A longitudinal resolution near-field infrared spectroscopy measurement system and method

Abstract

The application discloses a longitudinal resolution near-field infrared spectrum measurement system and method, and relates to the field of spectrum measurement. In the system, the half-reflection half-transmission lens in the interference measurement module transmits infrared light to a nano probe; the nano probe excites the near-field infrared light field formed by local near field to interact with a sample to be measured, so that a backscattering light signal is obtained; the half-reflection half-transmission lens combines the reference light and the backscattering light signal and generates interference to obtain a near-field interference signal; a control module controls a moving mirror to change an optical path difference in a step displacement manner, and changes the distance between the nano probe and the sample to be measured under the optical path difference corresponding to each step position of the moving mirror; a data acquisition module acquires a corresponding near-field interference signal group under each optical path difference; and a data processing module processes the near-field interference signal group to obtain longitudinal resolution infrared spectra of different depth layers. The application can realize the measurement of longitudinal resolution infrared spectra of different depth layers of a sample while maintaining broadband spectrum capability.

Description

Technical Field

[0001] This application relates to the field of spectral measurement, and in particular to a longitudinally resolved near-field infrared spectroscopy measurement system and method. Background Technology

[0002] Infrared spectroscopy can reflect information such as chemical bond vibrations, molecular configuration, and carrier response in materials, and is an important technique in materials analysis, device characterization, and non-destructive testing. As semiconductor devices, functional thin films, and composite materials develop towards multilayering, heterogeneity, and nanoscale, the key physical and chemical information of materials is often distributed in different depths below the sample surface. Infrared spectroscopy methods that rely solely on surface or equivalent volume responses are no longer sufficient to meet the needs for fine characterization of subsurface structures and interface properties.

[0003] To overcome the spatial resolution limitations of infrared spectroscopy, scattering-type scanning near-field infrared spectroscopy techniques, such as scattering-type scanning near-field optical microscopy (s-SNOM) and nanoscale Fourier-Transform Infrared Spectroscopy (nano-FTIR), excite and collect near-field enhanced infrared signals using nanoprobes. Combining interferometry and Fourier transform, they achieve nanoscale lateral-resolution infrared spectral measurements. These techniques typically employ a broadband infrared light source to excite the local near-field at the probe tip, introduce a Michelson interferometer to acquire interferograms, and then obtain the near-field infrared spectrum through Fourier transform. They are widely used in material composition analysis and nanoscale spectral imaging.

[0004] However, related near-field infrared spectroscopy measurement methods typically acquire interference signals under single or equivalent probe-sample interaction conditions. The measured spectral results are essentially a superposition of the probe's near-field and the material response within a certain depth range of the sample. Since the near-field electromagnetic interaction exhibits a continuous attenuation characteristic in the vertical direction, the contributions of different depth layers to the measurement signal are highly coupled, making it difficult to distinguish or quantitatively characterize the infrared spectral responses of different depth layers of the sample.

[0005] Furthermore, in related near-field infrared spectroscopy measurement methods, probe-sample distance is mainly used to achieve stable near-field coupling or enhance signal intensity, and a measurement mechanism that uses probe-sample distance as an independent information dimension for systematic sampling and encoding has not yet been established. Simultaneously, existing near-field infrared spectroscopy techniques typically only acquire a single interferogram and perform a Fourier transform, lacking a feasible technical path to combine multi-distance sampling, construct multi-dimensional data, and achieve longitudinally resolved spectral inversion. Therefore, in the analysis of multilayer thin films, composite materials, and samples with subsurface structures, it is difficult to achieve longitudinal resolution while maintaining broadband spectral capabilities.

[0006] Therefore, how to achieve longitudinally resolved infrared spectroscopy measurements of samples at different depths while maintaining broadband spectral capabilities has become an urgent problem to be solved. Summary of the Invention

[0007] The purpose of this application is to provide a longitudinally resolved near-field infrared spectroscopy measurement system and method, which can measure the longitudinally resolved infrared spectra of samples at different depths while maintaining broadband spectral capabilities.

[0008] To achieve the above objectives, this application provides the following solution.

[0009] In a first aspect, this application provides a longitudinally resolved near-field infrared spectroscopy measurement system, comprising: an infrared light source, an interferometric measurement module, a nanoprobe, a control module, a data acquisition module, and a data processing module; the interferometric measurement module comprises: a semi-reflective mirror and a moving mirror.

[0010] The infrared light source emits infrared light; the semi-reflective mirror reflects the infrared light to the moving mirror and transmits it to the nanoprobe; the moving mirror returns the reference light; the nanoprobe excites a local near field under the irradiation of the infrared light, forming a near-field infrared light field, which interacts with the sample to be tested to obtain a backscattered light signal containing information about the sample; the semi-reflective mirror also combines the reference light and the backscattered light signal to produce interference, resulting in a near-field interference signal; the control module controls the moving mirror to change the optical path difference in a step-shift manner, and changes the distance between the nanoprobe and the sample to be tested at each step position of the moving mirror corresponding to the optical path difference; the data acquisition module acquires the near-field interference signal group corresponding to each optical path difference, which includes near-field interference signals at multiple different static probe-sample distances; the data processing module performs data rearrangement, Fourier transform, and depth inversion processing based on the near-field interference signal groups corresponding to all optical path differences to obtain the longitudinally resolved infrared spectra of different depth layers of the sample to be tested.

[0011] In one embodiment, the data processing module includes: a matrix construction unit, configured to construct a two-dimensional sampling data matrix according to the optical path difference dimension and the distance dimension based on the near-field interferometric signal groups corresponding to all optical path differences; a data rearrangement unit, configured to rearrange the two-dimensional sampling data matrix according to the distance dimension and extract the interferogram corresponding to each static probe-sample distance; a Fourier transform unit, configured to perform Fourier transform on each of the interferograms to obtain the near-field infrared spectrum at each static probe-sample distance; and a depth inversion unit, configured to perform depth inversion on each of the near-field infrared spectra to obtain the longitudinally resolved infrared spectra of different depth layers of the sample under test.

[0012] In one embodiment, the infrared light source is a broadband infrared light source; the spectral coverage of the broadband infrared light source is 400 cm⁻¹. -1 ~4000cm -1 The output power of the broadband infrared light source is 0.1mW~20mW.

[0013] In one embodiment, the number of step points of the moving mirror is 128 to 4096; the maximum scanning range of the moving mirror is 50 μm to 300 μm; and the dwell time of the moving mirror at each step position is 0.1 ms to 100 ms.

[0014] In one embodiment, the number of static probe-sample distances is 2 to 100; the interval between the static probe-sample distances is 0.1 nm to 20 nm; and the switching time of the static probe-sample distances is 0.1 ms to 10 ms.

[0015] In one embodiment, the Fourier transform unit is specifically used to: perform DC component removal and window function weighting on each of the interferograms to obtain processed interferograms; perform Fourier transform on each processed interferogram to obtain Fourier transform results; and perform phase correction on the Fourier transform results to obtain the near-field infrared spectrum at each static probe-sample distance.

[0016] In one embodiment, the depth inversion unit is specifically used to: perform depth inversion on each of the near-field infrared spectra using a physical constraint-based inversion algorithm to obtain longitudinally resolved infrared spectra of different depth layers of the sample to be tested.

[0017] In one embodiment, the interferometric measurement module further includes a parabolic mirror; the parabolic mirror is disposed in the optical path between the semi-reflective lens and the nanoprobe.

[0018] Secondly, this application provides a longitudinally resolved near-field infrared spectroscopy measurement method, which is implemented using the aforementioned longitudinally resolved near-field infrared spectroscopy measurement system, and includes the following steps.

[0019] The optical path difference is changed by controlling the moving mirror to move in steps, and the distance between the nanoprobe and the sample is changed according to the optical path difference corresponding to each step position of the moving mirror.

[0020] Acquire the near-field interference signal group corresponding to each optical path difference, wherein the near-field interference signal group includes near-field interference signals at multiple different static probe-sample distances.

[0021] Based on the near-field interference signal groups corresponding to all optical path differences, data rearrangement, Fourier transform, and depth inversion processing are performed to obtain the longitudinally resolved infrared spectra of different depth layers of the sample under test.

[0022] In one embodiment, data rearrangement, Fourier transform, and depth inversion are performed on the near-field interferometric signal sets corresponding to all optical path differences to obtain the longitudinally resolved infrared spectra of the sample at different depths. Specifically, this includes: constructing a two-dimensional sampling data matrix according to the optical path difference dimension and the distance dimension based on the near-field interferometric signal sets corresponding to all optical path differences; rearranging the two-dimensional sampling data matrix according to the distance dimension to extract the interferogram corresponding to each static probe-sample distance; performing a Fourier transform on each interferogram to obtain the near-field infrared spectrum at each static probe-sample distance; and performing depth inversion on each near-field infrared spectrum to obtain the longitudinally resolved infrared spectra of the sample at different depths.

[0023] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a longitudinally resolved near-field infrared spectroscopy measurement system and method. In the interferometric measurement process, multiple static probe-sample distance sampling is introduced. The probe-sample distance is used as a depth encoding dimension and combined with the optical path difference (OPD) obtained by the moving mirror step scanning to construct two-dimensional sampling data of "optical path difference-distance" (i.e., multiple sets of near-field interferometric signal groups). By performing data rearrangement, Fourier transform and depth inversion processing on the two-dimensional sampling data, the longitudinally resolved infrared spectrum of the sample at different depth layers is measured while maintaining broadband spectral capability. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of a longitudinally resolved near-field infrared spectroscopy measurement system provided in an embodiment of this application.

[0026] Figure 2 A schematic diagram illustrating the principle of broadband near-field infrared interferometry provided in this application embodiment.

[0027] Figure 3 This is a schematic diagram of the construction of a two-dimensional sampling matrix for optical path difference-probe-sample distance provided in an embodiment of this application.

[0028] Figure 4 This is a schematic diagram of the longitudinal resolution infrared spectral output results provided in an embodiment of this application.

[0029] Figure labels: Nano probe—1, Semi-reflective lens—2, Moving mirror—3, Sample to be tested—4, Parabolic mirror—5. Detailed Implementation

[0030] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0031] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] In one exemplary embodiment, such as Figure 1 and Figure 2 As shown, a longitudinally resolved near-field infrared spectroscopy measurement system is provided, including: an infrared light source, an interferometric measurement module, a nanoprobe 1, a control module, a data acquisition module, and a data processing module. Figure 2 As shown, the interferometric measurement module includes a semi-reflective mirror 2 and a moving mirror 3. The infrared light source, the semi-reflective mirror 2, and the moving mirror 3 constitute a reference arm.

[0033] The infrared light source is used to emit infrared light; the semi-reflective mirror 2 is used to reflect the infrared light to the moving mirror 3 and transmit the infrared light to the nanoprobe 1; the moving mirror 3 is used to return the reference light; the nanoprobe 1 is used to excite a local near field under the irradiation of the infrared light, forming a near-field infrared light field, which interacts with the sample 4 to obtain a backscattered light signal containing information about the sample 4; the semi-reflective mirror 2 is also used to combine the reference light and the backscattered light signal, and generate interference to obtain a near-field interference signal; the control The module is used to control the moving mirror 3 to change the optical path difference in a step-displacement manner, and to change the distance between the nanoprobe 1 and the sample 4 under the optical path difference corresponding to each step position of the moving mirror 3; the data acquisition module is used to acquire the near-field interference signal group corresponding to each optical path difference, the near-field interference signal group including multiple near-field interference signals at different static probe-sample distances; the data processing module is used to perform data rearrangement, Fourier transform and depth inversion processing based on the near-field interference signal groups corresponding to all optical path differences to obtain the longitudinally resolved infrared spectra of different depth layers of the sample 4 under test.

[0034] In another exemplary embodiment of this application, such as Figure 2 As shown, the data acquisition module includes: a detector.

[0035] In another exemplary embodiment of this application, the data processing module includes: a matrix construction unit, configured to construct a two-dimensional sampling data matrix according to the optical path difference dimension and the distance dimension based on the near-field interferometric signal groups corresponding to all optical path differences; a data rearrangement unit, configured to rearrange the two-dimensional sampling data matrix according to the distance dimension and extract the interferogram corresponding to each static probe-sample distance; a Fourier transform unit, configured to perform Fourier transform on each of the interferograms to obtain the near-field infrared spectrum at each static probe-sample distance; and a depth inversion unit, configured to perform depth inversion on each of the near-field infrared spectra to obtain the longitudinally resolved infrared spectra of the sample under test at different depth layers.

[0036] In another exemplary embodiment of this application, the infrared light source is a broadband infrared light source; the spectral coverage of the broadband infrared light source is 400 cm⁻¹. -1 ~4000cm -1 The output power of the broadband infrared light source is 0.1mW~20mW.

[0037] In another exemplary embodiment of this application, the number of step points of the moving mirror 3 is 128 to 4096; the maximum scanning range of the moving mirror 3 is 50 μm to 300 μm; and the dwell time of the moving mirror 3 at each step position is 0.1 ms to 100 ms.

[0038] In another exemplary embodiment of this application, the number of static probe-sample distances is 2 to 100; the interval between the static probe-sample distances is 0.1 nm to 20 nm; and the switching time of the static probe-sample distances is 0.1 ms to 10 ms.

[0039] In another exemplary embodiment of this application, the Fourier transform unit is specifically used for: performing DC component removal and window function weighting processing on each of the interferograms to obtain processed interferograms; performing Fourier transform on each processed interferogram to obtain Fourier transform results; and performing phase correction on the Fourier transform results to obtain the near-field infrared spectrum at each static probe-sample distance.

[0040] In another exemplary embodiment of this application, the depth inversion unit is specifically used to: perform depth inversion on each of the near-field infrared spectra using a physical constraint-based inversion algorithm to obtain longitudinally resolved infrared spectra of different depth layers of the sample to be tested.

[0041] The inversion algorithm includes: convolutional neural network or equivalent inversion algorithm (e.g., parameterized inversion or regularized solution method based on forward model).

[0042] In another exemplary embodiment of this application, such as Figure 2 As shown, the interferometric measurement module further includes a parabolic mirror 5; the parabolic mirror 5 is disposed in the optical path between the semi-reflective lens 2 and the nanoprobe 1; the parabolic mirror 5 is used to focus the incident light beam to the tip of the nanoprobe 1.

[0043] The longitudinally resolved near-field infrared spectroscopy measurement system of the above embodiments mainly involves the following parts, which will be described in detail below.

[0044] (1) Broadband infrared near-field excitation and backscattered light signals were obtained by using an infrared light source and nano probe 1.

[0045] Specifically, a broadband infrared light source is used to irradiate the nanoprobe 1, which excites a local near-field enhanced infrared light field at the tip of the nanoprobe 1 and interacts with the sample 4 under test in the near field to form a backscattered light signal containing sample information.

[0046] Among them, the broadband infrared light source type can be a synchrotron infrared light source, a broadband thermal radiation source, or a broadband infrared supercontinuum light source; the spectral coverage range is 400cm. -1 ~4000cm -1 The preferred size is 600cm. -1 ~3000cm -1 Light source output power: 0.1mW~20mW, preferably 1mW~5mW.

[0047] Nanoprobe 1 type: metal or metal-coated atomic force microscope probe, preferably metal-coated silicon probe; nanoprobe 1 tip radius: 5nm~50nm, preferably 10nm~25nm.

[0048] (2) Interferometry and 3-step scanning of the moving mirror are achieved by using an interferometry module.

[0049] like Figure 2 As shown, the probe scattered light (i.e., backscattered light signal) returned by the nano probe 1 is introduced into the interferometric measurement module. After being combined with the reflected light (i.e., reference light) of the reference optical path by the semi-reflective lens 2, interference is generated at the detector. The control module controls the moving mirror 3 in the interferometric measurement module to change the optical path difference in a step displacement manner, so as to obtain the near-field interference signal under different optical path difference conditions.

[0050] Interference structure: The interferometric measurement module adopts a Michelson interferometer; the moving mirror 3 scanning method is step-scan; the number of optical path difference step points is 128 to 4096, preferably 512 to 2048; the maximum scanning range of optical path difference is 50 μm to 300 μm, preferably 100 μm to 200 μm; the dwell time of each optical path difference point is 0.1 ms to 100 ms, preferably 1 ms to 10 ms.

[0051] (3) The data acquisition module acquires the near-field interference signal at the distance between the multi-static probe and the sample under each optical path difference condition to realize distance encoding.

[0052] Specifically, the optical path difference (OPD) corresponds to each step position of the moving mirror 3. i The nanoprobe 1 is controlled to sequentially stop at multiple preset static probe-sample distances z. j And at each static probe-sample distance z j A near-field interference signal is collected at a location to achieve distance-coded sampling.

[0053] Number of static probes to samples N: 2 to 100, preferably 3 to 8; Static probe to sample sampling distance interval: 0.1 nm to 20 nm, preferably 0.5 nm to 5 nm; Distance control method: Atomic force microscopy feedback control or preset height control; Distance switching time: 0.1 ms to 10 ms, preferably ≤1 ms.

[0054] In this step, for each optical path difference (OPD) i A set of corresponding static probe-sample distances z can be obtained. j Near-field interference signal S(OPD) i ,z j ).

[0055] (4) The data processing module realizes the construction of the two-dimensional sampling data matrix and the data rearrangement.

[0056] All near-field interferometric signals acquired under conditions of optical path difference and static distance are organized according to the optical path difference dimension and the distance dimension to construct a two-dimensional sampling data matrix S(OPD,z), as follows. Figure 3 As shown, Figure 3 Each column corresponds to multi-distance sampling under the same optical path difference.

[0057] Then, the two-dimensional sampling data matrix is ​​rearranged according to the static distance dimension. The specific process of data rearrangement is as follows.

[0058] ①According to the static probe-sample distance z j ① Group the sampled data in the two-dimensional sampled data matrix; ② Extract the corresponding optical path difference (OPD) from each group of data. i ③ Arrange the sampling signals according to the optical path difference order to form a sequence corresponding to the static probe-sample distance z. j The interference signal sequence is used to determine the static probe-sample distance z. j The corresponding interferogram Sz j (OPD). The interferogram is used to characterize the broadband near-field interference signal under a defined probe-sample distance and serves as input data for subsequent Fourier transform processing.

[0059] (5) The data processing module performs Fourier transform to obtain the range-resolved near-field infrared spectrum.

[0060] For each static probe-sample distance, the corresponding interferogram Sz j (OPD) is subjected to Fourier transform to obtain the near-field infrared spectrum under the corresponding distance conditions. I (ω,z j ), where ω represents the infrared wavenumber.

[0061] Fourier Transform Method: Fast Fourier Transform; Optical Path Difference Resampling Method: Equal-Interval Resampling; Optional Processing: DC Component Removal, Window Function Weighting, Phase Correction; Output Spectral Resolution: 1cm -1 ~20cm -1 4cm is preferred. -1 ~8cm -1 .

[0062] (6) The data processing module realizes the inversion and output of longitudinally resolved infrared spectrum.

[0063] Near-field infrared spectra obtained under multiple static probe-sample distance conditions I (ω,z jDeep inversion was performed, and based on the probe-sample near-field electromagnetic interaction model, the longitudinally resolved infrared spectra of different depth layers of the sample were obtained. I (ω,depth), such as Figure 4 As shown.

[0064] Number of inversion depth layers: 2 to 20 layers, preferably 3 to 10 layers; resolvable depth range: 1 nm to 200 nm, preferably 5 nm to 100 nm; inversion method: convolutional neural network based on physical constraints or equivalent inversion algorithm; physical constraints (at least one of the following must be satisfied): non-negativity constraint of output spectrum, continuity constraint of spectrum of adjacent depth layers, and consistency constraint of forward model.

[0065] The inversion process of a convolutional neural network based on physical constraints is as follows.

[0066] The depth inversion unit in the data processing module acquires near-field infrared spectra under multiple static distance conditions. Then, the following inversion steps are performed to output the longitudinally resolved infrared spectrum. .

[0067] ① Input construction and normalization.

[0068] The near-field infrared spectra at various static distances were classified according to infrared wavenumber. Alignment (interpolation to uniformity if necessary) (Grid); stack the near-field infrared spectra at N aligned static distances along the distance dimension to form the input tensor. ,in The z-dimensional dimension represents the distance dimension, with z1, z2, z3…z N The input tensor is normalized by amplitude or reference spectrum (e.g., based on the spectral intensity of z=z1) to obtain the final input tensor. .

[0069] ②Network inference output.

[0070] Input tensor Input convolutional neural network The network output is the longitudinally resolved infrared spectrum of each depth layer. .

[0071] .

[0072] in depth Corresponding to d1, d2, d3…d M d M This indicates the Mth depth layer; M is the depth layer number, ranging from 2 to 20, preferably 3 to 10.

[0073] ③ Physical consistency constraints.

[0074] At least one physical constraint is imposed on the network output to improve interpretability and stability. The physical constraints include the following.

[0075] (a) Nonnegativity constraint: This can be achieved through ReLU or projection truncation. This represents the k-th depth layer.

[0076] (b) Continuity constraint: The spectral changes of adjacent depth layers are smooth, for example, by minimizing the following formula.

[0077] .

[0078] in, This represents the (k+1)th depth layer.

[0079] (c) Forward model consistency constraint: Substitute the output spectrum into the forward near-field model Generate predicted near-field infrared spectra .

[0080] .

[0081] and minimize Compared with the actual measured near-field infrared spectrum The error.

[0082] ④ Consistency correction and output.

[0083] If the consistency error exceeds a preset threshold, the data processing module can perform one or more "posterior corrections": fixing the initial value of the network output. Gradient descent or least squares fine-tuning is performed only on some parameters (such as depth weights or scale factors) to make the forward model output more closely match the measured spectrum; the final longitudinally resolved infrared spectrum is output. .

[0084] The inversion process of the equivalent inversion algorithm is as follows.

[0085] The specific process of the parameterized inversion or regularization solution method based on the forward model is as follows.

[0086] ① Constructing a forward model and discretization.

[0087] The sample is discretized into M depth layers d1, d2, d3…d in the longitudinal direction. M M ranges from 2 to 20, preferably from 3 to 10, and the unknown quantity to be determined is the longitudinally resolved infrared spectrum of each depth layer. .

[0088] A forward model is established based on the probe-sample near-field electromagnetic interaction to map the depth layer spectrum into measurable spectra at different static distances. .

[0089] .

[0090] in This is the near-field attenuation function, used to describe the contribution weight of the depth layer to measurements at different distances.

[0091] ② Wave number calculation or piecewise solution (feasible calculation process).

[0092] Solve the linear / nonlinear inverse problem independently for each wavenumber ω (or each band range).

[0093] .

[0094] in For constraint / regularity terms, It is a regularization coefficient (e.g., 0.1~10, preferably 1).

[0095] ③ Implementation of physical constraints and regularization terms.

[0096] Constraints / Regular Terms It must include at least one of the following.

[0097] (a) Nonnegativity constraint: This can be achieved using non-negative least squares or projected gradient.

[0098] (b) Depth continuity: minimize the following formula.

[0099] .

[0100] ④ Consistency verification and output.

[0101] The solution is obtained Then, the predicted spectrum was calculated by back-substituting the forward model. Compare the results with actual measurements; if the error exceeds the threshold, adjust accordingly. Alternatively, the near-field attenuation function can be solved repeatedly until consistency is achieved; the output longitudinally resolved infrared spectrum can be obtained. .

[0102] In one embodiment, the number of static probe-sample distances is 5; the static probe-sample distance range is 1 nm to 15 nm, with an interval of 3 nm; the number of moving mirror 3-step points is 1024; the maximum optical path difference scanning range is 800 μm; and the spectral resolution is approximately 4 cm⁻¹. -1 Inversion depth: 6 layers, depth range 5nm~80nm.

[0103] The above parameter combination achieves a better balance between signal-to-noise ratio, acquisition time and longitudinal resolution, but does not constitute a limitation on this embodiment.

[0104] This embodiment introduces a multi-static probe-sample distance sampling mechanism during broadband near-field infrared interferometry. The probe-sample distance is used as a depth encoding dimension and combined with the optical path difference information generated by the step scanning of the moving mirror 3 to construct a two-dimensional "optical path difference-distance" data matrix. Through data rearrangement, Fourier transform, and depth inversion processing, the longitudinally resolved infrared spectra of samples at different depths are obtained. This embodiment is applicable to multilayer thin film materials, composite materials, and samples with subsurface structures, enabling longitudinally resolved spectral measurements without damaging the sample.

[0105] The following are some specific examples illustrating the practical application of the longitudinally resolved near-field infrared spectroscopy measurement system.

[0106] Example 1:

[0107] (1) Sample.

[0108] Sample: Bilayer thin film sample, PMMA (top layer) / PS (bottom layer).

[0109] Preparation method: Spin coating / evaporation deposition of self-made or commercial standard sheets.

[0110] Thickness range: 10~200nm for a single layer; total thickness 20~400nm (feasible range).

[0111] (2) Instrument and system configuration.

[0112] Atomic force microscope (touch / tapping mode both acceptable, tapping preferred); broadband infrared light source (600~3000cm²). -1 ); Michelson interferometer with stepping mirror; detector (MCT or equivalent infrared detector); phase-locked loop / demodulation module (supporting high-order demodulation from 1st to 5th order); synchronous triggering and data acquisition of the stepping mirror and nanoprobe.

[0113] In this example, the interference signal acquired by the detector in the data acquisition module typically contains both near-field scattering and far-field background components. To enhance the signal-to-noise ratio of the near-field signal and suppress the far-field background, a phase-locked demodulation module can be used to demodulate the modulation component at the probe vibration frequency. Specifically, the atomic force microscope probe vibrates at a vibration frequency f in tapping mode, causing the near-field scattering signal to be nonlinearly modulated with the static probe-sample distance. Then, using the probe vibration frequency f or its higher-order harmonics nf (n=1~5, preferably n=2~4) as a reference, the detected interference signal is demodulated using phase-locked demodulation. The amplitude and phase signals of the corresponding orders are extracted as the effective components of the near-field interference signal and used as the static probe-sample distance z. j Near-field interference signal S(OPD) i ,z j The signal value is used for subsequent two-dimensional sampling data matrix construction, data rearrangement and Fourier transform processing.

[0114] (3) Experimental measurement procedures and parameters.

[0115] (i) System calibration and preheating.

[0116] Optical path alignment and interference contrast optimization (matching of reference arm and signal arm); Zero optical path difference (ZPD) positioning of moving mirror: coarse scanning is performed in the range of 0~20μm to determine the approximate position of the peak.

[0117] (ii) Set up a two-dimensional sampling grid.

[0118] OPD stepping points: 1024 points (512~2048 can be implemented); Maximum OPD range: 400μm (50~800μm can be implemented); Integration time per point: 5ms (1~20ms can be implemented); Number of static distances N: 5 (3~8 can be implemented).

[0119] Static distance sequence: z={1,4,7,10,13}nm (implementable from 0.5 to 20nm, with intervals of 1 to 5nm).

[0120] (iii) Two-dimensional sampling.

[0121] Stop the moving mirror of the interferometry module at OPD. i At each location, the probes are placed sequentially at z1, z2…z N After stabilization at each z-position, a sample is taken once to obtain the scattering interference signal S(OPD). i ,z j ), and record (OPD) i ,z j (time stamp), where time stamp represents a timestamp.

[0122] (4) Data processing and output.

[0123] (i) Form a two-dimensional sampling data matrix S(OPD,z).

[0124] (ii) The interferogram Sz(OPD) of each z is obtained by rearranging the z-dimensional components.

[0125] (iii) The following processing is performed on Sz(OPD): DC component removal, windowing function, Fourier transform, and phase correction.

[0126] (iv) Output near-field infrared spectra under different distance conditions I (ω,z).

[0127] Will I (ω,z) Input depth inversion unit, output I (ω,depth), where depth is 6 layers and the depth is 5~80nm.

[0128] Example 2:

[0129] Improve the number of recognizable layers and spectral resolution in depth inversion to make it applicable to complex samples such as multilayer films / interface layers.

[0130] Sample: Same as Example 1, any thin film / composite sheet can be used.

[0131] Key parameters different from Example 1: (1) Number of OPD points: 2048~4096 points (preferably 4096). (2) Maximum OPD range: 400~1600μm (preferably 1600μm). (3) Spectral resolution: 1~4cm -1 (Preferred size: 2cm) -1 (4) Number of static distances N: 8~10 (preferably 9). (5) Static distance sequence: z={0.8,1.6,2.8,4.2,6.0,8.2,11,14,18}nm. (6) Inversion output settings: depth layer number is 10~16 layers (preferably 12 layers), depth range is 3~120nm.

[0132] Example 3:

[0133] Sample: Same as Example 1, any thin film / composite sheet can be used.

[0134] Key parameters different from Example 1: (1) Number of OPD points: 256~512 points (preferably 512). (2) Maximum OPD range: 80~120μm (preferably 100μm). (3) Integration time per point: 1~2ms (preferably 1ms). (4) Number of static distances N: 3~4 (preferably 3). (5) Distance sequence: z={1,6,15}nm (1~25nm can be implemented). (6) Inversion output settings: depth layers 4~6 layers (preferably 5 layers).

[0135] Example 4:

[0136] This example demonstrates an extension from "point measurement" to "area hyperspectral imaging".

[0137] Sample: A patterned multilayer structure with localized buried structures (such as localized overlays, microstructures, and defect areas).

[0138] Imaging process: (1) Set the scanning area: 1μm×1μm (0.2~20μm can be implemented). (2) Number of pixels: 32×32 or 64×64. (3) Perform any two-dimensional sampling from Example 1, Example 2 and Example 3 on each pixel. (4) Obtain the image of each pixel. I (ω, depth). (5) Select the characteristic wavenumber ω0 and output The intensity map is obtained to obtain "depth-resolved spectral imaging". (6) Parameter examples: (i) Number of OPD points per pixel: 512. (ii) Number of static distances N: 5. (iii) Imaging time: determined by the number of pixels × the number of sampling points. (7) Output: Spectral intensity distribution of x~y~depth (or layered spectrum of x~y~ω). The spectral intensity distribution of x~y~depth is the three-dimensional distribution of spectral intensity of the sample at different positions and depths in the xy plane. The layered spectrum of x~y~ω is the spectral information corresponding to each position in the xy plane after layering by depth.

[0139] This embodiment proposes a novel concept based on existing broadband near-field infrared interferometry systems: it introduces multi-static probe-sample distance sampling during interferometry, using the static probe-sample distance as a depth encoding dimension. This distance is combined with the optical path difference information obtained from moving mirror step scanning to construct two-dimensional "optical path difference-distance" sampling data. Through data rearrangement, Fourier transform, and depth inversion processing, the separation and reconstruction of the infrared spectral responses of different depth layers of the sample are achieved. This embodiment enables longitudinally resolved infrared spectral measurements of multilayer and subsurface structures without damaging the sample, solving the problems of insufficient longitudinal resolution and difficulty in extracting depth information in current near-field infrared spectroscopy techniques. It provides a new technical means for the fine characterization of materials and devices.

[0140] Based on the same inventive concept, this application also provides a longitudinally resolved near-field infrared spectroscopy measurement method, which is implemented using the longitudinally resolved near-field infrared spectroscopy measurement system described above. The longitudinally resolved near-field infrared spectroscopy measurement method includes the following steps.

[0141] (1) Control the moving mirror 3 to change the optical path difference in a step displacement manner, and change the distance between the nano probe 1 and the sample 4 under the optical path difference corresponding to each step position of the moving mirror 3.

[0142] (2) Obtain the near-field interference signal group corresponding to each optical path difference, wherein the near-field interference signal group includes multiple near-field interference signals at different static probe-sample distances.

[0143] (3) Based on the near-field interference signal groups corresponding to all optical path differences, data rearrangement, Fourier transform and depth inversion processing are performed to obtain the longitudinal resolution infrared spectra of the sample 4 at different depths.

[0144] This step (3) specifically includes: constructing a two-dimensional sampling data matrix according to the optical path difference dimension and the distance dimension based on the near-field interference signal groups corresponding to all optical path differences; rearranging the data of the two-dimensional sampling data matrix according to the distance dimension to extract the interferogram corresponding to each static probe-sample distance; performing Fourier transform on each interferogram to obtain the near-field infrared spectrum under each static probe-sample distance; and performing depth inversion on each near-field infrared spectrum to obtain the longitudinally resolved infrared spectrum of the sample under test at different depth layers.

[0145] Compared with existing near-field infrared spectroscopy measurement techniques, this application achieves longitudinally resolved near-field infrared spectroscopy measurement based on multi-distance coded interferometry. By introducing multi-static probe-sample distance sampling in the broadband near-field interferometry process and combining it with the construction and depth inversion processing of two-dimensional sampling data, longitudinally resolved infrared spectroscopy measurement is realized. The specific advantages are as follows.

[0146] 1. Achieve longitudinal resolution infrared spectroscopy measurement of samples at different depths.

[0147] This application is able to obtain independent infrared spectral responses at different depths of the sample, rather than the superposition of responses at multiple depths as in the prior art, thus achieving longitudinally resolved infrared spectral measurements.

[0148] The main technical points are: (1) introducing multiple static probe-sample distance sampling under each optical path difference condition; (2) constructing an optical path difference-distance two-dimensional sampling data matrix; and (3) performing depth inversion based on near-field interference signal groups.

[0149] Since the electromagnetic interaction between the probe and the sample in the near field changes significantly in the vertical direction with the probe-sample distance, by obtaining the near field spectrum under multiple static distance conditions, the spectral contributions of different depth layers can be made to exhibit differentiated weights under different distance conditions, thereby providing a distinguishable information basis for subsequent depth inversion and realizing the separation of spectral responses of different depth layers.

[0150] 2. Improve the vertical information dimension while maintaining broadband spectral capabilities.

[0151] This application introduces a longitudinal information dimension without sacrificing broadband infrared spectral coverage and spectral resolution, thereby expanding the measurement results from a single spectral curve to a multi-distance spectral group that can be used for longitudinal inversion.

[0152] The main technical points are: (1) using broadband infrared light source and Michelson interferometry; (2) using a moving mirror to scan in three steps to obtain a complete interferogram; and (3) performing Fourier transform on the interferograms under different distance conditions.

[0153] By independently acquiring complete interferograms and performing Fourier transforms under each static distance condition, broadband spectral information can be preserved under each distance condition, ensuring that longitudinal encoding and spectral information do not interfere with each other, thereby increasing the longitudinal resolution dimension while maintaining spectral integrity.

[0154] 3. Achieve quantitative spectroscopic characterization of multilayer structures under non-destructive conditions.

[0155] This application can obtain infrared spectral information of different depth layers of multilayer structures without cutting, peeling or destructive preparation of the sample, and is suitable for non-destructive testing and online analysis scenarios.

[0156] The main technical points for achieving this effect are: (1) non-contact measurement based on near-field optical interaction; (2) extraction of depth information through distance coding and data inversion; and (3) longitudinal resolution can be achieved without changing the sample structure.

[0157] This application achieves longitudinal infrared spectral information encoding by changing the probe-sample distance and interferometric measurement parameters, while keeping the sample structure unchanged throughout the process. Therefore, longitudinal resolution spectral measurement of multilayer materials can be completed under non-destructive conditions.

[0158] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0159] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A longitudinally resolved near-field infrared spectroscopy measurement system, characterized in that, The longitudinally resolved near-field infrared spectroscopy measurement system includes: an infrared light source, an interferometric measurement module, a nanoprobe, a control module, a data acquisition module, and a data processing module; the interferometric measurement module includes: a semi-reflective mirror and a moving mirror; The infrared light source emits infrared light; the semi-reflective mirror reflects the infrared light to the moving mirror and transmits it to the nanoprobe; the moving mirror returns the reference light; the nanoprobe excites a local near field under the irradiation of the infrared light, forming a near-field infrared light field, which interacts with the sample to be tested to obtain a backscattered light signal containing information about the sample; the semi-reflective mirror also combines the reference light and the backscattered light signal to produce interference, obtaining a near-field interference signal; the control module controls the moving mirror to change the optical path difference in a step-shift manner, and changes the distance between the nanoprobe and the sample to be tested at each step position of the moving mirror corresponding to the optical path difference; the data acquisition module acquires the near-field interference signal group corresponding to each optical path difference, the near-field interference signal group including multiple near-field interference signals at different static probe-sample distances; the data processing module performs data rearrangement, Fourier transform, and depth inversion processing based on the near-field interference signal groups corresponding to all optical path differences to obtain the longitudinally resolved infrared spectra of different depth layers of the sample to be tested. The data processing module includes: The matrix construction unit is used to construct a two-dimensional sampling data matrix according to the optical path difference dimension and the distance dimension based on the near-field interference signal groups corresponding to all optical path differences. The data rearrangement unit is used to rearrange the two-dimensional sampling data matrix according to the distance dimension and extract the interferogram corresponding to each static probe-sample distance; The Fourier transform unit is used to perform Fourier transform on each of the interferograms to obtain the near-field infrared spectrum at each static probe-sample distance; The depth inversion unit is used to perform depth inversion on each of the near-field infrared spectra to obtain the longitudinally resolved infrared spectra of different depth layers of the sample under test. The Fourier transform unit is specifically used for: The interferograms are processed by removing DC components and weighting with window functions to obtain the processed interferograms. Perform Fourier transform on each processed interferogram to obtain the Fourier transform results; Phase correction was performed on the Fourier transform results to obtain the near-field infrared spectrum at each static probe-sample distance; The deep inversion unit is specifically used for: A physical constraint-based inversion algorithm was used to perform depth inversion on the near-field infrared spectra of each sample to obtain the longitudinally resolved infrared spectra of different depth layers of the sample under test.

2. The longitudinally resolved near-field infrared spectroscopy measurement system according to claim 1, characterized in that, The infrared light source is a broadband infrared light source; the spectral coverage of the broadband infrared light source is 400 cm⁻¹. -1 ~4000cm -1 The output power of the broadband infrared light source is 0.1mW~20mW.

3. The longitudinally resolved near-field infrared spectroscopy measurement system according to claim 1, characterized in that, The number of step points of the moving mirror is 128 to 4096; the maximum scanning range of the moving mirror is 50 μm to 300 μm; and the dwell time of the moving mirror at each step position is 0.1 ms to 100 ms.

4. The longitudinally resolved near-field infrared spectroscopy measurement system according to claim 1, characterized in that, The number of static probe-sample distances is 2 to 100; the interval between static probe-sample distances is 0.1 nm to 20 nm; and the switching time between static probe-sample distances is 0.1 ms to 10 ms.

5. The longitudinally resolved near-field infrared spectroscopy measurement system according to claim 1, characterized in that, The interferometric measurement module further includes a parabolic mirror; the parabolic mirror is disposed in the optical path between the semi-reflective lens and the nanoprobe.

6. A longitudinally resolved near-field infrared spectroscopy measurement method, characterized in that, The longitudinally resolved near-field infrared spectroscopy measurement method is implemented using the longitudinally resolved near-field infrared spectroscopy measurement system as described in any one of claims 1-5, and the longitudinally resolved near-field infrared spectroscopy measurement method includes: The optical path difference is changed by controlling the moving mirror to change the optical path difference in a step-like manner, and the distance between the nanoprobe and the sample to be tested is changed under the optical path difference corresponding to each step position of the moving mirror; Acquire the near-field interference signal group corresponding to each optical path difference, wherein the near-field interference signal group includes multiple near-field interference signals at different static probe-sample distances; Based on the near-field interference signal groups corresponding to all optical path differences, data rearrangement, Fourier transform, and depth inversion processing are performed to obtain the longitudinally resolved infrared spectra of different depth layers of the sample under test.

7. The longitudinally resolved near-field infrared spectroscopy measurement method according to claim 6, characterized in that, Based on the near-field interferometric signal sets corresponding to all optical path differences, data rearrangement, Fourier transform, and depth inversion processing are performed to obtain the longitudinally resolved infrared spectra of the sample at different depths, specifically including: Based on the near-field interference signal groups corresponding to all optical path differences, a two-dimensional sampling data matrix is ​​constructed according to the optical path difference dimension and the distance dimension. The two-dimensional sampling data matrix is ​​rearranged according to the distance dimension to extract the interferogram corresponding to each static probe-sample distance; Perform Fourier transform on each of the interferograms to obtain the near-field infrared spectrum at each static probe-sample distance; Depth inversion was performed on the near-field infrared spectra to obtain the longitudinally resolved infrared spectra of the sample at different depths.

Images