Semiconductor structure measurement method and system based on derivative measurement signals

Abstract

A method and system for measuring structural parameters characterizing a measurement target are presented herein, based on changes in the measurement signal value and estimated changes in the electrical properties, optical properties, or both of the measurement target, which result from perturbations of these properties. The electrical and optical properties of the measurement target are perturbed by inducing a change in the electric field at the measurement target under measurement. In preferred embodiments, the change in the electric field is induced by directing a modulated beam of illumination light at the measurement target under measurement. Both the change in the measurement signal value and the estimated changes in the electrical properties, optical properties, or both of the measurement target are quantified and supplied as input to the measurement model. In this manner, the measurement is based on the derivative of the measurement signal with respect to the electrical properties, optical properties, or both.

Description

[Technical Field]

[0001] [Cross-references to related applications] This patent application claims priority under Section 119 of the United States Patent Act based on U.S. Provisional Patent Application No. 63 / 420087, filed on October 28, 2022, entitled "Apparatus and Method for Acquiring the Mueller Matrix Derivatives and Modulated Ellipsometry Spectra from a Semiconductor Specimen," and incorporates the entire subject matter of that application into this application by reference.

[0002] The embodiments described relate to weighing systems and methods, and more specifically to methods and systems with improved measurement accuracy. [Background technology]

[0003] When manufacturing semiconductor devices, such as logic and memory devices, a series of processing steps are typically applied to the sample. These processing steps form various features and multiple layers of these semiconductor devices. For example, in the semiconductor manufacturing process known as lithography, patterns are generated on the semiconductor wafer. Additional examples of semiconductor manufacturing processes, though not limited to these, include chemical mechanical polishing, etching, deposition, and ion implantation. It is preferable to create multiple semiconductor devices on a single semiconductor wafer and then separate them into individual semiconductor devices.

[0004] Quantitative weighing processes are used in various stages of semiconductor manufacturing to detect defects on wafers and improve yield. Optical and X-ray weighing techniques offer the potential for high throughput without the risk of sample damage. Numerous techniques, such as scatterometry, ellipsometry, and reflectometry equipment and associated analytical algorithms, are widely used to elucidate the limiting dimensions, film thickness, composition, overlay, and other parameters of nanoscale structures.

[0005] As devices (e.g., logic and memory devices) move towards finer nanometer-scale dimensions, characterization becomes increasingly difficult. The incorporation of complex three-dimensional geometries and materials with diverse physical properties further exacerbates this difficulty. In some cases, the value of semiconductor devices is increasingly determined not only by speed but also by their energy efficiency. For example, energy-efficient consumer products are more valuable because they operate at lower temperatures for longer periods on fixed battery power. Another example is the demand for energy-efficient data servers to reduce their operating costs. Ultimately, there is a strong interest in reducing the energy consumption of semiconductor devices. Solutions include the use of high-k material layers and complex geometric structures, both of which contribute to the difficulty of characterization.

[0006] Modern semiconductor processes are employed to create complex structures. Complex measurement models with multiple parameters are needed to represent these structures and account for process and dimensional variations. In complex multi-parameter models, modeling errors are induced by parameter correlations and low measurement sensitivity for certain parameters. In addition, regression of complex multi-parameter models with a relatively large number of floating parameter values ​​is not easily handled from an information processing perspective.

[0007] In certain cases, multiple parameters are typically fixed in model-based measurements to reduce the impact of these error sources and to lessen the information processing effort. Fixing the values ​​of multiple parameters can improve computation speed and reduce the impact of parameter correlations, but it also leads to errors in parameter value estimation.

[0008] In some other examples, when performing a measurement, the local environment surrounding the metering target is treated with a purge gas flow containing a controlled amount of filler. A portion of this filler aggregates onto the metering structure, filling openings within structural features, openings between structural features, etc. The presence of the filler alters the optical properties of the metering structure compared to a measurement scenario without any filler in the purge gas. Model-based measurement is performed using an enhanced dataset containing measurement signals collected from a metering target having geometric features filled with filler. This reduces parameter correlations between floating measurement parameters and improves measurement accuracy. This reduces the information processing effort required to obtain model-based measurement results. Further details are described in Patent Document 1, acquired by KLA-Tencor Corporation in Milpitas, California, USA, and its entirety is incorporated into this application by reference. Unfortunately, applying fillers to wafers introduces several problems, including contamination of the wafer itself, contrast limitations induced by the fillers, a lack of flexibility in filler selection, increased system complexity, and increased risks due to contact with the wafer surface.

[0009] In other measurement examples, various forms of modulated spectroscopy, such as optically modulated reflectance and field reflectance spectroscopy, induce periodic changes in the electric field of the sample under test. This modulation of the electric field effectively causes modulation of the dielectric function of the sample material at the same frequency. The measured signal is usually expressed as the reflectance change ΔR divided by the normalized reflectance R. This measured signal ΔR / R reveals various characteristics related to electronic transitions in the sample material. In some cases, the measured signal ΔR / R becomes highly sensitive to the band structure of the sample material.

[0010] In some existing systems, reflectometry measurements are performed while modulating the intensity of a pump (excitation) beam supplied to the measurement site. Measurement of light reflected or scattered from the specimen in response to the probe beam is performed while the measurement site is illuminated by the pump beam. In some examples, the pump beam and the probe beam are the same beam. The modulated pump beam induces a change in the electric field at the specimen, and consequently modulates the reflectivity of the specimen under measurement. In some of these examples, the value of the parameter of interest, such as the band gap, is determined by directly examining the line shape of the measured reflectivity change, such as ΔR / R. In other examples, the value of the parameter of interest is determined based on the measured reflectivity change, such as ΔR / R, using a measurement model. [Prior art documents] [Patent Documents]

[0011] [Patent Document 1] U.S. Patent No. 10145674 [Patent Document 2] U.S. Patent Application Publication No. 2018 / 0059019 [Overview of the Initiative] [Problems that the invention aims to solve]

[0012] In existing systems, the changes induced in the internal optical properties of the measurement target by the modulated pump beam are not quantified and are not supplied as input to the measurement model. Rather, the measurement results are derived solely from observed changes in optical properties, such as reflectance. This method limits the range of parameters of interest that can be measured based on modulated reflectance data.

[0013] Currently, solving complex multi-parameter measurement models often requires unsatisfactory compromises. Current model reduction techniques frequently fail to arrive at measurement models that are both informationally manageable and sufficiently accurate. Furthermore, the complexity of multi-parameter models makes optimizing system parameter selection (e.g., wavelength, incidence angle, etc.) for each parameter of interest difficult or impossible.

[0014] Future metrology applications will face challenges arising from further refinement of resolution requirements, multi-parameter correlations, increasing geometric complexity, and the growing use of opaque materials. Therefore, developing high-throughput systems and methods for characterizing complex semiconductor structures, such as those incorporating high-k dielectric layers, would be advantageous. In particular, developing robust, reliable, and stable methods for in-line metrology of gate stacks (laminated structures) containing high-k dielectrics would be beneficial. Thus, improved methods and systems for semiconductor structure measurement are desired. [Means for solving the problem]

[0015] This invention presents a method and system for measuring structural parameters characterizing a measurement target, based on changes in the measured signal value and estimated changes in the electrical properties, optical properties, or both of the measurement target, which are caused by perturbations of those properties. The electrical and optical properties of the measurement target are perturbed by inducing a change in the electric field at the measurement target under measurement.

[0016] In preferred embodiments, the electric field change is induced by directing a modulated beam of illumination light onto the measurement target under measurement. Perturbations to the electrical and optical properties of the measurement target induce a change in the measurement signal value. Both the change in the measurement signal value and the estimated changes in the electrical properties, optical properties, or both of the measurement target are quantified and supplied as input to the measurement model. In this way, the measurement becomes based on the derivative of the measurement signal with respect to the electrical properties, optical properties, or both. In certain examples, measurements based on these derivative quantities make it possible to increase sensitivity to film and CD parameters while reducing the correlation between parameters characterizing the various materials present in the structure under measurement.

[0017] The methods and systems described herein are applicable to a wide range of non-contact, non-destructive measurement systems, such as optical, electronic, and X-ray measurement systems that operate in various signal modes, such as reflectometry, ellipsometry, scatterometry, pupil imaging, field imaging, and hyperspectral imaging.

[0018] In preferred embodiments, a modulated illumination beam is generated by a tunable laser-based illumination source, directing a selected beam energy towards the measurement target. In some embodiments, the beam energy of the modulated illumination beam is selected to break the correlation between the various materials present in the measurement-underlying structure. In certain examples, the various material layers have very low optical contrast but remarkable electrical properties, such as band gap contrast. In these examples, the photon energy of the modulated illumination beam is selected to lie between the band gaps of two different material layers present in the multilayer stack.

[0019] Generally, the illumination power of the light incident on the measurement spot can be varied between different illumination power levels in some periodic or aperiodic manner. In certain examples, the illumination light can be varied in a binary manner, such as on / off, following a sine wave between different illumination power levels, a square wave between different illumination power levels, and so on.

[0020] In certain examples, changes in the electrical properties, optical properties, or both of the measurement target are estimated based on the separate measurements of the monolayer film specimens. The electrical properties, optical properties, or both of each monolayer film specimen are measured without pump illumination light and with pump illumination light at a specified beam energy and beam power density level. The differences in the measured properties are changes induced in the electrical properties, optical properties, or both, and are related to the specified beam energy and beam power density level. In certain embodiments, the monolayer film specimens are located on the same wafer as the measurement target. In certain other embodiments, the monolayer film specimens are located on other wafers.

[0021] In the methods and systems described in this application, it is possible to improve the measurement of structural elements common to semiconductor manufacturing, such as material composition measurement, alloy composition measurement of compound semiconductors, material bandgap measurement, elucidation of the characteristics of semiconductor surfaces and interfaces, film layer property measurement, critical dimension measurement, etc. Included in the measurement applications are the measurement of structural elements with complex semiconductor structures such as 3D-VNAND structures and gate all around (GAA) structures, including front-end-of-line (FEOL) layers from oxide definition layers to high-k metal gate (HKMG) stacks. Included in the measurement applications are the measurement of structural elements comprising semiconductor materials, insulating dielectric materials, and conductive materials, including organic materials, inorganic materials, or combinations thereof.

[0022] Since the above is an overview, simplifications, generalizations, and detail omissions are included throughout; accordingly, as will be apparent to those skilled in the art (so-called persons skilled in the art), this overview is solely illustrative and in no way limiting. Other aspects, novel features, and advantages of the devices, apparatuses, and / or processes described in this application will become apparent in the non-limiting detailed description set forth herein.

Brief Description of the Drawings

[0023] [Figure 1]This diagram shows a silicon nitride specimen 10, which is illuminated by a beam of illumination light 11 having a photon energy greater than the band gap of the silicon nitride material. [Figure 2] This diagram depicts a multilayer material stack arranged on a silicon substrate, illuminated by a beam of illumination light having a photon energy greater than the bandgap of one stack material and less than the bandgap of another stack material. [Figure 3] This diagram illustrates a spectroscopic ellipsometry system, which modulates the electrical and optical properties of a measurement target and measures the structural parameters that characterize the target based on changes in the measured signal value and estimated changes in the optical properties of the target. [Figure 4] This figure shows a derivative-based measurement engine 160 in one embodiment. [Figure 5] This diagram depicts an illumination beam incident on a wafer in a specific orientation, described by the incident angle θ and the azimuth angle φ. [Figure 6] This flowchart illustrates a method 200 for measuring structural parameters that characterize a measurement target, and is based on estimated changes in the optical properties of the measurement target and changes in the measured signal value, as described in this application. [Modes for carrying out the invention]

[0024] The following provides background examples and several embodiments of the present invention, with detailed reference to those examples depicted in the accompanying drawings.

[0025] This invention presents a method and system for modulating the electrical and optical properties of a measurement target and measuring structural parameters that characterize the measurement target based on changes in the measurement signal value and estimated changes in the electrical properties, optical properties, or both of the measurement target.

[0026] The electrical and optical properties of the measurement target are perturbed by inducing a change in the electric field at the measurement target under measurement conditions. In preferred embodiments, this change in the electric field is induced by directing a modulated beam of illumination light at the measurement target under measurement conditions. The perturbation of the electrical and optical properties of the measurement target induces a change in the measurement signal value. Both the change in the measurement signal value and the change in the electrical properties, optical properties, or both are quantified and supplied as input to the measurement model. In this manner, the measurement is based on the derivative of the measurement signal with respect to electrical properties such as electron density, or optical properties such as optical dispersion, or both. In certain examples, measurements based on these derivative quantities can reduce the correlation between parameters characterizing the various materials present in the structure under measurement, while also improving sensitivity to film and CD parameters.

[0027] The measurement model estimates the values ​​of one or more parameters of interest that characterize one or more structural elements of the measurement target, based on both changes in the measurement signal value and changes in electrical properties, optical properties, or both. In this manner, the values ​​of one or more parameters of interest can be estimated by operating the measurement model on derivative information, i.e., changes in the measurement signal as a function of changes in electrical properties, optical properties, or both.

[0028] The methods and systems described herein are applicable to a wide range of non-contact, non-destructive measurement systems, such as optical, electronic, and X-ray measurement systems, that operate in various signal modes, such as reflectometry, ellipsometry, scatterometry, pupil imaging, field imaging, and hyperspectral imaging. By performing model-based measurements based on derivative information as described herein, correlations are broken, resulting in sensitivity to structural parameters that would otherwise not be accessible by non-contact, non-destructive measurement systems.

[0029] Generally, changes in both the electrical and optical properties of a semiconductor material are induced by incident illumination light. This incident illumination light generates electron-hole pairs when the photon energy of the light affecting the material is greater than the material's band gap. These electron-hole pairs induce plasma currents, thermal effects, or both, which in turn alter the material's optical properties, such as refractive index n and absorption coefficient k. The degree to which the electrical and optical properties change due to illumination light depends on the photon energy and power density of the incident illumination light, as well as the material itself. In certain embodiments, measurements are performed using various wavelengths, power densities, or both, thereby further breaking the correlation and increasing the measurement sensitivity.

[0030] Figure 1 shows a silicon nitride specimen 10, which is illuminated by a beam of illumination light 11 having a photon energy greater than the band gap of the silicon nitride material. As shown in Figure 1, the illumination light 11 is absorbed by the silicon nitride 10, generating electron-hole pairs within the material. These electron-hole pairs change the electrical properties of the material. Furthermore, these electron-hole pairs induce plasma currents and thermal effects, which in turn change the optical properties of the material, such as the refractive index n and the absorption coefficient k. Equations (1) and (2) represent the refractive index and absorption coefficient of the silicon nitride material 10 under the illumination light 11; in the equations, n0 is the normalized value of the refractive index when the illumination light 11 is not present, and Δn is the change in refractive index value due to the illumination light 11. Similarly, k0 is the normalized value of the absorption coefficient when the illumination light 11 is not present, and Δk is the change in absorption coefficient value due to the illumination light 11. n(λ) = n0(λ) + Δn(λ) (1) k(λ) = k0(λ) + Δk(λ) (2)

[0031] In certain cases, the change in the refractive index value becomes significant, and two different measurement signals are obtained by performing measurements with and without illumination light 11. In such cases, the difference between these two different measurement signals yields derivative information, that is, a change in the measurement signal that is a function of the change in optical properties.

[0032] The electrical and optical properties of a semiconductor material are modulated by modulating the power of the illumination light incident on the semiconductor material. This modulation of the illumination power can be binary, e.g., on / off, sinusoidal, or some other suitable temporal variation in illumination power. In certain embodiments, the modulation is achieved by electrically controlling the illumination power output of an illumination source. In other embodiments, the modulation is achieved by controlling the illumination power incident on the measured structure by mechanically blocking or deflecting the illumination light, for example, using a chopper.

[0033] In certain embodiments, a laser-based illumination source generates a modulated illumination beam directed toward the measurement target, thereby modulating the electrical and optical properties of the structure under measurement. In certain embodiments, the illumination beam is a single-wavelength beam. In other embodiments, the illumination beam is a broadband illumination beam.

[0034] In some embodiments, the beam energy, i.e., wavelength, of the modulated illumination beam is selected such that the correlation between the various materials present in the structure under measurement is broken. In preferred embodiments, a modulated illumination beam is generated by a tunable laser-based illumination source, directing the selected beam energy towards the measurement target.

[0035] In certain examples, modulation-based measurements are performed on multilayer stacks composed of various material layers. In some of these examples, the various material layers are chosen to have very low optical contrast but remarkable electrical contrast, such as in band gap. In these examples, the photon energy of the modulated illumination beam is selected so that it lies between the band gaps of two different material layers in the multilayer stack.

[0036] By modulating the pump beam with a photon energy greater than the band gap of a certain material, the pump light is absorbed by that specific material, and the measurement signal includes a modulation signal component originating from that material layer. Furthermore, by modulating the pump beam with a photon energy smaller than the band gap of another material in the multilayer stack, the pump light is not absorbed by that specific material, and the measurement signal does not include a modulation signal component originating from that material layer. In these embodiments, no significant optical contrast is provided by these different materials, but by selecting the photon energy of the pump beam using electronic properties, such as contrast in band gap, these different materials can be separated in the measurement signal. More specifically, the difference between the signal measured when the multilayer stack is illuminated by the pump light and the signal measured when the multilayer stack is not illuminated by the pump light causes the contribution of the non-absorbing material to be dropped, leaving a difference signal that depends solely on the absorbing layer. In this way, the non-absorbing material layer is effectively isolated from the measurement by this difference signal.

[0037] Figure 2 shows a multilayer material stack arranged on a silicon substrate, which is illuminated by a beam of illumination light having a photon energy greater than the bandgap of one of the stack materials and less than the bandgap of another of the stack materials.

[0038] In the multilayer material stack shown in Figure 2, alternating silicon oxide (SiO2) layers 22 and 24 and alternating silicon nitride (Si3N4) layers 21 and 23 are arranged on a silicon substrate 20. This type of thin film stack is generally called an oxide / nitride / oxide (ONO) film stack. In certain embodiments, each of the silicon nitride layer and silicon oxide layer is extremely thin, for example, each layer having a thickness of about 5 nanometers. The optical properties of SiO2 and Si3N4 are very similar, and as a result, when optical measurement tools are used, a strong correlation appears between the thickness, profile, and CD parameters that characterize the stacked structure created using both materials.

[0039] Although the optical contrast between the SiO2 layer and the Si3N4 layer is very low, the electrical contrast between these two materials is remarkable; that is, their electrical properties are quite different. For example, the band gap of SiO2 is approximately 9 eV, and the band gap of Si3N4 is approximately 5 eV. The multilayer material stack is illuminated by an illumination beam 25 having a photon energy greater than the band gap of the silicon nitride material and smaller than the band gap of the silicon oxide material, for example, with a photon energy of approximately 7 eV. In this example, the illumination light is absorbed exclusively by the Si3N4 layer. As a result, electron-hole pairs are generated in the Si3N4 layer, but not in the SiO2 layer. These electron-hole pairs induce plasma currents and thermal effects, which in turn change the optical properties of the Si3N4 layer, but not those of the SiO2 layer. In other words, the illumination beam 25 perturbs the optical properties of the Si3N4 layer, but not those of the SiO2 layer.

[0040] Furthermore, the difference between optical measurements performed without illumination light 11 and those performed with illumination light 11 shows virtually no contribution from the SiO2 material layer, while the Si3N4 material layer makes a significant contribution. In this example, the difference between these two different measurement signals yields derivative information, i.e., the change in the measurement signal as a function of the change in optical properties, and the derivative is near null for the SiO2 material layer and non-zero for the Si3N4 material layer. In general, a large number of derivative spectra can be generated by varying the pump beam wavelength.

[0041] Figure 3 shows a diagram of a spectroscopic ellipsometry system, which modulates the electrical and optical properties of a measurement target and measures the structural parameters that characterize the measurement target based on the estimated changes in the optical properties of the measurement target and the changes in the measured signal values.

[0042] Figure 3 shows an example 100 of a spectroscopic ellipsometer (SE) metering system that performs derivative SE measurement of one or more metering targets as described in this application. As shown in Figure 3, the metering system 100 has an SE subsystem 105 having an illumination source 110, the beam of SE illumination light 107 incident on the wafer 101 is generated by the illumination source 110. In certain embodiments, the illumination source 110 is a broadband illumination source that emits illumination light in ultraviolet, visible, and infrared spectra. In some embodiments, the illumination source 110 is a laser-sustained plasma (LSP) light source (also known as a laser-driven plasma light source). The pump laser of the LSP light source may be a continuous wave or a pulsed laser. The illumination source 110 may be a single light source or a combination of multiple broadband or discrete wavelength light sources. The light generated by the illumination source 110 includes or includes a continuous spectrum from ultraviolet to infrared (e.g., from vacuum ultraviolet to mid-infrared). In general, the illumination light source 110 can include any suitable light source, such as a supercontinuum (ultra-broadband) laser light source, an infrared helium-neon laser light source, an arc lamp, a globar® light source, and others.

[0043] In certain embodiments, the group of SE illumination light is broadband illumination light having a wavelength range of at least 500 nanometers. In one example, the broadband SE illumination light includes wavelengths less than 250 nanometers and wavelengths greater than 750 nanometers. Generally, the broadband SE illumination light includes wavelengths from 120 nanometers to 4200 nanometers. In certain embodiments, broadband illumination light including wavelengths greater than 4200 nanometers can be used. In certain embodiments, the illumination source 110 includes a deuterium light source that emits light at wavelengths ranging from 150 nanometers to 400 nanometers, an LSP light source that emits light at wavelengths ranging from 180 nanometers to 2500 nanometers, a hypercontinuum light source that emits light at wavelengths ranging from 400 nanometers to 4200 nanometers, a globar® light source that emits light at wavelengths ranging from 2000 nanometers to 20000 nanometers, and the like.

[0044] As shown in Figure 3, the SE subsystem 105 has an SE illumination subsystem configured to direct SE illumination light 107 to one or more structures formed on the wafer 101. As shown in the figure, this SE illumination subsystem has a light source 110, an illumination optical system 111A, one or more optical filters 111B, a polarizing member 112, an illumination field diaphragm 113, and an illumination pupil aperture diaphragm 114. As shown in Figure 3, the beam of SE illumination light 107 propagates from the illumination source 110 to the wafer 101, passing through the illumination optical system 111A, the optical filter(s) 111B, the polarizing member 112, the field diaphragm 113, and the aperture diaphragm 114. The SE illumination light 107 illuminates a portion of the wafer 101 over a certain measurement spot 108.

[0045] The illumination optical system 111A adjusts the brightness of the illumination light 107 to focus the SE illumination light 107 onto the measurement spot 108. One or more of the aforementioned optical filters 111B are used to control the light level, spectral output, or a combination thereof from the illumination subsystem. In certain examples, one or more multizone filters are used as the optical filters 111B. The polarizing member 112 generates a desired polarization state that will exit the illumination subsystem. In certain embodiments, this polarizing member may be a polarizer, a compensator, or both, and any suitable commercially available polarizing member may be included. The polarizing member may be fixed, rotatable to various fixed positions, or continuously rotatable. In the SE illumination subsystem shown in Figure 3, there is one polarizing member, but the SE illumination subsystem may have multiple polarizing members. The field diaphragm 113 controls the field of view (FOV) of the SE illumination subsystem and any suitable commercially available field diaphragm may be included. The aperture diaphragm 114 controls the numerical aperture (NA) of the SE illumination subsystem and can include any suitable commercially available aperture diaphragm. The SE illumination subsystem may have an illumination optical system 111A, optical filters (group) 111B, polarizing member 112, field diaphragm 113, and aperture diaphragm 114, of any type and arrangement, that are known in the field of spectroscopic ellipsometry.

[0046] The metering system 100 also includes a focusing optical subsystem configured to collect light generated by the interaction between one or more of the aforementioned structures and the incident SE illumination light 107. The beam of collected light 109 is collected from the measurement spot 108 by the focusing optical system 115. The collected light 109 passes through the focusing aperture diaphragm 116, polarizing element 117, and field diaphragm 118 of the focusing optical subsystem.

[0047] The focusing optical system 115 has some suitable optical element that focuses light from one or more structures formed on the wafer 101. The focusing aperture diaphragm 116 controls the NA of the focusing optical subsystem. The polarizing element 117 detects the desired polarization state described above. The polarizing element 117 is a polarizer or compensator. The polarizing element 117 can be fixed, rotatable to various fixed positions, or continuously rotatable. In the focusing subsystem shown in Figure 3, there is one polarizing element, but the focusing subsystem may have multiple polarizing elements. The focusing field diaphragm 118 controls the FOV of the focusing subsystem. The focusing subsystem collects light from the wafer 101 and directs that light through the focusing optical system 115, aperture diaphragm 116, and polarizing element 117 to focus it onto the focusing field diaphragm 118. In certain embodiments, the focusing field diaphragm 118 is used as a spectrometer slit for the spectrometer of the detection subsystem. However, the focusing field diaphragm 118 may be located at or near another spectrometer slit of the spectrometer of the detection subsystem. The focusing subsystem may have a focusing optical system 115, an aperture diaphragm 116, a polarizing element 117, and a field diaphragm 118, regardless of their type and arrangement, that are known in the field of spectroscopic ellipsometry.

[0048] As shown in Figure 3, the SE weighing system 100 has a pump illumination source 150 that generates pump illumination light 151 directed towards a measurement spot 108 on the wafer 101. The location of the pump illumination light 151 coincides with the SE illumination light 107 projected onto the surface of the wafer under measurement at the measurement spot 108. In this way, the optical properties of the structure measured by the SE illumination light 107 are modulated by the incident pump illumination light 151 at the measurement spot 108. As shown in Figure 3, a command signal 152 is sent to the pump illumination light source 150 shown in Figure 3. The command signal 152 incorporates various parameters necessary to characterize the desired pump illumination light 151. In a non-limiting example, the command signal 152 may incorporate the desired photon energy of the pump illumination light 151, the desired power density of the pump illumination light 151, the desired modulation frequency of the pump illumination light 151, the desired power waveform of the pump illumination light 151, etc. In response, the pump illumination light source 150 generates pump illumination light 151 according to the desired characteristics specified by the command signal 152.

[0049] Generally, the illumination power of the SE illumination light 151 incident on the measurement spot 108 can be varied between different illumination power levels in some periodic or aperiodic manner. In certain examples, the illumination light can be varied in a binary manner, for example, on / off, following a sinusoidal wave between different illumination power levels, following a square wave between different illumination power levels, and so on. In this manner, the reflectance, transmittance, or polarization of the structure being measured is alternated between the signal value without external perturbation of the optical properties and the signal value with perturbation of the optical properties of the structure being measured.

[0050] In the embodiment shown in Figure 3, a focusing optical subsystem directs light to the detector 119. The detector 119 generates an output in response to the light collected from one or more structures illuminated by the SE illumination subsystem at the measurement spot 108. One example of the detector 119 has a charge-coupled device (CCD) sensitive to ultraviolet and visible light (e.g., light with wavelengths from 190 nanometers to 860 nanometers). Another example of the detector 119 has a photodetector array (PDA) sensitive to infrared light (e.g., light with wavelengths from 950 nanometers to 2500 nanometers). However, generally, the detector 119 can have other detector technologies and arrays (e.g., position-sensing detectors (PSDs), infrared detectors, photovoltaic detectors, orthogonal cell detectors, cameras, etc.). Each detector converts the incident light into an electrical signal that indicates the spectral intensity of the incident light. In general, the detector 119 generates an SE measurement signal 103 that indicates the light detected on the detector 119.

[0051] Each directional component of the SE illumination beam 107 relative to the surface normal of the semiconductor wafer 101 can be described by any two arbitrary angular rotations of the wafer 101 relative to the illumination beam 107, and vice versa. For example, the orientation can be described relative to a coordinate system fixed to the wafer. The SE illumination beam 107 depicted in Figure 5 is incident on the wafer 101 in a specific orientation, and this specific orientation is described by the incident angle θ and the azimuth angle φ. The coordinate frame XYZ is fixed to the SE metering system (e.g., SE illumination beam 107), and the coordinate frame X'Y'Z' is fixed to the wafer 101. The Y axis is aligned in-plane with respect to the surface of the wafer 101. X and Z are not aligned with respect to the surface of the wafer 101. Z' is aligned with an axis perpendicular to the surface of the wafer 101, and X' and Y' are in a plane aligned with respect to the surface of the wafer 101. As depicted in Figure 5, the SE illumination beam 107 is aligned with the Z axis and therefore lies in the XZ plane. The incident angle θ describes the direction of the SE illumination beam 107 along the XZ plane with respect to the surface normal of the wafer. Furthermore, the azimuth angle φ describes the direction of the XZ plane with respect to the X'Z' plane. Together, θ and φ uniquely define the direction of the SE illumination beam 107 with respect to the surface of the wafer 101. In this example, the direction of the SE illumination beam with respect to the surface of the wafer 101 is described by rotation around an axis perpendicular to the surface of the wafer 101 (i.e., the Z' axis) and rotation around an axis aligned with the surface of the wafer 101 (i.e., the Y axis).

[0052] As shown in Figure 3, the SE weighing tool 100 has a sample positioning system 190, which is configured to align the sample 101 with respect to the illumination beam 107 and to orient the sample 101 over a wide range of incident angles and azimuth angles. In this way, the weighing system 100 collects measurement results for the sample 101 at any number of locations and orientations on the surface of the sample 101. In one example, the information processing system 130 sends a command signal (not shown) indicating the desired position of the sample 101 to the sample positioning system 190. In response, the sample positioning system 190 generates command signals for various actuators provided in the sample positioning system 190, thereby achieving the desired positioning of the sample 101.

[0053] In general, the sample positioning system 190 may have any combination of mechanical elements suitable for achieving desired linear and angular positioning performance, including, but is not limited to, a goniometer stage, a hexapod stage, an angular stage, and a linear stage.

[0054] Generally, an optical scatterometer, such as the SE weighing system 100, is configured to supply illumination light to the weighing target under measurement at some desirable angle of incidence and azimuth.

[0055] The optical properties of the structure illuminated by the pump illumination light 151 are modulated at the same frequency as the pump illumination light 151. In order to capture induced changes in the SE measurement signal, the spectrum must be collected quickly, i.e., at a frequency at least twice the modulation frequency of the pump illumination light, to avoid loss of signal information. Furthermore, the modulated SE measurement signal ΔSE is relatively smaller than the unmodulated SE measurement signal SE. That is, the SE measurement signal may be overwhelmed by optical and electrical noise. Fortunately, the modulated SE signal appears at a known frequency and can be detected using some suitable lock-in detection method. Lock-in detection allows a portion of the signal to be detected at its known modulation frequency or frequency group and to be discriminated from portions of the signal at other frequencies. Phase-sensing detection and lock-in amplification are common signal extraction techniques and can be employed to recover the modulated SE measurement signal from the detected measurement signal 103.

[0056] Extracting a modulated SE measurement signal using phase-sensing lock-in amplification requires at least one measurement at each wavelength over a time interval that includes both the moment the pump laser beam is present and the moment the pump laser beam is absent. Serial detection at each wavelength using a lock-in amplifier results in sequential measurements over the desired wavelength range, leading to longer measurement times.

[0057] In certain embodiments, signal multiplexing is employed to shorten the signal acquisition time by simultaneously measuring multiple wavelengths. During each modulation cycle, the detection system simultaneously measures each dispersed beam by multiplexing the signal readouts from these detectors. Furthermore, the detection system calculates the modulated SE signal from the signals of each wavelength.

[0058] The collected light is spatially dispersed on the detector array according to its wavelength. Light from individual narrow wavelength bands is incident on different pixels of that detector array. The optical signals detected in each narrow wavelength band are simultaneously measured by multiplexed readout of those pixels.

[0059] The detector's electronic circuit subsystem rapidly reads out modulated and unmodulated SE signals from each pixel of the detector array. Multiple measurements of the modulated signal are performed within a single period of the modulation cycle. In this way, the detector can simultaneously measure SE measurement signals in parallel across all desired wavelengths.

[0060] Generally, the detected SE measurement signal depends on the configuration of the SE subsystem. In certain embodiments, the detected SE measurement signal is an SE harmonic signal, such as {α,β}, {Ψ,Δ}, etc. In certain embodiments, the detected SE measurement signal is one or more elements of the Müller matrix representation of the SE measurement. Generally, any detected SE measurement signal that is sensitive to changes in optical properties caused by the modulated pump illumination light 151 is considered to be within the scope of this patent application.

[0061] The weighing system 100 also includes an information processing system 130 configured as a derivative-based measurement engine 160, the derivative-based measurement engine 160 being configured to estimate the values ​​of one or more parameters of interest that characterize the structure under measurement, based on changes in the measurement signal value and estimated changes in the optical properties of the structure, which are induced by modulation of the electrical and optical properties of the structure.

[0062] Figure 4 shows a derivative-based measurement engine 160 in one embodiment. As shown in Figure 4, the derivative-based measurement engine 160 has a modulation classifier module 161 and a derivative-based measurement module 163. Signal 165 indicates the characteristics of the pump illumination light 151 directed toward the measurement spot 108 and is sent to the modulation classifier 161. One example of signal 165 is a command signal 152 sent to the pump illumination light source 150 shown in Figure 3. In this example, signal 165 includes the desired photon energy of the pump illumination light 151 and the desired power density of the pump illumination light 151. Another example of signal 165 is generated by one or more optical sensors (not shown) arranged in the optical path between the pump illumination light source 150 and the wafer 101. In this example, signal 165 includes indication information for the photon energy of the pump illumination light 151 and indication information for the power density of the pump illumination light 151, which is measured by the optical sensors. In addition, signal 164 indicating the material present in the measured structure is sent to the modulation classifier module 161.

[0063] In response to signals 164 and 165, the modulation classifier module 161 generates a signal 162 indicating a change in the optical properties of the structure under measurement induced by the pump illumination light 151 incident on the structure under measurement, for example, {Δn, Δk}. In certain examples, the modulation classifier module 161 characterizes the changes in the optical properties of various semiconductor manufacturing materials induced by illumination at various photon energy and beam power density levels using a library database. In other examples, the modulation classifier module 161 characterizes the changes in the optical properties of various semiconductor manufacturing materials induced by illumination at various photon energy and beam power density levels using a model.

[0064] Changes in the optical properties of semiconductor materials are expected to depend on photon energy and beam power density, but are independent of geometry. In certain examples, single-layer samples, each having a separate layer of material, are prepared. Their optical properties, for example, {n,k}, are measured using conventional SE measurement while illuminating them with pump illumination light at multiple different photon energy levels and beam power density levels. The measured optical properties are compared with the measured values ​​of the optical properties without pump illumination. The difference between the optical properties measured at a specified pump illumination level and the optical properties measured without pump illumination is the induced change in the optical properties at each of those specified pump illumination levels, for example, {Δn,Δk}. In certain embodiments, these calibration values ​​are recorded and interpolated in a database so that the induced change in the optical properties can be identified for any specified pump illumination level. In such embodiments, the database is incorporated into the modulation classifier module 161. In other embodiments, the calibration values ​​are fitted to a model that identifies induced changes in optical properties for any specified pump illumination level. In such embodiments, the trained model is incorporated into the modulation classifier module 161.

[0065] In certain examples, one or more metering targets are arranged at different locations on the same semiconductor wafer as the structure under measurement. Each metering target is assumed to have a single film layer corresponding to each of the various materials present in its structure under measurement. Changes in the values ​​of one or more electrical or optical properties of each metering target, induced by fluctuations in the power of the pump illumination light, are measured. The pump illumination light power fluctuations for measuring each single film layer metering target are the same as those for measuring the structure under measurement. This allows the measured value changes for one or more electrical or optical properties of each metering target to be directly applied to model-based measurements of its structure under measurement.

[0066] As shown in Figure 4, a signal 162 indicating the induced change in the optical properties of the structure under measurement, for example, the values ​​of {Δn, Δk}, is sent to the derivative-based measurement module 162. In one example, the signal 162 contains two vectors. One vector represents the induced change Δn in the refractive index as a function of wavelength, and the other vector represents the induced change Δk in the absorption coefficient as a function of wavelength. In addition, an SE measurement signal 103 is sent to the derivative-based measurement module 163. These SE measurement signals are perturbations of the optical properties of the structure under measurement and fluctuate due to perturbations induced by the modulated pump illumination light 151. In certain examples, these SE measurement signals contain one or more elements of a Müller matrix. In such examples, the change in the value of one or more elements of the Müller matrix, and the change induced by its modulation, is supplied as input to the measurement model. In certain examples, these SE measurement signals become harmonic signals generated by the spectrometer. In such examples, the changes in the values ​​of these harmonic signals, and the changes induced by their modulation, are supplied as input to the measurement model.

[0067] Within the derivative-based measurement module 163, the derivative-based model estimates the values ​​of one or more parameters of interest that characterize the measured structure, based on changes in the SE measurement signal and induced changes in the optical properties of the measured structure, such as the values ​​of {Δn,Δk}.

[0068] In certain embodiments, the derivative-based measurement model is a trained machine learning (ML)-based model. The ML-based model is trained using design of experiments (DOE) datasets relating to DOE measurement targets with known parameters of interest. These DOE datasets include changes in the SE measurement signal and induced changes in the optical properties of the measured structure, such as the values ​​of {Δn, Δk}, related to the DOE measurement of those DOE measurement targets. In certain examples, the DOE measurement is a real SE measurement of a DOE measurement target with known parameters of interest. In some of these examples, these known parameters of interest are obtained by measurements using a highly reliable reference metrology system. In other examples, the DOE measurement is a simulated SE measurement of a simulated DOE measurement target with specified parameters of interest.

[0069] In certain embodiments, the derivative-based measurement model uses the values ​​of induced changes in the optical properties of the measured structure, e.g., {Δn, Δk}, and assumed values ​​of the parameter of interest to predict the normalized change of the SE measurement signal, e.g., (ΔSE / SE). * This is a physically based model associated with the system. The assumed values ​​of these parameters of interest are updated as part of the regression process until a sufficiently good fit is obtained between the predicted normalized change in the SE measurement signal and the actual normalized change in the SE measurement signal, for example (ΔSE / SE). Once a sufficiently good fit is obtained, the estimated values ​​of these parameters of interest are sent to memory, for example, memory 180.

[0070] The normalization change in the SE measurement signal 103 determines the change in any spectral ellipsometry signal, as expressed in equation (3); SE in the equation ON This is the SE measurement signal value when the power of the pump illumination light 151 is at its maximum, SE OFF This is the SE measurement signal value when the power of the pump illumination light 151 is at its minimum, for example, zero. ΔSE / SE=(SE ON -SE OFF ) / SE OFF (3)

[0071] As mentioned above, these SE signals include some suitable SE signals generated by the spectrometer of the SE measurement system, such as harmonic signals, one or more Müller matrix elements, etc.

[0072] The dielectric function ε of the material is characterized by a complex function having real coefficients ε1 and imaginary coefficients ε2, as expressed by equation (4). ε = ε1 + iε2 (4)

[0073] The predicted normalization change in the SE signal is a perturbation of the dielectric function, as expressed by equation (5), and is related to the perturbation induced by the pump illumination light; in the equation, a(ε1,ε2) and b(ε1,ε2) are two coefficients related to that dielectric function. (ΔSE / SE) * =a(ε1,ε2)Δε1+b(ε1,ε2)Δε2(5)

[0074] The coefficients a(ε1,ε2) and b(ε1,ε2) depend on both the material and geometry of the structure under optical modulation. However, the perturbations Δε1 and Δε2 of these dielectric function coefficients are related by the Kramers-Kronig relation and are independent of the geometry of the structure under measurement. For this reason, these perturbations can be determined, as described above, using the modulation classifier module 161, based on the known material and known properties of the pump illumination light 151.

[0075] Equation (5) relates the predicted normalization change in the SE signal to the perturbation of the dielectric function when the optical properties of a single material are perturbed by the pump illumination light 151. Equation (6) shows the relationship between the predicted normalization change in the SE signal and the perturbation of the dielectric function when the optical properties of d different materials are perturbed by the pump illumination light 151.

number

[0076] In the example represented by Equation (6), the structure under measurement has d different material layers, and the optical properties of these material layers are subjected to modulation by the pump illumination light 151. In one example, a Lorentz model of the dielectric function of a lattice structure having d material layers is assumed. In this example, Equation (6) can be rewritten and represented by Equation (7); where d is the number of separate modulation materials having a bandgap energy less than the modulation photon energy, A j is the dimensionless amplitude as a function of the photon energy of the pump illumination light and the material geometry such as CD and HT, Γ j is the broadening parameter expressed in electron volts, θ j is the line shape phase parameter expressed in radians, E j is the transition energy expressed in electron volts, and m is a parameter related to the dimension number of the transition. The value of m is 3 in the case of silicon.

Number

[0077] Equationation (7 Equation (7) is an approximate form of the derivative of the SE measurement signal under the assumption of the Lorentz model of the dielectric function.

[0078] The perturbation of the dielectric function can be expressed by the perturbation of the refractive index and the extinction coefficient as shown by Equation (8); where Δn is the change in the refractive index, Δk is the change in the extinction coefficient, and I is the intensity of the harmonic oscillator.

Number

[0079] Equation (8) can be rewritten and represented by Equations (9) to (12).

Number

Number

[0080] The values ​​of the parameters in equations (10) to (12) are known for many semiconductor materials. The unperturbed dielectric function can be expressed in a wide variety of forms, depending on the material, thickness, confinement effect, crystallinity, temperature, etc. The most common assumptions are the Lorentz assumption or the Gauss assumption. The Lorentz dielectric function is expressed by equation (13). ε = 1 + I / (EE g +iΓ) (13)

[0081] Under the assumed form expressed by equation (13), the modulation term in equation (12) can be expressed as shown by equations (14) to (15); however, y = (EE g Let ) / Γ be the case.

number

[0082] Assuming that intensity modulation is negligible, we can use equations (10) to (15) in (8) to arrive at the general form of the derivative spectrum under measurement.

[0083] In general, there are multiple models that describe dielectric functions. While the Lorentz model was mentioned earlier, generally speaking, all suitable models for dielectric functions are considered to fall within the scope of this patent application.

[0084] In general, the SE metering system described with reference to Figure 3 is not limited to this, but can be any form of spectroscopic ellipsometer, including compensator-rotating SE systems, polarizer-rotating SE systems, polarizer-rotating compensator-rotating SE systems, compensator-rotating compensator-rotating SE systems, etc. In addition, its derivative-based measurement technique can be applied to other ellipsometry systems that measure all or part of the sample Müller matrix using non-rotating solid-state devices, such as photoelastic modulators.

[0085] Generally, the electrical properties of a measurement target are perturbed by inducing a change in the electric field within the measurement target under measurement, which in turn changes the optical properties of the measurement target. As mentioned above, the change in the electric field is induced by directing a modulated beam of illumination light onto the measurement target under measurement. However, generally speaking, any technique suitable for changing the electric field is considered to be within the scope of this patent application. In certain embodiments, a wafer is placed between two electrodes, and a time-varying bias voltage is applied between these electrodes. This time-varying bias voltage induces modulation of the electric field in the wafer containing the measurement target. This electrical modulation induces modulation of the electrical and optical properties of the measurement target under measurement, which is then utilized for measurement purposes as described above.

[0086] The methods and systems described herein enable improvements in the measurement of structural elements common to semiconductor manufacturing, such as material composition measurement, alloy composition measurement of compound semiconductors, material bandgap measurement, characterization of semiconductor surfaces and interfaces, film layer characteristic measurement, and limit dimension measurement. Measurement applications include the measurement of structural elements with complex semiconductor structures, such as 3D-VNAND structures and gate-all-around (GAA) structures, including front-end-of-line (FEOL) layers from oxide fractionation layers to high-k metal gate (HKMG) stacks. Measurement applications also include the measurement of structural elements comprising semiconductor materials, insulating dielectric materials, and conductive materials, including organic materials, inorganic materials, or combinations thereof.

[0087] Generally, sensitivity can be improved and correlation reduced by selecting the modulation wavelength and power density of the pump illumination light for each measurement application.

[0088] As mentioned above, measurements of ONO thin-film stacks are improved by employing illumination pump light with a photon energy of more than 4.5 electron volts. In certain embodiments, measurements of ONO thin-film stacks are enhanced by employing a 244-nanometer solid-state continuous-wave laser as the pump illumination source.

[0089] In certain cases, silicon / silicon-germanium superlattice structures are measured using a pump illumination source that generates pump illumination light with photon energy less than 1.12 electron volts, i.e., the band gap of silicon.

[0090] In other examples, high-k / interface layer (HK / IL) gate stacks are measured using a pump illumination source that generates pump illumination light using the photon energy between the band gaps of hafnium oxide (HfO2) and silicon oxide (SiO2).

[0091] In general, the accuracy of measurements for complex semiconductor structures can be improved by combining various techniques for breaking the correlations between different contributing factors to the measured optical response described in this application. For example, by sequentially or in parallel analyzing the derivative-based measurement results of various structures at various wavelengths, incident angles, azimuth angles, or any combination thereof, structural features related to complex multilayer structures can be accurately decorrelated.

[0092] In a further embodiment, the wavelengths emitted by a measurement illumination source, such as illumination source 110, or a pump illumination source, such as illumination source 150, are selectable. In certain embodiments, illumination source 110 or illumination source 150 is used as an LSP light source, and by controlling it with an information processing system 130, the luminous flux in one or more selected spectral ranges is maximized. The plasma temperature and, consequently, the spectral range of the radiated light are controlled by the laser peak intensity in the target material. The laser peak intensity is varied by adjusting the pulse energy, pulse width, or both. As shown in Figure 3, the information processing system 130 sends a command signal 140 to illumination source 110, causing illumination source 110 to adjust the spectral range of various wavelengths emitted from illumination source 110. Similarly, the information processing system 130 sends a command signal 152 to illumination source 150, causing illumination source 150 to adjust the spectral range of various wavelengths emitted from illumination source 150. In one example, the illumination source 110 is an LSP light source, and by adjusting the pulse duration, pulse frequency, and target material composition of the LSP light source, a desired spectral range is achieved for the wavelengths emitted from the LSP light source. In another example, the illumination source 150 is a tunable laser-based light source, such as a tunable hypercontinuum laser light source, and by adjusting one or more operating parameters of the laser light source, a desired spectral range is achieved for the wavelengths emitted from the laser-based light source.

[0093] In certain cases, the derivative-based measurement engine 160 reads a file containing equations describing the shape and composition of the structure under measurement. In certain cases, this file is generated by a lithography simulator, such as PROLITH® software available from KLA Corporation in Milpitas, California, USA. Based on this application information, the derivative-based measurement engine automatically sets the parameter representation and constraints for the structural model.

[0094] Although the methods discussed in this application are described with reference to System 100, any optical or X-ray metering system configured to illuminate a sample and detect light scattered from the sample can be employed to carry out the exemplary methods described herein. Furthermore, any electronic metering system configured to illuminate a sample and detect scattered electrons from the sample can be employed to carry out the exemplary methods described herein. Examples of systems include angle-resolved reflectometers (i.e., beam profile reflectometers), angle-resolved ellipsometers (i.e., beam profile ellipsometers), scatterometers, spectroscopic reflectometers or ellipsometers, multiple illumination angle spectroscopic reflectometers or ellipsometers, Müller matrix spectroscopic ellipsometers (e.g., compensator rotation spectroscopic ellipsometers), single-wavelength ellipsometers, single-wavelength reflectometers, Lamans scatterometers, transmission small-angle X-ray scatterometers, reflection small-angle X-ray scatterometers, grazing-incident small-angle X-ray scatterometers, transmission electron microscopes, scanning electron microscopes, etc.

[0095] In a non-limiting example, an ellipsometer may be a single rotatable compensator, multiple rotatable compensators, a rotatable polarizer, a rotatable analyzer, a modulation element, or one having multiple modulation elements, or one without modulation elements.

[0096] It should be noted that the output from the source and / or target measurement system can be configured in a way that utilizes multiple technologies within that measurement system. In fact, the application can be configured to utilize any combination of available metrological subsystems, either within a single tool or across multiple separate tools.

[0097] A system for carrying out the methods described herein can be configured in a wide variety of ways. For example, a wide range of wavelengths (including visible, ultraviolet, and infrared), angle of incidence, polarization state, and coherence state may be taken into consideration. Alternatively, the system may have any of several separate light sources (e.g., directly coupled light source, laser-sustained plasma light source, etc.). Alternatively, the system may have elements (e.g., apodizer, filter, etc.) for adjusting the light directed towards or collected from a sample.

[0098] Generally, the optical dispersion characteristics of a semiconductor structure under measurement can be assumed to be isotropic. Under this assumption, the material parameters become scalar values. Alternatively, the optical dispersion characteristics of the semiconductor structure under measurement may be modeled more accurately and assumed to be anisotropic. Under this assumption, the material parameters become a matrix consisting of various values, rather than scalar values. Additional details regarding the handling of anisotropic structures under measurement are described in Patent Document 2, and its entirety will be incorporated into this application by reference.

[0099] Figure 6 illustrates a method 200 suitable for implementation by the weighing system 100 of the present invention. In some embodiments, it should be recognized that the data processing blocks of method 200 can be executed by having one or more processors of the information processing system 130 execute a pre-programmed algorithm. The following description is presented in the context of the weighing system 100, but in the view of the present application, the specific structural aspects of the weighing system 100 should be understood solely as illustrative and not as limiting.

[0100] In block 201, a structure created on a semiconductor wafer is illuminated during the measurement interval with a certain amount of measurement illumination light and a certain amount of pump illumination light. The power of the pump illumination light incident on the structure fluctuates over time during the measurement interval.

[0101] In block 202, the measurement signal related to the measurement of its structure is detected according to the amount of measurement illumination light and the amount of pump illumination light.

[0102] In block 203, changes in the values ​​of one or more electrical or optical properties of one or more types of materials present in its structure are estimated. These value changes are induced by power fluctuations of the pump illumination light.

[0103] In block 204, the change in the value of the detected measurement signal is estimated to be a change induced by power fluctuations of the pump illumination light.

[0104] In block 205, the values ​​of the parameters of interest that characterize the measured structure are estimated based on changes in the values ​​of one or more electrical or optical properties of one or more types of materials present in the structure, and changes in the values ​​of the detected measurement signals.

[0105] It should be recognized that the various steps described throughout this disclosure may be performed by a single computer system or by multiple computer systems. Furthermore, various subsystems, such as the spectroscopic ellipsometer 105, may have computer systems suitable for performing at least some of the steps described herein. Accordingly, the foregoing statements should be taken as illustrative examples only, and not as limitations on the present invention. Furthermore, one or more of the aforementioned information processing systems 130 may be configured to perform any other steps(s) that are included in any of the method embodiments described herein.

[0106] The information processing system 130 may include, but is not limited to, personal computer systems, mainframe computer systems, workstations, image computers, parallel processors, and any other information processing devices known in the art. Generally, the term "information processing system" can be broadly defined to encompass all devices having one or more processors that execute instructions obtained from a storage medium. Generally, the information processing system 130 can be integrated with a measurement system, such as measurement system 100, or alternatively, it can be separated from any measurement system. In accordance with this concept, the information processing system 130 can be remotely deployed and receive measurement data and user input from any measurement source and user input source, respectively.

[0107] For example, program instructions 134 embodying those described in this application can be transmitted on or stored on a carrier medium. A carrier medium may be a transmission medium, such as a wire, cable, or wireless transmission link. A carrier medium may also be a computer-readable medium, such as read-only memory, random-access memory, magnetic or optical disk, or magnetic tape. For example, as shown in Figure 3, program instructions 134 stored in memory 132 are transmitted to the processor 131 on bus 133. The program instructions 134 are stored in a computer-readable medium (e.g., memory 132). Examples of computer-readable media include read-only memory, random-access memory, magnetic or optical disk, or magnetic tape.

[0108] The term "limit dimension" as used in this application includes all limit dimensions of a structure (e.g., lower limit dimension, middle limit dimension, upper limit dimension, side wall angle, grid height, trench depth, etc.), limit dimensions between any two or more structures (e.g., distance between two structures), displacement between two or more structures (e.g., overlay displacement between overlapping grid structures, etc.), and dispersion properties of materials used in the structure or a part of the structure. Structures that may be included include three-dimensional structures, patterned structures, overlay structures, etc.

[0109] The terms "limit dimension application" and "limit dimension measurement application" as used in this application encompass all forms of limit dimension measurement.

[0110] The term "measuring system" as used in this application encompasses all systems employed with the aim, at least partially, to characterize a sample in some manner. However, such technical terms do not limit the technical scope of the term "measuring system" as used in this application. A weighing system can consist of an LED inspection tool, an edge inspection tool, a back inspection tool, a macro inspection tool, or a multimode inspection tool (which acquires data from one or more platforms simultaneously), as well as any other weighing or inspection tool.

[0111] This application describes various embodiments of semiconductor processing systems (e.g., inspection systems and lithography systems) that can be used to process samples. The term "sample," as used in this application, refers to one or more sites on a wafer, a reticle, or any other specimen that can be processed (e.g., printed, measured, or inspected for defects) by means known in the art. In certain examples, the sample comprises a single site having one or more measurement targets, and the simultaneous combined measurement of these targets is treated as a single sample measurement or reference measurement. In other examples, the sample is a collection of sites, and the measurement data relating to these collective measurement sites is a statistical collection of data relating to each of those sites. Furthermore, each of those sites may comprise one or more measurement targets relating to the sample or reference measurement.

[0112] As used in this application, the term "wafer" generally refers to a substrate formed from a semiconductor or non-semiconductor material. Examples include, but are not limited to, single-crystal silicon, gallium arsenide, and indium phosphide. Such substrates are commonly found and / or can be processed in semiconductor manufacturing equipment. In some cases, a wafer may consist only of a substrate (i.e., a bare wafer). Alternatively, a wafer may have one or more layers of different materials formed on the substrate. One or more layers formed on a wafer may be "patterned" or "unpatterned." For example, a wafer may contain multiple dies that have repeatable pattern features.

[0113] A "reticle" can be a reticle at any stage of the reticle manufacturing process or a finished reticle, and may or may not be released for use in semiconductor manufacturing equipment. A reticle, or "mask," is generally defined as a nearly transparent substrate on which a nearly opaque region is formed, and this region forms a pattern. The substrate may contain a glass material, such as amorphous SiO2. By placing the reticle on a wafer covered with resist and performing the exposure step of the lithography process, the pattern on the reticle can be transferred to the resist.

[0114] One or more layers formed on a wafer may or may not be patterned. For example, a wafer may have multiple dies, each having repeatable pattern features. The formation and processing of such material layers ultimately yield a finished device. A wide variety of devices can be formed on a wafer, and the term "wafer" as used in this application is intended to encompass any wafer on which any device known in the art is fabricated.

[0115] In one or more exemplary embodiments, the functions described can be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, these functions can be stored or transmitted on a computer-readable medium as one or more instructions or codes. Computer-readable mediums include both computer storage media and communication media, encompassing all media useful for transferring computer programs from one location to another. Storage media can be any available medium accessible by a general-purpose or dedicated computer. For example, but not limited to, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to transport or store desired program code means in the form of instructions or data structures, and is accessible by a general-purpose or dedicated computer or general-purpose or dedicated processor. Furthermore, any connection may be referred to as computer-readable medium. For example, if coaxial cables, fiber optic cables, twisted pair cables, digital subscriber lines (DSL), or wireless technologies such as infrared, radio frequency, or microwave are used to transmit software from a website, server, or other remote source, then those coaxial cables, fiber optic cables, twisted pair cables, DSL, or wireless technologies such as infrared, radio frequency, or microwave fall within the definition of a medium. The term "disk" as used in this application includes compact discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray® discs, as well as discs in which data is typically reproduced magnetically and discs in which data is reproduced optically using a laser. Combinations of the above are also naturally included within the framework of computer-readable media.

[0116] Although specific embodiments have been described above for teaching purposes, the teachings in this patent application have general applicability and are not limited to the specific embodiments described above. Therefore, various modifications, adaptations, and combinations can be made to the various features of the described embodiments without deviating from the technical scope of the invention described in the claims.

Claims

1. A weighing system, An illumination subsystem configured to illuminate a structure created on a semiconductor wafer with a certain amount of measurement illumination light during the measurement interval, A pump illumination subsystem configured to illuminate the structure with a certain amount of pump illumination light during the measurement interval, wherein the power of the pump illumination light incident on the structure fluctuates over time during the measurement interval. A detector configured to detect a measurement signal related to the measurement of the above structure in accordance with the amount of measurement illumination light and the amount of pump illumination light, It is an information processing system, This refers to a change in the values ​​of one or more electrical or optical properties of one or more types of materials provided in the aforementioned structure, and the change induced by the fluctuation in the power of the pump illumination light is estimated. The change in the value of the detected measurement signal, which is estimated to be a change induced by the fluctuation in the power of the pump illumination light, and Based on the changes in the values ​​of one or more electrical or optical properties of the one or more types of materials provided in the structure, and the changes in the values ​​of the detected measurement signals, the value of the parameter of interest that characterizes the structure under measurement is estimated. An information processing system configured in such a way, A weighing system equipped with [a specific feature / feature].

2. A metrology system according to claim 1, wherein a trained derivative-based measurement model is involved in the estimation of the value of the parameter of interest.

3. A weighing system according to claim 2, wherein the trained derivative-based weighing model is a machine learning-based model or a physical-based model.

4. A weighing system according to claim 2, wherein the parameter of interest is any of limit dimensions, film thickness, and material properties.

5. A weighing system according to claim 1, wherein the detected measurement signal relates to the measurement of the structure at multiple wavelengths, multiple incident angles, multiple azimuth angles, or any combination thereof.

6. A weighing system according to claim 1, wherein the pump illumination source of the pump illumination subsystem is a laser-based illumination source having wavelength-selectable output.

7. A weighing system according to claim 1, wherein there are multiple different materials within the structure under measurement, and the photon energy of the pump illumination light is between the band gap energy of a first of the multiple materials and the band gap energy of a second of the multiple materials.

8. A weighing system according to claim 1, wherein the weighing system is an optical weighing system, an electronic weighing system, or an X-ray weighing system.

9. A weighing system according to claim 1, wherein the estimation of the change in one or more electrical or optical properties of a material among the one or more types of materials provided in the structure is accompanied by the measurement of a change in one or more electrical or optical properties of a weighing target which is induced by the fluctuation of the power of the pump illumination light, wherein the weighing target is disposed on the semiconductor wafer at a location separate from the location of the structure under measurement, and the weighing target has a single film layer of the material.

10. A metering system according to claim 1, wherein the power of the pump illumination light is varied by mechanical modulation or electronic modulation.

11. A weighing system according to claim 10, wherein the modulation is sinusoidal modulation or square wave modulation.

12. A weighing system according to claim 1, wherein the detected measurement signal is the value of one or more Müller matrix elements, the value of one or more harmonic signals, or the value of one or more detected image signals.

13. A weighing system according to claim 1, wherein the structure under measurement is an array of partially fabricated transistor devices.

14. A metering system according to claim 1, wherein the fluctuation of the power of the pump illumination light is periodic or aperiodic.

15. It is a method, A structure created on a semiconductor wafer is illuminated with a certain amount of measurement illumination light and a certain amount of pump illumination light during the measurement interval, and the power of the pump illumination light incident on the structure fluctuates over time during the measurement interval. The measurement signal related to the measurement of the above structure is detected according to the amount of measurement illumination light and the amount of pump illumination light. This refers to a change in the values ​​of one or more electrical or optical properties of one or more types of materials provided in the aforementioned structure, and the change induced by the fluctuation in the power of the pump illumination light is estimated. The change in the value of the detected measurement signal, which is estimated to be a change induced by the fluctuation in the power of the pump illumination light, and Based on the changes in the values ​​of one or more electrical or optical properties of the one or more types of materials provided in the structure, and the changes in the values ​​of the detected measurement signals, the value of the parameter of interest that characterizes the structure under measurement is estimated. method.

16. A method according to claim 15, wherein the detected measurement signal relates to the measurement of the structure at multiple wavelengths, multiple incident angles, multiple azimuth angles, or any combination thereof.

17. A method according to claim 15, wherein the structure under measurement contains a plurality of different materials, and the photon energy of the pump illumination light is between the band gap energy of a first of the plurality of materials and the band gap energy of a second of the plurality of materials.

18. A method according to claim 15, wherein the estimation of the change in one or more values ​​of the electrical or optical properties of one of the one or more types of materials provided in the structure is accompanied by the measurement of a change in the one or more values ​​of the electrical or optical properties of a metering target, which is a change induced by the fluctuation of the power of the pump illumination light, wherein the metering target is disposed on the semiconductor wafer at a location separate from the location of the structure under measurement, and the metering target has a single film layer of the material.

19. A method according to claim 15, wherein the fluctuation of the power of the pump illumination light is periodic or aperiodic.

20. A weighing system, An illumination subsystem configured to illuminate a structure created on a semiconductor wafer with a certain amount of measurement illumination light during the measurement interval, A pump illumination subsystem configured to illuminate the structure with a certain amount of pump illumination light during the measurement interval, wherein the power of the pump illumination light incident on the structure fluctuates over time during the measurement interval. A detector configured to detect a measurement signal related to the measurement of the above structure in accordance with the amount of measurement illumination light and the amount of pump illumination light, A non-temporary computer-readable medium on which instructions are stored, and when those instructions are executed by one or more processors, that one or more processors This refers to a change in the values ​​of one or more electrical or optical properties of one or more types of materials provided in the aforementioned structure, and the change induced by the fluctuation in the power of the pump illumination light is estimated. The change in the value of the detected measurement signal, which is estimated to be a change induced by the fluctuation in the power of the pump illumination light, and Based on the changes in the values ​​of one or more electrical or optical properties of the one or more types of materials provided in the structure, and the changes in the values ​​of the detected measurement signals, the value of the parameter of interest that characterizes the structure under measurement is estimated. This refers to a non-temporary computer-readable medium, A weighing system equipped with [a specific feature / feature].

Images