Unexpected Phase Stability of Anatase TiO2 Nanocrystals at Ultrahigh Temperature via Surface Restructuring

Unexpected Phase Stability of Anatase TiO2 Nanocrystals at Ultrahigh Temperature via Surface Restructuring | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Unexpected Phase Stability of Anatase TiO 2 Nanocrystals at Ultrahigh Temperature via Surface Restructuring Yong Wang, Xiaoyun Guo, Yujing Zhang, Chao Yang, Yunhao Lu, Yang Ou, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6475787/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract The phase stability of nanocrystals is of great importance to its performance in different applications, and extensive studies have been devoted to understanding the nature of superheating/supercooling, to achieve the desired long-term phase stability in practice. While the fundamental mechanism remains elusive due to the absence of atomic-level insights into dynamic structural evolution of nanocrystals under extreme conditions. Herein, through in situ atomic level spherical aberration-corrected scanning transmission electron microscopy we revealed an abnormal phase stability of the individual single crystalline anatase TiO 2 at ultra-high temperatures governed by surface effects. The in situ atomic level observations at the exceptionally high temperatures show a highly anomalous phenomenon that the individual anatase TiO 2 nanorod single crystal could maintain the anatase phase at 1250°C, instead of turning into the rutile phase, which exceeded the reported anatase-to-rutile phase transition point nearly 650°C higher. As the temperature rises above the phase transition point, the surfaces of anatase TiO 2 nanorods undergo a series of atomic reconstructions, which not only reduce the total energy of the system, but also act as a kinetic “surface locking” effect preventing the rutile nucleation. This work shows that surfaces could have a critical effect on the phase stability of single crystalline nanomaterials, and the proposed “surface locking” mechanism opens up a new way for manipulating the thermal stability of nanocrystals. Physical sciences/Materials science/Nanoscale materials/Nanoparticles Physical sciences/Nanoscience and technology/Nanoscale materials/Nanoparticles In situ transmission electron microscopy anatase TiO2 thermal stability surface structure evolution surface reconstruction Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Main text Precisely controlling the materials’ physical and chemical properties, dominated by the crystalline phase structure that featured the periodic atomic arrangements with specific atomic bonds, symmetry, and thermodynamic parameters, through phase engineering has long been sought. Phase stability, particularly thermostability, which refers to the material's ability to maintain its physical and chemical properties under thermal stress, is extremely important for materials and devices, greatly affecting long-term stability and efficiency in applications 1 – 9 . When external conditions change, phase transitions occur caused by changes in the internal energy of nanomaterials, which is a common phenomenon widely exist nature and widely used by various technologies, such as water vaporization in the water cycle, structural transition of shape memory alloys, information storage, superconducting transition of superconductors 10 – 15 . The critical temperature and/or pressure point at which a phase transition occurs is referred to as the phase transition point driven by the free energy which is crucial for phase transition occurrence, typically serving as a crucial parameter to estimate phase thermostability. The phase thermostability and transition point are not only influenced by external environmental factors (temperature and pressure) but also by the structural parameters of the crystal itself 16 – 21 . Therefore, extensive studies have been devoted to understanding the nature of superheating/supercooling, and to design the specific structures with the desired phase stability 22 – 25 . Remarkable progress has been made in the study of phase transitions in bulk materials over the past few decades, and several methods have been successfully developed, such as embedded particles, alloy particles, self-assembled nanocrystals and so on 26 – 31 . Compared to the thermostability manipulation in bulk materials via interface engineering, much less is known about the contribution of surface effects to phase thermostability in nanomaterials, which are expected to play an increasingly dominant role with material size decreasing 16 – 18 . However, the current studies primarily focus on the surface atomic structure and properties itself, leaving a substantial gap in understanding whether and how to utilize surface effects to enhance thermostability 32 – 35 . The complexity of nanocrystal surface structures, which consist of facets, terraces, edges, and corners, makes it difficult to understand their interactions with the external environment fully 36 , 37 . On the other hand, the lack of in-situ structural evolution information at elevated temperatures—encompassing both surface and bulk changes—hampers establishing a clear relationship between thermal stability and surface structural evolution. Recent advancements in in-situ TEM have enabled real-time atomic-scale visualization of structural evolution, providing new insights into both phase transitions and surface reconstruction of nanomaterials 38 – 42 . These exciting advancements provide us a great opportunity to explore the atomic nature of materials supercooling/superheating. Herein, to clarify the surface effect on the phase transition and understand the nature of superheating of nanocrystals, we in situ explored the structural evolution of well-defined anatase TiO 2 nanocrystals at the atomic level via Cs-corrected scanning transmission electron microscopy (STEM), where micro-sized single crystalline anatase nanorods with various well-defined crystal surface facets were chosen as our research object. TiO 2 Anatase-to-Rutile transition, a classical phase transition system widely studied in the past decades, is widely believed to irreversibly occur at approximately 600°C 26 . While in the practical studies, the anatase TiO 2 samples showed distinct thermostabilities in the separated experiments, and there is no consensus on the mechanism of the unconventional thermostabilities. In our prior research, the atomic structure and the reconstruction mechanisms of a variety of anatase TiO 2 surfaces, such as (001), (100), (101), and (301), etc., were systematically studied via in situ TEM 43 – 47 . Based on the understanding of these surface properties, in this paper, we documented a highly anomalous phenomenon in this phase transition investigation that the anatase phase can still exist stably even when heated to 1250°C instead of turning into the rutile phase, which exceeded the reported phase transition point nearly 650°C higher. It is found that the surface of anatase TiO 2 forms a stable structure with a sequence of surface reconstructions taking place with the rise of temperature, resulting in a stable (101) crystal surface with lower surface energy. This low-energy surface plays a pivotal role in the phase thermal stability which significantly stabilized the metastable anatase phase via a “surface locking” mechanism, which kinetically inhibit surface rutile nucleation. To reveal the underlying mechanism of the unconventional stability of the TiO 2 nanocrystal, and to comprehensively estimate the influence of different surfaces on the phase transition, we chose the anatase TiO 2 nanorods as the sample and captured the in situ HRSEM images of the nanorods at different temperatures to show the morphology evolution of the anatase TiO 2 nanorods. The anatase TiO 2 nanorods were synthesized by a two-step hydrothermal method 48 (the details are shown in the experimental section), which is exposed to several different surfaces and has excellent crystallinity. Since there is a critical size in the anatase-to-rutile phase transition (ranges from 14 nm to 45 nm), the micro-sized TiO 2 nanocrystals used in the experiments are far beyond this critical size (in this size, rutile is more energetically favored than anatase), which could exclude the ‘‘size’’ issue during the experiments. In the experiments, we selected the well-defined individual nanocrystals that were not in contact with other nanocrystals to minimize the other structural factors involved in the experiments. The in situ heating experiments were performed to dynamically monitor the surface structural evolution of individual anatase TiO 2 nanorods, from 20°C to 1000°C under vacuum (10 − 5 Pa). The morphology of the selected anatase TiO 2 nanorod did not show a notable change from 20°C to 500°C (Fig. 1 a-b). When the temperature was beyond 600°C, the morphology of the nanorod also remained almost unchanged (Fig. 1 c-d), with length and width unchanged contrast 20°C (~ 650 nm; ~65 nm). While the surface of the nanorod started becoming no longer atomically flat, and some small protrusions appeared on the side surface of the nanorod, which indicated the possible surface reconstruction occurred. Above 700°C, the morphology of the TiO 2 nanorod exhibited significant change, the length became significantly shorter and the width slightly wider. The small protrusions on the side surfaces gradually evolved into larger protrusions and the tip of the nanorod became flatter (as shown in the enlarged images in Fig. 1 h-m). The observation of another individual nanorods showed a similar process up to 1000°C and 1200°C (Supplementary Figs. 1 and 2, Supplementary Video 1). To observe the atomic structural evolution of the anatase TiO 2 nanocrystals at elevated temperatures, we performed the in-situ heating experiments in a spherical aberration (Cs-) corrected scanning transmission electron microscope. The in-situ experiments were performed in vacuum and gas conditions via the Cs-corrected scanning transmission electron microscopes (Titan G2 80–200, Hitachi HF-5000), with micro-electromechanical system based heating holders (DENS solutions DH30 double-tilt heating holder, Hitachi single-tilt heating holder). The phase structure of anatase TiO 2 could transformed into rutile around ~ 600°C 26 . Since the TiO 2 crystal is very sensitive to electron beam (e-beam) irradiation, several measurements have been adopted to minimize the e-beam effect in the experiments. During the in-situ experiment, a very tiny dose (∼0.6 nA) was used, and the electron beam was turned on only when the photos were taken to reduce the influence of the electron beam on the sample 43 , 47 . Figure 2 a shows the schematic diagram of a typical as-synthesized anatase TiO 2 nanorod. The size of the nanorod is about 650 nm long and 65 nm wide. The axial direction of the nanorod is along the [001] direction, and the sidewall is mainly exposed by (100) surfaces with atomic flatness (Fig. 2 b). The tip of the nanorod is curved, which exposed by some high index surfaces, such as (301) facet with an orderly stepped structure containing succession of a (101) terrace separated by (001) step, (101) facet which is exposed by both 6-fold and 5-fold coordinated Ti atoms (Ti 6c and Ti 5c ), with 2- and 3-fold oxygen atoms (O 2c and O 3c ) located on the ridge of the sawtooth-like structure, (001) facet exposed by 5-fold coordinated Ti (Ti 5c ) atoms, 3-fold coordinated O atoms (O 3c ), and 2-fold coordinated O (O 2c ) atoms, etc. (Fig. 2 c-e), as well documented in our previous research 43 – 46 . The two-dimensional crystalline lattice 0.35 nm (101) and 0.37 nm (100) indicate the anatase phase. The prepared nanorods had good crystallinity, atomically flat surfaces, and few defects, which provided a good platform for in-situ study. To illustrate the phase transition process at the atomic level well, we choose the single nanorod in the heating experiment as the object, viewed along [010] zone axis. As reported by the previous literatures, the phase transition points were concentrated around 600°C, investigated by both ex-situ and in-situ routes 26 . While unexpectedly, we did not observe the phase transition above 650°C in our in-situ experiments. It is found that anatase phase TiO 2 nanorods could still maintain anatase phase even at a high temperature of 1250°C. Compared to the as-synthesized anatase TiO 2 nanorods, both the morphology and surface atomic structure show significant reconstruction at 1250°C (Fig. 2 a). The originally flat surface of the selected nanorod became uneven, and the initial atomically flat (100), (301) surfaces are reconstructed into protrusions formed by (101) facets with (001) surface transformed into the (1×4) reconstruction of ADM model (Fig. 2 f-i), which is consistent with the previous studies. While we carefully checked the HAADF STEM images in different areas, we did not find the nucleation of the rutile phase, even when the temperature exceeded the phase transition point of nearly 650°C. Considering the size of the nanorod is significantly larger than the critical size, the phase transition is expected to occur easily. Therefore, this observed unexpected ultra-stable anatase TiO 2 nanorod suggests an unconventional overheating phenomenon occurred in an individual single crystal, which was repeated in several cases (Supplementary Figs. 1, 2 and 3). Since the morphology and surface structure show significant change during the heating experiments, the in situ STEM images were acquired to illustrate the detailed atomic evolution process of surface structures. Since the surface of the nanorod is mainly exposed by (100), (301), (001), and (101) surfaces, the atomic structural evolution of each surface is decoupled. Initially, the (100) surface showed a bulk-truncated surface structure (Fig. 3 a). During the heating process, this atomically flat surface did not show visible change from room temperature to ~ 600°C (Fig. 3 a-d). When the temperature was beyond 600°C (Fig. 3 e-f), some fresh fuzzy islands began to appear on the surface, indicating that some atomic migration occurred. At 900°C (Fig. 3 g), the originally atomically flat (100) surface became obviously rough, with some fresh atomic columns appearing on the surface (white arrows) and some surface atomic columns disappearing (red arrows). With further temperature elevation, these small protrusions grew higher and became sharpened (Fig. 3 h), with (101) surface dominated. Finally, the (100) surface was completely reconstructed into well-arranged sharp (101) faceted protrusions (mean height: 1.2 nm; interval distance along [001]: 6.4 nm), as the temperature reached 1250°C (Fig. 3 i-j). Compared with the reconstruction of the (100) surface, directly evolving into (101) surface protrusion, the structural evolution process of the (301) surface was more complex (Fig. 4 a-j), which experienced two stages. Initially, the (301) surface showed an orderly stepped structure, which could be obtained by a succession of a (101) terrace with three titanium atom pairs, separated by a (001) step. In the first stage, this stepped (301) surface remained nearly unchanged from room temperature to 400°C (Fig. 4 a-b). At higher temperatures, the atoms on the surface of the nanorods gradually became unsharp and blurry. As indicated by the white arrow in Fig. 4 c, some atomic columns became blurred or even disappeared at 600°C, indicating that surface atomic migration occurred, as shown by the atoms circled in the green dotted box. The surface reconstruction became more obvious at 700°C (Fig. 4 d). As a result, at higher temperatures (750°C, 800°C), the (301) surface gradually transformed from a stepped (101) terrace (Fig. 4 b) into a different stepped structure (Fig. 4 e-f) consisting of (100) terraces and (101) facets, in the first stage. In the following heating process (the second stage), the newly formed (100) terraces also became unstable and tended to reconstruct into (101) surface-dominated protrusions as depicted in Fig. 3 . Therefore, at higher temperatures, with the formation of (101) faceted protrusions, the proportion of (100) surface significantly reduced (Fig. 4 g), and eventually disappeared (Fig. 4 h). The protrusions also grew larger and became more sharpened, eventually (301) surface was also completely transformed into a stepped structure dominated by the (101) surfaces at 1250°C (Fig. 4 i-j). Compared to the reconstructed (100) surface, the formed (101) protrusions by the reconstruction of the (301) surface have a preferential orientation proportion: (101) > (10 − 1). In addition, the (001) surfaces of the anatase TiO 2 nanorod tip were kept bulk-truncated structure from room temperature to 600°C, and then transformed into the (1×4) reconstruction by an ADM model, which could be stabilized at even 1250°C (Supplementary Fig. 4), as discussed in previous work 25 . At last, the structural evolution of (101) surface was also explored. While different from the significant reconstructions of other surfaces, the (101) remain stable and did not show significant reconstruction within 1250°C, which confirms that the surface of (101) is the thermodynamically stable surface of anatase TiO 2 (Supplementary Fig. 5). Overall, except the (101) surface, all the other crystal surfaces were reconstructed: (100) and (301) were reconstructed into stepped (101) surface; (001) was reconstructed via the ADM model. The final result is that the nanorods are reconstructed to form a stable surface envelope consisting of stepped (101) and (001) surfaces. To further reveal why and how these reconstructions happen during the reshaping of anatase TiO 2 nanorod, time-resolved STEM imaging was employed to track real-time mass transport dynamics. Initially, two distinct steps existed on the surface, as framed in dotted green lines in Fig. 5 a. Over time, the left step diminished, and even nearly vanishing at t = 9.5 s shown in Fig. 5 a. Whereas, the right step sharpened through atomic accumulation and migrated ~ 2.0 nm rightward (green dotted lines denote diffusion-mediated atomic additions). The diffusion process features synchronized atom depletion in Area 1 and the accumulation of atoms in area 2 and area 3, as visualized in Fig. 5 b-d (white and colored dotted lines mark pre- and post-diffusion atomic configurations, respectively). The detailed information was shown in the kinetics curve of the atomic diffusion in Fig. 5 f. On the one hand, the reduction of atoms in area 1 was the fastest and nonlinear, where the rapid migration of atoms along the atomic layer in the period of t = 2.5 ~ 4.5 s led to the rapid reduction of atoms layer by layer in area 1 shown in Fig. 5 b. On the other hand, the accumulation of atoms in area 2 and area 3 were different. In area 2, the increase of atoms was slow at first. After t = 4.5 s, atoms began to accumulate rapidly, indicating that nucleation was more difficult, but once nucleated, the step grew faster and gradually sharpened as shown in Fig. 5 c. At t = 9.5 s, the atoms rapidly accumulate 3 layers along the original atomic layer leading to the step sharper, with the lattice plane on both sides composed of (101) facets. In contrast, there was no significant accumulation of atoms in area 3 shown in Fig. 5 d, just the length of the layers added along the existing atomic layer ladder changed repeatedly over time. In addition, the number of atoms decreased in area 1 was much greater than the number of atoms increased in area 2, indicating that the excess atoms migrated rapidly layer by layer to the right along the atomic steps of area 3. Throughout the process, the rapid diffusion of surface atoms suggests that, under our experimental conditions, atoms could migrate freely. The diffusion process ultimately leads to atom accumulation at steps, forming sharp protrusions with low-surface-energy (101) facets, resulting in the morphological evolution depicted in Fig. 1 . The mass transportation direction is consistent with the structural evolution showed in Supplementary Video 3, where atoms migrate from the nanorod tip toward its midsection. Now the question comes up as to the nature of the ultrastability of the individual TiO 2 single crystal nanorod, which could exist at an extremely high temperature of ~ 1250°C, approximately ~ 650°C higher than its phase transition point without phase transition 26 . Notably, considering that the critical size between the anatase phase and rutile phase is in a range of ~ 14–45 nm 26 , 35 , 49 , 50 , the size of our sample far exceeds this range, it is anticipated the transition should be more favorable for the anatase to rutile phase transition. Thus, this unexpected overheating of the individual anatase single crystal nanorod represents an anomalous phenomenon, significantly contradicting the conventional understanding of TiO 2 stability. Interestingly, in contrast, in our additional experiments, a completely different phenomenon was found that the phase transition readily occurred when many TiO 2 nanorods stacked together. The nucleation of rutile crystallite formed at the multi-grain boundaries of adjacent particles, as shown in Supplementary Fig. 6. While, within a certain range, only a single rutile crystal nucleates and grows, and polycrystalline rutile formation is not observed, which also implies that kinetic nucleation is an exceedingly difficult process, requiring the stacking of anatase TiO 2 nanorods in a specific manner. Therefore, it could be concluded that the absence of phase transformation under high-temperature conditions is not due to a lack of source materials or limitations in mass transport, but rather is kinetically constrained by the failure to form rutile nuclei on the low-energy anatase TiO 2 surfaces. The phase transition from anatase to rutile has been studied extensively over the past 30 years, both theoretically and experimentally 26 , 35 , 51 – 57 . According to previous studies, rutile nucleation is the key step in the anatase-to-rutile phase transition, occurring via a (112) twin boundary in anatase 58 – 60 . In individual nanorod experiments, the formation kinetic of this nucleation site is extremely challenging. The (100) and (301) surfaces on the individual TiO 2 nanorod preferentially undergo reconstruction into the (101) surface rather than the (112) surface due to the higher surface energy (0.04 eV/Å 2 ) under high-temperature treatment, as described in Fig. 6 a-c. Surface energy calculations reveal that the (101) surface has the lowest energy (0.008 eV/Å 2 ), as shown in Fig. 6 c-d, making it the most stable surface. DFT + U calculations further confirm that \(\:{E}_{surface}\) (301) > \(\:{E}_{surface}\) (100) > \(\:{E}_{surface}\) (101), indicating a favorable pathway for this reconstruction (Fig. 6 d). After the reconstruction, the individual anatase TiO 2 nanocrystal is mainly exposed by the (101) surfaces. This surface with very low surface energy, significantly prevents the surface rutile nucleation for the elevated barrier. Additionally, the rapid diffusion of surface atoms avoids the atom accumulation on surface which further reduce nucleation probability. These combined effects significantly increase the nucleation barrier during the phase transition, resulting in the abnormal thermal stability of the anatase phase. We refer to this phenomenon as surface locking. This finding highlights the critical impact of surface on the stability of nanomaterials, which significantly affect the secondary phase nucleation and mass transport in the phase transformation of nanomaterials. The ultra-high thermal stability of anatase TiO 2 nanocrystals achieved through “surface locking” reveals the unique surface dynamics of nanoparticles and their profound impact on phase stability and transformation kinetics. Conclusions In conclusion, through an in situ atomic-scale exploration of the structural evolution of well-defined anatase TiO 2 nanocrystals, we revealed the abnormal ultra-high thermal stability of the individual anatase TiO 2 nanorod single crystal, which could maintain the anatase phase structure at 1250°C. Unexpectedly, in the heating process, different from the expected anatase-to-rutile phase transition, only the surface reconstructions and reshaping of the individual anatase TiO 2 nanorod occurred, with gradually reconstructed into the stable structures mainly exposed by (101) surfaces with deceased surface energy. The restructuring not only reduce the total energy of the system, but also kinetically inhibited the surface rutile nucleation, leading to the abnormal phase stability of the anatase TiO 2 at an ultra-high temperature. Our results reveal atomic-scale insights into the critical impact of surface atomic structure and restructuring on the phase stability of nanomaterials, and the proposed “surface locking” may serve as a fundamental principle for the phase stability manipulation of future functional nano crystalline materials design operating at higher temperatures. Materials and Methods 1. Synthesis of anatase TiO 2 nanorods: The anatase TiO 2 nanorods were synthesized via a facile hydrothermal method that was reported previously 48 . In a typical synthesis, 1 g Na-titanates (wet, isolated by centrifugation without drying) were dispersed into 40 mL deionized water solution, then transferred to a 50 mL Teflon-lined stainless-steel autoclave, and sealed. The autoclave was put into an oven, heated at 200°C for 24 h, and cooled naturally in air, producing white tetragonal faceted-nanorods (TFNR) precipitates. These white precipitates were isolated from the solution by centrifugation and subsequently washed with deionized water several times, and finally dried at 60°C in vacuum for 10 h. 2. In situ STEM observations: The in situ STEM experiments were conducted in an FEI Titan G 2 80–200 scanning transmission electron microscope (200 kV), with a spherical aberration (C s -) corrector and a high spatial resolution (~ 0.8 Å). Before the test, to ensure high-quality imaging, all low-order aberrations (up to order 5) have been calibrated to an acceptable level, such as C s < 0.5 µm, A1 (2-fold astigmatism) < 2 nm, A2 (3-fold astigmatism) < 20 nm, and B2 (coma) < 20 nm. The convergence angle employed for STEM imaging was ~ 21 mrad. The annular detection angle for high-angle annular dark field (HAADF) image is set to ~ 53–200 mrad, while for bright-field (BF) image, it is adjusted to ~ 0–20 mrad. During the in situ STEM experiments, the as-obtained TiO 2 nanorods are dispersed in ethanol and dropped on the heating chips, which are then transferred into the TEM chamber via a Wildfire D6 double tilt heating holder (DENS solutions). Instead of directly heating our sample to the elevated temperatures, in the step-by-step heating process, the temperature interval was set to 100°C with a heating rate of ~ 5°C s − 1 . TiO 2 is usually very sensitive to electron beam (e-beam) irradiation and it is easy to undergo irradiation decomposition under electron beam irradiation 61 , 62 . In the STEM mode, the beam radiation damage of the electron beam will be more severe because the electron beam is convergent and has a higher beam current density. In order to minimize the radiation damage of the electron beam to the maximum extent, we take a series of measures to reduce the radiation dose of the electron beam in the sample (∼0.6 nA). In particular, the target surface area was moved under the electron beam only during the imaging process. 3. The DFT calculations: The spin-polarized DFT calculations were performed using the Perdew-Burke-Ernzerh (PBE) functional to optimize all surface structures, implemented in the Vienna ab initio simulation package (VASP). Projector-augmented wave (PAW) potentials were used to represent core-valence interactions. To account for the localized Ti d states, we applied DFT + U with an effective U value of 4.1 eV. During optimization, atomic positions were relaxed until residual forces were below 0.02 eV/Å, and the convergence criterion for electronic self-consistent energy was set to 10 − 4 eV. A plane-wave cutoff of 400 eV was used to expand the Kohn-Sham valence states, and a vacuum layer of 10 Å was included to prevent interactions between periodic images. Surface energies of perfect anatase (001), (100), (101), (102), (103), (112), and (301) surfaces were calculated using (1×1) slab models with corresponding K-point grids up to (10×10×1), (10×4×1), (6×10×1), (3×10×1), (5×10×1), (8×4×1) and (2×9×1), respectively. The surface energy \(\:\left({E}_{surface}\right)\) is given by: $$\:{E}_{surface}=\frac{{E}_{slab}-{nE}_{bulk}}{2{A}_{slab}}$$ where \(\:{E}_{slab}\) is the total energy of the slab, \(\:{E}_{bulk}\) is the energy per atom in the bulk structure, n is the number of atoms in the slab, and \(\:{A}_{slab}\) is the surface area of the slab. The factor of 2 accounts for the two surfaces present in the slab model. Declarations Data availability All data needed to evaluate the conclusions are available in the main text or the supplementary information. Acknowledgements: We acknowledge the financial support of the National Key Research and Development Program (2023YFA1506904, 2022YFA1505500), the National Natural Science Foundation of China (52025011, 52422311, 22302173, 52171019, and 52473242)), the Zhejiang Provincial Natural Science Foundation of China (LR23B030004), the Fundamental Research Funds for the Central Universities. The authors thank Dr. Zhemin Wu and Dr Ruiyang You in Center of Electron Microscopy at Zhejiang University for STEM data analysis, and thank Jiahang Zhang for his contribution to the graphical illustration of the mechanistic model. Author contributions Y.W. initiated the project. Y.W. and W.Y. supervised this project. Y.W., and W.Y. designed the experiments. X.G. and W.Y. conducted the in situ STEM experiments. X.G. and Y.O. contributed to the sample preparation. Z.H., Y.Z., Y.L., C.Y., contributed to the theoretical calculations and explanation. B.Z., Y.J., H.Y., and Z.Z. participated in the analysis and discussion. X.G., and Y.Z. contributed equally to this work. References Yang, Y. et al . Amidination of ligands for chemical and field-effect passivation stabilizes perovskite solar cells. Science. 386, 898-902 (2024). Park, Y. M. et al . Engineering ligand reactivity enables high-temperature operation of stable perovskite solar cells. Science. 381, 209-215 (2023). Zhou, X., Li, X. Y. & Lu, K. Enhanced thermal stability of nanograined metals below a critical grain size. Science. 360, 526-530 (2018). Bu, T. L. et al . Lead halide–templated crystallization of methylamine-free perovskite for efficient photovoltaic modules. Science. 372, 1327-1332 (2021). Le, V. N. et al . Improved organic electrochemical transistor stability using solvent degassing and chemical doping. Nat. Electron (2025): https://doi.org/10.1038/s41928-024-01297-8. Han, G. G. D., Li, H. S. & Grossman, J. C. Optically-controlled long-term storage and release of thermal energy in phase-change materials. Nat. Commun. 8, 1446 (2017). Chen, J., Wang, Z.-K. & Liao, L.-S. Defect passivation and crystallization control of perovskite films for photovoltaic application. Mater. Today. Nano. 15 , 100118 (2021). Jia, Y. et al. Recent advances in energy storage and applications of form-stable phase change materials with recyclable skeleton. Carbon. Neutralization. 3 , 313–343 (2024). Tang, Z. et al. Tightened 1D/3D carbon heterostructure infiltrating phase change materials for solar–thermoelectric energy harvesting: Faster and better. Carbon. Energy. 5 , e281 (2023). Han, G. G. D., Li, H.S. & Grossman, J. C. Optically-controlled long-term storage and release of thermal energy in phase-change materials. Nat. Commun. 8 , 1446 (2017). Zhang, Y. N. et al . High-temperature superconductivity with zero resistance and strange-metal behaviour in La 3 Ni 2 O 7−δ . Nat. Phys. 20 , 1269–1273 (2024). Rao, F. et al . Reducing the stochasticity of crystal nucleation to enable subnanosecond memory writing. Science. 358 , 1423–1427 (2017). Ding, K. Y. et al . Phase-change heterostructure enables ultralow noise and drift for memory operation. Science. 366 , 210–215 (2019). Zewdie, G. M. et al. Chemical Design Principles for Cache-Type Sc–Sb–Te Phase-Change Memory Materials. Chem. Mater. 31 , 4008–4015 (2019). Atinafu, D. G. et al . Unveiling sustainable nano-enabled phase change materials for high thermal stability and energy storage capacity. J. Energy. Storage. 60 , 106650 (2023). Chen, Y. et al. Phase engineering of nanomaterials. Nat. Rev. Chem. 4 , 243–256 (2020). Li, W. B., Qian, X. F. & Li, J. Phase transitions in 2D materials. Nat. Rev. Chem. 6 , 829–846 (2021). Huang, H. H. et al. Recent progress of TMD nanomaterials: phase transitions and applications. Nanoscale. 12 , 1247–1268 (2020). Barnard, A. S. & Curtiss, L. A. Prediction of TiO 2 Nanoparticle Phase and Shape Transitions Controlled by Surface Chemistry. Nano Lett. 5 , 1261–1266 (2005). Zhang, H. Z. & Banfield, J. F. Structural Characteristics and Mechanical and Thermodynamic Properties of Nanocrystalline TiO 2 . Chem. Rev. 114 , 9613–9644 (2014). Sang, L. X., Zhao, Y. X. & Burda, C. TiO 2 Nanoparticles as Functional Building Blocks. Chem. Rev. 114 , 9283–9318 (2014). Cao, C. Z. et al. Bulk ultrafine grained/nanocrystalline metals via slow cooling. Sci. Adv. 5 , eaaw2398 (2019). Li, X. Y., Zhou, X. & Lu, K. Rapid heating induced ultrahigh stability of nanograined copper. Sci. Adv. 6 , eaaz8003 (2020). Xiao, X. M. et al. Superheating of grain boundaries within bulk colloidal crystals. Nat. Commun. 13 , 1599 (2022). Losurdo, M. et al. Thermally stable coexistence of liquid and solid phases in gallium nanoparticles. Nat. Mater. 15 , 995–1002 (2016). Hanaor, D. A. H. & Sorrell, C. C. Review of the anatase to rutile phase transformation. J. Mater. Sci. 46 , 855–874 (2011). Lu, K. & Li, Y. Homogeneous Nucleation Catastrophe as a Kinetic Stability Limit for Superheated Crystal. Phys. Rev. Lett. 80 , 4474–4477 (1998). Mei, Q. S. & Lu, K. Melting and superheating of crystalline solids: From bulk to nanocrystals. Prog. Mater. Sci. 52 , 1175–1262 (2007). Jin, Z. H., Gumbsch, P., Lu, K. & Ma, E. Melting Mechanisms at the Limit of Superheating. Phys. Rev. Lett. 87 , 055703 (2001). Shannon, R. D. & Pask, J. A. Kinetics of the Anatase-Rutile Transformation. J. Am. Ceram. Soc. 48 , 391–398 (1965). Zhang, L. et al . Superheating of Confined Pb Thin Films. Phys. Rev. Lett. 85 , 1484-1487(2000). Vittadini, A., Casarin, M. & Selloni, A. Chemistry of and on TiO 2 -anatase surfaces by DFT calculations: a partial review. Theor. Chem. Acc. 117 , 663–671 (2007). Lu, W. G. et al . Phase transition of individual anatase TiO 2 microcrystals with large percentage of (001) facets: a Raman mapping and SEM study. Phys. Chem. Chem. Phys. 25 , 3199–3210 (2023). Sheng, H.C., Xiao, B. B. & Jiang, X. B. Size dependence of the phase transition temperature of metal nanocrystals. Physica B. 667 , 415193 (2023). Zhang, H. Z. & Banfield, J. F. Phase transformation of nanocrystalline anatase-to-rutile via combined interface and surface nucleation. J. Mater. Res. 15 , 437–448 (2000). Gao, W. P., Hood, Z. D. & Chi, M. F. Interfaces in Heterogeneous Catalysts: Advancing Mechanistic Understanding through Atomic-Scale Measurements. Acc. Chem. Res. 50 , 787–795 (2017). Yang, X. F. et al . Single-Atom Catalysts: A New Frontier in Heterogeneous Catalysis. Acc. Chem. Res. 46 , 1740–1748 (2013). Yu, R. et al. Direct Subangstrom Measurement of Surfaces of Oxide Particles. Phys. Rev. Lett. 105 , 226101 (2010). He, M.-R., Yu, R. & Zhu, J. Reversible Wurtzite–Tetragonal Reconstruction in ZnO(101̅0) Surfaces. Angew. Chem. Int. Ed. 51 , 7744–7747 (2012). He, M.-R., Yu, R. & Zhu, J. Subangstrom Profile Imaging of Relaxed ZnO(101̅0) Surfaces. Nano Lett. 12 , 704–708 (2012). Yuan, W. T. et al . In situ manipulation of the active Au-TiO 2 interface with atomic precision during CO oxidation. Science. 371 , 517–521 (2021). Yuan, W. T. et al . Visualizing H 2 O molecules reacting at TiO 2 active sites with transmission electron microscopy. Science. 367, 428-430 (2020). Yuan, W. T. et al . In Situ STEM Determination of the Atomic Structure and Reconstruction Mechanism of the TiO 2 (001) (1 × 4) Surface. Chem. Mater. 29 , 3189–3194 (2017). Yuan, W. T. et al. Real-Time Observation of Reconstruction Dynamics on TiO 2 (001) Surface under Oxygen via an Environmental Transmission Electron Microscope. Nano Lett. 16 , 132–137 (2016). Yuan, W. T. et al . Unveiling the Atomic Structures of the Minority Surfaces of TiO 2 Nanocrystals. Chem. Mater. 30 , 288–295 (2018). Fang, K. et al . An Environmental Transmission Electron Microscopy Study of the Stability of the TiO 2 (1 × 4) Reconstructed (001) Surface. J. Phys. Chem. C. 123 , 21522–21527 (2019). Yuan, W. T. et al . Direct in-situ insights into the asymmetric surface reconstruction of rutile TiO 2 (110). Nat. Commun. 15 , 1616 (2024). Li, J. M. & Xu, D. S. Tetragonal faceted-nanorods of anatase TiO 2 single crystals with a large percentage of active {100} facets. Chem. Commun. 46 , 2301–2303 (2010). Gribb, A. A. & Banfield, J. F. Particle size effects on transformation kinetics and phase stability in nanocrystalline TiO 2 . Am. Mineral. 82 , 717–728 (1997). Zhang, H. Z. & Banfield, J. F. Thermodynamic analysis of phase stability of nanocrystalline titania. J. Mater. Chem. 8 , 2073–2076 (1998). Okeke, G., Hammond, R. B. & Antony, S. J. Influence of size and temperature on the phase stability and thermophysical properties of anatase TiO 2 nanoparticles: molecular dynamics simulation. J. Nanopart. Res. 15 , 1584 (2013). Satoh, N., Nakashima, T. & Yamamoto, K. Metastability of anatase: size dependent and irreversible anatase-rutile phase transition in atomic-level precise titania. Sci. Rep. 3 , 1959 (2013). Patra, S. et al . Phase stability frustration on ultra-nanosized anatase TiO 2 . Sci. Rep. 5 , 10928 (2015). Zhang, J. et al . Importance of the Relationship between Surface Phases and Photocatalytic Activity of TiO 2 . Angew. Chem. Int. Ed. 47 , 1766–1769 (2008). Mao, Q. et al . Sintering-Induced Phase Transformation of Nanoparticles: A Molecular Dynamics Study. J. Phys. Chem. C. 119 , 28631–28639 (2015). Alimohammadi, M. & Fichthorn, K. A. Molecular Dynamics Simulation of the Aggregation of Titanium Dioxide Nanocrystals: Preferential Alignment. Nano Lett. 9 , 4198–4203 (2009). Koparde, V. N. & Cummings, P. T. Phase Transformations during Sintering of Titania Nanoparticles. ACS Nano. 2 , 1620–1624 (2008). Zhu, S.-C., Xie, S.-H. & Liu, Z.-P. Nature of Rutile Nuclei in Anatase-to-Rutile Phase Transition. J. Am. Chem. Soc. 137 , 11532–11539 (2015). Penn, R. L. & Banfield, J. F. Morphology development and crystal growth in nanocrystalline aggregates under hydrothermal conditions: insights from titania. Geochim. Cosmochim. Acta. 63 , 1549–1557 (1999). Penn, R. L. & Banfield, J. F. Formation of rutile nuclei at anatase (112) twin interfaces and the phase transformation mechanism in nanocrystalline titania. Am. Mineral. 84 , 5–6 (1999). Chen, Q. L. et al. Imaging Beam-Sensitive Materials by Electron Microscopy. Adv. Mater. 32 , 1907619 (2020). Jiang, N. Electron beam damage in oxides: a review. Rep. Prog. Phys. 79 , 016501 (2015). Additional Declarations There is NO Competing Interest. Supplementary Files SuppplementaryInformation.docx Supplementary Information Supplementary Figs. 1-6 Movie1.mp4 Supplementary Movie1 Movie2.mp4 Supplementary Movie2 Movie3.mp4 Supplementary Movie3 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6475787","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":452038039,"identity":"a6842ab3-20ea-4159-8289-1adc52121c24","order_by":0,"name":"Yong Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3klEQVRIie3RvQrCMBSG4SOFdgm6xkH0EgqBIhT0VhIK7VKdHRwCQid/Vu9DcI4E2iXq2sFNcHHRxU0xegGxboJ5l285z3QAbLZfrgFA9ThfkCb/mvjiPRWIX8zkGY0PvdVenDCMQsa9rTATtYtDlJ+itRAxBpUwjobUTMo0IKkro2DDY1zLJOMY+RXIQ0ZkApo8qhFyHGSy57svwiuQplKBc59LihVEXZonJEOpmdSLKbkub7LfWChWXsZha+EpM+kIcLFexgHR9zNd472uzcG56O0DeOLTsc1ms/1pT7TaSU4AND78AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-9893-8296","institution":"Zhejiang University","correspondingAuthor":true,"prefix":"","firstName":"Yong","middleName":"","lastName":"Wang","suffix":""},{"id":452038040,"identity":"f5a7fa98-555f-4fcf-84b4-ac7a2e181288","order_by":1,"name":"Xiaoyun Guo","email":"","orcid":"","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyun","middleName":"","lastName":"Guo","suffix":""},{"id":452038041,"identity":"a8321fe1-2b4f-46bb-862b-0e1e547d82e0","order_by":2,"name":"Yujing Zhang","email":"","orcid":"","institution":"Beijing University of Chemical Technology","correspondingAuthor":false,"prefix":"","firstName":"Yujing","middleName":"","lastName":"Zhang","suffix":""},{"id":452038042,"identity":"1ade13d3-527d-4891-9ea4-ebf05b8841c2","order_by":3,"name":"Chao Yang","email":"","orcid":"","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Chao","middleName":"","lastName":"Yang","suffix":""},{"id":452038043,"identity":"36c3b3b0-e390-4c26-85c2-11c580d12f9c","order_by":4,"name":"Yunhao Lu","email":"","orcid":"https://orcid.org/0000-0001-6825-7206","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Yunhao","middleName":"","lastName":"Lu","suffix":""},{"id":452038044,"identity":"40a19bd8-d2ca-4a27-a2a3-b5e2482812bb","order_by":5,"name":"Yang Ou","email":"","orcid":"","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Ou","suffix":""},{"id":452038045,"identity":"3e2c7cfc-1580-46f5-a161-a5f657c63355","order_by":6,"name":"Beien Zhu","email":"","orcid":"","institution":"Shanghai Institute of Applied Physics, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Beien","middleName":"","lastName":"Zhu","suffix":""},{"id":452038046,"identity":"cb2c6ce6-df3b-4797-8355-902aed4d80cb","order_by":7,"name":"Ying Jiang","email":"","orcid":"","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Ying","middleName":"","lastName":"Jiang","suffix":""},{"id":452038047,"identity":"edafaf80-43e7-48d6-a4b8-3736ea022863","order_by":8,"name":"Zhong-Kang Han","email":"","orcid":"https://orcid.org/0000-0003-1489-6824","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Zhong-Kang","middleName":"","lastName":"Han","suffix":""},{"id":452038048,"identity":"324253c8-f51a-442c-bf35-644d6b72a901","order_by":9,"name":"Wentao Yuan","email":"","orcid":"https://orcid.org/0000-0001-7450-789X","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Wentao","middleName":"","lastName":"Yuan","suffix":""},{"id":452038049,"identity":"d67b561b-9da0-4581-867a-127fa1e3ea1f","order_by":10,"name":"Hangsheng Yang","email":"","orcid":"https://orcid.org/0000-0002-5968-9223","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Hangsheng","middleName":"","lastName":"Yang","suffix":""},{"id":452038050,"identity":"6cb471e7-f3e3-45ff-92c6-e0bd13f698e5","order_by":11,"name":"Yi Gao","email":"","orcid":"https://orcid.org/0000-0001-6015-5694","institution":"Shanghai Advanced Research Institute, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Gao","suffix":""},{"id":452038051,"identity":"e52f9b98-580c-4e13-8c69-784601fcfa46","order_by":12,"name":"Ze Zhang","email":"","orcid":"https://orcid.org/0000-0002-1739-9033","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Ze","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-04-18 04:00:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6475787/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6475787/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82060348,"identity":"448a6e71-c858-4a22-9fc7-7ca37cfb1a1b","added_by":"auto","created_at":"2025-05-06 11:31:34","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":699244,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe morphology evolution of anatase TiO\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e nanorod. \u003c/strong\u003eIn situ HRSEM images of a typical anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod in the heating process in vacuum (TEM column pressure: 5×10\u003csup\u003e-5\u003c/sup\u003e\u003csub\u003e \u003c/sub\u003ePa). \u003cstrong\u003ea-g,\u003c/strong\u003e A typical morphology evolution process of the anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod at different temperatures. \u003cstrong\u003eh-m\u003c/strong\u003e, The enlarged HRSEM images of red and green box areas marked in (\u003cstrong\u003ea\u003c/strong\u003e, \u003cstrong\u003ec\u003c/strong\u003e, \u003cstrong\u003ef\u003c/strong\u003e) show the detailed morphology evolution at 20 °C (\u003cstrong\u003ea\u003c/strong\u003e), 600 °C (\u003cstrong\u003ec\u003c/strong\u003e), and 900 °C (\u003cstrong\u003ef\u003c/strong\u003e), respectively. Red and green arrows indicate protrusions on the surface of the TiO\u003csub\u003e2\u003c/sub\u003e nanorods.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/a1726878a766e55f23eac08e.png"},{"id":82060347,"identity":"7e95983b-edcd-4251-8ad3-00d116e12ed5","added_by":"auto","created_at":"2025-05-06 11:31:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":620782,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIn situ HAADF STEM images show the structural evolution of TiO\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e nanorod at high temperatures in vacuum (Column pressure: 10\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e-5\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e Pa), viewed along the [010] direction.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e, Schematic diagram of morphology changes of a typical anatase TiO\u003csub\u003e2 \u003c/sub\u003enanorod before and after heating. \u003cstrong\u003eb-e\u003c/strong\u003e, The atomic level HAADF STEM images of (100) (\u003cstrong\u003eb\u003c/strong\u003e), (301) (\u003cstrong\u003ec\u003c/strong\u003e), (001) (\u003cstrong\u003ed\u003c/strong\u003e), and (101) (\u003cstrong\u003ee\u003c/strong\u003e) surfaces at 20 °C. The insets in (\u003cstrong\u003eb, d\u003c/strong\u003e) are the corresponding atomic structural models of (100), and (001), respectively (Ti, gray; O, red). \u003cstrong\u003ef\u003c/strong\u003e-\u003cstrong\u003ei\u003c/strong\u003e, The in situ atomic level HAADF STEM images show the surface reconstructions of (100) (\u003cstrong\u003ef\u003c/strong\u003e), (301) (\u003cstrong\u003eg\u003c/strong\u003e), (001) (\u003cstrong\u003eh\u003c/strong\u003e), and (101) (\u003cstrong\u003ei\u003c/strong\u003e) surfaces at 1250 °C. The insets in (\u003cstrong\u003eh, i\u003c/strong\u003e) are the atomic structural models of ADM model of (1×4) reconstructed (001) surface, and (101), respectively (Ti, gray; O, red).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/755cc390ac813bfa0f72e9b4.png"},{"id":82060349,"identity":"16da4334-b65d-4cf4-b153-b0552cbf935c","added_by":"auto","created_at":"2025-05-06 11:31:34","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":916629,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe structural evolution of anatase TiO\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e nanorod (100) surface.\u003c/strong\u003e \u003cstrong\u003ea-h\u003c/strong\u003e, The in situ HAADF STEM images show the atomic structural evolution of anatase TiO\u003csub\u003e2\u003c/sub\u003e (100) surface at different temperatures, in vacuum (TEM column pressure: 5×10\u003csup\u003e-5\u003c/sup\u003e\u003csub\u003e \u003c/sub\u003ePa).\u003cstrong\u003e i\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eThe atomic level HAADF STEM image acquired at 1250 °C. \u003cstrong\u003ej\u003c/strong\u003e, the enlarged image of (\u003cstrong\u003ei\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/fa8f48e967c3b45342965760.png"},{"id":82061153,"identity":"178cfc1e-2366-4eb0-8a4c-8c6b2328a348","added_by":"auto","created_at":"2025-05-06 11:39:34","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":985315,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe structural evolution of anatase TiO\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e nanorod (301) surface.\u003c/strong\u003e \u003cstrong\u003ea-h\u003c/strong\u003e, The in situ HAADF STEM images show the atomic structural evolution of anatase TiO\u003csub\u003e2\u003c/sub\u003e (301) surface at the different temperatures, in vacuum (TEM column pressure: 5×10\u003csup\u003e-5\u003c/sup\u003e\u003csub\u003e \u003c/sub\u003ePa).\u003cstrong\u003e i\u003c/strong\u003e, The atomic level HAADF STEM image acquired at 1250 °C. \u003cstrong\u003ej\u003c/strong\u003e, the enlarged image of (\u003cstrong\u003ei\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/7277e522effdf1d698c99edd.png"},{"id":82060350,"identity":"8df18ec3-c34b-44ae-9acc-032a43169574","added_by":"auto","created_at":"2025-05-06 11:31:34","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":465949,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSelected snapshots from a STEM video tracking the mass transportation during the dynamic process of surface diffusion of a typical anatase TiO\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e nanorod\u003c/strong\u003e (Supplementary Videos 2 and 3). The in situ STEM images were collected at 1000 °C in vacuum (TEM column pressure: 5×10\u003csup\u003e-5\u003c/sup\u003e Pa), viewed from [010] direction. The green dashed triangle frames and green arrows indicate the change in the position of the step tip. The dotted white lines and color lines indicate the positions of atomic arrangement on the surface at initial and after diffusion respectively. \u003cstrong\u003ea\u003c/strong\u003e, Sequential HAADF STEM images of the anatase TiO\u003csub\u003e2\u003c/sub\u003e surface, acquired at 0, and 9.5 s, collected from Supplementary Video 2. The yellow dotted lines show the moving direction and distance of the step enclosed by the green dashed triangle frame. \u003cstrong\u003eb-d\u003c/strong\u003e, The enlarged sequential HAADF STEM images of the anatase TiO\u003csub\u003e2\u003c/sub\u003e surface, acquired at 0, 2.5, 4.5, and 9.5 s, collected from Supplementary Video 2, corresponding to area 1 (\u003cstrong\u003eb\u003c/strong\u003e), area 2 (\u003cstrong\u003ec\u003c/strong\u003e), and area 3 (\u003cstrong\u003ed\u003c/strong\u003e) of (\u003cstrong\u003ea\u003c/strong\u003e). (\u003cstrong\u003ee\u003c/strong\u003e) Quantification of the changes in the number of atoms in area 1, area 2, and area 3 of (\u003cstrong\u003ea\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/6ca20312c558afa6646018fd.png"},{"id":82060353,"identity":"a58ea532-2f14-4591-9fa2-d9aca7176a8c","added_by":"auto","created_at":"2025-05-06 11:31:34","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":249380,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic diagram illustrating the surface reconstruction pathway. a-b\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eSchematic diagram of morphology changes of single anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod (\u003cstrong\u003ea\u003c/strong\u003e), and multiple nanorods (\u003cstrong\u003eb\u003c/strong\u003e) before and after heating. \u003cstrong\u003ec\u003c/strong\u003e, Surface energies of different surface cuts. \u003cstrong\u003ed\u003c/strong\u003e, Schematic diagram of the surface reconstruction path.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/834450ac75e6ce51010c9300.png"},{"id":86391574,"identity":"6bce5b95-dc9c-45be-b199-a468dbefcd29","added_by":"auto","created_at":"2025-07-10 06:58:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4710525,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/a6303654-8fce-482c-8384-228dfda8225a.pdf"},{"id":82060360,"identity":"7abe81fd-8022-4459-b59d-96f426457605","added_by":"auto","created_at":"2025-05-06 11:31:34","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4584610,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary Figs. 1-6\u003c/p\u003e","description":"","filename":"SuppplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/11f747abbf3874c2ac6e8e07.docx"},{"id":82060369,"identity":"1370ada4-a313-44b1-b1d3-35b33bf236f0","added_by":"auto","created_at":"2025-05-06 11:31:34","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":5066973,"visible":true,"origin":"","legend":"Supplementary Movie1","description":"","filename":"Movie1.mp4","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/89f179ea3d79b88c6e1fc2a6.mp4"},{"id":82061161,"identity":"19bd3b2b-8594-4738-83ba-ce0637e10a7d","added_by":"auto","created_at":"2025-05-06 11:39:34","extension":"mp4","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":15565580,"visible":true,"origin":"","legend":"Supplementary Movie2","description":"","filename":"Movie2.mp4","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/31e19cd010ac48742d28f0e0.mp4"},{"id":82060374,"identity":"4570d44a-1c1d-4b0c-8dbf-ae1f171c613c","added_by":"auto","created_at":"2025-05-06 11:31:35","extension":"mp4","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":13671158,"visible":true,"origin":"","legend":"Supplementary Movie3","description":"","filename":"Movie3.mp4","url":"https://assets-eu.researchsquare.com/files/rs-6475787/v1/df2abdd0313bf0de22de4e16.mp4"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"\u003cp\u003eUnexpected Phase Stability of Anatase TiO\u003csub\u003e2\u003c/sub\u003e Nanocrystals at Ultrahigh Temperature via Surface Restructuring\u003c/p\u003e","fulltext":[{"header":"Main text","content":"\u003cp\u003ePrecisely controlling the materials\u0026rsquo; physical and chemical properties, dominated by the crystalline phase structure that featured the periodic atomic arrangements with specific atomic bonds, symmetry, and thermodynamic parameters, through phase engineering has long been sought. Phase stability, particularly thermostability, which refers to the material's ability to maintain its physical and chemical properties under thermal stress, is extremely important for materials and devices, greatly affecting long-term stability and efficiency in applications\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5 CR6 CR7 CR8\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. When external conditions change, phase transitions occur caused by changes in the internal energy of nanomaterials, which is a common phenomenon widely exist nature and widely used by various technologies, such as water vaporization in the water cycle, structural transition of shape memory alloys, information storage, superconducting transition of superconductors\u003csup\u003e\u003cspan additionalcitationids=\"CR11 CR12 CR13 CR14\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. The critical temperature and/or pressure point at which a phase transition occurs is referred to as the phase transition point driven by the free energy which is crucial for phase transition occurrence, typically serving as a crucial parameter to estimate phase thermostability. The phase thermostability and transition point are not only influenced by external environmental factors (temperature and pressure) but also by the structural parameters of the crystal itself\u003csup\u003e\u003cspan additionalcitationids=\"CR17 CR18 CR19 CR20\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Therefore, extensive studies have been devoted to understanding the nature of superheating/supercooling, and to design the specific structures with the desired phase stability\u003csup\u003e\u003cspan additionalcitationids=\"CR23 CR24\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eRemarkable progress has been made in the study of phase transitions in bulk materials over the past few decades, and several methods have been successfully developed, such as embedded particles, alloy particles, self-assembled nanocrystals and so on\u003csup\u003e\u003cspan additionalcitationids=\"CR27 CR28 CR29 CR30\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Compared to the thermostability manipulation in bulk materials via interface engineering, much less is known about the contribution of surface effects to phase thermostability in nanomaterials, which are expected to play an increasingly dominant role with material size decreasing\u003csup\u003e\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. However, the current studies primarily focus on the surface atomic structure and properties itself, leaving a substantial gap in understanding whether and how to utilize surface effects to enhance thermostability\u003csup\u003e\u003cspan additionalcitationids=\"CR33 CR34\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. The complexity of nanocrystal surface structures, which consist of facets, terraces, edges, and corners, makes it difficult to understand their interactions with the external environment fully\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. On the other hand, the lack of in-situ structural evolution information at elevated temperatures\u0026mdash;encompassing both surface and bulk changes\u0026mdash;hampers establishing a clear relationship between thermal stability and surface structural evolution. Recent advancements in in-situ TEM have enabled real-time atomic-scale visualization of structural evolution, providing new insights into both phase transitions and surface reconstruction of nanomaterials\u003csup\u003e\u003cspan additionalcitationids=\"CR39 CR40 CR41\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. These exciting advancements provide us a great opportunity to explore the atomic nature of materials supercooling/superheating.\u003c/p\u003e \u003cp\u003eHerein, to clarify the surface effect on the phase transition and understand the nature of superheating of nanocrystals, we in situ explored the structural evolution of well-defined anatase TiO\u003csub\u003e2\u003c/sub\u003e nanocrystals at the atomic level via Cs-corrected scanning transmission electron microscopy (STEM), where micro-sized single crystalline anatase nanorods with various well-defined crystal surface facets were chosen as our research object. TiO\u003csub\u003e2\u003c/sub\u003e Anatase-to-Rutile transition, a classical phase transition system widely studied in the past decades, is widely believed to irreversibly occur at approximately 600\u0026deg;C\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. While in the practical studies, the anatase TiO\u003csub\u003e2\u003c/sub\u003e samples showed distinct thermostabilities in the separated experiments, and there is no consensus on the mechanism of the unconventional thermostabilities. In our prior research, the atomic structure and the reconstruction mechanisms of a variety of anatase TiO\u003csub\u003e2\u003c/sub\u003e surfaces, such as (001), (100), (101), and (301), etc., were systematically studied via in situ TEM\u003csup\u003e\u003cspan additionalcitationids=\"CR44 CR45 CR46\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. Based on the understanding of these surface properties, in this paper, we documented a highly anomalous phenomenon in this phase transition investigation that the anatase phase can still exist stably even when heated to 1250\u0026deg;C instead of turning into the rutile phase, which exceeded the reported phase transition point nearly 650\u0026deg;C higher. It is found that the surface of anatase TiO\u003csub\u003e2\u003c/sub\u003e forms a stable structure with a sequence of surface reconstructions taking place with the rise of temperature, resulting in a stable (101) crystal surface with lower surface energy. This low-energy surface plays a pivotal role in the phase thermal stability which significantly stabilized the metastable anatase phase via a \u0026ldquo;surface locking\u0026rdquo; mechanism, which kinetically inhibit surface rutile nucleation.\u003c/p\u003e \u003cp\u003eTo reveal the underlying mechanism of the unconventional stability of the TiO\u003csub\u003e2\u003c/sub\u003e nanocrystal, and to comprehensively estimate the influence of different surfaces on the phase transition, we chose the anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods as the sample and captured the in situ HRSEM images of the nanorods at different temperatures to show the morphology evolution of the anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods. The anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods were synthesized by a two-step hydrothermal method\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e (the details are shown in the experimental section), which is exposed to several different surfaces and has excellent crystallinity. Since there is a critical size in the anatase-to-rutile phase transition (ranges from 14 nm to 45 nm), the micro-sized TiO\u003csub\u003e2\u003c/sub\u003e nanocrystals used in the experiments are far beyond this critical size (in this size, rutile is more energetically favored than anatase), which could exclude the \u0026lsquo;\u0026lsquo;size\u0026rsquo;\u0026rsquo; issue during the experiments. In the experiments, we selected the well-defined individual nanocrystals that were not in contact with other nanocrystals to minimize the other structural factors involved in the experiments.\u003c/p\u003e \u003cp\u003eThe in situ heating experiments were performed to dynamically monitor the surface structural evolution of individual anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods, from 20\u0026deg;C to 1000\u0026deg;C under vacuum (10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e Pa). The morphology of the selected anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod did not show a notable change from 20\u0026deg;C to 500\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea-b). When the temperature was beyond 600\u0026deg;C, the morphology of the nanorod also remained almost unchanged (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec-d), with length and width unchanged contrast 20\u0026deg;C (~\u0026thinsp;650 nm; ~65 nm). While the surface of the nanorod started becoming no longer atomically flat, and some small protrusions appeared on the side surface of the nanorod, which indicated the possible surface reconstruction occurred. Above 700\u0026deg;C, the morphology of the TiO\u003csub\u003e2\u003c/sub\u003e nanorod exhibited significant change, the length became significantly shorter and the width slightly wider. The small protrusions on the side surfaces gradually evolved into larger protrusions and the tip of the nanorod became flatter (as shown in the enlarged images in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eh-m). The observation of another individual nanorods showed a similar process up to 1000\u0026deg;C and 1200\u0026deg;C (Supplementary Figs.\u0026nbsp;1 and 2, Supplementary Video 1).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo observe the atomic structural evolution of the anatase TiO\u003csub\u003e2\u003c/sub\u003e nanocrystals at elevated temperatures, we performed the in-situ heating experiments in a spherical aberration (Cs-) corrected scanning transmission electron microscope. The in-situ experiments were performed in vacuum and gas conditions via the Cs-corrected scanning transmission electron microscopes (Titan G2 80\u0026ndash;200, Hitachi HF-5000), with micro-electromechanical system based heating holders (DENS solutions DH30 double-tilt heating holder, Hitachi single-tilt heating holder). The phase structure of anatase TiO\u003csub\u003e2\u003c/sub\u003e could transformed into rutile around ~\u0026thinsp;600\u0026deg;C \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Since the TiO\u003csub\u003e2\u003c/sub\u003e crystal is very sensitive to electron beam (e-beam) irradiation, several measurements have been adopted to minimize the e-beam effect in the experiments. During the in-situ experiment, a very tiny dose (\u0026sim;0.6 nA) was used, and the electron beam was turned on only when the photos were taken to reduce the influence of the electron beam on the sample\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea shows the schematic diagram of a typical as-synthesized anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod. The size of the nanorod is about 650 nm long and 65 nm wide. The axial direction of the nanorod is along the [001] direction, and the sidewall is mainly exposed by (100) surfaces with atomic flatness (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). The tip of the nanorod is curved, which exposed by some high index surfaces, such as (301) facet with an orderly stepped structure containing succession of a (101) terrace separated by (001) step, (101) facet which is exposed by both 6-fold and 5-fold coordinated Ti atoms (Ti\u003csub\u003e6c\u003c/sub\u003e and Ti\u003csub\u003e5c\u003c/sub\u003e), with 2- and 3-fold oxygen atoms (O\u003csub\u003e2c\u003c/sub\u003e and O\u003csub\u003e3c\u003c/sub\u003e) located on the ridge of the sawtooth-like structure, (001) facet exposed by 5-fold coordinated Ti (Ti\u003csub\u003e5c\u003c/sub\u003e) atoms, 3-fold coordinated O atoms (O\u003csub\u003e3c\u003c/sub\u003e), and 2-fold coordinated O (O\u003csub\u003e2c\u003c/sub\u003e) atoms, etc. (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec-e), as well documented in our previous research \u003csup\u003e\u003cspan additionalcitationids=\"CR44 CR45\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. The two-dimensional crystalline lattice 0.35 nm (101) and 0.37 nm (100) indicate the anatase phase. The prepared nanorods had good crystallinity, atomically flat surfaces, and few defects, which provided a good platform for in-situ study.\u003c/p\u003e \u003cp\u003eTo illustrate the phase transition process at the atomic level well, we choose the single nanorod in the heating experiment as the object, viewed along [010] zone axis. As reported by the previous literatures, the phase transition points were concentrated around 600\u0026deg;C, investigated by both ex-situ and in-situ routes\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. While unexpectedly, we did not observe the phase transition above 650\u0026deg;C in our in-situ experiments. It is found that anatase phase TiO\u003csub\u003e2\u003c/sub\u003e nanorods could still maintain anatase phase even at a high temperature of 1250\u0026deg;C. Compared to the as-synthesized anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods, both the morphology and surface atomic structure show significant reconstruction at 1250\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). The originally flat surface of the selected nanorod became uneven, and the initial atomically flat (100), (301) surfaces are reconstructed into protrusions formed by (101) facets with (001) surface transformed into the (1\u0026times;4) reconstruction of ADM model (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef-i), which is consistent with the previous studies. While we carefully checked the HAADF STEM images in different areas, we did not find the nucleation of the rutile phase, even when the temperature exceeded the phase transition point of nearly 650\u0026deg;C. Considering the size of the nanorod is significantly larger than the critical size, the phase transition is expected to occur easily. Therefore, this observed unexpected ultra-stable anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod suggests an unconventional overheating phenomenon occurred in an individual single crystal, which was repeated in several cases (Supplementary Figs.\u0026nbsp;1, 2 and 3).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSince the morphology and surface structure show significant change during the heating experiments, the in situ STEM images were acquired to illustrate the detailed atomic evolution process of surface structures. Since the surface of the nanorod is mainly exposed by (100), (301), (001), and (101) surfaces, the atomic structural evolution of each surface is decoupled. Initially, the (100) surface showed a bulk-truncated surface structure (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). During the heating process, this atomically flat surface did not show visible change from room temperature to ~\u0026thinsp;600\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea-d). When the temperature was beyond 600\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee-f), some fresh fuzzy islands began to appear on the surface, indicating that some atomic migration occurred. At 900\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eg), the originally atomically flat (100) surface became obviously rough, with some fresh atomic columns appearing on the surface (white arrows) and some surface atomic columns disappearing (red arrows). With further temperature elevation, these small protrusions grew higher and became sharpened (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eh), with (101) surface dominated. Finally, the (100) surface was completely reconstructed into well-arranged sharp (101) faceted protrusions (mean height: 1.2 nm; interval distance along [001]: 6.4 nm), as the temperature reached 1250\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ei-j).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eCompared with the reconstruction of the (100) surface, directly evolving into (101) surface protrusion, the structural evolution process of the (301) surface was more complex (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea-j), which experienced two stages. Initially, the (301) surface showed an orderly stepped structure, which could be obtained by a succession of a (101) terrace with three titanium atom pairs, separated by a (001) step. In the first stage, this stepped (301) surface remained nearly unchanged from room temperature to 400\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea-b). At higher temperatures, the atoms on the surface of the nanorods gradually became unsharp and blurry. As indicated by the white arrow in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec, some atomic columns became blurred or even disappeared at 600\u0026deg;C, indicating that surface atomic migration occurred, as shown by the atoms circled in the green dotted box. The surface reconstruction became more obvious at 700\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed). As a result, at higher temperatures (750\u0026deg;C, 800\u0026deg;C), the (301) surface gradually transformed from a stepped (101) terrace (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb) into a different stepped structure (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee-f) consisting of (100) terraces and (101) facets, in the first stage.\u003c/p\u003e \u003cp\u003eIn the following heating process (the second stage), the newly formed (100) terraces also became unstable and tended to reconstruct into (101) surface-dominated protrusions as depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Therefore, at higher temperatures, with the formation of (101) faceted protrusions, the proportion of (100) surface significantly reduced (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eg), and eventually disappeared (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eh). The protrusions also grew larger and became more sharpened, eventually (301) surface was also completely transformed into a stepped structure dominated by the (101) surfaces at 1250\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ei-j). Compared to the reconstructed (100) surface, the formed (101) protrusions by the reconstruction of the (301) surface have a preferential orientation proportion: (101) \u0026gt; (10\u0026thinsp;\u0026minus;\u0026thinsp;1).\u003c/p\u003e \u003cp\u003eIn addition, the (001) surfaces of the anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod tip were kept bulk-truncated structure from room temperature to 600\u0026deg;C, and then transformed into the (1\u0026times;4) reconstruction by an ADM model, which could be stabilized at even 1250\u0026deg;C (Supplementary Fig.\u0026nbsp;4), as discussed in previous work\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. At last, the structural evolution of (101) surface was also explored. While different from the significant reconstructions of other surfaces, the (101) remain stable and did not show significant reconstruction within 1250\u0026deg;C, which confirms that the surface of (101) is the thermodynamically stable surface of anatase TiO\u003csub\u003e2\u003c/sub\u003e (Supplementary Fig.\u0026nbsp;5). Overall, except the (101) surface, all the other crystal surfaces were reconstructed: (100) and (301) were reconstructed into stepped (101) surface; (001) was reconstructed via the ADM model. The final result is that the nanorods are reconstructed to form a stable surface envelope consisting of stepped (101) and (001) surfaces.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo further reveal why and how these reconstructions happen during the reshaping of anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod, time-resolved STEM imaging was employed to track real-time mass transport dynamics. Initially, two distinct steps existed on the surface, as framed in dotted green lines in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea. Over time, the left step diminished, and even nearly vanishing at t\u0026thinsp;=\u0026thinsp;9.5 s shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea. Whereas, the right step sharpened through atomic accumulation and migrated\u0026thinsp;~\u0026thinsp;2.0 nm rightward (green dotted lines denote diffusion-mediated atomic additions). The diffusion process features synchronized atom depletion in Area 1 and the accumulation of atoms in area 2 and area 3, as visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb-d (white and colored dotted lines mark pre- and post-diffusion atomic configurations, respectively). The detailed information was shown in the kinetics curve of the atomic diffusion in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ef. On the one hand, the reduction of atoms in area 1 was the fastest and nonlinear, where the rapid migration of atoms along the atomic layer in the period of t\u0026thinsp;=\u0026thinsp;2.5\u0026thinsp;~\u0026thinsp;4.5 s led to the rapid reduction of atoms layer by layer in area 1 shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb. On the other hand, the accumulation of atoms in area 2 and area 3 were different. In area 2, the increase of atoms was slow at first. After t\u0026thinsp;=\u0026thinsp;4.5 s, atoms began to accumulate rapidly, indicating that nucleation was more difficult, but once nucleated, the step grew faster and gradually sharpened as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ec. At t\u0026thinsp;=\u0026thinsp;9.5 s, the atoms rapidly accumulate 3 layers along the original atomic layer leading to the step sharper, with the lattice plane on both sides composed of (101) facets. In contrast, there was no significant accumulation of atoms in area 3 shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ed, just the length of the layers added along the existing atomic layer ladder changed repeatedly over time. In addition, the number of atoms decreased in area 1 was much greater than the number of atoms increased in area 2, indicating that the excess atoms migrated rapidly layer by layer to the right along the atomic steps of area 3.\u003c/p\u003e \u003cp\u003eThroughout the process, the rapid diffusion of surface atoms suggests that, under our experimental conditions, atoms could migrate freely. The diffusion process ultimately leads to atom accumulation at steps, forming sharp protrusions with low-surface-energy (101) facets, resulting in the morphological evolution depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The mass transportation direction is consistent with the structural evolution showed in Supplementary Video 3, where atoms migrate from the nanorod tip toward its midsection. Now the question comes up as to the nature of the ultrastability of the individual TiO\u003csub\u003e2\u003c/sub\u003e single crystal nanorod, which could exist at an extremely high temperature of ~\u0026thinsp;1250\u0026deg;C, approximately\u0026thinsp;~\u0026thinsp;650\u0026deg;C higher than its phase transition point without phase transition\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eNotably, considering that the critical size between the anatase phase and rutile phase is in a range of ~\u0026thinsp;14\u0026ndash;45 nm\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e, the size of our sample far exceeds this range, it is anticipated the transition should be more favorable for the anatase to rutile phase transition. Thus, this unexpected overheating of the individual anatase single crystal nanorod represents an anomalous phenomenon, significantly contradicting the conventional understanding of TiO\u003csub\u003e2\u003c/sub\u003e stability. Interestingly, in contrast, in our additional experiments, a completely different phenomenon was found that the phase transition readily occurred when many TiO\u003csub\u003e2\u003c/sub\u003e nanorods stacked together. The nucleation of rutile crystallite formed at the multi-grain boundaries of adjacent particles, as shown in Supplementary Fig.\u0026nbsp;6. While, within a certain range, only a single rutile crystal nucleates and grows, and polycrystalline rutile formation is not observed, which also implies that kinetic nucleation is an exceedingly difficult process, requiring the stacking of anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods in a specific manner. Therefore, it could be concluded that the absence of phase transformation under high-temperature conditions is not due to a lack of source materials or limitations in mass transport, but rather is kinetically constrained by the failure to form rutile nuclei on the low-energy anatase TiO\u003csub\u003e2\u003c/sub\u003e surfaces.\u003c/p\u003e \u003cp\u003eThe phase transition from anatase to rutile has been studied extensively over the past 30 years, both theoretically and experimentally\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan additionalcitationids=\"CR52 CR53 CR54 CR55 CR56\" citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e. According to previous studies, rutile nucleation is the key step in the anatase-to-rutile phase transition, occurring via a (112) twin boundary in anatase\u003csup\u003e\u003cspan additionalcitationids=\"CR59\" citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e. In individual nanorod experiments, the formation kinetic of this nucleation site is extremely challenging. The (100) and (301) surfaces on the individual TiO\u003csub\u003e2\u003c/sub\u003e nanorod preferentially undergo reconstruction into the (101) surface rather than the (112) surface due to the higher surface energy (0.04 eV/\u0026Aring;\u003csup\u003e2\u003c/sup\u003e) under high-temperature treatment, as described in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea-c. Surface energy calculations reveal that the (101) surface has the lowest energy (0.008 eV/\u0026Aring;\u003csup\u003e2\u003c/sup\u003e), as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec-d, making it the most stable surface. DFT\u0026thinsp;+\u0026thinsp;U calculations further confirm that \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{surface}\\)\u003c/span\u003e\u003c/span\u003e(301) \u0026gt; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{surface}\\)\u003c/span\u003e\u003c/span\u003e (100) \u0026gt; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{surface}\\)\u003c/span\u003e\u003c/span\u003e (101), indicating a favorable pathway for this reconstruction (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ed). After the reconstruction, the individual anatase TiO\u003csub\u003e2\u003c/sub\u003e nanocrystal is mainly exposed by the (101) surfaces. This surface with very low surface energy, significantly prevents the surface rutile nucleation for the elevated barrier. Additionally, the rapid diffusion of surface atoms avoids the atom accumulation on surface which further reduce nucleation probability. These combined effects significantly increase the nucleation barrier during the phase transition, resulting in the abnormal thermal stability of the anatase phase. We refer to this phenomenon as surface locking. This finding highlights the critical impact of surface on the stability of nanomaterials, which significantly affect the secondary phase nucleation and mass transport in the phase transformation of nanomaterials. The ultra-high thermal stability of anatase TiO\u003csub\u003e2\u003c/sub\u003e nanocrystals achieved through \u0026ldquo;surface locking\u0026rdquo; reveals the unique surface dynamics of nanoparticles and their profound impact on phase stability and transformation kinetics.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eIn conclusion, through an in situ atomic-scale exploration of the structural evolution of well-defined anatase TiO\u003csub\u003e2\u003c/sub\u003e nanocrystals, we revealed the abnormal ultra-high thermal stability of the individual anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod single crystal, which could maintain the anatase phase structure at 1250\u0026deg;C. Unexpectedly, in the heating process, different from the expected anatase-to-rutile phase transition, only the surface reconstructions and reshaping of the individual anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod occurred, with gradually reconstructed into the stable structures mainly exposed by (101) surfaces with deceased surface energy. The restructuring not only reduce the total energy of the system, but also kinetically inhibited the surface rutile nucleation, leading to the abnormal phase stability of the anatase TiO\u003csub\u003e2\u003c/sub\u003e at an ultra-high temperature. Our results reveal atomic-scale insights into the critical impact of surface atomic structure and restructuring on the phase stability of nanomaterials, and the proposed \u0026ldquo;surface locking\u0026rdquo; may serve as a fundamental principle for the phase stability manipulation of future functional nano crystalline materials design operating at higher temperatures.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e1. Synthesis of anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods:\u003c/h2\u003e \u003cp\u003eThe anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods were synthesized via a facile hydrothermal method that was reported previously\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. In a typical synthesis, 1 g Na-titanates (wet, isolated by centrifugation without drying) were dispersed into 40 mL deionized water solution, then transferred to a 50 mL Teflon-lined stainless-steel autoclave, and sealed. The autoclave was put into an oven, heated at 200\u0026deg;C for 24 h, and cooled naturally in air, producing white tetragonal faceted-nanorods (TFNR) precipitates. These white precipitates were isolated from the solution by centrifugation and subsequently washed with deionized water several times, and finally dried at 60\u0026deg;C in vacuum for 10 h.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e2. In situ STEM observations:\u003c/h3\u003e\n\u003cp\u003eThe in situ STEM experiments were conducted in an FEI Titan G\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e 80\u0026ndash;200 scanning transmission electron microscope (200 kV), with a spherical aberration (C\u003csub\u003es\u003c/sub\u003e-) corrector and a high spatial resolution (~\u0026thinsp;0.8 \u0026Aring;). Before the test, to ensure high-quality imaging, all low-order aberrations (up to order 5) have been calibrated to an acceptable level, such as C\u003csub\u003es\u003c/sub\u003e \u0026lt; 0.5 \u0026micro;m, A1 (2-fold astigmatism)\u0026thinsp;\u0026lt;\u0026thinsp;2 nm, A2 (3-fold astigmatism)\u0026thinsp;\u0026lt;\u0026thinsp;20 nm, and B2 (coma)\u0026thinsp;\u0026lt;\u0026thinsp;20 nm. The convergence angle employed for STEM imaging was ~\u0026thinsp;21 mrad. The annular detection angle for high-angle annular dark field (HAADF) image is set to ~\u0026thinsp;53\u0026ndash;200 mrad, while for bright-field (BF) image, it is adjusted to ~\u0026thinsp;0\u0026ndash;20 mrad.\u003c/p\u003e \u003cp\u003eDuring the in situ STEM experiments, the as-obtained TiO\u003csub\u003e2\u003c/sub\u003e nanorods are dispersed in ethanol and dropped on the heating chips, which are then transferred into the TEM chamber via a Wildfire D6 double tilt heating holder (DENS solutions). Instead of directly heating our sample to the elevated temperatures, in the step-by-step heating process, the temperature interval was set to 100\u0026deg;C with a heating rate of ~\u0026thinsp;5\u0026deg;C s\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTiO\u003csub\u003e2\u003c/sub\u003e is usually very sensitive to electron beam (e-beam) irradiation and it is easy to undergo irradiation decomposition under electron beam irradiation\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e,\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. In the STEM mode, the beam radiation damage of the electron beam will be more severe because the electron beam is convergent and has a higher beam current density. In order to minimize the radiation damage of the electron beam to the maximum extent, we take a series of measures to reduce the radiation dose of the electron beam in the sample (\u0026sim;0.6 nA). In particular, the target surface area was moved under the electron beam only during the imaging process.\u003c/p\u003e\n\u003ch3\u003e3. The DFT calculations:\u003c/h3\u003e\n\u003cp\u003eThe spin-polarized DFT calculations were performed using the Perdew-Burke-Ernzerh (PBE) functional to optimize all surface structures, implemented in the Vienna \u003cem\u003eab initio\u003c/em\u003e simulation package (VASP). Projector-augmented wave (PAW) potentials were used to represent core-valence interactions. To account for the localized Ti \u003cem\u003ed\u003c/em\u003e states, we applied DFT\u0026thinsp;+\u0026thinsp;U with an effective U value of 4.1 eV. During optimization, atomic positions were relaxed until residual forces were below 0.02 eV/\u0026Aring;, and the convergence criterion for electronic self-consistent energy was set to 10\u003csup\u003e\u0026minus;\u0026thinsp;4\u003c/sup\u003e eV. A plane-wave cutoff of 400 eV was used to expand the Kohn-Sham valence states, and a vacuum layer of 10 \u0026Aring; was included to prevent interactions between periodic images. Surface energies of perfect anatase (001), (100), (101), (102), (103), (112), and (301) surfaces were calculated using (1\u0026times;1) slab models with corresponding K-point grids up to (10\u0026times;10\u0026times;1), (10\u0026times;4\u0026times;1), (6\u0026times;10\u0026times;1), (3\u0026times;10\u0026times;1), (5\u0026times;10\u0026times;1), (8\u0026times;4\u0026times;1) and (2\u0026times;9\u0026times;1), respectively.\u003c/p\u003e \u003cp\u003eThe surface energy \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left({E}_{surface}\\right)\\)\u003c/span\u003e\u003c/span\u003e is given by:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{E}_{surface}=\\frac{{E}_{slab}-{nE}_{bulk}}{2{A}_{slab}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{slab}\\)\u003c/span\u003e\u003c/span\u003e is the total energy of the slab, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{bulk}\\)\u003c/span\u003e\u003c/span\u003e is the energy per atom in the bulk structure, n is the number of atoms in the slab, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{A}_{slab}\\)\u003c/span\u003e\u003c/span\u003e is the surface area of the slab. The factor of 2 accounts for the two surfaces present in the slab model.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data needed to evaluate the conclusions are available in the main text or the supplementary information.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe acknowledge the financial support of the National Key Research and Development Program (2023YFA1506904, 2022YFA1505500), the National Natural Science Foundation of China (52025011, 52422311, 22302173, 52171019, and 52473242)), the Zhejiang Provincial Natural Science Foundation of China (LR23B030004), the Fundamental Research Funds for the Central Universities. The authors thank Dr. Zhemin Wu and Dr Ruiyang You in Center of Electron Microscopy at Zhejiang University for STEM data analysis, and thank Jiahang Zhang for his\u0026nbsp;contribution to the graphical illustration of the mechanistic model.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eY.W. initiated the project. Y.W. and W.Y. supervised this project. Y.W., and W.Y. designed the experiments. X.G. and W.Y. conducted the in situ STEM experiments. X.G. and Y.O. contributed to the sample preparation. Z.H., Y.Z., Y.L., C.Y., contributed to the theoretical calculations and explanation. B.Z., Y.J., H.Y., and Z.Z. participated in the analysis and discussion. X.G., and Y.Z. contributed equally to this work.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eYang, Y. et al\u003cem\u003e.\u003c/em\u003e Amidination of ligands for chemical and field-effect passivation stabilizes perovskite solar cells. \u003cem\u003eScience. \u003cstrong\u003e386,\u003c/strong\u003e\u003c/em\u003e 898-902 (2024).\u003c/li\u003e\n\u003cli\u003ePark, Y. M. et al\u003cem\u003e.\u003c/em\u003e Engineering ligand reactivity enables high-temperature operation of stable perovskite solar cells. \u003cem\u003eScience. \u003cstrong\u003e381,\u003c/strong\u003e\u003c/em\u003e 209-215 (2023).\u003c/li\u003e\n\u003cli\u003eZhou, X., Li, X. Y. \u0026amp; Lu, K. Enhanced thermal stability of nanograined metals below a critical grain size. \u003cem\u003eScience. \u003cstrong\u003e360,\u003c/strong\u003e\u003c/em\u003e 526-530 (2018).\u003c/li\u003e\n\u003cli\u003eBu, T. L. et al\u003cem\u003e.\u003c/em\u003e Lead halide\u0026ndash;templated crystallization of methylamine-free perovskite for efficient photovoltaic modules. \u003cem\u003eScience. \u003cstrong\u003e372,\u003c/strong\u003e\u003c/em\u003e 1327-1332 (2021).\u003c/li\u003e\n\u003cli\u003eLe, V. N. et al\u003cem\u003e.\u003c/em\u003e Improved organic electrochemical transistor stability using solvent degassing and chemical doping. \u003cem\u003eNat. Electron \u003c/em\u003e(2025): https://doi.org/10.1038/s41928-024-01297-8.\u003c/li\u003e\n\u003cli\u003eHan, G. G. D., Li, H. S. \u0026amp; Grossman, J. C. Optically-controlled long-term storage and release of thermal energy in phase-change materials. \u003cem\u003eNat. Commun. \u003cstrong\u003e8,\u003c/strong\u003e\u003c/em\u003e 1446 (2017).\u003c/li\u003e\n\u003cli\u003eChen, J., Wang, Z.-K. \u0026amp; Liao, L.-S. Defect passivation and crystallization control of perovskite films for photovoltaic application. \u003cem\u003eMater. Today. Nano.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 100118 (2021).\u003c/li\u003e\n\u003cli\u003eJia, Y. et al. Recent advances in energy storage and applications of form-stable phase change materials with recyclable skeleton. \u003cem\u003eCarbon. Neutralization.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 313\u0026ndash;343 (2024).\u003c/li\u003e\n\u003cli\u003eTang, Z. et al. Tightened 1D/3D carbon heterostructure infiltrating phase change materials for solar\u0026ndash;thermoelectric energy harvesting: Faster and better. \u003cem\u003eCarbon. Energy.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, e281 (2023).\u003c/li\u003e\n\u003cli\u003eHan, G. G. D., Li, H.S. \u0026amp; Grossman, J. C. Optically-controlled long-term storage and release of thermal energy in phase-change materials. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1446 (2017).\u003c/li\u003e\n\u003cli\u003eZhang, Y. N. et al\u003cem\u003e.\u003c/em\u003e High-temperature superconductivity with zero resistance and strange-metal behaviour in La\u003csub\u003e3\u003c/sub\u003eNi\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e7\u0026minus;\u0026delta;\u003c/sub\u003e. \u003cem\u003eNat. Phys.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 1269\u0026ndash;1273 (2024).\u003c/li\u003e\n\u003cli\u003eRao, F. et al\u003cem\u003e.\u003c/em\u003e Reducing the stochasticity of crystal nucleation to enable subnanosecond memory writing. \u003cem\u003eScience.\u003c/em\u003e \u003cstrong\u003e358\u003c/strong\u003e, 1423\u0026ndash;1427 (2017).\u003c/li\u003e\n\u003cli\u003eDing, K. Y. et al\u003cem\u003e.\u003c/em\u003e Phase-change heterostructure enables ultralow noise and drift for memory operation. \u003cem\u003eScience.\u003c/em\u003e \u003cstrong\u003e366\u003c/strong\u003e, 210\u0026ndash;215 (2019).\u003c/li\u003e\n\u003cli\u003eZewdie, G. M. et al. Chemical Design Principles for Cache-Type Sc\u0026ndash;Sb\u0026ndash;Te Phase-Change Memory Materials. \u003cem\u003eChem. Mater.\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 4008\u0026ndash;4015 (2019).\u003c/li\u003e\n\u003cli\u003eAtinafu, D. G. et al\u003cem\u003e.\u003c/em\u003e Unveiling sustainable nano-enabled phase change materials for high thermal stability and energy storage capacity. \u003cem\u003eJ. Energy. Storage.\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 106650 (2023).\u003c/li\u003e\n\u003cli\u003eChen, Y. et al. Phase engineering of nanomaterials. \u003cem\u003eNat. Rev. Chem.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 243\u0026ndash;256 (2020).\u003c/li\u003e\n\u003cli\u003eLi, W. B., Qian, X. F. \u0026amp; Li, J. Phase transitions in 2D materials. \u003cem\u003eNat. Rev. Chem.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 829\u0026ndash;846 (2021).\u003c/li\u003e\n\u003cli\u003eHuang, H. H. et al. Recent progress of TMD nanomaterials: phase transitions and applications. \u003cem\u003eNanoscale.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1247\u0026ndash;1268 (2020).\u003c/li\u003e\n\u003cli\u003eBarnard, A. S. \u0026amp; Curtiss, L. A. Prediction of TiO\u003csub\u003e2\u003c/sub\u003e Nanoparticle Phase and Shape Transitions Controlled by Surface Chemistry. \u003cem\u003eNano Lett.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1261\u0026ndash;1266 (2005). \u003c/li\u003e\n\u003cli\u003eZhang, H. Z. \u0026amp; Banfield, J. F. Structural Characteristics and Mechanical and Thermodynamic Properties of Nanocrystalline TiO\u003csub\u003e2\u003c/sub\u003e. \u003cem\u003eChem. Rev.\u003c/em\u003e \u003cstrong\u003e114\u003c/strong\u003e, 9613\u0026ndash;9644 (2014). \u003c/li\u003e\n\u003cli\u003eSang, L. X., Zhao, Y. X. \u0026amp; Burda, C. TiO\u003csub\u003e2\u003c/sub\u003e Nanoparticles as Functional Building Blocks. \u003cem\u003eChem. Rev.\u003c/em\u003e \u003cstrong\u003e114\u003c/strong\u003e, 9283\u0026ndash;9318 (2014). \u003c/li\u003e\n\u003cli\u003eCao, C. Z. et al. Bulk ultrafine grained/nanocrystalline metals via slow cooling. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, eaaw2398 (2019).\u003c/li\u003e\n\u003cli\u003eLi, X. Y., Zhou, X. \u0026amp; Lu, K. Rapid heating induced ultrahigh stability of nanograined copper. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, eaaz8003 (2020). \u003c/li\u003e\n\u003cli\u003eXiao, X. M. et al. Superheating of grain boundaries within bulk colloidal crystals. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 1599 (2022).\u003c/li\u003e\n\u003cli\u003eLosurdo, M. et al. Thermally stable coexistence of liquid and solid phases in gallium nanoparticles. \u003cem\u003eNat. Mater.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 995\u0026ndash;1002 (2016).\u003c/li\u003e\n\u003cli\u003eHanaor, D. A. H. \u0026amp; Sorrell, C. C. Review of the anatase to rutile phase transformation. \u003cem\u003eJ. Mater. Sci. \u003c/em\u003e\u003cstrong\u003e46\u003c/strong\u003e, 855\u0026ndash;874 (2011).\u003c/li\u003e\n\u003cli\u003eLu, K. \u0026amp; Li, Y. Homogeneous Nucleation Catastrophe as a Kinetic Stability Limit for Superheated Crystal. \u003cem\u003ePhys. Rev. Lett.\u003c/em\u003e \u003cstrong\u003e80\u003c/strong\u003e, 4474\u0026ndash;4477 (1998).\u003c/li\u003e\n\u003cli\u003eMei, Q. S. \u0026amp; Lu, K. Melting and superheating of crystalline solids: From bulk to nanocrystals. \u003cem\u003eProg. Mater. Sci.\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 1175\u0026ndash;1262 (2007).\u003c/li\u003e\n\u003cli\u003eJin, Z. H., Gumbsch, P., Lu, K. \u0026amp; Ma, E. Melting Mechanisms at the Limit of Superheating. \u003cem\u003ePhys. Rev. Lett.\u003c/em\u003e \u003cstrong\u003e87\u003c/strong\u003e, 055703 (2001).\u003c/li\u003e\n\u003cli\u003eShannon, R. D. \u0026amp; Pask, J. A. Kinetics of the Anatase-Rutile Transformation. \u003cem\u003eJ. Am. Ceram. Soc.\u003c/em\u003e \u003cstrong\u003e48\u003c/strong\u003e, 391\u0026ndash;398 (1965).\u003c/li\u003e\n\u003cli\u003eZhang, L. et al\u003cem\u003e.\u003c/em\u003e Superheating of Confined Pb Thin Films. \u003cem\u003ePhys. Rev. Lett.\u003c/em\u003e \u003cstrong\u003e85\u003c/strong\u003e, 1484-1487(2000).\u003c/li\u003e\n\u003cli\u003eVittadini, A., Casarin, M. \u0026amp; Selloni, A. Chemistry of and on TiO\u003csub\u003e2\u003c/sub\u003e-anatase surfaces by DFT calculations: a partial review. \u003cem\u003eTheor. Chem. Acc.\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 663\u0026ndash;671 (2007).\u003c/li\u003e\n\u003cli\u003eLu, W. G. et al\u003cem\u003e.\u003c/em\u003e Phase transition of individual anatase TiO\u003csub\u003e2\u003c/sub\u003e microcrystals with large percentage of (001) facets: a Raman mapping and SEM study. \u003cem\u003ePhys. Chem. Chem. Phys.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 3199\u0026ndash;3210 (2023).\u003c/li\u003e\n\u003cli\u003eSheng, H.C., Xiao, B. B. \u0026amp; Jiang, X. B. Size dependence of the phase transition temperature of metal nanocrystals. \u003cem\u003ePhysica B.\u003c/em\u003e \u003cstrong\u003e667\u003c/strong\u003e, 415193 (2023).\u003c/li\u003e\n\u003cli\u003eZhang, H. Z. \u0026amp; Banfield, J. F. Phase transformation of nanocrystalline anatase-to-rutile via combined interface and surface nucleation. \u003cem\u003eJ. Mater. Res.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 437\u0026ndash;448 (2000).\u003c/li\u003e\n\u003cli\u003eGao, W. P., Hood, Z. D. \u0026amp; Chi, M. F. Interfaces in Heterogeneous Catalysts: Advancing Mechanistic Understanding through Atomic-Scale Measurements. \u003cem\u003eAcc. Chem. Res.\u003c/em\u003e \u003cstrong\u003e50\u003c/strong\u003e, 787\u0026ndash;795 (2017).\u003c/li\u003e\n\u003cli\u003eYang, X. F. et al\u003cem\u003e.\u003c/em\u003e Single-Atom Catalysts: A New Frontier in Heterogeneous Catalysis. \u003cem\u003eAcc. Chem. Res.\u003c/em\u003e \u003cstrong\u003e46\u003c/strong\u003e, 1740\u0026ndash;1748 (2013).\u003c/li\u003e\n\u003cli\u003eYu, R. et al. Direct Subangstrom Measurement of Surfaces of Oxide Particles. \u003cem\u003ePhys. Rev. Lett.\u003c/em\u003e \u003cstrong\u003e105\u003c/strong\u003e, 226101 (2010).\u003c/li\u003e\n\u003cli\u003eHe, M.-R., Yu, R. \u0026amp; Zhu, J. Reversible Wurtzite\u0026ndash;Tetragonal Reconstruction in ZnO(101̅0) Surfaces. \u003cem\u003eAngew. Chem. Int. Ed.\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, 7744\u0026ndash;7747 (2012).\u003c/li\u003e\n\u003cli\u003eHe, M.-R., Yu, R. \u0026amp; Zhu, J. Subangstrom Profile Imaging of Relaxed ZnO(101̅0) Surfaces. \u003cem\u003eNano Lett.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 704\u0026ndash;708 (2012).\u003c/li\u003e\n\u003cli\u003eYuan, W. T. et al\u003cem\u003e.\u003c/em\u003e In situ manipulation of the active Au-TiO\u003csub\u003e2\u003c/sub\u003e interface with atomic precision during CO oxidation. \u003cem\u003eScience.\u003c/em\u003e \u003cstrong\u003e371\u003c/strong\u003e, 517\u0026ndash;521 (2021).\u003c/li\u003e\n\u003cli\u003eYuan, W. T. et al\u003cem\u003e.\u003c/em\u003e Visualizing H\u003csub\u003e2\u003c/sub\u003eO molecules reacting at TiO\u003csub\u003e2\u003c/sub\u003e active sites with transmission electron microscopy. \u003cem\u003eScience. \u003cstrong\u003e367,\u003c/strong\u003e\u003c/em\u003e 428-430 (2020).\u003c/li\u003e\n\u003cli\u003eYuan, W. T. et al\u003cem\u003e.\u003c/em\u003e In Situ STEM Determination of the Atomic Structure and Reconstruction Mechanism of the TiO\u003csub\u003e2\u003c/sub\u003e (001) (1 \u0026times; 4) Surface. \u003cem\u003eChem. Mater.\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 3189\u0026ndash;3194 (2017).\u003c/li\u003e\n\u003cli\u003eYuan, W. T. \u003cem\u003eet al.\u003c/em\u003e Real-Time Observation of Reconstruction Dynamics on TiO\u003csub\u003e2\u003c/sub\u003e(001) Surface under Oxygen via an Environmental Transmission Electron Microscope. \u003cem\u003eNano Lett.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 132\u0026ndash;137 (2016).\u003c/li\u003e\n\u003cli\u003eYuan, W. T. et al\u003cem\u003e.\u003c/em\u003e Unveiling the Atomic Structures of the Minority Surfaces of TiO\u003csub\u003e2\u003c/sub\u003e Nanocrystals. \u003cem\u003eChem. Mater.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 288\u0026ndash;295 (2018).\u003c/li\u003e\n\u003cli\u003eFang, K. et al\u003cem\u003e.\u003c/em\u003e An Environmental Transmission Electron Microscopy Study of the Stability of the TiO\u003csub\u003e2\u003c/sub\u003e (1 \u0026times; 4) Reconstructed (001) Surface. \u003cem\u003eJ. Phys. Chem. C.\u003c/em\u003e \u003cstrong\u003e123\u003c/strong\u003e, 21522\u0026ndash;21527 (2019).\u003c/li\u003e\n\u003cli\u003eYuan, W. T. et al\u003cem\u003e.\u003c/em\u003e Direct in-situ insights into the asymmetric surface reconstruction of rutile TiO\u003csub\u003e2\u003c/sub\u003e (110). \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1616 (2024).\u003c/li\u003e\n\u003cli\u003eLi, J. M. \u0026amp; Xu, D. S. Tetragonal faceted-nanorods of anatase TiO\u003csub\u003e2\u003c/sub\u003e single crystals with a large percentage of active {100} facets. \u003cem\u003eChem. Commun.\u003c/em\u003e \u003cstrong\u003e46\u003c/strong\u003e, 2301\u0026ndash;2303 (2010).\u003c/li\u003e\n\u003cli\u003eGribb, A. A. \u0026amp; Banfield, J. F. Particle size effects on transformation kinetics and phase stability in nanocrystalline TiO\u003csub\u003e2\u003c/sub\u003e. \u003cem\u003eAm. Mineral.\u003c/em\u003e \u003cstrong\u003e82\u003c/strong\u003e, 717\u0026ndash;728 (1997).\u003c/li\u003e\n\u003cli\u003eZhang, H. Z. \u0026amp; Banfield, J. F. Thermodynamic analysis of phase stability of nanocrystalline titania. \u003cem\u003eJ. Mater. Chem.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 2073\u0026ndash;2076 (1998).\u003c/li\u003e\n\u003cli\u003eOkeke, G., Hammond, R. B. \u0026amp; Antony, S. J. Influence of size and temperature on the phase stability and thermophysical properties of anatase TiO\u003csub\u003e2\u003c/sub\u003e nanoparticles: molecular dynamics simulation. \u003cem\u003eJ. Nanopart. Res.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1584 (2013).\u003c/li\u003e\n\u003cli\u003eSatoh, N., Nakashima, T. \u0026amp; Yamamoto, K. Metastability of anatase: size dependent and irreversible anatase-rutile phase transition in atomic-level precise titania. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 1959 (2013).\u003c/li\u003e\n\u003cli\u003ePatra, S. et al\u003cem\u003e.\u003c/em\u003e Phase stability frustration on ultra-nanosized anatase TiO\u003csub\u003e2\u003c/sub\u003e. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 10928 (2015).\u003c/li\u003e\n\u003cli\u003eZhang, J. et al\u003cem\u003e.\u003c/em\u003e Importance of the Relationship between Surface Phases and Photocatalytic Activity of TiO\u003csub\u003e2\u003c/sub\u003e. \u003cem\u003eAngew. Chem. Int. Ed.\u003c/em\u003e \u003cstrong\u003e47\u003c/strong\u003e, 1766\u0026ndash;1769 (2008).\u003c/li\u003e\n\u003cli\u003eMao, Q. et al\u003cem\u003e.\u003c/em\u003e Sintering-Induced Phase Transformation of Nanoparticles: A Molecular Dynamics Study. \u003cem\u003eJ. Phys. Chem. C.\u003c/em\u003e \u003cstrong\u003e119\u003c/strong\u003e, 28631\u0026ndash;28639 (2015).\u003c/li\u003e\n\u003cli\u003eAlimohammadi, M. \u0026amp; Fichthorn, K. A. Molecular Dynamics Simulation of the Aggregation of Titanium Dioxide Nanocrystals: Preferential Alignment. \u003cem\u003eNano Lett.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 4198\u0026ndash;4203 (2009).\u003c/li\u003e\n\u003cli\u003eKoparde, V. N. \u0026amp; Cummings, P. T. Phase Transformations during Sintering of Titania Nanoparticles. \u003cem\u003eACS Nano.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 1620\u0026ndash;1624 (2008).\u003c/li\u003e\n\u003cli\u003eZhu, S.-C., Xie, S.-H. \u0026amp; Liu, Z.-P. Nature of Rutile Nuclei in Anatase-to-Rutile Phase Transition. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e137\u003c/strong\u003e, 11532\u0026ndash;11539 (2015).\u003c/li\u003e\n\u003cli\u003ePenn, R. L. \u0026amp; Banfield, J. F. Morphology development and crystal growth in nanocrystalline aggregates under hydrothermal conditions: insights from titania. \u003cem\u003eGeochim. Cosmochim. Acta.\u003c/em\u003e \u003cstrong\u003e63\u003c/strong\u003e, 1549\u0026ndash;1557 (1999).\u003c/li\u003e\n\u003cli\u003ePenn, R. L. \u0026amp; Banfield, J. F. Formation of rutile nuclei at anatase (112) twin interfaces and the phase transformation mechanism in nanocrystalline titania. \u003cem\u003eAm. Mineral.\u003c/em\u003e \u003cstrong\u003e84\u003c/strong\u003e, 5\u0026ndash;6 (1999).\u003c/li\u003e\n\u003cli\u003eChen, Q. L. \u003cem\u003eet al.\u003c/em\u003e Imaging Beam-Sensitive Materials by Electron Microscopy. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 1907619 (2020).\u003c/li\u003e\n\u003cli\u003eJiang, N. Electron beam damage in oxides: a review. \u003cem\u003eRep. Prog. Phys.\u003c/em\u003e\u003cstrong\u003e79\u003c/strong\u003e, 016501 (2015).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"In situ transmission electron microscopy, anatase TiO2, thermal stability, surface structure evolution, surface reconstruction","lastPublishedDoi":"10.21203/rs.3.rs-6475787/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6475787/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe phase stability of nanocrystals is of great importance to its performance in different applications, and extensive studies have been devoted to understanding the nature of superheating/supercooling, to achieve the desired long-term phase stability in practice. While the fundamental mechanism remains elusive due to the absence of atomic-level insights into dynamic structural evolution of nanocrystals under extreme conditions. Herein, through in situ atomic level spherical aberration-corrected scanning transmission electron microscopy we revealed an abnormal phase stability of the individual single crystalline anatase TiO\u003csub\u003e2\u003c/sub\u003e at ultra-high temperatures governed by surface effects. The in situ atomic level observations at the exceptionally high temperatures show a highly anomalous phenomenon that the individual anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorod single crystal could maintain the anatase phase at 1250\u0026deg;C, instead of turning into the rutile phase, which exceeded the reported anatase-to-rutile phase transition point nearly 650\u0026deg;C higher. As the temperature rises above the phase transition point, the surfaces of anatase TiO\u003csub\u003e2\u003c/sub\u003e nanorods undergo a series of atomic reconstructions, which not only reduce the total energy of the system, but also act as a kinetic \u0026ldquo;surface locking\u0026rdquo; effect preventing the rutile nucleation. This work shows that surfaces could have a critical effect on the phase stability of single crystalline nanomaterials, and the proposed \u0026ldquo;surface locking\u0026rdquo; mechanism opens up a new way for manipulating the thermal stability of nanocrystals.\u003c/p\u003e","manuscriptTitle":"Unexpected Phase Stability of Anatase TiO2 Nanocrystals at Ultrahigh Temperature via Surface Restructuring","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-06 11:31:29","doi":"10.21203/rs.3.rs-6475787/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"7b1f4b8b-da7d-4dac-a8c9-82eca06a468b","owner":[],"postedDate":"May 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":48136269,"name":"Physical sciences/Materials science/Nanoscale materials/Nanoparticles"},{"id":48136270,"name":"Physical sciences/Nanoscience and technology/Nanoscale materials/Nanoparticles"}],"tags":[],"updatedAt":"2026-05-07T10:02:21+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-06 11:31:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6475787","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6475787","identity":"rs-6475787","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}