JP2014116170A - Charged particle beam device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a charged particle beam device in which a desired position can be irradiated accurately with a charged particle beam, even if the pressure state of a preliminary exhaust chamber changes.SOLUTION: A charged particle beam device includes a charged particle beam optical system for irradiating a sample with a charged particle beam, a vacuum chamber for maintaining the atmosphere around the sample, irradiated with the charged particle beam, in vacuum state, a preliminary exhaust chamber for evacuating the atmosphere around the sample after being introduced into the vacuum chamber, and a controller for controlling the charged particle beam optical system. A deflector for deflecting the irradiation position of the charged particle beam is also provided. During preliminary exhaust operation and/or leakage of the preliminary exhaust chamber, the controller controls the deflector so as to deflect the charged particle beam based on predetermined conditions.

Description

  The present invention relates to a charged particle beam apparatus that irradiates a sample with a charged particle beam, and more particularly to a charged particle beam apparatus including a preliminary exhaust chamber that evacuates a sample atmosphere.

  In the semiconductor market, semiconductor devices and the like have been highly integrated and miniaturized, and there has been a demand for improved performance of charged particle beam apparatuses that perform observation, measurement, and inspection of these samples. For example, a scanning electron microscope (SEM) which is an embodiment of a charged particle beam apparatus is required to have high resolution and measurement reproducibility.

  SEM irradiates a sample with an electron beam to excite atoms on the surface of the sample and generate secondary electrons with low energy that are emitted. When an electron beam is applied to the edge of a sample with irregularities such as a semiconductor circuit pattern, the amount of secondary electrons generated by the edge effect increases, and an image with a contrast depending on the irregularities is formed. However, since the secondary electron signal is very weak, the image obtained when the sample surface of the electron beam is irradiated once contains a lot of noise. Therefore, the same portion is scanned a plurality of times, and the signal image obtained at that time is averaged to remove noise and obtain a clear image.

  At this time, if the electron beam cannot be irradiated at the same location, the image obtained at that time will be blurred due to the shift of the scanning position, resulting in a resolution degradation or a difference in length measurement value.

  In addition, the processing speed of these apparatuses in the production line is also required at the same time. Therefore, a configuration and control are made so that a wafer to be inspected is prepared so that the next wafer can be inspected immediately during measurement and inspection.

  To irradiate the sample with an electron beam in SEM. The body including the sample chamber and the electron optical system is kept in a vacuum. Also, in order to replace the next wafer with a wafer that has been measured and inspected without wasting time, a preliminary exhaust chamber is provided to perform air release, wafer replacement, and vacuum exhaust processing during observation.

  At the time of preliminary exhaust in the preliminary exhaust chamber, since the vacuum pump is operating for preliminary exhaust, Patent Document 1 stops beam scanning for measurement so that the vibration of the vacuum pump does not affect the measurement. The technique to make it disclosed is disclosed. However, this beam scanning stop time is a dead time and becomes one of the causes of a decrease in processing speed.

  Patent Document 2 proposes a solution to the problem that the pre-exhaust chamber is deformed by a pressure change and the beam irradiation position is changed during vacuum exhaust or leak. Specifically, Patent Document 2 discloses a technique for canceling the tilt of the electron microscope body (column) caused by the pressure change in the preliminary exhaust chamber by tilting the sample chamber.

Japanese Patent Laid-Open No. 3-1556849 JP 2002-352760 A

  As disclosed in Patent Document 1, if the beam scanning is stopped while the vacuum pump is operating, a wasteful time will be generated accordingly, and the processing speed is reduced. Further, as disclosed in Patent Document 2, there is a method of canceling the tilt of the mirror body accompanying the deformation of the pressure in the preliminary exhaust chamber by reversely tilting the sample chamber. In measuring accuracy, it is difficult to deal with mechanical systems alone, and higher accuracy is required.

  Furthermore, the position of the center of gravity of the entire apparatus changes depending on the position of the sample stage in the sample chamber, and the tilt amount of the mirror body changes accordingly. The change in the amount of deviation of the irradiation position due to the stage position is about several nanometers, but it cannot be ignored in recent measurements that require high accuracy.

  In the following, a charged particle beam apparatus is proposed that aims to accurately irradiate a desired position with a charged particle beam even if the pressure in the preliminary exhaust chamber changes.

  As one aspect for achieving the above object, a charged particle beam optical system for irradiating a charged particle beam to a sample below, a vacuum chamber for maintaining an atmosphere around the sample irradiated with the charged particle beam in a vacuum, A charged particle beam apparatus comprising a preliminary exhaust chamber for evacuating an atmosphere around a sample to be introduced into the vacuum chamber and a control device for controlling the charged particle beam optical system, wherein the irradiation position of the charged particle beam is deflected And the control device controls the deflector so as to deflect the charged particle beam based on a predetermined condition at the time of preliminary exhaust and / or leakage of the preliminary exhaust chamber. A charged particle beam system is proposed.

  According to the above configuration, it is possible to accurately irradiate a charged particle beam at a desired position even when the pressure in the preliminary exhaust chamber changes.

The figure which shows embodiment of a scanning electron microscope system. The figure which shows the mirror body inclination when the pressure of a preliminary exhaust chamber changes, and the irradiation position shift of the charged particle beam which generate | occur | produces in connection with it. The figure which shows an example of the sample stage provided with the mark for monitoring the change of a beam irradiation position. The flowchart which shows a beam irradiation position correction amount calculation process. The flowchart which shows the process of acquiring an electron microscope image, performing beam irradiation position correction | amendment. The flowchart which shows the process of performing beam irradiation position calibration automatically. The figure which shows the locus | trajectory of the mark which moves within a visual field with the pressure change of the preliminary exhaust chamber at the time of preliminary exhaust.

  The following is an apparatus for measuring or observing a pattern based on a signal obtained by irradiation with a charged particle beam, in order to observe the sample, exhaust to enter and exit the sample chamber, and preliminary exhaust to release the atmosphere A charged particle beam apparatus having a chamber will be described. In particular, a charged particle beam apparatus will be described in which the charged particle irradiation position irradiated on the sample does not deviate from the target position even when the pressure in the preliminary exhaust chamber changes.

  The change in the pressure of the pre-exhaust chamber deforms the wall separating the pre-exhaust chamber and the sample chamber, and as a result, the upper surface of the sample chamber is also deformed. As a result, the electron source and electron installed in the sample chamber In some cases, the mirror including the optical system is tilted and the irradiation position shift of the electron beam occurs. Therefore, in the embodiment described below, a preliminary exhaust chamber is provided so as to accurately irradiate a target position on the sample and detect a good secondary electron image without depending on the position of the sample stage. The correlation between the change in the vacuum state, the position of the sample stage and the change in the position of the electron beam applied to the sample surface is measured in advance, and the electron beam is obtained from the correlation obtained from the measurement results according to the vacuum state and the position of the sample stage The electron beam irradiation position is corrected by a field movement deflector (electromagnetic deflector or electrostatic deflector).

  Instead of correcting the irradiation position of the electron beam as the correction method, when the acquired images are added and averaged, the images may be moved and averaged from the correlation to obtain an image for measurement and inspection.

  In addition, since the relationship between the tilt of the mirror and the degree of vacuum is mechanistic, it may change with time. For this reason, in normal measurement observation, when the vacuum level of the preliminary exhaust chamber is constant, the image displacement in the first scan start frame and the final scan frame at the measurement observation point at the measurement observation point and the preliminary exhaust chamber exhaust or Compare the image misalignment between the first frame and the last frame of the scan in the acquired image at the measurement observation point at the time of the leak. The secondary electron image can always be detected in a good state by re-measuring the correlation between the change in the position of the electron beam irradiated onto the sample surface, recalculating the correlation, and performing correction.

According to this embodiment, even when the pressure in the preliminary exhaust chamber changes, the position of the electron beam irradiated onto the sample does not shift, so that a good secondary electron image can be obtained even when the pressure in the preliminary exhaust chamber changes. The dead time can be shortened and the processing time can be improved. Hereinafter, an apparatus in which a preliminary exhaust chamber is provided in an SEM that is one embodiment of the charged particle beam apparatus will be described with reference to FIG. The main constituent members are the sample chamber 26, the mirror body 1, and the preliminary exhaust chamber 27, which are installed in a predetermined place by being held on the mounting table 28 via the vibration isolator 25.

  The sample chamber 26 forms a sealed space and is configured so that the inside can be maintained at a high vacuum. The mirror body 1 is attached to the top plate, and the preliminary exhaust chamber 27 is attached to the side wall. A sample stage 17 is provided in the interior.

  The mirror 1 (charged particle beam optical system) is mounted upright on the upper part of the sample chamber 26, is made of a cylindrical member without a bottom communicating with the inside, and has an electron source 2 near the top of the interior, and a primary acceleration electrode. The primary electrons 4 accelerated by 3 are converged by the condenser lens 5, limited to a necessary current by the diaphragm 6, focused on the wafer 9 on the sample stage 17 by the objective lens 19, and irradiated.

  The sample stage 17 on which the wafer 9 can be mounted is configured to be able to move an arbitrary amount in an arbitrary direction in a plane perpendicular to the central axis of the primary electrons 4, that is, in a horizontal plane. The sample stage 17 makes it possible to irradiate an arbitrary position on the wafer 9 with primary electrons.

  However, since the sample stage 9 cannot be positioned with an accuracy of several nanometers, the image shift coil 8 (field shift deflector) is used to correct the positioning error of the sample stage 9. In this case, for example, visual field correction is performed using a visual field specifying method called addressing. Addressing is a technique for searching a pattern (addressing pattern) that is relatively large with respect to a pattern to be measured and has a unique shape from a low-magnification image using a pattern matching method or the like. After the position of the addressing pattern is specified, the stage error is corrected by moving the visual field by the deflector to the measurement target pattern having a known positional relationship with the addressing pattern. This makes it possible to accurately perform beam irradiation at a desired irradiation location.

  After the location to be measured is accurately identified, the scanning coil 7 scans the primary electron 4 in the measurement observation area. At this time, secondary electrons 16 are generated from the wafer 9. The generated secondary electrons 16 are generated more in the edge portion having a surface inclined with respect to the beam than in the flat sample surface due to the edge effect, so that an image with a bright edge portion can be formed. The secondary electrons 16 emitted from the wafer 9 can be detected by the detector 10 and processed by the image processing 12 to form an image representing the surface shape from the relationship with the scanning XY signal.

  The image acquired in this way can be displayed on the display device 13 or can be taken into the control computer 14 and used for measurement of dimensions, measurement of pattern positions, observation of foreign matter, and the like.

  The gate valve 23 provided between the preliminary exhaust chamber 27 and the sample chamber 26 is configured to be openable and closable, and the sample is being exhausted while the preliminary exhaust chamber 27 is being evacuated from the atmosphere to the vacuum, or during a leak. It is configured to be closed when there is a pressure difference between the chamber 26 and the preliminary exhaust chamber 27. The preliminary exhaust chamber 27 is provided with an exhaust device 21 and a leak valve 22. The exhaust device 21 is used when performing vacuum exhaust, and the leak valve 22 is used when returning from the vacuum state to the atmosphere. A vacuum gauge 24 for monitoring the pressure in the preliminary exhaust chamber 27 is provided.

  The preliminary exhaust chamber 27 is loaded with a wafer to be inspected next, and after the observation inspection of the wafer 9 carried into the sample chamber is completed, the gate valve 23 is opened to open the wafer in the preliminary exhaust chamber and the sample chamber. The wafer 9 is replaced.

  Here, in order to transfer the wafer from the outside to the preliminary exhaust chamber 27 or to take out the wafer from the preliminary exhaust chamber, the gate valve 23 between the sample chambers 26 is closed, and the preliminary exhaust chamber 27 is air leaked by the leak valve 22, and the atmospheric pressure. Execute again. In order to open the gate valve 23 between the sample chambers 26, the preliminary exhaust chamber 27 is kept in vacuum.

  FIG. 2 shows how each part changes when the inside of the preliminary exhaust chamber 27 changes in vacuum. Here, as the wall between the preliminary exhaust chamber 27 and the sample chamber 26 is deformed, the top plate portion of the sample chamber 26 is mainly deformed, and the mirror body 1 is inclined to the left side of the drawing as shown in the figure. . The inclination of the mirror body 1 changes the position of the center of gravity of the entire apparatus as the position of the stage 17 moves. As a result, the amount of tilt changes due to the tilt of the entire sample chamber 26 mounted on the vibration isolator 25. When the mirror 1 is tilted in this way, the irradiation position of the primary electrons 4 on the wafer 9 is shifted from the original position, and this shift indicates that the secondary electron image moves on the screen display. If this is the case, there is a risk of measurement accuracy and erroneous inspection.

  In such an apparatus, the data collection method for correcting the irradiation position deviation of the primary electrons 4 on the wafer 9 is shown in FIG. 4, and the actual observation inspection flow using the correction data is shown in FIG. .

  As shown in FIG. 3, a plurality of measurement marks 29 are set in advance on the stage 9. The coordinates of each mark (Xi, Yi) (i = 1, 2,...) Are known coordinates, and the shape of the mark is XY of the primary electrons 4 and the amount of movement of each is easy to measure. . In this example, a cross pattern is used, but a dot shape may be used as long as the amount of movement is known.

  First, the stage 17 is moved so that the measurement mark 29 coordinates (X1, Y1) are directly below the primary electrons 4 (S101). Thereafter, the vacuum exhaust of the preliminary exhaust chamber is stopped, the leak valve is opened, and the leak is started (S102). After the start, the mark is scanned two-dimensionally with primary electrons so that the mark can be observed, and an image at that time is captured (S103). Next, the vacuum degree Pl and time tl of the preliminary exhaust chamber are taken in (S104). The primary electron scanning and the image capturing are repeated until the preliminary exhaust chamber leaks (S105).

  Next, the leak valve is closed, the pump is operated, and exhaust is started (S106). After that, as in the case of the leak, the mark is scanned with a primary electron in a two-dimensional manner so that the mark can be observed, and an image at that time is captured (S107). Next, the degree of vacuum Pe and time te in the preliminary exhaust chamber are taken in (S108). The primary electron scanning and image capture are repeated until the vacuum in the preliminary exhaust chamber reaches the target vacuum and exhaust is completed (S109).

  From the images thus obtained, the amount of movement of the mark on the image by the image matching process or the like based on the appearance of the mark in each image is (dxl, dyl) at the time of leak and (dxe, dye) at the time of exhaust. Is calculated. From the calculated values, the degree of vacuum Pl and Pe and the following relationship are obtained (S110).

(Dxl, dyl) = F1 (Pl) (Formula 1)
(Dxe, dye) = G1 (Pe) (Equation 2)
Ask for. The relationship may be an approximate expression or a table format.

  Next, the stage is moved so that the next mark (X2, Y2) is directly below the primary electron beam (S101), the same processing is performed, and F2 and G2 that are the relationship between the moving amount and the pressure are obtained in the same manner. .

  In this way, the relations Fn and Gn for all n marks on the stage are obtained. The value calculated in this way is stored in a storage medium built in the control computer 14 or the like. FIG. 5 shows a flow when an image is actually observed using (F1, F2... Fn) and (G1, G2... Gn) calculated as described above.

  The stage is moved to the observation coordinates (X, Y) so that the place to be observed is directly below the primary electrons (S201). Next, the correction processing program is requested to start the correction processing (S202). The correction processing program cancels the correction processing start waiting (S206), and takes in the state of whether the preliminary exhaust chamber is in the atmospheric state, the vacuum state, the exhaust state, the leak state, and the vacuum degree of the preliminary exhaust chamber (S207). ). If the correction process is in a leak state, the correction amount is calculated from the current stage coordinates, (F1, F2,... Fn) and the current degree of vacuum.

  First, a correction amount at each mark coordinate is calculated from (F1, F2... Fn) and the current degree of vacuum. Next, the correction amount is calculated by interpolation from the relationship between the observation coordinates (X, Y) and the coordinates (Xn, Yn) of the measurement mark 29 (S209). In the exhaust state, the correction amount is calculated from the current stage coordinates, (G1, G2,... Gn) and the current degree of vacuum. The correction amount is calculated by the same calculation as in the exhaust state (S210).

  In the case of a vacuum, no correction is performed. In the case of the atmospheric state, the difference (offset) between the position in the vacuum state and the atmospheric state is calculated in advance using the relationship of (F1, F2... Fn) or (G1, G2... Gn). . Next, a correction amount is calculated by interpolation from the relationship between the observation coordinates (X, Y) and the coordinates (Xn, Yn) of the measurement mark 29 (S211). Although a certain amount of addition during scanning does not affect image displacement during scanning, the result of scanning the target position value for observation inspection and the observation inspection object from the image and stage coordinates is a vacuum. Since it becomes a difference between the inside and the atmosphere, it is corrected.

  The control circuit 15 (control device) controls the current so that the correction amount calculated in this way is moved by the correction amount with respect to the image shift coil 8 (S212).

  This process is repeated until an end request is received from the observation inspection process. On the other hand, on the observation inspection processing side, after making a correction start request, scanning of an electron beam and a secondary electron image for observation inspection are taken in as usual (S203).

  By controlling in this way, even if the preliminary exhaust chamber is exhausted or leaked during observation scanning, the primary electrons can be scanned without being shifted on the observation inspection target on the sample.

  Thereafter, the observation inspection process issues a correction end request to the correction process (S204). The correction process receives this request, ends the correction process, and waits for a correction process start request again.

  In the observation inspection process, the secondary electron image is averaged by the image processing unit and the observation inspection process is performed (S205).

  According to the present embodiment, it is possible to acquire an image equivalent to when the preliminary exhaust chamber leaks or the vacuum degree of the preliminary exhaust chamber is constant even during exhaust. The apparatus of this embodiment selectively corrects the visual field position based on the correction amount as described above at the time of evacuation of the preliminary exhaust chamber 27 (a period from the atmosphere to a predetermined vacuum) or at the time of leakage. In this embodiment, the detection result of the vacuum gauge 24 (pressure gauge) and the correction amount at that time are obtained in advance, thereby making it possible to correct the visual field when the pressure in the preliminary exhaust chamber changes. Correction may be performed using an arithmetic expression or a table indicating the relationship between (time) and the correction amount. However, since there are samples that generate gas during evacuation, in order to perform accurate visual field correction according to deformation of the sample chamber 26, visual field correction based on pressure detection is performed as in the above-described embodiment. Is desirable.

  Furthermore, as illustrated in FIG. 7, when the movement trajectory 701 of the beam irradiation position is not linear, a pattern position 704 for each unit pressure value (time) is obtained, and the pattern position is approximated by a curve. The beam trajectory may be obtained. A mark 702 is a mark position before evacuation (before leakage), and a mark 703 is a mark position after evacuation (after leakage). By performing the visual field correction so as to be reversed along the moving trajectory, it is possible to perform an accurate visual field correction. When the amount of movement for each time differs, it is preferable to perform correction according to the movement speed for each time.

  The inclination of the mirror body 1 due to processing such as exhaust or leakage of the preliminary exhaust chamber 27 is mechanical and may change with time. Accordingly, FIG. 6 shows a flow for automatically calibrating during normal inspection observation.

  In the automatic measurement by the recipe, processing is performed on the observation inspection points stored in the recipe, but first, the observation inspection processing shown in FIG. 5 is performed by moving to the coordinates on the wafer specified in the recipe (S301). ). From the acquired image at the start of scanning and the final scanned image obtained at this time, the positional deviation amount (dx, dy) of the target observation inspection object is calculated by image processing (S302). Depending on the state of the preliminary exhaust chamber, it is added to the deviation during leak (sdxl, sdyl), the deviation during vacuum exhaust (sdxe, sdy), and the steady deviation (sdxn, sdyn).

  The observation inspection process (S301) is repeated until all observation inspection points are completed. The average deviation amount in each state is calculated from the added value of the deviation amounts thus calculated (S308). The average value of the calculated deviation amounts is compared with the average of the steady deviation, the deviation during leak, and the deviation during vacuum exhaust. When the preset allowable value is exceeded, the correction amount calculation process shown in FIG. 4 is executed. (S310). Compared to the steady-state deviation amount, the wafer manufacturing process is charged with primary electron irradiation and the image may drift, so the image drift caused by the process and the image drift caused by the preliminary exhaust chamber are distinguished. It is to do.

DESCRIPTION OF SYMBOLS 1 ... Mirror body, 2 ... Electron source, 3 ... Primary electron acceleration electrode, 4 ... Primary electron, 5 ... Condenser lens, 6 ... Aperture, 7 ... Scanning coil, DESCRIPTION OF SYMBOLS 8 ... Image shift coil, 9 ... Wafer (sample), 10 ... Detector, 11 ... Amplifier, 12 ... Image processor, 13 ... Image display apparatus, 14 ... Control computer, 15 ... control circuit, 16 ... backscattered electrons, 17 ... stage, 18 ... primary electron acceleration power source, 19 ... objective lens, 20 ... pump, 21 ... Exhaust device, 22 ... leak valve, 23 ... gate valve, 24 ... vacuum gauge, 25 ... vibration isolator, 26 ... sample chamber, 27 ... preliminary exhaust chamber, 28 ...・ Mounting base, 29 ... Mark for measurement

Claims (8)

A charged particle beam optical system for irradiating a charged particle beam, a vacuum chamber for maintaining the atmosphere around the sample irradiated with the charged particle beam in a vacuum, and a preliminary exhaust chamber for evacuating the atmosphere around the sample introduced into the vacuum chamber And a charged particle beam apparatus comprising a control device for controlling the charged particle beam optical system,
A deflector for deflecting the irradiation position of the charged particle beam is provided, and the control device deflects the charged particle beam based on a predetermined condition at the time of preliminary exhaust and / or leak of the preliminary exhaust chamber. Thus, the charged particle beam apparatus is characterized by controlling the deflector. In claim 1,
The control device controls the deflector so as to deflect the charged particle beam according to the pressure of the preliminary exhaust chamber at the time of preliminary exhaust in the preliminary exhaust chamber and / or at the time of leakage. Charged particle beam device. In claim 1,
The charged particle beam apparatus according to claim 1, wherein the control device controls the deflector so as to deflect the charged particle beam according to a preliminary exhaust time and / or a leak time of the preliminary exhaust chamber. In claim 1,
The control device controls the deflector so that the scanning position of the charged particle beam is constant regardless of the pressure of the preliminary exhaust chamber or the pressure change time. apparatus. In claim 1,
A charged particle beam apparatus characterized in that a sample stage for arranging a sample is provided in the vacuum chamber, and the sample stage is provided with a mark that can be imaged by scanning the charged particle beam. . In claim 5,
The charged particle beam apparatus according to claim 1, wherein the control device obtains a deflection condition of the deflector based on movement of the mark at the time of preliminary exhaust of the preliminary exhaust chamber and / or at the time of leakage. In claim 1,
The charged particle beam apparatus, wherein the control device adds and averages a plurality of images obtained by deflecting the charged particle beam based on the predetermined condition. In claim 1,
A scanning deflector for scanning the charged particle beam;
The control device obtains a deflection condition of the deflector when scanning by the scanning deflector is being performed and the change in the scanning position of the charged particle beam exceeds the allowable range while the pressure in the preliminary exhaust chamber is changing. A charged particle beam apparatus characterized by that.