Nanosheet characterization using power-dependent raman spectroscopy

Abstract

The present disclosure provides a method for characterizing a nanostructure comprising illuminating the nanostructure with a Raman spectroscopy laser at a first power level sufficient to generate significant free charge carriers, acquiring a first Raman spectrum, illuminating the nanostructure at a second power level lower than the first power level wherein the second power level generates negligible free charge carriers, acquiring a second Raman spectrum, and determining at least one material property based on a difference between the first and second Raman spectra. The nanostructure may comprise a nanosheet field effect transistor having a superlattice of alternating Silicon and Silicon- Germanium layers with each layer having thickness less than 10 nanometers. The material property may comprise strain, Germanium composition, free charge carrier lifetime, or diffusion properties.

Description

NANOSHEET CHARACTERIZATION USING POWER-DEPENDENT RAMAN SPECTROSCOPYCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority from US provisional patent serial number 63 / 727,211 filing date December 3, 2024, which is incorporated herein by reference.FIELD OF INVENTION

[0002] The present disclosure relates to semiconductor metrology and characterization techniques, and more particularly to a method and system for characterizing nanosheet structures using power-dependent Raman spectroscopy to determine material properties and provide vertical profiling of individual layers through analysis of nonlinear optical responses induced by free charge carriers.BACKGROUND

[0003] The fabrication of advanced front-end semiconductor logic transistors requires extremely stringent process control to achieve the dimensional, material and electrical characteristics demanded by modern device designs. Nanosheet and nanowire field effect transistors represent a class of three-dimensional semiconductor structures that pose particular challenges for characterization and metrology due to their complex multilayer architectures and nanoscale dimensions.

[0004] These nanostructures typically include superlattices of alternating semiconductor layers, such as Silicon and Silicon-Germanium, with individual layer thicknesses often less than 10 nanometers. The performance and reliability of devices based on such structures depend on precise control of various material properties including strain distribution, compositional uniformity, and electrical characteristics across each individual layer within the stack.

[0005] Current metrology approaches for nanostructure characterization each present limitations that constrain their effectiveness for comprehensive process control. Optical Critical Dimensions metrology provides fast dimensional characterization but exhibits poor sensitivity to material characteristics such as Germanium composition and strain, and lacks sensitivity to electrical properties. Traditional Raman Spectroscopy can provide information about average strain and Germanium composition, and offers some profiling capabilities when multiple excitation wavelengths are employed, but these profilingcapabilities remain limited and typically require multiple laser sources extending into the ultraviolet range.

[0006] Secondary Ion Mass Spectrometry offers high-quality profiling information for properties such as Germanium concentration and layer thickness, but is destructive in nature and typically applicable only to planar, non-patterned measurement sites. Electrical testing methods can characterize electrical properties through external contacts but can only be applied at advanced fabrication steps when contacts are available, cannot provide profiling information, and are limited to electrical properties alone.

[0007] The semiconductor industry continues to drive toward smaller feature sizes and more complex three-dimensional device architectures, creating an increasing demand for non- destructive, fast, and comprehensive metrology techniques capable of providing detailed characterization of individual layers within multilayer nanostructures. Such techniques would enable better process control during both development and high- volume manufacturing of advanced semiconductor devices.SUMMARY

[0008] This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

[0009] According to an aspect of the present disclosure, a Raman spectroscopy system for nanostructure characterization is provided and includes an illumination module that is configured to illuminate the nanostructure, during a set of Raman measurement iterations, with light beams that include a light beam of a high power level and a light beam of a low power level a sensing unit configured to acquire, during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; and a processing circuit configured to analyze at least differences between Raman spectra acquired in relation to the light beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

[0010] According to an aspect of the present disclosure there is provided a method for Raman spectroscopy for nanostructure characterization, the method includes illuminating, by an illumination module, the nanostructure, during a set of Raman measurement iterations, with light beams that include a light beam of a high power level and a light beam of a low power level, acquiring by a sensing unit and during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; analyzing by a processing circuit, at least differences between Raman spectra acquired in relation to the light beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

[0011] According to an aspect of the present disclosure there is provided a non- transitory computer readable medium that stores instructions for nanostructure characterization using Raman spectroscopy, wherein the execution of the instructions by a Raman spectroscopy system results in: illuminating, by an illumination module of the Raman spectroscopy system, the nanostructure, during a set of Raman measurement iterations, with light beams that include a light beam of a high power level and a light beam of a low power level, acquiring by a sensing unit of the Raman spectroscopy system and during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; analyzing by a processing circuit of the Raman spectroscopy system, at least differences between Raman spectra acquired in relation to the light beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

[0012] According to an aspect of the present disclosure, the high power level belongs to a non-linear power range in which there is a non-linear relationship between Raman spectra and power levels of the excitation light beams, wherein the low power levelbelongs to a linear power range in which there is a linear relationship between Raman spectra and the power levels of the excitation light beams.

[0013] According to an aspect of the present disclosure, the light beams include different excitation light beams of different power levels, the different power levels comprise the high power level and the low power level.

[0014] According to an aspect of the present disclosure, the light beams further comprise one or more measurement light beams.

[0015] The foregoing general description of the illustrative embodiments and the following detailed description thereof are merely exemplary aspects of the teachings of this disclosure and are not restrictive.BRIEF DESCRIPTION OF FIGURES

[0016] Non-limiting and non- exhaustive examples are described with reference to the following figures.

[0017] FIG. 1 illustrates a graph displaying Raman spectroscopy data for semiconductor materials with varying free charge carrier levels, according to aspects of the present disclosure.

[0018] FIG. 2A illustrates a graph showing Raman spectra that illustrate the effect of free-charge carriers on the Raman spectrum for silicon samples with different doping concentrations.

[0019] FIG. 2B illustrates a graph showing Raman spectra that illustrate the effect of free-charge carriers on the Raman spectrum for a silicon sample with different excitation energy.

[0020] FIG. 2C illustrates a graph showing an effect of free-charge carriers on the Raman signal of silicon through laser excitation, with Raman intensity plotted as a function of Raman shift in wavenumbers.

[0021] FIG. 3 illustrates an induced charge density profile in a nanosheet field effect transistor.

[0022] FIG. 4 illustrates a graph showing Raman peak intensity distributions comparing low power and high power excitation conditions for a nanosheet structure, according to aspects of the present disclosure.

[0023] FIG. 5 depicts an example of a Raman spectroscopy system.

[0024] FIG. 6 depicts an example of a method.

[0025] FIG. 7 depicts an example of a method.DETAILED DESCRIPTION

[0026] The following description sets forth exemplary aspects of the present disclosure. It should be recognized, however, that such description is not intended as a limitation on the scope of the present disclosure. Rather, the description also encompasses combinations and modifications to those exemplary aspects described herein.

[0027] According to an aspect of the present disclosure, a Raman spectroscopy system for nanostructure characterization is provided and includes an illumination module that is configured to illuminate the nanostructure, during a set of Raman measurement iterations, with light beams that include a light beam of a high power level and a light beam of a low power level a sensing unit configured to acquire, during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; and a processing circuit configured to analyze at least differences between Raman spectra acquired in relation to the light beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

[0028] According to an aspect of the present disclosure there is provided a method for Raman spectroscopy for nanostructure characterization, the method includes illuminating, by an illumination module, the nanostructure, during a set of Raman measurement iterations, with light beams that include a light beam of a high power level and a light beam of a low power level, acquiring by a sensing unit and during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; analyzing by a processing circuit, at least differences between Raman spectra acquired in relation to thelight beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

[0029] According to an aspect of the present disclosure there is provided a non- transitory computer readable medium that stores instructions for nanostructure characterization using Raman spectroscopy, wherein the execution of the instructions by a Raman spectroscopy system results in: illuminating, by an illumination module of the Raman spectroscopy system, the nanostructure, during a set of Raman measurement iterations, with light beams that include a light beam of a high power level and a light beam of a low power level, acquiring by a sensing unit of the Raman spectroscopy system and during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; analyzing by a processing circuit of the Raman spectroscopy system, at least differences between Raman spectra acquired in relation to the light beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

[0030] According to an aspect of the present disclosure, the high power level belongs to a non-linear power range in which there is a non-linear relationship between Raman spectra and power levels of the excitation light beams, wherein the low power level belongs to a linear power range in which there is a linear relationship between Raman spectra and the power levels of the excitation light beams.

[0031] According to an aspect of the present disclosure, the light beams include different excitation light beams of different power levels, the different power levels comprise the high power level and the low power level.

[0032] According to an aspect of the present disclosure, the light beams further comprise one or more measurement light beams.

[0033] According to an aspect of the disclosure the same light beam (for example the light beam of high power level - or any other light beam) may be an excitation light beam and a measurement light beam and there no need in two light beams.

[0034] For simplicity of explanation, it is assumed that there are separate excitation light beams and measurement light beams. Any reference to a excitation light beam and a corresponding measurement light beam should be applied mutatis mutandis to an excitation light beam that is also a measurement light beam.Nanostructure Samples and Technical Benefits

[0035] Referring to Figure 1, a nanosheet field effect transistor (NS-FET) structure may serve as a sample for characterization using the disclosed Raman spectroscopy methods. The nanosheet field effect transistor may include a superlattice arrangement deposited over a Silicon substrate. The superlattice may include alternating Silicon and Silicon-Germanium layers, where each layer may have a thickness of less than 10 nanometers. In some cases, the individual layers may have typical thicknesses ranging from 2 to 8 nanometers, providing a three-dimensional structure that requires precise dimensional, material, and electrical characterization during fabrication. The alternating layer structure creates distinct material interfaces with different optical absorption coefficients, strain states, and electrical properties that enable selective characterization through power-dependent excitation.

[0036] The multilayer nanostructure may be implemented in various configurations depending on the specific device requirements. In some cases, the nanostructure may include a standard nanosheet field effect transistor configuration with the alternating Silicon and Silicon-Germanium layers forming the active regions of the device. The alternating layer arrangement may provide different material properties that can be individually characterized using the disclosed Raman spectroscopy techniques. The Silicon layers may exhibit different strain states due to lattice mismatch with the Silicon- Germanium layers, creating measurable shifts in Raman peak positions that correlate with stress levels. The Germanium composition in the Silicon-Germanium layers may vary from 10% to 40% depending on device requirements, with each composition level producing distinct Raman spectral signatures.

[0037] In some cases, the nanostructure may be implemented as a Complementary FET (CFET) structure. The CFET structure may be different and more complex than the standard NS-FET configuration, but both structures may contain groups of stacked nanosheets. The CFET configuration may include p-type and n-type NS-FETs positionedone on top of the other in a vertical arrangement. This vertical stacking arrangement may provide enhanced device density and performance characteristics while maintaining the superlattice structure of alternating Silicon and Silicon-Germanium layers with individual layer thicknesses of less than 10 nanometers. The CFET structure may present additional metrology challenges due to the presence of both p-type and n-type regions with different doping profiles and electrical characteristics that require differentiated characterization approaches.

[0038] The superlattice arrangement may provide distinct material interfaces that create different optical and electrical properties within each layer. These property variations may enable the selective characterization of individual layers using wavelength-dependent penetration depths and power-dependent free charge carrier generation. The thin layer dimensions of less than 10 nanometers may allow for precise profiling of material properties across the vertical structure of the nanosheet field effect transistor. The interface regions between Silicon and Silicon-Germanium layers may exhibit unique strain gradients and compositional transitions that can be characterized through analysis of Raman peak broadening and asymmetry. The power-dependent approach enables differentiation between surface and buried layers by exploiting the depth-dependent generation and diffusion of photo-excited charge carriers.

[0039] The disclosed power-dependent Raman spectroscopy approach provides significant technical benefits over conventional metrology techniques. Unlike traditional Raman spectroscopy that provides only average material properties across the entire structure, the power-dependent approach enables vertical profiling of individual layers within multilayer nanostructures with depth resolution approaching 2-3 nanometers. This capability addresses a critical gap in semiconductor metrology where existing techniques either lack material sensitivity (optical critical dimensions) or are destructive in nature (secondary ion mass spectrometry). The non-destructive nature of the technique allows for in-line process monitoring during high-volume manufacturing (for example manufacturing at least tens of thousands of samples , while the ability to characterize electrical properties through free charge carrier analysis provides information typically only available through electrical testing at advanced fabrication steps. The technique offers measurement throughput comparable to conventional Raman spectroscopy whileproviding significantly enhanced information content, enabling simultaneous determination of strain, composition, and electrical properties with single-measurement efficiency. The power-dependent approach eliminates the need for multiple laser wavelengths extending into the ultraviolet range, reducing system complexity and cost while maintaining superior profiling capabilities.

[0040] Referring to Figure 2A, the effect of free charge carriers on Raman spectral response may be demonstrated through doping concentration data. Figure 3 may show Raman spectra for silicon samples with different doping concentrations, displaying Raman shift in wavenumbers on the horizontal axis and Raman intensity on the vertical axis. The spectral curves may correspond to different doping levels, including undoped samples and samples with carrier densities such as 5 1017, I xlO18, 5xl018, I xlO19, and 5xl019carriers per cubic centimeter. The systematic variation in spectral response with doping concentration demonstrates the fundamental sensitivity of Raman scattering to free charge carrier density, providing the physical basis for power- dependent characterization through laser-induced carrier generation.

[0041] The Raman spectra may exhibit characteristic peaks near 520 cm'1, with the peak shape and intensity varying as a function of doping concentration. As shown in Figures 2A and 2B, the presence of free charge carriers may create a distinct lineshape response in the Raman spectrum through Fano interference effects between discrete phonon modes and the continuum of electronic transitions. The spectral lineshape response may demonstrate how increasing carrier density affects the Raman spectral signature, with variations in peak shape, position, and intensity corresponding to different carrier concentrations. The asymmetric broadening and frequency shifts observed with increasing doping concentration provide quantitative measures that can be correlated with electrical properties such as carrier mobility and lifetime. The sensitivity to carrier density enables detection of electrical property variations across individual layers in multilayer nanostructures when combined with power-dependent excitation.Raman Spectroscopy System Architecture

[0042] Referring to Figure 5, a Raman spectroscopy system 100 for nanostructure characterization is illustrated. System 100 includes an illumination module 110 that is configured to illuminate the nanostructure, during a set of Raman measurement iterations,with different excitation light beams that include power-different excitation light beams of different power levels, and with different measurement light beams. The illumination module 110 may include laser sources, optical elements for beam shaping and directing, and power control components to achieve the required illumination conditions. The system architecture provides enhanced measurement capabilities by enabling systematic variation of excitation conditions to probe different physical regimes of charge carrier generation and transport within the nanostructure.

[0043] The Raman spectroscopy system 100 may be implemented in various configurations depending on the specific measurement requirements. In one embodiment, the system may utilize the same light source for both excitation and measurement beams, where a single laser provides the excitation energy for generating free charge carriers and simultaneously serves as the probe beam for Raman spectroscopy measurements. This single-source configuration may provide advantages in terms of system simplicity, cost reduction, and alignment stability, as both functions are inherently synchronized and colocated. The single-source approach enables real-time monitoring of charge carrier effects during Raman measurement, providing direct correlation between excitation conditions and spectral response without temporal delays or spatial misalignment issues.

[0044] Alternatively, system 100 may employ different light sources for excitation and measurement beams, where a first laser source generates high-power excitation light for free charge carrier generation and a second laser source provides lower-power measurement light for Raman spectroscopy. This dual-source configuration may offer enhanced flexibility by allowing independent optimization of excitation and measurement parameters, including different wavelengths, power levels, and temporal characteristics. The dual-source approach may enable decoupling of the charge carrier generation process from the Raman measurement process, potentially providing improved control over the measurement conditions and enhanced sensitivity to specific material properties. The temporal separation between excitation and measurement pulses in the dual-source configuration enables investigation of charge carrier dynamics and relaxation processes on timescales ranging from nanoseconds to microseconds.

[0045] System 100 further includes a sensing unit 120 configured to acquire, during the set, Raman spectra related to radiation received from the nanostructure following theillumination of the nanostructure by the different measurement light beams. The sensing unit 120 may include spectrometers, detectors, and associated optical components for collecting and analyzing the Raman-scattered light from the nanostructure sample. The sensing unit 120 may incorporate high-sensitivity photodetector arrays with low noise characteristics to enable detection of subtle spectral changes induced by varying charge carrier densities. The spectral resolution of the sensing unit 120 may be optimized to resolve peak shifts and lineshape changes on the order of 0.1 cm’1or better, providing the precision needed for quantitative material property determination.

[0046] System 100 includes a processing circuit 130 configured to analyze at least differences between Raman spectra acquired in relation to the power-different excitation light beams to determine material properties of the nanostructure. The processing circuit 130 may execute algorithms that compare spectral data from different power levels and apply models to extract material property information. The processing circuit 130 may implement advanced signal processing techniques including spectral deconvolution, peak fitting algorithms, and statistical analysis methods to extract quantitative information from complex spectral datasets. The processing capabilities may include real-time analysis for in-line process monitoring applications, with measurement-to-result times of less than 30 seconds for typical nanostructure characterization tasks.

[0047] The processing circuit 130 may be implemented using various types of computational hardware depending on the specific analysis requirements and performance specifications. In some cases, the processing circuit 130 may comprise a central processing unit (CPU) such as an Intel Core i7 or i9 processor, AMD Ryzen processor, or ARM-based processor for general-purpose computational tasks including spectral analysis algorithms, model fitting routines, and data management functions. The CPU implementation may provide flexibility for complex algorithmic operations and may be particularly suitable for sequential processing tasks such as iterative fitting algorithms and statistical analysis procedures.

[0048] In some cases, the processing circuit 130 may comprise a graphics processing unit (GPU) such as NVIDIA GeForce RTX series, NVIDIA Quadro series, AMD Radeon series, or specialized compute GPUs like NVIDIA Tesla or AMD Instinct processors for parallel processing of spectral data. The GPU implementation may provide significant acceleration for computationally intensive tasks such as electromagnetic fieldsimulations, charge carrier transport modeling, and machine learning algorithms used in spectral analysis. The parallel processing capabilities of GPUs may enable real-time analysis of large spectral datasets and may reduce processing time for complex modeling calculations from hours to minutes or seconds.

[0049] The processing circuit 130 may comprise field-programmable gate arrays (FPGAs) such as Xilinx Zynq series, Intel Altera series, or Microsemi SmartFusion devices for specialized signal processing applications requiring low latency and deterministic timing. FPGA implementations may provide customizable hardware acceleration for specific spectral processing algorithms and may enable real-time spectral analysis with microsecond-level response times. The reconfigurable nature of FPGAs may allow optimization of processing algorithms for specific nanostructure types or measurement protocols.

[0050] In some cases, the processing circuit 130 may comprise digital signal processors (DSPs) such as Texas Instruments TMS320 series, Analog Devices SHARC series, or ARM Cortex-M series processors optimized for signal processing operations. DSP implementations may provide efficient processing of spectral data with built-in mathematical functions for Fourier transforms, filtering operations, and statistical analysis. The specialized architecture of DSPs may enable power-efficient processing for portable or embedded measurement systems.

[0051] The processing circuit 130 may comprise arrays of processing circuits including multi-core CPU configurations, GPU clusters, or distributed computing systems for handling large-scale spectral analysis tasks. Multi-core CPU arrays may include configurations such as dual Intel Xeon processors, AMD EPYC processors, or ARM- based server processors providing 16, 32, 64, or more processing cores for parallel execution of analysis algorithms. GPU clusters may comprise multiple NVIDIA Tesla VI 00, Al 00, or Hl 00 GPUs connected through high-speed interconnects such as NVLink or InfiniBand for massively parallel processing of spectral datasets. Distributed computing arrays may include multiple processing nodes connected through Ethernet or specialized interconnects, enabling scalable processing capabilities for high-throughput measurement applications.

[0052] The processing circuit 130 may comprise hybrid processing architectures combining different types of processing units such as CPU-GPU combinations, FPGA- CPU systems, or heterogeneous computing platforms that leverage the strengths of different processing technologies. CPU-GPU hybrid systems may utilize CPUs for control and coordination tasks while offloading computationally intensive spectral analysis to GPUs for optimal performance. FPGA-CPU combinations may provide realtime signal processing capabilities through FPGAs while maintaining flexibility for algorithm development and system control through CPUs.

[0053] In some cases, the processing circuit 130 may comprise specialized artificial intelligence (Al) processors such as Google Tensor Processing Units (TPUs), Intel Neural Compute Sticks, or NVIDIA Jetson series processors optimized for machine learning applications in spectral analysis. Al processor implementations may provide enhanced capabilities for pattern recognition in spectral data, automated material property extraction, and adaptive measurement optimization based on learned patterns from previous measurements.

[0054] The processing circuit 130 may comprise cloud-based processing resources including Amazon Web Services (AWS) EC2 instances, Google Cloud Platform compute engines, or Microsoft Azure virtual machines that provide scalable computational resources for complex spectral analysis tasks. Cloud-based implementations may enable access to high-performance computing resources without requiring local hardware investments and may provide flexibility for handling varying computational loads based on measurement requirements.

[0055] System 100 further includes a control unit 140 for monitoring beam intensity and adjusting the illumination accordingly. The control unit 140 may include power monitoring sensors, feedback control electronics, and adjustment mechanisms to maintain precise control over the incident laser power. The control unit 140 may continuously monitor the actual laser power delivered to the sample and provide real-time corrections for any variations due to laser degradation, thermal effects, or changes in optical transmission. The control unit 140 may implement closed-loop feedback control with response times of less than 1 millisecond to maintain power stability during measurement sequences. The power monitoring system may utilize calibrated photodiodes or thermalpower sensors with traceability to national standards, ensuring measurement accuracy and repeatability across different systems and time periods.

[0056] The different power levels include a high power level that generates a significant density of free charge carriers in the nanostructure and a low power level that generates an insignificant density of free charge carriers in the nanostructure. The high power level may be within a non-linear power range in which there is a non-linear relationship between Raman spectra and power levels of the excitation light beams, while the low power level may be within a linear power range in which there is a linear relationship between Raman spectra and the power levels of the excitation light beams.

[0057] According to an embodiment, the transition between linear and non-linear regimes typically occurs when the photo-generated carrier density approaches 1016to 1017carriers per cubic centimeter, depending on the material properties and temperature. The high power level may generate carrier densities exceeding 1018carriers per cubic centimeter, sufficient to produce measurable changes in Raman spectrum (line shape) and intensity through Fano interference and carrier-induced absorption effects.

[0058] The different excitation light beams may include power-wavelength combination different excitation light beams of different combinations of power levels and wavelengths. The wavelengths may be selected from a group consisting of ultraviolet, green light, red light, blue light and infrared. The power-wavelength combination different excitation light beams may include multiple sub-sets of excitation light beams, where the sub-sets differ from each other by wavelength, and each sub-set includes excitation light beams that differ from each other by power. The wavelength selection may be optimized based on the optical absorption characteristics of the nanostructure materials, with shorter wavelengths providing enhanced surface sensitivity and longer wavelengths enabling deeper penetration. Typical wavelength ranges may include 325 nm, 405 nm, 488 nm, 532 nm, 633 nm, and 785 nm, each providing different penetration depths and sensitivity profiles for multilayer characterization.

[0059] The different excitation light beams may include polarization different excitation light beams of different polarization. The different excitation light beams may further include beam cross section different excitation light beams of cross section that include a charge diffusion dominant excitation light beam and a relaxation dominantexcitation light beam. The polarization control may enable selective excitation of different phonon modes and optimization of Raman scattering efficiency for specific crystallographic orientations. The beam cross section control may allow investigation of charge carrier transport properties by varying the spatial extent of excitation relative to the characteristic diffusion length of photo-generated carriers.

[0060] The processing circuit 130 may be configured to determine the material properties of the nanostructure by applying one or more models. The one or more models may include at least one of (a) an illumination model that models the illumination of the nanostructure, (b) a nanostructure model that models an impact of the illumination on the nanostructure, and (c) a collection and detection model that models the acquisition of the Raman spectra. The modeling approach may incorporate finite element analysis for electromagnetic field distribution calculations, drift-diffusion equations for charge carrier transport simulation, and quantum mechanical calculations for Raman scattering crosssections. The integrated modeling framework may enable prediction of measurement sensitivity and optimization of experimental parameters for specific metrology applications.

[0061] The processing circuit 130 may be configured to determine the material properties by simulating an electromagnetic field distribution inside the nanostructure, a local absorption and consequent electron hole excitation within the nanostructure, and an electron hole relaxation within the nanostructure. The processing circuit 130 may calculate a distribution of light in the nanostructure for given illumination parameters, simulate a generation of charge carriers in the nanostructure due to absorption of light, and simulate charge carriers consequent diffusion and drift due to internal electric fields within the nanostructure and recombination of the charge carriers. The simulation may incorporate material-specific parameters including absorption coefficients, carrier mobilities, diffusion constants, and recombination lifetimes to provide quantitative predictions of power-dependent spectral response.

[0062] When the nanostructure includes a sequence of nanosheets, the processing circuit 130 may be configured to determine the material properties based on an assumption that a steady-state density of charge carriers within the nanostructure is related to an absorption rate of the light inside the nanostructure. The steady-stateapproximation may be valid for continuous wave laser excitation when the measurement time is much longer than the carrier recombination lifetime, typically on the order of microseconds for silicon-based materials. The relationship between absorption rate and carrier density may be modulated by layer-specific recombination rates and inter-layer charge transfer processes.

[0063] The illumination module 110 may include multiple discrete lasers operating at different wavelengths. Multiple discrete lasers may provide wavelength flexibility for probing different regions of the nanostructure with varying penetration depths. Each discrete laser may be optimized for specific wavelength ranges and may include dedicated optics for each wavelength where needed, such as Raman filters or other optical elements. The different wavelengths may include wavelengths in a shortwavelength visible range and ultraviolet range for profiling different regions of the nanostructure. The discrete laser approach may provide superior power stability and spectral purity compared to tunable laser systems, with typical power stability better than 0.1% RMS over measurement periods of several hours.

[0064] In some cases, a tunable laser or supercontinuum laser may be used instead of multiple discrete lasers. A tunable laser system may provide more flexibility with choice of wavelengths, allowing continuous wavelength adjustment across a broad spectral range. A supercontinuum laser may generate a broad spectrum of wavelengths that can be selectively filtered for specific measurements. However, tunable and supercontinuum laser systems may have technical disadvantages compared to discrete laser systems, such as reduced power stability, increased complexity, or higher cost, which may make discrete laser systems more attractive for many applications. The wavelength tuning range for tunable laser systems may span 100-200 nm with tuning resolution of 0.1 nm or better, enabling fine optimization of penetration depth and sensitivity for specific nanostructure configurations.

[0065] The control unit 140 may be configured to control incident laser power on a sample to within 1% accuracy and provide continuous control over the incident laser power. The control unit 140 may maintain precise power levels needed to achieve reproducible measurements across the linear and nonlinear response regimes. The continuous control capability may allow smooth adjustment of power levels for acquiringspectra at different excitation conditions. The power control range may span three to four orders of magnitude, from microwatt levels for linear regime measurements to milliwatt levels for nonlinear regime characterization, with logarithmic or linear power stepping capabilities.

[0066] In some cases, the control unit 140 may be configured to control the incident laser power to within 0.3% accuracy. This enhanced accuracy may provide improved measurement precision for detecting subtle changes in spectral response due to varying free charge carrier densities. The higher accuracy power control may be particularly beneficial for quantitative analysis of material properties based on power- dependent spectral differences. The enhanced power control may be achieved through combination of laser current modulation and precision optical attenuators with closed-loop feedback control systems.

[0067] The control unit 140 may include a feedback mechanism configured to measure the incident laser power and correct for power drifts. The feedback mechanism may continuously monitor the actual laser power delivered to the sample and provide real-time corrections for any variations due to laser degradation, thermal effects, or changes in optical transmission. The feedback mechanism may include power monitoring sensors positioned in the optical path and control electronics that adjust laser output or optical attenuators to maintain the desired power level. The feedback system may compensate for power variations with time constants ranging from milliseconds for shortterm fluctuations to hours for long-term drift, ensuring measurement stability throughout extended characterization sequences.

[0068] The illumination module 110 may include optical elements configured to direct the excitation light to illuminate a nanostructure sample. The optical elements may include mirrors, lenses, beam splitters, and other components that shape and direct the laser beam to the sample location. The optical elements may be arranged to provide efficient light delivery while maintaining beam quality and spatial characteristics needed for precise measurements. The optical system may achieve numerical apertures ranging from 0.1 to 0.9 depending on the required spatial resolution and collection efficiency, with aberration correction to maintain diffraction-limited performance across the full field of view.

[0069] The optical elements may include beam shaping elements configured to control a point spread function of the excitation light on the nanostructure sample. The beam shaping elements may include apertures, spatial filters, beam expanders, or other optical components that modify the spatial distribution of the laser beam. The point spread function control may ensure consistent illumination conditions across different measurements and may enable adjustment of the illuminated area to optimize sensitivity to charge diffusion or recombination processes. The beam shaping capability may allow spot size variation from sub-micron to tens of microns, enabling investigation of charge carrier transport properties over different length scales relative to the characteristic diffusion length.

[0070] The sensing unit 120 may be configured to acquire Raman spectra from the nanostructure sample. The sensing unit 120 may be sensitive to the wavelengths of Raman-scattered light and may provide sufficient spectral resolution to detect changes in peak positions, widths, and intensities that correlate with material properties. The sensing unit 120 may include a spectrometer with appropriate gratings, filters, and photodetector arrays to capture the full Raman spectral range of interest. The spectral coverage may span 100 to 4000 cm’1Raman shift with resolution better than 1 cm’1, enabling simultaneous monitoring of multiple phonon modes and their power-dependent response characteristics.

[0071] The processing circuit 130 may implement model-based analysis using a modeling engine that can predict charge carrier excitation, diffusion processes, and spectrum impact. The modeling engine may allow optimization of measurement sequences by determining the most effective combination of power levels, wavelengths, and other parameters for specific metrology challenges. The modeling engine may also provide correct interpretation of the resulting spectral signals by relating observed spectral changes to underlying physical processes. The modeling capabilities may include Monte Carlo simulation of charge carrier transport, finite difference time domain electromagnetic field calculations, and quantum mechanical modeling of electron-phonon interactions to provide comprehensive understanding of the measurement physics.

[0072] The simulation model implemented by the processing circuit 130 may include an illumination model that models the illumination of the sample. The illumination modelmay calculate the electromagnetic field distribution inside the nanostructure for given illumination parameters such as wavelength, power, spot size, and angle of incidence. The illumination model may account for the optical properties of the different layers in the multilayer structure and their interfaces. The electromagnetic field calculations may incorporate the complex refractive indices of Silicon and Silicon-Germanium as functions of wavelength and temperature, interface roughness effects, and coherent interference phenomena in thin film structures.

[0073] The simulation model may include a sample model that models the impact of illumination on the sample, including generation of electron-hole pairs and formation of multiple virtual local Raman sources. The sample model may simulate the generation of charge carriers in the material due to absorption of light, their subsequent diffusion and drift due to internal electric fields, and their recombination processes. The sample model may treat different regions of the nanostructure as multiple virtual local Raman sources, each contributing to the overall spectral response based on local material properties and charge carrier densities. The charge carrier dynamics simulation may incorporate Auger recombination, Shockley-Read-Hall recombination, and radiative recombination mechanisms with temperature-dependent rate constants.

[0074] The simulation model may include a collection and detection model that models the signals sensed by the sensing unit 120. The collection and detection model may account for how Raman spectra generated by the multiple virtual local Raman sources are collected by the optical system and detected. The collection and detection model may include the effects of the collection optics, spectral filtering, and detector response characteristics. The model may incorporate the wavelength-dependent transmission of optical elements, detector quantum efficiency, and noise characteristics to provide accurate prediction of signal-to-noise ratios and measurement precision.

[0075] The processing circuit 130 may implement machine learning analysis tools that can be combined with the power-dependent Raman approach for nano-profiling of samples. Machine learning algorithms may be trained on spectral data to recognize patterns that correlate with specific material properties or structural characteristics. The machine learning analysis tools may enhance the ability to extract quantitative information from complex spectral datasets and may improve the accuracy of materialproperty determinations. The machine learning implementation may include neural networks, support vector machines, and random forest algorithms trained on both experimental data and simulation results to provide robust property extraction capabilities across diverse nanostructure configurations.

[0076] Dimensional metrology may be performed using Raman peak intensities which are sensitive to dimensional properties of the nanostructure. The measured intensities of Raman peaks may be directly related to dimensional properties such as nanosheet thickness and critical dimensions in patterned samples. The processing circuit 130 may analyze intensity variations across different power levels and wavelengths to extract dimensional information in addition to material properties. The dimensional sensitivity may arise from interference effects in thin film structures and volumedependent scattering contributions, enabling thickness measurements with precision approaching 0.1 nanometer for individual layers in multilayer stacks.

[0077] Examples of illumination modules and / or sensing units are illustrated in US patent application serial number 16 / 062,114 titled " Raman spectroscopy based measurements in patterned structures" which is incorporated herein in its entirety, US patent application serial number 16 / 613,448, titled " Raman spectroscopy based measurement system", US patent application serial number 18 / 452,494 titled "Accurate Raman spectroscopy" which is incorporated herein in its entirety, and US patent application serial number 18 / 003,801 titled " Systems and methods for optical metrology" which is incorporated herein in its entirety.

[0078] The illumination module 110 may include one or more radiation sources such as solid-state lasers, diode lasers, or gas lasers operating at wavelengths including 325 nm, 405 nm, 488 nm, 532 nm, 633 nm, and 785 nm., and / or may include one or more apertures for beam size control, and / or may include one or more polarizers for polarization state adjustment, and / or may include one or more mirrors for beam steering, and / or may include one or more beam splitters for separating excitation and collection paths, and / or may include one or more focusing lenses, and / or may include one or more collimating optics, and / or may include one or more neutral density filters for power attenuation and / or may include one or more sensors such as photodiodes for power monitoring, CCD or CMOS detector arrays for spectral detection, and / or may include oneor more thermal power sensors for calibration and / or may include one or more spectrometers with diffraction gratings, and / or may include one or more entrance slits, and / or may include one or more wavelength-selective filters for spectral analysis and / or may include one or more memory units (denoted 150 in figure 5) for storing spectral data, calibration parameters, and measurement protocols.

[0079] Memory unit 150 may include one or more types of storage devices including volatile memory such as dynamic random access memory (DRAM) and static random access memory (SRAM), non-volatile memory such as flash memory and electrically erasable programmable read-only memory (EEPROM), non-volatile random access memory (NVRAM), solid-state drives (SSD), and dynamic memory units that may provide high-speed data access for real-time spectral processing and analysis.

[0080] The illumination method 110 may illuminate the nanostructure at one or more angles of incidence and / or at one or more polarization and / or at one or more wavelengths and / or at one or more power levels.

[0081] Method for Nanostructure Characterization

[0082] Referring to Figure 6, a method 200 for nanostructure characterization using Raman spectroscopy is illustrated. Method 200 begins with step 210 of illuminating the nanostructure, during a set of Raman measurement iterations, with different excitation light beams that include power-different excitation light beams of different power levels, and with different measurement light beams. The method provides systematic characterization capabilities that enable extraction of layer-specific material properties from multilayer nanostructures through controlled variation of excitation conditions and analysis of resulting spectral changes.

[0083] Method 200 continues with step 220 of acquiring, during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure by the different measurement light beams. The different power levels include a high power level that generates a significant density of free charge carriers in the nanostructure and a low power level that generates an insignificant density of free charge carriers in the nanostructure. The spectral acquisition may be performed with integration times optimized for each power level to maintain consistent signal-to-noise ratios while avoiding detector saturation or thermal damage to the sample.

[0084] Method 200 proceeds with step 230 of analyzing at least differences between Raman spectra acquired in relation to the power-different excitation light beams to determine material properties of the nanostructure. The analyzing step 230 may involve applying various models and algorithms to extract material property information from the spectral differences. The analysis may incorporate statistical methods to quantify measurement uncertainty and provide confidence intervals for extracted material parameters, enabling process control applications with well-defined measurement precision.

[0085] Referring to Figure 4, determining properties of individual layers in the multilayer nanostructure may be based on analysis of differences between the first and second Raman spectra. Figure 4 may show Raman peak intensity distributions comparing low power and high power excitation conditions for a nanosheet structure. The figure may display intensity curves for contributions from a top nanosheet, middle nanosheet, bottom nanosheet, and the total combined response. The differential analysis approach enables separation of overlapping spectral contributions from different layers by exploiting the depth-dependent nature of charge carrier generation and transport.

[0086] As shown in Figure 4, under low power conditions, the contributions from the top, middle, and bottom nanosheets may be relatively balanced in the total spectrum. Under high power conditions, the contribution from the top nanosheet may be significantly increased compared to the middle and bottom nanosheets. The difference between the high power and low power measurements, after normalization to the laser power, may be dominated by the top nanosheet response to the induced charge carriers. This power-dependent selectivity enables extraction of top layer properties with enhanced sensitivity while maintaining information about the overall structure through the low power measurement.

[0087] The properties of individual layers may include strain, Germanium composition, free charge carrier lifetime, or diffusion properties for each layer. The first measurement may be used to characterize average properties across the structure, such as average Germanium composition or strain. Combining the high-power and low-power measurements may allow differentiation between top and buried regions of the multilayer structure. The strain determination may achieve precision of 0.01% or better through 1analysis of Raman peak frequency shifts, while Germanium composition measurements may provide accuracy of 1% absolute composition or better through peak intensity ratio analysis.

[0088] The measurement may be adapted to probe selectively charge relaxation or diffusion by controlling the measurement spot dimensions. When an illumination spot is small relative to a typical diffusion length, charge diffusion may become dominant as free charges can escape the laser spot region. The diffusion length may be defined as the square root of the product of diffusivity (D) and lifetime (tau), representing the typical distance a charge carrier may diffuse before recombination. For silicon- based materials, typical diffusion lengths may range from 1 to 100 micrometers depending on material quality and doping levels, enabling selective probing of transport properties through spot size control.

[0089] When using a large illumination spot relative to the diffusion length, charges may typically recombine before reaching the spot edge, making recombination processes more dominant than diffusion. A controlled aperture or other optical mechanism may be used to change the extent of the illumination spot, offering flexible measurement of free charge lifetime or diffusion properties. The spot size variation capability may enable determination of carrier mobility and lifetime with precision comparable to electrical measurement techniques while maintaining the non-destructive and spatially resolved advantages of optical characterization.

[0090] Additional measurements may vary excitation wavelength, polarization at illumination and collection, and angle of incidence to affect the field distribution inside the structure. Changing the polarization, angle of incidence, and azimuth may lead to different distributions of light inside the structure due to the nature of light-matter interaction. This may be particularly relevant for patterned structures that have different optical properties in different directions. The angular and polarization control may enable selective excitation of specific phonon modes and optimization of sensitivity to crystallographic orientation and strain tensor components.

[0091] Combining information from multiple measurements with different illumination parameters may allow extensive mapping and detailed characterization of the measured structure. Adding more measurement states may create additionalcombinations of contributions from each nanosheet, which may improve the individual measurements of concentration, lifetime, and other properties of each layer. The multiparameter measurement approach may enable determination of complete material property profiles across multilayer structures with layer-by-layer resolution approaching the physical thickness of individual nanosheets.

[0092] Different wavelengths may be used for Raman metrology and free charge excitation to break the coupling between excitation and measurement effects. When using the same laser wavelength for measuring the Raman spectrum and creating photo-excited charge carriers, the two effects may be inherently coupled. By exciting and measuring with different wavelengths, this coupling may be broken, allowing more control over the different contributions of individual nanosheets and better separation of their individual properties. The wavelength decoupling approach may enable independent optimization of charge carrier generation efficiency and Raman scattering sensitivity, providing enhanced measurement flexibility and improved property extraction accuracy.

[0093] Nonlinear absorption effects may be utilized where free charges increase light absorption, resulting in reduced field penetration at higher incident power. When free charges are excited, the light absorption may typically be increased, and consequently, increased incident power may result in reduced field penetration into the nanostructure. This nonlinear absorption may provide enhanced profiling selectivity, where at low power penetration may be large and an extended span of the structure may be measured, while for higher power penetration may be reduced and only near-surface regions may be measured. The nonlinear absorption mechanism may enable depth profiling with resolution better than the optical absorption length, providing enhanced surface sensitivity for characterizing thin surface layers and interfaces.

[0094] Referring to figure 2C which illustrates experimental power-dependent Raman response data may demonstrate the nonlinear optical response induced by free charge carriers in silicon. Figure 2C shows the effect of free-charge carriers on the Raman signal of silicon through laser excitation, with Raman intensity plotted as a function of Raman shift in wavenumbers. The graph may display three overlapping spectral curves representing measurements at different laser power levels: 81 mW, 27mW, and 5mW. The experimental data provides validation of the theoretical frameworkunderlying the power-dependent characterization approach and demonstrates the practical feasibility of the technique for semiconductor materials.

[0095] As shown in Figure 2C, all three curves may exhibit a peak centered around 520 cm'1, with the peak shape and intensity varying depending on the laser power used. The higher power measurements may show broader peak profiles and increased asymmetry compared to the lower power measurement. The 81 mW measurement may demonstrate the most pronounced spectral changes, while the 5mW measurement may represent conditions where free charge carrier effects are minimal. The systematic progression of spectral changes with increasing power provides quantitative evidence for the power-dependent response mechanism and enables calibration of the relationship between excitation conditions and charge carrier density.

[0096] The experimental data in Figure 2C may demonstrate that the Raman spectrum changes shape depending on the laser intensity rather than simply scaling in magnitude. At the 5mW power level, the spectrum may represent a linear response regime where the spectral shape remains relatively unchanged. At the 27mW and 81 mW power levels, the nonlinear response may become apparent through the broader peak profiles and asymmetric lineshapes caused by laser-induced free charge carrier generation. The spectral lineshape changes may be quantified through analysis of peak width, asymmetry parameters, and frequency shifts, providing multiple independent measures of charge carrier effects that enhance measurement robustness and accuracy.

[0097] Referring back to figure 6, method 200 may further include controlling a point spread function of the excitation light beams on the nanostructure. The point spread function distribution may be controlled to a fixed spatial extent, possibly using beam shaping elements throughout an optical path. In some cases, the spot size may be controlled to within approximately 1% accuracy to ensure consistent illumination conditions across measurements. The spatial control capability may enable investigation of charge carrier transport properties over different length scales and optimization of sensitivity to specific physical processes such as diffusion or recombination.

[0098] Method 200 may further include controlling incident laser power to an accuracy of less than 1%. The incident laser power control may provide continuous control over the incident laser power to maintain measurement consistency. A feedbackmechanism may be implemented to measure the incident laser power and allow correction for any power drifts due to degradation of laser power or optical transmission. The power control accuracy may be critical for quantitative analysis of power-dependent spectral changes and extraction of material properties with well-defined measurement uncertainty.

[0099] In some cases, the incident laser power may be controlled to an accuracy of less than 0.3%. This enhanced power control accuracy may provide improved measurement precision and repeatability for characterizing material properties based on power-dependent spectral changes. The enhanced accuracy may be particularly important for detecting subtle differences in material properties between similar samples or for monitoring small process variations in manufacturing environments.

[0100] The excitation light beams may have narrow and stable spectral linewidth to prevent smearing of sharp peaks in the Raman spectrum. The narrow spectral linewidth may be maintained to preserve the spectral resolution needed to detect subtle changes in peak positions and lineshapes that correlate with material properties. The stable spectral linewidth may ensure consistent measurement conditions across different power levels and measurement sequences. Typical laser linewidth requirements may be less than 0.1 cm'1to maintain spectral resolution sufficient for quantitative peak analysis and material property extraction.

[0101] The laser power may be stable with small residual instability accounted for by monitoring the laser power in parallel with the Raman measurements. A power monitoring system may measure the incident laser power continuously during spectral acquisition and scale the results accordingly to compensate for any power variations. This parallel monitoring approach may correct for power drifts and ensure accurate normalization of spectral data across different power levels. The power monitoring may achieve temporal resolution of microseconds or better to capture and correct for rapid power fluctuations during measurement sequences.

[0102] Method 200 may further include additional step 240 of illuminating the nanostructure with additional excitation light beams at different power levels and step 250 of acquiring additional Raman spectra. These additional measurements may provide more data points for analyzing the power-dependent response of the nanostructure andmay enhance the determination of material properties. The multi-point power series measurements may enable construction of power-response curves that provide enhanced sensitivity to material property variations and improved statistical confidence in extracted parameters.

[0103] Method 200 may include step 260 of determining the material properties based on multiple Raman spectra acquired at different power levels. The analysis of multiple spectra may enhance the determination of material properties by providing additional constraints for fitting models or algorithms used to extract material parameters. The combination of multiple power-dependent measurements may improve the accuracy and reliability of strain, Germanium composition, free charge carrier lifetime, or diffusion property determinations. The multi-spectrum analysis may enable detection of material property variations with precision approaching the fundamental limits imposed by measurement noise and sample heterogeneity.

[0104] Method 200 for profiling individual layers may further include illuminating the multilayer nanostructure with additional laser wavelengths having different penetration depths. The additional laser wavelengths may provide different sensitivity profiles across the multilayer structure, with each wavelength probing different combinations of layers based on its penetration characteristics. The selection of multiple wavelengths may be optimized to maximize the ability to distinguish contributions from individual layers. The wavelength selection strategy may be based on modeling of electromagnetic field distribution and absorption characteristics to ensure optimal sensitivity to each layer in the multilayer stack.

[0105] Method 200 may further include acquiring additional Raman spectra at the high power level and the low power level for each additional laser wavelength. Each wavelength may be used to acquire both high power and low power spectra, providing multiple sets of spectral differences that reflect the power-dependent response at different penetration depths. The additional spectral data may increase the amount of information available for analyzing individual layer properties. The multi-wavelength, multi-power measurement matrix may provide sufficient constraints to enable simultaneous determination of multiple material properties for each layer in complex multilayer structures.

[0106] Method 200 may further include combining analysis of differences between Raman spectra from multiple wavelengths to enhance profiling resolution of the individual layers. The combination of spectral differences from multiple wavelengths may provide additional constraints for separating the contributions of individual layers in the multilayer structure. The enhanced profiling resolution may enable more accurate determination of properties for each layer by reducing correlations between layer contributions and improving the ability to distinguish subtle differences in material characteristics across the vertical structure. The multi-wavelength analysis approach may achieve depth resolution better than 1 nanometer for individual layer characterization in optimized measurement configurations, enabling precise process control for advanced semiconductor manufacturing applications.

[0107] While method 6 illustrates method 200 that include illuminating the substrate, and obtaining Raman spectra, there may be provided a method for analyzing Raman spectra obtained by steps 210 and 220 of method 200.

[0108] Figure 7 illustrates method 300 for nanostructure characterization using Raman spectroscopy.

[0109] Method 300 includes step 310 of receiving Raman spectra. The Raman spectra were generated by executing step 210 and step 220 of method 200.

[0110] Step 310 is followed by step 230 of method 300. Step 320 includes analyzing at least differences between Raman spectra acquired in relation to the power-different excitation light beams to determine material properties of the nanostructure.

[0111] Method 300 is executed by a computerized system that includes one or more processing circuits and one or more memory and / or storage systems. The one or more processing circuit include one or more integrated circuits and maybe graphic processing units and / or central processing units, and the like.

[0112] A number of implementations have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the disclosure. Accordingly, other implementations are within the scope of the following claims.

[0113] Any reference to the term "comprising" or "having" should be applied, mutatis mutandis to "consisting of' or "essentially consisting of".

[0114] Any reference to a system should be applied, mutatis mutandis to a method executable by the system and / or should be applied, mutatis mutandis to a non-transitory computer readable medium that stores instructions executable by the system.

[0115] Any reference to a method system should be applied, mutatis mutandis to a system configured to execute the method and / or should be applied, mutatis mutandis to a non-transitory computer readable medium that stores instructions executable by the system.

[0116] Any reference to a computer readable medium should be applied, mutatis mutandis to a method executed based on instructions stored in the computer readable medium and / or should be applied, mutatis mutandis to a system configured toe xecute the instructions.

[0117] In the foregoing specification, the invention has been described with reference to specific examples of embodiments of the invention. It will, however, be evident that various modifications and changes may be made therein without departing from the broader spirit and scope of the invention as set forth in the appended claims.

[0118] Moreover, the terms "front, " "back, " "top, " "bottom, " "over, " "under " and the like in the description and in the claims, if any, are used for descriptive purposes and not necessarily for describing permanent relative positions. It is understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the invention described herein are, for example, capable of operation in other orientations than those illustrated or otherwise described herein.

[0119] Any arrangement of components to achieve the same functionality is effectively "associated" such that the desired functionality is achieved. Hence, any two components herein combined to achieve a particular functionality can be seen as "associated with " each other such that the desired functionality is achieved, irrespective of architectures or intermedial components. Likewise, any two components so associated can also be viewed as being "operably connected, " or "operably coupled, " to each other to achieve the desired functionality.

[0120] Furthermore, those skilled in the art will recognize that boundaries between the above described operations are merely illustrative. The multiple operations may be combined into a single operation, a single operation may be distributed in additionaloperations and operations may be executed at least partially overlapping in time. Moreover, alternative embodiments may include multiple instances of a particular operation, and the order of operations may be altered in various other embodiments.

[0121] While certain features of the invention have been illustrated and described herein, many modifications, substitutions, changes, and equivalents will now occur to those of ordinary skill in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.

Claims

CLAIMS1. A Raman spectroscopy system for nanostructure characterization, comprising: an illumination module that is configured to illuminate the nanostructure, during a set of Raman measurement iterations, with light beams that comprise a light beam of a high power level and a light beam of a low power level a sensing unit configured to acquire, during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; and a processing circuit configured to analyze at least differences between Raman spectra acquired in relation to the light beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

2. The Raman spectroscopy system according to claim 1, wherein the high power level belongs to a non-linear power range in which there is a non-linear relationship between Raman spectra and power levels of the excitation light beams, wherein the low power level belongs to a linear power range in which there is a linear relationship between Raman spectra and the power levels of the excitation light beams.

3. The Raman spectroscopy system according to claim 1, wherein the light beams comprise different excitation light beams of different power levels, the different power levels comprise the high power level and the low power level.

4. The Raman spectroscopy system according to claim 3, wherein the light beams further comprise one or more measurement light beams.

5. The Raman spectroscopy system according to claim 4, wherein the different excitation light beams comprise power-wavelength combination different excitation light beams of different combinations of power levels and wavelengths; wherein the processing circuit is further configured to analyze at least Raman spectra acquired in relation to the power- wavelength combination different excitation light beams to determine the material properties of the nanostructure.

6. The Raman spectroscopy system according to claim 5, wherein the wavelengths are selected of a group consisting of ultraviolet, green light, red light, blue light and infrared.

7. The Raman spectroscopy system according to claim 5, wherein power-wavelength combination different excitation light beams comprise multiple sub-sets of excitation light beams, the sub-sets of excitation light beams differ from each other by wavelength, wherein each sub-set comprises excitation light beams that differ from each other by power.

8. The Raman spectroscopy system according to claim 4, wherein the different excitation light beams comprise polarization different excitation light beams of different polarization; wherein the processing circuit is further configured to analyze at least Raman spectra acquired in relation to the polarization different excitation light beams to determine the material properties of the nanostructure.

9. The Raman spectroscopy system according to claim 4, wherein the different excitation light beams further comprise beam cross section different excitation light beams of cross section that comprise a charge diffusion dominant excitation light beam and a relaxation dominant excitation light beam, wherein the processing circuit is further configured to analyze at least Raman spectra acquired in relation to the beam cross section different excitation light beams to determine the material properties of the nanostructure.

10. The Raman spectroscopy system according to claim 4, wherein processing circuit is configured to determine the material properties of the nanostructure by applying one or more models.

11. The Raman spectroscopy system according to claim 10, wherein the one or more models comprise at least one of (a) an illumination model that models the illumination of the nanostructure, (b) a nanostructure model that models an impact of the illumination on the nanostructure, and (c) a collection and detection model that models the acquisition of the Raman spectra.

12. The Raman spectroscopy system according to claim 4, wherein the processing circuit is configured to determine the material properties of the nanostructure by simulating an electromagnetic field distribution inside the nanostructure, a local absorption and consequent electron hole excitation within the nanostructure, and an electron hole relaxation within the nanostructure.

13. The Raman spectroscopy system according to claim 4, wherein the processing circuit is configured to determine the material properties of the nanostructure by calculating a distribution of light in the nanostructure for given illumination parameters, simulating a generation of charge carriers in the nanostructure due to absorption of light, and simulating charge carriers consequent diffusion and drift due to internal electric fields within the nanostructure and recombination of the charge carriers.

14. The Raman spectroscopy system according to claim 4, wherein the nanostructure comprises a sequence nanosheets, wherein the processing circuit is configured to determine the material properties of the nanostructure based on an assumption that a steady-state density of charge carriers within the nanostructure is related to an absorption rate of the light inside the nanostructure.

15. The Raman spectroscopy system according to claim 4, wherein the different excitation light beams of different power levels impact charge carriers at different depths of the nanostructure and wherein the processing circuit is configured to determine material properties of the nanostructure at the different depths.

16. The Raman spectroscopy system according to claim 1, further comprising a control unit configured to monitor an intensity of at least one of the light beams and adjust an intensity of one or more next light beams based on the monitoring.

17. The Raman spectroscopy system according to claim 1, wherein the nanostructure belongs to a semiconductor wafer.

18. A method for Raman spectroscopy for nanostructure characterization, comprising: illuminating, by an illumination module, the nanostructure, during a set of Raman measurement iterations, with light beams that comprise a light beam of a high power level and a light beam of a low power level, acquiring by a sensing unit and during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; analyzing by a processing circuit, at least differences between Raman spectra acquired in relation to the light beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

19. The method according to claim 18, wherein the high power level belongs to a non-linear power range in which there is a non-linear relationship between Raman spectra and power levels of the excitation light beams, wherein the low power level belongs to a linear power range in which there is a linear relationship between Raman spectra and the power levels of the excitation light beams.

20. The method according to claim 18, wherein the light beams comprise different excitation light beams of different power levels, the different power levels comprise the high power level and the low power level.

21. The method according to claim 20, wherein the light beams further comprise one or more measurement light beams.

22. The method according to claim 21, wherein the different excitation light beams comprise power-wavelength combination different excitation light beams of different combinations of power levels and wavelengths; wherein the analyzing further comprises analyzing at least Raman spectra acquired in relation to the power-wavelength combination different excitation light beams to determine the material properties of the nanostructure.

23. The method according to claim 22, wherein the wavelengths are selected of a group consisting of ultraviolet, green light, red light, blue light and infrared.

24. The method according to claim 22, wherein power-wavelength combination different excitation light beams comprise multiple sub-sets of excitation light beams, the sub-sets of excitation light beams differ from each other by wavelength, wherein each sub-set comprises excitation light beams that differ from each other by power.

25. The method according to claim 21, wherein the different excitation light beams comprise polarization different excitation light beams of different polarization; wherein the analyzing further comprises analyzing at least Raman spectra acquired in relation to the polarization different excitation light beams to determine the material properties of the nanostructure.

26. The method according to claim 21, wherein the different excitation light beams further comprise beam cross section different excitation light beams of cross section that comprise a charge diffusion dominant excitation light beam and a relaxation dominant excitation light beam, wherein the analyzing further comprises analyzing at least Ramanspectra acquired in relation to the beam cross section different excitation light beams to determine the material properties of the nanostructure.

27. The method according to claim 21, wherein the analyzing comprises determining the material properties of the nanostructure by applying one or more models.

28. The method according to claim 28, wherein the one or more models comprise at least one of (a) an illumination model that models the illumination of the nanostructure, (b) a nanostructure model that models an impact of the illumination on the nanostructure, and (c) a collection and detection model that models the acquisition of the Raman spectra.

29. The method according to claim 21, wherein the analyzing comprises determining the material properties of the nanostructure by simulating an electromagnetic field distribution inside the nanostructure, a local absorption and consequent electron hole excitation within the nanostructure, and an electron hole relaxation within the nanostructure.

30. The method according to claim 21, wherein the analyzing comprises determining the material properties of the nanostructure by calculating a distribution of light in the nanostructure for given illumination parameters, simulating a generation of charge carriers in the nanostructure due to absorption of light, and simulating charge carriers consequent diffusion and drift due to internal electric fields within the nanostructure and recombination of the charge carriers.

31. The method according to claim 21 , wherein the nanostructure comprises a sequence nanosheets, wherein the analyzing comprises determining the material properties of the nanostructure based on an assumption that a steady-state density of charge carriers within the nanostructure is related to an absorption rate of the light inside the nanostructure.

32. The method according to claim 21 , wherein the different excitation light beams of different power levels impact charge carriers at different depths of the nanostructure and wherein the processing circuit is configured to determine material properties of the nanostructure at the different depths.

33. The method according to claim 18, further comprising monitoring, by a control unit, an intensity of at least one of the light beams and adjusting an intensity of one or more next light beams based on the monitoring.

34. The method according to claim 18, wherein the nanostructure belongs to a semiconductor wafer.

35. A non-transitory computer readable medium that stores instructions for nanostructure characterization using Raman spectroscopy, wherein the execution of the instructions by a Raman spectroscopy system results in: illuminating, by an illumination module of the Raman spectroscopy system, the nanostructure, during a set of Raman measurement iterations, with light beams that comprise a light beam of a high power level and a light beam of a low power level, acquiring by a sensing unit of the Raman spectroscopy system and during the set, Raman spectra related to radiation received from the nanostructure following the illumination of the nanostructure; wherein the high power level generates a significant density of free charge carriers in the nanostructure and the low power level generates an insignificant density of free charge carriers in the nanostructure; analyzing by a processing circuit of the Raman spectroscopy system, at least differences between Raman spectra acquired in relation to the light beam of the low power level and in relation to the light beam of the high power level to determine material properties of the nanostructure.

36. The non-transitory computer readable medium according to claim 35, wherein the high power level belongs to a non-linear power range in which there is a non-linear relationship between Raman spectra and power levels of the excitation light beams, wherein the low power level belongs to a linear power range in which there is a linear relationship between Raman spectra and the power levels of the excitation light beams.

37. The non-transitory computer readable medium according to claim 35, wherein the light beams comprise different excitation light beams of different power levels, the different power levels comprise the high power level and the low power level.

38. The non-transitory computer readable medium according to claim 37, wherein the light beams further comprise one or more measurement light beams.

39. The non-transitory computer readable medium according to claim 38, wherein the different excitation light beams comprise power- wavelength combination different excitation light beams of different combinations of power levels and wavelengths; wherein the analyzing further comprises analyzing at least Raman spectra acquired in relation to the power- wavelength combination different excitation light beams to determine the material properties of the nanostructure.

40. The non-transitory computer readable medium according to claim 39, wherein the wavelengths are selected of a group consisting of ultraviolet, green light, red light, blue light and infrared.

41. The non-transitory computer readable medium according to claim 39, wherein power-wavelength combination different excitation light beams comprise multiple subsets of excitation light beams, the sub-sets of excitation light beams differ from each other by wavelength, wherein each sub-set comprises excitation light beams that differ from each other by power.

42. The non-transitory computer readable medium according to claim 38, wherein the different excitation light beams comprise polarization different excitation light beams of different polarization; wherein the analyzing further comprises analyzing at least Raman spectra acquired in relation to the polarization different excitation light beams to determine the material properties of the nanostructure.

43. The non-transitory computer readable medium according to claim 38, wherein the different excitation light beams further comprise beam cross section different excitation light beams of cross section that comprise a charge diffusion dominant excitation light beam and a relaxation dominant excitation light beam, wherein the analyzing further comprises analyzing at least Raman spectra acquired in relation to the beam cross section different excitation light beams to determine the material properties of the nanostructure.

44. The non-transitory computer readable medium according to claim 38, wherein the analyzing comprises determining the material properties of the nanostructure by applying one or more models.

45. The non-transitory computer readable medium according to claim 44, wherein the one or more models comprise at least one of (a) an illumination model that models the illumination of the nanostructure, (b) a nanostructure model that models an impact of the31illumination on the nanostructure, and (c) a collection and detection model that models the acquisition of the Raman spectra.

46. The non-transitory computer readable medium according to claim 38, wherein the analyzing comprises determining the material properties of the nanostructure by simulating an electromagnetic field distribution inside the nanostructure, a local absorption and consequent electron hole excitation within the nanostructure, and an electron hole relaxation within the nanostructure.

47. The non-transitory computer readable medium according to claim 38, wherein the analyzing comprises determining the material properties of the nanostructure by calculating a distribution of light in the nanostructure for given illumination parameters, simulating a generation of charge carriers in the nanostructure due to absorption of light, and simulating charge carriers consequent diffusion and drift due to internal electric fields within the nanostructure and recombination of the charge carriers.

48. The non-transitory computer readable medium according to claim 38, wherein the nanostructure comprises a sequence nanosheets, wherein the analyzing comprises determining the material properties of the nanostructure based on an assumption that a steady-state density of charge carriers within the nanostructure is related to an absorption rate of the light inside the nanostructure.

49. The non-transitory computer readable medium according to claim 38, wherein the different excitation light beams of different power levels impact charge carriers at different depths of the nanostructure and wherein the processing circuit is configured to determine material properties of the nanostructure at the different depths.

50. The non-transitory computer readable medium according to claim 36, further comprising monitoring, by a control unit, an intensity of at least one of the light beams and adjusting an intensity of one or more next light beams based on the monitoring.

51. The non-transitory computer readable medium according to claim 36, wherein the nanostructure belongs to a semiconductor wafer.

Images