Thermoelectric materials can turn waste heat directly into electricity, but their efficiency is throttled by a stubborn property of most crystals: they conduct heat well. The key to better thermoelectrics is suppressing the lattice part of thermal conductivity, which is carried by phonons, the quantized vibrations of the atomic lattice. Tin telluride, SnTe, a lead-free cousin of the celebrated thermoelectric PbTe, already possesses strong lattice anharmonicity and low-energy optical modes that damp heat flow, yet pristine SnTe still conducts heat too well for practical use. A new study published in Advanced Science shows why the standard picture of phonon control in this material is incomplete, and how a subtle form of internal strain called chemical pressure reshapes individual phonon modes in ways that a simple volume argument cannot capture.
The research team, led by Hongchao Wang of Shandong University, used a model series of compounds in which five percent of the tin sites in SnTe were replaced with different elements: manganese, germanium, samarium, and lead. Each substituent changes the average size of the unit cell, creating either positive chemical pressure, which mimics compression, or negative chemical pressure, which mimics expansion. The team calibrated these pressures from refined lattice parameters using the third-order Birch–Murnaghan equation of state, obtaining values ranging from roughly minus 0.25 to plus 0.1 gigapascals. X-ray diffraction confirmed that every composition retained the face-centered cubic rock-salt structure with no detectable secondary phases, while electron microscopy and photoelectron spectroscopy verified dense microstructures and homogeneous incorporation of the substituents.
To establish a benchmark, the researchers first calculated how pristine SnTe responds to uniform physical pressure between 300 and 900 kelvin, using first-principles lattice dynamics combined with the Boltzmann transport equation. The results follow an elasticity-based relation closely: compression stiffens the lattice, hardens the phonon modes, and raises the lattice thermal conductivity, while expansion does the opposite. The pressure dependence is strongest near room temperature, where phonons are most sensitive to pressure-modified force constants, and weakens at high temperature as anharmonic scattering washes out the differences between pressure states. Calculations of three-phonon and four-phonon scattering rates showed that compression suppresses the high-scattering portion of the three-phonon distribution, while expansion broadens the four-phonon distribution and increases the population of strongly scattered phonons.
The surprise came when the experimental thermal conductivities of the substituted samples were placed on the same chemical-pressure scale. If chemical pressure were merely a disguised form of physical pressure, all samples would fall on the elastic reference curve. Most did not. The residuals from the elastic relation reached minus 27.5 percent for germanium substitution and minus 13.4 percent for samarium, with lead at minus 11.3 percent. The most striking comparison is between manganese and germanium: their calibrated chemical pressures are nearly identical, at 0.106 and 0.102 gigapascals respectively, yet their measured lattice thermal conductivities differ sharply, at 2.304 and 1.679 watts per meter-kelvin. A scalar pressure value, the study concludes, simply cannot predict how substitution alters heat transport.
To find out what actually happens to the phonons, the team deployed three complementary spectroscopic probes, each targeting a different region of the phonon spectrum. Terahertz time-domain spectroscopy accesses the zone-center transverse optical mode, in which tin and tellurium atoms move out of phase. Raman spectroscopy, aided by defect-activated scattering that relaxes the momentum selection rule, reveals zone-boundary optical phonons at the L points of the Brillouin zone. Brillouin spectroscopy detects long-wavelength acoustic phonons near 30 gigahertz, the collective vibrations that carry most of the heat. Together with first-principles calculations and spectral energy density analysis, these techniques resolved phonon frequencies, linewidths, spectral weights, and lifetimes mode by mode.
The terahertz measurements delivered the clearest anomaly. Under uniform physical pressure, the transverse optical mode shifts monotonically, softening under expansion and hardening under compression. Under chemical pressure, that regularity breaks down. Manganese, samarium, and lead substitution produce moderate softening, but germanium substitution drives a much larger frequency reduction accompanied by the broadest spectral linewidth in the series. The authors attribute this pronounced softening to local ferroelectric-like distortions around germanium atoms, an interpretation supported by ab initio molecular dynamics showing substantially larger off-centering in germanium-centered environments than in tin- or lead-containing models. Moreover, the line shapes of the terahertz peaks, analyzed with a Breit-Wigner-Fano model, revealed that manganese, germanium, and samarium substitution create clear asymmetries, indicating coupling between the discrete optical phonon and an overlapping continuum, most likely a defect-associated vibrational background, while pristine and lead-substituted samples remain nearly symmetric.
Raman spectroscopy extended the analysis to the zone boundary. The frequencies of the L-point transverse and longitudinal optical modes largely track the sign of the calibrated chemical pressure, hardening for manganese and germanium and softening for samarium and lead, consistent with the physical-pressure reference. The linewidths, however, tell a different story: the transverse mode broadens substantially for manganese, germanium, and lead, while the longitudinal mode broadens most for samarium. The broad two-phonon Raman feature also widens for every substituent. Density functional theory calculations of the two-phonon joint density of states showed why: positive chemical pressure reduces the number of momentum-compensated phonon pairs available for scattering, whereas negative chemical pressure increases them, redistributing the multiphonon phase space that governs relaxation.
Brillouin spectroscopy then connected these optical signatures to the heat carriers themselves. The acoustic phonon frequencies barely changed across the series, showing that light substitution has little effect on sound velocity. The acoustic phonon lifetimes, in contrast, varied dramatically, from about 550 picoseconds for manganese substitution down to roughly 300 picoseconds for germanium, with samarium and lead near 350 picoseconds. Chemical pressure, the results show, attacks heat transport through relaxation processes rather than through changes in acoustic velocity. Machine-learning-accelerated molecular dynamics on a 16,000-atom supercell, analyzed through spectral energy density maps, confirmed the mechanism: chemical pressure reshapes the spectral functions of low-energy optical modes, and because those modes sit close to the acoustic spectrum in SnTe, their frequency shifts and linewidth broadening increase the acoustic-optical spectral overlap, promote scattering, and shorten the lifetimes of the phonons that carry heat.
The broader lesson is that phonon engineering in anharmonic thermoelectrics cannot be reduced to a single number. Frequency shifts retain a partial pressure-like character, but linewidth broadening, Fano asymmetry, joint-density-of-states redistribution, and acoustic lifetime changes respond to the specific local perturbation each substituent introduces. Germanium substitution achieves the strongest suppression of lattice thermal conductivity near room temperature through pronounced softening and broadening of the transverse optical mode, while lead substitution dominates at high temperature through optical-mode softening. The study offers a mode-resolved framework for designing lower thermal conductivity in materials whose low-lying optical branches lie near the acoustic spectrum, a class that includes the cubic IV-VI thermoelectrics at the heart of modern waste-heat recovery research.
Subject of Research: Mode-resolved phonon dynamics under chemical pressure in SnTe thermoelectric materials
Article Title: Mode‐Resolved Phonon Dynamics Under Chemical Pressure in SnTe Thermoelectrics
Article References: Li, Z., Hu, J., Peng, P., Cao, P., Zhang, H., Zong, Y., Wu, Y., & Wang, H. (2026). Mode‐Resolved Phonon Dynamics Under Chemical Pressure in SnTe Thermoelectrics. Advanced Science, Article e78136. https://doi.org/10.1002/advs.78136
Image Credits: AI Generated
DOI: 10.1002/advs.78136
Keywords: SnTe, thermoelectrics, chemical pressure, phonons, lattice thermal conductivity, anharmonicity, terahertz spectroscopy, Raman spectroscopy, Brillouin spectroscopy, phonon linewidth, acoustic-optical scattering, ferroelectric distortion
