${\mathrm{U}}_3{\mathrm{Si}}_2$ is a promising accident-tolerant nuclear fuel, yet its elastic properties remain poorly understood because of significant discrepancies between experimental measurements, including bulk moduli ranging from $\sim 68\mathrm{GPa}$ (RUS) to $\sim 107\mathrm{GPa}$ (XRD), the origin of which has remained unresolved. In this work, we investigate the structural, thermodynamic, and elastic properties of ${\mathrm{U}}_3{\mathrm{Si}}_2$ from 0 to 1500 K using density functional theory in the $\mathrm{DFT}+U$ approach combined with machine-learning-assisted canonical sampling and interatomic potentials. The use of the PBEsol functional and a $U$ value of 2.0 eV enables us to simultaneously suppress the spurious structural distortion of previous calculations and reproduce the experimental equilibrium volume, which is a prerequisite for a reliable determination of the full tetragonal elastic tensor. We further demonstrate that explicit finite-temperature sampling is essential to capture the full thermal softening of the elastic constants, as static calculations at thermally expanded volumes underestimate this effect by up to 30% for certain elastic constants. Finally, we demonstrate that the apparent conflict between high pressure XRD and RUS bulk moduli arises from the pronounced elastic anisotropy of the tetragonal lattice combined with the isotropic approximations used in RUS data analysis and the effects of sample porosity. These findings provide a consistent set of temperature-dependent elastic properties for ${\mathrm{U}}_3{\mathrm{Si}}_2$ and resolve long-standing discrepancies between experimental measurements.

Antiperovskites share the same structure as perovskites, but allow completely different chemistries and nominal charge states of anions to be stabilized. This gives rise to many interesting phenomena, including septet superconductivity and topological crystalline insulating phases in these systems. Despite this, the work on the crystal structural trends in these compounds is more limited compared to perovskites. In this study, we consider the family of antiperovskite oxides with tetrel elements (Si, Ge, Sn, Pb) and alkaline earth metals (Ca, Sr, Ba), and perform a detailed study of their crystal structures using first-principles density functional theory. We show how tolerance factor arguments can be constructed to predict their structure in a way parallel to the perovskites, and furthermore, how heterostructuring (or cation-order) can be used to induce ferroelectricity in these systems which may provide an experimental knob to modify electronic structure. We conclude with a discussion of the electronic structure of antiperovskites, and show that they display interesting trends not observed in regular perovskites, including significant antibonding interactions between face-center ions, which might need to be taken into account when building effective electronic models of these compounds.

We report the evolution of the antiferromagnetic domain structure of epitaxial ${\mathrm{La}}_{0.45}{\mathrm{Sr}}_{0.55}{\mathrm{MnO}}_3$ (LSMO) ultrathin films with thickness in the range from 5–50 unit cells (uc), using x-ray photoemission electron microscopy (XPEEM). While the 5 uc thick LSMO shows no magnetic contrast down to $\sim 80\mathrm{K}$, thicker films display a multidomain antiferromagnetic configuration with a nonmonotonic variation in the characteristic domain size from 0.3 $\mu \mathrm{m}$ at 10 uc, to 3–5 $\mu \mathrm{m}$ at 15–16 uc and 1 $\mu \mathrm{m}$ for 50 uc. Post-growth annealing is found not to impact strongly the magnetic domain state. In contrast, for a 15 uc film grown on a substrate characterized by large atomic steps $(\sim 1\mu \mathrm{m})$, we observe much larger antiferromagnetic domains and the presence of net magnetic moment in the form of stripes with alternating contrast, assigned to the signal from the top uncompensated spins of the A-type antiferromagnetic state of LSMO. From the combined antiferromagnetic domain structure and the net magnetic moment contrast, we determine the exact orientation of the Néel vector, including at domain walls. We describe the antiferromagnetic domain size distribution in terms of interface and bulk contributions to the density of defects that pin the antiferromagnetic domain walls and which determine the equilibrium domain configuration. Our results demonstrate the impact of thickness and defects on the antiferromagnetic domain size and constitute a stepping stone in controlling the antiferromagnetic domain state required for oxide antiferromagnetic device applications.

Photoelectron excitations in oxide heterostructures routinely elicit both surface photovoltage and open-circuit voltage. Surface photovoltage measurements are hence often used as a proxy to gauge open-circuit voltage response. Here, we present a method for studying photovoltaic effects in oxide heterostructures by concurrently measuring, with an atomic force microscope, surface photovoltage $\left(\mathrm{\Delta }{V}_{\mathrm{SPV}}\right)$ and open-circuit voltage $\left(V_{\mathrm{OC}}\right)$ while varying illumination wavelength and intensity. This combined method is applied at super- and subband-gap illumination wavelengths on ${\mathrm{Bi}}_2{\mathrm{FeCrO}}_6/\mathrm{Nb}\text{−}{\mathrm{SrTiO}}_3$ heterostructures annealed under various oxidative and reductive conditions. The results show that $\mathrm{\Delta }{V}_{\mathrm{SPV}}$ consistently reflects interfacial photocarrier generation, whereas $V_{\mathrm{OC}}$ is more governed by photocarrier transport, sensitive to recombination. It is also shown that deep oxygen-vacancy levels act as recombination centers, whereas shallow cation-vacancy levels impact depletion region width. Experimental observations are consistent with DFT density of states of crystals containing isolated and coupled vacancies. Together, $\mathrm{\Delta }{V}_{\mathrm{SPV}}$ and $V_{\mathrm{OC}}$ spectroscopies—applied across multiple excitation energies and intensities—offer valuable insights into mechanisms governing photon-induced electrical power generation in photovoltaic oxide heterostructures.

Using first-principles calculations and ab initio molecular dynamics, we investigate the formation of atomic vacancies in the monolayer ${\mathrm{AsBiSe}}_3$ and its consecutive effect on electronic and mechanical properties. The calculated formation energies reveal that Se vacancies constitute the most stable defect. The energy barrier for the migration of these vacancies between neighboring atomic sites is 1.50 eV, surpassing the barrier energies for the migration of As (1.19 eV) and Bi (0.76 eV) vacancies, confirming their stability. Regarding mechanical properties, we have systematically investigated the mechanical response of all defect configurations considered in the ${\mathrm{AsBiSe}}_3$ monolayer. The presence of most-stable Se atomic vacancies does not destroy the semiconducting character of ${\mathrm{AsBiSe}}_3$ but it gives rise to midgap states that could impact the optical and transport characteristics of ${\mathrm{AsBiSe}}_3$, suggesting avenues for additional research.

Understanding gas adsorption on defect-engineered two-dimensional (2D) materials is essential for advancing applications in energy storage, catalysis, and sensing. In this work, first-principles calculations are employed to systematically investigate the adsorption behaviors of ${\mathrm{H}}_2,{\mathrm{O}}_2,{\mathrm{H}}_2\mathrm{O},\mathrm{CO}$, and ${\mathrm{CO}}_2$ at sulfur vacancy $V_{\text{S}}$ sites in both single-component transition metal dichalcogenides ${M\mathrm{S}}_2$ (M = W, Mo, Ta, Nb, and V) and high-entropy ${\left(\mathrm{WMoTaNbV}\right)}_{\mathrm{S}2}$ monolayers. In ${M\mathrm{S}}_2$ systems, $V_{\text{S}}$ introduces localized defect states that enhance surface reactivity, leading to diverse adsorption modes including physisorption, chemisorption, and dissociative adsorption. Notably, ${\mathrm{H}}_2$ dissociation is kinetically favorable on $V_{\text{S}}/{\mathrm{TaS}}_2$ and $V_{\text{S}}/{\mathrm{NbS}}_2$, while ${\mathrm{CO}}_2$ remains weakly physiosorbed across all single-metal systems. In contrast, high-entropy ${\left(\mathrm{WMoTaNbV}\right)}_{\mathrm{S}2}$ exhibits a continuous distribution of adsorption energies due to its diverse local coordination environments and synergistic multimetal interactions. This results in significantly reduced activation barriers for ${\mathrm{H}}_2$ dissociation, enhanced adsorption strength for ${\mathrm{O}}_2,{\mathrm{H}}_2\mathrm{O}$, and CO, and a transition of ${\mathrm{CO}}_2$ adsorption from physisorption to chemisorption. Electronic structure analyses reveal that these improvements originate from multicenter $d\text{−}p$ orbital hybridization and enhanced charge transfer at vacancy sites. Overall, the high-entropy design effectively tailors the adsorption landscape of 2D transition metal dichalcogenides, offering a promising strategy for developing advanced materials for hydrogen storage, ${\mathrm{CO}}_2$ activation, and gas-sensing applications.

Yttrium iron garnet (YIG, ${\mathrm{Y}}_3{\mathrm{Fe}}_5{\mathrm{O}}_{12}$) thin films are promising material systems for quantum magnonics due to their ultralow magnetic damping and the capability of building on-chip spin wave devices to be coupled with superconducting microwave circuits. However, the default substrate for epitaxial YIG thin-film growth, ${\mathrm{Gd}}_3{\mathrm{Ga}}_5{\mathrm{O}}_{12}$ (GGG), suffers from strong microwave absorption at cryogenic temperature due to the magnetic moment induced by the rare-earth element Gd. In this work, we study strained epitaxial YIG thin films grown on ${\mathrm{Y}}_3{\mathrm{Sc}}_2{\mathrm{Ga}}_3{\mathrm{O}}_{12}$ (YSGG), a rare-earth-free substrate compatible with cryogenic applications. Using cross-sectional scanning transmission electron microscopy along with strain mapping, we show the existence of spot-shape domain defects in the epitaxial YIG films and their connection with strain and stoichiometry, leading to drastically different magnetic properties such as perpendicular magnetic anisotropy, magnetic hysteresis loop, and magnetic damping. Applying high-resolution orientation mapping, we identify localized lattice rotations as the structural origins of these domains. We hypothesize that these domains are formed by a self-composed off-stoichiometry during the growth of the YIG thin films, as a mechanism to accommodate the relatively large lattice mismatch between YIG and YSGG. By elucidating the structure-property relationship in YIG/YSGG thin films, this work highlights a strain accommodation mechanism in complex oxides and provides guidance for future works on the applications of strained YIG thin films in quantum information.

Density functional/molecular dynamics (DF/MD) simulations on ${\mathrm{GeS}}_2$ with four compositions of silver $(2.5\%\text{–}20\%)$ provide valuable information on the structures and energetics of the amorphous material. Ag fills the available empty volume (cavities, voids) and interacts mainly with sulfur. The number of Ag-Ag contacts increases with increasing Ag composition, leading mainly to branched chains that partially disrupt the covalent Ge-S network. We use DF/MD simulations to study the energetics of Ag dissolution, the structures of $\mathrm{Ag}\text{−}{\mathrm{GeS}}_2$ alloys for differing Ag content, migration paths in the energy surfaces at 0 K, and liquid dynamics at 1100 K, close to the experimental melting point. We discuss the electronic structures of the optimized glass structures. DF/MD simulations can describe the dynamics of these systems at liquid temperatures, but are too short to allow the study of Ag migration at lower temperatures. We have used DF/MD trajectories and generated snapshots of high-$T$ liquids to develop a neural network potential (NNP) that allows simulations of these systems for tens of nanoseconds. These simulations are used for migration dynamics at $500–800\mathrm{K}$, leading to a linear behavior of the mean-square displacement on a logarithmic scale; the effective migration energy of Ag ($0.33-0.43\mathrm{eV}$, with little composition dependence) agrees well with the static DF migration barriers.

Planar defects, widely observed in diamond and boron nitride, have been shown to markedly enhance hardness and toughness, yet the mechanical role of inversion domain boundaries (IDBs) in metastable wurtzite boron nitride (w-BN) remains largely unexplored. In this work, we present a comprehensive investigation of the mechanical response of IDBs, combining first-principles cleavage and stacking fault energetics with the Peierls-Nabarro framework of dislocation resistance. To resolve the mechanical response specific to the IDB-affected region, localized energetics are introduced in terms of area-normalized energy changes evaluated for cleavage or sliding events confined to the IDB region. The IDB-affected region acts as a weakening origin: both cleavage and slip resistances, for deformation either across the IDB or along the IDB, are reduced relative to bulk w-BN. Consequently, the IDB vicinity becomes the preferred site for crack initiation and dislocation activation. Electronic-structure analyses further reveal suppressed deformation-induced charge transfer across the IDB, which reduces the energetic cost of bond distortion, and less negative ICOHP values indicate weakened covalent bonding at the IDB, thereby facilitating local failure. Together, these results provide a microscopic picture of how dense IDBs could contribute to the reduced strength and apparent hardness of w-BN, given that IDBs are ubiquitous in synthesized w-BN, even though an ideal, defect-free w-BN crystal may possess a high intrinsic hardness.

To address the issues of poor operational flexibility and ecological disturbance risks in traditional underwater robots, as well as the low manufacturing accuracy and low functional integration of current photo-responsive soft systems, in this study, we developed a poly(N-isopropylacrylamide)/multiwalled carbon nanotube (MWCNT) photothermal composite hydrogel and a biomimetic modular soft robot inspired by the giant river shrimp. We systematically clarified the effects of material composition and structural thickness on the expansion, mechanical, and photoresponsive properties of the hydrogel to screen out the optimal formulation. The high-precision manufacturing of the hydrogel structure was achieved through a custom digital micromirror device based on an ultraviolet three-dimensional printing system, and a specialized predehydration impregnation cycle was combined to ensure the uniform dispersion of MWCNTs in the complex hydrogel structure. The structure of the robot was analyzed through transient fluid-solid coupling velocity simulation and finite element method analysis. Experimental results demonstrated that the robot exhibited rapid response and multidirectional flexible movement capabilities. This bionic soft robot possesses multimode controllable movement, underwater cargo transportation and targeted placement capabilities, and has good cyclic stability. Compared with the previously reported similar light-driven hydrogel robots, its speed fluctuation during eight consecutive cycles is relatively small, and the average error of 10 repeated targeted placements is approximately 2 mm, which can meet the basic requirements for precise underwater operations at the microscale. In this work, we establish a comprehensive theoretical and technical framework for the development of underwater soft robot technology for applications in unstructured aquatic environments.

The design of high-entropy perovskite relaxor ferroelectrics has emerged as a promising strategy to enhance the performance of high-energy storage materials. However, the dynamics of dipole moments in the relaxor ground state remain unclear. To address this issue, we studied and compared the dielectric spectra of noncentrosymmetric, $P4mm$ $\left(\mathrm{B}{\mathrm{a}}_{1/4}\mathrm{P}{\mathrm{b}}_{1/4}\mathrm{B}{\mathrm{i}}_{1/4}{\mathrm{K}}_{1/4}\right)\mathrm{Ti}{\mathrm{O}}_3$, $\left(\mathrm{B}{\mathrm{a}}_{1/5}\mathrm{P}{\mathrm{b}}_{1/5}\mathrm{B}{\mathrm{i}}_{1/5}\mathrm{C}{\mathrm{a}}_{1/5}{\mathrm{K}}_{1/5}\right)\mathrm{Ti}{\mathrm{O}}_3$, and $\left(\mathrm{B}{\mathrm{a}}_{1/2}\mathrm{P}{\mathrm{b}}_{1/8}\mathrm{B}{\mathrm{i}}_{1/8}\mathrm{C}{\mathrm{a}}_{1/8}{\mathrm{K}}_{1/8}\right)\mathrm{Ti}{\mathrm{O}}_3$ and centrosymmetric, $Pm\overline{3}m$ $\left(\mathrm{B}{\mathrm{a}}_{1/5}\mathrm{P}{\mathrm{b}}_{1/5}\mathrm{S}{\mathrm{r}}_{1/5}\mathrm{R}{\mathrm{E}}_{1/5}{\mathrm{K}}_{1/5}\right)\mathrm{Ti}{\mathrm{O}}_3$ with $\mathrm{RE}=\mathrm{La}$, Nd, and Sm high-entropy compounds. Upon cooling, noncentrosymmetric compounds show a sharp ferroelectric transition ($T_C$) and an ergodic-to-paraelectric transition at the Burns temperature ($T_B$). In contrast, centrosymmetric compounds exhibit broader, diffuse dielectric peaks without a signal of long-range ferroelectric order. These features, arising from compositional disorder, are the fingerprint of relaxor ferroelectrics. Our analysis reveals that crystal symmetry correlates strongly with the dynamics of dipoles within polar nanoregions (PNRs). We studied PNRs dynamics using the Vogel-Fulcher (VF) model, fitting both the frequency $f$ of the permittivity peak maximum at the temperature $T_m$ and the characteristic relaxation time ${\tau }_c$ derived from the permittivity spectra (${\varepsilon }^{\prime }$ and ${\varepsilon }^{″}$), using a Debye-like function. For noncentrosymmetric compounds, the VF model was fitted to the frequency $f$ only in the paraelectric state, while the relaxation time ${\tau }_c$ does not show VF-type behavior. On the other hand, for centrosymmetric compounds, it was found that the relaxation time ${\tau }_c$ can be described with physically meaningful parameters, whereas fitting the frequency $f$ yielded fiscally unfeasible results using the VF model. These results define two distinct pathways in complex and highly disordered systems. On cooling, the symmetry breaking ($Pm\overline{3}m\text{−}P4mm$) is associated with a ferroelectric-ergodic-paraelectric transition, where PNRs freeze at $T_C$, characteristic of a relaxor ferroelectric, whereas the preservation of global symmetry ($Pm\overline{3}m$) correlates with an ergodic-nonergodic state at very low temperatures, where the PNRs become static below $T_{\mathrm{VF}}$, defining a canonical relaxor. The existence and nanoscale heterogeneity of PNRs were directly confirmed via piezoresponse force microscopy.

In this work, we report the discovery of a new crystal structure on the Ge-rich side of the Pr-Ge binary phase diagram. Using a high-temperature flux technique, we grew single crystals of ${\mathrm{Pr}}_9{\mathrm{Ge}}_{16}$, which adopt a previously unreported orthorhombic $Fdd2$ structure type featuring ordered Ge vacancies. We present the anisotropic magnetic properties and identify the crystallographic $b$ axis perpendicular to the crystal plane as the magnetic easy axis. Temperature-dependent resistivity measurements reveal metallic behavior with a distinct anomaly at $T_C$ = 14.3 K. Hall resistivity data indicate that electronlike carriers dominate, with a carrier concentration on the order of ${10}^{27}{\mathrm{m}}^{-3}$. The magnetic order is readily suppressed by a magnetic field of approximately 0.4 T applied along the easy $b$ axis.

Dislocation climb is one of the key processes controlling the mechanical response and microstructure evolution of materials at high temperatures. The mechanisms of this phenomenon are still unclear due to its complex nature, which involves interactions between point defects and dislocations. Here we employed large-scale classical molecular-dynamics (MD) simulations to shed light on vacancy-driven dislocation climb in fcc Ni. A Ni crystal with dissociated dislocations was simulated under high-temperature vacancy supersaturation for multiple nanoseconds. A particular advantage of this method is the ability to reveal the underlying mechanisms of dislocation climb without prior knowledge. The study included three different dislocation characters: pure edge, mixed ${30}^{\circ }$, and ${60}^{\circ }$. We found that jog-pair formation initiated with vacancy clustering at partial dislocations, followed by the alignment of vacancies into linear clusters along the $\langle 110\rangle$ direction. This stage was clarified by a detailed analysis of vacancy segregation energies at dissociated dislocations for vacancy clusters of different sizes and configurations. Subsequent stages of jog-pair formation involved local reduction of the stacking-fault region and further jog-pair growth due to vacancy accumulation. From the MD simulations, we traced the full dislocation climb process and compared it with classical descriptions of dislocation climb in fcc metals. Detailed analysis of the MD results allowed us to resolve and describe all stages of dislocation climb from the perspective of dislocation reactions. Based on the changes in dislocation positions, we estimated the climb velocity for the considered dislocations driven by the osmotic force.

Zeolites are microporous crystalline materials with diverse frameworks, widely used in industrial applications such as petroleum refining and molecular separation. Unlike most zeolites, EMT can be synthesized under mild conditions (at low temperatures and without the use of organic structure-directing agents), making it attractive for cost-effective and environmentally sustainable production. However, the specific synthesis conditions that selectively produce EMT rather than similar frameworks like FAU are not yet well established. In this work, we develop machine learning (ML) models to guide the discovery of synthesis conditions for EMT zeolites. Our dataset comprises 174 experimental synthesis attempts, recording reaction time, temperature, silica and alumina sources, Si/Al stoichiometric ratio, and other synthesis parameters. We apply both classical ML methods and pretrained foundation models to predict zeolite framework outcomes from these synthesis parameters. Feature importance analysis identifies critical parameters for EMT formation, validating known synthesis principles. Leveraging the ML models, we explore the synthesis space and identify six promising new conditions for EMT formation. Experimental validation confirms EMT crystallization in five cases, including two with Si/Al stoichiometric ratios outside the training dataset's range. Evaluation on independent literature-reported synthesis conditions further demonstrates the generalizability of the model. This work demonstrates a data-driven approach to accelerating zeolite synthesis, closing the loop between ML prediction and experimental validation.

Gallium arsenide (GaAs) doped with erbium (Er), a material of interest for optoelectronics and quantum information, has been studied for decades. Yet the formation of Er luminescence centers in the semiconductor host and their properties are still not well understood. Here we present a systematic investigation of Er-related defects in GaAs, including defect complexes consisting of Er and native point defects or oxygen impurities, using first-principles hybrid-functional defect calculations. We find that these defects have electronic structure and energetics that are generally asymmetric with respect to $n$- and $p$-type doping and tend to favor electron trapping. On the basis of the calculated defect levels, formation energies, and nonradiative carrier capture coefficients, we identify Er-related defects that are efficient as trap-assisted nonradiative recombination centers for ${\mathrm{Er}}^{3+}$ excitation under host photoexcitation or via minority carrier injection. Our results provide an understanding of why a particular defect center with Er coupled to two oxygen atoms is most efficient, and of the effects of $n$- and $p$-type doping and of the Er/O ratio on the formation of optically active Er centers and on the Er luminescence.

Antiperovskite ${\mathrm{Co}}_4\mathrm{N}$ with a large magnetic moment and high spin-polarization is a promising candidate for spintronic applications. However, its stabilization requires a careful attention due to the high formation enthalpy and N diffusion. This study demonstrates that Pd incorporation stabilizes antiperovskite ${\mathrm{Co}}_4\mathrm{N}$ thin films, reduces defects, and enables tuneable magnetization dynamic properties. Ferromagnetic resonance studies show that Pd doping in ${\mathrm{Co}}_4\mathrm{N}$ causes a systematic rise in Gilbert damping accompanied by a reduction in inhomogeneous broadening. Along with laboratory techniques, advanced synchrotron and neutron methodologies—hard x-ray photoelectron spectroscopy, extended x-ray absorption fine structure, and polarized neutron reflectivity have been utilized to elucidate the role of Pd doping on the structural, electronic, and magnetic properties of ${\mathrm{Co}}_4\mathrm{N}$. Thermal annealing studies revealed that Pd doping leads to enhanced thermal stability of ${\mathrm{Co}}_4\mathrm{N}$. Overall, this study presents an approach for achieving stabilization of antiperovskite ${\mathrm{Co}}_4\mathrm{N}$ by Pd doping along with enhanced structural and magnetic uniformity and tunable magnetization dynamic properties without a significant reduction of magnetic moment.

Temperature dependent Raman spectroscopy was employed on $\alpha \text{−}{\mathrm{In}}_2{\mathrm{O}}_3$ and $\alpha \text{−}{\mathrm{Al}}_2{\mathrm{O}}_3$ to determine the phonon decay factors associated with the decay of phonons into two and three phonons. Differently oriented corundum structured $\alpha \text{−}{\mathrm{In}}_2{\mathrm{O}}_3$ and $\alpha \text{−}{\mathrm{Al}}_2{\mathrm{O}}_3$ samples were analyzed by temperature dependent Raman spectroscopy between 80 and 300 K. The shift of the resonance frequency of the Raman active phonon modes $A_{1g}\left(1\right),$ $E_g\left(4\right), \text{and} E_g\left(5\right)$ was investigated by a model function including the thermal expansion of the lattice and the decay of the phonons into two and three phonons with lower energy. We obtain the phonon decay factors $A$ and $B$ representing the decay into two and three phonons, respectively. The findings are in reasonable agreement with the existing literature on $\alpha \text{−}{\mathrm{Al}}_2{\mathrm{O}}_3,$ $\alpha \text{−}{\mathrm{Ga}}_2{\mathrm{O}}_3,$ and $\alpha \text{−}{\mathrm{In}}_2{\mathrm{O}}_3$.

Epitaxial δ-NbN thin films (9–205 nm) were grown on MgO(001) and MgO(110) to examine how thickness-driven structural evolution controls normal-state transport and superconducting performance, and how these changes are linked to vacancy-mediated lattice evolution. Structural characterization shows that increasing thickness generally improves the microstructural coherence length. The room-temperature resistivity generally decreases with thickness when a single epitaxial registry is preserved. However, the 205-nm NbN(110) film, which exhibits a weak secondary (111) component, shows a higher resistivity despite its larger microstructural coherence lengths, indicating that resistivity is not governed by structural coherence alone. All measured films exhibit a negative temperature coefficient of resistivity, and the broad minimum in $\mathrm{d}{\rho }_{xx}/\mathrm{d}T$ appears for all samples around 50–80 K, which correlates more closely with the disorder parameter $k_F\mathcal{l}$ than with microstructural coherence lengths. Hall measurements indicate that the effective carrier density $n_{\mathrm{eff}}$ increases with thickness and correlates with the evolution of unit-cell volume. Together with density functional theory calculations showing that percent-level vacancy disorder can produce comparable changes in lattice volume, this supports a thickness-driven vacancy-linked lattice evolution accompanied by strain relaxation, coinciding with an increase of $T_c$ (8.5–12.3 K). The largest estimated upper critical field is 44.9 T for the 205-nm NbN(110) film, corresponding to a superconducting coherence length of 2.7 nm.