Feature creation in substrate supports

1 . A method comprising:

receiving a ceramic plate of a substrate support system;

determining a feature pattern for a surface of the ceramic plate;

performing laser material processing of the surface of the ceramic plate to form the feature pattern comprising a plurality of features on the surface of the ceramic plate; and

performing additional laser material processing to refine, based on determined performance data of the substrate support system, the plurality of features of the feature pattern.

2 . The method of claim 1 , further comprising:

determining the performance data of the substrate support system.

3 . The method of claim 2 , further comprising:

determining laser material processing parameters for the additional laser material processing based on the performance data, the laser material processing parameters comprising at least one of average power, pulse energy, pulse duration, dwell time, repetition rate, hatch distance, hatch type, or rastering type.

4 . The method of claim 2 , further comprising:

processing the performance data of the substrate support system using a trained machine learning model that outputs at least one of an updated feature pattern or values for one or more laser material processing parameters to be used for performing the additional laser material processing.

5 . The method of claim 4 , wherein the trained machine learning model is trained using training input data comprising historical performance data and training target output data comprising historical design data associated with the historical performance data.

6 . The method of claim 1 , further comprising:

performing laser material processing to remove at least a portion of the plurality of features; and

forming a new feature pattern on the surface of the ceramic plate.

7 . The method of claim 1 , further comprising:

applying a solvent soluble film to the surface of the ceramic plate prior to performing the laser material processing.

8 . The method of claim 1 , wherein surfaces of one or more features of the plurality of features comprise a freeform polynomial shape or a linear approximation of a freeform polynomial shape.

9 . The method of claim 1 , wherein surfaces of regions between one or more features of the plurality of features comprise a freeform polynomial shape or a linear approximation of a freeform polynomial shape.

10 . The method of claim 1 , further comprising:

determining performance data of the substrate support system;

determining a target profile for a bottom surface of the ceramic plate or an additional ceramic plate of the substrate support system based on the determined performance data; and

performing laser material processing to remove material from the bottom surface of the ceramic plate or the additional ceramic plate based on the determined target profile.

11 . A method comprising:

receiving a ceramic plate of a substrate support system;

determining a feature pattern for a surface of the ceramic plate;

performing laser material processing of the surface of the ceramic plate to remove at least a portion of a plurality of features of the feature pattern on the surface of the ceramic plate;

performing the laser material processing of the surface of the ceramic plate to form a new feature pattern comprising a plurality of features on the surface of the ceramic plate; and

performing additional laser material processing to refine, based on determined performance data of the substrate support system, the plurality of features of the feature pattern.

12 . The method of claim 11 , further comprising:

determining the performance data of the substrate support system.

13 . The method of claim 12 , further comprising:

processing the performance data of the substrate support system using a trained machine learning model that outputs at least one of an updated feature pattern or values for one or more laser material processing parameters to be used for performing the additional laser material processing.

14 . The method of claim 13 , wherein the trained machine learning model is trained using training input data comprising historical performance data and training target output data comprising historical design data associated with the historical performance data.

15 . The method of claim 11 , further comprising:

determining performance data of the substrate support system;

determining a target profile for a bottom surface of the ceramic plate or an additional ceramic plate of the substrate support system based on the determined performance data; and

performing laser material processing to remove material from the bottom surface of the ceramic plate or the additional ceramic plate based on the determined target profile.