Methods for characterizing tilt through microscopic examination

Abstract

This disclosure provides a method for tilt characterization. The method includes measuring a first tilt displacement of a structure based on a first arrangement of the structure. The structure is formed in a vertical direction on a horizontal plane of the product. A second tilt displacement of the structure is measured based on a second arrangement of the structure. The second arrangement is a horizontal flip of the first arrangement. A corrected tilt displacement is determined based on the first and second tilt displacements.

Description

Technical Field

[0001] This disclosure relates to microscopic examination, and more specifically, to correcting distortions in microscopic examination. Background Technology

[0002] Microscopy is the use of a microscope to observe objects or areas of objects that are not visible to the naked eye. Microscopy is widely used in a variety of applications, such as microelectronics inspection, in vitro diagnostics, cell sorting and counting (cell counting), 3D imaging, and other technological fields where small size and high precision are important. Optical microscopes and electron microscopes are among the most commonly used microscopes. Optical microscopes use visible light and transparent lenses to observe objects such as human hair, blood cells, etc. Electron microscopes (e.g., transmission electron microscopes (TEM)) use electron beams and electron lenses to focus electrons and can provide atomic resolution. TEM is particularly important for microelectronics inspection. Microelectronics technologies (e.g., three-dimensional (3D) NAND flash memory technology) typically require the fabrication of microstructures as small as a few nanometers or even smaller. Summary of the Invention

[0003] This disclosure provides a method for tilt characterization using a microscope.

[0004] According to several aspects, a method for characterizing tilt is provided. The method includes measuring a first tilt displacement of a structure based on a first arrangement of the structure. The structure is formed in a vertical direction on a horizontal plane of the product. A second tilt displacement of the structure is measured based on a second arrangement of the structure. The second arrangement is a horizontal flip of the first arrangement. A first corrected tilt displacement is determined based on the first and second tilt displacements.

[0005] In some embodiments, measuring a first tilt displacement of the structure includes capturing a first image of a first cross-section of the product. The first cross-section includes a structure arranged in a first manner. Image analysis of the first image is performed to measure the first tilt displacement of the structure.

[0006] In some embodiments, measuring the second tilt displacement of the structure includes flipping a first cross-section to obtain a flipped first cross-section, the flipped first cross-section comprising a second arrangement of the structure. A second image of the flipped first cross-section is captured. Image analysis of the second image is performed to measure the second tilt displacement of the structure.

[0007] In some embodiments, measuring the second tilt displacement of the structure includes capturing a second image of a second cross-section of the product. The second cross-section includes a structure arranged in a second manner. Image analysis of the second image is performed to measure the second tilt displacement of the structure.

[0008] In some embodiments, at least one of a first corrected tilt shift or a first distortion-induced shift is determined based on a scalar operation of a first tilt shift and a second tilt shift.

[0009] In some embodiments, at least one of a first corrected tilt shift or a first distortion-induced shift is determined based on vector operations of a first tilt shift and a second tilt shift.

[0010] In some embodiments, the formula is used. Determine the first corrected tilt displacement. This is the first corrected tilt displacement. It is the first tilt shift, and It is the second tilt shift.

[0011] In some embodiments, the formula is used. Determine the shift caused by the first distortion. It is the shift caused by the first distortion. It is the first tilt shift, and It is the second tilt shift.

[0012] In some embodiments, a third tilt displacement of the structure is measured based on a third arrangement of the structure. The third arrangement is substantially perpendicular to the first arrangement. A fourth tilt displacement of the structure is measured based on a fourth arrangement of the structure. The fourth arrangement is a horizontal flip of the third arrangement. A second corrected tilt displacement is determined based on the third and fourth tilt displacements.

[0013] In some embodiments, at least one of a second corrected tilt shift or a second distortion-induced shift is determined based on vector operations of a third tilt shift and a fourth tilt shift.

[0014] In some embodiments, the displacement caused by the tilt of the structure in the horizontal plane is determined based on a vector operation of a first corrected tilt displacement and a second corrected tilt displacement.

[0015] In some embodiments, the formula is used. Determine the displacement caused by the tilt of the structure in the horizontal plane. It is the displacement caused by the tilt of the structure in the horizontal plane. It is the first corrected tilt displacement, and It is the second corrected tilt displacement.

[0016] In some embodiments, the first formula is used. Determine the second corrected tilt shift and use the second formula. Determine the shift caused by the second distortion. It is the second corrected tilt displacement. It is the shift caused by the second distortion. It is the third tilt shift, and It is the fourth tilt shift.

[0017] In some embodiments, the tilt-induced displacement of the structure in the horizontal plane is determined based on a scalar operation of a first corrected tilt displacement and a second corrected tilt displacement.

[0018] In some embodiments, the product includes a vertical NAND memory device.

[0019] In some embodiments, the structure includes at least one of a word line contact, a channel, a dummy channel, or a gate line gap of a vertical NAND memory device.

[0020] In some embodiments, a first tilt displacement and a second tilt displacement of the structure are measured using a microscope.

[0021] In some embodiments, the microscope includes at least one of a transmission electron microscope (TEM), a scanning electron microscope (SEM), an X-ray microscope, an optical microscope, or a fluorescence microscope.

[0022] In some embodiments, the microscope includes a TEM. The first cross-section is obtained by ultrathin sectioning, cryo-ultrathin sectioning, focused ion beam, ion etching, tripod polishing, or electrochemical treatment.

[0023] In some embodiments, the first cross-section has a thickness of less than 200 nm. Attached Figure Description

[0024] The aspects of this disclosure are best understood from the following detailed description when read in conjunction with the accompanying drawings. Note that, in accordance with standard industry practice, the various features are not drawn to scale. In practice, the dimensions of the various features may be increased or decreased for clarity of discussion.

[0025] Figure 1 A schematic diagram of transmission electron microscopy (TEM) imaging without elliptic distortion is shown.

[0026] Figure 2 A schematic diagram of a TEM image with elliptic distortion is shown.

[0027] Figure 3A A right angle is shown.

[0028] Figure 3B , Figure 3C and Figure 3D It shows in Figure 3A When the orientation of the right angle is different, the ellipse of the right angle is distorted.

[0029] Figure 4A A top view of the semiconductor device is shown.

[0030] Figure 4B It shows along Figure 4A The vertical cross-section diagram intercepted by the intercepting line AA' in the figure.

[0031] Figure 5A , Figure 5B , Figure 5C , Figure 5D , Figure 5E , Figure 5F , Figure 5G and Figure 5H A schematic diagram illustrating the coupling and decoupling of analytical elliptic distortion and structural tilt according to an embodiment of the present disclosure is shown.

[0032] Figure 6A A cross-sectional view of a semiconductor device with elliptic distortion according to an embodiment of the present disclosure is shown.

[0033] Figure 6B Analysis of embodiments according to this disclosure is illustrated. Figure 6A A schematic diagram of elliptical distortion and structural tilt in the image.

[0034] Figure 6C The following diagram illustrates the presence of elliptic distortion according to embodiments of the present disclosure. Figure 6A A flipped cross-sectional view of a semiconductor device.

[0035] Figure 6D Analysis of embodiments according to this disclosure is illustrated. Figure 6C A schematic diagram of elliptical distortion and structural tilt in the image.

[0036] Figure 6E The illustration shows an embodiment of the present disclosure when elliptic distortion has been corrected. Figure 6A and Figure 6C Corrected cross-sectional view of the semiconductor device in the image.

[0037] Figure 6F An embodiment according to this disclosure is shown. Figure 6E A schematic diagram of the corrected structural tilt.

[0038] Figure 7 The results of tilt and distortion measurements for different samples according to embodiments of the present disclosure are shown.

[0039] Figure 8 A comparison of tilt measurement results between different characterization techniques according to embodiments of the present disclosure is shown.

[0040] Figure 9 A flowchart illustrating an exemplary process for tilt characterization according to an embodiment of the present disclosure is shown. Detailed Implementation

[0041] The following disclosure provides numerous different embodiments or examples of various features for implementing the provided subject matter. Specific examples of components and arrangements are described below to simplify this disclosure. Of course, these are merely examples and are not intended to be limiting. For example, in the following description, the formation of a first feature on or over a second feature may include embodiments in which the first and second features can directly contact each other, and may also include embodiments in which additional features can be formed between the first and second features such that the first and second features do not directly contact each other. Additionally, reference numerals and / or letters may be repeated in various examples. Such repetition is for simplicity and clarity and does not in itself indicate a relationship between the various embodiments and / or configurations discussed.

[0042] Furthermore, spatial relative terms, such as “below,” “under,” “down,” “above,” and “above,” are used herein for ease of description to describe the relationship between one element or feature and another element(s) or feature as shown in the figures. Spatial relative terms are intended to cover different orientations in the use or operation of the device other than those depicted in the figures. The device may be oriented in other ways (rotated 90 degrees or in other orientations), and the spatial relative descriptors used herein may be interpreted accordingly.

[0043] To ensure the reliability of microscopic examination data, accurate data measurements are essential. The microscope must be properly aligned, and aberrations must be understood and corrected. Distortion is an aberration that causes the imaged object to appear unnatural. In the presence of distortion, rays from each point on the object recombine at corresponding points in the image plane, but the magnification varies across the entire image plane. For example, elliptic distortion is a common distortion in transmission electron microscopy (TEM). When elliptic distortion occurs, the magnification differs in different radial directions; this can be referred to as "TEM with elliptic distortion" or "TEM image with elliptic distortion."

[0044] Elliptic distortion can occur for several reasons. One common cause is astigmatism. That is, when the symmetry of the TEM's lens system is disrupted, electrons can be focused to different degrees in different radial directions. Elliptic distortion can also occur due to a faulty camera (or detector). For example, even when the lens system is ideally aligned, detector misalignment or non-uniform pixel spacing can cause elliptic distortion.

[0045] Elliptic distortion is typically detected by examining the diffraction pattern of a sample in the presence of astigmatism. In the absence of elliptic distortion, the diffraction pattern is circular. With elliptic distortion, the diffraction pattern is an ellipse with major and minor axes perpendicular to each other. Magnification differs along these two axes, and elliptic distortion is usually analyzed along both axes. Similarly, for a problematic camera, images of circular objects will appear elliptical.

[0046] Figure 1 A schematic diagram of transmission electron microscopy (TEM) imaging without elliptic distortion is shown, and Figure 2 A schematic diagram of a TEM image with elliptic distortion is shown. In one example, Figure 2 The diffraction pattern of the sample when astigmatism occurs is shown. In another example, Figure 2 The image shown is from the problematic camera.

[0047] exist Figure 1 In the diagram, lines AC and BD represent the two perpendicular axes of circle 110. Point O is the intersection of lines AC and BD, and also the center of circle 110. Lines AC and BD are of equal length, meaning there is no distortion. Therefore, lines AD and AB are perpendicular to each other. An object with the shape of a square ABCD can have an undistorted TEM image because there is no distortion.

[0048] exist Figure 2 In the diagram, lines A'C' and B'D' can represent the major and minor axes of ellipse 220, respectively. Lines A'C' and B'D' are perpendicular to each other and intersect at point O'. In a non-restrictive example, line A'C' is undistorted, while line B'D' has 1% distortion (e.g., a 1% shortening of its length), resulting in ellipse distortion. Consequently, lines A'D' and A'B' are no longer perpendicular to each other. Here, line A'B' extends in the x-direction, and line A'E' extends in the z-direction. Therefore, ∠D'A'E' = 90° - ∠D'A'O' - ∠B'A'O' = 0.576°. Line A'D' is inclined relative to line A'E', which is perpendicular to line A'B'.

[0049] In one example, an object with the shape of a square ABCD is positioned such that line AB extends along the x-direction and line AD extends along the z-direction. Therefore, the object may have a first distorted TEM image due to elliptic distortion, which has the shape of a rhombus A'B'C'D'. Here, points A', B', C', and D' correspond to points A, B, C, and D, respectively. This can represent the tilting and shifting of line A'D' caused by distortion (hence also known as distortion-induced shift). In other words, This represents the orthogonal projection of line A'D' along line A'B'. Is it the length of the orthogonal projection or The scalar value of . Assume A'C' = AC and B'D' = 0.99BD, therefore, A'D' ≈ AD. More importantly, Therefore, the shift caused by the distortion of line AD is approximately 1% in the +x direction.

[0050] In another example, an object with the shape of a square ABCD is positioned such that line AB extends along a first direction of line A'C', and line AD extends along a second direction (not shown) of line B'D'. Because line A'C' is undistorted and line B'D' has a 1% length reduction, the object can have a second distorted TEM image with a rectangular shape. This rectangle has two undistorted sides corresponding to lines AB and DC and extending along the first direction. The rectangle also has two distorted sides corresponding to lines AD and BC and extending along the second direction. Both of these distorted sides have a 1% length reduction. Note that the second distorted TEM image has no distortion-induced displacement because the adjacent sides of the rectangle are perpendicular to each other. Furthermore, an object with the shape of a square ABCD can have various distorted TEM images depending on how the object is positioned relative to the major axis (e.g., line A'C') and minor axis (e.g., line B'D') of ellipse 220. Figures 3A-3D The lieutenant general will explain in more detail the effect of the orientation of an object.

[0051] Still referencing Figure 2 The TEM can have any type of elliptic distortion. In one example, line A'C' is undistorted, while line B'D' has 2% distortion (e.g., 2% shortening or 2% elongation). In another example, lines A'C' and B'D' are distorted independently. Additionally, other microscopes (e.g., scanning electron microscopes (SEM)) may also have similar image distortion, where the image is magnified to different degrees in different axial directions. Although similar image distortions may or may not be called elliptic distortion for a particular microscopy technique, the analysis herein and the techniques disclosed in this disclosure remain applicable. For illustrative purposes, the TEM is used herein as an example.

[0052] This article uses Figures 3A-3D This illustrates how ellipse distortion can distort angles. For example, a right angle may be distorted into an acute angle, a right angle, or an obtuse angle. Figure 3A It shows the vector and The right angle ∠klm is defined. Figure 3B , Figure 3C and Figure 3D The distortion of the right angle ∠klm is shown when the right angle ∠klm is oriented differently relative to the ellipse 220.

[0053] exist Figure 3B In the middle, ∠klm is placed such that Extending in the +x direction, and Extending in the +z direction. Therefore, ∠klm can have a first distorted TEM image ∠KLM, where K, L, and M correspond to k, l, and m, respectively. LK is tilted and has a displacement induced by distortion along the +x direction, which is consistent with... Figure 2 The line A'D' in the diagram is similar. Therefore, ∠KLM is in... Figure 3B In the example, it is an acute angle. In some examples, the angle of ∠KLM can be represented by a combination of a first distortion value along the first direction of line A'C' and a second distortion value along the second direction of line D'B'.

[0054] exist Figure 3C In the middle, ∠klm is placed such that Extending in the first direction of online A'C and Extending along the second direction of the line D'B'. As a result, ∠klm can have a second distorted TEM image ∠K'L'M', where K', L', and M' correspond to k, l, and m, respectively. Here, ∠K'L'M' remains a right angle. Note that although the second distorted TEM image has no angular change, an object with an angle of ∠klm can have a size reduction or enlargement corresponding to the distortion values ​​along the major and minor axes of the ellipse 220. Additionally, although not shown, ∠klm is positioned such that... Extending in the first direction of the online A'C and When the line D'B extends in the second direction, ∠klm can have another distorted TEM image that presents a right angle.

[0055] exist Figure 3D In the middle, ∠klm is placed such that Extending in the -z direction, and Extending in the +x direction. As a result, ∠klm can have a third distorted TEM image ∠K”L”M”, where K”, L”, and M” correspond to k, l, and m, respectively. L”M” is tilted, and ∠K”L”M” is in… Figure 3D In the example, it is an obtuse angle. In some examples, the angle ∠K”L”M” can be represented by a first distortion value along the first direction of line A'C' and a second distortion value along the second direction of line D'B'.

[0056] Apart from Figure 3B , Figure 3C and Figure 3D Beyond the examples provided, the right angle ∠klm can have other distorted TEM images depending on how it is placed and oriented relative to ellipse 220. Therefore, in cases of elliptic distortion, the angle measured by TEM can be greater than, equal to, or less than the actual angle. In some examples, the angle measured by TEM can be the angle between the contact structure of the substrate and the working surface in a vertical NAND memory device. This is the root cause of the unstable measurement results and unexpectedly large tilt results in the relevant examples. If elliptic distortion is not corrected, this unpredictability and instability will render TEM measurements unreliable.

[0057] In particular, for microelectronic applications (e.g., the aforementioned vertical NAND flash memory technology), elliptic distortion in the TEM can lead to measurement errors. In some semiconductor devices, such as vertical NAND memory devices, the measurement error caused by elliptic distortion may be larger than the dimension of interest. Therefore, it may be necessary to correct for elliptic distortion in the TEM to ensure data accuracy. Note that the vertical NAND memory device is used as an example to illustrate elliptic distortion correction techniques, and such techniques can be used during the fabrication of other semiconductor devices.

[0058] In some examples, vertical NAND memory devices include vertical structures such as channel structures, dummy channel structures, and contact structures. Tilting of the vertical structure can cause operational failures in vertical NAND memory devices. For illustrative purposes, contact structures are used as an example in the following description.

[0059] Figure 4A A top view of the semiconductor device 400 is shown. Figure 4B It shows along Figure 4A The image shows a vertical cross-sectional view of semiconductor device 400 taken by the intercept line AA'. Figure 4A and Figure 4B Both were obtained through TEM characterization. Figure 4A and Figure 4B In the example, semiconductor device 400 includes a vertical NAND memory device. Specifically, semiconductor device 400 may include a stack of alternating insulating and gate layers. The stack includes a core region and a stepped region. The core region may include multiple channel structures, while the stepped region may include multiple steps and contact structures. The stepped region may be covered by an insulating layer.

[0060] As shown in the figure, the semiconductor device 400 may include a stepped region comprising multiple steps 430 (e.g., shown via 430(a), 430(b), and 430(c)). Contact structures 410 are formed on the steps 430 to connect to the gates of vertically series-connected transistors. For example, contact structure 410(a) is formed on step 430(a) and is electrically connected to gate layer 441(a); contact structure 410(b) is formed on step 430(b) and is electrically connected to gate layer 441(b); and contact structure 410(c) is formed on step 430(c) and is electrically connected to gate layer 441(c). Generally, insulating layers 442 between gate layers 441, such as gate layers 441(a), 441(b), and 441(c), are stacked, so that gate layers 441 can be individually controlled.

[0061] The semiconductor device 400 also includes a plurality of dummy channels 420. These dummy channels 420 provide physical and mechanical support to the semiconductor device 400 when the sacrificial layer (not shown) is removed during device fabrication to form the gate layer 441. In other words, the dummy channels 420 prevent the semiconductor device 400 from collapsing.

[0062] refer to Figure 4A The contact structure 410 and the dummy channel 420 are spaced apart from each other and separated by an insulating material 451 (e.g., silicon oxide). However, for vertical NAND memory devices, as the number of vertically connected transistors continues to increase, the vertical structure becomes increasingly longer in the z-direction. Such a vertical structure may include the contact structure 410, the dummy channel 420, a channel (not shown), a gate line gap (not shown), etc. Tilting of the vertical structure may cause it to merge with another structure that should be separated from it, leading to failure.

[0063] For example, in Figure 4B In this configuration, contact structure 410(a) is tilted toward and merges with dummy channel 420. Specifically, during the etching operation used to form contact structure 410(a), the tilt may cause the etching to penetrate into the layer below gate layer 441(a). As a result, contact structure 410(a) forms a tail 411(a) extending through the layer below gate layer 441(a) (e.g., some other gate layer 441 and insulating layer 442). Tail 411(a) may cause gate layer short circuits and layer-to-layer leakage. Note that in Figure 4B In the example, both the insulating material 451 and the dummy channel 420 comprise silicon oxide. Although the dummy channel 420 is shown as a dashed rectangle, it should be understood that the dummy channel 420 is formed around the area represented by the dashed rectangle and may not have a strictly rectangular shape.

[0064] Such tilting of the vertical structure can significantly reduce device yield and may require close monitoring. In some examples, when the vertical structure (e.g., contact structure 410) exceeds 10 μm in length in the z-direction, tilting of the vertical structure cannot be characterized by some form of in-line characterization. Instead, in some examples, tilting can be characterized by imaging a cross-section of the product including the vertical structure using a TEM.

[0065] like Figure 1 , Figure 2 and Figures 3A-3D As discussed, elliptic distortion can distort TEM images by tilting the structure and / or changing its dimensions. Even when the TEM has been repeatedly adjusted, it may still exhibit 1% distortion in the axial direction for various reasons. This 1% distortion may result in a tilt of 0.576°, such as... Figure 2 As shown. This 0.576° tilt may in turn cause a 1% distortion-induced displacement (e.g., Figure 2 In For a contact structure approximately 10 μm long (e.g., 410), the distortion-induced displacement along the silicon substrate can be as large as 10 μm × 1% = 100 nm. Note that the tilt displacement measured by TEM characterization is generated by both the distortion-induced displacement of the contact structure and the actual tilt displacement. However, the actual tilt displacement of the contact structure is typically less than 100 nm, which is smaller than the distortion-induced displacement. Therefore, it may be necessary to decouple the distortion-induced displacement to obtain the actual tilt displacement (also known as the corrected tilt displacement).

[0066] According to some aspects, this disclosure provides a method for tilt characterization using a microscope. Specifically, based on a first arrangement of the structure, a first tilt displacement of the structure formed in a vertical direction on a horizontal plane of the product is measured using a microscope. Based on a second arrangement of the structure, a second tilt displacement of the structure is measured using a microscope. The second arrangement is a horizontal flip of the first arrangement, meaning that the second arrangement can be obtained by rotating the first arrangement 180° about a vertical axis (e.g., the z-axis). Next, a corrected tilt displacement is determined based on the first and second tilt displacements.

[0067] As used herein, “product” generally refers to any manufactured product or processed sample. A product can be an intermediate or final product. Products can include semiconductor devices, medical devices, and nanomaterials, among others. The techniques described herein can be used to characterize the structure of a product. In some examples, the structure can include channel structures, dummy channel structures, and contact structures, etc. For illustrative purposes, contact structures are used as examples in the following description.

[0068] The technique described herein decouples distortion-induced displacement from tilt displacement measured by TEM to obtain the actual tilt displacement (also known as the corrected tilt displacement). This decoupling can be accomplished through simple scalar or vector calculations (e.g., addition and subtraction). This decoupling is reliable even when the distortion-induced displacement is greater than the actual tilt displacement, and the corrected tilt displacement is accurate. Furthermore, this disclosure provides a technique for characterizing distortion-induced displacement caused by elliptic distortion. Therefore, distortion-induced displacement or elliptic distortion can be monitored over time for multiple samples to assess the state of the microscope.

[0069] The technique described in this paper enables accurate data measurements even in the presence of elliptic distortion. TEMs (especially older TEMs) may exhibit persistent elliptic distortion, which is economically infeasible to correct. For example, replacing a defective camera or lens system may be prohibitively expensive. The technique described in this paper addresses this problem, even if such persistent elliptic distortion occurs in every measurement.

[0070] In some embodiments, two corrected tilt displacements along two vertical directions in the horizontal plane are determined. The tilt-induced displacement of the structure in the horizontal plane is then obtained based on the combination of the two corrected tilt displacements. The tilt-induced displacement of the structure in the horizontal plane represents the orthogonal projection of the structure onto the horizontal plane.

[0071] In some embodiments, the structure may include word line contacts of a vertical NAND memory device (e.g., Figure 4B The contact structure 410 in the middle) or the dummy channel (e.g., Figure 4B This disclosure provides a method for characterizing the tilt of word line contacts using a TEM with elliptic distortion (such as a dummy channel 420 in a semiconductor device or other vertical structures).

[0072] Figure 5A , Figure 5B , Figure 5C , Figure 5D , Figure 5E , Figure 5F , Figure 5G and Figure 5H A schematic diagram illustrating the coupling and decoupling of elliptic distortion and structural tilt according to embodiments of the present disclosure is shown. Specifically, Figure 5A Based on the first arrangement of the structure, and Figure 5B The second arrangement is based on the structure. The second arrangement is a horizontal flip of the first arrangement. Therefore, Figure 5A It is considered an unflipped image, while Figure 5B regarded as Figure 5A The flipped image. Similarly, Figure 5D , Figure 5F and Figure 5HThey are respectively regarded as Figure 5C , Figure 5E and Figure 5G The image is flipped. Note that the x, y, and z axes describe the relative orientation of the object in the microscope and are associated with the microscope. Therefore, when the microscope remains stationary, the x, y, and z axes change when the image is flipped. Figure 5B , Figure 5D , Figure 5F and Figure 5H The coordinate axes remain unchanged. For example, the x, y, and z axes can be the coordinate axes of the image plane in the TEM. Flipping the object will produce a flipped image, but the coordinate axes of the image plane in the TEM will remain unchanged.

[0073] In this article, distortion-induced shift is denoted as And the actual tilt displacement (also known as the corrected tilt displacement) is labeled as The tilt displacement measured by TEM based on unflipped and flipped images is labeled as follows: and It will be shown as as well as

[0074] exist Figure 5A and Figure 5B In the example, and There is no elliptic distortion or actual tilt of the structure. As a result, and in other words, and therefore, and

[0075] exist Figure 5C and Figure 5D In the example, and There is no elliptic distortion, but the structure is tilted. As a result, and This means that the tilt displacement measured by TEM is the actual tilt displacement, unaffected by any elliptic distortion. In this case, and therefore, and

[0076] exist Figure 5E and Figure 5F In the example, and The structure is not tilted, but it exhibits elliptic distortion. As a result, and This means that the tilt displacement measured by TEM is purely a displacement caused by distortion. Similarly, in this case, and therefore, and

[0077] exist Figure 5G and Figure 5H In the example, and Elliptic distortion exists, and the structure is tilted. As a result, and also, and

[0078] refer to Figures 5A-5H , as well as It holds true under various circumstances.

[0079] According to aspects of this disclosure, scalar operations, rather than the vector operations described above, can also be used to determine the corrected tilt. and In the case of the same direction, and The direction is from and The relative value determines this. Specifically, in... Figures 5A-5H In the example, and Both extend in the +x direction. hour, Extends in the +x direction. hour, Extends in the -x direction. hour, Note that in and In the opposite direction, and In addition, this article uses and All vector analyses can be performed using... and The scalar analysis replacement. The main difference is that the skew caused by distortion and the actual skew direction need to be analyzed separately.

[0080] And in Figures 5A-5H In the example, and Both extend in the +x direction; in other examples, or At least one of them can extend in the -x direction. However, similar vector or scalar analysis can be applied, and as well as It still holds true. Furthermore, by and The error introduced by the assumption of parallelism to the x-direction is negligible. In the example, the word line contact (e.g., contact structure 410) has a length of 13,000 nm, and the error introduced by this assumption when measuring tilt displacement is less than 3 nm.

[0081] Figures 6A-6F An example of tilting a structure (e.g., word line contacts) using TEM is shown according to embodiments of the present disclosure. In one embodiment, Figure 6A An unflipped cross-sectional view of a semiconductor device 600 is shown in the presence of elliptic distortion. The semiconductor device 600 may include a vertical NAND memory circuit having a core region (not shown) and a stepped region 660 having multiple steps 630.

[0082] As shown in the figure, the semiconductor device 600 includes a stacked layer 640 of alternating gate layers 641 and insulating layers 642. Each step 630 includes one or more pairs of alternating gate layers 641 and insulating layers 642. The semiconductor device 600 also includes a plurality of contact structures 610 in a step region 660. Each contact structure 610 is connected to a corresponding gate layer 641 of a corresponding step 630.

[0083] Note that the embodiment of semiconductor device 600 is similar to... Figure 4A and Figure 4B The embodiment of the semiconductor device 400 is similar. For example, Figure 6A Gate layer 641 in the middle corresponds to Figures 4A-4B Gate layer 441 in the middle. Figure 6A The insulating layer 642 in the middle corresponds to Figures 4A-4B The insulating layer 442 in the middle. Figure 6A Step 630 in the middle corresponds to Figures 4A-4B Step 430. Figure 6A The contact structure 610 in the middle corresponds to Figures 4A-4B The contact structure 410 in the middle. Figure 6A The insulating material 651 in the middle corresponds to Figures 4A-4B The insulating material 451 is described above and will be omitted here for brevity. The semiconductor device 600 may also include a dummy channel (not shown) corresponding to the dummy channel 420.

[0084] As shown in the figure, the contact structure 610 does not strictly extend in the z-direction. In other words, the contact structure 610 is inclined. In one example, in Figure 6A The first tilt displacement measured in the middle Including in the +x direction First tilt displacement This represents the orthogonal projection of the contact structure 610 along the x-direction. This can be obtained by measuring the tilt displacement of a specific contact structure 610, or It can be the average or weighted average of multiple contact structures 610. First tilt displacement. It can be measured in various ways. For example, it can be obtained using image analysis software or image analysis algorithms. Alternatively, TEM users can print out Figure 6A The orthogonal projection is drawn manually and measured using a ruler or any other length measuring tool.

[0085] Figure 6C An embodiment according to this disclosure is shown. Figure 6A A flipped cross-sectional view of semiconductor device 600 in the image. That is to say... Figure 6A The contact structure 610 in the middle has a first arrangement, and Figure 6C The contact structure 610 has a second arrangement that is horizontally flipped as a first arrangement. In a non-limiting example, the coordinate axes x, y, and z can be coordinate axes of the image plane in the TEM. Therefore, the coordinate axes x, y, and z are in... Figure 6C The ellipse remains unflipped. In some examples, without elliptical distortion, Figure 6C It can be Figure 6A The mirror image. As a result, for Figure 6C The tilt shift measured by TEM will include 108 nm in the -x direction. However, in the case of elliptic distortion, Figure 6C The second tilt displacement measured in the middle Not equal to In one example Including in the +x direction

[0086] like Figures 5A-5H As shown, the tilt of the contact structure 610 in the TEM image may be caused by both elliptic distortion and structural tilt. It is very important that... and in, and These represent the displacement caused by distortion along the x-direction and the corrected displacement along the x-direction, respectively. Here, Including in the +x direction and Including in the +x direction therefore, Including in the +x direction and Including in the +x direction Figure 6B and Figure 6D Analysis of embodiments according to this disclosure is shown respectively. Figure 6A and Figure 6C A schematic diagram of elliptical distortion and structural tilt in the image.

[0087] Figure 6E The illustration shows an embodiment of the present disclosure when elliptic distortion has been corrected. Figure 6A and Figure 6B Corrected cross-sectional view of semiconductor device 600 in the image. Figure 6F An embodiment according to this disclosure is shown. Figure 6E A schematic diagram of the corrected structural tilt.

[0088] exist Figures 6A-6F The example illustrates tilted characterization of a structure (e.g., word line contacts) using TEM. It should be understood that other structures of vertical NAND memory circuits can also be characterized using the techniques described herein, such as dummy channels (e.g., 420), channels, gate line gaps, etc. Furthermore, structures can include other structures of semiconductor devices, structures of medical devices, and structures of nanomaterials, etc.

[0089] A horizontal plane is inherently two-dimensional. Tilt can typically occur in two perpendicular directions within a horizontal plane. Figures 6A-6F In the example, the tilt along the x-direction is considered and characterized. The tilt along the y-direction can be characterized similarly.

[0090] In some embodiments, firstly, the third tilt displacement of the contact structure 610 can be measured by TEM based on the third arrangement of the contact structure 610. The third arrangement is essentially perpendicular to the first arrangement. In one example, the first arrangement ( Figure 6A The first arrangement corresponds to the section in the xz plane, and the third arrangement corresponds to the section in the yz plane. Secondly, the fourth tilt displacement of the structure can be measured using a microscope based on the fourth arrangement of the structure. The fourth arrangement is a horizontal flip of the third arrangement. Again, based on the third tilt shift... and fourth tilt shift Determine the corrected tilt displacement along the y-direction. Similarly, and in, and These correspond to the distortion-induced displacement along the y-direction and the corrected tilt displacement along the y-direction, respectively. Additionally, the tilt-induced displacement of the structure in the horizontal plane can optionally be calculated, and this tilt-induced displacement is labeled as... in, It can be used and Sure And the direction of displacement caused by the tilt of the structure in the horizontal plane.

[0091] Figure 7 The diagram illustrates tilt and distortion measurements for different samples (e.g., 710, 720, 730, and 740) according to embodiments of the present disclosure. From left to right, "first tilt shift," "second tilt shift," "distortion-induced shift," and "corrected tilt shift" correspond to... Figures 5A-5H In and Note that bar charts cannot represent directions in three-dimensional space using coordinate axes. In this paper, positive values ​​correspond to shifts in the +x direction, and negative values ​​correspond to shifts in the -x direction.

[0092] For sample 710, and In the +x direction, where, and Through application In the +x direction. By applying... nm in the -x direction, It is shown as -50nm. Similarly, for sample 720, and In the +x direction, where, and therefore, In the +x direction, and In the -x direction, It is shown as -6nm.

[0093] Still referencing Figure 7 The distortion-induced shifts for samples 710, 720, 730, and 740 were 58 nm, 51 nm, 54 nm, and 49 nm, respectively. The corresponding changes are less than 10 nm, which is an order of magnitude smaller than the distortion-induced shift. Such a small change indicates high accuracy in tilt displacement measurement. Furthermore, the corrected tilt displacement of a TEM with 1% elliptic distortion is as accurate as that of an uncorrected TEM with 0.1% elliptic distortion.

[0094] Figure 8A comparison of tilt measurement results between different characterization techniques according to embodiments of the present disclosure is shown. In-line measurements by scanning electron microscopy (SEM) are labeled "HV-SEM". From left to right, the three values ​​correspond to... and Specifically, On the -x side. In the +y direction. The techniques described in this paper were also applied and labeled "TEM-Test 1" and "TEM-Test 2". In TEM-Test 1, In the -x direction. In the +y direction. In TEM-test 2, In the -x direction. In the +y direction.

[0095] Three sets of data were obtained for the same product using HV-SEM, TEM-test 1, and TEM-test 2. For example, in... Figure 8 As can be seen, the two sets of data obtained by TEM are consistent with each other, and the two sets of data obtained by TEM are very close to the set of data obtained by HV-SEM. The in-line SEM measurements can be used as a reference for comparison, and this demonstrates the accuracy of the technique provided in this disclosure.

[0096] Figure 9 A flowchart of a process 900 for tilt characterization according to an embodiment of the present disclosure is shown. Process 900 begins at step S910, which measures a first tilt displacement of the structure based on a first arrangement of the structure. The structure is formed in the vertical direction on the horizontal plane of the product.

[0097] In some embodiments, a first tilt displacement of the structure is measured using a microscope. In some embodiments, the microscope may include at least one of a transmission electron microscope (TEM), a scanning electron microscope (SEM), an X-ray microscope, an optical microscope, a fluorescence microscope, etc. In some embodiments, the product includes a substrate parallel to a horizontal plane, and the structure is formed on the substrate in a vertical direction. In some embodiments, the structure may include at least one of word line contacts, channels, dummy channels, or gate line gaps, among other examples.

[0098] In some embodiments, measuring a first tilt displacement of the structure includes capturing a first image of a first cross-section of the product using a microscope. The first cross-section includes a structure arranged in a first manner. Image analysis of the first image is then performed to measure the first tilt displacement of the structure. In one example, image analysis software or an image analysis algorithm may be used to measure the first tilt displacement. In another example, the first image may be printed out, allowing the orthogonal projection of the structure to be manually drawn and measured.

[0099] In a non-limiting example, the structure of the vertical NAND memory circuit (e.g., contact structure 610) is characterized by TEM. The first cross-section of the structure can be obtained by ultrathin slicing, cryo-ultrathin slicing, focused ion beam (FIB), ion etching, tripod polishing, or electrochemical treatment. The first cross-section can have a thickness of less than 200 nm, for example, 30-150 nm. The lateral dimension of the first cross-section depends on the fabrication method. For example, a thin-film TEM cross-section prepared by ultrathin slicing can have a lateral dimension on the order of millimeters, for example, 3 mm × 3 mm. A thin-film TEM cross-section prepared by focused ion beam (FIB) can have a lateral dimension of less than 50 μm, for example, 15 μm × 15 μm. In one example, the vertical NAND memory circuit is disposed on a horizontal substrate extending in the xy plane, and the first cross-section can be... Figure 6A Similarly, the vertical section in the xz plane.

[0100] At step S920, a second tilt displacement of the structure is measured, for example, by means of a microscope, based on a second arrangement of the structure. The second arrangement is a horizontal flip of the first arrangement.

[0101] In some embodiments, measuring the second tilt displacement of the structure includes flipping a first cross-section to obtain a flipped first cross-section, the flipped first cross-section comprising the second arranged structure. Next, a second image of the flipped first cross-section is captured using a microscope. Then, image analysis of the second image is performed to measure the second tilt displacement of the structure.

[0102] In one embodiment, the first cross-section is fixed to the sample holder, and the sample holder is rotated to flip the first cross-section. In another embodiment, the first cross-section is loaded onto the sample holder, but is not fixed to the sample holder. The sample holder is rotated to remove the first cross-section and flip it. The flipped first cross-section is then reloaded onto the sample holder. Note that in the case of TEM characterization, it may be necessary to remove the sample holder from the TEM before flipping the first cross-section. Therefore, it may be necessary to reinsert the sample holder into the TEM.

[0103] In some embodiments, measuring the second tilt displacement of the structure includes capturing a second image of a second cross-section of the product using a microscope. The second cross-section includes a structure arranged in a second manner. Image analysis of the second image is performed to measure the second tilt displacement of the structure. Note that, in the case of TEM characterization, the first and second cross-sections can be loaded onto a single sample loader, allowing measurements to be performed without removing the sample holder from the TEM and inserting the sample holder into the TEM between measurements.

[0104] Then, process 900 proceeds to step S930, in which a first corrected tilt shift is determined based on the first tilt shift and the second tilt shift. In some embodiments, the first corrected tilt shift is determined based on vector or scalar operations of the first and second tilt shifts. In some embodiments, the vector or scalar operation includes obtaining a combination of the first and second tilt shifts, for example, Figures 6A-6F The linear combination shown In some embodiments, the displacement induced by the first distortion is determined based on a combination of the first tilt shift and the second tilt shift, for example, Figures 6A-6F Linear combinations shown in

[0105] In some embodiments, a third tilt displacement of the structure is measured using a microscope-based third arrangement. The third arrangement is substantially perpendicular to the first arrangement. A fourth tilt displacement of the structure is measured using a microscope-based fourth arrangement. The fourth arrangement is a horizontal flip of the third arrangement. A second corrected tilt displacement can be determined based on the third and fourth tilt displacements.

[0106] Similar to step S910, measuring the third tilt displacement of the structure may include capturing a third image of a third cross-section of the product using a microscope. The third cross-section includes a third arrangement of the structure. In one example, the first cross-section is a vertical cross-section in the xz plane, and the third cross-section is a vertical cross-section in the yz plane. Therefore, the first corrected tilt displacement and the second corrected tilt displacement are along the x-direction and the y-direction, respectively.

[0107] In some embodiments, the second corrected tilt shift is determined based on vector or scalar operations of the third and fourth tilt shifts (e.g., obtaining a linear combination of the third and fourth tilt shifts). Alternatively, the second corrected tilt shift can be determined based on vector or scalar operations of the first and second corrected tilt shifts (e.g., ...). Figures 6A-6F Linear combinations shown in This determines the displacement caused by the tilt of the structure in the horizontal plane.

[0108] In some embodiments, the microscope is a SEM, which is similar to a TEM. As described above, the key differences will be explained. With SEM, a first cross-section can also be obtained to measure a first tilt displacement of the structure. The first cross-section has a certain thickness (e.g., 100 nm) and therefore has two sides. It may be necessary to expose the structure on at least one side of the first cross-section (e.g., contact structure 610) because electrons do not penetrate the first cross-section in SEM characterization. However, for TEM characterization, the structure can be hidden within the first cross-section or can be exposed because electrons can transmit through the first cross-section.

[0109] As used herein, “device” or “semiconductor device” generally refers to any suitable device, such as memory circuitry, a semiconductor chip (or die) having memory circuitry formed on a semiconductor chip, a semiconductor wafer having multiple semiconductor dies formed on a semiconductor wafer, a stacked layer of semiconductor chips, a semiconductor package including one or more semiconductor chips assembled on a packaging substrate, etc.

[0110] As used herein, “substrate” generally refers to an object processed according to the present invention. A substrate may include any material portion or structure of a device (particularly a semiconductor device or other electronic device) and may be, for example, a base substrate structure, such as a semiconductor wafer, a photomask, or a layer (e.g., a thin film) on or covering a base substrate structure. Therefore, a substrate is not limited to any particular base structure, underlayer, or overlayer (patterned or unpatterned), but is contemplated to include any such layer or base structure and any combination of layers and / or base structures. The specification may refer to specific types of substrates, but this is for illustrative purposes only.

[0111] The substrate can be any suitable substrate, such as a silicon (Si) substrate, a germanium (Ge) substrate, a silicon-germanium (SiGe) substrate, and / or a silicon-on-insulator (SOI) substrate. The substrate can include semiconductor materials, such as group IV semiconductors, group III-V compound semiconductors, or group II-VI oxide semiconductors. Group IV semiconductors can include Si, Ge, or SiGe. The substrate can be a bulk wafer or an epitaxial layer.

[0112] The foregoing has outlined features of several embodiments to enable those skilled in the art to better understand aspects of this disclosure. Those skilled in the art will understand that they can readily use this disclosure as a basis to design or modify other processes and structures for achieving the same purpose and / or realizing the same advantages of the embodiments described herein. Those skilled in the art will also recognize that such equivalent constructions do not depart from the spirit and scope of this disclosure, and that various changes, substitutions, and modifications can be made herein without departing from the spirit and scope of this disclosure.

Claims

1. A method for characterizing tilt, the method comprising: The first tilt displacement of the structure is measured based on the first arrangement of the structure, which is formed in the vertical direction on the horizontal plane of the product; A second tilt displacement of the structure is measured based on a second arrangement of the structure, wherein the second arrangement is a horizontal flip of the first arrangement; as well as The first corrected tilt shift and the first distortion-induced shift are determined based on scalar or vector operations of the first tilt shift and the second tilt shift. The measurement of the first tilt displacement of the structure includes: Capture a first image of a first cross-section of the product, wherein the first cross-section includes the structure of the first arrangement; and Perform image analysis on the first image to measure the first tilt displacement of the structure.

2. The method according to claim 1, wherein, Measuring the second tilt displacement of the structure includes: Flip the first cross section to obtain a flipped first cross section including the structure of the second arrangement; Capture a second image of the flipped first cross-section; and Perform image analysis on the second image to measure the second tilt displacement of the structure.

3. The method according to claim 1, wherein, Measuring the second tilt displacement of the structure includes: Capture a second image of a second cross-section of the product, wherein the second cross-section includes the structure of the second arrangement; and Perform image analysis on the second image to measure the second tilt displacement of the structure.

4. The method according to claim 1, further comprising using a formula = Determine the first corrected tilt shift, wherein: It is the first corrected tilt displacement. It is the first tilt shift, and This is the second tilt shift.

5. The method according to claim 1, further comprising using a formula = Determine the shift caused by the first distortion, wherein: The displacement is caused by the first distortion. It is the first tilt shift, and This is the second tilt shift.

6. The method according to claim 1, further comprising: A third tilt displacement of the structure is measured based on a third arrangement of the structure, the third arrangement being substantially perpendicular to the first arrangement; A fourth tilt displacement of the structure is measured based on a fourth arrangement of the structure, wherein the fourth arrangement is a horizontal flip of the third arrangement; as well as The second corrected tilt shift is determined based on the third tilt shift and the fourth tilt shift.

7. The method according to claim 6, further comprising: The vector operation of the third tilt shift and the fourth tilt shift determines at least one of the second corrected tilt shift or the second distortion-induced shift.

8. The method according to claim 7, further comprising: The displacement caused by the tilt of the structure in the horizontal plane is determined based on vector operations of the first corrected tilt displacement and the second corrected tilt displacement.

9. The method of claim 8, further comprising using a formula = + Determine the displacement caused by the tilt of the structure in the horizontal plane, wherein: It is the displacement caused by the tilt of the structure in the horizontal plane. It is the first corrected tilt displacement. It is the second corrected tilt shift, and = | + | = 。 10. The method of claim 7, further comprising using the first formula = Determine the second corrected tilt shift and use the second formula. = Determine the shift caused by the second distortion, wherein: It is the second corrected tilt shift. The shift is caused by the second distortion. It is the third tilt shift, and This is the fourth tilt shift.

11. The method of claim 6, further comprising: At least one of the second corrected tilt shift or the second distortion-induced shift is determined based on scalar operations of the third tilt shift and the fourth tilt shift.

12. The method according to claim 1, wherein, The products include vertical NAND memory devices.

13. The method according to claim 12, wherein, The structure includes at least one of the word line contacts, channels, dummy channels, or gate line gaps of the vertical NAND memory device.

14. The method of claim 1, further comprising measuring the first tilt displacement of the structure and the second tilt displacement of the structure by means of a microscope.

15. The method of claim 14, wherein: The microscope includes at least one of transmission electron microscopy (TEM), scanning electron microscopy (SEM), X-ray microscopy, optical microscopy, or fluorescence microscopy.

16. The method according to claim 15, wherein, The microscope includes a TEM, and the method further includes: The first cross section is obtained by ultrathin sectioning, cryo-ultrathin sectioning, focused ion beam, ion etching, tripod polishing, or electrochemical treatment.

17. The method according to claim 16, wherein, The first cross section has a thickness of less than 200 nm.

Images