Method of combining multiple sets of overlapping surface-profile interferometric data to produce a continuous composite map

• US 6,185,315 B1
• Filed: 09/15/1998
• Issued: 02/06/2001
• Est. Priority Date: 12/20/1996
• Status: Expired due to Term

• Current Assignee: Bruker Nano, Inc. (Bruker Corporation)
• Original Assignee: WYKO, Inc.
• Inventors: Becker, Brian W., Schmucker, Mark A.
• Primary Examiner(s): Mehta, Bhavesh
• Assistant Examiner(s): PATEL, KANJIBHAI B
• Application Number:
  US09/153,365
• Time in Patent Office: 875 Days
• Field of Search: 382/282, 382/283, 382/284, 382/154, 382/275, 382/286, 382/108, 382/168, 356/351, 356/357, 356/359, 356/376, 356/345, 702/166
• US Class Current: 382/108
• CPC Class Codes: G06T 11/00 2D [Two Dimensional] image generation

Abstract

A method of combining height profiles of adjacent sections of a test surface to produce a composite profile of the surface consists of taking successive measurements of adjacent sections of the surface of the test sample by sequentially placing them within the field of view of the instrument and profiling them by phase shifting or vertical scanning. The x-y translation of the microscope between successive measurements from one section to the next adjacent section of the surface being profiled is carried out by overlapping such sections, so that spatial continuity is maintained between measurements. The height data generated for each section are then combined to form a larger image corresponding to the entire surface tested and discontinuities and/or errors introduced by the x-y translation process are corrected by normalizing the overlapping portions to a common reference plane. A plane is fitted through each set of measured heights in the overlapping regions and the tip, tilt and offset of each fitted plane are corrected to produce matching overlapping height data in adjacent sections. The measured height data for the balance of each section are then also corrected by the same difference in tip, tilt and offset to obtain a continuous normalized image.

Claims

1. A method for using an interferometric profiler having an objective with a predetermined field of view to profile a test surface larger than said field of view, comprising the following steps:

(a) taking successive interferometric measurements of adjacent sections of the test surface by sequentially placing each section within the field of view of the objective and independently profiling each section to generate a corresponding height map; and

(b) combining a plurality of height maps generated in step (a) and normalizing said height maps to a common reference plane to form a composite map corresponding to the test surface.

2. The method described in claim 1, wherein step (a) is carried out by maintaining a region of overlap between sections, so that spatial continuity is retained between measurements.

3. The method described in claim 1, wherein the test surface is stationary and between successive measurements step (a) is carried out by translation of the objective with respect to the test surface.

4. The method described in claim 1, wherein the objective is stationary and between successive measurements step (a) is carried out by translation of the test surface with reference to the objective.

5. The method described in claim 3, wherein step (a) is carried out by maintaining a region of overlap between sections, so that spatial continuity is retained between measurements.

6. The method described in claim 4, wherein step (a) is carried out by maintaining a region of overlap between sections, so that spatial continuity is retained between measurements.

7. The method described in claim 1, wherein each of said adjacent sections of the test surface consists of a predetermined area of the test surface having a fixed geometry and containing a set of pixels corresponding to measured heights.

8. The method described in claim 2, wherein each of said adjacent sections of the test surface comprises a substantially rectangular section of the test surface containing a set of pixels corresponding to a matrix of measured heights.

9. The method described in claim 8, wherein the normalization in step (b) is carried out by

(c) positioning the objective over a predetermined initial section of the test surface and obtaining a measured height profile of said initial section by interferometric measurements;

(d) positioning the objective over a predetermined subsequent section of the test surface having a region of overlap with the initial section and obtaining a measured height profile of said subsequent section by interferometric measurements;

(e) positioning the objective over another predetermined subsequent section of the test surface having a region of overlap with any section previously measured and obtaining a measured height profile of said other predetermined subsequent section by interferometric measurements;

(f) repeating step (e) as required to cover a predetermined portion of the test surface by positioning the objective over a predetermined succeeding section of the test surface having a region of overlap with a preceding section of the test surface, and by obtaining a measured height profile of said succeeding section by interferometric measurements;

(g) selecting one of the sections with measured height profiles generated in steps (c) through (f) as a current reference section;

(h) fitting a reference plane equation through measured heights of said current reference section in a region of overlap with another section;

(i) fitting a correction plane equation through measured heights of said other section in said region of overlap;

(j) calculating a difference between said reference and correction plane equations at each pixel of said other section and adding said difference to a corresponding measured height for said pixel such as to produce a corrected height for each pixel of the other section, thereby generating a current composite section of the test surface having a current composite profile comprising said measured heights of the current reference section and said corrected heights of the other section; and

(k) repeating steps (h)-(j) to iteratively combine said current composite profile of the test surface generated by a last iteration with a height profile of another section thereof having a region of overlap with said current composite profile;

wherein at each iteration said current composite section is treated as the current reference section, and said current composite profile is treated as said measured heights of the current reference section.

10. The method described in claim 9, wherein said matrix has constant r,s dimensions for all of said adjacent sections, and wherein steps (c)-(f) and steps (h)-(k) are carried out by operating on adjacent sections sequentially either by row or by column.