31 results about "Critical layer" patented technology

A method for monitoring overlay alignment on a wafer that includes identifying a target machine, identifying a target process, identifying a plurality of critical layers, obtaining a plurality of overlay data from at least one of designated registration patterns on the wafer as baseline data, providing a plurality of reference overlay data, correlating the plurality of the reference overlay data with the baseline data to obtain overlay error, comparing the overlay error with specifications of the target machine, accepting the baseline data when the overlay error is within the specifications, and performing overlay alignment monitoring with the baseline data.

Distributed Bragg reflectors (DBRs), and VCSELs that use such DBRs, comprised of AlP layers on InP substrates. When grown on an InP substrate, if the critical layer thickness (tcrt) of AlP is greater than λ / 4nAlP, where nAlP is the index of refraction of InP and λ is the wavelength, then the DBR can be grown using alternating layers of InP and AlP, wherein the thickness of the AlP is less than the critical thickness. If the critical layer thickness (tcrt) of AlP is greater than λ / 4nAlP, then the DBR mirror is grown using alternating layers of InP and of an AlP / InP superlattice, wherein the AlP / InP superlattice is comprised of InP and of AlP wherein the thickness of the AlP is less than the critical thickness.

An optical semiconductor device includes an InP substrate; an active layer disposed above the InP substrate; a n-type semiconductor layer disposed below the active layer; and a p-type clad layer disposed above the active layer, wherein the p-type clad layer includes one or more p-type In1-xAlxP layers, the Al composition x of each of the one or more p-type In1-xAlxP layers is equal to or greater than a value corresponding to the doping concentration of a p-type dopant, and the absolute value of the average strain amount of the whole of the p-type clad layer is equal to or less than the absolute value of a critical strain amount obtained by Matthews' relational expression, using the entire layer thickness of the whole of the p-type clad layer as a critical layer thickness.

The invention relates to an optimal filling thickness control method for coal gangue and belongs to agricultural reclamation methods for the coal gangue. The control method comprises the steps: under the condition of selected compacting equipment, firstly, determining critical layering thickness of the coal gangue, then, determining the optimal layering thickness according to an object function model with minimal total compacting times, and finally, calculating filling super high amount; and during construction, carrying out filling by adopting a strip stacking method, and controlling the quality of filling and compacting by adopting a leveling method. The method has the advantages that uniform-thickness layered filling is carried out, and thus, the filling effect is the best; and the uniformity is guaranteed to the maximum, the compacting times are reduced to the maximum, the instability and non-uniform settlement of the gangue are lowered, and the leveling of farmland is guaranteed. Critical thickness control is carried out; during the layered filling, the filling thickness cannot exceed critical thickness, so that the control method is the most effective control method for disadvantageous engineering characteristics of the gangue, and disadvantageous characteristics such as weathering, swelling-shrinkage, spontaneous combustion and corrosion of the gangue are effectively controlled. The filling cost of the coal gangue is reduced; and on one hand, the engineering quality is guaranteed, and on the other hand, the engineering cost is reduced.

Disclosed is a critical layer stress adjusting method for solving cracking of a multi-layer film caused by electron beam deposition. Electron beam deposition technology and ion beam assisted deposition technology are combined to adopt, a stress critical layer is deposited by using the ion beam assisted deposition technology, and the other film layers except to the stress critical layer are deposited by using the electron beam deposition technology. The critical layer stress adjusting method can solve the problem that when a thicker film layer is deposited by the electron beams, the film layer is cracked because tension stress borne by the film layer is too high, and the difficulties of film system designing and practical preparation technologies are not increased.

The present invention relates to a design method of a digital mark in a test layout, comprising: step (a) first directly defining a non-key layer digital mark, wherein the non-key layer digital mark has a relatively large length and width; step (b) Multiple virtual patterns of key layers are inserted into the non-key layer digital mark area to form recognizable key layer digital marks. In order to solve the problem in the prior art that the special pattern of the digital mark causes process defects, the invention provides a brand-new layout digital mark design method. The digital mark design method defines a digital mark through non-critical layers, and then fills the digital mark area with a universally existing dummy pattern to form a recognizable uniform digital mark including all physical layers. A more uniform pattern can be obtained throughout the layout by the described setting.

The invention discloses a method for improving graphics critical dimension uniformity in a mask, which comprises the following steps in the technical process of photoetching: (1) adding a plurality of auxiliary graphics in areas in which virtual graphics are not permitted to be inserted in a domain before manufacturing the mask, thereby balancing difference of the graphics densities in different local areas; (2) manufacturing the mask in accordance with a revised domain which is added with the auxiliary graphics; and (3) forming chip graphics by the exposure of the photoetching technique. The invention can increase the graphic critical dimension in the chip unit on the mask and the uniformity of a section, thereby improving control of critical layers on a high wafer layer.

Provided is a Chinese chestnut branch embedding grafting method. When stock or young trees reach the grafting standard, grafting is performed, and the grafting method includes the steps of scion grafting, two-incision grafting, management performed on the current grafting year and pruning of live tree columns. Because the cortical layers of seedlings of nursery land and the cortical layers of the young trees which are planted newly are thin, the branch embedding grafting method can effectively protect the critical layers of branches and trunks, and the method is simple, practical and high in survival rate. Through the method, coppice shoots, generated after grafting, of the branches and the trunks can be removed in time or bud picking is performed as soon as possible, loss of tree liquid nutrient is reduced, and grafted buds can grow healthily and strongly. The method is particularly suitable for cultivating grafting seedlings in the nursery land and being used in a Chinese chestnut orchard where the young trees are newly planted, and the method can also be used for grafting of the branches and the trunks, with thin critical layers, of adult Chinese chestnut trees and grafting seedling cultivation of other fruit trees.

The invention discloses a method for controlling surrounding rock in gob-side entrydriving of a working face of a deep island. The method comprises the following steps of performing pre-cracking and top breaking before stoping of two adjacent working surfaces of the working face of the island, so as to avoid forming a hinge structure at the end surface of the working face, and avoid the hinged rock beam at the critical layer transferring the stoping stress of covering rock of the adjacent working faces to the working face; selecting a constant-drag large-deformation anchor bolt in a gob-side entrydriving support system, increasing the lengths of anchor bolts at the coal wall side and top plate side, deforming the anchor bolts to relieve pressure, and enabling anchor cables to transfer the concentrated stress of the surrounding rock to the farther physical coal wall; slantwise and downwards drilling pressure relief holes in the physical coal wall side with 3 to 4 cycle roof weighting paces in advance at the working face of the island; spraying slurry to the narrow coal column side. The method has the advantages that the stress concentration degree of the deep working face is effectively reduced, the drags of a bracket, a support column and support are reduced, the stoping efficiency of coal sources is improved, the safe and efficient stoping efficiency of the working face is also improved, and the major technical and economic benefits are realized.

The invention provides a method for determining key convolutional layer hyperparameters of a remote sensing classification convolutional neural network, comprising the following steps: construction of a convolutional neural network sample set; construction of a convolutional neural network structure; key layer hyperparameters of a convolutional neural network Determination: Select one of the convolutional layers as the key layer, preset the size of the convolution kernel of the key layer, and calculate the convolution scale; calculate the convolution step size according to the set rules according to the convolution kernel and convolution scale of the key layer; Set the convolution kernel of other convolutional layers as kernelsize, and the convolution step of other convolutional layers as 1; the subsequent downsampling size adopts mean downsampling or maximum downsampling. The invention proposes the concept of convolution scale based on the image input size and convolution kernel size, which is compatible with the remote sensing space scale, and on this basis, provides a joint determination key based on the input size and convolution scale The method of layer hyperparameters can reduce the time required for algorithm tuning and improve the classification accuracy of object-oriented remote sensing.