Semiconductor inspection device, electrical characteristic inspection device, and electrical characteristic inspection method

Abstract

The present disclosure improves sensitivity for determining normality / defect of an evaluation element having a main element, which is an inspection target, and a sub-element electrically connected to the main element. According to the present disclosure, a change characteristic is analyzed in which the luminance of an image of the evaluation element in an image of a sample acquired by scanning a pulsed charged particle beam over the sample under a predetermined irradiation condition and a predetermined intermittent condition changes depending on the irradiation condition or the intermittent condition of the pulsed charged particle beam, and a first feature amount and a second feature amount are extracted from the change characteristic. A setting is made as to which of the first feature amount and the second feature amount is a feature amount having a correlation with an electrical characteristic of the main element, and normality / defect of the evaluation element is determined on the basis of whichever of the first feature amount and the second feature amount is set as having the correlation with the electrical characteristic of the main element.

Description

Semiconductor testing apparatus, electrical characteristics testing apparatus, and electrical characteristics testing method

[0001] This disclosure relates to a semiconductor testing apparatus, an electrical characteristics testing apparatus, and an electrical characteristics testing method for testing the electrical characteristics of an evaluation element.

[0002] In the manufacturing process of semiconductor devices, the importance of inspecting the electrical characteristics of elements such as transistors, wiring, and contact plugs is increasing. One method for inline electrical characteristic inspection of semiconductor devices is the potential contrast method using scanning electron microscopy (SEM). The potential is generated on the sample surface due to charging caused by electron beam irradiation, and the brightness and darkness of the SEM image depend on the difference in surface potential, hence it is called potential contrast. The potential contrast method can identify electrical defects (opens, shorts) based on the difference in potential generated in the sample pattern. Patent document 1 discloses a method for inspecting electrical characteristics with high sensitivity by amplifying the difference in surface potential by the difference in transient changes in potential contrast caused by intermittent electron beam irradiation.

[0003] On the other hand, semiconductor wafers are equipped with Testing Element Groups (TEGs) for process control purposes. A TEG is a group of evaluation elements used to evaluate the elements that make up an integrated circuit, and to facilitate evaluation, it is formed independently for each evaluation unit (hereinafter referred to as "evaluation element"). In order not to reduce the integration density of the semiconductor wafer, it is often formed in scribe regions between chips. TEGs mounted on semiconductor wafers are also evaluated using in-line inspection and measurement equipment to evaluate the electrical characteristics of the evaluation elements. Patent Document 2 discloses a method for inspecting electrical characteristic defects using potential contrast with TEGs.

[0004] Japanese Patent Publication No. 2021-22440 Japanese Patent Publication No. 2010-45379

[0005] As described above, each TEG is formed separately and independently for each evaluation element, which is the unit of evaluation. However, in order for an evaluation element to be operational, it needs to be wired to supply power and signals to the element to be tested (hereinafter referred to as the "main element"). Therefore, an evaluation element is not a main element alone, but has a circuit configuration in which parasitic sub-elements are electrically connected. Consequently, the SEM image of the TEG reflects not the electrical characteristics of the main element that is actually to be tested, but the combined electrical characteristics of the main element and sub-elements. For this reason, when testing a TEG using the potential contrast method, the surface potential reflects the combined impedance of the main element and sub-elements, which may reduce the testing sensitivity of the main element or result in testing a sub-element other than the element that was intended to be tested.

[0006] This disclosure aims to improve the sensitivity of electrical characteristic testing of the main element that is originally intended for inspection.

[0007] A semiconductor inspection apparatus according to one embodiment of the present disclosure comprises: a stage mechanism on which a sample is placed; a charged particle optical system for irradiating the sample with a pulsed charged particle beam; a detector for detecting signal electrons emitted when the pulsed charged particle beam is irradiated onto the sample, and a detection system for forming an image of the sample based on the detection signal of the detector; and a control system for controlling the stage mechanism, the charged particle optical system, and the detection system to inspect the electrical characteristics of an evaluation element formed on the sample, wherein the evaluation element comprises a main element to be inspected and a sub-element electrically connected to the main element. The control system comprises: an apparatus control unit that controls the irradiation and intermittence conditions of the pulsed charged particle beam; a brightness analysis processing unit that analyzes the change characteristics of the brightness of the image of the evaluation element in the image of the sample, which change depending on the irradiation or intermittence conditions of the pulsed charged particle beam; a feature analysis unit that extracts a first feature quantity and a second feature quantity from the change characteristics; an inspection target feature quantity setting unit that sets which of the first feature quantity and the second feature quantity is correlated with the electrical characteristics of the main element; and a normal / defect determination unit that determines whether the evaluation element is normal or defective based on one of the first feature quantity and the second feature quantity which is set to be correlated with the electrical characteristics of the main element.

[0008] The present invention provides a semiconductor inspection apparatus capable of highly sensitively inspecting the electrical characteristics of the main element being inspected. Other challenges and novel features will become apparent from the description herein and the accompanying drawings.

[0009] This figure shows an example configuration of the semiconductor inspection apparatus of Example 1. This figure shows an example hardware configuration of the control device. This figure shows the inspection flow. This is an example structure of the TEG evaluation element. This is the equivalent circuit of the evaluation element 302. This is an example of SEM images acquired by changing the irradiation or intermittence conditions of the electron beam. This is a graph showing the dependence of the brightness L of a specified area on the interval time. This is a scatter plot of features acquired for the inspection location. This is an example of a Q-Q plot created for the time constant τ1. This is an example structure of the TEG evaluation element. This is the equivalent circuit of the evaluation element 702. This is a graph showing the dependence of the brightness L of a specified area on the interval time. This is an example of a Q-Q plot created for the first feature. This is an example of a Q-Q plot created for the second feature. This figure shows an example configuration of the semiconductor inspection apparatus of Example 3. This is a scatter plot of features acquired for the inspection location. This is a scatter plot of features acquired for the inspection location.

[0010] The semiconductor inspection apparatus, electrical characteristics inspection apparatus, and electrical characteristics inspection method of this disclosure will be described below with reference to the drawings.

[0011] Figure 1A shows an example of the configuration of a semiconductor inspection apparatus. The semiconductor inspection apparatus 100 includes a charged particle optical system (in this case, an electron optical system), a stage mechanism system, a detection system, and a control system that controls these.

[0012] The electron optical system includes, as its main components, an electron source 102, an intermittent irradiator 103 that pulses the electron beam from the electron source 102, a deflector 104 that scans the electron beam over the sample 108, and an objective lens 105 that focuses the electron beam onto the sample 108. The intermittent irradiator 103 is implemented to control the surface potential of the sample 108 caused by the charging from electron beam irradiation. As the electron source 102, a photo-excited electron source that can easily emit a pulsed electron beam from the electron source itself may be used. The stage mechanism system includes a stage 107 on which the sample 108 to be inspected is placed. The detection system includes a detector 106 that detects signal electrons emitted when the electron beam is irradiated onto the sample 108, and an image forming unit 111 that forms an image based on the irradiation position of the electron beam on the sample 108 and the output signal from the detector 106. The electron optical system, detector 106, and stage mechanism system are housed in a housing 101, and the inside of the housing 101 is a vacuum environment.

[0013] The electron optical system, stage mechanism system, and detection system are controlled by a control system. The control system includes a control device 120 and an electron beam control unit 110. Here, the set of controllers that control each component (electron source 102, intermittent irradiator 103, etc.) housed in the housing 101 is referred to as the electron beam control unit 110. The electron beam control unit 110 controls each component according to the control signals from the device control unit 121 that are output according to user settings from the input unit 131. Specifically, it controls each component housed in the housing 101 so that the electron beam irradiation conditions and intermittence conditions 103a set by the user are met.

[0014] In Figure 1A, the control device 120, which constitutes the control system, is shown with its functions represented by functional blocks. The control device 120 has many functions, but here, the device control unit 121, which controls the electron optical system, the stage mechanism system, and the detection system, and the functional block that performs inspection of semiconductor elements are shown.

[0015] The control device 120 is implemented as a computer, primarily comprising a processor (Central Processing Unit: CPU) 141, memory 142, storage device 143, input interface (I / F) 144, output I / F 145, communication I / F 146, and bus 147, as shown in Figure 1B. The processor 141 functions as a functional unit (functional block) that provides predetermined functions by executing processing according to a program loaded into the memory 142. The storage device 143 stores data and programs used by the functional unit. The input I / F 144 is connected to an input unit 131 such as a keyboard, pointing device, or operation panel, and the output I / F 145 is connected to an output unit 132 such as a display. The communication I / F 146 enables communication with other information processing devices via a network. These are connected to each other via the bus 147.

[0016] In the following explanation, when describing processing by a program, the program or functional unit may be described as the main component. However, the main hardware component is the processor, or the computer that includes such a processor. The computer, using the processor as appropriate, executes processing according to the program read into memory, utilizing resources such as memory and communication interfaces. Figure 1B shows an example of a CPU as the processor, but a GPU (Graphical Processing Unit) or the like may also be used. Furthermore, the processing to realize the function is not limited to software program processing; it can also be implemented with dedicated circuits. Applicable dedicated circuits include FPGAs (Field Programmable Gate Arrays) and ASICs (Application Specific Integrated Circuits). In addition, some of the functions of the control device 120, for example, some or all of the function that performs semiconductor element inspection, may be implemented as an application on cloud computing.

[0017] Figure 2 shows the inspection flow by the semiconductor inspection apparatus 100. Figure 3A shows an example of the structure of an evaluation element of the TEG to be inspected. The TEG 301 is composed of an evaluation element for evaluating an MTJ (Magnetic Tunnel Junction) element formed on a P-type substrate 313. In the evaluation element 302, the MTJ element 310 is formed on an insulating layer 315 and has a structure in which it is connected to an N-type layer 314 formed on the P-type substrate 313 by a plug 316. The MTJ element 310 also has a structure in which a tunnel insulating layer 312 is sandwiched between magnetic layers 311. The electrical characteristics of the evaluation element 302 are inspected to determine whether the MTJ element 310 is normal or defective. As an example of a defect, in the evaluation element 302b, the magnetic layers 311 of the MTJ element 310 are connected to each other at the side wall of the tunnel insulating layer 312.

[0018] When a pulsed electron beam is irradiated onto a normal evaluation element 302a and an SEM image is acquired, the potential contrast image obtained differs from the potential contrast image obtained when a pulsed electron beam is irradiated onto a defective evaluation element 302b and an SEM image is acquired. Due to the presence of this defect, a difference in surface potential occurs. This allows for the determination of whether the MTJ element 310 is normal or defective. However, in reality, the evaluation element 302 has a structure in which the MTJ element 310 is connected in series with a PN junction element formed by a P-type substrate 313 and an N-type layer 314 via a plug 316. What is actually observed in the potential contrast image is the potential contrast image of the evaluation element 302 in which the MTJ element 310, which is the element under inspection (hereinafter also called the "main element"), and the PN junction element, which is a parasitic element (hereinafter also called the "sub-element"), are connected in series. The potential contrast image does not only reflect the electrical characteristics of the main element, but also the combined electrical characteristics of the main and sub-elements, which leads to a decrease in sensitivity when inspecting the electrical characteristics of the main element (the MTJ element in the example of Figure 3A). In this embodiment, a feature quantity that reflects the electrical characteristics of the main element is extracted from the information obtained by combining the electrical characteristics of multiple elements, and normal / defective judgment is made from the extracted feature quantity, thereby improving the sensitivity of electrical characteristic inspection for the main element.

[0019] First, the TEG information setting unit 122 sets the information of the TEG evaluation element input from the input unit 131 (S101). The information of the evaluation element set here can include structural parameters related to the structure of the evaluation element (e.g., element size, etc.) and equivalent circuit information. As the equivalent circuit, for example, a SPICE model of the evaluation element that simulates the surface potential can be used. The information of the evaluation element set here is used for setting the first and second feature quantities described later, and for determining the feature quantities of the main element. The information of the evaluation element to be set is arbitrary, and one or more types of information can be registered. Figure 3B shows the equivalent circuit of the evaluation element 302 shown in Figure 3A. In the case of the evaluation element 302, both the main element 320 and the sub-element 321 can be represented as RC elements having a resistive component R and a capacitive component C. The main element 320 has a resistive component R ａ , capacitive component C ａ When it has such a property, the time constant τ1 is the resistance component R ａ and volume component C ａ It is calculated as the product of the two. Also, the sub-element 321 has a resistive component R ｂ , capacitive component C ｂ When it has the resistance component R, the time constant τ² is equal to the resistance component R. ｂ and volume component C ｂ It is calculated as the product of the two. In the case of the evaluation element 302, the relationship time constant τ1 < time constant τ2 holds. When setting an equivalent circuit as information for the evaluation element, information based on such a SPICE model can be registered.

[0020] Next, the device control unit 121 sets the irradiation conditions and intermittence conditions for the electron beam input from the input unit 131 (S102, S103), controls the electron beam control unit 110 to irradiate the sample 108 with a pulsed electron beam, and the image forming unit 111 forms an SEM image (S104). Examples of electron beam irradiation conditions include acceleration voltage, irradiation current, and irradiation position. Examples of electron beam intermittence conditions include pulse width and interval time between pulses. By pre-defining several combinations of predetermined irradiation conditions and predetermined intermittence conditions for the electron beam, registering each as an inspection recipe, and selecting one of them, the setting and resetting of the electron beam irradiation conditions and intermittence conditions can be easily performed.

[0021] The brightness analysis processing unit 127 analyzes the brightness of a specified area from the acquired SEM image (S105) and analyzes the dependence of the brightness on the irradiation conditions or intermittence conditions (S106). The specified area is specified by the user to determine whether the SEM image is suitable for inspection. Alternatively, one point of inspection may be arbitrarily selected from the inspection recipe. Figure 4 shows examples of SEM images acquired while changing the interval time, which is one of the intermittence conditions of the electron beam. SEM images 401 to 404 are SEM images acquired with interval times of 0.4 μs, 0.8 μs, 1.6 μs, and 3.2 μs, respectively, while keeping all other conditions the same, and the same MTJ element 310 of the evaluation element 302 is visible in the specified area 410. The brightness of the specified area 410 decreases as the interval time increases because the amount of discharge of charge from the MTJ element 310 due to irradiation with the pulsed electron beam increases as the interval time increases. Figure 5A is a graph showing the dependence of the luminance L of a specified area on the interval time. The horizontal axis represents the interval time, and the vertical axis represents the luminance of the specified area. Here, it is desirable for the user to visually judge whether the amplitude of the change in luminance dependent on the irradiation conditions or intermittence conditions is sufficient by displaying the analysis results on the output unit 132 (S107). Alternatively, a threshold for judging whether the amplitude of the change is good or bad may be predetermined, and the luminance analysis processing unit 127 may make the judgment. If the change is judged to be insufficient (failure), the process returns to step S102, and the electron beam irradiation conditions or intermittence conditions are reset. The device control unit 121 may select another inspection recipe with different electron beam irradiation conditions and intermittence conditions defined according to the reason for the failure. If the change is judged to be sufficient (good), the process proceeds to the next step S108.

[0022] The feature quantity setting unit 123 sets the first and second feature quantities input from the input unit 131 (S108). The first and second feature quantities are such that one correlates with the electrical characteristics of the main element and the other correlates with the electrical characteristics of the sub-element, and are features that can be extracted from the dependence of brightness on illumination conditions or intermittence conditions. The first and second feature quantities are set by the user based on the evaluation element information set in step S101.

[0023] For example, the first feature quantity is set as the time constant τ1, and the second feature quantity is set as the time constant τ2. In this case, the brightness L (target variable) of the image shown in Figure 5A of the MTJ element 310 can be expressed as (Equation 1) with the first feature quantity τ1 and the second feature quantity τ2 as explanatory variables.

[0024] Here, a is the coefficient and t is the interval time.

[0025] Next, the feature quantity to be inspected setting unit 124 sets which of the first and second feature quantities is to be inspected (S109). In the case of the evaluation element 302, it is known from its equivalent circuit that the time constant τ1 < time constant τ2. Therefore, based on the information set by the TEG information setting unit 122, the first feature quantity τ1 can be set as the feature quantity to be inspected. The feature quantity to be inspected setting unit 124 may also be specified by the user from the input unit 131. For example, suppose that in step S101, the information of the evaluation element set by the TEG information setting unit 122 does not include information about the magnitude of the time constants in the equivalent circuit, but the time constants τ1 and τ2 data measured for the evaluation element during the development stage of the element to be inspected (MTJ element) prior to the inspection in this embodiment were set. In such a case, the user may determine and set which time constant is the time constant of the element to be inspected based on the magnitude relationship of the measured time constants.

[0026] The luminance analysis processing unit 127 and the feature analysis unit 128 perform feature analysis at multiple locations on the sample based on the inspection recipe (S110). For the inspection locations specified in the inspection recipe, the luminance analysis processing unit 127 determines the change characteristics of the luminance L at the inspection location, which changes depending on the irradiation conditions or intermittence conditions of the pulsed electron beam, and by fitting (Equation 1) to these change characteristics, it obtains the first feature quantity τ1 and the second feature quantity τ2 for each inspection location. Figure 5B is a scatter plot with the horizontal axis as the time constant τ1 and the vertical axis as the time constant τ2.

[0027] Furthermore, the feature analysis unit 128 determines whether the inspection result is good or bad based on the data distribution of the feature to be inspected (in this case, the first feature τ1) (S111). The user may visually determine whether the data distribution is good or bad by displaying the data distribution of the feature to be inspected on the output unit 132, or the feature analysis unit 128 may make the determination based on predetermined criteria for determining whether the data distribution is good or bad. If the result is bad, the process returns to step S102 and the electron beam irradiation conditions or intermittence conditions are reset. The device control unit 121 may select another inspection recipe with different electron beam irradiation conditions or intermittence conditions defined according to the reason for the bad result. If the result is good, the process proceeds to the next step S112.

[0028] The normal / defective determination unit 126 determines whether a main element is normal or defective by statistically analyzing the data distribution of the time constant τ1 obtained for each inspection location, and stores the inspection results in the data storage unit 129 and outputs them to the output unit 132 (S112). For example, a Q-Q plot is created for the first feature quantity (time constant τ1) to distinguish between normal and defective main elements. Figure 6 is a Q-Q plot created for the time constant τ1. The horizontal axis is the time constant τ1, and the vertical axis is the standard deviation σ. A Q-Q plot is a graph used to compare data with a theoretical distribution and examine their similarity, and if they are similar, the plotted points will line up in a straight line. Figure 6 is a Q-Q plot created assuming that the frequency distribution of the time constant τ1 of a normal main element follows a normal distribution, and the theoretical distribution is a normal distribution. If there are no defective main elements, all the plotted points will line up in a straight line. In contrast, the Q-Q plot shown in Figure 6 is divided into a point group 601 showing characteristic distributions aligned along a first straight line and a point group 602 showing characteristic distributions aligned along a second straight line. It can be determined that either point group 601 or point group 602 represents the data distribution of a normal main element, while the other represents the data distribution of a main element with characteristics different from a normal main element, i.e., a main element with defects.

[0029] Regarding which of these point clouds represents the data distribution of a normal main element, it is determined according to the conditions set by the normal / defect determination condition setting unit 125. For example, in the development stage of the element to be inspected (MTJ element), information on the frequency distribution of the time constant τ1 measured for evaluation elements having normal main elements and evaluation elements having defective main elements is collected, and based on this information, the user sets the discrimination conditions. Alternatively, information for setting the discrimination conditions may also be obtained by inspecting whether the main element was finally normal or had a defect with respect to the time constant τ1 of the main element obtained by the method of this embodiment in another lot.

[0030] In the first embodiment, since both the main element and the sub - element in the evaluation element can be represented as RC elements, two feature quantities based on the equivalent circuit of the evaluation element can be set in association with the main element and the sub - element. However, the evaluation element to be inspected in the present disclosure is not limited to the case where both the main element and the sub - element are RC elements. As a second embodiment, an example in the case where the two feature quantities to be extracted are different feature quantities is shown. In the second embodiment, since the semiconductor inspection apparatus 100 and its inspection flow are the same as those in the first embodiment, the description will focus on the differences and overlapping descriptions will be omitted.

[0031] FIG. 7A shows an example of the evaluation element structure of the TEG to be inspected. The TEG 701 includes an evaluation element for evaluating a capacitor formed on the P - type substrate 313. In the evaluation element 702, the capacitor 710 is formed on the insulating layer 315 and has a structure connected to the N - type layer 314 formed on the P - type substrate 313 by the plug 316. Also, the capacitor 710 is a so - called MIM capacitor having a structure in which the insulating layer 712 is sandwiched by the metal layer 711. The electrical characteristics of the evaluation element 702 are inspected to determine whether the capacitor 710 is normal or defective. As an example of a defect, in the evaluation element 702b, the thickness of the insulating layer 712 of the capacitor 710 is thicker than the specification.

[0032] FIG. 7B shows the equivalent circuit of the evaluation element 702 shown in FIG. 7A. In the case of the evaluation element 702, the main element 720 can be represented as a capacitive element and the sub - element 321 can be represented as an RC element. The main element 720 has a capacitance component C ａIt has a sub-element 321 which has a resistive component R ｂ , capacitive component C ｂ When it has the resistance component R, the time constant τ² is equal to the resistance component R. ｂ and volume component C ｂ It is calculated as the product of the two. When setting an equivalent circuit as information for the evaluation element, information based on such a SPICE model can be registered.

[0033] Figure 8 is a graph obtained by the brightness analysis processing unit 127 analyzing the dependence of the brightness L of a specified region including the capacitor image on the interval time from the SEM image. The horizontal axis is the interval time, and the vertical axis is the brightness of the specified region. Note that Figure 8 shows graphs for both the normal case and the defective case. When the feature quantity setting unit 123 sets the amplitude A as the first feature quantity and the time constant τ of the evaluation element as the second feature quantity, the brightness L (objective variable) of the image of the capacitor 710 shown in Figure 7A can be expressed as (Equation 2) with the time constant τ (first feature quantity) and amplitude A (second feature quantity) as explanatory variables.

[0034] Here, t is the interval time.

[0035] In this case, unlike in Example 1, it is not clear from the brightness change characteristic model whether the first and second feature quantities correspond to the main element or the sub-element, respectively. Therefore, the feature quantity to be inspected setting unit 124 uses the data distribution of each feature quantity by the feature quantity analysis unit 128 to set which of the first and second feature quantities is the feature quantity to be inspected.

[0036] For example, suppose Figure 9A is the Q-Q plot for the first feature quantity τ, and Figure 9B is the Q-Q plot for the second feature quantity A. In the Q-Q plot of Figure 9A, all the plotted points are aligned on a single straight line, so it can be determined that the first feature quantity does not contain defect information. In contrast, the Q-Q plot of Figure 9B is divided into a group of points showing characteristic distributions aligned on different straight lines, so it can be determined that the second feature quantity contains defect information. In other words, in this case, the feature quantity to be inspected setting unit 124 sets the second feature quantity A as the feature quantity to be inspected.

[0037] In Example 1, the information of the evaluation element set by the TEG information setting unit 122 is used for discriminating the feature amount of the main element, and it was shown that information such as structural parameters (for example, the size of the element, etc.) regarding the structure of the evaluation element and an equivalent circuit are included. In Example 3, a case will be described which has a function of measuring the structural parameters of the main element from an SEM image and discriminating which of the first feature amount and the second feature amount has a correlation with the electrical characteristics of the main element based on the measured values.

[0038] FIG. 10 shows a configuration example of the semiconductor inspection apparatus of Example 3. The same reference numerals are given to the configurations equivalent to those of the semiconductor inspection apparatus in FIG. 1A, and duplicate explanations are omitted. The semiconductor inspection apparatus 100B shown in FIG. 10 includes a shape analysis unit 151 in the control device 120. The shape analysis unit 151 analyzes the structural parameters of the main element from the acquired SEM image. For example, it measures the diameter of the main element.

[0039] The inspection target feature amount setting unit 124 determines whether there is a correlation between the first feature amount and the second feature amount acquired by the feature amount analysis unit 128 and the structural parameters (for example, the diameter of the main element) acquired by the shape analysis unit 151, and sets the feature amount having a correlation as the inspection target feature amount.

[0040] FIGS. 11A and 11B show examples of scatter diagrams in which the first feature amount and the second feature amount are taken on the horizontal axis and the diameter of the main element as the structural parameter is taken on the vertical axis for the evaluation element 302 shown in FIG. 3A. In this case, it is estimated that the first feature amount τ1 has a correlation with the structural parameter, and it is estimated that the second feature amount τ2 has no correlation with the structural parameter. Based on this result, the inspection target feature amount setting unit 124 sets the first feature amount τ1 as the inspection target feature amount of the main element. This is effective when having knowledge about the inspection target element such that the shape of the main element is estimated as a factor causing a defect.

[0041] The above embodiments and modifications are described in detail for the purpose of making this disclosure easier to understand, and are not necessarily limited to having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment or modification with parts of another embodiment or modification, and it is also possible to add parts of other embodiments or modifications to the configuration of one embodiment or modification. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment or modification with other configurations.

[0042] 100, 100B: Semiconductor inspection equipment, 101: Housing, 102: Electron source, 103: Intermittent irradiator, 103a: Intermittent conditions, 104: Deflector, 105: Objective lens, 106: Detector, 107: Stage, 108: Sample, 110: Electron beam control unit, 111: Image forming unit, 120: Control device, 121: Device control unit, 122: TEG information setting unit, 123: Feature quantity setting unit, 124: Inspection target feature quantity setting unit, 125: Defect judgment condition setting unit, 126: Defect judgment unit, 127: Brightness analysis processing unit, 128: Feature quantity analysis unit, 129: Data storage unit, 131: Input unit, 132: Output unit, 141: Processor, 142: Memory, 143: Storage device, 144: Input I / F, 145: Output I / F, 146: Communication I / F, 147: Bus, 151: Shape analysis unit, 301, 701: TEG, 302, 702: Evaluation element, 310: MTJ element, 311: Magnetic layer, 312: Tunnel insulating layer, 313: P-type substrate, 314: N-type layer, 315: Insulating layer, 316: Plug, 320: Main element, 321: Sub-element, 401, 402, 403, 404: SEM image, 410: Designated area, 601, 602: Point cloud, 710: Capacitor, 711: Metal layer, 712: Insulating layer, 720: Main element.

Claims

1. The apparatus comprises: a stage mechanism on which a sample is placed; a charged particle optical system that irradiates the sample with a pulsed charged particle beam; a detection system that includes a detector for detecting signal electrons emitted when the pulsed charged particle beam is irradiated onto the sample, and forms an image of the sample based on the detection signal of the detector; and a control system that controls the stage mechanism, the charged particle optical system, and the detection system to inspect the electrical characteristics of an evaluation element formed on the sample, wherein the evaluation element comprises a main element to be inspected and a sub-element electrically connected to the main element, and the control system comprises: a device control unit for controlling the irradiation and intermittence conditions of the pulsed charged particle beam; a brightness analysis processing unit for analyzing the change characteristics of the brightness of the image of the evaluation element in the image of the sample, which change depending on the irradiation or intermittence conditions of the pulsed charged particle beam; a feature quantity analysis unit for extracting a first feature quantity and a second feature quantity from the change characteristics; and an inspection target feature quantity setting unit for setting which of the first feature quantity and the second feature quantity is correlated with the electrical characteristics of the main element. A semiconductor inspection apparatus comprising: a normal / defect determination unit that determines whether the evaluation element is normal or defective based on one of the first and second feature quantities set to have a correlation with the electrical characteristics of the main element; 2. The semiconductor inspection apparatus according to claim 1, wherein the feature analysis unit uses the brightness of the image of the evaluation element as the objective variable, extracts the first feature quantity and the second feature quantity using a model comprising a first explanatory variable indicating the electrical characteristics of the main element and a second explanatory variable indicating the electrical characteristics of the sub-element, and sets the first explanatory variable as the first feature quantity and the second explanatory variable as the second feature quantity.

3. The semiconductor inspection apparatus according to claim 2, wherein which of the first feature quantity and the second feature quantity is correlated with the electrical characteristics of the main element is set based on the measured electrical characteristics of the evaluation element formed on a sample different from the sample.

4. The semiconductor inspection apparatus according to claim 1, wherein the feature analysis unit uses the brightness of the image of the evaluation element as the objective variable, extracts the first feature and the second feature using a model of the first feature and the second feature, and sets which of the first feature and the second feature has a correlation with the electrical characteristics of the main element based on the data distribution of the first feature and the data distribution of the second feature.

5. The semiconductor inspection apparatus according to claim 1, wherein the feature analysis unit uses the brightness of the image of the evaluation element as the objective variable, extracts the first feature and the second feature using a model of the first feature and the second feature, and sets which of the first feature and the second feature is a feature that correlates with the electrical characteristics of the main element, based on whether or not there is a correlation with the structural parameters of the main element.

6. The semiconductor inspection apparatus according to claim 5, wherein the control system comprises a shape analysis unit that measures the structural parameters of the main element from an image of the sample.

7. The semiconductor inspection apparatus according to claim 1, wherein the control system comprises a normal / defective determination condition setting unit that sets determination conditions for determining whether the evaluation element is normal or defective, the normal / defective determination unit determines whether the evaluation element is normal or defective based on a data distribution of feature quantities set to have a correlation with the electrical characteristics of the main element, and the determination conditions are set based on the measured electrical characteristics of the evaluation element formed on a sample different from the sample.

8. In the semiconductor inspection apparatus according to claim 1, if the brightness analysis processing unit determines that the change characteristics are insufficient, or if the feature analysis unit determines that the data distribution of the feature quantities set to have a correlation with the electrical characteristics of the main element is poor, the apparatus control unit readjusts the irradiation conditions and / or intermittence conditions of the pulsed charged particle beam.

9. An electrical characteristic inspection apparatus for inspecting the electrical characteristics of an evaluation element formed on a sample, wherein the evaluation element comprises a main element to be inspected and a sub-element electrically connected to the main element, a brightness analysis processing unit that analyzes the change characteristics of the brightness of the image of the evaluation element in an image of the sample obtained by scanning the sample with a pulsed charged particle beam under predetermined irradiation conditions and predetermined intermittence conditions, which change depending on the irradiation conditions or intermittence conditions of the pulsed charged particle beam, a feature analysis unit that extracts a first feature quantity and a second feature quantity from the change characteristics, an inspection target feature quantity setting unit that sets which of the first feature quantity and the second feature quantity is a feature quantity that correlates with the electrical characteristics of the main element, and a normal / defect determination unit that determines whether the evaluation element is normal or defective based on one of the first feature quantity and the second feature quantity that is set to correlate with the electrical characteristics of the main element.

10. The electrical characteristic inspection device according to claim 9, wherein the feature analysis unit uses the brightness of the image of the evaluation element as the objective variable, extracts the first and second feature quantities using a model comprising a first explanatory variable indicating the electrical characteristics of the main element and a second explanatory variable indicating the electrical characteristics of the sub-element, and sets the first explanatory variable as the first feature quantity and the second explanatory variable as the second feature quantity.

11. The feature analysis unit, wherein the feature analysis unit uses the brightness of the image of the evaluation element as the objective variable, extracts the first feature and the second feature using a model of the first feature and the second feature, and sets which of the first feature and the second feature is correlated with the electrical characteristics of the main element based on the data distribution of the first feature and the data distribution of the second feature, in an electrical characteristic inspection device.

12. The feature analysis unit, wherein the feature analysis unit uses the brightness of the image of the evaluation element as the objective variable, extracts the first feature and the second feature using a model of the first feature and the second feature, and sets which of the first feature and the second feature is a feature that correlates with the electrical characteristics of the main element, based on whether or not there is a correlation with the structural parameters of the main element.

13. The electrical characteristic inspection device according to claim 9, further comprising a normal / defective determination condition setting unit for setting determination conditions for determining whether the evaluation element is normal or defective, wherein the normal / defective determination unit determines whether the evaluation element is normal or defective based on a data distribution of feature quantities set to have a correlation with the electrical characteristics of the main element, and the determination conditions are set based on the measured electrical characteristics of the evaluation element formed on a sample different from the sample.

14. An electrical characteristics inspection method for inspecting the electrical characteristics of an evaluation element formed on a sample using an electrical characteristics inspection device, wherein the electrical characteristics inspection device comprises a brightness analysis processing unit, a feature analysis unit, an inspection target feature setting unit, and a normal / defect determination unit, the evaluation element has a main element that is the object of inspection and a sub-element electrically connected to the main element, the brightness analysis processing unit analyzes the change characteristics of the brightness of the image of the evaluation element in an image of the sample obtained by scanning the sample with a pulsed charged particle beam under predetermined irradiation conditions and predetermined intermittence conditions, which change depending on the irradiation conditions or intermittence conditions of the pulsed charged particle beam, the feature analysis unit extracts a first feature and a second feature from the change characteristics, and the inspection target feature setting unit sets which of the first feature and the second feature has a correlation with the electrical characteristics of the main element. An electrical characteristic inspection method in which the normal / defect determination unit determines whether the evaluation element is normal or defective based on one of the first and second characteristic quantities set to have a correlation with the electrical characteristics of the main element.

15. The electrical characteristic inspection method according to claim 14, wherein the feature analysis unit uses the brightness of the image of the evaluation element as the objective variable, extracts the first and second feature quantities using a model comprising a first explanatory variable indicating the electrical characteristics of the main element and a second explanatory variable indicating the electrical characteristics of the sub-element, and sets the first explanatory variable as the first feature quantity and the second explanatory variable as the second feature quantity.

16. The electrical characteristic inspection method according to claim 14, further comprising a normal / defective determination condition setting unit for setting determination conditions for determining whether the evaluation element is normal or defective, wherein the normal / defective determination unit determines whether the evaluation element is normal or defective based on a data distribution of feature quantities set to have a correlation with the electrical characteristics of the main element, and the determination conditions are set based on the measured electrical characteristics of the evaluation element formed on a sample different from the sample.

Images