TW201732280A - Method of imaging defects using an electron microscope - Google Patents

Abstract

There is provided a method of using an electron microscope (10) to image atomic level defects associated with interfaces between different materials. The impact energy of an electron beam (20) incident on a sample (12) is selected based on the expected potential energy at an interface between at least two materials, a back-scattered electron image simulated and the electron energy iteratively adjusted by a bias voltage (44) applied to the sample until a plane of origin of defects within the sample is reached.

Description

Method of imaging defects using an electron microscope

The present invention relates to a method of imaging a defect using an electron microscope such as a scanning electron microscope (SEM), and more particularly to a method of imaging a defect in a junction in an electronic device.

Modern electronic and spintronic devices generally rely on characteristics at the junction to create their electronic properties. The junction can be formed as a region in the film structure where two different materials are in contact, or where the microstrips of the two different materials pass over or under each other. Such junctions can be produced in an interface that is nearly atomically flat. However, this leads to the problem of uniform electrical conduction across the interface, since the presence of a pinhole or other defect is not a short circuit, or a region of high resistance, thereby impairing the conduction through the junction. In the case of such defects, it is difficult to discern which defects occur in the process and where they occur. It is extremely difficult to properly evaluate these defects, which requires the creation of an ultra-thin cross section of the junction area for investigation by transmission electron microscopy. The preparation of the sample may take at least one day, and because of the process required to thin the sample, there is often uncertainty as to whether the imaged person in the thinned cross section represents the original defect.

According to a feature of the present invention, there is provided a method of imaging an atomic level defect using an electron microscope, the interface defects between different materials generally having a few atoms to a single atom, The method includes changing an impact energy of an incident electron beam by repeatedly adjusting an electron energy adjustment region of the sample that is being imaged, and evaluating each adjustment of the impact energy of the incident electron beam. A simulated backscattered electronic image until the defect originates within a single or discrete plane along which the sample originates. By repeatedly adjusting the impact energy of the incident electron beam in this manner, an image of an electron microscope can be obtained which exhibits defects originating along a plane buried in the sample, wherein the defects It is not obscured by imaging the deeper, more prominent features of the sample. This allows defects within the sample, and thus associated interface defects between different materials, to be found and imaged at the depth at which they occur, and allows the defects to be evaluated and analyzed without Thin the physical program required for the sample. Such non-destructive identification of such defects allows for an assessment of the quantified and the total area of a junction-related defect by calculating the regions in which the defects occur. Energy dispersive X-ray (EDX) analysis can be performed to identify the composition of the defects themselves, since the exact location of the defects is known, and thus the EDX can be correlated with the defect itself, rather than It may be impaired because it may be a background reading of other features within the sample.

The method can further include repeatedly adjusting the electron energy adjustment region until the impact energy of the incident electron beam matches the energy associated with a defect associated with an interface between the materials.

Preferably, the method further comprises selecting an impact energy of the incident electron beam based on an expected potential energy of an interface between the at least two materials to be imaged This defect is related to the interface. This is preferably accomplished by using a look-up table or by estimating a potential barrier at an interface between different materials that is known as an interface barrier. The current-voltage characteristic is a model used to calculate this potential from current-voltage characteristics. Alternatively, modeling of simulation techniques such as electronic flight simulation software can be utilized to select the impact energy required to reach a penetration depth at which the interface between different materials abuts.

Different types of interfaces between different materials are known to have different barrier energies. The defect potential will be a barrier similar to the theoretical or calculated interface, but not the same. By selecting the initial energy of the incident electron beam based on the barrier between the interfaces, the final voltage required to image the plane from which the defects originate can be further adjusted from the repeated adjustments. Quickly identified, these repetitive adjustments allow for a slight adjustment of the energy of the electron beam until it matches the potential energy of the defect.

Although a singular reference is made to a defect, it should be recognized that in any given junction of two or more materials adjacent to each other at an interface, there will generally be a plurality of defects, typically having The same type, and therefore is derived from the same plane. These defects are generally at the atomic level of a size ranging from a few atoms to as low as a single atom.

In general, the first scan will not show discrete defects because the potential energy of the defect itself is not exactly known. However, the initial scan provides a starting point from which the simulated backscattered image is evaluated prior to adjusting the impact energy. Typically, the scan of the impact energy and the repeated adjustments will be repeated until discrete features originating along a plane are visible within the simulated backscattered image.

Once the plane from which the defects originated has been identified, and thus the defect After the potential energy has been identified, the same type of sample can be evaluated at the selected impact energy. This enables non-destructive testing of samples of electronic wafers in a production line facility such that a continuous throughput of wafers can be routinely and non-destructively evaluated for quality control purposes.

Preferably, the electron energy adjustment region is set by applying a bias voltage to the sample, for example by applying the voltage to a device on which the sample is mounted.

Energy filtration can be performed in relation to the electrons generated from the sample. This allows electrons having energy in a desired range to be selected, and higher energy electrons generated closer to the surface of the sample can be filtered out to avoid blurring the image of defects at a deeper plane. This is to ensure that the characteristics of the most relevant electrons are not imaged, but that the deeper features of the less energetic electrons within the sample are selectively imaged.

Preferably, the bias voltage is positively and negatively adjustable, thereby permitting fine adjustment of the impact energy of the incident electron beam through acceleration or deceleration as the electrons approach the sample.

The bias voltage is preferably between 0 and ±300 V and can be adjusted with a step of 1 mV to allow for a fine adjustment of the impact energy.

Preferably, the energy of the incident electron beam is initially selected in the range of 0 to 5 keV because the interface barrier is typically 1 keV or less.

Thus, in accordance with another feature of the invention, there is provided a method of imaging a defect associated with an interface between different materials using an electron microscope. The method includes: i) selecting an initial energy of an electron beam incident on a sample containing the defect based on potential energy at an interface between different materials, wherein the interface includes but is not limited to a metal/semiconductor, a ferromagnetic metal/non-magnetic metal, an organic material/metal, a thin film deposited on a substrate, such as an Fe film deposited on a GaAs substrate, and a multilayer interface, for example, found on a tunnel junction surface and In particular, the Fe/MgO/Fe interface: ii) using the electron beam to perform a scan of the sample; iii) simulating a backscattered electronic image produced by the electron beam; iv) evaluating the simulated backscattering An electronic image to determine whether backscattered electrons originate from a plane or layer within the sample; v) adjust an electron energy adjustment region proximate the sample to change an impact energy of the electron beam; Steps ii through v until reaching a plane as the source of the defect indicated by the simulated backscattered electronic image.

Preferably, the electron generation that simulates the backscattering is performed using an electronic flight simulation software.

Preferably, the electron energy adjustment region is set by applying a bias voltage to the sample, for example by applying the voltage to a device on which the sample is mounted.

Energy filtration can be performed in relation to the electrons generated from the sample. This allows electrons generated closer to the surface of the sample to be filtered out to avoid blurring the image of defects at a deeper plane.

Preferably, the bias voltage is positively and negatively adjustable, thereby permitting fine adjustment of the impact energy of the incident electron beam through acceleration or deceleration as the electron approaches the sample.

The bias voltage is preferably between 0 and ±300 V, and can be used with 1 mV. The steps are adjusted to allow for a subtle adjustment of the impact energy.

Preferably, the energy of the incident electron beam is initially selected in the range of 0 to 1 keV because the interface barrier is typically 1 keV or less.

The method of the present invention allows a buried plane within a sample to be imaged, thereby allowing defects at the plane from which the defect originated to be imaged.

10‧‧‧Scanning electron microscope

12‧‧‧ sample

14‧‧‧Sample table

18‧‧‧Electronic gun

20‧‧‧Electron beam

22‧‧‧ Concentrating lens

24‧‧‧ Objective

26, 26'‧‧‧ coil

30‧‧‧Scanning coil

32‧‧‧Electronic detector above

34‧‧‧ secondary electronic detectors above

36‧‧‧Retractable backscattered electronic detector

38‧‧‧Electronic detector below

40‧‧‧Energy filter

44‧‧‧DC power supply

45‧‧‧ Controller

46, 47, 48, 49, 50 ‧ ‧ steps

51‧‧‧Electronic wafer (horizontal GMR device)

52‧‧‧NiFe line

54‧‧‧Cu wire

56‧‧‧Connected

60‧‧‧ Defects

62‧‧‧Substrate

70‧‧‧Digital image of backscattering

72‧‧‧ SEM image

74‧‧‧ hole

80‧‧‧ Simulated backscattering images

The invention will now be described by way of example with reference to the accompanying drawings in which: FIG. 1 is a schematic block diagram of a scanning electron microscope (SEM) used to carry out the method of the invention; A schematic view of a portion of the sample being analyzed; Figure 3 is a flow chart depicting the method; Figure 4 is a cross section of the same junction; and Figures 5 (a) and (b) are shown for two different electrons An electronic image of the backscattered voltage of the voltage and the corresponding SEM image.

1 is a scanning electron microscope 10 (typically a field emission scanning electron microscope) that is used to analyze a sample 12 mounted on a sample stage 14 in accordance with the method of the present invention (eg, a Defects within the electronic chip). The scanning electron microscope 10 includes an electron gun 18 that produces an electron beam 20 that is focused by a collecting lens 22 and a semi-in objective lens 24 including coils 26, 26'. Above the sample (specimen) 12. Scanning coil 30 is deflecting beam 20 so that the electron beam can be scanned two-dimensionally in sample 12 Above. Typically, the scanning electron microscope 10 is computer controlled and a display device is used to display images to the user. Typically beam 20 initially has an energy greater than 5 keV to maintain high resolution and avoid aberrations as beam 20 passes through lens 22, wherein the energy of beam 20 is further adjustable by objective lens 24 to approximately 1 keV. To avoid sample damage.

The detectors 32, 34, 36, 38 are configured to detect electrons having different properties resulting from the interaction of the primary electron beam 20 with the sample 12. The detectors 32, 34 are disposed between the objective lens 24 and the scanning coil 30, wherein the detector 32 is an upper electronic detector, and the detector 34 is an upper secondary electronic detector. . A collapsible backscattered electron detector 36 and a lower electronic detector 38 are disposed between the objective 24 and the sample 12.

The upper electronic detector 32 is typically associated with an energy filter 40 such that when a bias voltage is applied to the energy filter 40, it can detect reflected electrons from the sample 12 having an energy in excess of 1 eV. The upper secondary electronic detector 34 detects reflected electrons below 1 eV.

The sample stage 14 is connected to a thermally stable DC power supply 44 that produces a voltage between ±0.01V and 300V, which can be adjusted by a controller 45 in steps of 1 mV. Adjust the voltage applied to the station. Power supply 44 is used to apply a positive or negative bias voltage to sample 12, thereby adjusting the energy at which electron beam 20 strikes the sample. A negative bias voltage will reduce the energy of the beam 20, which reduces the penetration, and a positive bias voltage will increase the energy of the beam 20, which is increased throughout.

2 shows a portion of an electronic wafer 51 comprising two NiFe wires 52 bridged by a Cu wire 54 that is orthogonal to the NiFe wires. Junction 56 It is formed where the lines cover each other. Where the junctions are defective, the electrical performance of the wafer 51 will be compromised, so it is important to be able to analyze the defect to identify where the defect occurred in the process and thus modify the process to avoid defects. Previously, damaged joints have been evaluated by cross-sectional transmissive electron microscopy involving the preparation of a sample of the cross-section of the joints by slicing, and other preparations by penetrating The technique of electron microscopy to examine very thin samples. However, preparing such a sample is time consuming and requires a complicated technique, which may cause additional damage and strain to the sample of the cross section, thus making it difficult to identify the problem that originally caused the joint failure.

In the present method, a non-destructive method of analyzing buried junctions within a sample is proposed such that the locations of defects close to the interface interface can be identified and, if desired, such defects The composition can be analyzed and/or the defects can be quantitatively evaluated. Any type of electrical interface can be evaluated, especially in junctions in electronic and spintronic devices, junctions between different types of semiconductors such as n-type and p-type germanium, for example in The junction between the metal and the non-metal of CoFe and MgO in the magnetic tunneling junction, and the organic junction of, for example, graphene/metal or carbon nanotube/metal.

A method of imaging a defect within a sample occurs as follows: a sample 12 of a broken or damaged electronic wafer, such as an unknown defect, is placed on the sample stage 14. Typically the defect is at an electron or spintronic junction, although it may be at an organic junction such as graphene/metal or carbon nanotube/metal. This defect occurs at an interface close to the different structures buried under the surface of the sample and may be a hole or an impurity. The exact depth at which the defect occurred is unknown.

The initial electron beam 20 is decelerated before entering the objective lens 24 to facilitate production. A beam of between 0.3 and 3 keV is produced, and is typically 1 keV for good resolution.

The electron beam 20 will penetrate the sample 12 to a depth based on the beam energy, wherein the backscattered electrons are produced due to the characteristics within the penetration depth. The microscope image is characterized by overlapping at different depths within the sample. Although defects can be seen, the depth positions of the defects within the sample 12 cannot be accurately determined without the use of transmission electron microscopy to analyze the ultra-thin cross-section.

In order to find and image defects at the depth at which defects occur within the sample, without images from below this depth and thus deeper within the sample, the electron beam incident on the sample needs to have a similar The energy of the relatively clean junction interface of the defect is therefore required to be of the order of 1 keV or less. Therefore, in order to image defects close to the interface interface, the energy of the incident electron beam between leaving the objective lens 24 and reaching the sample 12 needs to be modified. To achieve this, a bias voltage is applied to the sample stage 14 using the power supply 44 and the controller 45 to adjust the energy when the electron beam 20 is incident on the sample.

By decelerating near the sample stage 14, the beam impact energy can be matched to the energy of the defect, and thus the beam penetration depth can be controlled to match the defects at or near the junction interface. The vertical position of the place.

The sample stage 14 is biased at a DC voltage between ±0.01 and 300V to adjust the electron velocity and may be positive or negative biased. The initial voltage required for the incident beam is fitted by fitting a current-voltage characteristic with a standard model, such as Simmon's for tunneling junctions, from the height of the interface. Estimate to choose. This provides an indication of the barrier at the interface of the junction, which typically represents the area of the lowest resistance, for example, of the aperture.

The initial voltage can also be by, for example, "Small World Electron Flight Simulator's simulation software is used to estimate the required penetration depth and hence the beam impact energy using the known scattering parameters of the different layers in the sample. The deceleration potential in this model can be adjusted To calculate an initial voltage is the voltage required to reach the potential depth of a region that is close to the interface at the intersection of the layers.

The initial voltage can also be selected from a lookup table based on the nature of the sample and the type of defect expected to occur. Examples of typical interface barriers are given in Table 1, and it can be seen that these ranges are between 1 keV and 0 keV.

A flow chart showing the steps within the method is shown in FIG. The initial selection of the voltage for the incident beam occurs at step 46. A scan of sample 12 is then performed at the selected voltage to produce a two-dimensional image of a plane spanning an unknown depth X within the sample (step 47), and incorporated in the selected electron An electronic flight simulation of the backscattered electrons under the energy of the beam determines the coarse penetration depth of the beam and the depth from which the backscattered electrons are generated. A suitable simulation soft system contains the "Small World Electron Flight Simulator".

By performing a simulation, it is recognized that the backscattered electrons emitted from the sample are being It is possible to generate where it came from. If we wish to image defects close to the same surface, then an electron beam with minimal penetration is desirable. If we need to image close to the buried junction, the simulation indicates if the correct depth has arrived. If necessary, the secondary electrons emitted from the closer surface can be filtered out and thus not used in the microscope image.

The simulated backscattered image is evaluated (step 48) to see if the beam penetration coincides with the plane of interest in which the defect of interest within the sample 12 is located. Initially, the incident beam energy is the energy that is unlikely to match the defect, and the depth X represents a plane that is too far out of the defect to be imaged within the sample. In this example, the energy of the incident beam is not close enough to the energy of the defect to image the defect itself, and the backscattered electron detector receives backscatter from a number of features across many layers. electronic. The simulation will show electrons originating from multiple depths of backscatter, as seen, for example, in Figure 5(a). When the electron beam 20 is too far within the sample, the defects associated with the junction are not visible due to other more prominent portions of the sample being imaged.

The bias voltage and, if necessary, the energy filter are then adjusted (step 49) to reduce the energy of the incident beam, thereby scanning a plane at a shallower depth closer to the surface of the sample 12. .

The repeated tuning of the bias voltage is performed to adjust the voltage with a step between 0.01 and 300V. The scanning and simulation of the backscattered electronic image is repeated, as seen in the bottom dashboard of Figure 5(a), until the backscattered simulated image shows that the backscattered electrons are from one along the sample. The inner plane is shown in the bottom dashboard of Figure 5(b), and the discrete features can be seen in the backscatter simulation, which typically extends from the source plane toward the sample surface. At this point, the energy of the electron beam at the sample matches the deficiency. The energy trapped, and the defects in a plane are clearly imaged and can be seen in the image of the scanning electron microscope. The combination of the image of the backscattered electrons and the combination of the bias voltages that are finely controlled at the sample allows for the selection of a particular depth associated with the defect. We can know which depth the defect is located without having to prepare the ultra-thin section of the sample for destruction of the sample by imaging by transmission electron microscopy. To improve resolution, additional energy filtering can be applied to the detectors, and thus backscattered electrons having a selected energy range are selected.

Once the energy of the electron beam at the sample matches the occlusion energy of the defect, the image of the two-dimensional scanning electron microscope represents a sample characteristic that is not deeper than a plane from which the defect originated. If additional energy filtering is used, higher energy secondary electrons generated from the surface of the sample can be filtered from the image, which enhances the visibility of defects in a plane within the sample 12.

Once the planes of the sources of the defects have been identified, non-destructive analysis occurs at step 50, wherein the total area of the defects can be determined by measuring the contrast in the image that is different from the area of the clean interface. Evaluate to produce quantified measurements of one of the affected areas. EDX analysis of backscattered electrons in this plane can also be performed to identify defective compositions.

In the case where the possible imaging voltage for the defect within the sample is unknown, an initial voltage of 1 keV applied to the stage 14 without an additional bias will then be selected, an initial scan will be It is performed and the electronic simulation of the backscatter will be evaluated. The bias voltage will be negatively or positively adjusted until discrete defects are visible in the electronic simulation of the backscatter, as this indicates that the plane from which the defect originated has been identified.

An example of imaging defects using this method will now be given with reference to Figures 2, 4 and 5. Figure 2 is a view showing a portion of a lateral GMR device 51 consisting of two 100 nm x 100 nm Ni ₈₀ Fe ₂₀ / Cu junctions, using high vacuum sputtering and lithography with stripped electron beams. It was fabricated in which the bottom 30 nm thick NiFe nanowire 52 was fabricated first, followed by a top 70 nm thick Cu wire 54. A junction 56 is formed where the lines cover each other. The device 51 is characterized by a conventional four-probe method which confirms that the resistance is ~5 kΩ/μm ² . The device is then intentionally exposed to a large current to cause damage and thus defects, which results in an increased resistance of a few MΩ/μm ² .

Figure 4 is a diagram used to illustrate the locations of the defects and how they are imaged. Defect 60 occurs along a plane Y, the exact location of which is unknown. The material of the covered wires 52, 54 placed on the substrate 62 is known to have a barrier between zero and 1 keV at its interface (symmetric with the barrier of the interface), and thus the objective 22 The voltage is initially adjusted to a voltage of 1 kV to produce an electron beam 20 of 1 keV energy incident on the wafer 51, wherein no bias voltage is applied to the stage 14. The backscattered electronic simulation is performed and the backscattered simulated image 70 (see Figure 5(a)) shows electrons from a plurality of depths. The difference in contrast in the SEM image 72 above the backscattered trajectory indicates the difference in conductivity of the sample, where the black dot represents the hole 74 in the underlying NiFe structure, and a possible The joint is broken. Thus, in this example, the 1 keV system represents the plane X across the sample, which indicates that the electron beam has penetrated further into the bottom NiFe wire, but the backscattered electrons are effectively effective because they do not have scattering. The length is prominently generated from these defects.

A continuous scanning is performed by using the objective lens to reduce the energy of the incident beam to about 200 eV, and then applying a bias voltage to the stage 14 for fine adjustment The entire step is to reduce the energy of the incident electron beam and to ensure that the beam penetrates less deeply within the sample. In Figure 5(b), only the surface morphology and any defects close to the surface are observed due to the relatively deep beam penetration.

Continuous adjustment of this voltage may be necessary to fine tune the imaged plane, but when the correct voltage is identified to match the potential of the defect, for example, the simulated backscatter shown in Figure 5(b) The trajectory in image 80 is visible, which clearly depicts the extent and depth of the defect within the uppermost Cu layer 54. By tuning the bias voltage applied to the stage, the location and depth of defects in any given interface can be identified.

Using this method, deeper defects in the same sample can also be imaged. Therefore, in order to image the defect at the interface (plane Z) between the lines 52, 54, an energy filter lower than -500 V can be utilized for the upper electron detector 34 to filter out the surface generated close to the sample. Secondary electronics. This allows the defects buried in the sample 12 to be clearly examined without the surface morphology from the coverage of the secondary electrons, which blurs the defects that may be present in the sample depth.

Energy dispersive X-ray chemical analysis (EDX) can be performed to identify the exact chemical composition of the defect, and thus whether the defect has been due to atomic segregation, formation of a compound, or in certain layers. The change in thickness occurs. This helps to identify where the defect is being produced in the process, and by knowing that the image seen on the screen exactly represents a defect in a lateral plane that straddles and is embedded within the sample, Quantitative measurements of the area of the affected junction surface can be performed.

For a sample in which the defect is expected to be well below 1 keV, then an additional energy filter may have to be placed between objective 24 and sample 12 to exploit the bias The voltage of the electron beam is first reduced to about 300 eV before the voltage is applied to produce a fine adjustment.

Accordingly, the present method provides a non-destructive method for evaluating the properties of buried junctions in an electronic device. By combining the simulated backscattered image and the subtle adjustment of the beam energy close to the sample, the imaging depth can be selected to facilitate a plane in which the imaging defect occurs without other features from below the plane. Overlapping images and know the depth of the plane within the sample. By controlling the beam energy, a contrast or local junction resistance can be observed for a controlled depth, which is representative of the junction properties. By combining this with chemical analysis, the source of any defects can be identified and the process can be modified to remove defects.

The incident energy of the beam controls the penetration of the electron beam into the sample, and the tuning of the bias voltage applied to the stage 14 ensures that the beam energy can be closely matched to between the defect and the clean interface The potential difference ensures that the defect itself can be imaged. Without the subtle tuning of such impact energy, accurate information about these defects cannot be obtained in a non-destructive manner.

For a production line facility, typically a defective sample will be taken and the bias voltage will be finely tuned over successive scans to identify the plane from which the defects originate. Precise voltage. Once the bias voltage required for the sample is identified, the process can perform a continuous non-destructive quantitative evaluation of the sample in a production line to determine if each sample is defective. If no defects are identified, the sample passes a quality assurance test.

46, 47, 48, 49, 50 ‧ ‧ steps

Claims (27)

An electron microscopy method for imaging atomic level defects associated with interfaces between different materials, the method comprising: repeatedly adjusting an electron energy adjustment region of a sample that is being imaged and for Each adjustment of the impact energy of the incident electron beam evaluates an electronic image that simulates backscattering until the source plane of the defect is found to change the impact energy of an incident electron beam. According to the method of claim 1, the method of using an electron microscope, wherein the depth of the source plane of the defect is obtained from the simulated backscattered electron image. The method of using an electron microscope according to claim 1 or 2, further comprising repeatedly adjusting the electron energy adjustment region until the impact energy of the incident electron beam matches and is related to an interface between the materials. The energy of a defect is up. The method of using an electron microscope according to any one of claims 1 to 3, wherein the simulated backscattered electron image is produced by an electronic flight simulation software. The method of using an electron microscope according to any one of claims 1 to 4, further comprising selecting an impact energy of the incident electron beam according to an expected potential energy at an interface between the at least two materials. According to the method of claim 5, wherein the expected potential energy at an interface is calculated by a look-up table or by a current-voltage characteristic from the sample at an interface The model of potential energy is obtained. An electron microscopy method according to claim 1 or 2 wherein an electron flight simulation soft system is used to select the impact energy. According to the method of claim 7, the method of using an electron microscope, wherein the electron is flying The line simulation soft system uses the scattering parameters of the different materials. A method of using an electron microscope according to any one of the preceding claims, wherein the electron energy adjustment region is set by applying a bias voltage to the sample. A method of using an electron microscope according to any one of the preceding claims, wherein the energy filtration is performed in relation to electrons generated from the sample. The method of using an electron microscope according to claim 9 or 10, wherein the bias voltage is adjustable in positive and negative values. The method of using an electron microscope according to any one of claims 9 to 11, wherein the bias voltage is between 0 and ±300V. A method of using an electron microscope according to any one of the preceding claims, wherein the energy of the incident electron beam is initially selected in the range of 0 to 5 keV. A method of using an electron microscope according to any one of the preceding claims, further comprising performing an energy dispersive X-ray analysis to identify a defective composition. A method of using an electron microscope according to any one of the preceding claims, further comprising the step of quantifying a total area covered by the defect. A method of imaging defects associated with an interface between different materials and imaged using an electron microscope, the method comprising: i) based on a potential at an interface between different materials Energy to select an initial energy of an electron beam incident on a sample containing the defect, ii) using the electron beam to perform a scan of the sample; iii) simulating a backscattering by the electron beam Electronic image; iv) evaluation of the backscattered electron image simulated to determine whether backscattered electrons are sourced From a plane in the sample; v) adjusting an electron energy adjustment region close to the sample to change a collision energy of the electron beam; vi) repeating steps ii to v until reaching the back as simulated The source plane of the defect indicated by the scattered electronic image. The method of imaging defects according to claim 14 of the patent application, wherein the electronic flight simulation soft system simulates the backscattered electronic image. The method of imaging defects according to claim 16 or 17, wherein the expected potential energy at an interface is calculated using a lookup table or by a current-voltage characteristic from the sample at an interface A model of potential energy is obtained. The method of imaging defects according to claim 18, wherein the electronic flight simulation soft system is used to select the initial energy of the electron beam. The method of imaging defects according to any one of claims 15 to 19, wherein the interface is one of: metal/semiconductor; ferromagnetic metal/nonmagnetic metal; organic material/metal; deposited on a substrate Film; multilayer interface; tunneling junction. The method of imaging defects according to any one of claims 15 to 20, wherein the electron energy adjustment region is set by applying a bias voltage to the sample. The method of imaging defects according to any one of claims 15 to 21, wherein the energy filtration is performed in relation to electrons generated from the sample. The method of imaging defects according to claim 21 or 22, wherein the bias voltage is adjustable in positive and negative values. The method of imaging defects according to any one of claims 21 to 23, wherein The bias voltage is between 0 and ±300V. The method of imaging defects according to any one of claims 16 to 24, wherein the energy of the incident electron beam is initially selected in the range of 0 to 5 keV. The method of imaging defects according to any one of claims 16 to 25, further comprising performing energy dispersive X-ray analysis to identify a defective composition. The method of imaging defects according to any one of claims 16 to 26, further comprising an analysis of quantifying a total area covered by the defects.