Nitride halides ZrNCl and HfNCl

ZrNCl, HfNCl

Also called ZrNCl, HfNCl, β-ZrNCl, layered metal nitride chlorides

van der Waals crystal superconductor

A band insulator that becomes a superconductor when you put electrons into it – by chemistry or by a gate – and the fewer electrons you add, the higher the transition temperature. That is the opposite of what a simple theory expects, and it makes these crystals one of the clearest places to watch superconductivity evolve from ordinary overlapping pairs towards tightly bound ones. Doped HfNCl superconducts at 25.5 K, high for a material with so few carriers.

Cell: Hexagonal, a = 3.60 Å
Atoms per cell: 6
Zr–N bonds: 2.10 and 2.34 Å
Zr–Cl bond: 2.77 Å
Height: 6.26 Å between the outer atom centres

Two buckled zirconium–nitrogen honeycombs stacked on top of each other and capped on both faces by chlorine: Cl–Zr–N–N–Zr–Cl. On its own the layer is a band insulator. Lithium or other electron donors placed in the van der Waals gap between the chlorine faces dope the zirconium–nitrogen slab and make it superconduct, and ionic-liquid gating does the same thing electrostatically – which made ZrNCl an early test of whether gating and intercalation give the same superconductor. One layer of bulk β-ZrNCl (Istomin, Köhler and Simon, Physica C 319, 219, 1999; COD 1521250): a = 3.60 Å, Zr–N 2.10 and 2.34 Å, Zr–Cl 2.77 Å.

Data sheet

Show all 10 entriesHide the data sheet

Kind of sheet: van der Waals crystal
Peels off a layered crystal, one sheet at a time
Behaves as: superconductor
Band gap: A wide-gap band insulator when undoped; adding electrons turns it metallic and then superconducting
Thickness: ~0.9 nm per layer
Structure: Double honeycomb [M₂N₂] slabs capped on both sides by chlorine, so each layer is Cl–Zr–N–N–Zr–Cl; the chlorine faces of neighbouring layers meet across a van der Waals gap, and guests such as lithium sit in that gap
Space group: ZrNCl (β): bulk R3m, monolayer P3m1
HfNCl (β): bulk R3m, monolayer P3m1
Bulk: the Materials Project crystal linked below. Monolayer: the 2DMatPedia layer, as a periodic slab
Stability: The undoped crystals are stable enough to cleave and handle, but intercalated phases are air- and moisture-sensitive: lithium-doped samples are prepared and measured under inert conditions.
First made: Lithium intercalation into β-ZrNCl reported in 1984 (Yamanaka and colleagues); superconductivity at 25.5 K in doped HfNCl in 1998 (Yamanaka and colleagues); ion-gated surface superconductivity in 2010 (Ye and colleagues)
Materials Project: ZrNCl (β)
HfNCl (β)
Computed crystal structure and properties
2DMatPedia: ZrNCl
HfNCl
Computed monolayer, exfoliated from a bulk crystal of the same compound

Compare with other materials

Key properties

Electron doping by lithium intercalation makes ZrNCl and HfNCl superconducting; HfNCl reaches 25.5 K
The transition temperature rises as carrier density falls, together with the coupling strength and the magnetic susceptibility, while the density of states stays roughly constant
Ionic-liquid gating superconducts the surface alone, giving a clean superconducting sheet thinner than one unit cell, with a dome-shaped phase diagram
That gated sheet shows a metallic ground state in a perpendicular magnetic field rather than an insulating one
Tuning carrier density by intercalation drives the system across the crossover between BCS pairing and Bose–Einstein condensation of tightly bound pairs
The lithium intercalation itself is reversible and electrochromic – the crystals change colour as electrons go in

How it is made

Bulk crystals by chemical vapour transport of the nitride chloride
Electron doping by lithium intercalation, chemically with n-butyllithium or electrochemically, often co-intercalating solvent molecules
Ionic-liquid gating of a cleaved surface, which dopes a single layer without adding any atoms

Uses, and how close they are

Model system for superconductivity at low carrier density and the BCS–BEC crossoverlab
Gate-controlled superconducting deviceslab

Readiness runs lab → prototype → pilot → deployed.

Open problems

Why does the transition temperature rise as electrons are removed – are magnetic or valley fluctuations doing the pairing?
Is the pairing state the same in the intercalated bulk and in the ion-gated surface layer?
What is the nature of the metallic ground state seen in a magnetic field, and is it a genuine bosonic metal?
Can the doped layer be isolated as a free-standing 2D crystal rather than being made at a surface or between layers?

Going deeper

Short notes for specialists. Choose a lens in the header and yours comes first.

For theoreticians · your lens

A doped band insulator with a low carrier density and weak screening, which is exactly where the standard Migdal–Eliashberg treatment starts to fail. Experiments find the coupling strength and susceptibility growing as the carrier density falls while the density of states stays put, pointing to pairing helped by electronic fluctuations rather than by phonons alone, and the same tuning knob carries the system into the BCS–BEC crossover where the pair size approaches the interparticle spacing. Calculations have to handle the two-dimensional honeycomb slab, the dielectric environment of the van der Waals gap and the guest ions in it at the same time.

For experimentalists · your lens

State how the electrons were added: chemical intercalation with co-intercalated solvent gives a different interlayer spacing, and therefore different two-dimensionality, from ionic-liquid gating of a single surface. Intercalated samples degrade in air, so cleave, dope and measure without breaking the inert chain. Because the interesting regime is low doping, quote carrier density with the transition temperature – a Tc without a carrier density says little in this system.

For engineers · your lens

Not a device material: superconductivity needs either reactive intercalated lithium or an ionic-liquid gate, and the transition temperatures are still cryogenic. Its value is as a laboratory for how superconductivity behaves when there are very few carriers, which is the regime that gated 2D devices increasingly work in.

In the research tracks

Theory frontiers

Superconductivity in the 2D limit

Growth methods

Chemical vapour transport (CVT)

Key references

Lithium intercalation and electrochromism in β-ZrNCl layered crystalYamanaka et al. · Chemistry Letters 13, 1403 (1984)cited by 13doi:10.1246/cl.1984.1403
Superconductivity at 25.5 K in electron-doped layered hafnium nitrideYamanaka et al. · Nature 392, 580 (1998)cited by 428doi:10.1038/33362
Enhancement of pairing interaction and magnetic fluctuations toward a band insulator in an electron-doped LixZrNCl superconductorKasahara et al. · Physical Review Letters 103, 077004 (2009)cited by 54doi:10.1103/PhysRevLett.103.077004
Liquid-gated interface superconductivity on an atomically flat filmYe et al. · Nature Materials 9, 125 (2009)cited by 596doi:10.1038/nmat2587
Metallic ground state in an ion-gated two-dimensional superconductorSaito et al. · Science 350, 409 (2015)cited by 323doi:10.1126/science.1259440
Gate-controlled BCS-BEC crossover in a two-dimensional superconductorNakagawa et al. · Science 372, 190 (2021)cited by 126doi:10.1126/science.abb9860