Testing system and testing method

Abstract

The purpose of the present invention is to provide a testing system capable of efficiently specifying a measurement condition or a testing condition for a semiconductor device. This testing system compares a measured value with a design value of a semiconductor device so as to estimate a failure mode indicating a failure type of an electrical characteristic that is the measured value, and on the basis of the degree of failure indicated by the measured value and the failure type indicated by the failure mode, extracts a testing point to be tested on the semiconductor device (see figure 1).

Description

Inspection system and inspection method

[0001] The present invention relates to a technique for inspecting semiconductor devices.

[0002] One method for inspecting semiconductor devices uses images obtained by irradiating a semiconductor device with an electron beam (e.g., CD-SEM (Critical Dimension SEM)). For example, the level of potential of a semiconductor device can be obtained as a voltage contrast image, which can be used to measure or inspect the electrical characteristics of the semiconductor device.

[0003] Japanese Patent Application Laid-Open No. 2006-124994 (Patent Document 1) describes an inspection technique using a voltage contrast image. The document aims to provide a failure analysis apparatus capable of efficiently performing layer analysis of semiconductor integrated circuit devices using the absorption potential method. The document describes the following technology: "In a failure analysis apparatus 10, an end point search unit 2 searches for a wiring route end point of pattern data of a conductive layer to be observed based on input information about the conductive layer to be observed. An expected image generation unit 3 determines the contrast brightness of the pattern data according to the layer of the end point to generate an expected voltage contrast image. A comparison unit 5 compares the expected voltage contrast image stored in the expected image storage unit 4 with the observed voltage contrast image stored in the observed image storage unit 1. A comparison result image generation unit 6 converts the match / mismatch information output from the comparison unit 5 into a comparison result image. A display unit 7 displays the comparison result image" (see Abstract).

[0004] Japanese Patent Application Laid-Open No. 2007-155449

[0005] In conventional techniques such as those described in Patent Document 1, optimal electronic and optical conditions (measurement conditions or inspection conditions) must be specified in advance to evaluate the electrical characteristics of a device under test. However, specifying these optimal conditions requires a great deal of time and know-how.

[0006] The present invention has been made in view of the above-mentioned problems, and has as its object to provide an inspection system that can efficiently identify measurement conditions or inspection conditions for semiconductor devices.

[0007] The inspection system according to the present invention compares design values ​​and measurement values ​​of a semiconductor device to estimate a failure mode that indicates a type of failure in the electrical characteristics of the measurement values, and extracts inspection points on the semiconductor device that should be inspected based on the degree of failure indicated by the measurement values ​​and the type of failure indicated by the failure mode.

[0008] The inspection system according to the present invention can perform optimal inspection and measurement suited to the object of observation. It can also shorten the time required to determine the inspection and measurement conditions. Other objects, configurations, advantages, etc. of the present invention will become apparent with reference to the following embodiments.

[0009] 4A is a block diagram of an inspection system according to a first embodiment. FIG. 4B is a process flow diagram for a computer system 50 to extract inspection point data. FIG. 4C is a schematic diagram for explaining an example of a method for estimating a failure mode. FIG. 4D is a diagram showing an example of a contrast image representing the measurement results of each device. FIG. 4E is a diagram showing the measurement results shown in FIG. 4A as a numerical table. FIG. 4F is a block diagram showing the configuration of a charged particle beam device 21 according to a first embodiment. FIG. 4G is a diagram showing an example of a user interface provided by a computer system 50 via a display unit 155. FIG. 4H is a block diagram of an inspection system according to a second embodiment. FIG. 4I is a flowchart for explaining the operation of the inspection system according to the second embodiment. FIG. 4I is a schematic diagram for explaining a method for a condition calculation unit 55 to calculate inspection conditions. FIG. 4I is a specific example of processing by the condition calculation unit 55. FIG. 4I is a diagram showing an example of calculation of inspection conditions based on a failure mode. FIG. 4H is a block diagram of an inspection system according to a third embodiment. FIG. 4I is a flowchart for explaining the operation of the inspection system according to the third embodiment. FIG. 4I is a data flow of processing performed by a computer system 50. FIG. 4H is a block diagram of an inspection system according to a fourth embodiment. FIG. 4I is a diagram showing an example of a grading result output by a failure mode determination unit 56. FIG. 4I is a flowchart for explaining the operation of the inspection system according to the third embodiment. 10 is a flowchart for outputting optimal analysis conditions to an analysis device based on an estimated failure mode.

[0010] 1 is a configuration diagram of an inspection system according to a first embodiment of the present invention. The system illustrated in FIG. 1 includes design data describing layout data and / or symbolic data formats (e.g., GDSII (GDS2), GL1, OASIS, map files, or other formats for storing design data structures) of an integrated circuit, an Electronic Design Automation (EDA) tool, and a computer system 50.

[0011] 1 may further be connected to other measurement devices 20, such as an optical inspection device, an OCD (Optical Critical Dimension), a CD-SEM, an EBI (Electron Beam Inspection) device, and device monitor information for an etching device or a film deposition device. These devices (tools) are connected to each other via a bus so that they can communicate with each other.

[0012] The computer system 50 may include software for calculating the electrical characteristics, a storage medium for storing the software, and one or more processors for executing the software. Each device constituting the system may be equipped with a computer system, or multiple devices may be controlled by a single workstation.

[0013] The computer system 50 in FIG. 1 includes a memory 51. The memory 51 stores measurement values ​​40, design values ​​41, and calculation values ​​42 output from the EDA tool. The design values ​​41 are values ​​described in design data for the semiconductor device. The calculation values ​​42 are various values ​​calculated by simulation using the EDA tool. The measurement values ​​40 are the results of measuring an observation target using a measurement device 20 or the like. The memory 51 stores coordinates on the semiconductor device as well as the measurement values ​​40, design values ​​41, and calculation values ​​42 at those coordinates. A set of the measurement values ​​40, design values ​​41, and calculation values ​​42 is defined as data 45. The data 45 stored in the memory 51 is input to a failure mode estimation unit 52 in the computer system 50.

[0014] The failure mode estimation unit 52 includes a process simulator. Based on the input data 45, the process simulator estimates the shape and material state of the object of observation (a semiconductor device in this example), such as the etching depth and the amount of misalignment with the underlying structure. If the shape or material is defective, information about the defect state is described as failure mode data 46. In addition to the above processing, the failure mode estimation unit 52 calculates electrical characteristic values, including resistance and capacitance values, at specified coordinates.

[0015] The failure mode data 46 of each observation target calculated by the failure mode estimation unit 52 is input to the inspection point extraction unit 54. The inspection point extraction unit 54 extracts coordinates to be inspected (inspection point data 47) according to conditions input by the user (inspection time, class range to be inspected, etc.), and outputs the results to the charged particle beam device 21 (inspection device). The inspection point data 47 calculated by the inspection point extraction unit 54 may be provided not only to the charged particle beam device 21, but also to an optical inspection device, a film formation device, a post-process tester, etc.

[0016] 2 is a process flow diagram for extracting inspection point data by the computer system 50. Each step in FIG. 2 will now be described.

[0017] (FIG. 2: Step S101) Two or more data including the measurement values ​​40 and the design values ​​41 are input to the memory 51 from among (a) measurement values ​​40 including shape data acquired by the SEM (Tool 2 and Tool 11 in FIG. 1), (b) layout data of the integrated circuit including pattern and circuit information and / or design values ​​41 including symbol data format, and (c) monitor data 48 of the manufacturing equipment (for example, the stage temperature of the etching equipment, the emission spectrum during the etching process, and the type and pressure of the exhaust gas from the film forming equipment).

[0018] (FIG. 2: Step S102) The failure mode estimation unit 52 uses a process simulator based on the information input in S101 to calculate failure mode data 46. There may be two or more failure mode data 46 (i.e., failure modes at multiple locations may be calculated collectively). A specific example of a method for estimating a failure mode will be described later.

[0019] (FIG. 2: Step S103) The inspection point extraction unit 54 calculates inspection point data 47 based on the failure mode data 46 calculated in S102. Finally, the inspection point data 47 is output to the charged particle beam device 21 (inspection device). A specific example of a method for calculating the inspection point data 47 will be described later.

[0020] 3 is a schematic diagram illustrating an example of a method for estimating a failure mode. The upper part of FIG. 3 is a side cross-sectional view of a semiconductor device. The failure mode represents a type of failure that occurs in a semiconductor device. The failure mode can be estimated using design data (design values ​​41) of the three-dimensional shape of the semiconductor device and the results of actually measuring the semiconductor device (measurement values ​​40).

[0021] The location shown in the left diagram of Figure 3 has a relatively large opening diameter exposing the lower layer, so the etching solution comes into contact with the lower layer through the opening during the etching process, making it prone to over-etching. Therefore, if a defect occurs at this location, it is relatively likely that the defect is over-etching (i.e., over-etching can be identified as a possible defect type). This leads to deterioration of bonding performance. The failure mode estimation unit 52 predetermines a three-dimensional shape like the one shown in the left diagram of Figure 3 based on design data, and can estimate that over-etching has occurred at that location if the measurement value 40 deviates from the normal range. That is, the failure mode estimation unit 52 estimates that a failure mode of over-etching (or a failure mode of bonding performance degradation) has occurred at this location.

[0022] The position shown in the right diagram of Figure 3 has a relatively small opening diameter, so it can be said that there is a relatively high probability of a void defect occurring. This results in deterioration of contact performance. The failure mode estimation unit 52 predetermines the three-dimensional shape shown in the right diagram of Figure 3 based on design data, and if the measurement value 40 deviates from the normal range, it can estimate that a void defect has occurred at that position. In other words, the failure mode estimation unit 52 estimates that a failure mode: void defect (or failure mode: contact performance degradation) has occurred at that position.

[0023] FIG. 4A shows an example of the measurement results obtained by each device, expressed as a contrast image. Each image was acquired at measurement points 1 to 4. As shown by the grid in FIG. 4A, a measurement point may include multiple coordinates on a semiconductor device. It should be noted that FIG. 4B, which will be described later, describes the measurement values ​​for each coordinate included in each measurement point.

[0024] TOOL2 in Figure 1 is a CD-SEM, which measures the gate width of a transistor. The size of the gate width is represented by the shade of the Point 1 image. TOOL3 in Figure 1 is an ellipsometer (film thickness measurement device), which measures the film thickness of the gate film. The size of the film thickness is represented by the shade of the Point 2 image. TOOL11 in Figure 1 is a CD-SEM, which measures the amount of misalignment between the active region and the contact hole. The amount of misalignment is represented by the shade of the Point 3 image. TOOL12 in Figure 1 is a film deposition device, which monitors and measures the stage temperature during film deposition. The stage temperature is represented by the shade of the Point 4 image.

[0025] FIG. 4B shows the measurement results shown in FIG. 4A described as a numerical table. The failure mode estimation unit 52 receives this data as measurement values ​​40. The design data includes a process margin set at each measurement point. The inspection point extraction unit 54 calculates the deviation from the process margin for data associated with each observation object. The inspection point extraction unit 54 calculates the normal probability of the measurement point based on the deviation between the calculated deviation and the process margin. The inspection point extraction unit 54 further classifies the performance (class) of the observation object based on the severity of the estimated failure mode and the deviation. Since a low-class class (high degree of defect) is assumed to require inspection, the inspection point extraction unit 54 extracts positions classified as low-class as inspection points. The class can be set arbitrarily by the user. The user can also specify which grade range to inspect, and the inspection point may be set based on the time required for inspection.

[0026] In principle, the inspection point extraction unit 54 extracts, as inspection points, positions where there is a large deviation from the process margin (a class with a low normal rate). However, even when the deviation from the process margin is large, the influence of the deviation may be large or small depending on the failure mode (defect type). Therefore, the inspection point extraction unit 54 may extract inspection points by taking the failure mode into account in addition to the deviation amount. For example, it is possible to predefine the influence of the deviation from the process margin for each failure mode and correct the normal rate based on the influence and the actual deviation amount (e.g., subtract the influence degree x the deviation amount from the normal rate). Other appropriate methods may also be used. By considering the influence of the failure mode in addition to the deviation amount, it is possible to more appropriately extract positions to be inspected.

[0027] 5 is a block diagram showing the configuration of a charged particle beam device 21 according to embodiment 1. The charged particle beam device 1 includes an electron optical system 11, a stage mechanism system 12, a light irradiation system 13, a control system 14, and an operation system 15.

[0028] The electron optical system 11 is composed of an electron gun 111, a deflector 112, an electron lens 113, and an electron detector 114. The stage mechanism system 12 is composed of an XYZ stage 121 on which a sample 122 is placed. The interior of the housing of the electron optical system 11 is controlled to a high vacuum, and the stage mechanism system 12 is installed therein. The light irradiation system 13 includes a light source 131 and a light adjustment unit 132, and light is irradiated onto the sample 122 via a light introduction unit 133. The control system 14 is composed of an electron gun control unit 1411, a deflection signal control unit 142, an electron lens control unit 143, a detector control unit 144, a stage position control unit 145, a light control unit 146, a control message unit 147, and a calculation unit 148. The control message unit 147 writes and controls control values ​​to each control unit based on input information input from a sequence control unit 151.

[0029] An electron beam accelerated by the electron gun 111 is focused by an electron lens 113 and irradiated onto a sample 122. The irradiation position on the sample 122 is controlled by a deflector 112. The electron beam is controlled in accordance with the acceleration voltage, irradiation current, deflection conditions, and electron lens conditions set by a measurement item setting unit 152.

[0030] The control message unit 147 is a functional block that controls the components of the charged particle beam device 1. The control message unit 147 sends operational commands to the detector control unit 144, the electron gun control unit 141, and the like, based on, for example, observation conditions input from the sequence control unit 151. Each control message system controls the stage mechanism system 12 and moves the sample 122 to a predetermined position, based on, for example, light and electron conditions input from the sequence control unit 151. The control message unit 147 controls the detection process of emitted electrons by the electron detector 114 by controlling the supply of power and control signals to the electron detector 114 via the detector control unit 144. Based on the conditions input in the measurement item setting unit 152, the control message unit 147 sends information on light irradiation conditions, such as wavelength, light intensity, and irradiation timing, to the light control unit 146, thereby controlling the operation of the light source 131, the light adjustment unit 132, and the like. More specifically, for example, the light control unit 146 controls the light intensity and wavelength of the light irradiated from the light source 131. The light control unit 146 instructs the light adjustment unit 132 to adjust the traveling direction and polarization of the light emitted from the light source 131. The light control unit 146 may be implemented, for example, by manual operation, or by a program executed on a personal computer equipped with a processor such as a CPU. The control message unit 147 may be configured, for example, by an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).

[0031] Light emitted from the light source 131 is irradiated onto a position on the surface of the sample 122 via the light introduction section 133. In this embodiment, a laser capable of irradiating multiple wavelengths or a laser capable of irradiating a single wavelength is installed as the light source 131 for each required wavelength. The light introduction section 133 has a slit configuration, which allows the light shape to be controlled arbitrarily. This makes it possible to control the irradiation location depending on the irradiation location and pattern.

[0032] In addition to the above, the operation system 15 also includes an output unit 154 and a display unit 155. The output unit 154 outputs the processing results of the calculation unit 148 in, for example, an appropriate data format. The display unit 155 displays the processing results of the calculation unit 148 on, for example, a screen of a user interface described below.

[0033] The device information input unit 153 of the operation system 15 is a block with the function of inputting the structure and circuit information of the sample to be observed. By inputting CAD information, etc., used in the device pattern design, the size and electrical circuit information of each pattern are associated, and the irradiation position of the electron beam and light is controlled accordingly. The input format of the pattern circuit information and size is not particularly specified, as long as it provides information such as the pattern coordinates, circuit information, and connection relationships of each pattern. For example, CAD data or text data that indicates the coordinates and circuit structure may be used. Circuit designation of each pattern is possible even without using CAD data. An example of the designation procedure is as follows: (1) An SEM image of the inspection area is acquired; (2) The signal obtained from the electron detector 114 is converted into gradation information by the processing unit and calculation unit, and the information is displayed on the display unit 155 via the output unit 154; (3) The user assigns each pattern type on the GUI (described later), and the information is sent to the device information input unit 153. This flow also allows the type of pattern to be designated for each observation position.

[0034] The measurement item setting unit 152 references the device information input by the device information input unit 153 and generates the light and electron beam conditions required to achieve the inspection items specified on the GUI. The sequence control unit 151 creates a measurement sequence based on the light and electron beam conditions set by the measurement item setting unit 152. The sequence control unit 151 generates an irradiation sequence and transmits an instruction signal to the control system. When the test item to be inspected is selected by the measurement item setting unit 152, a light and electron beam irradiation sequence is generated. At this time, the user can also customize the light and electron beam conditions on the GUI, which will be described later. By selecting the sequence visualization button on the GUI, the flow of the light and electron beam irradiation sequence can also be visually confirmed. The user may also set the light and electron beam irradiation timing as desired.

[0035] The control messenger unit 147 controls the detector control unit 144, the electron gun control unit 141, the deflection signal control unit 142, the electron lens control unit 143, the stage position control unit 145, the light control unit 146, etc. in accordance with the sequence specified by the sequence control unit.

[0036] The light control unit 146 controls light parameters such as the wavelength, irradiation amount, and peak power of the light. The light control unit 146 includes a light source 131 and a light adjustment unit 132. The light source 131 may be a white light source, a semiconductor diode laser, a solid-state laser, or other light sources. The light adjustment unit 132 adjusts the light parameters, such as the light amount, wavelength, polarization plane, irradiation angle onto the sample, pulsed laser / CW (continuous wave) laser, and repetition rate (in the case of a pulsed laser). The light adjustment unit 132 also has the function of controlling and irradiating the light irradiation timing according to the sequence information generated by the measurement item setting unit 152 and the sequence control unit 151. The mechanism having the light irradiation timing function may be a mechanical shutter, an electro-optical element or an acousto-optical element for blocking, or another mechanism capable of realizing optical switching control.

[0037] The electron optical system 11 irradiates an electron beam under the acceleration voltage, irradiation current, deflection conditions, and electron lens conditions set by the measurement item setting unit 152. The timing of electron beam irradiation is controlled by controlling the timing of irradiation and non-irradiation based on the sequence generated by the measurement item setting unit 152 and the sequence control unit 151. The electron beam accelerated by the electron gun 111 is focused by the electron lens 113 and irradiated onto the sample 122. The position at which the electron beam is irradiated onto the sample 122 and the observation magnification are controlled by the deflector 112.

[0038] 6 shows an example of a user interface provided by the computer system 50 via the display unit 155. The user interface (GUI) can present, for example: (a) a charged particle beam irradiation condition input section for inputting irradiation conditions for an electron beam for observation; (b) a light irradiation condition input section for inputting light irradiation conditions; (c) a histogram showing the relationship between the normal rate and the number of inspection points having that normal rate; (d) an SEM observation image of the semiconductor device; and (e) the result of estimating a failure mode for each coordinate on the semiconductor device.

[0039] Summary of First Embodiment The inspection system according to the first embodiment acquires design values ​​and measurement values ​​from previous processes in the manufacturing process of semiconductor devices, and can output inspection points by estimating the failure mode of the element to be evaluated based on that information. This makes it possible to reduce the inspection time. The inspection system according to the first embodiment estimates the failure mode based on data from one lot, but comparison between lots may also be used when performing failure estimation.

[0040] In general, the performance of semiconductor device manufacturing equipment has an in-plane distribution, and therefore the types and degrees of failure modes vary even within the same semiconductor device. Therefore, it is difficult to identify a failure mode in a semiconductor device that has multiple failure modes. The inspection system according to the first embodiment can estimate the failure mode for each coordinate on the semiconductor device, even for such a semiconductor device.

[0041] Second Embodiment In a second embodiment of the present invention, a system will be described that outputs optimal inspection and measurement conditions based on the failure mode estimated according to the first embodiment, and performs inspection and measurement according to the output conditions.

[0042] 7 shows a block diagram of an inspection system according to embodiment 2. Instead of or in combination with the inspection point extraction unit 52 described in embodiment 1, the computer system 50 includes a condition calculation unit 55. The condition calculation unit 55 calculates observation conditions to be used when the charged particle beam device 21 observes the semiconductor device based on the failure mode data 46 estimated by the failure mode estimation unit 52, and outputs the results to the charged particle beam device 21. The charged particle beam device 21 performs inspection and measurement using the observation conditions.

[0043] 8 is a flowchart illustrating the operation of the inspection system according to embodiment 2. Steps S101 to S102 are the same as those in embodiment 1. Steps S105 to S106 are performed instead of or in combination with step S103.

[0044] (FIG. 8: Step S105) The condition calculation unit 55 receives the failure mode data 46. At this time, the failure mode data 46 stores information such as resistance values, capacitance values, pattern structure information, and defect occurrence locations. Based on the failure mode data 46, the condition calculation unit 55 calculates measurement conditions for the charged particle beam and electromagnetic wave, an inspection and measurement sequence, and the like (i.e., inspection conditions). A specific method for calculating the inspection conditions will be described later.

[0045] (FIG. 8: Step S106 ) The charged particle beam device 21 performs an inspection at each coordinate based on the conditions calculated by the condition calculation unit 55 .

[0046] FIG. 9 is a schematic diagram illustrating a method by which the condition calculation unit 55 calculates the inspection conditions. FIG. 9 is the same as FIG. 3 except that the inspection conditions are calculated. The condition calculation unit 55 acquires a predefined correspondence relationship between the failure mode and the inspection conditions, and can determine the inspection conditions based on the correspondence relationship. Data describing the correspondence relationship may be stored in advance in the memory 51. For example, the inspection conditions can be determined as follows:

[0047] The condition calculation unit 55 estimates the resistance and capacitance values ​​at the inspection position when a defect occurs based on the calculated values ​​42. It calculates electron beam irradiation conditions suitable for observing the estimated resistance and capacitance values. It also determines the inspection point and observation magnification based on the aperture diameter and defect occurrence probability. It also determines the light irradiation conditions to be used during observation based on the estimated failure mode. For example, when observing bonding performance degradation due to over-etching, infrared light is irradiated at a power of 10 mW or less, and when observing contact performance degradation due to void defects, ultraviolet light is irradiated at a power of 10 mW or more. These inspection conditions are stored in the memory 51 in advance as a correspondence relationship between the basic conditions (e.g., type of light) and the failure mode, and specific numerical values ​​are calculated based on the measured values ​​40, the design values ​​41, and the calculated values ​​42. Inspection conditions may be specified for each failure mode category (described later), and the inspection conditions corresponding to the category may be acquired.

[0048] 10A shows a specific example of processing by the condition calculation unit 55. Measurement Points 1 to 4 are the same as those in FIGS. 4A and 4B. The failure mode estimation unit 52 receives the measurement results at each measurement point. Here, it is assumed that the same measurement results as those in FIG. 4B have been received. The failure mode estimation unit 52 calculates the normal rate based on the amount of deviation from the process margin, for example, as in the first embodiment. As in the first embodiment, the failure mode estimation unit 52 estimates the type of failure based on the three-dimensional shape at the inspection position, identifies the degree of influence of the amount of deviation, and estimates the failure mode based on this.

[0049] FIG. 10B is a diagram illustrating the results of calculating inspection conditions based on failure modes. In this example, the failure mode estimation unit 52 classifies failure modes into 26 types, from A defects to Z defects. A defects are the best, and Z defects are the worst. The failure mode estimation unit 52 identifies the failure mode with the highest percentage for each inspection position. For example, the darkest grid in FIG. 10B has A defects, and the percentage of A defects in all defects is 85%. Since multiple types of defects may be included in the same grid, the total percentage may exceed 100% (159.2% in FIG. 10B).

[0050] The condition calculation unit 55 acquires inspection conditions corresponding to the failure mode classification of the inspection position. When the inspection position includes multiple failure modes (classifications), the inspection conditions corresponding to the most common failure mode are adopted in principle. However, if there is one of the multiple failure modes that has a large effect on the three-dimensional shape of the inspection position, the inspection conditions corresponding to the failure mode with the largest effect may be adopted regardless of the proportion of the failure mode in number. If the degree of influence is taken into consideration when estimating the failure mode, the failure mode itself includes the degree of influence, so basically it is sufficient to adopt the inspection conditions corresponding to the failure mode.

[0051] Summary of Second Embodiment The inspection system according to the second embodiment estimates the optimal inspection and measurement conditions for each inspection and measurement point based on the estimated failure mode. This makes it possible to perform inspections using the optimal inspection conditions for the object to be observed without requiring a great deal of time or know-how.

[0052] <Embodiment 3> In embodiment 3 of the present invention, a system will be described that outputs optimal inspection and measurement conditions based on an estimated failure mode, performs inspection and measurement according to the output conditions, and identifies or outputs the possibility of a process (a process prior to inspection and measurement) in which a failure occurred based on the inspection and measurement results.

[0053] 11 shows a block diagram of an inspection system according to the third embodiment. In addition to the configuration shown in the second embodiment, a failure mode determination unit 56 is provided. The charged particle beam device 21 performs an inspection using the inspection conditions calculated by the condition calculation unit 55 and outputs the results as inspection measurement data 49. The failure mode determination unit 56 uses the inspection measurement data 49 and the failure mode data 46 to identify the failure mode that actually occurred and estimate the process in which the failure mode occurred. At this time, not only data from the same lot is used, but also data from different lots may be included in the data for statistical analysis.

[0054] 12 is a flowchart illustrating the operation of the inspection system according to embodiment 3. Steps S101 to S106 are the same as those in embodiment 2, and steps S107 and S108 are performed after step S106.

[0055] (FIG. 12: Step S107) The failure mode determination unit 56 identifies a failure mode corresponding to each observation point within the wafer surface based on the inspection measurement data 49 and the failure mode data 46 obtained from the charged particle beam device 21.

[0056] (FIG. 12: Step S108) The failure mode determination unit 56 outputs or identifies the process in which the failure mode occurred based on the failure mode identified in S107. For example, the failure mode determination unit 56 can obtain in advance a correspondence relationship between the combination of inspection values ​​that constitute the failure mode (including the combination of the failure modes themselves if multiple failure modes occur) and the manufacturing process that has the potential to cause the failure mode, and can estimate the process in which the failure mode occurred based on the correspondence relationship. At this time, by recording the probability of the failure mode occurring together with the correspondence relationship, the probability can also be estimated.

[0057] (FIG. 12: Step S108: Supplement) For example, the tendency of in-plane variation in the film thickness of a semiconductor device affects the probability of failure in the process of forming that film. Therefore, the probability of failure can be estimated in advance based on the tendency of variation, and the estimated result can be defined in advance as a correspondence table for this step. Alternatively, if the failure mode involves multiple processes, the probability of failure occurring for each of those processes is known to a certain extent, and this can be defined in advance in the correspondence table as a probability for this step. By using this information, it is possible to estimate the process that caused the failure mode and the probability that that process caused the failure mode.

[0058] 13 shows a data flow of processing performed by the computer system 50. There may be multiple failure modes estimated by the failure mode estimation unit 52 for the same inspection position. This situation occurs, for example, between failure modes with similar properties. In this case, the condition calculation unit 55 calculates inspection conditions for each failure mode, and the charged particle beam device 21 performs inspection using the respective inspection conditions. The failure mode determination unit 56 may acquire each inspection result data 49 and compare them to re-estimate the failure mode that actually occurred.

[0059] <Summary of Third Embodiment> The inspection system according to the third embodiment estimates the manufacturing process in which the failure mode occurred and the probability that the process caused the failure mode based on the estimated failure mode. By feeding back the estimation result to the manufacturing process, defects can be quickly improved.

[0060] Fourth Embodiment In a fourth embodiment of the present invention, a system will be described in which inspection and measurement are performed using estimated optimal inspection conditions, and the quality of the inspection points is evaluated based on the inspection results.

[0061] 14 shows a block diagram of an inspection system according to embodiment 4. In addition to the configuration described in embodiment 3, a failure mode determination unit 56 grades the inspection results based on the inspection measurement data 49 received from the charged particle beam device 21. The rest is the same as embodiment 3.

[0062] FIG. 15 shows an example of the grading results output by the failure mode determination unit 56. Measurement points 1 to 4 and the measurement results at each measurement point are the same as those in FIGS. 4A and 4B. The failure mode determination unit 56 acquires the inspection result data 49 for each inspection point. The failure mode determination unit 56 grades the inspection points according to the magnitude of the impact that the inspection results have on the performance of the semiconductor device. For example, the impact that the defect has on the performance of the semiconductor device can be evaluated based on the severity of the defect type represented by the failure mode and the degree of the defect (severity of the defect) represented by the inspection result data 49. For example, if a failure mode (defect type) with a high severity occurs and the degree of the defect is high in the inspection results, the impact on performance can be evaluated as being large.

[0063] When determining the impact on performance from the severity of the defect type and the degree of the defect, the relationship between these can be defined in advance as a data table, and the impact can be estimated based on this. Alternatively, the know-how and other empirical knowledge of the device manufacturer can be entered in advance in the data table, and the impact can be estimated based on this. These methods can also be combined.

[0064] 16 is a flowchart illustrating the operation of the inspection system according to the embodiment 3. Steps S101 to S106 are the same as those in the embodiment 2, and step S109 is performed after step S106.

[0065] (FIG. 16: Step S109) The failure mode determination unit 56 grades each inspection measurement point based on the inspection measurement data 49 and the estimation data 46 obtained from the charged particle beam device 21. The grades may be set by the user, or may be calculated using any of the above methods.

[0066] Summary of Fourth Embodiment The inspection system according to the fourth embodiment performs inspection and measurement using inspection and measurement conditions determined based on an estimated failure mode, and ranks the inspection results based on the impact on device performance. This allows the estimated failure mode to be used not only to estimate optimal inspection conditions but also in the overall evaluation of the inspection results.

[0067] Fifth Embodiment In a fifth embodiment of the present invention, a system will be described that performs an inspection using estimated optimal inspection and measurement conditions, and outputs optimal analysis conditions to an analysis device that analyzes the cause of a failure mode based on the inspection results.

[0068] 17 is a block diagram of an inspection system according to the fifth embodiment. The failure mode corresponds to a failure type that is likely to occur in a semiconductor device based on the three-dimensional shape of that shape. The failure mode determination unit 56 identifies the failure type that corresponds to the estimated failure mode based on this correspondence. The type of analysis device suitable for observing a specific failure type is known in advance (e.g., for a pattern width failure, a Critical Dimension SEM is used, etc.). The failure mode determination unit 56 then notifies the analysis device suitable for analyzing the cause of the defect of the position / depth to be analyzed, analysis conditions, etc. The rest is the same as in the third and fourth embodiments.

[0069] 18 is a flowchart for outputting optimal analysis conditions to an analysis device based on an estimated failure mode. Steps S101 to S107 are the same as those in the third embodiment. Step S110 is performed after step S107.

[0070] (FIG. 18: Step S110) Based on the inspection results, the failure mode determination unit 56 identifies the process that caused the defect and also identifies an analysis device suitable for analyzing the defect. The failure mode determination unit 56 notifies the analysis device of information such as the position / depth to be analyzed, the analysis conditions, and the process that caused the defect.

[0071] <Summary of Fifth Embodiment> The inspection system according to the fifth embodiment feeds forward detailed information about the source and location of a failure mode to an analysis device used for detailed analysis when a failure occurs. This reduces the time required to determine analysis conditions and improves the accuracy of failure analysis.

[0072] <Modifications of the present invention> The present invention is not limited to the above-described embodiment, and includes various modifications. For example, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment, without departing from the spirit of the invention. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with another configuration.

[0073] In the above embodiments, examples of electrical characteristics inspected by the inspection system include, but are not limited to, the resistance value of a semiconductor device, the capacitance value of a semiconductor device, the frequency characteristics of a semiconductor device, the response speed of a semiconductor device, the amount of leakage current of a semiconductor device, the current-voltage characteristics of a switching element provided in a semiconductor device, and the threshold voltage of a switching element provided in a semiconductor device.

[0074] In the above embodiments, each functional unit of the computer system 50 can be configured by hardware such as a circuit device that implements that function, or by a computing device such as a CPU (Central Processing Unit) that executes software that implements that function.

[0075] Measurement and inspection both mean the same thing in that they measure the electrical characteristics of a semiconductor device, and unless otherwise specified, they may be considered synonymous in the present invention.

[0076] 50: Computer system 52: Failure mode estimation unit 54: Inspection point extraction unit 55: Condition calculation unit 56: Failure mode determination unit

Claims

1. An inspection system for inspecting a semiconductor device, comprising: a computer system for inspecting the semiconductor device based on measurement values ​​describing the results of measuring the electrical characteristics of the semiconductor device using a charged particle beam, wherein the computer system receives design values ​​and the measurement values ​​of the semiconductor device, wherein the computer system estimates a failure mode representing a type of failure of the electrical characteristics of the measurement values ​​by comparing the design values ​​with the measurement values, and wherein the computer system extracts inspection points on the semiconductor device to be inspected based on the degree of failure represented by the measurement values ​​and the type of failure represented by the failure mode, and outputs the results.

2. The inspection system according to claim 1, characterized in that the computer system estimates defect types that may occur on the semiconductor device as defect type candidates from design data of the semiconductor device, and the computer system estimates the defect mode that may occur on the semiconductor device based on the defect type candidates and the degree of defect represented by the measurement value.

3. The inspection system described in claim 1, characterized in that: the computer system obtains, as the design value, a process margin indicating a range that is acceptable as a normal range of the measurement value; the computer system identifies the degree of defect represented by the measurement value based on the degree to which the measurement value deviates from the process margin; and the computer system determines a priority for performing inspection based on the severity of the defect type represented by the defect mode and the degree of defect represented by the measurement value, and extracts the inspection points to be inspected according to the priority.

4. The inspection system according to claim 1, further comprising an inspection device that inspects the semiconductor device by irradiating the semiconductor device with the charged particle beam, wherein the computer system uses the estimated failure mode to estimate optimal inspection conditions for the inspection device when inspecting the semiconductor device, and the inspection device inspects the semiconductor device using the estimated inspection conditions.

5. The inspection system according to claim 4, characterized in that the computer system obtains a predetermined first correspondence relationship between the failure mode and the optimal inspection condition, and the computer system estimates the optimal inspection condition according to the first correspondence relationship.

6. The inspection system described in claim 5, characterized in that the computer system classifies the degree of defect based on the severity of the defect type represented by the failure mode and the degree of defect represented by the measurement value, the first correspondence relationship specifies the relationship between the failure mode and the optimal inspection conditions for each of the classifications, and the computer system estimates the optimal inspection conditions corresponding to the classification in accordance with the first correspondence relationship.

7. The inspection system according to claim 1, further comprising an inspection apparatus for inspecting the semiconductor device by irradiating the semiconductor device with the charged particle beam, wherein the computer system identifies a failure mode that has occurred in the semiconductor device based on an inspection result by the inspection apparatus, and wherein the computer system uses the identified failure mode to estimate a manufacturing process of the semiconductor device that caused the failure mode in the semiconductor device.

8. The inspection system according to claim 7, wherein the computer system obtains a second correspondence between a failure mode occurring in the semiconductor device and the manufacturing process that caused the failure mode, the second correspondence further describing a probability that the manufacturing process caused the failure mode, and the computer system estimates the manufacturing process that caused the failure mode in the semiconductor device and the probability thereof according to the second correspondence.

9. The inspection system according to claim 4, characterized in that, when the computer system estimates a plurality of failure modes for the same inspection position on the semiconductor device, the computer system estimates the inspection conditions for each of the estimated failure modes, the inspection equipment performs inspection using each of the inspection conditions, and the computer system re-estimates the failure mode occurring at the same inspection position based on the inspection results for each of the inspection conditions.

10. The inspection system according to claim 1, further comprising an inspection device that inspects the semiconductor device by irradiating the semiconductor device with the charged particle beam, the computer system acquiring inspection results from the inspection device, and the computer system grading the degree of defect for each inspection position on the semiconductor device according to the degree of defect represented by the inspection results.

11. The inspection system according to claim 10, wherein the computer system grades the degree of failure for each inspection position on the semiconductor device based on the severity of the failure type represented by the failure mode and the degree of failure represented by the inspection result.

12. The inspection system according to claim 1, wherein said computer system notifies an analysis device that analyzes the cause of said failure mode of the location on said semiconductor device where said failure mode has occurred and said failure mode.

13. The inspection system according to claim 12, characterized in that the computer system identifies an analysis device suitable for analyzing the cause of the failure mode based on the defect type represented by the failure mode and the defect position and defect depth represented by the measurement value, and the computer system notifies the identified analysis device of the position on the semiconductor device where the failure mode has occurred and the failure mode.

14. The inspection system according to claim 1, wherein the electrical characteristics are physical quantities used when designing the semiconductor device, and the electrical characteristics include at least any of the following: a resistance value of the semiconductor device, a capacitance value of the semiconductor device, a frequency characteristic of the semiconductor device, a response speed of the semiconductor device, an amount of leakage current of the semiconductor device, a current-voltage characteristic of a switching element provided in the semiconductor device, and a threshold voltage of a switching element provided in the semiconductor device.

15. An inspection method for inspecting a semiconductor device, comprising the steps of: inspecting the semiconductor device based on measurement values ​​describing results of measuring electrical characteristics of the semiconductor device using a charged particle beam; in the inspecting step, receiving design values ​​and the measurement values ​​of the semiconductor device; in the inspecting step, estimating a failure mode representing a failure type of the electrical characteristics of the measurement values ​​by comparing the design values ​​with the measurement values; and in the inspecting step, extracting inspection points on the semiconductor device to be inspected based on the degree of failure represented by the measurement values ​​and the failure type represented by the failure mode, and outputting the results.

Images