Optical measuring method for semiconductor wafer including a plurality of patterns and method of manufacturing semiconductor device using optical measurement

Patent metadata

Field	Value
Publication number	US-10410937-B2
Application number	US-201816035991-A
Country	US
Kind code	B2
Filing date	Jul 16, 2018
Priority date	Dec 8, 2017
Publication date	Sep 10, 2019
Grant date	Sep 10, 2019

Abstract

A method of manufacturing a semiconductor device comprising: obtaining a raw light signal by selecting a predetermined wavelength band of light reflected from a wafer on which a plurality of patterns are formed; converting the raw light signal into a frequency domain; obtaining a first detection signal having a first frequency band from the raw light signal converted into the frequency domain; obtaining a second detection signal having a second frequency band from the raw light signal converted into the frequency domain, the second frequency band being different from the first frequency band; obtaining a representative value using the first detection signal, the representative value representing a profile of the plurality of patterns; and obtaining a distribution value using the second detection signal, the distribution value representing a profile of the plurality of patterns using the second detection signal. The method may include determining whether the representative value and the distribution value are within predetermined ranges respectively; and performing a following step of manufacturing the semiconductor device when the representative value and the distribution value are within respective predetermined ranges.

First claim

What is claimed is:

1. A method of manufacturing a semiconductor device, the method comprising:
obtaining a raw light signal by selecting a predetermined wavelength band of light reflected from a wafer on which a plurality of patterns are formed;
converting the raw light signal into a frequency domain;
obtaining a first detection signal having a first frequency band from the raw light signal converted into the frequency domain;
obtaining a second detection signal having a second frequency band from the raw light signal converted into the frequency domain, the second frequency band being different from the first frequency band;
obtaining a representative value using the first detection signal, the representative value representing a profile of the plurality of patterns;
obtaining a distribution value using the second detection signal, the distribution value representing a profile of the plurality of patterns using the second detection signal;
determining whether the representative value and the distribution value are within predetermined ranges respectively; and
performing a following step of manufacturing the semiconductor device when the representative value and the distribution value are within respective predetermined ranges.

2. The method of claim 1 , wherein the second frequency band is higher than the first frequency band, and
wherein the following step is a layer deposition step, a dicing step or a packaging step.

3. The method of claim 2 , further comprising:
filtering a third frequency band having higher frequencies than the second frequency band from the raw light signal converted into the frequency domain,
wherein the first detection signal and the second detection signal are obtained from the raw light signal converted into the frequency domain after the third frequency band is filtered.

4. The method of claim 1 , wherein the first detection signal and the second detection signal are obtained using at least one of a moving average filter and a fast Fourier transform (FFT) filter.

5. The method of claim 1 , wherein the first detection signal is converted into a wavelength domain to calculate the representative value.

6. The method of claim 5 , wherein a predetermined modeling technique is applied to the first detection signal converted into a wavelength domain to calculate the representative value.

7. The method of claim 1 , further comprising:
generating first simulation signals by selecting the predetermined wavelength band of light reflected from each of a plurality of sample regions including a plurality of sample patterns having different profiles;
obtaining second simulation signals by converting the first simulation signals into the frequency domain and selecting the second frequency band from the first simulation signals converted into the frequency domain; and
generating a matching function defining a relationship between eigenvalues of the second simulation signals and actual distribution values of dimensional parameters of sample patterns determining the profiles of the sample patterns of the plurality of sample regions.

8. The method of claim 7 , further comprising:
calculating an eigenvalue of the second detection signal; and
determining the distribution value of the plurality of patterns using the eigenvalue of the second detection signal and the matching function.

9. The method of claim 8 , further comprising:
determining an offset value and a slope from the matching function, the offset value and the slope defining a relationship between the eigenvalues of the second simulation signals and the actual distribution values of the sample patterns; and
calculating the distribution value of the plurality of patterns using the eigenvalue of the second detection signal, the offset value, and the slope.

10. The method of claim 1 , wherein the representative value is a median value or an average value of one of dimensional parameters of the plurality of patterns determining the profile, and the distribution value is a variance or a standard deviation of one of the dimensional parameters.

11. The method of claim 1 , wherein the raw light signal is obtained from at least one of a reflectance spectrum and a phase difference spectrum according to a wavelength of light reflected from the wafer.

12. The method of claim 1 , wherein the representative value is a median value or an average value of at least one of a width, a length, and a height of the plurality of patterns and an interval between the plurality of patterns, and
the distribution value is a variance or a standard deviation of at least one of the width, the length, and the height of the plurality of patterns and the interval between the plurality of patterns.

13. The method of claim 1 , wherein the raw light signal is generated from light reflected from the wafer using a spectroscopic ellipsometer (SE).

14. A method of manufacturing a semiconductor device, the method comprising:
obtaining a raw light signal defined in a frequency domain using light reflected from a measurement region including a plurality of patterns;
dividing the raw light signal into a first detection signal having a first frequency band, a second detection signal having a second frequency band, and a third detection signal having a third frequency band;
obtaining a representative value representing a profile of the plurality of patterns by converting the first detection signal into a wavelength domain; and
obtaining a distribution value representing a profile of the plurality of patterns by inserting a value of the second detection signal into a matching function provided in advance;
determining whether the representative value and the distribution value are within predetermined ranges respectively; and
performing a following step of manufacturing the semiconductor device when the representative value and the distribution value are within respective predetermined ranges.

15. The method of claim 14 , wherein the profile is determined by dimensional parameters including at least one of a width, a height, and a length of the plurality of patterns and an interval between the plurality of patterns.

16. The method of claim 14 , wherein the representative value is a median value or an average value of at least one of dimensional parameters of the plurality of patterns determining the profile, and the distribution value is a variance or a standard deviation of at least one of the dimensional parameters.

17. The method of claim 14 , wherein the matching function is a function defining a relationship between distribution values of dimensional parameters determining a profile of a plurality of sample patterns included in each of a plurality of sample regions and eigenvalues of a plurality of simulation signals obtained from the plurality of sample regions.

18. The method of claim 14 , wherein the first frequency band is lower than the second frequency band, and the second frequency band is lower than the third frequency band.

19. The method of claim 18 , wherein the third detection signal is determined to be a noise component.

20. A method of manufacturing a semiconductor device, the method comprising:
generating first simulation signals by selecting a predetermined wavelength band of light reflected from each of a plurality of sample regions, each of the plurality of sample regions including a plurality of sample patterns;
obtaining second simulation signals by converting the first simulation signals into a frequency domain and selecting a predetermined

Assignees

• Samsung Electronics Co Ltd

Inventors

• Park Jang Ik
• Kim Bong Seok
• Kim Souk
• Yang Yu Sin
• Lee Soo Seok

Classifications

• H10P74/203 Primary

  Structural properties, e.g. testing or measuring thicknesses, line widths, warpage, bond strengths or physical defects · CPC title

• H10P54/00

  Cutting or separating of wafers, substrates or parts of devices · CPC title

• H10W10/17

  formed using trench refilling with dielectric materials, e.g. shallow trench isolations · CPC title

• H10W10/014

  using trench refilling with dielectric materials, e.g. shallow trench isolations · CPC title

• H10P74/238 Primary

  comprising acting in response to an ongoing measurement without interruption of processing, e.g. endpoint detection or in-situ thickness measurement · CPC title