Quantitative Defect–Property Correlations in Ti3C2Tx MXenes via Precursor-Controlled Defect Engineering
Tufail Hassan, Doyeon Lee, Shabbir Madad Naqvi, Myungjae Kim, Jung-Min Oh, Sang Woon Park, Aamir Iqbal, Soo Yeong Cho, Zhiwang Hao, Noushad Hussain, Zubair Khalid, Shakir Zaman, Xiangmeng Kong, Ki-Min Roh, Hanjung Kwon, Chong Min Koo
TL;DR
Researchers engineered defect structures in MXenes to achieve high performance in conductivity, thermal properties, and stability.
Contribution
A new method for precise defect control in MXenes is introduced, linking defect structures to multifunctional performance.
Findings
Defect minimization in MXenes achieved electrical conductivity of 26,000 S cm−1 and thermal conductivity of 57 W m−1 K−1.
Defect-controlled MXenes showed EMI shielding of 90.5 dB and Joule heating of 263 °C at 1.5 V.
Defect-minimized MXene retained ~90% optical absorption after 4 months in dilute dispersion.
Abstract
Titanium vacancies (VTi), carbon vacancies (VC), and substitutional oxygen (SO) defects were precisely tuned in TiC and Ti3AlC2 MAX phases by adjusting C and Al feed ratios, yielding Ti3C2Tx MXenes with systematically varied defect densities.Defect minimization resulted in excellent multifunctional performance, including electrical conductivity of 26,000 S cm−1, thermal conductivity of 57 W m−1 K−1, infrared emissivity of 0.05, EMI shielding of 90.5 dB (at 10 µm), Joule heating of 263 °C (at 1.5 V), and activation energy of 72 kJ mol−1.The defect-minimized MXene exhibited excellent oxidation stability, retaining ~90% optical absorption after 4 months in dilute dispersion (0.02 mg mL−1).This study establishes a comprehensive quantitative framework linking precursor-derived defect structures to electrical, thermal, optical, and environmental stability of MXenes. Titanium vacancies (VTi),…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMXene and MAX Phase Materials · Electromagnetic wave absorption materials · Advanced Antenna and Metasurface Technologies
Introduction
Lattice defects play a central role in determining the physicochemical properties of two-dimensional (2D) materials. Vacancies, atomic substitutions, and lattice distortions modulate charge transport, phonon scattering, and light–matter interactions, thereby directly affecting the electrical, thermal, and optical performance of nanoscale systems [1–3]. As a result, defect engineering has emerged as a powerful strategy to tailor the functional behavior of 2D materials for applications in electronics, optoelectronics, thermal management, and electromagnetic shielding etc. However, for many 2D materials, particularly MXenes synthesized through multistep solid-state processes, precise control over defect formation and quantitative understanding of their correlation with functional properties remain limited, largely due to the challenges of manipulating defects across multiple synthesis stages and their inherent structural complexity [1–3].
MXenes—a large family of transition metal carbides and nitrides with the general formula M_n+1_X_nTx—have attracted significant attention due to their high metallic conductivity, low IR emissivity, tunable structure and composition, and versatile surface chemistry [4, 5]. In this formula, M denotes an early transition metals, X represents carbon and/or nitrogen, Tₓ refers to surface terminations (e.g., –OH, –O, –F), and n is an integer ranging from 1 to 4. MXenes are typically synthesized by selective etching of A element from layered MAX phases (Mn+1_AX_n_), where A is a group IIIA or IVA element such as Al or Ga. Their multifunctionality has enabled diverse applications including electromagnetic interference (EMI) shielding, Joule heating, infrared (IR) camouflage, and thermal management [6–8]. However, their multistep synthesis process—comprising high-temperature MAX phase sintering followed by chemical etching process—inevitably introduces a range of structural imperfections, including M site vacancies, X site vacancies, oxygen substitutional defects, and associated compressive lattice strain [7, 9, 10]. These lattice defects degrade electronic and thermal transport, while simultaneously accelerating oxidative degradation, particularly under aqueous or humid environments [11, 12].
Recent studies suggest that such defect formation is strongly coupled to the chemistry and quality of the starting precursors. For instance, using graphite instead of lampblack as the carbon source during Ti_3_AlC_2_ MAX synthesis improves the crystallinity and electrical conductivity in the resulting Ti_3_C_2_T_x_ MXene [13]. Oxygen incorporation during TiC precursor synthesis leads to substitutional defects that persist into MAX phase and final MXene, particularly due to titanium’s strong oxygen affinity [7, 9]. Similarly, incorporating excess aluminum during Ti_3_AlC_2_ MAX phase sintering improves Ti/C atomic diffusion and scavenges oxygen, thereby producing high crystallinity MAX phase and Ti_3_C_2_T_x_ MXenes exhibiting high electrical conductivity of 20,000 S cm^−1^ [10, 14]. As a result, the chemistry and quality of precursors influenced electrical conductivity and electromagnetic shielding performance of final MXene products. Beyond degrading functional performance, lattice defects strongly affect environment stability [11, 15, 16]. Structural imperfections serve as active sites for hydrolysis and oxidation, enabling Ti atoms at defect-adjacent positions to undergo rapid transform into TiO_2_. This compromises structural integrity and properties [11, 12, 17, 18]. While these observations highlight the influence of precursor chemistry and processing on defect evolution, systematic defect control and quantitative defect–property correlations remain largely unexplored.
To address this challenge, we build upon our recently introduced precursor-driven synthesis concept, here advancing it into a refined framework for deterministic control over in-plane defects in Ti_3_C_2_T_x_ MXenes. By rationally tuning the carbon-to-titanium ratio during ball-milling-assisted TiC synthesis and modulating Al content during Ti_3_AlC_2_ MAX phase formation, we achieve fine regulation of titanium vacancies (VTi), carbon vacancies (VC), oxygen substitutional defects (SO), and associated compressive lattice strain (ε) across synthesis stages. This improved methodology not only yields MXenes with the lowest defect densities reported to date, but also enables the fabrication of a systematic series of MXene samples with varied defect levels—providing, for the first time, a comprehensive mapping of defect density against electrical, thermal, optical, and environmental stability.
Through comprehensive structural and spectroscopic analysis, we trace the defect propagation from precursors to MAX to MXene, establishing quantitative correlations between defect types, lattice strain, and functional properties including electrical/thermal conductivities, IR emissivity, EMI shielding, Joule heating, and thermal insulation and camouflage. Furthermore, we construct time–temperature–performance retention maps that reveal how defect density governs oxidation kinetics. These results demonstrate that precursor chemistry is the primary determinant of nanoscale defect landscapes, introducing a generalizable design principle for synthesizing defect-minimized, high-performance, and oxidation-resistant MXenes.
Experimental Section
Materials and Chemicals
Titanium oxide (TiO_2_, 99.5%, 400 mesh), used as the precursor for TiC synthesis, was purchased from Junsei Chemical Co., Ltd. Lithium fluoride (LiF, 98.5%), titanium (Ti, 99.5%, 325 mesh), graphite (99.8%, 325 mesh), and aluminum (Al, 99.5%, 325 mesh) powders were obtained from Alfa Aesar. Hydrochloric acid (HCl, 37%) and hydrofluoric acid (HF, ACS reagent grade, 48%) were sourced from Sigma-Aldrich. All chemicals were employed as received, with no additional purification. Deionized water with a specific resistivity of 10^6^ Ω cm was consistently used throughout all procedures. A polypropylene membrane (Celgard; pore size, 0.064 µm) facilitated the preparation of free-standing MXene films by vacuum-assisted filtration.
Synthesis
The detailed synthesis procedures and characterization methods are provided in the Supporting Information as follows: TiC synthesis (Note S1), Ti_3_AlC_2_ synthesis (Note S2), and Ti_3_C_2_T_x_ MXene synthesis (Note S3). Additional methodological details include the determination of lattice parameters and lattice strain (Note S4), quantitative defect analysis (Note S5), first-principles calculations of carrier density, carrier mobility, and electrical conductivity measurements (Note S6), thermal conductivity evaluation (Note S7), infrared emissivity measurements (Note S8), Joule heating performance tests (Note S10), and oxidation kinetics determination (Note S11).
Characterizations
Surface and cross-sectional morphologies of MXene flakes and films were investigated by field-emission SEM (JSM-7600F, Japan) fitted with energy-dispersive spectroscopy (EDS). X-ray diffraction (XRD) measurements (D8, Bruker, USA) were carried out using Cu Kα radiation, scanning over a 3° to 70° 2θ range at a rate of 3° per min through a 10 × 10 mm^2^ window slit. Atomic-resolution high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) (JEM-2100F, Japan), operated at 200 kV, provided atomic-resolution structural information of individual MXene flakes. The delaminated Ti_3_C_2_T_x_ MXene flakes were dispersed in ethanol to form a dilute suspension and drop-cast onto lacey carbon–coated copper grids, followed by drying under vacuum. Images were acquired with the MXene flakes oriented along the [001] c-axis, enabling direct visualization of the in-plane Ti atomic lattice. HAADF-STEM conditions were used to provide Z-contrast imaging, where missing Ti lattice sites appear as dark spots, allowing reliable identification and statistical quantification of Ti vacancies. For HAADF-STEM analysis of the MAX phase, the finely crushed powder was dispersed in ethanol, allowed to settle down the large particles, and the supernatant was dropped-cast onto a lacey carbon Cu grid. Imaging was performed at thin edges and electron-transparent regions to minimize multiple scattering effects. The oxygen concentrations in the carbide powders were measured by oxygen elemental analysis (TCH-600, LECO Corporation). The precisely weighed samples of carbide powder were loaded into graphite crucibles and treated at high temperature in an impulse furnace. During this process, the oxygen presents in the carbide reacted with the graphite, generating carbon monoxide (CO) and carbon dioxide (CO_2_) gases. These gases were expelled from the furnace by a helium carrier gas, whose flow rate was regulated using a mass flow controller. To obtain accurate oxygen content readings, the gas mixture was passed through a heated catalyst bed, ensuring complete oxidation of CO to CO_2_. The total oxygen content was subsequently determined by measuring the produced CO_2_ with a nondispersive infrared (NDIR) detector. Similarly, the carbon concentrations in TiC and MAX were measured by carbon elemental analysis (CS-600, LECO Corporation). The total carbon content was determined by oxidizing all carbon species to CO_2_ at elevated temperatures using a C/S determinator, with oxygen employed as the combustion carrier gas. The released CO_2_ was subsequently quantified using a second NDIR detection system (see supporting information for more details). UV–Vis absorbance measurements were performed with a V-770 instrument (Jasco), covering the wavelength range of 200–1000 nm. Electrical conductivity was measured by means of a four-pin probe (MCP-TP06P PSP) interfaced with a Loresta-GP meter (Model MCP-T610, Mitsubishi Chemical, Japan). Inductively coupled plasma mass spectrometry (ICP-MS, PerkinElmer NexION 2000B) was used to measure the concentration of dissolved titanium (Ti) in the acidic solution after 24-h etching process. Raman spectroscopy was carried out using a LabRAM Soleil spectrometer (Horiba) with a 532 nm laser to analyze the ID and IG bands. The ID/IG ratio was used to evaluate the degree of structural disorder in the carbon lattice of the synthesized MXene samples. Infrared images and temperature data were acquired using an IR camera (FLIR A310, FLIR, Sweden) with an emissivity of 0.94. Detailed information on the theoretical calculations, defect analysis, thermal conductivity, IR emissivity, EMI shielding, and Joule heating measurements are provided in the Supporting Information. The density of the films was measured using a Sartorius Quintix 224-1SKR precision analytical balance with a density determination setup, following the Archimedes immersion method in accordance with ISO 1183-1:2019 (KS ISO 1183-1:2019). The sample was weighed in air and then while fully immersed in deionized water as the reference liquid at room temperature and the corresponding liquid density was used to calculate the film density from the buoyancy-induced mass difference.
Results and Discussion
Precursor-Guided Synthesis and Defect Evolution from TiC to Ti3AlC2 MAX to Ti3C2Tx MXenes
To elucidate how precursor chemistry dictates the formation and inheritance of in-plane defects in MXenes, we established a multistep synthesis framework in which each stage—TiC formation, Ti_3_AlC_2_ MAX phase growth, and subsequent etching to form Ti_3_C_2_T_x_ MXenes—is chemically engineered to regulate defect evolution (Fig. 1a). By deliberately adjusting the carbon stoichiometry and aluminum activity during the solid-state reactions, we achieve deterministic control over the formation of titanium vacancies (VTi), carbon vacancies (VC), substitutional oxygen (SO), and associated lattice strain (ε).Fig. 1. Precursor-controlled synthesis and in-plane defect analysis from TiC to Ti_3_AlC_2_ MAX to Ti_3_C_2_T_x_ MXenes. a Schematic diagram depicting the propagation of in-plane defects from TiC and Ti_3_AlC_2_ MAX phases to the resulting Ti_3_C_2_T_x_ MXene. Carbon vacancies (VC) within the MAX phase induce the removal of adjacent Ti atoms during etching, producing surface Ti vacancies (VTi) in MXenes. These structural defects exert a pronounced effect on the functional characteristics of MXenes. (b–d) XRD patterns illustrating precursor and product phases: b TiC–1/2/3, c Ti_3_AlC_2_–1/2/3/4, and d Ti_3_C_2_T_x–1/2/3/4 MXenes. e Lattice strain evolution across TiC, Ti_3_AlC_2, and Ti_3_C_2_T_x–1/2/3/4 samples are presented along a lattice parameter. f Schematic of V_C and substitutional oxygen (S_O_) defects within TiC and Ti_3_AlC_2_ phases. g–j Atomic-resolution HAADF-STEM images of Ti_3_C_2_T_x–1/2/3/4 flakes, with red circles marking in-plane Ti vacancy (V_Ti) defects. (Scale bar: 2 nm). k Quantified V_Ti_ defect concentrations in TiC, Ti_3_AlC_2_, and Ti_3_C_2_T_x_–1/2/3/4, as determined by HAADF-STEM image analysis. For comparison, data regarding Ti edge defects are also provided
To modulate defect populations at the earliest stage, TiC was synthesized by high-energy-ball-milling-assisted carbothermal reduction of TiO_2_ at 1500 °C, using varying carbon-to-TiO_2_ molar ratios (2.4, 2.7, and 3.1), yielding three distinct precursors (TiC–1/2/3) with systematically different stoichiometries and crystallinities (Note S1, Fig. S1, and Table S1). X-ray diffraction (XRD) analysis confirmed the synthesis of phase-pure TiC with the rock-salt structure (Fm–3m), and there was a consistent shift in the (200) diffraction peak toward lower 2θ values (from 41.92° to 41.76°), consistent with lattice expansion and lower strain accompanying increased carbon content (Fig. 1b) [9, 19]. The resulting lattice parameter along a increased from 4.30 to 4.32 Å, and compressive lattice strain decreased from − 0.51% to − 0.20%, indicating a reduction in the concentration of in-plane lattice defects (Fig. 1e, Note S4).
Quantitative defect analysis demonstrated that increasing the carbon ratio effectively suppressed major in-plane defects: SO decreased from 5.8% to 1.15%, VC from 11.1% to 3.65%, and VTi from 0.48% to 0.09% (Figs. 1f and S2, Table S1). These observations were validated by oxygen elemental analysis, carbon elemental analysis, and atomic-resolution high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) combined with selected area electron diffraction (SAED) (Note S5). The results indicate that carbon-rich synthesis conditions facilitate the formation of stoichiometric TiC while inhibiting oxygen incorporation and carbon deficiency—both primary causes of in-plane defects [9].
To investigate how these defect characteristics propagate during MAX phase formation, we synthesized Ti_3_AlC_2_–1/2/3 MAX phases from the respective TiC–1/2/3 precursors under stoichiometric condition (TiC:Ti:Al = 2:1:1). In addition, Ti_3_AlC_2_–4 MAX phase was prepared from the least defective TiC-3 under non-stoichiometric condition (TiC:Ti:Al = 2:1:2.2) to examine the effect of excess Al (Note S2, Fig. S3, Table S2). All MAX samples preserved the characteristic hexagonal P6_3_/mmc symmetry corresponding to (111), (200), and (220), while XRD peak shifts (e.g., (104) reflection) indicated expansion in both a and c lattice constants (a = 3.0607–3.0673 Å, c = 18.4991–18.5931 Å) alongside a decrease in lattice strain (ε) from − 0.30% to − 0.08% (along the a-axis) and from − 0.65% to − 0.14% (along the c-axis) as precursor quality improved (Fig. 1c, e, Note S4).
This structural evolution was accompanied by a reduction in defect concentrations: SO from 6.3% to 0.6%, VC from 4.6% to 2.78%, and VTi from 0.54% to 0.06% (Figs. 1f and S4, Table S2). Notably, the MAX phase Ti_3_AlC_2_-5, synthesized from the most defective TiC-1 under same non-stoichiometric conditions (TiC:Ti:Al = 2:1:2.2), exhibited a significantly higher lattice strain (ε = − 0.15, − 0.62 along the a- and c-axis, respectively) compared to Ti_3_AlC_2_-4 (from TiC-3), and achieved a much lower yield (57.3%) due to the formation of substantial Al_2_O_3_ and intermetallic TiAl_3_ impurities (Fig. S5, Table S2). These findings underscore the essential roles of carbon stoichiometry and Al activity in defect modulation during MAX phase synthesis. Carbon serves both as a key structural element and as a reducing agent, while surplus molten aluminum increases Ti/C diffusion and functions as an oxygen scavenger by forming Al_2_O_3_, which further suppresses in-plane defects [9, 10]. In case of TiC, the carbon sublattice occupies 1 out of 2 total lattice sites, whereas in Ti_3_AlC_2_, carbon occupies only 2 out of 6 total lattice sites. Consequently, even for samples of comparable structural quality, the absolute carbon vacancy concentration (VC) in Ti_3_AlC_2_-1 appears smaller (~ 5%) than that in TiC-1 (~ 11%). However, when VC is normalized by the fraction of carbon (X) sites in each lattice, the effective carbon vacancy concentrations in TiC and Ti_3_AlC_2_ become comparable, indicating consistent vacancy retention during the TiC to Ti_3_AlC_2_ phase transformation.
Subsequent etching of the MAX phases using an aqueous LiF/HCl/HF solution at 35 °C produced Ti_3_C_2_T_x–1/2/3/4 MXenes, yielding lateral flake sizes of 3–5 µm (Note S3, Fig. S6) [20]. The XRD patterns demonstrated systematic downshifts in the (002) reflection, corresponding to an increase in interlayer d-spacings (from ~ 0.93 to ~ 1.26 nm) from Ti_3_C_2_Tx–1 to − 4, which verifies both effective etching and delamination (Fig. 1d) [21]. TEM combined with SAED patterns verified the preservation of hexagonal symmetry inherited from the original MAX phases (Fig. S7), and enabled quantification of the lattice strain, which reduced from − 4.24% in Ti_3_C_2_Tx–1 to − 0.17% in Ti_3_C_2_Tx_–4 (along a-axis) (Fig. 1e, Note S4, Table S3), in agreement with improvements in lattice order attributed to effects of the precursor.
Although direct observation of VC and S_O_ in MXenes is hindered by both low atomic contrast and the presence of surface terminations [17], VTi densities were determined by atomic-resolution HAADF-STEM imaging along the c-axis (Fig. 1g–j). The measured VTi concentrations declined from 3.06% in Ti_3_C_2_T_x–1 to 0.12% in Ti_3_C_2_Tx–4, which is substantially higher than those found in their respective MAX and TiC precursors (Fig. 1k, Note S5, Table S3). This pronounced increase implies that structural defects within the MAX phase weaken local Ti bonding, thereby promoting Ti extraction during the etching process and increasing vacancy formation [10]. Inductively coupled plasma spectroscopy (ICP) analysis of the etchant after 24 h revealed a decrease in dissolved Ti content, dropping from 3% for Ti_3_C_2_Tx–1 to 0.29% for Ti_3_C_2_Tx–4, which further corroborates the HAADEF-STEM observations (Fig. S8). Notably, edge-site Ti dangling atoms (Ti edge defects) persisted at a minimal and nearly unchanged level (~ 0.01%) across all examined samples, independent of MXene type, since the lateral dimensions of the MXene flakes were consistently within the range of approximately 3–5 µm (Fig. 1k green spectra, Note S5). Consequently, in-plane defects—rather than edge defects—are identified as the predominant defect type influencing the properties of MXenes. Raman spectroscopy provided additional support for these observations. The I_D/I_G_ ratio—a metric associated with carbon disorder—was reduced from 0.83 (for Ti_3_C_2_T_x–1) to 0.26 (for Ti_3_C_2_Tx–4), indicating a decrease in V_C and S_O_ concentrations (Fig. S9) [22, 23].
Taken together, these findings confirm that in-plane defect densities in Ti_3_C_2_T_x_ MXenes are highly tunable via careful precursor-level selection and control. The nature of the final MXene’s defect state predominantly reflects the chemical composition and structural characteristics of both the TiC and Ti_3_AlC_2_ precursors. Furthermore, the cost-effective TiO_2_-based carbothermal approach enables precise defect management and provides a scalable strategy for producing high-quality MXenes with superior functional properties [9].
Defect–Property Relationship in Ti3C2Tx MXene
Electronic Transport Properties
To determine the impact of in-plane defects on charge transport, both computational and experimental analyses were performed over the full synthesis pathway from TiC to Ti_3_AlC_2_ and Ti_3_C_2_T_x. Density functional theory (DFT) simulations investigated how increasing concentrations of VTi, VC, and SO affect carrier density and mobility in TiC and Ti_3_AlC_2 (Note S6). With elevated defect densities, both carrier density and mobility showed significant decreases (Fig. 2a, b), resulting in calculated conductivity dropping for both TiC and Ti_3_AlC_2_ as a function of VTi, VC, and SO (Figs. 2c and S10, Tables S4–S6) [24, 25], in accordance with the classical transport relationship σ = neμ (where σ is conductivity, n is electron carrier density, e is charge of an electron (1.602 × 10^−19^ C), and μ is mobility). For Ti_3_C_2_T_x_ MXenes, where theoretical predictions are complicated by 2D confinement and complex surface terminations, electrical conductivity was experimentally measured using a four-point probe on approximately 10 µm-thick free-standing films. Conductivity values decreased from 26,000 (Ti_3_C_2_T_x–4) to 13,120, 10,200, and 7,550 S cm^−1^ for Ti_3_C_2_Tx–3, –2, and –1, respectively, correlating with defect density (Fig. 2c, green spectra). VTi defects lower free-electron density by removing electron-rich Ti atoms, while VC and SO defects intensify electron scattering, collectively degrading charge transport. These findings demonstrate that defect engineering derived from the precursor is highly effective for controlling electronic transport in MAX phases and MXenes, highlighting the importance of in-plane defect management to achieve high conductivity.Fig. 2. Influence of in-plane defects on the properties of Ti_3_C_2_Tx_ MXene films. a Theoretical electron carrier density and b electron mobility of TiC– and Ti_3_AlC_2_–1/2/3/4 samples with various defect concentrations. c Electrical conductivity of TiC (DFT calculations), Ti_3_AlC_2_ MAX phases (DFT calculations), and Ti_3_C_2_T_x_ MXenes (experimental measurements) with different defect concentrations. d In-plane thermal conductivity (black) and thermal diffusivity (blue), and e infrared emissivity of Ti_3_C_2_T_x–1/2/3/4 film samples with different defect concentrations. f Radar chart summarizing three core performance metrics of Ti_3_C_2_Tx_ MXenes with different defect densities. Each performance metric is normalized to its respective maximum value. Comparison of g electrical conductivity, h thermal conductivity, and i IR emissivity values with previous literature indicating the highly crystalline Ti_3_C_2_T_x_–4 MXene outperform all the previous literature. (Color figure online)
Thermal Conductivity
Thermal conductivity is a key material property that indicates a material’s capacity for heat transfer in response to a temperature gradient [26]. As the defect density was reduced from Ti_3_C_2_T_x_–1 to –4, thermal diffusivity (α) increased from 13.31 to 16.46, 17.24, and 20.70 mm^2^ s^−1^, resulting in corresponding increases in thermal conductivity (κ) to 33.1, 43.0, 45.3, and 57.0 W m⁻^1^ K⁻^1^ (Fig. 2d, Note S7, Table S7). This progression mirrors the rise in electrical conductivity and is consistent with the Wiedemann–Franz law (κ = LσT, where L is the Lorenz number (2.44 × 10^–8^ W Ω K^−2^) [27], σ is electrical conductivity, and T is absolute temperature), indicating that free electrons are primarily responsible for both charge and heat transport [27]. Additionally, lattice defects function as phonon-scattering centers, which further diminish thermal conductivity. These results verify that reducing in-plane defects leads to enhanced electronic and thermal transport in MXenes [28, 29].
Infrared Emissivity
Infrared emissivity (ε) measures a material’s capacity to emit infrared (IR) radiation compared to an ideal blackbody [30]. The emissivity was observed to decrease systematically from 0.23 to 0.05 across Ti_3_C_2_T_x–1 to –4 (Figs. 2e and S11, Note S8), confirming an inverse relationship with electrical conductivity [31]. Ti_3_C_2_Tx_–4, characterized by a reduced defect concentration, demonstrates enhanced electrical conductivity and increased free-electron density, which results in stronger IR reflection and lower emissivity [32]. These results underline that engineering the defect structure is an effective approach to improve IR reflectivity, opening new pathways for energy-saving thermal management and infrared camouflage [32].
A radar plot comparing the electrical, thermal, and infrared properties of Ti_3_C_2_T_x–1 through –4 demonstrates the critical influence of defect density on MXene properties (Fig. 2f). Ti_3_C_2_Tx_–4, which possesses the lowest defect density, delivers exceptional performance results—including an electrical conductivity of 26,000 S cm⁻^1^, thermal conductivity of 57 W m⁻^1^ K⁻^1^, average infrared emissivity of 0.05, exceeding those of its counterparts in every evaluated property (Fig. 2f). Notably, these results also surpass those of almost all previously reported MXene systems (Fig. 2g–i, Tables S8–S10), further substantiating that deliberate control of defect structure is essential for significantly improving the electrical, thermal, and infrared properties necessary for different applications.
Defect–Effect on Electromagnetic Shielding, Joule Heating, and Thermal Camouflage of Ti3C2Tx MXene
Electromagnetic Shielding
Electromagnetic interference (EMI) shielding is essential for safeguarding sensitive electronic equipment against external electromagnetic disturbances, ensuring dependable function in fields including aerospace, telecommunications, medical devices, and military systems [33, 34]. The total shielding effectiveness (SE_T_) of Ti_3_C_2_T_x_ films (10 µm thick) was investigated in the X-band (8.2–12.4 GHz), and a substantial rise was recorded from 66.1 dB (Ti_3_C_2_T_x–1) to 90.5 dB (Ti_3_C_2_Tx–4) as the defect density was reduced (Figs. 3a, b and S12, Note S9), achieving attenuation levels up to 99.9999999% [33]. Simon’s formula shows that shielding effectiveness is inherently correlated with electrical conductivity (Note S9) [35]. An elevated defect density compromises conductivity, subsequently reducing the reflection (SE_R) and absorption (SE_A_) contributions and decreasing overall SE_T_. On the other hand, Ti_3_C_2_T_x–4 delivers superior EMI shielding at a comparable thickness, highlighting that minimizing defects is fundamental in the development of lightweight, high-efficiency shielding materials for future generations of electronic and communication technologies.Fig. 3. Influence of in-plane defects on the EMI shielding, Joule heating, and thermal camouflage capability of Ti_3_C_2_Tx_ MXene films. a-b EMI SE_T_, SE_R_, and SE_A_ of Ti_3_C_2_T_x–1/2/3/4 in the X-band frequency window (8.2–12.4 GHz). c The steady-state temperature of the Ti_3_C_2_Tx–1/2/3/4 as a function of the square of voltage. d Joule heating capability of Ti_3_C_2_Tx–1/2/3/4 and the associated steady-state surface temperature and power density at 1.5 V. e Infrared thermal images depicting surface temperature versus a background temperature of 100 °C, demonstrating IR camouflage effect. f Temperature versus time data obtained from the e indicating the surface temperature against the 100 °C and respective reduction in radiation temperature of Ti_3_C_2_Tx–1/2/3/4. g Radar chart summarizing three core performance metrics of Ti_3_C_2_Tx_ MXenes with different defect densities. Each performance metric is normalized to its respective maximum value. Comparison of h EMI (SE_T_) versus thickness, and i Joule heating temperature versus applied voltage with previous literature indicating the highly crystalline Ti_3_C_2_T_x_–4 MXene outperform all the previous literature
Joule Heating
Joule heating—the process of converting electrical energy into thermal energy—is vital for applications including de-icing, wearable heaters, and thermal management [36]. MXenes have garnered considerable attention in this field because of their fast and effective heating response [37, 38]. A wide range of equilibrium temperatures was achieved at safely applied voltages up to 1.5 V. As the voltage increased from 0.5 to 1.5 V, the Ti_3_C_2_T_x–1/2/3/4 films exhibited a clear rise in temperature. Furthermore, the stabilized temperatures at steady state displayed a distinct linear dependence on the square of the applied voltage (Fig. 3c), highlighting the dominant influence of voltage in governing equilibrium heating behavior. The Ti_3_C_2_Tx_–1/2/3/4 MXene films showed gradually increasing saturation temperatures of 129, 171, 192, and 263 °C, along with heating power densities of 1.95, 2.25, 2.4, and 3.15 W cm^−2^, respectively,S14 when subjected to a low applied voltage of 1.5 V (Figs. 3d, S13, S14, and Note S10). Both the saturation temperature and power density increased further with higher applied voltages (Figs. S13 and). Collectively, these findings convincingly illustrate that minimizing defect density significantly improves Joule heating performance, predominantly by raising electrical conductivity and thus allowing higher power output under a given applied voltage.
Infrared Camouflage
Infrared (IR) camouflage plays a critical role in concealing objects from thermal imaging and IR detection systems, supporting important applications in military, aerospace, and surveillance fields [30, 39]. To assess thermal camouflage effectiveness, Ti_3_C_2_T_x_ MXene films were tested on a 100 °C hot plate, with surface temperatures recorded via IR thermography. The apparent surface temperature dropped from 43.5 °C (Ti_3_C_2_T_x–1) to 35.1 °C (Ti_3_C_2_Tx–4) (Fig. 3e), which corresponded with the noted decrease in emissivity. Ti_3_C_2_Tx–4, which has the lowest emissivity of 0.05, reflected 95% of incident IR radiation due to its elevated electrical conductivity and higher number of free electrons, efficiently reducing thermal emission. These outcomes underscore the potential of defect-engineered MXenes as promising advanced materials for applications in IR stealth and camouflage. The reduction in radiation temperature is a key parameter for evaluating thermal camouflage capability of a material. Notably, the Ti_3_C_2_Tx–4 exhibited a significantly higher reduction in the radiation temperature of 64.9 °C against background temperatures of 100 °C than that of Ti_3_C_2_Tx_–1/2/3 exhibiting 56.5, 60.3, and 62.8 °C, respectively (Figs. 3f and S15).
A radar plot comparing the multifunctional performance capability of Ti_3_C_2_T_x–1 through –4 demonstrates the critical influence of defect density on material performance. Ti_3_C_2_Tx_–4, which possesses the lowest defect density, delivers exceptional performance results—including EMI shielding effectiveness of 90.5 dB at 10 µm thickness, a Joule heating saturation temperature of 263 °C (at 1.5 V), and reduction in the radiation temperature of 64.9 °C against background temperatures of 100 °C due to its very low infrared emissivity of 0.05—exceeding those of its counterparts in every evaluated property (Fig. 3g). Notably, these performance matrices also surpass those of almost all previously reported MXene systems in term of EMI shielding, and joule heating (Fig. 3h, i, Tables S11–S12), further substantiating that deliberate control of defect structure is essential for developing high-performance MXenes customized for cutting-edge electronic, thermal, optical, and camouflage applications.
Defect Density–Oxidation Stability Relationship
Crystal defects have been widely recognized as active sites that facilitate oxidative degradation in MXenes [40]; however, the quantitative relationship between in-plane defect density and oxidation kinetics has not been systematically investigated. To fill this gap, we conducted a comprehensive study on the environmental stability of Ti_3_C_2_T_x_ MXenes with controlled in-plane defect concentrations (Ti_3_C_2_T_x–1 to Ti_3_C_2_Tx_–4), emphasizing their oxidation mechanisms in dilute aqueous dispersions (0.02 mg mL^−1^). Utilizing such low concentrations accelerates oxidation by substantially increasing the interaction of the MXene surface with water and oxygen molecules.
Freshly prepared Ti_3_C_2_T_x_ dispersions exhibited a distinctive greenish hue, which transitioned to whitish shades as oxidation progressed (Fig. 4a) [41]. After 30 days of storage under ambient conditions, Ti_3_C_2_T_x–4—fabricated with the lowest defect density—maintained its initial appearance and flake structure, whereas Ti_3_C_2_Tx–1 was fully converted to TiO_2, as verified through TEM and SAED analyses (Fig. 4b) [42]. Intermediate samples (Ti_3_C_2_T_x–2 and –3) experienced partial degradation, clearly demonstrating that the extent of oxidation depends on defect density. To obtain quantitative insight into the oxidation pathway, we monitored the normalized absorbance associated with the surface plasmon resonance at 780 nm, which serves as a spectral marker for delaminated MXene flakes [43]. All the tested samples showed an exponential decrease in absorbance over time, with higher defect densities corresponding to a more rapid decline (Figs. 4c and S16–S19, Note S11). This degradation process followed first-order reaction kinetics, and rate constants (k), determined by fitting the data to an exponential model, were found to decrease as both the defect concentration and storage temperature were lowered (Fig. 4d, Table S13) [44–46]. Using temperature-dependent kinetics, we determined the activation energies (Eₐ) of oxidation through the Arrhenius equation (Note S11) [43]. Ti_3_C_2_Tx–1 exhibited the lowest activation energy (Eₐ) at 32.3 kJ mol⁻^1^, while Ti_3_C_2_Tx–4 displayed a much higher barrier of 72.7 kJ mol^−1^ (Fig. 4e), clearly demonstrating a quantitative relationship between in-plane defect density and oxidation resistance. MXenes with higher structural order and reduced defect density showed significantly higher oxidation stability.Fig. 4. Oxidation kinetics of Ti_3_C_2_Tx–1/2/3/4 MXenes. a Digital images of dilute aqueous dispersions (0.02 mg mL⁻^1^) of Ti_3_C_2_Tx–1/2/3/4, displayed as freshly prepared samples (top) and after 30 days of ambient storage (bottom), demonstrate progressively intensified color changes resulting from oxidation. b TEM images, together with corresponding SAED patterns (insets), of aged Ti_3_C_2_Tx_ flakes stored for 30 days, reveal structural degradation that correlates with different defect densities. (Scale barsin SAED patterns are 10 nm^−1^.) c Normalized absorbance intensity at 780 nm is plotted over time for dispersions stored at 3 °C. Dashed lines represent first-order exponential decay model fits. d Oxidation reaction rate constants for Ti_3_C_2_T_x–1/2/3/4 dispersions as a function of storage temperature. e Activation energies for oxidation of Ti_3_C_2_Tx–1/2/3/4 are plotted, emphasizing the relationship between structural disorder and oxidative reactivity. f–i Contour plots map showing the degree of oxidation for Ti_3_C_2_Tx–1/2/3/4, depicting changes as a function of both temperature and time. The red line indicates the threshold of 90% retention in absorbance at 780 nm. Importantly, Ti_3_C_2_Tx_–4 (0.02 mg mL⁻^1^), which exhibits the lowest defect density and greatest crystallinity, preserves over 90% of its original structure for four months and thus exhibits superior oxidation resistance. These continuous lines in the plots indicate fitting to the first-order kinetic equation, I = I0 e^−kt^, where I is the peak intensity at 780 nm after a certain time, I0 is the initial intensity, t is the time, and k is the rate constant. (Color figure online)
Mechanistically, in-plane defects—such as carbon vacancies and oxygen substitutions—cause destabilization of adjacent Ti atoms, thereby increasing their susceptibility to hydrolysis and oxidation. These sites preferentially adsorb water and oxygen molecules, thus promoting acid-catalyzed oxidation pathways that initiate structural degradation [43]. Moreover, charge redistribution at defects accelerates redox reactions, with electrons accumulating near carbon vacancies and holes near Ti vacancies, which enhances oxidation of neighboring atoms and facilitates Ti^4+^ diffusion and TiO_2_ nucleation [11, 17, 18, 42].
To capture the degradation behavior, we generated time–temperature contour plots (Fig. 4f–i) that map the onset and progression of degradation phenomena. Dark blue areas represent chemically stable states, while dark red corresponds to complete degradation. The bold red line delineates the 10% oxidation threshold (i.e., 90% absorbance retention), sets as the criterion for stability. Ti_3_C_2_T_x–1 underwent rapid degradation, reaching 10% oxidation after only 0.45 days, while Ti_3_C_2_Tx–2 and –3 required 2.4 and 5.9 days, respectively. By contrast, Ti_3_C_2_Tx_–4 stayed stable for 120 days in ambient conditions.
Oxidation was found to be further reduced at higher dispersion concentrations [47]. At 8 mg mL^−1^, Ti_3_C_2_T_x_–4 maintained its green coloration and layered morphology even after four months of storage, indicating almost complete suppression of oxidative decomposition (Fig. S20a, b). This phenomenon is attributed to the limited supply of oxygen and water per MXene flake in more concentrated dispersions [47].
Collectively, these results highlight the predominant influence of in-plane lattice defects on the oxidation kinetics of Ti_3_C_2_T_x_ MXenes. Lowering the density of these defects significantly increases the activation energy for degradation, thereby improving environmental stability and extending material lifetimes. In conjunction with optimized storage approaches, defect engineering at the precursor stage provides an effective and scalable method for generating MXenes with high oxidation resistance, supporting their long-term use in aqueous or humid settings.
Oxidation-Induced Property Degradation in Ti3C2Tx MXenes
To quantitatively evaluate the consequences of oxidation on the functional performance of Ti_3_C_2_T_x_ MXenes, we established systematic correlations between the extent of oxidation and crucial physical properties, including electrical conductivity, thermal conductivity, EMI shielding efficiency, and Joule heating output. All measurements were conducted on Ti_3_C_2_T_x_–4 samples possessing distinct, controlled levels of oxidation in highly dilute aqueous dispersions (0.02 mg mL^−1^), which accelerate oxidative deterioration and enable precise monitoring of performance degradation.
The oxidation degree was determined as the normalized UV–Vis absorbance peak intensity at 780 nm (I/I0), which provides a reliable and quantitative metric for evaluating MXene preservation in dispersion. Free-standing MXene films were produced via vacuum-assisted filtration from dispersions displaying I/I0 values between 1.0 and 0.2, corresponding to successive stages of structural degradation.
As oxidation progressed the films underwent evident visual and structural modifications: The metallic luster characteristic of fresh samples faded (Fig. S21a), and increased surface roughness was observed as a result of TiO_2_ by-product formation (Fig. S21b). XRD measurements identified the development of anatase TiO_2_ peaks ((101) and (200) reflection) while displaying a weakened Ti_3_C_2_T_x_ (002) reflection (Fig. S21c). Cross-sectional SEM analysis directly indicated the accumulation of TiO_2_ crystals and a simultaneous reduction of MXene flakes as the level of oxidation grew (Fig. 5a).Fig. 5. Correlation between oxidation degree and functional property degradation in Ti_3_C_2_T_x_ MXenes. a Cross-sectional SEM images of Ti_3_C_2_T_x–4 films prepared from dispersions with different oxidation degrees, as determined by normalized UV–Vis absorbance at 780 nm (I/I0 = 1.0, 0.8, 0.5, and 0.2), which illustrate both structural degradation and TiO_2 formation. b Quantitative relationship between normalized absorbance intensity (I/I0 at 780 nm) and normalized property metrics: electrical conductivity (σ/σ0), EMI shielding effectiveness (SE_T_/SE_T0_), Joule heating temperature, and thermal conductivity (κ/κ_0_). c–f Contour plots depicting changes in normalized c electrical conductivity, d thermal conductivity, e EMI shielding effectiveness (SE_T_), and f Joule heating performance in diluted Ti_3_C_2_T_x_–4 dispersions (0.02 mg mL⁻^1^) as functions of storage temperature and elapsed time. The color scale indicates retention percentages of initial property values, with red contours delineating the 90% retention level. These continuous lines in the plots indicate fitting to the first-order oxidation kinetic equation, I = I0* e*^−kt^, where I is the peak intensity at 780 nm after a certain time, I0 is the initial intensity, t is the time, and k is the rate constant
This structural evolution led to marked deterioration in functional properties. As I/I0 fell from 1.0 to 0.2, the 10 µm-thick films showed pronounced drops in performance: Electrical conductivity declined from 26,000 → 284 S cm^−1^, thermal conductivity fell from 57 → 5.4 W m^−1^ K^−1^, EMI shielding SE_T_ dropped from 89.2 → 10.6 dB, and Joule heating saturation temperature decreased from 263 to 49.9 °C (Fig. 5b).
These pronounced losses primarily result from the transformation of highly conductive Ti_3_C_2_T_x_ to insulating TiO_2_, drastically reducing both carrier density and mobility. Additionally, oxidation-driven disorder undermines interlayer registry and causes greater interlayer spacing, further intensifying electron and phonon-scattering effects. Because electrical conductivity underpins thermal conductivity, EMI shielding, and Joule heating capabilities, all properties experience proportional deterioration (Fig. S22a–d). Notably, the decrease in EMI shielding effectiveness is significantly slower than the decline in electrical conductivity during oxidation. This behavior is attributed to the formation of TiO_2_ between MXene layers, which enhances interfacial polarization, dipolar relaxation losses, and multiple internal reflections, partially compensating for the loss of electrical conductivity [48–50].
To our knowledge, this represents the first comprehensive study quantitatively correlating the extent of oxidation with MXene property degradation. Such findings facilitate devising a unified model that relates oxidation levels to performance loss, supporting predictive assessments of environmental stability. For clarity, time–temperature–performance contour diagrams were developed for each property (Fig. 5c–f). Blue regions signify > 90% retention of initial properties, while red regions show < 10% retention. The demarcated bold red curve visualizes the 90% retention boundary. At room temperature, Ti_3_C_2_T_x–4 retained 90% of its original attributes over: 75 days (electrical conductivity), 117 days (thermal conductivity), 135 days (EMI shielding), and 113 days (Joule heating). These data underscore the remarkable environmental robustness of Ti_3_C_2_Tx_–4, even when subjected to accelerated oxidation.
Moreover, increasing the dispersion concentration to 8 mg mL^−1^ led to dramatic suppression of oxidation. Under these conditions, Ti_3_C_2_T_x_–4 maintained: > 97% electrical conductivity, > 95% thermal conductivity, > 98% EMI shielding, Joule heating, and IR shielding performance after 120 days of storage in air (Fig. S23a–e), demonstrating near-complete resistance to environmental degradation.
Collectively, these findings provide a quantitative basis for linking structural oxidation to the progressive loss of functional properties in MXenes. This established relationship offers essential design principles for developing high-performance, durable MXenes, highlighting the importance of engineering precursor-level defects to enhance environmental stability and ensure consistent performance in practical applications.
Conclusion
Through a precursor-guided synthesis strategy, we have successfully fabricated Ti_3_C_2_T_x_ MXenes with systematically regulated in-plane defects. By tailoring carbon stoichiometry during TiC synthesis and controlling aluminum content during Ti_3_AlC_2_ MAX phase formation, we achieved deterministic modulation of titanium and carbon vacancies, substitutional oxygen, and the resulting lattice strain across synthesis stages. This refined methodology enables the production of MXenes with unprecedentedly low defect densities and excellent multifunctional performance, including an electrical conductivity of 26,000 S cm^−1^, thermal conductivity of 57 W m^−1^ K^−1^, electromagnetic shielding effectiveness of 90.5 dB at 10 µm, Joule heating performance of 263 °C at 1.5 V, ultralow infrared emissivity (0.05), and markedly enhanced oxidation resistance with an activation energy of 72 kJ mol^−1^. Furthermore, this study establishes—for the first time—a comprehensive and quantitative correlation between defect density, lattice strain, oxidation kinetics, and multifunctional performance. The constructed time–temperature–property retention maps provide predictive guidelines for the long-term stability of MXenes under realistic conditions. Overall, our findings demonstrate that precursor chemistry is the governing factor for nanoscale defect evolution, introducing a generalizable design principle for defect-minimized, high-performance, and oxidation-resistant MXenes. This work bridges a long-standing gap in understanding defect–property relationships and provides a robust foundation for the reliable integration of MXenes into advanced electronics, optoelectronics, thermal management, and EMI shielding applications.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 15.4 MB)