High Processing Efficiency Research Articles

Implementation of Grey Wolf, Multi-Verse and Ant Lion Metaheuristic Algorithms for Optimizing Machinability of Dry CNC Turning of Annealed and Hardened UNIMAX® Tool Steel

Advances in machining technology and materials science impose the identification of optimal settings for process-related parameters to maintain high quality and process efficiency. Given the available resources, manufacturers should determine an advantageous process parameter range for their settings. In this work, the machinability of a special tool steel (UNIMAX® by Uddeholm, Sweden) under dry CNC turning is investigated. The working material is examined under two states; annealed and hardened. As major machinability indicators, main cutting force Fz (N) and mean surface roughness Ra (μm) were selected and studied under different values for the cutting conditions of cutting speed, feed rate, and depth of cut. A systematic experimental design was established as per the response surface methodology (RSM). The experimental design involved twenty base runs with eight cube points, four center points in the cube, six axial points, and two center points in the axial direction. Corresponding statistical analysis was based on analysis of variance and normal probability plots for residuals. Two regression models referring to main cutting force and surface roughness for both the annealed and hardened states of the material were developed and used as objective functions for subsequent evaluations by three modern meta-heuristics under the goal of machinability optimization, namely multi-objective grey wolf algorithm, multi-objective multi-verse algorithm and multi-objective ant lion algorithm. All algorithms were found capable of providing beneficial Pareto-optimal solutions for both main cutting force and surface roughness simultaneously whilst regression models achieved high correlation among input variables and optimization responses.

A Bio-Inspired Probabilistic Neural Network Model for Noise-Resistant Collision Perception.

Bio-inspired models based on the lobula giant movement detector (LGMD) in the locust's visual brain have received extensive attention and application for collision perception in various scenarios. These models offer advantages such as low power consumption and high computational efficiency in visual processing. However, current LGMD-based computational models, typically organized as four-layered neural networks, often encounter challenges related to noisy signals, particularly in complex dynamic environments. Biological studies have unveiled the intrinsic stochastic nature of synaptic transmission, which can aid neural computation in mitigating noise. In alignment with these biological findings, this paper introduces a probabilistic LGMD (Prob-LGMD) model that incorporates a probability into the synaptic connections between multiple layers, thereby capturing the uncertainty in signal transmission, interaction, and integration among neurons. Comparative testing of the proposed Prob-LGMD model and two conventional LGMD models was conducted using a range of visual stimuli, including indoor structured scenes and complex outdoor scenes, all subject to artificial noise. Additionally, the model's performance was compared to standard engineering noise-filtering methods. The results clearly demonstrate that the proposed model outperforms all comparative methods, exhibiting a significant improvement in noise tolerance. This study showcases a straightforward yet effective approach to enhance collision perception in noisy environments.

Wide-Spectrum Antireflective Properties of Germanium by Femtosecond Laser Raster-Type In Situ Repetitive Direct Writing Technique

A femtosecond laser raster-type in situ repetitive direct writing technique was used for the fabrication of anti-reflective microhole structures in Germanium (Ge) in the visible near-infrared range (300–1800 nm). This technique builds a layer of microstructured arrays on the surface of Ge, enabling Ge to exhibit excellent anti-reflective properties. The large-area micro-nanostructures of Ge were fabricated using femtosecond laser raster-type in situ repetitive direct writing. Ge microstructures are characterized by their structural regularity, high processing efficiency, high reproducibility, and excellent anti-reflective properties. Experimental test results showed that the average reflectance of the Ge microporous structure surface in the range of 300–1800 nm was 2.25% (the average reflectance of flat Ge was 41.5%), and the lowest reflectance was ~1.6%. This microstructure fabrication drastically reduced the optical loss of Ge, thus enhancing the photothermal utilization of Ge. The many nanoburrs and voids in the Ge microporous structure provided excellent hydrophobicity, with a hydrophobicity angle of up to 133 ± 2° (the hydrophobicity angle of flat Ge was 70 ± 2°). The high hydrophobicity angle allows for strong and effective self-cleaning performance. The femtosecond laser raster-type in situ repeatable direct writing technology has many desirable properties, including simplicity, high accuracy, flexibility, and repeatability, that make it one of the preferred choices for advanced manufacturing. The Ge micro-nanostructured arrays with excellent optical anti-reflective properties and hydrophobicity have become an attractive alternative to the current photo-thermal absorbers. It is expected to be used in many applications such as solar panels, photovoltaic sensors, and other optoelectronic devices.

Pool boiling heat transfer performance of micro-embossing molds for the fabrication of polymer wicks

As flexible heat dissipation devices are urgently needed for flexible electronics, polymers with the advantages of good flexibility, insulation, and low density have attracted wide attention. It would be interesting to fabricate thermal functional structures on polymers to overcome their low thermal conductivity. This study aims to design and fabricate a micro-embossing mold to process polymer three-dimensional microgroove wicks (3D-MWs), which are applied in flexible heat pipes and achieve high thermal conductivity. Three kinds of tungsten steel molds were processed by using two wire electrical discharge machining processes (WEDM), two electrical discharge shaping machining processes (EDSM), and a combination of WEDM and EDSM. The 3D-MWs of the tungsten steel mold were reprinted onto the polymer films by a two-step embossing process. To investigate the effects of microstructures on boiling heat transfer performances, tungsten steel molds fabricated by different methods were tested to reflect the heat dissipation potential of the polymer 3D-MWs. The results showed that the 3D-MWs fabricated by the combination of WEDM and EDSM exhibit good boiling heat transfer performance and high processing efficiency, with a critical heat flux of 1040.1 kW/m2 and a maximum heat transfer coefficient of 38.4 kW/m2·°C. These values are 46% and 611% higher than those of the smooth tungsten steel surface, respectively, making it a superior choice for micro-embossing molds.

Empowering In-Network Classification in Programmable Switches by Binary Decision Tree and Knowledge Distillation

Given the high packet processing efficiency of programmable switches (e.g., P4 switches of Tbps), several works are proposed to offload the decision tree (DT) to P4 switches for in-network classification. Although the DT is suitable for the match-action paradigm in P4 switches, the range match rules used in the DT may not be supported across devices of different P4 standards. Additionally, emerging models including neural networks (NNs) and ensemble models, have shown their superior performance in networking tasks. But their sophisticated operations pose new challenges to the deployment of these models in switches. In this paper, we propose Mousikav2 to address these drawbacks successfully. First, we design a new tree model, i.e., the binary decision tree (BDT). Unlike the DT, our BDT consists of classification rules in the form of bits, which is a good fit for the standard ternary match supported by different hardware/software switches. Second, we introduce a teacher-student knowledge distillation architecture in Mousikav2, which enables the general transfer from other sophisticated models to the BDT. Through this transfer, sophisticated models are indirectly deployed in switches to avoid switch constraints. Finally, a lightweight P4 program is developed to perform classification tasks in switches with the BDT after knowledge distillation. Experiments on three networking tasks and three commodity switches show that Mousikav2 not only improves the classification accuracy by 3.27%, but also reduces the switch stage and memory usage by 2.00 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times$</tex-math> </inline-formula> and 28.67%, respectively. Code is available at https://github.com/xgr19/Mousika.

Bis-tridentate Ir(III) Phosphors and Blue Hyperphosphorescence with Suppressed Efficiency Roll-Off at High Brightness.

Narrowband blue emitters are indispensable in achieving ultrahigh-definition OLED displays that satisfy the stringent BT 2020 standard. Hereby, a series of bis-tridentate Ir(III) complexes bearing electron-deficient imidazo[4,5-b]pyridin-2-ylidene carbene coordination fragments and 2,6-diaryloxy pyridine ancillary groups were designed and synthesized. They exhibited deep blue emission with quantum yields of up to 89% and a radiative lifetime of 0.71 μs in the DPEPO host matrix, indicating both the high efficiency and excellent energy transfer process from the host to dopant. The OLED based on Irtb1 showed an emission at 468 nm with a maximum external quantum efficiency (EQE) of 22.7%. Moreover, the hyper-OLED with Irtb1 as a sensitizer for transferring energy to terminal emitter v-DABNA exhibited a narrowband blue emission at 472 nm and full width at half-maximum (FWHM) of 24 nm, a maximum EQE of 23.5%, and EQEs of 19.7, 16.1, and 12.9% at a practical brightness of 100, 1000, and 5000 cd/m2, respectively.

High-resolution non-line-of-sight imaging based on liquid crystal planar optical elements

Abstract Non-line-of-sight (NLOS) imaging aims at recovering hidden objects located beyond the traditional line of sight, with potential applications in areas such as security monitoring, search and rescue, and autonomous driving. Conventionally, NLOS imaging requires raster scanning of laser pulses and collecting the reflected photons from a relay wall. High-time-resolution detectors obtain the flight time of photons undergoing multiple scattering for image reconstruction. Expanding the scanning area while maintaining the sampling rate is an effective method to enhance the resolution of NLOS imaging, where an angle magnification system is commonly adopted. Compared to traditional optical components, planar optical elements such as liquid crystal, offer the advantages of high efficiency, lightweight, low cost, and ease of processing. By introducing liquid crystal with angle magnification capabilities into the NLOS imaging system, we successfully designed a large field-of-view high-resolution system for a wide scanning area and high-quality image reconstruction. Furthermore, in order to reduce the long data acquisition time, a sparse scanning method capitalizing on the correlation between measurement data to reduce the number of sampling points is thus proposed. Both the simulation and experiment results demonstrate a &gt;20 % reduction in data acquisition time while maintaining the exact resolution.

Hybrid optical convolutional neural network with convolution kernels trained in the spatial domain

Although many hybrid convolutional neural networks using optical convolution processors have been demonstrated for high energy efficiency and fast image processing speed, the need for accurate modeling of the optical convolution hardware and large Fourier kernel sizes result in a long and energy-intensive training process. Here, we demonstrate a hybrid convolutional neural network based on an optimized optical convolution processor—the system uses kernels trained in the spatial domain and compensates the optical path mismatch that arises from the reflection of digital micromirror devices for convolution computation. The spatial-domain convolution kernels are Fourier transformed and then binarized before being programmed into a digital micromirror device in the optical convolution processor. This method minimizes overhead from the optical convolution process in the training of the hybrid convolutional neural network, and it can improve the energy efficiency and reduce the training time of the hybrid optical convolutional neural network. For convolution computation using digital micromirror devices with grayscale images, bit-plane-wise convolution is also presented. We tested the hybrid neural network on MNIST, Fashion-MNIST, and CIFAR-10 color datasets and obtained classification accuracies of 98.7%, 85.8%, and 57.8%, respectively.

Application of response surface and artificial neural network optimization approaches for exploring methylene blue adsorption using luffa fiber treated with sodium chlorite

Wastewater often laden with colored impurities, presents a substantial environmental contamination risk, necessitating the exploration of effective pollutant removal methods. Clay particles have emerged as promising adsorbents for this purpose. This study delves into the intricacies of methylene blue removal in a dilute environment, utilizing pre-treated luffa fiber with sodium chlorite to eliminate the reactive pigment from aqueous solutions. The adsorption optimization is achieved by applying both response surface methodology (RSM) and an artificial neural network (ANN) architecture. The investigation focuses on four key parameters—sodium chlorite concentration, activation time, temperature, and their combined impact on methylene blue adsorption. Descriptive statistics discern optimal operational conditions for efficiently extracting methylene blue from sewage water solutions. Notably, the ANN models exhibit exceptional optimization accuracy, yielding a remarkable adsorption capacity of 133.58 mg/g. This surpasses the experimental (124.36 mg/g) and RSM (126.97 mg/g) values, indicating a 6.90 % improvement over experimental results and a 4.94 % increase over RSM outcomes. These findings underscore the high efficiency of the adsorption process, positioning it as a promising method for effectively removing ionic anthocyanin contaminants in sewage water treatment.

Fundamental Study of Press Molding Method for CFRP Preform Using a 3D Printer

Carbon fiber reinforced plastic (CFRP) is a composite material with high specific strength and is applied to transportation and aviation equipment. However, conventional processing methods require large-scale production apparatus or a high level of dexterity that only comes with extensive experience which makes it difficult to achieve high processing efficiency. The objective of this study is to develop a novel method for forming thermos-plastic CFRP (CFRTP) preforms implementing a 3D printer for press molding. Applying this method offers the advantage that continuous carbon fibers can be formed on a free-form surface. It also reduces the manufacturing time and operator skill required. The goal of this research is to establish a method for molding a free-form surface composed of continuous fibers by employing a 3D-printed preform designed to match the unfolded polygonised diagram of the free-form surface. Previous research introduced an unfolding approach for converting a three-dimensional shape to a plane surface based on a computer-aided design and manufacturing (CAD/CAM) system, enabling the generation of an unfolding diagram that maintains the continuity of fiber tow. Furthermore, the validity of unfolded diagram was confirmed by reproducing the objective three-dimensional shape from the unfolded diagram using thermos-setting CPRP (CFRTS) tow prepreg. In this study, the viability of the proposed molding process using CFRTP preform fabricated by a 3D printer was verified and an assessment of the formability of the molded parts was conducted.

Green processing via surface diffuse atmospheric plasma to enhance the dyeing performance on polylactic acid fabric.

Poor dyeing performance has been a major defect of polylactic acid (PLA) fibers, which is caused by the lack of active chemical groups in PLA, and hinders the widespread use of this biodegradable material. Most of the existing chemical modification methods are not environmentally friendly and produce effluents. Herein, we report a green, efficient and continuous method to process PLA fibers via surface diffuse atmospheric plasma for the improvement of its hydrophilicity and dyeing performance. PLA fibers were processed via atmospheric plasma for grafting oxygen-containing functional groups, such as carboxyl, to achieve hydrophilicity and, meanwhile, strengthen the binding interactions with various dye molecules via covalent bonds, ionic bonds, or hydrogen bonds. In addition, different mechanisms of improving the dyeing performance on plasma-modified PLA fibers with different kinds of dyes have been discussed. This approach of material modification involves no chemical additives and has high processing efficiency, showing the potential applicability of green treatment to products in various fields.

Analysis of processing efficiency, surface, and bulk chemistry, and nanomechanical properties of the Monel® alloy 400 after ultrashort pulsed laser ablation

Monel® alloy 400 has excellent corrosion resistance and finds applications in marine industries. The processing of marine components requires high processing efficiency and a quality finish. Hence, this research aims to investigate the effects of the laser processing parameters such as laser fluence, scanning velocity, hatching distance, and the scanning pass on the ablation rates and efficiency, chemistry, and nanomechanical properties of the Monel® alloy 400 after pulsed picosecond (ps) laser ablation. From the experimental findings, the ablation depth increases as the laser fluence increases while decreasing as the scanning velocity increases. Surface roughness was noted to increase as the laser fluence increased. The findings demonstrated that the ablation rate increases as laser fluence increases while ablation efficiency decreases. Energy dispersive x-ray spectroscopy (EDX) showed that the elemental composition of laser-ablated zones is almost similar to that of the polished sample. X-ray spectroscopy (XPS) shows that the outer layer on the surface of Monel® alloy 400 is composed of NiO and CuO. The hardness and Young’s modulus of the laser-processed alloy were found to be less than those of the bulk material. This study can be used to establish optimal processing parameters for the ultrafast ps laser processing of materials to achieve high ablation efficiency with a high-quality surface finish for industrial applications.

Photocatalytic degradation of methylene blue dye using biogenic copper oxide nanoparticles and its degradation pathway analysis

In order to remove the toxic organic dyes from the environment, biogenic inorganic semiconductor nanoparticle based photocatalyst has drawn attention due to its low cost, simplicity, eco-friendliness, high efficiency and ease of process. Herein, Copper oxide nanoparticles (CuO NPs) were prepared by using Syzygium Cumini seed extract as reducing as well capping agent. The formed CuO NPs were characterized by X-ray diffraction (XRD), Fourier Transform Infrared Spectroscopy (FT-IR), B.E.T isotherm, UV–Visible spectroscopy, Scanning Electron Microscopy (SEM) and X-ray Photoelectron Spectroscopy (XPS) analysis. The CuO NP’s photocatalytic activities have been analyzed with methylene blue (MB) degradation. The effect of the catalyst for MB degradation has been optimized at various catalyst amount, dye concentration and pH environments. The results showed that, the CuO NPs can effectively degrade 97 % of MB in 60 min, following pseudo first order kinetics, while the rate constant was found to be 0.06 min−1. The present catalyst material showed 95 % MB degradation with similar morphology of freshly prepared catalyst even after five consecuetive reuse cycles and worked effectively over a wide pH range. The degradation pathway of MB was studied through the HR-MS technique.

A Review of the Preparation of Porous Fibers and Porous Parts by a Novel Micro-Extrusion Foaming Technique.

This review introduces an innovative technology termed "Micro-Extrusion Foaming (MEF)", which amalgamates the merits of physical foaming and 3D printing. It presents a groundbreaking approach to producing porous polymer fibers and parts. Conventional methods for creating porous materials often encounter obstacles such as the extensive use of organic solvents, intricate processing, and suboptimal production efficiency. The MEF technique surmounts these challenges by initially saturating a polymer filament with compressed CO2 or N2, followed by cell nucleation and growth during the molten extrusion process. This technology offers manifold advantages, encompassing an adjustable pore size and porosity, environmental friendliness, high processing efficiency, and compatibility with diverse polymer materials. The review meticulously elucidates the principles and fabrication process integral to MEF, encompassing the creation of porous fibers through the elongational behavior of foamed melts and the generation of porous parts through the stacking of foamed melts. Furthermore, the review explores the varied applications of this technology across diverse fields and imparts insights for future directions and challenges. These include augmenting material performance, refining fabrication processes, and broadening the scope of applications. MEF technology holds immense potential in the realm of porous material preparation, heralding noteworthy advancements and innovations in manufacturing and materials science.

Two-step separation strategy with mixed-mode chromatography to purify recombinant human serum albumin from Pichia pastoris culture broth

Mixed-mode chromatography (MMC) with the specially-designed ligands shows good adsorption selectivity for protein separation. In this study MMC was explored to purify recombinant human serum albumin (rHSA) from Pichia pastoris culture broth, which combined bind-elute and flow-through modes. Four mixed-mode resins were evaluated for their adsorption capacities of HSA. It was found that two home-made MMC resins showed special performances. IC-B-6FF resin with indole-3-acetic acid-cysteine ligand had high adsorption and selectivity towards HSA, and SFA-B-6FF resin with sulfapyridine ligand showed minimal binding to HSA. Therefore, a two-step MMC separation strategy was designed that IC-B-6FF was used to capture rHSA directly from yeast culture broth and then SFA-B-6FF resin was used to polish the remaining impurities by flow-through mode. The results revealed that the purity of rHSA reached 95.1% with a yield of 81.2% after the two-step separation. The removal rate of host cell proteins (HCPs) was 99.8% with final HCP level as low as 50 ppm. The results demonstrate that MMC combining bind-elute and flow-through modes has a promising potential to separate protein with high selectivity and process efficiency.

Advanced techniques in quartz wafer precision processing: Stealth dicing based on filament-induced laser machining

This research aims to achieve the high-quality and high-efficiency processing of quartz crystal through cutting experiments performed on quartz wafers, leveraging the capabilities of an ultraviolet (UV) femtosecond laser (fs-laser). The study integrates two crucial techniques, filament-induced laser machining (FILM) and stealth dicing (SD), to introduce an innovative approach named stealth dicing based on filament-induced laser machining (SDFILM). Using a harmonic generator, a conventional ∼1030 nm, 270 fs infrared (IR) fs-laser is transformed into a ∼ 343 nm UV fs-laser, enabling FILM on quartz wafers. Then, following a systematic investigation of laser parameters and processing variables on the machined quartz grooves, the optimally machined grooves take on a distinctive “Y” shape and yield effortlessly upon gentle mechanical force, akin to SD. Finally, an optimized laser focus strategy is introduced, resulting in a near-vertical “|” type cutting pattern on the quartz wafers. This research serves as a new method in the field of quartz processing, shedding light on new possibilities in the precision manufacturing of hard and brittle crystal materials.

Compound mechanical and chemical-mechanical polishing processing technique for single-crystal silicon carbide

In this study, a novel polishing approach was proposed, that combined nanodiamond mechanical polishing (MP) with silicon dioxide chemical-mechanical polishing (CMP) to achieve the high-quality and high-efficiency processing of single-crystal silicon carbide (SiC). A compound-polishing experiment was conducted to evaluate the material-removal rate (MRR) and surface roughness. The results showed that MP had a high MRR and poor surface quality, whereas CMP obtained a better surface quality; however, its MRR was low. Using surface roughness as the research target, the relationship between MP and CMP was then obtained through a combined process optimisation analysis. Consequently, the compound-polishing method was revealed in terms of the material-removal mechanism, with the surface quality being related to the formation rate of the oxidation reaction layer. Finally, the effectiveness of the compound-polishing method was experimentally verified. For the same surface accuracy, the polishing efficiency of this method improved by 30 % compared to that of CMP. The results proved the proposed compound-polishing approach to be beneficial for enhancing the machining accuracy and efficiency of a single-crystal SiC substrate.

Defective ZnIn2S4 Nanosheets for Visible-Light and Sacrificial-Agent-Free H2O2 Photosynthesis via O2/H2O Redox.

H2O2 photosynthesis has attracted great interest in harvesting and converting solar energy to chemical energy. Nevertheless, the high-efficiency process of H2O2 photosynthesis is driven by the low H2O2 productivity due to the recombination of photogenerated electron-hole pairs, especially in the absence of a sacrificial agent. In this work, we demonstrate that ultrathin ZnIn2S4 nanosheets with S vacancies (Sv-ZIS) can serve as highly efficient catalysts for H2O2 photosynthesis via O2/H2O redox. Mechanism studies confirm that Sv in ZIS can extend the lifetimes of photogenerated carriers and suppress their recombination, which triggers the O2 reduction and H2O oxidation to H2O2 through radical initiation. Theoretical calculations suggest that the formation of Sv can strongly change the coordination structure of ZIS, modulating the adsorption abilities to intermediates and avoiding the overoxidation of H2O to O2 during O2/H2O redox, synergistically promoting 2e- O2 reduction and 2e- H2O oxidation for ultrahigh H2O2 productivity. The optimal catalyst displays a H2O2 productivity of 1706.4 μmol g-1 h-1 under visible-light irradiation without a sacrificial agent, which is ∼29 times higher than that of pristine ZIS (59.4 μmol g-1 h-1) and even much higher than those of reported photocatalysts. Impressively, the apparent quantum efficiency is up to 9.9% at 420 nm, and the solar-to-chemical conversion efficiency reaches ∼0.81%, significantly higher than the value for natural synthetic plants (∼0.10%). This work provides a facile strategy to separate the photogenerated electron-hole pairs of ZIS for H2O2 photosynthesis, which may promote fundamental research on solar energy harvest and conversion.

A facile strategy to fabricate lignocellulose-based slow-release fertilizers via a high-performance treatment of rice straw using deep eutectic solvents

Lignin-based slow-release fertilizers (SRFs) have attracted widespread attention due to their ability to enhance nutrient utilization efficiency and reduce environmental pollution in agricultural production. However, the extraction and separation processes of lignin from biomass sources are intricate, involving substantial quantities of non-reusable toxic reagents. Here, a sustainable and eco-friendly approach using deep eutectic solvents (DES) was employed to treat rice straw, effectively dissolving the lignin present. Subsequently, the in-situ lignin regeneration was facilitated through the addition of a zinc chloride solution. The regenerated lignin was tightly wrapped around and connected to cellulose micro/nanofibers, forming a homogeneous slurry. A simple coating technique was employed to uniformly coat urea particles with the lignocellulosic slurry, yielding lignocellulose-based SRFs. Results revealed that the nutrient release of the lignocellulose-based coated fertilizers in water exceeded 56 days. A pot trial demonstrated that the application of lignocellulose-based SRFs significantly promoted the growth of rice and improved grain yield (by 10.7 %) and nitrogen use efficiency (by 34.4 %) compared to the urea treatment in rice production. Furthermore, the DES demonstrated consistently high efficiency in biomass processing even after four cycles of reuse. This green strategy offers a novel approach for the preparation of SRFs coating materials, promoting agricultural sustainability.

A surface-normal photodetector as nonlinear activation function in diffractive optical neural networks

Optical neural networks (ONNs) enable high speed, parallel, and energy efficient processing compared to their conventional digital electronic counterparts. However, realizing large scale ONN systems is an open problem. Among various integrated and non-integrated ONNs, free-space diffractive ONNs benefit from a large number of pixels of spatial light modulators to realize millions of neurons. However, a significant fraction of computation time and energy is consumed by the nonlinear activation function that is typically implemented using a camera sensor. Here, we propose a novel surface-normal photodetector (SNPD) with an optical-in–electrical-out (O–E) nonlinear response to replace the camera sensor that enables about three orders of magnitude faster (5.7 µs response time) and more energy efficient (less than 10 nW/pixel) response. Direct efficient vertical optical coupling, polarization insensitivity, inherent nonlinearity with no control electronics, low optical power requirements, and the possibility of implementing large scale arrays make the SNPD a promising O–E nonlinear activation function for diffractive ONNs. To show the applicability of the proposed neural nonlinearity, successful classification simulations of the MNIST and Fashion MNIST datasets using the measured response of SNPD with accuracy comparable to that of an ideal ReLU function are demonstrated.