Automated root cause analysis for defect detection during fabrication processes of semiconductor structures

Patent metadata

Field	Value
Publication number	US-12045969-B2
Application number	US-202017034640-A
Country	US
Kind code	B2
Filing date	Sep 28, 2020
Priority date	Oct 1, 2019
Publication date	Jul 23, 2024
Grant date	Jul 23, 2024

Abstract

A method includes obtaining at least one 2-D image dataset of semiconductor structures formed on a wafer including one or more defects during a wafer run of a wafer using a predefined fabrication process. The method also includes determining, based on at least one machine-learning algorithm trained on prior knowledge of the fabrication process and based on the at least one 2-D image dataset, one or more process deviations of the wafer run from the predefined fabrication process as a root cause of the one or more defects. A 3-D image dataset may be determined as a hidden variable.

First claim

What is claimed is:

1. A method, comprising:
during a wafer run of a wafer using a predefined fabrication process, obtaining a 2-D image dataset of 3-D semiconductor structures formed on the wafer, the wafer comprising a defect of one of the 3-D semiconductor structures;
using a machine-learning algorithm to determine, based on the 2-D image dataset, a probability distribution of multiple 3-D image datasets imaging the defect, the probability distribution assigning probabilities to each of the multiple 3-D image datasets;
responsive to the probability distribution of the multiple 3-D image datasets comprising equal probabilities for two or more of the multiple 3-D image datasets due to ambiguity of a projection the 3-D semiconductor structures into an imaging plane of the 2-D image dataset, obtaining a further 2-D image dataset of the 3-D semiconductor structures to reduce the ambiguity, the 2-D image dataset and the further 2-D image dataset depicting the 3-D semiconductor structures using different poses; and
based on at least one of the 3-D image datasets, determining a process deviation of the wafer run from the predefined fabrication process as a root cause of the defect of the one of the 3-D semiconductor structures.

2. The method of claim 1 , wherein the machine-learning algorithm is trained on the prior knowledge of the predefined fabrication process.

3. The method of claim 1 , further comprising, prior to the wafer run and during a development phase of the predefined fabrication process, using invasive imaging to train the machine-learning algorithm, wherein the invasive imaging yields 3-D image datasets of the semiconductor structures.

4. A device, comprising:
control circuitry configured to:
during a wafer run of a wafer using a predefined fabrication process, obtain a 2-D image dataset of 3-D semiconductor structures formed on the wafer, the wafer comprising a defect of one of the 3-D semiconductor structures;
use a machine-learning algorithm to determine, based on the 2-D image dataset, a probability distribution of multiple 3-D image datasets imaging the defect, the probability distribution assigning probabilities to each of the multiple 3-D image datasets;
responsive to the probability distribution of the multiple 3-D image datasets comprising equal probabilities for two or more of the multiple 3-D image datasets due to ambiguity of a projection the 3-D semiconductor structures into an imaging plane of the 2-D image dataset, obtain a further 2-D image dataset of the 3-D semiconductor structures to reduce the ambiguity, the 2-D image dataset and the further 2-D image dataset depicting the 3-D semiconductor structures using different poses; and
based on at least one of the 3-D image data sets, determine a process deviation of the wafer run from the predefined fabrication process as a root cause of the defect of the one of the 3-D semiconductor structures.

5. A system, comprising:
control circuitry configured to:
during a wafer run of a wafer using a predefined fabrication process, obtain a 2-D image dataset of 3-D semiconductor structures formed on the wafer, the wafer comprising a defect of one of the 3-D semiconductor structures; and
use a machine-learning algorithm to determine, based on the 2-D image dataset, a probability distribution of multiple 3-D image datasets imaging the defect, the probability distribution assigning probabilities to each of the multiple 3-D image datasets;
responsive to the probability distribution of the multiple 3-D image datasets comprising equal probabilities for two or more of the multiple 3-D image datasets due to ambiguity of a projection the 3-D semiconductor structures into an imaging plane of the 2-D image dataset, obtain a further 2-D image dataset of the 3-D semiconductor structures to reduce the ambiguity, the 2-D image dataset and the further 2-D image dataset depicting the 3-D semiconductor structures using different poses; and
based on at least one of the 3-D image datasets, determine a process deviation of the wafer run from the predefined fabrication process as a root cause of the defect of the one of the 3-D semiconductor structures; and

production equipment configured to implement the predefined fabrication process.

6. The system of claim 5 , further comprising an image source.

7. The system of claim 5 , further comprising at least one microscope selected from the group consisting of an optical microscope and a scanning electron microscope.

8. One or more machine-readable hardware storage devices comprising instructions that are executable by one or more processing devices to perform operations comprising the method of claim 1 .

9. A system comprising:
one or more processing devices; and
one or more machine-readable hardware storage devices comprising instructions that are executable by the one or more processing devices to perform operations comprising the method of claim 1 .

10. The system of claim 5 , further comprising:
an image source; and
an interface,
wherein the image source is configured to provide the 2-D image dataset to the interface, and the interface is configured to provide the 2-D image dataset to the control circuitry.

11. The system of claim 5 , further comprising:
an optical microscope; and
an interface to the interface,
wherein the optical microscope is configured to provide the 2-D image dataset to the interface, and the interface is configured to provide the 2-D image dataset to the control circuitry.

12. The system of claim 11 , further comprising:
a scanning electron microscope; and
an interface to the interface,
wherein the scanning electron microscope is configured to provide the 2-D image dataset to the interface, and the interface is configured to provide the 2-D image dataset to the control circuitry.

13. The system of claim 5 , further comprising:
a scanning electron microscope; and
an interface to the interface,
wherein the scanning electron microscope is configured to provide the 2-D image dataset to the interface, and the interface is configured to provide the 2-D image dataset to the control circuitry.

14. The method of claim 1 , further comprising using an image source to collect the 2-D image dataset.

15. The method of claim 14 , further comprising using production equipment to implement the predefined fabrication process.

16. The method of claim 1 , further comprising using an optical microscope to collect the 2-D image dataset.

17. The method of claim 16 , further comprising using a scanning electron microscope to collect the 2-D image dataset.

18. The method of claim 17 , further comprising using production equipment to implement the predefined fabrication process.

19. The method of claim 1 , further comprising using a scanning electron microscope to collect the 2-D image dataset.

20. The method of claim 1 , further comprising using production equipment to implement the predefined fabrication process.

Assignees

• Zeiss Carl Smt Gmbh

Inventors

• Neumann Jens Timo
• Foca Eugen
• Pichumani Ramani
• Srikantha Abhilash
• Wojek Christian
• Korb Thomas
• Correa Joaquin

Classifications

• H10P74/238

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

• G06T2207/20081

  Training; Learning · CPC title

• G06T2207/30148

  Semiconductor; IC; Wafer · CPC title

• G06T2207/10061

  from scanning electron microscope · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title