Method and apparatus for classifying defects occurring at or near a surface of a smooth substrate

A method for classifying defects occurring at or near a surface of a smooth substrate, comprising:

• (a) defining a plurality M of different idealized types of defects that can occur at the surface of the substrate, such that a given defect occurring at the surface can be described at least approximately by a size parameter d and by one of said M idealized types;
• (b) providing a database containing a plurality N of different sets of data for each said idealized type of defect, each set of data comprising a magnitude S of a signal from a light collector versus size parameter, the N different sets for each defect type corresponding to N different predetermined test configurations in which a beam of light having predetermined characteristics is impinged on the surface of the substrate at a predetermined incident angle and light emanating from the surface is collected by the light collector at a predetermined location above the surface and the intensity of the collected light is measured, the different predetermined test configurations being the same for each idealized defect type;
• (c) testing the substrate with each of said N different predetermined test configurations so as to derive N signal magnitudes S1 to SN for a defect to be sized and classified;
• (d) for each idealized defect type, using the database to determine at least N sizes d1 to dN corresponding to the N signal magnitudes S1 to SN, calculating an average size <d>, and using the database to determine signal magnitudes <S1> to <SN> that correspond to the average size <d>;
• (e) for each idealized defect type, calculating a deviation parameter a representing a combined deviation between the signal magnitudes S1 to SN and the determined signal magnitudes <S1> to <SN>, so as to derive M deviation parameters σ1 to σM corresponding to the M idealized defect types;
• and
• (f) classifying the defect as being of the idealized type having the smallest deviation parameter σ, and determining a size of the defect based on the average size <d> corresponding to said idealized type.

A method for determining the material of which a particle occurring on a surface of a smooth substrate is made and for determining a size of the particle, comprising:

• (a) defining a plurality of different materials of which the particle could be made;
• (b) providing a database containing a group of N different sets of data for each said material, each set of data comprising a magnitude S of a signal from a light collector versus a particle diameter d, the N different sets for each material corresponding to N different predetermined test configurations in which a beam of light having predetermined characteristics is impinged on the surface of the substrate at a predetermined incident angle and light emanating from the surface is collected by the light collector at a predetermined location above the surface and the intensity of the collected light is measured, the different predetermined test configurations being the same for each material;
• (c) testing the substrate with each of said N different predetermined test configurations so as to measure N signal magnitudes S1 to SN for a particle to be identified and sized;
• (d) comparing the measured signal magnitudes S1 to SN with the database and identifying the material of the particle based on the group whose data most closely matches the measured signal magnitudes S1 to SN;
• and
• (e) determining the size of the particle based on the group identified in step (d).

An apparatus for determining the material of which a particle occurring on a surface of a smooth substrate is made and for determining a size of the particle, comprising:

• a light source operable to create a light beam and impinge the light beam on the surface of the substrate with a predetermined incident angle relative to the surface and with a predetermined wavelength and predetermined polarization;
• a light collector system disposed above the surface of the substrate and operable to collect, at one or more locations, light emanating from a particle on the surface as a result of the impingement of the light beam thereon and further operable to create a signal for each said location having a magnitude indicative of intensity of the collected light, the light collector system defining a separate channel for carrying the signal associated with each said location;
• means for changing a test configuration of the apparatus so as to define a plurality N of different predetermined test configurations that, when tested, yield N measured signal magnitudes S1 to SN for a particle to be identified and sized;
• and
• a storage medium storing a database containing a group of N different sets of data for each of a plurality M of predetermined materials of which the particle could be made, each set of data comprising a magnitude S of a signal versus a particle diameter d, the N different sets for each material corresponding to the N different predetermined test configurations.

An apparatus for classifying defects occurring at or near a surface of a smooth substrate, comprising:

• a light source operable to create a light beam and impinge the light beam on the surface of the substrate with a predetermined incident angle relative to the surface and with a predetermined wavelength and predetermined polarization;
• a light collector system disposed above the surface of the substrate and operable to collect, at one or more locations, light emanating from a defect on the surface as a result of the impingement of the light beam thereon and further operable to create a signal having a magnitude indicative of intensity of the collected light for each said location, the light collector system defining a separate channel for carrying the signal associated with each said location;
• means for changing a test configuration of the apparatus so as to define a plurality N of different predetermined test configurations that, when tested, yield N measured signal magnitudes S1 to SN for a defect to be identified and sized;
• a storage medium storing a database containing a group of N different sets of data for each of a plurality M of predetermined idealized defect types that can occur at the surface of the substrate, each set of data comprising a magnitude S of a signal versus a defect size parameter d, the N different sets for each idealized defect type corresponding to the N different predetermined test configurations;
• and
• a computer connected with the storage medium and light collector system and operable to access each group in the database corresponding to each idealized defect type and to determine, based on the data sets in each group, a plurality of sizes d1 to dN for each idealized defect type corresponding to the N measured signal magnitudes S1 to SN, calculate an average size <d> for each idealized defect type, and determine signal magnitudes <S1> to <SN> that correspond to the average size <d> for each idealized defect type.