Computer-aided design and analysis of circuits and semiconductor – Nanotechnology related integrated circuit design
Reexamination Certificate
2001-10-30
2004-03-09
Smith, Matthew (Department: 2825)
Computer-aided design and analysis of circuits and semiconductor
Nanotechnology related integrated circuit design
C716S030000, C430S030000, C430S311000, C438S016000, C700S121000
Reexamination Certificate
active
06704920
ABSTRACT:
FIELD OF THE INVENTION
This invention is generally in the field of process control techniques, and relates to a method and system for controlling a process of manufacturing patterned structures, such as photolithography and etching processes.
BACKGROUND OF THE INVENTION
The currently common methods for process control in photolithography, particularly micro-lithography, are based on the use of CD-SEM. The latter is a stand-alone tool, which performs measurements of critical dimensions (minimal lateral dimensions of a pattern) for creating Statistical Process Control (SPC) trend charts for further monitoring thereof. One of these methods involves creating a so-called “Focus-Exposure Matrix” (FEM), produced by varying the focus and exposure (energy) parameters of the lithography from field to field within the wafer, thereby producing a two-dimensional array of fields spanning a range of these parameters. By determining CD in each of the FEM fields, optimal values of the focus and exposure, as well as their allowed tolerance (process window), are determined for each specific process.
Recently, tools based on scatterometry have been developed, which provide for higher accuracy and repeatability, faster measurement, smaller volume and lower cost, as compared to CD-SEM tools. Such scatterometry-based tools are disclosed, for example, in U.S. Pat. Nos. 5,867,276 and 5,963,329; and in the following publication: “
Specular Spectroscopic Scatterometry in DUV Lithography”,
Xinhui Niu et al, SPIE Vol. 3677, SPE Conference on Metrology, Inspection and Process Control for Microlithography XIII, pp. 159-168.
Scatterometry is a method by which the optical signature (spectral response) of a periodic structure is measured. The signature can be obtained by measuring the optical properties of a structure (reflectance or ellipsometric parameters) as a function of one or more light parameters, e.g., the angle of incidence, polarization or wavelength. Due to the periodicity of the structure, it is possible to theoretically calculate the signature of a given sample using exact models thereof (e.g., utilizing a Rigorous Couple Wave Theory (RCWT)). Processing is thus performed by correlating the measured signature to theoretically calculated signatures, while fitting the structure's parameters. This fitting method suffers from such drawbacks as long calculation time, in-adequacy to real-time calculations, and the need for detailed knowledge about the structure (e.g., optical constants) that is required as input to the model. The problem of long calculation time is usually overcome by preparing a library of pre-calculated signatures. This procedure, however, requires a long setup time. The detailed knowledge about the structure, in many cases, also requires preliminary setup processes, such as material characterization. Additionally, the measurement is limited to periodic structures that do not usually exist within the die, thus requiring fabricating special test structures and correlating the measurements on these test structures to measurements taken within the die. Yet another problem is the complicated, sometimes indirect relation between the process parameters (e.g., focus and exposure) and the profile parameters, rendering the attempt to control the process by modifying process parameters based on profile information, which is difficult to implement. These problems impede the application of scatterometry-based systems as a production tool, specifically for integrated monitoring that require a fast feedback for process control.
According to another technique, disclosed in the article “
Phi-Scatterometry for On-line Process Control”,
N. Benesch et al, AEC/APC Symposium XII, Lake Tahoe, Nev., USA, Sep. 23-28, 2000, the signatures measured in different fields of a Focus-Exposure Matrix are classified using a neural network (NN) under those found within the control limits and those found outside of them. In other words, this technique provides only “pass”/“fail” information which allows a Process Alarm to be operated. However, no quantitative information is provided, therefore feedback to the process (adjusting the working parameters of the processing tool in a closed loop control) is impossible.
SUMMARY OF THE INVENTION
There is accordingly a need in the art to facilitate the control of a process of manufacturing patterned structures, particularly micro-lithography, by providing a novel control method and system. The present invention introduces a methodology that starts with identifying those major process parameters whose variation affects the process results. This new methodology also directly exploits the dependence of the measured signature on the process parameters, without requiring any model having predictive capabilities with regard to the way this dependence is manifested.
The invention is particularly useful for controlling a lithography process, wherein focus and exposure are among the dominant factors affecting the lithographical profile (critical dimensions, wall angle, etc.). These parameters are usually considered in order to control the lithography process and keep the resulting profile within the required control limits. The new methodology bypasses the main limitations inherent in conventional scatterometry as presented above.
In the description below, the following terms as used:
The term “parametric matrix” or in short “matrix” used herein signifies a set of patterned structures (wafers) and/or fields created within the structure(s), that were fabricated using different values of one or more working parameters of the process to be controlled. Consequently, the term “matrix field” or in short “field” signifies one specific part of a parametric matrix, being a wafer or part of a wafer, having a specific value or set of values of the working parameters. All fields are supposed to include equivalent measurement sites, not necessarily in the same locations.
The term “measurement site” or in short “site” refers to a specific location found within each matrix field where the signature measurement is actually being taken.
The term “signature” signifies an optical response of the structure to predetermined incident light. Such a signature may be measured as a diffraction of light interacting with the structure as a function of a light parameter such as wavelength (spectrum), angle of incidence, ellipsometry, etc. The term “signature” refers to the total optical information that can be attained from a certain field, including several measurements taken at different measurement conditions, different measurement tools and/or at different measurement sites within the same field.
The term “reference tool results” signifies the results of one or more measurements applied to the parametric matrix or a part thereof by reference tools other than the measurement apparatus of the present invention.
The term “reference data” refers to all data available in order to perform the setup process (training of the NN), including mainly but not only signatures measured on a group of matrix fields and reference tool results from corresponding fields, as well as the processing conditions of the same field and any other sort of information available on these fields.
The term “control window” or “process window” signifies a range or ranges of one or more working (process) parameters providing desired process results.
The term “merit function” refers to a function that gets two signatures as input and results with a single number that is some measure of the “distance” between the two input signatures.
There is thus provided according to one aspect of the present invention, a method of controlling a process to be applied to a patterned structure in a production run, the method comprising the steps of:
(i) providing reference data including data representative of diffraction signatures corresponding to a group of different fields in a structure similar to said patterned structure in the production line, and data representative of a control window for the process parameters corresponding to a signature repres
Brill Boaz
Cohen Yoel
Browdy and Neimark P.L.L.C
Lin Sun James
Nova Measuring Instruments Ltd.
Smith Matthew
LandOfFree
Process control for micro-lithography does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process control for micro-lithography, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process control for micro-lithography will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3247297