An object of the present invention is to provide a pattern measuring apparatus which performs high-accuracy concavity/convexity determination (e.g., distinguishing between a line segment and space) while simultaneously reducing the dose of a beam falling onto a pattern to be measured. To attain the object, this invention proposes a pattern measuring apparatus which specifies a pattern in a measurement object area by scanning a tilted bean with respect to another area different from the measurement object area and then performs measurement based on the pattern-specifying result. With such arrangement, it becomes possible to perform measurement without the risk of wrong pattern designation while lowering the dose of a beam hitting the measurement object area.
|
1. A pattern measuring apparatus, comprising:
an arithmetical unit for determining a pattern size by using a signal to be obtained based on scanning of a charged particle beam emitted from a charged particle source, and
a beam tilting deflector for irradiating the charged particle beam while tilting this beam with respect to an ideal light axis thereof, and
wherein the arithmetical unit specifies a kind of a pattern formed on a sample based on a signal to be obtained by obliquely scanning the beam in an area of the sample being different from an area used to obtain the pattern size.
9. A pattern measuring apparatus having an arithmetical unit for determining a pattern size by using a signal to be obtained based on scanning of a charged particle beam emitted from a charged particle source, wherein the apparatus comprises a beam tilting deflector for irradiating the charged particle beam while tilting this beam with respect to an ideal light axis thereof and wherein the arithmetical unit specifies a kind of a pattern formed on a sample based on a signal to be obtained by obliquely scanning the beam in an area of the sample being different from an area used to obtain the pattern size, thereby to measure the pattern thus specified.
2. The pattern measuring apparatus according to
3. The pattern measuring apparatus according to
4. The pattern measuring apparatus according to
5. The pattern measuring apparatus according to
6. The pattern measuring apparatus according to
7. The pattern measuring apparatus according to
8. The pattern measuring apparatus according to
10. The pattern measuring apparatus according to
|
The present invention relates generally to pattern measuring apparatus which perform pattern measurement based on pattern concavity/convexity determination or the like, and more particularly to a pattern measuring apparatus that performs pattern identification of line segments and gap spaces formed on a semiconductor wafer.
A scanning electron microscope (SEM) is one form of charged particle beam apparatus, which enables observation of fine objects. The SEM has a variety of applications, including measurement and inspection of circuit patterns that make up highly integrated semiconductor devices. Incidentally, with recent advances in miniaturization of semiconductor devices, circuit patterns increase in integration density. In particular, this has often led to appearance of ultrafine circuit patterns, known as line-and-space patterns having successive line patterns, wherein lines and gap spaces are hardly distinguishable from each other in interval or pitch. Additionally, with advances in semiconductor microfabrication technology, observation images of a line and space are becoming more similar in look to each other.
An increase in visual similarity of line and space results in the risk that one of them is mistaken for the other. More specifically, even when an image was acquired in order to measure the width of a line, it can happen in practice that the width of a space is measured by mistake. Japanese Patent Literature JP-A-2004-251674 (its corresponding United States Patent Application Publication is US 2004-A-0222375) discloses therein a technique for distinguishing between line and space portions on the basis of expansion of hem parts of a plurality of peaks contained in a differential waveform of a profile, which has been obtained by applying projection processing to an image acquired. However, sensed images tend to decrease in contrast due to further miniaturization in recent years; so, it is difficult in some cases to extract the difference in brightness of hem parts of profile peaks. Once peaks overlap each other, discrimination can be failed.
JP-A-2006-332069 (its corresponding U.S. Pat. No. 6,872,943) discloses a technique for performing concavity/convexity determination with high accuracy by irradiating a tilted beam and emphasizing a difference between peaks of two, right and left edges of a line.
The techniques as disclosed in the above-cited Patent Literatures are to enable distinction of lines and spaces that have traditionally been difficult to be distinguished over each other. Especially, the technique taught by JP-A-2006-332069 is capable of emphasizing a difference between the right and left edges of a line, thus making it possible to enhance the success rate of concavity/convexity determination. However, as the beam is irradiated obliquely with respect to the pattern being measured, an image to be obtained by such beam irradiation is an oblique image, which is unsuitable for use in accurate measurement of pattern sizes. Although it is also permissible to perform size measurement by scanning a beam for the measurement use (beam with no tilt) after having completed the concavity/convexity determination, a need is felt to apply to the pattern under measurement both the beam scanning for concavity/convexity determination and the beam scanning for size measurement.
By taking into consideration the adhesion of electrification charge due to beam irradiation and pattern shrink, it is desirable that the beam irradiation onto the measurement object pattern stay less. Proposed below is a pattern measuring apparatus capable of performing high-accuracy concavity/convexity determination (e.g., line/space distinction) while at the same time suppressing the irradiation amount of a beam falling onto a pattern under measurement.
To attain the foregoing object, in accordance with one aspect of this invention, a pattern measuring apparatus is provided which specifies a pattern in an object area under measurement by scanning a tilted beam with respect to an area different from the measurement object area and executes measurement based on the specified result.
With the arrangement stated above, it becomes possible to perform measurement without the risk of wrong pattern designation while simultaneously lowering the dose of a beam hitting the measurement area.
Other objects, features, and advantages of the present invention will become apparent from the following description of embodiments of the present invention provided in relation to the accompanying drawings.
An explanation will now be given below of an apparatus for measuring pattern sizes by using either image data obtained by an image sensing device or waveform data, examples of which apparatus include a pattern measuring apparatus that executes desired measurement by accurately specifying a target pattern to be measured.
A scanning electron microscope (SEM) which irradiates an electron beam onto a semiconductor circuit pattern and detects secondary electrons to be derived therefrom to thereby evaluate the shape of a circuit pattern that is the object of interest will be described below as an example of the image sensor device, although other charged particle beam devices, such as a focused ion beam (FIB) device or like equipment, may alternatively be employed as the imager device.
In the embodiment below, an example will be described which tilts an electron beam within an area different from the measurement object (i.e., area unused for length measurement) in the process of acquiring a length measurement image that is subjected to pattern measurement and uses the information to be obtained based on irradiation of such tilted beam to perform concavity/convexity determination. By the beam tilting, the electron beam is caused to hit one side-wall of a circuit pattern rather than its top surface; so, this sidewall increases in width of white band when compared to top-down images. As the white band varies in width in a way depending on a tilt angle of the electron beam, it becomes easier to specify an edge on one side with the white band getting thick. A sidewall onto which the electron beam is irradiated is a sidewall of convex portion. Thus, if this edge is specifiable, it becomes possible to specify the positions of convex and concave portions with respect to this edge.
More concretely, in the case of tilting the beam by using the swing-back action of an objective lens, it is understood that when the sidewall is used as a reference, a convex part (line) is positioned in the beam swing-back direction for beam tilt whereas a concave part (space) is placed in the reverse direction to the swingback direction. In this embodiment, the concavity/convexity determination is achieved by evaluation of the electron beam's tilt angle and the extensity of the white band at that time. Additionally, in areas used for circuit pattern length measurement, it is possible to perform length measurement with high accuracy because image pickup is done in the state that the tilt angle is returned to vertical beam angle.
In accordance with the embodiment explained below, the line-and-space concavity/convexity determination is carried out by irradiating a tilted electron beam with respect to an area which is not used for length measurement whereby it becomes possible to achieve high-accuracy line-and-space concavity/convexity determination in a short period of time, which is substantially the same as an image pickup time consumed to simply sense a length measurement image.
In addition, it is no longer necessary to perform a plurality of scanning operations with the tilt angle varied. Thus, the irradiation amount or “dose” of the electron beam is lowered. This makes it expectable to reduce chargeup and shrink of a sample, also known as workpiece. The intended image is acquirable by a single scan. It is thus unnecessary to execute positioning between an oblique image and length measurement image. This makes it possible to perform the concavity/convexity determination in a simple and easy manner.
Practical embodiments will be described in detail with reference to the accompanying drawings below.
(Overview of Scanning Electron Microscope)
An explanation will first be given of a scanning electron microscope (SEM) in accordance with one embodiment of the invention. As shown in
The electron source 101 is the one that emits a primary electron beam 108. The acceleration electrode 102 is for accelerating primary electrons. The focusing lens 103 is to perform focusing of the primary electron beam. The deflector 104 performs two-dimensional (2D) scanning deflection of primary electrons. The objective lens 105 focuses primary electrons onto the sample wafer 106. A stage 107 is for stably mounting the sample thereon.
The detector 110 detects an emission electron signal 109 of secondary electrons generated from the sample. The digitizer 111 digitizes a detected signal. A deflection control unit 112 controls a deflection amount or the like in the deflector. Respective ones of these components are connected to the system control unit 113 via a bus 119.
The pattern measuring apparatus have other components, including the image memory 116 for storage of image data, the image generator 115 that performs image creation processing, the image processing unit 114 that performs processing of an image acquired, the recipe unit 117 that stores a recipe(s) containing therein inspection conditions, data entry devices for giving instructions to the device, such as a keyboard and a pointing device called the mouse, the I/O unit 118 made up of a display monitor and/or a printer for outputting data from the device, and the result storage unit 120 for storing measurement and inspection results. These components are interconnected together by the bus 119.
This embodiment device also functions to form a profile based on detected secondary electrons or reflection electrons or else. This profile is formed by execution of projection processing based on an electron detection quantity at the time of performing one-dimensional (1D) or two-dimensional (2D) scanning of primary electrons or based on brightness information of an acquired image. The profile obtained is for use in size measurement of patterns fabricated on a semiconductor wafer, for example.
The SEM of this embodiment further includes a beam tilting deflector 805 shown in
(SEM Image Acquisition)
Next, SEM image acquisition in the embodiment device will be explained with reference to
Firstly, the electron source 101 emits primary electron beam 108, which is accelerated by the acceleration electrode 102. Subsequently, rays of this beam are gathered together by the focusing lens 103, deflected by the deflector 104, and thereafter focused again by the objective lens 105 to fall onto the sample wafer 106 at a target location on the top surface thereof. By the irradiation of this primary electron beam 108, an emission electron signal 109 is generated from the sample surface, the intensity of which signal is detected by the detector 110 and is then converted by the digitizer 111 into a digital image signal.
The signal as output from the digitizer 111 is sent to the image generator unit 115, which performs processing including pixel multiplication, digital filtering and others, thereby providing processed digital image data which is stored in the image memory 116.
Then, at the image processor unit 114, frame addition-average processing is applied to the digital image data as read out of the image memory 116 in a manner that one input line is processed at a time.
Additionally, by irradiating the electron beam while simultaneously tilting it by the beam tilting deflector 805 shown in
As the half-value width of an edge signal in the image pickup direction becomes larger in this way, use of this phenomenon makes it possible to achieve the line-and-space concavity/convexity determination based on an increase or decrease of the half-value width.
More specifically, when the width of a profile that was obtained by tilting the electron beam with respect to its light axis increases in the electron beam scanning direction, it is decidable that a portion with formation of a sidewall (edge) in the electron beam tilting direction exists at a location corresponding to the profile. In other words, when the measured extensity of a detection signal in the linear scan direction (i.e., X direction in the case of 2D scanning) becomes larger than that in the case of irradiating the electron beam along its light axis, it is determinable that an edge surface exists on the electron beam tilt direction side. Conversely, when the width of the profile becomes smaller, it is determinable that the object of interest is a portion corresponding to a rear surface (opposite side) when viewing from the tilt direction of the electron beam. That is to say, it can be seen that in cases where one of two peaks appearing in the event of obliquely scanning the electron beam becomes larger in width whereas the other becomes smaller, either a concave or convex portion exists therebetween. Additionally in this state, when a peak with its increasing width exists on the electron beam tilt side, it is decidable that a concave portion is present; when a peak with its decreasing width exists on the electron beam tilt side, a convex portion exists.
A further feature of this embodiment is that image pickup is performed while moving the tilt direction relative to an image region which is not used for length measurement of the sample as shown in
In accordance with this embodiment, there is no need to perform repeatedly the scanning for concavity/convexity determination and the scanning for length measurement. This suppresses the irradiation amount or “dose” of such electron beam. Thus it is possible to reduce damage to the sample.
After having completed the scanning stated above, the arithmetical unit executes concavity/convexity determination based on the information extracted from a tilted-beam irradiation area and the above-stated judgment criterion (at step 506, 507). Checking whether or not the concavity/convexity determination is succeeded is as follows: a white band (peak) value of a sidewall opposing the beam coming from the beam tilt direction (swing-back direction by means of the objective lens), for example, is compared with a predetermined value; when the former is greater than or equal to the latter, a decision of success is made. This decision of successful completion of the concavity/convexity determination may alternatively be made in the case of a peak value difference from a sidewall on the opposite side is larger than or equal to a prespecified value. Still alternatively, in case the beam is tilted in different directions, one of which is for an upper side region of the length measurement object area and the other of which is for a lower side region thereof as shown in
In case it is decided that the concavity/convexity determination was completed successfully, a length measurement value is calculated based on a preset measurement condition (at step 508). For example, if the measurement condition is to measure the width of a line pattern, then measure the size of a distance between edges of this pattern. In this case, a length measurement box is set at both edges of the pattern to be measured; then, measure a peak-to-peak distance of brightness profile within the box.
In the case of the concavity/convexity determination being judged to be in fail, an error message is issued to prompt a system operator(s) to make a decision as to what kind of action should be taken thereafter (step 509).
A result of superposition of a concavity/convexity determination profile and acquired image is shown in
The switching between the tilted beam irradiation state and normal beam irradiation state based on the beam deflection at a locking position is executable by turn-on/off control of the beam tilting deflector-theoretically, no appreciable irradiation position is moved. Thus, it is possible to minimize the risk of pattern selection mistake.
Owing to the ability to achieve the concavity/convexity determination using one frame information, there is no need to perform multiple scanning operations. It is thus possible to prevent throughput degradation while at the same time suppressing pattern shrink due to excessive beam irradiation. This makes it expectable to improve the reproducibility of length measurement.
An explanation will next be given of an example which performs concavity/convexity determination of a workpiece or sample wafer by a technique other than the beam tilt method, which utilizes the swing-back action of an objective lens.
In the example of
Since the sizes of line width and gap space width are predefined by design data, it is also possible by the concavity/convexity determination to readily specify a pattern at a position which is moved by a predetermined distance from the pattern that was identified by the concavity/convexity determination.
As the addressing pattern 1204 and the measurement object area 1201 are in a known positional relationship, it is possible by specifying the position of addressing pattern 1204 to accurately perform the view movement toward the measurement area 1201. In this embodiment, prior to moving the view to the measurement area, the view is shifted to the concavity/convexity decision area 1205, thereby executing concavity/convexity determination (at step 1106). Since the positional relation between the measurement area 1201 and concavity/convexity decision area 1205 is also known, the view movement of from the addressing pattern 1204 to concavity/convexity decision area 1205 is also executable based on the known information. In a case where the concavity/convexity determination is judged to be succeeded, the view is shifted to the measurement area 1201 (step 1107); then, execute length measurement (step 1108). If the concavity/convexity determination is failed, any one of the following corrective actions is taken (at step 1109): measurement is performed again; measurement is skipped; and, an error message is issued to prompt the system operator to take remedial action for avoiding occurrence of measurement interruption accidents.
With the above-stated arrangement, it becomes possible to perform the intended measurement based on accurate pattern identification by means of the concavity/convexity determination, without having to perform excessive bean irradiation onto the measurement object area 1201. Thus it is possible to reduce the risk of beam irradiation damages with respect to the object being measured, thereby suppressing the influence, such as sample shrink or the like. This leads to achievement of a technical advantage as to improvement in length measurement reproducibility.
Although the accuracy of the view movement by the deflector is high enough to enable execution of precise concavity/convexity determination as far as the positional relationship of patterns is approximate to that defined in the design data, undesired determination with half-pitch deviation can still be made, resulting in a line pattern being misjudged to be a gap space. In this respect, an example capable of attaining more accurate specifiability will be explained, which specifies a deviation of view field on a display screen for the concavity/convexity determination use and correct such deviation accurately to thereby situate the view in the measurement area.
This example is arranged so that the arithmetical unit evaluates a deviance between an image of concavity/convexity determination area and its corresponding design data and shifts the view field by a distance equivalent to such deviance during positioning of the view in the measurement area.
By performing deviance correction using an on-screen display image of the concavity/convexity decision area in the way stated above, it becomes possible to accurately locate a view field for the length measurement use at the exact position. The view position correction may be replaced with a process of adjusting the position of a length measurement box by a distance corresponding to the deviance stated supra.
Although in this embodiment the concavity/convexity determination is performed based on the information gained from the concavity/convexity decision area 1003, this may be modified in various ways, one example of which is as follows: another concavity/convexity decision area 1004 is provided to permit the decision of success in concavity/convexity determination to be made when it is verified that the both areas are consonant with each other in concavity/convexity determination results. Additionally, the concavity/convexity determination and the image acquisition of concavity/convexity decision area may be arranged so that these are performed after having acquired an image of length measurement object area and yet before execution of the length measurement.
Although the above-stated arrangement is the one that concerns concavity/convexity determination of patterns, autofocusing may also be implemented simultaneously during scanning of a concavity/convexity decision area (at step 1105).
However, in the case of Δx1 being large in value, a deviance Δf 1503 takes place between a focus position 1501 of concavity/convexity decision area and a focus position 1502 of length measurement object area as shown in
In this case, in light of the fact that Δx1 is a known value, a properly focused image is acquirable by preparing in advance a focus position correction table 1504 shown in
By using the above-stated method, it is no longer necessary to perform the scanning for the autofocusing separately, thus making it possible to obtain an advantage as to improvement of throughput in addition to the improvement in accuracy of concavity/convexity determination.
The above-stated pattern distinguishing method using a tilted beam is employable not only for convexo-concave patterns, such as line-and-space patterns, but also for the specifying of circuit patterns that are fabricated by self-alignment double patterning (SADP) techniques.
Accordingly, it is possible by performing pattern measurement based on the oblique image acquisition and position detection, which have been discussed in the embodiments stated supra, to accurately perform the intended size measurement of one set of circuit patterns chosen as the length measurement object, without mistaking the measurement position.
While in the embodiments 1-2 the concavity/convexity determination is performed using the half-value width of brightness profile, the SADP process is such that period judgment is performed using a peak value(s) of brightness profile and a position(s) thereof because of the fact that the visual brightness (luminance value) varies as shown in
The oblique image's peak position is obtained with the length measurement image's peak value being as a reference. In case the oblique image and length measurement image are different in pixel number from each other, their native values are not directly usable for comparison. Therefore, this comparison is done after having applied normalization to the brightness profiles, followed by execution of peak position detection (step 1406).
The peak position detection may be performed by various methods other than the above-stated method, including self-correlation using a single oblique image and intercorrelation of two oblique images.
The SADP-formed pattern shown in
In light of the fact that the interval of peak positions detected in each brightness profile is expected to have the periods λ1, λ2 and λ3 as shown in
Finally, a peak position of length measurement image that is the closest to the peak position of oblique image detected in step 1406 becomes at the measurement position of the length measurement object (at step 1407), followed by execution of length measurement (step 1408).
With the arrangement stated above, it is possible to perform measurement based on accurate pattern identification relying upon the cycle judgment which has been difficult by mere use of a length measurement image, without having to excessive beam irradiation onto the measurement object area.
It is needless to say that in this embodiment also, effects and advantages of the aforesaid embodiments are expectable.
It should be further understood by those skilled in the art that although the foregoing description has been made on embodiments of the invention, the invention is not limited thereto and various changes and modifications may be made without departing from the spirit of the invention and the scope of the appended claims.
Maeda, Tatsuya, Ikeda, Mitsuji, Koyama, Susumu, Nishihama, Hiroshi
Patent | Priority | Assignee | Title |
Patent | Priority | Assignee | Title |
6872943, | Jul 12 2001 | Hitachi, Ltd. | Method for determining depression/protrusion of sample and charged particle beam apparatus therefor |
8063363, | Mar 31 2008 | HERMES MICROVISION INCORPORATED B V ; ASML NETHERLANDS B V | Method and apparatus for charged particle beam inspection |
8497475, | Mar 09 2009 | HERMES MICROVISION INCORPORATED B V ; ASML NETHERLANDS B V | Method and apparatus for charged particle beam inspection |
20040222375, | |||
20110147586, | |||
JP2004251674, | |||
JP2006332069, | |||
JP4996206, |
Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
Mar 07 2013 | Hitachi High-Technologies Corporation | (assignment on the face of the patent) | / |
Date | Maintenance Fee Events |
Date | Maintenance Schedule |
Dec 31 2016 | 4 years fee payment window open |
Jul 01 2017 | 6 months grace period start (w surcharge) |
Dec 31 2017 | patent expiry (for year 4) |
Dec 31 2019 | 2 years to revive unintentionally abandoned end. (for year 4) |
Dec 31 2020 | 8 years fee payment window open |
Jul 01 2021 | 6 months grace period start (w surcharge) |
Dec 31 2021 | patent expiry (for year 8) |
Dec 31 2023 | 2 years to revive unintentionally abandoned end. (for year 8) |
Dec 31 2024 | 12 years fee payment window open |
Jul 01 2025 | 6 months grace period start (w surcharge) |
Dec 31 2025 | patent expiry (for year 12) |
Dec 31 2027 | 2 years to revive unintentionally abandoned end. (for year 12) |