Unveiling Anisotropic in‐Plane Thermal Conductivity and Thermal Expansion Coefficient in Low‐Symmetry Few‐Layer ReS2 Manab Mandal, Prahalad Kanti Barman, Susmita Jana, Saroj Poudyal, Kaushalya Kumari, Sourav Mondal, Abhishek Misra, Birabar Ranjit Kumar Nanda, Pramoda Kumar Nayak, Kanikrishnan Sethupathi Abstract ReS2 belongs to a unique class of 2D materials with low structural symmetry and extremely weak interlayer bonding, which give rise to intriguing thermal properties, like in‐plane anisotropic thermal conductivity (ATC) and anisotropic thermal expansion coefficient (ATEC). This study provides experimental evidence of in‐plane ATC and ATEC in few‐layer ReS2, revealed through simultaneous two‐mode Raman thermometry. In a single optical configuration, ATC of ReS2 has been determined by simultaneous analysis of two anisotropic Raman modes (III and V) corresponding to in‐plane crystallographic axes. For the five‐layer (5L) ReS2, the extracted thermal conductivities are ∼ 28.78 (‘b’‐axis) and ∼ 22.25 (‘a’‐axis) Wm−1K−1, respectively while the corresponding TECs are ∼ 6.1 × 10−6 and ∼ 4.04 × 10−6 K−1. As the layer thickness varies from 5 to 8L, the ATC and ATEC ratios monotonically increase from 1.29 and 1.52 to 1.36 and 1.94, respectively. The ab initio calculations reveal that the directional in‐plane anisotropy in phonon dispersion, group velocity, and Grüneisen parameter gives rise to ATC and ATEC. This work paves the way for exploring in‐plane ATC and ATEC in other low‐symmetry 2D materials, with potential implications for device applications. Simultaneous two‐mode Raman thermometry reveals strong in‐plane anisotropic thermal conductivity and thermal expansion coefficients in low‐symmetry few‐layer ReS2. By decoupling thermal transport along the crystallographic a‐ and b‐axes within a single optical configuration, this study provides robust experimental evidence and ab initio theoretical insights into nanoscale directional heat flow, paving the way for advanced thermal management in 2D materials. [ADVS-9999-e77734-g002.webp] Introduction Symmetry constitutes a crucial role in the fundamental laws of nature [, , , ], governing classical equations of motion, conservation laws, quantum selection rules, and exchange interactions [, , ]. In condensed matter, symmetry dictates a broad range of mechanical, electronic, and optical responses, including stress‐strain relations, charge transport, conductivity, refractive index, and the nature of allowed nonlinear processes [, , ]. Crystals with high structural symmetry typically exhibit isotropic electronic, optical, and thermal properties, which suppress phenomena such as even‐harmonic generation, chirality, and birefringence [, , ]. In contrast, lowering the crystal symmetry gives rise to anisotropy, where a given physical property becomes axial [, ]. The 2D Van der Waals (vdW) layered systems present an ideal platform for exploring such effects due to their intrinsic structural anisotropy arising from the strong intralayer covalent bonding and weak interlayer vdW interactions []. This anisotropy facilitates a variety of exotic light‐matter interactions, leading to exciton, phonon, edge, and moiré polaritons [, , ]. While the majority of 2D vdW materials exhibit isotropic behavior, certain systems with in‐plane low symmetry, such as ReX2 (X = S, Se), demonstrate biaxial anisotropy [, , , , ]. In these materials, the non‐orthogonal orientation of in‐plane excitonic and phononic resonances induces a wavelength‐ and frequency‐dependent rotation of the principal axes, further enriching their optical and vibrational characteristics [, , , ]. In this context, symmetry lowering and anisotropy not only influence optical processes but also play a crucial role in thermal transport. In‐plane structural anisotropy governs phonon dispersion, scattering mechanisms, thermal expansion, and heat conduction pathways, thereby giving rise to strongly direction‐dependent TC and TEC. Among other 2D TMDCs, the 1T' stable phase of ReS2 is one of the promising candidates to exhibit strong thermal anisotropy. Although anisotropic optical and electrical properties of layered ReS2 have been extensively studied [, , , ] [], systematic investigations of its thermal transport, particularly in the 2D limit, remain unexplored. In the bulk limit, ReS2 exhibits in‐plane thermal conductivities of (70±18) and (50±13) Wm−1K−1, corresponding to an anisotropy ratio of 1.4 []. Other low‐symmetry materials, like Td‐WTe2 and Black Phosphorus (BP), show notable in‐plane ATC. Td‐WTe2 exhibits ATCs of 0.743 and 0.639 Wm−1K−1 along the zigzag and armchair axes, respectively []. Luo et al. reported that, for BP films thicker than 15 nm, the armchair and zigzag thermal conductivities are ∼ 20 and ∼ 40 Wm−1K−1, respectively []. In previous studies, ATC has been determined by analyzing specific Raman modes (power‐ and temperature‐dependent evolution) along either the zigzag or armchair directions [, ]. However, in those studies, the laser polarization had to be precisely aligned with the zigzag and armchair orientations to accurately probe the ATC. In contrast, our measurement approach is simpler and fundamentally different. Here, we first identify the Raman modes associated with vibrations along the in‐plane crystallographic axes. Consequently, these modes exhibit anisotropic heat propagation along the respective axes. By employing a single‐step Raman measurement (rather than armchair or zigzag), we probe two Raman modes simultaneously to quantify the material's thermal anisotropy. The anisotropy in ReS2 is distinct from that in other well‐known anisotropic materials such as SnSe2, PdTe2, and BP, which typically exhibit uniaxial or biaxial in‐plane anisotropy. ReS2 (ReSe2) features a non‐orthogonal basis leading to wandering or tilting of optical axes within the plane. In this work, we perform a systematic vibrational analysis to elucidate the layer‐dependent anisotropic behavior due to the wandering of the characteristic vibrational axes in ReS2 crystals. In this work, we investigated in‐plane anisotropic thermal transport (ATT) in the triclinic 1T′ layered ReS2 system. We experimentally probed the ATT through Raman thermometry [, , ], and also validated the observation with first‐principles calculations and symmetry analysis. Prior to the determination of in‐plane ATC and ATEC in few‐layer ReS2, we performed angle‐resolved polarized Raman (ARPR) Spectroscopy to identify the in‐plane crystallographic orientations. Then, Raman thermometry (power‐ and temperature‐dependent) was employed to extract the in‐plane ATC and ATEC in suspended ReS2 flakes, having thicknesses of 5–8L. This study investigates the in‐plane ATC and ATEC in few‐layer ReS2 using combined experimental and theoretical approaches, and systematically analyzes the thickness dependence of anisotropy. To the best of our knowledge, this study also provides the first experimental demonstration of anisotropic TECs in an anisotropic 2D system. We believe that this study opens up a new paradigm in anisotropic 2D systems for further optoelectronic and thermoelectric device applications. Results and Discussion In‐Plane Anisotropy Unlike group VI TMDCs, the group VII compound like ReX2 (X = S, Se) possesses additional electrons that promote the formation of Re‐Re bonds, driving a Peierls distortion [, , ]. This distortion transforms the ideal trigonal (1T) lattice into a distorted 1T′ configuration, lowering the symmetry to a triclinic structure having a space group P1 ̄, in which the principal crystallographic axes are non‐orthogonal. Figure illustrates the crystal structure of monolayer (1L) ReS2, consisting of a Re‐layer sandwiched between two S‐layers, belonging to the C i point group preserving inversion symmetry [, ]. The Re‐Re diamond chain direction is usually defined as the ‘b’‐axis (Re − Re ∥), and the other in‐plane principal axis is designated as the ‘a’‐axis (Re − Re ∠), having an angle of either 60° (or 120°) with the ‘b’‐axis. To investigate the in‐plane anisotropic thermal conductivity, as well as the thermal expansion coefficient of few‐layer suspended ReS2, the flakes were transferred onto a patterned (array of holes with a diameter ∼ 5 μm) SiO2(290 nm)/Si substrate by the dry transfer method (see the methods section). Figure (upper panel) illustrates the optical microscopy (OM) image of a suspended ReS2 flake (6L). Topography and height profile (confirmed by atomic force microscopy (AFM)) are shown in the middle and lower panels of Figure . For other thicknesses (5L, 7L, and 8L), the OM images, AFM topography, and corresponding height profiles are shown in Figure . According to the group‐theoretical symmetry analysis, the 36 Γ‐point phonon modes decompose into the irreducible representations Γ = 18(Ag + Au​), where the Ag modes are Raman active. The 18 Raman‐active Ag modes of ReS2 are observed in the frequency range of 125–450 cm−1. The Raman spectra of ReS2 flakes (5–8L) are shown in Figure . For simplicity, the observed peaks are sequentially labeled as modes I‐XVIII. It should be noted that ReS2 has two different polymorphs, named AA and AB [, ]. In our measurement, we have chosen the AA phase, which was confirmed through the peak separation (∼13 cm−1) of modes I and III (see Table ) [, ]. Among the 18 Raman‐active modes, we primarily focused on mode III (∼151 cm−1) and mode V (∼212 cm−1), which capture the essence of Re‐Re atomic chain vibrations (Re − Re ∠ and Re − Re ∥) along the crystallographic ‘a’ and ‘b’‐axes, respectively, as illustrated by the corresponding vibrational eigenvectors in Figure . The resultant atomic displacements of modes III and V are predominantly aligned with these major crystal axes, making them highly sensitive to in‐plane anisotropy. Prior to the measurement of ATC, we performed ARPR for a few layers of suspended ReS2 for the confirmation of crystallographic axes along with the in‐plane anisotropy. Due to the reduced symmetry, all Raman modes exhibit distinct polarization dependencies. Among them, the relative directions of the maximum‐intensity polarization of the Raman modes III (∼151 cm−1) and V (∼212 cm−1) can be used to determine the ‘a’ and ‘b’‐axes directions. Figure shows the polar intensity distribution of modes III and V (for 6L‐ReS2) in the ARPR measurement, which shows an angular separation of ∼39° between the two modes. For other thicknesses, this angular separation varies with layer number, being ∼36°, 42°, and 45° for 5L, 7L, and 8L, respectively, as shown in Figure . The intensity distribution was fitted with Raman tensor analysis (Supporting Information Note I). The maximum intensity variation of these two modes (for mode III at 330° and mode V at 0°) for the specific case of 6L‐ReS2 is shown separately in Figure . The polarization maxima of Raman modes III and V for different polarizer orientations across 5–8L samples are summarized in Figure . Although the crystallographic ‘a’ and ‘b’‐axes of ReS2 are separated by 60°, the intensity maximum angular separation of modes III and V is less than 60°. Similar observations have also been reported in earlier studies [, , ]. This could be due to the mismatch between the optical axis and crystallographic axis, as explained by Novosolve et al. []. Our observations also align with the previous results. Therefore, the orientation of mode III and mode V maxima will represent the optical anisotropy in the in‐plane orientation of a specific thickness of ReS2. As a result, by monitoring the frequency shifts of phonon modes III and V as a function of laser power and temperature, the in‐plane anisotropic thermal conductivity of few‐layer ReS2 can be determined along the ‘a’ and ‘b’‐axes, respectively. In‐Plane Anisotropic Thermal Conductivity Experimental Results One of the most widely employed non‐contact techniques to investigate the thermal properties of 2D materials is Raman thermometry, where a focused laser beam serves as a localized heat source, which increases the local temperature of the sample, and the resulting temperature rise is monitored through Raman spectroscopy. The Raman‐based thermometry relies on the inelastic scattering of the excitation laser, where the frequency shift of a temperature‐calibrated Raman mode is employed as a sensitive probe of the local temperature in the suspended sample. Therefore, the power‐ and temperature‐dependent Raman studies have been carried out to determine the thermal conductivity. Herein, Raman thermometry was performed using a 632.8 nm continuous‐wave (CW) laser as the excitation source. The laser was focused to a spot size of r0 ≈ 0.76 μm (see spot size calculation [], Figure ) at the center of suspended ReS2 flakes with thicknesses ranging from 5–8L. The fraction of incident laser power absorbed by the 2D material plays a decisive role in the effective heating and thereby influences the thermal conductivity values determined from the optothermal Raman technique. For anisotropic materials, the absorption of incident laser power along the crystallographic axis (‘a’ or ‘b’‐axis) differs due to a change in the refractive index along the principal optical axis []. In few‐layer ReS2, anisotropic absorption behavior is evident across the examined energy range, stemming from the combined effects of refractive index anisotropy and thickness‐dependent optical absorption (see Figure and Table ). To further validate the anisotropic absorption behavior, heat propagation simulations using the finite element method were performed (see Figure and Note II). Therefore, to estimate the in‐plane ATC in ReS2, we have analyzed modes III and V with the variation of absorbed power and temperature, which leads to the anisotropic heat propagation along the two in‐plane crystallographic axes ‘a’ and ‘b’. In Figure , the peak frequencies of modes III and V in the 6L‐ReS2 (suspended) flake are shown as a function of incident laser power. Similarly, the results on flakes with different layer numbers (5L, 7L, and 8L) are shown in Figure . An increase in laser power corresponds to a higher local temperature rise (ΔT = θRaman), leading to a red shift of both modes. Notably, the first‐order power coefficient for mode V (− 0.9 × 10−2 cm−1 μW−1) exhibits a larger red shift compared to mode III (− 0.655 × 10−2 cm−1μW−1), indicating its stronger temperature sensitivity as a function of absorbed laser power, PA, as depicted in Figure (for other thicknesses, see Figure and Table ). In previous studies, the in‐plane ATC of BP [] and Td‐WTe2 [] was determined by analyzing a single Raman mode (∼ 464 cm−1 for BP and ∼212 cm−1 for Td‐WTe2). In these approaches, the sample was either aligned along the zigzag or armchair crystallographic axes, or the laser polarization was oriented along these axes, to extract the ATC. In contrast, our approach for ReS2 offers a distinct advantage: neither sample alignment along crystallographic axes nor laser polarization matching is required. This is because Raman modes III and V intrinsically correspond to vibrations along the in‐plane crystallographic ‘a’ and ‘b’ axes. Thus, by independently analyzing the power‐ and temperature‐dependent phonon shifts of these two modes in a single measurement, the ATC along both axes can be directly obtained. For validation of our method against the earlier approach on Td‐WTe2, we performed Raman measurements (for 8L) with the laser polarization aligned along the ‘a’ and ‘b’ axes and compared the results with those obtained using unpolarized excitation for modes III and V under power‐dependent conditions. Both results showed excellent agreement (see Figure ). Hence, we conclude that our method offers a more convenient and straightforward way to extract the ATC in ReS2. Temperature‐dependent Raman measurements were conducted on 6L‐ReS2 to investigate the phonon frequency shifts of modes III and V in the temperature range 275–475 K, as shown in Figure (for other thicknesses, see Figure ). To minimize laser‐induced heating effects, the laser power was kept to a minimum during the temperature‐dependent study. The individual Raman spectra (in Figure ) were fitted using a multi‐Lorentzian function to accurately determine the peak positions. As the temperature increases, the phonon interacts more due to the anharmonic term in the lattice potential, which causes a softening of optical phonons that leads to a red shift in phonon modes []. For 6L‐ReS2, the first‐order temperature coefficient of Mode V(− 0.815 × 10−2 cm−1K−1) exhibits a larger red‐shift than Mode III(− 0.455 × 10−2 cm−1K−1), which is shown in Figure (for other thicknesses, see Figure and Table ). This mismatch eventually leads to the in‐plane ATC along the optic axes. Mode V is associated with weaker and more anisotropic bonds, yields a higher value of the Grüneisen parameter, and stronger acoustic phonon coupling. Therefore, it enhances the anharmonic phonon‐phonon interactions that drive a greater temperature‐dependent frequency softening for mode V compared to mode III. Using these temperature coefficients, the laser power‐induced red‐shift of the Raman modes III and V is translated into the corresponding local temperature rise (θRaman) as a function of the absorber laser power (PA). In Figure , the local temperature rise along modes III and V shows distinct anisotropy with varying absorbed power (for other thicknesses, see Figure ). To extract the in‐plane ATC along both axes, linear fits of modes III and V were applied to the extracted ΔT as a function of absorbed laser power (PA ), yielding the slope ∂T∂PA, which was subsequently used in the following expression [], (1) κa,b=α12πt∂T∂PAa,b−1lnRr0 where t represents the thickness of the ReS2 flake, R and r 0 are the radii of the hole and laser spot, respectively, and α is a dimensionless prefactor expressed as a function of R/r 0. Under our experimental conditions, α ∼ 1 []. Using Equation (), the in‐plane thermal conductivities along the ‘a’ and ‘b’‐axes were extracted for ReS2 flakes with thicknesses ranging from 5L–8L, as presented in Figure . With increasing thickness, the thermal conductivity shows an upward trend (along ‘a’ and ‘b’‐axes κa5L = 22.25 ± 1.59 Wm−1K−1 to κa8L = 37.4 ± 2.59 Wm−1K−1 and κb5L = 28.78 ± 2.03 Wm−1K−1 to κb8L = 50.83 ± 3.50 Wm−1K−1), primarily due to the suppression of phonon‐boundary scattering. The total relative error in κ was derived by propagating statistical and systematic experimental errors (see Note ). Furthermore, the anisotropy ratio (β) also increases with thickness (β5L ∼ 1.29 to β8L ∼ 1.36), indicating that the influence of intrinsic phonon anisotropy progressively outweighs that of extrinsic scattering, as illustrated in Figure . The in‐plane ATC along ‘b’ and ‘a’‐axes and their anisotropic ratio (β) for different thicknesses have been tabulated in Table . Theoretical Aspect The strong in‐plane anisotropy arising from the distorted 1T' crystal structure directly influences the phonon dispersion and, consequently, the thermal conductivity of ReS2. To quantitatively assess the degree of thermal anisotropy in layered ReS2, we first examine its phonon behavior in both the monolayer and bulk limits. The high‐symmetry paths of interest constructed within the Brillouin zones (BZ) of these systems, along the in‐plane reciprocal axes (Γ‐M1 and Γ‐M2 directions), are illustrated in Figure . The phonon spectra along these two directions, presented in Figure , respectively, exhibit distinctly different slopes and reveal directional inhomogeneity in both systems that extends across the full frequency range. The low‐energy optical modes (< 240 cm−1) are mostly dominated by the vibrations due to Re atoms, as seen by the phonon density of states in Figure . Particularly, the dispersion of modes III and V, associated with the axial vibrations, primarily dominated by asymmetrically buckled Re‐Re chains along the ‘a’ and ‘b’‐axes, are highlighted in Figure . As referred in the previous section, the distorted trigonal symmetry of the ReS2 crystal provides a coupled vibrational basis [], where the real‐space vibrations along the Re‐Re chain (Re − Re ∥) and those at a certain angle to the Re‐Re chain (Re − Re ∠) can be associated with the phonon behavior along the reciprocal directions Γ‐M2 and Γ‐M1, respectively. To verify this correspondence, we calculated the angular separation of the vibrational eigenvectors for modes III and V, which was measured to be ∼ 36° for the monolayer and ∼ 42° for the bulk. These values are in excellent agreement with the experimental results obtained in this work. The phonon group velocity (vg ), extracted from the slopes of the dispersion curves, directly reflects the lattice anisotropy. The frequency‐dependent average vg captures the cumulative anisotropy in the reciprocal space, which is written as [, ], (2) vavω=vi2vω=Δq2π3∑λ,i(viλ)2viλδω−ωλ∑λgλω where, viλ is the i‐th component of the group velocity of phonon mode λ at a wavevector q and g λdenotes the phonon density of states. There is an overall contrast in vav coming from vibrations alongRe − Re ∥ andRe − Re ∠, mostly in the low‐frequency region (< 240 cm−1), with a larger directional difference in the bulk crystal than the monolayer, as seen in Figure , respectively. For comparison, the experimentally calibrated modes III and V yield vav values of ∼ 427 (434) m/s, ∼ 295 (436) m/s for monolayer, and ∼ 325 (472), ∼ 212 (473) m/s for bulk along Γ‐M2 (Γ‐M1). This thickness‐dependent behavior is further examined by considering AA‐stacked ReS2 systems up to 5L, as shown in Figure . Since our study focuses only on the in‐plane anisotropy, while additional layers introduced along the crystallographic c direction introduce very limited change in vav along the two considered in‐plane directions. However, the difference in the absolute values of vav between Γ‐M2 and Γ‐M1, suggests that vibrational anisotropy is more pronounced once we move toward the bulk limit. The difference in absolute values of vav suggests that vibrational anisotropy is more pronounced if we move toward the bulk limit. Within the range of ∼ 240 cm−1, higher vav along Γ‐M1 suggests that phonons in this direction of Re − Re ∠ experience less scattering and therefore make stronger contributions to the cumulative thermal conductivity. However, above this threshold (> 240 cm−1), there is a drop in vav , primarily due to strong scattering from lone‐pair electron repulsion of the lighter S atoms []. The TC calculated from phonon‐scattering theory (Figure ) closely follows the trends observed in vav . At room temperature (300 K), the thermal conductivity values are approximately 15.3 (16.4) W m− 1 K− 1 for the monolayer and about 42.4 (46.2) W m− 1 K− 1 for the bulk, along Γ‐M2 (Γ‐M1), respectively. The anisotropy ratio (β), calculated as TCΓ−M1:TCΓ−M2, is ∼1.05 for the monolayer, increasing up to ∼1.1 for the bulk. This indicates a higher degree of directionality of the planar lattice dynamics in the bulk case. These results are consistent with experimental observations revealing that TC in both systems is highly anisotropic, which strengthens with increasing layer number and temperature. Anisotropic Thermal Expansion Coefficient The thermal expansion coefficient (TEC) in a 2D system can be described through the quasi‐harmonic approximation (QHA). Within this framework, if all vibrational modes are initially treated as harmonic, the free energy of the system can be expressed analytically. The TEC can be derived as follows [, ], (3) α=1l02∂2E∂l20∑k,mγk,mcvk,m Here, the subscript ‘0’ denotes the ground state, l represents the lattice constant, and E is the total energy of the system. The term cv (k, m) corresponds to the contribution to the heat capacity from the mth phonon branch at wavevector k. The quantity γ k,m is the Grüneisen parameter, which quantifies the degree of phonon anharmonicity, defined as [], (4) γk,m=−l0ω0,k,m∂ωk,m∂l0 where, ω k ,m ​ denotes the phonon frequency of the mth branch at wavevector k, while ω0, k , m ​ represents the corresponding frequency in the ground state configuration. Theoretical estimation of the average Grüneisen parameter,γ av (ω m )was obtained by performing a frequency‐dependent density regression over the two high‐symmetry directions of the BZ as, (5) γavωm=∑kγk,mδωm−ωk,mgωm where g(ω m ) is the phonon density of states. As shown in Figure , the low‐frequency optical phonon modes exhibit strong temperature‐dependent volume responses in the quasi‐anharmonic regime, which gradually saturate in the higher‐frequency regime. As already established through phonon dispersion, the vibrations along the closely spaced diamond‐shaped Re − Re ∥ and Re − Re ∠ are the origin of thermal anisotropy in layered ReS2. These dynamically distinct vibrations give rise to inhomogeneous electron localization along the ‘a’ and ‘b’‐axes, which increases with layer thickness. This leads to the enhancement of γ av (ω m )along the Γ‐M2 than the Γ‐M1, which replicates the behavior of TEC along the crystallographic ‘b’ and ‘a’‐axes, respectively. In addition, the mode‐specific Grüneisen parameters (γ k,m ) for modes III and V along the selected BZ path exhibit distinctly different slopes between the monolayer and bulk ReS2, as highlighted in Figure . To elucidate the influence of anisotropic TEC on the phonon frequency shifts, we carried out a detailed symmetry and perturbation analysis of few‐layer ReS2. Based on the basis functions of the Ci point group, the deformation potential was constructed by using the corresponding character table (see Table and Supporting Information Note ). Owing to the inherent anisotropy of ReS2, the stress‐ and temperature‐dependent phonon frequency shifts along the ‘a’ and ‘b’‐axes can be expressed as, (6) Δωa,bAg=Ka,bσa,b+Aa,bΔT where Kb and Ka are the stress coefficients along ‘b’ and ‘a’‐axes, respectively. The stress coefficients along both axes (K b ∼ ‐0.56 cm−1GPa−1 and K a ∼ ‐0.428 cm−1GPa−1) for few‐layer ReS2 were calculated by using the gauge factor values from the report []. Ab and Aa are the temperature coefficients along ‘b’ and ‘a’‐axes, for the suspended region, respectively. σ a,b is the in‐plane (along either ‘a’ or ‘b’‐axis) thermal stress within the ReS2 flakes. The hole has a diameter (D) of ∼ 5 μm. Since the suspended few‐layer ReS2 possesses a very high aspect ratio (D/t ∼756‐1154), any compressive stress present can be relaxed through buckling. Therefore, the suspended region is expected to remain nearly stress‐free (for details, see Supporting Information Notes and Figure ). The strain‐stress relations of the flakes supported by SiO2(290 nm)/Si substrate can be expressed as, (7) εSiO2a,b=αa,bΔT+1−νEa,bσa,b=αSiO2ΔT where, εSiO2a,b is the in‐plane (along either ‘a’ or ‘b’‐axis) thermal strain within the ReS2 flakes on the SiO2(290 nm)/Si substrate. α a,b and αSiO2 are the in‐plane TEC of ReS2 flakes (along either ‘a’ or ‘b’‐axis) and SiO2, respectively and v represents the Poisson ratio of ReS2. Ea and Eb are the Young's moduli along the ‘a’ and ‘b’‐axes. The anisotropic Young's modulus of few‐layer ReS2, corresponding to both the principal in‐plane axes (Ea ∼ 201 GPa and Eb ∼ 127 GPa), was adopted from the earlier report []. So, the in‐plane (along either ‘a’ or ‘b’‐axis) thermal stress can be expressed as, (8) σa,b=Ea,bαSiO2−αa,b1−νΔT By substituting the Equation () into the Equation (), we obtained the temperature‐dependent frequency shift along ‘a’ or ‘b’‐axis as, (9) ΔωSiO2a,b=Ka,bαSiO2−αa,bEa,b1−νΔT+Aa,bΔT where, ΔωSiO2aandΔωSiO2b are the temperature‐dependent frequency shifts of substrate (SiO2/Si) supported ReS2 flakes along ‘a’ and ‘b’‐axes, respectively. The temperature coefficients ASiO2a,ASiO2b,Aa,andAb can be measured from the temperature‐dependent Raman experiments (see Figures , and Table ). Therefore, the TEC along ‘a’ or ‘b’‐axis can be expressed as a function of temperature coefficients, (10) αa,b=αSiO2−1−νEa,bKa,bASiO2a,b−Aa,b Using Equation (), the in‐plane TEC along ‘a’ and ‘b’‐axes were extracted for ReS2 flakes with thicknesses ranging from 5–8L, as presented in Figure . With increasing thickness, the TEC shows a decreasing trend (along ‘a’ and ‘b’‐axes αa5L = (4.04 ± 0.09)×10−6 K−1 to αa8L = (2.02 ± 0.18)×10−6 K−1 and αb5L = (6.1 ± 0.22)×10−6 K−1 to αb8L = (3.88 ± 0.22)×10−6 K−1), primarily due to the absence of interlayer vdW interaction [, ]. Furthermore, the anisotropy ratio(δ) also increases with thickness (δ5L ∼ 1.51 to δ8L ∼ 1.92), as illustrated in Figure . The in‐plane anisotropic TEC along ‘b’ and ‘a’‐axes and their anisotropic ratio(δ)for different thicknesses are summarized in Table . In summary, the ATEC in few‐layer ReS2 originates from the inherent in‐plane structural anisotropy associated with the directional Re‐Re chain, resulting in distinct thermal expansion along Re − Re ∥ and Re − Re ∠. Conclusion In summary, we probed the in‐plane ATC and ATEC of suspended few‐layer ReS2 flakes using Raman spectroscopy and analyzed the thickness‐dependent variation of the anisotropic ratio. This is the first experimental demonstration of ATEC in 2D layered materials. Following the confirmation of the crystallographic axis of ReS2 through ARPR measurements, thermal conductivity (along ‘a’ and ‘b’‐axis) was extracted from Raman thermometry data. The extracted thermal conductivity of ReS2 along ‘a’‐axis is κa5L = 22.25 ± 0.29 Wm−1K−1 to κa8L = 37.4 ± 0.32 Wm−1K−1 for 5–8L thicknesses. Whereas, along ‘b’‐axis the values varies from κb5L = 28.78 ± 0.23 Wm−1K−1 to κb8L = 50.83 ± 0.38 Wm−1K−1 corresponding to 5–8L. Depending on the thickness, this yields an ATC ratio (β) in the range of ∼1.29 to 1.36. The microscopic origin of in‐plane ATC can be attributed to the phonon dispersion characteristics, group velocities, and scattering rates associated with lattice vibrations along Re − Re ∥ and Re − Re ∠, as obtained from first‐principles calculations. Also, we have estimated ATECs along two different axes (along ‘a’ and ‘b’‐axes αa5L = (4.04 ± 0.09) × 10−6 K−1 to αa8L = (2.02 ± 0.18) × 10−6 K−1 and αb5L = (6.1 ± 0.22) × 10−6 K−1 to αb8L = (3.88 ± 0.22) × 10−6 K−1) for different layer thicknesses (5–8L). Based on symmetry analysis, we demonstrated that the TEC influences the Γ‐point optical phonon frequency through the combined effects of in‐plane thermal stress and anisotropic temperature coefficients, originating from thermal mismatch and free expansion. The estimated ATEC ratio (δ) ranges from ∼1.52 to 1.91, depending on the thickness. The calculated average Grüneisen parameter elucidates this ATEC by quantifying the thermal response to volume changes from the monolayer to the bulk limit. This finding provides a foundation for constructing anisotropic optoelectronic devices in other low‐symmetry 2D systems. Experimental Section Sample Fabrication High‐quality single‐crystal ReS2 (commercially obtained from 2D Semiconductors) was used for all measurements. Bulk ReS2 crystals were mechanically exfoliated onto a SiO2(290 nm)/Si substrate using the standard scotch tape method. In order to achieve suspended few‐layer ReS2 flakes, circular holes with a diameter of ∼ 5 μm were patterned on the SiO2(290 nm)/Si substrate by electron beam lithography (EBL) followed by reactive ion etching (RIE). The exfoliated ReS2 flakes were then transferred onto the array of holes using a standard dry transfer method. Optical Microscopy (OM) and Atomic Force Microscopy (AFM) Bright‐field OM imaging was carried out using a Nikon ECLIPSE LV100ND microscope, equipped with a 100× objective lens (NA = 0.95). During image acquisition, the exposure time was fixed at 100 ms, and the white balance was calibrated using a predefined R: G: B ratio. The surface morphology of few‐layer ReS2 was further examined by AFM (Park Systems NX10) employing an Al‐coated cantilever (PPP‐NCHR). All the AFM measurements were performed in non‐contact mode with a scan rate of 0.3 Hz. Raman Thermometry Temperature‐ and power‐dependent Raman spectra were acquired using a confocal Raman spectrometer (LabRam 800, Horiba Jobin Yvon) operated in backscattering geometry with a 50× long working distance objective (NA = 0.5). A 632.8 nm laser was employed as the excitation source, yielding a spot size of ∼1.53 μm on the sample surface (see Figure , Supporting Information, for spot‐size determination). Spectra were collected using a 1800 lines/mm grating. For temperature‐dependent measurements, the samples were mounted in a liquid‐nitrogen‐cooled vacuum cryostat (Linkam). The flakes were affixed to a thin glass plate using silver paste to ensure efficient thermal coupling with the stage, and a stabilization time of 10 min was maintained at each set temperature. Raman measurements were performed on both supported and suspended regions of the flakes, yielding consistent results (see Supporting Information). The temperature rise was quantified as ΔT(θRaman)=(υPA−υPA=0)χT), where υPA=0 represents the intercept obtained from the linear fit of Raman shift as a function of absorbed power. Computational Methods In this work, first‐principles density functional theory (DFT) calculations are carried out using the VASP package with a plane‐wave energy cut‐off of 450 eV. The projector‐augmented wave (PAW) method within the generalized gradient approximation (GGA‐PBE) is employed to describe the Re: {s, d} and S: {p} valence electrons. For the bulk and monolayer 12‐atom unit cell of ReS2, the Brillouin zone (BZ) is sampled using a 7 × 7 × 1 Monkhorst–Pack k‐point grid. Structural relaxations are performed until the interatomic forces are reduced below 0.01 eV/Å. To investigate the harmonic phonon spectra, second‐order interatomic force constants (IFCs) are obtained using density functional perturbation theory (DFPT) in a 3 × 3 × 1 supercell, as implemented in the Phonopy package []. The lattice thermal conductivity (κ L ) is then evaluated by explicitly considering phonon–phonon scattering processes within the anharmonic approximation. Anharmonic IFCs are computed by including only three‐phonon interactions, using the Phono3py package []. To limit computational cost, phonon–phonon interactions beyond 6 Å are neglected by applying the cut‐off pair method, which is well suited for the distorted trigonal (P1 ̄) phase of both bulk and monolayer ReS2. The harmonic and anharmonic IFCs are used as inputs for solving the phonon Boltzmann transport equation (BTE) beyond the relaxation time approximation (RTA) on a dense 35 × 35 × 1 q‐point grid (convergence tests are also verified at finer grids). Within the RTA framework, the lattice thermal conductivity(κ L ) is expressed as: κL=1NV∑λκLT=1NV∑λCλυλ⊗υλτλ where, the sum is taken over all phonon modes λ of the system. Here, N and V are the number of phonon wavevectors and the unit cell volume, respectively. C λand τλdenote the mode‐specific heat capacity and single‐mode relaxation time, while, υλ=∂ωλ∂q, represents the phonon group velocity along the lattice directions. To quantify anisotropy in the phonon response, the mode Grüneisen parameter(γ i ) is also calculated, which captures the sensitivity of phonon frequencies to volume changes along different BZ directions. This parameter plays an essential role in understanding the thermal expansion behavior in these systems.