As an important computing operation, photonic matrix–vector multiplication is widely used in photonic neutral networks and signal processing. However, conventional incoherent matrix–vector multiplication focuses on real-valued operations, which cannot work well in complex-valued neural networks and discrete Fourier transform. In this paper, we propose a systematic solution to extend the matrix computation of microring arrays from the real-valued field to the complex-valued field, and from small-scale (i.e., 4 × 4) to large-scale matrix computation (i.e., 16 × 16). Combining matrix decomposition and matrix partition, our photonic complex matrix–vector multiplier chip can support arbitrary large-scale and complex-valued matrix computation. We further demonstrate Walsh-Hardmard transform, discrete cosine transform, discrete Fourier transform, and image convolutional processing. Our scheme provides a path towards breaking the limits of complex-valued computing accelerator in conventional incoherent optical architecture. More importantly, our results reveal that an integrated photonic platform is of huge potential for large-scale, complex-valued, artificial intelligence computing and signal processing.
Integrated photonics is widely regarded as an important post-Moore’s law research direction. However, it suffers from intrinsic limitations, such as lack of control and satisfactory photonic memory, that cannot be solved in the optical domain and must be combined with electronics for practical use. Inevitably, electronics and photonics will converge. The photonic fabrication and integration technology is gradually maturing and electronics-photonics convergence (EPC) is experiencing a transition from device integration to circuit design. We derive a conceptual framework consisting of regulator, oscillator, and memory for scalable integrated circuits based on the fundamental concepts of purposeful behavior in cybernetics, entropy in information theory, and symmetry breaking in physics. Leveraging this framework and emulating the successes experienced by electronic integrated circuits, we identify the key building blocks for the integrated circuits for EPC and review the recent advances.
Spatial confinement is a simple and cost-effective method for enhancing signal intensity and improving the detection sensitivity of laser-induced breakdown spectroscopy (LIBS). However, the spatial confinement effects of LIBS under different pressures remains a question to be studied, because the pressure of the ambient gas has a significant influence on the temporal and spatial evolution of plasma. In this study, spatial confinement effects of LIBS under a series of reduced air pressures were investigated experimentally, and the plasma characteristics under different air pressures were studied. The results show that the reduced air pressure can lead to both earlier onset and weakening of the enhancement effect of the spatial confinement on the LIBS line intensity. When the air pressure drops to 0.1 kPa, the enhancement effect of the emission intensity no longer comes from the compression of the reflected shock wave on the plasma, but from the cavity’s restriction of the plasma expansion space. In conclusion, the enhancement effect of spatial confinement technology on the LIBS is still effective when the pressure is reduced, which further expands the research and application field of spatial confinement technology.
Energy harvesting and light detection are key technologies in various emerging optoelectronic applications. The high absorption capability and bandgap tunability of organic semiconductors make them promising candidates for such applications. Herein, a poly(3-hexylthiophene-2,5-diyl) (P3HT):indene-C60 bisadduct (ICBA) bulk heterojunction-based organic photodiode (OPD) was reported, demonstrating dual functionality as an indoor photovoltaic (PV) and as a high-speed photodetector. This OPD demonstrated decent indoor PV performance with a power conversion efficiency (PCE) of (11.6 ± 0.5)% under a light emitting diode (LED) lamp with a luminance of 1000 lx. As a photodetector, this device exhibited a decent photoresponsivity of 0.15 A/W (green light) with an excellent linear dynamic range (LDR) of over 127 dB within the optical power range of 3.74 × 10-7 to 9.6 × 10-2 W/cm2. Furthermore, fast photoswitching behaviors could be observed with the rising/falling times of 14.5/10.4 µs and a cutoff (3 dB) frequency of 37 kHz. These results might pave the way for further development of organic optoelectronic applications.
Sodium metal batteries (SMBs) are receiving broad attention due to the high specific capacity of sodium metal anodes and the material abundance on earth. However, the growth of dendrites results in poor battery performance and severe safety problems, inhibiting the commercial application of SMBs. To stabilize sodium metal anodes, various methods have been developed to optimize the solid electrolyte interphase (SEI) layer and adjust the electroplating/stripping behavior of sodium. Among the methods, developing anode host materials and adding electrolyte additives to build a protective layer are promising and convenient. However, the understanding of the interaction process between sodium metal and those organic materials is still limited, but is essential for the rational design of advanced anode hosts and electrolyte additives. In this study, we use copper(II) hexadecafluorophthalocyanine (F16CuPc), and copper(II) phthalocyanine (CuPc), as model systems to unravel the sodium interaction with polar functional groups by in-situ photoelectron spectroscopy and density functional theory (DFT) calculations. It is found that sodium atoms prefer to interact with the inner pyrrolic nitrogen sites of CuPc, while they prefer to interact with the outer aza bridge nitrogen atoms, owing to Na-F interaction at the Na/F16CuPc interface. Besides, for the both organic molecules, the central Cu(II) ions are reduced to Cu(I) ions by charge transfer from deposited sodium. The fluorine-containing groups are proven to promote the interaction process of sodium in organic materials, which sheds light on the design of functional interfaces in host materials and anode protective layers for sodium metal anodes.
The slab effective index difference between the transverse-electric (TE) and transverse-magnetic (TM) polarizations was utilized to obtain complete photonic bandgap (CPBG) in a silicon nitride ( SixNy) photonic crystal slab. For this, coincident frequency range in the TE photonic bandgap (PBG) and TM PBG, which denotes the CPBGs of the slab, must be found with the same structure. Through adjusting the effective index pair of TE and TM polarizations by changing the thickness of the SixNy core layer, and also optimizing the structure parameters within the photonic crystal plane, a large normalized CPBG of 5.62% was theoretically obtained in a slab of SixNy with a refractive index of 2.5. Moreover, based on the obtained CPBG, a microcavity which could support both TE and TM polarization was theoretically demonstrated. The cavity modes for different polarizations were both well confined, which proved the reliability of the CPBG. In addition, using the same method, the lowest refractive index of SixNy on silica slab for a CPBG could be extended to as low as 2. The results indicate that there is potential for development of various high-performance CPBG devices based on SixNy slab technology.
Simultaneously realizing the optical and electrical properties of organic materials is always challenging. Herein, a convenient and promising strategy for designing organic materials with integrated optoelectronic properties based on cocrystal engineering has been put forward. By selecting the fluorene (Flu) and the 7,7′,8,8′-tetracyanoquinodimethane (TCNQ) as functional constituents, the Flu-TCNQ cocrystal prepared shows deep red emission at 702 nm, which is comparable to the commercialized red quantum dot. The highest electron mobility of organic field-effect transistor (OFET) based on Flu-TCNQ is 0.32 cm2 V-1s-1. Spectroscopic analysis indicates that the intermolecular driving force contributing to the co-assembly of Flu-TCNQ is mainly charge transfer (CT) interaction, which leads to its different optoelectronic properties from constituents.
The development of computer vision technology provides a possible path for realizing intelligent control of road sweepers to reduce energy waste in urban street cleaning work. For garbage segmentation of seven categories under road scene, we introduce an efficient deep-learning-based method. Our model follows a lightweight structure with a feature pyramid attention (FPA) module employed in the decoder to enhance feature integration at multi-levels. Besides, a similarity guidance (SG) module is added to the decoder branches, which calculates the cosine distance between learned prototypes and feature maps to guide the segmentation results from a metric learning perspective. Our model has less than 3 M parameters and can run at over 65 FPS in an RTX 2070 GPU. Experimental results demonstrate that our method can yield competitive results in terms of speed and accuracy trade-off, with overall mean intersection-over-union (mIoU) reaching 0.87 and 0.67, respectively, on two garbage data sets we built. Besides, our model can perform acceptable category-balanced segmentation from less than 20 annotated images per category by introducing the SG module.
With the rapid growth of computer science and big data, the traditional von Neumann architecture suffers the aggravating data communication costs due to the separated structure of the processing units and memories. Memristive in-memory computing paradigm is considered as a prominent candidate to address these issues, and plentiful applications have been demonstrated and verified. These applications can be broadly categorized into two major types: soft computing that can tolerant uncertain and imprecise results, and hard computing that emphasizes explicit and precise numerical results for each task, leading to different requirements on the computational accuracies and the corresponding hardware solutions. In this review, we conduct a thorough survey of the recent advances of memristive in-memory computing applications, both on the soft computing type that focuses on artificial neural networks and other machine learning algorithms, and the hard computing type that includes scientific computing and digital image processing. At the end of the review, we discuss the remaining challenges and future opportunities of memristive in-memory computing in the incoming Artificial Intelligence of Things era.
Metasurfaces, also known as 2D artificial metamaterials, are attracting great attention due to their unprecedented performances and functionalities that are hard to achieve by conventional diffractive or refractive elements. With their sub-wave-length optical scatterers, metasurfaces have been utilized to freely modify different characteristics of incident light such as amplitude, polarization, phase, and frequency. Compared to traditional bulky lenses, metasurface lenses possess the advantages of flatness, light weight, and compatibility with semiconductor manufacture technology. They have been widely applied to a range of scenarios including imaging, solar energy harvesting, optoelectronic detection, etc. In this review, we will first introduce the fundamental design principles for metalens, and then report recent theoretical and experimental progress with emphasis on methods to correct chromatic and monochromatic aberrations. Finally, typical applications of metalenses and corresponding design rules will be presented, followed by a brief outlook on the prospects and challenges of this field.
Co3O4 nanomaterials as electrodes have been studied widely in the past decade due to their unique structural characteristics. However, their performance does not yet reach the level required for practical applications. It is, nevertheless, an effective strategy to synthesize hybrid electrode materials with high energy density. Herein we prepare Co3O4@NiMoO4 nanowires by a two-step hydrothermal method. The as-obtained sample can be directly used as cathode material of supercapacitors; with specific capacitance of 600 C/g at 1 A/g. An assembled capacitor delivers an energy density of 36.1 Wh/kg at 2700 W/kg, and retains 98.2% of the initial capacity after 8000 cycles.
The recently reported non-fullerene acceptor (NFA) Y6 has been extensively investigated for high-performance organic solar cells. However, its charge transport property and physics have not been fully studied. In this work, we acquired a deeper understanding of the charge transport in Y6 by fabricating and characterizing thin-film transistors (TFTs), and found that the electron mobility of Y6 is over 0.3–0.4 cm2/(V·s) in top-gate bottom-contact devices, which is at least one order of magnitude higher than that of another well-known NFA ITIC. More importantly, we observed band-like transport in Y6 spin-coated films through temperature-dependent measurements on TFTs. This is particularly amazing since such transport behavior is rarely seen in polycrystalline organic semiconductor films. Further morphology characterization and discussions indicate that the band-like transport originates from the unique molecule packing motif of Y6 and the special phase of the film. As such, this work not only demonstrates the superior charge transport property of Y6, but also suggests the great potential of developing high-mobility n-type organic semiconductors, on the basis of Y6.
A stable mode-locked laser was demonstrated using a newly developed zinc phthalocyanine (ZnPc) thin film as passive saturable absorber (SA) in ytterbium-doped fiber laser (YDFL). The ZnPc thin film was obtained using a casting method and then inserted between the two fiber ferrules of a YDFL ring cavity to generate mode-locked pulses. The resulting pulsed laser operated at a wavelength of 1034.5 nm having a repetition rate of 3.3 MHz. At pump power of 277 mW, the maximum output power and pulse energy are achieved at 4.92 mW and 1.36 nJ, respectively. ZnPc has a high chemical and photochemical stability, and its significance for use as a potential SA in a mode-locked laser is reported in this work.
