Classifications

• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F30/00—Computer-aided design [CAD]
  • G06F30/30—Circuit design
  • G06F30/39—Circuit design at the physical level
  • G06F30/398—Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM]
• • G—PHYSICS
  • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
  • G03F—PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR
  • G03F1/00—Originals for photomechanical production of textured or patterned surfaces, e.g., masks, photo-masks, reticles; Mask blanks or pellicles therefor; Containers specially adapted therefor; Preparation thereof
  • G03F1/68—Preparation processes not covered by groups G03F1/20 - G03F1/50
  • G03F1/82—Auxiliary processes, e.g. cleaning or inspecting
  • G03F1/84—Inspecting
• • H—ELECTRICITY
  • H01—ELECTRIC ELEMENTS
  • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
  • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
  • H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
  • H01L21/027—Making masks on semiconductor bodies for further photolithographic processing not provided for in group H01L21/18 or H01L21/34
• • H—ELECTRICITY
  • H01—ELECTRIC ELEMENTS
  • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
  • H01L22/00—Testing or measuring during manufacture or treatment; Reliability measurements, i.e. testing of parts without further processing to modify the parts as such; Structural arrangements therefor
  • H01L22/20—Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps

Definitions

• the present invention refers to a method for generating high density registration maps for masks.
• the present invention refers to a system for generating high density registration maps for masks.
• the present invention refers to a computer program product disposed on a non-transitory computer readable medium.
• a mask (may also be referred to as a photomask or reticle) is a device that physically stores a pattern. The pattern is transferred to a wafer by lithography.
• Mask registration metrology and mask inspection have traditionally been decoupled from one another due to their inherently conflicting requirements.
• Mask registration is usually implemented using a stepping approach and involves positioning the reticle under the imaging optics for a period of time to image through focus steps.
• the position of the reticle is held to tight absolute accuracy bounds by regulating the temperature of the measurement chamber very tightly and using high precision displacement metrology.
• Such an approach while guaranteeing tight bounds on absolute accuracy does not lend itself to a high throughput thus limiting the number of points on the reticle that can be measured.
• US patent 8,582,1 13 discloses a device for determining the position of a structure on an object in relation to a coordinate system.
• the object is placed on a measuring table which is movable in one plane.
• At least one optical arrangement is provided which comprises an illumination apparatus for reflected light illumination and/or transmitted light illumination.
• 7,823,295 disclose devices or methods for determining the positions of structures on a mask.
• TDI Time Delay Integration
• Absolute position accuracy is less important during mask inspection since the primary motive is to detect and classify defects on the mask.
• Image swaths from a mask inspection are also divided up into sub-patches which are realigned algorithmically to remove low-frequency image shifts (such as those due to temperature fluctuations) further reducing absolute accuracy.
• Mask inspection systems are disclosed in US patent US 8,855,400, US patent application US 2014/0217298, US patent US 8,498,468 or US patent US 7,564,545 B2.
• a modular array can include a plurality of TDI sensor modules, each TDI sensor module having a TDI sensor and a plurality of localized circuits for driving and processing the TDI sensor.
• the plurality of TDI sensor modules can be positioned to capture a same inspection region or different inspection regions. Spacing of the sensor modules can be arranged to provide 100% coverage of the inspection region in one pass or for fractional coverage requiring two or more passes for complete coverage.
• the present systems or methods for mask registration metrology or mask inspection systems do not provide full mask registration map measurements. Metrology systems alone are not fast enough to cover the full mask.
• the method further comprises passing the information about the anchor point measurement including position and image render parameters from the registration tool to the inspection tool for improved accuracy.
• a system for generating high density registration maps for masks the system comprises:
• a data preparation software module which generates a plurality of anchor points, a plurality of sample points, a plurality of weights and at least one first recipe and at least one second recipe;
• a registration tool connected to the data preparation module to determine data for positions of the anchor points on the mask with regard to the at least one first recipe.
• the registration tool can provide additional data learned from the mask (e.g., image rendering model) to the inspection too (or data fusion module) for improved accuracy.
• the advantage of the inventive method and system is a higher density registration map of a reticle is obtained which in turn covers the increasing demands on both overlay and CD uniformity as feature sizes shrink. As a result, the entire mask is inspected to be within the mask registration error budget leading to no good masks get rejected and no bad masks get accepted.
• a graphical representation of the registration map of the mask is displayed on a display.
• the graphical representation shows the corrected set of registration points, wherein each registration point is provided with an error bar.
• the sample points, the anchor points and the weights are determined based on expected measurement error on both metrology and inspection tools.
• the generated number of anchor points is less than the generated number of sample points.
• approximately 10 3 anchor points are generated and/or approximately 10 6 sample points are generated.
• the generated sample points may be up to 10 8 or even larger.
• the sample points, measured by the inspection tool are cast over the entire mask by the data fusion module, according to the generated weights, into a mask coordinate frame as established by the registration tool to obtain the registration map of the mask.
• the previously determined weights are used to determine the influence of a specific anchor point on the adjacent sample points in the mask coordinate frame.
• bounds are established for potential errors between sample points according to a predetermined interpolation scheme.
• the predetermined interpolation is realized by using influence functions.
• a user can regrid the displayed registration map over the sample points over a different set of points.
• the different set of points is on a regularly spaced grid
• the data preparation module has at least a first input for providing mask design data in order to search for the appropriate anchor points as well as sample points.
• the design data for anchor and sample points are rendered in the registration tool and the inspection tool for position measurement.
• a second input of the data preparation module provides a noise model for the registration tool and the inspection tool.
• a first recipe module is connected to an anchor point output of the data preparation module and connected to an input of the registration tool.
• a second recipe module is connected to a sample point output of the data preparation software module and connected to an input of the inspection tool.
• the data fusion software module is configured to take the data of the measured positions of the anchor points via the output of the registration tool. Via the output of the inspection tool the data of the measured sample points are taken. A corrected set of registration points is generated along with the weights.
• a display is connected to the data fusion module for displaying bounded interpolation errors between anchor points over the entire mask.
• the number of anchor points is less than the number of sample points.
• a computer program product which is disposed on a non-transitory computer readable medium.
• the computer program product comprises computer executable process steps operable to control a computer to: obtain positions of a plurality of anchor points in a mask coordinate system measured by a registration tool according to a predetermined recipe for the registration tool; obtain positions of a plurality of sample points as well as anchor points in the mask coordinate system measured by an inspection tool according to a predetermined recipe for the inspection tool; and calculate a correction function for sample points from the weight of anchor points and measured positions of the anchor points in both metrology and inspection tools.
• the correction function is applied to the sample points to provide a corrected registration map for the full mask.
• the weights, the recipe for the registration tool and the recipe for the inspection tool are obtained from a data preparation software module.
• the data of the measured positions of the anchor points and the measured positions of the sample points are used to generate along with the weights a corrected set of registration points bounded interpolation errors between anchor points over the entire mask.
• the invention seeks to enable full mask registration map measurement. Metrology systems are not fast enough to cover the full mask. Inspection systems are not accurate enough for registration metrology.
• the invention proposes a way to combine both a metrology system and an inspection system in order to gain full mask registration mapping of masks.
• the key advantage of the invention is the ability of the customer to obtain densely populated registration maps without any additional inspection or registration overhead and using existing capital equipment.
• the only additional requirement is that of the data preparation module and the data fusion module.
• the pre-processing and postprocessing is realized with adequate software modules along with modifications to existing software of the registration tool and the inspection tool to enable the data gathering as required.
• a novel feature of the present invention is the creation of a high-density registration map using a combination of (a few) anchor points from a mask registration tool and a larger number of sample points from the mask inspection tool. Furthermore, a novel feature is the use of a data preparation module (pre-processor) allowing the determination of appropriate locations (positions) of anchor points and sample points and the weights for the influence functions of the anchor points to achieve maximum accuracy in the final dense registration map.
• the use of a data fusion module (post-processor) is new, which casts the sample points in the coordinate frame of the mask imparted by the registration tool.
• the algorithms are used to bound interpolation errors between anchor points, and thus the entire mask is new. This permits the decoupling of the selection of anchor points which might be dependent on the mask design and the output data which might be use-case dependent.
• High density registration maps of masks are becoming very important as the features (structures) on masks continue to shrink and requirements on wafer overlay become tighter.
• the registration of the masks with respect to one another affects both CD uniformity and overlay and hence is a key metric in ensuring adequate yields in a semiconductor fabrication.
• the emergence of multi-patterning has placed significant demands on mask overlay even within a single layer.
• the use of these high- density registration maps is multi-pronged.
• the invention allows a feedback to the mask writer.
• the acceptance or rejection as well as the qualification of a mask for the fabrication is enhanced. A feedforward of the mask to the scanner is possible.
• Figure 1 is a schematic view of a mask (reticle, photomask) with a plurality of patches;
• Figure 2 is a schematic enlarged view of a single patch with a plurality of randomly distributed anchor points
• Figure 3 is a schematic view of a mask with swaths defined by an inspection tool
• Figure 4 is a schematic view of the data preparation module with the inputs and the outputs;
• Figure 5 is a schematic setup of the inventive system for generating high density registration maps for masks
• Figure 6 is a sparse registration map of a mask with the error vectors with the X- coordinate component and the Y-coordinate component determined by the registration tool of the system;
• Figure 7 is an image of a mask taken by the inspection tool with a plurality of
• Figure 8 is a possible influence function, showing the weight an anchor point has on adjacent sample points
• Figure 9 is a further possible influence function, showing the weight an anchor point has on adjacent sample points.
• Figure 10 is a graphical representation of a corrected dense set of registration points error vectors on a mask.
• FIG. 1 shows a schematic representation of a mask 2 which has a plurality of patches 3 formed thereon, which encompass the structures (not shown) to be imaged on a wafer (not shown). The patches 3 are arrange on the mask in the x-coordinate X direction and the y-coordinate y direction on the mask 2
• FIG. 2 is a schematic enlarged view of a single patch 3, wherein a plurality of anchor points 5 is defined within the patch 3.
• the random distribution of the anchor points 5 shown here should not be regarded as a limitation of the invention. It is clear for a skilled person that the anchor points 5 can be arranged as well on a uniformly spaced grid in the x-coordinate X direction and the y-coordinate y direction on the mask 2.
• the anchor points can consist of specially designed targets, or on-device pattern or an arbitrary mix of them.
• Figure 3 is a schematic representation of a mask 2 with the plurality of patches 3 placed thereon.
• the inspection by the inspection tool 30 is carried with a scanning approach using a TDI sensor (not shown) for example. Absolute accuracy is less important during mask inspection, since the primary motive is to detect and classify defects on the mask 2.
• the scanning approach provides image swaths 6 from the mask 2 which also can be divided up into sub-patches which are realigned algorithmically to remove low-frequency image shifts (such as those due to temperature fluctuations). Temperature fluctuations would further reduce absolute accuracy.
• a care area 7 on the mask 2 defines the area in which the high density registration map for a mask 2 is generated. The care area 7 could be as well the entire surface of the mask 2.
• FIG 4 is a detailed view of a data preparation module 10 which is used in the inventive system 100 as shown in Figure 5.
• the data preparation module 10 is a preprocessing module that essentially generates the required recipes for mask inspection as well as mask registration. These recipes are in a form suitable for the integration with a first recipe generation module for 32 for the inspection tool 30 and a second recipe generation module 22 for the registration tool 20.
• the data preparation module 10 also generates weights 17 suitable for use in a data fusion module 40, which is used for further enhancement of the result quality/uncertainty.
• the data preparation module 10 also has at least a first inputl 1 and at least a second input 12.
• the data preparation module 10 receives a data base rendered image of the mask via the first inputl 1.
• the rendered image is for both the registration tool 20 and the inspection tool 30.
• the data preparation module 10 receives noise models for both the registration tool 20 and the inspection tool 30.
• the first input 1 1 and the second input 12 to the data preparation module 10 are used essentially as a constrained optimizer.
• the data preparation module 10 is connected with its first output 34 to the first recipe generation module 34.
• the second output 24 of the data preparation module 10 is connected to the second recipe generation module 22.
• the first recipe generation module 32 and the second recipe generation module 22 are considered as constrained optimizers.
• the second recipe generation module 22 generates anchor points 5 for the registration tool 20 and with the generated recipe carries out registration of the anchor points 5.
• the first recipe generation module 32 generates sample points (not shown) for the inspection tool 30. With the generated recipe the inspection tool 30 carries out the inspection of the sample points on the mask 2. The inspection is carried out according to the recipe determined by the second recipe generation module 22. It should be noted that neither the anchor points 5 nor the sample points need to be on uniformly spaced grids.
• the locations of these points as well as the weights 17 are determined by an evaluation of the expected measurement error on each of the metrology an inspection tools at those locations as well as consideration for overlay hotspots, etc. over a care area 7 of the mask 2.
• the registration tool 20 is connected to the data preparation module 10 via the second recipe generation module 22.
• the inspection tool 30 is connected to the data preparation module 10 via the first recipe generation module 32.
• the inspection tool 30 then does the inspection and generates the various patches 3 that are evaluated to determine relative positions of the sample points with respect to the sample points on the same/adjacent swaths 6. All these measurements are done on the coordinate system 8 of the mask 2 as determined by the inspection tool 30.
• the inspection tool 30 with augmented software can generate registration data as well as inspection data.
• the registration data from the inspection tool 30 is large in number (on the order of a million points per mask), but typically more limited in absolute accuracy than the registration tool 20.
• the registration tool 20 has an augmented software to generate registration data for the anchor points 5.
• the anchor points 5 are typically about a thousand in number, but the registration tool 20 can measure their location to the demanding accuracies required by the next few nodes of semiconductors.
• the mask coordinate frame is established by the registration tool 20. Simultaneously, bounds are also established for potential errors between sample points according to a predetermined interpolation scheme. The customer can then choose to regrid the registration map over the sample points over a different set of points (on a regularly spaced grid, for example).
• a data fusion module 40 is connected to the registration tool 20, the inspection tool 30 and the data preparation module 10.
• the weights 17 were previously determined by the data preparation module 10.
• the weights 17 are used as influence functions (see figure 8 and figure 9) to determine the influence of an anchor point 5 (registration of the anchor point 5 is determined by the registration tool 20) on the adjacent sample points (determined by the inspection tool 30).
• the anchor points 5 are shown with error bars 15, indicating the registration deviation in the x-coordinate direction X and in the y-coordinate direction Y.
• the image of the mask 2 with the error bars 15 is shown on a display 19, wherein colored areas 16 provide an indication that within the colored areas 16 with the same color the error bars 15 are mainly oriented within a predetermined direction range.
• Figure 7 is an image 37 of a mask 2 taken by the inspection tool with a plurality of swaths 6.
• the image 37 contains the sample points which are distributed over the entire mask 2.
• the data fusion module 40 receives a data output 26 of the registration tool 20 and a data output 36 of the inspection tool 30.
• the data from the inspection tool 30 contain the sample points over the entire mask 2. These sample points are then combined (or fused) in the data fusion module 40 with the positions of the anchor points 5, which are determined by registration tool 20 in the mask 2 coordinate system 8.
• the data preparation module 10 also generates weights 17 suitable for use in the data fusion module 40 used for further enhancement of the result quality/uncertainty. Two possible weights (influence functions) are shown in Figure 8 and Figure 9.
• the weights 17 illustrate the influence that an anchor pinot 5 has on adjacent sample points in the x-coordinate X direction and in the y-coordinate Y direction.
• the whole process of determining the locations or positions on the anchor points 5 and their weights 17 are implemented in the data preparation module 10 (see Figure 5).
• the data preparation module 10 takes into account the noise models of the registration tool 20 and the inspection tool 30 as well as the image pattern on the mask 2.
• Figure 10 is a graphical representation of a corrected set of registration points with error bars 15 on a mask 2.
• the data fusion module 40 carries out the post-processing, that takes the registration data from both the registration tool 20 (anchor points 5) and the inspection tool 30 (swaths 6) along with the weights 17 generated by the data preparation module 10.
• the anchor points 5 are placed on a regular grid.
• the output of the data fusion module 40 provides a corrected set of registration points 18, which are as well placed on a regular grid. These registration points 18 may (but are not limited to) essentially remain at the same locations as those generated by the inspection tool 30 and have the (weighted) corrections from the registration tool 20 applied to them.
• the data fusion module 40 Along with a registration map 50, the data fusion module 40 also generates error bars 15 for registration accuracy for the region between each registration point 18 and its neighbors, thus guaranteeing bounded accuracy figures of merit over the entire mask 2. It is believed that the method and system of the present disclosure and many of its attendant advantages will be understood by the foregoing description, and it will be apparent that various changes may be made in the form, construction and arrangement of the components without departing from the disclosed subject matter or without sacrificing all of its material advantages. The form described is merely explanatory. List of Reference Numerals:

Landscapes

• Engineering & Computer Science (AREA)
• Physics & Mathematics (AREA)
• Computer Hardware Design (AREA)
• General Physics & Mathematics (AREA)
• Theoretical Computer Science (AREA)
• Manufacturing & Machinery (AREA)
• Power Engineering (AREA)
• Microelectronics & Electronic Packaging (AREA)
• Geometry (AREA)
• General Engineering & Computer Science (AREA)
• Evolutionary Computation (AREA)
• Condensed Matter Physics & Semiconductors (AREA)
• Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
• Preparing Plates And Mask In Photomechanical Process (AREA)
• Image Processing (AREA)
• Exposure And Positioning Against Photoresist Photosensitive Materials (AREA)
• Testing Or Measuring Of Semiconductors Or The Like (AREA)

Priority Applications (4)

Applications Claiming Priority (2)

Related Child Applications (1)

Publications (1)

Family

ID=54241286

Family Applications (1)

Country Status (6)

Cited By (5)

Families Citing this family (1)

Citations (5)

Family Cites Families (14)

• 2015
  • 2015-04-02 KR KR1020167030599A patent/KR102330732B1/ko active Active
  • 2015-04-02 JP JP2016560399A patent/JP6570010B2/ja active Active
  • 2015-04-02 CN CN201580018393.1A patent/CN106165065B/zh active Active
  • 2015-04-02 TW TW104111017A patent/TWI640843B/zh active
  • 2015-04-02 WO PCT/US2015/024060 patent/WO2015153872A1/en not_active Ceased
  • 2015-07-09 US US14/795,576 patent/US20150310160A1/en not_active Abandoned

Patent Citations (5)

Also Published As

Similar Documents

Legal Events