Over the past two decades, the research interest in neuron science and artificial intelligence (AI) has increased dramatically,1–3 as progress in neuron science is expected to lead to a higher level of understanding the brain, and eventually the creation of smarter artificial neural networks (ANNs).4,5 This co-dependence of technology and scientific research in biology will inevitably end with the merging of natural and artificial systems. However, such an endeavor remains a formidable task, as our current state of engineering does not allow for seamless integration of those two naturally dissimilar systems. Toward this aim, biomimicry6 is regarded as an innovative solution in biofabrication and biological applications, introducing models and designs from nature.7–9 Plenty of collaborative brain projects have been proposed around the globe, including the Brain Initiative,10 the Human Brain Project,11,12 China Brain Project,13 Brain Mapping by Integrated Neurotechnologies for Disease Studies,14 and the Brain Alliance.15 These projects have stimulated an intense research field to create various two-dimensional (2D) and three-dimensional (3D) biomimetic neuron interfaces (BNIs),16–19 where biological neuron tissues could be integrated into an artificial extracellular matrix20–22 to rebuild, understand,23–26 and improve neuronal functions in neuron tissue engineering,21,27–30 and also enable physicists to achieve novel BNIs toward the development of ANNs.6,31,32
Recent development of artificial neural networks (ANNs) and inverse design methods have demonstrated their prospective significance for planar diffractive lens design, with a plethora of optical lenses designed for wavelengths ranging from visible to Thz wavelengths. However, previous research to design planner diffractive lenses only considers the maximum intensity in the focus area or its derivatives as the optimization function, leaving the intensity outside the focus area unconsidered. We proposed and investigated a two-dimensional (2D) physics-driven ANN method assisted by the negative Pearson correlation coefficient (NPCC) to design microlenses with varied focusing distances, which takes the entire 2D intensity distribution at the focus plane as an optimization function. Taking advantage of 3D two-photon nanolithographic technology, sub-micrometer thickness microlenses with varied focusing lengths are designed and fabricated, achieving an average focusing efficiency of around 35%, and an average focusing spot size of about 1 µm. Furthermore, a microlens array (19 by 19 microlenses with a total size of 4 mm2) with a curved focusing plane was fabricated and integrated into a CMOS sensor, achieving direct object imaging under incoherent white light illumination. Our results demonstrate that the NPCC is a very useful optimization function for designing planar diffractive lenses, and the use of NPCC in ANNs is of great potential for the future design of functional diffractive optical elements in optics and nanophotonics.
Perovskite quantum dots (PQDs), with their excellent optical properties, have become a leading semiconductor material in the field of optoelectronics. However, to date, it has been a challenge to achieve the three-dimensional high-resolution patterning of perovskite quantum dots. In this paper, an in situ femtosecond laser-direct-writing technology was demonstrated for three-dimensional high-resolution patterned CsPbBr3 PQDs using a two-photon photoresist nanocomposite doped with the CsPbBr3 perovskite precursor. By adjusting the laser processing parameters, the minimum line width of the PQDs material was confirmed to be 98.6 nm, achieving a sub-100 nm PQDs nanowire for the first time. In addition, the fluorescence intensity of the laser-processed PQDs can be regulated by the laser power. Our findings provide a new technology for fabricating high-resolution display devices based on laser-direct-writing CsPbBr3 PQDs materials.
The manipulation of light at the micrometer scale has been the core technology for the recent development of diffractive optics, especially for optical displays and diffractive neural networks (NNs). However, the factors that influence the diffraction efficiency of the diffractive surfaces have not been well understood. Taking advantage of a simplified physics-driven neural network model and two-photon nanolithography (TPN) technology, we theoretically and experimentally investigated the factors that influence the diffraction efficiency of the diffractive surfaces, such as diffraction propagation distances and a variety of diffraction patterns with different geometrical features, the result of which enables the generation of two-dimensional images with high diffraction efficiencies through the diffractive surfaces with a thickness of less than 1 μm and a size of 100 μm by 100 μm. The demonstrated results are of great significance for optical light manipulation in optical display and computational imaging using free-space diffractive optical elements.
Three-dimensional two-photon nanolithography (3D TPN) enables the mask-free fabrication of arbitrary 3D microstructures, which makes it a powerful tool for creating complex micrometer-scale diffractive optical elements (DOEs) in micro-optics and nanophotonics. Among these, the development of multiwavelength achromatic diffractive microlenses has emerged as a critical challenge, as conventional methods often rely on computationally intensive full-wave Maxwell simulations (e.g., finite-difference time-domain, FDTD) or complex meta-atom fabrication processes. Here, we propose an optimization framework that integrates the optical path difference equation into the Rayleigh-Sommerfeld diffraction equation, enabling direct optimization of height distribution for multiwavelength diffractive microlenses with significantly reduced computational demands compared to FDTD-based approaches. As a demonstration, we designed and fabricated a dual-wavelength (green and near-infrared) height-optimized achromatic microlens using 3D TPN, achieving excellent agreement between experimental characterization and theoretical calculations. Our approach offers a computationally efficient strategy for advancing multiwavelength DOE design, with potential applications in integrated photonics and bioimaging.