摘要:
A method is provided for calibrating a model of a lithographic process that includes defining a parameter space of lithographic model parameters that are expected in an integrated circuit layout. The parameter space is defined according to bin values of a lithographic model parameter that span the range from a predetermined minimum and maximum value of the model parameter. The bin values may be incremented uniformly between the maximum and minimum parameter values, or may be distributed according to a weighting. The lithographic model is calibrated to an initial calibration test pattern. The resulting simulated calibration pattern is evaluated to determine whether the model parameter space is adequately populated. If the parameter space is over or under populated, the calibration pattern is modified until the calibration pattern test values adequately populate the parameter space, so that the final calibrated lithographic process model will more reliably predict images over the full range of image parameters.
摘要:
Setting final dimensions while protecting against the possibility of merging shapes is provided by performing a decomposition of tolerance bands onto a plurality of masks for use in a multi-exposure process. This allows the maximum process latitude between open and short failure mechanisms, while also providing a mechanism to enforce strict CD tolerances in critical regions of a circuit. The decomposition enables co-optimizing various types of shapes placed onto each mask along with the source used to print each mask. Once the tolerance bands are decomposed onto the two or more masks, standard tolerance-band-based data preparation methodologies can be employed to create the final mask shapes.
摘要:
A method of designing an integrated circuit is provided in which the design layout is optimized using a process model until the design constraints are satisfied by the image contours simulated by the process model. The process model used in the design phase need not be as accurate as the lithographic model used in preparing the lithographic mask layout during data prep. The resulting image contours are then included with the modified, optimized design layout to the data prep process, in which the mask layout is optimized using the lithographic process model, for example, including RET and OPC. The mask layout optimization matches the images simulated by the lithographic process model with the image contours generated during the design phase, which ensures that the design and manufacturability constraints specified by the designer are satisfied by the optimized mask layout.
摘要:
Setting final dimensions while protecting against the possibility of merging shapes is provided by performing a decomposition of tolerance bands onto a plurality of masks for use in a multi-exposure process. This allows the maximum process latitude between open and short failure mechanisms, while also providing a mechanism to enforce strict CD tolerances in critical regions of a circuit. The decomposition enables co-optimizing various types of shapes placed onto each mask along with the source used to print each mask. Once the tolerance bands are decomposed onto the two or more masks, standard tolerance-band-based data preparation methodologies can be employed to create the final mask shapes.
摘要:
A method is provided for calibrating a model of a lithographic process that includes defining a parameter space of lithographic model parameters that are expected in an integrated circuit layout. The parameter space is defined according to bin values of a lithographic model parameter that span the range from a predetermined minimum and maximum value of the model parameter. The bin values may be incremented uniformly between the maximum and minimum parameter values, or may be distributed according to a weighting. The lithographic model is calibrated to an initial calibration test pattern. The resulting simulated calibration pattern is evaluated to determine whether the model parameter space is adequately populated. If the parameter space is over or under populated, the calibration pattern is modified until the calibration pattern test values adequately populate the parameter space, so that the final calibrated lithographic process model will more reliably predict images over the full range of image parameters.
摘要:
A methodology to improve the through-process model calibration accuracy of a semiconductor manufacturing process using lithographic methods by setting the correct defocus and image plane position in a patterning process model build. Separations of the optical model and the photoresist model are employed by separating out the adverse effects of the exposure tool from the effects of the photoresist. The exposure tool is adjusted to compensate for the errors. The methodology includes a determination of where the simulator best focus location is in comparison to the empirically derived best focus location.
摘要:
Embodiments of the present invention provide a method for performing lumped-process model calibration. The method includes creating a plurality of sub-process models for a set of sub-processes; creating a lumped-process-model incorporating said set of sub-processes; calculating a first set of output patterns from a set of test patterns by using said plurality of sub-process models; calculating a second set of output patterns from said set of test patterns by using said lumped-process-model; and adjusting process parameters used in said lumped-process-model to calculate said second set of output patterns to match said first set of output patterns. A computer system for performing the lumped-process model calibration is also provided.
摘要:
A methodology to improve the through-process model calibration accuracy of a semiconductor manufacturing process using lithographic methods by setting the correct defocus and image plane position in a patterning process model build. Separations of the optical model and the photoresist model are employed by separating out the adverse effects of the exposure tool from the effects of the photoresist. The exposure tool is adjusted to compensate for the errors. The methodology includes a determination of where the simulator best focus location is in comparison to the empirically derived best focus location.
摘要:
Embodiments of the present invention provide a method for performing lumped-process model calibration. The method includes creating a plurality of sub-process models for a set of sub-processes; creating a lumped-process-model incorporating said set of sub-processes; calculating a first set of output patterns from a set of test patterns by using said plurality of sub-process models; calculating a second set of output patterns from said set of test patterns by using said lumped-process-model; and adjusting process parameters used in said lumped-process-model to calculate said second set of output patterns to match said first set of output patterns. A computer system for performing the lumped-process model calibration is also provided.
摘要:
A method for laying out process dummy cells in relationship to inside memory cells of a memory array includes (a) calculating an initial process performance parameter for the memory array; (b) changing dummy cell layout configuration for a layer electrically connected to inside cells; (c) applying lithographic simulation and yield model for both the inside memory cells and the changed layout configuration process dummy cells; and (d) repeating steps (b) and (c) until yield is maximized. Checks may be performed to ensure that there is enough room to make the change and that there is no significant adverse effect to neighboring circuits. The process performance parameter may be yield or a process window for the inside memory cells.