Metal grows more conductive as it cools, until a few stray magnetic atoms turn the trend around and resistance begins to climb below a threshold known as the Kondo temperature. Physicists have long understood why this happens, yet until now they could only approximate its exact behavior. Researchers at Caltech and Yale have, for the first time, calculated the effect precisely in a real material, a step that hints at a future in which the quantum behavior of complex materials can be foretold by computation before anyone sets up an experiment.
Caltech and Yale Team Predicts the Kondo Effect From First Principles, Without Lab Data
Predicting a quantum material without first running the experiment
Mark
So what's actually new here? The Kondo effect has been known for decades.
Mimi
The cause was known. What wasn't possible was calculating it precisely for a real material from fundamental physics alone. Before, people leaned on experiments and simplified models.
Mark
And how did they get around the complexity?
Mimi
They treated a single magnetic atom like a large molecule, using chemistry techniques, and then applied basic quantum laws with mathematical shortcuts and a lot of computing time.
Luke
The headline number is 100 times more accurate. What does that hold up to?
Mimi
The source says in some cases, compared with current simplified modeling. It isn't a blanket figure.
Luke
So we don't know the typical improvement.
Mimi
Not from this reporting, no.
Mark
Why does it matter beyond this one effect?
Mimi
Chan calls these materials a prototype on the way to high-temperature superconductors and quantum magnets, which could matter for power grids and computing.
Luke
That's the aspiration, though. They only studied the simplest metal and magnetic atom pairings, and nothing here shows the method scales to those harder materials. The researchers say scaling is the next step.
Mark
So it's a proof of concept.
Mimi
Yes, and a validated one, checked against lab measurements.
The Pulse
- Metal that should be growing easier to conduct electricity as it cools suddenly grows stubborn when magnetic impurities are present.
- The many-body problem at the heart of the effect is so tangled that even the most powerful supercomputers have struggled to predict it.
- Researchers borrowed ab initio techniques from chemistry, modeling a single magnetic atom in full and refining the results through repeated calculation.
- Validated against laboratory measurements, the method proved up to 100 times more accurate than simplified models, and it now stands as a prototype for predicting high-temperature superconductors and quantum magnets without prior lab data.
Metal grows more conductive as it cools, until a few stray magnetic atoms turn the trend around and resistance begins to climb below a threshold known as the Kondo temperature. Physicists have long understood why this happens, yet until now they could only approximate its exact behavior. Researchers at Caltech and Yale have, for the first time, calculated the effect precisely in a real material, a step that hints at a future in which the quantum behavior of complex materials can be foretold by computation before anyone sets up an experiment.
Cool an ordinary metal and it conducts electricity more easily, since electrons slip past less thermal agitation. Add a few magnetic atoms, though, and below a certain temperature, the Kondo temperature, the trend reverses and resistance starts to rise again. Physicists have understood the cause for a long time: electrons gather around the magnetic impurities, cancelling their magnetism while blocking the flow of current. What they could never do was calculate the effect with full precision.
Researchers at Caltech and Yale now report that they have done exactly that in a real material, publishing the work in Science. Earlier efforts leaned on experiments and simplified models, and neither matched the exactness of the new approach. The obstacle is a many-body problem, where temperature, electron spin, and magnetism entangle at tiny scales, a problem described as beyond even the most powerful supercomputers.
The team modeled a single magnetic atom in full, using ab initio techniques borrowed from chemistry and treating the atom almost as a large molecule. They applied the basic laws of quantum physics, used mathematical shortcuts, and repeated the calculations many times. When they checked their numbers against laboratory measurements, the agreement held, and in some cases the method was 100 times more accurate than current simplified models. Caltech physicist Garnet Chan said some complicated materials can now be predicted purely through computation.
For this first test the researchers chose the simplest metal and magnetic atom pairings, which Chan called a baby step. The larger aim is to reach high-temperature superconductors and quantum magnets, materials that could reshape power grids and computing and that carry the most intricate electron behavior. Caltech chemist Linqing Peng expressed hope that faithful quantum descriptions of real materials are coming within reach. Much work remains, but the next step is to scale the method to more complicated materials, so that scientists can study quantum behavior without first needing lab data.
Cool a piece of ordinary metal and it gets better at carrying current. With less thermal energy in the way, electrons move more smoothly, and resistance falls. Add a few magnetic atoms to that metal, though, and below a certain threshold, called the Kondo temperature, the trend reverses and resistance begins to climb again.
That reversal is the Kondo effect, one of the more important phenomena in quantum physics. Physicists have understood its cause for a long time. Electrons crowd around the magnetic impurities in the metal. In doing so they cancel out the magnetism of the atoms and, at the same time, impede the flow of electricity.
What they have not been able to do is calculate it with full precision. Researchers at the California Institute of Technology (Caltech) and Yale University now report that they have, in a real material, and the work has been published in Science. Until now, pinning down the effect has meant leaning on real-world experiments and simplified models, neither of which produced predictions as exact as the new method's.
The difficulty lies in what physicists call a many-body problem. Temperature, electron spin, and magnetism all interact at extremely small scales as the effect takes hold. The reporting describes the full problem as beyond even the most powerful supercomputers, which is why predicting when the resistance will flip, for different pairings of metal and magnetic impurity, has been so tricky.
The team took an ab initio approach, meaning one built from first principles. They borrowed techniques from chemistry to model a single magnetic atom in full, without simplification, treating it as though it were a larger molecule. Then, using mathematical shortcuts and a great deal of computer time, they applied the basic laws of quantum physics to that atom and repeated the calculations many times to tighten their predictions.
To test the results, they compared their calculated values with laboratory measurements of the Kondo effect. The comparison supported the method's accuracy; in some cases it was 100 times more accurate than current simplified modeling. Caltech physicist Garnet Chan said it is now possible to predict the properties of some complicated materials purely through computation, without referring to experiments.
The researchers chose the most basic combinations of metals and magnetic atoms for this first test. Chan called these materials a baby step, a prototype problem on the way to more complex phenomena such as high-temperature superconductors and quantum magnets. Those are the kinds of materials that promise to change systems like power grids and large-scale computing, and they also tend to have the most complex electron behavior, which is where this approach could help.
Caltech chemist Linqing Peng said faithful predictive quantum descriptions of the full chemical complexity of real materials are coming within reach, and that predicting the material-specific behavior of correlated electrons from first principles is becoming realistic, even in some of the hardest classes of quantum materials.
Much work remains. The next step is to scale the method up to more complicated materials, which would give scientists a way to study quantum behavior without needing lab data first.
Notable Quotes
It is now possible to predict the properties of some complicated materials purely through computation without referring to experiments.— Garnet Chan, physicist, Caltech
These first materials are a baby step, or a prototype problem, on the way to high-temperature superconductors and quantum magnets.— Garnet Chan, Caltech (paraphrased)
Want the full story?
Read the original at ScienceAlert ↗