Method and system for backside plan view sheet preparation
1. A method for processing a sample comprising at least a substrate layer and a device layer using a charged particle beam, comprising:
removing at least a portion of the substrate layer to obtain a sample surface;
scanning a region of interest (ROI) associated with the sample surface with an electron beam;
flowing a first gas to the ROI to spontaneously etch the scanned ROI; and
scanning the etched ROI with the electron beam in response to the device layer not being exposed in the ROI.
2. The method of claim 1, further comprising: acquiring a sample image including the etched ROI, and determining that the device layer is not exposed in the ROI based on the sample image.
3. The method of claim 2, wherein the sample image is a Scanning Electron Microscope (SEM) image and a beam current of the electron beam used to scan the ROI is higher than a beam current used to acquire the sample image.
4. The method of claim 2, wherein the sample image is an SEM image and a beam energy of the electron beam used to scan the ROI is higher than a beam energy used to acquire the sample image.
5. The method of claim 2, wherein a duration for scanning the ROI is longer than a duration for acquiring the sample image.
6. The method of any of claims 2-5, wherein determining that the device layer is not exposed in the ROI based on the sample image comprises determining that the etched ROI is not flat based on the sample image.
7. The method of any one of claims 2 to 5, wherein determining, based on the sample image, that the device layer is not exposed to the ROI comprises determining that a ratio between an area of the device layer exposed to the ROI and an area of the ROI is not greater than a threshold ratio.
8. The method of claim 1, wherein scanning the ROI with the electron beam comprises scanning the ROI with the electron beam so as to deposit or dope the ROI with carbon.
9. The method of claim 1 or 8, wherein flowing the first gas to the sample surface comprises flowing the first gas for a predetermined first predetermined duration.
10. The method of claim 9, wherein removing at least a portion of the substrate layer comprises flowing a second gas toward the sample surface for a second duration to etch the sample surface, the second duration being longer than the first duration.
11. The method of claim 1, wherein removing at least a portion of the substrate layer comprises milling the substrate layer with a focused ion beam.
12. A method for producing a plan view lamella using a charged particle beam, comprising:
extracting a sample from a workpiece with a focused ion beam, the sample comprising at least a device layer and a substrate layer;
removing at least a portion of the substrate layer from the back side of the sample to obtain a sample surface; and
alternately scanning a region of interest (ROI) associated with the sample surface with an electron beam and flowing a gas toward the scanned ROI until the device layer is exposed within the ROI, wherein the scanned ROI is spontaneously etched by the gas.
13. The method of claim 12, further comprising: determining that the device layer is exposed within the ROI in response to a ratio between an area of the device layer exposed in the ROI and an area of the ROI being greater than a threshold ratio.
14. The method of claim 12 or 13, wherein the sample comprises a plurality of device layers, and the exposed device layer is a last device layer of the plurality of device layers; and the method further comprises: removing at least one device layer of the plurality of device layers from the back side of the sample after the last device layer is exposed within the ROI.
15. The method of claim 12, further comprising: forming the plan view sheet based on the sample with the exposed device layers; and imaging the plan view lamella using a transmission electron microscope.
16. The method of claim 12 or 15, further comprising: removing at least one device layer of the plurality of device layers from a front side of the sample, the front side being opposite the back side.
17. A system for processing a sample comprising at least a substrate layer and a device layer, comprising:
a first column for forming a focused ion beam;
a second column for forming an electron beam;
a lower chamber coupled with both the first column and the second column;
a gas supply system coupled to the lower chamber; and
a controller having instructions stored in a non-transitory memory, the controller configured to:
removing at least a portion of the substrate layer to obtain a sample surface;
scanning a region of interest (ROI) associated with the sample surface with the electron beam;
flowing gas to the ROI via the gas supply system to spontaneously etch the scanned ROI; and is
Scanning the etched ROI with the electron beam in response to the device layer not being exposed in the ROI.
18. The system of claim 17, wherein the controller comprises instructions such that the sample is further milled from a workpiece using the focused ion beam prior to removing the at least a portion of the substrate layer to obtain the sample surface.
19. The system of claim 17, wherein the sample comprises a plurality of device layers, the device layers being final device layers adjacent to the substrate layer, and the controller is further configured to remove one or more device layers with the focused ion beam in the presence of an etch assist gas after the final device layers in the ROI are exposed.
20. The system of any one of claims 17 to 19, wherein the controller is further configured to flow the gas to the scanned ROI after scanning the etched ROI with the electron beam.
Background
Transmission Electron Microscopy (TEM) requires that the sample be sufficiently thin so that electrons transmitted through the sample can be used to form an image. One method of preparing TEM samples is to extract the sample by milling the workpiece using a Focused Ion Beam (FIB). The extracted sample was then thinned from the front and back to form a thin TEM sample, i.e., a thin sheet. In order to perform failure analysis on a microelectronic device, such as a microelectronic device having a 3D-NAND structure, a plan view sheet may be prepared. The plan view sheet has front and back surfaces that extend parallel to the device layers and can be used to view specific layers of the 3D-NAND structure.
Franco et al, US 2018/0350558A1, disclose a method of making a flat view sheet. Wherein the back side of the lifted sample is thinned using FIB with or without the presence of an etch assist gas. Applicants recognized, however, that to ensure that the back side of the sheet is parallel to the device layer, the sample back side surface formed prior to thinning with FIB must be parallel to the device layer. One way to ensure that the sample backside surface is parallel to the device layer is to use, for example, XeF2The etching gas spontaneously etches the back side of the sample. XeF2The silicon substrate may be selectively etched while leaving the device layer. However, applicants recognized that when preparing samples having a 3D-NAND structure, XeF2The polysilicon in the device layer may be over etched. As a result, the etched surface may be uneven. Furthermore, device layers adjacent to the substrate layer may be damaged during the etching process.
Disclosure of Invention
In one embodiment, a method for processing a sample comprising at least a substrate layer and a device layer using a charged particle beam comprises: removing at least a portion of the substrate layer to obtain a sample surface; scanning a region of interest (ROI) associated with a surface of the sample with an electron beam; flowing a first gas to the ROI to spontaneously etch the scanned ROI; and scanning the etched ROI with an electron beam in response to the device layer not being exposed in the ROI. In this way, a plan view sheet having a flat back surface can be formed from the processed sample.
It should be understood that the summary above is provided to introduce in simplified form some concepts that are further described in the detailed description. It is not meant to identify key or essential features of the claimed subject matter, the scope of which is defined uniquely by the claims that follow the detailed description. Furthermore, the claimed subject matter is not limited to implementations that solve any disadvantages noted above or in any part of this disclosure.
Drawings
FIG. 1 illustrates a dual beam system according to some embodiments of the inventions.
Fig. 2 illustrates a method for forming a sheet.
FIG. 3A shows a workpiece having multiple device layers.
Fig. 3B shows a sample taken from a workpiece.
Fig. 3C shows the extracted sample.
FIG. 4 shows a portion of a device layer of a 3D-NAND structure.
Fig. 5 is a flow chart for processing the backside of a sample extracted from a workpiece.
Fig. 6 is an image of the back side of a sample with the last device layer exposed.
Fig. 7 shows signals received during a layer reduction process.
Like reference numerals refer to corresponding parts throughout the several views of the drawings.
Detailed Description
The following description relates to systems and methods for preparing the backside of a planar view lamella for imaging in a Transmission Electron Microscope (TEM). The preparation may be performed in the dual beam system shown in fig. 1. The dual beam system has a first column for forming an ion beam and a second column for forming an electron beam.
A particular device layer of the workpiece may be inspected based on a high resolution TEM image of a plan view slice prepared from the workpiece. In one example, a workpiece having a 3D-NAND structure may include tens of layers of Integrated Circuit (IC) chips fabricated on a silicon substrate. Fig. 4 is a TEM image showing a part of a device layer of the 3D-NAND structure. X-y cross sections of a plurality of vertical structures can be observed in the TEM images. Each memory cell 400 includes concentric layers of materials including silicon dioxide, polysilicon, and silicon nitride. The vertical structure extends in the z-direction. The device layers in the 3D-NAND structure may be separated from each other by a spacer material such as silicon dioxide. A metric of the device structure (e.g., memory cell) in a particular device layer can be determined from the TEM image.
The plan view sheet may be formed according to the method shown in fig. 2. A sample is milled and extracted from the workpiece. As shown in fig. 3A-3B, the multiple device layers in the extracted sample extend parallel to the surface of the workpiece. The front surface of the extracted sample, as shown in fig. 3C, is a portion of the top surface of the workpiece. On the back side of the extracted sample, the final or bottom device layer is embedded under at least one substrate (e.g., a silicon substrate). In some examples, a spacer layer, such as a silicon dioxide layer, may be located between the substrate layer and the last device layer. There are no other device layers between the last device layer and the substrate layer. Material on the back and front sides of the extracted sample is removed to obtain a planar view slice with specific device layers exposed from both the back and front sides within a region of interest (ROI). For example, to obtain the backside of the plan-view slice, the material covering the last device layer (e.g., substrate and spacer layers) is removed to expose the last device layer. One or more device layers may then be removed from the back side of the sample by a subtractive process.
To form a planar view slice with a flat and large ROI, the exposed last device layer must be flat and parallel to the device layers. A method for removing the substrate is to use XeF2The sample backside was etched spontaneously. Due to XeF2The silicon is selectively etched so that a portion of the final device layer may remain after etching. However, it is difficult to control XeF2Exposure time of etching: short exposure times can result in under-etching of the silicon substrate; while long exposure times may overetch some materials, such as polysilicon in the device layer. Underetching can reduce the area of the exposed last device layer, thereby reducing the ROI. On the other hand, over-etching may result in unevenness and damage of the device layer. For example, device layers close to the silicon substrate, such as 0-5 device layers far from the silicon substrate, cannot be observed.
One method for preparing the flat surface is shown in FIG. 5Method of viewing the back of a sheet. The back side of the sample may be thinned globally using Focused Ion Beam (FIB) and/or spontaneous etching using an etching gas. Alternately scanning the ROI relative to the back side of the sample with a high current electron beam and using XeF when the exposed back sample surface is close to the final device layer2And (4) spontaneous etching. When XeF is removed from the lower chamber2Performing electron beam scanning and performing XeF without electron beam irradiation2And (6) etching. The scan-etch sequence may be performed in multiple iterations until most of the final device layers within the ROI are exposed. Here, when a device or device structure (e.g., a memory cell in a 3D-NAND structure) in the last device layer is exposed, the last device layer is exposed. In one example, the iteration can be terminated when a ratio between an area of the last device layer exposed in the ROI and a total area of the ROI is greater than a threshold ratio. In another example, the iteration can be terminated when the exposed final device layer is flat and planar. The duration of the spontaneous etch within each iteration may be constant and predetermined. Scan-etch iterations remove material covering the last device layer on the back of the sample. If a spacer layer is present between the substrate layer and the last device layer, both the substrate layer and the spacer layer are removed during the iteration. After the last device layer is exposed, one or more device layers on the back side of the sample may be removed by a subtractive process to obtain a planar view foil back side.
During backside plan view sheet preparation, SEM images of the backside of the sample may be taken for monitoring the preparation process. The beam current for SEM imaging is lower than the beam current for scanning the ROI in the scan-etch sequence. The beam current for SEM imaging can be below 1 nA, while the beam current for ROI scanning can be above 1 nA. In one example, the beam current for SEM imaging is 100 pA to 1 nA, and the beam current for ROI scanning is 2-4 nA. Furthermore, to observe the sample surface, the beam energy used for SEM imaging may be lower than the beam energy used to scan the ROI in the scan-etch sequence. For example, the beam energy for SEM imaging is below 10 kV (e.g., 2-10 kV), while the beam current for ROI scanning is above 10 kV (e.g., 10-20 kV).
By using high current electronsThe beam scans the ROI, which may be deposited and/or doped with carbon. In some examples, a carbon source may be provided in the lower chamber. By using e-beam deposition/doping of the ROI, a preferential etch rate between the device structure in the device layer and the material in the other layers (e.g., substrate layer and spacer layer) is achieved during the etch process immediately after scanning. For example, in XeF2During etching, the silicon substrate is etched at a higher rate than the device structure. The preferential etch rate may be caused by a thin carbon layer preferentially deposited over the non-uniform device material of the device structure, as compared to a relatively uniform substrate layer during electron beam scanning. In addition, electron beam scanning may result in enhanced adsorption of carbon species onto the device layer. The intrinsic mechanism of this enhancement may be due to material differences, topological considerations, or a combination thereof. Due to the preferential etch rate, material overlying the last device layer (e.g., substrate and spacer layers) may be selectively removed during scan-etch iterations while preserving device structure near the substrate layer. As a result, a large and flat final device layer is exposed in the ROI.
Fig. 6 is an SEM image of the back side of a sample with the last device layer exposed. Samples with exposed final device layers can be debulked using FIB with or without an etch assist gas to obtain a planar view wafer backside. The sample backside surface formed using the method of fig. 2 is flat and parallel to the device layer, as reflected by the signals received during the backside subtractive process shown in fig. 7 below.
Turning to fig. 1, fig. 1 shows a dual beam system 110. A dual beam system includes an ion beam that is vertical or tilted a few degrees relative to the plane of the workpiece and an electron beam having an axis that is also tilted, e.g., 52 degrees, relative to the axis of the ion beam. In some embodiments, the ion beam and the electron beam are aligned such that the fields of view of the two beams coincide to within a few microns. Both the ion beam and the electron beam may be used to image and/or process the sample.
The dual beam system 110 includes an ion column 111 for generating an ion beam. The ion column 111 includes an ion source 114, an extraction electrode 115, an electrostatic optical system 117, and an electrostatic deflection plate 120. An ion beam 118 generated by the ion source 114 isThe workpiece 122 is deflected by the electrostatic deflector 120 prior to irradiation. The workpiece 122 is positioned on a movable X-Y-Z stage 124 within a lower chamber 126. The lower chamber 126 may be evacuated using a turbomolecular and mechanical pumping system 168 under the control of the pumping controller 130. The vacuum system provides, for example, about 5 x 10 in the lower chamber 126-8Bracket and 5 x 10-4Vacuum between the trays. When etch assist, etch delay, or deposition precursor gases are used, the chamber background pressure may be raised, for example, to about 1 × 10-5And (4) supporting.
A high voltage power supply 134 is connected to the ion source 114 and appropriate electrodes in the ion column 111 for forming and directing the ion beam 118 downward. A deflection controller and amplifier 136, operating according to a prescribed pattern provided by a pattern generator 138, is coupled to the deflection plate 120, whereby the ion beam 118 can be controlled to trace a corresponding pattern on the workpiece 122. In some systems, the deflection plate is placed before the final lens.
The ion source 114 typically provides a metal ion beam of gallium, but other ion sources, such as multi-spike or other plasma ion sources, may be used. The ion source 114 is typically capable of focusing into a sub-tenth micron wide beam at the workpiece 122 for modifying the workpiece 122 by ion milling, enhanced etching, material deposition, or for imaging the workpiece 122. A charged particle multiplier 140 for detecting secondary ions, or for example a secondary electron detector 140 as a detector for detecting secondary electron emissions for imaging, is connected to a signal processor 142 in which the signal from the charged particle multiplier 140 is amplified, converted to a digital signal, and signal processed. The resulting digital signal will display an image of the workpiece 122 on the monitor 144.
An electron column 141 is also provided along with a power supply and control unit 145 with the dual electron beam system 110. By applying a voltage between the cathode 152 and the anode 154, an electron beam 143 is emitted from the cathode 152. The electron beam 143 is focused to a fine spot by means of a condenser lens 156 and an objective lens 158. The electron beam 143 is scanned in two dimensions over the workpiece by means of deflection coils 160. The operation of the condenser lens 156, the objective lens 158 and the deflection coil 160 is controlled by the power supply and control unit 145. When the electrons in the electron beam 143 hit the surface of the workpiece 122, secondary electrons and backscattered electrons are emitted. Accordingly, these electrons are detected by SE detector 1240 or backscattered electron detector 162. The analog signal generated by SE detector 140 or backscattered electron detector 162 is amplified and converted to digital luminance values by signal processor unit 142. The resulting digital signal may be displayed on the monitor 144 as an image of the workpiece 122.
The micromanipulator 147 may include a precision motor 148 to provide X, Y, Z and theta control of a portion 149 located within the vacuum chamber. The micromanipulator 147 may be equipped with different end effectors for manipulating small objects, such as lifting a sample cut from the workpiece 122. In the embodiments described herein, the end effector is a stylet 150.
The door 170 is opened for insertion of the workpiece 122 onto the X-Y stage 124, which may be heated or cooled, and if used, for maintenance of the internal air supply reservoir. The doors are interlocked so that they cannot be opened if the system is under vacuum. A gas delivery system 146 extends into the lower chamber 126 for introducing and directing gaseous vapors toward the workpiece 122.
A system controller 119 controls the operation of the various parts of the dual beam system 110. Through the system controller 119, a user may scan the ion beam 118 or the electron beam 143 in a desired manner through commands input into a user interface (not shown). The system controller 119 may also contain computer readable memory 121 and may control the dual beam system 110 according to data or computer readable instructions stored in the memory 121 to implement the methods described herein.
The above-described apparatus and system may utilize a high-precision beam placement method for local navigation. Furthermore, it should be recognized that elements, aspects, or embodiments may be implemented via computer hardware or software, or a combination of both. The methods may be implemented in a computer program using standard programming techniques, including a computer-readable storage medium configured with the computer program, where the storage medium so configured causes a computer to operate in a specific and predefined manner according to the methods and figures described in this specification. Each program may be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language. Also, the program may be run on an application specific integrated circuit programmed for this purpose.
Further, the method may be implemented in any type of computing platform, including but not limited to a personal computer, a microcomputer, a mainframe, a workstation, a network or distributed computing environment, a computer platform separate from, integrated with, or in communication with a charged particle tool or other imaging device, and the like. Aspects may be embodied in machine-readable code stored on a storage medium or device, whether removable or integrated with a computing platform, such as a hard disk, optical read and/or write storage media, RAM, ROM, etc., so that the storage medium or device may be read by a programmable computer for configuring and operating the computer when the storage medium or device is read by the computer to perform the procedures described herein. Also, the machine-readable code or portions thereof may be transmitted over a wired or wireless network. When such media contain instructions or programs for implementing the above-described steps in conjunction with a microprocessor or other data processor, the embodiments described herein may include these and other various types of computer-readable storage media. Embodiments may also include the computer itself when programmed according to the methods and techniques described herein.
A computer program may be applied to input data to perform the functions described herein to transform the input data to generate output data. The output information is applied to one or more output devices, such as a display monitor. In some implementations, the transformed data may represent physical and tangible objects, including generating a particular visual depiction of the physical and tangible objects on a display.
As indicated, some embodiments may also utilize a charged particle beam, such as an ion beam or an electron beam, in order to image the sample using the particle beam. Such charged particles used to image a sample may inherently interact with the sample, resulting in some degree of physical transformation. Moreover, throughout this specification, discussions utilizing terms such as "computing," "determining," "measuring," "generating," "detecting," "forming," or the like, also refer to the action and processes of a computer system, or similar electronic device, that manipulates and transforms data represented as physical quantities within the computer system into other data similarly represented as physical quantities within the computer system or other information storage, transmission or display devices.
Ion beams and electron beams are described herein as examples of charged particle beams used to image or process a workpiece. Other charged particle beams may be used, such as a laser beam, or some other shape of ion beam, such as an ion beam from a liquid metal ion source.
FIG. 2 illustrates a method 200 of preparing a plan view lamina using the dual beam system of FIG. 1. The plan view sheet is prepared for inspecting a device layer of a workpiece including a plurality of device layers. The device layer may have a 3D-NAND structure.
At 202, a sample is extracted from a workpiece. The sample may be milled from the workpiece first using the FIB and then lifted/extracted from the workpiece for further processing using a micromanipulator (such as micromanipulator 147 of fig. 1). Fig. 3A-3C illustrate the process of sample extraction.
In fig. 3A-3B, the workpiece 310 is oriented with a length and width extending in an x-y plane and a height extending along a z-axis. The workpiece includes one or more device layers fabricated over a substrate 307. The substrate 307 may be silicon. The device layer extends in the x-y plane. The device layers may be separated from each other by spacer layers. The spacer layer may be formed of silicon dioxide. In FIG. 3A, three device layers (301-303) are shown as an example. For a 3D-NAND structure, the number of device layers may be tens of layers or more than a hundred layers. The top surface 308 of the workpiece 310 on the front side is parallel to the device layers. In one example, the last or bottom device layer 303 is in direct contact with the substrate 307. In another example, the last device layer 303 is separated from the substrate 307 by a spacer layer. The front side of the workpiece is indicated by arrow 305 (z direction) and the back side of the workpiece is indicated by arrow 306 (opposite to the z direction).
Fig. 3B illustrates one method for cutting a sample 330 from a workpiece 310. The workpiece 310 is first undercut from the opposite direction by two intersecting ion beam cuts 321 and 322. The ion beam then cuts sides 323 and 324. The sample 330 may be lifted from the workpiece 310 by the probe for further processing.
Fig. 3C shows an extracted or lifted sample 330. The sample 330 includes a plurality of device layers (331-333) on the front side 305. The front surface from which the sample is taken is part of the top surface 308 of the workpiece 310. Sample 330 includes at least one substrate layer 334 on back side 306. The wedge-shaped substrate layer 334 must be removed to expose the final device layer. The sample with the exposed last device layer may also be de-laminated to remove one or more device layers from the front and/or back side to obtain a planar view slice.
Turning back to fig. 2, at 204, the back side of the extracted sample is processed to form the back side of the plan view sheet. The back side process removes the wedge-shaped substrate layer (e.g., substrate layer 334 of fig. 3C) and exposes the last level device layer (e.g., last level device layer 333 of fig. 3C) adjacent to the substrate layer. The wafer backside preparation may also include using FIB to de-layer one or more device layers from the sample backside with or without an etch assist gas. The details of the backside processing are described in fig. 5.
At 206, the front side of the extracted sample is processed to form the front side of the plan view sheet. Front side processing may include reducing or removing a predetermined number of device layers from the front side using FIB. An etch assist gas may optionally be provided with the FIB. In one example, a predetermined number of device layers are removed to expose a particular device layer embedded in the sample using a facing FIB.
In some examples, step 206 is omitted and only the sample backside is processed to inspect the top device layer.
At 208, a high resolution TEM image of the plane view slice is acquired. Metrics for particular device layers may be determined based on the TEM image. The TEM images may be acquired in the dual beam system of fig. 1, or alternatively in a different TEM system.
In this manner, the device layers in the planar view sheet are parallel to the back surface of the sheet and the front surface of the sheet. The plan view lamellae provide a larger ROI in which certain device layers can be imaged. Further, the particular device layer may be a device layer proximate to the substrate layer.
Fig. 5 illustrates a method 500 for processing the back side of a lift-off sample to form the back side of a plan view sheet. First, the lift-off sample is thinned entirely by milling and/or etching. When the exposed sample back surface is close to the last level device layer, the sample back within the ROI is alternately scanned and spontaneously etched using a high current electron beam until most of the last level device layer within the ROI is exposed. One or more device layers may then optionally be removed from the back side of the sample.
At 501, the back side of the sample is thinned entirely to obtain the sample back side surface. The sample may be thinned globally using FIB milling and/or spontaneous etching. The bulk thinning process can remove a majority of the wedge-shaped substrate layer (such as substrate layer 334 of fig. 3C) on the back of the sample and provide a flat surface for further processing. Step 502-508 illustrates an example workflow for bulk thinning of the backside of a sample.
At 502, the back side of the extracted sample is milled using FIB. Milling can be performed using on-edge FIB, where the angle of the incident ion beam to the sample surface is less than 45 degrees. The sample backside may be milled using FIB with or without an etch assist gas. The milling process can be monitored based on an image of the back side of the sample. In one example, the image is formed by secondary electrons collected during milling. In another example, the image may be taken after the ion beam is scanned across the back surface of the sample. FIB milling may be terminated when the milled surface is proximate to the final device layer. For example, FIB milling is terminated when the milled surface is within a first threshold distance from the final device layer. In another example, the FIB milling is terminated after removal of the first thickness of the substrate. The thickness of the first substrate may be determined based on the total thickness of the sample and the thickness of the device layer.
At 504, XeF2Flowed to the back of the sample to spontaneously etch the silicon substrate without the assistance of a charged particle beam. In one example, XeF2And/or amount of and/or for XeF2The duration of the flow may be determined based on the second thickness of the substrate to be removed. The second thickness of the substrate to be removed may be estimated based on the thickness of the removed substrate, the total sample thickness, and the thickness of the device layer.
At 506, XeF is caused to2After flowing to the sample, SEM images of the back surface of the sample were taken to monitor the spontaneous etching process. SEM images can be taken with beam currents below 1 nA. For example, the beam current for SEM imaging is 100 pA. When the chamber pressure is reduced and XeF is removed from the lower chamber2At this time, SEM images were taken.
At 508, the method 500 determines whether the exposed sample backside is within a second threshold distance from the last device layer based on the image acquired at 506. At 502, the second threshold distance is less than the first threshold distance. In some examples, the second threshold distance may be zero. That is, due to XeF2Etching exposes a portion of the last device layer. The spontaneous etching is terminated if the exposed sample backside is within a second threshold distance from the last device layer. Otherwise, provide XeF2To further etch the back side of the sample.
At 510, method 500 checks whether the final device layer is exposed in the ROI relative to the sample back surface. The area of the ROI can be smaller than the area of the sample back surface. For example, as shown in FIG. 6, the area of the ROI 601 is less than the cross-sectional area of the workpiece in the x-y plane. Thus, the plan view foil has at least one thick edge for supporting the thin, reduced thickness region of the foil. In one example, the ROI may be determined based on an image of the back surface of the sample. The ROI can be a region of the sample back surface that is non-uniform or non-planar.
The last device layer within the ROI is exposed when the devices or device structures of the last device layer are not covered by the substrate or spacer layers. In one example, the last device layer A when exposed in the ROIDevice layerArea and ROI A ofROIIs greater than a threshold ratio, the final device layer is exposed. Final device layer A when exposed in ROIDevice layerArea and ROI A ofROIIs not greater than a threshold ratioExposing the final device layer. The area of the last device layer exposed within the ROI can be estimated based on the most recently acquired image of the back side of the sample. The image may be an SEM image acquired with a low beam current such as a beam current below 1 nA. The image may be taken at step 506 or step 516. In one example, the threshold ratio is greater than 90%. In another example, the threshold ratio is greater than 95%. If the last device layer is exposed in the ROI, method 500 moves to 518 to further de-layer one or more device layers from the back side of the sample. If the last device layer is not exposed in the ROI, method 500 moves to 512. The ROI is alternately scanned with a high current electron beam and spontaneously etched until most of the device final layer is exposed.
In another example, the last device layer is exposed when the last device layer within the ROI is flat. The flatness may be determined based on the most recently acquired SEM image. For example, any variation in surface uniformity observed in SEM images may indicate unevenness.
At 512, an ROI associated with the back surface of the sample is scanned with a high current electron beam. By scanning the ROI with an electron beam, carbon can be deposited or doped into the scanned area. At 506 or 516, the beam current for scanning the ROI is higher than the electron beam current for SEM imaging. For example, the beam current for ROI scanning is greater than 1 nA, while the beam current for imaging is less than 1 nA. As an example, the electron beam energy for scanning the ROI was 10 kV, and the beam current was 3.2 nA. At 506 and 516, the beam energy used to scan the ROI is higher than the beam energy used for SEM imaging. At 506 and 516, the duration for scanning the ROI is longer than the duration for acquiring the SEM image. In one example, the ROI is scanned without detecting scattered electrons. That is, an SEM image is not formed by scanning the ROI with an electron beam. In another example, an SEM image of the ROI is acquired based on scattered electrons received during the ROI scan.
At 514, the scanned ROI is spontaneously etched. Scanned ROI utilization XeF2Spontaneous etching is performed for a predetermined duration. The duration of the etching process can be determined by the interaction cross section of the XeF2 molecules with the substrate material. In one example, the utilization heightThe current electron beam scans the ROI for one minute. The scanned ROI was then etched spontaneously for one minute. XeF in step 5142The duration of the stream may be shorter than XeF in 5042The duration of the flow.
At 516, an image is taken that includes the etched ROI. The image may be an SEM image taken at parameters similar to the SEM image taken at 506. For example, the beam current and energy are lower than the beam current and energy used for ROI scanning in 512. The beam energy is reduced to reduce the electron penetration depth, thereby exciting more surface electrons to better view the etched surface.
At 518, device layers within the ROI are optionally reduced using FIB to remove a predetermined number of layers. The subtractive process may be performed by scanning the ROI with a facing FIB with or without an etch assist gas. The angle between the facing FIB and the bare sample surface within the ROI is greater than 45 degrees. The etch assist gas may be Methyl Nitroacetate (MNA) or an MNA-like gas. In one example, the etch assist gas comprises methyl nitroacetate. In other examples, the etch assist gas is a combination of one or more of methyl acetate, ethyl nitroacetate, propyl acetate, propyl nitroacetate, ethyl nitroacetate, methyl oxoacetate, or methyl oxoacetyl chloride. The subtractive process may be monitored based on secondary electrons or level currents recorded during each FIB scan of the ROI.
In this way, a large area of the final device layer can be exposed. The exposed final device layer is planar due to the selective etching of the carbon deposited/doped surface. Furthermore, damage to the device layer close to the substrate layer is prevented.
FIG. 6 shows an SEM image of a sample backside of the last device layer with a 3D-NAND structure exposed in the ROI 601 after multiple scan-etch iterations (steps 510-514 of FIG. 5). The pattern of the final device layer can be seen with high contrast in a large area ROI. Arrow 602 points to a region near the edge of ROI 601 covered by the silicon substrate. A subtractive layer may be further performed within the ROI 601 to remove one or more device layers.
Fig. 7 shows a graph generated based on secondary electrons and stage current collected from the back of the sample during the subtractive process. The sample includes multiple device layers of a 3D-NAND structure. The device layers are removed using FIB from the last device layer. The gray curve 701 is the total gray level of an image formed by secondary electrons. The higher the gray level, the smaller the number of secondary electrons corresponding to the collection. The level current curve 702 is the total level current sensed during each scan of the FIB over the ROI.
Both the gradation curve 701 and the level current curve 702 decrease from the first peak at the time point of 4 seconds and then increase again from the time point of 60 seconds or so. The first peak around the 4 second time point corresponds to the subtractive layer of the final device layer. The second peak around the 80 second time point corresponds to the subtractive layer for the next to last device layer. Curves 701 and 702 decrease from 4 seconds to 23 seconds as the subtractive process progresses in the final device layer. The low gray scale and level currents from 23 seconds to 65 seconds indicate that the spacer layer of the 3D-NAND structure is being removed. The smooth rise of both curves from 60 to 80 seconds to the second peak indicates that the sample surface is flat and parallel to the device layer.
The technical effect of scanning the ROI with an electron beam prior to spontaneous etching of the sample is to reduce the etch rate of at least a portion of the device layer and prevent over-etching of the final device layer. The technical effect of scanning the ROI with a high current electron beam is to deposit/dope the ROI with carbon. The technical effect of imaging the sample surface with a low energy electron beam during the sheet back side processing is to obtain an image of low penetration depth. The technical effect of repeating the scan-etch process is that the substrate layer is selectively etched while the device remains in the device layer.
In one embodiment, a method for processing a sample comprising at least a substrate layer and a device layer using a charged particle beam, the method comprising: removing a portion of the substrate layer to obtain a sample surface; scanning a region of interest (ROI) associated with a surface of the sample with an electron beam; flowing a first gas to the ROI to spontaneously etch the scanned ROI; and scanning the etched ROI with an electron beam in response to the device layer not being exposed in the ROI. In a first example of the method, the method further includes acquiring a sample image including the etched ROI, and determining that the device layer is not exposed in the ROI based on the sample image. A second example of the method optionally includes the first example, and further comprising, wherein the sample image is a Scanning Electron Microscope (SEM) image, and the beam current of the electron beam used to scan the ROI is higher than the beam current used to acquire the sample image. A third example of the method optionally includes one or more of the first and second examples, and further comprising wherein the sample image is an SEM image and the beam energy of the electron beam used to scan the ROI is higher than the beam energy used to acquire the sample image. A fourth example of the method optionally includes one or more of the first through third examples, and further comprising wherein the duration for scanning the ROI is longer than the duration for acquiring the sample image. A fifth example of the method optionally includes one or more of the first through fourth examples, and further comprising, wherein determining that the device layer is not exposed in the ROI based on the sample image comprises determining that the etched ROI is not flat based on the sample image. A sixth example of the method optionally includes one or more of the first through fifth examples, and further comprising, wherein determining that the device layer is not exposed to the ROI based on the sample image comprises determining that a ratio between an area of the device layer exposed to the ROI and an area of the ROI is not greater than a threshold ratio. A seventh example of the method optionally includes one or more of the first through sixth examples, and further includes wherein scanning the ROI with the electron beam includes scanning the ROI with the electron beam to deposit or dope the ROI with carbon. An eighth example of the method optionally includes one or more of the first through seventh examples, and further comprising, wherein flowing the first gas to the sample surface comprises flowing the first gas for a predetermined first predetermined duration. A ninth example of the method optionally includes one or more of the first through eighth examples, and further comprising wherein removing a portion of the substrate layer comprises flowing a second gas to the sample surface for a second duration to etch the sample surface, the second duration being longer than the first duration. A tenth example of the method optionally includes one or more of the first through ninth examples, and further includes wherein removing a portion of the substrate layer includes milling the substrate layer with a focused ion beam.
In one embodiment, a method of producing a plan view lamella using a charged particle beam includes extracting a sample from a workpiece with a focused ion beam, the sample including at least a device layer and a substrate layer; removing a portion of the substrate layer from the back side of the sample to obtain a sample surface; and alternately scanning a region of interest (ROI) associated with the surface of the sample with the electron beam and flowing a gas to the scanned ROI until the device layer is exposed within the ROI, wherein the scanned ROI is spontaneously etched by the gas. In a first example of the method, the method further includes determining that the device layer is exposed within the ROI when a ratio between an area of the device layer exposed in the ROI and an area of the ROI is greater than a threshold ratio. A second example of the method optionally includes the first example, and further includes wherein the sample includes a plurality of device layers, and the exposed device layer is a last device layer of the plurality of device layers; and the method further comprises: at least one device layer of the plurality of device layers is removed from the back side of the sample after the last device layer is exposed within the ROI. A third example of the method optionally includes one or more of the first and second examples, and further includes, forming a plan view sheet based on the sample with the exposed device layer; and imaging the plan view lamella with a transmission electron microscope. A fourth example of the method optionally includes one or more of the first through third examples, and further includes removing at least one of the plurality of device layers from a front side of the sample, the front side being opposite the back side.
In one embodiment, a system for processing a sample comprising at least a substrate layer and a device layer, the system comprises a first column for forming a focused ion beam; a second column for forming an electron beam; a lower chamber coupled with both the first column and the second column; a gas supply system coupled to the lower chamber; and a controller having instructions stored in a non-transitory memory, the controller configured to: removing a portion of the substrate layer to obtain a sample surface; scanning a region of interest (ROI) associated with a surface of the sample with an electron beam; flowing gas to the ROI via a gas supply system to spontaneously etch the scanned ROI; and scanning the etched ROI with an electron beam in response to the device layer not being exposed in the ROI. In a first example of the system, the system further comprises, wherein the instructions are included such that the sample is further cut from the workpiece using the focused ion beam before removing the portion of the substrate layer to obtain the sample surface. A second example of the system optionally includes the first example, and further includes wherein the sample includes a plurality of device layers, the device layer being a last device layer adjacent to the substrate layer, and the controller is further configured to remove one or more device layers with the focused ion beam in the presence of the etch assist gas after exposing the last device layer in the ROI. A third example of the system optionally includes one or more of the first and second examples, and further comprising wherein the controller is further configured to flow gas to the scanned ROI after scanning the etched ROI with the electron beam.
- 上一篇:石墨接头机器人自动装卡簧、装栓机
- 下一篇:嵌套式扫描探针显微镜