Spatial light modulator device for the modulation of a wave field with complex information

Abstract

A three-dimensional light modulator, of which the pixels are combined to form modulation elements. Each modulation element can be coded with a preset discrete value such that three-dimensionally arranged object points can be holographically reconstructed. The light modulator is characterized in that assigned to the pixels of the modulator are beam splitters or beam combiners which, for each modulation element, combine the light wave parts modulated by the pixels by means of refraction or diffraction on the output side to form a common light beam which exits the modulation element in a set propagation direction.

Claims

1. A complex spatial light modulator comprising: a polarization phase modulator which modifies incident light and separates the incident light into a first beam having a first polarization and a first phase, and a second beam having a second polarization and a second phase, and which outputs the first beam and the second beam; and a beam synthesizer comprising a prism structure formed of an optical anisotropic material, the optical anisotropic material having a first refractive index with respect to the first beam having the first polarization and a second refractive index, different from the first refractive index, with respect to the second beam having the second polarization, where the beam synthesizer combines the first beam and the second beam and outputs a beam.

2. The complex spatial light modulator according to claim 1, further comprising a first polarizer, disposed on an optical path before the polarization phase modulator, which transforms a polarization of a beam incident thereon into a beam having the first polarization.

3. The complex spatial light modulator according to claim 1, wherein the polarization phase modulator comprises: a phase-type spatial light modulator comprising a first pixel that modulates a phase of a portion of the incident light and outputs a beam having the first phase and a second pixel that modulates a phase of a portion of the incident light and outputs a beam having the second phase.

4. The complex spatial light modulator according to claim 1, wherein the beam synthesizer comprises: a first prism array comprising at least one prism element that has a light-incident surface which is normal to an optical axis of the incident light, and a first inclined surface which is inclined with respect to the light-incident surface; and a second prism array that is spaced apart from the first prism array, and that comprises at least one prism element having a light-exit surface which is parallel with the light-incident surface, and a second inclined surface facing away from the first inclined surface and which is inclined with respect to the light-incident surface.

5. The complex spatial light modulator according to claim 4, wherein the first prism array is formed of an isotropic material and the second prism array is formed of an optical anisotropic material having a first refractive index, with respect to the first beam having the first polarization, that is substantially the same as a refractive index of the surrounding material, and having a second refractive index, with respect to the second beam having the second polarization, that is chosen to be different to the first refractive index.

6. The complex spatial light modulator according to claim 4, wherein an angle by which the first inclined surface and the second inclined surface are inclined with respect to the light-incident surface is such that an optical path of the first beam is changed due to refraction at the first inclined surface and refraction at the second inclined surface, and thus the optical path of the first beam is modified, by the beam synthesizer, to be the same as an optical path of the second beam which passes through the first inclined surface and the second inclined surface without refraction.

7. The complex spatial light modulator according to claim 1, wherein the beam synthesizer comprises: a first prism array comprising at least one prism element that has a light-incident surface which is normal to an optical axis of the incident light, and a first inclined surface which is inclined with respect to the light-incident surface; a second prism array that is spaced apart from the first prism array, wherein the second prism array comprises at least one prism element having a second inclined surface facing away from the first inclined surface and which is inclined with respect to the light-incident surface, and a light-exit surface which is parallel to the light-incident surface; and a structure which fills a region between the first prism array and the second prism array.

8. The complex spatial light modulator according to claim 7, wherein the structure is formed of an isotropic material, and where the first prism array is formed of an isotropic material and the second prism array is formed of an optical anisotropic material with a first refractive index, with respect to the first beam having the first polarization, and with a second refractive index, with respect to the second beam having the second polarization, that is different from the first refractive index.

9. The complex spatial light modulator according to claim 8, wherein the beam combiner comprises deflection layers, where a first light deflection layer changes the path of the first modulation light into a direction of the path of the second modulation light, where a second deflection layer changes the path of the first modulation light so that the first modulation light is outputted of the beam combiner parallel to the second modulation light.

10. The complex spatial light modulator according to claim 9, wherein the an angle by which the first inclined surface and the second inclined surface are inclined with respect to the light-incident surface is such that an optical path of the first beam is changed due to refraction at the first inclined surface and refraction at the second inclined surface, and thus the optical path of the first beam is modified, by the beam synthesizer, to be the same as with an optical path of the second beam which passes through the first inclined surface and the second inclined surface without refraction.

11. The complex spatial light modulator according to claim 1, wherein the beam synthesizer comprises: a first prism that has a light-incident surface which is normal to an optical axis of the incident beam, and a first inclined surface which is inclined with respect to the light-incident surface; a second prism that has a second inclined surface which is inclined with respect to the light-incident surface, and a light-exit surface which is parallel to the light incident surface, which is spaced apart from the first prism along a direction of the optical axis, and which is offset from the first prism; and a structure which fills a region between the first prism and the second prism.

12. The complex spatial light modulator according to claim 11, wherein the structure is formed of an isotropic material; and the first prism is formed of an isotropic material and the second prism array is formed of an optical anisotropic material with a first refractive index, with respect to the first beam having the first polarization, and with a second refractive index, with respect to the second beam having the second polarization, that is different from the first refractive index.

13. The complex spatial light modulator according to claim 11, wherein an angle by which the first inclined surface and the second inclined surface are inclined with respect to the light-incident surface is such that an optical path of the first beam is changed due to refraction at the first inclined surface and refraction at the second inclined surface, and thus the optical path of the first beam is modified, by the beam synthesizer, to be the same as with an optical path of the second beam which passes through the first inclined surface and the second inclined surface without refraction.

14. A holographic three-dimensional image display device comprising, a light source unit; the complex spatial light modulator of claim 1; and a controller which controls the complex spatial light modulator, to modulate a light beam from the light source unit according to three-dimensional information.

15. The holographic three-dimensional image display device according to claim 14, wherein the light source unit outputs a beam that is polarized in the first polarization direction.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Now, there are a number of possibilities for embodying and continuing the teachings of the present invention. To this end, reference is made on the one hand to the dependent claims that follow claim 1, and on the other hand to the description of the preferred embodiments of this invention below including the accompanying drawings. Generally preferred physical forms and continuations of the teaching will be explained in conjunction with the description of the preferred embodiments of the invention and the accompanying drawings. The Figures are schematic drawings, where

(2) FIG. 1 shows a detail of spatial light modulator device of an embodiment according to this invention,

(3) FIG. 2 shows a first embodiment of the optical multiplexing means for spatial multiplexing of exiting modulated light wave portions with an array of micro-prisms and a volume grating,

(4) FIG. 3 shows a second embodiment of the optical multiplexing means for spatial multiplexing of exiting modulated light wave portions with an array of micro-prisms and a volume grating, where diffracted light is used,

(5) FIG. 4 shows a third embodiment of the optical multiplexing means for spatial multiplexing of exiting modulated light wave portions, where diffracted light is used and where the non-diffracted light is filtered out by a spatial frequency filter with an aperture mask,

(6) FIG. 5 shows a fourth embodiment of the optical multiplexing means for spatial multiplexing of exiting modulated light wave portions with a polarising light wave splitter,

(7) FIGS. 6 to 8 show embodiments of the optical multiplexing means with a polarising beam splitter which is compensated in respect of changes in the light wavelengths,

(8) FIG. 9 illustrates the functional principle of a prior art polarisation grating according to cit. [2],

(9) FIG. 10 shows the beam path in a volume grating (non-symmetrical),

(10) FIG. 11 shows the beam path in an arrangement with polarisation gratings (symmetrical),

(11) FIG. 12 shows an embodiment of the present invention,

(12) FIGS. 13 and 14 each show a further embodiment of the present invention,

(13) FIG. 15 illustrates space division multiplexing of phase pixels which are combined to form complex-valued pixels,

(14) FIG. 16 shows a neutral density apodisation function (left) and a laterally offset, colour-sensitive apodisation function (right), and

(15) FIGS. 17 to 22 each show a further embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

(16) Since all optical multiplexing means for each of the modulator cells of the modulation array have the same structure, only a single modulator cell of the modulation array will be shown in the drawings referred to below in order to keep the drawings simple and comprehensible.

(17) For the same reason, the optical multiplexing means will be described below with the example of a modulation array with regularly structured modulator cells, where each modulation element comprises two adjacent modulator cells of the modulation array. A typical example of such a spatial light modulator device is a spatial phase-modulating light modulator designed to implement the above-mentioned two-phase encoding method. Generally, the structure can also correspond to a modulation element which comprises more than two modulator cells.

(18) The following embodiments can also be adapted in a comparable way to amplitude modulation. In the latter case, a phase-shifting optical layer would additionally be required for at least one modulator cell per modulation element. If the bias encoding method is employed, a fix phase shift of ?/2 is required for one of the two modulator cells, and if the Burckhardt encoding method is employed, phase shifts of 2?/3 and 4?/3 are required for two of the three modulator cells.

(19) FIG. 1 shows a modulation element ME with a first modulator cell P01 and a second modulator cell P02, both of which being arranged next to each other in a modulation array. A light wave field LW which is capable of generating interference illuminates the modulation element ME in the modulation array. A modulator control unit CU encodes each modulator cell P01, P02 with a phase component of a complex hologram value, so that each modulator cell P01, P02 emits a discretely modulated light wave portion LWP.sub.1 and LWP.sub.2, respectively, with parallel optical axes a.sub.01, a.sub.02 in a direction D in order to generate a holographic reconstruction. According to this invention, an array of optical multiplexing means is arranged as close as possible to the modulator cells P01, P02. The optical multiplexing means comprise a structure of wave deflection means U1, U2, which are spatially assigned to the modulator cells P01, P02. The wave deflection means U1, U2 have optical axes which differ from each other and which are oriented in respect of each other such that the light wave portions LWP.sub.1 and LWP.sub.2, which come from the same modulation element ME, are combined in the array of optical multiplexing means and form a wave multiplex of a modulated common light wave portion LWP.sub.0 with a common optical axis a.sub.0.

(20) According to a preferred embodiment of the present invention, the array of optical multiplexing means comprises an optical plate unit of stacked optical plates. The optical plates can for example comprise multiple transparent polymer layers with a presettable optical propertyin particular birefringence.

(21) FIG. 2 shows a first embodiment of such a plate unit, which comprises a micro prism array PA which provides for the modulator cells P01, P02 of each modulation element ME a micro-prism which realises a desired optical wave deflection function for the modulator cells P01, P02. This optical plate unit also combines the light wave portions LWP.sub.1 and LWP.sub.2 of the modulation elements to form a wave multiplex of a modulated common light wave portion LWP.sub.0. This is achieved in that a volume hologram BG, also known as Bragg hologram, is additionally arranged in the optical path of the optical plate unit. This volume hologram BG has the task of preventing an intersection of the propagating light wave portions LWP.sub.1 and LWP.sub.2 and to lead the two light wave portions LWP.sub.1 and LWP.sub.2, which have been modulated by the modulator cells P01 and P02 of a modulation element, into direction D without any difference in the optical path lengths. The volume hologram BG is encoded such that is directs light waves with defined wavelengths tightly in a greatly limited deflection angle or exit angle. Any light wavelengths which are required for a colour reconstruction must be considered as defined wavelengths, e.g. the colours red, green and blue.

(22) FIG. 3 shows a second embodiment of the optical plate unit of FIG. 2. The two embodiments differ in the provision or angle of incidence of the light wave field LW which is capable of generating interference. In the embodiment according to FIG. 3, the light which is capable of generating interference hits the spatial light modulator device or the modulator cells P01, P02 at an oblique angle to the optical axis, so thatas a consequence of the oblique angle of incidencethe first diffraction order can be used for reconstruction. In the embodiment according to FIG. 2, the light wave field which is capable of generating interference hits the spatial light modulator device parallel with the optical axis, so that the zeroth diffraction order can be used for reconstruction.

(23) Referring to FIG. 4, an additional telescopic filter array (TFA) with an aperture mask AP between two afocally arranged lens arrays systems L1, L2 allows to suppress undesired light portions, e.g. those of neighbouring spatial diffraction orders in respect of the direction of incidence of the light wave field of the 0th diffraction order or of unused periodicity intervals. At the same time, the two afocally arranged lens arrays systems L1, L2 allow the fill factor of the modulator cells of the modulation element ME in the modulation array to be raised due to an optical magnification.

(24) FIG. 5 shows another embodiment of the present invention, where a polarising light wave splitter Pol combines the light wave portions of each modulation element. The optical multiplexing means for spatial multiplexing use a plate with polarisation elements Spol and Ppol, which assign to each light wave portion of a modulator cell P01, P02 in the modulation element a discrete light polarisation, combined with a birefringent coplanar plate BP, which assigns to all modulated light wave portions LWP.sub.1, LWP.sub.2 of a modulation element a discrete inclined optical axis. The optical axes of all light wave portions are inclined in respect of each other and the strength of the coplanar plate BP is chosen such that all light wave portions are superposed at its exit-side interface.

(25) A polarising light wave splitter, as shown in FIG. 6, is very sensitive to changes in the wavelength which is chosen to generate the holographic reconstruction. A lateral offset which is dependent on the wavelength of the used light and a change in the phase relation of the light are obtained.

(26) FIGS. 6 and 7 show two embodiments which illustrate the fundamentals of realising a self-compensating beam splitter double plate according to FIG. 8. Vg1 and Vg2 denote volume gratings which serve as beam splitters.

(27) The distance d between the two parallel grating planes must be d=a/(2.Math.cos(?/2)), i.e. 0.57735 ?m per ?m modulator cell width, in order to achieve a complete superposition of the light wave portion TE of modulator cell 1 and light wave portion TM of modulator cell 2, both of which having the width a, downstream of the planar polarising beam splitter Vg2.

(28) Assuming 50 ?m wide modulator cells, a thickness d=28.87 ?m can be achieved with the 0?/60? geometry of the polarising beam splitters, while in comparison with that a Savart plate must have a minimum thickness of 385.8 ?m if ?n=0.2 is to be obtained.

(29) The pointing vectors of the polarised light wave portions TE and TM will be parallel downstream of the polarising beam splitter if they were parallel upstream of the polarising beam splitter. The parallelism of the exiting beams should therefore not be a problem here.

(30) However, wavelength fluctuations of the light are problematic. Given a modulator cell width of 30 ?m and, consequently, a thickness of the polarising beam splitter double plate of 17.32 ?m, a wavelength deviation of ??=1 nm will result in a relative phase difference between the two superposed modulator cells of about 2?/10. In order to solve this problem, it is possible to choose a polarising beam splitter geometry with lower diffraction angle.

(31) A possible polarising beam splitter deflection geometry (with polarising beam splitters Vg1, Vg2) is 0?/48.2?, as shown in FIG. 6. The light wave portion TE is deflected, while the light wave portion TM is not deflected in this example. FIG. 7 illustrates a 0?/41.2? polarising beam splitter (Vg1, Vg2) which transmits TE-polarised light without deflection, while it diffracts or deflects TM-polarised light.

(32) Given a modulator cell width of a=50 ?m and a maximum permitted distance to a plane of EW prisms (not shown), as regards cross-talking among neighbouring modulator cells, of D.sub.max=5?a=250 ?m, it follows ?.sub.min=arctan(a/D.sub.max)=arctan(0.2)=11.31?. At ?n=0.2, the Savart plate achieves about 7.4?.

(33) Since the possible polarising grating beam splitter geometries are mathematically terms of a series, there are also usable angles in a range around 11?. The required refractive index variation is then very high though, i.e. usage of 11? as polarising beam splitter geometry is deemed rather unrealistic.

(34) However, a polarising beam splitter geometry of 0?/33.557? is not unlikely to be realised in practice, where there must still be a refractive index variation reserve for RGB multiplexing. At a stability of the wavelength of ??=1 nm, this geometry would correspond with a relative phase difference of the combined modulator cell beams of ??<2?/20.

(35) There are a number of possibilities to compensate the effect of a possibly drifting key wavelength.

(36) One possibility is to use the generated summed signal of the two combined modulator cells in order to compensate the phase shift simply and easily during operation. For this, the phase of one modulator cell can for example be shifted such that as a result a certain target intensity is achieved. This produces a value for a corrective phase to be introduced.

(37) Further, it is possible to introduce a set of phase shifts, i.e. at least three, in order to determine with the help of phase-shifting interferometry the relative phase of the combined modulator cell with an accuracy of <2?/512.

(38) In displays which only comprise few light sources, it presents itself to use two diodes per wavelength and light source which have spectrally different characteristics. If the characteristic lines are known, then the wavelength can be determined with an accuracy of <0.1 nm from the signals of the diodes. This principle is for example employed in the wavelength measuring device WaveMate? supplied by the company Coherent.

(39) When the key wavelength is known, the relative phase to be set in the combined modulator cells can be corrected directly if the key wavelength drifts. This should result in a remaining error of <2?/256 when setting the relative phase in the combined modulator cells.

(40) The above-mentioned approaches for online correction can be combined with each other in order to improve the measurement accuracy and thus to compensate the effect of a wavelength drift. Irrespective of that, a laser can also be stabilised to ??<0.1 nm.

(41) FIG. 8 shows an embodiment of a compensated polarising beam splitter. Here, TM-polarised light is diffracted and TE-polarised light is not with the first two beam splitters Vg1 and Vg2, which compares to the arrangement shown in FIG. 7. Further, TE-polarised light is diffracted and TM-polarised light is not with the other two beam splitters Vg3 and Vg4, which compares to the arrangement shown in FIG. 6. The distance between the individual beam splitters Vg1 to Vg4 can here be less than in the embodiments shown in FIGS. 6 and 7, because only half the beam offset must be achieved for the TE- and TM-polarised light.

(42) FIG. 8 further shows how a compensation of a drift of the light wavelength can be achieved. The compensation of ??.sub.rel(??) is based on the fact that this effect is equally distributed over the combined modulator cells. Since the required retardation layers (one structured and one plane, unstructured) are only about 1.5 ?m thick, the resultant distance between SLM and polariser WGP is DD<2a (applies to a ?50 ?m), even if the thickness of the volume gratings Vg1, Vg2, Vg3 and Vg4, which are arranged in four planes, cannot be neglected (each about 10 ?m thick). If a=70 ?m, then the resultant thickness DD<a. If a=20 ?m, DD is less than 3a, which makes it still possible for small modulator cells to use polarisers other than wire grid polarisers.

(43) Referring to FIG. 6, a ?/2 plate is drawn between the modulator cell 2 and the volume grating Vg1. Providing a ?/2 plate becomes necessary when the light which falls on the modulator cells 1, 2 only has one presettable polarisation, e.g. a linear TE polarisation. In that case, the polarisation of the light which passes through the modulator cell 2 is turned by the ?/2 plate by 90 degrees, so that the light which passes through the modulator cell 1 is given a polarisation that is perpendicular to that of the light which passes through the modulator cell 2. If the light which falls on the modulator cells 1, 2 already has a perpendicular polarisation, the provision of a ?/2 plate between the modulator cell 2 and the volume grating Vg1 is not necessary. In other words, it is important that the light which passes through the modulator cell 1 has a differente.g. perpendicularpolarisation than the light which passes through the modulator cell 2, so that the light which passes through one modulator cell is deflected by the volume grating Vg1, and the light which passes through the other modulator cell is not deflected by the volume grating Vg1. What has been said above also applies to FIGS. 7, 8, 13, 14 and 17 in a similar way.

(44) The angular geometry does not have to be met with an accuracy of <0.05?. An error of 0.1? is uncritical. If D<a, angle errors of 0.3? are uncritical, i.e. even if a measurement is necessary to compensate the resultant effects. This is due to the fact that the portion of non-diffracted light is blocked in the plane of the apodisation filter APF if D<a.

(45) However, it is necessary or at least recommended to accept the sandwich part which comprises two volume gratings of like geometry as such. A lateral offset of the wave fronts which are to be superposed is unproblematic, because the fill factor of the apodisation filter APF is smaller than the fill factor of the phase-modulating SLM, i.e. FF.sub.APO<FF.sub.SLM. The dimension of the light-emitting area is thus constant, this area is sufficiently homogeneously illuminated and comprises only common superposed wave front portions, i.e. a lateral offset of as much as 5% would be unproblematic. In other words, a lateral offset of the light wave portions which leave the modulation element ME can be compensated with the help of a filter or shutter, e.g. an apodisation filter APF with a defined transmittance characteristic or an aperture mask with a defined mask geometry, which is arranged downstream of that modulation element ME. This can also be applied to the embodiments according to FIGS. 5 to 7.

(46) Refractive Beam Combination with Lenses or Prisms

(47) Now, another possibility of refractive beam combination will be described which is based on the use of lenses and/or prisms, or lenticulars and/or prism arrays.

(48) FIG. 18 illustrates an embodiment and shows in a top view a detail of an optical system 100 which comprises a lenticular L and a prism array P. A lens 102 of the lenticular L and a prism 104 of the prism array P are both assigned to two pixels of the SLM (not shown in FIG. 18). The drawing shows the beams 106, 108 coming from two mutually assigned pixels, a lens 102 of the lenticular L and a prism 104 of the prism array P. The pixel pitch is denoted by p, the diameter of one beam 106, 108 upstream of the lens 102 is denoted by a, and the distance between lenticular L and prism array P is denoted by d.

(49) Lens 102 focuses the light of each beam 106, 108 and converges the two beams 106, 108. The distance d is somewhat smaller than the focal length of lens 102, so that the focused beams 110, 112 are a small distance D apart in the plane of the prism array P. The two beams 110, 112 hit different sides of a prism 104. The prism angle is chosen such that the beams 114 substantially run in the same direction downstream of the prism. The drawing shows the double angle of divergence 20 and the double beam waist 2w.

(50) In this arrangement the two beams 106, 108 do not fully converge, but remain a narrow distance D apart. However, this distance is much smaller than the original distance, which equals the pixel pitch. Therefore, the difference in the optical path length of the light towards the edge of a diffraction order is much smaller, which greatly improves the reconstruction quality.

(51) Now, a numerical example will be provided under the simplifying assumption that the distance d between lenticular L and prism array P equals the focal length f, i.e. d=f. Further, the beams are assumed to be Gaussian beams. The pixel pitch is p=50 ?m. The distance of the beams is to be minimised from p=50 ?m to D=p/10=5 ?m. The beam waist is chosen such that D=2.Math.w.

(52) The following relations apply:
?*w=?/? (beam parameter product of a Gaussian beam=ratio of divergence and beam waist of a Gaussian beam)
a=2?*f
D=2w

(53) If with the help of this arrangement the distance p of the beams is reduced from 50 ?m to 5 ?m, then a focal length f=0.31 mm will result for a wavelength of 500 nm. The radius of the lenses would thus be about 0.15 mm at a lens pitch of 0.1 mm.

(54) Lenticulars L and prism arrays P are optical components which can be made and aligned in large sizes. They serve to substantially reduce the distance of the two beams 106, 108 and thus to improve the reconstruction quality.

(55) FIG. 19 illustrates another embodiment and shows in a top view a detail of an optical system 100 which comprises two prism arrays P1 and P2 and a spacer glass plate G with a thickness d. The drawing shows two beams 106, 108 coming from mutually assigned SLM pixels (not shown), said beams comprising perpendicular polarisation directions after having passed a structured retardation plate (not shown).

(56) The first prism array P1 is made of an isotropic material. In contrast, the second prism array P2 is made of a birefringent material. One direction of polarisation is transmitted as an ordinary beam 108, 112 without being deflected, while the perpendicular direction of polarisation is deflected as an extraordinary beam 106, 110. This is similar to the birefringent lenticulars which are used by the company Ocuity for switchable 2D/3D displays. The ordinary refractive index is chosen to be equal to the refractive index of the surrounding material. In contrast, the extraordinary refractive index is chosen to be different, so that the extraordinary beam is deflected.

(57) The lower beam 108 passes through the prism array P1 without being deflected, because it hits the planar interface. On entry into the spacer glass plate G it is denoted by 112; it is not deflected by the prism array P2 either, because it has the direction of polarisation of an ordinary beam. The upper beam 106 is deflected by both prism arrays, P1 and P2, because it is the extraordinary beam. Both beams 106, 108 are thus combined and leave the optical system in the form of a superposed light beam 114 in the same direction.

(58) Now, a numerical example will be given with a pixel pitch of p=50 ?m. The thickness of the glass plate is assumed to be d=500 ?m. In this arrangement, the upper beam 106 must be deflected in each prism array P1 by ?=5.7?. For small angles, the following relation applies:
?=(n.sub.1/n.sub.2?1)*?
where ? is the prism angle and n.sub.1 and n.sub.2 are the refractive indices of the prism P1 and of the surrounding material, i.e. the glass G. Typical values are n.sub.1=1.65 and n.sub.2=1.5, i.e. there is a refractive index difference of ?n=0.15. This results in a required prism angle of ?=57?.

(59) The company Ocuity has already produced birefringent lenticulars of a size of several inches for an application that is different from the one described here. A sandwich of commercially available prism array, spacer glass plate and birefringent prism array can thus me made in a large size in order to achieve beam combination.

(60) The light wave multiplexing means can thus comprise a lens means and a prism means (see FIG. 18). The light 106 which passes though a first modulator cell can be focused by the lens means in a first region in a plane that lies downstream of the lens means in the direction of light propagation. The light 108 which passes through a second modulator cell can be focused by the lens means in a second region in that plane. The prism means is arranged at the plane. The prism means is designed such that the light of the first region is deflected by the prism means into a first presettable direction, and the light of the second region is deflected into a second presettable direction. The first and the second presettable direction are substantially identical. The first region is arranged at a distance to the second region. The lens means comprises a lenticular L, and the prism means comprises a prism array P.

(61) The light wave multiplexing means according to FIG. 19 comprises a first prism means and a second prism means. The light 106 which has passed a first modulator cell can be deflected by the first prism means into a first direction. The light 108 which has passed a second modulator cell is not deflected. The first prism means is followed in the direction of light propagation by the second prism means at a defined distance d. The second prism means is designed such that the light 110 which has been deflected by the first prism means can be deflected by the second prism means into a presettable direction. The light 112 which has not been deflected is not deflected by the second prism means.

(62) The second prism means comprises a prism array P2 with birefringent prism elements. The light 106 which passes through the first modulator cell is polarised such that it can be deflected by a birefringent prism element of the second prism means. The light 108 which passes through the second modulator cell is polarised such that it is not deflected by the second prism means.

(63) The first prism means comprises a prism array P1 with prism elements. The prism elements are arranged such that only the light 106 which passes though the first modulator cell is assigned to a prism element, and that the light 108 which passes though the second modulator cell is not assigned to a prism element.

(64) FIG. 20 shows another embodiment of the present invention. Here, the light wave multiplexing means comprises at least two birefringent media SV1, SV2. One birefringent medium SV1 is arranged upstream of the modulator cells 1, 2, seen in the direction of light propagation, and another birefringent medium SV2 is arranged downstream of the modulator cells 1, 2. The birefringent media SV1, SV2 each have a presettable optical property. The optical property of the birefringent medium SV1, which is arranged upstream of the modulator cells 1, 2, is chosen such that a first portion of the light is deflected by a first defined angle towards the first modulator cell 1. In an upper section of FIG. 20, the beam diameter of this light portion is indicated by dotted lines. Two further beams are drawn in below, and all beams shall be construed to be like this across the entire surface of the element. Another portion of the light is not deflected. The beams of these light portions are indicated by full lines. The optical property of the birefringent medium SV2, which is arranged downstream of the modulator cells 1, 2, is chosen such that the other portion of the light is deflected by a second defined angle and that the first portion is not deflected. The optical property of the two birefringent media SV1, SV2 shall in particular be understood to be the orientation of the optical axis or major axis of the respective birefringent medium SV1, SV2. The optical axes of the two birefringent media SV1, SV2 are indicated by double arrows and have substantially the same orientation. There are other thinkable configurations, where the orientation of the optical axes of the two birefringent media SV1, SV2 do not lie in the drawing plane of FIG. 20. Although it is generally possible that the light which runs towards the first birefringent medium SV1 is not polarised, it is preferably provided that the light which falls on the first birefringent medium SV1 has a defined linear polarisation.

(65) The two birefringent media SV1, SV2 in FIG. 20, but also the birefringent media SP, SP1, SP2 and SP3 in FIGS. 13, 14, 17 and 21, have substantially coplanar interfaces.

(66) Referring to FIG. 20, a retardation plate in the form of a ?/2 plate is arranged between the two birefringent media SV1, SV2. This retardation plate turns the direction of polarisation of the light which passes through the modulator cells 1, 2 by 90 degrees.

(67) An aperture mask BA is arranged upstream of the first birefringent medium SV1, seen in the direction of light propagation, said aperture mask being designed such that the non-deflected portion of the light, which would propagate towards each modulator cell 1, is blocked out. In other words, the aperture mask BA comprises individual apertures which have substantially the same cross-sectional area as the modulator cells 1, 2. Now, the aperture mask is positioned such that every other modulator cell, i.e. all modulator cells 1 are covered so that no light falls on them. This is to prevent that non-deflected light passes through the modulator cells 1. The individual components are shown separately in FIG. 20 in order to keep the drawing comprehensible. However, the components can be combined in the form of a sandwich, i.e. be in direct contact with each other.

(68) In the arrangement shown in FIG. 20, the distance between the modulator cells 1, 2 and a further optical element which is arranged downstream of the birefringent medium SV2 (e.g. a deflection prism cell arrangement or an apodisation filter, not shown in FIG. 20) can preferably be reduced compared with an arrangement for example as shown in FIG. 17. The arrangement shown in FIG. 20 is particularly preferable for beam combination of spectrally broad-band light, but can also be used for spectrally narrow-band light. An arrangement shown in FIG. 20 serves to realise a symmetrical beam splitting and beam combination, which can serve to minimising the deviation of the optical path lengths on the one hand and/or of the superposed, i.e. combined wave fronts on the other. It can thus be achieved that the diffraction patterns of the two superposed modulator cells 1, 2 comprise the same intensity and phase distributions (except the orthogonality of the polarisation state) at the point of exit of the light modulator device. This is a major aspect for a high-quality hologram reconstruction, if such a light modulator device is used in a holographic display. Similarly, minimising cross-talking of light which passes through two adjacent modulator cells 1, 2 of the arrangement is another important aspect of high-quality hologram reconstruction.

(69) FIG. 22 shows another embodiment which serves to realise a similar function as the embodiment shown in FIG. 20. The embodiment shown in FIG. 20 uses refractive components, namely the two birefringent media SV1 and SV2. In contrast, the embodiment shown in FIG. 22 uses diffractive components, namely the deflection layers Vg1, Vg2, Vg3 and Vg4 shown in the drawing, which are realised in the form of volume gratings. The deflection layers Vg1, Vg2 are arranged upstream of the modulator cells 1, 2, seen in the direction of light propagation. The deflection layers Vg3, Vg4 are arranged downstream of the modulator cells 1, 2, seen in the direction of light propagation. The light which falls on the first deflection layer Vg1, i.e. which is not blocked by the cover B, is not polarised but shows a homogeneous distribution of individual polarisation portions, or is has a defined polarisation state, e.g. a linear polarisation.

(70) The first deflection layer Vg1 is designed such that the light is split up into two partial beams. The one partial beam is substantially not deflected and it is linearly polarised, i.e. has for example a TE polarisation; it is indicated by dotted lines in the drawing. The other partial beam is deflected by a defined angle and it is also linearly polarised, but has for example a TM polarisation; it is indicated by broken lines in the drawing. The second deflection layer Vg2 is arranged parallel to the first deflection layer Vg1, and it is designed such that the light which has not been deflected is not deflected and the light which has been deflected by the defined angle is deflected by another angle. The absolute values of the two deflection angles are substantially identical, namely 60?. The direction of polarisation of the light which has not been deflected is turned by 90 degrees by the structured retardation plate, which is realised in the form of a ?/2 plate and which is arranged downstream of the second deflection layer Vg2. Consequently, the light which passes through the modulator cells 1, 2 has a substantially identical polarisation state.

(71) The modulator cells 1, 2 are designed such that they can modify the phase of the light which interacts with them. A further structured retardation plate in the form of a ?/2 plate is arranged between the modulator cells 1, 2 and the third deflection layer Vg3, said plate turning the direction of polarisation of the light which passes through the modulator cell 2 by 90 degrees. The light falls on the third deflection layer Vg3, which is designed such that the light which passes through the modulator cell 2 is substantially not deflected, and the light which passes through the modulator cell 1 is deflected by a defined angle. The fourth deflection layer Vg4 is arranged parallel to the third deflection layer Vg3, and it is designed such that the light which has not been deflected by the third deflection layer Vg3 is not deflected and the light which has been deflected by the third deflection layer Vg3 by the defined angle is deflected by another angle. The absolute values of the two further deflection angles are substantially identical. In this respect, the light beams which pass though the two modulator cells 1, 2 are thus combined and propagate substantially in the same direction. If the two modulator cells 1, 2 realise substantially the same phase value, the optical path lengths of the two partial beams are substantially identical.

(72) There are modulator cells 1, 2, or SLMs, which do not require a defined entry polarisation. In that case, it is possible to omit the structured retardation plate upstream of the modulator cell plane and to replace the second structured retardation plate, which is arranged immediately downstream of the modulator cell plane, by an unstructured retardation plate, i.e. an unstructured half-wavelength plate.

(73) For an RGB presentationi.e. when using light of different wavelengthsit is possible to expose three different volume gratings, each being adapted to an individual wavelength, in an interleaved manner in each of the deflection layers Vg1-Vg4. The arrangement shown in FIG. 22 is of course also thinkable complemented in the form of columns, lines or matrices, namely when the components which are shown in FIG. 22 continue above and below and/or out of the drawing planevery much like in FIG. 20.

(74) The light wave multiplexing means are typically arranged immediately downstream of the modulator cells of the modulation array, seen in the direction of light propagation, in the drawings. However, it is also possible to arranged the light wave multiplexing means shown in the drawings at a different position. For example, another optical component can be arranged between the modulation array and the light wave multiplexing means. A light wave multiplexing means as shown in the drawings and as claimed in the claims, can thus be arranged downstream of that further optical component, seen in the direction of light propagation. Such a further optical component can for example be an illumination unit as disclosed in documents DE 10 2009 028 984.4 or PCT/EP2010/058619. The light which is injected into this illumination unit can for example leave at right angles to its surface (which is arranged parallel to the modulation array) and propagate onto a reflection-type modulation array. Once the light which comes from the illumination unit has been modulated by the modulator cells of the modulation array and reflectedfor example by a reflective layer of the modulation array, the modulated light passes through the illumination unit substantially without being deflected and then falls on the light wave multiplexing means. In this case, the light wave multiplexing means is arranged on the side of the illumination unit which faces away from the modulation array. To enable the light which is modulated and reflected by the modulation array to pass through the illumination unit without any obstructions, a film is provided between the illumination unit and modulation array which turns the direction of polarisation for example by 45? whenever the light passes through it.

(75) Finally, it must be said that the embodiments described above shall solely be understood to illustrate the claimed teaching, but that the claimed teaching is not limited to these embodiments.

CITATIONS

(76) [1] Chulwoo Oh and Michael J. Escuti: Achromatic polarization gratings as highly efficient thin-film polarizing beamsplitters for broadband light, Proc. SPIE, vol. 6682, no. 628211, 2007 [2] Jihwan Kim et al.: Wide-angle nonmechanical beam steering using thin liquid crystal polarization gratings, Proc. SPIE, vol. 7093, no. 709302, 2008