Engineering and developing cost-effective, wavelength-selective photodetectors (PDs) with high responsivity and long-term stability remains a critical challenge in optoelectronics. In this work, we presented a scalable strategy for engineering porous silicon (PS) PDs functionalized with femtosecond-laser-ablated silver nanoparticles (Ag-NPs) using a spray-time-controlled deposition approach. Ag-NPs exhibited strong colloidal stability with a negative surface charge of − 35 mV by zeta potential, and deposited onto PS using optimized spray durations of 250 s (PS- Ag1-NPs) and 350 s (PS- Ag2-NPs). FESEM analysis revealed that Ag incorporation induced a clear size-dependent evolution (13.97 nm for PS- Ag1-NPs and 19.83 nm for PS- Ag2-NPs) and increased PS layer thickness from 2.27 to 2.65 µm and 2.78 µm, respectively. XRD and FTIR confirmed SiO2 and AgO formation, with intensities rising with Ag content. Graphite electrodes provided stable ohmic contacts, with thermal annealing at 130 °C for 30 min further improving adhesion and stability. Voltage-dependent I-V characteristics were recorded from 0–5.2 V for all devices, followed by wavelength-resolved photocurrent measurements across the UV-Red spectrum at 2.8, 4.0, and 5.2 V. Ag functionalization substantially reconfigured the spectral response: Under 5.2 V bias, PS- Ag1-NPs PD exhibited a peak response in the green region (530 nm, 1.7 mW/cm2), achieved a sensitivity of 310.07
We examine the equatorial capture shadow of a phenomenological charged Konoplya–Zhidenko-type metric with Δ =r^2+a^2-2Mr+Q^2-ϵ /r . The charge term is introduced in Kerr–Newman form; no field-equation derivation is asserted. Using unstable spherical photon orbits that satisfy angular admissibility and the full radial-capture condition, we calculate contacts of projected orbit images. For the theoretical benchmark (a/M,ϵ /M^3)=(2,0.5) , the cusp-removal contact reaches the celestial axis at Q_off/M≃ 1.3520 . A sampling of spin and deformation values shows how this contact charge varies within the ansatz. In the moderate-spin family (a/M,ϵ /M^3)=(0.9,0.5) , the shadow is smooth over 0≤ Q/M≤ 0.6 , develops a cusp near Q/M=0.9518 , and loses it near Q/M=1.3313 . The cusp morphology therefore depends on the chosen parameter family; the values are theoretical controls, not observational constraints.
The present study explores the cosmological dynamics of an LRS Bianchi Type-I universe in the context of f(R) gravity, focusing on vacuum solutions. Two anisotropic models are constructed by considering different functional forms of f(R), where plane-symmetric LRSBI metrics incorporate anisotropy in the expansion rates, maintaining directional differences among Hubble parameters and scale factors throughout cosmic evolution. While such anisotropic assumptions in general relativity typically lead to a decelerating universe, functional forms of f(R) can support both accelerating and decelerating expansion branches, depending on the corresponding model parameters. To constrain the model parameters, we perform a comprehensive statistical analysis using Markov chain Monte Carlo (MCMC) methods with Hubble, BAO, and Pantheon datasets. The resulting observational constraints are interpreted at the level of the effective mean-background expansion. Model I and II constraints are illustrated through posterior contour and error bar plot, and the corresponding statistical values are summarized in tables.
This study presents machine-learning-guided screening of 40 catalyst formulations for the hydrogenation of CO2 to propane, spanning platinum-group metals, chromium oxides, transition metals, and bimetallic combinations across 450–625 °C and 0.1–0.5 MPa, as a greenhouse-gas mitigation strategy yielding a versatile fuel for gas cutting and liquefied petroleum gas (LPG) blending. CO2 conversion ranged 8.1–19.5
Stochastic resonance (SR) is a powerful nonlinear technique for amplifying weak signals via noise, widely utilized in early fault detection. However, traditional SR systems often suffer from output saturation effects, which severely limit amplification capability and degrade detection performance. To address this issue, this paper proposes a composite Adversarial Unsaturated Tri-stable Stochastic Resonance (AUTSR) model. By introducing a bounded correction derived from the Woods–Saxon profile in the outer region, AUTSR mitigates output saturation over a finite operating range while retaining polynomial confinement at large displacements. Furthermore, an adaptive fault detection scheme is established by integrating the genetic algorithm (GA) for optimal parameter tuning. Simulations and experiments on bearing fault signals show that AUTSR achieves higher output signal-to-noise ratios (SNRs) and larger peak-to-interference differences than classical tri-stable stochastic resonance (TSR), linear piecewise tri-stable stochastic resonance (LPTSR), and quadratic piecewise tri-stable stochastic resonance (QPTSR) in the investigated cases. These improvements facilitate characteristic fault frequency detection in noisy environments.
