Charge carrier scattering channels in graphite bridging its valence and conduction band offer an efficient Auger recombination dynamic to promote low energy charge carriers to higher energy states. It is of importance to answer the question whether a large number of charge carriers can be promoted to higher energy states to enhance the quantum efficiency of photodetectors. Here, we present an experimental demonstration of an effective Auger recombination process in the photo-excited nano-graphite film. The time-resolved hot carrier thermalization was analyzed based on the energy dissipation via the Auger scattering channels. We split the Auger recombination occurrence centered at 0.40 eV energy state into scattering and recombination parts, for characterizing the scattering rate in the conduction band and the recombination rate toward the valence band. The scattering time with respect to the energy state was extracted as 8 ps · eV⁻¹, while the recombination time with respect to the energy state was extracted as 24 ps · eV⁻¹. Our study indicates a 300 fs delay between the hot carrier recombination and generation, leading to a 10⁵ ps⁻¹ · cm⁻³ Auger scattering efficiency. The observed duration for the Auger recombination to generate hot carriers is prolonged for 1 ps, due to the hot carriers energy relaxation bottleneck with optical-phonons in the nano-graphite. The presented analytic expression gives valuable insights into the Auger recombination dynamic to estimate its most efficient energy regime for mid-infrared photodetection.

Aiming at the demand for rapid detection of highway pavement damage, many deep learning methods based on convolutional neural networks (CNNs) have been developed. However, CNN methods with raw image data require a high-performance hardware configuration and cost machine time. To reduce machine time and to apply the detection methods in common scenarios, the CNN structure with preprocessed image data needs to be simplified. In this work, a detection method based on a CNN and the combination of the grayscale and histogram of oriented gradients (HOG) features is proposed. First, the Gamma correction was employed to highlight the grayscale distribution of the damage area, which compresses the space of normal pavement. The preprocessed image was then divided into several unit cells, whose grayscale and HOG were calculated, respectively. The grayscale and HOG of each unit cell were combined to construct the grayscale-weighted HOG (GHOG) feature patterns. These feature patterns were input to the CNN with a specific structure and parameters. The trained indices suggested that the performance of the GHOG-based method was significantly improved, compared with the traditional HOG-based method. Furthermore, the GHOG-feature-based CNN technique exhibited flexibility and effectiveness under the same accuracy, in comparison to those deep learning techniques that directly deal with raw data. Since the grayscale has a definite physical meaning, the present detection method possesses a potential application for the further detection of damage details in the future.

Phosphorene, one of the graphene counterparts, is believed to have promising potential to be utilized in nanoelectronics due to its significant properties. Phosphorene has a nonplanar puckered structure with high anisotropy, which enables the elastic strain or external field to tune its electronic structure. In this work, we propose a nanodevice model based on an armchair phosphorene nanoribbon (APNR) with normal-metal electrodes and study the tuning effect of elastic strain and electric field on the electronic transport properties. We first confirm that the APNR can be driven to be of metallic conduction with linear dispersion around the Fermi level, by applying a critical compressive strain. After applying a perpendicular electric field, the APNR turns out to be a band insulator. Furthermore, we calculate the dc conductance and density of states (DOS) of the nanodevice, where the APNR is connected to normal-metal electrodes. The numerical results show that in the absence of an electric field, the nanodevice possesses peak values of conductance and DOS at the Fermi level. Once the electric field is applied, a gap emerges around the Fermi level in the conductance, which suggests that the nanodevice is turned off by the external electric field. Our investigation on the present system could be useful in the development of a field-effect nanodevice based on monolayer phosphorene.

The spin-valley-related electronic properties of quasi-one-dimensional kagome lattices with intrinsic spin-orbit coupling are studied, based on the tight-binding formalism. Three types of kagome-lattice nanoribbons along the x-direction with various geometric boundaries are proposed, including two symmetric nanoribbons and one asymmetric one. It is found that two nonequivalent Dirac cones and helical edge states exist in all the three types of kagome-lattice nanoribbons at 1/3 filling. Among them in the asymmetric nanoribbon, the spin and valley are found to be locked to each other due to inversion symmetry breaking, resulting in spin-valley polarized edge states. Band structure and probability density of wave function show that the spin-up/-down edge states locate at the K/K′ valley, with opposite propagation direction at the upper and lower boundaries. Spin-resolved real-space local current confirms the spin-valley polarized helical edge state in the asymmetric nanoribbon. The device application of the asymmetric kagome-lattice nanoribbon is worth further investigation.

In this work, we propose a road pavement damage detection deep learning model based on feature points from a local minimum of grayscale. First, image blocks, consisting of the neighborhood of feature points, are cut from the image window to form an image block dataset. The image blocks are then input into a convolutional neural network (CNN) to train the model, extracting the image block features. In the testing process, the feature points as well as the image blocks are selected from a test image, and the trained CNN model can output the feature vectors for these feature image blocks. All the feature vectors will be combined to a composite feature vector as the feature descriptor of the test image. At last, the classifier of the model, constructed by a support vector machine (SVM), gives the classification as to whether the image window contains damaged areas or not. The experimental results suggest that the proposed pavement damage detection method based on feature-point image blocks and feature fusion is of high accuracy and efficiency. We believe that it has application potential in general road damage detection, and further investigation is desired in the future.

In this work, we propose a Y-shaped nanojunction consisting of zigzag silicene nanoribbons (ZSNRs), and study the charge/spin transport based on a tight-binding approach, Landauer formalism and non-equilibrium Green’s function method. The topological states with various edge modes, such as quantum spin Hall insulator, quantum Hall insulator and band insulator, can all be engineered by the electric field and photoirradiation in ZSNRs. Through tuning the edge states of the two output terminals, three types of charge/spin router can be achieved in the present nanojunction. The first is a charge current switcher, which switches the charge current from one output terminal to the other. The second is a spin current router, which shunts the spin-up and spin-down current into different output terminals. The last is a spin filter, which filters the spin-polarized current into a specific output terminal. The proposed Y-shaped nanojunction may have potential applications in electronic and spintronic nanodevices in the future.