Apparatus and method for measuring substrates

Abstract

To provide a method and an apparatus that are improved to measure a multi-layer substrate.SOLUTION: A method includes a step (110) of manufacturing a substrate having a pattern on the surface of a top layer, a step (120) of measuring the substrate, particularly the pattern, by using at least one measurement technique, a step (130) of creating a simulation of the substrate by using a measurement result from the measurement of the substrate, a step (140) of comparing the measurement result with a simulation result from the simulation of the substrate, and a step (150) of calculating a parameter of another substrate when the measurement result corresponds to the simulation result. When there is a deviation between the measurement result and the simulation result, step (130) optimizes the simulation and creates the simulation of the substrate again by using the measurement result from the measurement of the substrate (120).SELECTED DRAWING: Figure 1a

Description

[Technical Field]

[0001] The present invention relates to an apparatus and method for measuring a substrate.

[0002] The significant reduction in the critical dimension (CD) of patterns fabricated on substrates using photolithography or nanoimprint lithography is increasing the challenges for measurement and inspection techniques in the manufacturing process.

[0003] Conventional imaging methods, such as optical microscopy, are limited to pattern dimensions exceeding half the optical wavelength, usually well over 100 nm, due to the Abbe resolution limit—diffraction-limited resolution. Three-dimensional characterization of patterns with these limit dimensions is only possible to a limited extent using optical microscopy. Geometric parameters such as pattern width, pattern height, edge angle, and three-dimensional pattern roughness are becoming increasingly important as patterns become smaller. Alternative methods for measuring patterns, such as transmission electron microscopy (TEM), scanning electron microscopy (SEM), or atomic force microscopy (AFM), are too labor-intensive (time-consuming) for the semiconductor industry and therefore unsuitable for process monitoring and mass production. Similarly, large-area profilometry is only to a limited extent suitable for determining topology.

[0004] In conventional techniques, traditional imaging methods such as optical microscopy are combined with non-imaging measurement methods using light wave scattering (scatterometry).

[0005] However, conventional imaging methods are omitted because the limiting dimensions of patterns fabricated on substrates using methods such as photolithography or nanoimprint lithography have been significantly reduced.

[0006] Conventional techniques use traditional polarization analysis (ellipsometry) to measure layer thickness and optical material properties such as refractive index and reflectivity. To avoid damaging the substrate or wafer, it is necessary to measure the coating without destroying the layer. In particular, spectral polarization analysis and reflectometry are established measurement systems for process control and process optimization in the semiconductor industry. Essentially, relative changes or deviations are detected. Programs required for measurement, simulation, and evaluation of simple systems are well-known in the prior art.

[0007] A detailed overview of the subject of spectroscopic polarization analysis and polarimetry for analyzing materials and systems in the nm range, including thin films, is provided, for example, in J. Nanopart. Res. (2009) 11: 1521-1554, and will not be discussed in further detail here.

[0008] For example, U.S. Patent No. 6,912,056 describes an apparatus and method for measuring a multilayer on a substrate. The apparatus includes a measurement unit for measuring reflected light, which is reflected by the substrate on which a multilayer is formed. A plurality of recipe data is input by an input unit, and each of the plurality of recipe data corresponds to a plurality of virtual multilayers, one of the virtual multilayers being initially assumed to be the multilayer actually formed on the substrate. Using the set of the plurality of recipe data, a control unit calculates a plurality of theoretical spectra, each of the plurality of theoretical spectra indicating at least one thickness of each constituent layer of a virtual multilayer assumed to be the multilayer actually formed on the substrate, and the measured spectrum is compared with the plurality of theoretical spectra, whereby a provisional thickness of the multilayer is determined. The calculated provisional thickness needs to be within an allowable range, and if not, after changing the recipe data, the selection of the provisional thickness is repeated. Physical information includes the refractive index and absorption coefficient of each constituent layer.

[0009] U.S. Patent No. 7,196,793 also uses data measured by spectroscopic ellipsometry, such as polarization changes of radiation (Ψ(λ i ) and Δ(λ i )) to characterize a thin bilayer system on a substrate, and compares these data with simulated spectra. The simulated spectra are adapted in a model until the difference between the simulated values (Ψ M (λ i )) and Δ M (λ i )) and the measured values (Ψ E (λ i )) and Δ E (λ i )) is minimized, where the layer thickness d (best) and the angle of incidence Φ (best) are changed.

[0010] U.S. Patent No. 7,289,219 and U.S. Patent No. 7,502,101 describe the use of polarimetry-based optical wave scattering measurement to measure the limiting dimensions of periodic patterns on wafers or semiconductor components.

[0011] U.S. Patent No. 7,268,876 describes the use of a minimization algorithm based on nonlinear regression or the Levenberg-Marquardt method to characterize the removal or deposition of the outermost layer of a sample in situ using spectroscopic polarization analysis.

[0012] In "Aetherometry for nanoimprint lithography," Journal of Vacuum Science & Technology B, Vol. 34, No. 06K503, 2016, by Zhu, R., Brueck, SRJ, Dawson, N., Busani, T., Joseph, P., Singhai, S., and Sreenivasan, SV, variable angle scatterometry is used to characterize patterns fabricated using nanoimprint lithography. These patterns are wire grid polarizers (WGP) and photoresist grids. The RCWA algorithm was used for model generation.

[0013] In "aetherometry for in situ measurement of pattern reflow in nanoimprinted polymers" by Patrick, HJ, Germer, TA, Ding, Y., Ro, HW, Richter, LJ, and Soles, CL, Applied Physics Letters, Vol. 93, No. 233105, 2008, spectroscopic polarization analysis is used to characterize lattice patterns made of polystyrene produced by NIL.

[0014] The above method was used to detect changes in the pattern after heat treatment (annealing). The RCWA algorithm was used for model generation. The results were compared with AFM measurements.

[0015] Reflective polarization analysis methods are well-established, particularly in the measurement of thin layers in the semiconductor industry, and therefore will be briefly explained. In conventional techniques, single layers are typically characterized. Here, for example, the reflection of linearly polarized, parallel, and monochromatic light is measured in a three-phase system of substrate / layer / air. In a three-phase system, reflection and refraction occur at two interfaces. In the case of reflection in a (multi)layer system, reflection and refraction must be considered at each individual phase boundary. From the measurable characteristic quantities of the polarization analysis method describing the polarization change—the loss angle Ψ and phase difference Δ—the complex refractive index specific to the layer system material can be determined.

number

[0016] In contrast to monochromatic ellipsometry, variable angle spectral ellipsometry (VASE) provides broad wavelength coverage. Large amounts of data or information allow for the calculation of more accurate models. Using VASE, the following parameters, in particular, can be measured or calculated for organic and inorganic materials: • Layer thickness ranging from sub-nanometers to several micrometers • Surface roughness • Refractive index • Electrical conductivity • Absorption rate • Polymerization state • Composition of the mixture ·defect • Optical anisotropy • Material doping · Form

[0017] However, in the prior art, in most cases, individual transparent or translucent thin layers or bilayer systems are measured. When a multilayer system is inspected, there is no patterned surface. Due to the complexity, in the prior art, either a multilayer system or a periodic pattern is measured. In the case of a multilayer system, it is a prerequisite in the prior art that the surface is smooth. A further problem in the prior art is that, in many cases, the accuracy of the simulation from the measurement of complex samples is insufficient. It becomes even more difficult when the pattern also has a more complex pattern shape. In that case, the adjustment of the measurement and simulation is still a problem.

[0018] The problem to be solved by the present invention is to solve the problems of the prior art, and in particular, to present an improved method and an improved apparatus for measuring a multilayer substrate.

[0019] The above problems are solved by the subject matter of the independent claims. Advantageous developments of the present invention are described in the dependent claims. All combinations of at least two of the features presented in the description, the claims, and / or the drawings are also included within the scope of the present invention. When a range of values is described, the values within the described limits are also disclosed as limiting values and shall be claimable in any combination.

[0020] The present invention is, in particular, in a method for measuring a multilayer substrate having at least one pattern having a critical dimension, in particular a surface pattern having a critical dimension, The method has at least the following steps, in particular the following sequence, namely, · a step of manufacturing a substrate having a plurality of layers, the substrate having, in particular, a pattern, in particular a pattern on the surface of the uppermost layer, and the dimensions of the plurality of layers, in particular the dimensions of the pattern, being known, • A step of measuring the substrate, particularly the pattern, using at least one measurement technique, • A step to create a simulation of the substrate using the measurement results from the substrate measurement, • A step of comparing the measurement results with the simulation results from the circuit board simulation, If there is a discrepancy between the measurement results and the simulation results, the simulation is optimized and the simulation of the board is regenerated using the measurement results from the board measurement, or If the measurement results correspond to the simulation results, the next step is to use the generated simulation to calculate further substrate parameters. This relates to a method having

[0021] Advantageously, the generated simulations can be used to optimize the desired layer thickness, pattern, and material.

[0022] The present invention further relates to an apparatus for measuring a multilayer substrate having at least one pattern having limit dimensions, and in particular a surface pattern having limit dimensions, The device is • A means for measuring the substrate, particularly the pattern, using at least one measurement technique, • A means of creating a simulation of a substrate using measurement results from substrate measurements, • A means of comparing measurement results with simulation results from circuit board simulations, • A means for optimizing the simulation and regenerating the simulation of the substrate using the measurement results from the substrate measurement, A means for evaluating and optimizing further substrates (1,1',1'') by reconstructing layer parameters and / or pattern parameters using the generated simulation based on measurement results from further substrate (1,1',1'') measurements. This relates to an apparatus having the following characteristics.

[0023] Advantageously, it can measure thin transparent or translucent layers, or double-layer or multi-layer systems, and can also measure patterned surfaces. In this way, it becomes possible to measure complex substrates with high precision.

[0024] In particular, it is proposed to provide simultaneous reconstruction of layer parameters and pattern parameters by combining measurement and simulation. The amount of information increases by adding multiple measurement variables and / or multiple measurement methods. Such combined measurement techniques are preferably used with RCWA as the calculation method, enabling the characterization of complex samples containing multiple layers and patterns by acquiring diffraction and phase information and topology information. The newly used techniques provide results with realistic computational performance and with acceptable computation time for process monitoring.

[0025] Alternative computational methods for electromagnetic simulations include, for example, the FDTD (Finite Difference Time Domain) method and the FE (Finite Element Method) method.

[0026] Preferably, the measurement technique includes the following techniques: • VUV-UV-Vis-NIR angle-tunable spectroscopy polarization analysis (VASE) in either reflection or transmission mode, with a measurement range from vacuum ultraviolet (VUV) to near-infrared (NIR), i.e., 146 nm to 1700 nm. • VASE (Variable Angle Spectroscopic Polarization Analysis) in either reflection or transmission mode, with a spectral measurement range of 1.7 μm to 30 μm. • (Polarized) reflectance • (Polarized) light wave scattering measurement method ·UV-Vis spectroscopy ·THz spectroscopy At least one of these, preferably exactly one.

[0027] These measurement techniques are known in the prior art and will not be described in further detail. In particular, in the IR or MIR (mid-infrared) range, in addition to reflectance or transmission measurements, measurements by attenuated total reflection in ATR (attenauted total reflection) mode are also possible (ATR spectroscopy).

[0028] A configuration using a spectrophotometric polarization analyzer is preferred and is used as the first measurement technique in the first embodiment of the present invention.

[0029] Preferably, the angle of incidence and / or wavelength and / or polarization state are changed and measured.

[0030] Preferably, RCWA (Rigorous coupled-wave analysis) is used to create the simulation.

[0031] Preferably, according to the present invention, only one measurement technique is used, in which the incident angle and wavelength, which are independent measurement variables, and the polarization state are changed and measured.

[0032] In further embodiments where the system to be inspected is not particularly complex, the angle is not changed.

[0033] In a third embodiment of the present invention, where the system to be inspected is very complex, a second measurement technique is used in addition to angle-variable spectroscopic polarization analysis, and a third measurement technique is also used as needed, and so on. The number and types of measurement techniques used depend on the substrate to be inspected and must be selected case-specific during the model generation process. According to the present invention, combining wavelength-resolved measurement techniques with angle-resolved measurement techniques, such as optical wave scattering measurement or polarization analysis, improves the accuracy of the simulation.

[0034] The following optical properties are present over a wide spectral range: • Refractive index (n) • Absorption coefficient (k) • Real and imaginary parts of the dielectric function (ε1, ε2) Absorption coefficient (α) • Real and imaginary parts (σ1, σ2) of complex photoconductivity • Optical anisotropy It is possible to make a decision.

[0035] These optical properties are well known to those skilled in the art and will not be described in further detail.

[0036] Preferably, the means for measurement includes at least one optical device, in particular a polarization analyzer and / or a reflectance analyzer and / or an optical wave scattering analyzer and / or a spectrometer.

[0037] Preferably, the apparatus includes at least one data processing unit and at least one data processing system for processing and storing data acquired from means for measuring a substrate.

[0038] Preferably, the means for measurement comprises at least one radiation source, in particular a laser or broadband radiation source, at least one monochromator, at least one polarizer, at least one compensator, at least one substrate holder, at least one analyzer, and at least one detector, wherein at least one polarizer allows for adjustment of a selected elliptic polarization state, in particular linear or circular.

[0039] Preferably, all means for measuring the substrate are located within the apparatus.

[0040] Method Steps In particular, the present invention describes a multi-step method for characterizing a multilayer system having (surface) patterned portions:

[0041] In the first step, a sufficiently large number of measurements are performed on a selected known system (hereinafter also referred to as a substrate), i.e., a manufactured sample. The sample may be a multilayer system having or not having patterns or surface patterns. According to the present invention, in particular, wavelength-resolved and / or angle-resolved measurements are performed to measure and modify the polarization state.

[0042] The selected sample is measured using at least one measurement technique depending on its complexity, and preferably, all components for performing various measurement techniques are provided within the apparatus according to the present invention. Individual instrument components can be replaced, added, or omitted as needed. Adding multiple measurement methods increases the amount of information recorded in the reference signature. In a less preferred alternative embodiment, the sample to be measured is transferred to a further measuring device, thereby allowing further measurements to be performed using various measurement techniques.

[0043] According to the present invention, appropriate measurement techniques for acquiring information include, in particular, optical wave scattering measurement, polarization analysis, reflectance measurement, spectroscopy, and / or diffraction measurement. Measurements can be performed, for example, with variable polarization of the radiation being measured, changes in the angle of incidence, and changes in wavelength. According to the present invention, in particular, combining wavelength-resolved measurement techniques and angle-resolved measurement techniques leads to improved simulation accuracy. Furthermore, in order to acquire additional information and data, measurements can be performed not only in the reflectance mode but also in the transmission mode, depending on the type of sample and the measurement technique.

[0044] The measurement technique is also selected based on the optical properties of the individual layers of the layer system. For example, within a specific wavelength range, one layer may be sufficiently transparent while one or more other layers absorb or reflect more strongly.

[0045] In a further step, a suitable model is created based on the recorded data, preferably using RCWA (Exact Coupled Wave Analysis) as the calculation method. A newly developed complex simulation algorithm is used for model generation. The simulation allows for the consideration of various influences in the measurement by comparing the measurement results with the simulation results for a known sample. For this purpose, model-based measurements are performed. If the measured sample consists of multiple layers and (surface) patterns, the complexity of the system and the number of parameters to be determined increase.

[0046] According to the present invention, iterative steps of measurement, model generation or model optimization, and simulation are required. If the measurement results and simulation results do not agree within an acceptable range, the model needs to be further optimized. If the measurement results and simulation results agree within an acceptable range, the simulation can be used for further analysis of the sample.

[0047] A configuration using a spectrophotometric polarization analyzer, such as VASE, is a preferred configuration according to the present invention and is used as the first measurement technique in the first embodiment of the present invention. The number and type of measurement techniques used must be determined individually for each system to be characterized. The measurement techniques must provide experimental measurement data that is sensitive to many of the parameters of interest without excessive correlation between the parameters. Examples of parameters include, for example, height and width in the case of a surface pattern, and layer thickness in the case of the nth layer.

[0048] RCWA is used to calculate the diffraction of a diffraction grating, and the sample is divided into several separate layers. The RCWA algorithm, according to the present invention, makes it possible to determine the limit dimensions of the pattern being inspected. The limit dimensions of the pattern being inspected are, for example, the height or depth of the pattern, the width and length of the pattern, angles (e.g., sidewall angles), the thickness of the residual layer, and the surface roughness. Measurements can be performed on periodic positive patterns and / or periodic negative patterns.

[0049] If a specific parameter is changed in the substrate to be characterized, this change should cause a change in the spectral recording. If multiple different parameters cause the same change in the experimental recording, the correlation is too high, making a unique correspondence impossible or difficult. In such cases, the selection of measurement techniques needs further optimization. The possibility of correlation with unknown parameters presents another challenge in model generation. High reproducibility in the manufacturing process of the sample to be tested is a prerequisite.

[0050] According to the present invention, correlation analysis and sensitivity analysis are performed, in particular, to evaluate the quality of the reconstruction based on the measurements performed.

[0051] In a further step, an optimized model is used to characterize an unknown sample, which requires mapping the sample to a known sample system. Layer and pattern dimensions are reconstructed by comparing measured and simulated spectra. The simulated spectra are used as a dataset to reconstruct the parameters being sought.

[0052] These parameters are, for example, as follows: • Layer thickness ranging from sub-nanometers to several micrometers • Surface roughness • Refractive index • Electrical conductivity • Absorption rate • Polymerization state • Composition of the mixture ·defect • Optical anisotropy • Material doping ·form • Limit dimensions of the pattern to be inspected. Limit dimensions of the pattern to be inspected include, for example, the height or depth of the pattern, the width and length of the pattern, angles (e.g., sidewall angles), the thickness of the residual layer, and the surface roughness.

[0053] The developed and optimized models can be used not only to calculate desired parameters for further similar substrates and similar (layer) materials, but also to optimize desired layer thicknesses, patterns, and materials. Therefore, for example, layer thickness or pattern dimensions can be optimized using the models according to the present invention based on desired parameter quantities.

[0054] Substrates having multilayer systems and / or (surface) patterned areas A sample means a substrate that has been processed or treated using methods known in the semiconductor industry, and in particular means a substrate that has been coated, and / or imprinted, and / or bonded, and / or etched, and / or treated with plasma, and / or treated with light, for example by laser, or master stamps, working stamps, and microfluidic assemblies are also understood as samples.

[0055] A substrate or semiconductor substrate is understood to be a round, semi-finished product, particularly in the semiconductor industry, that has not yet been individualized. A wafer is also understood to be a substrate. A substrate can have any arbitrary diameter, but the diameters of substrates are, in particular, 1 inch, 2 inches, 3 inches, 4 inches, 5 inches, 6 inches, 8 inches, 12 inches, or 18 inches or more. In special embodiments, a substrate may also have a rectangular shape, or at least a shape other than circular.

[0056] The sample to be characterized includes, among other things, one or more of the following components and / or coatings: • Resists, especially photoresists • Anti-sticking layer (ASL) • For example, the initial layer such as an adhesion promoter layer (primer) • Polymer layer • Working stamp materials (soft stamp materials or hard stamp materials) for imprint and nanoimprint methods. • Master stamp materials Patterned master stamps or hard stamps, and patterned hard materials, manufactured by electron beam lithography and / or chemical etching processes. • Layers with imprinted patterns • Patterned coating • Semiconductor layer Oxide layer

[0057] The method according to the present invention is not limited to the above-mentioned samples, and is generally suitable for multilayer systems having or not having patterned sections with limiting dimensions, as long as the sample can be measured using at least one of the measurement techniques according to the present invention (polarization analysis, optical wave scattering measurement, spectroscopy, diffraction measurement, and reflection measurement).

[0058] Preferred uses In addition to determining patterns with layer thickness and critical dimensions, analytical methods (particularly RCWA) can be used for the following applications: • Characterization of multilayer systems • Error analysis and error detection (failure analysis) • Characterization of the anti-adhesion layer and the initial layer. • Monitoring the chemical stability of materials such as resists, work stamping materials, initial layers, and anti-adhesion materials. • Monitoring of wear or corrosion of materials such as resists, work stamp materials, master stamp materials, initial layers, and anti-adhesion materials. • Monitoring of environmental wear of materials such as resists, work stamp materials, master stamp materials, initial layers, and anti-adhesion materials. • Monitoring of oxidation or reduction processes of materials such as resists, working stamp materials, master stamp materials, initial layers, and anti-adhesion materials. For example, characterization of the chemical composition of materials such as resists, working stamp materials, master stamp materials, initial layers, and anti-adhesion materials. For example, characterization of the irradiation stability of materials such as resists, working stamp materials, master stamp materials, initial layers, and anti-adhesion materials. For example, evaluation of the thermal stability characteristics of materials such as resists, working stamp materials, master stamp materials, initial layers, and anti-adhesion materials. For example, characterization of the durability (aging degradation) of materials such as resists, work stamp materials, master stamp materials, initial layers, and anti-adhesion materials. For example, characterization of the electrical conductivity of materials such as resists, working stamp materials, master stamp materials, initial layers, and anti-adhesion materials. Generally, characterization of the degree of curing of resists, work stamp materials, and imprint materials. • Characterization of the miscibility of materials, such as resists or work stamp materials. • Characterization of mixtures and monitoring of separation • Evaluation of the isotropic properties of materials • Monitoring of coating processes such as resist coating, working stamp material layer, master stamp material, initial layer, and anti-adhesion coating, and layer formation and / or film formation. Characterization of microfluidic assemblies of 1D, 2D, and 3D diffractive optical elements (DOEs) with limit dimensions (printed directly on transparent substrates or polymer masters), imprinted NIL patterns, work stamps, high-temperature imprinted nanopatterns, lithographic patterns, and microfluidic assemblies. Characterization and testing of patterned master stamps, patterned hard stamps, and patterned hard materials produced by electron beam lithography and / or chemical etching processes. • Characterization of the spatial uniformity of patterns with limit dimensions • Characterization of layer formation and layer growth mechanisms (e.g., Vollmer-Weber growth patterns) Characterization of the shrinkage behavior of microfluidic assemblies, imprinted NIL patterns, work stamps, high-temperature imprinted nanopatterns, lithography patterns, and 1D, 2D, and / or 3D diffractive optical elements (DOEs) with limiting dimensions. Characterization of additional coatings on microfluidic assemblies, on imprinted NIL patterns, on work stamps, on master stamps, on high-temperature imprinted nanopatterns, on lithography patterns, and on 1D, 2D, and / or 3D diffractive optical elements (DOEs) with limit dimensions. • Monitoring and characterization of cleaning processes for 1D, 2D, and / or 3D diffractive optical elements (DOEs) with limit dimensions, imprinted NIL patterns, work stamps, master stamps, high-temperature imprinted nanopatterns, lithography patterns, and microfluidic assemblies. Identification of defects (e.g., voids) in 1D, 2D, and / or 3D diffractive optical elements (DOEs) with limiting dimensions, imprinted NIL patterns, work stamps, master stamps, high-temperature imprinted nanopatterns, lithography patterns, microfluidic assemblies, and other multilayer systems. • Characterization of the etching process in situ

[0059] More generally, the method according to the present invention can be used for quality control of manufactured samples having multilayer systems and / or patterned sections. For example, the quality of work stamps for nanoimprint lithography is characterized. These work stamps may consist of multiple layers having varying thicknesses and materials. One of the multiple layers is patterned (e.g., a soft stamp material layer). Several parameters may be important for quality. Quality control may be performed immediately after manufacturing and / or after a predetermined time interval to control wear or deterioration over time, for example, during use.

[0060] The method according to the present invention can be used to monitor manufactured products and optimize the manufacturing process until desired characteristics, i.e., desired parameters, are reproducibly achieved. The spatial uniformity of the selected parameters can also be determined, for example, by the method according to the present invention and can be used as a selection criterion.

[0061] The applications of the present invention are not limited to the above-mentioned multilayer systems (with or without patterned sections).

[0062] Device The apparatus includes optical devices such as a polarization analyzer and / or a reflectance analyzer and / or an optical wave scattering analyzer and / or a spectrometer, and a data processing unit or data processing system for processing and storing data acquired from the optical devices.

[0063] The essential components of the optical device are at least one radiation source (e.g., a laser or broadband radiation source), at least one monochromator, at least one polarizer, at least one compensator, a sample holder, at least one analyzer, and at least one detector. The polarization optics allow for adjustment of a selected elliptically polarized state (linear, circular, etc.). Measurements are not limited to reflective mode but can also be performed in transmission mode or ATR mode. In the case of ATR measurements, a substrate and / or an ATR crystal and a corresponding ATR holder for the ATR optics are added as additional or alternative components.

[0064] According to the present invention, in particular, measurements are performed using wavelength-resolved and angle-resolved methods, in which case the polarization state can be changed. If the system to be inspected is not very complex, the wavelength and polarization state are sufficient information, so the angle is not changed.

[0065] A configuration using a spectroscopic polarization analyzer, preferably employing angle-variable spectroscopic polarization analysis, is a preferred configuration according to the present invention and is used as the first measurement technique in the first embodiment of the present invention.

[0066] The selected sample can be measured using multiple measurement techniques depending on its complexity, and preferably, all components for performing various measurement techniques are provided within the apparatus according to the present invention. Individual instrument components can be replaced, added, or omitted as needed. According to the present invention, the amount of information in the recorded reference signature increases by adding multiple measurement methods. In an alternative embodiment, the sample to be measured is transferred to a further measuring device, thereby allowing further measurements to be performed using various measurement techniques.

[0067] Typical components of a polarization analyzer include, for example, a light source, a polarizer, possibly a compensator (e.g., a λ / 4 plate), a sample holder, an analyzer, a monochromator if necessary, and a detector.

[0068] Further advantages, features, and details of the present invention will become apparent from the following description of preferred embodiments and based on the drawings. [Brief explanation of the drawing]

[0069] [Figure 1a] This is a flowchart showing the method steps of an exemplary embodiment of the method according to the present invention. [Figure 1b] This is a flowchart having method steps of an exemplary method according to the present invention. [Figure 2a] This is a cross-sectional view of a substrate having a multilayer system and a surface patterned portion having a periodic positive pattern. [Figure 2b] This is a cross-sectional view of a second substrate having a multilayer system and a surface patterned portion having a periodic negative pattern. [Figure 2c] This is a cross-sectional view of a third substrate having a multilayer system and a surface patterned portion having a periodic trapezoidal positive pattern. [Figure 2d] These are plan views of four exemplary embodiments of periodic patterns (rectangle 7, linear 7''', circular 7IV, and irregular shape 7V). [Figure 3] This is a schematic diagram of the optical components of the apparatus according to the present invention in an exemplary embodiment.

[0070] In each figure, the same components or components with the same function are given the same reference numeral. Figures 2 and 3 are not drawn to scale in order to improve the illustration.

[0071] Figure 1a shows a flowchart of the method according to the present invention. A comparison is made between measurement results and simulation results for a known sample (a multilayer system having a patterned (structured) part). The method requires iterative steps of measurement 120, model generation 130 or model optimization 140, and further simulation.

[0072] In the first method step 110, a substrate (hereinafter also referred to as a sample) having multiple thin layers and (surface) patterned (structured) sections is manufactured. Thus, the system to be inspected is known and used for model generation and model optimization. If necessary, multiple precisely known reference substrates are manufactured and their measurement results are used to verify and optimize the developed simulation model.

[0073] In the second method step 120, the substrate is irradiated with electromagnetic radiation at a predetermined angle of incidence, and the reflected radiation is measured, for example, depending on the wavenumber and / or angle. This measurement is not limited to reflection but may also be performed by transmission. Multiple measurement techniques can be used to improve the reliability and / or accuracy of the calculation method. Suitable measurement techniques for obtaining information include, among others, optical scattering, polarization analysis, reflectance, spectroscopy, and / or diffraction. These various measurement techniques are based, among others, on measuring the light reflected or diffracted in micro-patterned (microstructured) and / or (sub)-nanopatterned (structured) samples as a function of instrument parameters such as the angle of incidence or the wavelength of light, in which case the polarization dependence of the measured quantity is utilized. Similarly, individual layers of a multilayer system are also detected directly or indirectly by the reflection and / or diffraction and / or scattering of light at the interfaces.

[0074] For the development of the simulation model, the entire multilayer system is measured after manufacturing is complete, or each individual layer is manufactured step by step and measured sequentially, depending on the layer thickness and refractive index of each known and defined sample.

[0075] In the first embodiment, each layer of the multilayer system has a layer thickness greater than 20 nm (20 nm or more), and the refractive index of each layer is known. In this embodiment, the entire multilayer system is measured after manufacturing is complete. For example, an imprint stamp for imprint lithography or nanoimprint lithography having a patterned imprint layer.

[0076] In the second embodiment, the sample includes very thin layers having a thickness of less than 20 nm with a known refractive index. In the case of very thin layers, particularly layers with thicknesses ranging from less than nm to sub-nm, each individual layer is manufactured and measured before the next layer is manufactured on top of it. In this embodiment, for example, the sample is measured after each layer of the ASL layer has been deposited. All measurement data of the sample measured layer by layer are considered for model generation.

[0077] If necessary, layers with a thickness exceeding 20 nm can be individually measured during the sample manufacturing process, depending on the refractive index and existing material information.

[0078] In the third embodiment, the sample includes an intermediate layer having a thickness of more than 20 nm, but with an unknown refractive index. In this embodiment, measurements are performed each time an individual layer with a thickness of more than 20 nm is deposited, and measurements are also performed on the entire multilayer system after manufacturing is complete.

[0079] In polarization analysis, polarization changes are described using measurable polarization analysis characteristic quantities—loss angle Ψ and phase difference Δ. Since optical parameters cannot be directly determined using Ψ and Δ, it is necessary to develop parameterized models for the sample system to be examined. To calculate the interaction between multilayer systems, as well as nanopatterns and micropatterns, and light, according to the present invention, RCWA (exact coupled-wave analysis) is preferably used as the calculation method. RCWA is used to calculate the diffraction of a diffraction grating, where the sample is divided into multiple individual layers. This model concept was developed and complemented by the present invention.

[0080] An advanced form of the present invention advantageously allows for the simultaneous evaluation of the diffraction of an incident (plane) wave in a multilayer system and the diffraction of an incident (plane) wave in a pattern using a simulation model from step 130 of the method. Advantageously, the multilayer system and the non-planar layers, i.e., the pattern, can be identified with very high reliability and accuracy using polarization analysis methods and polarized light.

[0081] In the third method step 130, the selected dataset of measurement variables is used for model generation, which enables the calculation of a simulation of a multilayer system having a (surface) patterned portion. It is required to minimize the deviation between experimental data and simulation data (140). For example, for a complex system according to the present invention, such as the one illustrated in Figure 2a, which has multiple layers and surface patterns, the model developed in the method according to the present invention can accurately describe the sample physically. This enables highly reliable simulations.

[0082] In the fourth method step 140, a comparison is made between the measured and simulated results for the known sample prepared in step 110. This requires iterative steps of measurement, model creation or model optimization (model fitting), and further simulation. The goal of the adjustment is to make the model, i.e., the generated dataset, fit the measured dataset (i.e., experimental data) as best as possible. If this is still not the case, the model is further optimized in method step 130. If it is the case, the developed model can be used in method step 150 to identify the desired parameters.

[0083] The mathematical analysis of the developed model system can be used additionally, as needed, particularly in the development of general model systems.

[0084] Figure 1b shows a flowchart of the method according to the present invention when the completed simulation model according to the present invention is used for a multilayer system having (surface) patterned parts. Here, after system optimization, the developed model can be used for routine simulation. System parameters are predetermined. According to the present invention, in method step 120, a known sample—i.e., the number of flat and / or non-planar (i.e., patterned) layers and the material of the layers are known—is measured, and after the experimental dataset is compared with the generated dataset (140), the desired parameters are identified (150).

[0085] Figure 2a shows a cross-sectional view of a substrate 2 according to the present invention having multiple thin coatings 3-6 and a surface patterned portion 7. The number and thickness of the coatings are not limited to the embodiments shown in Figures 2a-2c. The thickness of the coatings is not drawn to scale for the sake of illustration. Figures 2a-2c show similar embodiments having different surface patterned portions 7, 7', 7'', respectively. The last coatings 6, 6', 6'' may consist of a photoresist or imprint material patterned using, for example, lithography or nanoimprint lithography. The patterns 7, 7', 7'' have dimensions in the nanometer range. The surface 6o, 6'o, 6''o of the patterned coatings defines the thickness of the residual layer.

[0086] Figure 2a shows a multilayer system 1 having a surface patterned section with a periodic positive pattern 7.

[0087] Figure 2b shows a further embodiment of the multilayer system 1' having a surface patterned portion with a periodic negative pattern 7'.

[0088] Figure 2c shows a third embodiment of a multilayer system 1'' having a surface patterned portion with a periodic trapezoidal positive pattern 7''.

[0089] When a surface is patterned with micropatterns and / or nanopatterns and / or subnanometer patterns, incident light strikes these (mostly) periodic surface coating patterns, and these surface coating patterns can represent an optical diffraction grating. The limiting dimensions of the inspected pattern 7,7',7'' include the pattern height or depth, the pattern width and length, angles, e.g., the angle of the sidewalls, the thickness of the residual layer, and the surface roughness.

[0090] Figure 2d is a plan view consisting of multiple sections, showing further possible surface patterns according to the present invention of the sample 7,7''',7 IV , and 7 VThis is shown in comparison. According to the present invention, a model is developed that enables highly reliable characterization of complex multilayer systems having (surface) patterned parts according to the patterns in Figures 2a to 2d. Pattern 7 is a quadrilateral, particularly a square. Pattern 7''' is a periodic linear surface coating pattern. In a further embodiment of the present invention, pattern 7 IV It is circular. (Pattern 7 in Figure 2d) V Patterns with relatively complex or irregular shapes, as shown above, are not a problem for the method according to the present invention and are detected and correctly reproduced in model generation. The patterning unit is not limited to the illustrated embodiment.

[0091] According to the present invention, the non-planar layer having a patterned portion is the uppermost layer 6,6',6'' of the multilayer system 1,1',1''. In an alternative embodiment, the non-planar layer having a patterned portion is located between two layers in the multilayer system. For example, the patterned imprint material is coated with an ASL coating after imprinting for use as a working stamp. In a further alternative embodiment, the multilayer system includes two or more non-planar layers having patterned portions.

[0092] Figure 3 shows the optical components of the apparatus 13 according to a first embodiment of the present invention. The polarizer (P) 9 converts unpolarized light from the radiation source 8 into linearly polarized light. The radiation is reflected by the sample 1 and then passes through the analyzer (A) 10. The electromagnetic radiation is elliptically polarized when reflected by the sample 1. The analyzer 10 changes the polarization of the reflected electromagnetic radiation again, and the electromagnetic radiation then strikes the detector (D) 11. In the first preferred embodiment, a polychromatic radiation source is used, and therefore, a selected wavelength range is used in the measurement method. In an alternative embodiment, monochromatic radiation is used, preferably a laser as the radiation source. Multiple radiation sources can be provided simultaneously in the apparatus and / or can be replaced as needed.

[0093] Further optical components include, for example, optical filters, compensators (e.g., λ / 2 plates), monochromators, and various tunable optical attenuators, which can be used as needed depending on the measurement technique and / or wavelength range. These components are known to those skilled in the art and will not be described in further detail.

[0094] The measurement technique according to the present invention differs in terms of the arrangement and type of optical components. The analyzer 10 can be configured to rotate, for example.

[0095] In contrast to monochromatic polarization analysis, angle-variable spectral polarization analysis (VASE) provides broad wavelength coverage. This improves the amount of information in the measured data and the accuracy of the simulation. A goniometer enables variable angle measurement. A preferred embodiment combines wavelength-resolved and angle-resolved measurements, which, according to the present invention, leads to improved simulation reliability. According to the present invention, this combination is performed using VASE as the preferred measurement technique.

[0096] Furthermore, to obtain additional information and data as needed, measurements can be performed not only in reflection mode but also in transmission mode, depending on the type of sample and measurement method.

[0097] The apparatus 13 according to the present invention includes an optical device and a data processing unit 12 for processing and storing data acquired from the optical device.

[0098] A housing device (not shown) is used to house and secure the sample or substrate. In a special embodiment, the housing device is movable in the Z direction as needed. Furthermore, the housing device is rotatable and / or tiltable.

[0099] The containment device can be heated and temperature-controlled within a temperature range of 0°C to 1000°C, preferably 0°C to 500°C, more preferably 0°C to 400°C, and most preferably 0°C to 350°C. Alternatively, the containment device may be cooled by a cooling device. For example, in the first embodiment, the containment device can be cooled within a temperature range of -196°C to 0°C. The temperature of the containment device is adjustable by a temperature control device. The containment device may further include sensors (not shown) capable of measuring physical and / or chemical properties. These sensors may be, for example, temperature sensors.

[0100] In a further preferred embodiment of the containment device, or even in an independent embodiment, the containment device includes a liquid cell that enables measurement under liquid conditions. In a particular embodiment, the liquid cell is a flow cell. Thus, multilayer systems having or not having (surface) patterned areas can be measured in a liquid environment. According to the present invention, in a particular application, the electrochemical responsiveness of a multilayer system can be characterized using the liquid cell. The liquid cell may be configured as an electrochemical cell having a reference electrode, a counter electrode, and an optical window for measurement by spectroscopic polarization analysis.

[0101] The apparatus 13 according to the present invention is advantageously operable in a vacuum or under ambient pressure in a gas atmosphere. Preferably, the gas atmosphere is an inert gas atmosphere, such as nitrogen (N2). Therefore, it is possible to inspect multilayer systems having patterned sections that are vulnerable to, for example, moisture or oxygen.

[0102] The apparatus 13 according to the present invention is preferably evacuable and heatable. The apparatus has means for introducing one or more gaseous components. A loading device, preferably a sluice, allows for the loading of a sample. In an alternative embodiment, the apparatus can be configured to allow measurements to be performed in situ.

[0103] Instead of the embodiment shown in Figure 3, alternative embodiments are conceivable that allow for a combination of measurement methods, particularly spectroscopic measurement and / or polarization analysis or optical wave scattering measurement. According to the present invention, all the measurement techniques already mentioned are conceivable.

[0104] The computer-aided data processing system 12 stores and processes data acquired from the optical device in order to simulate the multilayer system according to the present invention (with or without a patterning section) using a simulation algorithm further developed by the present invention. The simulation model according to the present invention enables, for the first time using the proposed method, simultaneous and highly reliable detection and characterization of multiple thin layers and (surface) patterns. [Explanation of Symbols]

[0105] Substrate / multilayer system with 1,1',1'' (surface) patterned area 2 Refractive index n s Substrate base material 3. First layer having refractive index n1 4. A second layer having a refractive index n2 5. Refractive index n3 or n n The third layer (nth layer) has 6,6',6'' Surface patterned area and refractive index n O The uppermost layer has 60,6'0,6''0 Top layer surface 7,7',7'',7''',7 IV ,7 V pattern 8 Radiation source 9 Polarizer 10 Photometer 11 detectors 12 Control and Computing Units 13 Optical device according to the present invention 110 Method Steps 120 Method Steps 130 Method Steps 140 Method Steps 150 Method Steps

Claims

1. A method for measuring a multilayer substrate (1,1',1''), This method is Step (110) of manufacturing a substrate (1, 1', 1'') having multiple layers (2, 3, 4, 5, 6, 6', 6''), wherein the parameters of the multiple layers are known, A step of processing or treating the substrate (1, 1', 1''), wherein in the processing or treatment, the substrate (1, 1', 1'') is coated and / or imprinted and / or bonded and / or etched and / or treated with plasma and / or treated with light. Step (120) of measuring the substrate (1, 1', 1'') using at least one measurement technique, Step (130) of creating a simulation of the substrate using the measurement results from the measurement of the substrate (1, 1', 1''), Step (140) of comparing the measurement results with the simulation results from the simulation of the substrate (1, 1', 1''), If a deviation exists between the measurement result and the simulation result, the simulation is optimized (130), and the simulation of the substrate is regenerated using the measurement result from the measurement of the substrate (1,1',1'') (130). The steps include evaluating other substrates (1, 1', 1'') by reconstructing the parameters of the multiple layers using the newly generated simulation, A method characterized by having at least one of the following.

2. The aforementioned measurement techniques include the following: VUV-UV-Vis-NIR angle-variable spectral polarization analysis (VASE) in either reflection mode or transmission mode, with a measurement range from vacuum ultraviolet (VUV) to near-infrared (NIR), i.e., 146 nm to 1700 nm. VASE (Variable Angle Spectroscopic Polarization Analysis) in either reflection or transmission mode, with a spectral measurement range of 1.7 μm to 30 μm. (Polarized) reflectance measurement (Polarized) light wave scattering measurement method UV-Vis spectroscopy THz spectroscopy At least one of the following: The method according to claim 1.

3. The angle of incidence and / or wavelength and / or polarization state are changed and measured. The method according to any one of claims 1 to 2.

4. A mathematical algorithm is used to create the aforementioned simulation. The method according to any one of claims 1 to 3.

5. RCWA (Exact Coupled Wave Analysis) is used to create the simulation. The method according to claim 4.

6. In a device for measuring multilayer substrates (1,1',1''), The device is A means (120) for measuring the substrate (1, 1', 1'') using at least one measurement technique, Means for processing or treating the substrate (1, 1', 1''), wherein in the processing or treatment, the substrate (1, 1', 1'') is coated and / or imprinted and / or bonded and / or etched and / or treated by plasma and / or treated by light, A means (130) for creating a simulation of the substrate using the measurement results from the measurement of the substrate (1, 1', 1''), Means (140) for comparing the measurement results with the simulation results from the simulation of the substrate (1, 1', 1''), Means (130) for optimizing the simulation (130) and for regenerating the simulation of the substrate (1,1',1'') using the measurement results from the measurement of the substrate (1,1',1''), A means for evaluating and optimizing further substrates (1,1',1'') by reconstructing layer parameters and / or pattern parameters using a simulation of the substrate that has been regenerated based on measurement results from measurements of other substrates (1,1',1''), and A device having.

7. The means for measuring includes at least one optical device. The apparatus according to claim 6.

8. The device is At least one data processing unit, A data processing system that processes and stores data acquired from means for measuring the substrate (1, 1', 1'') and The apparatus according to claim 6 or 7, having the following features.

9. The measuring means comprises at least one radiation source, at least one monochromator, at least one polarizer, at least one compensator, at least one substrate holder, at least one analyzer, and at least one detector. The at least one polarizer allows for adjustment of the selected elliptically polarized state. The apparatus according to any one of claims 6 to 8.

10. All means for measuring the substrate (1, 1', 1'') are located within the apparatus. The apparatus according to any one of claims 6 to 9.

Images