This invention relates to metrology systems and methods in determining features of sub-micron structural elements such as lines, contacts, trenches and the like, of measure objects such as but not limited to semiconductors wafers, reticles.
Integrated circuits are very complex devices that include multiple layers. Each layer may include conductive material, isolating material while other layers may include semi-conductive materials. These various materials are arranged in patterns, usually in accordance with the expected functionality of the integrated circuit. The patterns also reflect the manufacturing process of the integrated circuits.
Integrated circuits are manufactured by complex multi-staged manufacturing processes. During this multi-staged process resistive material is (i) deposited on a substrate/ layer, (ii) exposed by a photolithographic process, and (iii) developed to produce a pattern that defines some areas to be later etched.
Various metrology, inspection and failure analysis techniques evolved for inspecting integrated circuits both during the fabrication stages, between consecutive manufacturing stages, either in combination with the manufacturing process (also termed “in line” inspection techniques) or not (also termed “off line” inspection techniques). Various optical as well as charged particle beam inspection tools and review tools are known in the art, such as the VeraSEM™, Compluss™ and SEMVision™ of Applied Materials Inc. of Santa Clara, Calif.
Manufacturing failures may affect the electrical characteristics of the integrated circuits. Some of these failures result from unwanted deviations from the required dimensions of the patterns. A “critical dimension” is usually the width of a patterned line, the distance between two patterned lines, the width of a contact and the like.
One of the goals of metrology is to determine whether the inspected objects includes deviations from these critical dimensions. This inspection is usually done by charged particles beam imaging that provide the high resolution required to measure said deviations.
A typical measured structural element is a line that has two opposing sidewalls. The measurement of the bottom width of the line involves measuring the top width of the line as well as measuring its sidewalls.
Measurement of a structural element line critical dimensions using only a top view (in which the electron beam that scans the line is perpendicular to the substrate) may result in faulty results, especially when one of the sidewalls has a negative sidewall angle such that an upper end of the sidewall obscures a lower end of that sidewall.
In order to address said inaccuracies CD-SEM tools that enable electronic tilt of an electron beam were introduced. NanoSem 3D of Applied Materials from Santa Clara, is a fully automated CD-SEM that has a column that allows electronic tilting as well as mechanical tilting of the scanning electron beam to scan the wafer surface with various tilt angles from several directions.
Critical dimension measurement may involve illuminating a test object by multiple tilted beams and processing the detected waveforms to define critical dimensions.
Multiple measurements have some disadvantages. First, they reduce the throughput of the inspection system, especially when the measurement involves changing the tilt of scanning electron beam. Such a change may require a de-Gauess stage, as well as an electron beam stabilization stage. A further disadvantage of multiple measurements results from degradation (for example shrinkage and carbonization) of the measured structural element, as well as unwanted charging of the measured structural element.
Due to various reasons, such as process variations, measurement inaccuracies and the like, the height and accordingly the measured height of structural elements varies across measured objects. In order to determine the structural height of a measured structural element there is a need to perform at least two measurements of said structural element, at two different tilt angles. The height of the structural element may be estimated, usually in response to multiple height measurements of structural element across the measured object. The estimation may be associated with height measurement errors, as well as estimation errors, that can effect the critical dimension measurements.
The invention provides various scanning schemes that enable to selectively reduce the amount of measurements required for determining cross sectional features of structural elements.
The invention provides methods and systems that allow accurate measurement of cross-dimensional features, with high accuracy, without knowing the height of the structural element.
The invention provides methods and systems that allow measurement of the whole cross-section of structural elements with a reduced number of scans and amount of measurements.
The invention provides a method for determining a cross sectional feature of a measured structural element having a sub-micron cross section, the cross section is defined by an intermediate section that is located between a first and a second traverse sections. The method includes a first step of scanning a portion of at least one reference structural element and a portion of the measured structural element to determine one or more relationships between the at least one reference structural element and the measured structural element. The first scanning step may be followed by additional scanning steps. According to an embodiment of the invention at least one additional scanning step is mandatory. According to other embodiments of the invention one or more additional scanning steps are optional, and their execution may be responsive to a fulfillment of various conditions, such as an uncertainty associated with the results of the previous scan. The uncertainty may be related to a steepness or orientation of a portion of a section of a measured structural element, to a relationship between a width of a section of the measured element and to a width of an electron beam that scans that section. The conditions may also reflect the required signal to noise ratio of the measurement, the accuracy of the measurement, and, optionally or alternatively, to the topography of the measured object
If additional scans are performed, additional relationships may be determined. Usually, the scan are preformed at different scan states.
According to an aspect of the invention one or more reference structural elements are provided. If multiple reference structural elements are provided structural element as a result of an interaction with the electron they may be positioned on both sides of a measured structural element, but this is not necessarily so.
The invention provides a system for determining a cross sectional feature of a structural element having a sub-micron cross section, the cross section is defined by an intermediate section that is located between a first and a second traverse sections, the system includes means for directing an electron beam towards an inspected object, at least one detector that is positioned such as to detect electrons emitted from the beam; and a processor, coupled to the at least one detector. The system is capable of performing at least one measurement, at one or more tilt states, to determine one or more relationships between one or more structural element and portions of a measured structural element. At least two relationship helps to determine the cross sectional feature.
In order to understand the invention and to see how it may be carried out in practice, a preferred embodiment will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which:
a is a schematic illustration of a critical dimension scanning electron microscope, in accordance with an embodiment of the invention;
b is a perspective view of an objective lens according to another embodiment of the invention;
a-2b illustrate various lines and reference structural elements;
a-3c are schematic illustration of waveforms that represent a relatively wide positively oriented traverse section, a relatively narrow traverse section and a negative oriented traverse section;
A typical CD-SEM includes an electron gun, for generating an electron beam, deflection and tilt units as well as focusing lens, for enabling scanning of a specimen with an electron beam, that may be in a certain tilt condition, while reducing various aberrations and misalignments. Electrons, such as secondary electrons that are omitted as result of an interaction between the specimen and the electron beam are attracted to a detector that provides detection signals that are processed by a processing unit. The detection signals may be used to determine various features of the specimen, as well as form images of the inspected specimen.
The invention may be implemented on CD-SEMs of various architectures that may differ from each other by the amount of their parts as well as the arrangement of said parts. For example the amount of deflection units, as well as the exact structure of each unit may vary. The CD-SEM may include in-lens as well as out of lens detectors or a combination of both.
A block diagram of a critical dimension scanning electron microscope (CD-SEM) 100 is shown schematically in
In-lens detector 16 is able to detect secondary electrons that escape from the specimen 105 at a variety of angles with relatively low energy (3 to 50 eV). Measurements of scattered or secondary corpuscles from a specimen can be conducted with detectors in the form of scintillators connected to photomultiplier tubes or the like. Since the way of measuring the signals does not influence the inventive idea in general, this is not to be understood as limiting the invention. It is noted that the CD-SEM may include, additionally or alternatively, at least one out-of lens detector.
Detection signals are processed by a processing unit (that may be a part of controller 33, but this is not necessarily so) that may have image processing capabilities and is able to process the detection signals in various manners. A typical processing scheme includes generating a waveform that reflects the amplitude of the detection signal versus the scan direction. The waveform is further processed to determine locations of at least one edge, and other cross sectional features of inspected structural elements.
The different parts of the system are connected to corresponding supply units (such as high voltage supply unit 21) that are controlled by various control units, most of them are omitted from the figure for simplifying the explanation. The control units may determine the current supplied to a certain part, as well as the voltage.
CD-SEM 100 includes a double deflection system that includes deflection units 110 and 111. Thus, the beam tilt introduced in the first deflection unit 110, can be corrected for in the second deflection unit 111. Due to this double deflection system, the electron beam can be shifted in one direction without introducing a beam tilt of the electron beam with respect to the optical axis.
b is a perspective view of an objective lens 120 according to another embodiment of the invention. In
Modern CD-SEMs are able to measure structural elements that have cross sections that have sub-micron dimensions, with an accuracy of several nanometers. The size of these cross sections is expected to reduce in the future, as manufacturing and inspection processes continue to improve.
Various features of the cross section may be of interest. These feature may include, for example: the shape of the cross section, the shape of one or more sections of the cross section, the width and/or height and/or angular orientation of the cross section sections, as well as the relationship between cross section sections. The feature can reflect typical values, as well as maximal and/or minimal values. Typically the width of the bottom of a line is of interest, but this is not necessarily so and other features may be of interest.
a illustrates a perspective as well as a cross sectional view of line 210 and a reference structural element, such as bump 250. Line 210 has a cross section 230 that includes a top section 234 and two substantially opposing traverse sections 222 and 226 (that correspond to the top section 214 as well as to two sidewalls 212 and 216 of line 210) that are both positively oriented at substantially opposing angles, such that the bottom of the line is not obscured by the top section 214.
Bump 250 is much smaller that line 210 and has at least one point, such as certain point 251, that is located substantially at the same plane as the lower points 233 and 237 of first and second traverse sections 232 and 236. Bump 250 is located such as to be viewed by electron beams that are used to scan the first traverse section as well as electron beams used to scan the second traverse section. The location of bump 250, and especially the relationship between its height (which is usually much smaller than the height of line 210) and the distance between bump 250 and line 210 are calculated to allow the bump 250 to be viewed by both type of electron beams. The distance can be responsive to the maximal tilt angle of the scanning electron beam. As a rule of thumb the distance between bump 250 and line 210 has to be greater than Hmax×Tangents (Amax) whereas Hmax being a maximal height of line 210 and Amax being a maximal tilt angle of a scanning electron beam.
The inventors found out that a bump can be generated relatively quickly using a charged electron beam that interacts with an object to provide a bump based upon the carbonization phenomena. For example—the inventors generated a 0.5 micron long carbonization line in less than a minute. Very good results were achieved by generating the bump by scanning the object multiple times (such as but not limited to 10-50 scans) with a relatively low current electron beam (such as but not limited to 20, 50, 100 pA and the like). The inventors also found that such bumps can be achieved by using various acceleration voltages, whereas using 200 v provided better results than using 500 v or 1000 v. Longer bumps usually required to increase the amount of scans. It is noted that inaccuracies of the bumps can reflect the accuracy of the critical measurements. The inventors found that the added error is less than 1 nano-meter, and that using several bumps and averaging the results for all bumps may improve the precision of the measurement
b illustrates a cross section 230′ of another line 210′ that has a top section 234, a first traverse section 232 that is positively oriented and a negatively oriented second traverse section 236.
In some cases, a line may be positioned between first and second reference structural elements (e.g., two bumps similar to bump 250). The distance between the reference structural elements may be measured by performing multiple scans of said reference structural elements.
The inventors found that using reference elements that are positioned at opposing sides of a measured structural element may be useful in various cases, such when a single structural element cannot be viewed by electron beams that scan the measured structural element from both its sides. This can occur other structural elements are positioned in a very close proximity to the measured structural element, such that to obscure the reference structural element.
Thus, in a dense array of lines 210, reference structural elements (e.g., bump 250) may be positioned between pairs of lines 210.
In other embodiments of the invention multiple reference multiple structural elements are positioned in proximity to a measured structural element, such as to allow multiple measurements of the relationship between the measured structural element and the multiple reference structural elements and to allow statistical processing of the multiple relationships to provide better results. Said statistical process may average out or otherwise reduce errors and/or inaccuracies associated with a single measurement. This may result from an increment of the overall signal to noise ratio. In some cases an array of structural elements (e.g., similar to bump 250) may be positioned at a first side of a line (e.g., similar to line 210), while in other cases an array of reference structural elements may be positioned on both sides of a measured structural element. The bottom CD of a line can be responsive to multiple distances measured with respect to these structural elements.
It is noted that although the above discussion refers to line shaped measured structural elements and reference structural elements, the method and system are applicable to determine cross sectional features (such as top CD, bottom CD, maximal CD, and the like) of various structural elements, such as contacts, recesses and the like. The reference structural elements may have other shapes.
a-3c are schematic illustration of waveforms 250-252 that represent a relatively wide positively oriented traverse section, a relatively narrow traverse section and a negative oriented traverse section. As can seen from these figures the waveform portion that is associated with steep sidewalls, as well as negative oriented sidewalls is relatively narrow and corresponds to the width of the scanning electron beam.
Referring to
Step 470 is followed by step 472 of calculating at least three points—a first point that is characterized by a maximal derivative value, a lower point and an upper point that are characterized by a predefined derivative values.
Step 472 is followed by step 474 of determining a location point in response to an intersection between a height threshold and a line that is drawn between the upper and lower points.
According to an embodiment of the invention the relationship between a reference element and a traverse section is actually the relationship between a certain point of the reference structural element and a location point of that traverse section. The certain point of the reference element may also be a location point.
It is noted that other points, such as upper point, intermediate point, lower point and/or other points of waveform 410 may be used in determining a relationship between a traverse section and a reference structural element.
Method 500 measures a feature of a structural element. The measured structural element is a part of a measured object, such as a wafer, die or dice. If appropriate reference structural elements are not positioned at the vicinity of the measured element they may be added by use of prior art methods for adding and/or removing material, such as Focused Ion Beam (FIB) based methods, Electron beam or even laser based methods. These methods may also require the provision/injection of gas/plasma in manner known in the art. A system that implements some of these techniques is the SEMVision G2 FIB of Applied Materials Inc. of Santa Clara, Calif.
Method 500 starts at step 520 of scanning, at a first tilt state, a first portion of a reference structural element and at least the first traverse section of the measured structural element, to determine a first relationship between the reference structural element and the first traverse section.
Conveniently, the first relationship is a distance between a certain point of the reference structural element and a first edge of the first traverse section. As illustrates by
Conveniently, the bottom of the measured structural element and the certain point of the reference structural element are substantially located on the same plane. Usually the height of the reference structural element, and especially the height of a certain reference point, is much smaller than a height of the measured structural element.
It is noted that the invention can be adapted to handle other shaped and/or sized structural elements. The height of the reference structural element may be taken into account, for example by using elementary geometrical equations, for compensating for a height difference between the bottom of the measured structural element and the reference structural element.
Step 520 is followed by step 530 of scanning, at a second tilt state, a second portion of a reference structural element and at least the second traverse section of the measured structural element, to determine a second relationship between the reference structural element and the second traverse section.
Step 530 is followed by step 540 of determining a cross sectional feature of the measured structural element in response to the first and second relationships.
According to another embodiment of the invention, if one or more additional relationships were determined during steps 520 and 530 then the determination of step 540 is further responsive to at least one of the additional relationships.
It is noted that during a tilt stage the electron beam may be tilted in relation to an imaginary plane that is perpendicular to the measured object. The tilt angle can be positive or negative. According to an embodiment of the invention during the first tilt state the electron beam is tilted at a positive angle while during the second tilt state the electron beam is tilted at a negative angle.
According to another embodiment of the invention during the first tilt state the electron beam is tilted at a negative angle while during the second tilt state the electron beam is tilted at a positive angle.
According to an embodiment of the invention during the first or second tilt state the electron beam is tilted at substantially zero degrees.
It is noted that large tilt angles, such as 15 degrees and even more, can be achieved by using prior art tools such as the NanoSem 3D or the VERASem of Applied Materials Inc.
Method 600 starts by step 620 of scanning, at a first tilt state, a first portion of a reference element and at least the first traverse section of the measured structural element, to determine a first relationship between the reference structural element and the first traverse section.
Step 620 is followed by query step 625 of determining whether to perform an additional scanning. If the answer is yes method 600 proceeds to step 630, else the method proceeds to step 640.
The determination of step 625 may be responsive to various parameters such as but not limited to top, bottom, maximal and/or minimal width of the measured structural element, estimated or measured orientation of a traverse section and/or estimated or measured width of a traverse section. The estimation can involve processing of detection signals, comparison with detection signals originating from other structural elements. The comparison may involve die to die comparison, die to data base comparison and the like.
An additional scanning step may be required if one traverse section (or both) is (are) suspected to be either steep (for example above a steepness threshold) or to be negatively oriented. This orientation may be estimated from waveforms acquired during the detection of signals resulting from the scan of the measured structural element. Steep traverse sections, as well as negative oriented traverse sections are associated with certain waveforms. The inventors found that a traverse section is suspected if the width of it substantially equals the width of the electron beam. The inventors also found the a comparison of waveforms to previously recorded waveforms of known elements may also be used.
Step 630 includes scanning, at a second tilt state, a second portion of a reference element and at least the second traverse section of the measured structural element, to determine a second relationship between the reference structural element and the second traverse section.
Step 630 is followed by step 640 of determining a cross sectional feature of the measured structural element in response to at least the first relationship.
According to another embodiment of the invention, if one or more additional relationships were determined during steps 620 and 630 then the determination of step 640 is further responsive to at least one of the additional relationships.
Method 700 starts at step 720 of scanning, at a first tilt state, at least a point of a first reference structural element and at least the first traverse section of the measured structural element, to determine a first relationship between the first reference structural element and the first traverse section.
Step 720 is followed by step 730 of scanning, at a second tilt state, at least a point of a second reference structural element and at least the second traverse section of the measured structural element, to determine a second relationship between the second reference structural element and the second traverse section.
Step 730 is followed by step 740 of determining a cross sectional feature of the measured structural element in response to the first and second relationships.
According to an embodiment of the invention the measured structural element is positioned between the first and second reference structural elements.
The relationship (usually distance and optionally height difference) between the first and second reference element feature must be measured/estimated. Inaccuracies in this measurement may effect the measurement of the cross sectional feature.
Step 700 may include a preliminary step of either estimating or receiving this reference relationship, but it may also include a step 710 of measuring it. The measurement may take place in locations other than the location in which the measured element is being measured, so that various contamination/carbonization. Shrinkage effects will not effect the measurement of the measured structural element
Step 710 may include one or more scan of an area that includes the first and second points of the first and second reference elements and the measured structural element. Step 710 may also include preventing the electron beam to illuminate the measured structural element. This can be implemented by blanking the beam while it is supposed to illuminate the measured structural element.
Method 800 starts at step 820 of scanning, at a first tilt state, a portion of a reference element and at least the first and second traverse sections, to determine at least one relationship between the reference element and the at least one scanned traverse element. During this step the electron beam can be substantially perpendicular to the measured object that includes the measured structural element
Step 820 is followed by query step 825 for determining whether an additional scanning step is required for determining an additional relationship between the reference structural element and at least one traverse section. If the answer is positive method 820 proceeds to step 830, else it proceeds to step 840. This determination may be responsive to an estimated width of a traverse section, to an estimated orientation of a traverse section, to an estimated cross sectional feature of the measured structural element, and the like.
Step 830 includes performing at least one additional scanning step. Conveniently, the tilt state differs from the first tilt state of step 820.
Step 840 involves determining a cross sectional feature of the measured structural element in response to at least one relationship.
Method 900 starts by step 920 of scanning, at a first tilt state, first portions of a set of reference structural elements and at least the first traverse section of the measured structural element, to determine a first set of relationships between reference structural elements of the set of reference structural elements and the first traverse section.
Step 920 is followed by step 930 of scanning, at a second tilt state, second portion of the set of reference structural elements and at least the second traverse section of the measured structural element, to determine a second set of relationships between reference structural elements of the set of reference structural elements and the second traverse section.
Step 930 is followed by step 940 of determining a cross sectional feature of the measured structural element in response to the first and second sets of relationships. Step 940 may include statistical processing of the relationships of the first set to provide a first relationship. It may also include statistical processing of the relationships of the second set to provide a second relationship.
According to another embodiment of the invention, method 900 may include a query step for determining a necessity of additional scanning steps, in addition to step 920, and to determine whether step 930 (or even additional scanning steps) is required.
Is noted that the intermediate section, which may be a top section in the case of an elevated structural element, may be determined from each of the scanning steps. It is further noted that given the first and second traverse section cross sectional features the cross section of the structural element as well as any feature (such as but not limited to top CD, bottom CD, maximal CD) of said cross section can be determined. A typical cross sectional feature is the horizontal projection of a traverse section. In cases where the tilt angle is relatively small it is assumed that the tilt angle is approximately equal to the Tangents of this angle.
It is noted that some of the measurements may be repeated, and that additional tilted scans of the structural element (with the same and/or differing tilt angles) may be performed for various reasons, such as averaging out statistical noise, and the like. Accordingly methods 400 and 500 may include multiple measurements of one or more cross sectional features, even if the height of the structural elements in known or estimated and even if a certain cross sectional feature was measured.
The present invention can be practiced by employing conventional tools, methodology and components. Accordingly, the details of such tools, component and methodology are not set forth herein in detail. In the previous descriptions, numerous specific details are set forth, such as shapes of cross sections of typical lines, amount of deflection units, etc., in order to provide a thorough understanding of the present invention. However, it should be recognized that the present invention might be practiced without resorting to the details specifically set forth.
Only exemplary embodiments of the present invention and but a few examples of its versatility are shown and described in the present disclosure. It is to be understood that the present invention is capable of use in various other combinations and environments and is capable of changes or modifications within the scope of the inventive concept as expressed herein.
This application claims the priority of U.S provisional application 60/486,566, filed 11 Jul. 2003.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/US04/22347 | 7/12/2004 | WO | 10/26/2006 |
Number | Date | Country | |
---|---|---|---|
60486566 | Jul 2003 | US |