MEASURING DEVICE AND SIGNAL PROCESSING METHOD
The described method for carrier beam devices addresses signal interference issues by using pre-measured dark pulse distributions to set stable threshold and amplitude discrimination values, enhancing measurement accuracy and image quality in scanning electron microscopes.
Patent Information
- Application Number
- DE112018008108
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2018-12-18
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2038-12-18
AI Technical Summary
Existing carrier beam devices, such as scanning electron microscopes, face challenges in achieving high-precision measurements of three-dimensional semiconductor structures due to signal interference from dark pulses and fluctuations in SiPM elements, leading to degraded signal-to-noise ratios and inaccurate image representation.
A measuring device and signal processing method that utilize a photoelectric conversion element and a signal processing unit to distinguish the amplitude of detected pulses based on pre-measured dark pulse amplitude distributions, setting threshold values and amplitude discrimination to improve signal processing accuracy.
This approach enables highly accurate measurements by stabilizing threshold and amplitude discrimination values, reducing dark pulse interference, and enhancing signal-to-noise ratios, resulting in improved image quality.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
Technical area
[0001] The present invention relates to a measuring device, and more particularly, relates to a signal processing technique of a charged particle beam device (which is a type of measuring device) that radiates a charged particle beam and detects generated second charged particles, etc., to obtain an image, thereby performing measurement. Technical background
[0002] A charged particle beam device such as a scanning electron microscope or a mass analyzer is used as a device for evaluating and measuring semiconductor devices. In recent years, semiconductor device structures have been miniaturized and processed in three dimensions, and the evaluation values required by semiconductor device manufacturers are diversifying. In particular, with the three-dimensional processing of device structures, there is a need to measure the bottom surface dimension of a hole or groove shape on a semiconductor substrate with high accuracy to improve yield.
[0003] In the above-mentioned measurement using a scanning electron microscope, when a sample is irradiated with a primary electron beam, signal electrons with different energies are emitted in different directions due to the interaction between the electrons and the sample. The signal electrons, as second charge carriers, contain different information about the sample depending on their emission energies and emission angles, and discriminatory detection of the signal electrons is essential for various measurements.
[0004] Generally, signal electrons are classified as secondary electrons emitted with an energy of 50 eV or less, and backscattered electrons emitted with an energy greater than 50 eV and close to the energy of the primary electron beam. Secondary electrons are sensitive to the surface shape or electric potential of the sample and are effective for measuring a dimension of a surface structure, such as the pattern width of a semiconductor device structure. For three-dimensional structures such as a hole and a groove, secondary electrons cannot escape from the hole or groove because they are absorbed by the sidewalls, and detection and measurement cannot be performed.On the other hand, as disclosed in Patent Literature 1, reflected electrons contain information about a composition and a three-dimensional shape of a sample, and it is possible to obtain information about a three-dimensional structure, composition differences between a surface and a bottom surface, and the like. Since the reflected electrons have high energy, the reflected electrons can escape from a hole and a groove via sidewalls, and signals from the bottom of the hole and groove structure can be detected and measured. Patent Literature 1 discloses an example in which a semiconductor element such as a silicon photomultiplier (hereinafter referred to as SiPM) is used to detect the reflected electrons.
[0005] Patent Literature 2 discloses the use of solid state devices (SSD) in a matrix arrangement for processing the secondary electrons.
[0006] Patent Literature 3 describes, among other things, a device comprising: (a) a first material layer disposed on a first surface of a support structure and configured to generate secondary electrons in response to incident charged particles impinging on the first layer, the first layer including an opening configured to allow a portion of the incident charged particles to pass through the opening; and (b) a second material layer disposed on a second surface of the support structure and separated from the first layer by a distance of 0.5 cm or more.
[0007] Patent Literature 4 describes a spectroscopic device comprising a sample holder for holding a sample, a source for generating an input beam, an illuminator for directing the beam to irradiate the sample, a detector array for detecting an X-ray flux emanating from the sample in response to the irradiation, and a computer processor for executing an automated process in the device. The invention particularly relates to a situation where the process is performed in a charged particle microscope.
[0008] Patent Literature 5 describes a method for examining a sample with a charged particle microscope. Patent Literature 6 describes embodiments related to an apparatus and method for improving the gain calibration of photosensors in a PET system. Citation listPatent literature Patent Literature 1: JP2013 / 541799 A Patent literature 2: US 2017 / 0 309 445 A1 Patent literature 3: US 2012 / 0 068 068 A1 Patent literature 4: US 2016 / 0 189 922 A1 Patent literature 5: US 2012 / 0 292 503 A1 Patent literature 6: US 2014 / 0 110 573 A1 Summary of the inventionTechnical problem
[0009] As described above, in a charged particle beam device, a semiconductor element such as a photodiode or SiPM, which can complement reflected electrons near a sample, and a photoelectric conversion element such as a photomultiplier are suitable for high-precision shape measurement. In particular, the SiPM is characterized in that a detected signal amplitude changes linearly in proportion to the number of input photons, and photon counting is possible through amplitude discrimination. The SiPM outputs a current corresponding to the number of photons of incident light by combining multiple avalanche photodiodes (APDs) of a basic element in an array and applying a bias voltage to enter a Geiger mode.
[0010] When photons are radiated to multiple pixels in such a SiPM at almost the same time, the output pulses are superimposed according to the number of photons, and the amplitude is large. In addition, the SiPM generates a dark pulse due to a thermally generated dark current carrier, which degrades the S / N even when no incident photon is present. Most of the amplitude values correspond to the amplitude of the current generated when a photon is input to the APDs. Therefore, to improve the S / N of a signal output from the SiPM and obtain a good SEM image, it is necessary to process the output pulses taking these SiPM characteristics into account. In addition, since a large number of SiPMs are arranged in the device to improve the SN, it is necessary to consider the variation between SiPM elements and the DC offset of a detection circuit.
[0011] An object of the invention is to provide a charged particle beam device and a signal processing method that solve the above-mentioned problems and enable highly accurate measurement. Further objects will become clear in the description of the embodiments. Solution to the problem
[0012] The above-mentioned object is achieved by the independent claims. These claim a measuring device including, among other things, a photoelectric conversion element and a signal processing unit configured to receive a detected pulse from the photoelectric conversion element, which includes a dark pulse and a signal pulse output in response to an input photon. The signal processing unit distinguishes the amplitude of the detected pulse based on the amplitude distribution of the dark pulse of the photoelectric conversion element, which was acquired in advance. An example of the measuring device is a charged particle beam device.
[0013] Additionally, the claims claim a signal processing method comprising, among other things: detecting in advance an amplitude distribution of a dark pulse of a photoelectric conversion element installed in a device; receiving from the photoelectric conversion element a detected pulse including a dark pulse and a signal pulse output in response to an input photon; and discriminating the amplitude of the received detected pulse based on the amplitude distribution of the dark pulse. Beneficial effect
[0014] According to the invention, a highly accurate measurement is possible by setting a threshold value and an amplitude discrimination value using the amplitude distribution of a dark pulse. Brief description of the drawings [ Fig. 1] Fig. 1 is a schematic view of a scanning electron microscope according to a first embodiment. [ Fig. 2] Fig. 2 is a diagram showing an example of a dark pulse histogram according to the first embodiment. [ Fig. 3A] Fig. 3A is a diagram showing an operation flow chart of the scanning electron microscope according to the first embodiment. [ Fig. 3B] Fig. 3B is a diagram showing an example of a photon distribution during measurement by the scanning electron microscope according to the first embodiment. [ Fig. 4] Fig. 4 is a schematic view of a scanning electron microscope according to a second embodiment. [ Fig. 5A] Fig. 5A is a diagram showing an operation flow chart of the scanning electron microscope according to the second embodiment. [ Fig. 5B] Fig. 5B is a diagram showing an example of an amplitude distribution of a dark pulse at a certain time according to the second embodiment. [ Fig. 6] Fig. 6 is a schematic view of a scanning electron microscope according to a third embodiment. [ Fig. 7] Fig. 7 is a diagram showing an example of a table of a threshold value and an amplitude discrimination value according to the third embodiment. [ Fig. 8] Fig. 8 is a diagram showing an operation flow chart of the scanning electron microscope according to the third embodiment. Description of the embodiments
[0015] The following embodiments describe a case where the invention is specifically applied to a scanning electron microscope which is a charged particle beam device, and the invention is not limited to the scanning electron microscope and can also be applied to other charged particle beam devices. First embodiment
[0016] The first embodiment is an embodiment of a scanning electron microscope including a photoelectric conversion element and a signal processing unit configured to receive, from the photoelectric conversion element, a detected pulse including a dark pulse and a signal pulse output in response to an input photon. The signal processing unit distinguishes the amplitude of the detected pulse based on the amplitude distribution of the dark pulse of the photoelectric conversion element, which has been acquired in advance.
[0017] The first embodiment is an embodiment of a signal processing method including detecting in advance an amplitude distribution of a dark pulse of a photoelectric conversion element installed in a scanning electron microscope, receiving from the photoelectric conversion element a detected pulse including the dark pulse and a signal pulse output in response to an input photon, and discriminating the amplitude of the received detected pulse based on the amplitude distribution of the dark pulse.
[0018] The configuration and operation of the first embodiment will be described with reference to Fig. 1, Fig. 2, Fig. 3A and Fig. 3B described. Fig. 1 shows a configuration example of the scanning electron microscope of the first embodiment. An electron gun 101 is arranged in an electron microscope column 100 located in a vacuum environment. A primary electron beam 102 emitted from the electron gun 101 travels along an optical axis of a primary electron beam. A trajectory of the primary electron beam 102 is adjusted by a deflector 105 and is converged onto a sample 109 by an objective lens 107. In addition, the material 109 is placed on a stage provided in the electron microscope. A negative voltage is applied to the sample 109, and the primary electron 102 impacts the sample 109 with an energy lower than the energy generated by the electron gun 101.Reflected electrons 108 and secondary electrons 103 generated from the sample due to the collision of the primary electron 102 fly in the electron microscope column 100 according to respective emission energies and emission angles.
[0019] When the reflected electrons 108 collide with a scintillator 136, the reflected electrons 108 are converted into light. The light is converted into a detection current by a semiconductor element such as a photodiode or SiPM, or a photoelectric conversion element 106a such as a photomultiplier, and becomes a detected signal 111. The detected signal 111 is hermetically transmitted from an in-vacuum wiring 110 to a signal processing unit 115 via a signal preprocessing unit 114. As described later, the signal preprocessing unit 114 generates an amplitude distribution 117 and a bias voltage 112 of the photoelectric conversion element 106a based on a dark pulse in a certain period of time. The dark pulse is a pulse generated by multiplying a thermally generated dark current carrier.
[0020] The signal processing unit 115 includes a dark pulse removal unit 119, an image generation unit 121, a photon counting unit 134, and a control unit 123. The detected signal 111 is converted into a digital signal by an analog-to-digital converter (ADC) in the signal preprocessing unit 114 during a measurement and is input to the signal processing unit 115. The signal processing unit 115 can be implemented by executing a program of a central processing unit (CPU).
[0021] The photon counting unit 134 of the signal processing unit 115 distinguishes the signal amplitude of the digitized detected signal 111 based on an amplitude discrimination value 140 from the control unit 123 and counts the number of photons. In addition, the dark pulse removal unit 119 removes a dark pulse from the detected signal 111 based on a threshold value 141 from the control unit 123. The image generation unit 121 generates an image based on the detected signal 111. The image from the image generation unit 121 is adjusted by a luminance contrast adjustment unit 120 and then displayed as a measurement observation test image on a user interface 132 of a display unit of a computer 131.In this specification, the signal preprocessing unit 114, which detects the amplitude distribution of the dark pulse in advance, and the signal processing unit 115 may be collectively referred to as a signal processing unit. An analog detection unit 122 will be described later.
[0022] A procedure for measuring reflected electrons in the scanning electron microscope, which is Fig. 1 is shown with reference to the dark pulse histogram of Fig. 2, the operating schedule of Fig. 3A and the photon distribution example during a measurement of Fig. 3B described.
[0023] If a company in the course of Fig. 3A (S1), the primary electron beam 102 is blocked to block the photons (S2), and a standard bias voltage (Vs) 112 is applied to the photoelectric conversion element 106a (S3). The dark pulse is output by applying the bias voltage from the photoelectric conversion element 106a. Based on the output dark pulse, the amplitude distribution of the dark pulse at a specific time is determined as shown in Fig. 2, ie, the dark pulse histogram, is generated (S4). A horizontal axis of the dark pulse histogram shows a dark pulse level (mV), and a vertical axis shows a frequency (times).
[0024] As in Fig. As shown in Figure 2, by observing the amplitude distribution of the dark pulse at a certain time, at least a peak value P1 of a pulse amplitude corresponding to a one-photon input, a peak value P2 of a pulse amplitude corresponding to a two-photon input, and a peak value K1 of circuit noise generated in the signal pre-processing unit 114 are observed in the amplitude distribution. The control unit 123 of the signal processing unit 115 receives the amplitude distribution 117 from the signal pre-processing unit 114 and sets the threshold value (T1) 141 for a dark pulse and a plurality of amplitude discrimination values (Dn) 140 (S5). In the present embodiment, the threshold value 141 is set by T1 = (A1 + A2) / 2 from the peak value A1 of P1 and the peak value A2 of P2.In addition, the amplitude discrimination value 140 is calculated by the amplitude discrimination values Dn = {A1 × (3 - 2n) + A2 × (2n - 1)} / 2 (n ≥ 2) from the peak values of n photons and the peak values of (n + 1) photons. In addition, the control unit 123 generates a control signal 142 for controlling the bias voltage 112 of the photoelectric conversion element 106a and outputs the control signal 142 to the signal preprocessing unit 114.
[0025] Then, a primary beam is emitted by the scanning electron microscope during an actual measurement, and a photon containing information about an object to be measured, such as a semiconductor wafer, is input to the photoelectric conversion element 106a (S6). An example of the photon distribution during a measurement is shown in Fig. 3B shown.
[0026] In the photon distribution, which in Fig. As shown in Fig. 3B, the dark pulse removal unit 119 removes pulses equal to or smaller than the dark pulse based on the previously calculated threshold value T1 (S7). In addition, the photon counting unit 134 discriminates the amplitude of the detected pulse and performs photon counting of how many photons the detected pulse corresponds to based on the previously calculated amplitude discrimination values D1 to Dn (S8). Further, the image generation unit 121 generates an image based on a photon count and adjusts the luminance and contrast of the generated image through program processing (S9). This adjustment can be performed by digital signal processing such that a luminance contrast distribution of the obtained image matches a reference luminance contrast distribution that was previously generated and stored.The adjusted image is then displayed in the user interface 132 on a screen of the computer 131 (S10). The computer 131 can execute functions of the luminance contrast adjustment unit 120 and also of the signal processing unit 115.
[0027] As described above, an operation main body of the operation flow described with reference to Fig. 3A, the signal preprocessing unit 114 includes an amplitude distribution generation unit, and the signal processing unit 115 includes the control unit 123. When this operation is performed by digital signal processing through a program, the program may be stored, for example, in a storage medium in the control unit 123, a storage unit in the computer 131, or a storage device of a server of an external network to which the computer 131 is connected.
[0028] Additionally, the luminance contrast adjustment of the image can be performed using the output of the analog detection unit 122, to which an analog detected signal 111 is input, in addition to the digital luminance contrast adjustment described above. In this case, the analog detection unit 122 performs analog detection at the standard bias voltage Vs (U1) after the sequence S6, and the luminance contrast distribution is generated from a detection result. The luminance contrast distribution is compared with a reference luminance contrast distribution such that the luminance and contrast are adjusted (U2).That is, the luminance contrast adjustment is performed by returning to the sequence S3 to make the luminance contrast distribution agree with the reference luminance contrast distribution, and adjusting the bias voltage 112 from Vs to Vt, and then the sequences S6, S7, S8, and S10 described above are performed.
[0029] In addition, the invention is not limited to the operation procedure described above and can be carried out as follows: S6 → U1 → U2 → S2 → S3 → S4 → S5 → S6 → S7 → S8 → S10, after the start (S1) of the process in Fig. 3A.
[0030] According to the first embodiment described above, the amplitude distribution of the dark pulse is measured in a state where no photon is input, and the threshold value and the amplitude discrimination value are adjusted so that a stable threshold value and a stable amplitude discrimination value are obtained regardless of the object to be measured. The SN of the formed image can be improved by removing the dark pulse. In addition, the SN of the formed image can be improved by performing photon counting on the pulse output amplitude output from the photoelectric conversion element based on the amplitude discrimination value. Second embodiment
[0031] The configuration and operation of a scanning electron microscope provided with a plurality of photoelectric conversion elements of the second embodiment will be described with reference to Fig. 4, Fig. 5A and Fig. 5B. Fig. Figure 4 shows a general schematic configuration of the scanning electron microscope of the present embodiment. The same components as in Fig. 1 for describing the first embodiment will be omitted.
[0032] The present embodiment is an embodiment in which a plurality of photoelectric conversion elements 106, ie, four photoelectric conversion elements, are used (note that only two photoelectric conversion elements 106a / b are shown in an illustrated context). Furthermore, in the present embodiment, four signal preprocessing units 114a / b and four bias voltage generation units 113a / b are provided. For the illustrated context, Fig. 4, two systems are used. However, the invention is not limited to two systems, and multiple systems corresponding to the number of photoelectric conversion elements 106 may be installed. The output corresponding to detected signals 111a / b is input from signal preprocessing units 114a / b to a signal preprocessing unit 114c. The signal preprocessing unit 114c includes an ADC 116 and an amplitude distribution generation unit 135.
[0033] The amplitude distribution generation unit 135 of the signal preprocessing unit 114c is a block that generates the amplitude distribution 117 of the dark pulse at a specific time in advance and outputs the generated amplitude distribution 117 to the control unit 123 of the signal processing unit 115. The control unit 123 compares the amplitude distribution 117 with a stored reference distribution and outputs the control signal 142 for controlling a bias voltage 112a / b generated by a SiPM bias voltage generation unit 113a / b.
[0034] The detected signals 111a and 111b, converted into digital signals by the ADC 116, are input to the signal processing unit 115. The photon counting unit 134, the dark pulse removal unit 119, an image generation block 121, and the control unit 123 of the signal processing unit 115 also have the function of processing all the detected signals 111a / b of the plurality of photoelectric conversion elements 106a / b. In the configuration of the present embodiment, the amplitude distribution generation unit 135 and the signal processing unit 115, installed downstream of the ADC 116, can be implemented by executing a program of a CPU, as in the first embodiment.
[0035] The operation of the scanning electron microscope, which is equipped with several ch photoelectric conversion elements, which are Fig. 4, is carried out using the flow chart of Fig. 5A. The primary electron beam 102 is blocked (S2), and one of the plurality of photoelectric conversion elements 106 is selected (S3). Here, four systems (four channels, hereinafter referred to as ch) of photoelectric conversion elements 106 are provided as an example, but the invention is not limited thereto.
[0036] First, ch 1 is selected, and the standard bias voltage (Vs) 112 is applied to the selected ch of the plurality of photoelectric conversion elements 106 (S4). The dark pulse is output from the selected ch of the plurality of photoelectric conversion elements 106 by applying a bias voltage, and the amplitude distribution of the dark pulse at a specific time is measured as shown in Fig. 2 is generated (S5). By observing the amplitude distribution at a specific time, at least the peak value P1 of the pulse amplitude corresponding to a one-photon input, the peak value P2 of the pulse amplitude corresponding to a two-photon input, and the peak value K1 of the circuit noise generated in the signal preprocessing unit 114 are observed in the amplitude distribution 117. This amplitude distribution is compared with peak positions S1 and S2 of one photon and two photons of the reference amplitude distribution shown in Fig. 5B, and the bias voltage 112 is adjusted such that an amplitude difference (A2 - A1 = S2 - S1) between the peak positions is equal (S4).
[0037] Here, one reason why the bias voltage 112 is not adjusted such that A1 = S1 or A2 = S2 is that an error due to the DC offset occurs in the threshold value T1 and the amplitude discrimination value Dn when the bias voltage 112 is adjusted such that A1 = S1 or A2 = S2. This is because the signal preprocessing unit 114 that processes the pulse output from each ch has the DC offset due to element fluctuations and temperature characteristics. In addition, the bias voltage 112 may be adjusted such that A1 = S1 or A2 = S2 if a function is provided to remove the DC offset generated by the signal preprocessing unit 114 or the like in advance.If the dark pulse distribution matches the reference distribution, the threshold value T1 and the amplitude discrimination value Dn are set by the threshold value 141 from the peak value A1 of P1 and the peak value A2 of P2 by T1 = (A1 + A2) / 2. Additionally, the amplitude discrimination value 140 can be calculated by the amplitude discrimination value Dn = {A1 × (3 - 2n) + A2 × (2n - 1)} / 2 (n ≥ 2) from the peak values of n photons and the peak values of (n + 1) photons (S7). The above sequence from S3 to S7 is performed for each of the four ch, and the threshold value and amplitude discrimination value of each of the four ch are obtained, respectively (S8).
[0038] Then, the primary beam 102 is radiated through the scanning electron microscope for actual measurement, and a photon containing information about the sample 109, which is an object to be measured such as a semiconductor wafer, is input into each ch of the plurality of photoelectric conversion elements 106 (S9). An example of a photon distribution of each ch during measurement is shown in Fig. 3B. In this distribution, pulses equal to or smaller than the dark pulse are removed based on the threshold value T1 shown above (S10), the amplitude of the detected pulse is discriminated based on the amplitude discrimination value Dn, and photon counting is performed based on how many photons the detected pulse corresponds to (S11). After that, photon counting results of each ch are combined (S12), and the image generation unit 121 generates an image, adjusts the luminance and contrast (S13), and displays the image (S14). In addition, also in the present embodiment, luminance contrast adjustment can be performed using the analog detection unit 122 provided in each ch, as in the first embodiment.
[0039] According to the present embodiment, since the threshold value T1 and the amplitude discrimination value Dn of each ch of the plurality of photoelectric conversion elements 106 are obtained using the reference amplitude distribution, there is an effect that the fluctuation of each ch of the plurality of photoelectric conversion elements 106, the fluctuation of the signal preprocessing unit 114, and the DC offset are not affected. In addition, by adjusting the threshold value and the amplitude discrimination value using the dark pulse, it is possible to obtain a stable threshold value and a stable amplitude discrimination value regardless of the object to be measured. The SN of the formed image can be improved by removing the dark pulse.Additionally, the SN of the generated image can be improved by performing photon counting on the pulse output amplitude output from the plurality of photoelectric conversion elements based on the amplitude discrimination value. Furthermore, by adjusting the bias voltage of each element so that the dark pulse distribution of each of the plurality of arranged SiPM elements matches the reference value, it is possible to reduce the influence of inter-element variation and the DC offset of a detected signal processing circuit. Third embodiment
[0040] The configuration and operation of a scanning electron microscope using a LUT of the third embodiment will be described with reference to Fig. 6 to Fig. 8 described. Fig. Fig. 6 is a diagram showing a schematic configuration of the scanning electron microscope of the present embodiment. The same configuration units as in Fig. 1 for describing the first embodiment will be omitted.
[0041] The signal preprocessing unit 114 according to the present embodiment provides the bias voltage 112 from the SiPM bias voltage generation unit as shown in Fig. 4, to a plurality of voltages, generates the amplitude distribution 117 of the dark pulse at each bias voltage 112, and creates an LUT 118 of the threshold value 141 and the amplitude discrimination value 140 based on the amplitude distribution 117. In the present embodiment, the amplitude distribution generation unit 135, the LUT 118, and the signal processing unit 115 can be implemented by executing a program of a CPU, as in the first and second embodiments. Fig. 7 shows a configuration example of LUT 118.
[0042] First, the operation procedure of the scanning electron microscope of the third embodiment will be described with reference to the flowchart of Fig. 8. First, the primary electron beam 102 is blocked (S2), and the standard bias voltage (Vs) 112 is applied to the plurality of photoelectric conversion elements 106 (S3). The dark pulse is output from the plurality of photoelectric conversion elements 106 by applying the bias voltage, and an amplitude distribution of the dark pulse at a specific time, as shown in Fig. 2 is generated (S4). By observing the amplitude distribution at a specific time, at least the peak value P1 of the pulse amplitude corresponding to a one-photon input, the peak value P2 of the pulse amplitude corresponding to a two-photon input, and the peak value K1 of the circuit noise generated in the signal preprocessing unit 114 are observed in the amplitude distribution. From these amplitude distributions, the threshold value (T1) 141 of the dark pulse and the amplitude discrimination value (Dn) 140 are set (S5). Here, the threshold value 141 is set by T1 = (A1 + A2) / 2 from the peak value A1 of P1 and the peak value A2 of P2. In addition, the amplitude discrimination value 140 can be calculated by the amplitude discrimination value Dn = {A1 × (3 - 2n) + A2 × (2n - 1)} / 2 (n ≥ 2) from peak values of n photons and peak values of (n + 1) photons.
[0043] Then, the bias voltage 112 of the SiPM is changed in a range from VI to Vs to Vh (S6). After the change, the process returns to sequence S4 to generate the amplitude distribution 117. The threshold value and the amplitude discrimination value are calculated in S5, and the bias voltage 112 is further changed in S6. Afterward, the LUT 118 of the threshold value and the amplitude discrimination value for the bias voltage 112 is generated in sequence S8.
[0044] Fig. 7 shows a configuration example of the LUT 118. As shown in the same figure, when the bias voltage 112 is changed a predetermined number of times (S7), a one-photon peak level A1 (V) and a two-photon peak level A2 (V) corresponding to a plurality of control voltages (the bias voltages 55 to 65) are stored, and the threshold value and the amplitude discrimination value corresponding to the respective bias voltages can be obtained (S8).
[0045] Then, during the actual measurement, the primary beam 102 is radiated by the scanning electron microscope in the present embodiment, photons containing information about the sample 109, which is an object to be measured, such as a semiconductor wafer, are input to the plurality of photoelectric conversion elements 106 (S9), and the bias voltages of the photoelectric conversion elements 106 are set to the standard voltage Vs (S10). An example of a photon distribution during a measurement, as shown in Fig.3 is obtained. In this distribution, pulses equal to or smaller than the dark pulse are removed based on the threshold value T1 determined according to the stored contents of the LUT 118 (S11), the amplitude of the detected pulse is discriminated based on the amplitude discrimination value Dn, and photon counting of how many photons the detected pulse corresponds to is performed for each ch (S12).
[0046] Next, an image is generated, and the luminance and contrast are adjusted (S13). In this adjustment, it is determined whether the luminance contrast distribution of the generated image matches the reference luminance contrast (S14). If the luminance contrast distribution of the generated image does not match the reference luminance contrast, the bias voltage 112 is adjusted by the control signal from the control unit 123 (S15), the threshold value and the amplitude discrimination value corresponding to the adjusted bias voltage 112 are derived from the LUT 118 (S16), and the process is sequentially repeated from the sequence S11. For bias voltages not included in the LUT 118, the threshold value and the amplitude discrimination value can be derived by linear interpolation or the like. If the luminance histogram distributions match (Yes) in S14, a wafer measurement is started (S17).
[0047] According to the present embodiment, by preparing the table of the threshold value and the amplitude discrimination value in advance using the dark pulse, the threshold value can be easily read when the bias voltage value is changed. Using the table, for example, the luminance histogram can be easily adjusted. In addition, if the table of the threshold value and the amplitude discrimination value is prepared for different temperatures of the plurality of photoelectric conversion elements 106, stable measurement can be performed by resetting the threshold value and the amplitude discrimination value using this table when the plurality of photoelectric conversion elements 160 or the signal preprocessing unit 114 change depending on the temperature.
[0048] The invention is not limited to the embodiments described above and includes various modified embodiments. For example, the above-mentioned embodiments have been described in detail for the purpose of facilitating a better understanding of the invention and are not necessarily limited to those having all the configurations of the description. In the first to third embodiments, examples in which the invention is applied to a charged particle beam device represented by a scanning electron microscope are described, but the invention is not limited to the examples. For example, the invention can be applied to a plurality of photoelectric conversion elements used in an inspection device that radiates or uses an X-ray beam (an electromagnetic wave in the broad sense), such as a CT device or a PET device, which is an example of a measuring device.Of course, these measuring devices include part or all of the above-described embodiments, such as an irradiation unit (corresponding to the electron gun 101 of the above-mentioned embodiments) that radiates the X-ray beam (an electromagnetic wave in the broad sense), and a stage for arranging a measurement target or a person to be measured (corresponding to the sample 109 of the above-mentioned embodiments).
[0049] In addition, the signal preprocessing unit 114 may include a CPU in addition to the ADC described above, and may execute a program to implement the "unit" described above. In addition, the analog detection unit 122 and the luminance contrast adjustment unit 120 may be implemented by a computer having a CPU. In addition, the signal preprocessing unit 114, the signal processing unit 115, the analog detection unit 112, the luminance contrast adjustment unit 120, and the computer 131 (which may be collectively referred to as a processing subsystem hereinafter. However, as described above, the processing subsystem does not necessarily include all of them) may be mounted on the same substrate or mounted separately on multiple substrates. In addition, it is understood that each component having a CPU may include a memory.
[0050] Furthermore, although an example of creating a program that implements part or all of the configurations, functions, and computing described above is mainly described, it is natural that part or all of them can be implemented by hardware, for example, by designing an integrated circuit. In other words, all or part of the functions of the signal processing unit can be implemented by an integrated circuit such as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA) instead of the program. List of reference symbols 100 electron microscope column 101 Electron Gun 102 primary electrons 105 Deflector 106, 106a, 106b Photoelectric conversion element 107 Objective lens 108 Reflected electron 109 Sample 110 In-vacuum wiring 111 Detected signal 112 Preload 114, 114a, 114b, 114c Signal preprocessing unit 115 Signal processing unit 116 analog-to-digital converters (ADC) 117 Amplitude discrimination 119 Dark Pulse Removal Unit 121 Image generation block 120 Luminance contrast adjustment unit 131 computers 123 Control unit 132 User interface 133 Measurement observation test image 134 photon counting unit 135 Amplitude distribution generation unit 118 Table (LUT) 140 amplitude discrimination value 141 Threshold 142 Control signal
Claims
[1] Measuring device comprising: a photoelectric conversion element (106, 106a, 106b), a signal processing unit (115) configured to receive from the photoelectric conversion element (106, 106a, 106b) a detected pulse (111) containing a dark pulse and a signal pulse output in response to an input photon, the measuring device characterized by , that the signal processing unit (115) is configured to detect an amplitude distribution of the dark pulse in a state in which no light is incident on the photoelectric conversion element (106, 106a, 106b); using the amplitude distribution of the dark pulse to provide an amplitude A1, corresponding to a 1-photon input, and an amplitude A2, corresponding to a 2-photon input, and to distinguish the amplitude of the detected pulse based on the amplitude A1, corresponding to a 1-photon input, and the amplitude A2, corresponding to a 2-photon input. [2] The measuring device according to claim 1, wherein the signal processing unit (115) is configured to set a threshold configured to remove the dark pulse included in the detected pulse based on the amplitude distribution of the dark pulse. [3] The measuring device according to claim 2, wherein the signal processing unit (115) is configured to set an amplitude discrimination value of the detected pulse based on the amplitude distribution of the dark pulse. [4] The measuring device according to claim 3, wherein the signal processing unit (115) is configured to schematically set the threshold value T1 as (A1 + A2) / 2 and to set the amplitude discrimination value Dn between n photons and n + 1 photons based on an amplitude A1 corresponding to one photon of the dark pulse and an amplitude A2 corresponding to two photons in the amplitude distribution of the dark pulse substantially to Dn = {A1 × (3 - 2n) + A2 × (2n - 1)} / 2 (n ≥ 2). [5] The measuring device according to claim 4, wherein the signal processing unit (115) is configured to set a plurality of bias voltages (112) configured to drive the photoelectric conversion element, set the threshold value T1 and the amplitude discrimination value Dn for each bias voltage (112), and store the threshold value T1 and the amplitude discrimination value Dn in a look-up table (LUT). [6] A measuring device according to claim 1, wherein a silicon photomultiplier (SiPM) is used as the photoelectric conversion element (106, 106a, 106b). [7] Measuring device according to claim 1, wherein the photoelectric conversion element contains a plurality of photoelectric conversion elements (106, 106a, 106b), and the signal processing unit (115) includes a plurality of bias voltage generating units each configured to drive the plurality of photoelectric conversion elements (106, 106a, 106b). [8] The measuring device according to claim 7, wherein the signal processing unit (115) is configured to detect the amplitude distribution of the dark pulse of each of the plurality of photoelectric conversion elements (106, 106a, 106b) in advance, adjust bias voltages (112) generated by the bias voltage generation units so that the reference distribution and the detected amplitude distribution of the dark pulse coincide with each other, distinguish the amplitude of the detected pulse based on the amplitude distribution of the dark pulse after the adjustment, and remove the dark pulse included in the detected pulse. [9] The measuring device according to claim 8, wherein the signal processing unit (115) is configured to remove the dark pulse corresponding to each of the plurality of photoelectric conversion elements (106, 106a, 106b), count the detected amplitude-discriminated pulse, and combine all the counting results to generate an image. [10] Signal processing method comprising: detecting in advance an amplitude distribution of a dark pulse of a photoelectric conversion element (106, 106a, 106b) installed in a device; Receiving a detected pulse from the photoelectric conversion element (106, 106a, 106b) including the dark pulse and a signal pulse output in response to an input photon; detecting the amplitude distribution of the dark pulse in a state in which no light is incident on the photoelectric conversion element (106, 106a, 106b); Providing an amplitude A1, corresponding to a 1-photon input, and an amplitude A2, corresponding to a 2-photon input, from the amplitude distribution of the dark pulse, and Distinguishing the amplitude of the received detected pulse based on the amplitude A1, corresponding to a 1-photon input, and the amplitude A2, corresponding to a 2-photon input. [11] A signal processing method according to claim 10, further comprising: Setting a threshold configured to remove the dark pulse contained in the detected pulse based on the amplitude distribution of the dark pulse. [12] A signal processing method according to claim 11, further comprising: Setting an amplitude discrimination value of the detected pulse based on the amplitude distribution of the dark pulse. [13] A signal processing method according to claim 12, further comprising: Setting the threshold value T1 schematically as (A1 + A2) / 2 and setting the amplitude discrimination value Dn between n photons and n + 1 photons essentially to Dn = {A1 × (3 - 2n) + A2 × (2n - 1)} / 2 (n ≥ 2) based on an amplitude A1 corresponding to one photon of the dark pulse and an amplitude A2 corresponding to two photons in the amplitude distribution of the dark pulse.
Citation Information
Patent Citations
Electron detectors including a tightly coupled scintillator-photomultiplier tube combination, electron microscopes and X-ray detectors using the same.
JP2013541799A
Charged particle detectors
US20120068068A1
Charged-particle microscopy with occlusion detection
US20120292503A1
Method and apparatus for photosensor gain and scintillation crystal optical coupling monitoring in radiation detectors
US20140110573A1
Charged particle microscope with improved spectroscopic functionality
US20160189922A1
