Methods to improve 3D printing using two-color polymerization

Abstract

A dual-color photoinitiator-mediated 3D printing process comprising the photopolymerization of at least one monomer at a desired spatial position on an object plane (1) in a liquid comprising the least one monomer and at least one photo initiator system capable of initiating photopolymerization of the at least one monomer by providing one or more first light beams and a second light beam having different wavelengths to the desired spatial position thereby activating the at least one photo initiator system and wherein the first and second light beams are movable relative to the desired spatial position characteriyed in that the first light beam is provided in form of a light sheet to the object plane (1).

Description

[0001] METHODS TO IMPROVE 3D PRINTING USING TWO-COLOR POLYMERIZATION [1] The photoinitiator system in this invention is designed to enable precise spatial control during the additive manufacturing process. It achieves this by utilizing a dual-wavelength irradiation method, where the photoresist hardens only at the intersection of two light beams within a very short time interval. This system includes at least one photoinitiator that can transition into a ‘dormant state’ upon absorption of the first photon and then into an active state upon absorption of the second photon. The invention includes arrangements and methods to make printing by two-color polymerization more manageable, faster, and to improve resolution and printable volume. In addition, the improved methods are complemented with new applications. BACKGROUND [2] Dual-color photoinitiator-mediated printing is known from DE102018009916A1. The photoinitiator system absorbs a photon from the first light source (wavelength λ1), transitioning from its ground state to an excited state. This excited state is referred to as the activated but ‘dormant’ state, because it does not immediately lead to polymerization. The dormant state has a very short lifetime, typically less than 1 millisecond. During this time, the photoinitiator remains in the dormant state, but does not initiate the polymerization process. The photoinitiator system returns with an exponential decay into the non-activated state. If a second photon from the second light source (wavelength λ2) is absorbed within the dormant state’s lifetime, the photoinitiator system transitions to an active state. This active state generates radicals that initiate the polymerization of the monomer in the photoresist, leading to hardening at the specific location of dual irradiation. [3] The requirement for dual-wavelength absorption within a short time frame ensures that polymerization occurs only at the precise intersection of the two light beams. This spatial selectivity allows for high-resolution 3D structuring, with feature sizes potentially smaller than 1 micrometer. Photoinitiator systems containing α-Diketones or photochromic compounds are used. The printing of a 3D object is generated point by point or plane by plane. The quality of the printing result is determined by the quality of the structured intersection of the two light sources within the printing volume. This intersection is restructured and moved relative to the object to be printed, finally resulting in a polymerized 3D object which can be rinsed and optionally post cured. The lifetime of the dormant state limits the speed of the restructuring and movement of the intersection. A long dormant state lifetime causes a ‘smear’ of the printing result. The shape of the structured intersection is also compromised by the shape of the involved beam-waist, i.e. by the Abbe criterion and the Rayleigh length of focused beams. In case of a plane-by-plane printing, a light sheet-based activation of the dormant state is used. Light sheets are known to be limited in size at a given thickness. These limitations also restricts the relation between the smallest printable detail and the overall dimension of the objects to be printed. [4] Two-color photoinitiators form the basis for Xolography, a dual-color photoinitiator- mediated 3D printing process (referred to as DCPMP, https: / / www.xolo3d.com / ; Regehly, M. et al. Xolography for linear volumetric 3D printing. Nature 588, 620–624, 2020). This technology is used primarily for polymerization inside a cuvette to create 3D objects that are rinsed and post-processed after being taken out of the cuvette for polymerization (DE102021101164A1, WO2021089090A1). The lifetime of the ‘dormant state’ is increased in the Xolography process to a time frame of one to several seconds, utilizing ‘dormant states’ of the photoinitiator system, which are ground states. [5] In general, DCPMP processes require a photoinitiator, which is brought into an active state by a first radiation. By applying a second radiation, the activated photoinitiator initiates the polymerization of a monomer. In other words, polymerization (i.e., printing) occurs only at the location where the first and second radiation interact with the photoinitiator. The printing process is performed within a liquid, comprising the monomer(s) and the photoinitiator. Since the polymer / the printed object has a different density than the surrounding liquid, the printed object will start to move in the surrounding non-polymerized liquid, therefore changing position, which jeopardizes printing quality. In order to prevent floating of the printed object and maintain printing quality / resolution, only highly viscous fluids can be used. [6] The term ‘polymerization’ is used generally in this text to describe the process, which is used to generate a structure in 3D. It should include all types of structuring mediated by an activated molecule, which might be photoinitiation, photo-crosslinking, photocleaving used to bind or unbind molecules locally. It could be also applied to further 3D-functionalize already printed objects. 3D-printing also includes a ‘negative’ structuring where a crosslinked rigid or gel-like printing medium is rendered liquid by the printing process or processes following the dual color printing process, where chemical links are dissociated. For example cleavable crosslinkers may be part of the ‘polymerization system’. [7] The present invention discloses advantageous modifications of the utilized printer and the printing process and solves the compromises of the state of the art dual color photoinitiator printing process. The printing volume can be enlarged and the lifetime of the dormant state of the photoinitiator system can be increased, maintaining a high resolution printing result within a short time. [8] Object of the invention is a dual-color, photoinitiator-mediated 3D printing process comprising the photopolymerization of at least one monomer at a desired spatial position on an object plane (1) in a liquid comprising the least one monomer and at least one photo initiator capable of initiating photopolymerization of the at least one monomer by providing one or more first light beams and a second light beam having different wavelength at the desired spatial position thereby activating the at least one photo initiator thereby initiating photopolymerization of the at least one monomer, wherein the first and second light beams are movable relative to the desired spatial position characterized in that the first light beam is provided in form of a light sheet to the object plane (1). [9] Further embodiments of the invention are directed to a new method to generate and utilize a dynamic light sheet called TWI (Tilted waist illumination) for 3D printing. The dual- color photoinitiator-mediated 3D printing process can be used for 3D printing on silicon surfaces or for printing lost molds, both with a large printing area. The printing plane is then tilted with respect to the surfaces, in contrast to the known stereolithographic printers, where the printing is performed layer by layer and where the layers are parallel to a substrate, starting directly on the substrate. The tilt enables a continuous printing process, where stripes of a 3D- structure or multiple objects could be printed without a process interruption.

[0010] The term “object plane (1)” refers to the plane where the polymerization takes place, i.e. is the actual printing plane.

[0011] The term “photo initiator” refers not necessarily to a single compound acting as photo initiator, but may refer to a system or cascade of compounds which in the end initiate photopolymerization of the monomers.

[0012] Further embodiments of the invention are disclosed in the depended claims.

[0002] GLOSSARY

[0013] Labeling of figure items 1 Object plane (XY) 2 Excitation pupil plane (YZ) 3 Propagation – detection plane (XZ) (PD-plane) 4 Excitation laser beam 5 Scan ocular (SO) 6 Scanner 7 Detection objective 8 X’, Tilted waist direction ^̂ 9 Tilt angle ^ 10 Object area imaged 11 Area in the object plane, which is illuminated by the beam waist 12 Y’, perpendicular to X’ and Z = Z’, direction corresponding to the rows of the image sensor 13 Pupil of the scan ocular (5) 14 Axicon 15 Lens 16 Intermediate waist 17 Intermediate waist length ^′ 18 Intermediate waist to optical axis tilt angle ^′ 19 Intermediate cone half angle ^^20 First propagation distance 21 Second propagation distance 22 Radial beam displacement 23 Detector 24 Sectional aperture 25 Series of individual beams 26 Excitation objective 27 Center of the TWI sheet / object area 28 Begin / end of the TWI sheet / object area 29 Printing chamber 30 Mirror 31 Detection cone 32 Area on the optic contributing to the TWI beam 33 X, axis of the scanning eyepiece 34 Y axis perpendicular to X and Z 35 Z = Z’ axis 36 Tilted waist length 37 Foci of the individual beamlets contributing to the TWI beam 38 Means to form an excitation wavefront and power / amplitude density distribution of the laser beams 39 Out of plane TWI tilt angle 100 Z direction moveable substrate 101 Printing solution 102 Light sheet of activation 103 Printed object 104 Light projector 105 Cuvette 106 Moving active zone 107 Stable active zone 108 Coupling prism 109 Motion at an angle with respect to printing geometry 110 Vertical printing plane 111 Printed structure 112 Printing movement 113 Scheimpflug geometry 114 Green Bessel beam 115 Support 116 Supply of printing medium 117 Waste removal 118 Synchronous movement and robotic handling 119 Projection lens 120 F-Theta lens 121 Laser line generator 122 Polygon scanner 123 Image inverter 124 Printing medium 125 Wafer with CMOS-circuity 126 Coupling plane 127 Gel 128 Connecting tube to microfluidic system 129 Cells / organoids / embryos 130 Lost mold 131 Sinter material 132 Vibration 133 UV excitation profile 134 Density of photoinitiator in dormant state 135 Gauß intensity distribution 136 Flat top TWI beam Labeling of figure process steps P1 Starting point: lost mold printed using Xolography P2 Filling the lost mold with sinter material (e.g., metal, glass, polymer, ceramic) P3 Optional use of force or vibration to compress P4 Sintering to harden object and remove lost mold P5 Start P6 During printing P7 End of printing process BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1: Setup for printing on a substrate.

[0015] Figure 2: Setup for printing on a substrate within the cuvette.

[0016] Figure 3: Printing at high resolution into a disposable.

[0017] Figure 4: Setup for printing with simplified geometry.

[0018] Figure 5: Setup for printing with simplified geometry, adding an additional visible beam (e.g., Bessel-type) to the UV-laser.

[0019] Figure 6: Setup for continuous printing and assembly.

[0020] Figure 7: Sequential printing of thick layers to generate large objects at high resolution.

[0021] Figure 8: Setup to eliminate the requirement for the time constant of the photoinitiator. a: Optical path. b: Cuvette.

[0022] Figure 9: Combining CMOS-circuits with microfluidics.

[0023] Figure 10: A microfluidic access to a FEP-formed multi-well sample holder. Cells, organoids or embryos cultured inside the wells are supplied with desired nutrients or stimulants via the microfluidic system.

[0024] Figure 11: Using the 3D printed objects as a lost mold.

[0025] Figure 12: Printing on a substrate with angled light sheet.

[0026] Figure 13: Non symmetric conversion of the photoinitiator

[0027] Figure 14: General geometrical layout of the optical system

[0028] Figure 15: Geometry of the elongated excitation beam waist (8)

[0029] Figure 16: Illumination plane with a tilted series of illumination beamlets (25) according to the invention

[0030] Figure 17: (a and b) Bessel beams according to the prior art, (c) An embodiment of the invention

[0031] Figure 18: Geometry of a continuous TWI beam in the object area (10)

[0032] Figures 19 and 20: Simulated data for the y-z cross section of the TWI beam together with vertical and horizontal central profiles

[0033] Figure 21: Effect of the dormant state life time and scan speed on the dynamic density of the photoinitiator in the dormant state

[0034] Figure 22: Example where the invention is used to produce a robust STED deexcitation of out of plane excitation of a TWI excitation beam

[0035] Figure 23: A simple but functional version of the tilted waist beam (TWI) in a quasi- continuous form

[0036] Figure 24: Excitation light is imaged from the pivot point of the scanner (6) to the back- focal plane of an excitation objective (26)

[0037] Figure 25: The collimation of the excitation light can be adapted at the plane of the scanner introducing a variable propagation distance (21)

[0038] Figure 26: A variant that does not use a scan ocular

[0039] Figure 27: Generation of TWI beam and scan geometry with dual-mirror scanner

[0040] Figure 28: Homogenization of the TWI beam with an aspheric element. DESCRIPTION OF THE INVENTION

[0041] The task of the invention is to enable printing of large objects with high resolution with a good homogeneity of resolution of the printing process and at a high speed. The following aspects of the invention are discussed individually: Printing on a substrate

[0042] In general, the planar substrate can be a part of a cavity for holding the liquid and may be a transparent glass sheet, a transparent plastic sheet o or a transparent foil or film.

[0043] In DCPMP (dual color photoinitiator mediated printing), a viscous medium is required to enable support-free printing. The disadvantage is that rinsing fine structures after the printing process to remove residual medium is complicated or in some cases even impossible. The viscosity hinders printing small capillary objects. This is especially required in application of microfluidics and for biomedical applications, e.g., for bioprinting.

[0044] In one embodiment of the invention, printing is done on a supporting substrate, while less viscous medium is used. For this purpose, the geometry of the system is rotated from horizontal to vertical, comparable to the digital light processing (DLP) printing technique actually offered by Xolo GmbH. In a material bath, a support (e.g., a cover glass) is moved upwards along the z axis. On its surface, the activating light sheet activates a thin volume, and coming from the bottom of the cuvette, the projector polymerizes the medium in the thin activated volume. The thin volume activated by the light sheet is hereinafter referred to as “printing volume” or “volume”.

[0045] Figure 1 shows a schematic view of a printing geometry where the light sheet (102) enters to the cavity or cuvette (105) from the side and is oriented parallel to the surface of the movable substrate (100) and the structured light projected from a projector (104) enters the cuvette from the bottom. This setup does not solve the task of printing a large area since it is difficult to generate the light sheet parallel and close to a large planar surface (100).

[0046] In one embodiment of the invention, shown in Figure 2, a planar substrate (100) is positioned inside a cuvette (105), containing the printing solution (101). The orientation of the substrate is tilted with respect to the light sheet (102) activating the photoinitiator. A horizontal movement of the active zone (depicted in Figure 2a), synchronized with the movement of the cuvette synchronously with the substrate, is utilized to keep the thinnest part of the light sheet close to the surface. Alternatively, a movement (depicted in Figure 2b) along or parallel to the surface of the substrate is utilized to keep the thinnest part of the light sheet close to the surface. The movement depicted in Figure 2a and 2b could be utilized at the same time. The slicer (a software, which generates the pattern to be projected) has then to be modified to adapt the ‘shear’ generated by the movement of the substrate along its surface. In both procedures shown in Figure 2a and b, the printing volume is limited since the path of the light sheet within the printing solution is linked to the size of the object. Both printing beams enter the printing volume through one planar substrate

[0047] In one embodiment of the invention, the printing optics for the first and second radiation are coupled to a disposable or cuvette with a planar bottom transparent surface (Figure 3). Coupling a prism (108) approximately matched in refraction index to the printing medium in the disposable enables illumination under an angle with respect to the bottom of the cuvette. A suitable angle of incidence might be 45 degrees. Both the UV sheet (102) and the projected pattern (104) enter the cuvette through one planar surface. Moving the disposable parallel (109) to the planar bottom generates a print where the patterns are sheared against each other. Other combinations of angles for the coupling of the beams to the bottom of the substrate can also be chosen using a modified prism and angular arrangement maintaining the approximate orthogonality of the beams in the disposable. The printing process can be realized by one or several subsequent movements. The printing area is then only limited by the area of the planar bottom surface. As the bottom surface can be also used as a support for the printed objects, they are stabilized during the printing process. The disposable might be equipped with structures before the printing process is performed. This might be the case for microtiter plates of microfluidic devices with provided connectors. These structures can be assembled directly to the printed objects. It would be advantageous to add an imaging readout to the printer to calibrate the position of existing structure to be integrated with respect to the printed objects and to monitor the printing process.

[0048] The advantages are: 1. The printing area is not limited. 2. Printing in low viscosity medium is possible. Printed structures can be mounted directly on the substrate. 4. A structure that is already present in the disposable (as for example a microtiter plate or microfluidic connectors) could be directly connected. This would also enable washing after the printing process. 5. The light sheet thickness could be minimized.

[0049] Another embodiment of the invention is shown in Figure 12 where the light sheet enters the cuvette from the side with a small tilt angle introduced by prisms (a) or mirrors (b). The process of printing ist illustrated in (c).

[0050] When printing structures into microtiter plates, e.g., for cell culture applications, it is favorable that the UV light shines vertically through the bottom of the microtiter plate, so that the print plane is perpendicular to the bottom of the plate. This minimizes the length of the used part of UV light sheet and thus enables thinner light sheets. This is realized in another embodiment of the invention shown in Figure 4. It is achieved by directing a Scheimpflug projection (113) of the DLP chip (104) onto the printing plane (110)).

[0051] A more advanced variant not shown in the Figures would be to use a blaze grating approximately in the object plane of the Scheimpflug optics. The DLP then has to be imaged onto the grating. Adding this diffractive element helps to optimize the overall transmission of the setup.

[0052] In another embodiment, a second light beam in form of a Bessel beam (114) is added to the UV laser forming the light sheet. This would create an additional printing modality, shown in Figure 5. The UV photoconversion along one line perpendicular to the bottom of the disposable would be accompanied by a green straight intensity maximum at a well-defined fraction of the beam path in the printing medium. Switching the additional green laser on would instantaneously generate polymerization along the line. This could be used for fast printing of vertical walls or fill volumes. It could be for example possible to structure the overall plate initially and then add a high resolution imprinting. The method could be also simplified by just using a standard visible (green) gaussian beam.

[0053] The printing geometry could be also modified in such a way that the Scheimpflug projection is provided from two opposite sides (not shown in the Figures). The effort would be reasonable especially if the printing is done through one high-NA objective. It would be then more practicable to mimic the Scheimpflug by means of an Spatial Light Modulator (SLM). In cases of a reflective SLM, the light emitted from the SLM has to be mirrored in two opposite sides of the back focal plane. At very high resolution printing it would even help to generate a constructive interference of the two images generated by the SLM in a plane containing the optical axis. This plane would then be the printing plane (110).

[0054] For very high resolution prints, both beams can also be directed through a single objective with a high numerical aperture (not depicted). In this case, the UV beam and the projection beam would have to be directed through different opposite locations of the entrance pupil. The objective can then be either perpendicular to the bottom of the microtiter plate or at a suitable angle to it. In all cases, the slicer for generating the DLP mask must be adjusted to match the keystone distortion of the Scheimpflug image.

[0055] This embodiment of the invention, depicted in Figure 6, expands the idea to print inside a disposable to a continuous printing process. Here all described advantages are still valid, but a delamination of the printed object from the substrate, which can be for example a FEP foil, is utilized. A robotic handling would be advantageous here to be able to add and mount printed objects to existing parts, e.g., adding an interposer to a cartridge or a chip. Small printed parts can be filtered out from the waste channel of the printing media. Optionally, an additional imaging light sheet, not shown in the figure, is used, which acts as a 3D-profilometer. This allows scanning the object (e.g., a cartridge) precisely and to adapt the printing geometry and timing subsequently within an uninterrupted process. Imaging can also provide quality control data for the final assembly. In addition, new features can be integrated by printing on a printed circuit board to add electrical contact pads to a microfluidic channel.

[0056] Figure 7 explains another embodiment of the invention where a variant of the method of printing on a substrate through a planar surface is utilized. Here, the UV light sheet (102) and the structured visible light (104) are transmitted through the bottom of a foil band. The foil band is optionally synchronized in movement with a substrate (115). The substrate is moved at a certain distance in parallel to the foil. Large layers of the 3D object are printed sequentially by moving the substrate repeatedly at stepwise increased elevated layers parallel and synchronized with the foil. The DLP light selects the structure to be printed and connects the structure directly to the substrate and, after the first layer is printed, subsequently to the previous layers. This allows to print large objects. The advantage here is that the density of the photoinitiator can be increased until substantial absorption of the UV light takes place within the layer that is just to be printed. Thus the UV light will not penetrate deep into the earlier polymerized layers above. This allows to print with photoinitiators with a low contrast, i.e., a high UV cross-section of the already activated photoinitiators. Another advantage is the capability to print large objects with a thin UV activation focus, since the Rayleigh-length can be reduced to the dimension of the layer thickness. The procedure to print large objects layer by layer can be realized with the different methods to generate dual color photoinitiator mediated printing through one planar substrate disclosed in this invention. How to utilize photoinitiators with short intermediate state lifetime

[0057] In another embodiment of the invention, depicted in Figure 8, the requirement for the time constant of the dormant state of the photoinitiator is circumvented by an adapted optical setup. In Xolography, the simultaneous illumination of the volume with an UV light sheet and the structured pattern from the DLP (green light) is achieved with a time constant of the chemistry in the range of seconds. Faster and more confined printing is possible with a shorter lifetime of the activated state of the photoinitiator. Short lifetimes can be achieved by utilizing excited triplet states with lifetimes in the range of µs to ms, as known from the state of the art.

[0058] The "sectioning" produced by the UV light sheet depends on the speed of the cuvette movement and the time constant of the dormant state. The density function is asymmetric (134) to the illumination plane ((133) see Figure 13). If the light sheet is generated with a line type illumination and scanned using a polygon scanner, while the DLP illumination is planar, the time constant of the dormant state must be significantly larger than the time it takes for the scanners to scan the plane. One aspect of the invention is to synchronously illuminate the DLP with a laser line. Then the line that is projected to the printing plane, which is illuminated by the UV beam. The two required beams both illuminate a line-shaped area on the printing plane. Then the UV line would be overlapped or followed by a line of illumination emitted from the corresponding line type area of the DLP. It is then possible to reduce the required time constant of the intermediate state of the photoinitiator to the range of a few microseconds. The lifetime of the intermediate state would then only depend on the geometry and scanning speed in the plane. Polymerization would thus be symmetrical to the plane illuminated by the UV light. This enables the use of entirely different classes of switchable photoinitiators, whose "intermediate state" involves no conformational change, but rather a long-lived electronic state. Long live triplet-state systems become a viable option in this scenario and all types of photoinitiator systems with short dormant state lifetimes known from the state of the art can be used. As a result, the system could be developed for higher resolution and significantly faster printing times. The Figure 8b illustrates the potential situation inside the cuvette (105) and provides a schematic layout. It would be ideal to generate the line illumination by using the same polygon scanner (122) for both beams (121, 102).

[0059] In a different embodiment not shown in the figures, the function of DLP might be replaced by means to structure the line illumination directly. This might be realized by a fast acousto-optical deflector generating a structured line of visible light by modulation combined with deflection. This would then be inserted prior to the rotating polygon scanner. The dwell times of the deflected UV beam at a single point of the sample could be reduced to sub µs values. Improved homogeneity and increased dual color photoinitiator activation life time by using a tilted waist illumination (TWI)

[0060] The reduction of the time-constant of the dual color photoinitiator paired with a synchronized interaction of a ‘line-type’ UV activation within the printing plane with a structured ‘line-type’ polymerization beam requires higher intensities of the polymerization beam, which is realized by focusing the light onto the line type illumination area. Additionally, the photoinitiator with short time constant is not accumulated in the activated stage. Thus the UV laser power and or the power of the polymerization beam and / or the concentration of the photoinitiator need to be adapted. It is thus desirable to enlarge the area of interaction into a widened line of illumination, where the effective widening only affects the width of the UV beam, keeping the sectioning of the beam still good. This enlarged width of the line-type activation area could be in the range between a few µm and multiple mm. The scanning of the activation area is ideally approximately along the direction of the enlarged width of the activation area. A reasonable interaction time of the structured polymerization light with the activation area would be approximately the width of the area divided by the speed of scanning. As an example: If the width would be 50 µm and the speed of scanning would be 50 mm / s the interaction time would be 1 ms. Thus a dual color photoinitiator could be utilized with a life time of the intermediate state of approximately 1 ms. If on the other hand the width would be only 2 µm available with a Bessel beam a reasonable interaction time and thus lifetime would be 40 µs only. The factor introduced by enlarging the width could thus either be used to enlarge the life time of the photoinitiator or to increase the scanning speed of the synchronized structured planar polymerization. Enlarging the photoinitiator lifetime leads to an accumulation of the activated photoinitiator. Thus, the concentration of the photoinitiator could be reduced.

[0061] Another aspect of the dual color polymerization process is the fact that the UV light is absorbed on its way through the printing media containing the photoinitiator. It would be desirable the be able to shape the excitation light in a way such that the density of activated photoinitiator is approximately constant along the ‘line-type’ area.

[0062] Both aspects – the adaptation of the interaction time by increasing the width of the interaction zone and the unsolved problem of absorption of the UV light along the propagation direction – can be solved by the embodiment described in the following section.

[0063] In an embodiment of the invention, a tilted-waist-illumination light sheet (TWI) is utilized to activate an laterally enlarged and laterally scanned line-type area within the printing volume and to synchronize the structured polymerization light with the lateral scanning. The scanned tilted light sheet could then be synchronized with a perpendicular illumination with a structured activation light, e.g., with the use of a DLP-projector or light structured by a fast scanning acousto-optical deflector. The cross-section of the TWI activation beam and the projected polymerization beam can be rendered identical over the full length of the TWI beam. The intensity of the TWI beam can be adapted along the beam without infringing the volume which is closer to the source. A completely homogenized printing is possible then. TWI light sheet activation and imaging could be combined to guarantee a long homogenic activation and a high resolution quality control of the printed volume.

[0064] Definition of a TWI light sheet according to the invention: The TWI light sheet is an electromagnetic propagating field which concentrates the light to a straight elongated area along one direction x’(8). The concentration of the light measured, e.g., with its FWHM (Full Width at Half Maximum) differing along the perpendicular directions y’(12) and z’(35), resulting in an enlarged FWHM in the direction y’(12) by a factor between 1.5 to 1000 with respect to the FWHM in the direction z’(35). The TWI beam is a coherent or incoherent superposition of conical waves. These individual waves are not rotational symmetric with respect to x’ and are in most of the embodiments also asymmetric with respect to the x’-z’ plane. One aspect of the invention is that the generated power density in the elongated area can be rendered almost constant along the direction x’ taking both an initial laser beam profile and a potential absorption of the printing media into account. Some embodiments realize this optimized power distribution starting with common laser beam profiles with reasonable simple means. The conic angle and intensities of the individual waves may vary locally with respect to the location along x’. In contrast to the TWI beam, a Bessel beam is rotation-symmetric with respect to its propagation axis and generates a line focus with rotational symmetry and strong side lobes. Analyzing the Fourier-space of the electromagnetic wave reveals difference of the invention from the state of the art. Figure 17 shows a Bessel beam (a) and a sectioned Bessel beam (b), e.g., accordingly used in light sheet microscopy, whereas Figure 17c shows one embodiment of the invention where 13 shows schematically the distribution in the Fourier space (equals an intensity distribution in the pupil plane of a scan ocular and the lower left and right sketches show the intensity distribution in real space along z’ and within the x’-z’ plane (see Figure 14 a and b). Variant b in Figure 17 could be used according to the invention, but generates an unwanted interference structure along the y’ direction. This can be overcome by using non- coherent illumination of the left and right arc in Fourier space. Variant c is the simplest form of a TWI beam. Here, the Fourier space is calculated in a plane tilted against the x’-z’ plane to show that the illumination distribution in the pupil of a real scan ocular (F-Theta lens), which is represented by the Fourier space, could be utilized in its full diameter (height), generating a beam waist according to the invention which is tilted against the optical axis of the scan ocular. The resulting geometry is shown in Figure 14 (a and b), where the tilt of the propagation direction x (33) with respect to the axis of the TWI beam x’ (8) is depicted. The shape of the beam is shown in Figure 18 a and b, where (a) shows how the beam is propagating through the TWI area oriented along x’ (8) from one side to the other and (b) shows three simulated cross sections from the start, the center and the end of the TWI beam (27, 28) and 10 is the printing plane .

[0065] Variants of implementation of the invention are claimed, which realize cost effective means for generating the excitation beam. Variants of the invention also address the need for flexibility in terms of printing speed, efficiency and sectioning. It provides easy methods to axially confine the excitation beam to the size of the object and to the magnification of the objective used for the structured illumination with the second polymerizing light.

[0066] Further variants describe different technical implementations of the invention. The TWI beam can be implemented using refractive lenses and / or aspheric elements, combinations of refractive and diffractive optical elements (DOEs), such as DOEs in reflection or transmission, or programmable SLMs. DOEs or SLMs could also be used in a double-pass geometry to redistribute the intensity profile of the incoming beam and define the required wavefront. Other variants of the invention relate to methods for exploiting different coherence behaviors or different polarization patterns. The method always utilizes a relative movement between the TWI beam and the object to be printed and a synchronization to the structured illumination region illuminated by the second light to define the object. Means for relative movement are state of the art can be realized by means of galvanometric or MEMS mirror scanners or also by a sliding type of object or beam movement. New approaches are realized according to the invention, where the excitation beam forming the TWI beam is reflected at least two times by a set of movable mirrors. The at least two movable mirrors are moved synchronously or asynchronously or shifted together in a manner that the TWI beam is displaced. In case of a beam scanning or movement, the confinement and synchronization to the printing region illuminated by the second light can be realized by synchronization with a projector (DLP) or with synchronized scanning slit type illumination of the projector, fast acousto-optical or galvanometric or polygon scanners. Synchronization is also possible if the sample is moved relative to the TWI beam.

[0067] To generate a light sheet, an excitation laser beam (4) can be scanned by means of a scanner (6) and a scanning ocular (5) in the form of a generally telecentric F-theta objective or an aplanatic objective such that the direction of propagation lies in the image plane (1) and the scanning movement displaces the beam laterally in the image plane (see Figure 14).

[0068] The axis of the scanning eyepiece and the normal of the image plane (1) (usually the axis of the illumination optics of the projector) define a plane (3) that is perpendicular to the pupil plane (2) of the scanning ocular and perpendicular to the printing plane. This plane will hereinafter be referred to as the propagation-projection plane (PP-plane). For the definitions of the coordinate system, see Figure 14.

[0069] The invention relates to the design of an excitation beam, which is asymmetric with respect to the PP plane. In the nomenclature given by the Figures 14 this plane is defined as the x’z’ plane. To be clear: the invention includes all potential tilt angles (09) between the axis of the TWI sheet illumination optics x and the direction of the TWI sheet x’. Thus also x = x’ is included. This asymmetry can relate to the intensity and or phase and / or to the polarization and or to the coherence of the excitation laser. The excitation light generates a beam waist that is extended along a straight line along x’ that is located in the image plane. The TWI beam can also be generated by a coherent symmetric excitation beam with respect to the PP-Plane. The main aspect is the non-rotational symmetry with respect to the beam waist axis, which generates a laterally enlarged area, which is activated by the first light.

[0070] The waist of the beam might have a tilt angle ^ (09) with respect to the excitation direction x. This tilt angle is half the cone angle of the conical wave. The tilt angle can have angles ranging between 0 and 45 or even 60 degrees, ideally between 0 and 30 degrees according to the invention (see Figure 15). The printing area (10), which is illuminated by structured light is oriented according to the elongation of the beam tail (8) and therefor is rotated with respect the normal of the printing plane (1) in case of ^ (9) being not zero. It is advantageous to introduce a new coordinate system ^^, ^^, ^′ (see (8) for ^′,(12) for ^′and (35) ^^= ^ in Figure 14 to Figure 17. The plane defined by ^^, ^′ differ from the PP-plane in case of ^ ≠ 0. One aspect of the invention is that the excitation light contributing to the TWI beam is radiating from one half-space of the coordinate system ^^, ^^, ^′, separated by the ^^, ^′ plane. This means that in the 2D Fourier space with respect to the ^^, ^′-plane only half of the plane shows non-zero values.

[0071] In one embodiment not shown in a figure of the invention the excitation beam generates a series of foci (37, beamlets), aligned along the beam waist (08), where the foci might not be limited to gaussian foci, but could be also foci with an extended depth of focus (EDOF) or short Bessel-like foci. The separation of the foci might be as large as necessary that they do not show substantial interference with each other in the vicinity of the beam waist forming a ‘light chain’ of well separated beams. The intention of this embodiment of the invention is that the light of an individual beamlet transversally passes the line of the beam waist within the object plane. The photoactivation generated by the excitation of the individual beamlets in the vicinity of the beam waist is then accumulated over the activation lifetime of the dual color photoinitiator system and is able to finally generate an elongated and in y’ direction widened activated area (11). This is depicted in Figure 16. The sum of the contributions from all beamlets leads to an activation area, which are not limited by the Rayleigh length of the individual beamlet. The fact that the beamlets transversally pass the line which connects all foci breaks the symmetry of the overall excitation beam which is the subsummation of all beamlets. The series of beamlets are focused in a series of foci. These are transversally and axially displaced in a row. The invention includes this discrete but also a continuous variant, using conical waves and all intermediate variants. This is elucidated in the following section.

[0072] In a more general description of the discrete embodiment of the invention, the excitation beam electrical field Eexcan be described as: ^^^− ^^^^^^ an in the sample are, ^⃗ is the position vector and ^̂ is a unit vector along the direction of the waist and the lists of real values of ^^and ^^thus generate translated beamles assigned to the positions ^⃗ − ^^^̂ with individual phases ^^. One would typically chose of ^^and ^^to generate equidistant positions and constant phase differences and ^^,^independent of i. The excitation beam intensity distribution is then proportional to |^^^(^⃗)|^. In an incoherent case, e.g., if a broad band excitation source is used or if the distance|^^^̂|is large enough to render interferences small, it would be also sufficient to just add the intensities of the individual beamlets.

[0073] The choice of the phase between the beamlets can be chosen according the overall homogeneity of the excitation in case that there is substantial interference of the beamlets with each other within the printing area.

[0074] In another embodiment of the invention a continuous realization of the excitation beam according to claim 1 is proposed. Eq. 1 can be rewritten in a continuous form: $%Eq. 2 ^&^^(^⃗)=$'^& ^^^⃗ − ^^̂^^"!⃗ #^⃗(^ ^( )

[0075] Here")⃗*is the wave number vector along the waist direction ^̂. In case ^⃗ ="0⃗would be assigned to the center of the printing area address the useful elongation of the excitation beam. For the ease of limiting the phase propagation is assumed to be linear. A more simplified description can be elucidated if the integral is extended to infinity. This shows prominent restrictions: Eq. 3 ^.0^^(^)⃗ =,0^^(^⃗ − ^^)̂^^"!⃗ #^⃗(^

[0076] Normalizations are not taken into account also in the following equations. The property of the field is that of a freely propagating wave. To take this into account, it makes sense to carry out a coordinate transformation in which the direction of the beam waist is along a coordinate axis x' of the new coordinate system (the direction of")⃗*). At the same time, one can write the equation with respect to the other two coordinates (y',z') of the Fourier transformed form: Eq. 4 ^1^^2^^= 0, ).03^, )4^5 =,0^1^2^^, )3^, )4^5^^!67^^(^′

[0077] With the use of the propagation operator with )8=29:& ;as the wavenumber of the excitation laser beam in the medium with refraction index : one can reassign the integral to the values of ^1^in one plane.

[0078] The integral shows only non-zero values if )8− )37− )47= )^^. Thus Eq. 6 )8^− )^7^= )37^+ )47^= BC:D^.

[0079] The wavefront of the wave generated by Eq. 3 has a conical shape since in the Fourier space all values disappear for k values that do not lie on the circle with )8^− )^7 ^= )37 ^+ )47 ^in the coordinate system where x’ is the coordinate axis of the beam waist. If now the wave in the center of the field ^12^^= 0, )3^, is propagated by adding a propagation ^ in the direction of the beam waist equation Eq. 7 ^1^^2^^F, )3^, )4^5 = ^^^^^^ ^!

[0081] non- equation. Eq. 8 shows that this generic setup of the excitation beam with an arbitrary wave ^^(^)⃗ always generates a non-diffracting beam. This means that the intensity is constant in the direction of the propagation axis ^̂. In case that ^^(^)⃗ is chosen to generate a waist, the intensity distribution of this waist is constant along the propagation axis. The parameter to be chosen is the angle of the wave front conus

[0082] The other parameter to be chosen is the intensity distribution of ^^(^)⃗. The beam waist direction would be typically within the printing plane and a symmetry of the intensity with respect to the object plane (^^= ^ = 0) would be advantageous. It is also advantageous to define the generating beam in polar coordinates ^^(^)⃗ = ^^(M, N, ^^) with M = 0 and M = 9 relates to the printing plane. Then a symmetric beam would be generated, if ^^(M, N, ^^) = ^^(−M, N, ^^).

[0083] Bessel beams and sectioned Bessel as described in [COMMUNICATIONS PHYSICS | https: / / doi.org / 10.1038 / s42005-020-00458-3 (Luise Köbele, Alexander Rohrbach) Fig.1b)] are examples for such beams where the propagation axis of the beam waist is colinear with the optical axis of the excitation objective and the wave has rotational / mirror symmetry with respect to the (PP-plane (3), see Figure 14). This means that in the state of the art for light sheet microscopy beams with symmetry with respect to the Propagation-projection-plane (PP-plane (3), see Figure 14) and with respect to the object plane obey ^^(M, N, ^^) = ^^(−M, N, ^^) and aspect of the invention reduces TWI beam to values of the waist is only from one side and thus very limited or zero interference occurs within the waist in ^′ (and also ^) direction.

[0085] If ^̂ is the unit vector along the waist in the coordinate system depicted in Figure 14, the conical phase in the pupil could be written as ^⃗ ^⃗ a scan`= (^`, ^, ^) with ^̂ = (cos ^ , sin ^ , 0) ^⃗ = ^`^ ^ ^ _(^`Da: ^ + ^ BCD ^)^+ (^ Da: ^)^+ (^ BCD ^)^)8Da: I =(^`BCD ^ + ^ Da: ^))8BCD I + _(^`Da: ^ + ^ BCD ^)^+ ^^)8Da: I q. 14 tan M =47E437=,^d efg h.3 ije h

[0087] The aperture of the scan ocular limits the z values which can be utilized. This depends on the numerical aperture Eq. 15 kl^^= : Da: m

[0088] of the scan ocular and its focal length nop. In a paraxial approximation ^TU^= : noptan m , with ^`= −: nop, where : is the refraction index of the printing medium. The maximum overall numerical aperture can be utilized if the wavefront in the apex of the pupil, i.e. at the point Eq. 16 ^q⃗F^^= (−: nop, 0, : noptan m ) points towards the center of the printing area. This is the case if the partial derivative r this point. of Eq.13 the following formula for ^sU^can be derived which maximizes the use of the pupil: Eq. 17 Da: ^sU^=^−(^J: m)^(BCD I)^

[0090] Note that useful of ^sU^can only derived from Eq. 17 if I ≥ m. The area of the pupil contributing to the waist is limited to the region of interest in the image plane has an elliptical shapes in the pupil plane (13).

[0091] It is also part of the invention that a lower portion of the provided numerical aperture is used (klu^U$< kl^^). In this case ^ could still be calculated using Eq. 17 with a modified value of m according Fehler! Verweisquelle konnte nicht gefunden werden. inserting klu^U$. In this cases it is also versatile to use lower values of ^ which also include ^ = 0.

[0092] In the case where I ≥ m and an optimized ^sU^can be calculated according to Eq. 17, Eq. 13, one can calculate the maximum value inserting ^q⃗F^^from Eq. 16 in Eq. 14 leading to vUw x respect to the PP-plane (^ = 0) is broken. This is always the case if ^ ≠ 0 as e.g. outlined in Eq. 17. But this is also the case if ^ = 0 and if the intensity is asymmetric to the PP-plane as outlined in Eq. 10. One main aspect is illustrated in Figure 18 (c) along with Bessel (a) and sectioned Bessel beams (b), where the wave distribution in the Fourier plane (13) is shown together with the intensity distribution in the center of the printing plane in a plane parallel to the excitation pupil plane (E-plane (02), see Figure 14). It is obvious that the choice of the intensity distribution in the pupil on the elliptical arc illumination line and the local angular distribution affects the length of the waist along x’, sectioning capability and intensities of potential side lobes of the central intensity maximum. An intensity modulation within the printing plane can be suppressed and the excitation scheme is robust against a jitter of the scanner with respect to the synchronized motion of the second light. There is no principal limitation of the length of the waist. Limitations relating to the classical Rayleigh length known from the focus of a gaussian beam are not present. The full potential of the scan ocular (5) in the direction perpendicular to the object plane (01) to provide sectioning can be exploited if Eq. 17 is considered.

[0094] Figure 19 shows calculated properties of the TWI beam. Here, an excitation wavelength of 405 nm, a printing medium containing the dual color photoinitiator system with an refraction index n = 1.5, kl^^= 0.5, I = 30.0° and ^sU^= 23.3° according to Eq.17 is used. According to Eq. 18 MsU^= 41.8° . Figure 19 (a) shows the intensity distribution of the TWI beam in the y-z plane with the central beam profiles in ^ and ^ direction. (b) shows the intensity profile in ^ direction as an integral over the range −0.6µ^ < ^ < 0.6µ^ which is an appropriate width of illumination with the second light for this parameters. A Gauss profile fitted to the profile shows a FWHM of 350 nm in z-direction. Figure 19(c) shows the intensity in the Fourier space as already shown schematically in Figure 17(c). The ellipse segment connects the upper and lower apex of the pupil. The TWI beam leaves notches on top and below the beam waist free of excitation light. The full angle of the notches is 180° − 2MsU^. If, as shown in Figure 19(d), the intensity is smoothed by a truncated Gaussian intensity distribution with respect to the azimuthal angle M; side-lobs could be suppressed without a big loss of sectioning. All other parameters are identical to those used for the calculation represented by Figure 19 (a-c).

[0095] A version for macro printing is shown in Figure 20: The parameters are 405 nm, n=1.5), kl^^= 0.13, I = 17.2° and ^ = 16.5° with an illumination width of −3.5µ^ < ^ < 3.5µ^, and according to Eq. = 17.0°. The sectioning is 1.41 µm FWHM Figure 20 (d) is a variant with gaussian leading to a side lobe reduced light TWI light sheet and a sectioning of 1.5 µm FWHM. The simulations of the TWI excitation beam profile, shown in the Figures 19 and 20, do not take a finite dormant state lifetime in combination with a scanning speed into account.

[0096] One aspect which has not been taken into account is the temporal accumulation of activation of the dual color photoinitiator system. Given that the intermediate or dormant state of the photoinitiator is declining exponentially over time after being activated by the TWI beam with a time constant of ^^o, a lateral scan speed of the TWI beam of ^^^^in the direction of ^ is combined into a ‘spatial decay’ of the dormant state D^o= ^^o^^^^. The actual distribution of the dual color photoinitiator can be calculated by a convolution of the 3D-TWI intensity distribution with a one dimensional decay curve with the decay constant D^o. It is obvious that reasonable results of the parameters of ^^oand ^^^^have to be chosen such that D^ois in the order of magnitude of the width of the TWI beam. Enlarging the width could therefore be beneficial for a larger scan speed or for utilizing larger dormant state life times ^^o. The amount of photoinitiator in the dormant state scales with ^^ounless saturation is reached. Figure 22 shows the influence of an enlarged D^oon distribution of the activated photoinitiator in the y-z plane. The confinement of the 3D-printing is maintained over a large range of values of D^o.

[0097] In another embodiment of the invention STED is utilized: The Excitation-TWI beam at wavelength 1 is used to excite the sample whereas a second modified STED-TWI beam at an additional wavelength 2 is used to deplete the activation of the photoinitiator system by deexcitation of the excited photoinitiators. The modification of the second beam is a 9 step in the phase at M = 0 leading to a destructive interference in the plane of the TWI beam of the wavelength 1. The cone angles I and azimuthal intensity distributions could be selected differently for Excitation-TWI and STED-TWI beam. This could be used to optimize the impact of the STED-TWI beam. One could for example use a Gaussian envelope for the Excitation- TWI beam to generate a Gaussian focus which is than squeezed down by the two symmetric maxima of the STED-TWI beam resulting in a side lobe free high resolution sectioning. Figure 22 shows a STED-TWI beam cross-section (a) and ^-profile (b). A Gaussian envelope was additionally applied. (c) shows an Excitation-TWI beam with the identical parameters but depleted by the STED-TWI beam (b) and an exemplary depleted TWI profile with strong saturation of the STED TWI beam. There is no principal limitation in the potential length of the TWI beam. Adding a truncated Gaussian envelope in M – direction especially to the STED-TWI beams helps to generate a robust deexcitation since the unwanted minima between the sidelobes are reduced.

[0098] Figure 23 (a) and (b) shows a simple but functional version of the tilted waist beam (TWI). The incoming collimated beam hits an axicon (14) at an off-axis position (22). An intermediate beam waist (16) is formed with a length (17) which depends on the width of the incoming beam. At this position, two sectional apertures (24) can be inserted to confine the intermediate waist. The central beam propagates at an intermediate cone half angle ^^(19) with respect to the axis of the axicon. The following lens (15) is placed at a potentially tilted angle ^′ (18). This angle might have values between 0 and ^^. The distance of the lens to the center of the intermediate beam waist (20) is in the current example the focal length of the lens. The following propagation distance (21) leads to a segment of a circular or elliptic focal line on the scanner (6) and telecentric illumination, if the scanner is placed in the focal plane of the lens (15). The beam is than reflected towards the scan ocular (5) and generates the image (TWI) of the intermediate beam waist within the printing area (1). The tilt angle of the TWI beam with respect to the optical axis of the scan ocular scales with ^′ (18) and the magnification introduced by the lenses (15) and (5). The beam is always asymmetric with respect to the PP-Plane independently of the choice of ^′. The term Tilted Waist Illumination (TWI) refers to this asymmetry which elucidates the fact that the waist is illuminated predominantly from one side within the printing plane (1).

[0099] One realization of a TWI beam setup is shown Figure 24, where a microscope objective (26) is used to generate the TWI beam. Here, the position where the scanner (6) is located is imaged into the back focal plane of the excitation objective (26) by means of intermediate imaging optics (5,17,5). The placement of the pivot point in the pupil of the objective (26) important for a telecentric motion of the TWI beam. The limitation of the TWI beams can also be achieved by apertures (24), as shown in Figure 23 (a) , near one of the intermediate image planes (17).

[0100] There are other variants not shown here. One might use two scanners with parallel axis realizing a beam geometry, where a pivot point can be addressed, which lies at a distance to the surface of the scanners. With this method the use of intermediate imaging optics could be omitted in a case where a microscope objective is used with a pupil position inside the objective.

[0101] A person skilled in the art can adapt the scanning scheme in various variants: Scanning could also be realized by a relative movement of the excitation lens (26 or 05) including parts of the excitation optics with respect to the object synchronously with the optics which image the second light to the printing plane.

[0102] Further variants of the invention includes variations of the conus angle I(^^)along the waist and means to dynamically change those variation. This variable conus angle has to be inserted in

[0103] Eq. 11. One solution depicted in Figure 25 is to leave the telecentric condition in the object space of the scan ocular (5) or excitation objective (26). This allows the beam waist to be concentrated or expanded in the direction of ^^to adapt to the size of the object to be printed. This would be also advantageous if a magnification changer or changing of the illumination objectives for the second light are implemented. The conus angle I(^^) is related to the useful width of the active detection region.

[0104] It is also part of the invention to change the characteristic of the TWI beam dynamically especially to change kl^^towards smaller values kl^`^.

[0105] The wave according to the invention can be generated by means of various means. Among those are Spatial-light-modulators (SLM) either in one or in double path as well as diffractive optical elements (DOE). Classic optical elements to be used are also axicons. Axicons can also be replaced by conical mirrors and the optics imaging the intermediate waist could be also replaced by off axis mirrors, pairs of spherical mirrors or toroidal mirrors.

[0106] In another embodiment of the invention more than one excitation laser source is used. This might be helpful if compositions of materials should be printed. The shaping of the wavefront could be done by using separated diffractive optical elements of different areas on the same diffractive optical element. It is advantageous that the single sided elliptical shape of the wavefront is relatively small. This makes it possible to utilize reasonably small diffractive optical elements for shaping several laser sources. The combination of the light sources could be done by dielectric beam combiners, prisms of just geometrically as the angle I is chosen individually with identical values to ^ for each wavelength. Than the elliptical shaped beams hit the scanner plane at different positions but identical positions of the TWI beam would be obtained.

[0107] In another embodiment not shown, at least two excitation TWI laser light beams (4) have the same wavelength are used to incoherently illuminate the excitation beam waists (TWI) from opposite directions. This can be achieved by splitting one excitation laser light beam (4). The advantage is that a more homogenous illumination is achievable in cases where the printing media is not perfectly clear and causes scattering and beam distortion. This could be the case if the printing medium contains cells and tissue like structures should be printed.

[0108] In another embodiment of the invention shown in Figure 26, the use of a scan ocular is omitted. The beam generated by an axicon (14) or other means to provide the required wavefront and intensity distribution. The TWI beam is scanned by means of a two reflections on the same scanner mirror (08). Here a variant is shown where the beam hits the scanner (08) from opposite sites. The intermediate reflections on the mirrors (30) and the second reflection on the scanner leads to an compensation of the scan angle. Thus a pure parallel movement of the TWI beam is achieved. The setup is thus very simple and does not suffer from distortions of a scan ocular. High TWI beam numerical apertures are achievable for a large field of view. This setup might be ideal for macro-printing. An alternative embodiment shown in Figure 27 utilizing only one scanner can be realized by rotating a set of two mirrors opposite to each other with one scanner. The two reflections can also be positioned to reflect the beam back to the same side of the scanner with the identical compensational effect on the angle of the TWI beam leaving the scanner after the second reflection.

[0109] The TWI beam can also be homogenized by aspherical elements such as n aspherical conical mirror as shown in Figure 28. Here a slight modulation of the angle Iwith respect to the position within the TWI beam but transforming e.g. gaussian angular distribution of the source into a flat top TWI beam along its length (17).

[0110] In another embodiment not shown in the figures the scan movement of the TWI beam is performed by a synchronous movement of mirrors between the wave front forming elements (axicon, aspheric mirror, SLM..) and the printing plane. The mirrors are set to not change the overall distance to the printing plane. Potential realizations of such an arrangement are obvious for skilled people. The slit type structured illumination might also be generated by moving the projection optics synchronously with the TWI beam. The pattern of the projector might need to be shifted on the projector synchronously with the movement of the projection system.

[0111] The invention can also be applied if the angle between the excitation and projection objective differs from 90°. Here a compensation of the distortion due to a tilted printing plane has to be utilized. This can cause keystone distortion of the image. The problem here is that the lines simultaneously illuminated by the TWI beam in the printing volume might rotate while scanning. One embodiment of the invention also deals with this restriction. The TWI could be adapted to a keystone distorted projection of the second light into the printing plane. It is possible to adapt the TWI beam while scanning in the printing plane to be able to synchronize with line illumination with the second light by leaving the telecentric condition in the image space of the scan ocular (see Figure 24)

[0112] The invention also includes the use of dual color light sheet based printing by generation of the light sheet and perform illumination with the second light through the same objective (see similar setup for microscopy Dunsby, C. Optically sectioned imaging by oblique plane microscopy. Opt. Express 16, 20306 (2008)) USE OF THE INVENTION

[0113] In another embodiment, DCPMP (dual color photoinitiator mediated printing) is used to directly print on processed silicon wafers with integrated electronic circuits (Figure 9). Usually, this requires complex assembly technology that can be overcome by directly printing onto a processed wafer. One example is a microfluidic bubble jet generator with corresponding contact pads and current amplifiers on the silicon substrate with printed microfluidic circuits that are ideally adjusted to the electronic circuitry on silicon by application of DCPMP. Another application is the integration of electronic pressure sensors that are located directly under the microfluidic channels. A monitoring setup like a 3D-profilometer by additional imaging light sheets can be utilized to precisely adjust the printed microfluidic channel to the preprocessed silicon substrate. Another application could be to use processed microfluidic MEMS-wafers to print an optically transparent lid on top of microfluidic channels that were etched into the silicon substrate. On wafer level this is usually done by bonding a glass wafer on top of the processed silicon wafer. By application of DCPMP, complex bonding processes can be overcome. In addition, add more functionalities can be added to the device, e.g., by printing integrated optical components next to the microfluidic channels. An interposer and optical elements can be integrated subsequently.

[0114] FEP based sample holders are the gold standard for live microscopy of larger tissues (>100µm) using single-photon fluorescence microscopy. Imaging is possible with light sheet microscopy. A Swiss company, Viventis Microscopy (https: / / viventis-microscopy.com), has developed a system that allows multi-view light-sheet imaging over many days on a variety of organoids, spheroids, embryos. In one embodiment of the invention, a system comparable to this can be microfluidically equipped using dual color 3D printing technology to further pattern the sample holders (Figure 10). Here, in the FEP channel, a vascular system patterned with gel could be printed. This could be perfused along the direction of the groove. One could imagine various geometries for printing. It is possible to connect the inlet and outlet of each well with a printed tube structure. The row of wells can only be one-dimensional because imaging is done from the opposite sides with high-NA dipping objectives (not shown in the figure The geometry is ideally accessible with a TWI based light sheet generation. If I is chosen properly the TWI beam could be parallel to the bottom of the FEP well and two TWI beams are advantageous to be installed and enter the volume from opposite directions. The scan direction could be vertical and projection and microscopic detection is possible from opposite directions to the side of the FEP well. Both TWI beams would be parallel.

[0115] In one embodiment, 3D-printed objects can be utilized to form a lost mold (Figure 11 with process steps P1 to P4). DCPMP technique offers the possibility of printing molds that are not demoldable. Classically, non-demoldable molds are made of sand, are produced around wax positives in the melt-out process with clay or fireclay. Embedded Styrofoam positives are directly replaced by the liquid metal. The molds are then referred to as a lost mold. Molds that cannot be demolded can also be made of silicone, so that the objects can then be removed from the mold. The mold can then be recycled. The printing material can be removed by burning, dissolving, chemicals or enzymatic digestion. Lost forms can be used in connection with sintering (e.g. metal alloys, plastic granules, PTFE) or firing processes (ceramics, glass). In this case, the molds printed would burn without residue at high temperatures. It may be necessary to first press or vibrate the mold together with the filled sintering material in order to compact the sintering material. Potential materials can be: 1. Glass (Glasomer) (https: / / www.glassomer.com / ) 2. Ceramics (https: / / www.ikts.fraunhofer.de / content / dam / ikts / downloads / jahresberichte / jb2020 / 14_Freeform_Injection_Molding_Keramik_Spritzguss_mit_verl orenen_Formen.pdf) 3. Ceramics and metals (https: / / www.3d-model.com / de-de / material)

[0116] In the LIGA process, lost microstructure molds made of plastic (usually PMMA) are filled with metal with a galvanometric procedure. The plastic is then removed. With DCPMP, LIGA can be implemented much more cost-efficient and with almost any object. For example, metallic 3D spring elements and valves can be created or microreactors can be generated. Electrolytes can be added to the printing material so that conductivity is provided.

Claims

CLAIMS 1. Dual-color photoinitiator-mediated 3D printing process comprising the photopolymerization of at least one monomer at a desired spatial position on an object plane (1) in a liquid comprising the least one monomer and at least one photo initiator capable of initiating photopolymerization of the at least one monomer by providing one or more first light beams and a second light beam having different wavelength at the desired spatial position thereby activating the at least one photo initiator thereby initiating photopolymerization of the at least one monomer, wherein the first and second light beams are movable relative to the desired spatial position characterized in that the first light beam is provided in form of a light sheet to the object plane (1).

2. Dual-color photoinitiator-mediated 3D printing process according to claim 1 characterized in that the light sheet is provided by propagating one or more first light beams with a wavefront and a power / amplitude density distribution into one or more continuous or modulated or discontinuous excitation beam waist (TWI) in the object plane (1).

3. Dual-color photoinitiator-mediated 3D printing process according to claim 1 or 2 characterized in that the second light beam illuminates the area excited by the excitation beam waists (TWI) and the relative movement of the second light beam is synchronized with relative movement of the excitation beam waist (TWI).

4. Dual-color photoinitiator-mediated 3D printing process according to any of claim 1 to 3 characterized in that the excitation beam waist (TWI) has a power / amplitude density distribution in the shape of at least one circular or elliptical arc of 10 - 179° in the Fourier transformation of the wavefront and amplitude distribution with respect to the x’(8),z’(35) plane of the volume activated by the light sheet.

5. Dual-color photoinitiator-mediated 3D printing process according to claim 4 characterized in that the elliptical or circular power / amplitude distribution has an extended radial structure forming a distribution of predominantly conical waves interfering to a axially confined and or modulated or discontinuous excitation beam waists (TWI) in the object plane (1).

6. Dual-color photoinitiator-mediated 3D printing process according to any of claims 1 to 5 characterized in that at least two excitation laser light beams (4) having the same wavelength are used as first light beams to incoherently illuminate the excitation beam waists (TWI) from opposite directions.

7. Dual-color photoinitiator-mediated 3D printing process according to any of claims 1 to 6 characterized in that a second excitation beam waist (TWI) having a intensity minimum in the object plane (1) is provided to deexcite the dual color photoinitiator system resulting in a STED light sheet mode.

8. Dual-color photoinitiator-mediated 3D printing process according to any of claims 1 to 7 characterized in that the excitation beam waist TWI and the second light beam is generated through one common objective (7).

9. Dual-color photoinitiator-mediated 3D printing process according to any of claims 1 to 8 characterized in that the light sheet has a propagation direction perpendicular to the second light beam.

10. Dual-color photoinitiator-mediated 3D printing process according to any of claims 1 to 9 characterized in that the first and second light beam are coupled to the object plane (1) by a prism, optionally with an immersion liquid.

11. Dual-color photoinitiator-mediated 3D printing process according to any of claims 1 to 10 characterized in that the first and / or second light beam is provided through a Scheimpflug optical system.

12. Dual-color photoinitiator-mediated 3D printing process according to any of claims 1 to 10 characterized in that the second light beam is provided in part through a Scheimpflug optical system and in part coaxial with the first light beam using a Bessel or Gauss optical system.

13. Dual-color photoinitiator-mediated 3D printing process according to any of claim 1 to 12 characterized in that the object plane (1) is illuminated through a continuously moving transparent film.

14. Dual-color photoinitiator-mediated 3D printing process according to any of claim 1 to 13 characterized in that the first and / or second light beam is provided adjacent to the object plane (1) through a polygon or galvanometric or translational scanner.

15. Dual-color photoinitiator-mediated 3D printing process according to any of claim 1 to 14 characterized in that the first and / or second light beam is provided using one or more scanners provided with mirrors and where the first and / or second light beam is reflected at the one or more mirrors of the scanner more than one time.

Images