Preparation method of ultrathin TEM sample

Abstract

The invention relates to the technical field of semiconductors, and provides a preparation method of an ultrathin TEM sample. The method comprises the following steps: establishing an XYZ three-dimensional coordinate system; performing primary longitudinal cutting treatment on the to-be-tested device to obtain a first longitudinal section TEM sample; determining and marking a target structure area in the TEM sample with the first longitudinal section, and preparing a first protection layer on the first surface, the second surface and the sixth surface of the TEM sample with the first longitudinal section to obtain a first protection sample; the first protection sample is rotated by 90 degrees along the Y axis and the X axis in sequence, a second protection layer is prepared on the second surface and / or the sixth surface of the first protection sample, and a second protection sample is obtained; and performing secondary longitudinal cutting treatment on the second protection sample to obtain a second longitudinal section TEM sample. The ultrathin TEM sample with the thickness of 40 nm or below is prepared, two longitudinal sections of the device to be detected can be observed, the morphology of the device can be observed from more dimensions, and the failure reason of the device can be accurately analyzed.

Description

Technical Field

[0001] This application relates to the field of semiconductor technology, and in particular to a method for preparing ultrathin TEM samples. Background Technology

[0002] In the process of manufacturing chips from wafers, transmission electron microscopy (TEM) is usually used to observe the morphology of the devices and analyze the causes of their failures.

[0003] Current TEM sample preparation methods typically cause device deformation when the thickness is reduced to around 100 nm. Further thinning to around 80 nm can easily lead to severe deformation due to insufficient mechanical stress, making it impossible to prepare ultrathin TEM samples (thickness below 40 nm). Furthermore, TEM samples prepared using existing methods can usually only be observed from one longitudinal section of the device, making it impossible to observe the device's morphology from multiple dimensions and accurately analyze the cause of its failure. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method for preparing ultrathin TEM samples, which aims to solve the problem that the existing technology cannot prepare ultrathin TEM samples (thickness below 40 nm) and can only observe one longitudinal section of the device, making it impossible to observe the morphology of the device from more dimensions and accurately analyze its failure cause.

[0005] To achieve the above-mentioned objectives, the technical solution adopted in this application is as follows: In a first aspect, embodiments of this application provide a method for preparing ultrathin TEM samples, including: Establish an XYZ three-dimensional coordinate system, wherein the XY plane of the XYZ three-dimensional coordinate system is parallel to the first surface of the device under test, the XZ plane is parallel to the second surface of the device under test, and the YZ plane is parallel to the third surface of the device under test; the XY plane intersects with the XZ plane and the YZ plane and is perpendicular to each other, and the first surface intersects with the second surface and the third surface and is perpendicular to each other. The device under test is subjected to a longitudinal section using a focused ion beam to obtain a first longitudinal section TEM sample; wherein the incident direction of the focused ion beam is parallel to the third surface of the device under test. The target structural region in the first longitudinal section TEM sample is identified and marked, and a first protective layer is prepared on the first surface, the second surface and the sixth surface of the first longitudinal section TEM sample to obtain a first protective sample; wherein the second surface and the sixth surface are opposite and parallel, and the target structural region is located in the space between the second surface and the sixth surface; After rotating the first protective sample 90 degrees along the Y-axis and X-axis respectively, a second protective layer is prepared on the second surface and / or the sixth surface of the first protective sample to obtain the second protective sample. The second protective sample is subjected to a second longitudinal section using a focused ion beam to obtain a second longitudinal section TEM sample; wherein the incident direction of the focused ion beam is perpendicular to the second surface of the second protective sample, and the thickness of the second longitudinal section TEM sample is less than 40 nm.

[0006] Compared with the prior art, the beneficial effects of the technical solution provided in this application include at least the following: First, a first longitudinal section TEM sample is obtained by performing a longitudinal sectioning process on the device under test (DUT). Then, a first protective layer is prepared on the first, second, and sixth surfaces of the first longitudinal section TEM sample to obtain a first protective sample. The first protective sample is then rotated 90 degrees sequentially along the Y-axis and X-axis, and a second protective layer is prepared on the second and / or sixth surfaces of the first protective sample to obtain a second protective sample. Finally, the second protective sample is subjected to a second longitudinal sectioning process to obtain a second longitudinal section TEM sample. Through this method, ultra-thin TEM samples with a thickness of less than 40 nm can be obtained, and both longitudinal sections of the DUT can be observed. This allows for observation of the device morphology from more dimensions and accurate analysis of its failure causes, which is beneficial for improving the accuracy and reliability of the TEM failure analysis results of the DUT. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 This is a schematic flowchart of an ultrathin TEM sample preparation method provided in an embodiment of this application; Figure 2 This is a schematic diagram of an XYZ three-dimensional coordinate system provided in an embodiment of this application; Figure 3 This is a schematic diagram of the ion beam incident direction and the placement of the device under test in the first longitudinal sectioning process of the ultrathin TEM sample preparation method provided in the embodiments of this application; Figure 4 This is a schematic diagram of rotating the first protective sample 90 degrees clockwise along the Y-axis, as provided in an embodiment of this application. Figure 5 The embodiments provided in this application will Figure 4 A schematic diagram of the first protected sample rotating 90 degrees clockwise along the X-axis; Figure 6 This is a schematic diagram of rotating the first protective sample 90 degrees counterclockwise along the Y-axis, as provided in an embodiment of this application. Figure 7 The embodiments provided in this application will Figure 6 A schematic diagram of the first protective sample rotated 90 degrees counterclockwise along the X-axis; Figure 8 The embodiments provided in this application will Figure 6 A schematic diagram of the first protective sample rotated 90 degrees clockwise along the X-axis; Figure 9 This is a schematic diagram of the first longitudinal section TEM sample in Embodiment 1 of this application; Figure 10 This is a schematic diagram of the deposition of the first protective layer on the first, second, and sixth surfaces of the first longitudinal section TEM sample in Embodiment 1 of this application; Figure 11 This is a schematic diagram of rotating the first protective sample sequentially along the Y-axis and X-axis by 90 degrees each in Embodiment 1 of this application; Figure 12 This is a schematic diagram of depositing a second protective layer on the second or sixth surface of the first protective layer, as provided in Embodiment 1 of this application; Figure 13 This is a schematic diagram of the process of extracting the target sample from the second protected sample and transferring and fixing it to the side wall of the copper pillar in Embodiment 1 of this application; Figure 14 This is a schematic diagram of the secondary longitudinal processing of the target sample in Embodiment 1 of this application; Figure 15 This is a TEM analysis image of the second longitudinal section TEM sample in Embodiment 1 of this application; Figure 16 It is by Figure 15 The TEM rendering image obtained by converting the TEM analysis image corresponding to the target structure in the image; Figure 17 This is a schematic diagram of rotating the first protected sample 90 degrees along the Z-axis in Comparative Example 1; Figure 18 This is a schematic diagram of the process of transferring and fixing the sample to be longitudinally cut to the top of the copper column in Comparative Example 1. Figure 19 This is a schematic diagram of the secondary longitudinal processing of the sample to be longitudinally cut in Comparative Example 1. Figure 20 This is a TEM analysis image of the longitudinal section TEM sample of Comparative Example 1; Figure 21 It is by Figure 20 The TEM rendering image obtained by converting the TEM analysis image corresponding to the target structure in the image; Figure 22 This is a schematic diagram of rotating the first protective sample 90 degrees counterclockwise along the Z-axis in Comparative Example 3. Detailed Implementation

[0009] To make the technical problems, technical solutions, and beneficial effects to be solved by this application clearer, the following detailed description is provided in conjunction with embodiments. It should be understood that the specific embodiments described herein are only for explaining this application, but the implementation of this application is not limited thereto.

[0010] As the prices of electronic products decrease and their market penetration increases, smaller semiconductor devices can reduce material waste and energy consumption in the production process, meeting the needs of environmental protection and sustainable development, thereby prompting companies to promote the miniaturization of device sizes.

[0011] As the process dimensions of semiconductor devices continue to shrink, traditional TEM sample preparation methods typically cause devices to gradually deform when the thickness is reduced to around 100 nm. When the thickness is further reduced to around 80 nm, the devices are prone to severe deformation due to insufficient mechanical stress, making it impossible to prepare ultra-thin TEM samples (thickness below 40 nm). Furthermore, the prepared TEM samples can only be observed from one longitudinal section of the device, making it impossible to observe the morphology of the device from multiple dimensions and accurately analyze the cause of its failure.

[0012] In view of this, embodiments of this application provide a method for preparing an ultrathin TEM sample. First, a first longitudinal section TEM sample is obtained by longitudinally slicing the device under test. Then, a first protective layer is prepared on the first, second, and sixth surfaces of the first longitudinal section TEM sample to obtain a first protective sample. Next, the first protective sample is rotated 90 degrees sequentially along the Y-axis and X-axis, and a second protective layer is prepared on the second and / or sixth surfaces of the first protective sample to obtain a second protective sample. The second protective sample is then longitudinally sliced ​​a second time to obtain a second longitudinal section TEM sample. This second longitudinal section TEM sample is an ultrathin TEM sample with a thickness of less than 40 nm, and it allows observation of both longitudinal sections of the device under test. This allows for observation of the device's morphology from more dimensions and accurate analysis of its failure causes, which is beneficial for improving the accuracy and reliability of the TEM failure analysis results of the device under test.

[0013] Figure 1 This is a schematic flowchart of an ultrathin TEM sample preparation method provided in an embodiment of this application. Please refer to... Figure 1 The ultrathin TEM sample preparation method provided in this application includes the following steps: Step S101: Establish an XYZ three-dimensional coordinate system, wherein the XY plane of the XYZ three-dimensional coordinate system is parallel to the first surface of the device under test, the XZ plane is parallel to the second surface of the device under test, and the YZ plane is parallel to the third surface of the device under test; the XY plane intersects with the XZ plane and the YZ plane and is perpendicular to each other, and the first surface intersects with the second surface and the third surface and is perpendicular to each other.

[0014] The device under test 200 can be a semiconductor device, such as a chip or wafer.

[0015] Figure 2 This is a schematic diagram of an XYZ three-dimensional coordinate system provided in an embodiment of this application. For an example, please refer to... Figure 2 A three-dimensional XYZ coordinate system is established with the sample center of the device under test (DUT) 200 as the origin. The XY plane of this system is parallel to the first surface of DUT 200 (the plane marked ①, i.e., the upper side of DUT 200), the XZ plane is parallel to the second surface of DUT 200 (the plane marked ②, i.e., the front side of DUT 200), and the YZ plane is parallel to the third surface of DUT 200 (the plane marked ③, i.e., the left side of DUT 200). The XY plane intersects and is perpendicular to the XZ and YZ planes, and the first surface intersects and is perpendicular to the second and third surfaces. DUT 200 also includes a fourth surface (the plane marked ④, i.e., the lower side of DUT 200), a fifth surface (the plane marked ⑤, i.e., the right side of DUT 200), and a sixth surface (the rear side of DUT 200) that is opposite to and parallel to the second surface.

[0016] Step S102: The device under test is longitudinally sectioned using a focused ion beam to obtain a first longitudinal section TEM sample; wherein the incident direction of the focused ion beam is parallel to the third surface of the device under test.

[0017] Combination Figure 3 A single longitudinal section refers to using a focused ion beam (FIB) to perform longitudinal cutting from the first surface (the plane marked ①) of the device under test 200 towards the fourth surface (the plane marked ④, i.e., the lower side of the device under test 200). The incident direction of the focused ion beam is parallel to the third surface (the plane marked ③, i.e., the left side of the device under test 200).

[0018] Step S103: Determine and mark the target structural region in the first longitudinal section TEM sample, and prepare a first protective layer on the first surface, second surface and sixth surface of the first longitudinal section TEM sample to obtain a first protective sample, wherein the target structural region is located in the space between the second surface and the sixth surface.

[0019] As an example, please refer to Figure 3The first longitudinal section TEM sample prepared above is scanned using SEM (Scanning Electron Microscopy) to confirm its target structural region and mark it. For example, the coordinates of the target structural region can be determined by combining the morphological features (such as device shape) of the first longitudinal section TEM sample with the required slicing drawings, and a high-precision marking reference point (such as a "T" shaped notch) can be etched at the target structural region using an electron beam. This target structural region is located in the space between the second and sixth surfaces. This ensures that subsequent longitudinal slicing can accurately extract the target structural region of the device under test 200, avoiding subsequent analysis failures due to incomplete longitudinal sections caused by offset (such as only cutting to the edge of the target structural region of the device under test 200 or missing the target structural region). Next, after marking the target structural region in the first longitudinal section TEM sample, a first protective layer is prepared on the first, second, and sixth surfaces of the first longitudinal section TEM sample to obtain the first protective sample.

[0020] The target structural region generally refers to the failure location or structural / functional abnormality location of the device under test.

[0021] Step S104: After rotating the first protective sample 90 degrees along the Y-axis and X-axis respectively, a second protective layer is prepared on the second surface and / or the sixth surface of the first protective sample to obtain the second protective sample.

[0022] As an example, please refer to Figures 3-4 First, rotate the first protective sample 90 degrees clockwise along the Y-axis. At this point, the third surface of the first protective sample (the plane marked ③) is flipped to the upper side position, and the first surface (the plane marked ①) is flipped to the right side position. Next, please refer to... Figure 5 The first protective sample is rotated 90 degrees clockwise along the X-axis. At this point, the fifth surface (plane marked ⑤) of the first protective sample is flipped to the front side, the second surface (plane marked ②) is flipped to the upper side, and the sixth surface is flipped to the lower side. Next, a second protective layer is prepared on the second surface (plane marked ②) of the first protective sample to obtain the second protective sample. Specifically, the second protective layer can be prepared only on the second surface of the first protective sample at the location corresponding to the target structural region. This reduces the deposition area and time, avoids the problem of excessive heat introduction leading to sample bending due to prolonged deposition, and improves sample preparation efficiency.

[0023] As another example, please refer to Figure 3 , Figure 6Rotate the first protective sample 90 degrees counterclockwise along the Y-axis. At this point, the right side of the first protective sample (the plane marked ⑤) is flipped to the upper side position, and the third surface (the plane marked ③) is flipped to the lower side position. Next, please refer to... Figure 7 Next, the first protective sample is rotated 90 degrees counterclockwise along the X-axis. At this point, the second surface (the plane marked ②) of the first protective sample is flipped to the lower side, the sixth surface is flipped to the upper side, and the fifth surface (the plane marked ⑤) is flipped to the front side. Then, a second protective layer is prepared on the sixth surface of the first protective sample to obtain the second protective sample. Specifically, the second protective layer can be prepared only on the sixth surface of the first protective sample at the location corresponding to the target structure region. This reduces the deposition area and time, avoids the problem of sample bending caused by excessive heat introduced over a long period, and improves sample preparation efficiency.

[0024] As yet another example, please refer to Figure 3 , Figure 6 Rotate the first protective sample 90 degrees counterclockwise along the Y-axis. At this point, the fifth surface (the plane marked ⑤) of the first protective sample is flipped to the upper side position, and the third surface (the plane marked ③) is flipped to the lower side position. Next, please refer to... Figure 8 Next, the first protective sample is rotated 90 degrees clockwise along the X-axis. At this point, the third surface of the first protective sample is flipped to the front side, the second surface is flipped to the upper side, and the sixth surface is flipped to the lower side. Then, a second protective layer is prepared on the second surface of the first protective sample to obtain the second protective sample. Specifically, the second protective layer can be prepared only on the second surface of the first protective sample at the location corresponding to the target structural region. This reduces the deposition area and time, avoids the problem of excessive heat introduction leading to sample bending due to prolonged deposition, and improves sample preparation efficiency.

[0025] As another example, please refer to Figure 3 , Figure 6 Rotate the first protective sample 90 degrees counterclockwise along the Y-axis. At this point, the fifth surface (the plane marked ⑤) of the first protective sample is flipped to the upper side position, and the third surface (the plane marked ③) is flipped to the lower side position. Next, please refer to... Figure 8Next, the first protective sample is rotated 90 degrees clockwise along the X-axis. At this time, the third surface of the first protective sample is flipped to the front side position, the second surface (the plane marked ②) is flipped to the upper side position, and the sixth surface is flipped to the lower side position. Then, a second protective layer is prepared on the second surface of the first protective sample. Afterwards, the first protective sample with the second protective layer deposited on the second surface is rotated 180 degrees counterclockwise or clockwise along the Y-axis. At this time, the sixth surface of the first protective sample is flipped to the upper side position, and the second surface is flipped to the lower side position. A second protective layer is then prepared on the sixth surface of the first protective sample, resulting in the second protective sample. Through the above method, a second protective layer can be prepared on the second and sixth surfaces of the first protective sample to better protect the target structural region located in the space between the second and sixth surfaces, thereby preventing deformation during the secondary longitudinal cutting process and facilitating the thinning of the device under test to the desired thickness (below 40 nm).

[0026] Step S105: The second protective sample is subjected to a second longitudinal sectioning process using a focused ion beam to obtain a second longitudinal section TEM sample; wherein, the incident direction of the focused ion beam is perpendicular to the second surface of the second protective sample, and the thickness of the second longitudinal section TEM sample is less than 40 nm.

[0027] Secondary longitudinal cutting refers to using a focused ion beam (FIB) to perform longitudinal cutting along the upper side to the lower side of the device under test 200. The incident direction of the focused ion beam is parallel to the YZ plane in the XYZ three-dimensional coordinate system.

[0028] The technical solution provided in this application involves preparing a first protective layer on the first, second, and sixth surfaces of a first longitudinal section TEM sample. Then, the first protective sample is rotated 90 degrees along the Y-axis and then 90 degrees along the X-axis. A second protective layer is then prepared on the second and / or sixth surfaces of the first protective sample. This effectively prevents damage to the target structural region of the device under test (DUT) from bombardment by a focused ion beam during subsequent secondary longitudinal sectioning. Simultaneously, it increases the mechanical strength of the first longitudinal section TEM sample, preventing deformation during secondary longitudinal sectioning. This facilitates thinning the DUT to the desired thickness (below 40 nm). Furthermore, the resulting ultrathin TEM sample allows for observation of the morphology of the target structural region and analysis of its failure causes through two longitudinal sections, improving the accuracy and reliability of the TEM failure analysis results for the DUT.

[0029] In some embodiments, a first protective layer can be prepared on the first surface of the first longitudinal section TEM sample using electron beam deposition, wherein the voltage is 2 kV and the current is 1~2.5 nA.

[0030] Preferably, the first protective layer is a platinum layer (Pt layer). The thickness of the first protective layer is 0.8~2 µm.

[0031] Pt is a solid at room temperature. The operating temperature for electron beam deposition or ion beam deposition is 38 ℃ to 42 ℃, with a preheating time of 5 min. The Pt layer is very hard and resistant to chemical corrosion, which can effectively improve the mechanical strength of the sample and prevent deformation during secondary longitudinal cutting. Since electron beam deposition of Pt layers is faster than ion beam deposition, electron beam deposition of Pt layers is preferred in practical applications to improve the preparation efficiency of the first protective layer.

[0032] Tungsten (W) is a solid at room temperature. Electron beam deposition or ion beam deposition is performed at 50 °C, with a relatively long preheating time, typically 15 min. Its resistivity is lower than that of Pt. Because the W layer is very hard, it effectively improves the mechanical strength of the sample, thus preventing deformation due to ion beam bombardment during secondary longitudinal cutting. Carbon (C) is a solid at room temperature. Electron beam deposition or ion beam deposition can be performed at room temperature, with a fast preheating time, typically 3 min.

[0033] However, since the deposition time of the W layer is relatively long, it is easy to introduce too much heat energy, which can cause the sample to bend. On the other hand, the deposition time of the Pt layer or C layer is shorter, which can avoid introducing too much heat energy and causing the sample to bend. Therefore, in practical applications, the first protective layer is preferably a Pt layer or a C layer, with the Pt layer being the most preferred.

[0034] In some embodiments, the second protective layer electron beam can be prepared on the second or sixth surface of the first protective sample using ion beam deposition, wherein the voltage is 8 kV and the current is 61 pA.

[0035] Preferably, the second protective layer is a platinum layer. The thickness of the second protective layer is 1.5~5 µm.

[0036] It should be noted that the thickness of the first and second protective layers can be flexibly adjusted according to the desired thinning thickness or material of the device under test. Generally, the smaller (larger) the desired thinning thickness, the larger (smaller) the thickness of the first and second protective layers. This can provide sufficient mechanical support for the sample, avoid deformation of the sample due to ion beam bombardment during the secondary longitudinal sectioning process, and at the same time take into account the efficiency of sample preparation.

[0037] The technical solution provided in this application involves preparing a first protective layer with a thickness of 0.8–2 µm on the first, second, and sixth surfaces of a TEM sample in a first longitudinal section using electron beam deposition. Then, the TEM sample with the first protective layer deposited is rotated 90 degrees along both the Y and X axes, and a second protective layer with a thickness of 1.5–5 µm is prepared on its second or sixth surface using ion beam deposition. The low-damage characteristics of electron beam deposition prevent sample overheating, and the first protective layer provides support for ion beam deposition, resulting in a dense and uniform second protective layer. This improves the mechanical strength of the sample and prevents damage and deformation caused by ion beam bombardment during subsequent secondary longitudinal sections, enabling the preparation of ultrathin TEM samples with a thickness below 40 nm.

[0038] It is easy to understand that when an ion beam deposits a protective layer directly on the surface of a TEM sample, it will cause damage to the sample surface. When observing TEM samples with a thickness of 40 nm or more, this surface damage has little impact on the observation results. However, when observing TEM samples with a thickness of 40 nm or more, this damage will affect the observation results. Moreover, the smaller the size of the TEM sample, the greater the impact. For example, when an ion beam deposits a protective layer directly on the surface of a 10 nm TEM sample, the damage to the sample will lead to inaccurate observation results.

[0039] Therefore, this application first prepares a first protective layer on the first surface, second surface and / or sixth surface of the first longitudinal section TEM sample by electron beam to obtain a first protective sample with a thickness of 0.8~2 µm. Then, after rotating the first protective sample 90 degrees along the Y-axis and X-axis respectively, a second protective layer is prepared on the second surface and / or sixth surface of the first protective sample to obtain a second protective sample with a thickness of 1.5~5 µm.

[0040] It's easy to understand that compared to ion beam deposition of the protective layer, electron beam preparation of the first protective layer is faster and ensures non-destructive processing, thus guaranteeing the efficiency and quality of TEM sample preparation. However, because electron beam preparation of the protective layer is faster, taking the preferred Pt layer (hard and corrosion-resistant) as an example, when the thickness of the prepared protective layer is greater than 2 µm, the deposition time is longer, which will introduce excessive heat energy and cause the sample to bend.

[0041] Therefore, after preparing a first protective layer on the first, second, and sixth surfaces to obtain a first protective sample, the first protective sample is rotated 90 degrees sequentially along the Y-axis and X-axis, and then a second protective layer is prepared on the second and / or sixth surfaces of the first protective sample by ion beam deposition to obtain a second protective sample. This avoids sample damage and allows for the preparation of samples with the target thickness based on the required thickness.

[0042] In summary, an electron beam deposition layer is first prepared relative to the first protective layer. Electron beam deposition ensures non-destructive and efficient processing, but the time required cannot be too long. Therefore, ion beam deposition is superimposed, which keeps the overall time short (ensuring preparation efficiency) and ensures no damage. This allows for the preparation of a sufficiently thick second protective layer on the second and / or sixth surfaces, thereby ensuring the final preparation of an ultrathin TEM sample with a thickness of less than 40 nm. This allows for the observation of two longitudinal sections of the device under test, enabling observation of the device morphology from more dimensions and accurate analysis of its failure causes. This is beneficial for improving the accuracy and reliability of TEM failure analysis results for the device under test.

[0043] In some embodiments, a second longitudinal section is performed on the second protected sample using a focused ion beam to obtain a second longitudinal section TEM sample, including: Extract the target sample containing the target structure region from the second protective sample and fix the target sample on the support device; The target sample is subjected to a second longitudinal section using a focused ion beam to obtain a second longitudinal section TEM sample, wherein the incident direction of the focused ion beam is perpendicular to the second surface of the target sample.

[0044] The supporting device can be a copper mesh or a copper column.

[0045] The target sample covers a large area, and the target structural region is usually enclosed within it.

[0046] Please see Figure 5 The target sample is positioned in the same location as the second protective sample. A robotic arm can be used to extract the target sample, and its fourth surface can be fixed to a support device (such as the sidewall of a copper pillar) using electron beam deposition or ion beam deposition. Next, a focused ion beam is used to perform a second longitudinal sectioning of the target sample, obtaining a second longitudinal section TEM sample. The incident direction of the focused ion beam is perpendicular to the second surface of the target sample (the plane marked ②).

[0047] In some embodiments, a second longitudinal section is performed on the second protected sample using a focused ion beam to obtain a second longitudinal section TEM sample, including: The second protective sample is finely cut using a high-energy focused ion beam to obtain a finely cut longitudinal section sample; wherein the incident direction of the high-energy focused ion beam is perpendicular to the second surface of the second protective sample. A low-energy focused ion beam is used to refine the finely cut longitudinal section sample to obtain a second longitudinal section TEM sample; wherein the incident direction of the low-energy focused ion beam is perpendicular to the second surface of the second protective sample.

[0048] In some embodiments, in the step of finely slicing the second protective sample using a high-energy focused ion beam to obtain a finely sliced ​​longitudinal section sample, the voltage of the high-energy focused ion beam is 30 kV, the beam current is 300 pA, and the etching step size is 20-30 nm. The thickness of the finely sliced ​​longitudinal section sample is approximately 150 nm.

[0049] For example, a high-energy focused ion beam with a high voltage of 30 kV and a low beam current of 300 pA can be used to gradually etch the second protective sample with a step size of 20-30 nm. At the same time, SEM is used to observe in real time until the longitudinal section containing the target structure region is about to be exposed (the shadow of the target structure region can be vaguely seen through SEM electron beam observation), and the second protective sample is initially thinned to a certain thickness (usually around 150 nm).

[0050] In some embodiments, in the step of using a low-energy focused ion beam to refine a finely cut longitudinal section sample to obtain a second longitudinal section TEM sample, the voltage of the low-energy focused ion beam is 2-8 kV.

[0051] By using a low-energy focused ion beam to refine the finely cut longitudinal section sample, the thickness of the finely cut longitudinal section sample can be further reduced, the damage layer can be reduced, the residue generated by etching can be removed, and the flatness of the cross section can be ensured to meet the preset requirements, so as to obtain an ultrathin TEM sample (thickness below 40 nm).

[0052] The technical solution and effects of this application will be further described in detail below with reference to specific embodiments.

[0053] Example 1 The ultrathin TEM sample preparation steps provided in this embodiment are as follows: S1. The device under test is longitudinally sectioned using a focused ion beam to obtain the first longitudinal section TEM sample (e.g., Figure 9 (As shown), then, SEM scanning is performed on the first longitudinal section TEM sample. Combining the required slicing drawing and the morphological characteristics of the first longitudinal section TEM sample (such as device shape), the coordinate position of the target structural region is determined. Then, an electron beam is used to etch a high-precision marker reference point (such as...) in the target structural region of the first longitudinal section TEM sample. Figure 9 The “T”-shaped markings in the diagram ensure that the slicing path accurately samples the target structural region, avoiding incomplete longitudinal sections due to offset (such as only cutting to the edge of the target structural region or missing the target structural region). Next, a first protective layer (green part in the figure) is prepared on the target structural region on the first surface (upper side), second surface (front side), and sixth surface (rear side) of the first longitudinal section TEM sample, resulting in the first protected sample (e.g., Figure 10 (As shown).

[0054] S2, please refer to Figure 5 , Figures 10-11 After rotating the first protective sample, which has a first protective layer deposited thereon, 90 degrees clockwise along the Y-axis and X-axis respectively, a second protective layer (Pt layer) is prepared on the second surface of the first protective sample at the position corresponding to the target structure region using a low-voltage ion beam of 8 kV and a low current of 61 pA, thus obtaining the second protective sample (e.g., Figure 12 (As shown). By depositing a second protective layer on the second surface at the location corresponding to the target structural region, the mechanical strength of the target structural region can be increased, preventing damage or deformation of the target structural region due to ion beam bombardment during subsequent secondary longitudinal cutting processes.

[0055] S3. Use an ion beam to cut out the target sample containing the target structure region from the second protected sample, and use a robotic arm ( Figure 13 The "Lift-out" in the text refers to the robotic arm that extracts the target sample and transfers it to the copper column. Figure 13 The "Cu-grid" in the text refers to the sidewalls of a copper mesh / copper pillar, and is processed by ion beam deposition or electron beam deposition. Figure 13 In this context, "GIS" stands for ion beam or electron beam deposition. This fixes the fourth surface of the target sample to the sidewall of the copper pillar (e.g., ...). Figure 13 (As shown).

[0056] S4, Combination Figure 14 First, a high-energy ion beam (30 kV, 300 pA) is used to gradually etch the target sample in steps of 20–30 nm, while simultaneously observing it in real time using SEM, until the longitudinal section of the target structural region of the target sample is about to be exposed (the shadow of the target structural region can be faintly seen through SEM electron beam observation). The target sample is then initially thinned to a certain thickness (approximately 150 nm) to obtain a finely cut longitudinal section sample. Next, a low-energy ion beam (2–8 kV) is used to trim the finely cut longitudinal section sample to reduce the damage layer and remove etching residues, ensuring that the cross-sectional flatness meets the preset requirements, resulting in a second longitudinal section TEM sample (thickness 30–35 nm). The high-energy and low-energy ion beams are incident in the same direction, both perpendicular to the second surface of the target sample, which is parallel to the XY plane of the XYZ three-dimensional coordinate system.

[0057] S5. Place the second longitudinal section TEM sample obtained in step S4 into a dedicated TEM carrier. Use a TEM device to emit a high-energy electron beam to penetrate the second longitudinal section TEM sample (e.g., ...). Figure 14As shown, the copper pillar is located on the left side of the sample, and a high-energy electron beam penetrates the second longitudinal section of the TEM sample from front to back. Images of the device are acquired in bright-field, high-resolution, or STEM modes. Imaging effects are optimized by adjusting parameters such as accelerating voltage and objective aperture. Using TEM software or specialized analysis tools, dimensional parameters such as diameter, height, and structural morphology of the device are measured, and composition and distribution are analyzed using EDS and EELS. The images observed by the TEM are shown below. Figures 15-16 As shown. Among them, Figure 15 This is a TEM analysis image of the second longitudinal section TEM sample in Embodiment 1 of this application; Figure 16 It is by Figure 15 The TEM drawing image is obtained by converting the TEM analysis image corresponding to the target structure (red box part).

[0058] Comparative Example 1 The TEM sample preparation steps provided in this comparative example are as follows: S1 is the same as step S1 in Example 1.

[0059] S2, such as Figure 17 As shown, the first protective sample is rotated 90 degrees along the Z-axis, and the target structural region in the first protective sample is determined using an electron beam (low voltage 2~5 KV, low beam current 0.1~0.2 nA). A robotic arm is then used to extract the sample to be longitudinally cut, which contains the target structural region, and this sample is deposited and fixed onto the top of a copper pillar (e.g., Figure 18 (As shown).

[0060] S3, Combination Figure 19 First, a high-energy ion beam (30 kV, low current 300 pA) is used to gradually etch the sample to be longitudinally sectioned a second time, with a step size of 20-30 nm. Simultaneously, SEM is used for real-time observation until the longitudinal section of the target structural region of the sample is about to be exposed (the shadow of the target structural region can be faintly seen through SEM electron beam observation). The sample is then initially thinned to a certain thickness (approximately 150 nm) to obtain a finely cut longitudinal section sample. Next, a low-energy ion beam (2-8 kV) is used to refine the finely cut longitudinal section sample to reduce the damage layer and remove etching residues, ensuring that the cross-sectional flatness meets the preset requirements, resulting in a longitudinal section TEM sample (approximately 80 nm thick, which cannot be further thinned). The high-energy and low-energy ion beams are incident in the same direction, parallel to the second surface of the target sample, which is perpendicular to the XY plane of the XYZ three-dimensional coordinate system.

[0061] S4. Place the longitudinal section TEM sample obtained in step S3 onto a dedicated TEM carrier. Use the TEM equipment to emit a high-energy electron beam that penetrates the longitudinal section TEM sample, acquiring images of the device in bright-field, high-resolution, or STEM modes. Optimize the imaging effect by adjusting parameters such as accelerating voltage and objective aperture. Measure the device's diameter, height, structural morphology, and other dimensional parameters using the TEM's built-in software or specialized analysis tools, and analyze the composition and distribution using EDS and EELS. The images observed by the TEM are shown below. Figures 20-21 As shown. Among them, Figure 20 This is a TEM analysis image of the longitudinal section TEM sample of Comparative Example 1; Figure 21 It is by Figure 20 The TEM drawing image is obtained by converting the TEM analysis image corresponding to the target structure (red box part).

[0062] from Figure 18 , Figure 19 As can be seen, the sample to be longitudinally sectioned in Comparative Example 1 is a columnar structure with insufficient mechanical stress. During the secondary longitudinal sectioning process, the top, longitudinal section, and both sides of the sample deform due to ion beam bombardment, thus the sample thickness can only be reduced to 80 nm and cannot be further reduced. For some advanced process devices smaller than 70 nm, the preparation method of Comparative Example 1 cannot produce high-quality TEM samples for effective TEM analysis.

[0063] from Figures 13-14 As can be seen, the target sample in Example 1 has a block structure, which significantly increases mechanical stress. Simultaneously, the ion beam bombardment is only applied to the top and cross-section of the target sample, preventing deformation due to ion beam bombardment. This allows the sample thickness to be reduced from 80 nm to below 40 nm, and even below 20 nm. For some advanced process devices smaller than 80 nm, the preparation method in Example 1 can produce high-quality TEM samples for effective TEM analysis.

[0064] from Figures 15-16 , Figures 20-21 It can be seen that the morphological features of the target structural region of the TEM sample prepared by Example 1 are clear, and the thickness of the key target region of the sample is controllable, so effective TEM analysis can be performed; while the morphological features of the target structural region of the TEM sample prepared by Comparative Example 1 are relatively blurry, and the thickness of the key target region of the sample is uncontrollable, so effective TEM analysis cannot be performed.

[0065] Comparative Example 2 The TEM sample preparation steps provided in this comparative example are as follows: S1 is the same as step S1 in Example 1.

[0066] S2, please refer to Figure 11 After rotating the first protective sample 90 degrees clockwise along the Y-axis and X-axis respectively, the target sample containing the target structure region is cut from the second protective sample using an ion beam. The target sample is then extracted using a robotic arm and transferred to the sidewall of the copper pillar, and then deposited using either ion beam deposition or electron beam deposition. Figure 13 In this context, "GIS" stands for ion beam or electron beam deposition. This fixes the fourth surface of the target sample to the sidewall of the copper pillar (e.g., ...). Figure 13 (As shown).

[0067] S3. First, a high-energy ion beam (30 kV, low current 300 pA) is used to gradually etch the target sample in steps of 20–30 nm, while simultaneously observing it in real time using SEM, until the longitudinal section of the target structural region of the target sample is about to be exposed (the shadow of the target structural region can be faintly seen through SEM electron beam observation). The target sample is then initially thinned to a certain thickness (approximately 150 nm) to obtain a finely cut longitudinal section sample. Next, a low-energy ion beam (2–8 kV) is used to trim the finely cut longitudinal section sample to reduce the damaged layer and remove etching residues, ensuring that the cross-sectional flatness meets the preset requirements, resulting in a longitudinal section TEM sample (approximately 70 nm thick, which cannot be further thinned). The high-energy and low-energy ion beams are incident in the same direction, both perpendicular to the XY plane of the XYZ three-dimensional coordinate system.

[0068] S4. Referring to step S5 of Example 1, the longitudinal section TEM sample obtained in step S3 was subjected to TEM analysis. The results showed that the morphological features of the target structural region were relatively blurry, and the thickness of the sample in the key target region was uncontrollable, making it impossible to perform effective TEM analysis.

[0069] Comparative Example 3 The TEM sample preparation steps provided in this comparative example are as follows: S1 is the same as step S1 in Example 1.

[0070] S2, please refer to Figure 3 , Figure 17 and Figure 22 The first protective sample is rotated 90 degrees counterclockwise along the Z-axis. At this time, the third side (the plane marked ③) of the first protective sample is flipped to the front side, the second surface (the plane marked ②) is flipped to the right side, and the sixth surface (opposite to and parallel to the second surface) is flipped to the left side. Then, a second protective layer (Pt layer) is prepared at the position corresponding to the target structure region on the second surface and / or the sixth surface to obtain the third protective sample.

[0071] S3. Using an ion beam, a sample containing the target structure region is cut from the third protective sample to be longitudinally cut again. A robotic arm is used to extract this sample and transfer it to the sidewall of the copper pillar. The sample is then deposited and fixed to the top of the copper pillar (e.g., ...). Figure 18 (As shown).

[0072] S4, Combination Figure 19 First, a high-energy ion beam (30 kV, low current 300 pA) is used to gradually etch the sample to be longitudinally sectioned a second time, with a step size of 20-30 nm. Simultaneously, SEM is used for real-time observation until the longitudinal section of the target structural region of the sample is about to be exposed (the shadow of the target structural region can be faintly seen through SEM electron beam observation). The sample is then initially thinned to a certain thickness (approximately 150 nm) to obtain a finely cut longitudinal section sample. Next, a low-energy ion beam (2-8 kV) is used to refine the finely cut longitudinal section sample to reduce the damage layer and remove etching residues, ensuring that the cross-sectional flatness meets the preset requirements, resulting in a longitudinal section TEM sample (approximately 60 nm thick, which cannot be further thinned). The high-energy and low-energy ion beams are incident in the same direction, both perpendicular to the XY plane of the XYZ three-dimensional coordinate system. Figure 22 As shown, the high-energy ion beam and the low-energy ion beam have the same incident direction, both parallel to the second surface of the target sample, and the second surface of the target sample is perpendicular to the XY plane of the XYZ three-dimensional coordinate system.

[0073] S5. Referring to step S5 of Example 1, the longitudinal section TEM sample obtained in step S4 is subjected to TEM analysis. The results show that the morphological features of the target structural region are relatively blurry, and the thickness of the sample in the key target region is uncontrollable, making it impossible to perform effective TEM analysis.

[0074] The comparison results between Example 1 and Comparative Examples 1-3 show that Example 1 first prepares a first protective layer on the first, second, and sixth surfaces of the TEM sample in the first longitudinal section. Then, the first protective sample is rotated 90 degrees along the Y-axis and then 90 degrees along the X-axis, and a second protective layer is prepared on the second and / or sixth surfaces of the first protective sample. Finally, the ion beam incident direction is perpendicular to the second surface of the target sample for slicing and thinning, which can reduce the device under test to the desired thickness (below 40 nm). In contrast, Comparative Example 1 rotates the first protective sample 90 degrees along the Z-axis, does not prepare a second protective layer, and directly performs a second longitudinal sectioning process (slicing and thinning with the ion beam incident direction parallel to the second surface of the target sample). This can only reduce the device under test to about 80 nm, and cannot further reduce the device under test to the desired thickness (below 40 nm). In Comparative Example 2, after rotating the first protective sample 90 degrees clockwise along the Y-axis and X-axis respectively, without preparing a second protective layer, a second longitudinal sectioning process was performed (slicing and thinning with the ion beam incident direction perpendicular to the second surface of the target sample). This only reduced the device under test to about 70 nm, and could not be further reduced to the desired thickness (below 40 nm). In Comparative Example 3, after rotating the first protective sample 90 degrees counterclockwise along the Z-axis, a second protective layer was prepared on the second and / or sixth surface of the first protective sample, and then a second longitudinal sectioning process was performed (slicing and thinning with the ion beam incident direction parallel to the second surface of the target sample). This also only reduced the device under test to about 60 nm, and could not be further reduced to the desired thickness (below 40 nm). Therefore, only by following the technical solution provided in the embodiments of this application, firstly preparing a first protective layer on the first, second, and sixth surfaces of the first longitudinal section TEM sample, and then rotating the first protective sample 90 degrees along the Y-axis and then 90 degrees along the X-axis, and preparing a second protective layer on the second and / or sixth surfaces of the first protective sample, can the device under test be thinned to the desired thickness (below 40 nm), and both longitudinal sections of the device under test can be observed. This allows for observation of the device's morphology from more dimensions and accurate analysis of its failure causes, which is beneficial for improving the accuracy and reliability of the TEM failure analysis results of the device under test.

[0075] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for preparing ultrathin TEM samples, characterized in that, include: A three-dimensional XYZ coordinate system is established, wherein the XY plane of the XY coordinate system is parallel to the first surface of the device under test, the XZ plane is parallel to the second surface of the device under test, and the YZ plane is parallel to the third surface of the device under test; the XY plane intersects with and is perpendicular to the XZ and YZ planes, and the first surface intersects with and is perpendicular to the second and third surfaces. The device under test is subjected to a longitudinal section using a focused ion beam to obtain a first longitudinal section TEM sample; wherein the incident direction of the focused ion beam is parallel to the third surface of the device under test. The target structural region in the first longitudinal section TEM sample is identified and marked, and a first protective layer is prepared on the first surface, the second surface and the sixth surface of the first longitudinal section TEM sample to obtain a first protective sample; wherein the second surface and the sixth surface are opposite and parallel, and the target structural region is located in the space between the second surface and the sixth surface; After rotating the first protective sample 90 degrees along the Y-axis and X-axis respectively, a second protective layer is prepared on the second surface and / or the sixth surface of the first protective sample to obtain the second protective sample. The second protective sample is subjected to a second longitudinal section using a focused ion beam to obtain a second longitudinal section TEM sample; wherein the incident direction of the focused ion beam is perpendicular to the second surface of the second protective sample, and the thickness of the second longitudinal section TEM sample is less than 40 nm.

2. The method for preparing ultrathin TEM samples according to claim 1, characterized in that, A first protective layer is prepared on the first surface, the second surface, and the sixth surface of the first longitudinal section TEM sample to obtain a first protected sample, comprising: A first protective layer was prepared on the first, second, and sixth surfaces of the first longitudinal section TEM sample using electron beam deposition, wherein the voltage was 2 kV and the current was 1~2.5 nA.

3. The method for preparing ultrathin TEM samples according to claim 2, characterized in that, The first protective layer is a platinum layer; the thickness of the first protective layer is 0.8~2 µm.

4. The method for preparing ultrathin TEM samples according to claim 1, characterized in that, A second protective layer is prepared on the second or sixth surface of the first protective sample to obtain a second protective sample, comprising: A second protective layer electron beam is prepared on the second or sixth surface of the first protective sample using ion beam deposition, wherein the voltage is 8 kV and the current is 61 pA.

5. The method for preparing ultrathin TEM samples according to claim 4, characterized in that, The second protective layer is a platinum layer; the thickness of the second protective layer is 1.5~5 µm.

6. The method for preparing ultrathin TEM samples according to claim 1, characterized in that, The second protected sample was subjected to a second longitudinal section using a focused ion beam to obtain a second longitudinal section TEM sample, including: Extract the target sample containing the target structure region from the second protected sample, and fix the target sample on the support device; The target sample is subjected to a second longitudinal section using a focused ion beam to obtain a second longitudinal section TEM sample, wherein the incident direction of the focused ion beam is perpendicular to the second surface of the target sample.

7. The method for preparing ultrathin TEM samples according to claim 1, characterized in that, The second protected sample was subjected to a second longitudinal section using a focused ion beam to obtain a second longitudinal section TEM sample, including: The second protective sample is finely slicing using a high-energy focused ion beam to obtain a finely sliced ​​longitudinal section sample; wherein the incident direction of the high-energy focused ion beam is perpendicular to the second surface of the second protective sample. The finely cut longitudinal section sample is refined using a low-energy focused ion beam to obtain a second longitudinal section TEM sample; wherein the incident direction of the low-energy focused ion beam is perpendicular to the second surface of the second protective sample.

8. The method for preparing ultrathin TEM samples according to claim 7, characterized in that, In the step of using a high-energy focused ion beam to finely cut the second protected sample to obtain a finely cut longitudinal section sample, the voltage of the high-energy focused ion beam is 30 kV, the beam current is 300 pA, and the etching step size is 20~30 nm.

9. The method for preparing ultrathin TEM samples according to claim 7, characterized in that, In the step of using a low-energy focused ion beam to refine the finely cut longitudinal section sample to obtain a second longitudinal section TEM sample, the voltage of the low-energy focused ion beam is 2-8 kV.

10. The method for preparing ultrathin TEM samples according to claim 7, characterized in that, The thickness of the finely cut longitudinal section sample is 150 nm.

Images