Four Layers of Nickel, Superconducting Without the Squeeze
A deep dive into one paper: what was known before, what the researchers did, what the numbers say and what they don't.
The paper: Ambient-pressure superconductivity in bulk single crystals of La5Ni4O10 (arXiv)
Chapters
Also linked
Transcript
The script the two AI voices read from. The times are worked out from the script, not heard, so a line can start a few seconds early or late.
Read the whole episode
Cold open
Host 1Can you make a nickel-based crystal that carries electricity with zero resistance while sitting at ordinary air pressure, as a solid lump rather than a film thinner than a soap bubble?
Host 2A group in Stuttgart says yes: they grew crystals of a lanthanum nickel oxide with four nickel layers per block, and once chemically stripped of some oxygen, they superconduct below about eighteen kelvin. That makes it only the second nickelate known to do this in bulk without being squeezed.
Host 1Welcome to the Hastrup Review of Science, where artificial intelligence reads real science, so you don't have to pretend you understood the paper.
Host 2Today it is one paper, on a crystal called La five Ni four O ten, from the Max Planck Institute for Solid State Research. It is a materials chemistry story as much as a physics story.
Why superconductors
Host 1Before the chemistry, can you remind me why anyone hunts for superconductors in the first place?
Host 2Because a superconductor carries current with no loss at all and pushes magnetic fields out of itself, which is what makes hospital scanners and some power cables work. The catch, in the wider picture of the field, is that every known one only does it when very cold.
Host 1And eighteen kelvin, is that cold in everyday terms?
Host 2Very cold: it is about two hundred and fifty five degrees below zero Celsius, so you need liquid helium or a cryocooler. Nobody will put this in a fridge, and the paper does not pretend otherwise.
Host 1So why do physicists get excited about nickel compounds in particular?
Host 2Because of copper, as a matter of history. The copper oxides, or cuprates, found in the late nineteen eighties, still hold the record for superconductivity at ordinary pressure, and nobody fully agrees on why they work. Nickel sits right next to copper in the periodic table, so for decades people hoped a nickel oxide built the same way would superconduct too and help crack the copper puzzle.
Host 1Did that hope ever pay off, or was it all wishful thinking?
Host 2It paid off in pieces. Thin films of a nickel oxide superconducted in twenty nineteen, and in twenty twenty three a nickel oxide with two layers reached about eighty kelvin, but only under huge pressure. The paper sums up the state of play bluntly: nickelate superconductivity has lived either in films held in shape by a substrate, or in samples crushed in a pressure cell.
The layer ladder
Host 1You keep saying layers. What does a layer mean in one of these crystals?
Host 2Picture flat sheets of nickel and oxygen stacked like pages, with a sheet of lanthanum and oxygen acting as a spacer between bundles of pages. The family is called the Ruddlesden Popper series, and the number n tells you how many nickel oxygen sheets are in each bundle before the next spacer.
Host 1And changing n changes how the material behaves?
Host 2Dramatically. With one layer you get a magnetic insulator, two and three layers give metals that superconduct under pressure, and the limit of infinitely many layers is an ordinary metal. The paper points out that n also sets the charge on each nickel atom, so the layer count acts like a dial on the electrons.
Host 1How can simply stacking more sheets change the charge on the nickel atoms?
Host 2It is bookkeeping: the spacer layers and the oxygen fix the total charge, and spreading that over more nickel atoms per bundle shifts the average. In the full oxygen form the nickel charge runs from two plus at one layer toward three plus at infinity, following three minus one over n.
Host 1So where did the ladder stop when people tried to make these as real lumps of crystal?
Host 2At three layers. Four and five layer versions are not stable when you make them the normal way, in a furnace at air pressure; at best they show up as mistakes inside crystals of their neighbours. Almost everything known about them came from thin films, where even basic facts like the oxygen content had to be inferred, and the substrate's strain always muddied the picture.
The open question
Host 1You mentioned a chemically stripped version. What happens to these crystals when you take oxygen out?
Host 2You pull out the oxygen atoms that sit above and below each nickel, so each nickel is left in a flat square of four oxygens, like the cuprates. That is called a square planar nickelate, and its nickel charge becomes one plus one over n.
Host 1Why should that particular number matter?
Host 2Because in films, superconductivity has turned up for layer counts from about four to eight, which is exactly where that formula puts the nickel charge in the range thought to favour pairing. At n equals four, the nickel charge works out to one and a quarter.
Host 1So if it worked in films, what stopped people from just doing it in bulk crystals?
Host 2Several things. Bulk versions of the original nickelate made by adding strontium stayed insulating even up to about fifty gigapascals, the stripping reaction tends to crack crystals into domains, and the bulk two and three layer stripped versions never superconducted. So the open question was where, along this ladder, bulk superconductivity at normal pressure first switches on.
Host 1Was anything known in bulk before this paper?
Host 2Very recently, yes: another group reported the five layer member in bulk, made under high pressure, and found it superconducts after reduction. This paper calls its own four layer crystal the second such case, so it is extending a fresh result rather than starting from nothing.
How they made it
Host 1How do you actually make a crystal that refuses to form at normal pressure?
Host 2You squeeze it while it grows. They used a multianvil press, where several hardened anvils push on a small ceramic octahedron from all sides, and ran it at seven gigapascals, around seventy thousand times atmospheric pressure.
Host 1And at that pressure, what temperature did they need to grow the crystal?
Host 2They heated the sealed platinum capsule to sixteen hundred degrees Celsius, so the mixture melts, then cooled it evenly to fourteen hundred and eighty degrees over eight hours, and then quenched it to room temperature. Only after that did they slowly let the pressure off and pick crystals out of the pellet under a microscope.
Host 1So is pressure the knob that decides whether you get four layers or five?
Host 2No, and that is one of the more useful findings for other chemists. Dropping to five gigapascals still gave the four layer phase; what decided it was the starting recipe. A lanthanum to nickel ratio of six to five gave four layers, and seven to six gave five layers, so pressure opens the door and the recipe picks the room.
Host 1Is there anything special about how they prepared that starting recipe?
Host 2They used a sol gel route: dissolving the ingredients in nitric acid with citric acid, drying it to a gel, burning off the organics, then firing in oxygen at eight hundred degrees. The point is to mix the lanthanum and nickel atoms evenly at the start, which they say is essential for a single clean phase.
Host 1How do they know the crystal really has four layers per block, rather than a jumble of threes and fives?
Host 2X-ray diffraction. The pattern showed sharp, well separated spots with no sign of intergrowth, and the crushed crystals fitted a single phase. Going from four to five layers, the repeat distance along the stack grew by three point eight angstroms, which is what one extra nickel oxygen sheet should add.
Host 1And how big are these crystals, roughly?
Host 2Tiny but real: typically two hundred by two hundred micrometres across and one to two hundred micrometres thick, about the size of a grain of fine sand. Each weighs around two hundredths of a milligram.
Host 1Before any stripping, how did the four layer crystal itself behave electrically?
Host 2Metallic at high temperature, with a resistance that falls roughly in line with temperature, plus a reproducible kink near one hundred and fifty kelvin that looks like the density wave transitions seen in the two and three layer cousins. A density wave is a frozen ripple in the electrons or their spins, and it often competes with superconductivity. The five layer crystal was about ten times less resistive and showed no such kink, so it looks too metallic for that instability.
Taking out the oxygen
Host 1Then how did they take the oxygen out without wrecking the crystal?
Host 2With calcium hydride, a powder that grabs oxygen. They sealed crystals and two hundred milligrams of it in an evacuated quartz ampoule and baked it at three hundred degrees Celsius for one hundred or one hundred and thirty hours, either with crystals lying on the powder or held apart so only the vapour reached them.
Host 1How can they be sure the oxygen actually left, and not just a bit of it?
Host 2Three independent checks. Chemical analysis by electron beam at about twenty spots kept the lanthanum to nickel ratio at five to four, and weighing the crystals while heating them to destruction gave an oxygen count of twelve point nine before and nine point six after, against nominal values of thirteen and ten. The crystal also shrank along the stack by an amount that sits on a straight line with the known two and three layer cases.
Host 1Nine point six instead of ten, does that mean the stripping went slightly too far?
Host 2The authors read it as essentially complete reduction, with the sample close to starting to decompose; the electron beam figure carried an error larger than that gap. What it does tell you is that there is little room between enough and too much, which matters for the results.
Host 1Did the crystals survive in one piece?
Host 2They did, according to diffraction and electron microscope images, which the authors contrast with the heavy domain formation seen when reducing the plain perovskite with no spacer layers at all. One known nuisance remains: small inclusions of nickel oxide turn into specks of magnetic nickel metal, which they had to measure and subtract.
What they found
Host 1So what did they find when they cooled the stripped crystals down?
Host 2Superconductivity, with the same transition near eighteen kelvin in every superconducting crystal, whatever the reduction time, method or sample mass. That sameness is why they treat eighteen kelvin as a property of the material rather than of the recipe.
Host 1How do you even tell that a crystal weighing a hundredth of a milligram has become a superconductor?
Host 2One way is magnetism: a superconductor expels magnetic field, so in a sensitive magnetometer it shows up as a strong negative signal. From that signal they estimate that more than sixty percent of the sample's volume is shielding the field in the best crystals.
Host 1But couldn't a thin superconducting skin on the outside fake that kind of shielding?
Host 2It could, which is why they also measured heat. When a material becomes superconducting in bulk, its heat capacity jumps at the transition, and that jump scales with how much of the material takes part.
Host 1And what did the heat measurement say?
Host 2A clear jump, with the electronic part rising sharply below about eighteen and a half kelvin. Below that, a leftover normal signal implies that about fifty five percent of the sample is superconducting, and the size of the jump is close to what standard theory predicts for that fraction.
Host 1So if I have this right, the magnet test and the heat test independently say roughly half or a bit more of the crystal superconducts?
Host 2Exactly that: about sixty percent from magnetism and about fifty five from heat, and varying the fit only moved the heat figure between fifty three and fifty seven. Two unrelated methods agreeing is the core of their claim that this is bulk superconductivity.
Host 1What about the obvious test, does the electrical resistance actually drop to zero?
Host 2Partly. With their standard cured silver contacts the resistance fell by about ninety eight percent, leaving a residue of roughly two percent that they attribute to the surface decomposing in air. One sample with uncured contacts did reach zero, and in the best crystal the onset was seventeen point two kelvin, with half the resistance gone at sixteen point five.
A word from our sponsor
Host 1A word from our sponsor.
Host 2This episode is brought to you by the Instant Pot, now in a limited edition for the materials chemist at home: the Instant Pot Multianvil.
Host 1I own the prototype, and my stew comes out monoclinic, with no intergrowth of yesterday's soup.
Host 2Each Multianvil reaches seven gigapascals on the yoghurt setting and quenches dinner from sixteen hundred degrees in one beep. The makers have not paid for this, have not been consulted and, as far as we know, have never heard of it.
Host 1Only forty exist, each sealed in platinum, and slow cook removes the apical oxygen from lasagne.
Host 2Leftovers are superconducting below eighteen kelvin with a shielding fraction above sixty percent. Lids levitate over the hob until Thursday.
Three dimensional pairing
Host 1After those heat jumps and resistance drops, what did they learn about the shape of the superconductivity?
Host 2That it is surprisingly three dimensional. They applied magnetic fields up to fourteen tesla, either along the nickel sheets or across them, and tracked how much field it takes to kill superconductivity.
Host 1Why would the direction of the magnetic field make any difference?
Host 2Because in a strongly layered superconductor, a field lying along the sheets barely hurts it, while a field crossing the sheets destroys it easily, so the ratio between the two becomes huge. Here the ratio was only about two point four, and it stayed finite as they approached the transition, instead of shooting up as it does in thin films.
Host 1So in plain words, the electrons pair up across the spacer layers, not only within each four layer block?
Host 2That is their reading. The size of a pair in the stacking direction, around one point two to one point eight nanometres, is bigger than the zero point six five nanometre gap between neighbouring blocks, so the blocks are coupled. They call it anisotropic but three dimensional.
Host 1Were there any other numbers from the field tests worth knowing?
Host 2Two. Extrapolating for fields along the sheets gives about sixty three tesla to destroy superconductivity at zero temperature, roughly double the simple limit of about thirty two tesla set by electron spin, which makes it an interesting test sample for very high field labs. The lower critical field, where field first starts to leak in, was about a thousand and sixty oersted.
Host 1You said the four layer crystal had a kink at one hundred and fifty kelvin before stripping. Does the stripped one have anything like that?
Host 2It does: every stripped sample shows a resistance upturn somewhere around ninety to a hundred kelvin, which they describe as either a partly gapped density wave or a metal to insulator like transition. It is hysteretic, meaning it differs on heating and cooling, ignores a fourteen tesla field, and its strength varies a lot, from weak in two samples to a fourfold rise in resistance in another.
Why it matters
Host 1So what does all this actually change, for physicists and for the rest of us?
Host 2For physicists, the principle is that the layer count alone can dial the nickel charge into the superconducting range, with no chemical substitution, no pressure cell and no substrate. The authors call this self doping and note it has no direct analogue in the cuprates.
Host 1Why is having a real lump of crystal so much better than having a thin film?
Host 2Because many of the best tools need bulk. The authors name neutron scattering, muon spin rotation and photoemission on cleaved surfaces, all hard or impossible on a film or on a sample inside a diamond anvil cell, and they admit the crystals must grow larger for neutrons.
Host 1And for an ordinary person, does this bring cheaper scanners or lossless power lines any closer?
Host 2Not directly. At eighteen kelvin it is far colder than the copper oxides, which in the wider picture already work above the boiling point of liquid nitrogen. Its value is as a clean model system for understanding why nickel oxides pair up at all.
Caveats
Host 1What would be the strongest thing I could wrongly conclude from this paper?
Host 2That the material is fully superconducting. About half the crystal is, by both heat and magnetism, and the heat transition is broad, with a main peak near sixteen kelvin and a shoulder near eleven, which they put down to uneven oxygen removal rather than an exotic two gap state.
Host 1How touchy is the oxygen stripping step in practice?
Host 2Quite touchy. Crystals of about a hundredth of a milligram reduced for one hundred hours in direct contact gave the best shielding, larger crystals did worse because the reaction is limited by diffusion, and extending to one hundred and thirty hours cut the signal to about a third. The vapour only method gave a few good crystals and mostly weak ones.
Host 1And the zero resistance, is that solid?
Host 2It is the softest point. Zero resistance appeared in one sample with imperfect uncured contacts, while the properly contacted samples kept a roughly two percent residue, which the authors blame on surface decay in air. The bulk claim leans on heat and magnetism more than on resistance.
Host 1Is there anything else a careful reader might trip over in the paper itself?
Host 2A few loose ends. The text gives the upturn as near ninety eight kelvin in one place and around ninety in another, and the two volume fractions are quoted as fifty five and fifty six percent. It is a preprint, so these are the kind of details peer review usually tidies up, and the origin of the upturn itself is openly unresolved.
Host 1Does the paper say why that upturn matters for the superconductivity?
Host 2Briefly, in its closing line: the authors argue that superconductivity living alongside this transition places the compound on the low doping side of the superconducting dome, by analogy with the cuprates. That is an interpretation, not something they measured directly.
What to watch
Host 1So what should we keep an eye out for next?
Host 2Three things the authors themselves set up. First, climbing the ladder to six, seven and eight layers in bulk, to see how the transition temperature and dimensionality change toward the infinite layer limit.
Host 1What are the second and third things worth watching?
Host 2Second, swapping lanthanum for other rare earths at fixed layer count, as a kind of chemical pressure that might raise the transition temperature. Third, squeezing the unstripped four layer crystal, since its one hundred and fifty kelvin kink resembles the two and three layer compounds that superconduct under pressure.
Host 1If someone gets the stripping step under control, what would a better result look like?
Host 2A sharper transition, a superconducting fraction close to one hundred percent, and clean zero resistance with ordinary contacts. Bigger crystals would then let the neutron and muon people in, which is where the question of how the electrons pair could actually be tested.
Sign-off
Host 1So, in one sentence, what did this paper show?
Host 2That a four layer lanthanum nickel oxide can be grown as bulk crystals under pressure, stripped of oxygen without falling apart, and then about half of it superconducts at eighteen kelvin at normal pressure, in a surprisingly three dimensional way. Modest numbers, but a cleaner platform than this corner of the field had before.
Host 1That was the Hastrup Review of Science, and every weekend this same feed carries the roundup, the week's new papers in one short episode.
More from this week
- Sprint Caffeine, Lost Lean Mass and a Too-Eager Atlantic CollapseEpisode 14 · 10 min
- Caffeine on the Blocks, How Much Makes Sprinters FasterDeep dive · 18 min
- Weight-Loss Drugs and the Lean Mass That Goes With the FatDeep dive · 21 min
- The Waist the Scale Misses in Young PeopleDeep dive · 20 min
- Which Cholesterol Drugs Prevent a First Heart AttackDeep dive · 16 min
- Do Climate Models Give Up on the Atlantic Too Easily?Deep dive · 21 min