cycle. Another application is magnetic energy conversion in thermomagnetic motors/generators, which can transfer waste heat into kinetic/electric energy. Gadolinium metal is the standard reference material for the application of the MCE. However, it has a limited MCE with a second-order magnetic transition. Several intermetallic material systems with first-order magnetic transition resulting in a giant MCE have been discovered, including La(Fe,Si)13 based alloys, MnFeP(As, Ge, Si) alloys and Ni-Mn-based Heusler alloys. To design a magnetocaloric material that is suitable for applications, first of all, requires an estimated recipe, which can be obtained from the phase diagram. Secondly, an appropriate synthesis route should be chosen. Thirdly, the stoichiometry of the material should be optimised to avoid impurity phases. For the energy conversion applications, the desired material should preferentially be in the vicinity of the border between a first-order magnetic phase transition (FOMT) and a second-order magnetic phase transition (SOMT). If it is a FOMT or SOMT, the formula can be adjusted by changing the heat treatment, the element ratios and introducing new elements, until the transition is close to the critical point (CP). Finally, the transition temperature needs to be checked to see it is in the designed working temperature range. If not, the recipe needs to be adjusted until an optimised material is found. Experimental diagrams of the ferromagnetic transition temperature (TC) and the thermal hysteresis as a function of composition were constructed in the (Mn,Fe)2(P,Si) system as a guide to estimate suitable compositions for applications. The structure change across the magnetic phase transition is coupled with the thermal hysteresis of the magnetic transition in the experimental diagram. Both Mn-rich samples and Fe-rich samples with a low Si concentration were found to show a low hysteresis that can form promising candidates for applications in a thermomagnetic motor. The effect of V substitution for Fe is investigating in the Mn0.7Fex-zVzP0.6Si0.4 alloys. The (Mn,Fe)1.91(P,Si) stoichiometry was chosen as a starting point to obtain the smallest impurity content. For an increasing V content the a axis expands and the c axis shrinks (together with the c/a ratio), whereas the unit-cell volume remains about constant. The ferromagnetic transition temperature TC decreases with increasing V content. In the Mn0.7Fe1.18V0.03P0.6Si0.4 compounds, 93% of saturation magnetisation at 5 K was reached in an applied magnetic field of 0.5 T, which makes this compound a promising candidate for low-field applications. The heat treatment clearly affects the amount of the impurity phase, and thereby the composition of the main phase. In this case, oven-cooled samples contain a larger impurity phase fraction than the quenched samples, which results in a lower transition temperature. The currently applied methods to classify FOMT and SOMT materials were applied and compared using a series of samples Mn13Fe0.7P1-ySiy (y = 0.4, 0.5 and 0.6). The FOMT samples are easy to categorise. Every criterion shows that y = 0.4 and 0.5 sample is FOMT materials. However, the SOMT and CP samples are problematic. In this thesis, different criteria were found to result in different conclusions for the y = 0.6 sample. From the latent heat, the y = 0.6 is predicted to undergo a FOMT. From the XRD data and the field dependence of TC, the y = 0.6 sample is right on the CP. However, based on the Arrott plots, the gradual field dependence of the entropy change and the newly proposed field exponent n, the y = 0.6 sample is a SOMT material (but in close proximity to the CP). The structural, magnetic and electronic properties of LaFe11.8-bCobSi1.2 (b = 0.25, 0.69 and 1.13) compounds are studied. With increasing Co content, the material is tuned from a FOMT towards a SOMT, TC increases, and the thermal hysteresis remains neglectable. In the unit cell, the most remarkable change in bond length is between the 8b and 96i sites and for one of the bonds between two neighbouring 96i sites. The negative thermal expansion across the transition correlates with the angle change in the orientation of the cage formed by the atoms on the 96i sites within the cubic unit cell. The experimental electron density maps reveal how the cage rotates within the cubic primitive cell. The samples with a smaller Co content show a larger change in the electron density compared to the sample with the highest Co content when TC is crossed. The choice of synthesis method plays an important role in the physical properties of the prepared materials. For lab-scale samples, the most common way to synthesise (Mn,Fe)2(P,Si) compounds is ball milling. For Ni-Mn based Heusler alloys, the most common synthesis route is arc-melting. In this thesis, ball milling was applied to synthesise Ni-Mn based Heusler alloys. The advantage of ball milling is that the annealing time can be shortened. Based on the optimised sample fabrication, the maximum magnetisation can be tuned by adjusting the Ni/Mn and Mn/Sn ratios. Introducing small amounts of cobalt and aluminium leads to a significant increase in the magnetisation.