Abstract: Classifying wafers using Collaborative Learning. An initial wafer classification is determined by a rule-based model. A predicted wafer classification is determined by a machine learning model. Multiple users can manually review the classifications to confirm or modify, or to add user classifications. All of the classifications are input to the machine learning model to continuously update its scheme for detection and classification.

Abstract: A machine learning model for each die for imputing process control parameters at the die. The model is based on wafer sort parametric measurements at multiple test sites across the entire wafer, as well as yield results for the wafer. This allows for a better analysis of outlier spatial patterns leading to improved yield results.

Abstract: A sequence of models accumulates r-squared values for an increasing number of variables in order to quantify the importance of each variable to the prediction of a targeted yield or parametric response.

Abstract: An IC that includes a contiguous standard cell area with a 4×3 e-beam pad that is compatible with advanced manufacturing processes and an associated e-beam testable structure.

Abstract: Scheme for detection and classification of semiconductor equipment faults. Sensor traces are monitored and processed to separate known abnormal operating conditions from unknown abnormal operating conditions. Feature engineering permits focus on relevant traces for a targeted feature. A machine learning model is built to detect and classify based on an initial classification set of anomalies. The machine learning model is continuously updated as more traces are processed and learned.

Abstract: A characterization vehicle may include a first test circuit and a second test circuit located on separate panels of a panelized printed circuit (PC) board. The first test circuit may be fabricated in accordance with a first plurality of design parameters. The second test circuit may be fabricated in accordance with a second plurality of design parameters. The first plurality of design parameters and the second plurality of design parameters may be chosen in accordance with a design of experiment (DOE) concerning one or more design rules or design trade-offs such that at least two corresponding design parameters from the first and second test circuits have identical values, and at least two corresponding design parameters from the first and second test circuits have different values.

Abstract: A maintenance tool for semiconductor process equipment and components. Sensor data is evaluated by machine learning tools to determine when to schedule maintenance action.

Abstract: A method for processing a semiconductor wafer uses non-contact electrical measurements indicative of at least one tip-to-side short or leakage, at least one corner short or leakage, and at least one via open or resistance, where such measurements are obtained from non-contact pads associated with respective tip-to-side short, corner short, and via open test areas.

Abstract: Described are test structures and methods for measuring silicon thickness in fully depleted silicon-on-insulator technologies.

Abstract: Described is a method for implementing a snap to capability that enables the manufactured of a valid pattern in a semiconductor device, based upon an originally invalid pattern.

Abstract: Testing data is evaluated by machine learning tools to determine whether to include or exclude chips from further testing.

Abstract: An IC includes a contiguous standard cell area with first, second, and third TS-GATE-short-configured test area geometries disposed therein. In some embodiments, the contiguous standard cell area may further include: fourth and fifth TS-GATE-short-configured test area geometries, and/or other test area geometries, such as tip-to-tip-short, tip-to-side-short, diagonal-short, corner-short, interlayer-overlap-short, via-chamfer-short, merged-via-short, snake-open, stitch-open, via-open, or metal-island-open.

Abstract: Described here is an apparatus and method of testing a vertical (3D) semiconductor memory structure coupled between word lines and bit lines, by means of a direct connections of a plurality of test pads to word lines and bit lines of the memory structure on memory product wafer. Such connections are created by modified patterns of metal lines through contacts and vias created on the memory product wafer. The described apparatus and method are used for detecting electrical continuity (opens and shorts) in the memory structure, calculating resistance of selected word lines or bit cell strings, or performing more complex tests of memory bit cell transistors. The result of this detection can then be used to find defective regions or memory cells in the semiconductor memory structure. Such a testing device may be referred to as a direct testing system.

Abstract: Techniques for measuring and/or compensating for process variations in a semiconductor manufacturing processes. Machine learning algorithms are used on extensive sets of input data, including upstream data, to organize and pre-process the input data, and to correlate the input data to specific features of interest. The correlations can then be used to make process adjustments. The techniques may be applied to any feature or step of the semiconductor manufacturing process, such as overlay, critical dimension, and yield prediction.

Abstract: Disclosed is a system and method for performing direct memory characterization of memory cells in a memory array using peripheral transistors. A memory array is fabricated using a mask layer defining routing for a set of first stage periphery transistors electrically connected to the word lines of the memory array. A revised mask is used for defining a different routing for a set of second stage periphery transistors including different characteristics than the first stage periphery transistors. Testing is conducted by applying a simulated Erase signal to the nonvolatile memory cells and determining which cells are erased. Based on this test, certain characteristics of the first and/or second stage periphery transistors can be identified that provide improved conditions for the nonvolatile memory cells. A product chip can be manufactured using modified versions of the first stage periphery transistors that incorporate the characteristics that provide the improved condition(s).

Abstract: Disclosed is a system and method for collecting trace data of integrated circuits from the back-end assembly tools and using yield, reliability, and burn-in data to distinguish good circuit traces from bad ones. Described further is an system and method for implementing a heuristic mapping of trace data for distinguishing between good or bad traces in an Internet-based or offline application. The result of this detection can then be used for yield improvement or for burn-in reduction where for example burn-in chips having “good” circuit traces are subjected to thermal stress for less time than for chips identified as having “bad” circuit traces.

Abstract: Described is a CBCM technique that works only with PMOS transistors or only with NMOS transistors. Specifically, a method of monitoring performance of an integrated circuit device using a CBCM technique is disclosed, the method comprising: providing a metrology structure having a pseudo-inverter comprising a pull-up pull-down transistor switch, wherein the transistor switch comprises a pull-up transistor and a pull-down transistor of the same type; charging and discharging a device under test (DUT) coupled to the pseudo-inverter using a non-overlapping clock; measuring capacitance of the DUT with a gate voltage of the pull-up transistor at a preset value; and, using the value of the measured capacitance to estimate a dimension of a structure in the integrated circuit device. The non-overlapping clock is generated by: turning the pull-down transistor off when the pull-up transistor is on; and, turning the pull-down transistor on when the pull-up transistor is off.

Abstract: An apparatus and method for testing two-terminal memory elements organized as a cross-point memory array. The apparatus allows functional testing of two-terminal memory elements organized as a cross-point memory array, and built in a short flow manufacturing process. The proposed apparatus substantially eliminates the use of any type of additional active or passive switches, selectors, or decoders. A large number of memory elements of various memory types including planar (two dimensional) or three dimensional memory structures can be tested without the need of manufacturing selectors or running the full flow process.

Abstract: The present invention relates to IC chips containing a mixture of standard cells obtained from an original set of design rules and enhanced standard cells that are a variant of the original set of design rules and methods for making the same.

Abstract: Improved processes for manufacturing wafers, chips, or dies utilize in-line data obtained from non-contact electrical measurements (“NCEM”) of fill cells that contain structures configured target/expose a variety of open-circuit, short-circuit, leakage, or excessive resistance failure modes. Such processes may involve evaluating Designs of Experiments (“DOEs”), comprised of multiple NCEM-enabled fill cells, in at least two variants, all targeted to the same failure mode(s).