Extreme ultraviolet spectrograph calibration method based on extreme ultraviolet diffraction

Abstract

The invention discloses an extreme ultraviolet spectrograph calibration method based on extreme ultraviolet diffraction, and relates to the technical field of spectrograph calibration, and the technical scheme is characterized in that a desktop type XUV light source generation device is used for generating XUV light beams; a diffraction device is arranged on the propagation path of the XUV light beam, so that the XUV light beam is diffracted; taking the FID spectral line of the helium atom 2p-1s as an energy reference point, and deducing part of unknown parameters of the diffraction device and the spectrometer according to the separation distance of + / -1 level diffraction beams measured at the position corresponding to the energy; in combination with a Gaussian fitting method, the positions of 0-level light beams and + / -1-level diffraction light beams at other positions on the detector are obtained, and according to a diffraction formula, energy corresponding to all positions on the XUV spectrometer is inverted; calibration precision is ensured by correcting copper mesh placement angle deviation and CCD image angle deviation. According to the invention, XUV light is diffracted by using the copper net, and global continuous calibration of the energy axis of the spectrometer is realized by combining with a characteristic spectral line with a known wavelength.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of spectrometer calibration, more particularly, it relates to an extreme ultraviolet spectrometer calibration method based on extreme ultraviolet light diffraction. BACKGROUND

[0002] Since the birth of attosecond pulse, it has become a revolutionary tool for studying the process of light and matter interaction. Because of the super wide spectrum and ultra short duration of attosecond pulse, it can realize the function of detecting ultrafast micro dynamic process which femtosecond laser cannot do. Experimentally, the commonly used means to realize this function is attosecond transient absorption technology, the core of which is infrared pump-attosecond detection technology based on Mach-Zehnder interferometer. By detecting the change of XUV (extreme ultraviolet) transient absorption spectrum of the material under the action of infrared pulse and attosecond pulse at different delays, the photoionization, energy level transition and coherent state evolution process can be analyzed. Due to the characteristics of attosecond pulse in time domain and frequency domain, the development of attosecond transient absorption experiment puts forward two challenges to researchers:

[0003] The first is the ultra-short time resolution. Because the response rate of current experimental instrument electronic components cannot reach attosecond level, and the intensity of attosecond pulse is relatively low, the pump-probe technology, that is, the attosecond transient absorption technology, has become a widely used means to solve this problem. In this technology, the key to improve the time resolution is to realize shorter attosecond pulse.

[0004] The second is the ultra-high energy resolution. The XUV band attosecond pulse covers a wider energy range than the femtosecond pulse, which means that it can detect more and more complex physical processes, which requires the XUV spectrometer to accurately distinguish different energy photons, cover a wide detection range, and analyze the millielectron volt (meV) absorption linewidth.

[0005] In the experiment of attosecond transient absorption based on desktop XUV light source, the XUV spectrometer is usually composed of flat-field grating (FFG), micro-channel plate (MCP), phosphor screen (as an imaging accessory of MCP, which does not affect the imaging space relationship and can be ignored) and CCD camera in sequence and placed in a vacuum environment. Although the FFG has been accurately calibrated and set with parameters when it leaves the factory, there will be inevitable errors when the self-made XUV spectrometer is installed. In addition, a slight change in object distance or incident angle will bring great difference in imaging effect and imaging position. In addition, the design and production of flat-field grating expects that the photons in the working wavelength range are focused on the detection surface in order of wavelength size, but in fact it is difficult to achieve uniform distribution in the same plane. This is because the FFG with working wavelength in the extreme ultraviolet band often needs to be incident at a grazing angle. This is to avoid strong absorption of the material to the XUV photons to ensure sufficient diffraction intensity. In this case, the Rowland circle condition and the paraxial approximation fail. At this time, the generalized grating equation corresponds to the approximate relationship,

[0006]

[0007] Wherein, r, r', a, b, R represent the object distance, the image distance, the incident angle, the exit angle and the radius of curvature of the grating respectively. Combined with the grating equation d(sin a + sin b) = m l, we can draw a rough conclusion: for the grazing incidence grating, the focal plane is usually approximated as a parabola or a high-order curve. That is, the distribution of photons on the detection surface according to the wavelength is not uniform, and the cumulative error of the experimental equipment installation in the experiment will exacerbate this phenomenon. Based on the above reasons, the XUV spectrometer must be accurately calibrated before use.

[0008] The existing calibration methods mainly include the following:

[0009] Standard light source calibration method: using standard light sources such as mercury lamp, argon lamp and laser plasma to generate emission lines for calibration. This method depends on the stability of the standard light source, and there are limited types of XUV band standard light sources, and the spectral lines may overlap, resulting in calibration error;

[0010] Metal absorption edge calibration method: using the absorption edge of metal foil (such as aluminum foil, titanium foil) as a reference point. However, the metal absorption edge is generally wide, resulting in low energy resolution, making it difficult to achieve high-precision calibration, and if a wide energy range is to be covered, multiple metals need to be combined, which is cumbersome and costly;

[0011] Plasma emission line calibration method: generating plasma emission lines by high-power laser bombarding solid targets for calibration. This method requires a high-power laser, and the experimental device is complex and expensive, and the plasma spectrum is easily disturbed by the environment, resulting in calibration error;

[0012] Electron beam ion trap (EBIT) emission line calibration method: using the characteristic line of high charge state ions generated by the EBIT device for calibration. This method has high precision, but the equipment is large and expensive, only a few laboratories have the conditions, and it is difficult to popularize;

[0013] Instrument comparison method: compare the to-be-calibrated spectrometer with a high-precision benchtop spectrometer or a synchrotron radiation source. This method relies on external high-precision instruments, and the calibration cost is high and the process is complicated, and the error of this method is the superposition of the error of the to-be-calibrated instrument and the instrument used for calibration, which will further increase the calibration error;

[0014] In-situ calibration method based on atomic lines: using the known absorption / emission characteristics of gas targets (such as excited states of helium and Fano resonance of argon) for calibration; this method simplifies the process and has high resolution, but the calibration accuracy is limited by the line width, and continuous energy calibration cannot be achieved.

[0015] The common shortcomings of existing calibration methods can be summarized as follows:

[0016] (1) Dependence on external reference source: Most methods require standard light sources or samples, which makes it difficult to maintain stable experimental light paths and introduces large errors; (2) Limited energy coverage: There are few standard spectral lines in the XUV band, making it difficult to achieve full-range calibration and continuous energy calibration; (3) Complexity and cost: High-energy lasers, EBIT, and other equipment are expensive and difficult to operate and maintain.

[0017] Therefore, the present application aims to provide an extreme ultraviolet spectrometer calibration method based on extreme ultraviolet diffraction to solve the above problems. SUMMARY

[0018] The purpose of the present application is to provide an extreme ultraviolet spectrometer calibration method based on extreme ultraviolet diffraction, which uses a copper mesh to diffract XUV light and combines known wavelength characteristic spectral lines to achieve full-range continuous calibration of the energy axis of the spectrometer.

[0019] The above technical purpose of the present application is achieved by the following technical solution: an extreme ultraviolet spectrometer calibration method based on extreme ultraviolet diffraction, comprising the following steps:

[0020] S1, using a desktop XUV light source generation device to generate an XUV light beam;

[0021] S2, setting a diffraction device on the propagation path of the XUV light beam, so that the XUV light beam is diffracted to produce a 0th order main beam and a diffraction beam;

[0022] S3, using the FID spectrum line of helium atom 2p→1s as a known energy reference point, and deriving unknown parameters of the diffraction device and the spectrometer from the separation distance of the +1 order diffraction beam measured at the energy corresponding position;

[0023] S4, combining the Gaussian fitting method to obtain the positions of the 0 order beam and the +1 order diffraction beam on the detector at other positions, and inversely deriving the corresponding relationship between the XUV spectrometer horizontal axis and the energy according to the diffraction formula;

[0024] S5, correcting the deviation of the copper mesh placement angle and the CCD image angle to ensure the calibration accuracy.

[0025] The application is further provided that: the specific process of generating the XUV light beam by using the desktop XUV light source generating device in the step S1 is: a near-infrared femtosecond pulse is generated by using a titanium: sapphire-based chirped pulse amplification system and a fiber pulse compression system based on a gas pressure gradient distribution, and then the near-infrared femtosecond pulse is focused into a gas cell to generate an XUV light beam with high harmonic characteristic or supercontinuum spectrum characteristic.

[0026] The application is further provided that: the diffraction device in the step S2 is a carrier copper mesh in the TEM experiment, and the specifications include: a center distance of 16.5 microns, a rib width of 6 microns, a pore diameter of 10.5 microns, a mesh number of 2000, a whole diameter of the copper mesh of 3.1 mm, and a mesh covering diameter of 3.0 mm.

[0027] The application is further provided that: the diffraction formula in the step 3 is:

[0028] mλ=d·sin(tan -1 (Δx / a))

[0029] Wherein, m represents the diffraction order, d is the grid spacing, λ is the wavelength, a is the distance from the copper mesh to the detection surface, and Δx is the distance between the 0 order beam and the diffraction beam on the detection surface.

[0030] The application is further provided that: the energy difference of the FID spectrum line of helium atom 2p→1s in the step S3 is 20.964eV, and the data is from NIST.

[0031] The application is further provided that: the Gaussian fitting method in the step S4 is used to determine the spot center position of the diffraction beam at different energies to improve the calibration accuracy.

[0032] The application is further provided that: the correction of the copper mesh placement angle deviation in the step S5 is realized by a multi-copper mesh clamping device, the clamping device is installed on a vacuum electric translation table, the center of the copper mesh is aligned with the XUV light beam by adjusting the position of the translation table, there is a certain included angle between each copper mesh on the clamping device, and the selection of the copper mesh angle is determined by the parallel relationship between the center connecting line of the diffraction light spot in the imaging result of the CCD and the FID spectrum line.

[0033] The application is further provided that: the CCD image angle correction in the step S5 is corrected by the divergence angle of the FID spectrum line, and the correction angle is 1.25°.

[0034] In summary, the application has the following beneficial effects:

[0035] 1. The application breaks through the limitation of traditional discrete reference points (such as metal absorption edge, characteristic spectrum line) by using the unique mapping relationship between the spacing (Δx) of the 0th and ±1st order beams generated by copper mesh diffraction and the XUV wavelength, and establishing a continuous calibration model of the energy axis, realizing uninterrupted energy calibration in the spectral coverage range of 16-40eV, and significantly improving the calibration accuracy and coverage range;

[0036] 2. The application directly uses the helium atom FID spectrum line (20.964eV) in the experimental system as the initial calibration reference without relying on high-cost equipment (such as EBIT, synchrotron radiation source) or complex standard light source (such as laser plasma), and completes the calibration combined with the desktop XUV light source (high harmonic or supercontinuum spectrum), greatly simplifying the experimental process and reducing the cost;

[0037] 3. The application dynamically adjusts the position and angle of the copper mesh through the vacuum electric translation table and the multi-copper mesh clamping device, combines with the aluminum film / zirconium film filter design, ensures the purity and stability of the XUV light beam, avoids the system disturbance caused by introducing external light source or frequently adjusting the grating parameters and experimental samples, and guarantees the stability and repeatability of the ultrafast experiment (such as attosecond transient absorption);

[0038] 4. The helium atom FID spectrum line has the characteristics of narrow linewidth (the resolution reaches the order of 10meV) and large divergence angle, can accurately locate the initial calibration point of the energy axis, the angle-broadening function fitted according to the Lorentz line type under different angles of the CCD image can correct the deflection error of the detector, in addition, the separation of the FID spectrum line and the high harmonic / supercontinuum spectrum without light path adjustment can be realized in the experiment, so as to reduce the interference of the FID large divergence angle spectrum line on the calibration result, and the energy calibration accuracy of the method is better than that of the traditional metal absorption edge method (the resolution is improved by more than 10 times);

[0039] 5、The present application can be used with metal absorption edge calibration method or gas characteristic line calibration method, further improve the calibration accuracy through multi-technology fusion; adapt to different XUV light sources (such as high harmonic, isolated attosecond pulse) and spectrometer configuration (flat field grating, grazing incidence structure), meet the dual needs of desktop experiment and large device;

[0040] 6、The present application only needs a single characteristic spectrum (such as FID spectrum) to complete the initial calibration, without frequent sample replacement or light path adjustment, significantly shortens the calibration time; suitable for ultrafast science, plasma diagnosis, material science and other fields, provides a standardized solution for the wide application of XUV spectrometer.

[0041] 7、The structure of the present application can meet two experiments, if the 0th order light beam is directly used to carry out attosecond transient absorption experiment, the method is in-situ, and the 0th order light beam incident spectrum can be inverted according to +1 or-1 order (only one beam is generally reserved in the experiment) diffracted light beam, so as to reduce the experimental noise; if the 0th order light beam through the copper net is not used, but the original XUV pulse is used to carry out the experiment, since the diffraction device only involves the movement of the vacuum electric translation table, does not have any influence on other parts of the light path and the propagation direction of the light path, and does not need to adjust and replace the sample, so the strict stability of the light path can still be maintained. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 It is a flowchart of an extreme ultraviolet spectrometer calibration method based on extreme ultraviolet diffraction in the embodiment of the present application;

[0043] Figure 2 It is a schematic diagram of the energy transmittance and dispersion of different metal films in the embodiment of the present application;

[0044] Figure 3 It is a schematic diagram of the distribution of light beams on the detection surface after the copper net diffracts the XUV light beams in the embodiment of the present application;

[0045] Figure 4 It is an image of the CCD image when the XUV light beams are diffracted by the copper net and the CCD image without the copper net in the embodiment of the present application;

[0046] Figure 5 It is a schematic diagram of the relationship between photon energy and separation distance in the embodiment of the present application;

[0047] Figure 6 It is a schematic diagram of the integral result of FID broadening and a schematic diagram of the angle-FID broadening relationship in the embodiment of the present application;

[0048] Figure 7 It is a schematic diagram of the relationship between the object distance and the incident angle of FFG at 40nm in the embodiment of the present application;

[0049] Figure 8 is a schematic view of a multi-copper mesh clamping device and a copper mesh in an embodiment of the present application, and the direction marks of the notch on the clamping device and the copper mesh are shown in the figure;

[0050] Figure 9 is the extraction result and calibration result of the center position of the light spot in an embodiment of the present application;

[0051] Figure 10 is a schematic view of a double optical gate (DOG) in an embodiment of the present application;

[0052] Figure 11 is a schematic view of the light path of an attosecond transient absorption experiment;

[0053] Figure 12 is a schematic view of an XUV spectrometer;

[0054] Figure 13 is a schematic view of a near-infrared femtosecond long pulse and a near-infrared femtosecond short pulse generation device;

[0055] Figure 14 is the relationship between the horizontal axis coordinate of the CCD picture and the propagation distance of the light beam. DETAILED DESCRIPTION

[0056] The following will be described in detail in combination with the accompanying Figures 1-14 The present application will be further described in detail.

[0057] Embodiment: An extreme ultraviolet spectrometer calibration method based on extreme ultraviolet diffraction, comprising the following steps:

[0058] S1, using a desktop XUV light source generation device to generate an XUV light beam.

[0059] In an embodiment of the present application, the XUV light source is realized by a self-built HHG-based desktop XUV pulse generation device. The chirped pulse amplification system (Femtopower Pro HE) based on titanium: sapphire produced by FemtoLaser company outputs a long pulse with a center wavelength of 800 nm, a pulse width of 25 fs, and an energy of 4.0 mJ. The long pulse is focused by a focusing mirror with a focal length of 500 mm and then introduced into a hollow core fiber filled with helium. The inner diameter of the optical fiber is 300 μm, and the gas pressure in the optical fiber is gradiently distributed from 3 mbar to 2500 mbar. The advantage of designing the gas pressure as a gradient distribution is that it can avoid the phenomenon of energy reduction and spectral blue shift of the infrared pulse at the entrance due to the high gas density. The long pulse is compressed by the gas-filled hollow core fiber and dispersed by a series of chirped mirrors, and a negative chirp few-cycle near-infrared pulse with a center wavelength of 750 nm, a pulse width of 8 fs, and an energy of 2.0 mJ is output. The purpose of keeping the output pulse negative chirp is to compensate for the dispersion introduced by subsequent propagation in air. The XUV light source in the present application can be generated by a long pulse or a short pulse, and its generation device is as followsFigure 13 .

[0060] The simplified optical path diagram for the subsequent attosecond transient absorption experiment is as follows: Figure 11 As shown. The few-period pulse is split at 45° by a beam splitter with a reflected light intensity: transmitted light intensity ratio of 9:1, resulting in Beam 1 and Beam 2. Their functions are to generate an attosecond probe beam and a pump infrared beam, respectively. Beam 1 passes through a set of dual optical gates and enters the vacuum environment, where it is focused into the gas cell GAS1. There, it interacts with the helium gas, undergoing the HHG process to generate an attosecond pulse. The dual optical gates consist of a polarization gate and a dichroic gate, as shown... Figure 10 As shown. The polarization gate consists of a half-wave plate and a quarter-wave plate, while the dichroic gate consists of a BBO crystal with a specially designed thickness. Since the BBO crystal is a negative uniaxial crystal, while the quartz crystal used as the quarter-wave plate is a positive uniaxial crystal, the quarter-wave plate in the polarization gate in DOG is a combination of a specially designed quartz plate and a BBO crystal. These two elements compensate for each other, achieving the effect of a quarter-wave plate. After the attosecond pulse is refocused by the tire lens, it passes through a metal film (usually a zirconium or aluminum film) to filter out the infrared pulse and further compensate for dispersion. The energy transmittance and dispersion of different metal films are as follows... Figure 2 As shown, Beam 2 is reflected to a beam combiner after passing through a time-delay control device. The attosecond pulse and the infrared pulse are combined at the beam combiner, which is mounted on a vacuum motorized frame. Behind the beam combiner is a vacuum thruster used to advance a mirror to the center of the optical path. The beam is reflected by this mirror and then incident on a CCD camera. The spatiotemporal coincidence of the two beams is determined by the center positions of the images formed by the two infrared pulses on the CCD camera when the metal film is not placed in the optical path (the attosecond pulse path contains a residual infrared pulse). The time-delay control device and the motorized vacuum frame housing the beam combiner are adjusted to maximize the coincidence of the center positions of the images and the interference intensity, ensuring optimal spatial and temporal coincidence. Since the dispersion difference between the infrared pulse and the attosecond pulse is extremely small in a vacuum environment, and the metal film has little effect on the group velocity of the beam, the spatiotemporal coincidence relationship determined by this method can be considered equally applicable to both the infrared pump pulse and the attosecond detection pulse. Both were focused on the gas target GAS2, and the subsequent XUV spectrometer was used to detect the transient absorption spectrum of the attosecond pulse.

[0061] S2. A diffraction device is set up in the propagation path of the XUV beam to cause the XUV beam to diffract, producing the 0th order main beam and the diffracted beam.

[0062] In the embodiment of the present application, the device for diffracting XUV is a carrier copper mesh for TEM experiment produced by Gilder Company in the United Kingdom, and the XUV light beam is further propagated in the optical path after being diffracted by the copper mesh after being focused twice by the gold-coated tire mirror. This device is applied in the experiment by the Thomas Pfeifer research group earlier, and the noise in the attosecond transient absorption experiment is reduced by using the diffraction ±1 order light beam to reverse the 0 order light beam. The schematic diagram of the XUV spectrometer is shown in Figure 12 The FFG is installed on the vacuum electric optical mirror frame produced by New Port Company, so that the FFG can be finely adjusted in the vacuum environment. The distance from the focus point of the secondary focusing of the XUV pulse to the FFG is the object distance r, the incident angle is a, the diffraction angle is b1 to b2, the diffraction image distance is r', the axis in the normal direction of the FFG is defined as X, and the axis parallel to the plane of the FFG is defined as Z.

[0063] The copper mesh has the following specifications: the center distance is 16.5 microns, the rib width is 6 microns, the aperture is 10.5 microns, the mesh number is 2000, the overall diameter of the copper mesh is 3.1 mm, and the diameter of the mesh covering part is 3.0 mm. Before or after the copper mesh, an aluminum film or a zirconium film is used to filter the near-infrared light, and the energy transmittance and dispersion of different metal films are shown in Figure 2 Due to the close distance between the position of the metal film and the copper mesh and the small divergence angle of the XUV pulse compared with the infrared pulse, the peripheral XUV pulse or infrared pulse will not sweep around the metal film. After the XUV light beam is focused twice by the tire mirror, it is dispersed in order of photon energy by the flat-field grating (FFG) and is incident on the MCP, and then is displayed by the thermoelectric water-cooled CCD camera. The size of the MCP panel is a circle with a diameter of 90 mm, the size of the CCD negative is 22022 microns x 17628 microns, and the pixel of the PNG format picture directly generated by the signal detected by the CCD is 3388 x 2712.

[0064] S3, the FID spectrum of the helium atom 2p→1s is used as a known energy reference point, and the separation distance of the ±1 order diffraction light beam measured at the energy corresponding position is used to derive part of the unknown parameters of the diffraction device and the spectrometer.

[0065] In the present invention and similar experimental apparatus, the spectral signal detected is given by the CCD camera, however, the actual detection surface is the MCP. Since the area of the CCD is much smaller than the MCP, and the two have different shapes, there is a scaling relationship between the image presented on the MCP and the image presented on the CCD camera. For example, for the 0th and 1st order beams of a certain wavelength diffracted from a grating, if the separation distance between them in the CCD image is n pixels, then the actual separation distance between them on the detection surface is n*δx iCCD . Where δx iCCD represents the actual distance on the MCP that a CCD pixel spans. Since each pixel of the CCD is square, δx iCCD is applicable in both the horizontal and vertical axes. In addition, since the lens used in our CCD camera is a long-focus lens, the imaging distortion caused by the transmission and refraction of the lens module can be basically ignored. Since the phosphor screen after the MCP only has an electron-photon conversion relationship with the MCP, and the two are the same size and are placed in parallel, the phosphor screen does not scale the imaging result, so we do not need to consider it. In our experiment, δx iCCD is determined as follows: since the CCD camera imaging range covers the left and right edges of the phosphor screen, and the edges are clearly visible when the background light is strong, and the size of the left and right edges of the phosphor screen is consistent, then the number of pixels that the left and right edges of the phosphor screen span in the camera's imaging result can be used to get the value of δx iCCD . Of course, there is a more direct way, that is, a ruler is placed tightly behind the phosphor screen, so that the geometric relationship between the CCD imaging result and the phosphor screen can be more finely determined. In this way, the Δx i corresponding to each horizontal axis coordinate on the CCD image can be determined, and since the edges of the MCP are determined, the position on the MCP corresponding to each pixel on the CCD (within the MCP range) can be determined.

[0066] It is noted that the propagation distance d of the light beam is not a constant value according to the configuration of the spectrometer. Fortunately, since the geometric parameters of the spectrometer and the optical path are determined, and the mapping relationship between the pixels of the CCD image and the MCP coordinates can be established by the method described above, the propagation distance d of each position of the horizontal axis of the CCD image can be given. However, there is still a problem to be noted: the distance from the FFG to the MCP is easy to determine, because the configuration of the spectrometer is relatively compact, and the positions of the various components are fixed; however, the copper mesh for diffraction is installed on a motorized translation stage, which is relatively far from the FFG, which leads to a certain error in the direct measurement of the distance of the two. This distance can be calibrated under the following conditions: if the photon energy corresponding to a certain coordinate on the horizontal axis of the CCD image is known, and the separation distance of the position is also determined, then the propagation distance d from the copper mesh to the position on the MCP where the photon energy is located can be accurately obtained. Thus, the distance from the copper mesh to the FFG is calibrated. In this way, the propagation distance corresponding to each coordinate of the horizontal axis of the CCD image can be given, and the result is shown in FIG. 8. Figure 14

[0067] In the embodiment of the present application, the FID spectrum of the helium atom 2p→1s is used to determine the known energy. Since the FID process has a long duration in time, it will appear a very narrow spectrum in the energy domain, and the divergence angle of the FID beam is large, which makes it easy to separate from the high-order harmonic or IAP and easy to identify. The energy level difference between the two energy levels is 20.964eV, and the data comes from NIST. The energy level structure of the helium atom is relatively simple, and the experimental difficulty and theoretical calculation difficulty are low, and there is sufficient existing data to prove the credibility of the energy level difference.

[0068] S4, the positions of the 0-order light beam and the ±1-order diffracted light beam at the detector at other positions are obtained by combining the Gaussian fitting method, and the corresponding relationship between the horizontal axis of the XUV spectrometer and the energy is inversely calculated according to the diffraction formula.

[0069] In the embodiment of the present application, when there is no copper mesh to diffract the XUV in the experimental optical path, the image of the XUV light beam displayed on the CCD is a classic discrete high-order harmonic spectrum, or the IAP (in other words, the XSC) presents a relatively smooth continuous spectrum. When the copper mesh diffracts the XUV light beam, the XUV light is mainly separated into five beams in space, which are the 0-order light propagating along the original path, and the ±1-order light propagating along the grid lines of the copper mesh in four directions with a certain diffraction angle, which is approximately distributed as shown in FIG. 9. Figure 3 ​The CCD images of the original XUV beam without copper mesh and the diffraction images with copper mesh are shown in FIG. 2 (wherein subgraph a is the CCD image with copper mesh, and subgraph b is the CCD image without copper mesh). According to the diffraction formula: Figure 4

[0070] mλ = d·sin(tan -1 (Δx / a))

[0071] wherein m represents the diffraction order, d is the grid spacing, λ is the wavelength, a is the distance from the copper mesh to the MCP, and Δx is the spacing (i.e. the separation distance) of the 0th order beam and the diffraction beams in the MCP plane. According to the diffraction formula, the positions of photons of different energies in the observation plane are inconsistent, and the relative position relationship is shown in FIG. 3. It can be seen that the XUV diffraction beams 1 and 3 do not change in the vertical direction, and the diffraction beams 2 and 4 are not considered in the present application. Figure 5

[0072] According to the diffraction formula, the expression of Δx can be obtained:

[0073] Δx = a·tan(sin -1 (mλ / d))

[0074] For different wavelengths, we have:

[0075] Δx i = a·tan(sin -1 (mλ i / d))

[0076] Considering that there may be errors in the installation of the copper mesh, causing the diffraction beams 1 and 3 not to propagate completely in the vertical plane, the separation distance of the diffraction beams 1 and 3 in the vertical direction is the projection of the actual separation distance, and if the deflection angle of the copper mesh is θ:

[0077] Δx i ′ = a·tan(sin -1 (mλ i / d))·cosθ

[0078] Combining E = hc / λ, we have,

[0079]

[0080] ​​Where h is Planck's constant, c is the speed of light, m, d are known quantities, and the deflection angle θ is eliminated by calibration. At this time, by determining the separation distance (in units of pixels) of the 0th order beam and the ±1st order beams on the CCD image, the corresponding photon energy can be directly obtained.

[0081] The horizontal axis of the CCD is designated as the x-axis and the vertical axis as the y-axis. At the position with spectral distribution, the center of the spot can be extracted from each y-direction pixel axis according to the intensity distribution of the axis, thereby determining the separation distance. The determination of the center of the spot is performed using Gaussian fitting, because the spatial intensity of the spot is generally considered to be Gaussian distributed, and the operation of the flat-field grating on the light beam is to focus in the horizontal direction and arrange in order according to the energy size, and no processing is performed in the vertical direction. Therefore, Gaussian fitting can more accurately restore the vertical coordinates of the center spot and the spot center of the ±1st order diffraction spot. Since the 1st order diffraction beam has a certain oblique incidence angle, a linear component is added in the Gaussian fitting.

[0082] Since this method only needs one point of accurate energy to calibrate all positions covered by the spectrum, it greatly simplifies the step of determining the positions of multiple known energies mentioned in the first part, and changes the requirement of this step from determining as many energy-known and high-resolution spectral lines within the spectral instrument measurement range as possible to determining one energy-known high-resolution spectral line within the spectral range, which greatly reduces the experimental difficulty and meets the demand of full-range continuous calibration of XUV spectrum in experiments.

[0083] In addition, the known terms and unknown terms are arranged, and the expression of E i can be converted to,

[0084]

[0085] Thus, A and B can be determined according to the two characteristic lines without determining the grid spacing of the copper mesh.

[0086] S5, by correcting the copper mesh placement angle deviation and the CCD image angle deviation, ensuring the calibration accuracy.

[0087] In the embodiment of the present application, the angle correction of the CCD image: since the deviation of the CCD image obtained in the experiment is the sum of the deviations of each part of the experimental light path (such as light path deviation, FFG angle deviation), the final image obtained in the experiment is often not the expected one arranged horizontally on the CCD in the order of different energies and distributed vertically, and the spectral lines of the same energy may have a certain deflection angle on the CCD, which needs to be corrected when processing the data, and the correction method is similar to the calibration of the spectrometer. Some research groups use metal films to correct the CCD image of the equal-energy absorption edge of the continuous spectrum. In the present application, since the FID has a relatively large divergence angle, it is easier to determine this deflection angle, so it is not necessary to use other methods for correction. For a CCD image, it is rotated within a certain angle range, and the Lorentz line fitting is performed on the FID concentrated region to obtain the broadening at this angle. By comparing the broadening at different angles, the rotation angle at which the broadening is the smallest is determined as the deflection angle of the CCD image, and the integral result of the FID region and the angle-broadening relationship are as shown in Figure 6 .

[0088] In the embodiment of the present application, the copper mesh placement angle deviation correction: since the diffraction result of the XUV light beam by the flat-field grating is not uniformly distributed, for example, 40 nm XUV, the position presented on the CCD is greatly affected by the angle and the object distance, as shown in Figure 7 When the copper mesh is placed at a certain angle, the diffraction beams 1 and 3 do not completely propagate in the vertical plane, and a deviation angle is generated. It is extremely complex to statistically analyze the influence of the deviation angle on the diffraction distance and the diffraction angle of the XUV light at all energies. At the same time, this deflection will adversely affect the imaging accuracy of the XUV spectrometer. In order to avoid this situation as much as possible, a multi-copper mesh multi-angle clamping device is designed, and the structure is as shown in Figure 8 . The clamping device is installed on a vacuum electric translation stage parallel to the experimental platform. According to the change of the CCD image, the position of the electric translation stage is adjusted so that the center of the copper mesh is aligned with the XUV light beam. Each copper mesh on the clamping device is installed on the clamping device in a certain angle difference order, and the difference is determined by the etched mark of the copper mesh and the mark line on the clamping device. The flatness relationship of the final copper mesh is determined according to the parallel relationship between the center line of the diffraction beam spot and the FID spectral line.

[0089] According to the corrected experimental results, first, the center position of the FID spectral line is determined on the CCD according to the pixel value. In the results obtained by the experimental light path of the present application, the pixel value of the energy position is 1450. In order to avoid the influence of the FID signal on the subsequent Δx i(mainly determined by the central energy of the high harmonic or supercontinuum spectrum, since the FID spectrum has a larger divergence angle, compared with the high harmonic, it has a more special spatial structure, so it is not considered to use the FID signal to determine Δx i The parameters such as pressure are fine-tuned to make the FID signal as weak as possible without changing the optical path, so as to obtain a relatively pure high harmonic signal. The high harmonic signal is integrated at the pixel for a certain broadening, and based on the integration result, the longitudinal axis pixel value corresponding to the spot center is fitted, and the fitting method adopts Gaussian fitting. The spot center positions of the processed beams along the pixel distribution and the obtained CCD transverse axis-energy and transverse axis-wavelength relationship diagram are as shown in Figure 9 The fitting curves of the two are also plotted in the figure (where (a) is a schematic diagram of the spot center position extraction result, and (b) is a schematic diagram of the calibration result, wherein the orange line is the fitting curve).

[0090] Similarly, for the XUV continuous spectrum generated by the XUV source, that is, IAP, the same method is used for calibration to obtain the corresponding energy of all positions in the spectral coverage. On the experimental optical path, unlike the generation of high harmonics, the generation of XUV supercontinuum requires that the few-cycle near-infrared pulse pass through a double optical gate (DOG) before entering the vacuum cavity, as shown in Figure 10 The double optical gate is composed of a 1 / 2 wave plate and a 1 / 4 wave plate, wherein the 1 / 4 wave plate is composed of a quartz wave plate and a BBO of a certain thickness. Since the XUV intensity is lower when generating supercontinuum, its signal-to-noise ratio is lower, and the error in determining the Gaussian spot center position is greater. A better solution is to calibrate the supercontinuum in combination with the results of the high harmonic calibration.

[0091] The specific embodiments are only an explanation of the present application, and are not a limitation of the present application. Those skilled in the art can make modifications to the embodiments without creative contribution after reading the specification, as long as the modifications are within the scope of the claims of the present application.

Claims

1. An extreme ultraviolet spectrometer calibration method based on extreme ultraviolet light diffraction, characterized in that: The method comprises the following steps: S1, generating an XUV light beam by using a desktop XUV light source generating device; S2, setting a diffraction device on the propagation path of the XUV light beam to diffract the XUV light beam to generate a 0th order main light beam and a diffraction light beam; S3, using the FID spectrum line of helium atom 2p→1s as a known energy reference point, and deriving the partial unknown parameters of the diffraction device and the spectrometer according to the separation distance of the ±1st order diffraction light beam measured at the energy corresponding position; S4, combining a Gaussian fitting method to obtain the positions of the 0th order light beam and the ±1st order diffraction light beam on the detector at other positions, and inversely calculating the corresponding relationship between the XUV spectrometer horizontal axis and the energy according to the diffraction formula; S5, correcting the copper mesh placement angle deviation and the CCD image angle deviation to ensure the calibration accuracy.

2. The method of calibrating an extreme ultraviolet spectrometer based on extreme ultraviolet diffraction of claim 1, wherein: The specific process of generating the XUV light beam in the step S1 is as follows: a titanium: sapphire-based chirped pulse amplification system and a fiber pulse compression system based on gas pressure gradient distribution are used to generate near-infrared femtosecond pulses, and then the near-infrared femtosecond pulses are focused into a gas cell to generate an XUV light beam with high harmonic characteristic or supercontinuum spectrum characteristic.

3. The method of calibrating an extreme ultraviolet spectrometer based on extreme ultraviolet diffraction of claim 1, wherein: The diffraction device in the step S2 is a carrier copper mesh in the TEM experiment, and the specifications include: a center distance of 16.5 microns, a rib width of 6 microns, a pore diameter of 10.5 microns, a mesh number of 2000, a whole copper mesh diameter of 3.1 mm, and a mesh covering part diameter of 3.0 mm.

4. The method of calibrating an extreme ultraviolet spectrometer based on extreme ultraviolet diffraction of claim 1, wherein: The diffraction formula in the step 3 is as follows: mλ = d - sin(tan -1 (Δx / a)) wherein m represents the diffraction order, d is the grid spacing, λ is the wavelength, a is the distance from the copper mesh to the detection surface, and Δx is the interval of the 0th order light beam and the diffraction light beam on the detection surface.

5. The method of calibrating an extreme ultraviolet spectrometer based on extreme ultraviolet diffraction of claim 1, wherein: The energy difference of the FID spectrum line of helium atom 2p→1s in the step S3 is 20.964 eV, and the data is from NIST.

6. The method of calibrating an extreme ultraviolet spectrometer based on extreme ultraviolet diffraction of claim 1, wherein: The Gaussian fitting method in the step S3 is used to determine the spot center position of the diffraction light beam to improve the calibration accuracy.

7. The method of calibrating an extreme ultraviolet spectrometer based on extreme ultraviolet diffraction of claim 1, wherein: The correction of the copper mesh placement angle deviation in the step S5 is realized by a multi-copper mesh clamping device, the clamping device is installed on a vacuum nanometer electric translation stage, the position of the translation stage is adjusted to align the center of the copper mesh with the XUV light beam. There is a certain included angle between each copper mesh on the clamping device, and the selection of the copper mesh angle is determined by the parallel relationship between the center line of the diffraction light beam spot in the imaging result of the CCD and the FID spectrum line.

8. The method of calibrating an extreme ultraviolet spectrometer based on extreme ultraviolet diffraction of claim 1, wherein: The CCD image angle correction in the step S5 is performed by the divergence angle of the FID spectrum line, and the correction angle is 1.25°.

Images