Favorable half-Heusler structure of synthesized TiCoSb alloy: a theoretical and experimental study

This study combines experimental characterization and first-principles calculations to identify the most favorable half-Heusler structure of synthesized TiCoSb alloy and evaluate its thermoelectric properties.

The Gist

Imagine you have a magical box that can turn waste heat (like the heat from your car engine or a laptop) directly into electricity. Scientists are always hunting for the best materials to build these boxes. One of the most promising candidates is a material called TiCoSb (a mix of Titanium, Cobalt, and Antimony).

However, there's a catch: this material can arrange its atoms in a few different ways, like a set of Lego bricks that can be snapped together in different patterns. Each pattern gives the material different superpowers. The big question was: Which pattern is the "real" one that the scientists actually made in their lab, and which one works best for making electricity?

This paper is the story of how the researchers solved this mystery using a mix of "detective work" and "supercomputer simulations."

1. The Four Suspects (The Atomic Arrangements)

Think of the TiCoSb material as a tiny, 3D city made of three types of citizens: Titanium (Ti), Cobalt (Co), and Antimony (Sb). In a perfect city, they stand in specific spots. But because atoms are tiny and jiggly, they might swap places.

The researchers looked at four possible city layouts (called "Wyckoff positions"):

• Type I: Ti, Sb, and Co in one specific order.
• Type II: A different order.
• Type III: Another variation.
• Type IV: The fourth arrangement.

They didn't know which one they had built in the lab. They had to find the "real" one.

2. The Detective Work (Experimental Evidence)

First, the scientists built the material using a method called "arc melting" (basically melting the ingredients together with a super-hot electric arc, like a lightning bolt in a jar).

Then, they used two main tools to figure out the layout:

• X-Ray Diffraction (The Crystal Scanner): They shot X-rays at the material. The X-rays bounced off the atoms and created a unique pattern of dots, like a barcode. The researchers tried to match this barcode against the four "suspect" layouts.
  • The Result: The barcode matched Type IV perfectly. The other three layouts didn't fit the data well.
• Microscopes (The Magnifying Glass): They used powerful microscopes (SEM and TEM) to look at the material's grain and check the atomic spacing. This confirmed that the material was indeed a crystal and that the atoms were arranged exactly as the X-rays suggested.

3. The Supercomputer Simulation (The Virtual Lab)

To be absolutely sure, the researchers ran a massive simulation on a supercomputer. They built virtual models of all four layouts and calculated their energy.

• The Analogy: Imagine balancing a ball on top of a hill. If the ball rolls down, it's unstable. If it sits in a deep valley, it's stable.
• The Result: The computer showed that Type IV was the deepest valley. It had the lowest energy, meaning it was the most stable and natural way for these atoms to arrange themselves. The other three layouts were like unstable hills; the atoms wouldn't stay there naturally.

4. Why Does This Matter? (The Power of the Material)

Once they confirmed that Type IV was the winner, they asked: "Is this the best version for making electricity?"

• The "Traffic" of Electrons: In this Type IV structure, the material acts like a p-type semiconductor. Think of electricity as traffic. In this material, the "cars" moving are actually holes (empty spots where a car could be), and they move very efficiently.
• The Heat Barrier: A big problem with heat-to-electricity materials is that they usually let heat escape too easily (like a leaky bucket). The researchers found that their Type IV material has very low thermal conductivity. It's like putting a thick wool blanket around the heat source, keeping the heat trapped so it can be converted into electricity instead of just leaking away.
• The Result: The material has a high "Power Factor" (a score for how good it is at making power), especially at high temperatures (like 500°C to 900°C).

The Big Takeaway

This paper is a victory for precision. By combining real-world lab experiments with powerful computer simulations, the team proved that the TiCoSb alloy they made isn't just any random mix of atoms. It has a very specific, stable structure (Type IV) that makes it a p-type semiconductor with excellent potential for turning waste heat into clean energy.

In short: They found the perfect "Lego pattern" for this material, proved it's the most stable one, and showed that it's a strong contender for future green energy technology.

Technical Summary

1. Problem Statement

TiCoSb is a promising half-Heusler (HH) thermoelectric (TE) material for mid-temperature applications (500K–900K). However, the thermoelectric efficiency ($ZT$) of HH alloys is highly sensitive to their crystal structure and atomic ordering. While TiCoSb is known to crystallize in the $C1b$ structure (space group $F\bar{4}3m$), the specific Wyckoff positions of the constituent atoms (Ti, Co, and Sb) within the unit cell have been reported inconsistently in literature. Different atomic arrangements (e.g., MgAgAs-type vs. inverse structures) lead to drastically different electronic and thermal transport properties. There is a lack of systematic studies correlating the actual synthesized structure with its theoretical electronic properties and thermoelectric performance. This study aims to resolve the ambiguity regarding the most probable atomic configuration of synthesized TiCoSb and evaluate its thermoelectric potential based on that specific structure.

2. Methodology

The study employs a combined experimental and theoretical approach:

• Synthesis:
  • TiCoSb ingots were prepared via arc melting of high-purity Ti, Co, and Sb (with 2% excess Sb to compensate for evaporation).
  • The ingots were sealed in a vacuum and annealed at 1173 K for 5 days to ensure homogeneity.
• Experimental Characterization:
  • X-Ray Diffraction (XRD): Performed at room temperature. Data was analyzed using Rietveld refinement (FullProf software) considering four distinct structural models (Types I–IV) with different atomic assignments to Wyckoff positions (4a, 4b, 4c, 4d).
  • Microstructural Analysis: Scanning Electron Microscopy (SEM) with Energy Dispersive X-ray Spectroscopy (EDS) confirmed stoichiometry. Transmission Electron Microscopy (TEM) and Selected Area Electron Diffraction (SAED) were used to verify crystallinity, grain size, and Bragg reflection planes.
  • Thermal Properties: Lattice thermal conductivity ($\kappa_L$) was estimated using the Debye-Waller factor derived from Rietveld refinement and the Slack equation.
• Theoretical Investigation:
  • First-Principles Calculations: Performed using Density Functional Theory (DFT) via the FP-LAPW method in the Quantum Espresso package.
  • Structural Optimization: Structural energy was minimized for the four probable Wyckoff configurations using the Murnaghan equation of state.
  • Electronic & Transport Properties: Partial Density of States (PDOS), band structures, and thermoelectric parameters (Seebeck coefficient $S$, electrical conductivity $\sigma$, electronic thermal conductivity $\kappa_e$) were calculated using BoltzTraP2 based on the rigid band approximation.

3. Key Contributions

• Resolution of Structural Ambiguity: The study definitively identifies the most probable crystal structure of synthesized TiCoSb among four competing models by correlating experimental Rietveld refinement statistics with theoretical energy minimization.
• Identification of p-type Nature: The research confirms the specific atomic ordering that results in p-type semiconducting behavior, a critical factor for TE device design.
• Comprehensive Property Correlation: It links the specific atomic arrangement (Wyckoff positions) directly to the electronic band gap, effective mass, and thermoelectric transport coefficients.
• Low Thermal Conductivity Estimation: The study reports a significantly lower lattice thermal conductivity for the identified structure compared to previously reported theoretical and experimental values.

4. Key Results

• Structural Identification:
  • Experimental: Rietveld refinement showed that Type IV structure provided the best fit to the experimental XRD data (lowest statistical parameters: $R_{wp} = 11.4$, $\chi^2 = 9.07$).
  • Theoretical: DFT calculations confirmed that the Type IV structure possesses the minimum structural energy at equilibrium volume.
  • Atomic Configuration: The most favorable structure corresponds to the space group $F\bar{4}3m$ with the following Wyckoff positions:
    • Sb: 4a $(0, 0, 0)$
    • Ti: 4b $(1/2, 1/2, 1/2)$
    • Co: 4c $(1/4, 1/4, 1/4)$
• Electronic Structure:
  • The Type IV structure is confirmed as a p-type semiconductor with a direct band gap of 1.09 eV (located at the $\Gamma$ point).
  • The Fermi level ($E_f$) is shifted toward the valence band.
  • The relative effective mass ($m^*/m_e$) was calculated to be 1.75.
  • PDOS analysis reveals that the valence band is dominated by Co-3d states, while the conduction band is dominated by Ti-3d states.
• Thermoelectric Properties:
  • Lattice Thermal Conductivity ($\kappa_L$): Estimated at 4.07 W/mK, which is notably lower than many previously reported values for TiCoSb.
  • Seebeck Coefficient ($S$): The material exhibits intrinsic semiconducting behavior at 300 K. Theoretical calculations predict a maximum $S$ of ~376 $\mu$V/K at 500 K and ~675 $\mu$V/K at 800 K.
  • Power Factor (PF): The power factor ($S^2\sigma$) shows a significant increase starting from 500 K, indicating high potential for mid-to-high temperature applications.
  • Relaxation Time ($\tau$): Estimated using the Drude model, showing temperature dependence consistent with semiconductor scattering mechanisms.

5. Significance

This study provides a crucial benchmark for the TiCoSb half-Heusler system. By rigorously establishing that the synthesized material crystallizes in the Type IV (Sb-4a, Ti-4b, Co-4c) configuration, it resolves conflicting literature reports. The confirmation of a p-type nature with a wide band gap (1.09 eV) and low lattice thermal conductivity validates TiCoSb as a robust candidate for thermoelectric energy conversion in the 500–900 K range. The methodology of combining precise Rietveld refinement with first-principles energy optimization offers a reliable framework for identifying the true ground-state structures of complex intermetallic alloys, which is essential for optimizing their thermoelectric efficiency ($ZT$).