Plasma processing apparatus
Provided is a plasma processing apparatus including a processing unit in which a sample is plasma processed and which includes a monitor (optical emission spectroscopy) that monitors light emission of plasma, wherein the processing unit includes a prediction model storage unit that stores a prediction model predicting a plasma processing result, and a control device in which the plasma processing result is predicted by using a prediction model selected based on light emission data and device data as an indicator of state change of the processing unit.
1. A plasma processing apparatus comprising:
a processing chamber configured to perform plasma processing of a sample;
a monitor that monitors light emission of plasma;
a memory storing a plurality of multivariate prediction models each predicting a plasma processing result; and
a controller configured to
select one of said plurality of multivariate prediction models having a highest similarity of partial correlation data and a monitor variable based on a partial correlation of an apparatus data, a partial correlation of light emission data of plasma, and a partial correlation of each of said plurality of multivariate prediction models, the apparatus data including data obtained from the processing chamber comprising a parameter specifying a plasma processing condition;
predict a processing result of plasma processing of a subsequent one or more samples using the selected multivariate prediction model;
adjust a condition of said plasma processing based on said prediction that uses said selected multivariate prediction model; and
perform said plasma processing using the adjusted condition.
2. The plasma processing apparatus according to claim 1 ,
wherein the plasma processing apparatus further comprises a prediction analyzer configured to construct at least one of said plurality of multivariate prediction model in which a variable of monitoring data having a partial correlation with the plasma processing result is specified as an explanatory variable.
3. The plasma processing apparatus according to claim 1 , wherein
the controller is configured to correct a value of said parameter, according to a correction amount obtained based on a difference between a predictive value obtained by using the prediction model and a target value.
4. The plasma processing apparatus according to claim 1 , wherein the apparatus data includes the number of samples which are plasma processed after wet cleaning.
5. The plasma processing apparatus according to claim 1 , further comprising:
a prediction analyzer configured to construct the prediction model in a which a variable of monitoring data having a partial correlation with the plasma processing result is specified as an explanatory variable.
6. The plasma processing apparatus according to claim 2 , wherein
the explanatory variable includes a wavelength of light emission of plasma and a physical quantity related to the apparatus data, and
the apparatus data is an indicator of state change of the processing system.
7. The plasma processing apparatus according to claim 2 , wherein
the partial correlation is a partial correlation between the plasma processing result and the variable of the monitoring data, for each lot that is a unit of the plasma processing.
8. A plasma processing apparatus comprising:
a processing chamber configured to perform plasma processing of a sample;
a monitor that monitors light emission of plasma;
a memory storing a plurality of multivariate prediction models each predicting a plasma processing result; and
a controller configured to
select one of said plurality of multivariate prediction models having a highest similarity of partial correlation data and a monitor variable based on statistics of an apparatus data, statistics of light emission data, and statistics of said one selected multivariate prediction model, the apparatus data including data obtained from the processing chamber comprising a parameter specifying a plasma processing condition;
predict a processing result of plasma processing of a subsequent one or more samples using the selected multivariate prediction model;
adjust a condition of said plasma processing based on said prediction that uses said selected multivariate prediction model; and
perform said plasma processing using the adjusted condition.