Determination of substrate layer thickness with polishing pad wear compensation
A method of training a neural network includes obtaining two ground truth thickness profiles a test substrate, obtaining two thickness profiles for the test substrate as measured by an in-situ monitoring system while the test substrate is on polishing pads of different thicknesses, generating an estimated thickness profile for another thickness value that is between the two thickness values by interpolating between the two profiles, and training a neural network using the estimated thickness profile.
1. A polishing system, comprising:
a platen to support a polishing pad;
a carrier head to hold a substrate and bring the substrate into contact with the polishing pad;
an in-situ monitoring system to generate a signal that depends on a thickness of a conductive layer on the substrate while the conductive layer is being polished by the polishing pad; and
a controller configured to
receive a measure of a pre-polish thickness of the conductive layer,
obtain an initial signal value from the in-situ monitoring system at a beginning of a polishing of the conductive layer,
determine an expected signal value for the conductive layer based on the pre-polish thickness,
calculate a gain based on the initial signal value and the expected signal value,
determine a polishing pad thickness value from the gain using a gain function,
receive signals from the in-situ monitoring system during polishing of the conductive layer to generate a plurality of measured signals for a plurality of different locations on the layer,
determine a plurality of thickness values for the plurality of different locations on the layer from the plurality of measured signals;
generate, for each location of at least some of the plurality of different locations, a corrected thickness value for the location to provide a plurality of corrected thickness values by processing at least some of the plurality of thickness values through a neural network, wherein the at least some of the plurality of thickness values and the polishing pad thickness value are input to the neural network and the corrected thickness values are output by the neural network, and
at least one of detecting a polishing endpoint or modifying a polishing parameter based on the plurality of corrected thickness values.
2. The system of claim 1 , wherein the neural network includes a plurality of input nodes, a plurality of output nodes, and a plurality of intermediate nodes, wherein the at least some of the plurality of thickness values are applied to at least some of the input nodes, wherein the polishing pad thickness value is applied directly to an intermediate node from the plurality of intermediate nodes, and wherein at least some of the plurality of output nodes output the plurality of corrected thickness values.
3. The system of claim 1 , wherein the in-situ monitoring system comprises an eddy current monitoring system.
4. The system of claim 1 , the system further comprising an in-line metrology system to perform a pre-polish measurement to provide the measure of the pre-polish thickness.
5. The system of claim 1 , wherein the gain function comprises a linear function.
6. The system of claim 1 , wherein the controller is configured to determine the plurality of thickness values by calculating a thickness using a function giving measured signal as a polynomial function of thickness.
7. The system of claim 1 , wherein the controller is configured to calculate the gain as a ratio of the initial signal value to the expected signal value.
8. A method of controlling polishing, comprising:
receiving a measure of a pre-polish thickness of a conductive layer on a substrate;
bringing the conductive layer on the substrate into contact with a polishing pad in a polishing system and commencing polishing;
obtaining an initial signal value from an in-situ monitoring system at a beginning of the polishing of the conductive layer;
determining an expected signal value for the conductive layer based on the pre-polish thickness;
calculating a gain based on the initial signal value and the expected signal value;
determining a polishing pad thickness value from the gain using a gain function;
receiving signals from the in-situ monitoring system during polishing of the conductive layer to generate a plurality of measured signals for a plurality of different locations on the layer;
determining a plurality of thickness values for the plurality of different locations on the layer from the plurality of measured signals;
generating, for each location of at least some of the plurality of different locations, a corrected thickness value for the location to provide a plurality of corrected thickness values by processing at least some of the plurality of thickness values through a neural network, wherein the at least some of the plurality of thickness values and the polishing pad thickness value are input to the neural network and the corrected thickness values are output by the neural network, and
at least one of detecting a polishing endpoint or modifying a polishing parameter based on the plurality of corrected thickness values.
9. The method of claim 8 , wherein receiving the measure of a pre-polish thickness includes measuring the pre-polish thickness at a stand-alone metrology station.
10. The method of claim 8 , wherein receiving the measure of a pre-polish thickness includes measuring the pre-polish thickness at an in-line metrology station of the polishing system.
11. The method of claim 8 , wherein the in-situ monitoring system comprises an eddy current monitoring system.
12. The system of claim 8 , wherein the gain function comprises a linear function.
13. The method of claim 8 , comprising determining the plurality of thickness values by calculating a thickness using a function giving measured signal as a polynomial function of thickness.
14. The method of claim 8 , comprising calculating the gain as a ratio of the initial signal value to the expected signal value.