Method for monitoring a semiconductor fabrication process for processing a substrate

• US 6,643,559 B2
• Filed: 02/12/2002
• Issued: 11/04/2003
• Est. Priority Date: 08/12/1999
• Status: Expired due to Fees

• Current Assignee: Infineon Technologies AG
• Original Assignee: Infineon Technologies AG
• Inventors: Bell, Ferdinand
• Primary Examiner(s): Picard, Leo
• Assistant Examiner(s): KOSOWSKI, ALEXANDER J
• Application Number:
  US10/074,823
• Publication Number:
  US 20020119668A1
• Time in Patent Office: 630 Days
• Field of Search: 700/121, 204/192.33, 204/192.13, 438/9, 438/16, 356/72, 356/316, 216/59, 216/60, 216/67
• US Class Current: 700/121
• CPC Class Codes: B24B 37/013 Devices or means for detecting lapping completion
• CPC Class Codes: H01J 37/32935 Monitoring and controlling tubes by information coming from the object and/or discharge
• CPC Class Codes: H01L 22/20 Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps
• CPC Class Codes: H01L 22/26 Acting in response to an ongoing measurement without interruption of processing, e.g. endpoint detection, in-situ thickness measurement endpoint detection arrangements in CMP apparatus B24B37/013, in discharge apparatus H01J37/32
• CPC Class Codes: H01L 2924/00 Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
• CPC Class Codes: H01L 2924/0002 Not covered by any one of groups H01L24/00, H01L24/00 and H01L2224/00

Abstract

The invention relates to a method for monitoring a production process, whereby several models are used for detecting a finish point. The results of the model are subsequently compared with one another and the best model therefrom is used in other production processes to detect a finish point. The inventive method provides the advantage that process changes resulting from chamber contaminations or sensor drift are compensated for by selecting the best model, thereby ensuring reliable finish point detection even in case of unfavorable process conditions.

Claims

1. A method for monitoring a plasma process, which comprises:

using a first model to determine an end point of a first plasma process that is performed in a plasma;

defining the first model with an algorithm, with a termination criterion, and with at least one predetermined measurement quantity that can be determined during the first plasma process and that is based on an intensity of at least one predetermined emission wavelength of the plasma;

configuring the algorithm such that when the algorithm is applied to the predetermined measurement quantity that is determined, the algorithm provides a decision quantity which, upon comparison with the termination criterion, serves for determining the end point of the first plasma process;

performing the first plasma process by using a plasma-excited gas in a plasma chamber, by introducing a substrate, which will be treated, into the plasma chamber, and by allowing the substrate to interact with the plasma-excited gas in the plasma chamber;

during the first plasma process, determining the predetermined measurement quantity for the first model to thereby obtain a measured quantity;

applying the algorithm of the first model to the measured quantity and determining the decision quantity;

comparing the decision quantity with the termination criterion prescribed by the first model and terminating the first plasma process when the termination criterion is met;

using a second model for comparatively determining the end point of the first plasma process;

defining the second model with an algorithm, with a termination criterion, and with at least one predetermined measurement quantity that can be determined during the first plasma process and that is based on an intensity of at least one predetermined emission wavelength of the plasma;

using an additional monitoring function for continuously assessing the first model and the second model;

if the end point that was determined with the second model has a higher significance than the end point that was determined with the first model, then using the second model to determine an end point of a second plasma process succeeding the first plasma process;

measuring intensities of a plurality of emission wavelengths of the plasma during the first plasma process;

using the intensities of the plurality of the emission wavelengths as measurement quantities and continuously storing the measurement quantities in a data processing system;

also using the data processing system for identifying the end point of the first plasma etching process;

performing the first plasma process as a plasma etching process and providing the plasma-excited gas as a dry etching gas that etches at least parts of the substrate;

providing the substrate with an insulating layer; and

etching contact holes using the plasma etching process.

2. The method according to claim 1, which comprises:

using the second model during the second plasma process;

performing the second plasma process in a plasma;

using a third model for comparatively determining the end point of the second plasma process;

defining the third model with an algorithm, with a termination criterion, and with at least one predetermined measurement quantity that can be determined during the second plasma process and that is based on an intensity of at least one predetermined emission wavelength of the plasma of the second plasma process; and

if the end point that was determined with the third model has a higher significance than the end point that was determined with the second model, then using the third model for determining an end point of third plasma processes.

3. The method according to claim 1, which comprises:

determining a significance of the end point that was determined with the first model and determining a significance of the end point that was determined with the second model by comparing a temporal development of the measurement quantity based on the predetermined emission wavelength of the first model with a temporal development of the measurement quantity based on the predetermined emission wavelength of the second model.

4. The method according to claim 1, which comprises:

determining a significance of the end point that was determined with the first model and determining a significance of the end point that was determined with the second model by comparing the decision quantity that was determined by the algorithm of the first model with a decision quantity that is determined by the algorithm of the second model.

5. The method according to claim 4, which comprises:

comparing the decision quantity that was determined by the algorithm of the first model with the termination criteria of the first model to obtain a first result;

comparing the decision quantity that was determined by the algorithm of the second model with the termination criteria of the second model to obtain a second result; and

determining the significance of the end point that was determined with the first model and determining the significance of the end point that was determined with the second model by comparing the first result with the second result.

6. The method according to claim 1, which comprises:

basing the measurement quantity of the first model and the measurement quantity of the second model on a common emission wavelength; and

determining a measure of contamination of the plasma chamber by comparing the measurement quantity of the first model and the measurement quantity of the second model.

7. The method according to claim 1, which comprises:

making the insulation layer from silicon oxide.

8. The method according to claim 1 which comprises:

etching the contact holes to have different depths in the insulation layer.

9. The method according to claim 1, which comprises:

using a rotating plasma during the first plasma process.

10. The method according to claim 9, which comprises:

using a rotating plasma during the second plasma process.

11. The method according to claim 1, which comprises:

obtaining the measured quantity by forming a mean value over a predetermined period of time.

12. The method according to claim 11, which comprises:

using a rotating plasma during the first plasma process; and

setting the predetermined period of time to correspond to at least one circulation period of the rotating plasma.

13. The method according to claim 1, which comprises:

using the algorithm of the first model to determine a feature selected from the group consisting of a position of a local maximum, a local gradient, and a point of inflection of a curve representing a temporal development of the measurement quantity of the first model to thereby yield the decision quantity; and

using the algorithm of the second model to determine a feature selected from the group consisting of a position of a local maximum, a local gradient, and a point of inflection of a curve representing a temporal development of the measurement quantity of the second model to thereby yield a decision quantity.

14. The method according to claim 1, wherein the first model differs from the second model at least in terms of a feature selected from the group consisting of the predetermined measurement quantity of the first model and the predetermined algorithm of the first model.

15. The method according to claim 1, which comprises:

using stored measurement quantities from preceding fabrication processes for determining a significance of the first model and of the second model.