A long-wave infrared pulse spectrum real-time measurement method and device

Abstract

A real-time measurement method and apparatus for long-wave infrared pulse spectra based on nonlinear frequency upconversion is disclosed. The method utilizes a non-oxide semiconductor crystal as the nonlinear medium and employs a narrowband pulse with a center wavelength in the range of 2-2.5 μm as the sampling light. This pulse is summed with the long-wave infrared pulse to be measured, and the resulting summed pulse is received and detected by a near-infrared real-time spectrometer. The combination of the selected sampling light wavelength and the selected crystal generates an ultra-wide phase-matching bandwidth. Through a linear frequency mapping relationship, the spectrum of the long-wave infrared pulse to be measured can be inverted from the detected summed pulse spectrum. By analyzing the spectrum acquired by the near-infrared spectrometer in real time, the spectrum of the long-wave infrared pulse to be measured can be directly and in real-time reconstructed. This invention overcomes the limitations of Fourier transform spectrometers and traditional grating spectrometers in real-time measurement, achieving real-time, mechanical-scan-free measurement of long-wave infrared pulse spectra without requiring a specific repetition frequency for the long-wave infrared pulse.

Description

Technical Field

[0001] This invention belongs to the field of laser technology, specifically relating to a real-time measurement method and device for long-wave infrared pulse spectra based on nonlinear frequency upconversion, which is suitable for rapid acquisition and analysis of the spectra of ultrashort laser pulses in the long-wave infrared band (8-15 μm). Background Technology

[0002] Since its inception in the mid-20th century, laser technology has profoundly influenced the development of scientific research and engineering, and is widely used in fields such as spectral analysis, environmental monitoring, and biomedicine. With the continuous advancement of laser technology, the scope of spectral research has expanded from the traditional visible and near-infrared bands to the mid-wave infrared (MWIR, 3-8 μm) and long-wave infrared (LWIR, 8-15 μm) regions. Techniques such as difference frequency generation (DFG), optical parametric oscillation (OPO), and optical parametric amplification (OPA) based on nonlinear optical frequency conversion have become the main means of generating long-wave infrared ultrashort laser pulses. However, the development of real-time detection technology for long-wave infrared pulse spectra faces many challenges. The relative scarcity of materials and devices for the long-wave infrared band, and the insufficient maturity and stability of existing materials and devices, significantly restrict the development of real-time detection technology for long-wave infrared pulse spectra, causing its development to lag behind corresponding laser generation technologies.

[0003] Currently, direct detection methods for long-wave infrared spectra mainly include Fourier transform infrared spectrometers and traditional grating spectrometers. However, both have inherent limitations, making it difficult to achieve real-time analysis and measurement of long-wave infrared pulses. Fourier transform infrared spectrometers rely on the precise displacement of a moving mirror to acquire interferograms, but this mechanical motion is susceptible to environmental vibrations, leading to accumulated phase errors and degraded spectral resolution. Furthermore, when the pulse repetition frequency is ≤1kHz, the number of pulses received in a single scan is extremely small, resulting in insufficient light signal accumulation and a sharp drop in the signal-to-noise ratio, making effective signal extraction impossible and thus failing to meet the requirements of dynamic spectral measurements. Traditional grating spectrometers require high-cost, cryogenically cooled semiconductor detectors (such as HgCdTe, InSb, PbSe, etc.) to achieve wavelength resolution and complete the full-spectrum scan measurement through the mechanical rotation of the grating. Under this scanning mechanism, a single complete measurement often takes several minutes to tens of minutes. Therefore, how to achieve real-time measurement of long-wave infrared pulse spectra has become a key technical challenge urgently needing to be solved in this field, which is crucial for the optimization and adjustment of laser systems and the rapid analysis of experimental results.

[0004] In the field of molecular fingerprinting, there exists an infrared upconversion detection technique. This technique aims to overcome the limitations of traditional infrared detectors, such as low sensitivity and high cost. Its core lies in mapping the infrared light signal carrying the spectral characteristics of the sample to the visible light band through a nonlinear frequency upconversion process (using a specific crystal material), and then detecting it using a high-performance, low-cost visible light detector. By comparing the spectral differences with and without the sample, the characteristic spectral lines of the sample can be obtained. It is worth noting that this technique measures the spectral characteristics of the sample, not the spectrum of the infrared pulse itself; therefore, the bandwidth requirement for the upconversion process is not high. This technique is currently widely used in the 2-3 μm band, and the commonly used upconversion process is a frequency doubling process or a sum-frequency process with near-infrared lasers such as those at 1 μm. The commonly used crystal is periodically polarized lithium niobate. However, limited by the transparency range and phase-matching bandwidth of the crystal material, this technique is difficult to directly apply to the measurement of long-wavelength infrared pulse spectra. Summary of the Invention

[0005] To overcome the limitations of existing Fourier transform infrared spectrometers and grating spectrometers in achieving real-time, high-fidelity spectral measurements in the long-wave infrared band, this invention provides a real-time measurement method and apparatus for long-wave infrared pulse spectra based on nonlinear frequency up-conversion. This method constructs an ultra-wideband sum-frequency process to map the spectrum of the long-wave infrared pulse to be measured to the near-infrared band. A mature near-infrared real-time spectrometer is used for detection, and the original spectrum is obtained by inverting the wavelength mapping relationship, thus achieving real-time acquisition of long-wave infrared pulse spectra without mechanical scanning.

[0006] The principle of this invention is to focus a long-wave infrared pulse into a nonlinear crystal using a long-focal-length concave mirror. The focused long-wave infrared pulse then interacts with a near-infrared sampling pulse processed by a narrowband filter within the nonlinear crystal, generating a sum-frequency pulse. The selected near-infrared sampling pulse, combined with the chosen nonlinear crystal, produces an ultra-wide phase-matching bandwidth, enabling full mapping of the spectrum of the long-wave infrared pulse under test. This sum-frequency pulse is then detected using a real-time near-infrared spectrometer, and the original spectrum of the long-wave infrared pulse is obtained by inverting the specific wavelength mapping relationship.

[0007] The technical solution of the present invention is as follows: A real-time measurement method for long-wave infrared pulse spectra based on nonlinear frequency up-conversion is characterized by the following steps: S1. Sampling wavelength selection: Select a narrowband laser pulse with a center wavelength in the range of 2-2.5 μm as the sampling pulse; S2. Sum-frequency process creation: A non-oxide semiconductor crystal is selected as the nonlinear medium, and the long-wave infrared pulse to be measured is subjected to a sum-frequency interaction with the sampling pulse to generate a sum-frequency pulse; the wavelength of the sampling pulse and the material combination of the non-oxide semiconductor crystal satisfy the ultrawide phase matching condition, so that at the crystal cutting angle, all wavelength components of the long-wave infrared pulse to be measured can effectively participate in the sum-frequency process, thereby forming a mapping relationship between the spectrum of the long-wave infrared pulse to be measured and the spectrum of the sum-frequency pulse; S3. Sum-frequency spectral acquisition: The spectrum of the sum-frequency pulses is measured using a near-infrared real-time spectrometer; S4. Spectral data inversion: Based on the frequency linear mapping relationship, the spectrum of the long-wave infrared pulse to be measured is obtained by inversion.

[0008] Furthermore, the bandwidth of the sampling pulse determines the wavelength mapping accuracy from the long-wave infrared pulse to the sum-frequency pulse; when the sampling pulse is a broadband pulse, it needs to be narrowbanded after step S1 and before step S2.

[0009] Furthermore, the non-oxide semiconductor crystal includes, but is not limited to, one of AgGaS2, BaGa2GeSe6, GaSe, or LiGaTe2; the synergistic selection of the sampling pulse wavelength and the non-oxide semiconductor crystal material ensures that, at a fixed crystal cutting angle, the phase-matching bandwidth covers the long-wave infrared range to be measured.

[0010] The near-infrared sampling pulse can be generated by the optical path that generates long-wave infrared pulses, or it can be obtained by a fourth harmonic of the long-wave infrared pulse, or introduced by other means.

[0011] The narrowband processing unit includes, but is not limited to, bulk material chirping, narrowband filter filtering, and grating pair frequency selection. Taking bulk material chirping as an example, by adjusting the amount of introduced dispersion, the spectral width of the near-infrared sampling pulse that overlaps with the time of the ultra-short and long-wave infrared pulses can be changed, effectively achieving the effect of spectral narrowing.

[0012] The sum-frequency process requires a large phase-matching bandwidth to ensure that the spectrum of the long-wave infrared pulse to be measured is mapped one by one to the spectrum of the near-infrared sum-frequency pulse.

[0013] The nonlinear crystal is transparent to the long-wave infrared pulse, near-infrared sampling pulse, and sum-frequency pulse under test, and has a large effective nonlinear coefficient. It can be cut according to a specific phase-matching angle, and broadband upconversion can be achieved without adjusting the crystal angle during use.

[0014] The wavelength mapping formula is: 1 / λ 和频 = 1 / λ 取样 + 1 / λ 长波红外 (Conservation of energy).

[0015] The real-time measurement device for long-wave infrared pulse spectroscopy that implements the above measurement method comprises a near-infrared sampling pulse, a time delay line, a narrowband processing unit, a focusing concave mirror, a reflector, a nonlinear crystal, a sum-frequency pulse, an imaging lens, a real-time spectrometer, and an inversion calculation module.

[0016] The focusing concave mirror focuses long-wave infrared pulses into the nonlinear crystal, thereby increasing the light intensity.

[0017] The time delay line is used to adjust the relative delay between the long-wave infrared pulse and the near-infrared sampling pulse, ensuring that the two overlap in time within the nonlinear crystal.

[0018] The technical effects of this invention are as follows: By designing the wavelength and bandwidth of the sampling light and selecting a suitable nonlinear crystal, a broadband sum-frequency process is constructed to map the entire spectrum of the long-wave infrared pulse to be measured to the near-infrared band. Sum-frequency spectral data are acquired in real time in the near-infrared band, and then the spectrum of the long-wave infrared pulse can be obtained in real time according to the wavelength mapping relationship of the sum-frequency process. This process has a simple structure, requires no mechanical scanning, and has real-time measurement capabilities. This invention can not only be directly applied to the spectral characterization of long-wave infrared pulses, but also to measuring the absorption spectra of infrared samples, analyzing transient chemical reactions, and detecting multi-component gases, demonstrating strong compatibility. Attached Figure Description

[0019] Figure 1 This is the core working principle of real-time measurement of long-wave infrared pulse spectra based on nonlinear frequency upconversion in the embodiments of the present invention.

[0020] Figure 2 This is a schematic diagram of a real-time measurement device for long-wave infrared pulse spectra based on nonlinear frequency upconversion, according to an embodiment of the present invention.

[0021] Figure 3 Phase matching bandwidth diagrams of different crystals near 9 μm at two sampling wavelengths: 1030 nm and 2.3 μm.

[0022] Figure 4 Phase-matching bandwidth diagram of 2.3 μm and 9 μm sum frequencies in AgGaS2 crystal.

[0023] Figure 5 Upconversion measurement inversion spectra corresponding to different sampling pulse chirps under Gaussian spectroscopy.

[0024] Figure 6 Upconversion measurement inversion spectra corresponding to different sampling pulse chirp amounts under single-concave modulation spectra.

[0025] Figure 7Upconversion measurement inversion spectra corresponding to different sampling pulse chirp amounts under dual-concave modulation spectra.

[0026] Figure 8 Upconversion measurement inversion spectra corresponding to different sampling pulse chirp amounts under three-concave modulation spectra.

[0027] The numbers in the figure are as follows: 1 is the long-wave infrared pulse to be measured, 2 is the near-infrared sampling pulse, 3 is the time delay line, 4 is the narrowband processing unit, 5 is the focusing concave mirror, 6 is the reflecting mirror, 7 is the nonlinear crystal, 8 is the sum-frequency pulse, 9 is the imaging lens, 10 is the real-time spectrometer, and 11 is the inversion calculation module. Detailed Implementation

[0028] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments, but the present invention is not limited to the following embodiments. All equivalent substitutions or modifications made based on the technical solutions of the present invention should fall within the protection scope of the present invention.

[0029] Figure 1 This diagram illustrates the principle of real-time measurement of long-wave infrared pulse spectra based on nonlinear frequency upconversion. A wide-spectrum, narrow-pulse-width long-wave infrared pulse to be measured interacts with a narrow-spectrum, wide-pulse-width sampling pulse in a nonlinear crystal, generating a sum-frequency pulse in the near-infrared band. The spectral bandwidth of the sampling pulse is much smaller than that of the long-wave infrared pulse to be measured. Under the condition of broadband phase matching, different wavelength components in the long-wave infrared pulse to be measured can interact with the sampling pulse, thus establishing a one-to-one mapping between the long-wave infrared pulse spectrum and the sum-frequency pulse spectrum. By measuring the spectrum of this sum-frequency pulse using a mature near-infrared real-time spectrometer and combining it with the energy conservation principle, the spectrum of the original long-wave infrared pulse to be measured can be retrieved, achieving real-time measurement of the long-wave infrared pulse spectrum.

[0030] like Figure 2 As shown, this embodiment provides a real-time measurement device for long-wave infrared pulse spectra based on nonlinear frequency upconversion, including a long-wave infrared pulse to be measured 1, a near-infrared sampling pulse 2, a time delay line 3, a narrowband processing unit 4, a focusing concave mirror 5, a reflector 6, a nonlinear crystal 7, a sum-frequency pulse 8, an imaging lens 9, a real-time spectrometer 10, and an inversion calculation module 11.

[0031] The long-wave infrared pulse 1 to be tested is first incident on the time delay line 3, which consists of a pair of high-reflectivity right-angle mirrors. The optical path is controlled by a precision displacement stage to finely adjust the relative time delay between the long-wave infrared pulse to be tested and the near-infrared sampling pulse, ensuring that the two pulses achieve effective time overlap within the nonlinear crystal. The long-wave infrared pulse adjusted by the delay line is reflected by the focusing concave mirror 5 (preferably an off-axis parabolic mirror) and focused onto the center of the nonlinear crystal 7 to obtain a higher peak power density and enhance the nonlinear frequency conversion efficiency.

[0032] After passing through the narrowband processing unit 4, the near-infrared sampling pulse 2 is reflected by the reflector 6 and incident on the nonlinear crystal 7 in a small-angle non-collinear manner, ensuring that it spatially overlaps with the focused long-wave infrared pulse 1 to be measured inside the crystal.

[0033] Inside the nonlinear crystal 7, two pulses interact with each other to generate a sum-frequency pulse 8. After being emitted, the sum-frequency pulse 8 is imaged by the imaging lens 9 and projected onto the real-time spectrometer 10 for spectral measurement. The sum-frequency spectral data acquired by the real-time spectrometer 10 is input to the inversion calculation module 11. This module performs inversion calculations based on a preset wavelength mapping relationship (based on the law of conservation of energy), and combines the wavelength mapping lookup table with the calibrated curve to finally output and display the original spectrum of the long-wave infrared pulse 1 to be measured.

[0034] Narrowband processing unit 4 is a non-essential unit, only applicable when the bandwidth of the near-infrared sampling pulse itself exceeds the required spectral resolution. This unit can be omitted when the sampling pulse itself already possesses a sufficiently narrow spectral bandwidth. In one embodiment, narrowband processing unit 4 is implemented using bulk material chirping, i.e., a bulk material with a large group velocity dispersion is introduced into the propagation path of the sampling pulse, causing the near-infrared sampling pulse to be temporally stretched into a chirped pulse. At this time, different frequency components in the sampling pulse are temporally separated, so that only the portion of the near-infrared spectrum that temporally overlaps with the ultra-short and long-wave infrared pulse participates in the sum-frequency interaction within the nonlinear crystal, thereby achieving spectral narrowing of the sampling pulse in an equivalent sense. By changing the thickness of the bulk material, the introduced group delay dispersion parameter can be continuously adjusted, achieving precise control of the spectral width of the sampling pulse participating in the sum-frequency process. The following embodiments will use bulk material chirping as the narrowband processing unit.

[0035] An example is provided. The core of this invention lies in constructing a sum-frequency process that can achieve ultra-wideband phase matching at a fixed crystal cutting angle. Calculation results show that when the sampling pulse wavelength is around 1 μm, conventional monolithic nonlinear crystals struggle to meet the phase matching bandwidth required for long-wave infrared pulse broadband measurements at a fixed angle. Although this can be achieved by introducing additional auxiliary techniques, this increases the system complexity. However, when the sampling pulse wavelength is selected in the range of 2-2.5 μm, various non-oxide semiconductor crystals can provide significantly extended phase matching bandwidth at a fixed angle. Figure 3 This study compares the phase-matching bandwidths of four crystals (AgGaS2, BaGa2GeSe6, GaSe, and LiGaTe2) around 9 μm under sampling pulse wavelengths of 1030 nm and 2.3 μm. The crystal length was set to 0.4 mm for all calculations, and type I phase matching was used for all cases. The phase matching method for AgGaS2 and GaSe was: 9 μm (o-light) + sampling light (o-light) → sum-frequency light (e-light); the phase matching method for BaGa2GeSe6 and LiGaTe2 was: 9 μm (e-light) + sampling light (e-light) → sum-frequency light (o-light). At a sampling wavelength of 1030 nm, the phase-matching angles for the four crystals were 39.9°, 31°, 16°, and 33.6°, respectively; at a sampling wavelength of 2.3 μm, the phase-matching angles were 33.3°, 23.5°, and 11.4°, respectively. 26.1°, the shaded area in the figure represents the spectrum of the actual 9 μm output system, and the vertical axis represents the normalized intensity value of the phase matching efficiency function at different wavelengths. It can be seen that the phase matching bandwidth obtained using a 2.3 μm sampling pulse is significantly wider, capable of covering the spectral range of the actual long-wave infrared pulse system. In a specific embodiment, a 0.4 mm thick AgGaS2 crystal is selected as the nonlinear sum-frequency crystal, with a phase matching method of 9 μm (e-ray) + 2.3 μm (o-ray) → sum-frequency (e-ray), a phase matching angle of 38.2°, and an effective nonlinear coefficient of approximately 14.7 pm / V. For example... Figure 4 As shown, under this phase-matching condition, the sum-frequency process exhibits high conversion efficiency in the range of approximately 6-14 μm, enabling effective mapping of the wide spectrum of long-wave infrared pulses.

[0036] Application examples are as follows. Under the aforementioned crystal (i.e., an AgGaS2 crystal with a thickness of 0.4 mm) and optical path configuration, the measurement performance of the method of this invention was verified through numerical simulation. The center wavelength of the near-infrared sampling pulse was set to 2.3 μm, the initial pulse width to 110 fs, and the peak light intensity to 10 GW / cm². 2The long-wave infrared pulse to be tested has a center wavelength of 9 μm, a pulse width of 91 fs, and a peak intensity of 10 MW / cm². 2 Based on the nonlinear coupled-wave equations, the sum-frequency process is numerically calculated using the distributed Fourier method combined with the Runge-Kutta algorithm. By changing the group delay dispersion parameter introduced into the sampling pulse, the changes in the upconversion measurement inversion spectrum under different chirp conditions are analyzed.

[0037] like Figure 5 As shown, when the spectrum of the long-wave infrared pulse to be measured has an ideal Gaussian distribution, the group delay dispersion parameter of the sampled pulse is set to 0 fs. 2 5000 fs 2 10000 fs 2 and 30000 fs 2 The chirp levels correspond to different degrees of chirp. It can be seen that when the sampling pulse introduces no chirp or has a small chirp, due to the broad spectrum of the sampling pulse itself, different long-wave infrared frequency components will simultaneously interact with multiple frequency components in the sampling pulse, resulting in significant broadening and distortion of the retrieved long-wave infrared spectrum. As the group delay dispersion parameter gradually increases, the effective spectral bandwidth of the sampling pulse participating in the interaction gradually decreases. The upconversion measurement inversion spectrum shows a significant improvement in the agreement between the peak position, spectral width, and overall profile and the original Gaussian spectrum, indicating that narrowing the sampling pulse spectrum plays a crucial role in achieving high-fidelity spectral mapping.

[0038] Based on this, to simulate the absorption or modulation characteristics commonly found in actual long-wave infrared pulse spectra, a single concave modulation structure was introduced into the original Gaussian spectrum as a new input for the long-wave infrared pulse spectrum to be measured. The modulation position was 9 μm, the modulation depth was 0.8, and the modulation width was 300 nm. The group delay dispersion parameter of the sampling pulse was set to 0 fs. 2 25000fs 2 50000 fs 2 and 200,000 fs 2 , Figure 6 The results of upconversion measurement inversion spectra under different group delay dispersion parameters are presented. It can be seen that when the sampling pulse chirp is small, the concave structures in the inversion spectrum are significantly weakened or even difficult to resolve; however, as the group delay dispersion parameter increases to 200,000 fs... 2 The upconversion measurement inversion spectrum can not only accurately reproduce the location of the depression, but its depth and shape are also in good agreement with the original long-wave infrared pulse spectrum, indicating that the method of the present invention has the ability to accurately distinguish and reconstruct a single absorption feature.

[0039] Two concave modulation structures were further introduced into the original Gaussian long-wave infrared pulse spectrum to simulate multi-absorption peak or multi-component spectral conditions. One concave modulation position was 9 μm, modulation depth was 0.9 μm, and modulation width was 200 nm; the other concave modulation position was 8.5 μm, modulation depth was 0.5 μm, and modulation width was 200 nm. The corresponding upconversion measurement inversion spectral results are as follows: Figure 7 As shown, it can be observed that when the chirp is insufficient, the two concave features are prone to aliasing in the inversion spectrum, leading to structural ambiguity or mutual interference. However, when a large group delay dispersion is introduced into the sampling pulse, the positions of the two concave features in the inversion spectrum can be clearly distinguished, and their relative depth relationship is basically consistent with the original long-wave infrared pulse spectrum, indicating that the method can still maintain high spectral resolution under multi-feature spectral conditions. Next, a complex spectral model containing three concave modulation structures was constructed. The modulation positions of the three concave features are 9 μm, 8.5 μm, and 10 μm, respectively, with modulation depths of 0.9, 0.8, and 0.7 μm, and modulation widths of 200 nm for each. The corresponding upconversion measurement inversion spectral results are shown below. Figure 8 As shown. The results indicate that, with a group delay dispersion parameter of 200000 fs... 2 When the inverted spectrum is obtained, it can reproduce the overall distribution characteristics of multiple depression structures in the original long-wave infrared pulse spectrum well, and the positional relationship between the depressions is well preserved. This shows that even under the condition of relatively complex spectral structure, by reasonably controlling the chirp of the sampling pulse, the method of the present invention can still achieve stable and high-fidelity inversion of long-wave infrared pulse spectrum.

[0040] The above embodiments demonstrate that the real-time measurement method and apparatus for long-wave infrared pulse spectra based on nonlinear frequency upconversion proposed in this invention, through reasonable control of the wavelength and bandwidth of near-infrared sampling pulses and in combination with a nonlinear crystal that satisfies the broadband phase-matching condition, can construct an ultra-wideband sum-frequency process, achieving a high-fidelity mapping from long-wave infrared pulse spectra to near-infrared sum-frequency spectra. By real-time measurement of the sum-frequency pulse spectrum and inversion calculation based on the wavelength mapping relationship, the original spectral information of the long-wave infrared pulse to be measured can be obtained. Therefore, this invention achieves real-time acquisition of long-wave infrared pulse spectra without the need for mechanical scanning. It features a simple overall structure, clear parameter control, good stability and applicability, and can be applied to various long-wave infrared spectroscopy measurement scenarios.

Claims

1. A real-time measurement method for long-wave infrared pulse spectra based on nonlinear frequency up-conversion, characterized in that, Includes the following steps: S1. Sampling wavelength selection: Select a narrowband laser pulse with a center wavelength in the range of 2-2.5 μm as the sampling pulse; S2. Sum-frequency process creation: A non-oxide semiconductor crystal is selected as the nonlinear medium, and the long-wave infrared pulse to be measured is subjected to a sum-frequency interaction with the sampling pulse to generate a sum-frequency pulse; the wavelength of the sampling pulse and the material combination of the non-oxide semiconductor crystal satisfy the ultrawide phase matching condition, so that at the crystal cutting angle, all wavelength components of the long-wave infrared pulse to be measured can effectively participate in the sum-frequency process, thereby forming a mapping relationship between the spectrum of the long-wave infrared pulse to be measured and the spectrum of the sum-frequency pulse; S3. Sum-frequency spectral acquisition: The spectrum of the sum-frequency pulses is measured using a near-infrared real-time spectrometer; S4. Spectral data inversion: Based on the frequency linear mapping relationship, the spectrum of the long-wave infrared pulse to be measured is obtained by inversion.

2. The real-time measurement method according to claim 1, characterized in that, The bandwidth of the sampling pulse determines the wavelength mapping accuracy from the long-wave infrared pulse to the sum-frequency pulse; when the sampling pulse is a broadband pulse, it needs to be narrowbanded after step S1 and before step S2.

3. The real-time measurement method according to claim 1, characterized in that, The non-oxide semiconductor crystal has high transmittance for all three beams of light involved in the interaction; the synergistic selection of the sampling pulse wavelength and the non-oxide semiconductor crystal material ensures that, at a fixed crystal cutting angle, the phase-matching bandwidth covers the long-wave infrared range to be measured.

4. The real-time measurement method according to claim 2, characterized in that, The narrowing process employs a chirped narrowing method, which introduces group delay dispersion through the bulk material to transform the broadband sampling pulse into a chirped pulse. This ensures that only the spectral components that overlap temporally with the ultra-short and long-wave infrared pulses participate in the sum-frequency interaction within the nonlinear crystal, thereby effectively achieving spectral narrowing of the sampling pulse.

5. The real-time measurement method according to claim 4, characterized in that, By changing the thickness or insertion length of the bulk material, the introduced group delay dispersion parameter can be continuously adjusted to achieve precise control over the spectral width of the sampling pulses involved in the sum-frequency interaction, thus matching different spectral measurement resolution requirements.

6. The real-time measurement method according to claim 4, characterized in that, By changing the relative delay between the sampling pulse and the long-wave infrared pulse, the center wavelength of the sampling pulse involved in the sum-frequency interaction can be continuously adjusted to meet the design requirements of sum-frequency processes in different bands.

7. The real-time measurement method according to claim 4, characterized in that, The narrowbanding process can also be achieved using narrowband filter filtering or grating pair frequency selection.

8. The real-time measurement method according to claim 1, characterized in that, The repetition frequency of the sampling pulse is an integer multiple of the long-wave infrared pulse to be measured, and n≥1.

9. The real-time measurement method according to claim 1, characterized in that, The long-wave infrared pulse to be tested and the sampling pulse are incident on the non-oxide semiconductor crystal at a small angle and in a non-collinear manner, so that the generated sum-frequency pulse is naturally separated from the two incident beams in the spatial propagation path.

10. A real-time measurement device based on long-wave infrared pulse spectrum using nonlinear frequency up-conversion, characterized in that, include: Near-infrared sampling pulse (2), time delay line (3), narrowband processing unit (4), focusing concave mirror (5), reflecting mirror (6), nonlinear crystal (7), sum frequency pulse (8), imaging lens (9), real-time spectrometer (10) and inversion calculation module (11). The nonlinear crystal (7) is located at the intersection of the outgoing light from the focusing concave mirror (5) and the reflecting mirror (6), and is a non-oxide semiconductor crystal cut at a fixed angle. The nonlinear crystal (7) is combined with a near-infrared sampling pulse (2) with a center wavelength in the range of 2-2.5 μm, and has an ultra-wide phase matching bandwidth covering the long-wave infrared range to be measured under the fixed cutting angle. The long-wave infrared pulse (1) to be measured is focused into the nonlinear crystal (7) by the focusing concave mirror (5) after the optical path is adjusted by the time delay line (3); The near-infrared sampling pulse (2) is narrowed by the narrowband processing unit (4) and then reflected by the mirror (6) into the nonlinear crystal (7), where it interacts with the focused long-wave infrared pulse to generate a sum-frequency pulse (8). The sum-frequency pulse (8) is imaged by the imaging lens (9) to the real-time spectrometer (10). The real-time spectrometer (10) collects the spectral data of the sum-frequency pulse, and the inversion calculation module (11) inverts the original spectrum of the long-wave infrared pulse (1) to be measured based on the collected spectral data.

11. The real-time measurement device for long-wave infrared pulse spectra based on nonlinear frequency up-conversion according to claim 9, characterized in that, The nonlinear crystal (7) has high transmittance for all three beams of light involved in the interaction. The crystal length ranges from 0.1 to 2 mm. The cutting angle is optimized according to the type I or type II phase matching conditions. It is combined with the near-infrared sampling pulse (2) with a center wavelength in the range of 2-2.5 μm to achieve an ultra-wide phase matching bandwidth covering the long-wave infrared range to be measured under a fixed cutting angle.

12. The real-time measurement device for long-wave infrared pulse spectra based on nonlinear frequency up-conversion according to claim 9, characterized in that, The narrowband processing unit (4) is a bulk material dispersion introduction module, which introduces group delay dispersion to turn the near-infrared sampling pulse into a chirped pulse, thereby achieving equivalent spectral narrowing; the bulk material dispersion introduction module includes a set of infrared optical materials with replaceable thickness or adjustable insertion length, the infrared optical materials including silicon and germanium; or, the narrowband processing unit (4) is a narrowband filter or grating pair spatial filtering structure.

13. The real-time measurement device for long-wave infrared pulse spectra based on nonlinear frequency up-conversion according to claim 9, characterized in that, The long-wave infrared pulse (1) to be tested and the near-infrared sampling pulse (2) after narrowband processing are incident on the nonlinear crystal (7) at a small angle and non-collinearly, so that the generated sum-frequency pulse (8) is naturally separated from the long-wave infrared pulse (1) to be tested and the near-infrared sampling pulse (2) in the spatial propagation path.

14. The real-time measurement device for long-wave infrared pulse spectra based on nonlinear frequency up-conversion according to claim 9, characterized in that, The inversion calculation module (11) has a pre-stored wavelength mapping lookup table or calibration curve, which is obtained by calibrating a narrowband light source with a known wavelength.

Images