Method and apparatus for measuring lateral depth in microstructures

Abstract

The method of the invention applies spectroscopic measurements to determine a lateral recess depth in a sidewall of a microstructure. The structure is formed on a larger substrate and the sidewall is in an upright position with respect to the substrate, and the recess extends substantially parallel to the substrate. The recess can be an etch depth. An energy beam is directed at the structure. An incident beam falling on the structure produces a spectral response. The response includes one or more peaks associated with one or more materials of the substrate and the structure. A parameter indicative of the lateral recess depth is derived from the one or more peaks, the parameter having a previously established one-to-one relationship with the depth. The depth is derived from the measured parameter using the previously established relationship.

Description

Technical Field

[0001] This invention relates to the field of metrology for patterned microstructures produced in semiconductor processing. In the context of this specification, a microstructure is a structure having dimensions of various characteristics, such as the width and length of fins, on the order of nanometers to one micrometer or more. Background Technology

[0002] The semiconductor fabrication industry has been driven by scaling of devices, cells, and functions to maintain steady growth in performance and cost per unit area, making device architectures below 10 nm a reality for high-volume manufacturing (HVM). However, due to their extremely small nanoscale dimensions, the fabrication of these devices requires precise control over their geometry, especially the critical dimension (CD). Meanwhile, technology nodes at 4 nm and above will see an increased demand for 3D metrology of buried information. In the case of lateral cavity etching, a certain amount of material is removed laterally, for example, beneath nitride and / or oxide stacks, which significantly hinders precise control over etching depth parameters. This monitoring is typically accomplished through metrological steps involving critical dimension scanning electron microscopy (CD-SEM) or optical critical dimension (OCD), but CD-SEM is insensitive to this buried information, and OCD relies on complex models where many geometric parameters of the compound (oxide, nitride, polysilicon, etc.) stacks may be related to the cavity depth, making the latter impossible to determine uniquely by fitting. Summary of the Invention

[0003] The objective of this invention is to provide a solution to the problem stated in the preceding paragraph.

[0004] The method of this invention uses spectroscopic measurements to determine the depth of a lateral recess in the sidewalls of a microstructure. The structure is formed on a larger substrate, and the sidewalls are in an upright position relative to the substrate, with the recess extending substantially parallel to the substrate. The recess can be an etch depth obtained by etching a first layer relative to two adjacent layers, each oriented parallel to the substrate, with the etching process progressing inward from the sidewalls. According to the method of this invention, an energy beam is directed onto the substrate. The incident beam falling on the structure produces a spectral response that is captured and processed by a detector and a processing unit, respectively. This response includes one or more peaks associated with one or more materials of the substrate and the structure. According to the invention, a parameter representing the depth of the lateral recess is derived from said one or more peaks, the parameter having a previously established one-to-one relationship with said depth. The depth is derived from the measured parameter using the previously established relationship.

[0005] This method is rapid and, in most embodiments, non-invasive, thus representing a way to perform metrology on buried structures without the drawbacks of prior art such as CD-SEM or OCD. This method can be used for in-line measurement of lateral depth in semiconductor fabrication lines.

[0006] The present invention particularly relates to a method for measuring the depth of a lateral recess in a microstructure located on a substrate, the structure having an elongated sidewall in an upright position relative to the substrate, the structure including at least one recess formed inward from the sidewall and extending substantially parallel to the surface of the substrate, the method comprising the steps of:

[0007] - Direct the energy beam at this structure.

[0008] - Measure the spectral response generated by the interaction between the incident beam and the structure.

[0009] - Detect one or more peaks in the response and derive parameter values ​​representing the indentation depth from the one or more peaks, the parameters having a previously established one-to-one relationship with the indentation depth.

[0010] - Use the previously established relationship to derive the depression depth from the detected parameter values.

[0011] According to one embodiment, the structure includes at least one first layer oriented substantially parallel to the surface of the substrate, the layer being sandwiched between two adjacent layers, and wherein the first layer is recessed relative to the adjacent layers, the recess being formed inward from the sidewall.

[0012] According to one embodiment, the first layer is formed of a first material, while the adjacent layer is formed of a second material different from the first material.

[0013] According to one embodiment:

[0014] - The substrate is a silicon substrate.

[0015] - The structure is a fin-shaped structure with two elongated sidewalls, wherein the first layer and the adjacent layer extend between the two sidewalls.

[0016] - The first material is SiGe and the second material is Si.

[0017] - The energy beam is a laser beam and the spectral response is a Raman response.

[0018] The fin structure comprises a silicon substrate portion consistent with the substrate and a stack of silicon nanosheets on the substrate portion, the silicon nanosheets being separated from each other and from the substrate portion by SiGe layers recessed relative to the silicon nanosheets.

[0019] This parameter is the ratio between the intensity peaks in the Raman spectrum associated with the vibrations of silicon atoms in the SiGe layer and the intensity peaks associated with the vibrations of silicon atoms in the silicon substrate, substrate portion, and / or silicon nanosheets.

[0020] - The previously established relationship is linear.

[0021] According to one embodiment, the recess depth is an etching depth formed in a process in which the first layer is progressively etched relative to adjacent layers, the etching process progressing inward from the sidewall.

[0022] According to one embodiment, the energy beam is a laser beam whose light is polarized along the length of the sidewall, and whose spectral response is a Raman spectral response.

[0023] According to one embodiment, the energy beam is an electron beam, and the spectral response is an energy-dispersive X-ray spectral response.

[0024] According to one embodiment, the energy beam is an X-ray beam, and the spectral response is an X-ray photoelectron spectral response.

[0025] According to one embodiment, the energy beam is an ion beam, and the spectral response is a secondary ion mass spectrometry response.

[0026] According to one embodiment, the previously established relationship is a linear relationship.

[0027] The present invention equivalently relates to an apparatus for performing the method according to the invention, the apparatus comprising:

[0028] - Energy beam source

[0029] - Detector,

[0030] - A processing unit configured to perform the following steps when the method described above is applied to the structure:

[0031] Obtain the spectral response from the detector.

[0032] The value of this parameter is determined based on the spectral response.

[0033] The indentation depth is derived from the obtained parameter values ​​and from the previously established one-to-one relationship.

[0034] According to one embodiment, the source is a laser source and the detector and processing unit are configured for Raman spectroscopy.

[0035] This invention also relates to computer program products configured to run on a processing unit and perform the aforementioned steps.

[0036] This invention is equivalent to the use of a method according to the invention for performing online measurement of lateral recess depth in a semiconductor processing line.

[0037] The present invention is equivalent to the use of an apparatus according to the invention for performing online measurement of recess depth in a semiconductor processing line. Attached Figure Description

[0038] Figures 1a to 1d The progressive stages of an etching process are shown, which is applied to a fin structure to which the method of the present invention is applied.

[0039] Figure 2 Typical Raman spectra obtained from the structure shown in Figure 1 by the method of the present invention are shown.

[0040] Figure 3 The relationship between the ratio of the two characteristic peak intensities from the spectrum and the etching time is shown.

[0041] Figure 4 It shows Figure 3 The same ratio is represented in the text, but now varies with the relationship between the etch depths.

[0042] Figure 5a and 5b The method of the present invention is shown to be applied to a fork-shaped fin structure array.

[0043] Figure 6 It shows that for Figure 5a and 5b The relationship between various parameters and etching depth of the fork-shaped structure shown is derived using different spectral techniques. Detailed Implementation

[0044] In semiconductor reliability and characterization, spectroscopy is used as a tool to measure mechanical stress, composition, doping, and phase. Well-known spectroscopic techniques include X-ray photoelectron spectroscopy (XPS), energy-dispersive X-ray spectroscopy (EDS), and Raman spectroscopy.

[0045] The method of the present invention allows for the measurement of lateral depth based on spectral measurements, wherein a representative parameter can be derived from the response, which is in a previously established one-to-one relationship with the depth, thereby allowing the depth to be derived from the measurement.

[0046] This invention was validated based on an etching process, but generally relates to measuring the depth of a lateral recess, regardless of the process used to produce that depth. The method can also be used, for example, to progressively measure the decreasing depth of the lateral cavity while the cavity is being filled with a filler material.

[0047] The following experiments have been conducted and used as proof of concept for this invention. A fin-shaped structure has been produced, the cross-section of which is... Figure 1a As shown in the figure, the structure includes a substrate portion 2 formed of silicon, which is aligned with a silicon substrate 10. The width of the structure is approximately 120 nm. The length of the structure, measured in a direction perpendicular to the plane of the figure, is several micrometers. The structure has two sidewalls 11, as shown in the figure, which may be perpendicular to the surface of the substrate 10, but may also be slightly inclined toward the vertical center plane of the fin structure 1. Generally, the sidewalls are in an upright position relative to the substrate 10.

[0048] Structure 1 further includes a SiGe layer 3 with a thickness of approximately 9 nm, which creates a multilayer stack of silicon nanosheets 4, each also having a thickness of approximately 9 nm. The silicon nitride portion 5 above this structure represents the photolithographic mask used to produce the fin structure 1. Structure 1 was produced by depositing consecutive thin layers of SiGe 3 and Si 4 on a silicon substrate 10, followed by forming a photolithographic mask 5 and etching down onto either side of the mask to a depth of approximately 105 nm, measured from the upper surface of the silicon substrate 10. The Si and SiGe layer stack has found applications in the fabrication of superlattice structures in FinFET transistors.

[0049] The SiGe layer 3 is then etched away by a selective etching process, which can be any process known in the art for this purpose, wherein the SiGe is progressively etched inward from the sidewall 11 of the fin structure 1 relative to the substrate portion 2, the mask portion 5, and the silicon nanosheet 4. Figure 1b , 1c Figures 1 and 1d show the structures at etching times of 10 s, 20 s, and 30 s, respectively. The lateral etching depths at these time points are approximately 4 nm, 10 nm, and 19 nm, respectively. At each of these etching depths, Raman spectroscopy measurements were performed using a laser beam perpendicularly oriented to the substrate 10 on which the fin structure 1 rests. The laser has a wavelength of 532 nm and is polarized along the length of the fin structure 1. The laser is focused at a point with a diameter of approximately one micrometer (i.e., the laser beam diameter itself is approximately one micrometer), which overlaps with the entire width of the fin structure 1. The measurement time at each depth is 100 s. Generally, when applying the Raman spectroscopy-based embodiments of the present invention, the measurement time range can be from short periods (e.g., between 5 s and 10 s) to several minutes, depending on the laser power or other parameters.

[0050] Figure 2 The Raman response at the start of the etching process is shown, corresponding to... Figure 1aThe structure is shown. The strong peak 15 at 521 cm⁻¹ originates from Si-Si vibrations in the substrate 10, the substrate portion 2, and / or the silicon nanosheet 4, but simultaneously, a different accompanying peak 16 appears at a frequency slightly lower than the main 521 cm⁻¹ peak. This feature originates from Si-Si vibrations in the SiGe material. The intensity of this second peak varies with the remaining volume of the SiGe material and is therefore directly related to the etching depth. Figure 3 The normalized Si-Si co-peak intensity, which varies with etching time, is shown as the ratio between the co-peak intensity 16 and the main peak intensity 15. The ratio is shown to decrease non-linearly with increasing etching time. However, when the total etching depth is used as the abscissa parameter, as... Figure 4 As shown, the relationship becomes linear. This linear relationship indicates a one-to-one relationship between the ratio and the etching depth, meaning that the etching depth can be determined when the ratio is known. This relationship thus allows the etching depth to be determined by measuring the Raman spectrum and deriving the value of the ratio of the accompanying peak intensity to the main peak intensity from the spectrum. This method thus allows for monitoring the etching depth by performing periodic Raman measurements during the etching process.

[0051] Using the SiGe / Si ratio as a depth-related parameter is itself related to the use of Raman spectroscopy, since it has been found that both the SiGe correlation peak 16 and the Si correlation peak 15 increase with increasing etching depth, which is unexpected and can be attributed to the combined effect of the decrease in both the reflectivity and absorptivity of the sample. However, the Si peak rises faster than the SiGe correlation peak, so that the ratio exhibits a dependence on decreasing etching depth.

[0052] A similar relationship between Raman peak ratio and indentation depth has been verified in nanosheet structures with similar dimensions to those shown in Figure 1, but these nanosheet structures consist of stacked SiO2 sheets spaced apart from each other by polycrystalline silicon sheets, and the polycrystalline silicon sheets have various indentation depths relative to the SiO2 sheets.

[0053] However, the indentation correlation parameter can differ from the stated ratio, as is the case when using other spectroscopic techniques (see below). In many cases, the indentation correlation parameter can simply be a specific peak intensity in the spectrum or the integrated area beneath that peak, which directly represents the amount of a particular material in the irradiated structure. This parameter then decreases with increasing indentation depth. The peak or integrated peak area is preferably based on the spectrum expressed in arbitrary units (au) (e.g., ...). Figure 2 As shown), the unit is independent of the specific measurement conditions applied during the measurement.

[0054] The method of the present invention is therefore not limited to the set of conditions limiting the above-described experiments. The method is not merely applicable to fin structures 1 in which the etched layer extends between two sidewalls 11 of the structure, but is generally applicable to any structure having at least one sidewall that is upright relative to the substrate on which the structure is formed, and at least one lateral recess extending along that sidewall. The method can be applied to parallel fin arrays, each fin recessed from the side as described above. In the latter case, the beam width can illuminate several fins of the array. For example, the array of the above-described fin structures can have a spacing of approximately 400 nm, such that a laser beam with a diameter of 1 micrometer illuminates two adjacent fins of the array. When the recess-related parameter is the ratio of the two peaks as described above, the number of irradiated fins is factored into that ratio. However, when the peak itself or the integral area under that peak is used as the depth-related parameter, the number of irradiated fins may need to be taken into account when calculating the recess depth in each fin.

[0055] The fins can be narrower than the structure used in this experiment, for example, having a width on the order of 20 nm. As mentioned above, the parameter chosen in this experiment is the ratio of the accompanying peak intensity to the main peak intensity. This is true for the specific case of etching a SiGe layer relative to Si. Other parameters may be applicable for other geometries and materials. This method can also be applied, for example, to similar superlattice structures as shown in Figure 1, but with SiO2 and amorphous Si replacing Si and SiGe, respectively. The one-to-one relationship is preferably linear, which allows the relationship to be derived over a large lateral depth range based on only a few calibration measurements. However, when the relationship is not linear, it can also be determined over a large depth range through more thorough calibration. The method is therefore not limited to a linear relationship between this parameter and the lateral depth.

[0056] This invention was validated based on an etching process, but generally relates to measuring the depth of lateral recesses regardless of the process used to produce that depth. The method can also be used, for example, to progressively measure the decreasing depth of lateral recesses, as the recesses are filled with a filler material. For example, when the structure of FIG1 is used as a superlattice, the recesses can be filled with different materials (e.g., TiN) to create spacers between adjacent silicon layers. The method of this invention can be used to measure the decreasing depth based on a spectral response that is sensitive to the progressively increasing volume of the different materials in the recesses. In the specific case of TiN-filled spacers, the decreasing depth can be detected, for example, by the method of this invention, without using Raman spectroscopy, which is unsuitable for TiN, but instead using, for example, EDS (see below).

[0057] This invention is not limited to the use of an incident laser beam and analysis via Raman spectroscopy. When using this combination, the wavelength of the incident light allows it to probe the entire region of interest. It is also important that the laser is polarized along the length of one or more sidewalls of the structure (in which transverse recesses are formed)

[0058] Several alternative energy beam and spectroscopic measurement techniques applicable to the present invention are summarized below. Using EDS, the energy beam directed at the structure is an electron beam, and the response is measured by detecting the X-rays produced by electrons from this energy beam that excite inner-shell electrons in the structure. These inner-shell electrons are then released and emit X-rays at energy levels characteristic of the excited material. The spectral response is therefore an X-ray intensity spectrum that varies depending on the X-ray energy. For Figures 1a-1d The structure exhibits a Ge-related peak in the spectrum that is sensitive to etching depth; the peak itself, or the integrated area beneath it, decreases with increasing recess depth. The landing energy of the electron beam is preferably higher than approximately 4 keV. Successful experiments have been performed on the structure shown in Figure 1 using a landing energy of 5.7 keV and a measurement time of 60 s.

[0059] Using XPS, the energy beam directed at the structure is an X-ray beam. The material is ionized by the ejection of inner-shell electrons, which are detected and their energies characterize the material. The spectral response is now a spectrum of emitted electron intensity (electron count) that varies with the binding energy of the inner-shell electrons in various materials, taking into account that the kinetic energy of the emitted electrons equals the X-ray energy (known) minus the binding energy. Successful experiments have been performed on the structure in Figure 1 using an aluminum-based X-ray source that generates X-rays at an energy of 1.48 keV, showing that the 3d Ge-related peak in the spectrum is sensitive to the etching depth at a binding energy of approximately 30 eV, and that the peak itself or the integral area under the peak decreases with increasing recess depth. The measurement time in this case was 2 hours. However, the use of XPS does not require the removal of the hard mask 5.

[0060] Another alternative is to use secondary ion mass spectrometry (SIMS), where the incident energy beam is an ion beam that sputters material from the structure as emitted ions, layer by layer. The emitted ions are detected and analyzed according to their mass. The mass is determined, for example, by measuring the time of flight of the ionized material. The spectral response is thus a spectrum of ion counts that vary with the time of flight or mass of the various materials in the structure. However, this technique is invasive because each upper layer is sputtered before a signal associated with the material at the recessed location is detected. Consequently, the measurement time is also longer than the techniques described above. However, successful experiments have been performed on the structure in Figure 1, showing that the Ge-related peaks (Ge-, SiGe-, Ge-2) in the SIMS spectrum are sensitive to the recess depth, and that the integral area under each peak decreases with increasing recess depth. The measurement time can be one hour or longer, during which material is progressively sputtered from the top of the structure. When the SiGe layer 3 was sputtered away during the continuous phase of the total sputtering time, the recorded spectrum allowed for the determination of different etch depths of the SiGe layer, because the Ge-related peak intensity varied with the depth of the corresponding recess (i.e., depended on the depth of the recess). Figures 1a to 1d Which of the structures shown is subjected to the sputtering beam?

[0061] Other existing spectroscopic techniques known in the art can be used in the methods and apparatus of this invention, such as XRF (X-ray fluorescence), XRD (X-ray diffraction analysis) and RBS (Rutherford backscattering spectroscopy).

[0062] The energy beam is not necessarily oriented perpendicular to the surface of the substrate 10. Optimal beam orientation may depend on the spectroscopic technique used.

[0063] This invention is not limited to determining the depth of a recess that can be obtained by removing a first material layer (e.g., SiGe layer 3 in FIG. 1) sandwiched between two layers of a second material different from the first material (such as silicon substrate portion 2 and Si layer 4 in FIG. 1). The recess can be formed in a structure formed of a single material, provided that the applied spectroscopic technique is sensitive to volumetric changes dependent on the depth of the recess.

[0064] The method of this invention can be applied online, for example, integrated into semiconductor manufacturing processes that include the lateral etching process described above. In this way, the method enables rapid and (in most cases) non-invasive measurement of the etching depth at various times during the etching process, which allows for online verification of the etching process.

[0065] This method can be performed using known spectroscopic tools, including energy beam sources such as laser sources or electron beam sources, detectors, and processing units for numerically processing the acquired spectra. However, the invention also relates to specific tools for performing the method of the invention, comprising some basic components, wherein, when applying a combination of laser and Raman spectroscopy, the processing unit has access to a previously established relationship between parameters (e.g., the ratio between the accompanying peak intensity and the main peak intensity in the case of Si / SiGe layer stacks) and the recess depth. This relationship can be stored in a memory incorporated into or accessible to the processing unit. The processing unit is further configured to perform the following steps:

[0066] - Obtain the spectral response from the detector.

[0067] - The value of this parameter is determined based on the spectral response obtained from the detector.

[0068] - Derive the depression depth from the acquired parameter values ​​and the previously established one-to-one relationship.

[0069] Preferably, the processing unit is a computer programmed to perform the above steps. The invention also relates to such a computer program.

[0070] The method of this invention is also applicable to so-called fork-shaped structures, such as... Figure 5a and 5b As shown. Arrays of these forked structures have been provided, and have a critical size CD of approximately 50 nm and a spacing p of approximately 90 nm. As is known in the art, the forked structure is a fin-like structure comprising nanosheets as shown in Figure 1, but with vertical walls separating the nanosheets on either side of the central vertical plane of the fin. Figure 5a The forked fins 20 shown have been fabricated on a silicon substrate 10. Each fin comprises alternating Si nanosheets 21 and SiGe nanosheets 22. The top SiGe nanosheets have a greater thickness than the SiGe nanosheets below. Vertical walls 23 comprise SiN and additional layers, which may not be shown in this figure but are well known in the art. STI (shallow trench isolation) oxide layers separate these structures. The SiGe is etched relative to Si to produce… Figure 5b The lateral etched structure is shown. Both the non-etched and etched structures undergo the method of this invention at various stages of the etching process corresponding to different etch depths using various different spectroscopic techniques.

[0071] Figure 6Normalized characteristic parameters varying with etching depth are shown for four spectroscopic techniques: Raman (curve 30), EDS (curve 31), XPS (curve 32), and SIMS (curve 33), along with the ratio of the SiGe volume at each depth to the SiGe volume at the start of the etching process (curve 34). The parameters used vary depending on the spectroscopic technique. For example, for Raman curve 30, the ratio of the SiGe-related peak to the Si-related peak in the spectrum is used, as referenced above. Figure 2 The explanation is as follows: For EDS and XPS, this parameter is the integrated intensity of the typical Ge-related peak (i.e., the area under that peak) representing the SiGe volume in the spectrum. For SIMS, three different Ge-related peak results represent the SiGe volume. Figure 6 The SIMS-related parameters used are based on the integral intensity of one of the three peaks measured during a sputtering time of approximately 1 hour.

[0072] XPS curve 32 shows an outlier at the zero-depth measurement point. This is attributed to the thin SiO2 layer overlay. However, aside from this data point, all these techniques produce a clear one-to-one relationship between the parameters and the etching depth, making it possible to determine the etching depth by measuring the parameters under consideration.

[0073] Although the invention has been illustrated and described in detail in the accompanying drawings and the foregoing description, such illustrations and descriptions should be considered illustrative or exemplary rather than restrictive. Other variations of the disclosed embodiments may be understood and implemented by those skilled in the art from a study of the drawings, this disclosure, and the appended claims when implementing the claimed invention. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite articles "a" or "an" do not exclude a plural. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that combinations of these measures cannot be advantageously used. Any reference numerals in the claims should not be construed as limiting the scope.

Claims

1. A method for measuring the depth of a lateral recess in a microstructure located on a substrate, the microstructure having elongated sidewalls in an upright position relative to the substrate, the microstructure including at least one recess formed inwardly from the sidewalls and extending parallel to a surface of the substrate, the method comprising the steps of: Direct the energy beam at the microstructure. The spectral response generated by the interaction between the incident beam and the microstructure was measured. One or more peaks in the spectral response are detected, each peak representing an amount of a specific material in the microstructure, and a parameter value representing the depth of the indentation is derived from the one or more peaks, the parameter value having a previously established one-to-one relationship with the depth of the indentation. The previously established relationship is used to derive the indentation depth from the detected parameter values.

2. The method of claim 1, wherein the microstructure includes at least one first layer oriented substantially parallel to the surface of the substrate, the layer being sandwiched between two adjacent layers, and wherein the first layer is recessed relative to the adjacent layers, the recess being formed inward from the sidewall.

3. The method of claim 2, wherein the first layer is formed of a first material, and the adjacent layer is formed of a second material different from the first material.

4. The method of claim 3, wherein: The substrate is a silicon substrate. The microstructure is a fin-shaped structure with two elongated sidewalls, wherein the first layer and the adjacent layer extend between the two elongated sidewalls. The first material is SiGe and the second material is Si. The energy beam is a laser beam and the spectral response is a Raman spectral response. The fin structure includes a silicon substrate portion consistent with the substrate, and a stack of silicon nanosheets on the silicon substrate portion, the silicon nanosheets being spaced apart from each other and from the silicon substrate portion by SiGe layers recessed relative to the silicon nanosheets. The parameter value is the ratio between the intensity peak value in the Raman spectrum related to the vibration of silicon atoms in the SiGe layer and the intensity peak value related to the vibration of silicon atoms in the silicon substrate, the substrate portion, and / or the silicon nanosheets. The previously established relationship is linear.

5. The method of claim 2, wherein the recess depth is an etching depth formed in a process in which the first layer is progressively etched relative to the adjacent layer, the etching progressing inward from the sidewall.

6. The method of claim 1, wherein the energy beam is a laser beam, the light of the laser beam is polarized along the length of the sidewall, and wherein the spectral response is a Raman spectral response.

7. The method of claim 1, wherein the energy beam is an electron beam, and wherein the spectral response is an energy-dispersive X-ray spectral response.

8. The method of claim 1, wherein the energy beam is an X-ray beam, and wherein the spectral response is an X-ray photoelectron spectral response.

9. The method of claim 1, wherein the energy beam is an ion beam, and wherein the spectral response is a secondary ion mass spectrometry response.

10. The method of claim 1, wherein the previously established relationship is a linear relationship.

11. An apparatus for performing the method as described in any one of claims 1 to 10, comprising: Energy beam source Detector, A processing unit configured to apply the method as described in any one of claims 1 to 10 to a microstructure.

12. The apparatus of claim 11, wherein the energy beam source is a laser source and the detector and the processing unit are configured for Raman spectroscopy.

13. A computer-readable storage medium storing instructions configured to run on a processing unit and perform the method of claim 1.

14. The use of the method as claimed in any one of claims 1 to 10, wherein the method is used to perform an online measurement of lateral recess depth in a semiconductor processing line.

15. Use of the apparatus as claimed in claim 11 or 12, wherein the apparatus is used to perform online measurement of recess depth in a semiconductor processing line.

Images