Non-centered axisymmetric catadioptric optical system

Abstract

The invention relates to a non-centered axisymmetric catadioptric optical system, comprising a front block (50) having first and second reflective surfaces (S1, S2) configured to form an axisymmetric collimated beam (20) that is inclined relative to its optical axis (10), and reflected as a coaxial collimated beam (25) by a third reflective surface (S3). The first reflective surface (S1) has a nominal surface formed by rotating a conical generatrix extending radially from the optical axis (10). The second reflective surface (S2) has a nominal surface formed by rotating a straight or parabolic generatrix, having a normal orthogonal to the optical axis (10).

Description

Description Title: Non-center axisymmetric catadioptric optical system TECHNICAL FIELD [ooi] The field of the invention is that of catadioptric optical systems, in particular those adapted to multispectral, confocal, Raman microscopy, fluorescence microscopy or photoluminescence microscopy. PREVIOUS STATE OF THE ART

[0002] Among optical systems, there are essentially dioptric systems. These consist of a set of lenses made of materials whose refractive index varies with wavelength. Each lens thus induces chromatic aberration that must be corrected. To this end, lenses made of different materials are generally combined so that the dispersion of their refractive indices at least partially compensates for each other.

[0003] These dioptric optical systems generally offer good brightness, which increases with their numerical aperture. However, aberrations are likely to be significant at high numerical apertures. These can be corrected by adding lenses and / or by correcting their surfaces.

[0004] Therefore, whether correcting chromatic aberrations or other aberrations at high numerical apertures, a large number of lenses, typically more than 10, is generally required. Dioptric optical systems are thus usually bulky and heavy. Furthermore, because chromatic aberration-corrected dioptric optical systems incorporate materials with different coefficients of thermal expansion, they are poorly suited to applications involving large temperature variations, such as imaging samples in a cryostat or for space applications. For these latter applications, the weight and lack of compactness are also problematic.

[0005] Catoptric optical systems, on the other hand, are inherently achromatic. Advantageously, these consist of axisymmetric mirrors, meaning their surfaces are formed by rotating a generatrix around an axis. Such a mirror can be produced relatively easily by precision turning, for example, using a lathe equipped with a diamond cutting tool. The generatrix can for example, it can be conical. Mirrors are generally polished to achieve a surface of optical quality.

[0006] When all the foci of the generating elements are aligned on an optical axis of the axisymmetric catoptric optical system, it is said to be centered. Examples include the objectives of Cassegrain telescopes and Schwarzschild microscopes. Any other axisymmetric catoptric optical system that does not meet this definition is said to be non-centered, or decentered. This latter category notably includes axisymmetric catoptric systems with a conical generating mirror where one of the foci is located off the optical axis.

[0007] All catoptric systems are brighter the fewer mirrors they have, which leads to greater compactness and lightness.

[0008] Centered axisymmetric catoptric systems generally suffer from central pupil obstruction when they have more than one mirror, due to the folding of the optical axis upon itself. This can negatively impact the modulation transfer function at certain spatial frequencies, or even generate angular or spatial blind spots.

[0009] This drawback can be overcome by adopting an off-center catoptric system design. However, the mirrors used in this type of optical system generally have optical characteristics sensitive to small deformations, making them difficult to manufacture with the precision required to achieve the same optical performance as centered systems. Their relative positions are also difficult to adjust, for example, to compensate for machining or polishing errors. The field of view may, in some cases, be reduced. [ooio] An axisymmetric catadioptric optical system can be viewed as an axisymmetric catadoptric optical system to which one or more diopters may be associated. A catadoptric system is a special type of catadiptric system that does not include any diopters. Preferably, when present, these diopters are arranged and configured to introduce little or no chromatic aberration, for example, by ensuring that light rays pass through them at near-normal incidence, and / or that the thicknesses of the materials they traverse are small. They may, for example, perform a mechanical function, such as holding the mirrors in place, or an optical function, such as shaping a beam or correcting field curvature. DESCRIPTION OF THE INVENTION [ooii] The invention aims to remedy, at least in part, the drawbacks of the prior art, and more particularly to provide a compact, non-centered, axisymmetric catadioptric optical system with good resolution, low sensitivity to temperature variations, and greater tolerance to machining errors. According to advantageous embodiments, its resolution is close to the diffraction limit for a large field of view, for example, + / - 5 pm. [ooi2] For this purpose, the object of the invention is a non-centered axisymmetric catadioptric optical system, comprising: an optical axis, a front block including a collimator block and a third reflective surface in the shape of a truncated cone centered on the optical axis. [ooi3] The collimator block is such that it has a focus, a first reflective surface and a second reflective surface, and is configured to form an axisymmetric collimated beam inclined with respect to the optical axis by successive reflections on the first reflective surface and the second reflective surface when an isotropic point light source is placed at the focus.

[0014] The third reflective surface is configured to direct the axisymmetric collimated beam parallel to the optical axis, thus generating a coaxial collimated beam. [ooi5] The collimator block is further such that: the first reflecting surface has a nominal surface formed by revolution of a first conical generatrix around the optical axis, a geometric focus of the first generatrix being coincident with the focus of the collimator block, the first generatrix extending radially from a vertex of the first reflecting surface located on the optical axis; the second reflecting surface has a nominal surface formed by revolution around the optical axis, of a second right or parabolic generatrix, having a normal orthogonal to the optical axis.

[0016] Some preferred but not limiting aspects of this optical system are as follows.

[0017] The axisymmetric collimated beam can form a cone of light with an apex angle greater than or equal to 130°.

[0018] The first reflecting surface may have an optical correction of its nominal surface area to ensure uniformity of the irradiance of the axisymmetric collimated beam at the level of the second reflecting surface. The second reflecting surface may have an optical correction of its nominal surface area to ensure uniformity of the phase of the axisymmetric collimated beam after reflection from the second reflecting surface.

[0019] The optical system may further include a monolithic block made of a single material. The first reflective surface and the second reflective surface may be one or both surfaces of the monolithic block.

[0020] The monolithic block may consist of a central block and / or a peripheral block surrounding the central block. The peripheral block may be mounted with adjustable translation parallel to the optical axis relative to the central block. The first and third reflective surfaces may be surfaces of the central block. The second reflective surface may be a surface of the peripheral block.

[0021] The monolithic block may include a centering block attached to the central block, the centering block and the peripheral block may include corresponding surfaces of revolution, which may allow one to be guided relative to the other in translation parallel to the optical axis.

[0022] The monolithic block can be made of metal, the front block can be purely catoptric, and the centering block can be a spider.

[0023] The monolithic block can be made of a transparent material. The centering block can be a disc into which part of the central block can be fitted.

[0024] The centering block can extend the central block, and the central block can include a truncated cone-shaped recess to accommodate the third reflective surface.

[0025] The first generator can be a portion of a parabola, and the second generator can be a straight line segment parallel to the optical axis.

[0026] The optical system may further include a rectifier block which may comprise two head-to-tail axicon lenses, centered on the optical axis, the rectifier block being able to be arranged to receive the coaxial collimated beam. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other aspects, objects, advantages and features of the invention will become more apparent from the following detailed description of preferred embodiments thereof, given by way of non-limiting example, and made with reference to the accompanying drawings in which: Figure 1A is a geometric representation in meridian section of a catoptric and axisymmetric frontal block; Figure 1B is a geometric representation in meridian section of an advantageous variant of the frontal block of Figure 1A; Figure 1C is a geometric representation in meridian section of another variant of the front block of Figure 1A; Figure 2 is a schematic cross-sectional view of a first example of a monolithic block having a front block similar to that of any one of Figures 1A, 1B or 1C; Figure 3 is a schematic cross-sectional view of a second example of a monolithic block having a front block similar to that of any one of Figures 1A, 1B or 1C; Figure 4 is a schematic perspective view of a third example of a monolithic block having a front block similar to that of any one of Figures 1A, 1B or 1C; Figure 5 is a schematic view of an embodiment of a non-centered axisymmetric catadioptric optical system having a front block similar to that of any one of Figures 1A, 1B or 1C; Figure 6A is a spot diagram obtained with a frontal block similar to that of Figure 1B;Figure 6B is a spot diagram obtained with the front block of Figure 6A, but with a constant material removal error during the machining of one of its reflective surfaces; Figure 6C is a spot diagram obtained with the front block of Figure 6B, after axial adjustment of the position of the reflective surface exhibiting the machining defect; Figure 7A is a geometric representation of a front block similar to that of Figure 1B; Figure 7B is a geometric representation of the front block of Figure 7A, in which an advantageous optical correction has been applied to certain nominal surfaces; Figure 8A illustrates an optical correction obtained according to the method illustrated in Figure 7B, applied to a first reflective surface of the front block; Figure 8B illustrates an optical correction obtained according to the method illustrated in Figure 7B, applied to a second reflective surface of the front block;Figure 9A is a ray tracing in the acceptance cone of a frontal block similar to that of Figure 1B, from an isotropic point light source; Figure 9B is a spot diagram obtained with an optical system similar to that of Figure 5 and the ray tracing of Figure 9A; Figure 9C is a spot diagram obtained with the ray tracing of Figure 9A and the optical system of Figure 9B, for which some of its reflective surfaces have been corrected; Figure 10A represents spot diagrams obtained with an optical system similar to that of Figure 5 and the ray tracing of Figure 9A, for 5 different positions of the light source; Figure 10B represents spot diagrams obtained with the optical system of Figure 10A, for the same 5 positions of the light source, but in which some reflective surfaces have been corrected; Figure 10C is a spot diagram obtained with an optical system similar to that of Figure 10A, for the same 5 positions of the light source, but with a different numerical aperture and magnification and in which some reflective surfaces have been corrected. DETAILED DESCRIPTION OF SPECIFIC METHODS OF IMPLEMENTATION

[0028] In the figures and throughout the description, the same reference numerals represent identical or similar elements. Furthermore, the various elements are not drawn to scale to ensure clarity. Moreover, the different embodiments and variants are not mutually exclusive and may be combined. Unless otherwise stated, the terms "approximately," "around," and "in the order of" mean within 10%, and preferably within 5%. Furthermore, the terms "between ... and ..." and equivalents mean inclusive of the bounds, unless otherwise specified.

[0029] Machine tools used for machining optical surfaces typically include a cutting tool, such as a diamond tip, manipulated by a nanometer actuator to achieve high precision. Machining precision refers to the smallest displacement of the cutting tool that the machine tool can control. Accuracy, another indicator of machining quality, is defined as the uncertainty in the cutting tool's positioning relative to the target surface. Accuracy depends on inherent machine tool capabilities, such as the position of the diamond tip relative to the machine tool axis in turning operations, but also on other factors like local heating, the materials being machined, and wear.

[0030] As an example, the inventors determined that it is necessary to control the local temperature during machining to better than 2°C, in order to obtain an uncertainty of placement of two points on a machined surface, 50 cm apart, less than or equal to one hundredth of a millimeter, with a material commonly used to make a mirror, such as glass.

[0031] In a centered axisymmetric catoptric optical system, it is generally possible to partially compensate for machining errors by adjusting the relative positions of its reflecting surfaces, and consequently their focal points, during assembly. The compensations achievable through such adjustment are more limited when the optical system is not centered.

[0032] As an example, an axisymmetric reflecting surface with a conical generatrix and a focus located off the optical axis has a distinct focus in each meridian plane including the optical axis. All these foci together form a focal line that loops back on itself around the optical axis. A machining error distorts this focal line, for example, homothetically in the case of radially uniform excess material removal. Thus, for most non-centered axisymmetric catoptric optical systems, simply shifting the reflecting surfaces does not restore the expected conjugate relationships between them, since this would require applying an inverse deformation to the focal line.

[0033] The non-centered axisymmetric optical system of the invention is advantageous in that a translation parallel to the optical axis of a second reflective surface of the system partially compensates for any lack of accuracy in its machining. This is made possible by a shape and arrangement of this second reflective surface, specific to the invention and resulting from the combined configuration of all the reflective surfaces. This second reflective surface is characterized in particular by having a nominal surface formed by revolution around the optical axis of a second straight or parabolic generatrix, having a normal orthogonal to the optical axis.

[0034] This configuration allows all reflective surfaces to be grouped into a particularly compact central block, making it easy to insert into a cryogenic chamber, for example. Its height, measured parallel to the optical axis, can be less than or equal to 3 cm, or even less than or equal to 2 cm. Its working distance can be less than or equal to 5 mm.

[0035] The optical system of the invention is homothetic, and therefore particularly well-suited for applications in extreme environments, for example, for analyses at extremely low temperatures, typically down to 4 K or less. In the space sector, it can be used to track luminous objects such as stars, including the sun, which can be useful for navigation.

[0036] By adding an illumination channel comprising a field diaphragm and another field diaphragm in a conjugate plane of the sample to be analyzed in the imaging channel, the optical system of the invention is particularly suited to confocal microscopy, for example for the analysis of the photoluminescence emitted by an individual color center such as a vacancy-nitrogen pair in a diamond crystal.

[0037] In the description, a reflective surface has three distinct characteristics, designated respectively as the nominal surface, the corrected surface, and the realized surface. The nominal surface is the base surface used to establish the conjugate relationships in the optical system. The corrected surface corresponds to the nominal surface to which an optical correction has been applied to correct certain defects in the optical system, such as geometric aberrations. Finally, the realized surface is the surface obtained after machining, and possibly polishing, using the corrected surface as the target.

[0038] Optical correction is a two-dimensional deviation applied to the nominal surface in a direction normal to it, limited to low spatial frequencies, for example strictly less than 1 / 4 Å, and low amplitudes, for example strictly less than Å / 4, where Å is the average wavelength of the optical system's working range. Thus, the maximum deviation between the corrected surface and the nominal surface can typically be strictly less than Å / 4, measured perpendicular to the nominal surface.

[0039] For an optical system, the diffraction limit designates the theoretical resolution, that is, the maximum resolving power that would be obtained with an ideal optical system of the same numerical aperture and in an identical geometric and photometric configuration (defined, for example, by its focal points, entrance and exit pupils, principal planes, or other geometric characteristics), this optical system being notably free of aberrations and apodization. Such an ideal optical system, when it has a circular pupil, has an impulse response (also called the point spread function, a literal translation of the English "Point Spread Function" (PSF)), corresponding to an Airy disk with a diameter at the first zero equal to 1.22 Å / NA, where Å is the wavelength and NA is the numerical aperture.

[0040] In this description, an optical system is said to be diffraction-limited in a given object or image field when, for any point in that field, the diameter of the smallest circle containing 95% of the energy of the image of that point differs by less than 10% from that of the circle containing 95% of the energy of the Airy disk obtained for the same aperture. digital. When the image is at infinity, this comparison can be made in the focal plane of an ideal lens without affecting the exit pupil of the optical system.

[0041] Specific embodiments will be described relating to a non-centered axisymmetric catadioptric optical system 1, some with diopters, others without. They include an optical axis 10 and a collimator block delivering an axisymmetric collimated beam inclined with respect to this axis when a point light source is placed at its focus, i.e., in an infinite conjugate relation. However, this collimator block is also suitable for imaging an object point at a finite distance, in a finite conjugate relation; these embodiments can be optimized for this type of relation.

[0042] In all figures, the embodiments are represented according to idealized geometric forms, it being understood that the corresponding realized forms may undergo modifications compatible with the invention, due to limitations of the manufacturing process, thus a salient edge or a summit point may be blunt once made, an optical surface may have asperities or roughness such as grooves produced by the translation of the tool during machining.

[0043] Figures 1A, 1B, and 1C illustrate different variants of a front block 50 of the optical system 1. This block includes a third reflective surface S3 and the collimator block. The latter comprises a first reflective surface S1 and a second reflective surface S2. In these figures, only the corresponding nominal surfaces are shown. These figures illustrate different embodiments in which the second reflective surface S2 has a generally concave, convex, or cylindrical shape.

[0044] The respective nominal surfaces of the first and second reflective surfaces S1, S2 are formed by revolution of, respectively, a first generatrix G1 and a second generatrix G2 around the optical axis 10. The nominal surface of the third reflective surface S3 is a truncated cone centered on the optical axis 10. It is therefore an axisymmetric surface obtained by revolution of a third generatrix G3 around the optical axis 10, this third generatrix G3 being a straight line segment inclined with respect to the optical axis.

[0045] The first and third reflective surfaces S1, S3 have flared shapes in opposite directions along the optical axis 10, the third reflective surface S3 being located above the first reflective surface S1 according to an arbitrary orientation convention.

[0046] Here and for the remainder of the description, we define a three-dimensional orthogonal direct frame (X, Y, Z), where the X and Y axes form a plane perpendicular to the optical axis 10, The X-axis is oriented in the cutting planes, and the Z-axis is oriented parallel to the optical axis 10, from the first reflective surface S1 to the third reflective surface S3. In the following description, the terms "vertical" and "vertically" refer to an orientation substantially parallel to the Z-axis, and the terms "horizontal" and "horizontally" refer to an orientation substantially parallel to the (X, Y) plane. Furthermore, the terms "lower" and "upper" refer to increasing positioning as one moves away from the first reflective surface S1 along the +Z direction. A top view is a view along the -Z direction.

[0047] In figures 1A, 1B and 1C, the abscissas X and the alts Z are given in mm.

[0048] For clarity, the first, second and third generators G1, G2, G3 have been extended beyond the corresponding nominal surfaces on the geometric cross-sectional representations of figures 1A, 1B and 1C.

[0049] In Figure 1A, the first generatrix G1 is a portion of an ellipse and the second generatrix G2 is a portion of a parabola. One of the geometric foci of this ellipse, Fi,1, coincides with a point F on the optical axis 10, which constitutes an optical focus F of the collimator block. A second geometric focus, Fi,2, of the ellipse is located off the optical axis 10, in the plane of section shown in Figure 1A. As it revolves around the optical axis 10, the second geometric focus, Fi,2, generates a focusing line, Li,2, which loops back on itself around the optical axis. This line, Li,2, forms a horizontal circle centered on the optical axis 10.

[0050] The first generator G1 extends radially from the optical axis 10. Thus, one end of the first generator G1 remains fixed on the optical axis 10 during its revolution to form the first reflecting surface S1. This point of fixation generates a point singularity of the first reflecting surface S1, forming a vertex of this surface located on the optical axis. The vertex of the first reflecting surface S1 points towards the focal point F. This first reflecting surface S1 is capable of reflecting the light from the focal point F towards the second reflecting surface S2. The vertex has an elevation strictly greater than that of the focal point F along the Z-axis. The distance separating this vertex from the focal point F defines the working distance.

[0051] The second reflective surface S2 surrounds the first and third reflective surfaces S1, S3 in top view. In any meridian plane, the orthogonal projections of the first, second and third reflective surfaces S1, S2, S3 onto the optical axis 10 follow each other in that order along the +Z direction, with possible overlaps.

[0052] The second generator G2 has a normal orthogonal to the optical axis 10. This normal defines a horizontal median plane during the revolution of the second generator G2 around the optical axis 10. The geometric focus F2,i of the second generator G2 coincides with the second geometric focus of the first generator G1. As it revolutions around the optical axis 10, the geometric focus p2,i traverses the focal line LI,2. This line LI,2 therefore constitutes a focal line LF2 of the second reflecting surface S2. Thus, any light beam originating from the focus F converges towards the focal line LF2 and is subsequently collimated after reflection on the second reflecting surface S2.

[0053] More precisely, when an isotropic point light source is placed at the focal point F, it generates an incident beam on the first reflective surface S1. This beam is reflected from the first reflective surface S1 towards the second reflective surface S2, illuminating it on both sides of the median plane. By reflection from the first reflective surface S1, the incident beam is focused uniformly towards the focal line LI,2. Since this focal line is a focal line of the second reflective surface S2, the beam is collimated after reflection into an axisymmetric collimated beam.

[0054] An isotropic point light source is a point source that emits a uniform intensity in all directions of space. In an optical optimization tool, such a source can be simulated by a uniform angular density of rays in 4TT steradians, each ray being assumed to carry the same energy. Without further specification, the emitted light can be polarized or unpolarized, can cover any spectral band, and can cover the directions of space simultaneously or successively.

[0055] The axisymmetric collimated beam 20 is such that, in each meridian plane including the optical axis 10, it forms two collimated subbeams propagating in different directions, symmetrical to each other with respect to the optical axis 10. Here, the two subbeams are themselves symmetrical to each other with respect to the optical axis 10 (mirror images of each other with respect to the axis). The direction of propagation of each subbeam makes a non-zero angle with the optical axis 10, that is, it is inclined with respect to this axis.

[0056] Thus, the axisymmetric collimated beam 20 is a symmetrical cone of light rotating about the optical axis 10, propagating towards a region of the optical axis 10 opposite the focus F with respect to the horizontal median plane generated by the normal to the second generator G2. The incident beam and the axisymmetric collimated beam 20 propagate mainly in two distinct half-planes delimited by this median plane.

[0057] The axisymmetric collimated beam 20 is reflected on the third reflective surface S3 into a coaxial collimated beam 25 propagating parallel to the optical axis 10. In a horizontal cutting plane, the coaxial collimated beam 25 has an substantially annular shape.

[0058] Since the second generator G2 has a normal orthogonal to the optical axis 10, it follows that the second reflective surface S2 has a substantially vertical arrangement, which allows the horizontal footprint of the optical system 1 to be reduced.

[0059] It also follows from this arrangement of the second reflective surface S2 that the incident beam strikes it at a shallow angle of incidence relative to the horizontal median plane. This compensates for a systematic machining error by translating the second reflective surface S2 parallel to the optical axis 10 relative to the first reflective surface S1.

[0060] This low incidence, combined with the essentially vertical orientation of the second reflecting surface S2, logically results in the cone of light formed by the axisymmetric collimated beam 20 having a large apex angle, which is greater the smaller the angle of the beam incident on the second reflecting surface S2. For example, the apex angle of the axisymmetric collimated beam 20 is greater than or equal to 130°, or even greater than or equal to 150°.

[0061] For example, if we assume a systematic radial offset 5r of the cutting tool relative to the axis of revolution during the machining of the second reflective surface S2, this has the effect of moving this surface S2 away from the optical axis 10 by a distance equal to 5r. The consequence is that the incident beam is offset in Z relative to its expected position on the second reflective surface S2 by a distance of 5r.tan(0 m ), where 0 m is the average angle of incidence of the beam incident on S2. It is then possible to achieve good centering by translating the second reflective surface S2 parallel to the optical axis 10 by the same amount. Thus, this machining error can be compensated to a first order by translating S2 parallel to the Z-axis. The necessary offset is smaller the smaller the angle of incidence 0 m is small. This angle of incidence 0 m It could be in the order of 15°.

[0062] This compensation is all the more effective as all the rays of the incident beam tend to have the same angle of incidence on the second reflecting surface S2. Indeed, due to its topology, the second reflecting surface S2 introduces a phase relationship between the axisymmetric collimated beam 20 and the incident beam which varies spatially. It is therefore important that the footprint of the incident beam on the second reflecting surface S2 also be preserved, in order to recover the desired phase shift to be applied to the axisymmetrically collimated beam 20 by the second reflecting surface S2. This ideal case is approached when the incident beam tends to be collimated in each meridian plane, at the level of the second reflecting surface S2. This corresponds to making the eccentricity of the first generator G1 tend towards 1 and making the focal length of the second generator G2 tend towards infinity. This advantageous limiting situation is illustrated in Figure 1B.

[0063] A low angle of incidence on a reflective surface offers other advantages, including greater tolerance for local slope errors generated by machining. This involves both minimizing the average angle of incidence and increasing the local radius of curvature at every point on the reflective surface.

[0064] For example, inventors have established that a local radius of curvature at each point greater than or equal to 1 cm allows for a local slope accuracy of less than or equal to 100 prad. This tolerance value for the local slope generally corresponds to the resolution adopted by optical optimization software for the design and correction of optical systems. It is also an accuracy achievable through diamond machining processes. It should be noted that for this type of machining, a large local radius of curvature greater than or equal to 1 cm is also advantageous for minimizing the risk of groove formation.

[0065] Based on these considerations, the focal length of the second generator G2 is chosen, for example, to be greater than or equal to 5 mm, or even greater than or equal to 10 mm. The eccentricity of the first generator G1 is preferably chosen to be greater than or equal to 0.9, for example, 0.94.

[0066] Now, variants of the frontal block 50 of figure 1A will be described in relation to figures 1B and 1C. Only the differences with figure 1A are explicitly described.

[0067] In Figure 1B, the first generator G1 is a portion of a parabola and the second generator G2 is a straight line segment parallel to the optical axis 10. The second reflecting surface S2 is therefore a portion of a cylinder whose axis coincides with the optical axis 10. The geometric focus of the parabola Fi,i is located at the focus F of the optical system 1.

[0068] The orientation of the axis of the first parabolic generator G1 defines the angle of incidence 0 m of the beam incident on the second reflective surface S2. This angle incidence 0 m is equal to the angle that the director of the parabola makes with the optical axis 10. 0m is for example equal to 15°.

[0069] In Figure 1C, the first generator G1 is a portion of a hyperbolic branch. The second generator G2 is a parabola. The first generator G1 has a first geometric focus Fi,1 (not shown) around which it wraps (the set of points minimizing the ratio of the distance to the focus Fi,1 with respect to the respective distances of these points to the two directrs). By its revolution around the optical axis 10, the second geometric focus FI,2 of the first generator G1 generates the focusing line LI,2, which loops back on itself around the optical axis. As in Figure 1A, the focal line LF2 of the second reflecting surface S2 coincides with the focusing line LI,2.

[0070] The focal length of the second generator G2 is chosen for example to be greater than or equal to 40 mm, and the eccentricity of the first generator G1 is preferably chosen to be greater than or equal to 1.05, for example equal to 1.07.

[0071] Figures 2 to 4 illustrate advantageous examples of integration of the first, second and third reflective surfaces S1, S2, S3 shown in Figures 1A to 1C, in a monolithic block 40. In these examples, these reflective surfaces S1, S2, S3 are respective surfaces of the monolithic block 40, possibly metallized independently of each other.

[0072] The monolithic block 40 is advantageously made entirely of a single material. It may be a single piece or the result of assembling several elements joined together, for example by adhesive bonding, welding, or other permanent fastening methods. It may include coatings or thin layers made of other materials, particularly surface metallizations. These coatings or thin layers have thicknesses that do not affect the expansion of the monolithic block 40 during temperature changes.

[0073] In the examples shown in Figures 2, 3, and 4, the axisymmetric collimated beam 20 and the coaxial collimated beam 25 each exhibit a substantially uniform irradiance in a cross-section orthogonal to their propagation direction when the isotropic point source is placed at the focus F of the optical system 1. This is achieved by applying optical corrections to the first and second reflecting surfaces S1 and S2. Thus, the rays emanating from the source appear uniformly distributed in the cross-sectional planes of these figures, particularly in comparison with those of Figures 1A to 1C. Consequently, in the examples in Figures 2 to 4, the corrected areas of the first and second reflecting surfaces S1 and S2 deviate from their corresponding nominal areas, although this is not strictly necessary.

[0074] In Figure 2, the monolithic block 40 comprises a central block 40.1, a centering block 40.2, and a peripheral block 40.3. The centering block 40.2 is made of a material transparent in the wavelength range in which the optical system 1 is intended to operate, such as, for example, silica or borosilicate glass. It has a mechanical function but, due to its specific arrangement, must be adapted to transmit the coaxial collimated beam 25.

[0075] The first and third reflective surfaces, S1 and S3, are surfaces of the central block 40.1, here metallized. The second reflective surface, S2, is a cylindrical surface of the peripheral block 40.3, also metallized in this example. S2 is located on an internal vertical wall of the peripheral block 40.3, but it could be located on an external wall if the peripheral block 40.3 were made of a transparent material. In this example, the peripheral block 40.3 is a cylinder whose axis coincides with the optical axis 10, but more complex shapes are possible.

[0076] The central block 40.1 is integral with the centering block 40.2, the latter being fitted into the centering block 40.2 at a specific joint 40.5 in the monolithic block 40. A male portion of the central block 40.1 is fitted into a corresponding female portion of the centering block 40.2 at a central region centered on the optical axis 10. These male and female portions can be bonded or press-fitted to make them integral. Advantageously, as in this case, the central block 40.1 is a fully symmetrical mechanical part of revolution about the optical axis 10, with the possible exception of its male portion fitting into the centering block 40.2.

[0077] The peripheral block 40.3 is mounted with adjustable translation D zparallel to the optical axis 10 with respect to the central block 40.1. The peripheral block 40.3 has an internal wall 43 of substantially cylindrical shape with its axis coinciding with the optical axis 10. The centering block 40.2 has an external wall 42 of substantially cylindrical shape with the same axis. The diameters of these internal 43 and external 42 walls are substantially identical and adjusted to allow the centering block 40.2 to slide parallel to the optical axis 10, for example according to a centering adjustment H6 / h5 ensuring very precise positioning.

[0078] The centering block 40.2 is held in place by fastening means which bind it securely to the peripheral block 40.3. As an example, the inner wall 43 and / or the outer wall 42 may be provided with one or more grooves into which glue is infiltrated after adjustment.

[0079] The 40.2 centering block has a disc shape with two substantially flat and parallel horizontal transmission faces. It is traversed from one side to the other. by the coaxial collimated beam 25 through these faces. It is arranged on one side of the optical system 1 opposite the focal point F with respect to the central block 40.1. This arrangement is preferable to having it on the same side as the focal point F because, in that case, fixing the central block 40.1 would degrade the optical performance of the optical system 1. This arrangement also allows for a satisfactory working distance (distance separating the object from the first optical element). The working distance here is 4 mm.

[0080] This arrangement of the centering block 40.2 requires that it be transparent across the wavelength range of the coaxial collimated beam 25. For example, it has a transmission of 90% or greater in this range. The two transmission faces of the centering block 40.2 are substantially orthogonal to the optical axis 10, and therefore substantially perpendicular to the direction of the coaxial collimated beam 25. Thus, it introduces little or no chromatic aberration, even though its material exhibits a non-zero refractive index dispersion with respect to wavelength. This embodiment, and all those described, are suitable for operation over a wide spectral range, for example, between 500 nm and 1000 nm.

[0081] The adjustment of the relative positions of the centering block 40.2 and the peripheral block 40.3, and consequently of the second reflecting surface S2 relative to the first reflecting surface S1, can be performed on an optical bench. A light source is placed at the focal point F. Micrometer screws can be used to translate the peripheral block 40.3 parallel to the optical axis. The wavefront quality of the coaxial collimated beam 25 is then observed, for example, by interferometric wavefront analysis or by a Shack-Hartmann method. Alternatively, it is possible to base the adjustment on the Strehl ratio or the spread of the PSF obtained at the focal point of a lens or optical system placed opposite the coaxial collimated beam 25.

[0082] The peripheral block 40.3 is moved until an optimal result is obtained (approximately uniform wavefront, high Strehl ratio, or reduced PSF spread). Once the adjustment is complete, the peripheral block 40.3 is fixed to the centering block 40.2.

[0083] Figures 6A to 6C illustrate a fit using the second method (spreading of the PSF). These figures are ray-tracing spot diagrams obtained at the focus F of a frontal block 50 similar to that of Figure 1B (first generator G1 parabolic and second generator G2 in the shape of a straight line segment). An incoherent source is placed at infinity and exhibits a divergence of + / - 0.005°. In each figure, a solid circle represents the extent of the Airy disk at its first zero, obtained with the numerical aperture of the optical system 1. Each point represents the impact of a ray in the horizontal focal plane containing the focus F.

[0084] For these three figures 6A to 6C, optical corrections have been applied to the first and second reflecting surfaces S1, S2 to uniformize the irradiance of the axisymmetric collimated beams 20 and coaxial collimated beams 25. The abscissas and ordinates are given in pm.

[0085] In Figure 6A, we observe that all the rays are contained within the Airy disk. Optical system 1 is therefore diffraction-limited for an incidence parallel to the optical axis. This optically corrected optical system 1 exhibits few or no aberrations.

[0086] In Figure 6B, a deformation representative of a systematic radial offset 5r of 10 pm of the cutting tool relative to the axis of revolution at the time of machining was applied to the second reflective surface S2. This axis of revolution coincides with the optical axis 10. The resolution is considerably degraded compared to the situation in Figure 6A, with numerous rays impacting the focal plane outside the solid circle representing the Airy disk. The PSF exhibits a spread with a diameter approximately equal to 14 pm.

[0087] In Figure 6C, the second reflective surface S2 has been displaced by 5z = 5r.tan(0 m) relative to the first reflective surface S1, along the +Z direction. This translation has allowed us to recover good resolution, close to that obtained in Figure 6A, with few ray impacts outside the solid circle. Thus, the translational adjustment of the second reflective surface S2 has compensated for the machining accuracy error.

[0088] Figures 7A and 7B illustrate an optical correction method for the first and second reflecting surfaces S1 and S2, enabling the production of axisymmetric collimated beams 20 and coaxial beams 25 with substantially uniform irradiance in their respective cross-sections. A result of this correction method is shown in Figures 8A and 8B.

[0089] Figure 7A shows two extreme rays emanating from the focal point F, intercepting the entrance pupil of the optical system 1 on two diametrically opposite edges thereof, and two central rays, substantially parallel to the optical axis 10, intercepting the first reflecting surface S1 on either side of its vertex in the plane of section. Two intermediate rays, symmetrical to each other in the plane of section with respect to the optical axis 10, are also shown. These intermediate rays make an angle ei = OjEmax with the optical axis 10, where ai is a real number between 0 and 1. ma x = arcsin(NA) where the angle of opening that the extreme rays make with the optical axis 10, NA being the numerical opening of the optical system 1.

[0090] Each central and extreme ray intercepts the second reflective surface S2 at, respectively, a height zo and a height z max along the Z-axis. The intermediate rays intercept this surface S2 at a height Zj. After reflection on the first reflective surface S1, the central, intermediate, and extreme rays have angles respectively equal to 0°, 1 / 2 and 0°. ma x with the Y axis.

[0091] The first generator G1 being a parabola with geometric focus F, the extreme, central and intermediate rays are parallel to each other after reflection on the first reflecting surface S1 in their respective propagation half-planes delimited by the optical axis 10. The second reflecting surface S2 being cylindrical without optical correction, these rays also propagate parallel within the axisymmetric collimated beams 20 and coaxial 25.

[0092] In the situation shown in Figure 7A, the angles 0o, 0j and 0 ma x are therefore equal. Due to the shapes and arrangement of the first and second reflective surfaces 51 and S2, there is no constant proportionality factor linking, for any pair of rays, the differences in their dimensions (Zj - Zj) at the level of the second reflective surface 52 to their angular emission deviations (Ej - Ej) at focus F. Thus for an isotropic point source placed at focus F, the irradiance of the axisymmetric collimated beam 20 is not uniform at the level of the second reflective surface S2.

[0093] To standardize this irradiance, the procedure illustrated in Figure 7B is followed. The value ai is incremented progressively, and the local slope of the first reflective surface S1 is corrected step by step at the point of impact of the corresponding intermediate ray so that it is reflected at an angle 0'j, allowing it to reach the second reflective surface S2 at the elevation z'j = aj.(z max- zo) + zo. Following this operation, the angles of incidence 0'j on the second reflecting surface S2 are no longer constant. It is therefore necessary to correct the wavefront of the axisymmetric collimated beam 20 so that its subbeams in the cutting plane are collimated. The necessary phase correction is achieved by deforming the second reflecting surface S2. These successive corrections of the first and second reflecting surfaces S1, S2 can be repeated as many times as necessary in an iterative process until a satisfactory result is obtained.

[0094] For this embodiment and the others presented in the description, the monolithic block 40 is advantageously made entirely of the transparent material of the centering block 40.2. The optical systems illustrated in Figures 1A to 1C and shown in Since the components in the assemblies of figures 2 to 4 are invariant under homothety, the fabrication of the monolithic block 40 in a single material ensures good optical performance over a wide temperature range, as the expansion of the single material has the same homothetic effect for all reflective surfaces S1, S2, S3. Thus, an optical system 1 limited by diffraction at a given temperature remains so over a wide range of temperature variations, making it suitable for space or cryogenic applications.

[0095] A scaling-invariant optical system is one that preserves the angular orientation of the rays propagating through it, on either side of each of its optical surfaces, when a global scaling is applied, particularly that resulting from thermal expansion when all the elements are made of the same material. For a collimator, this means that a beam collimated by it before scaling remains collimated after scaling.

[0096] Figure 3 illustrates a first variant of the realization of the monolithic block 40. Only the differences with figure 2 are explicitly described.

[0097] Here, the central block 40.1 is integral with the centering block 40.2 because the latter extends at least one upper portion of the former, so that together they form a single mechanical component. As in the example in Figure 2, the centering block 40.2 is disc-shaped, here without any visible boundary with the central block 40.1 at its central portion.

[0098] The central block 40.1 includes a clearly defined element 41 centered on the optical axis 10. This element has an internal wall that houses the third reflecting surface S3. This wall can be a truncated cone. Advantageously, as in this case, it is a cone arranged so that the extreme rays reach approximately its apex after reflection from the second reflecting surface S2. Thus, the coaxial collimated ring beam 25 has an internal diameter that is approximately zero.

[0099] A cone is a limiting case of a truncated cone, in which the smaller of its two bases has a diameter tending towards 0, or even zero. [ooioo] When the recess 41 is filled with a material having a lower refractive index than the material constituting the central block 40.1, the axisymmetric collimated beam 20 can be reflected by total internal reflection on the third reflective surface S3. Alternatively, the walls of the recess 41 can be metallized to form the third reflective surface S3. This latter extends here to the apex of the cone formed by the recess 41. [ooioi] To facilitate the manufacture of the central block 40.1, it can be made in two parts, the upper part receiving the recess 41 and a lower part closing it, this lower part receiving the first reflective surface S1. If necessary, the lower and upper parts are made to be joined together, for example by gluing.

[0102] Figure 4 illustrates a second variant of the realization of the monolithic block 40. Only the differences with figure 2 are described explicitly.

[0103] The front block 50 is purely catoptric. In this variant, the centering block 40.2 can be made of an opaque material. Here it is made of metal. It is a spider (or "spider" in English) comprising, in this example, four arms distributed angularly and uniformly in a plane parallel to the (X, Y) plane.

[0104] The branches can have any shape; here they are in a circular arc in top view. The branches can be guided in translation by straight vertical trenches arranged in the inner wall 43 of the peripheral block 40.3. Alternatively, they can be fixed by their outer ends to a cylinder of the centering block 40.2, comprising the outer wall 42 corresponding to the inner wall 43 of the peripheral block 40.3.

[0105] Figure 5 is an example of an axisymmetric non-centered optical system 1 comprising a rectifier block 70 adapted to invert the exit pupil of the front block 50. This front block 50 can be any of those described in connection with Figures 1A to 1C, and 2 to 4. The rectifier block 70 transforms the coaxial collimated beam 25 into an output beam 26.

[0106] The erecting block 70 aims to correct the radial inversion of the exit pupil of the frontal block 50. This pupil is radially inverted in the sense that the marginal rays from focus F are located in a central region of the coaxial collimated beam 25, and the paraxial rays in a peripheral region of this beam 25.

[0107] The rectifier block 70 comprises a first axiconical lens 71 and a second axiconical lens 72. These two axiconical lenses 71, 72 are arranged end-to-end along the optical axis 10. Each axiconical lens 71, 72 has a substantially flat face opposite a conical face whose apex is centered on the optical axis 10. In this example, the axiconical lenses 71, 72 are identical. Their respective flat faces are arranged opposite each other, their respective conical faces being separated from each other by the flat faces of the first and second axiconical lenses 71, 72.

[0108] The first axiconical lens 71 is arranged upstream of the coaxial collimated beam 25, relative to the second axiconical lens 72. It has a marginal focus Fm located upstream of the conical face of the second axiconic lens 72, this in order to reverse the pupil of the front block 50. By this arrangement, any ray of the coaxial collimated beam 25 propagating along a straight line parallel to the optical axis 10 upstream of the rectifier block 70, propagates downstream along a straight line also parallel to the optical axis 10, these two lines extending in Z on either side of the optical axis 10 in the same meridian plane.

[0109] The marginal focus Fm of an axiconical lens corresponds to the point of intersection of the marginal optical rays parallel to its optical axis; that is, the rays furthest from this axis that can propagate within the optical system in which the axiconical lens is arranged. Here, the marginal rays are the rays of the coaxial collimated beam 25 originating from the isotropic point source placed at the focus F, intercepting the edge of the pupil of the optical system 1. Preferably, the marginal focus Fm is located inside the second axiconical lens 72, on the optical axis 10 near the apex of its conical face. Thus, the output beam 26 has a virtually zero internal diameter at the exit of the rectifier block 70. Its internal diameter increases as the second axiconical lens 72 moves away from the first axiconical lens 71 and its marginal focus Fm. [oom] In this configuration (identical axiconic lenses and marginal focus Fm located upstream of the conical face), the erecting block 70 introduces little or no angular chromatic aberration, although edge iridescence of the output beam 26 may be observed, its extent decreasing as the apex angle of the conical face increases. The apex angle, or aperture angle, is the internal angle separating the intersections of the conical face with a meridian plane containing its axis. This apex angle is preferably between 90° and 179°, for example, 170°.

[0112] Preferably, the diameter of the first and second axiconic lenses 71, 72 does not limit the pupil of the optical system 1. It is, for example, between 1 cm and 5 cm. These axiconic lenses 71, 72 can be made of glass.

[0113] Figures 9B and 9C show ray distributions within a cross-section of the output beam 26, obtained with an isotropic point light source placed at the focus F of an optical system 1 similar to that illustrated in Figure 5. Figure 9A shows the angular distribution of rays emitted by the light source, limited to those intercepting the entrance pupil of the optical system 1, i.e., limited to the acceptance cone of the optical system 1. In this figure, the abscissas and ordinates are given as fractions of the aperture angle of the optical system 1. In Figures 9B and 9C, the abscissas and ordinates are given in mm. For these simulations, the front block 50 is of the type of that in figure 1 B (first reflective surface S1 parabolic, second reflective surface S2 cylindrical).

[0114] In Figure 9B, no optical correction is applied to the first and second reflective surfaces S1, S2. Aberrations of the front block 50 and a non-uniform irradiance of the coaxial collimated beam 25 lead to the formation of an output beam 26 exhibiting a higher irradiance on its central part than on its outer periphery.

[0115] In Figure 9C, respective optical corrections were applied to the first and second reflective surfaces S1, S2 according to the method taught in connection with Figures 7A and 7B. A substantially uniform ray distribution is obtained within the output beam 26, including a blind spot at the center with a diameter less than 5% of the diameter of the output beam 26. The application of these optical corrections has the effect of widening the output beam 26 from approximately 5 mm to approximately 8.5 mm.

[0116] Figures 10A to 10C illustrate the widening of the object field obtained through optical corrections, such as those implemented according to the method described in connection with Figures 7A and 7B. These three figures represent spot diagrams obtained with a uniform point source, as illustrated in Figure 9A, in the image plane of an ideal converging lens with a focal length of 200 mm placed opposite the output beam 26. The diagrams are established for 5 positions of the light source, located in a plane parallel to the (X, Y) plane including the focus F, at 0 pm, 1 pm, 2 pm, 4 pm and 8 pm respectively from this focus along the X-axis. The optical system 1 is similar to that described in connection with Figures 1B and 5 (first reflecting surface S1 parabolic, second reflecting surface S2 cylindrical).

[0117] The abscissa and ordinate are given in mm. The first zero of the Airy spot obtained for the same aperture as optical system 1, and a wavelength of 532 nm, is represented by a solid circle.

[0118] The spot diagrams in Figures 10A and 10B are obtained with the same optical system 1, except that optical corrections have been applied to the first and second reflecting surfaces S1, S2 in the case of Figure 10B. The magnification and numerical aperture are respectively equal to 40 and 0.7 for Figures 10A and 10B, and to 30 and 0.6 for Figure 10C.

[0119] In Figure 10A, no optical correction has therefore been applied to the first and second reflective surfaces S1, S2. It can be seen that the object field is practically zero in size. As soon as the light source moves away from the focus F, the the impulse response spreads rapidly to extend well beyond the region defined by the Airy disk.

[0120] In Figures 10B and 10C, it is estimated that optical system 1 is limited by diffraction within an object field of + / - 5 pm.

[0121] As an example, figures 8A and 8B show the optical corrections applied to the first reflective surface S1 (figure 8A) and to the second reflective surface S2 (figure 8B) of optical system 1 which were used to obtain the spot diagrams of figure 10B.

[0122] Curves 101 and 111 (dashed lines) represent respectively the first generatrix G1 and the second generatrix G2 in a plane parallel to the (Y, Z) plane – the cutting plane of figures 1A to 1C, and 2 to 5. Curves 102 and 112 (solid lines) represent the corrected surfaces of the first and second reflective surfaces S1 and S2, respectively. These are symmetrical about revolutions around the optical axis 10; the solid curves therefore represent their respective generatrices.

[0123] The radial coordinates in mm in the +Y direction are given on the x-axis. The Z-axis is given in mm.

[0124] Figure 8B shows that the optical correction is sigmoidal in shape, meaning that the radial difference between the corrected area and the nominal area of ​​the second reflecting surface S2 follows a sigmoidal distribution as a function of the z-coordinate. It has been established that the optimal optical correction of the second reflecting surface S2 is sigmoidal regardless of whether the second generatrix G2 is a straight line segment or a portion of a parabola. The radial distance is defined as the distance along a straight line orthogonal to the optical axis 10.

[0125] Curves 103 and 113 (lines of dots) represent the local radius of curvature in mm (logarithmic scales along the second ordinate axis in Figure 8A and the second abscissa axis in Figure 8B). It can be seen that the local radius of curvature of the corrected surfaces of the first and second reflective surfaces S1 and S2 is greater than 1 cm at every point, except for a region of the first reflective surface S1 located within a radius approximately 200 pm from its vertex. Thus, good accuracy of the local slope can be obtained after machining.

[0126] Specific embodiments have just been described. Various variations and modifications will be apparent to those skilled in the art. For example, all corrected surfaces in the presented embodiments are symmetrical of revolution so that they can be easily produced by turning. Those skilled in the art may, however, find it advantageous to to apply other types of optical corrections leading to non-axisymmetric corrected surfaces.

Claims

DEMANDS 1. Non-centred axisymmetric catadioptric optical system (1), comprising: o an optical axis (10), o a front block (50) comprising a collimator block and a third reflective surface (S3) in the shape of a truncated cone centred on the optical axis (10), - the collimator block being such that it comprises a focus (F), a first reflective surface (S1) and a second reflective surface (S2), and such that it is configured to form an axisymmetric collimated beam (20) inclined with respect to the optical axis (10) by successive reflections on the first reflective surface (S1) and the second reflective surface (S2) when an isotropic point light source is placed at the focus (F); - the third reflective surface (S3) being configured to direct the axisymmetric collimated beam (20) parallel to the optical axis (10), thus generating a coaxial collimated beam (25); - the collimator block being further such that: > the first reflective surface (S1) has a nominal surface formed by revolution of a first conical generatrix (G1) around the optical axis (10), a geometric focus of the first generatrix (G1) being coincident with the focus (F) of the collimator block, the first generatrix (G1) extending radially from a vertex of the first reflective surface (S1) located on the optical axis (10); > the second reflective surface (S2) has a nominal surface formed by revolution around the optical axis (10), of a second straight or parabolic generatrix (G2), having a normal orthogonal to the optical axis (10).

2. Optical system (1) according to claim 1, in which the axisymmetric collimated beam (20) forms a cone of light with an apex angle greater than or equal to 130°.

3. Optical system (1) according to claim 1 or 2, wherein the first reflective surface (S1) has an optical correction of its nominal surface so as to uniformize the irradiance of the axisymmetric collimated beam (20) at the level of the second reflective surface (S2), and wherein the second reflective surface (S2) has an optical correction of its nominal surface so as to uniformize the phase of the axisymmetric collimated beam (20) after reflection on the second reflective surface (S2).

4. Optical system (1) according to any one of the preceding claims further comprising a monolithic block (40) made of a single material, in which the first reflective surface (S1) and the second reflective surface (S2) are surfaces of the monolithic block (40).

5. Optical system (1) according to claim 4, wherein the monolithic block (40) comprises a central block (40.1) and a peripheral block (40.3) surrounding the central block (40.1), the peripheral block (40.3) being mounted adjustable in translation parallel to the optical axis (10) relative to the central block (40.1), the first and third reflective surfaces (S1, S3) being surfaces of the central block (40.1), and the second reflective surface (S2) being a surface of the peripheral block (40.3).

6. Optical system (1) according to claim 5, wherein the monolithic block (40) comprises a centering block (40.2) integral with the central block (40.1), the centering block (40.2) and the peripheral block (40.3) comprising corresponding surfaces of revolution (42, 43), allowing one to be guided relative to the other in translation parallel to the optical axis (10).

7. Optical system (1) according to claim 6, wherein the monolithic block (40) is made of metal, the front block (50) is purely catoptric and the centering block (40.2) is a spider.

8. Optical system (1) according to claim 6, wherein the monolithic block (40) is made of a transparent material, and wherein the centering block (40.2) is a disc inside which a part of the central block (40.1) is fitted.

9. Optical system (1) according to claim 6, wherein the monolithic block (40) is made of a transparent material, the centering block (40.2) extends the central block (40.1), and the central block (40.1) has a recess (41) in the shape of a truncated cone hosting the third reflective surface (S3).

10. Optical system (1) according to any one of the preceding claims, wherein the first generator (G1) is a portion of a parabola, and the second generator (G2) is a straight line segment parallel to the optical axis (10).

11. Optical system (1) according to any one of the preceding claims further comprising a rectifier block (70) comprising two head-to-tail axicon lenses (71, 72) centered on the optical axis (10), the rectifier block (70) being arranged to receive the coaxial collimated beam (25).

Images