MnxGa: Understanding a magnet in the hope of designing better magnets
Invited
Abstract
Generation and usage of electric power both depend on electric motors (generators are electric motors run backwards) and “electric” motors are really magnetic motors as they rely on the Lorentz force (I × B) for their action. Generating the magnetic field is key to the efficiency of the machine. For large-scale applications, copper wire wound on an iron core has been the method of choice. Cheap, reliable, but power hungry. The chance discovery of Nd2Fe14B in the late 80s led to motors with permanent magnet cores that draw zero power. Improvements in motor designs are now migrating to generators, trading up-front capital costs for long-term efficiency gains and putting pressure on permanent magnet supplies.
Nd2Fe14B is definitely a mature product and the search for its successor is following two distinct paths. The traditional shake’n’bake approach where “promising” combinations of elements are mixed in the hope of finding something new is being supplanted by first-principles density functional theory (DFT) calculations supported by ever more powerful computers. Designing a material from scratch by exploring possibilities on a computer before trying to make them. While a DFT search may ultimately be more efficient, the approach needs to be validated against known systems. We need to know that we are getting the properties right.
Tetragonal MnxGa (1.2 < x 1.8) shows how this might work. The neutron contrast between Mn and Ga means that we can fit the crystal structure accurately, and then determine the magnetic structure. DFT calculations confirm the crystal and magnetic structures (atomic positions, moments, coupling) and even yield quantitative agreement with measured anisotropy fields. Departures at low Mn compositions have been traced to disorder − the real materials are not “perfect”. Adding disorder to the model removes the disagreement, further validating the calculations.
As confidence in the models improves, searches for real “designer magnets” become possible.
Nd2Fe14B is definitely a mature product and the search for its successor is following two distinct paths. The traditional shake’n’bake approach where “promising” combinations of elements are mixed in the hope of finding something new is being supplanted by first-principles density functional theory (DFT) calculations supported by ever more powerful computers. Designing a material from scratch by exploring possibilities on a computer before trying to make them. While a DFT search may ultimately be more efficient, the approach needs to be validated against known systems. We need to know that we are getting the properties right.
Tetragonal MnxGa (1.2 < x 1.8) shows how this might work. The neutron contrast between Mn and Ga means that we can fit the crystal structure accurately, and then determine the magnetic structure. DFT calculations confirm the crystal and magnetic structures (atomic positions, moments, coupling) and even yield quantitative agreement with measured anisotropy fields. Departures at low Mn compositions have been traced to disorder − the real materials are not “perfect”. Adding disorder to the model removes the disagreement, further validating the calculations.
As confidence in the models improves, searches for real “designer magnets” become possible.
–
Presenters
-
Dominic Ryan
Physics, McGill University, McGill University
Authors
-
Dominic Ryan
Physics, McGill University, McGill University
-
Zaven Altounian
Physics, McGill University
-
Xubo liu
Physics, McGill University
-
Chad Boyer
Canadian Neutron Beam Centre, Chalk River Laboratories