The present disclosure generally relates to determining a process condition in a semiconductor process using attribute-relative process conditions. An example is a method of forming an integrated circuit (IC). First and second historical process conditions are obtained. The first historical process conditions are of previous semiconductor processing corresponding to a target value of a process attribute for forming the IC, and the second historical process conditions are of previous semiconductor processing corresponding to variable values of the process attribute. Attribute-relative process conditions are calculated. Each attribute-relative process condition is based on the first historical process conditions and the second historical process conditions that correspond to a respective given value of the variable values. An average process condition is determined from a subset of the attribute-relative process conditions. A process condition of a subsequent semiconductor process is set based on the average process condition.

A method of generating training spectra for training of a neural network includes generating a plurality of theoretically generated initial spectra from an optical model, sending the plurality of theoretically generated initial spectra to a feedforward neural network to generate a plurality of modified theoretically generated spectra, sending an output of the feedforward neural network and empirically collected spectra to a discriminatory convolutional neural network, determining that the discriminatory convolutional neural network does not discriminate between the modified theoretically generated spectra and empirically collected spectra, and thereafter, generating a plurality of training spectra from the feedforward neural network.

A substrate assembling device (1) includes a first end effector 10 attached to a first arm (3), a second end effector 20 attached to a second arm (3), and a controller 4. The second end effector 20 includes a pair of grippers 22 configured to grip a second substrate 102, and a placing part 23 where threaded elements are placed. The controller 4 is adapted to control operations of the first arm and the second arm to position the second substrate 102 on a first substrate 101 while gripping the second substrate 102 by using the pair of grippers 22 of the second end effector 20, and hold the threaded element placed on the placing part 23 of the second end effector 20 and fasten the held threaded element, by using the first end effector 10, to join the first substrate 101 and the second substrate 102 together.

A method, apparatus, and system for manufacturing a composite part. A set of reference locations is identified for a set of fiducial markers on a composite ply from a ply shape model for the composite part. The set of fiducial markers is created at the set of reference locations on the composite ply. The composite ply is cut to have a shape defined by the ply shape model.

A method of generating training spectra for training of a neural network includes measuring a first plurality of training spectra from one or more sample substrates, measuring a characterizing value for each training spectra of the plurality of training spectra to generate a plurality of characterizing values with each training spectrum having an associated characterizing value, measuring a plurality of dummy spectra during processing of one or more dummy substrates, and generating a second plurality of training spectra by combining the first plurality of training spectra and the plurality of dummy spectra, there being a greater number of spectra in the second plurality of training spectra than in the first plurality of training spectra. Each training spectrum of the second plurality of training spectra having an associated characterizing value.

Methods and systems for laser etching substrates to fine tune antennas for wireless communication are provided. A method includes laser etching an antenna element design into a substrate material. The antenna element design is for receiving conductive material to form an antenna structure. The method also includes laser etching a first area of the substrate material to change an intrinsic property of the substrate material in order to control an electrical characteristic of the antenna structure.

Methods and apparatus for the in-situ measurement of metrology parameters are disclosed herein. Some embodiments of the disclosure further provide for the real-time adjustment of process parameters based on the measure metrology parameters. Some embodiments of the disclosure provide for a multi-stage processing chamber top plate with one or more sensors between process stations.

A method of generating training spectra for training of a neural network includes measuring a first plurality of training spectra from one or more sample substrates, measuring a characterizing value for each training spectra of the plurality of training spectra to generate a plurality of characterizing values with each training spectrum having an associated characterizing value, measuring a plurality of dummy spectra during processing of one or more dummy substrates, and generating a second plurality of training spectra by combining the first plurality of training spectra and the plurality of dummy spectra, there being a greater number of spectra in the second plurality of training spectra than in the first plurality of training spectra. Each training spectrum of the second plurality of training spectra having an associated characterizing value.

Some implementations described herein provide a deposition tool that includes a grounding component between an edge ring of a substrate stage and a pumping plate component. The grounding component includes a grounding strap having a deformation region. The deformation region includes a recessed edge to reduce a likelihood of the grounding strap rubbing against a surface of the pumping plate component during operation of the deposition tool. Material properties of the grounding strap may reduce a likelihood of plastic deformation of the grounding strap during repeated cycling. In this way, an amount of particulates dislodged from the surface of the pumping plate component may be decreased to improve a yield of semiconductor product fabricated using the deposition tool. Furthermore, a frequency of servicing the grounding component may be decreased to decrease a downtime of the deposition tool and increase a throughput of semiconductor product fabricated using the deposition tool.

A system includes a light emitter attached to a destination chamber, the light emitter to emit a collimated light beam across an entrance to the destination chamber. The system includes an end effector attached to a distal end of an arm of a robot. The system includes a two-dimensional (2D) area sensor disposed on the end effector at a location that coincides with the collimated light beam while the end effector reaches within the destination chamber. The 2D area sensor is to detect a location of the collimated light beam incident on a surface of the 2D area sensor and transmit, to a controller coupled to the robot, sensing data including the location.