Project

Permanent Magnets

Permanent magnets are indispensable elements of our daily lives, crucial for a wide range of key technologies in the pursuit of carbon neutrality. They find application in electric vehicles, wind turbines, and various domestic appliances. However, the current European production of permanent magnets heavily relies on costly imported rare-earth materials, which pose challenges in terms of recycling and have substantial environmental impacts. Yet the European Union in general is 100% dependent on foreign suppliers for 14 out of 27 crucial raw materials. The European Innovation Council is now supporting a Europe-wide research project on new magnetic materials that do not require these critical raw materials.

Magnets, also known as permanent magnets, are materials capable of retaining magnetic fields even after being magnetised. Their history dates back to the discovery of Lodestones in the 6th century BC, and since then, they have found diverse applications. The term “magnet” originates from Magnesia, Greece, where the first lodestone was mined. One of the earliest recorded uses of permanent magnet material was in the creation of the “south pointer” or compass, which essentially involved shaping a lodestone into a spoon-like form.

The maximum energy product (BHmax) of permanent magnets is a critical parameter that determines their overall performance and versatility. It serves as a key indicator of a magnet’s ability to store and release magnetic energy, making it an important factor in various applications. By understanding the concept of maximum energy product, it is possible to design and optimise magnet systems for diverse fields such as electronics, transportation, renewable energy, and more.

Magnetic Refrigeration

Energy efficient cooling is becoming one of the most critical challenges for our society. With energy demand for refrigeration and air conditioning expected to soar in the coming decades, addressing this issue is of paramount importance. All refrigeration technologies share a common feature: the use of a refrigerant that undergoes a state change, thereby changing the temperature. To meet global climate change targets, stringent regulations such as the EU’s F-Gas Regulation and the United Nations Environment Programme’s Kigali Amendment have been implemented, restricting the use of high-GWP refrigerants in conventional gas compression technology. However, the available alternatives are often problematic because they are either toxic, explosive, or inefficient. In this context, magnetic refrigeration is emerging as the only viable alternative cooling technology that not only eliminates the need for harmful refrigerant gases, but also significantly reduces energy requirements and, consequently, carbon dioxide emissions.

The magnetocaloric effect (MCE) is a phenomenon that results from the partial alignment of the magnetic moments of a material when subjected to an external magnetic field. This alignment results in a decrease in the entropy of the magnetic moments. When a magnetic field is applied under adiabatic conditions, the decrease in magnetic entropy is offset by an increase in crystal lattice entropy, causing the material to heat up (step 1). The excess heat is then removed, returning the material to its initial temperature, but now in a magnetised state (Step 2). When the magnetic field is removed, the opposite effect occurs and the material cools (Step 3). During this cooling phase, the material can absorb heat from a target being cooled (Step 4), and the cycle can be restarted. By utilising the magnetocaloric effect in a cyclic operation, it becomes possible to design refrigeration machines capable of efficient and effective cooling.

Compositionally Complex Alloy Design Concept

In the quest for materials with enhanced properties and performance, the concept of Compositionally Complex Alloys (CCAs) has emerged as a promising frontier in materials science and engineering. This concept is derived from the high-entropy alloys (HEAs) design strategy, which is a new alloying strategy that combines multiple principal elements in relatively high (often equiatomic) concentrations. The CCA concept does not have a strict limitation on the near equal ratio of the different elements, but provides a higher design freedom and enables developing sustainable magnets with excellent mechanical properties, corrosion resistance and shapeability. CCAs represent a departure from traditional alloys by incorporating multiple key elements in nearly equal proportions, resulting in intricate and unique atomic structures. This intentional complexity enables CCAs to exhibit remarkable properties such as superior strength, enhanced corrosion resistance, and exceptional thermal stability. The design and synthesis of compositionally complex alloys has become a subject of intense research, driven by the desire to develop new materials with tailored functionalities for a wide range of applications, from aerospace and automotive to renewable energy and electronics.

Computational Material Design and Machine Learning

The success of Machine Learning (ML) lies in its remarkable ability to describe intricate patterns of materials, and these sophisticated ML models can be systematically optimised through learning from the vast data of HEAs. These exceptional features position ML as a highly promising tool to tackle the challenges faced by theoretical modelling in the context of HEAs. One of the most significant advantages of ML is its impressive calculation speed, which surpasses that of Density Functional Theory (DFT) and approaches the efficiency of empirical potentials. This heightened efficiency empowers researchers to simulate materials containing millions of atoms with near DFT accuracy, significantly expanding the possibilities beyond the limitations of conventional DFT methods. The aforementioned merits of ML have paved the way for a groundbreaking data-driven paradigm in HEAs research, as evident in its triumphant application to the design of High-Entropy Invar alloys, showcased in Figure 1.

Figure 1. An active learning framework for the targeted composition design and discovery of HEAs, which combines machine learning models, DFT calculations, thermodynamic simulations and experimental feedback. The iteration is repeated until the discovery of Invar alloys [1].

We will use micromagnetics and multi-objective optimization for materials design. Micromagnetics is a field of physics that deals with predicting magnetic behaviours at sub-micrometer length scales. It involves using software to generate synthetic microstructures based on data from structural investigations. These microstructures are then used to simulate the influence of the microstructure on magnetic and magneto-mechanical properties. The goal is to create a model for predicting the properties of magnetic materials by combining data from simulations and experiments. The model uses methods such as partial least square regression to make fast predictions based on chemical composition and microstructural features. A genetic algorithm is used to optimise materials according to different target properties.

Reference:

[1] Rao Z, Tung P Y, Xie R, et al. Machine learning–enabled high-entropy alloy discovery[J]. Science, 2022, 378(6615): 78-85.