Voltage contrast method for semiconductor inspection using low voltage particle beam

• US 6,344,750 B1
• Filed: 01/08/1999
• Issued: 02/05/2002
• Est. Priority Date: 01/08/1999
• Status: Expired due to Term

• Current Assignee: Applied Materials Incorporated
• Original Assignee: Schlumberger Technologies Incorporated (SLB)
• Inventors: Kanai, Kenichi, Lo, Chiwoei Wayne
• Primary Examiner(s): Karlsen, Ernest
• Assistant Examiner(s): TANG, MINH NHUT
• Application Number:
  US09/227,395
• Time in Patent Office: 1,124 Days
• Field of Search: 324/765, 324/158.1, 324/71.3, 324/752, 324/753, 324/750, 324/751, 324/754, 324/501, 250/310, 250/307, 250/397, 250/492.2
• US Class Current: 324/754.22
• CPC Class Codes: G01R 31/307 of integrated circuits
• CPC Class Codes: H01J 2237/2594 Measuring electric fields or potentials
• CPC Class Codes: H01J 2237/2817 Pattern inspection
• CPC Class Codes: Y10S 977/849 with scanning probe

Abstract

Defects in a patterned substrate are detected by inspection with a charged particle beam inspection tool which generates an image of a portion of the patterned substrate and compares the image with a reference in order to identify any defects in the patterned substrate. Parameters of the tool are optimized to improve image uniformity and contrast, particularly voltage contrast. Prior to imaging an area of the substrate, the tool charges an area surrounding the image area to eliminate or reduce the effects caused by asymmetrical charging in the surrounding area. The tool alternates between charging the surrounding area and imaging the image area to produce a plurality of images of the image area, which are then averaged. The result is a highly uniform image with improved contrast for accurate defect detection.

Claims

1. A method of detecting defects in a patterned substrate comprising the acts of:

directing a charged particle beam onto the substrate;

scanning the beam across the substrate;

optimizing parameters of the beam to improve uniformity and contrast of a resulting image and an image acquisition speed;

acquiring an image of a first area of the substrate from charged particles from the substrate, including charging a second area of the substrate and imaging the first area, the second area encompassing the first area; and

comparing the acquired image with a reference to identify defects in the patterned substrate.

2. The method of claim 1, wherein optimizing the parameters includes optimizing at least one of a scan area size, scan speed, beam dose, beam current, beam energy, beam spot size, wafer chuck bias voltage, charge control plate bias voltage, energy filter voltage, and scan direction relative to a circuit pattern, and wherein the parameters being optimized for the first and second areas need not be the same.

3. The method of claim 1, wherein optimizing the parameters improves a voltage contrast of the image such that a background of the image has the same degree of contrast throughout the image.

4. The method of claim 1, wherein optimizing the parameters improves a voltage contrast of the image such that a circuit on the substrate has the same degree of contrast as other circuits with the same underlying connections.

5. The method of claim 1, wherein optimizing the parameters improves a voltage contrast of the image such that a circuit on the substrate has the same voltage contrast consistency when located anywhere within the first area.

6. The method of claim 1, wherein optimizing the parameters improves a voltage contrast of the image such there is a greater variation between the lightest and darkest areas of the image.

7. The method of claim 1, wherein the reference is any one of a stored image in a database, another patterned substrate, and a different portion of the same patterned substrate.

8. The method of claim 1, wherein the beam is produced from an electron-beam source.

9. The method of claim 8, further comprising the acts of emitting a focused beam from a primary electron gun for imaging the first area and emitting a flood beam from a flood gun for charging the second area.

10. The method of claim 9, further comprising the acts of scanning at a high amplitude when the flood beam is on and scanning at a lower amplitude when the focused beam is on.

11. The method of claim 1, further comprising the acts of:

alternating between charging the second area and imaging the first area to acquire a plurality of images of the first area; and

averaging the plurality of images.

12. The method of claim 11, wherein averaging the plurality of images includes generating a rolling average of the plurality of images.

13. The method of claim 11, wherein averaging the plurality of images includes simple arithmetic averaging.

14. A method of detecting defects in a patterned substrate comprising the acts of:

directing a charged particle beam onto the substrate;

scanning the beam across the substrate;

optimizing parameters of the beam to improve uniformity and voltage contrast of a resulting image and an image acquisition speed, wherein optimizing the parameters includes;

creating a performance matrix including second area sizes and beam doses ranging from a low second area size and a low beam does, respectively, at which the image acquisition speed is optimized, to a high second area size and a high beam dose, respectively, at which the best voltage contrast image is produced, the performance matrix indicating a voltage contrast quality and a time required to perform a charge operation at each respective second area size and beam does; and

selecting from the performance a particular second area size and beam dose based on a desired voltage contrast quality, such that there is a greater variation between the lightest and darkest areas of the image;

acquiring an image of a first area of the substrate from charged particles from the substrate, including charging a second area of the substrate and imaging the first area, the second area encompassing the first area; and

comparing the acquired image with a reference to identify defects in the patterned substrate.

15. A method of detecting defects in a patterned substrate comprising the acts of:

directing a charged particle beam onto the substrate;

scanning the beam across the substrate;

optimizing parameters of the beam to improve uniformity and voltage contrast of a resulting image and an image acquisition speed, wherein optimizing the parameters includes;

creating a performance matrix with at least one of a scan area size, scan speed, beam does, beam current, beam energy, beam spot size, wafer chuck bias voltage, charge control plate bias voltage, energy filter voltage, and scan direction relative to a circuit pattern, the performance matrix indicating a voltage contrast quality and a time required to perform a charge operation at each setting; and

selecting from the performance matrix a particular setting based on a desired voltage contrast quality, such that there is a greater variation between the lightest and darkest areas of the image;

acquiring an image of a first area of the substrate from charged particles fiom the substrate, including charging a second area of the substrate and imaging the first area, the second area encompassing the first area; and

comparing the acquired image with a reference to identify defects in the patterned substrate.

16. A method of detecting defects in a patterned substrate comprising the acts of:

directing a charged particle beam onto the substrate;

scanning the beam across the substrate;

optimizing parameters of the beam to improve uniformity and contrast of a resulting image and an image acquisition speed;

acquiring an image of a first area of the substrate from charged particles form the substrate, including removing a negative charge from a second area of the substrate and imaging the first area, the second area encompassing the first area; and

comparing the acquired image with a reference to identify defects in the patterned substrate.

17. The method of claim 16, wherein removing the negative charge from the second area of the substrate avoids a voltage contrast switch from positive to negative in the first area with increased magnification.

18. The method of claim 16, wherein removing the negative charge from the second area of the substrate avoids an appearance of charging-induced “

ghost features”

on the acquired image.