Model-based measurement of semiconductor device features with feed forward use of data for dimensionality reduction

• US 7,716,003 B1
• Filed: 07/16/2007
• Issued: 05/11/2010
• Est. Priority Date: 07/16/2007
• Status: Active Grant

• Current Assignee: Kla-Tencor Technologies Corporation (KLA Corporation)
• Original Assignee: Kla-Tencor Technologies Corporation (KLA Corporation)
• Inventors: Wack, Daniel C., Bareket, Noah, Ratner, Edward R., Veldman, Andrei, Hench, John
• Primary Examiner(s): Barbee; Manuel L
• Application Number:
  US11/879,227
• Time in Patent Office: 1,030 Days
• Field of Search: 702/58, 702/117, 702/118, 702/127, 702/155, 702/159, 702170-172, 356/625, 356/369, 700/121
• US Class Current: 702/117
• CPC Class Codes: G01N 21/4788 Diffraction for sizing particles G01N15/0205
• CPC Class Codes: G01N 21/956 Inspecting patterns on the surface of objects contactless testing of electronic circuits G01R31/308; testing currency G07D; manufacturing processes per se of semiconductor devices implementing a measuring step H01L22/10

Abstract

The present application discloses a new technique which reduces the dimensionality of a feature model by re-use of data that has been obtained by a prior measurement. The data re-used from the prior measurement may range from parameters, such as geometrical dimensions, to more complex data that describe the electromagnetic scattering function of an underlying layer (for example, a local solution of the electric field properties).

Claims

1. A method of model-based measurement of semiconductor device features, the method comprising:

identifying a substrate to process;

performing process steps to form at least one feature on the identified substrate;

performing measurements at multiple locations on the identified substrate;

performing computations to determine at least one parameter at said multiple locations on the identified substrate;

storing data for said at least one parameter, associated locations, and substrate identification in a database;

re-identifying the substrate to process;

performing process steps to form at least one additional feature on the identified substrate;

performing measurements at said multiple locations on the identified substrate;

retrieving said at least one parameter which was previously determined from the database; and

using said at least one parameter determined prior to said re-identification to reduce a dimensionality required for real-time regression to determine at least one additional parameter at said multiple locations on the identified substrate.

2. The method of claim 1, wherein the measurements comprise optical diffraction measurements.

3. The method of claim 1, wherein said real-time regression requires less computing time than real-time regression with a larger number of variable parameters.

4. An apparatus for model-based measurement of semiconductor device features, the apparatus comprising:

means for identifying a substrate to process;

at least one processing tool for performing process steps to form at least one feature on the identified substrate;

a measuring tool for performing measurements at multiple locations on the identified substrate;

a data processing system for performing computations to determine at least one parameter at said multiple locations on the identified substrate;

a data storage module for storing data for said at least one parameter, associated locations, and substrate identification in a database;

a data retrieval module for retrieving said at least one parameter which was previously determined from the database,

wherein the means for identifying is configured to subsequently re-identify the substrate to process,

wherein the at least one processing tool is configured to subsequently perform process steps to form at least one additional feature on the identified substrate,

wherein the measuring tool is configured to subsequently perform measurements at said multiple locations on the identified substrate, and

wherein the computing system is configured to use said at least one parameter determined prior to said re-identification to reduce a dimensionality required for real-time regression to determine at least one additional parameter at said multiple locations on the identified substrate.

5. The apparatus of claim 4, wherein the measurements comprise diffraction measurements.

6. The apparatus of claim 5, wherein said real-time regression requires less computing time than real-time regression with a larger number of variable parameters.

7. A method of model-based measurement of semiconductor device features, the method comprising:

performing process steps to form at least one feature on a substrate;

performing process steps to form at least one additional feature on the substrate in a device region but not test locations;

performing measurements at a test location;

performing computations to determine at least one parameter at the test location;

performing measurements at a target region of interest in the device region;

retrieving said at least one parameter for the test location; and

using said at least one parameter determined from the test location to reduce a dimensionality required for real-time regression to determine at least one additional parameter at said target region of interest.

8. The method of claim 7, wherein the test location is selected so as to be closely located to the target region of interest.

9. The method of claim 8, wherein the measurements comprise optical diffraction measurements.

10. The method of claim 9, wherein said reduced-dimensionality real-time regression requires less computing time than real-time regression with a larger number of variable parameters.

11. An apparatus for model-based measurement of semiconductor device features, the method comprising:

at least one processing tool for performing process steps to form at least one feature on a substrate and for performing process steps to form at least one additional feature on the substrate in a device region but not test locations;

a measuring system for performing measurements at a test location and for performing measurements at a target region of interest in the device region; and

a computing system configured to perform computations to determine at least one parameter at the test location, to retrieve said at least one parameter for the test location, and to use said at least one parameter determined from the test location to reduce a dimensionality required for real-time regression to determine at least one additional parameter at said target region of interest.

12. The apparatus of claim 11, wherein the computing system is configured to select a test location so as to be closely located to the target region of interest.

13. The apparatus of claim 12, wherein the measurements comprise diffraction measurements.

14. The apparatus of claim 13, wherein said reduced-dimensionality real-time regression requires less computing time than real-time regression with a larger number of variable parameters.