Author: raoam488

A new microfabrication technology is bridging the gap between intricate optical designs and large-scale manufacturing of optics for wafers. This innovation combines additive microfabrication with step-and-repeat nanoimprint lithography, paving the way for mass production of advanced micro-optical components.

The demand for sophisticated optical sensors in smartphones, tablets, wearables, and other mobile devices is surging, driving the need for cost-effective miniaturization of optical components. Wafer-level optics manufacturing, which utilizes semiconductor-style processes, offers a promising solution.

Step-and-repeat nanoimprint lithography (S&R NIL) now enables the scaling of highly detailed optical structures, initially designed at the die level, to full 300mm wafers. This is achieved by replicating a master mold multiple times across a substrate, creating master templates that maintain the fidelity of the original design. This S&R master then facilitates the production of working stamps for subsequent wafer-level manufacturing. This approach is significantly faster and less costly than traditional methods, enabling the mass production of sophisticated optical designs previously limited to niche applications.

Key to creating these initial master molds is two-photon grayscale lithography (2GL), a high-precision 3D printing technology. 2GL allows for the fabrication of complex micro-optical elements with exceptional shape accuracy and extremely smooth surfaces, surpassing traditional methods like diamond turning or electron-beam lithography. It offers rapid design iterations and cost-effective microfabrication of 2.5-dimensional structures.

By combining the design freedom and precision of 2GL with the efficient replication of UV-nanoimprint lithography (UV-NIL) and S&R mastering, a complete manufacturing chain is established. This process begins with creating a master using 2GL, scaling it to the wafer level through S&R mastering, and finally enabling high-volume wafer-level production using NIL.

The S&R mastering process ensures high precision with alignment within 250 nanometers and comprehensive process control, maintaining an optimal environment throughout dispensing, imprint, curing, and demolding. Working stamps, replicated from the S&R master, are used for the final imprint, minimizing wear on the expensive master and allowing for quick and cost-effective replacement of defective stamps during mass production.

While polymer shrinkage during replication can cause slight dimensional changes, these are predictable and can be compensated for in the master design, particularly facilitated by the flexibility of 2GL. The 2GL method allows for rapid prototyping and direct testing of designs on glass substrates, enabling early design validation and quicker iteration cycles. This combined approach significantly reduces the time required to bring innovative optical designs from concept to mass market, not only for advanced micro-optics but also for other 2.5D structures like sensor components in microfluidic chips. Ultimately, this technology breakthrough overcomes previous limitations in micro-optics manufacturing, paving the way for improved performance and new applications in consumer electronics and beyond.

Source link

Continental Eyewear has launched its Star Fit collection, a new line of eyewear designed for children aged one to eight. The launch took place at 100% Optical. This collection is a product of Millmead Optical Group and utilizes research data from Dr. Alicia Thompson, Director of Education, Research and Professional Development at ABDO. Dr. Thompson’s research informed the design to improve the fit of spectacles for young children, moving away from simply scaling down adult frames.

James Conway, CEO of Millmead Optical Group, acknowledged that previously their children’s eyewear was essentially smaller versions of adult frames, not specifically designed for children’s head shapes. Dr. Thompson’s research, born from practical frustrations with ill-fitting children’s frames, involved a seven-year PhD study measuring 1300 children aged from birth to 16. The study considered ethnicity, gender and whether the child had Down syndrome, and the resulting data is publicly accessible.

Millmead Optical Group adopted this data early in their design process, working with Dr. Thompson to understand the needs of dispensing opticians, including frame adjustability. Conway highlighted the importance of frame fit, particularly with the rise of myopia management lenses in children’s eyewear. The Star Fit collection balances technical requirements, aesthetics, and comfort in its design.

The collection features 16 SKUs comprising eight models in two colours, using eco-acetate and recycled metal, all equipped with flex hinges. Dr. Thompson explained her research findings indicating that children have a low and wide bearing surface, requiring lower bridge positions and wider angles in frame design. The research also provided data on children’s head widths and ethnic variations, informing adjustability features in the new frames. The frames incorporate flex hinges, cylindrical cores for easier cutting, and notched tips. Dr. Thompson emphasized the goal of creating stable and comfortable frames for children, addressing the common issue of frames sliding down children’s noses due to designs based on adult nasal profiles. The Star Fit collection aims to counteract this problem through the incorporation of comprehensive dimensional data into its design.

Source link

Researchers have developed a novel computational approach for designing advanced Terahertz (THz) optical elements, overcoming significant challenges in manipulating these waves for broadband applications. The team’s method centers on a sophisticated understanding of scalar diffraction, meticulously accounting for how THz waves change as they pass through specially structured surfaces and air. This precise modelling is crucial for creating devices that work effectively across a range of THz frequencies.

Designing broadband THz diffractive optical elements is inherently complex because it requires multiple frequencies to be accurately steered and constructively interfere at a target point. Theoretically, achieving perfect efficiency is impossible due to the nature of wave interference. However, this new research successfully integrates these physical limitations into a computational design process.

To tackle the intricate optimization required, the researchers implemented a refined algorithm called Gradient Descent Assisted Binary Search (GDABS). This algorithm, built upon existing binary search techniques, operates under the principle that effective solutions are abundant within the design possibilities. The algorithm iteratively tweaks the height of individual pixels in the optical element’s structure, assessing the impact of each change on the device’s performance using a Figure of Merit. By incorporating a gradient descent approach, the GDABS algorithm efficiently navigates the design space, converging rapidly to an optimal solution. The use of symmetry in the design further accelerated the process. Notably, this new GDABS method significantly reduces computation time compared to traditional techniques, while ensuring the optimized design represents a highly effective solution.

The capability of this algorithm was demonstrated through the design of three distinct THz optical components. First, a high-performance broadband spherical lens was created, capable of focusing a wide range of THz frequencies with minimal aberrations. This lens was physically fabricated using common 3D printing techniques and its performance was experimentally validated, showing good agreement with computational predictions. Furthermore, simulations confirmed the lens’s polarization insensitivity, as expected from the scalar diffraction model.

Second, the researchers designed spectral splitters, analogous to optical gratings, which separate different THz frequencies into spatially distinct beams. Different splitter designs were created to achieve both regular and arbitrarily spaced frequency separation, showcasing the versatility of the design approach. Interestingly, designs for arbitrary frequency separation exhibited enhanced spectral resolution.

Finally, a complex on-axis broadband transmissive hologram was designed to project a THz image of Mickey Mouse. This hologram demonstrated excellent image fidelity across the targeted broadband frequencies, with high transmission efficiency.

Further investigation into the robustness of these designs revealed a strong tolerance to fabrication imperfections, specifically random variations in pixel height expected from standard 3D printing processes. This indicates the practical viability of the designed THz elements for real-world applications. Despite limitations in measurement facilities which introduced some experimental inaccuracies, the overall results qualitatively validated the design methodology. This advancement paves the way for more sophisticated and efficient THz technologies across various fields, from imaging and sensing to communications.

Source link

Researchers have developed tiny, flat lenses known as metalenses as alternatives to bulky traditional glass lenses for manipulating light. These metalenses, thinner than a micron and covered with nanoscale objects, can change light properties and be precisely engineered to focus light. This technology allows for significant miniaturization in optical devices, including microscopes, cameras, VR headsets, optical sensors for IoT, and enhancements for optical fibers. Stacking metalenses is possible without significantly increasing size, enabling complex devices like spectrometers and polarimeters.

A significant breakthrough has addressed chromatic aberration, a problem where white light splits into different wavelengths focusing at varying distances. A single metalens can now focus all wavelengths of white light to a single point. Further advancements include metalenses correcting other image aberrations like coma and astigmatism.

The manufacturing process of metalenses, utilizing semiconductor industry equipment, promises to reduce the cost of optical components and potentially integrate optical and electronic components on the same chip. However, current challenges include the high precision required for nanoscale element placement on centimeter-scale chips and limitations in light transmission efficiency compared to traditional lenses, which affects applications like full-color imaging. Their small size also limits light capture, making them currently unsuitable for high-quality photography.

Despite these limitations, metalenses are expected to be integrated into smaller sensors, diagnostic tools like endoscopes, and optical fibers in the coming years. This potential has attracted research support from government agencies and companies like Samsung and Google, with the startup Metalenz aiming to commercialize the technology soon.

Source link

Nanophotonics is advancing technological innovations across various fields, necessitating intricate photonic microstructures for nanoscale light manipulation. Designing these microstructures is complex and often evades conventional optimization methods. Optical metasurfaces offer a solution by miniaturizing optical components onto a single plane. Their functionality stems primarily from their structure and orientation rather than the materials themselves. These surfaces, built from sub-wavelength structures, precisely control light’s phase, polarization, mode, and spectrum, enabling the creation of custom light patterns for integrated photonics applications such as light coupling, gratings, and lenses.

The design of a metasurface hinges on the size, orientation, shape, and arrangement of its meta-atoms, the individual scattering units. While analytical and numerical methods exist to predict metasurface electromagnetic behavior, they are often slow and limited when meta-atom dimensions approach the light’s wavelength. Numerical simulations like FEM, FDTD, and FIT, though used, rely on trial-and-error, hindering the design of complex devices.

To address these limitations, researchers are exploring machine-learning approaches, particularly Deep Neural Networks (DNNs). DNNs offer faster and more effective routes to design metasurfaces. These algorithms can model intricate, non-linear relationships, creating nanophotonic simulators that bypass time-consuming conventional methods. This includes developing both ‘forward models’ that predict electromagnetic responses from given designs, and ‘inverse models’ that determine design parameters for desired responses.

One specific application being explored is photonic beam engineering using meta-scatterer distributions in planar waveguides. The ability to create tailored beam profiles like Gaussian, focused, and collimated beams enhances the efficiency of integrated opto-fluidic sensors used for analyte excitation in on-chip fluorescence and IR spectroscopy. This integrated approach offers advantages including scalability for large-scale field of view by using many excitation sources, compact and cost-effective hardware, and automation of biological analyses. Furthermore, it can lead to improved grating couplers for photonic integrated circuits, maximizing power transmission through fiber couplers. This technology also paves the way for edge emitter laser sources to be used in vertical light excitation applications, relevant to LiDAR and 3D depth sensing.

Recent studies highlight the increasing interest in using DNNs for metasurface design. Research has demonstrated the effectiveness of Artificial Neural Networks for RF and microwave designs, Multi-layer perceptron models for studying photonic crystal dispersion, and hybrid EM optimization methods enhanced with machine learning for predicting metamaterial permittivity. Generative Neural Networks and Generative Adversarial Networks are also being applied to efficiently explore design spaces and optimize metasurface topology, showing the broad adoption of machine learning in this field to create advanced nanophotonic devices.

In a specific design, a planar metasurface with a 5×5 array of meta-scatterers is being used to diffract engineered light beams from a photonic waveguide made of silicon nitride (SiN) on a silicon dioxide-on-silicon wafer. This design focuses on controlling light diffraction through key parameters: periodicity of the scatterers or grating period, the gap between scatterers described by the gap factor, the height or etch depth of the scatterers, and the size of individual scatterers, particularly their width in the direction perpendicular to light propagation. These parameters influence the diffraction angle and energy leakage of the light beam, ultimately shaping the output beam profile. The relationship between these design parameters and the resulting light diffraction is being exploited to develop machine learning models capable of solving the inverse problem: determining the necessary metasurface structure to achieve a desired light diffraction pattern.

Source link

Optical coherence tomography, or OCT, is a medical imaging technique that has become increasingly vital in recent years. It uses light waves to capture high-resolution, three-dimensional images from within optical scattering media, such as biological tissues. Think of it as ultrasound, but using light instead of sound, allowing for much finer detail.

Initially developed for ophthalmology, OCT has revolutionized the diagnosis and management of various eye diseases including macular degeneration, glaucoma, and diabetic retinopathy. Its ability to provide cross-sectional views of the retina and other ocular structures with micron-level resolution offers clinicians unprecedented insights into tissue morphology. Doctors can now visualize retinal layers, measure nerve fiber thickness, and assess anterior chamber angles with greater precision than ever before.

Beyond the eye, the versatility of OCT is expanding its reach into other medical specialties. Cardiology is exploring OCT for imaging within blood vessels to guide stent placement and assess plaque buildup. Dermatology is using it to examine skin lesions for cancer diagnosis, and even dentistry explores its potential for detecting early caries. Research is ongoing to improve OCT technology, focusing on increasing imaging speed, penetration depth, and image quality.

Scientists are actively working on advanced techniques like adaptive optics OCT to correct for distortions and enhance resolution even further. Wavefront shaping methods are being investigated to improve imaging through dense tissues, potentially unlocking new diagnostic capabilities in areas previously inaccessible with standard OCT. Innovations in interferometer designs are also leading to more compact, cost-effective, and robust OCT systems, making this valuable technology more widely available.

The ongoing research and development in optical coherence tomography promise even broader applications in the future. As technology advances, OCT is poised to become an even more indispensable tool for medical diagnostics and research, improving patient care across a range of specialties.

Source link

Researchers have developed a new terahertz (THz) spectroscopic system that incorporates 3D-printed neural networks to manipulate and analyze THz light. The system utilizes a titanium-sapphire laser to generate ultrashort optical pulses, which are then used to produce and detect THz radiation. One part of the laser beam activates a plasmonic photoconductive nano-antenna array, the THz emitter, generating pulses of THz light. These THz pulses are then directed and focused onto a detector, another high-sensitivity plasmonic nano-antenna array, using off-axis parabolic mirrors. The remaining part of the optical laser beam is used to trigger the THz detector after passing through a delay line, allowing for time-domain measurements.

The detected signal, representing the interaction of THz and optical fields, is amplified and processed to obtain a time-domain signal. This system boasts a high signal-to-noise ratio, exceeding 90 decibels, and can operate across a broad bandwidth of up to 5 THz. Measurements are taken rapidly, with each trace captured in 5 seconds and averaged over 10 pulses within a 400 picosecond time window.

A key feature of this development is the integration of computationally designed diffractive neural networks. Fabricated using high-resolution 3D printing, these networks are positioned within the THz beam path, between the emitter and detector. These networks, characterized by a 1×1 centimeter input aperture, are designed to perform specific optical functions on the THz beam, such as spectral filtering. The design process involves complex simulations based on the Rayleigh-Sommerfeld equation of diffraction, allowing for precise control over the amplitude and phase modulation of the THz waves as they propagate through the 3D-printed layers.

The neural networks are digitally designed and optimized using machine learning techniques, considering the material properties and fabrication constraints of the 3D printing process. A training process is employed to refine the network parameters for desired spectral responses, balancing factors like signal power and spectral resolution. The resulting 3D structures, constructed from materials with specific refractive properties in the THz range, act as sophisticated optical elements capable of performing complex THz signal processing tasks. This advancement paves the way for new possibilities in THz imaging, sensing, and communication technologies.

Source link

Researchers have developed a novel approach using deep learning to accelerate and optimize the design of nanoantennae, crucial components in advanced optical devices known as metasurfaces. These metasurfaces, engineered with meticulously designed nanostructures, can manipulate light in unprecedented ways.

The team generated a comprehensive dataset of 1500 aluminum nanoantenna designs, categorized into rectangles, double-arcs, rectangle-circle pairs, and rectangle-square pairs, each with varying dimensions. These designs, visualized as 2D images, were paired with simulated optical responses representing how efficiently they convert left circular polarized light to right circular polarized light across a spectrum of wavelengths (400nm to 800nm). The pixel intensity in the design images encodes the thickness of a dielectric spacer layer, a critical factor influencing performance.

To streamline the design process, two deep learning models were employed. A Simulation Neural Network (SNN), functioning as a forward design tool, was trained to predict the optical response of a given nanoantenna design. This model, structured with convolutional and fully connected layers, learns the complex relationship between the design’s geometry and its optical behavior, enabling rapid prediction of performance.

In tandem, a conditional Generative Adversarial Network (cGAN) was implemented for inverse design. This model is capable of generating nanoantenna designs based on a user-specified desired optical response. The cGAN architecture involves a generator, which creates designs from random noise and the desired optical response, and a discriminator, which distinguishes between real and generated designs. Through adversarial training, the generator learns to produce realistic and effective designs that match the requested optical performance.

To further refine the design process, the researchers introduced a cyclical generation framework. This innovative system integrates the SNN and cGAN models with a pseudo-genetic algorithm (pGA). Beginning with a batch of user-defined optical responses, the cGAN generates initial designs, which are then evaluated by the SNN to predict their optical performance. The pGA then selects the best-performing designs based on accuracy metrics like Mean Squared Error (MSE) and cosine similarity, which measures the similarity in the shape of the optical response. These top designs, along with their associated optical responses, are incorporated back into the training dataset, iteratively improving the models’ ability to generate and predict optimal designs in subsequent cycles. This cyclical refinement process effectively combines forward and inverse design, leading to highly optimized nanoantenna structures.

Evaluations of the models demonstrated high accuracy, with cosine similarity used to ensure the models captured the sharp variations and nuances in the optical responses. The training process was conducted efficiently on standard hardware, highlighting the practical applicability of this approach.

This deep learning-based framework significantly accelerates the design and optimization of nanoantennae for metasurfaces. By automating and optimizing the design process, this method paves the way for faster development and deployment of advanced optical technologies with tailored functionalities. The cyclical generation framework offers a powerful tool for creating complex nanostructures for a wide range of optical applications.

Source link

Scientists have developed a new mathematical approach to better understand and design electrostatic lenses, crucial components in technologies like electron microscopes. Electrostatic lenses use electric fields to focus beams of charged particles, similar to how optical lenses focus light. These lenses are essential for high-resolution imaging and analysis in various scientific fields.

Researchers have introduced a novel method for calculating the potential distribution within these lenses, specifically focusing on systems with multiple electrodes. This new formulation is particularly valuable because it addresses the complexities of real-world lens systems where electrodes can have varying thicknesses and gaps between them.

The team’s work centers around using advanced mathematical techniques, including quintic spline equations and matrix formulations, to solve fundamental physics equations that govern the behavior of electric fields. This approach allows for a more precise and efficient way to determine the potential distribution along the axis of the lens, a critical factor in lens performance.

By developing this refined mathematical model, scientists and engineers can design electrostatic lenses with improved accuracy and performance. This advancement could lead to enhanced electron microscopes with better resolution and capabilities, benefiting research in materials science, biology, and nanotechnology. The new method provides a more detailed and accurate tool for lens design, paving the way for future innovations in charged particle beam technology.

Source link