Back to results
US 12,504,696 B2Granted
Method for monitoring process variation index
AssigneeSamsung Electronics Co., Ltd.
InventorsDahan Kang, Youngseok Kim
Life of this patent
Estimating the term…
Abstract
A method for monitoring a process variation index includes operations of: obtaining a target parameter to be monitored and a reference parameter used to increase goodness of fit among structural parameters predicted by measuring a structure in a specific location of a wafer; obtaining a reference parameter set in a reference model; and calculating a process variation index capable of confirming a structural change of the structure according to a change in process conditions using the structural parameter and the reference parameter.
Claims18
A method for monitoring a process variation index, the method comprising operations of:
- performing a first process on a wafer to manufacture a semiconductor device;
- obtaining measurement data regarding a structure in a specific location of the wafer which is obtained after performing of the first process;
- predicting structural parameters by simulating the measurement data using a simulation tool, wherein the structural parameters are predicted by measuring the structure in the specific location of the wafer;
- obtaining a target parameter to be monitored;
- obtaining a reference parameter used to increase goodness of fit from among the structural parameters, wherein the reference parameter is set in a reference model;
- calculating the process variation index, which is capable of confirming a structural change of the structure according to a change in process conditions, using the target parameter and the reference parameter,
- wherein the process variation index satisfies the following equation:
- PCI
- =
- (
- 1
- -
- MP
-
- 1
- -
- NP
-
- 1
- R
-
- A
-
- 1
- )
- ·
- (
- 1
- -
- MP
-
- 2
- -
- NP
-
- 2
- R
-
- A
-
- 2
- )
-
- …
-
- SF
- ·
- RL
- wherein MP1 and MP2 denote predicted structural parameters, NP1 and NP2 denote reference parameters, which are preset in the reference model and correspond to the predicted structural parameters, respectively, RA1 and RA2 denote allowable ranges preset in the reference model and correspond to the predicted structural parameters, respectively, SF denotes a scale factor to make a magnitude of goodness of fit (GOF) within a certain range, RL denotes whether a respective one of the predicted structural parameters deviates from an allowable range and a range limit indicating whether a structural parameter deviates from the allowable range;
- obtaining process variation indices corresponding to a plurality of structures at a plurality of locations on each of a plurality of wafers;
- calculating an average value by averaging the process variation indices;
- and
- depending on the average value, performing a second process to manufacture the semiconductor device, or modifying the reference model.
50 more elements
A method for monitoring a process variation index, the method comprising operations of:
- performing a first process on a wafer to manufacture a semiconductor device;
- obtaining first measurement data, which is obtained from a structure sampled in a specific location of the wafer;
- obtaining a structural parameter, which is obtained as a result of a structural analysis of the structure;
- training a machine learning model using the first measurement data and the structural parameter as training data;
- predicting a structural parameter from second measurement data, which is obtained from the structure using the trained machine learning model;
- obtaining a structural parameter, which is set in a reference model, as a reference parameter;
- and
- calculating the process variation index capable of confirming a structural change of the structure using the predicted structural parameter and the reference parameter,
- wherein the process variation index satisfies the following equation:
- PCI
- =
- (
- 1
- -
- M
-
- P
-
- 1
- -
- N
-
- P
-
- 1
- R
-
- A
-
- 1
- )
- ·
- (
- 1
- -
- M
-
- P
-
- 2
- -
- N
-
- P
-
- 2
- R
-
- A
-
- 2
- )
-
- …
-
- SF
- ·
- RL
- wherein MP1 and MP2 denote predicted structural parameters, NP1 and NP2 denote reference parameters, which are preset in the reference model and correspond to the predicted structural parameters, respectively, RA1 and RA2 denote allowable ranges preset in the reference model and correspond to the predicted structural parameters, respectively, SF denotes a scale factor to make a magnitude of goodness of fit (GOF) within a certain range, RL denotes whether a respective one of the predicted structural parameters deviates from an allowable range and a range limit indicating whether a structural parameter deviates from the allowable range, and
- wherein the operation of training the machine learning model includes operations of:
- obtaining a plurality of process variation indices using a plurality of output data corresponding to a plurality of structures at a plurality of locations on each of a plurality of wafers;
- and
- calculating an average value by averaging the plurality of process variation indices;
- and
- depending on the average value, performing a second process to manufacture the semiconductor device, or modifying the reference model.
62 more elements
A method for monitoring a process variation index, the method comprising operations of:
- performing a first process on a wafer to fabricate a semiconductor device;
- obtaining a structural parameter, which is predicted by measuring a structure in a specific location of the wafer;
- obtaining a reference parameter, which is set in a reference model;
- obtaining an allowable range, which is set in the reference model;
- and
- calculating the process variation index capable of confirming a structural change of the structure according to a change in process conditions using the structural parameter, the reference parameter, and the allowable range,
- wherein the process variation index satisfies the following equation:
- PCI
- =
- (
- 1
- -
- M
-
- P
-
- 1
- -
- N
-
- P
-
- 1
- R
-
- A
-
- 1
- )
- ·
- (
- 1
- -
- M
-
- P
-
- 2
- -
- N
-
- P
-
- 2
- R
-
- A
-
- 2
- )
-
- …
-
- SF
- ·
- RL
- wherein MP1 and MP2 denote predicted structural parameters, NP1 and NP2 denote reference parameters, which are preset in the reference model and correspond to the predicted structural parameters, respectively, RA1 and RA2 denote allowable ranges preset in the reference model and correspond to the predicted structural parameters, respectively, SF denotes a scale factor to make a magnitude of goodness of fit (GOF) within a certain range, RL denotes whether a respective one of the predicted structural parameters deviates from an allowable range and a range limit indicating whether a structural parameter deviates from the allowable range,
- wherein the allowable range is a numerical range of the structural parameter, and is preset in the reference model, and
- wherein the reference model is a model formed based on parameters obtained as a result of a structural analysis of the structure sampled in the specific location of the wafer;
- obtaining process variation indices corresponding to a plurality of structures at a plurality of locations on each of a plurality of wafers;
- calculating an average value by averaging the process variation indices;
- depending on the average value, performing a second process to manufacture the semiconductor device, or modifying the reference model.
59 more elements
Dependencies are inferred from the claim text.