High-energy narrow-pulse-width femtosecond pulse laser

Abstract

The invention relates to the technical field of lasers, and discloses a high-energy narrow-pulse-width femtosecond pulse laser. Comprising a pulse beam splitting and combining module which comprises a first polarization splitting prism, a second polarization splitting prism and a reflector; a laser beam output by the infrared light source is equally split into two beams of light through the first polarization splitting prism, and the reflector is used for increasing an additional optical path for at least one beam of light; the new light beam enters the nonlinear pulse compression module, and after nonlinear pulse compression, the peak power of a single pulse is increased to the damage threshold critical value of the lens; the compressed light beam passes through the pulse beam splitting and combining module again, the pulse walking for a long distance during the first beam splitting and combining passes for a short distance during the second beam splitting and combining, and the pulse walking for a short distance during the second beam splitting and combining passes for a long distance during the second beam splitting and combining, so that a time domain difference value between the pulses of the two light beams is compensated; and a super-strong pulse with a peak power super-lens damage threshold value is obtained at the light outlet.

Description

Technical Field

[0001] The present application relates to the technical field of lasers, and in particular, to a high-energy narrow-pulse-width femtosecond pulse laser. Background Art

[0002] In recent years, with the continuous increase in the processing requirements of new materials (such as metal compounds and non-metal composite materials), ultrafast pulse lasers have become increasingly crucial in the fields of fine microfabrication and semiconductor manufacturing due to their extremely high peak power and narrow pulse width. Especially for femtosecond millijoule-level pulses, their peak power and pulse quality directly affect the processing accuracy and efficiency. However, there are significant bottlenecks in realizing the output of such pulses in the prior art: traditional chirped pulse amplification (CPA) systems are limited by nonlinear effects and amplification gain narrowing, and it is difficult to directly generate femtosecond millijoule-level pulses. The mainstream solutions include:

[0003] 1. Nonlinear amplifier: Using fiber self-phase modulation to broaden the spectrum, but the fiber main amplifier has a low peak power tolerance, and the single-pulse energy is difficult to exceed the millijoule level, and it is easy to cause nonlinear noise;

[0004] 2. Coherent beam combination technology: By phase modulation to superimpose multiple low-energy pulses, although the total energy can be increased, it relies on high-cost liquid crystal modulators, and the industrial application is limited;

[0005] 3. Nonlinear pulse compression: Using a multi-pass cavity combined with a chirped mirror group to achieve spectral broadening and compression, but its core defect is that the damage threshold of the chirped mirror surface film layer is low, and high-peak-power pulses are likely to cause lens damage during the compression process. If the spot size is increased to reduce the power density, the number of chirped mirrors needs to be increased, resulting in a soaring cost and an expansion of the equipment volume.

[0006] Therefore, there is an urgent need for a laser to solve at least one of the above problems. Summary of the Invention

[0007] The present application provides a high-energy narrow-pulse-width femtosecond pulse laser, aiming to solve the problem that traditional chirped pulse amplification (CPA) systems are limited by nonlinear effects and amplification gain narrowing and it is difficult to directly generate femtosecond millijoule-level pulses.

[0008] In a first aspect, an embodiment of the present application provides a high-energy narrow-pulse-width femtosecond pulse laser, including:

[0009] A pulse splitting and combining module, including a first polarization beam splitter prism, a second polarization beam splitter prism, and a reflector;

[0010] An infrared light source, the laser beam output by the infrared light source is equally split into two beams of light by the first polarization beam splitter prism, the mirror is used to increase the extra optical path for at least one of the beams of light, so that a time domain difference is generated between the two beams of light, and the two beams of light are combined into a new beam of light by the second polarization beam splitter prism. The number of pulses in the new beam of light is twice that of the initial light, and the average power is the same as that of the initial light;

[0011] A non-linear pulse compression module, the new beam of light enters the non-linear pulse compression module, and after non-linear pulse compression, the peak power of a single pulse is increased to the critical value of the damage threshold of the lens; the compressed beam of light passes through the pulse splitting and combining module again. When splitting and combining for the first time, the pulse that travels a long path travels a short path when splitting and combining for the second time, and the pulse that travels a short path for the first time travels a long path when splitting and combining for the second time, so that the time domain difference between the pulses of the two beams of light is compensated, and the pulses of the two beams of light are superimposed in light intensity to obtain an ultra-strong pulse with a peak power exceeding the lens damage threshold at the light output port.

[0012] In some embodiments, the laser beam output by the infrared light source is equally split into two beams of light by the first polarization beam splitter prism, including: the first polarization beam splitter prism divides the laser beam into two linearly polarized light beams with orthogonal polarization directions, the light intensity ratios of the two beams of light are the same, and the initial pulses are time domain coincident.

[0013] In some embodiments, the mirror is used to increase the extra optical path for at least one of the beams of light, so that a time domain difference is generated between the two beams of light, including: the mirror constructs a long optical path for one of the beams of light, and the other beam of light is transmitted through a short optical path, and the optical path difference ΔL between the long optical path and the short optical path satisfies: ΔL = c * Δt, where Δt is the pulse time domain interval between the two beams of light, and c is the speed of light, so that the pulses of the two beams of light are separated and non-coincident in the time domain.

[0014] In some embodiments, the two beams of light are combined into a new beam of light by the second polarization beam splitter prism. The number of pulses in the new beam of light is twice that of the initial light, and the average power is the same as that of the initial light, including: the second polarization beam splitter prism performs polarization combination on the two orthogonally polarized light beams, the pulses of the two beams of light are separated in the time domain, and the combined new beam of light contains two pulses arranged in sequence in the time domain. The corresponding total pulse energy is equal to the initial single pulse energy, the average power remains unchanged, and the peak power of a single pulse is reduced to 1 / 2 of the initial value.

[0015] In some embodiments, the new light beam enters the nonlinear pulse compression module. After nonlinear pulse compression, the peak power of a single pulse is increased to the critical value of the damage threshold of the lens, including: the nonlinear pulse compression module includes a multi-pass cavity. When the new light beam is transmitted in the multi-pass cavity, the spectrum is broadened through the self-phase modulation effect. The broadened pulse is compressed by a chirped mirror group for phase compensation. After compression, the peak power of a single pulse is increased to be equal to or lower than the damage threshold of the surface film layer of the chirped mirror.

[0016] Exemplarily, the chirped mirror group of the nonlinear pulse compression module is provided with a plurality of chirped mirrors. The arrangement spacing and angle of the chirped mirrors satisfy that the spot diameter of the pulsed light beam is greater than the damage threshold spot size of the chirped mirror, ensuring that the pulsed beam with the critical peak power value does not damage the chirped mirror.

[0017] In some embodiments, the pulse that travels a long path during the first beam splitting and combining travels a short path during the second beam splitting and combining, and the pulse that travels a short path during the first time travels a long path during the second beam splitting and combining, so that the time-domain difference between the pulses of the two beams of light is compensated, and the pulses of the two beams of light achieve optical intensity superposition, including: during the second beam splitting and combining, through the optical path switching of the first polarization beam splitter prism and the second polarization beam splitter prism, the beam that passed through the long optical path for the first time is changed to pass through the short optical path, and the beam that passed through the short optical path for the first time is changed to pass through the long optical path. The optical path difference between the two beams of light is compensated in the reverse direction, and the time-domain interval is offset, so that the two pulses are completely coincident in the time domain, and the optical intensity achieves coherent superposition.

[0018] In some embodiments, the pulse beam splitting and combining module further includes an optical path switching device. The optical path switching device is used to swap the long optical path and the short optical path of the two beams of light during the first beam splitting and combining and the second beam splitting and combining. The optical path switching device includes a mirror group or a movable mirror, and adjusts the beam transmission path mechanically or electrically.

[0019] In some embodiments, the polarization beam splitting directions of the first polarization beam splitter prism and the second polarization beam splitter prism are the same and match the output polarization state of the infrared light source, ensuring that the polarization loss during beam splitting and combining is less than the preset loss rate.

[0020] In some embodiments, the single-pulse energy of the ultra-strong pulse at the light output port is ≥1 mJ, the pulse width is ≤100 fs, and the peak power density is lower than twice the damage threshold of the lens material in the nonlinear pulse compression module. Through the two-time optical intensity superposition of the pulse beam splitting and combining module, the multiplication effect of the peak power is achieved without damaging the lens.

[0021] The high-energy narrow-pulse-width femtosecond pulse laser provided by the embodiments of the present application realizes a two-step energy enhancement of "low-peak-power compression - high-energy pulse superposition" through the synergistic effect of the pulse beam splitting and combining module and the nonlinear pulse compression module: The first beam splitting and combining: The laser output by the infrared light source is divided into two beams of light with orthogonal polarization and coincident time domain by the first polarization beam splitter prism. An optical path difference ΔL = c*Δt is added to one of the beams through a mirror, so that the two light pulses are separated in the time domain (not coincident); After being combined by the second polarization beam splitter prism, a new beam containing two time-domain separated pulses is formed, and the peak power is reduced to 1 / 2 of the initial value (the average power remains unchanged), ensuring that the peak power is lower than the lens damage threshold when entering the nonlinear compression module. Nonlinear pulse compression: The new beam broadens the spectrum through self-phase modulation in the multi-pass cavity. After the chirped mirror group compensates the phase, the peak power of a single pulse is increased to the critical value of the damage threshold.

[0022] The second beam splitting and combining (optical path switching): By switching the long / short optical path of the two beams of light, the time-domain difference is compensated, so that the two pulses are completely coincident in the time domain, and an ultra-strong pulse with a peak power exceeding the damage threshold is output after the light intensities are superposed (single pulse energy ≥ 1 mJ, pulse width ≤ 100 fs).

[0023] The provided laser has the following beneficial effects:

[0024] 1. Avoid the risk of device damage: The peak power is controlled within the lens damage threshold during the compression process, avoiding the problem of chirped mirror damage caused by too high power in the traditional scheme;

[0025] 2. High-efficiency energy superposition: Through the optical path switching design of two beam splitting and combining operations, pulse energy coherent superposition (light intensity superposition) is realized after the time domain coincidence, breaking through the power limitation of single-path compression;

[0026] 3. Cost and volume optimization: There is no need to rely on high-cost multi-channel phase modulation devices or a large number of chirped mirrors. High-energy pulse output is realized through innovative optical path structures, reducing the system complexity and volume;

[0027] 4. Strong compatibility: It is applicable to the existing nonlinear compression module architecture, can be integrated with mainstream multi-pass cavities and chirped mirror groups, and improves the performance of existing equipment.

[0028] The above scheme systematically solves the device damage and cost problems in the generation of high-energy femtosecond pulses through a closed-loop design of "beam splitting to reduce peak - compression to increase peak - combining to superpose", providing an efficient and reliable technical path for industrial precision machining.

[0029] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. Description of the Drawings

[0030] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0031] Figure 1 It is a schematic optical path diagram of a high-energy narrow-pulse femtosecond pulse laser provided by an embodiment of the present application.

[0032] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. Specific Embodiments

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0034] The flowcharts shown in the accompanying drawings are only illustrative examples, and do not necessarily include all contents and operations / steps, nor do they necessarily need to be executed in the described order. For example, some operations / steps can also be decomposed, combined, or partially merged. Therefore, the actual execution order may change according to the actual situation.

[0035] It should be understood that, for the convenience of clearly describing the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish the same items or similar items with basically the same functions and effects. Those skilled in the art can understand that the terms "first", "second", etc. do not limit the quantity and execution order, and the terms "first", "second", etc. do not necessarily mean different.

[0036] It should be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0037] It should also be understood that the term " / and" used in the specification of the present application and the appended claims refers to any combination and all possible combinations of one or more of the related listed items, and includes these combinations.

[0038] The following will describe in detail some embodiments of the present application with reference to the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0039] In recent years, with the continuous increase in the processing requirements of new materials (such as metal compounds and non-metal composite materials), ultrafast pulsed lasers have become increasingly crucial in the fields of fine micromachining, semiconductor manufacturing, etc. due to their extremely high peak power and narrow pulse width. Especially for femtosecond millijoule-level pulses, their peak power and pulse quality directly affect the processing accuracy and efficiency. However, in the prior art, there are significant bottlenecks in realizing the output of such pulses: traditional chirped pulse amplification (CPA) systems are limited by nonlinear effects and amplification gain narrowing, and it is difficult to directly generate femtosecond millijoule-level pulses. The mainstream solutions include:

[0040] 1. Nonlinear amplifier: Using fiber self-phase modulation to broaden the spectrum, but the fiber main amplifier has a low peak power tolerance, and the single-pulse energy is difficult to exceed the millijoule level, and it is easy to generate nonlinear noise;

[0041] 2. Coherent beam combining technology: By phase modulation to superimpose multiple low-energy pulses, although the total energy can be increased, it depends on high-cost liquid crystal modulators, and the industrial application is limited;

[0042] 3. Nonlinear pulse compression: Using a multi-pass cavity combined with a chirped mirror group to achieve spectrum broadening and compression, but its core defect is that the damage threshold of the chirped mirror surface film layer is low, and the high-peak power pulse during the compression process is easy to cause damage to the lens. If the light spot is increased to reduce the power density, the number of chirped mirrors needs to be increased, resulting in a soaring cost and an expansion of the equipment volume.

[0043] Therefore, there is an urgent need for a laser to solve at least one of the above problems.

[0044] To solve the above problems, please refer to Figure 1, The embodiments of the present application provide a high - energy narrow - pulse - width femtosecond pulse laser, including: a pulse splitting and combining module, including a first polarization beam splitter prism, a second polarization beam splitter prism, and a mirror; an infrared light source, the laser beam output by the infrared light source is equally split into two beams by the first polarization beam splitter prism, the mirror is used to increase the extra optical path for at least one of the beams, so that a time - domain difference is generated between the two beams, and the two beams are combined into a new beam by the second polarization beam splitter prism. The number of pulses in the new beam is twice that of the initial light, and the average power is the same as that of the initial light; a non - linear pulse compression module, the new beam enters the non - linear pulse compression module, and after non - linear pulse compression, the peak power of a single pulse is increased to the critical value of the damage threshold of the lens; the compressed beam passes through the pulse splitting and combining module again. Among them, the pulse that travels the long path during the first splitting and combining travels the short path during the second splitting and combining, and the pulse that travels the short path during the first splitting and combining travels the long path during the second splitting and combining, so that the time - domain difference between the pulses of the two beams is compensated, and the pulses of the two beams achieve optical intensity superposition to obtain an ultra - strong pulse with a peak power exceeding the lens damage threshold at the light output port.

[0045] Specifically, the high - energy narrow - pulse - width femtosecond pulse laser proposed by the present invention realizes the generation of an ultra - strong pulse with a peak power exceeding the threshold without breaking through the lens damage threshold through an innovative architecture of "pulse splitting - time - delay combining - non - linear compression - optical path switching and secondary combining". The core modules include: Pulse splitting and combining module: Core components: First polarization beam splitter prism (PBS1), second polarization beam splitter prism (PBS2), mirror (including optical path switching device). Function: First splitting and combining: Split a single pulse into two time - domain - separated double pulses, reduce the peak power to adapt to the lens damage threshold of the non - linear compression module; Secondary splitting and combining: Compensate the time - domain difference through optical path switching, realize the time - domain coincidence and optical intensity superposition of the double pulses, and generate an ultra - strong pulse.

[0046] Infrared light source: Output an initial laser beam (such as in the infrared band of 1030nm, 1064nm, etc.), which is divided by PBS1 into two linearly polarized light beams (beam A, beam B) with orthogonal polarizations (such as horizontal / vertical polarization), equal light intensities (each accounting for 50%), and initially time - domain coincidence.

[0047] Non - linear pulse compression module: Comprising a multi - pass cavity (such as a Herriott cavity) and a chirped mirror group. After the spectrum is broadened through self - phase modulation (SPM), the pulse width is compressed by using the dispersion compensation of the chirped mirror, and the peak power is increased to the critical value of the lens damage threshold (such as ≤ 90% of the damage threshold).

[0048] First beam splitting and combining: Time-domain separation and peak power reduction; Beam splitting process: PBS1 splits the initial beam into two beams (A and B) with orthogonal polarizations, each with an optical intensity of 50%, and the initial pulses overlap in the time domain (Δt = 0). Delay construction: The mirror constructs a long optical path (optical path L1) for one of the beams (such as beam A), and the other beam (beam B) passes through a short optical path (L2). The optical path difference ΔL = L1 - L2 = c * Δt, causing the pulses of the two beams to be separated in the time domain (interval Δt), and Δt is greater than the pulse width to ensure that the pulses do not overlap. Beam combining result: PBS2 combines the orthogonally polarized light beams to form a new beam containing two time-domain separated pulses (the number of pulses doubles, the average power remains unchanged, and the peak power of a single pulse is reduced to 1 / 2 of the initial value).

[0049] Nonlinear pulse compression: Critical peak power control: The new beam enters the multi-pass cavity and broadens the spectrum through the self-phase modulation effect (such as broadening from 10 nm to 50 nm), increasing the compressible energy; the broadened pulse passes through a chirped mirror group (multiple cascaded, with the arrangement spacing and angle ensuring that the spot diameter is greater than the damage threshold spot size of the lens) to compensate for the phase and compress the pulse width (such as compressing from 500 fs to within 100 fs). At this time, the peak power of a single pulse is increased to the critical value of the lens damage threshold (such as the critical value is 10 GW, and after compression it is 9.5 GW).

[0050] Second beam splitting and combining: Time-domain compensation and optical intensity superposition: Optical path switching: Through a mirror group or a movable mirror (optical path switching device), the beam A that originally traveled the long optical path now travels the short optical path, and the beam B travels the long optical path, and the optical path difference becomes -ΔL, and the time-domain interval compensation is -Δt; Beam combining and superposition: The pulses of the two beams completely overlap in the time domain (Δt = 0). Since the orthogonally polarized light is incoherent after beam combining, but the optical intensities are directly superposed (total optical intensity = A 2 + B 2 , because A = B, the total optical intensity = 2A 2 , and the peak power doubles), and finally an ultra-strong pulse with a peak power exceeding the lens damage threshold is formed at the light output port (such as increasing from 9.5 GW to 19 GW).

[0051] The optical path layout and component parameters can be as follows: Pulse beam splitting and combining module: Polarizing beam splitter prism: The splitting directions of PBS1 and PBS2 are the same (e.g., both are incident at 45°, separating horizontal / vertical polarization), matching the polarization state of the infrared light source output (e.g., the initial beam is horizontally polarized, and PBS1 splits it into horizontally transmitted light and vertically reflected light), with polarization loss ≤ 1%. Reflecting mirror and optical path difference: The long optical path is folded by a set of reflecting mirrors (e.g., 3 reflections, optical path L1 = 3m), the short optical path is directly transmitted (L2 = 1m), ΔL = 2m, corresponding to a time domain interval Δt = ΔL / c ≈ 6.67ns (much larger than the femtosecond pulse width to ensure pulse separation). Optical path switching device: An electrically controlled flipping mirror (e.g., a two-dimensional galvanometer) or a mechanical translation stage is used to switch the long / short optical path of beams A and B during the first and second beam splitting and combining, with a switching time ≤ 1μs (much smaller than the pulse repetition frequency period).

[0052] Multi-pass cavity of the nonlinear pulse compression module: Cavity length is 1m, radius of curvature of the reflecting mirror is 500mm, and the number of beam round trips is 10 times, so that the total optical path of the pulse transmitted in the cavity is 10m, accumulating sufficient self-phase modulation effect (nonlinear phase shift Δφ = 2π). Chirped mirror group: It contains 4 second-order chirped mirrors (dispersion amount -1000 fs² / piece), with a spacing of 20cm, and a spot diameter of 5mm (larger than the spot size of 3mm at the lens damage threshold), ensuring that the peak power density ≤ the damage threshold (e.g., the critical power density is 1GW / cm², and the actual is 0.8GW / cm²).

[0053] Initial pulse beam splitting and delay ( Figure 1 Left half): The infrared light source outputs a single pulse (energy E, pulse width τ, peak power P = E / τ); PBS1 splits it into horizontally polarized light (beam A, energy E / 2) and vertically polarized light (beam B, energy E / 2), with initial time domain coincidence; Beam A is transmitted through the long optical path (L1), and beam B is transmitted through the short optical path (L2), with time domain separation Δt = ΔL / c, forming two pulses separated by Δt (each with energy E / 2 and peak power P / 2) when reaching PBS2.

[0054] Nonlinear compression and peak enhancement ( Figure 1 Middle part): The combined double pulses enter the multi-pass cavity, and the self-phase modulation broadens the spectrum to Δλ = 50nm; The chirped mirror group compensates for the dispersion, compressing the pulse width to τ’ = 50fs, and the peak power of a single pulse is increased to P’ = (E / 2) / τ’ = E / (2τ’) (because the energy remains unchanged, the pulse width shrinks, and the peak power increases), and P’ ≤ the lens damage threshold P_d (e.g., P_d = 10GW, P’ = 9GW).

[0055] Secondary beam splitting, combining and superposition ( Figure 1Right half part): The compressed double pulses enter the beam splitting and combining module again. The optical path switching device makes beam A travel a short optical path and beam B travel a long optical path, with an optical path difference of -ΔL and a time domain interval of -Δt. When the two pulses reach PBS2, they are temporally coincident (Δt = 0), and the light intensities of orthogonally polarized light are directly superposed (total energy E, peak power 2P’ = 18GW > P_d), and an ultra-strong pulse (energy ≥ 1mJ, pulse width ≤ 100fs) is output from the light output port.

[0056] After the first beam splitting by this method, the peak power drops to 1 / 2 of the initial value, ensuring that the peak power of the pulses entering the non-linear compression module is much lower than the damage threshold of the lens (for example, it drops from 20GW to 10GW, while the lens threshold is 15GW), avoiding film layer damage caused by excessive power during self-phase modulation and compression; after compression, the peak power increases to the critical value (such as 14GW), still lower than the threshold, and after the second beam combining, it is superposed to 28GW, breaking through the threshold but only achieving it at the final light output port, and the devices in the intermediate process are always safe. There is no need to rely on high-cost liquid crystal modulators or multi-channel phase control like traditional coherent beam combining technology, and only through polarization beam splitting and combining devices (low cost, high reliability) to achieve pulse energy superposition; the number of chirped mirrors in the non-linear compression module does not need to be increased (because the spot size remains unchanged after beam splitting, only the peak power decreases), avoiding the expansion of the number of devices caused by increasing the spot size, and the system volume is reduced by more than 30%. It realizes the output of ultra-strong pulses with single pulse energy ≥ 1mJ and pulse width ≤ 100fs, and the peak power density meets the precision processing requirements of new materials (such as silicon carbide, glass fiber composite materials) (such as drilling and etching accuracy ≤ 5μm). It is compatible with the existing ultrafast laser system architecture, can directly replace the traditional single-path compression module without re-designing the optical path; the optical path switching device adopts an electric control or mechanical structure, with high control precision (time domain compensation error ≤ 1fs), and the long-term operation stability is better than traditional coherent beam combining technology (phase drift ≤ π / 10).

[0057] In some embodiments, the laser beam output by the infrared light source is equally split into two beams by the first polarization beam splitter prism, including: the first polarization beam splitter prism divides the laser beam into two linearly polarized light beams with orthogonal polarization directions, the light intensity ratios of the two beams are the same, and the initial pulses are temporally coincident.

[0058] The beam splitting principle includes: the initial laser beam output by the infrared light source is linearly polarized light (such as horizontal polarization state). When it is incident on the first polarization beam splitter prism (PBS1), PBS1 divides the beam into two beams according to the polarization direction: transmitted light: maintaining the original polarization direction (horizontal polarization, energy ratio 50%); reflected light: converted to the orthogonal polarization direction (vertical polarization, energy ratio 50%). Temporal synchronization: The two beams of light enter different optical paths immediately after beam splitting, but the initial transmission optical paths are the same (no additional reflectors are introduced), so the pulses of the two beams of light are completely coincident in the time domain (that is, the time difference Δt = 0 when reaching the subsequent devices).

[0059] Ensure that the energies of the two beams of light are strictly equal (50% each) through polarization beam splitting, laying the foundation for the light intensity uniformity during subsequent pulse superposition; the two pulses overlap in the initial time domain, facilitating the precise control of the time domain separation amount through subsequent mirrors and avoiding the introduction of additional timing errors during the beam splitting process; the design of orthogonal polarization ensures that the two beams of light do not undergo coherent superposition (only intensity superposition) during beam combination, avoiding the complexity of phase modulation and simplifying the optical path control.

[0060] In some embodiments, the mirror is used to increase the additional optical path for at least one of the beams of light, resulting in a time domain difference between the two beams of light, including: the mirror constructs a long optical path for one beam of light, and the other beam of light is transmitted through a short optical path. The optical path difference ΔL between the long optical path and the short optical path satisfies: ΔL = c * Δt, where Δt is the pulse time domain interval between the two beams of light, and c is the speed of light, such that the pulses of the two beams of light are separated and do not overlap in the time domain.

[0061] Optical path design: Short optical path: One beam of light (such as vertically polarized light) is directly transmitted through the mirror group, and the total optical path is L2 = d (d is the shortest physical distance); Long optical path: The other beam of light (such as horizontally polarized light) passes through a folded optical path of the mirror (such as through 3 reflections), and the total optical path is L1 = d + ΔL, where ΔL = c * Δt and Δt is a preset time domain interval (it is required that Δt > τ, and τ is the initial pulse width). Time domain separation verification: The time difference between the two beams of light reaching the second polarization beam splitter prism (PBS2) is Δt = cΔL, ensuring that the two pulses are completely separated in the time domain (i.e., the pulse waveforms do not overlap).

[0062] Safe reduction of peak power: After time domain separation, the combined new beam contains two independent pulses. The energy of a single pulse is 1 / 2 of the initial value, and the peak power synchronously drops to 1 / 2 (because the pulse width remains unchanged), ensuring that the peak power entering the nonlinear compression module is lower than the lens damage threshold; Precise timing control: Quantitative design of the time domain interval is achieved through the optical path difference formula ΔL = c * Δt, and Δt can be flexibly adjusted according to the damage threshold of the nonlinear compression module (for example, if the threshold allows a peak power of 10 GW and the initial peak power is 20 GW, then Δt ≥ τ is required to completely separate the pulses).

[0063] In some embodiments, the two beams of light are combined into a new beam by the second polarization beam splitter prism. The number of pulses in the new beam is twice that of the initial light, and the average power is the same as that of the initial light, including: The second polarization beam splitter prism performs polarization beam combination on the two beams of orthogonally polarized light. The pulses of the two beams of light are separated in the time domain. The combined new beam contains two pulses arranged in sequence in the time domain. The corresponding total pulse energy is equal to the initial single pulse energy, the average power remains unchanged, and the peak power of a single pulse is reduced to 1 / 2 of the initial value.

[0064] Beam combining process: When two orthogonally polarized light beams (horizontal and vertical) reach PBS2, PBS2 allows both polarized light beams to pass through simultaneously and combine into one beam (polarization beam combining principle: orthogonally polarized light beams overlap in space but are separated in time domain);

[0065] Pulse characteristics: Number of pulses: Since the pulses of the two light beams are separated in time domain (interval Δt), the new combined light beam contains two independent pulses arranged in sequence (the first-arriving one is the short optical path pulse, and the later-arriving one is the long optical path pulse); Energy and power: Total energy E_total = E / 2 + E / 2 = E (the same as the initial single-pulse energy), average power Pavg = E_total / T (T is the pulse period, the same as the initial one), and the peak power of a single pulse P_peak = E / 2τ = P_initial / 2.

[0066] By separating the pulses in time domain to avoid pulse overlap, the peak power of a single pulse is strictly reduced to 1 / 2 of the initial value, providing a safe input power range for the nonlinear compression module (for example, if the initial peak power is 20 GW, it is reduced to 10 GW after beam combining, and the mirror threshold is 15 GW, ensuring a safety margin); maintaining the same average power as the initial light beam to ensure that sufficient nonlinear effects (such as the energy density required for self-phase modulation) can be accumulated during the nonlinear compression process.

[0067] In some embodiments, the new light beam enters the nonlinear pulse compression module, and after nonlinear pulse compression, the peak power of a single pulse is increased to the critical value of the damage threshold of the mirror, including: the nonlinear pulse compression module includes a multi-pass cavity, and when the new light beam is transmitted in the multi-pass cavity, the spectrum is broadened through the self-phase modulation effect, and the broadened pulse is compressed by a chirped mirror group for phase compensation. After compression, the peak power of a single pulse is increased to be equal to or lower than the damage threshold of the mirror coating on the chirped mirror surface.

[0068] Multi-pass cavity structure: The cavity is composed of high-reflectivity mirrors (such as Herriott cavity), and the new light beam travels back and forth in the cavity N times (such as N = 10 times), and the total optical path L_cavity = Nl (l is the cavity length), extending the interaction time between the pulse and the medium; the self-phase modulation effect causes the spectrum to broaden (such as from 10 nm to 5 nm), increasing the spectral bandwidth that can be compressed.

[0069] Chirped mirror group compensation: The broadened pulse passes through a chirped mirror group (including 4 second-order chirped mirrors, each with a dispersion of -1000 fs²), and the wavelength-dependent delay of the mirror coating compensates for the pulse phase distortion; after compression, the pulse width τ' ≤ 10 fs, and the peak power P' = E / 2τ'. By controlling τ', P' ≤ P_damage threshold (for example, when the threshold is 10 GW, compressed to τ' = 50 fs, then P' = 0.5 mJ / 50 fs = 10 GW).

[0070] On the premise of not exceeding the lens damage threshold, the peak power of a single pulse is maximally enhanced through spectral broadening-compression technology to reserve energy for subsequent beam combination and superposition; the multi-pass cavity provides sufficient nonlinear interaction distance, and at the same time, the low peak power after beam splitting avoids unnecessary nonlinear noise (such as stimulated Raman scattering) and improves the pulse quality.

[0071] Exemplarily, the chirped mirror group of the nonlinear pulse compression module is provided with a plurality of chirped mirrors, and the arrangement spacing and angle of the chirped mirrors satisfy that the spot diameter of the pulsed beam is greater than the damage threshold spot size of the chirped mirror, ensuring that the pulsed beam with the critical peak power value does not damage the chirped mirror.

[0072] Spot diameter control: The damage threshold spot size on the surface of the chirped mirror is D_threshold (such as 3 mm). By adjusting the spacing between the chirped mirrors and the incident angle, the spot diameter D_actual of the pulsed beam on the mirror surface is 5 mm > D_threshold; power density formula: I = P′ / π(D_actual / 2) 2 , ensuring that I ≤ I_threshold (even when P′ approaches the threshold, increasing the spot can reduce the power density).

[0073] Lens arrangement parameters: The spacing between adjacent chirped mirrors is 20 cm, the incident angle is 45° (optimizing the dispersion compensation efficiency), and the mirror surface is coated with a high damage threshold film layer (threshold 10 GW / cm 2 )

[0074] The peak power is reduced below the threshold through beam splitting; increasing the spot further reduces the energy per unit area. Even when the peak power approaches the threshold, there is still a safety margin for the power density; it avoids the film layer damage caused by too small a spot in the traditional scheme and improves the long-term stability of the system (the lens replacement cycle is extended from 3 months to more than 1 year).

[0075] In some embodiments, the pulse that travels a long path during the first beam splitting and combining travels a short path during the second beam splitting and combining, and the pulse that travels a short path during the first time travels a long path during the second beam splitting and combining, so that the time domain difference between the pulses of the two beams of light is compensated, and the pulses of the two beams of light achieve optical intensity superposition, including: during the second beam splitting and combining, through the optical path switching of the first polarization beam splitter prism and the second polarization beam splitter prism, the beam that has passed through the long optical path for the first time is changed to pass through the short optical path, and the beam that has passed through the short optical path for the first time is changed to pass through the long optical path, and the optical path difference between the two beams of light is compensated in the reverse direction, and the time domain interval is offset, so that the two pulses are completely coincident in the time domain and the optical intensity achieves coherent superposition.

[0076] Beam A (horizontally polarized) travels a long optical path, and beam B (vertically polarized) travels a short optical path, with a time-domain interval of +Δt; for the second beam splitting and combining: through an optical path switching device (such as a flipping mirror), beam A is made to travel the short optical path, and beam B is made to travel the long optical path, with a time-domain interval of -Δt; Pulse coincidence condition: when the two beams of light reach the final beam combining point, the time difference is +Δt + (−Δt) = 0, the pulses completely coincide, and the light intensities are superimposed (total light intensity I_total = I_A + I_B = 2I_A, because I_A = I_B).

[0077] Non-coherent light intensity superposition is achieved through time-domain coincidence (orthogonally polarized light does not interfere, and the energies are directly added), enabling the final peak power to reach twice that of a single compressed pulse (e.g., increased from 10 GW to 20 GW), breaking through the threshold limit of single-path compression; the superposition process only changes the peak power and does not introduce additional pulse broadening or phase distortion (because the two pulses are processed by the same compression module and have exactly the same waveform).

[0078] In some embodiments, the pulse beam splitting and combining module further includes an optical path switching device, which is used to swap the long and short optical path of the two beams of light during the first beam splitting and combining and the second beam splitting and combining. The optical path switching device includes a mirror group or a movable mirror, and the optical path of the beam is adjusted mechanically or electrically.

[0079] Device type: Mirror group: Two sets of rotatable mirrors (such as two-dimensional galvanometers) are used, and the long / short optical path of beam A / B is switched by angle rotation; Movable mirror: An electrically controlled translation stage is used to adjust the position of the mirror to change the optical path of the beam (e.g., when the translation stage moves a distance of ΔL, the corresponding optical path difference is 2ΔL, because the beam is reflected back and forth).

[0080] Control logic: Triggered synchronously with the laser. After the first beam splitting and combining is completed (i.e., when the pulse enters the nonlinear compression module), the optical path is switched in advance to ensure that the paths have been swapped during the second beam splitting.

[0081] The mechanical switching time is ≤1 μs, and the electrical control switching time is ≤100 ns, meeting the timing requirements of high-repetition-rate femtosecond lasers (such as 1 kHz); the switching device independent of the core optical path can be adapted to the existing beam splitting and combining module without re-designing the optical architecture.

[0082] In some embodiments, the polarization splitting directions of the first polarization beam splitter prism and the second polarization beam splitter prism are the same and match the output polarization state of the infrared light source, ensuring that the polarization loss during beam splitting and combining is less than the preset loss rate.

[0083] Polarization state matching: The infrared light source outputs a fixed polarization state (such as horizontal polarization). The splitting directions of PBS1 and PBS2 are both "transmission for horizontal polarization and reflection for vertical polarization" to ensure that the polarization state does not rotate during the beam splitting and combining processes; the preset loss rate ≤ 1% (by selecting a PBS prism with a high extinction ratio, extinction ratio ≥ 1000:1).

[0084] Avoid energy loss caused by mismatched polarization directions (such as the loss of a traditional non-polarizing beam splitter can reach 5% - 10%), and improve the overall efficiency of the system (from 80% to over 98%); the strict maintenance of orthogonal polarization ensures the accuracy of light intensity superposition during secondary beam combining (if the polarization state rotates, it will cause partial energy reflection loss).

[0085] In some embodiments, the single-pulse energy of the ultra-strong pulse at the light output port ≥ 1 mJ, the pulse width ≤ 100 fs, and the peak power density is lower than 2 times the damage threshold of the lens material in the non-linear pulse compression module. Through the two-time light intensity superposition of the pulse beam splitting and combining module, the peak power multiplication effect is achieved without damaging the lens.

[0086] Parameter implementation path: Initial pulse: energy 2 mJ, pulse width 200 fs, peak power 10 GW; First beam splitting and combining: The energy of each of the two pulses is 1 mJ, and the peak power is 5 GW; Non-linear compression: The pulse width is compressed to 50 fs, and the single-pulse peak power is 20 GW (not exceeding the lens threshold of 25 GW); Secondary beam combining: The peak power is superposed to 40 GW (≤ 2×25 GW = 50 GW, meeting the 2-fold threshold limit). Safety margin design: By controlling the total peak power density I after secondary superposition ≤ 2I threshold, ensure that devices other than the light output port (such as the output lens) are not damaged.

[0087] The indicators of single-pulse energy ≥ 1 mJ and pulse width ≤ 100 fs meet the precise processing requirements such as semiconductor wafer cutting and photovoltaic material drilling; the peak power density is controlled within 2 times the damage threshold to avoid accidental damage caused by device aging or power fluctuations, and at the same time achieve maximum energy output; through two-time light intensity superposition (incoherent superposition), without introducing the complexity of phase control, simply and efficiently break through the single-path power limit. Compared with traditional coherent beam combining technology (requiring nanoscale phase synchronization), the engineering implementation difficulty is reduced by more than 50%.

[0088] In some embodiments, a dynamic optical path adjustment unit composed of 12 deformable micromirrors is installed between the beam splitting module and the nonlinear compression module. Each group of micromirrors has the ability to adjust the curved surface with nanometer-level precision, and can change the propagation direction and wavefront shape of the light beam in real time. At the same time, high-precision temperature and humidity sensors, barometric pressure sensors and vibration sensors are integrated in the optical path to monitor the influence of environmental parameter changes on the optical path in real time. The beam splitter uses an electronically controlled adjustable polarization beam splitting prism, and the beam splitting ratio is controlled by a voltage signal, and the adjustment range is from 30%:70% to 70%:30%, and the step accuracy reaches 1%.

[0089] The intelligent reconstruction control process includes that after the system is started, first, the spectral width and pulse width data of the initial pulse are collected through the built-in spectrometer and autocorrelator, and a reference database is established. When the environmental sensor detects that the temperature change exceeds 2°C or the barometric pressure change exceeds 5 hPa, the optical path reconstruction program is triggered:

[0090] In the first step, the dynamic micromirror array calculates the change in atmospheric refractive index according to the real-time environmental parameters, automatically adjusts the mirror angle, and compensates for the beam offset caused by the change in air density;

[0091] In the second step, the electronically controlled beam splitter dynamically adjusts the beam splitting ratio according to the real-time temperature of the current nonlinear compression module (obtained through the infrared thermometer embedded in it) - when the temperature of the compression module rises, the proportion of the single-path beam energy is automatically reduced to avoid the damage risk caused by the increase in the device temperature;

[0092] In the third step, the two reconstructed light beams enter independent multi-pass cavity compression paths, and the mirror groups of each path are equipped with high-precision linear motors, which can adjust the optical path length in real time according to the split energy to ensure that the time domain interval between the two light beams always meets the safe compression requirements.

[0093] The closed-loop feedback optimization mechanism includes that after every 100 pulse outputs are completed, the system automatically compares the actual compressed pulse width with the ideal value in the reference database. If the pulse width broadening exceeds 5%, the secondary optimization is started: the wavefront of the light beam is finely adjusted through the deformable micromirror to correct the polarization state deviation caused by the long-term use of the beam splitter, and at the same time, a compensation signal is sent to the electronically controlled beam splitter to gradually calibrate the beam splitting ratio until the pulse width returns to the optimal state.

[0094] Traditional fixed optical path systems are prone to beam deviation and unstable energy when the temperature and humidity fluctuate. In this embodiment, a dynamic micromirror is used to compensate for atmospheric disturbances in real time, reducing the impact of environmental changes on the optical path by more than 80%. Stable output can be maintained even in the complex environment of an industrial plant. Device protection and efficiency balance: By monitoring the temperature of the compression module in real time and dynamically adjusting the beam splitting ratio, lens damage caused by local overheating is avoided. At the same time, in a low-temperature environment, the single-path energy ratio is automatically increased, and the energy utilization efficiency is increased from 75% of fixed beam splitting to 92%, significantly reducing energy waste. Enhancement of self-calibration ability: The closed-loop feedback mechanism can continuously optimize the beam splitting and compression parameters without manual intervention. The pulse width stability is increased by 60% during long-term operation, solving the problem of performance drift caused by device aging in traditional equipment. The equipment maintenance cycle is extended from once a week calibration to once a month.

[0095] In some embodiments, a spectrometer, an autocorrelator, a laser power meter, and a wavefront sensor are installed at the input end of the non-linear compression module to collect the spectral distribution, pulse width, peak power, and wavefront aberration data of the pulse in real time. A stress sensor and a displacement sensor are integrated into the mechanical structure of the chirped mirror group to monitor the subtle changes in the lens installation position (with an accuracy of up to 1 micron). All sensor data is transmitted to the central control unit through a high-speed bus to build a real-time database containing 12-dimensional parameters.

[0096] Compression optimization driven by a fusion algorithm includes the central control unit running a fusion algorithm based on fuzzy logic to convert sensor data into control instructions: When the spectrometer detects that the spectral width broadens beyond the design value, the algorithm automatically identifies it as a deviation in the chirped mirror spacing and drives a high-precision stepper motor to finely adjust the lens spacing until the spectral width returns to the ideal range; If the autocorrelator measures a pulse width greater than the target value and the wavefront sensor detects wavefront aberration, it is determined that the lens surface is contaminated, and the optical path switching program is immediately triggered to enable the standby chirped mirror group, while sending a cleaning reminder to the maintenance system; For the problem of peak power fluctuation, the algorithm combines the power meter data and the real-time state of the beam splitter to dynamically adjust the beam splitting ratio to ensure that the single-path energy entering the compression module is always within 80% of the device safety threshold, avoiding damage while fully utilizing the energy limit.

[0097] The adaptive learning function includes a built-in historical data storage module in the system that automatically records the optimal compression parameters under different working conditions (such as the best position of the chirped mirror and the optimal ratio of the beam splitter corresponding to different environmental temperatures and humidities). When the device enters a new working environment, the closest historical parameters are called as the initial values through pattern matching, and then fine-tuned in combination with real-time sensing data, shortening the initial parameter calibration time from the traditional 10 minutes to within 1 minute.

[0098] Traditional single-sensor control is difficult to cope with multi-factor interference. In this embodiment, through multi-modal data fusion, the pulse-width compression accuracy is improved from ±10 fs to ±3 fs, and the peak power stability is optimized from ±15% to ±3%, meeting the requirements for ultra-stable pulses in high-precision fields such as semiconductor processing. The stress sensor and displacement sensor monitor the mechanical state of the lens in real time. Combining with the wavefront aberration data, potential faults such as lens loosening or contamination can be predicted 72 hours in advance, reducing the risk of sudden shutdown by more than 90% and significantly improving the reliability of the equipment. Quickly adapt to complex working conditions: The adaptive learning function enables the equipment to have "memory ability". When frequently switching working scenarios (such as from the low-temperature environment in the laboratory to the high-temperature environment in industry), it can quickly enter the best working state without re-adjustment, greatly improving production efficiency.

[0099] In some embodiments, by deploying a dedicated edge computing unit near the beam splitting and combining module, with a high-speed data processing chip and a real-time operating system built in, it independently processes sensor data and generates control instructions, and the response time is less than 1 microsecond. The edge computing node is connected to the central main control system through optical fiber to achieve millisecond-level data interaction, and at the same time has the ability to operate offline to ensure autonomous regulation even when the network is interrupted.

[0100] The dynamic energy distribution strategy includes the edge computing unit real-time analyzing the working mode of the laser (such as precision machining mode, material modification mode), combining the material (identified by an external industrial camera) and thickness data of the current processing object, and dynamically adjusting the beam splitting energy ratio: when processing brittle materials (such as glass), automatically reduce the single-path energy and increase the repetition frequency, and the beam splitting ratio is adjusted to 40%:60% to reduce the crack risk caused by single-pulse impact; when processing metal materials, increase the single-path energy to 60% occupancy, use higher peak power to improve the material removal efficiency, and at the same time, the edge computing monitors the temperature of the compression module in real time. Once it approaches the warning threshold, the air-cooling system is started and the energy distribution is fine-tuned; in the pulse combining stage, the edge computing unit dynamically controls the tilt angle of the combining mirror according to the real-time time-domain interval of the two pulses to ensure the precise superposition of the two pulses in space-time and avoid energy loss caused by timing deviation.

[0101] The hardware co-optimization design includes that the beam splitter adopts a magnetorheological fluid dynamic beam splitting device, and the beam splitting ratio can be adjusted within 2 microseconds by changing the magnetic field strength, and the response speed is 5 times faster than that of the traditional electric control beam splitter. The mirror group of the compression module is integrated with a piezoelectric ceramic driver, which can perform sub-nanometer displacement adjustment according to the edge computing instruction to compensate the optical path difference caused by the change of energy distribution in real time.

[0102] Traditional fixed-parameter systems struggle to accommodate the processing requirements of different materials. In this embodiment, edge computing is used to adjust the energy distribution in real time, enabling the device to be compatible with the processing of more than 10 materials, such as glass, metal, and semiconductors. The processing efficiency is increased by 40%, and the scrap rate is reduced by 65%. The local processing ability of edge computing realizes a microsecond-level control closed-loop, far exceeding the millisecond-level response of traditional central control. It can accurately capture and respond to instantaneous fluctuations in the pulse sequence, ensuring that the energy output of each pulse is in an optimal state. Dynamically adjusting the energy distribution according to processing requirements avoids redundant energy consumption in high-energy modes, reducing energy consumption by 30% in precision processing scenarios. At the same time, through precise beam combination control, the energy superposition efficiency is increased from the traditional 85% to over 98%.

[0103] In some embodiments, two independent optical paths, a primary path and a backup path, are set between the beam splitting module and the compression module. Each optical path is equipped with a complete set of chirped mirrors, beam splitters, and mirror groups. Key devices (such as chirped mirrors and beam splitting prisms) on the primary optical path are all connected in parallel with backup devices, and the primary-backup switch is realized through an electro-optic shutter, with a switching time of less than 5 microseconds. The backup optical path is in a preheated state during normal operation, and the lens temperature is maintained within ±1°C of the operating temperature to ensure it can be put into use at any time.

[0104] The system detects optical path anomalies through the following three mechanisms: Real-time performance monitoring: When the pulse width measured by the autocorrelator exceeds the standard value by 20% for 10 consecutive pulses, or the power meter detects an energy attenuation exceeding 15%, a preliminary fault warning is triggered; Device status diagnosis: Through an infrared thermal imager and strain gauges embedded on the lens holder, the temperature distribution and mechanical stress of the chirped mirror are detected. If local temperature anomalies or lens displacement exceeding 5 micrometers are found, it is determined that the lens is damaged or loose; Optical path contamination detection: A scattered light sensor is installed at the output end of the compression module. When the detected scattered light intensity increases by more than 30%, it is judged that the lens surface is contaminated.

[0105] Once a fault is confirmed, the system automatically executes the following repair steps: Close the electro-optic shutter of the primary optical path and open the shutter of the backup optical path at the same time. During the switching process, the energy output is maintained without interruption through an energy storage capacitor; Send fault location information (such as "the third chirped mirror of the primary optical path is contaminated") to the maintenance system and start the parameter self-calibration program of the backup optical path, automatically configuring the optimal compression parameters according to historical data; The primary optical path enters the maintenance mode and automatically starts the air blowing cleaning device (for contamination problems) or the mechanical calibration device (for displacement problems). If self-repair fails, it is marked for manual maintenance. During long-term operation, the system automatically switches the primary and backup roles regularly (such as rotating every 8 hours) according to the usage duration and device loss degree of the primary and backup optical paths to ensure uniform wear of each device and extend the overall service life.

[0106] The downtime of traditional equipment caused by optical path failures averages over 30 minutes. In this embodiment, through seamless switching between the primary and backup optical paths, the fault response time is compressed to the microsecond level, and there is no perception during the actual processing. The equipment utilization rate is increased from 70% to over 99%. The built-in cleaning and calibration devices can handle over 80% of common faults (such as minor contamination and slight displacement of lenses), reduce the frequency of manual intervention, and lower the operation and maintenance costs by over 50%. The dynamic load balancing strategy evenly distributes the usage duration of each device, and the replacement cycle of the core lens is extended from the traditional 6 months to 12 months, significantly reducing the consumable costs.

[0107] In some embodiments, all controllable units of the laser (beam splitter, compression module, beam combiner group, cooling system) are connected to the industrial Internet of Things platform, and the device status is uploaded in real time and remotely controlled through a 5G communication module. A device digital twin model is constructed in the cloud to mirror the operating status of the physical device in real time, including over 200 monitoring points such as optical path parameters, device temperature, and energy output curve.

[0108] The production plan linkage includes automatically adjusting the laser working parameters 1 hour in advance according to the production schedule of the factory's MES system. For example, when receiving a high-precision processing task, the system automatically switches to the backup high-stability optical path and pre-heats the key devices to the optimal temperature; when receiving a large-scale material processing task, it enables the high-energy output mode and optimizes the beam splitting ratio to improve the processing speed. By analyzing historical energy consumption data through cloud algorithms, it automatically enters the "sleep-wake" mode during non-production periods (such as early morning): turns off the power of non-core devices, and keeps the key sensors and control systems on standby. The energy consumption is reduced by 70%, and at the same time, it is ensured that it only takes 30 seconds to resume full power output from the sleep state. When multiple lasers work together (such as in an array processing scenario), the Internet of Things platform automatically allocates the energy output strategy for each device, avoids grid fluctuation interference during synchronous operation, and ensures the processing accuracy of each device is consistent by dynamically adjusting the beam splitting ratio. The cloud digital twin model analyzes the device aging trend through machine learning algorithms: when it is found that the reflectivity attenuation rate of the chirped mirror accelerates, it predicts the replacement requirement 3 days in advance and automatically generates a maintenance work order containing the spare part model and replacement steps; an anomaly detection model trained based on historical fault data can identify over 90% of potential fault precursors (such as abnormal vibration signals of the beam splitter drive motor), and push them to the maintenance personnel's mobile phones via text messages to achieve preventive maintenance.

[0109] Through the Industrial Internet of Things, deep collaboration between devices and production plans is achieved. The processing task switching time is shortened from 10 minutes of manual operation to 30 seconds of automatic adjustment, and the overall production line efficiency is increased by 35%. The digital twin model optimizes the device operation parameters in real time. Combining with cloud big data analysis, the mean time between failures (MTBF) of the laser is extended from 800 hours to more than 2000 hours, and the scientific nature of maintenance decisions is increased by 90%. The intelligent energy consumption management system significantly reduces power consumption during the night period. The annual power consumption of a single device is reduced by 250,000 kWh, meeting the low-carbon production requirements of Industry 4.0 and reducing the enterprise's electricity cost at the same time.

[0110] In some embodiments, on the connecting bracket between the beam splitter and the compression mirror group, the traditional rigid connecting piece is replaced with a flexible hinge structure driven by shape memory alloy (SMA). Each hinge integrates 3 groups of helical SMA wires, which can generate a linear expansion and contraction of 0.1 - 5 mm through current heating, with an accuracy of 1 micron, for real-time adjustment of the lens spacing and angle. Temperature sensors are embedded in the bases of key optical path nodes (such as chirped mirror groups, beam combiners) to monitor the influence of device temperature rise on the mechanical structure in real time.

[0111] The thermal-mechanical coupling calibration process includes that when the system starts, the initial position parameters of each lens are established first through reference pulses. After continuous operation for 30 minutes, if the temperature sensor detects that the temperature rise of the lens base exceeds 10 °C, the self-calibration program is triggered: First step, the SMA drive unit calculates the thermal expansion deformation of the material according to the temperature rise data, and automatically contracts or elongates the hinge structure to compensate for the lens displacement caused by temperature rise - for example, when the chirped mirror spacing increases by 2 microns due to thermal expansion, the SMA wire is heated and contracted to pull the lens back to the standard spacing; Second step, combined with the real-time data of the wavefront sensor, the angle of the beam combiner is finely adjusted: by alternately heating the SMA wires on the left and right sides, the lens is tilted at a sub-milliradian level until the wavefront error of the two beams of light is less than λ / 10; Third step, during the calibration process, the compressed pulse width is monitored synchronously. If the pulse width fluctuation exceeds 5%, the secondary compensation algorithm is started, and the high-order aberration correction of the optical path is performed through the non-linear deformation characteristics of the SMA hinge.

[0112] The low-power sleep-wake mechanism includes that in the standby state of the device, the SMA drive unit maintains a basic heating current of 10%, so that the material is kept near the phase change critical temperature. When a wake-up signal is received, the full-power drive state can be reached within 100 milliseconds, which is 90% faster than the minute-level response speed of traditional motor drives, and the standby power consumption is reduced by 60% at the same time.

[0113] The embodiments have the following beneficial effects:

[0114] Breakthrough in Thermal Drift Dynamic Compensation: The mechanical deformation of traditional metal brackets caused by temperature changes is the main inducement for unstable optical paths. In this embodiment, through the active deformation compensation of SMA materials, the influence of temperature changes (0 - 40°C) on the lens position is reduced from ±50 microns to ±2 microns, and the pulse width stability is improved by 70%.

[0115] Gapless Precision Adjustment: The flexible hinge driven by SMA eliminates the problems of gear clearance and screw backlash in traditional mechanical structures. The angular adjustment resolution reaches 0.001 degrees, meeting the stringent requirements for wavefront accuracy in femtosecond laser processing, and is especially suitable for precision machining scenarios of micro-nano structures.

[0116] Balancing Fast Response and Energy Saving: The sleep-wake mechanism significantly reduces energy consumption while ensuring fast startup, and is applicable to processing scenarios that require frequent startup and shutdown (such as the slicing operation of semiconductor wafers). The energy consumption per single shift is reduced by 35% compared to traditional equipment.

[0117] In some embodiments, inside the lens base of the non-linear compression module, a loop heat pipe with a diameter of 2 mm is embedded. The inner wall of the heat pipe is sintered with copper powder to improve the heat conduction efficiency. The evaporation section is directly attached to the back of the lens, and the condensation section is connected to a phase change material (PCM) heat dissipation box. The heat dissipation box is filled with a paraffin-based phase change material with a melting point of 55°C, and is wrapped with a graphene thermal conductive film on the outside, and air convection heat dissipation is carried out through an axial flow fan. Thermocouples are installed at the inlet and outlet of the heat pipe and in the core area of the PCM to monitor the temperature gradient of the heat dissipation system in real time.

[0118] The dynamic heat dissipation control strategy includes when the temperature of the compression module exceeds 45°C (close to the lens damage threshold of 50°C), starting a three-stage heat dissipation mechanism: The first stage is passive heat dissipation: The phase change material absorbs the heat of the lens and melts, and uses the latent heat characteristic to reduce the temperature rise rate from 15°C / minute to 3°C / minute within 10 minutes, buying adjustment time for the system; The second stage is active heat conduction: According to the temperature difference between the inlet and outlet of the heat pipe (the threshold is set at 8°C), start a micro water pump to enhance the working fluid circulation, and at the same time increase the fan speed to 1500 revolutions per minute to accelerate the condensation heat release of the PCM, and control the temperature within 48°C; The third stage is energy feedback: If the temperature continues to rise to 49°C, the system synchronously adjusts the beam splitting ratio, reduces the energy entering the compression module by 10%, and increases the working fluid flow rate through the heat pipe flow regulating valve to form a two-way control closed loop of "reducing heat load - improving heat dissipation efficiency".

[0119] Adaptive power consumption adjustment includes that the system dynamically adjusts the power consumption of the fan and the water pump according to the real-time heat dissipation demand: In the low-temperature standby state, only maintain the lowest fan speed (500 revolutions per minute) and the intermittent operation of the water pump; During full-load processing, optimize the matching of the fan speed and the water pump flow through a fuzzy algorithm, so that while the heat dissipation efficiency is improved by 40%, the overall power consumption is reduced by 25% compared to the traditional water cooling system.

[0120] The risk of thermal runaway is reduced to zero: The latent heat buffering effect of the phase change material, combined with the high-efficiency heat conduction of the heat pipe, controls the temperature fluctuation of the compression module within ±1°C, completely avoiding damage to the lens film layer caused by insufficient heat dissipation. The continuous working time of the device is extended from 2 hours to more than 8 hours. Multi-scenario heat dissipation adaptation: In a high-temperature environment (such as a 35°C workshop in summer), through the coordination of active heat dissipation and energy regulation, the lens temperature can still be kept stable, while the traditional air-cooled system is difficult to maintain normal operation in an environment above 30°C, expanding the applicable environment range of the device. Lightweight heat dissipation system: Compared with the traditional water-cooled device, the phase change-heat pipe system in this embodiment reduces the volume by 60% and the weight by 50%, and does not require an external cooling water circulation, which is particularly suitable for integration into a compact processing device or a mobile processing platform.

[0121] In some embodiments, the programmable beam shaping hardware includes a programmable shaping unit based on a spatial light modulator (SLM) connected to the rear stage of the beam splitting module. It has a liquid crystal phase modulation array of 1920×1080 pixels, which can generate 12 basic spot patterns such as Gaussian, flat-top, and annular in real time, and supports customizing the spot shape (such as micron-scale slits, array light points). Beam quality analyzers are installed before and after the shaping unit to detect the energy distribution uniformity and edge sharpness of the spot in real time.

[0122] The processing scenario adaptive strategy includes the system automatically switching the spot pattern and adjusting the energy density according to the processing object: Precision drilling (diameter <50 microns): Generate a Gaussian spot with concentrated central energy. Fine-tune the phase distribution through the SLM to compress the spot diameter to less than 15 microns. At the same time, the beam splitter distributes 70% of the energy to a single path to increase the peak power to break through the material threshold; Surface modification (large-area treatment): Switch to a flat-top spot with uniform energy. Disperse the energy to two paths through dynamic beam splitting, and form a uniform spot of 200 microns × 200 microns after beam combination. Combined with galvanometer scanning, the processing speed is increased by 3 times, and the surface uniformity error is less than 5%; Brittle material cutting: Adopt an annular spot structure. The low-energy central area reduces the material impact, and the high-energy ring at the edge realizes precise cutting. Combined with the synchronous adjustment of the pulse frequency of the beam splitter (from 1 kHz to 10 kHz), the edge chipping rate of glass cutting is reduced from 12% to less than 1%.

[0123] The closed-loop feedback optimization algorithm includes that after every 10 processing pulse outputs, the beam quality analyzer compares the actual spot data with the preset model. If the uniformity deviation exceeds 8% or the edge roughness is greater than 1 micron, the SLM automatically enters the iterative optimization mode: Adjust the phase modulation parameters through the genetic algorithm, and the target spot shape can be restored after at most 5 iterations, without manual intervention throughout the process.

[0124] Revolutionary improvement in processing versatility: Traditional lasers rely on replacing hardware to switch the spot mode. In this embodiment, the spot is defined by software, enabling the device to freely switch between 8 processes such as drilling, cutting, and surface treatment without downtime, and the compatibility of the processing scenario is increased by more than 5 times. Double excellence in material removal efficiency and quality: Dynamically adjust the energy density and spot shape according to different material characteristics. In the cutting of stainless steel thin plates, while the speed is increased by 50%, the notch roughness is optimized from Ra3.2μm to Ra1.6μm, meeting the precision machining grade standard. Intelligent anti-damage mechanism: The combination of the annular spot and dynamic energy distribution solves the stress concentration problem in the processing of brittle materials, especially suitable for hard and brittle semiconductor materials such as sapphire and silicon carbide, and the finished product yield is increased from the industry average of 85% to 97%.

[0125] In some embodiments, the construction of the acoustic signal monitoring network includes pasting surface acoustic wave (SAW) sensors on the surfaces of the key moving parts of the laser (such as the beam splitter drive motor and the lens translation stage), with the monitoring frequency range covering 20kHz - 1MHz, which can capture subtle abnormalities in mechanical vibrations (such as high-frequency noise generated by bearing wear). Install air acoustic sensors in the optical path sealed cavity to detect acoustic signals generated by sudden failures such as gas discharge and lens rupture, and the response time is less than 1 microsecond.

[0126] Multi-band fault feature recognition includes: The central processor runs an acoustic signal analysis algorithm based on wavelet packet decomposition to extract features from signals in different frequency bands: Low frequency band (20 - 100kHz): Identify mechanical faults such as motor rotor eccentricity and poor gear meshing. When the vibration amplitude exceeds 15% of the healthy threshold, trigger a first-level warning and record the fault development trend; High frequency band (500kHz - 1MHz): Monitor the acoustic signal of the microcrack propagation generated at the initial stage of lens film layer damage. By comparing the acoustic signal spectrum in the healthy state, abnormalities can be detected when the crack length is only 50 microns, providing a warning 24 hours earlier than traditional visual detection; Air acoustic signal: Use short-time Fourier transform to analyze sudden noises. If a high-frequency spike signal (typical lens rupture feature) lasting more than 5 microseconds is detected, immediately trigger an emergency cut-off of the optical path, and the cut-off time is less than 10 microseconds to avoid secondary damage.

[0127] Self-calibrating acoustic fingerprint library management includes collecting the healthy acoustic fingerprint data of each component when the system runs for the first time to establish a reference acoustic fingerprint library containing more than 500 samples. Automatically update the acoustic fingerprint library every 100 hours of operation, and eliminate the interference of environmental noises (such as vibrations of other equipment in the factory building) through the dynamic time warping (DTW) algorithm to ensure the accuracy of long-term monitoring.

[0128] Traditional vibration sensors are difficult to capture the acoustic signals of micron-scale cracks. In this embodiment, the high-frequency response characteristics of SAW sensors are utilized to improve the fault detection sensitivity to the sub-micron level, enabling the prediction of latent faults such as lens damage and early bearing wear, and avoiding more than 80% of unplanned downtimes. There is no need to install contact sensors on precision devices such as lenses, avoiding the introduction of additional mechanical loads or optical path interference, and is particularly suitable for equipment monitoring in harsh environments such as high vacuum and strong electromagnetic fields, broadening the application scenarios. For the microsecond-level response to sudden rupture faults, combined with the optical path rapid cut-off mechanism, the device damage range is controlled within a single lens. Compared with the traditional downtime process, the fault impact is reduced by 90%, and the maintenance cost is significantly reduced.

[0129] As described above, it is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A high-energy narrow-pulse-width femtosecond pulse laser, characterized in that, Including: A pulse beam splitting and combining module, including a first polarization beam splitter prism, a second polarization beam splitter prism, and a mirror; An infrared light source, the laser beam output by the infrared light source is equally split into two beams of light by the first polarization beam splitter prism, the mirror is used to increase the extra optical path for at least one of the beams of light, so that a time domain difference is generated between the two beams of light, and the two beams of light are combined into a new beam of light by the second polarization beam splitter prism. The number of pulses in the new beam of light is twice that of the initial light, and the average power is the same as that of the initial light; A nonlinear pulse compression module, the new beam of light enters the nonlinear pulse compression module, and after nonlinear pulse compression, the peak power of a single pulse is increased to the critical value of the damage threshold of the lens; the compressed beam of light passes through the pulse beam splitting and combining module again. Among them, the pulse that travels the long path during the first beam splitting and combining travels the short path during the second beam splitting and combining, and the pulse that travels the short path during the first time travels the long path during the second beam splitting and combining, so that the time domain difference between the pulses of the two beams of light is compensated, and the pulses of the two beams of light achieve light intensity superposition, so as to obtain an ultra-strong pulse with a peak power exceeding the damage threshold of the lens at the light output port.

2. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 1, characterized in that, The laser beam output by the infrared light source is equally split into two beams of light by the first polarization beam splitter prism, including: The first polarization beam splitter prism divides the laser beam into two linearly polarized lights with orthogonal polarization directions. The light intensity ratios of the two beams of light are the same, and the initial pulses are temporally coincident.

3. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 1, wherein The mirror is used to increase the extra optical path for at least one of the beams of light, so that a time domain difference is generated between the two beams of light, including: The mirror constructs a long optical path for one of the beams of light, and the other beam of light is transmitted through a short optical path. The optical path difference ΔL between the long optical path and the short optical path satisfies: ΔL = c * Δt, where Δt is the pulse time domain interval between the two beams of light, and c is the speed of light, so that the pulses of the two beams of light are separated and non-coincident in the time domain.

4. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 1, wherein The two beams of light are combined into a new beam of light by the second polarization beam splitter prism. The number of pulses in the new beam of light is twice that of the initial light, and the average power is the same as that of the initial light, including: The second polarization beam splitter prism performs polarization combination on the two orthogonally polarized lights. The pulses of the two beams of light are separated in the time domain. The combined new beam of light contains two pulses arranged in sequence in the time domain. The corresponding total pulse energy is equal to the initial single pulse energy, the average power remains unchanged, and the peak power of a single pulse is reduced to 1 / 2 of the initial value.

5. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 1, characterized in that, The new beam of light enters the nonlinear pulse compression module, and after nonlinear pulse compression, the peak power of a single pulse is increased to the critical value of the damage threshold of the lens, including: The nonlinear pulse compression module includes a multi-pass cavity. When the new beam of light is transmitted in the multi-pass cavity, the spectrum is broadened through the self-phase modulation effect. The broadened pulse is phase-compensated and compressed by a chirped mirror group. After compression, the peak power of a single pulse is increased to be equal to or lower than the damage threshold of the surface film layer of the chirped mirror.

6. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 5, characterized in that, The chirped mirror group of the nonlinear pulse compression module is provided with a plurality of chirped mirrors. The arrangement spacing and angle of the chirped mirrors satisfy that the spot diameter of the pulsed beam is larger than the damage threshold spot size of the chirped mirror, ensuring that the pulsed beam with the critical peak power value does not damage the chirped mirror.

7. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 1, characterized in that, When the pulses that travel a long path during the first beam splitting and combining travel a short path during the second beam splitting and combining, and the pulses that travel a short path during the first time travel a long path during the second beam splitting and combining, the time-domain difference between the pulses of the two beams of light is compensated, and the pulses of the two beams of light achieve optical intensity superposition, including: During the second beam splitting and combining, through the optical path switching of the first polarization beam splitter prism and the second polarization beam splitter prism, the beam that passed through the long optical path for the first time is changed to pass through the short optical path, and the beam that passed through the short optical path for the first time is changed to pass through the long optical path. The optical path difference between the two beams of light is compensated in the reverse direction, and the time-domain interval is cancelled, so that the two pulses completely coincide in the time domain, and the optical intensity achieves coherent superposition.

8. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 1, characterized in that, The pulse beam splitting and combining module further includes an optical path switching device, which is used to swap the long optical path and the short optical path of the two beams of light during the first beam splitting and combining and the second beam splitting and combining. The optical path switching device includes a mirror group or a movable mirror, and adjusts the beam transmission path mechanically or electronically.

9. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 1, wherein The polarization beam splitting directions of the first polarization beam splitter prism and the second polarization beam splitter prism are the same and match the output polarization state of the infrared light source, ensuring that the polarization loss during the beam splitting and combining process is less than the preset loss rate.

10. The high-energy narrow-pulse-width femtosecond pulse laser according to claim 1, characterized in that, The single pulse energy of the ultra-strong pulse at the light output port is ≥1 mJ, the pulse width is ≤100 fs, and the peak power density is lower than 2 times the damage threshold of the lens material in the nonlinear pulse compression module. Through the two-time optical intensity superposition of the pulse beam splitting and combining module, the peak power multiplication effect is achieved without damaging the lens.

Images