High-accuracy pattern shape evaluating method and apparatus
AssigneeHitachi High-Technologies Corporation
InventorsAtsuko Yamaguchi, Hiroshi Fukuda, Osamu Komuro, Hiroki Kawada
Life of this patent
Estimating the term…
Abstract
A quantity (or dispersion value) of a distribution of edge position due to random noise is expected to be reduced statistically to 1/N when N edge position data items are averaged. Using this property, the single page image is averaged in a vertical direction with various values of parameter S, and then the edge roughness index is calculated. The S-dependence of the edge roughness index is analyzed and a term of a dispersion value directly proportional to 1/S is determined as being due to noise.
Claims18
A pattern shape evaluating apparatus comprising:
- a scanning electron microscope detecting a secondary electron or reflected electron by irradiation of an electron beam to a specimen formed with line pattern, and retrieving a two-dimensional distribution data of the secondary electron signal or the reflected electron signal;
- and
- a data processing apparatus for processing the two-dimensional distribution data;
- wherein the data processing apparatus retrieves a plurality of edge points data of the line pattern from the two-dimensional distribution data;
- generates data set of the edge points with different averaging parameter that defines a range of averaging to longitudinal direction of the line pattern, by calculating an averaging process to the data set within a range defined by the averaging parameter;
- calculating a standard deviation of the roughness of the data set corresponding to each averaging parameter;
- and
- calculating a noise-caused bias of the roughness included based on a relationship between the standard deviation and the averaging parameter.
A critical dimension measurement SEM equipped with a function calculating a roughness, including:
- a scanning electron microscope for obtaining a data set of edge points used for the calculation of the roughness;
- a data processing apparatus that calculates a roughness;
- wherein said data processing apparatus has function comprised of:
- averaging the edge points within a range defined by a predetermined averaging parameter;
- calculating a standard deviation of the roughness using the data set of the averaged edge points;
- calculating a relationship between the standard deviation and the averaging parameter;
- and
- calculating a noise-caused bias in the roughness based on the relationship between the standard deviation and the averaging parameter.
Dependencies are inferred from the claim text.