Crystal field tuned spin-flip luminescence in NiPS3

This paper resolves the debate surrounding the nature of NiPS3's sharp photoluminescence by combining substitution experiments and theoretical calculations to demonstrate that the emission originates from a crystal-field-tuned spin-flip transition between a triplet ground state and a singlet excited state, thereby establishing a fundamental link between the material's optical properties and its magnetic order.

The Gist

Imagine a crystal called NiPS₃ as a tiny, layered city made of atoms. In this city, the "residents" are Nickel atoms, and they have a special habit: they like to hold hands with their neighbors in a very specific, orderly pattern called antiferromagnetism. This means the residents organize themselves into two opposing teams (spins pointing up and down) that cancel each other out, creating a quiet, magnetic "ground state."

For a long time, scientists were puzzled by a strange behavior in this city. When they shined a specific light on it and cooled it down, the crystal would glow with a very sharp, bright light (photoluminescence) at a specific energy level (1.475 eV).

The Great Mystery: Is the Light Magnetic?

The big question was: Is this glow caused by the magnetic "teamwork" of the residents?

Previous theories suggested that the light was a direct result of the magnetic order. The logic was simple: The glow only appears when the temperature is low enough for the magnetic teams to form (below 155 K). Therefore, the glow must be a "magnetic signal." Some even thought the light was a complex, collective dance of electrons and holes (called Zhang-Rice states) moving freely through the crystal.

The Experiment: Changing the Neighborhood

To solve this mystery, the researchers decided to play a game of "what if" by changing the residents and the environment of the crystal city. They created two types of modified crystals:

1. The "Zn" Swap (Replacing the Nickel): They swapped some magnetic Nickel residents with non-magnetic Zinc residents.

   • The Result: This weakened the magnetic teamwork (lowering the temperature at which the teams form).
   • The Surprise: Even though the magnetic order got weaker, the glow stayed strong. It got a little dimmer and fuzzier, but it didn't disappear. This is like turning down the volume on a radio but the music still playing clearly.

2. The "Se" Swap (Changing the Ligands): They swapped the Sulfur neighbors (the "walls" of the city) with Selenium neighbors.

   • The Result: This actually strengthened the magnetic teamwork (raising the temperature at which teams form).
   • The Shock: Despite the magnetic order getting stronger, the glow vanished completely.

The Conclusion: If the light were purely a result of the magnetic order, the "Se" swap should have made the light brighter, and the "Zn" swap should have killed it. Since the opposite happened, the researchers concluded: The light is not a magnetic signal. The magnetic order might influence the light, but it is not the cause of it.

The Real Cause: The "Spin-Flip" Trick

So, what is the light? The paper explains it using a concept from chemistry called Crystal Field Theory.

Think of the Nickel atom as a musician with a specific set of instruments (electron energy levels). The "walls" of the city (the Sulfur atoms) press on the musician, changing the pitch of the instruments. This is the Crystal Field.

• The Ground State: The musician is usually playing a "Triplet" tune (a specific, magnetic rhythm).
• The Excited State: When hit with light, the musician jumps to a "Singlet" tune (a non-magnetic rhythm).
• The Trick: Usually, jumping from a Triplet to a Singlet is forbidden by the rules of physics (like trying to walk through a wall). However, in this specific crystal, the "walls" (the crystal field) are tuned just right to make this forbidden jump possible. This is called Spin-Flip Luminescence.

The researchers used a "map" called a Tanabe-Sugano diagram (which is like a musical score showing how the notes change as the room gets bigger or smaller) to prove that the energy of the glow matches exactly with this "Spin-Flip" jump.

Why Did the "Se" Swap Kill the Light?

When they swapped Sulfur for Selenium, the "walls" of the city changed. Selenium atoms are bigger and hold hands more tightly with the Nickel. This changed the "pitch" of the instruments (the energy levels).

The researchers found that this change pushed the "forbidden" Singlet tune too close to another "allowed" tune. When they got too close, the musician stopped playing the sharp, bright "Spin-Flip" note and started playing a different, blurry, and silent note instead. The light didn't die because the magnetic order got stronger; it died because the acoustics of the room changed so the specific trick could no longer be performed.

The Final Verdict

The paper concludes that the sharp, bright light in NiPS₃ is not a magical magnetic phenomenon. Instead, it is a localized trick performed by a single Nickel atom, made possible only because the surrounding crystal "walls" are tuned to a very specific strength.

• The Analogy: Imagine a singer who can only hit a high note if the room is a specific size. If you change the room's size (by swapping atoms), the singer might hit a different note or stop singing, even if the audience (the magnetic order) is still cheering.
• The Takeaway: The light is a "Spin-Flip" event, a known phenomenon in chemistry, but it is rare to see it so clearly in a solid crystal. The magnetic order of the crystal is just a bystander that happens to be present when the trick works, not the magician pulling the rabbit out of the hat.

This discovery provides a "template" for finding other materials that can do this trick, which could be useful for future technologies that need to control light and magnetism together, but the paper strictly focuses on explaining what the light is, not on building devices with it yet.

Technical Summary

Technical Summary: Crystal Field Tuned Spin-Flip Luminescence in NiPS₃

Problem Statement
The nature of the sharp photoluminescence (PL) observed at 1.475 eV in the layered antiferromagnet NiPS₃ (below its Néel temperature, $T_N \approx 155$ K) has been a subject of significant debate. Previous interpretations have varied, proposing the emission as a coherent, delocalized Zhang-Rice charge transfer excitation involving hybridized Ni and S orbitals, or conversely, as a localized d-d excitation on the Ni ion. A central point of contention is whether this emission is fundamentally linked to the long-range magnetic order or if the magnetic state merely influences the transition's visibility through symmetry breaking. Additionally, the origin of a secondary non-radiative peak at 1.498 eV and the specific mechanism allowing a spin-forbidden transition to become optically active remain unclear.

Methodology
To resolve these ambiguities, the authors employed a combined approach of experimental spectroscopy and theoretical modeling:

1. Sample Fabrication: Single crystals of NiPS₃ were synthesized via physical vapor transport. To probe the dependence of the luminescence on magnetic and structural parameters, the authors created substitutional variants:
   • Metal Substitution: Ni was partially replaced by non-magnetic Zn ($Ni_{1-x}Zn_xPS_3$).
   • Ligand Substitution: S was partially replaced by Se ($NiPS_{3-x}Se_x$).
2. Experimental Characterization:
   • Magnetization: Measured using a SQUID magnetometer to determine the Néel temperature ($T_N$) as a function of substitution.
   • Photoluminescence (PL) and Reflectance: Temperature-dependent PL and reflectance contrast measurements were performed on bulk flakes (down to 4 K) to track the intensity, linewidth, and energy of the 1.475 eV feature and the 1.498 eV peak.
3. Theoretical Calculations:
   • Tanabe-Sugano (TS) Analysis: Used to map the $d^8$ Ni²⁺ energy levels in an octahedral crystal field, correlating experimental peaks with specific electronic transitions.
   • Charge Transfer Multiplet (CTM) Calculations: Performed using the Quanty code to model the NiS₆ cluster. This included configuration interaction between $d^8$, $d^9\underline{L}$, and $d^{10}\underline{L}^2$ configurations, accounting for intra-atomic Coulomb interactions, crystal field splitting, and charge transfer gaps.

Key Results

1. Decoupling of PL from Magnetic Order:

   • Zn Substitution: Replacing Ni with Zn reduced $T_N$ (consistent with suppressing magnetic exchange) but did not suppress the 1.475 eV PL, even at 14% substitution. This contrasts with previous reports where Cd substitution suppressed PL at 10% substitution.
   • Se Substitution: Replacing S with Se increased $T_N$ (due to enhanced superexchange via Se 4p orbitals) but completely suppressed the 1.475 eV PL at just 8% substitution.
   • Conclusion: The presence or absence of the PL signal does not correlate directly with the existence of long-range magnetic order ($T_N$), refuting the hypothesis that the emission is a magnetic excitation.

2. Identification of the Transition:

   • TS diagrams and CTM calculations identify the 1.475 eV emission as a spin-flip luminescence corresponding to a transition from the triplet ground state ($^3A_2$) to a singlet excited state ($^1E$).
   • The transition is formally spin- and symmetry-forbidden but becomes observable due to symmetry-breaking mechanisms (likely antiferromagnetic ordering or spin-orbit coupling) and mixing with nearby spin-allowed states ($^3T_1$).

3. Mechanism of Suppression:

   • The suppression of PL upon Se substitution is attributed to changes in the crystal field and covalency. Se substitution increases the Ni-Se bond length and covalency, effectively shifting the system toward a lower crystal field strength in the TS diagram.
   • This shift brings the $^1E$ and $^3T_1$ levels closer, facilitating a crossing where the $^3A_2 \to ^3T_1$ (spin-allowed) transition becomes lower in energy or gains spectral weight. This leads to enhanced non-radiative decay and broadening, quenching the sharp spin-flip luminescence.

4. Nature of the 1.498 eV Peak:

   • The peak at 1.498 eV (23 meV above the main PL) is identified not as a double-magnon sideband, but as a phonon sideband. This energy matches a known IR-active phonon mode in NiPS₃, explaining its non-radiative nature and high intensity.

5. Line Broadening and Temperature Dependence:

   • The PL linewidth broadens and intensity decreases as temperature increases, vanishing near $T_N$. This is interpreted not as a loss of the excitation mechanism, but as thermal damping and the loss of the specific low-temperature ground state that facilitates the narrow emission, similar to behavior observed in other transition metal fluorides (e.g., KNiF₃).

Significance and Claims
The paper claims to provide a conclusive explanation for the 1.475 eV emission in NiPS₃, establishing it as a localized crystal field transition (spin-flip luminescence) rather than a charge-transfer or collective magnetic excitation. The authors emphasize that while the magnetic ground state influences the visibility of the transition (likely via symmetry breaking), the excitation itself is not magnetic in origin.

The study highlights that the "brightness" of such spin-flip luminescence is highly sensitive to the crystal field environment and covalency, which can be tuned via chemical substitution. This finding serves as a template for identifying and designing similar layered magnetic materials where optical excitations are coupled to magnetic ground states, a property of potential interest for opto-spintronic applications and the optical manipulation of spins, without overstating immediate device implementations. The work resolves the controversy regarding the nature of the exciton and clarifies the distinct roles of metal vs. ligand substitution in tuning these optical properties.