METHOD AND DEVICE FOR DETERMINING THE POSITIONS OF MOLECULES IN A SAMPLE

Description

TECHNISCHES GEBIET DER ERFINDUNG

[0001] The invention relates to a method and a device for determining positions of spaced-apart molecules in one or more spatial directions in a sample, in particular according to a MINFLUX method. STAND DER TECHNIK

[0002] State-of-the-art localization and tracking methods for single molecules based on the MINFLUX method are described, for example, in patent applications DE 10 2011 055 367 A1, WO 2015 / 052186 A1, patent specification DE 10 2013 114 860 and the publication Balzarotti F, Eilers Y, Gwosch KC, Gynnå, A, Westphal V, Stefani F, Elf J, Hell SW "Nanometer resolution imaging and tracking of fluorescent molecules with minimal photon fluxes", arXiv:1611.03401 [physics.optics] (2016).

[0003] Essentially, a sample is scanned with fluorophores with an excitation light intensity distribution that exhibits a local intensity minimum in the center surrounded by intensity maxima. Such intensity distributions are known, for example, from STED (Stimulated Emission Depletion) microscopy.

[0004] In MINFLUX microscopy, however, the sample is scanned stepwise using the minimum of the excitation light distribution. First, a single fluorophore is located using an independent method, and the position of the fluorophore is roughly estimated in a pre-localization step. Then, the central minimum of the excitation light distribution is successively positioned at several positions near the estimated position, and an emission signal is measured for each position, preferably by counting individual fluorescence photons with a confocal point detector. The measured fluorescence intensity, or the number of photons counted per unit time, depends for each position of the excitation light distribution on the distance between the minimum of the excitation light distribution and the actual location of the fluorophore: the closer the fluorophore is to the minimum, the lower the fluorescence signal.In this way, the position of a single fluorophore can be determined with extremely high accuracy (down to 1 nm), especially iteratively. State-of-the-art methods, for example, use modified maximum likelihood estimators.

[0005] A further advantage of MINFLUX microscopy, in addition to its very high accuracy, is its high photon efficiency: On average, a significantly smaller number of excitation photons are required to localize a fluorophore than, for example, single-molecule localization using PALM / STORM microscopy, which reduces bleaching and phototoxicity (in the case of living samples). This results from the fact that with each iteration, the minimum of the excitation light distribution typically approaches the actual fluorophore position – with each step, the fluorophore is exposed to fewer photons.

[0006] STED microscopy is based on the depletion of fluorescence by stimulated emission: By irradiating STED light of a suitable wavelength, fluorophores can be specifically (re)converted from the excited state to the ground state, in which the fluorophores no longer fluoresce.

[0007] In STED microscopy, an approximately Gaussian excitation light distribution is typically superimposed on a, for example, donut-shaped STED light distribution with a local minimum, whereby the maximum of the excitation light distribution coincides with the minimum of the STED light distribution. This selectively de-excites fluorophores located off-center by the STED light and thus do not contribute to the fluorescence signal. In this way, the effective resolution can be reduced well below the diffraction limit.

[0008] The article "Stimulated emission depletion microscopy with array detection and photon reassignment" by Wang Wensheng et al. (Optics and Lasers in Engineering, Elsevier, Amsterdam, NL, Vol. 129, February 18, 2020, doi: 10.1016 / J.OPTLASENG.2020.106061) describes a STED microscopy technique in which fluorescence light emitted by emitters in the sample is spatially detected in a detection plane using an array of avalanche photodiodes (APDs). The APD signals are combined using pixel reassignment to improve the signal strength and thus the signal-to-noise ratio while maintaining the same spatial resolution.

[0009] Some prior art publications have already proposed a combination of MINFLUX and similar single-molecule localization methods with STED microscopy.

[0010] For example, patent applications DE 10 2017 104 736 A1 and EP 3 372 989 A1 describe the superposition of an excitation light distribution with a local maximum with a STED light distribution with a local minimum. The sample is scanned by shifting the STED distribution with the STED minimum, and the position of the fluorophore is determined from the measured fluorescence intensity values ​​at different positions of the STED intensity distribution using a fitting algorithm. The STED light can suppress additional fluorescence light caused by other fluorophores located in the vicinity of the fluorophore to be localized. However, the described method is significantly less photon-efficient, especially compared to the conventional MINFLUX method, since the effective emission (detection) point spread function (ESF) resulting from the superimposed light distributionsPoint Spread Function (PSF) has a maximum in the center instead of a minimum (as in MINFLUX microscopy).

[0011] WO 2020 / 198750 A1 also discloses a method in which a sample is labeled with a first and a second fluorophore, wherein the excitation and emission wavelengths of the two fluorophores differ. In a 4π microscope arrangement (i.e. an arrangement in which the light beams are focused onto the sample by two objectives positioned in the axial direction on opposite sides of the sample, whereby these light beams at least partially interfere with each other), the position of the first fluorophore is determined at short intervals, e.g. by MINFLUX nanoscopy, and the position of the sample relative to the beam path of the microscope is adjusted based on the determined position so that the first fluorophore is always in the center of the field of view during the experiment.By imaging the second fluorophore relative to the first fluorophore, the second fluorophore is then exposed to a Gaussian excitation light distribution in the range of the excitation wavelength of the second fluorophore, which is superimposed with a donut-shaped STED light distribution that affects exclusively the fluorescence of the second fluorophore. This method can be used, for example, to localize and track RNA polymerase complexes labeled with the second fluorophore in biological cells relative to specific gene loci labeled with the first fluorophore.

[0012] Furthermore, patent EP 3 055 674 B1 and patent application US 2014 / 0042340 A1 disclose a MINFLUX-based single-molecule tracking method in which an excitation light distribution with a local minimum and a STED distribution with a local minimum are concentrically superimposed. The width and shape of the minimum of the STED distribution essentially correspond to the width and shape of the excitation distribution, except for a slight wavelength-dependent deviation. Due to the additional STED light, the fluorophores are only excited by the excitation light in a narrow region around the center. This eliminates the need to isolate a fluorophore across the entire field of view; instead, isolation only needs to be achieved for the narrower region of the STED minimum to track the movement of a fluorophore.

[0013] An analogous combination of the MINFLUX and STED techniques is also disclosed in Chinese patent application CN 111024658 A, according to which a sample is scanned together with concentrically arranged excitation and STED light distributions to localize individual fluorophores in the sample.

[0014] Using the described methods, additional STED light can be used, among other things, to reduce disturbing background fluorescence during MINFLUX localization of individual fluorophores.

[0015] However, these methods have the disadvantage that the effective PSF of the fluorescence emission light (detection light) is influenced by both the excitation light distribution and the STED light distribution. This results in a reduced gradient of the effective excitation and a reduced detection efficiency. Furthermore, these methods require a relatively good knowledge of the position of the molecule to be localized. AUFGABE DER ERFINDUNG

[0016] The invention is therefore based on the object of providing a method and a device for determining positions of spaced-apart molecules in one or more spatial directions in a sample, which reduces the background fluorescence in the MINFLUX localization in an improved manner with regard to the disadvantages of the prior art explained above. LÖSUNG

[0017] The object of the invention is achieved by a method having the features of independent claim 1 and a device having the features of claim 13. Dependent claims 2 to 12 relate to advantageous embodiments of the method, and dependent claims 14 and 15 relate to advantageous embodiments of the device. BESCHREIBUNG DER ERFINDUNG

[0018] The method according to a first aspect of the invention serves to determine positions of spaced-apart molecules in one or more spatial directions in a sample.

[0019] The spaced-apart molecules are particularly designed to emit emission light of a second wavelength when the sample is illuminated with excitation light of a first wavelength, wherein the emission light can be, for example, fluorescent light.

[0020] The light-emitting molecules are spaced apart from one another, i.e., isolated. In other words, the molecules have such an average distance from one another that they can be localized separately using the method according to the invention. Since the method according to the invention is, in particular, a non-diffraction-limited localization method, the average distance can, in particular, be below the wavelength of the excitation light and / or the deactivation light. In some applications, however, it can nevertheless be advantageous if the average distance corresponds at least to the wavelength of the excitation light and / or the deactivation light. In this case, it is particularly possible to pre-localize the individual molecules using diffraction-limited methods such as PALM / STORM microscopy with camera imaging with lower resolution.

[0021] The isolation of the molecules can be achieved, in the case of fluorophores (also referred to as 'fluorescent dyes' in the context of this application), for example, by labeling sample molecules with a defined amount of fluorophores, resulting in a desired labeling density of the sample. Alternatively, it is also possible, for example, to label the sample with a higher density of fluorophores if not all fluorophores emit light simultaneously. For this purpose, the blinking behavior of the fluorophores can be exploited, i.e., the property of spontaneously switching, with a certain probability, from an active state in which fluorescent light is emitted upon irradiation with excitation light, to an inactive state in which no fluorescent light is emitted upon irradiation with excitation light. Of course, a subset of the fluorophores can also be specifically converted into the active or inactive state, e.g.by irradiating activation light (also called switching light) or deactivation light of a suitable wavelength (especially when using so-called switchable fluorophores) in order to isolate the active fluorophores.

[0022] The determination of positions of the molecules is also referred to here as localization and includes in particular the determination of the position with an accuracy in the range of 1 nm to 100 nm, more particularly 1 nm to 10 nm.

[0023] The position can be determined in one spatial direction, in two spatial directions, or in three spatial directions. In particular, the two spatial directions run perpendicular to an optical axis along which the excitation light and, in particular, also the deactivation light is irradiated onto the sample, i.e., the two directions form a plane (xy plane) perpendicular to the optical axis, in particular a focal plane in the sample. In the case of a position determination in three spatial directions, the third spatial direction runs, in particular, parallel to the optical axis (i.e., in the z direction). Of course, other spatial directions are also possible, e.g., planes arranged obliquely to the optical axis.

[0024] The procedure includes the following steps: Generating a plurality of light distributions, each light distribution having a local intensity minimum and adjacent intensity increase regions, the light distributions comprising an excitation light distribution and a deactivation light distribution, illuminating the sample with the excitation light distribution and the deactivation light distribution, detecting photons emitted by the molecule for different positionings of the excitation light distribution, and deriving the position of the molecule based on the photons detected for the different positionings of the excitation light distribution.

[0025] The light distributions (i.e. spatial intensity distributions of the light) each have a local intensity minimum, i.e. one point or several adjacent points in space at which a minimum light intensity prevails. The minimum light intensity can be zero or almost zero, i.e. not greater than a background light intensity. The local intensity minima of the excitation light distribution and the deactivation light distribution lie in particular in a focal plane in the sample. In particular, the excitation light distribution and / or the deactivation light distribution are symmetrical, with the respective local intensity minimum lying in a center of the respective light distribution. Of course, the excitation light distribution and / or the deactivation light distribution can also each have several local minima, e.g. in the case of periodic light distributions such asLine patterns or two-dimensional or three-dimensional grids.

[0026] The light distributions are generated, in particular, by generating a coherent light beam (i.e., an excitation light beam and a deactivation light beam) for each light distribution, e.g., from a laser source. The light beams are each amplitude- or phase-modulated and focused into the sample, in particular using an objective lens. Due to the amplitude or phase modulation, the light distributions described above, with the local minima, are then produced at the focus. The light beams can be amplitude- or phase-modulated separately or together. For example, a so-called 2D donut or 3D donut can be generated by phase modulation.

[0027] In the context of this specification, the term 'excitation light' means light which has a suitable wavelength to excite the spaced-apart molecules, in particular the fluorophores, in the sample, so that they emit photons, in particular fluorescence.

[0028] In the context of this specification, 'deactivation light' refers to light that prevents the spaced-apart molecules from emitting photons when exposed to excitation light. For this purpose, the deactivation light can, for example, convert the molecules from an excited state to a ground state or a vibrational state above the ground state, e.g., through stimulated emission. Alternatively, it is also possible for the deactivation light to place the molecules into an inactive state, e.g., a dark state, in which they cannot emit photons. Furthermore, the deactivation light according to the present invention could also prevent the molecules from changing from an inactive state to an active state. In the context of this specification, the deactivation light can also be referred to as fluorescence-preventing light and is, in particular, stimulation light or STED light.

[0029] According to the method according to the invention, the excitation light distribution and the deactivation light distribution illuminate the sample, ie the excitation light distribution illuminates at least a part of the sample in which the molecule to be localized is located.

[0030] The emitted photons are detected in particular with a point detector such as an avalanche photodiode (APD). single photon avalanche diode, SPAD), a photomultiplier tube (PMT), a hybrid detector, or an area detector comprising several point detectors arranged in an image plane, e.g., of the types mentioned above. This allows, for example, a count rate, i.e., a number of detected photons per unit time, to be determined, which can be treated analogously to an emission intensity for a single molecule.

[0031] To reposition the excitation light distribution, the excitation light beam can be shifted relative to the sample, for example with one or more electro-optical deflection units (EOLs). electro optical deflector, EOD), acousto-optical deflection units (English acousto optical deflector, AOD), galvanometric scanners, micromirror arrays or light modulators (English). spatial light modulators, SLM). As an alternative to beam deflection, a sample holder to which the sample is attached can of course also be repositioned relative to the excitation light beam, for example using piezoelectric elements.

[0032] The emitted photons are recorded for each position of the excitation light distribution, and the position of the molecule is determined from the positions of the excitation light distribution and the associated emission light intensities or photon count rates. This can be done, for example, using a MINFLUX method. This method exploits the fact that the closer the molecule is to the local minimum of the excitation light distribution, the fewer photons are emitted by it. The position of the molecule can be estimated, for example, using a possibly modified maximum likelihood estimator, in particular as known from the prior art.

[0033] The method according to the invention is characterized in that the local minimum of the excitation light distribution is arranged successively at a plurality of scanning positions within a scanning range, wherein the light intensity of the deactivation light corresponds at most to the saturation intensity in a capture range associated with the scanning range, in which the position of the molecule can be clearly derived from the scanning positions and the associated detected photons. In particular, the light intensity of the deactivation light also corresponds at most to the saturation intensity in the scanning range.

[0034] The term 'saturation intensity' refers to the light intensity of the deactivation light at which 50% of the molecules, especially fluorophores, are deactivated (e.g. in the case of STED light by stimulated emission).

[0035] If the molecule is located within the capture region, it can be uniquely localized by evaluating the photon counts for the respective scanning positions. Outside the capture region, however, several estimated positions of the molecule can result from the determined photon counts. This means, for example, that a maximum likelihood estimator determined based on the photon counts and scanning positions will deliver unique values, provided the molecule is actually located within the capture region. The capture region is smaller than the scanning region, but can, depending on the pattern used, be smaller. set of targeted coordinates, Depending on the sampling position (STC) and, in particular, the position estimator used, the capture area can be equal to or larger than the sampling area. Furthermore, the capture area can be an area (e.g., in the focal plane for 2D localization) or a volume (for 3D localization).

[0036] A maximum intensity of the deactivation light in the capture region, which corresponds at most to the saturation intensity, can be achieved, for example, by a broader deactivation light distribution (e.g., by adjusting the intensity or phase modulation) compared to the excitation light distribution. The excitation light distribution can be repositioned, for example, within a broader range around the minimum of a stationary deactivation light distribution, or the deactivation light distribution can be repositioned together with the excitation light distribution.

[0037] By means of the method according to the invention it can be achieved in particular that an effective detection point spread function (PSF), in particular a width of the effective detection point spread function (e.g. defined as full width at the position of half maximum, engl. full width at half maximum, FWHM) is largely unaffected by the deactivation light distribution, regardless of where the molecule is actually located within the capture region.

[0038] The term "effective detection PSF" refers to the distribution of the emission light emitted by the molecule after excitation with the excitation light, which results from diffraction by the optical system. The detection PSF here essentially results from a combination of the excitation PSF and the PSF of the deactivation light. However, in the method according to the invention, this PSF may be distorted by the fact that the excitation PSF and the deactivation PSF are not arranged concentrically to one another, thus resulting in a deviation from the position of the center of a projection of the detection pinhole, i.e., a non-confocal arrangement.

[0039] In state-of-the-art STED microscopy (STED light corresponds to deactivation light), the effective detection PSF is deliberately narrowed by the STED light in order to increase the resolution below the diffraction limit.

[0040] In contrast, the light distributions according to the present invention are designed such that the effective detection PSF is essentially only influenced by the excitation light distribution. The detection PSF is therefore not intended to be narrower in the method according to the invention. Rather, the deactivation light is intended to suppress background emission from other fluorophores located in the region of the excitation light distribution or in its vicinity in order to increase the signal-to-background ratio. The fact that the deactivation light distribution has only a minimal effect on the effective PSF of the photon emission has the advantage of enabling the suppression of background fluorescence while simultaneously enabling simpler and more precise determination of the position of the molecule.

[0041] According to one embodiment, the light intensity of the deactivation light at the scanning positions corresponds to a maximum of 50% of the saturation intensity, in particular a maximum of 45% of the saturation intensity, further in particular a maximum of 40% of the saturation intensity, further in particular a maximum of 35% of the saturation intensity, further in particular a maximum of 30% of the saturation intensity, further in particular a maximum of 25% of the saturation intensity, further in particular a maximum of 20% of the saturation intensity, further in particular a maximum of 15% of the saturation intensity, further in particular a maximum of 10% of the saturation intensity, further in particular a maximum of 5% of the saturation intensity.

[0042] According to one embodiment, the deactivation light is STED light, i.e. light that puts the molecules, in particular fluorophores, into a vibrational state above the ground state by stimulated emission, from which they transition to the ground state without emitting a photon.

[0043] According to a further embodiment, the excitation light distribution is repositioned while the deactivation light distribution remains stationary, so that the local minimum of the excitation light distribution is positioned several times differently within a vicinity of the local intensity minimum or a region around the local intensity minimum of the deactivation light distribution. The region around the local intensity minimum, in which the excitation light distribution is repositioned, has a width such that the effective detection PSF of the photon emission in the region is not influenced or only minimally influenced by the deactivation light.

[0044] By keeping the deactivation light distribution stationary, a less complex and, in particular, more compact optical setup can be achieved, since no means for simultaneously rapidly shifting the excitation light beam and the deactivation light beam are required. For example, the excitation light can be shifted using acousto-optic deflectors (AOD), while the deactivation light beam is shifted more slowly, e.g., by a galvanometric scanner.

[0045] According to a further embodiment, light distributions resulting from interference of coherent light are used to determine the positions. This is the case, for example, with phase modulation of laser light in a plane conjugated to the objective pupil to generate the light distributions at the focus.

[0046] According to a further embodiment, the plurality of light distributions is generated using at least one beam-shaping device, in particular at least one light modulator, further in particular with a plurality of switchable pixels. The at least one light modulator can, for example, be a phase-modulating so-called spatial light modulator (SLM). These modulators have, in particular, an active surface with a blaze grating, wherein the switchable pixels are arranged on the active surface, and wherein the individual pixels can be set to a desired phase value by a switching process using liquid crystals. By generating phase patterns such as vortex patterns or concentric rings and circular disks using the switchable pixels, intensity distributions such as 2D donuts and 3D donuts can be generated in the focal plane. In particular, the at least one light modulator is arranged in a plane conjugate to an objective pupil. As an alternative to a light modulator, a phase plate onto which the phase pattern is impressed can also be provided.

[0047] According to a further embodiment, a region between two local maxima of the deactivation light distribution adjacent to the local intensity minimum of the deactivation light distribution is further extended at least in one spatial direction, in particular by 10% or more, in particular by 20% or more, in particular by 30% or more, in particular by 40% or more, in particular by 50% or more, in particular by 60% or more, in particular by 70% or more, in particular by 80% or more, in particular by 90% or more, in particular by 100% or more, in particular by 150% or more, in particular by 200% or more, in particular by 300% or more,further, in particular, by 400% or more, further, in particular, by 500% or more, than a corresponding area between two local maxima of the excitation light distribution adjacent to the local minimum of the excitation light distribution. The width of the deactivation light distribution can be achieved, for example, by changing the overall intensity of the deactivation light and / or by changing the phase pattern imposed on the deactivation light beam. The wider deactivation light distribution compared to the excitation light distribution advantageously ensures that the effective detection PSF is only minimally influenced by the deactivation light distribution during repositioning of the excitation light distribution.

[0048] According to a further embodiment, a light intensity of the deactivation light within the scanning region in which the excitation light distribution, in particular the local minimum of the excitation light distribution, is repositioned, is 10% or less, in particular 5% or less, more particularly 4% or less, more particularly 3% or less, more particularly 2% or less, more particularly 1% or less, more particularly 0.5% or less, more particularly 0.4% or less, more particularly 0.3% or less, more particularly 0.2% or less, more particularly 0.1% or less, of a maximum light intensity of the deactivation light. The maximum light intensity of the deactivation light refers to the intensity value at a local intensity maximum, which prevails in particular at the position of a local maximum adjacent to the local minimum.In this way, background suppression can be achieved with only minimal impact of the deactivation light distribution on the effective detection PSF.

[0049] According to a further embodiment, a plurality of scanning steps are performed, wherein the local minimum of the excitation light distribution in each of the scanning steps is positioned at a plurality of scanning positions of a respective scanning region, wherein the respective scanning region of each scanning step has a smaller area or a smaller volume than the scanning regions of the preceding scanning steps, and the deactivation light distribution is adapted in at least some of the scanning steps, in particular depending on the area or the volume of the respective scanning region. In particular, the capture region of the respective scanning step assigned to the respective scanning region also has a smaller area or a smaller volume than the capture region of the preceding scanning step.In particular, the method can carry out further scanning steps in which the area or volume of the scanning region (and in particular also the area or volume of the capture region) remains constant, in particular towards the end of a localization of a molecule.

[0050] According to a further embodiment, a total intensity of the deactivation light distribution is adapted as a function of the area or volume of the respective scanning area, wherein in particular the total intensity is greater the smaller the area or volume of the respective scanning area is.

[0051] According to a further embodiment, a shape of the deactivation light distribution is adapted depending on the area or volume of the respective scanning region, wherein, in particular, the half-width of the deactivation light distribution is smaller the smaller the area or volume of the respective scanning region. For this purpose, for example, the phase pattern used for phase modulation of the deactivation light beam can be modified between the scanning steps, for example by controlling the pixels of a light modulator.

[0052] According to a further embodiment, the scanning area is a circle or a sphere with a radius around an estimated position of the molecule, wherein the radius has a first value in an initial scanning step in which the excitation light distribution is repositioned several times within the scanning area, and wherein in particular the excitation light distribution is repositioned within the scanning area following the initial scanning step in at least one further scanning step, wherein the radius in the at least one further scanning step has a second value that is smaller than the first value. This corresponds, for example, to the iterations of a MINFLUX method. In particular, the minimum of the excitation light distribution is determined in each of the scanning steps successively at a plurality of positions of a coordinate pattern arranged within the respective scanning area (also referred to as set of targeted coordinates, STC, or targeted coordinate pattern, TCP) around an estimated position of the fluorophore. In particular, after each scanning step, an estimated position of the fluorophore is determined from the emitted photons and the corresponding positions of the minimum of the excitation light distribution, whereby in each scanning step, the position of the molecule estimated in the previous scanning step defines the center of the scanning area.

[0053] According to a further embodiment, the deactivation light distribution remains constant during the repositioning of the excitation light distribution. This means that, in particular, the total light intensity and the shape of the deactivation light distribution remain constant. In particular, the total intensity and the shape of the deactivation light distribution remain constant over a plurality of scanning steps, in particular over all scanning steps. Alternatively, the deactivation light distribution can be changed during the repositioning of the excitation light distribution (or between position changes of the excitation light distribution), this means that, in particular, the shape and / or the total intensity of the deactivation light distribution can change.

[0054] According to a further embodiment, after deriving the position of the molecule, the position of at least one further molecule in the sample is determined by illuminating the further molecule with the excitation light distribution, detecting photons emitted by the further molecule for different positions of the excitation light distribution, and deriving the position of the further molecule on the basis of the photons detected for the different positions of the excitation light distribution, wherein, before determining the position of the further molecule, the local minimum of the excitation light distribution and the local minimum of the deactivation light distribution are positioned, in particular simultaneously, at an expected position of the further molecule. The excitation light distribution and the deactivation light distribution can be determined, for example, jointly with a scanning device, e.g.a galvanometric scanner, whereby in particular an expected position of the further molecule in the focal plane is arranged in the center of an image of a detection pinhole (ie a confocal arrangement is realized with respect to the expected position of the molecule).

[0055] According to a further embodiment, a pre-localization of the molecule is carried out, wherein an initial estimated position of the molecule is determined, and wherein the sample is illuminated during the pre-localization, in particular only with the excitation light distribution (i.e., not with the deactivation light distribution). In particular, the pre-localization comprises locating the molecule, i.e., identifying a light emitter in a specific sample volume by detecting photons above a background emission. In particular, the localization is carried out on the basis of the initial estimated position, wherein, furthermore, in particular, a center point of the initial scanning area is arranged at the initial estimated position. The pre-localization does not have to be based on the MINFLUX method, but can also be based on an independent positioning method with lower accuracy. E.g.Pre-localization can be performed using wide-field fluorescence microscopy with camera imaging. Another possibility is confocal scanning of a sample area with an approximately Gaussian focus of the excitation light and detecting the emission light with a point detector. According to a further embodiment, during pre-localization, the sample is illuminated with an excitation light distribution with a local minimum (in particular, the same excitation light distribution as in MINFLUX localization), whereby emission light is detected in a position-dependent manner and the initial position of the molecule is determined based on the position-dependent detection. Position-dependent detection can be performed in particular by shifting a projection of a detection pinhole into the sample plane while maintaining the same position of the excitation light distribution (so-called 'pinhole orbit scan'). For example,The projection of the detection pinhole can be shifted using a first scanning device, e.g., a galvanometric scanner, while a second scanning device, e.g., one or more EODs, corrects the position of the excitation light beam so that the position of the excitation light distribution remains constant. Alternatively, a distribution of the emission light in a detector plane can be determined using an area detector (e.g., an array of avalanche photodiodes) to determine the initial position of the molecule.

[0056] According to a further embodiment, the sample is illuminated with the deactivation light distribution during pre-localization (e.g., during a pinhole orbit scan or position-dependent detection with an array detector). This advantageously reduces background fluorescence, particularly from planes located along the optical axis above or below the focus of the excitation light, thereby increasing the accuracy of pre-localization. In particular, a shape of the deactivation light distribution and / or an intensity of the deactivation light is adjusted between pre-localization and the subsequent localization steps (e.g., MINFLUX steps). In some cases, however, it may also be advantageous to illuminate the sample only with the excitation light during pre-localization and only illuminate the sample with the deactivation light during the subsequent localization steps.

[0057] According to a further embodiment, the excitation light distribution is shifted relative to the sample using an electro-optical deflector, while the deactivation light distribution is shifted relative to the sample using a galvanometric scanner. According to a further embodiment, both the excitation light distribution and the deactivation light distribution are shifted relative to the sample using an electro-optical deflector, in particular together.

[0058] According to a further embodiment, the excitation light distribution is formed by irradiating an excitation light beam onto the sample along an optical axis and / or the deactivation light distribution is formed by irradiating a deactivation light beam onto the sample along an optical axis, in particular along the same optical axis.

[0059] According to a further embodiment, the deactivation light distribution is formed as a 2D donut. A 2D donut is understood to mean, in particular, a light distribution that has a (single) central intensity minimum, in particular a zero point of the intensity, wherein the intensity minimum has local intensity maxima adjacent to the intensity minimum in a plane perpendicular to the optical axis. In particular, a 2D donut can be formed in a plane conjugated to the objective pupil by phase modulation with a vortex pattern (from 0 to 2n · π in the circumferential direction relative to the optical axis, where n is a natural number), e.g., using a phase plate or an SLM. The phase-modulated light beam is, in particular, circularly polarized.

[0060] According to a further embodiment, the 2D donut is generated by phase modulation of the deactivation light with a first phase pattern, wherein the first phase pattern has a phase shift in a circumferential direction with respect to the optical axis, in particular continuously, from 0 to 2n · π rising phase, where n is a natural number which is greater than 1, where in particular n ≥ 5. This means that a vortex pattern is generated which ranges from phase values ​​of 0 to phase values ​​greater than 2π, namely 4π, 6 π , 8 πor 10π. This advantageously allows a wider region of low light intensity to be generated around the local minimum in the focal plane, which makes it easier to arrange the scanning positions in a region of the deactivation distribution in which the saturation intensity prevails. The excitation light distribution can be generated, for example, by phase modulation of the excitation light with a phase pattern that has a phase increasing in the circumferential direction, in particular continuously, from 0 to 2π. A vortex pattern with a phase increasing from 0 to 10π or more has proven to be particularly advantageous, since this phase pattern creates a deactivation light distribution with a particularly wide region of low intensity.

[0061] According to a further embodiment, the deactivation light distribution is designed as a 3D donut. A 3D donut (also known as bottle beam ), has a (single) central intensity minimum, in particular a zero point of the intensity, which is surrounded by local intensity maxima in all three spatial directions. Thus, the 3D donut forms a minimum surrounded on all sides by intensity increases. The 3D donut can be generated, for example, by modulation with a first phase pattern that has concentric rings (or rings and circular disks) of constant phase, which have a phase difference of πA 3D donut-shaped deactivation light distribution is particularly advantageous for the method according to the invention, since it has a relatively wide range of low light intensity in the focal plane (in the xy direction), so that the effective emission PSF is little influenced by the deactivation light. In particular, the first phase pattern has a circular disk and a first ring arranged concentrically around the circular disk, a second ring arranged concentrically around the first ring, and a third ring arranged concentrically around the second ring, wherein the circular disk and the first ring, the first ring and the second ring, and the second ring and the third ring have a phase difference of π This advantageously results in an intensity distribution stretched in the z-direction.

[0062] According to a further embodiment, the deactivation light beam is focused into the sample by means of an objective lens, wherein a beam cross-section of the deactivation light beam is adjusted, in particular reduced, such that a pupil of the objective lens is illuminated from below, so that the deactivation light distribution is stretched in the direction of the optical axis, i.e., in the axial direction. The plane of the pupil is a Fourier plane of the focal plane in the sample. The axial stretching of the deactivation light distribution has the advantage that the area in which the deactivation light suppresses the background emission (in particular, background fluorescence) by other molecules in the vicinity of the molecule to be localized is expanded in the axial direction. The interference of the signal by the background fluorescence is particularly pronounced in the axial direction, so that the quality of the localization can be significantly improved by this embodiment.Even if the overall intensity of the deactivation light remains constant when stretching the deactivation light distribution, thereby reducing the light intensity at certain points, this lower intensity may still be sufficient to effectively suppress the background emission, so that the extension of the axial range has a positive effect despite this effect.

[0063] The beam cross-section of the deactivation light beam can be adjusted, for example, using an aperture stop, particularly an adjustable one, which specifically illuminates the lens from below. Alternatively, a stop can be placed in the beam path upstream of the beam-shaping device (especially the light modulator, e.g., the SLM) to narrow the deactivation light beam before it hits the beam-shaping device. Another option, which ensures no light is lost at a stop, is to narrow the beam cross-section using a zoom lens.

[0064] According to a further embodiment, the beam cross-section of the deactivation light beam is adapted by adapting an active surface of a beam-shaping device, in particular a light modulator (e.g., SLM). The active surface of the beam-shaping device is designed to impart a phase pattern, in particular an adjustable one, to a light beam that passes through the active surface or is reflected and / or diffracted at the active surface. The active surface has, in particular, a blaze grating for diffracting the deactivation light beam and a plurality of pixels, in particular liquid crystal-based pixels, wherein a phase value of the pixels is adjustable, such that an adjustable phase pattern is superimposed on the blaze grating in order to impose a phase distribution on the diffracted light beam.

[0065] According to a further embodiment, an orientation of the blaze grating in an outer region of the active surface of the beam-shaping device is adjusted in order to adjust the beam cross-section of the deactivation light beam, wherein in particular a diffraction order of the blaze grating is adjusted. The adjustment of the blaze grating leads in particular to light incident on the outer region being decoupled from a beam path so that it does not illuminate the sample. The outer region of the active surface encloses in particular an inner region of the active surface, on which in particular the first phase pattern is displayed in order to generate the deactivation light distribution. Thus, the beam cross-section of the deactivation light beam is reduced to the light that enters the beam path from the inner region of the active surface.

[0066] According to a further embodiment, a second phase pattern is generated on the active surface of the beam-shaping device, in particular in addition to the first phase pattern. The deactivation light distribution is stretched by phase modulating the deactivation light beam with the second phase pattern.

[0067] According to a further embodiment, the second phase pattern is designed as a ring extending in a circumferential direction to the optical axis, in particular wherein the ring has a plurality of segments, wherein adjacent segments each have a phase difference of πto each other. Such a phase pattern leads to a stretching of a light intensity distribution in the focus, which is generated by a first phase pattern arranged within the ring in a plane conjugate to the objective pupil. In particular, the second phase pattern is arranged concentrically around the first phase pattern, further in particular on the outer region of the active surface, wherein the first phase pattern is formed on the inner region of the active surface. For example, the first phase pattern can be a further ring arranged concentrically around a circular disk, wherein the further ring and the circular disk have a phase difference of πIn this way, for example, an elongated 3D donut can be generated if suitable area ratios are maintained. This is particularly well suited for the method according to the invention, since it has a relatively wide range of low light intensity in the focal plane (in the xy direction), so that the effective detection PSF is little influenced by the deactivation light, but covers an extended area in the axial direction (z direction) to effectively suppress background fluorescence. The excitation light distribution can, for example, be designed as a 3D donut ( bottle beam) be designed without additional axial stretching.

[0068] According to a further embodiment, both the first phase pattern and the second phase pattern have a circumferential gradient from 0 to n · 2 πincreasing phase values ​​(vortex pattern), wherein the first phase pattern is arranged on a circular disk and the second phase pattern is arranged on a ring arranged concentrically around the first phase pattern, and wherein the first phase pattern and the second phase pattern have a phase difference of a · π Such a phase pattern is described in patent application WO 2019 / 221622 A1. By adjusting the ratio between the outer radii of the circular disk (first phase pattern) and the ring (second phase pattern), the extension of the resulting light distribution in the direction of the optical axis and the width of the distribution can be flexibly adjusted.

[0069] According to a further embodiment, the first phase pattern comprises a circular disk and a first ring arranged concentrically around the circular disk, wherein the second phase pattern comprises a second ring arranged concentrically around the first ring and a third ring arranged concentrically around the second ring, and wherein the circular disk and the first ring, the first ring and the second ring as well as the second ring and the third ring each have a phase difference of π to each other. In this way, a 3D donut stretched in the z-direction can also be generated with suitable surface ratios.

[0070] According to a further embodiment, a further, third phase pattern is generated on the active surface of the beam-forming device, in particular in addition to the first and / or the second phase pattern, wherein the third phase pattern is in particular at least partially superimposed on the first phase pattern. The third phase pattern forms in particular a circular disk with a plurality of segments, wherein adjacent segments each have a phase difference of π to each other. This advantageously allows additional light (e.g., secondary maxima) to be generated in the focal plane in a peripheral region around the focal intensity distribution, thereby improving the suppression of background emission. The third phase pattern overlaps with the first phase pattern, particularly in a central partial region, or is arranged radially within the first phase pattern.

[0071] According to a further embodiment, the first phase pattern has the shape of a circular disk with a circumferentially vortex-shaped phase change from 0 to 2 n · π rising phase, wherein the third phase pattern is formed from a plurality of circular disk segments extending in the center of the circular disk over a partial area of ​​the circular disk, wherein the first phase pattern in the segments is given an additional phase jump of + π 2 or − π 2 In this way, in combination with a circularly polarizing element, a (possibly broadened) 2D donut with additional secondary maxima can be created.

[0072] According to a further embodiment, the third phase pattern has the shape of a circular disk with a plurality of segments, wherein adjacent segments have a phase difference of π from one another, wherein the first phase pattern has a first ring arranged concentrically around the third phase pattern and a second ring arranged concentrically around the first ring, and wherein the first ring and the second ring have a phase difference of π from one another. By maintaining suitable area ratios of the first and second rings and in particular of the segments, a 3D donut with additional secondary maxima can be generated in the focal plane in this way.

[0073] According to a further embodiment, the deactivation light beam is at least approximately formed as a Bessel beam. Such a light beam can be generated, for example, by focusing the deactivation light with an axicon or by phase modulating the deactivation light with a phase pattern, in particular an annular one. The generation of a hollow Bessel-like STED beam is known from the prior art (see, for example, P. Zhang, P. Goodwin, J. Werner: 'Fast, super resolution imaging via Bessel beam stimulated emission depletion microscopy', Opt. Expr. 22, 12398 (2014) ); W. Yu, Z. Ji, D. Dong, X. Yang, Y. Xiao, ).

[0074] A second aspect of the invention relates to a device, in particular a microscope, for determining the positions of spaced-apart molecules in one or more spatial directions in a sample, in particular according to a method according to the first aspect. The device comprises at least one light source for generating an excitation light beam and a deactivation light beam, in particular a STED light beam, and at least one beam-shaping device for forming an excitation light distribution from the excitation light beam and a deactivation light distribution from the deactivation light beam, wherein the excitation light distribution and the deactivation light distribution each have a local intensity minimum and adjacent intensity increase regions.Furthermore, the device comprises an optical arrangement for illuminating the sample with the excitation light distribution and the deactivation light distribution, at least one detector for detecting photons emitted by the molecule for different positionings of the excitation light distribution, and a computing unit for deriving the position of the molecule based on the photons detected for the different positionings of the excitation light distribution.

[0075] The device has, in particular, at least one beam deflection device which is designed to arrange the local minimum of the excitation light distribution successively at a plurality of scanning positions within a scanning area, wherein the device further has a control unit which is designed to adjust the light intensity of the deactivation light in a capture area assigned to the scanning area, in which the position of the molecule can be clearly derived from the scanning positions and the associated detected photons, to a maximum of the saturation intensity, in particular a maximum of 50% of the saturation intensity, further in particular a maximum of 45% of the saturation intensity, further in particular a maximum of 40% of the saturation intensity, further in particular a maximum of 35% of the saturation intensity, further in particular a maximum of 30% of the saturation intensity, further in particular a maximum of 25% of the saturation intensity,further in particular to a maximum of 20% of the saturation intensity, further in particular to a maximum of 15% of the saturation intensity, further in particular to a maximum of 10% of the saturation intensity, further in particular to a maximum of 5% of the saturation intensity, of the deactivation light. To this end, the control unit controls in particular the light source that generates the deactivation light or a separate device for adjusting the intensity of the deactivation light.

[0076] The device is particularly designed to form the excitation light distribution and the deactivation light distribution during the various positionings of the excitation light distribution such that an effective detection PSF, in particular a width of the effective detection PSF, is unaffected by the deactivation light distribution.

[0077] According to one embodiment, the at least one beam deflection device is designed to reposition the excitation light distribution relative to the deactivation light distribution, in particular wherein the at least one beam deflection device is designed to position the local minimum of the excitation light distribution multiple times differently within an environment of the local intensity minimum of the deactivation light distribution.In particular, the at least one beam deflection device comprises a scanning device, in particular a galvanometric scanner, which is designed to jointly reposition the excitation light distribution and the deactivation light distribution relative to the sample, wherein the at least one beam deflection device further comprises a lateral beam deflection device, in particular comprising at least one electro-optical or acousto-optical deflection unit, which is designed to reposition the excitation light distribution relative to the deactivation light distribution.

[0078] According to a further embodiment, the at least one beam deflection device is designed to reposition the excitation light distribution while the deactivation light distribution is left stationary, so that the local minimum of the excitation light distribution is positioned several times differently within an environment of the local intensity minimum of the deactivation light distribution.

[0079] According to a further embodiment, the at least one beam deflection device is designed to jointly reposition the excitation light distribution and the deactivation light distribution.

[0080] According to a further embodiment, the at least one beam-shaping device is designed to form a region between two local maxima of the deactivation light distribution adjacent to the local intensity minimum of the deactivation light distribution, at least in one spatial direction, further, in particular at least 10% further, further, in particular at least 20% further, extended than a corresponding region between two local maxima of the excitation light distribution adjacent to the local minimum of the excitation light distribution.

[0081] According to a further embodiment, the device is designed to carry out a plurality of scanning steps, wherein the at least one beam deflection device is designed to position the local minimum of the excitation light distribution in each of the scanning steps at a plurality of scanning positions of a respective scanning area, wherein the respective scanning area of ​​each scanning step has a smaller area or a smaller volume than the scanning areas of the preceding scanning steps, and wherein the at least one beam shaping device orthe control unit is designed to adapt the deactivation light distribution in at least some of the scanning steps as a function of the area or volume of the respective scanning region, wherein the at least one beam shaping device is designed to adapt an overall intensity and / or a shape of the deactivation light distribution as a function of the area or volume of the respective scanning region.

[0082] According to a further embodiment, the at least one beam-shaping device or the control unit is designed to keep the deactivation light distribution constant during the repositioning of the excitation light distribution.

[0083] According to a further embodiment, the device is designed to determine the position of at least one further molecule in the sample after deriving the position of the molecule, wherein the excitation light source and the at least one beam shaping device are designed to illuminate the further molecule with the excitation light distribution, and wherein the detector is designed to detect photons emitted by the further molecule for different positionings of the excitation light distribution, and wherein the computing unit is designed to derive the position of the further molecule on the basis of the photons detected for the different positionings of the excitation light distribution, wherein the at least one beam deflection device is designed toBefore determining the position of the further molecule, the local minimum of the excitation light distribution and the local minimum of the deactivation light distribution are positioned at an expected position of the further molecule.

[0084] According to a further embodiment, the device comprises a first beam-shaping device, in particular a first light modulator or a first phase plate, for generating the excitation light distribution and a second beam-shaping device, in particular a second light modulator or a second phase plate, for generating the deactivation light distribution independently of the generation of the excitation light distribution. Alternatively, the light distributions can also be generated, for example, by two spatially separated regions of a single light modulator.

[0085] According to a further embodiment, the beam-shaping device has an active surface for generating a phase pattern, wherein the beam-shaping device is designed in particular as a light modulator or phase plate.

[0086] According to a further embodiment, the at least one beam-shaping device is designed to form the deactivation light distribution as a 2D donut.

[0087] According to a further embodiment, the at least one beam-shaping device is designed to generate the first phase pattern, the second phase pattern and / or the third phase pattern on its active surface.

[0088] According to a further embodiment, the first phase pattern has a phase shift in a circumferential direction with respect to the optical axis, in particular continuously, from 0 to 2 n · π rising phase, especially where nis a natural number which is greater than 1. In this case, the device has in particular a circularly polarizing element, in particular a λ / 4 plate, in a beam path between the beam-shaping device and the objective.

[0089] According to a further embodiment, the device comprises a control unit, wherein the control unit is designed to control the beam-shaping device such that the first phase pattern is modified, in particular such that a form of the deactivation light distribution is adapted as a function of the area or volume of the respective scanning region, in particular in the scanning steps.

[0090] According to a further embodiment, the at least one beam-shaping device is designed to form the deactivation light distribution as a 3D donut.

[0091] According to a further embodiment, the at least one beam-shaping device is designed to generate a first phase pattern on its active surface, which has concentric rings of constant phase, which have a phase difference of π with respect to one another.

[0092] According to a further embodiment, the device has an objective for focusing the deactivation light beam (and in particular also the excitation light beam) into the sample, wherein the device further has an adjustable aperture stop in a beam path between the deactivation light source and the beam shaping device or a zoom optics for reducing the beam cross-section of the deactivation light beam, wherein the device has a control unit which is designed to control the aperture stop or the zoom optics such that a beam cross-section of the deactivation light beam is adapted such that a pupil of the objective is illuminated from below, so that the deactivation light distribution is stretched in the direction of the optical axis (i.e. in the axial direction).

[0093] According to a further embodiment, the device comprises an optical unit configured to form the deactivation light beam at least approximately as a Bessel beam. In particular, the optical unit comprises an axicon or an SLM for forming or beam-shaping the deactivation light beam as a Bessel beam.

[0094] According to a further embodiment, the control unit is designed to adapt the active surface of the beam-shaping device, in particular of the light modulator, such that the beam cross section of the deactivation light beam is adapted, in particular reduced, by adapting the active surface.

[0095] According to a further embodiment, the control unit is designed to adapt an orientation of a blaze grating in an outer region of the active surface of the beam-shaping device, so that the beam cross section of the deactivation light beam is adapted, in particular reduced.

[0096] Advantageous developments of the invention emerge from the patent claims, the description, and the drawings. The claims are not to be understood as meaning that only those objects, devices, or methods that each have all or none of the features of a subclaim in addition to the features of independent claims 1 and 13 can be possible developments of the invention. Rather, further developments can result from features mentioned in the description and inferred from the drawings, which can be effective individually or cumulatively. KURZBESCHREIBUNG DER FIGUREN

[0097] In the following, the invention is further explained and described with reference to preferred embodiments shown in the figures. Fig. 1 shows schematically the sequence of the method according to the invention; Fig. 2 schematically shows several steps of a MINFLUX localization procedure; Fig. 3 shows an apparatus for determining positions of a molecule according to a first embodiment of the present invention; Fig. 4 shows an apparatus for determining positions of a molecule according to a second embodiment of the present invention; Fig. 5 shows a beam-shaping device as part of an inventive device according to an embodiment of the invention; Fig. 6A-D show different phase patterns for generating deactivation light distributions and corresponding contour line diagrams generated by simulations; Fig. 7A-D shows further phase patterns for generating deactivation light distributions and corresponding simulated contour line diagrams; Fig. 8A-D show plots of an excitation light distribution and various deactivation light distributions along a direction in the focal plane; Fig. 9A-D show another phase pattern for generating a deactivation light distribution as well as various plots of the deactivation light distribution generated by the phase pattern and an excitation light distribution; Fig. 10A-E show another phase pattern for generating a deactivation light distribution as well as various plots of the deactivation light distribution generated by the phase pattern compared to a bottle beam distribution. FIGURENBESCHREIBUNG

[0098] Fig. 1 illustrates the flow of the method according to the invention as a block diagram. The method is, in particular, a MINFLUX localization with additional deactivation (e.g., STED) light.

[0099] In step 101, an excitation light distribution 100 and a deactivation light distribution 300, in particular a STED light distribution, are generated. Both light distributions have a local intensity minimum 110, 310 and adjacent intensity increase regions 120, 320 (see also Fig. 8 ). Around the respective local intensity minimum 110, 310, the light distributions each have a region of relatively low light intensity. This region of the deactivation light distribution 300 is in particular wider than the corresponding region of the excitation light distribution 100. The deactivation light distribution 300 in particular forms a shell around a region of the excitation light distribution 100.

[0100] According to step 102, the sample 20 is illuminated with the excitation light distribution 100 and the deactivation light distribution 300, so that a molecule M to be localized in the sample 20 is excited by the excitation light and emits photons due to it, while the deactivation light suppresses the emission of background photons from other molecules M in the sample 20.

[0101] In step 103, photons emitted by the molecule are detected for different positionings of the excitation light distribution 100 with a detector, e.g., a point detector arranged confocally to the focus.

[0102] Finally, in step 104, the position of the molecule M is determined based on the photons detected for the various positionings of the excitation light distribution 100, e.g., by means of a maximum likelihood estimator. The method is carried out in particular such that the excitation light distribution 100 and the deactivation light distribution 300 are configured during the various positionings of the excitation light distribution 100 such that an effective detection PSF, in particular a width of the effective emission point spread function, is unaffected by the deactivation light distribution 300, i.e., is only influenced by the excitation light distribution. This improves the localization while simultaneously suppressing the background emission. Of course, the representation of the Fig. 1 as a block diagram does not mean that the individual steps of the process must be carried out one after the other.

[0103] The Fig. 2 shows an embodiment of the method according to the invention, which is a MINFLUX localization, which is carried out in particular in several iterations, wherein the minimum 110 of the excitation light distribution 100 is arranged in each iteration at a plurality of scanning positions 201 in a scanning range 200 around a previously determined estimated position of the molecule M. The scanning range 200 is defined as a circle with a radius L around an estimated position of the molecule to be localized. The position estimate selected in the initial step as the center of the scanning range 200 can be determined using an independent method (e.g. PALM / STORM microscopy, confocal scan with Gaussian excitation light distribution or pinhole orbit scan). Thereafter, in particular, the position estimate determined in the previous MINFLUX iteration is used as the center of the circle.

[0104] The sampling positions 201 of an iteration form a so-called set of target coordinates (engl. set of targeted coordinates, STC, or target coordinate pattern, TCP). The individual scanning positions 201 can, in particular, lie on a circle with radius L around the estimated position of the molecule M, wherein the scanning positions 201 are, in particular, arranged symmetrically around the center point. Furthermore, in particular, the estimated position of the molecule M is also used as the scanning position 201. Analogous to the example shown in two dimensions, the sample 20 can be scanned in three dimensions with scanning positions 201 that lie on or within a sphere with radius L in order to perform a three-dimensional localization of the molecule M (3D-MINFLUX). To shift the excitation light distribution 100 (and, in particular, also the deactivation light distribution 300) in the z-direction (along the optical axis), a deformable mirror, for example, can be used.In particular, within the scanning area 200 there is a capture area 210, wherein the molecule M can be clearly localized by means of the method according to the invention, provided that it is located within the capture area 210.

[0105] As in Fig. 2 As shown, the radius L can become smaller in each step, so that the accuracy of the position estimate also increases in each step, at least as long as the molecule M is located in the correspondingly reduced scanning area 200. The method is typically carried out until the molecule M loses its emissivity (e.g. due to bleaching or a transition to a dark state in the case of fluorophores) or, if possible, until the accuracy of the position estimate approaches a limit value resulting from the signal-to-noise ratio. The entire process is then repeated, in particular, with several additional molecules M. From the multiple localizations, a high-resolution image of the sample with the several localized molecules M can then be calculated. The described reduction in the scanning area 200 also results in a smaller capture area 210 than in the previous iteration step.

[0106] In Fig. 2 Furthermore, the deactivation light distribution 300 is schematically illustrated as a circle arranged concentrically around the scanning area 200. At the position marked by the circle, the intensity of the deactivation light can correspond, for example, to the saturation intensity or a lower value.

[0107] The deactivation light distribution 300 can, as in Fig. 2 As shown, the excitation light distribution 100 remains stationary at least during one MINFLUX iteration, and in particular also across multiple MINFLUX iterations, while the excitation light distribution 100 is moved within a region of lower deactivation light intensity around the local minimum 310 of the deactivation light distribution 300 in order to control the scanning positions 201. This has the advantage that fast beam deflection devices such as EODs only need to be provided for the excitation light beam, which reduces the complexity of the device and enables a more compact design. The deactivation light distribution 300 is then arranged, in particular at the beginning of each new localization of a molecule M, in the center of a projection of a detection pinhole into the focal plane in a corresponding sample area, in particular with a galvanometric scanner (slower than the EODs but present anyway).In particular, the deactivation light distribution 300 can be adjusted between the iteration steps such that an area in which, for example, the saturation intensity of the deactivation light or a lower light intensity prevails becomes smaller as the scanning area 200 is reduced in each iteration step (see . Fig. 2 This can be achieved in particular by controlling the overall intensity of the deactivation light or by different beam shaping (e.g., different phase patterns on an SLM).

[0108] Alternatively, the excitation light distribution 100 can also be moved together with the deactivation light distribution 300, so that in each iteration, the local minima of the two light distributions are arranged together one after the other at the scanning positions 201. Due to the required scanning speed, this is carried out in particular by means of a joint beam deflection with electro-optical or acousto-optical deflectors.

[0109] In the Figuren 3 und 4 Embodiments of a device 1 according to the invention for determining positions of a molecule M in the form of a MINFLUX microscope with an additional deactivation (in particular STED) light source are shown. The basic principle of the device is first explained with reference to the Fig. 3 Subsequently, differences in the embodiment according to the Fig. 4 compared to the embodiment according to Fig. 3 received.

[0110] The Fig. 3 shows a device 1 with a first light source 11, in particular a laser source, for generating a coherent excitation light beam A and a second light source 12, in particular a laser source, for generating a coherent deactivation light beam D. The light from the first light source 11 has a suitable wavelength for exciting the molecules M in the sample, and the second light source 12 generates light of a wavelength that is suitable for deactivating the molecules M, so that the molecules can no longer emit photons when exposed to the excitation light (e.g. depletion by stimulated emission, i.e. STED). The wavelengths of the excitation light and the deactivation light differ in particular from one another. As an alternative to two separate light sources 11, 12, the device 1 can also have just one light source, in particular a laser source, which generates both the excitation light and the deactivation light.For this purpose, the light source can, for example, generate a continuum of wavelengths from which excitation light and deactivation light of suitable wavelengths can then be obtained, for example by means of optical filters.

[0111] In the Fig. 3 In the example shown, the excitation light beam A and the deactivation light beam D are coupled into a common beam path at a first, in particular dichroic, beam splitter 13a. The light beams then pass together through a lateral beam deflection device 15, which consists of two beam deflection units 15a, 15b arranged in series in the beam path. The beam deflection units 15a, 15b can be, for example, electro-optical deflectors, wherein the first beam deflection unit 15a is designed to deflect the light beams passing through it in a first direction x perpendicular to the optical axis OA, and wherein the second beam deflection unit 15b is designed to deflect the light beams passing through it in a second direction y perpendicular to the optical axis OA and perpendicular to the first direction.The lateral beam deflection device 15 has, in a previously known manner, in particular further optical components, in particular lenses for focusing the light beams, which are not shown here for the sake of clarity.

[0112] At a second, in particular dichroic, beam splitter 13b, the excitation light beam A and the deactivation light beam D are then split into two parallel beam paths. The excitation light beam A strikes a first beam-shaping device 16a, in particular a first light modulator, further in particular a phase-modulating spatial light modulator SLM with switchable pixels. By diffracting the excitation light beam A at a blaze grating on an active surface of the light modulator, a phase pattern adjusted by the switchable pixels is imprinted on the excitation light beam A, resulting in the excitation light distribution 100 with the local minimum 110 in the focal plane of the sample 20 through interference. The phase-modulated excitation light beam A is recoupled into the main beam path via a first mirror 14a and a second mirror 14b.

[0113] In an analogous manner, the deactivation light beam D is directed via a third mirror 14c onto a second beam-shaping device 16b, in particular a second light modulator, which imparts a preset phase pattern thereto, which may differ in particular from the phase pattern of the first beam-shaping device 16a. The phase-modulated deactivation light beam D is then deflected via a fourth mirror 14d and then combined with the phase-modulated excitation light beam A via a third, in particular dichroic, beam splitter 13c. Of course, a configuration is also possible in which the deactivation light beam D is coupled into the main beam path via mirrors and the excitation light beam A is combined with the deactivation light beam D via a beam splitter.

[0114] As an alternative to the first light modulator 16a and the second light modulator 16b, phase plates with fixed phase patterns can also be used, for example, through which the respective light beams typically pass (instead of diffraction at a grating on the active surface of the light modulator) in order to impart the respective phase patterns to them. In addition, optical elements which influence the polarization direction of the respective light beams, e.g., λ / 2 plates (not shown), can be arranged in front of the beam-shaping devices 16a, 16b. Furthermore, particularly when imposing a vortex phase pattern to generate a 2D donut, a circularly polarizing element (e.g., a λ / 4 plate) can be provided in the beam path behind the beam-shaping devices 16a, 16b in order to achieve the desired light distribution in the focus (not shown).

[0115] The beam-shaping devices 16a, 16b are each arranged in a plane perpendicular to the optical axis OA, which is conjugated to a pupil of the objective 19. Particularly in the case of light modulators, a slight deviation from this position can also be corrected by a corresponding adjustment of the phase pattern.

[0116] By means of the beam shaping devices 16a, 16b, different phase patterns are imposed on the excitation light beam A and the deactivation light beam D, in particular so that in the focal plane for the deactivation light distribution 300, a wider region of low intensity (e.g., below the saturation intensity) around the local minimum 310 results than for the excitation distribution 100. For example, for a 2D MINFLUX localization, the excitation light distribution 100 could be designed as a 2D donut, which is generated by a vortex phase distribution between 0 and 2π, while the deactivation light distribution is designed as a wider 2D donut, which is generated by a vortex phase distribution between 0 and 4π, 6π, 8π, or 10π. Likewise, for example,It is possible that for a 3D MINFLUX localization an excitation light distribution 100 and a deactivation light distribution 300 are used, each in the form of a 3D donut, wherein the 3D donut of the deactivation light is additionally axially stretched.

[0117] An axial beam deflection device 17, e.g., comprising a deformable mirror, is also located in the common beam path of the excitation light beam A and the deactivation light beam D. Using the axial beam deflection device 17, the excitation light beam A and the deactivation light beam D can be shifted together in the z-direction, i.e., parallel to the optical axis OA, e.g., to control scanning positions 201 in the sample in the z-direction for 3D localization of the molecule M.

[0118] Finally, a scanning device 18, e.g., a galvanometric scanner, is arranged in the beam path to scan the excitation light beam A and the deactivation light beam D together across the sample 20. The excitation light beam A and the deactivation light beam D are focused by an objective 19 with a pupil 19a into a sample 20 containing spaced-apart molecules M to be localized. At least the objective 19 and the scanning device 18 form, in particular, an optical arrangement 2 for illuminating the sample 20 with the excitation light distribution 100 and the deactivation light distribution 300. The excitation light beam A and the deactivation light beam D enter the objective 19 along an optical axis.

[0119] Due to the excitation with the excitation light, the molecules emit photons (e.g., fluorescent light). The light E emitted by the sample 20 is descanned by the scanning device 18 and coupled into a detection beam path via a fourth, in particular dichroic, beam splitter 13d. Located in this detection beam path is, in particular, a pinhole 21, which is arranged confocally to the focus of the excitation and deactivation light beam in the sample 20. Behind the pinhole 21, a detector 22 is arranged, in the example shown here, a point detector, e.g., an avalanche photodiode or a photomultiplier, which detects and, in particular, counts the photons emitted by the molecule M.

[0120] Furthermore, the device 1 comprises a computing unit 23, which is in electrical connection or data connection with the detector 22 and is designed to determine the position of the molecule M from the photons detected by the detector 22 and the associated positions of the local minimum 110 of the excitation light distribution 100, e.g., by means of a maximum likelihood estimator implemented in software or hardware on the computing unit 23. In order to obtain the positions of the excitation light distribution 100 associated with the detected photon numbers or photon count rates, the computing unit 23 receives, in particular, a signal from the control unit 24 or, e.g., directly from the lateral beam deflection device 15, the axial beam deflection device 17, and / or the scanning device 18.

[0121] The control unit 24 is further designed to control the lateral beam deflection device 15, the axial beam deflection device 17, the scanning device 18 and the beam shaping devices 16a and 16b.

[0122] With the Fig. 3 In the device 1 shown, the method according to the invention is carried out in particular such that the excitation light beam A and the deactivation light beam D are moved together in a plurality of MINFLUX iterations by means of the lateral beam deflection device 15 (and optionally also by means of the axial beam deflection device 17) relative to the sample 20. In this case, the scanning positions 201 (see Fig. 2 ) with the local minimum 110 of the excitation light distribution 100, and the photons emitted by the molecule M are detected for each sampling position 201 by means of the detector 22. Subsequently, for each MINFLUX iteration, an estimated position of the molecule M is determined by means of the computing unit 23 from the detected photons and the associated positions of the excitation light distribution 100.

[0123] Fig. 4 shows a device 1 according to the invention, which differs from the device shown in Fig. 3 shown device 1 in that the lateral beam deflection device 15 and the axial beam deflection device 17 are arranged in a partial beam path, via which the excitation light beam A is directed onto the sample 20. The deactivation light beam D, however, is guided on a parallel partial beam path without corresponding beam deflection devices. The light beams A, D are split into the partial beam paths by means of the second, in particular dichroic, beam splitter 13b and recombined via the third, in particular dichroic, beam splitter 13c. In the corresponding partial beam paths are also the Fig. 3 described beam shaping devices 16a,16b.

[0124] When using the Fig. 4 In the illustrated embodiment of the device 1, in particular before the localization of a molecule M, the excitation light beam A and the deactivation light beam D are directed jointly by means of the scanning device 18 onto a region of the sample 20 in which the molecule M to be localized is suspected according to a pre-localization. The excitation light beam A is then moved over the sample by means of the lateral beam deflection device 15 and optionally also by means of the axial beam deflection device 17 in order to target the scanning positions 201 with the local minimum 110 of the excitation light distribution 100. During this time, the deactivation light distribution 300 remains, in particular, stationary.

[0125] The Fig. 5 shows a schematic representation of a second beam-shaping device 16b for phase modulation of the deactivation light beam D in cross-section. The second beam-shaping device has an active surface 160 with a blaze grating, which diffracts the incident deactivation light beam D in a desired order and thus couples it into the further beam path, so that the deactivation light beam is focused by the objective 19 into the sample 20. Furthermore, a phase pattern with the blaze grating can be superimposed on the active surface, so that a corresponding phase distribution is impressed on the deactivation light beam D, which forms the deactivation light distribution 300 at the focus through interference. The active surface 160 is, in particular, a plane conjugated to the pupil 19a of the objective 19 (see Fig. 3-4 ). The phase pattern can be generated, in particular, in the form of controllable pixels, whereby a phase value can be specified for each pixel. The phase pattern and, in particular, the orientation of the blazed grating can be controlled by means of the control unit 24.

[0126] In the Fig. 5 Furthermore, an outer region 161 of the active surface 160 and an inner region 162 of the active surface 160 enclosed by the outer region 161 are schematically shown.

[0127] By means of the control unit 24, the orientation of the blaze grating in the outer region 161 can be changed such that the deactivation light beam D is not coupled into the beam path, but rather, for example, into a beam trap (not shown). In this way, the beam cross-section of the deactivation light beam D can be reduced such that the pupil 19a of the objective 19 is illuminated from below, thus stretching the deactivation light distribution 300 in the direction of the optical axis OA (i.e., axially).

[0128] In another embodiment, the first phase pattern for generating the deactivation light distribution 300 can be displayed in the inner region 162 and a second phase pattern can be displayed in the outer region 161, which effectively leads to a reduction in the beam cross section of the deactivation light beam D and thus to an axial stretching of the deactivation light distribution 300.

[0129] In the Figuren 6 bis 7 und 9 bis 10 are phase patterns ( Fig. 6A-D , above) are shown, which can be generated, for example, on an active surface 160 of a beam-shaping device 16a, 16b, in order to modulate the phase distribution of the excitation light beam A or the deactivation light beam D, in particular the deactivation light beam D, so that a corresponding light intensity distribution with a local minimum is created at the focus of the respective light beam in the sample 20 by interference. Below the phase patterns, contour line diagrams of corresponding simulated light distributions are shown, each as a section through the focal plane (xy plane, perpendicular to the optical axis, where the z coordinate is zero, Fig. 6A-D , center) and as a section in the yz-plane, where the x-coordinate is zero ( Fig. 6A-D , below). The location coordinates in Fig. 6A-D Middle and bottom are each given in micrometers. The phase patterns are generated in a plane conjugate to the plane of the pupil 19a of the objective lens 19.

[0130] Fig. 6A (above) shows a first phase pattern 610 consisting of a first ring 602, which is arranged concentrically around a circular disk 601, wherein the first ring 602 and the circular disk 601 have a phase difference of π to each other. With such a phase pattern, a 3D donut (also known as bottle beam denoted) with a central local minimum 310.

[0131] In the Figuren 6B bis 6D (above) also shows a first phase pattern 610, which has a first ring 602 arranged concentrically around a circular disk 601, wherein the circular disk 601 and the first ring 602 have a phase jump of π. However, the radius of the first ring 602 and the circular disk 601 is each smaller than in the first phase pattern 610 according to Fig. 6A (above). In addition, Fig. 6B-6D Another second phase pattern 611 is shown, which has a second ring 603 arranged concentrically around the first ring 602 of the first phase pattern 610. This second ring 603 has a plurality of segments 604 that have alternating phase values ​​that correspond to the phase value of the first ring 602 and the phase value of the circular disk 601. Accordingly, the adjacent segments 604 on the second ring 603 have phase jumps of π relative to one another. In the example shown here, all segments 604 have identical dimensions and are thus arranged symmetrically around the first ring 602.

[0132] As can be seen from the comparison between the contour line diagrams in Fig. 6A (below) and Fig. 6B (below), the additional second phase pattern 611 with the segmented second ring 603 leads to an extension of the 3D donut in the axial direction (z-direction). In contrast, there is only a slight change in the light intensity distribution in the focal plane ( Fig. 6A und 6B , center). The stretching of a 3D donut-shaped deactivation light distribution 300 in the z-direction is advantageous for the method according to the invention, since it allows the background emission from regions located above and below the focus in the z-direction to be effectively suppressed.

[0133] Furthermore, from the Figuren 6C und 6D It can be seen that by widening the second ring 603 in the radial direction inwards and simultaneously reducing the size of the circular disk 601, an even greater extension in the z-direction can be achieved than for the Fig. 6A In the case shown in Fig. 6D In the configuration shown, a broadening of the central local minimum 310 in the focal plane is also visible ( Fig. 6D , middle), which can also have a beneficial effect on the suppression of background fluorescence.

[0134] In the Figuren 7A und 7B Further phase patterns for phase modulation of the deactivation light beam are shown. Fig. 7C-D show corresponding contour line diagrams of the light intensity distribution obtained at the focus in the XZ plane, where Fig. 7C the in Fig. 7A shown phase pattern and Fig. 7D the in Fig. 7B shown phase pattern. Fig. 7A shows as well as Fig. 6A a first phase pattern consisting of a circular disk 601 and a first ring 602 arranged concentrically around the circular disk 601 with a phase difference of π Accordingly, Fig. 7C the contour line diagram of a 3D donut.

[0135] According to Fig. 7B In addition to the circular disk 601 and the first ring 602, a second ring 603 and a third ring 605 are provided, with adjacent rings each having a phase difference of π to each other. This advantageously results in an intensity distribution at the focus that is stretched in the z-direction ( Fig. 7D ). The scales are given in µm.

[0136] Fig. 8 A-D show plots along a direction in the focal plane (e.g., along the x-coordinate) of a simulated 2D donut-shaped excitation light distribution 100 with a wavelength of 642 nm, each overlaid with simulated 2D donut-shaped deactivation light distributions 300 of different shapes and widths with a wavelength of 775 nm. The x-axis scale is in micrometers. The excitation light distribution 100 has a central local minimum 110, and the deactivation light distributions 300 each have at least one local minimum 310. Adjacent to the local minima 110, 310, the distributions 100, 300 have intensity increase regions 120, 320 and at least two local maxima 130, 330 each.The excitation light distribution 100 was generated by phase modulating the excitation light beam A with a vortex-shaped phase pattern in a plane conjugate to the objective pupil, which has phase values ​​gradually increasing clockwise from 0 to 2π in a circumferential direction relative to the optical axis. Additionally, the excitation light beam A was circularly polarized to the left, resulting in the excitation light distribution 100 in the focal plane through interference.

[0137] The Fig. 8A The deactivation light distribution 300 shown was generated by phase modulation of the deactivation light beam 100 with the same vortex-shaped phase pattern, also with left-hand circular polarization. The local minima 110, 310 of the distributions 100, 300 are located at the same position here. The deactivation light distribution 300 according to Fig. 8A is slightly wider than the excitation light distribution 100 due to the higher wavelength.

[0138] The Figuren 8B bis 8D In contrast, the deactivation light distributions 300 shown were generated with vortex-shaped phase patterns which have gradually increasing first phase patterns between 0 and 4π ( Fig. 8B ), 0 and 6 π ( Fig. 8C ) or 0 and 10π ( Fig. 8D ). As a result, with increasing maximum phase value, the deactivation light distributions 300 have an increasing width, i.e., a greater distance between the local maxima 130, 330 adjacent to the local minimum 110, 310, and an increasingly broader region of relatively low light intensity between the local maxima 130, 330. This region is particularly broad for a maximum of 10π. Advantageously, the deactivation light of these distributions reduces the background fluorescence but does not significantly affect the detection PSF of the emission light.

[0139] The Fig. 8B The deactivation light distribution 300 shown has a further local maximum 330 at the position of the minimum of the excitation light distribution 100. This effect results from the polarization direction of the light in this special phase pattern (vortex pattern from 0 to 4π).

[0140] In Fig. 9A is a first phase pattern 610 for generating the Fig. 9B und 9D The deactivation light distribution 300 shown is superimposed with a third phase pattern 612 for generating secondary maxima. The first phase pattern 610 is vortex-shaped with a phase increasing in the circumferential direction from 0 to 10π. This pattern is superimposed with a third phase pattern 612 in the form of a circular disk 601 divided into segments 604, wherein the segments 604 have a phase difference of + π 2 or − π 2 The outer region of the superimposed phase pattern, on which the multiple vortex-shaped phase progression is best visible, forms a first ring 602, which is arranged concentrically around the circular disk 601. The segments 604 with positive and negative phase jumps to the first ring 602 are arranged alternately. Accordingly, adjacent segments 604 in the circumferential direction each have a phase difference of π on.

[0141] With the phase pattern shown, a 2D donut-shaped deactivation light distribution 300 with a wide range of low intensity around the central local minimum 310 (see Fig. 8D ) and additional secondary maxima of the light intensity. These secondary maxima advantageously suppress background emission in areas farther from the focus.

[0142] Fig. 9B shows a plot of the deactivation light distribution 300 along the x-coordinate. In Fig. 9D a 2D plot of the deactivation light distribution 300 in the xy plane is shown and the Fig. 9C shows for comparison an xy-plot of a 2D donut in Fig. 8A shown deactivation light distribution 300.

[0143] In Fig. 10A A further first phase pattern 610 for phase modulation of the deactivation light is shown, which is superimposed with a second phase pattern 611 for stretching the deactivation light distribution 300 in the z-direction and with a third phase pattern 612 for generating secondary maxima. The first phase pattern 610 has a first ring 602 and a second ring 603 arranged concentrically around the first ring 602, wherein the first ring 602 and the second ring 603 have a phase difference of πso that a 3D donut is formed at the focus. The second phase pattern 611 is formed by a third ring 605 arranged concentrically around the second ring 603, which is divided into alternating segments 604, wherein the segments 604 have alternating phase differences of 0 or 1 relative to the second ring 603. π The third phase pattern 612 forms a circular disk 601 arranged concentrically within the first ring 602, which is in turn divided into segments 604, which have alternating phase differences of 0 or π have.

[0144] Fig. 10B und 10C show a corresponding by the in Fig. 10A shown phase pattern generated deactivation light distribution 300 in focus, where Fig. 10B an xy-section (focal plane) and Fig. 10C represents an xz-section. In the Figuren 10D und 10E are for comparison sections through a regular 3D donut, ie, bottle beam (see phase pattern Fig. 6A and 7A ). All scales are in µm.

[0145] As is particularly evident from Fig. 10B As can be seen from the third phase pattern 612, secondary maxima are created, which have a beneficial effect on the suppression of the background emission by the deactivation light. In addition, the deactivation light distribution 300 is stretched in the z-direction by the second phase pattern 611, as can be seen from Fig. 10C is evident. BEZUGSZEICHENLISTE

[0146] 1Device for determining the position of a molecule 2Optical arrangement 11Excitation light source 12Deactivation light source 13aFirst beam splitter 13bSecond beam splitter 13cThird beam splitter 13dFourth beam splitter 14aFirst mirror 14bSecond mirror 14cThird mirror 14dFourth mirror 15Lateral beam deflection device 15aFirst beam deflection unit 15bSecond beam deflection unit 16aFirst beam shaping device 16bSecond beam shaping device 17Axial beam deflection device 18Scanning device 19Objective lens 19aPupil 20Sample 21Pinhole 22Detector 23Calculation unit 24Control unit 100Excitation light distribution 101Generating a plurality of light distributions 102Illumination with the excitation light distribution 103Detecting emitted photons 104Deducing the position of the molecule 110Local minimum of the excitation light distribution 120Intensity increase region of the excitation light distribution 130Local maximum of the excitation light distribution 160Active surface 161Outer region162Inner area 200Scanning area 201Scanning position 210Capture area 300Deactivation light distribution 310Local minimum of the deactivation light distribution 320Intensity increase area of ​​the deactivation light distribution 330Local maximum of the deactivation light distribution 601Circular disk 602First ring 603Second ring 604Segment 605Third ring 610First phase pattern 611Second phase pattern 612Third phase pattern AExcitation light beam DDeactivation light beam EEmitted light LRadius MMolecule OAptical axis

Claims

1. A method for determining positions of molecules (M) spaced apart from one another in one or more spatial directions in a sample (20) comprising the steps of: - generating (101) a plurality of light distributions, each light distribution comprising a local intensity minimum (110, 310) and intensity increasing regions (120, 320) adjacent thereto, the light distributions comprising an excitation light distribution (100) and a deactivation light distribution (300), particularly a STED light distribution, - illuminating (102) the sample (20) with the excitation light distribution (100) and the deactivation light distribution (300), - detecting (103) photons emitted by the molecule (M) for different positionings of the excitation light distribution (100), and - deriving (104) the position of the molecule (M) based on the photons detected for the different positionings of the excitation light distribution (100), wherein the local minimum (110) of the excitation light distribution (100) is successively arranged at a plurality of scanning positions (201) within a scanning region (200), characterized in that the light intensity of the deactivation light in a catch region (210) assigned to the scanning region (200), in which the position of the molecule (M) can be unambiguously derived from the scanning positions (201) and the associated detected photons, corresponds at most to a saturation intensity.

2. The method according to claim 1, characterized in that the excitation light distribution (100) is repositioned while the deactivation light distribution (300) is left stationary so that the local minimum (110) of the excitation light distribution (100) is positioned differently more than once within a vicinity of the local intensity minimum (310) of the deactivation light distribution (300).

3. The method according to claim 1, characterized in that the excitation light distribution (100) and the deactivation light distribution (300) are repositioned together.

4. The method according to one of the preceding claims, characterized in that a region between two local maxima (330) of the deactivation light distribution (300) adjacent to the local intensity minimum of the deactivation light distribution (300) is extended further, particularly at least 10% further, at least in one spatial direction than a corresponding region between two local maxima (130) of the excitation light distribution (100) adjacent to the local minimum (110) of the excitation light distribution (100).

5. The method according to one of the preceding claims, characterized in that a plurality of scanning steps is performed, wherein the local minimum of the excitation light distribution (100) in each of the scanning steps is positioned at a plurality of the scanning positions (201) of a respective scanning region (200), wherein the respective scanning region (200) of each scanning step comprises a smaller area or volume than the scanning regions of the preceding scanning steps, and the deactivation light distribution (300) is adjusted in at least a part of the scanning steps depending on the area or volume of the respective scanning area (200), wherein a total intensity and / or a shape of the deactivation light distribution (300) is adjusted depending on the area or volume of the respective scanning area (200).

6. The method according to one of the preceding claims, characterized in that the deactivation light distribution (300) remains constant during repositioning of the excitation light distribution (100).

7. The method according to one of the preceding claims, characterized in that the deactivation light distribution (300) is formed as a 2D donut or a 3D donut.

8. The method according to claim 7, characterized in that the 2D donut is generated by phase modulation of the deactivation light with a first phase pattern (610), the first phase pattern (610) comprising a phase increasing in a circumferential direction with respect to an optical axis, particularly continuously, from 0 to 2π · n, wherein n is a natural number greater than 1.

9. The method according to one of the preceding claims, characterized in that the deactivation light distribution (300) is formed by irradiating a deactivation light beam (D) onto the sample (20) along an optical axis (OA), wherein the deactivation light beam (D) is focused into the sample (20) by means of an objective (19), and wherein a beam cross-section of the deactivation light beam (D) is adjusted such that a pupil (19a) of the objective lens (19) is under-illuminated so that the deactivation light distribution (300) is stretched in the direction of the optical axis (OA).

10. The method according to claim 9, characterized in that the beam cross-section of the deactivation light beam is adjusted by adapting an active surface (160) of a beam shaping device (16a, 16b), particularly a light modulator.

11. The method according to claim 10, characterized in that a. an orientation of a blazed grating in an outer region (161) of the active surface of the beam shaping device (16a, 16b) is adjusted, or b. a second phase pattern (611) is generated on the active surface of the beam shaping device (16a, 16b), particularly wherein the second phase pattern (611) is formed as a ring (603) extending in a circumferential direction to the optical axis (OA), particularly wherein the ring comprises a plurality of segments (604), wherein adjacent segments (604) respectively comprise a phase difference of π to each other.

12. The method according to claim 9 or 10, characterized in that the deactivation light beam (D) is formed at least approximately as a Bessel beam, particularly by means of an axicon or a light modulator.

13. A device (1), particularly a microscope, for determining positions of molecules (M) spaced apart from one another in one or more spatial directions in a sample (20), particularly according to a method according to one of the claims 1 to 12, comprising a. at least one light source (11,12) for generating an excitation light beam (A) and a deactivation light beam (D), particularly a STED beam, b. at least one beam shaping device (16a, 16b) for forming an excitation light distribution (100) from the excitation light beam (A) and a deactivation light distribution (300) from the deactivation light beam (D), wherein the excitation light distribution (100) and the deactivation light distribution (300) each comprise a local intensity minimum (110, 310) and intensity increasing regions (120, 320) adjacent thereto, c. an optical arrangement (2) for illuminating the sample (20) with the excitation light distribution (100) and the deactivation light distribution (300), d. at least one detector (22) for detecting photons emitted by the molecule (M) for different positionings of the excitation light distribution (100), and e. a computing unit (23) for deriving the position of the molecule (M) based on the photons detected for the different positionings of the excitation light distribution (100), wherein the device (1) comprises at least one beam deflection device (15a, 15b, 17) configured to successively arrange the local minimum of the excitation light distribution (100) at a plurality of scanning positions (201) within a scanning region (200), characterized in that the device comprises a control unit (24) configured to adjust the light intensity of the deactivation light in a catch region (210) assigned to the scanning region (200), in which the position of the molecule (M) can be unambiguously derived from the scanning positions (201) and the associated detected photons, to at most a saturation intensity.

14. The device (1) according to claim 13, characterized in that the at least one beam deflection device (15a, 15b, 17) is configured to reposition the excitation light distribution (100) relative to the deactivation light distribution (300) and to position the local minimum (110) of the excitation light distribution (100) differently more than once within a vicinity of the local intensity minimum (310) of the deactivation light distribution (300).

15. The device (1) according to claim 13 or 14, characterized in that the device (1) comprises a first beam shaping device (16a) for generating the excitation light distribution (100) and a second beam shaping device (16b) for generating the deactivation light distribution (300) independently of the generation of the excitation light distribution (100).