Method for setting and adjusting process parameters to maintain acceptable critical dimensions across each die of mass-produced semiconductor wafers
First Claim
Patent Images
1. A wafer processing system comprising:
- (a) a mass-production line that processes wafers that are to have a mask-defined circuit layout, said processing occurring on a mass-production basis;
(b) sample-and-measure means for sampling processed wafers flowing through the mass-production line and for measuring and storing the values of critical dimensions of the sampled wafers;
(c) a machine-implemented, matched process model that is calibrated by use of the stored values of critical dimensions of the sampled wafers to model the behavior of the mass-production line with respect to the mask-defined circuit layout;
(d) consultor means for consulting the matched process model to predict the statistical behavior of the mass-production line for large numbers of wafers at various settings of process parameters and various tolerance allowances in those process parameters; and
(e) expert control means, operatively coupled to the sample-and-measure means, to the consultor means, and to the mass-production line, for detecting undesired variances in the statistical distribution of the processed wafer outflow of the mass-production line, for querying the consultor means to identify the most likely contributing causes for said undesired variances, and for responsively adjusting the settings of process parameters and/or the tolerance allowances of those process parameters that are identified as being the most likely contributing causes for said undesired variances;
(d.1) wherein said consultor means includes a best leveraging analyzer.
0 Assignments
0 Petitions
Accused Products
Abstract
A method and system are disclosed for: (a) matching a machine-implemented process simulator with an actual fabrication line, (b) using the matched model to simulate the statistical results of mass production by the modeled production line, (c) using the model to predict cross-reticle variance from collected data for in-scribe features, (d) using the model to decompose the variance contributions of each process parameter and identify the more prominent contributors, and (e) using the model to identify the process parameter adjustments which would provide best leverage when taken one at a time.
244 Citations
39 Claims
-
1. A wafer processing system comprising:
-
(a) a mass-production line that processes wafers that are to have a mask-defined circuit layout, said processing occurring on a mass-production basis; (b) sample-and-measure means for sampling processed wafers flowing through the mass-production line and for measuring and storing the values of critical dimensions of the sampled wafers; (c) a machine-implemented, matched process model that is calibrated by use of the stored values of critical dimensions of the sampled wafers to model the behavior of the mass-production line with respect to the mask-defined circuit layout; (d) consultor means for consulting the matched process model to predict the statistical behavior of the mass-production line for large numbers of wafers at various settings of process parameters and various tolerance allowances in those process parameters; and (e) expert control means, operatively coupled to the sample-and-measure means, to the consultor means, and to the mass-production line, for detecting undesired variances in the statistical distribution of the processed wafer outflow of the mass-production line, for querying the consultor means to identify the most likely contributing causes for said undesired variances, and for responsively adjusting the settings of process parameters and/or the tolerance allowances of those process parameters that are identified as being the most likely contributing causes for said undesired variances; (d.1) wherein said consultor means includes a best leveraging analyzer. - View Dependent Claims (2, 3, 4, 5)
-
-
6. A wafer processing system comprising:
-
(a) a mass-production line for processing one or more lots of wafers, each wafer being processed to form a predefined circuit layout repeatedly thereon, the predefined circuit layout having at least one critically-dimensioned feature that defines operational characteristics of the corresponding circuit, wherein the mass-production line has a plurality of controllable process parameters with each said controllable process parameter having a respective mean setting and a respective variance as seen by in-process wafers of said one or more lots of wafers; (b) sample-and-measure means for sampling in-process wafers flowing through a given part of the mass-production line and for measuring and recording dimensions of prespecified features of the sampled wafers; (c) a machine-implemented, matched process model that is defined by use of the recorded dimensions to model the statistical behavior of the mass-production line at least with respect to the dimensions of prespecified features of the sampled wafers, (c.1) wherein the matched process model includes programmable, mean-and-variance modeling means for modeling respective mean settings and respective variances of a respective one or more of the controllable process parameters of said mass-production line; and (d) consultor means for temporarily modifying respective combinations of modeled mean setting and modeled variance of a respective one or more of the controllable process parameters so as to thereby cause the temporarily modified process model to predict the statistical behavior of the mass-production line, at least with respect to the dimensions of the prespecified features, for large numbers of wafers, given the modified combinations of respective mean settings and respective variances. - View Dependent Claims (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)
-
-
25. A wafer processing system comprising:
-
(a) a mass-production line for processing a plurality of wafers to produce from each wafer a plurality of devices, each said wafer being processed by the line to form a predefined layout of features repeatedly on the wafer, the predefined layout having at least one critically-dimensioned feature that defines operational characteristics of each correspondingly produced device of the wafer, wherein the mass-production line has a plurality of controllable process parameters with each said controllable process parameter having a respective mean and a respective variance as seen by in-process devices of each of said wafers; (b) sampling means for obtaining from in-process wafers flowing through the mass-production line, sample measurements of prespecified features of the in-process wafers; (c) a machine-implemented, process model that models the statistical behavior of the mass-production line at least with respect to said prespecified features, said model having one or more modifiable inputs corresponding to a respective one or more of the controllable process parameters of the mass-production line; and (d) a machine-implemented consultor means for temporarily disturbing one or more of the modifiable inputs of said model so as to thereby cause the temporarily disturbed process model to predict the statistical behavior of the mass-production line, at least with respect to the measured features given the temporary disturbing of said modifiable inputs. - View Dependent Claims (26, 27, 28, 29, 30, 31, 32, 33)
-
-
34. A machine-implemented simulation system comprising:
-
(a) a slow-to-evaluate first simulator; (b) a fast-to-evaluate second simulator that can be programmed to generate second results substantially equal to first results of the first simulator in less time than does the slow-to-evaluate first simulator when the so-programmed second simulator is given input parameters within a predefined range; (c) bootstrapping means, responsive to results of the first simulator, for programming the fast-to-evaluate second simulator to generate said second results in response to input parameters within said predefined range; and (d) substituting means for substituting the programmed second simulator in place of the slow-to-evaluate first simulator. - View Dependent Claims (35, 36, 37, 38, 39)
-
Specification