System, method and apparatus for improving accuracy of RF transmission models for selected portions of an RF transmission path

Patent metadata

Field	Value
Publication number	US-9652567-B2
Application number	US-201414519081-A
Country	US
Kind code	B2
Filing date	Oct 20, 2014
Priority date	Oct 20, 2014
Publication date	May 16, 2017
Grant date	May 16, 2017

Abstract

Systems and methods for determining an RF transmission line model for an RF transmission system includes generating a baseline RF transmission line model characterizing the RF transmission system. A plasma RF voltage, RF current, RF power and/or a corresponding RF induced DC bias voltage is calculated from the baseline RF transmission line model. An end module including the electrostatic chuck, a plasma and an RF return path is added to the baseline RF transmission line model to create one or more revised RF transmission line models. A revised plasma RF voltage, a revised plasma RF current, a revised plasma RF power and/or a corresponding revised RF induced DC bias voltage is calculated from each of the revised baseline RF transmission line models. The revised RF transmission line models are scored to identify a best fitting revised RF transmission line model as a complete RF transmission line model.

First claim

The invention claimed is:

1. Method for determining an RF transmission line model for a complete RF transmission system comprising:
generating a baseline RF transmission line model characterizing the RF transmission system, the baseline RF transmission line model having a plurality of stages ending at an input to an electrostatic chuck;
calculating at least one of a plasma RF voltage (Vrf), and/or a plasma RF current (Irf) and/or a plasma RF power (Prf) from the baseline RF transmission line model based on an empirically measured RF induced DC bias voltage;
adding an end module to the baseline RF transmission line model to create at least one revised RF transmission line model, the end module including the electrostatic chuck, a plasma and an RF return path;
calculating at least one of a revised plasma RF voltage (V′rf) and/or a revised plasma RF current (I′rf) and/or a revised plasma RF power (P′rf) and/or a corresponding revised RF induced DC bias voltage from each of the at least one revised baseline RF transmission line models;
scoring each one of the at least one revised RF transmission line models to identify a best fitting revised RF transmission line model; and
recording the best fitting revised RF transmission line model as a complete RF transmission line model.

2. The method of claim 1 , wherein adding an end module to the baseline RF transmission line model includes:
selecting an initial range for each of equivalent serial and shunt RLC values representing the end module;
dividing the selected ranges into a selected number of subdivisions to identify test values corresponding to each subdivision for each serial and shunt RLC values;
testing each combination of the test values for each of the serial and shunt RLC values to identify a plurality of local minimums;
recording each of the plurality of local minimums; and
selecting additional ranges for each of equivalent serial and shunt RLC values representing the end module when a selected threshold number of local minimums is not met.

3. The method of claim 2 , wherein selecting the initial range for each of equivalent serial and shunt RLC values representing the end module includes selecting a range of equivalent serial and shunt RLC values including the equivalent serial and shunt RLC values in at least one other stage of the RF transmission system.

4. The method of claim 2 , wherein selecting the initial range for each of equivalent serial and shunt RLC values representing the end module includes selecting random ranges of values for each equivalent serial and shunt RLC values.

5. The method of claim 2 , wherein dividing the selected ranges into the selected number of subdivisions to identify test values corresponding to each subdivision for each serial and shunt RLC values includes dividing each of the selected ranges into an equal number of subdivisions for each of the ranges of the serial and shunt RLC values.

6. The method of claim 2 , wherein dividing the selected ranges into the selected number of subdivisions to identify test values corresponding to each subdivision for each serial and shunt RLC values includes dividing each of the selected ranges into non-equal numbers of subdivisions for each of the ranges of the serial and shunt RLC values.

7. The method of claim 2 , wherein selecting additional ranges for each of equivalent serial and shunt RLC values representing the end module includes selecting a multiple of the initial range for at least one of each of equivalent serial and shunt RLC values representing the end module.

8. The method of claim 2 , wherein selecting additional ranges for each of equivalent serial and shunt RLC values representing the end module includes selecting a larger range for at least one of each of equivalent serial and shunt RLC values representing the end module.

9. The method of claim 1 , wherein adding an end module to the baseline RF transmission line model includes:
selecting an initial random value for each of equivalent serial and shunt RLC values representing the end module;
testing each combination of the test values for each of the serial and shunt RLC values to identify a gradient;
selecting new test values for each of the serial and shunt RLC values corresponding to the gradient toward a local minimum;
recording the local minimum when a local minimum is identified; and
selecting additional random values for each of equivalent serial and shunt RLC values representing the end module when the threshold number of local minimums is not met.

10. The method of claim 9 , wherein testing each combination of the test values for each of the serial and shunt RLC values to identify the gradient includes for each one of the serial and shunt RLC values:
selecting one of the serial and shunt RLC values;
selecting a corresponding first gradient test value offset a selected positive increment from the selected one of the serial and shunt RLC values;
selecting a corresponding second gradient test value offset a selected negative increment from the selected one of the serial and shunt RLC values; and
calculating a gradient using the first gradient test value and the second gradient test value.

11. The method of claim 9 , wherein testing each combination of the test values for each of the serial and shunt RLC values to identify the gradient includes for each one of the serial and shunt RLC values:
selecting one of the serial and shunt RLC values;
selecting at least one of a corresponding first gradient test value offset a selected positive increment from the selected one of the serial and shunt RLC values and/or a corresponding second gradient test value offset a selected negative increment from the selected one of the serial and shunt RLC values; and
calculating a gradient using at least two of:
the first gradient test value and/or
the second gradient test value and/or
the selected one of the serial and shunt RLC values.

12. The method of claim 1 , wherein scoring the revised RF transmission line models to identify the best fitting revised RF transmission line model includes:
selecting at least one bias model to test each revised RF transmission line model;
testing each revised RF transmission line model with each of the selected at least one bias model;
storing test results for each of the selected at least one bias model for each of the revised RF transmission line model; and
scoring the stored test results according to accuracy of the revised plasma RF voltage (V′rf) and/or a revised plasma RF current (I′rf) and/or a revised plasma RF power (P′rf) and/or a corresponding revised RF induced DC bias voltage corresponding to each of the at least one revised baseline RF transmission line models.

13. The method of claim 1 , wherein calculating at least one of a plasma RF voltage (Vrf), and/or the plasma RF current (Irf) and/or the plasma RF power (Prf) and/or the corresponding RF induced DC bias voltage from the baseline RF transmission line model includes calculating the corresponding RF induced DC bias voltage from the baseline RF transmission line model; and
wherein calculating at least one of the revised plasma RF voltage (V′rf) and/or the revised plasma RF current (I′rf) and/or the revised plasma RF power (P′rf) and/or a corresponding revised RF induced DC bias voltage from each of the at least one revised baseline RF transmission line models includes calculating the corresponding revised RF induced DC bias voltage from each of the at least one revised baseline RF transmission line models.

14. The method of claim 13 , wherein scoring each one of the at least one revised RF transmission line model to identif

Assignees

• Lam Res Corp

Inventors

• Valcore Jr John C
• Povolny Henry

Classifications

• H05H1/46

  using applied electromagnetic fields, e.g. high frequency or microwave energy (H05H1/26 takes precedence) · CPC title

• H01J37/32926

  Software, data control or modelling · CPC title

• H01J37/32183

  Matching circuits · CPC title

• H01J37/32174 Primary

  Circuits specially adapted for controlling the RF discharge · CPC title

• G06F30/367

  Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods · CPC title