Wire-grid polarizers (WGPs) require precise control of nanoscale metallic structures to achieve high polarization selectivity while maintaining optical transmission. Conventional fabrication commonly relies on blanket metal deposition followed by lithographic patterning and etching, increasing process complexity and placing stringent requirements on dimensional control. In this study, an additive fabrication approach based on nanoimprint-guided electroforming is developed for direct formation of nickel (Ni) WGPs on indium tin oxide (ITO)-coated glass. Nanoimprint lithography (NIL) defines periodic resist trenches, and selective Ni electroforming from the exposed ITO at the trench bottoms directly converts the resist template into a functional metallic wire grid without blanket metal deposition or subsequent metal etching. Ni WGPs with pitches of 320, 210, and 140 nm were fabricated to investigate the relationships among electroforming conditions, cross-sectional geometry, and polarization performance. Increasing the electroforming duration promotes Ni growth but eventually produces lateral broadening, resulting in coupled changes in wire height, width, metal filling fraction, and optical aperture. The corresponding optical measurements show a clear balance between transmittance and extinction ratio (ER). Finite-difference time-domain (FDTD) simulations further show that cross-sectional geometry strongly influences the optical response, with a trapezoidal profile providing higher calculated transmittance and ER than a broadened-top profile. Argon plasma post-treatment modifies the upper wire geometry and increases the interwire aperture, although prolonged treatment also reduces the Ni wire height and affects the exposed ITO surface. The P140 structures exhibit the highest polarization selectivity, reaching approximately 65% transmittance and an ER of approximately 1300 near 850 nm. The combined experimental and numerical results establish grating pitch, electroforming-controlled wire dimensions, and post-treatment-controlled cross-sectional geometry as interconnected parameters governing the optical performance of electroformed Ni WGPs. Nanoimprint-guided electroforming therefore provides an additive approach for direct fabrication and geometrical control of nanoscale metallic wire grids.
In a five-axis machining center, the worktable-bottom crossbeam bolted joint interface is a primary load-bearing and force-transmission path, and degradation of its contact stiffness can deteriorate machining-region dynamics and thereby reduce machining accuracy. This study quantifies how distributed stiffness degradation at this multi-bolt worktable-bottom crossbeam interface affects the dynamic relative motion between the tool and worktable. The interface is represented by orthotropic virtual-material layers, and a dimensionless scale factor r is applied to their normal and tangential moduli. A healthy-state finite element baseline is checked against the measured first natural frequency; modal scanning, mode tracking based on modal assurance criterion (MAC), and harmonic response analysis are then used to identify degradation-sensitive frequency bands. Within the 28-34 Hz band, the simulated mean tool-worktable relative-displacement frequency response function (FRF) peak is higher at r = 0.6 than at r = 1.0. Whole-machine impact tests compare nominal-torque and 60%-nominal-torque assembly conditions using H1-estimated acceleration FRFs, magnitude-squared coherence, and acceleration-to-displacement conversion. Within 28-34 Hz, the reduced-torque assembly yields peak ratios of 1.20-1.22 and band response ratios of 1.47-1.48 for the spindle response and spindle-worktable relative responses, reproducing the simulation-predicted band and direction of response change. The proposed numerical-experimental procedure provides condition-sensitive, precision-relevant frequency-domain features for assessing the dynamic condition of critical multi-bolt interfaces in five-axis machining centers.
To reveal the stiffness evolution mechanism of rotation worktable under complex loading boundaries and enhance load-bearing robustness, a spatial six-dimensional rigid-flexible coupling semi-analytical static stiffness modeling method for the cradle-type dual-axis rotary table (CDRT) is proposed, and a reliability-based design optimization (RBDO) is conducted. First, Timoshenko beam theory is employed to discretely characterize the structural flexibility of the pendulum frame and A-axis shell. Meanwhile, von Kármán large deflection theory and Hertz contact theory are utilized to characterize the nonlinear contact boundaries of the brake mechanism and support bearings, respectively. Subsequently, the accuracy of the proposed model is verified through static loading tests and finite element analysis. Based on the established model, the C-axis lifting effect under a brake-clamped state is revealed, and the influence of key parameters on system stiffness characteristics is investigated. Furthermore, considering the random uncertainties of structural parameters in practical engineering, an RBDO framework is constructed, achieving a collaborative improvement in the static load safety factor of the support bearings and the overall reliability level of the machine. The semi-analytical model and RBDO framework constructed in this paper provide a theoretical basis for the performance evaluation and forward design of CDRT and similar multi-axis equipment.
Atomic force microscopy depends on probe-tip geometry, yet commercial nominal specifications often diverge from the actual shape of an individual tip. We present an automated pipeline for quantifying the bluntness exponent d — the slope of the power-law tip profile z=Brd, where d=1 is a cone and d=2 a paraboloid — of AFM probe tips from SEM images, using Bayesian linear regression (BLR), total least squares, and SAM2-assisted segmentation. The pipeline was validated against published measurements for two silicon probes and extended to three diamond-coated probe images spanning sharp and blunt conditions. Phase 1 reproduced all reference values within stated tolerances. Phase 2 classified the sharp diamond-coated probe (d≈1.50, R≈5.6 nm) and blunt probe (d≈2.85) into distinct condit ion groups; Phase 3 SAM2 segmentation revised the blunt-probe exponent to d=3.388 (R≈114.5 nm), the primary reported value. Phase 3 added analytical 95% credible intervals, confirmed the two diamond-coated probes are statistically distinct, and showed that the 594kX image yields a more precisely constrained exponent (d=1.498 vs d=1.856) than the 300kX image of the same probe. A depth-dependence analysis demonstrated that d is not constant across the measurable tip profile for the probes examined: the near-contact zone exponent (8–30nm) differs significantly from the full-profile value (8–85nm) — a finding inaccessible from ordinary least squares and directly relevant to nanoindentation and nanomachining. The workflow provides a reproducible route from SEM image to metrologically traceable tip-shape inference.
For wafer thinning processes utilizing the self-rotating grinding method, research on ground surface quality has primarily focused on improving wafer surface shape accuracy and surface roughness, while neglecting the formation of surface waviness profiles. As medium-to-low frequency errors in wafer surface, excessive waviness profile characteristics can also significantly increase subsequent polishing and adversely affect processing efficiency. To suppress the waviness error on the ground wafer surface, this study investigates the generation mechanism of the ground surface shape in wafer self-rotation grinding, considering both grinding wheel runout and grinding speed ratio. First, a control method for the wafer surface shape is proposed based on an independent adjustment strategy for the single-spindle inclination angle, and a corresponding surface shape generation model is established. Furthermore, by comprehensively considering grinding kinematics, grinding wheel contact characteristics, and material removal properties, the mapping relationship from micro-level runout errors to nano-level waviness errors was elucidated, revealing the generation mechanism of surface waviness profiles caused by grinding wheel axial runout. Subsequently, a simulation model of the ground surface waviness profile is established by combining the wafer surface shape generation model with the waviness profile mapping mechanism. This model reveals the distribution trends of waviness profile characteristics and was experimentally validated. Finally, a waviness profile suppression strategy based on the mechanism of profile phase-misalignment superposition is proposed. Simulation and experimental results demonstrate that adopting a non-integer rotational speed ratio effectively suppresses waviness profile characteristics on the ground surface, significantly improving ground surface flatness.
