Defect classification and source analysis for semiconductor equipment

Abstract

Defects on a substrate comprising electronic components can be classified with a computational defect analysis system that may be implemented in multiple stages. For example, a first stage classification engine may process metrology data to produce an initial classification of defects. A second stage classification engine may use the initial classification, along with manufacturing information and/or prior defect knowledge to output probabilities that the defects are caused by one or more potential sources.

1 - 40 . (canceled) 
     
     
         41 . A defect analysis computational system comprising:
 (a) one or more processors;   (b) program instructions for executing on the one or more processors, the program instructions defining:
 one or more first stage defect classification engines configured to:
 receive metrology data acquired for defects on a substrate comprising electronic devices or partially fabricated electronic devices, wherein the metrology data comprises image data and composition data, and 
 produce one or more first stage defect classifications from the metrology data; and 
 
 a second stage defect classification engine configured to:
 receive the one or more first stage defect classifications produced by the one or more first stage defect classification engines, 
 receive manufacturing information comprising data about (i) manufacturing equipment that conducts a fabrication process on the substrate and produces the defects on the substrate, (ii) the fabrication process, (iii) one or more materials on the substrate, and/or (iv) prior defect classification for the manufacturing equipment and/or fabrication process, 
 determine, using the one or more first stage defect classifications and the manufacturing information, one or more sources of the defects on the substrate, and 
 output a likelihood of the defects being caused by a first source associated with the manufacturing equipment, one or more materials on the substrate, and/or fabrication process. 
 
   
     
     
         42 . The defect analysis computational system of  claim 41 , wherein at least one of the one or more first stage defect classification engines comprises one or more neural networks. 
     
     
         43 . The defect analysis computational system of  claim 42 , wherein the one or more neural networks comprise one or more first layers and one or more second layers, and the one or more first layers are configured to receive image data and the one or more second layers are configured to receive spectra data and outputs from the one or more first layers. 
     
     
         44 . The defect analysis computational system of  claim 43 , wherein the one or more first layers are convolutional layers and the one or more second layers are fully connected layers. 
     
     
         45 . The defect analysis computational system of  claim 41 , further comprising a single first stage defect classification engine configured to receive the image data and the composition data. 
     
     
         46 . The defect analysis computational system of  claim 41 , wherein the metrology data additionally comprises wafer map data. 
     
     
         47 . The defect analysis computational system of  claim 41 , wherein the metrology data comprises metadata pertaining to an inspection tool used in obtaining the metrology data. 
     
     
         48 . The defect analysis computational system of  claim 41 , wherein the one or more first stage defect classifications comprise a morphology classification for the defects on the substrate. 
     
     
         49 . The defect analysis computational system of  claim 41 , wherein the one or more first stage defect classifications comprise a composition classification for the defects on the substrate. 
     
     
         50 . The defect analysis computational system of  claim 41 , wherein the first stage defect classification comprises a wafer map classification for the defects on the substrate. 
     
     
         51 . The defect analysis computational system of  claim 41 , wherein the manufacturing information comprises data about a plurality of materials on the substrate and prior defect classifications for a manufacturing equipment and/or fabrication process. 
     
     
         52 . The defect analysis computational system of  claim 41 , wherein the second stage defect classification engine is additionally configured to further classify the defects on the substrate. 
     
     
         53 . The defect analysis computational system of  claim 41 , wherein the second stage defect classification engine is additionally configured to provide suggested corrective actions to reduce generation of defects on the substrate and/or reduce occurrences of defects on substrates processed in the future. 
     
     
         54 . The defect analysis computational system of  claim 41 , wherein the second stage defect classification engine is configured to determine the one or more sources of defects on the substrate by using a Bayesian analysis. 
     
     
         55 . The defect analysis computational system of  claim 41 , wherein the metrology data was obtained in situ during the fabrication process. 
     
     
         56 . The defect analysis computational system of  claim 41 , wherein the one or more first stage defect classification engines are further configured to receive sensor data selected from the group consisting of environmental conditions of the fabrication process, change in the mass of the substrate during the fabrication process, mechanical forces experienced during the fabrication process, and combinations thereof, and wherein the second stage defect classification engine is further configured to determine the one or more sources of defects on the substrate using the sensor data. 
     
     
         57 . A computational method of analyzing defects, the method comprising:
 receiving metrology data acquired for defects on a substrate comprising electronic devices or partially fabricated electronic devices, wherein the metrology data comprises image data and composition data;   producing one or more first stage defect classifications from the metrology data;   receiving manufacturing information comprising data about (i) manufacturing equipment that conducts a fabrication process on the substrate and produces the defects on the substrate, (ii) the fabrication process, (iii) one or more materials on the substrate, and/or (iv) prior defect classification for the manufacturing equipment and/or fabrication process,   determining, using the one or more first stage defect classifications and the manufacturing information, one or more likely sources of the defects on the substrate, and   outputting a likelihood of the defects being caused by a first source associated with the manufacturing equipment, one or more materials on the substrate, and/or fabrication process.   
     
     
         58 . The method of  claim 57 , further comprising producing at least one of the one or more first stage defect classifications using one or more neural networks. 
     
     
         59 . The method of  claim 58 , wherein the one or more neural networks comprise one or more first layers and one or more second layers, and the one or more first layers are configured to receive image data and the one or more second layers are configured to receive spectra data and outputs from the one or more first layers. 
     
     
         60 . The method of  claim 59 , wherein the one or more first layers are convolutional layers and the one or more second layers are fully connected layers. 
     
     
         61 . A computer program product for analyzing defects, the computer program product comprising a non-transitory computer readable medium on which is provided computer executable instructions for:
 receiving metrology data acquired for defects on a substrate comprising electronic devices or partially fabricated electronic devices, wherein the metrology data comprises image data and composition data;   producing one or more first stage defect classifications from the metrology data;   receiving manufacturing information comprising data about (i) manufacturing equipment that conducts a fabrication process on the substrate and produces the defects on the substrate, (ii) the fabrication process, (iii) one or more materials on the substrate, and/or (iv) prior defect classification for the manufacturing equipment and/or fabrication process,   determining, using the one or more first stage defect classifications and the manufacturing information, one or more likely sources of the defects on the substrate, and   outputting a likelihood of the defects being caused by a first source associated with the manufacturing equipment, one or more materials on the substrate, and/or fabrication process.   
     
     
         62 . The computer program product of  claim 61 , the non-transitory computer readable medium comprising further computer executable instructions for producing at least one of the one or more first stage defect classifications using one or more neural networks. 
     
     
         63 . The computer program product of  claim 62 , wherein the one or more neural networks comprise one or more first layers and one or more second layers, and the one or more first layers are configured to receive image data and the one or more second layers are configured to receive spectra data and outputs from the one or more first layers. 
     
     
         64 . The computer program product of  claim 63 , wherein the one or more first layers are convolutional layers and the one or more second layers are fully connected layers.