Method and system for processing three-dimensional measurement data from a scanning probe microscope system, and computer program product.

Abstract

The present invention relates to a method for processing three-dimensional measurement data from a scanning probe microscopy system to provide three-dimensional topographic image data, as well as a method for performing scanning probe microscopy, a scanning probe microscopy system, a computer-implemented method, and a computer program product. The method comprises receiving 3D measurement data acquired by the SPM system into a processing device, the 3D measurement data typically mapping the topography of a structure provided by the substrate. The method further comprises digitally constructing model image data using shape contour data of the shape, and a regression operation to fit the model image data to the 3D measurement data. This results in a composite model image of the 3D measurement data that can be used as a topographic image.

Description

Technical Field

[0001] The present invention is directed to a method for processing three-dimensional measurement data from a scanning probe microscope system to provide three-dimensional topography image data, as well as a method for performing scanning probe microscopy, a scanning probe microscope system, a method implemented on a computer, and a computer program product.

Background Art

[0002] Scanning probe microscopy (SPM) forms a set of techniques that utilize various interactions between a probe tip and a surface to image on-surface and / or sub-surface topography with nanometer-level resolution and accuracy. SPM tools are applied during process steps, for example, to provide 3D images of semiconductor devices on a wafer in order to enable process control and yield management in high-volume manufacturing (HVM) of semiconductor logic and memory chips. For these applications, it is important to achieve a high throughput with a satisfactory level of accuracy in order to enable the use of this technology for this purpose. Generally, by achieving a high throughput, the SPM technology can provide added value in more application fields, technical areas, and industrial processes.

[0003] The raw output of a scanning probe microscope (SPM) system typically includes three-dimensional (3D) measurement data. This can be on-surface data related to the topography of the sample's surface structure, or sub-surface three-dimensional measurement data related to the topography of structures and patterns beneath the surface. For example, sub-surface measurements allow for verification of overlays and alignments in semiconductor devices during and after manufacturing, enabling the manufacturing process to achieve a satisfactory rate of correctly manufactured and functioning semiconductor devices while minimizing losses. In the SPM imaging process, one of the current bottlenecks from the perspective of achieving high throughput is the processing of raw 3D measurement data before providing a topographic image. Data processing involves many individual steps, from noise reduction and artifact identification and removal to preparing the results for subsequent processes in which the results will be used (e.g., for metrological purposes). [Overview of the project] [Problems that the invention aims to solve]

[0004] Therefore, further process improvements are needed to increase efficiency from a data processing perspective. [Means for solving the problem]

[0005] Some aspects of this disclosure relate to methods for processing three-dimensional measurement data from a scanning probe microscope system in order to provide three-dimensional topographic image data. Other or further aspects relate to scanning probe microscope systems, and therefore, still other or further aspects relate to computer-implemented methods, computer program products, for processing three-dimensional measurement data from a scanning probe microscope system.

[0006] Typically, a scanning probe microscope system includes a substrate carrier (sometimes referred to as a sample) for supporting the substrate. The system may also include a probe comprising a cantilever and a probe tip for manipulating the surface of the substrate. By scanning the substrate surface in this manner, the system in use acquires three-dimensional measurement data. For example, in some embodiments, when the probe tip contacts (or attempts to contact) the substrate surface, the precise position in three orthogonal dimensions in a coordinate system (e.g., the X, Y, and Z axes in Cartesian coordinates; the r, θ, and φ axes in spherical coordinates; or the r, θ, and z axes in cylindrical coordinates) may be registered. In other or further embodiments, three-dimensional measurement data may be acquired by applying an acoustic signal to either or both the probe tip and the substrate, and recording the phase and / or amplitude of a return signal indicating subsurface topography. The acquisition of three-dimensional measurement data by scanning the surface of the substrate with a probe tip may be carried out in various ways, and in this respect, the present invention is not limited to a specific measurement mode.

[0007] Typically, a method for processing three-dimensional measurement data described herein includes receiving the three-dimensional measurement data obtained by a scanning probe microscope system by scanning the surface of the substrate into a processing device. This three-dimensional measurement data typically maps the topography of one or more structures provided by the substrate. In a very general form, this may include the detected 3D coordinates of the precise position of the probe tip when in contact with the surface, but other forms of 3D topographic data may be received in this step, as described. The method further includes digitally constructing model image data by the processing device using shape contour data of one or more shapes. Here, the model image data is constructed by either or both of the contour data and the three-dimensional measurement data to mimic the topography of one or more structures provided by the substrate. The matching does not necessarily need to be accurate at this stage, but those skilled in the art will understand that a better match between the model image data and the three-dimensional measurement data can facilitate subsequent steps in the process. Subsequently, a regression operation is iteratively performed to fit the model image data to the three-dimensional measurement data. This results in a composite model image of the three-dimensional topography of the substrate from the fitted model image data. According to the method of the present invention, the composite model image is provided as three-dimensional topographic image data of the substrate's topography in the output (e.g., via the output). This makes it possible, for example, according to some embodiments, to perform measurement on the topography of one or more structures.

[0008] According to the present invention, the three-dimensional measurement data is modeled with an equivalent composite image having floating parameters. For example, in some embodiments, the digital construction is performed by reconstructing the received three-dimensional measurement data from building blocks. These building blocks are provided by shapes digitally determined by their contour data, e.g., standard three-dimensional shapes. The shapes or parts thereof are used to reconstruct the three-dimensional measurement data as accurately as possible. This may include, in some further or different embodiments, convenient combinations, unitings, or subtractions of shapes to reconstruct, for example, more complex three-dimensional shapes. Also, in some embodiments, the shapes are used only partially, for example, by a suitable truncation. In principle, various types of operations may be applied to the shapes during the digital construction process. For example, in some embodiments, arranging or resizing the contour data includes at least one of the following: translating, rotating, inflating, deflating, flattening, stretching, compacting, truncating, or resizing the contour data of one or more shapes; or combining, merging, uniting, or subtracting the contour data of at least two of the shapes.

[0009] Subsequently, iterative regression processing allows the model image data to be fitted to the three-dimensional measurement data to generate, for example, a composite model image of the three-dimensional topography of the substrate. The main advantage of this method is that it can be performed far more efficiently than conventional data processing methods. Moreover, this method provides log data of the operation, from which it becomes clear which parts of the image are related to artifacts and which parts are actually related to the topography of the substrate structure. For example, in other embodiments or further embodiments, the method further includes fitting the model image data to include measurement artifact data, where the measurement artifact data represents image modification resulting from obtaining the three-dimensional measurement data by the scanning probe microscope system. For example, some of these measurement artifacts can be included in relatively simple forms, such as noise, spikes, tilts, background slopes, or tip-sample interaction artifacts.

[0010] With regard to the latter, if the shape of the probe tip (or a good estimate thereof) and the direction of movement of the probe tip are provided as input to the processing device, it is quite possible to modify the model image data to correct the chip shape artifact. This relates, for example, to the fact that in actual SPM images, sharp edges are typically not drawn as sharp edges, but rather as a convolution of the probe tip shape and edge shape in the direction of movement of the probe tip relative to the sample. By making the various modifications suggested above, not only is a better fitting obtained between the three-dimensional measurement data and the composite model image, but shape characteristics related to the added artifacts and those that are not actual features of the measured topography can be immediately identified. This additional data is obtained incidentally in this process. In some of these embodiments, the measurement artifact data includes at least one element from the group that includes noise data for adding noise to the composite model image, probe tip shape modification data for correcting the composite model image for artifacts in the three-dimensional measurement data caused by the shape of the probe tip, and system data for correcting the composite model image for system characteristics, such as parameter offset, measurement limits, measurement range, or filter effects.

[0011] In other or further embodiments, during or after the digital construction, the model image data may be adapted to match one or more image characteristics of the three-dimensional measurement data, such as the number of pixels or the pixel size. As can be understood, while constructing the composite model image, the pixel resolution and / or pixel size may be set to any desired value. However, these values ​​are set to be equivalent to the corresponding values ​​of the three-dimensional measurement data, thereby sampling according to the characteristics of the measurement. In this case, each pixel of the composite model image corresponds to a single measurement point of the probe tip.

[0012] In other or further embodiments, the topography of one or more structures provided by the substrate includes on-surface topography of the structures on the surface of the substrate; or sub-surface topography of the structures beneath the surface of the substrate; or both. As previously stated, this method is not limited to the application of surface topography measurements but can be applied to various types of SPM measurements. In some specific embodiments, the topography of one or more structures provided by the substrate includes both surface topography and sub-surface topography, where the sub-surface topography is projected onto the on-surface topography in the composite model image. This allows for a direct combination of on-surface and sub-surface measurements to perform, for example, overlay and alignment detection.

[0013] In some embodiments, the contour data for one or more shapes is obtained from at least one of a shape database, a data file, or user input. However, in other embodiments or further embodiments, the contour data may be generated by a processor or obtained in a different manner, for example, by analysis of a designed topography. In some other embodiments or further embodiments, the shape may be arbitrarily selected from the group including blocks, cubes, spheres, ovals, solid or hollow cylinders, pyramids (e.g., triangular pyramids, square pyramids, pentagonal pyramids, hexagonal pyramids, heptagonal pyramids, octagonal pyramids, or other polygonal pyramids), rods, cuboids, cones, hemispheres, prisms (e.g., triangular prisms, square prisms, pentagonal prisms, hexagonal prisms, heptagonal prisms, octagonal prisms, or other polygonal prisms), octahedrons, rhombuses, stellated prisms, tetrahedrons, dodecahedrons, ellipsoids, icosahedrons, parallelepipeds, zonal polyhedra, or other three-dimensional shapes.

[0014] In other or further embodiments, the contour data for each of the one or more shapes is characterized by one or more floating parameters, where each floating parameter of an individual shape has a parameter value that defines an individual dimension of one or more dimensions of the individual shape; where, during the digital construction, the processing device sets the parameter value of one or more of the floating parameters to an initial value; and during the iterative execution of the regression operation, the parameter value of the one or more floating parameters is modified to perform the fitting. The selection and adjustment of the parameter values ​​of the floating parameters makes it easy to change the shape characteristics and size of various shapes used for constructing the model image data. The regression can be performed using any suitable regression algorithm, for example, by applying the sum of distances between the measured data and the model image data.

[0015] In a further aspect of the present invention, a method is provided for performing scanning probe microscopy on the surface of a substrate using a scanning probe microscopy system, the scanning probe microscopy system comprising a substrate carrier for supporting the substrate and a probe comprising a cantilever and a probe tip for scanning the surface of the substrate to acquire three-dimensional measurement data of the topography of the one or more structures provided by the substrate, wherein the method comprises processing the three-dimensional measurement data to provide a composite model image including three-dimensional topographic image data as an output, wherein the processing is performed using a method according to any one or more of the claims. In other or further embodiments, such a method may further include performing measurement measurements on the topography of the one or more structures, at least one portion of which measurement measurements are performed on the composite model image.

[0016] More specifically, a scanning probe microscope system is provided for providing three-dimensional topographic image data of one or more structures provided by a substrate, wherein the scanning probe microscope system comprises a substrate carrier for supporting the substrate and a probe comprising a cantilever and a probe tip for scanning the surface of the substrate to acquire the three-dimensional measurement data, the system further comprising a processing device and a memory, the memory being suitable for storing instructions that, when executed by the processing device, cause to perform at least one of the methods described herein.

[0017] In another view, a computer-implemented method for processing three-dimensional measurement data from a scanning probe microscope system in order to provide three-dimensional topographic image data, wherein the scanning probe microscope system comprises a substrate carrier for supporting a substrate and a probe comprising a cantilever and a probe tip for scanning the surface of the substrate to acquire the three-dimensional measurement data, the method receiving the three-dimensional measurement data obtained by the scanning probe microscope system by scanning the surface of the substrate, wherein the three-dimensional measurement data maps the topography of one or more structures provided by the substrate; and the computer-implemented method for generating model image data using shape contour data of one or more shapes, the computer-implemented method The above method is provided, which includes digitally constructing, for example, the model image data by matching it with three-dimensional measurement data to mimic the topography of the one or more structures provided by the substrate, where the model image data is constructed by either or both arranging and / or resizing the contour data; iteratively performing regression operations by the processing device of the computer to fit the model image data to the three-dimensional measurement data to obtain a composite model image of the three-dimensional topography of the substrate from the fitted model image data; and providing the composite model image as three-dimensional topographic image data of the topography of the substrate in the output, so as to enable performing measurement measurements against the topography of the one or more structures.

[0018] In a further view, the computer program product is provided which includes instructions that cause a computer processing device to perform one of the methods described herein when the program is executed by the processing device of the computer.

[0019] The present invention will be further elucidated by the description of some specific embodiments while referring to the appended drawings. The detailed description herein provides examples of possible implementations of the present invention, but is not considered to describe the only embodiments falling within the scope. The scope of the present invention is defined by the claims, and this specification should be regarded as illustrative without limiting the present invention.

Brief Description of Drawings

[0020] [Figure 1] FIG. 1 schematically shows a scanning probe microscope apparatus according to an embodiment that can be used in the method of the present invention. [Figure 2] FIG. 2 schematically shows a scanning probe microscope apparatus according to an embodiment that can be used in the method of the present invention. [Figure 3] FIG. 3 schematically shows the representation of a three-dimensional image and a model image of a substrate surface obtained in an embodiment of the present invention. [Figure 4A] FIG. 4A schematically shows the digital construction of model image data according to one embodiment. [Figure 4B] FIG. 4B schematically shows the digital construction of model image data according to one embodiment. [Figure 4C] FIG. 4C schematically shows the digital construction of model image data according to one embodiment. [Figure 4D] FIG. 4D schematically shows the digital construction of model image data according to one embodiment. [Figure 5] FIG. 5 schematically shows a method according to an embodiment of the present invention.

Modes for Carrying Out the Invention

[0021] Figures 1 and 2 schematically illustrate scanning probe microscope (SPM) systems 1 and 1' according to embodiments of the present invention. Systems 1 and 1' are suitable for use in methods according to the present invention by comprising a processing device 26 and a memory 27 and / or 27' suitable for storing instructions, which, when executed by the processing device 26, cause the processing device 26 to perform the methods described herein according to one or more embodiments of the present invention. In Figures 1 and 2, the processing device 26 is shown as a single entity in each of the SPM systems 1 and 1'. However, those skilled in the art will understand that while all of the described method steps may be implemented using a single processing device 26, the processing device 26 may be implemented using multiple elements that jointly perform the described method steps. Thus, the processing device 26 may comprise a cluster of multiple processing devices, or it may be implemented by various entities that individually perform certain (partial) steps and implement in emphasis of the present invention. Furthermore, as illustrated, memory 27 may be internal memory 27, external memory 27', or another external entity reachable via data communication network 29. Systems 1 and 1' may include a communication unit 28 for communicating with the data communication network 29. In Figures 1 and 2, the data storage element 27, data processing element 26, and data communication element 28 are all shown as part of the analysis unit 25 of systems 1 and 1'. While this is often the implementation, it is not always the case (as already mentioned above). The illustration of a single entity 25 in Figures 1 and 2 is solely for the purpose of avoiding unnecessarily complicating the diagrams.

[0022] In the drawings, elements that are technically and functionally equivalent, i.e., elements that perform the same or similar functions in the same or similar manner with respect to the invention described herein, may be designated by the same reference numerals or by the same reference numerals with a prime symbol (') or a subnumber (such as "-1", "-2",...). These entities may have different natures. For example, the memories 27 in FIGS. 1 and 2 may each be the internal memory of the systems 1 and 1', but they may have different technical natures or may be implemented (e.g., connected or controlled) in different ways. On the other hand, from the perspective of the invention, they may each provide a function that enables the storage of data or operation instructions for the processing device 26. Moreover, the memory 27' in FIG. 2, although this element performs (or is capable of performing) the same function as or in a certain embodiment the same function as the internal memory 27 of the system 1', is different in that this memory 27' (which may be implemented as a server or as an externally stored data file or database) is an external memory rather than an internal memory, but does not deviate from its function in the embodiments of the present invention. The above is merely an example and is similarly applicable to other entities described below. In principle, unless otherwise specifically stated in any part of this specification, any entity or element described is assumed to be capable of being implemented in different ways in alternative embodiments.

[0023] In the systems 1 and 1' shown in Figures 1 and 2, the substrate carrier 3 is configured to support a substrate or sample 7 to be inspected by the SPM systems 1 and 1'. The substrate carrier 3 is configured to move the substrate 7 in a plane parallel to the carrier 3, by means of or connected to an actuator (not shown). This is typically referred to as the XY plane of the system. In addition to the ability to move the substrate 7 in the X and Y directions, the substrate carrier 3 is connected to a metrology frame 5 that provides a fixed base for the system. The SPM systems 1 and 1' further include a scanning head 15 that is movable in the Z direction. Each scanning head 15 includes a tip holder 16 capable of holding a probe tip comprising a probe 10 that forms the sensing element of the SPM systems 1 and 1'. The probe 10 comprises a cantilever 12 and a probe tip 11. The probe tip 11 typically has a very sharp tip that enables extremely accurate scanning and measurement (with nanometer precision) on the surface 8 of the substrate 7.

[0024] In addition to the above, the scanning head 15 typically further includes a sensor system for determining the precise position of the cantilever relative to the scanning head. For example, the sensing system in Figures 1 and 2 consists of an optical beam deflection (OBD) unit provided by a laser unit 20 for generating a light beam 23 and a four-quadrant optical sensor 21 for receiving the reflected beam 23. The beam 23 is directed by the optical system to the back of a probe tip 11 having a specular reflective surface. The specular reflective surface of the probe tip 11 reflects the beam 23 to the optical detector 21. Any displacement of the probe tip 11 results in a displacement of the illumination position of the beam 23 on the optical detector 21. In the four-quadrant optical detector 21, the light spot formed by the beam 23 is preferably set, by default, to be located exactly in the center of the detector's four different quadrants. Therefore, equal portions of the light spot illuminate each quadrant of the four-quadrant optical detector 21. It can be understood that even if the laser beam 23 does not actually align precisely in this manner, the principle of detecting the displacement of the light spot remains the same. All that is needed to detect the displacement is for each of the four quadrants of the photodetector 21 to receive a portion of the light from the beam 23.

[0025] When the probe tip 11 is displaced relative to the scanning head 15, which is equipped with a laser unit 20, the light spot formed on the optical detector 21 is slightly displaced such that the ratio between different regions illuminated by the light spot changes in each quadrant of the optical detector 21. This allows the precise position and / or orientation of the probe tip 11 relative to the scanning head 15 to be determined. Furthermore, since the Z position of the probe 10 applied by the Z actuator 18 is also known (from the control data of the Z actuator 18), the orientation and position of the probe tip 11 in the Z direction can be determined. Moreover, the XY position of the probe tip 11, which indicates its position relative to the sample 7 in the plane of the substrate carrier 3, is known from the actuator of the substrate carrier 3. In this manner, at each position in the XY plane, the precise orientation of the probe tip 11 in the Z direction can be determined in combination with the Z level applied via the Z actuator 18.

[0026] In the system 1 shown in Figure 1, the above information makes it possible to determine with great precision the exact height of the surface 8 at each point in the XY plane. Therefore, the surface topography 9 consisting of various structures on the surface 8 of the sample 7 can be accurately determined and mapped to provide a topography map. At each point in the XY plane parallel to the sample surface 8, the exact Z level of the surface 8 can be determined from data for controlling the Z level actuator 18 and data from the optical beam detector optical sensor 21. This data can be stored, for example, in the memory 27 of the system 1 as three-dimensional measurement data. In order to provide a topography map of the 3D topography of the surface 8, in addition to the above analysis of the data from the Z level actuator 18 and the optical detector 21, several other processing steps need to be performed. These include, for example, noise reduction and identification and removal of measurement artifacts.

[0027] Similarly, in the SPM system 1' of Figure 2, the position and orientation of the probe tip 11 are acquired and registered by combining the measurements at the OBT detector formed by the laser unit 20 and the optical detector 21 with Z-level data obtained from the actuator control data of the Z-level actuator 18. In particular, in the system 1' of Figure 2, subsurface measurements of the topography 9' of the structure 13 located below the surface 8 of the sample 7 are performed. For this purpose, the substrate carrier 3 further includes a vibration actuator that enables vibration 4 to be applied to the sample 7 from below. In an alternative SPM system, the vibrational signal 4 may be applied in a different manner, for example, by the vibration actuator on or to the side of the surface 8 of the sample 7. It is also possible to apply the vibrational signal via the probe 10 using a vibration transducer on the scanning head 15, or via the probe tip 11 by periodic output intensity fluctuations of the laser beam 23 provided by the laser unit 20. Such alternative methods for applying a vibrational signal 4 to the sample 7 are described in the literature and will not be further discussed herein. The present invention may be applied to measurements obtained from an SPM system, for example, SPM system 1 which performs surface topography measurements or SPM system 1' which performs subsurface measurements of topography 9'. The topography 9' is provided by a subsurface structure 13 which may be, for example, a preceding layer during the manufacturing of a semiconductor device. In this last example, it is typically important to be able to detect whether the superposition of subsequent layers in the semiconductor device is accurate enough to produce a functional semiconductor device, or whether the critical dimension is specified with the correct tolerance. Therefore, measurements of surface and subsurface structures are ideally applied, for example, in an industrial environment, during and after the manufacturing of a semiconductor device.It will be understood that being able to perform such measurements very quickly and accurately during the manufacturing process is particularly useful, as it allows them to be performed as frequently as needed. This ultimately improves the yield of operational and properly functioning semiconductor devices, thereby reducing the loss of faulty devices. It will be further understood that, in order to perform these measurements accurately in the dimensions of modern semiconductor devices, at least equivalent precision is required for imaging. Since optical methods are typically diffraction-limited, and the use of scanning electron microscopes (SEMs) is often undesirable due to their potential for destructiveness, scanning probe microscopy offers the optimal choice for performing this type of measurement during semiconductor device manufacturing.

[0028] In the system shown in Figure 2, a vibroacoustic input signal applied to the substrate 7 via the substrate carrier 3 is picked up by the surface 8 of the sample 7 via the probe tip 11. Due to the vibration, the Z level of the probe tip 11 is periodically displaced at the frequency applied via the acoustic signal 4. Various acoustic measurement techniques are known in the art, among others, for example, the heterodyne method, which uses an ultra-high frequency gigahertz acoustic signal encompassing two frequencies in the gigahertz band. The difference between the two applied frequencies in the gigahertz band is relatively small, typically in the megahertz band. By using the principle of heterodyne mixing of signals, a low-frequency signal at the difference frequency (in the megahertz band) can be picked up by the probe tip 11 and identified by analysis of the output signal obtained via the optical beam deflector unit 21. Again, this measurement data needs to be further analyzed to obtain three-dimensional topographic data from it. For example, amplitude and phase data may be acquired, and, as described above, identification of measurement artifacts and noise reduction should be performed.

[0029] As will be understood, in both the SPM system 1 in Figure 1 and the SPM system 1' in Figure 2 (incidentally, they may be integrated into a single SPM system capable of performing both whole-surface and subsurface topography measurements), data processing to acquire a three-dimensional image of the desired topography is relatively complex and requires processing of the large amount of data acquired from the system. For this reason, conventional SPM systems are relatively slow, and their application to industrial environments can be limited at best. These conventional systems are typically too slow to perform the desired measurements frequently enough to enable modifications to the manufacturing process or improve the yield of properly functioning semiconductor devices. In the present invention, this is a result of the method described herein, which can be implemented using either of the systems, for example, as shown in Figures 1 and 2.

[0030] Figure 3 shows an exemplary surface topography image obtained using an SPM system, e.g., SPM1. The surface topography image 30 shows an element 31 having a surface 32 with various holes 33 at regular positions. In the lower right corner of Figure 3, a model image 40 of the element 31 obtained using a method according to an embodiment of the present invention is shown. The model image 40 shows a model structure 41 having a surface 42 with similar holes 43. As is clear from Figure 3, the critical dimensions 46 (CDx, CDy) in the XY direction, the pitch 45 (pitch_x, pitch_y) in the XY direction, and the depth 47 of each hole 43 can be obtained directly from the model structure 41. Verification of these parameters and their tolerances is desirable, for example, in industrial processes where it is necessary to verify whether the overlay of an element is correct, it is sufficient to create a model image, e.g., model image 40, to perform this task. The main advantage is that it is easier to construct model image data 40 using a construction data processing algorithm than to perform all data processing steps to generate image 30 from the coarse measurement data that forms the basis of image 30. Moreover, as can be seen from the topography image 30, it is not so easy to obtain accurate measurements of the parameters 45, 46, and 47 mentioned above from image 30. For example, as is evident from the topography image 30, when considering critical dimensions 46 in the XY direction, the measurement error in these parameters is larger when these parameter values ​​are determined from the topography image 30 compared to when the parameter values ​​for the same critical dimension parameters 46 are determined based on the model image 40.

[0031] It might be pointed out that since the model image 40 is not real data but rather an estimate, the exact parameter values ​​cannot be determined if the parameter values ​​are obtained from the model image 40. However, the present invention is based on the insight that by digitally constructing the model image data and subsequently fitting this model to real data, the parameter values ​​determined in this manner more accurately reflect those actual values ​​on the sample 7 than the values ​​directly determined from the topography image 30. This is because the fitting process performed by regression analysis converges to the optimal match between the model image 40 and the topography image 30. In this optimal match, the differences between the images are minimized not just at one point but across the entire image. Therefore, by fitting the model image 40 to the topography image 30, the complete three-dimensional measurement data underlying the image 30 is used to provide the optimal match. For example, when the parameter values ​​for critical dimensions 46 are determined directly from the topography image 30, the estimates are typically based on, for example, multiple measurements, but not complete three-dimensional measurement data. Therefore, by fitting the model image 40 to the topography image 30, a very accurate estimate of the parameter values, for example, the parameter value of the critical dimension 46, is provided.

[0032] Figure 3 in the upper left corner also shows exemplary measurement artifacts 35 and 35'. The ring-shaped artifact 35 is, for example, not one of the actual structures present in the measured topography, but rather a result of pre-processing the measurement by the SPM. Depending on the application, it is possible to add such artifacts to the image 40 obtained from the construction algorithm according to some embodiments of the present invention. Noise can also be intentionally added to obtain a better match with the original three-dimensional measurement data.

[0033] Figures 4A to 4D schematically illustrate how the model image 40 is obtained in the digital construction algorithm. Figure 4A again schematically illustrates the model image 40 based on the model image data. In embodiments of the present invention, this model image data is obtained by constructing shape contour data of one or more known shapes in a database, for example, in memory 27. For example, the image 40 in Figure 4A can be obtained from various shapes (and many other potential shapes) shown in Figure 4B. Each of these shapes can be precisely described by its systematic characteristics. For example, the basic shape 50 is a square plate of a certain thickness. Each of the dimensions of this square plate 50, for example, not only its thickness but also its width and length, can be provided as floating parameters. Floating parameters are parameters that can be changed during the execution of the digital construction algorithm and during the fitting of the model image data in subsequent processes. For example, to obtain the image 40, initial values ​​or initial estimates of the dimensions of the square plate of the image 40 are provided or generated in the system and then improved during the execution of the digital construction algorithm and during the subsequent fitting. Other shapes shown in Figure 4B include a disc 51, a prism 52, a cube 53, or a three-dimensional cross 54. In fact, any three-dimensional shape can be used to model a particular structure or part thereof, for example, to obtain model image data. The examples in Figure 4B are merely some exemplary three-dimensional shapes. In memory 27 or 27', shape contour data for shapes 50-54 are stored, including their floating parameters and optionally, various other data. In the digital construction process applied in the present invention, the processing device 26 may select, for example, a square plate 50 as a basic shape to model a structure 31, and generate starting values ​​for its floating parameters (e.g., reflecting its typical size).Subsequently, as shown in Figure 4C, the contour data of the disk 51 is acquired, and for each of the holes 43 in image 40, the shape 51 is subtracted at a particularly defined position on the square-shaped plate 50. Similar to Figure 4C, the disk 51 may include, for example, floating parameters such as depth or height h and diameter D. In this case as well, the processing device 26 can generate their starting values ​​to subtract the shape from the shape contour data of the square-shaped plate 50. To accurately perform the subsequent fitting, the floating parameters of each shape and their relative positions may maintain the floating parameters of the complementary shape in Figures 4C and 4D. Finally, in Figure 4D, an initial rough estimate of the structure 60 with holes 62 is shown, forming model image data to be used as input to the fitting process.

[0034] Figure 5 schematically illustrates the method of the present invention, including the implementation of the fitting process. In step 80, three-dimensional measurement data 36 is acquired from the SPM system 1. The three-dimensional measurement data is acquired by scanning the surface 8 of such a shape 7, or by performing subsurface measurements of a subsurface structure as described herein, for example. In step 82, shape contour data of multiple shapes is acquired from memory, data files, databases, or other data repositories 27 or 27'. Using these, the digital construction algorithm shown in Figure 4 is executed to generate model image data 60 by appropriately combining and shaping the shape contour data. For example, to obtain any arbitrary complementary shape 60, parts of shapes 50-56 may be added to or subtracted from other shapes 50-56. Shapes may be moved relative to each other, rotated in various directions, appropriately truncated, enlarged, compressed, distorted, twisted, or subjected to any other suitable operation in order to digitally construct the desired model image data. This ultimately yields model image data 40, which is provided as input to step 85. In step 85, the model image data 40 is compared with the three-dimensional measurement data 36 obtained from system 1. To enable the comparison, the model image data is used to generate topographic data modeled at a similar resolution to the three-dimensional measurement data 36. The comparison in step 85 is performed by obtaining the difference between the two data files, and then in step 86, the difference between the model image data 40 and the three-dimensional measurement data 36 is reduced by appropriate shape adjustment and fitting.

[0035] For example, in step 85, both the three-dimensional measurement data 36 and the model image data 40 are received by the processing system, and in step 86, the model image data 40, particularly its floating parameters, are modified to obtain a better match between the model image data 40 and the three-dimensional measurement data 36. In step 86, the model image data may be rotated, stretched, enlarged, reduced, translated, or displaced, or certain individual floating parameters may be slightly modified to obtain a correct fit, for example. An idiomatic regression algorithm may be applied to do this, but it is also possible to train a self-learning data processing model (AI model) to do this job. In step 88, it is determined whether the processing in step 86 has improved or worsened the match between the three-dimensional image data 36 and the model image data 40. The algorithm in fitting step 84 continues until the difference is sufficiently small, or until the minimum difference between the model image data 40 and the three-dimensional measurement data 36 is obtained (for example, if the difference has not decreased after subsequent steps). Next, the model image data is provided in the output 90 of the method as a composite model image 70 of the three-dimensional shape of the substrate 7. The composite model image 70 generates a great deal of information about the modeled structure. For example, if each of the floating parameters is now defined and it is known exactly which parts of the image are related to measurement artifacts and to what extent this effect is present, then the critical dimensions or the pitch between the holes 43 can also be precisely determined. Noise in the image can be easily removed by switching it on or off, and the composite model image 70 of the surface topography of sample 7 makes it possible to determine and monitor manufacturing parameters and overlays very effectively during the manufacturing of semiconductor devices.

[0036] Terms used to describe specific embodiments are not intended to limit the invention. Where used herein, the singular forms “a,” “an,” and “the” are intended to also include the plural unless the context clearly indicates otherwise. The word “and / or” includes any and all combinations of one or more of the associated enumerated items. The words “comprises” and “comprising” will be understood to identify the presence of a described feature but not to exclude the presence or addition of one or more other features. Where a particular step of a method is referred to as following another step, it will be further understood that, unless otherwise specified, it may follow the other step directly or one or more intermediate steps may be performed before performing the particular step. Similarly, where a connection between structures or components is described, it will be understood that, unless otherwise specified, this connection may be established directly or through an intermediate structure or component.

[0037] The present invention is described more fully below this specification with reference to the accompanying drawings illustrating embodiments of the invention. In the drawings, absolute and relative sizes of systems, components, layers, and areas may be exaggerated for clarity. Embodiments may be described with reference to schematic and / or cross-sectional views of idealized embodiments and intermediate structures of the invention, in some cases. In this specification and drawings, similar numbers refer to similar elements throughout. Relative terms and their derivatives should be interpreted as referring to orientations as described therein or as shown in the drawings discussed. These relative terms are for convenience of the specification and do not require that the system be constructed or operated in a particular direction unless otherwise specified.

[0038] The present invention is described in terms of several specific embodiments. It will be understood that the embodiments shown in the drawings and described herein are for illustrative purposes only and are not intended to limit the invention in any way or by any means. The operation and configuration of the invention are considered to be evident from the above description and the accompanying drawings. It will be apparent to those skilled in the art that the invention is not limited to the embodiments described herein and that modifications to be considered within the scope of the appended claims are possible. Furthermore, the kinematic inverse transformation is considered to be essentially disclosed and within the scope of the invention. Moreover, any components and elements of the various embodiments disclosed may be combined with or incorporated into other embodiments without departing from the scope of the invention as defined in the claims, where deemed necessary, desirable, or preferred.

[0039] In the claims, no reference numeral should be construed as limiting the claims. When used in this specification or the appended claims, the words “includes” and “equipped” should be construed as comprehensive, not exclusive or exhaustive. Accordingly, when used in this specification, the expression “includes” does not exclude the existence of other elements or processes in addition to any element or process described in any claim. When used in this specification or the appended claims, the expression “consistes of” should be construed as comprehensive, “consistes of at least” rather than exhaustive enumeration. Furthermore, the words “one” ("a" and "an") should not be construed as limiting to “only one,” but rather as meaning “at least one,” and not excluding plural. Features not specifically or expressly described or claimed may be included additionally to the configuration of the present invention to that extent. Unless otherwise stated, the claimed or disclosed devices or parts thereof may be combined with each other or divided into further parts without departing from the claimed invention. The phrase "means for..." should be interpreted as "components configured for..." or "members constructed in...", and should be understood to include equivalents for the disclosed structures. The use of expressions such as "essential," "preferred," and "particularly preferred" is not intended to limit the invention. Additions, deletions, and modifications within the scope of those skilled in the art can generally be made without departing from the spirit and scope of the invention as defined by the claims. The invention may be carried out in ways other than those specifically described herein and is limited only by the appended claims.

Claims

1. A method for processing three-dimensional measurement data from a scanning probe microscope system in order to provide three-dimensional topographic image data, wherein the scanning probe microscope system comprises a substrate carrier for supporting a substrate and a probe comprising a cantilever and a probe tip for scanning the surface of the substrate to acquire the three-dimensional measurement data, and the method is The three-dimensional measurement data obtained by the scanning probe microscope system by scanning the surface of the substrate is received by a processing device, wherein the three-dimensional measurement data maps the topography of one or more structures provided on the substrate. A model image data is digitally constructed by the processing device using shape contour data of one or more shapes, wherein the model image data is constructed by either arranging or resizing the contour data, or both, and is matched with three-dimensional measurement data to mimic the topography of the one or more structures provided by the substrate. A regression operation is repeatedly performed by the processing device to fit the model image data to the three-dimensional measurement data, thereby obtaining a composite model image of the three-dimensional topography of the substrate from the fitted model image data; and, To enable measurement of the topography of the one or more structures, the composite model image is provided as output as three-dimensional topographic image data of the substrate's topography. The method, including the method described above.

2. The method according to claim 1, wherein arranging or resizing the contour data includes translating, rotating, expanding, contracting, flattening, stretching, compressing, truncating, or resizing the contour data of one or more shapes; or combining, merging, combining, or subtracting at least two of the contour data of the shapes.

3. The method according to claim 1 or 2, wherein the contour data of the one or more shapes is obtained from at least one of a shape database, a data file, or user input.

4. The method according to any one of claims 1 to 3, further comprising fitting the model image data to include measurement artifact data, wherein the measurement artifact data indicates image correction resulting from obtaining the three-dimensional measurement data by the scanning probe microscope system.

5. The method according to claim 4, wherein the measurement artifact data includes at least one element from the group including noise data for adding noise to the composite model image, probe tip shape correction data for correcting the composite model image for artifacts in the three-dimensional measurement data caused by the shape of the probe tip, and system data for correcting the composite model image for system characteristics, such as parameter offset, measurement limit, measurement range, or filter effect.

6. The method according to any one of claims 1 to 5, wherein during or after the digital construction, the model image data is adapted to match one or more image characteristics of the three-dimensional measurement data, such as the number of pixels or the pixel size.

7. The method according to any one of claims 1 to 6, wherein the topography of the one or more structures provided by the substrate includes surface topography of the structures on the surface of the substrate, subsurface topography of the structures below the surface of the substrate, or both.

8. The method according to claim 7, wherein the topography of the one or more structures provided by the substrate includes both surface topography and subsurface topography, and in the composite model image, the subsurface topography is projected onto the surface topography.

9. The method according to any one of claims 1 to 8, wherein the one or more shapes are elements selected from the group including blocks; cubes; spheres; ovals; solid or hollow cylinders; pyramids, e.g., triangular pyramids, square pyramids, pentagonal pyramids, hexagonal pyramids, heptagonal pyramids, octagonal pyramids or other polygonal pyramids; rods; cuboids; cones; hemispheres; prisms, e.g., triangular prisms, square prisms, pentagonal prisms, hexagonal prisms, heptagonal prisms, octagonal prisms or other polygonal prisms; octahedrons; rhombuses; stellated prisms; tetrahedrons; dodecahedrons; ellipsoids; icosahedrons; parallelepipeds; zonal polyhedra, or other three-dimensional shapes.

10. The method according to any one of claims 1 to 9, wherein the contour data for each of the one or more shapes is characterized by one or more floating parameters, where each floating parameter of the individual shape has a parameter value that defines an individual dimension of one or more dimensions of the individual shape; wherein, during the digital construction, the processing device sets the parameter value of one or more of the floating parameters to an initial value; and during the iterative execution of the regression operation, the parameter value of one or more of the floating parameters is modified to perform the fitting.

11. A method for performing scanning probe microscopy on the surface of a substrate using a scanning probe microscopy system, wherein the scanning probe microscopy system comprises a substrate carrier for supporting the substrate and a probe comprising a cantilever and a probe tip for scanning the surface of the substrate to acquire three-dimensional measurement data of the topography of the one or more structures provided by the substrate, wherein the method comprises processing the three-dimensional measurement data to provide a composite model image including three-dimensional topographic image data as an output, the processing being performed using the method according to any one of claims 1 to 10.

12. The method according to claim 11, further comprising performing measurements on the topography of the one or more structures, wherein at least one portion of the measurements is performed on the composite model image.

13. A scanning probe microscope system for providing three-dimensional topographic image data of one or more structures provided by a substrate, wherein the scanning probe microscope system comprises a substrate carrier for supporting the substrate and a probe comprising a cantilever and a probe tip for scanning the surface of the substrate to acquire the three-dimensional measurement data, the system further comprising a processing device and a memory, the memory being suitable for storing instructions causing the processing device to perform at least one of the methods according to claim 1 or claim 11 when executed by the processing device.

14. A computer-implemented method for processing three-dimensional measurement data from a scanning probe microscope system in order to provide three-dimensional topographic image data, wherein the scanning probe microscope system comprises a substrate carrier for supporting a substrate and a probe comprising a cantilever and a probe tip for scanning the surface of the substrate to acquire the three-dimensional measurement data, and the method is The three-dimensional measurement data obtained by the scanning probe microscope system by scanning the surface of the substrate is received by the computer processing device, wherein the three-dimensional measurement data maps the topography of the one or more structures provided on the substrate; Model image data is digitally constructed by the processing device of the computer using shape contour data of one or more shapes, and is matched with three-dimensional measurement data to mimic, for example, the topography of the one or more structures provided by the substrate, wherein the model image data is constructed by either arranging or resizing the contour data, or both; To fit the model image data to the three-dimensional measurement data, a regression operation is repeatedly performed by the processing device of the computer to obtain a composite model image of the three-dimensional topography of the substrate from the fitted model image data; and, To enable measurement of the topography of the one or more structures, the composite model image is provided as output as the three-dimensional topographic image data of the substrate's topography. The method, including the method described above.

15. A computer program product comprising an instruction that, when the program is executed by a computer processing device, causes the processing device to execute at least one of the methods according to claim 1 or the method according to claim 11.

Images