JP2008070155A - Preparation method for observing sample for transmission electron microscope - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To correct surface inclination errors of an observation sample, in the installation of the sample on FIB processing sample stand, with high accuracy, in the manufacture of the sample for TEM for thinning a plate by the FIB processing. <P>SOLUTION: Linear marks are respectively provided to the surface of the sample to be observed and the oblique surface part in the vicinity thereof; and when the marks are not related linearly or the marks are not set in parallel relation, an FIB processing surface is set, inclined from a predetermined position, with respect to the linear positional relation of the marks of respective parts; and by adjusting the linear positional relation so that the relation is connected linearly, by using the rotary mechanism of the sample stand, inclination of the surface of the sample is corrected. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a method for producing an observation sample for observation using a transmission electron microscope (TEM), and in particular, a focused ion beam (FIB) for processing on the surface of a target sample. The present invention relates to a method for forming a thin TEM observation region for an accurate TEM in an observation sample according to an intended purpose by using the FIB.

  Up to now, the following two types of methods have been mainly performed for the observation sample preparation method (TEM sample preparation method) for a transmission electron microscope using FIB.

  One is that the target sample is processed to a thickness of about 10 to several tens of μm by a high-speed rotating outer peripheral blade processing device (dicing saw, dicer), and then the thinned portion is focused on a focused charged particle beam (focusing). This is a method of further thinning by processing to a thickness of several tens to 100 nm by an ion beam (FIB) (here, this method is called a dicing saw method) (for example, Patent Documents 1, 2, and 3).

  This dicing saw method will be described with reference to FIG. In FIG. 1A, a TEM sample 1 having a convex cross section is cut out from a semiconductor device, for example, with a dicing saw. The length a of the side surface portion 2 of the TEM sample 1 is 1.5 mm, for example, and the cross-sectional thickness b of the cross-sectional convex portion 3 is 0.04 mm (40 μm), for example. Such a TEM sample 1 is attached to a mesh 4 (for example, a metal mesh of 3 mmφ) which is a TEM sample support using an adhesive. Next, in order to thin the TEM sample 1 in a desired direction using the FIB in the desired observation place 5 in the cross-sectional convex part 3, the inclination of the TEM sample in the FIB apparatus, for example, Adjustment is performed by the sample rotation mechanism 6 or the like.

  FIG. 1B shows an enlarged view of the vicinity of the observation place 5. In this figure, the depth from the surface of the convex section 3 by the FIB (for example, focused Ga ion beam) scanning process as shown by A in the figure with respect to the observation portion 5 of the convex section 3 (cross section thickness b). The thin plate portion 8 having a thickness c (for example, 50 nm) and a depth d is formed by performing a double-side etching process on a predetermined position of the side surface portion 7 of the convex section in the direction. Here, TEM observation is performed on the thin plate portion 8 from the vertical direction B.

  Another method related to TEM sample preparation using FIB is that when separating any part such as a semiconductor device, three-dimensional processing of FIB from the beginning without using the above-mentioned dicing saw or means for breaking. This is a method of separating and processing a required portion by a technique (here, this method is referred to as a microsampling method) (for example, Patent Document 4).

  This microsampling method will be described with reference to FIGS. In the sample arrangement schematic diagram of FIG. 2A, a TEM sample 10 is set on a sample stage 9 in the FIB apparatus. Using the FIB (for example, focused Ga ion beam) indicated by A, the periphery including the TEM observation position 11 in the TEM sample 10 is irradiated and etched. As shown in the figure obtained by the SIM (Scanning Ion Microscope) after irradiation / etching in FIG. 2B, the length e () in which the periphery including the TEM observation place 11 is isolated by four square holes. For example, a separation sample 12 of 20 μm is defined (in this case, for example, a bridge portion D for holding the separation sample after punching the bottom is formed at a position indicated by D in the drawing), and FIG. As shown in the SIM diagram of c), the TEM sample 10 is inclined by 45 to 60 degrees, for example, and the bottom 13 of the separated sample 12 is punched out by FIB.

  Next, as shown in the SIM diagram of FIG. 2 (d), the inclination of the TEM sample 10 is returned to the original position, and the tip of the metal probe 14 is brought into contact with the contact portion E of the separated sample 12, for example, the right end surface. A deposited film of, for example, tungsten, carbon, or platinum is formed at the contact portion E by the function of forming the beam-induced deposited film, and the probe 14 and the separated sample 12 are connected. Thereafter, the bridge portion D is cut by FIB processing, and the separation sample 12 attached to the tip of the probe 14 is separated from the TEM sample 10.

  As shown in the schematic diagram of FIG. 3E, the separated sample 12 thus obtained is transported by using the probe 14 on the mesh 15 whose end face is flattened and brought into contact with the end face 16 of the mesh. 3 (f) is a SIM view of the end face 16 of the mesh 15 observed from above, but as shown in this figure, the upper side surface of the separation sample 12 and a part of the end face 16 are shown in FIG. Similarly to the above, the deposited film F is formed and fixed to the mesh 15 here.

  Next, the probe 14 is cut from the separated sample 12 by FIB processing, for example, at a cutting place G in the drawing. The schematic diagram in FIG. 3G shows a state in which the deposited film F formed on the side surface of the separation sample 12 and a part of the end surface 16 is fixed on the end surface 16 of the mesh 15.

Then, as in the case shown in FIG. 1B, the TEM sample having the TEM observation place is produced by forming a thin plate portion of, for example, a thickness of 50 nm from the separation sample 12 by FIB processing.
Japanese Patent Laid-Open No. 5-180739 JP-A-8-261894 JP-A-8-261898 Japanese Patent Laid-Open No. 5-52721

  In the dicing saw method, the TEM sample is fixed to the mesh with an adhesive, but in general, the sample side surface and the mesh surface are not necessarily parallel (or the sample surface and the mesh surface are not necessarily perpendicular). The required direction of the sample for use is not necessarily suitable for the direction of FIB processing. In other words, the sample may be arranged with the axis that should be perpendicular to both the ion beam direction axis and the TEM observation direction axis of the FIB being inclined, and as a result, the TEM observation surface of the thin plate after the FIB processing is As the processing depth increases from the surface of the sample, the surface may deviate from the required observation surface.

  Conventionally, it is a side surface portion that is a cut surface of a TEM sample cut by a dicing saw method (that is, the side surface portion 2 of the TEM sample side surface 2 or the cross-sectional convex portion 3 in FIG. The inclination 6 of the sample was corrected using the reference plane as the reference plane. When the depth d of the thin plate portion 8 in FIG. 1B is as shallow as about 1 μm or less, for example, the TEM observation result (of the thin plate portion) in the depth direction of the observation sample is There was no.

  However, for example, with regard to a trench capacitor type DRAM, in particular, a product having a trench capacitor with a large aspect ratio that has been developed and commercialized recently, let's observe the formation state of the trench capacitor part in a cross section in the depth direction. Consider the case. FIG. 4 is a diagram schematically showing a cross section in the depth direction of a trench capacitor forming portion of a typical trench capacitor type DRAM. The trench capacitor 17 is formed, for example, by forming a pore having a depth of f (for example, 10 μm), for example, a diameter of 200 nm on a Si substrate and covering the sidewall of the hole with, for example, an ONO film with a thickness of several nanometers. Filled to form a trench type capacitor, which is periodically formed at a predetermined interval.

  4 is a cross-sectional view in the depth direction of the trench capacitor portion, which has the same cross-sectional configuration as when the cross-sectional convex portion 5 is viewed from the direction indicated by C in FIG. It is assumed that they are formed deeply and periodically from the convex surface of the convex section 5 in the downward direction. At this time, in FIG. 1B, the thin plate portion 8 is formed to a depth d equal to the depth f of the trench capacitor (for example, about 10 μm), and a cross section in the depth direction of the trench capacitor 17 is to be observed by TEM. In this case, it is necessary to adjust the inclination angle of this sample more accurately (to FIB) and to process with FIB so that the cross section in the depth direction of the trench capacitor 17 appears uniformly over the depth f (about 10 μm). is there.

  If the depth direction of the thin plate portion 8 for TEM observation of the thickness c (for example, 50 nm thickness) of the thin plate portion is shifted from the depth direction of the trench capacitor 17 as shown in FIG. Even if the TEM observation of the cross section of the trench capacitor 17 is possible in the vicinity of the surface H, the TEM observation in a deep place cannot be performed. That is, in the preparation of a sample by FIB for TEM observation of such an observation portion having a particularly high aspect ratio, a more accurate adjustment of the sample arrangement angle with respect to the FIB axis is important.

  The situation that the adjustment of the sample arrangement angle with respect to the accurate FIB axis is important also occurs in the manufacturing method using the microsampling method. 3 (e), (f), and (g), when the separation sample 12 is fixed to the mesh 15 with a deposited film, first, the “TEM sample 10 is inclined by 45 to 60 degrees, for example, and the separation sample 12 is tilted. The bottom surface of the separated sample 12 is cut obliquely as described in “Punching the bottom 13 of the substrate with FIB”. For this reason, it is difficult to stand upright and fix to the end face 16 of the mesh so that the processing axis on which the FIB is incident is accurately perpendicular to the surface of the separated sample 12, and it is actually inclined. It is often fixed by the situation.

  Therefore, as described in the case of the dicing saw method, the thin plate portion for TEM observation is provided so that the cross section in the depth direction of the element formed deep from the surface (high aspect ratio) can be covered in the entire depth direction. It becomes very difficult to process accurately with FIB. That is, a means for adjusting the tilt angle of the sample with respect to the FIB incident axis, which can ensure the TEM observation direction and the TEM observation range accurately, which is effective in such a case, is important.

  Therefore, an object of the present invention is to prepare a TEM observation sample in FIB, particularly when a mark is arranged on the surface of the observation sample linearly or periodically so that it can be detected by a SIM or the like. In view of the above, there is provided a method for correcting and adjusting the tilt angle of the processed sample with respect to the FIB processing direction, which is deep in the depth direction from the surface, in accordance with the demand for high-precision processing.

  By applying this method, even when FIB processing is performed deeper than the surface for forming a TEM sample, it is possible to accurately determine the FIB direction by correcting the tilt of the observation sample. The TEM observation surface can be observed over a long span in the depth direction (for example, a depth direction cross section of a deeply buried trench capacitor from the vicinity of the surface to the vicinity of the capacitor bottom).

The purpose of the present invention is to
In the observation sample preparation method for a transmission electron microscope using a focused ion beam,
On each surface of the observation target sample having a flat portion and a stepped portion, it has a linearly connected or discrete mark,
This is made possible by a method for preparing an observation sample for a transmission electron microscope, in which the inclination of the surface of the observation target sample with respect to the focused ion beam is corrected from the positional relationship of the respective marks.

  Further, the step portion is a slope or a step bottom surface.

  The sample to be observed is a semiconductor device, and the mark is a device formation region formed on a semiconductor substrate.

  And the said mark is the groove | channel or hole previously processed and formed in the said mutually adjacent plane part and level | step difference part.

  FIB processing for TEM sample formation is carried out particularly deeper than the surface, and even when the required processing location is accurately expressed even at a deep processing location, the observation sample can be obtained by applying the method of the present invention. The inclination can be corrected, and the FIB direction can be accurately determined and processed. For this reason, for example, the depth direction cross section of the deeply buried trench capacitor can be observed by TEM over a long span in the depth direction from the vicinity of the surface to the vicinity of the bottom of the capacitor.

(First embodiment)
The case where a sample for TEM observation of the cross section in the depth direction of the trench capacitor of the trench capacitor type DRAM is described with reference to FIG.

  The trench capacitors are usually arranged deeply from the chip surface into the substrate and periodically as viewed from the surface. As described with reference to FIG. 1, an observation place of the DRAM sample (a trench capacitor portion cut out so that a cross section in the depth direction can be observed) is cut out by a dicing saw method, and pasted on a mesh, and then a TEM observation place Using a FIB, slope formation etching is performed on the surface at a place away from the (thin plate processing place).

  A specific example is shown in FIG. FIG. 5A is a schematic diagram of a cross section of the trench capacitor 17 portion similar to that shown in FIG. 4, which is an extension of the TEM observation place and is away from the place, for example, FIG. In (a), it is assumed that the vicinity of the end of the convex section 3 is shown. Then, after affixing to the mesh as shown in FIG. 1 (b), before forming the thin plate portion 8 by FIB processing in the FIB apparatus, the end portion of the cross-sectional convex portion 3 is diagonally cut by the FIB. Processing is performed to form a slope region 19 adjacent to the surface region 18 as shown in FIG.

  The processing angle and processing length (range) of the slope region 19 may be set within a range sufficient for the subsequent angle adjustment determination.

  Next, FIGS. 5B and 5C are schematic views when the surface side of FIG. 5A is observed from directly above the sample (that is, the FIB incident surface).

  In order to observe the state of the sample placed on the sample stage, the apparatus for FIB processing of the sample usually has secondary particles (ions and electrons) emitted by scanning irradiation of the sample surface with a focused ion beam. ) To obtain a sample observation image. That is, it is a function of SIM (scanning ion microscope) or SEM (scanning electron microscope). By using these functions, it is easy to observe the location of the trench capacitor 17 (cross section in the arrangement direction) on the substrate as shown schematically in FIGS. 5B and 5C. . That is, FIGS. 5B and 5C show a trench capacitor including a surface region 18 and a slope region 19 partially processed in the depth direction, which are observed from a plane perpendicular to the FIB irradiation direction. It is an observation image of a sample surface showing a periodic arrangement direction cross section.

  The observed cross section of each trench capacitor 17 in the slope region 19 is seen in a deeper section from the surface as it is in the left direction (the direction away from the boundary between the surface region 18 and the adjacent slope region 19). It will be.

  Therefore, the arrangement direction line (virtual line) L1 of the periodically arranged cross section of the trench capacitor observed in the surface region 18 and the arrangement direction line (imaginary line) of the slope region 19 shown in FIG. When the line L2 is on a straight line as shown in the figure, in this region where the trench capacitors 17 are periodically arranged, the depth direction of each trench capacitor 17 is perpendicular to the sample surface being observed. (I.e., the FIB incident direction).

  However, as shown in FIG. 5C, the alignment direction line L1 of the periodically arranged cross-sections of the trench capacitors observed in the surface region 18 and the alignment direction line L2 of the inclined region 19 are shown in FIG. In the case where they are not in a straight line, the depth direction of each trench capacitor 17 does not coincide with the direction perpendicular to the observed sample surface (that is, the incident direction of the FIB), and the depth of the trench capacitor 17 As the direction becomes deeper, the width of the deviation from the FIB incident position becomes larger.

  If the thin plate portion processing by FIB is performed deeply in the TEM observation place in such a situation, the situation described in FIG. 4, that is, the TEM observation of the cross section in the depth direction of the trench capacitor 17 cannot be performed in the deep portion of the thin plate portion. Such a situation will occur.

  Therefore, when the observation as shown in FIG. 5C is made, the TEM observation is performed so that the direction line L1 of the trench capacitor observed in the surface region 18 and the alignment direction line L2 of the inclined region 19 coincide with each other. By adjusting the tilt of the sample for use with the rotating mechanism 20 of the sample stage mechanism on which the mesh is mounted, the direction of the FIB incident on the surface of the sample for FIB processing is corrected, and deep processing is performed in the intended direction. It is possible to form a thin plate portion for TEM observation (having been processed correctly to a cross section in the depth direction of the trench capacitor at a deep position).

  In the above description, the arrangement direction (virtual) line L1 of the trench capacitors 17 in the surface region 18 and the arrangement direction (virtual) line L2 of the trench capacitors 17 in the slope region 19 have been described. Obviously, in order to obtain information on the line (virtual) line of the capacitors 17, it is not limited to the method of forming the slope, and the line direction of the trench capacitors (or deeply formed marks and marks) in the surface region. It is only necessary to form a stepped surface having a predetermined width in a step shape at a predetermined depth and compare the line in the direction of the trench capacitor (or a deeply formed mark or mark) on that surface. As the step surface, for example, a square or circular vertical hole can be formed by FIB, and the bottom surface thereof can be used as the step surface.

(Second embodiment)
Next, an example will be described in which, when a TEM observation sample is formed by the above-described microsampling method, a portion to be observed is accurately adjusted in the depth direction to be thinned and suitable for TEM observation.

  First, according to FIG. 2A and FIG. 2B, the separated sample 12 connected by the bridge portion D is defined. Next, as shown in the SIM diagram of FIG. 2 (c), the TEM sample 10 is inclined, for example, 45 degrees to 60 degrees, and the bottom 13 of the separated sample 12 is punched out by FIB. As described above, since the bottom surface of the separated sample 12 is cut obliquely, it is possible to fix the separated sample 12 to the end face 16 of the mesh by standing upright so that the processing surface on which the FIB is incident is accurately vertical. He stated that it was a difficult task and was often fixed in a tilted situation.

  An example in which the sample surface and the end face of the mesh are easily adjusted to be parallel (that is, so that the FIB enters the sample surface accurately perpendicularly) will be described below.

  First, the sample surface and the sample table surface are fixed on the sample table so that they are parallel to each other. FIG. 6A shows the observation state at this time. In the state where the separation sample 12 of FIG. 2B is not separated from the TEM sample 10 by the bridge portion D, the surface ( This surface is fixed in parallel with the surface of the sample stage) and is a schematic view when observed by SIM or SEM. At this time, the observation surface of the TEM sample 12 is in a state parallel to the sample table surface (that is, FIB is incident perpendicularly to the TEM sample surface).

  Here, a slope region 22 having a certain region that is in contact with the surface region 21 and gradually becomes deeper is formed by using the FIB at a location away from the observation location 11 where the separation sample 12 is thinned by FIB. To do. Of course, the width of the slope region 22 does not have to be the same as the width of the separated sample 12, and may be, for example, a square hole shape having a sloped bottom surface.

  After forming the slope region 22, using the FIB, as shown in the figure, the groove line L3 and the groove line L4 that form straight grooves are formed so as to pass through the surface region 21 and the slope region 22 continuously. To do. As a result, as shown in the figure, the groove line L3 and the groove line L4 are continuous straight lines. In this way, a reference mark (mark) is formed.

  The separation sample 12 in which the groove lines L3 and L4 are thus formed is separated from the TEM sample 11 by the method shown in FIGS. 2 and 3, and a schematic view of an example when the surface side is observed when mounted on the mesh. Is FIG. 6B. In this observation example, the groove line L3 of the surface region 21 and the groove line 22 of the slope region 22 are observed with a SIM or SEM so as to be bent at the boundary.

  This is because the surface of the separation sample 12 in this case is not the same as when the groove lines L3 and L4 are formed in FIG. 6A (that is, the observation surface of the TEM sample is parallel to the sample stage). It is determined that it is tilted from the direction of processing (the surface is not perpendicular to the FIB). Therefore, the original orientation of the separated sample can be corrected by changing the inclination using the mesh rotation mechanism 23 so that L3 and L4 are in a straight line. Thereafter, the corrected separation sample 12 is subjected to FIB processing to make a thin plate, and a thin plate portion for TEM observation having an appropriate depth direction can be obtained.

  In the above example, the slope region 22 gradually deeper in contact with the surface region 21 is formed, but the bottom region is formed with a steep step from the boundary of the surface region, and both groove lines are formed as shown in FIG. It may be formed continuously in the region. In FIG. 6B corresponding to this case, the groove lines L3 and L4 are not in contact with each other, and the both groove lines are not in parallel, so that they are in contact with each other and become parallel (straight) by rotation correction.

  In the above example, a linear groove formed by FIB processing is used as a reference mark. However, the present invention is not limited to this. Basically, it may be a set of marks that form a virtual straight line. For example, circular holes are formed so as to form a virtual straight line discretely, and the above adjustment is performed using this virtual line. Just do it.

(Third embodiment)
In the first embodiment, an example in which FIB processing for performing TEM observation in the depth direction of the trench capacitor forming portion in the trench capacitor type DRAM is shown. In the present embodiment, for example, in a trench capacitor type DRAM, a so-called planar TEM observation is performed over a wide range with a uniform sample depth on the formation state of the gate electrode periodically formed on the surface of the device in a planar direction. An example of FIB processing will be shown.

  FIG. 7 is a diagram for explaining sample preparation for planar TEM observation. FIG. 7A schematically shows a state in which the TEM sample 24 cut out by the dicing saw method is mounted on the end face of the mesh 4 as in FIG. 1A.

  The fact that the TEM sample 24 is convexly processed with a dicer from only one side surface indicates that the surface side of the TEM sample 24 on which the gate electrode is formed is not processed, but is ground only from the opposite substrate side. . Here, A represents the FIB incident direction for further thin plate processing. FIG. 7B is a schematic diagram when the mesh of FIG. 7A is viewed from the opposite side, that is, when the plane on which the gate electrode is formed is the front, and the observation place 25 is FIB. The cutting part is thinned only by processing from the back side, and TEM observation (planar TEM observation) is performed from the direction B perpendicular to the thin plate side surface.

  In order to perform TEM observation (that is, planar TEM observation) from the surface side of the gate electrode formed on the device surface periodically in a wide range under uniform conditions, the electrode formation surface (surface side) ), And it is necessary to accurately form a thin plate having a uniform thickness from the back side (substrate side) by FIB. That is, the TEM sample needs to be arranged so that the FIB incident direction is exactly parallel to the gate electrode formation surface (TEM observation surface).

  FIGS. 8C and 8D are schematic diagrams for explaining a method therefor. FIG. 8C shows the gate electrode formation surface of the device when viewed from the direction B in FIG. 7B and shows that the gate electrodes 27 are periodically arranged.

  In this situation, the slope formation process is performed at a place away from the observation place using the FIB, and the surface region 28 (that is, the region in which the TEM sample 24 in FIG. 7A is in a flat surface state when viewed from the A direction). And a slope region (similarly, a region in which the slope is processed so as to cut deeper as the TEM sample 24 in FIG.

  FIG. 8D shows a schematic diagram when the observation sample processed in this way is viewed from above (that is, the TEM sample 24 of FIG. 7A is viewed from the A direction).

  At this time, the arrangement state of the cross section of the gate electrode 27 on the substrate 30 can be observed. By comparing the direction line (virtual line) L5 of the cross section of the gate electrode 27 in the surface region 28 with the direction line (virtual line) L6 of the cross section of the gate electrode 27 in the surface region 29, the TEM sample 25 enters the FIB. It can be determined whether or not it is accurately positioned with respect to the direction, and if each direction line is not parallel, the tilt of the sample can be corrected so that it can be observed parallel using a mesh rotation mechanism. Become.

  By correcting the tilt of the sample in this way and performing the FIB processing accurately, the same observation conditions are applied to the TEM observation of the gate electrodes 27 arranged in the downward direction in FIG. Can be used for plane observation. In this example, the formation of the slope region has been described, but it goes without saying that the present invention is not limited to this, as described in the first example.

  By applying the method of the present invention described above, a TEM sample capable of TEM observation over a wide range having a depth of 10 μm or more can be easily produced.

  The following supplementary notes are further disclosed with respect to the embodiments including the above examples.

(Appendix 1)
In the observation sample preparation method for a transmission electron microscope using a focused ion beam,
On each surface of the observation target sample having a flat portion and a stepped portion, it has a linearly connected or discrete mark,
An observation sample preparation method for a transmission electron microscope, wherein an inclination of the surface of the observation target sample with respect to the focused ion beam is corrected based on a positional relationship between the respective marks.

(Appendix 2)
The observation sample preparation method for a transmission electron microscope according to appendix 1, wherein the stepped portion is an inclined surface or a stepped bottom surface.

(Appendix 3)
The observation sample preparation method for a transmission electron microscope according to appendix 1 or 2, wherein the observation target sample is a semiconductor device, and the mark is a device formation region formed on a semiconductor substrate.

(Appendix 4)
The observation sample preparation method for a transmission electron microscope according to appendix 1 or 2, wherein the mark is a groove or a hole formed in advance in the planar portion and the step portion adjacent to each other.

(Appendix 5)
The observation sample preparation method for a transmission electron microscope according to appendix 4, wherein the groove or hole is processed and formed by FIB.

(Appendix 6)
The observation sample preparation method for a transmission electron microscope according to appendix 3, wherein the device formation region is a trench capacitor region.

(Appendix 7)
The observation sample preparation method for a transmission electron microscope according to appendix 3, wherein the device formation region is an electrode region.

The figure explaining the TEM sample preparation method by the dicing saw method The figure explaining the TEM sample preparation method by the micro sampling method (the 1) Diagram explaining TEM sample preparation method by microsampling method (Part 2) Cross-sectional view of trench capacitor in depth direction The figure explaining the method of this invention (1st Example) Diagram for explaining the method of the present invention (second embodiment) The figure explaining the method of this invention (3rd Example and the 1) The figure explaining the method of this invention (3rd Example and the 2)

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 10, 24 TEM sample 2, 26 Side surface part 3 Section convex part 4, 15 Mesh 5, 11, 25 Observation place 6, 20, 23 Sample rotation mechanism 7 Side part of sectional convex part 8 Thin plate part 9 Sample stand 12 Separation Sample 13 Bottom 14 Probe 16 End face 17 Trench capacitor 18, 21, 28 Surface region 19, 22, 29 Slope region
27 Gate electrode 30 Substrate

Claims (5)

In the observation sample preparation method for a transmission electron microscope using a focused ion beam,
On each surface of the observation target sample having a flat portion and a stepped portion, it has a linearly connected or discrete mark,
An observation sample preparation method for a transmission electron microscope, wherein an inclination of the surface of the observation target sample with respect to the focused ion beam is corrected based on a positional relationship between the respective marks.   2. The observation sample preparation method for a transmission electron microscope according to claim 1, wherein the step portion is an inclined surface or a step bottom surface.   The observation sample preparation method for a transmission electron microscope according to claim 1, wherein the observation target sample is a semiconductor device, and the mark is a device formation region formed on a semiconductor substrate.   3. The observation sample preparation method for a transmission electron microscope according to claim 1, wherein the mark is a groove or a hole formed in advance in the planar portion and the step portion adjacent to each other.   The observation sample preparation method for a transmission electron microscope according to claim 4, wherein the groove or hole is formed by processing using FIB.

Images