STATISTICAL MODEL-BASED METROLOGY
First Claim
1. A method comprising:
- receiving a first amount of measurement data associated with measurements of a first plurality of sites on a surface of a semiconductor wafer with known variations of at least one process parameter, structure parameter, or both, wherein the first amount of measurement data is derived from measurements performed by at least one metrology technique;
determining an expected response model of each of the at least one known process parameters, structure parameters, or both;
determining an input-output measurement model based at least in part on the first amount of measurement data; and
training the input-output measurement model based on parameter values determined from the expected response model.
1 Assignment
0 Petitions
Accused Products
Abstract
Methods and systems for creating a measurement model based on measured training data are presented. The trained measurement model is used to calculate process parameter values, structure parameter values, or both, directly from measured data collected from other wafers. The measurement models receive measurement data directly as input and provide process parameter values, structure parameter values, or both, as output. The measurement model enables the direct measurement of process parameters. Measurement data from multiple targets is collected for model building, training, and measurement. In some examples, the use of measurement data associated with multiple targets eliminates, or significantly reduces, the effect of under layers in the measurement result, and enables more accurate measurements. Measurement data collected for model building, training, and measurement, may be derived from measurements performed by a combination of multiple, different measurement techniques.
105 Citations
20 Claims
-
1. A method comprising:
-
receiving a first amount of measurement data associated with measurements of a first plurality of sites on a surface of a semiconductor wafer with known variations of at least one process parameter, structure parameter, or both, wherein the first amount of measurement data is derived from measurements performed by at least one metrology technique; determining an expected response model of each of the at least one known process parameters, structure parameters, or both; determining an input-output measurement model based at least in part on the first amount of measurement data; and training the input-output measurement model based on parameter values determined from the expected response model. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A system comprising:
-
a metrology tool including an illumination source and a detector configured to perform measurements of a target structure; and a computing system configured to; receive a first amount of measurement data from the metrology tool, the first amount of data associated with measurements of a first plurality of sites on a surface of a semiconductor wafer with known variations of at least one process parameter, structure parameter, or both; determine an expected response model of each of the at least one known process parameters, structure parameters, or both; determine an input-output measurement model based at least in part on the first amount of measurement data; and train the input-output measurement model based on parameter values determined from the expected response model. - View Dependent Claims (12, 13, 14, 15, 16, 17)
-
-
18. A method comprising:
-
receiving a first amount of measurement data associated with measurements of a semiconductor target on a first semiconductor wafer; determining at least one process parameter value, at least one structural parameter value, or both, based on a fitting of the first amount of measurement data to a trained input-output measurement model; and storing any of the at least one process parameter value, the at least one structural parameter value, or both, in a memory. - View Dependent Claims (19, 20)
-
Specification