WO1998015855A1

WO1998015855A1 - Converter of natural white light into light with planar polarization

Info

Publication number: WO1998015855A1
Application number: PCT/US1997/018038
Authority: WO
Inventors: Brahim Dahmani; Alexander Galpern; Sergey Kuchinsky; Vitaly I. Sukhanov
Original assignee: Corning Incorporated
Priority date: 1996-10-08
Filing date: 1997-10-03
Publication date: 1998-04-16

Abstract

A converter of natural white light into light of planar polarization has a divider (3) of said natural white light into two beams of mutually orthogonal polarization, each consisting essentially of planar polarized light of p-polarization or s-polarization; a rotator (6) for turning the plane of polarization of one of the two beams, and means (4, 5, 7) for combining the two emerging beams with the same polarization. According to the invention, the divider (3) consist of a stack of alternating parallel plates of two distinct optical media and refractive indices, the angle of incidence of the beam arriving from the natural white light source (1) being equal to or greater than the Brewster angle as determined by the refractive indices of the two optical media. The invention finds industrial applicability as a source of polarized light in video image projectors.

Description

CONVERTER OF NATURAL WHITE LIGHT INTO LIGHT WITH PLANAR

POLARIZATION

Background of the Invention

The present invention relates to a converter of natural white light into light with planar polarization, and more particularly to such a converter which includes a) a beam source of natural white light, b) a divider of said beam into a reflected beam and a transmitted beam, these two beams each consisting essentially of one of the two components (P, s) with planar polarization of the light beam coming from the source, the two planes of polarization being orthogonal with respect to one another, c) a rotator in order to turn the plane of polarization of one of the beams, coming from the divider, to the plane parallel to the plane of polarization of the other beam, and d) some means for combining the two emerging beams with the same polarization.

Such a converter is known from the article entitled "A polarization transforming optics for high luminance LCD projector" by Shikama et al . , pages 64-67 of the transactions of the Congress Eurodisplay 9.', Amsterdam, Holland (September 25-27, 1990) . Such a converter is useful, in particular, for increasing the luminance of the images formed by video image projectors of the type which have a matrix of liquid crystal cells arranged between crossed polarizer and analyzer. The projection beam which passes through this assembly is provided by a source of non- Polarized natural -light. In order to prevent absorption of light by the polarizer, the aforementioned article proposes Converting all of the light coming from the source into light with planar polarization, parallel to the plane of polarization of the polarizer, in front of it. To do this, the converter proposed in the article includes a beam - - divider which transmits the polarized component linearly, with the electrical field vector parallel to the plane of incidence (component p) , and which reflects the polarized component linearly, with the electrical field vector perpendicular to the plane of incidence (components), of the beam provided by the source. A half-wave plate causes the plane of polarization of the reflected light (component s) to turn by π/2, this light is parallel to the plane of polarization of the beam p transmitted by the prism. The latter is combined with the beam which emerges from the half-wave plate, in order to illuminate the polarizer, whose plane of polarization is oriented parallel to that of these two beams. One thus increases the quantity of light transmitted by the input polarizer of", the matrix of liquid crystal cells.

The polarizing beam divider used in thv arrangement described above can be produced by the cutting of birefringent crystals or by deposition under vacuum of interferential dielectric layers on the hypotenuse sides, facing one another, of two rectangular prisms, which makes it an expensive optical element, which heavily burdens the price of a video image projector equipped with a light converter incorporating such a prism. Furthermore, by deposition of interferential layers, it is difficult to reach a high level of polarization (ratio of the intensities of the components p and s) .

The present invention aims to produce a converter of natural white light into light with planar polarization, which does not have these disadvantages and which can be used in a video image projector intended for the general public, a projector whose cost of production must consequently be as low as possible. Summary of the Invention

This aim of the invention is reached, as are others which appear upon reading of the following description, with a converter of the type described in the preamble of the - present description, which is remarkable in that the beam divider which equips this converter consists of a stack of parallel plates of two distinct optical media and which alternate with refractive index nj. and n_m respectively. The thickness of each of the these plates being much greater than the lengths of coherence of the light coming from the source, the angle of incidence, θ of the beam coming from the source on the second plate of the stack being such that:

n, θ ≥ θ_h = arctg—

θ_b being the Brewster angle of incidence on said second plate, made up of the optical medium with index n_x. The use of such a stack of plates allows one to produce the converter, according to the invention, for a relatively low cost, while giving a very good performance in terms of the level of polarization and conversion yield, particularly, as will be seen further on. According to one embodiment of the invention, the angle of incidence, θ of the beam coming from the source on the second plate of the stack is such that θ = θ_b. The transmitted beam which emerges from the rotator illuminates, according to the same incidence a planar reflector parallel to the plates of the stack of plates, in such a way that the beams reflected by the two stacks combine to form a single beam with planar polarization of the type s.

According to another embodiment of the invention, the angle of incidence, θ of the beam on the second plate of the stack is such that θ_b < θ < θ₂, in which θ₂ is the angle of incidence of the beam coming from the source on the second plate, for which the intensity of the reflected fraction of the component s of the light is equal to the intensity of the transmitted fraction of the component p. It is thus possible to obtain a high level of reflection of the component s from the light of the source, on the stack of plates, while advantageously minimizing the number of these plates, as will appear subsequently.

Detailed Description of the Drawings

- Figure 1 is a diagram of one embodiment of the converter according to the invention,

- Figure 2 is a graph and Figure 3 is a diagram which explain the functioning of the stack of plates incorporated in the converter of Figure 1, - Figure 4 is a diagram of another embodiment of the converter according to the invention, and

- Figures 5-8 are graphs illustrating the performances of the converter according to the invention.

Detailed Description of the Invention

Reference is made to Figure 1 of the appended drawing, in which it appears that the converter according to the invention has source 1 of natural white light, that is to say not polarized, which is collimated by lens 2 in order to illuminate an extensive surface area of stack 3 of plates which will be described in more detail in connection with Figures 2 and 3, and which constitutes a polarizing beam divider. The stack of plates 3 reflect a part of the light received to mirror 4 and transmit the other part of this light, through polarization rotator 6, to another stack of plates 5, parallel to stack 3 and functioning in the same way. In turn, this stack 5 reflects a part of the light to mirror 7 and transmits another part of it, preferably to an optical medium (not represented) which absorbs the transmitted light.

According to a particular application of the converter of this invention, to the projection of video images, the light reflected by mirrors 4 and 7 illuminates matrix 8 of liquid crystal cells functioning as light valves. These cells are controlled individually so that each ensures the formation of an element (or pixel) of a visible image projected using an objective (not represented) on a screen (not represented) arranged to the right of matrix 8, from the point of view of Figure 1. For this purpose, the converter of this invention illuminates matrix 8 with a beam of light with planar polarization, polarizing element 9 called analyzer being conventionally attached to matrix 8 on the side which is opposite to that illuminated by the light coming from source 1. As is well known in the technique, analyzer 9, whose axis of polarization is oriented at 90° with respect to the plane of polarization of the incident light, extinguishes or allows the light to pass through each of the liquid crystal cells of matrix 8, according to the state of electrical excitation of these cells, in such a way that an image is displayed on the projection screen. As a variant, the axis of polarization of the analyzer could be parallel to that of the light coming from the converter. In this case, each cell is designed to allow the light to pass in the absence of excitation and to block che light in the presence of excitation.

As represented in more detail in Figure 3, stack 3 and, in the embodiment of Figure 1, stack 5 consist of plates with parallel surfaces 10!, 10₂, 10₃, etc... of an optical medium with refractive index n_l7 two successive plates 10ι, 10_i+i being separated by a plate with parallel surfaces such as Hi, I 2 H3, etc... of a second optical medium with index rim, the thickness of each of these plates being much greater than the wavelength of the visible light used, which prevents any stray interferential effect. As an illustrative and non-limiting example, plates 10ι can be made of glass, plates Hi are plates of air, plates 10_^. and Hi having a thickness of approximately 0.1 mm.

More generally, when the medium which separates two plates 10i has an index n_m different from the medium in which the whole stack of plates 3 is immersed, it is understood that the angle θ mentioned in the present description is that of the index of the beam illuminating any one of plates IO2-IO4, beginning with the second 10₂ and not that of the incidence of the beam on the first plate 10ι as represented in the drawing in which, for the sake of clarity of the figures, all plates are considered to be immersed in the same medium.

It is known that natural non-polarized light can be considered to be formed by two components p and s with planar polarization, the planes of polarization of the two components being perpendicular to one another. When natural light illuminates a planar diopter separating two optical media with different refractive indexes, such as surface 10ι of stack 3, the components p and s are reflected and transmitted differently by this diopter and depend on the angle of incidence θ of the beam coming from source 1. The graph of Figure 2 illustrates, for a glass plate 10ι with refractive index ni = 1.5, placed in the air, the variations of the coefficients of reflection r_p and r_s of the components p and s of the incident beam. It appears on this graph that the coefficient of reflection r_s of the component s of the light is always higher than the coefficient of reflection r_p of the component p. At the Brewster angle of incidence in the vicinity of 57° in the illustrative example chosen, r_p is zero, which means that the reflected light is then made up of component s radiation alone, component p being entirely transmitted. By stacking such interfaces, as represented in Figure 3, the total quantity of reflected light of type s increases at each interface, in order to reach an overall reflection coefficient R_s such that:

2-V.r.

R. =^■ \ + {2N- \)r_s in which N is the number of glass plates 10 (i from 1-N) , this coefficient R_s tending towards 1 when N increases. Advantageously, according to the invention, one chooses θ = θ_b so that only the light of component s is reflected by stack 3 to mirror . The transmitted light is then essentially made up of component p and of a small part, not reflected by stack 3, of component s. Polarization rotator- 6, consisting, for example, of a half-wave plate causes the plane of polarization of component p to turn by 90° and then take on type s. The small fraction of component s transmitted by stack 3 takes on the type p. Stack of plates 5, of the same type as stack 3, practically entirely reflects the light of type s which it receives to mirror 7. Mirrors 4 and 7 then both send back light only of type s on matrix 8 of liquid crystal cells. The small residue of light of type p which arrives at stack 5 is transmitted by this stack to an absorbent medium (not represented), for example. This quantity of absorbed light being small, one sees that the yield of the conversion carried out by the converter of this invention is high.

By thus using practically all the light coming from source 1 to illuminate matrix 8, rather than that which conventionally comes out of a highly absorbent polarizing filter placed up stream from the matrix, one sees that it is possible to increase the illumination of this matrix considerably and therefore, the luminance of the images formed by projection of light through it.

This increase obviously depends strictly on the capacity of stack 3 and to a lesser extent on that of stack 5 to reflect a large fraction of component s of the light of the source. The graph of Figure 5 illustrates the variations of the overall reflection coefficient R_s as a function of the number N of glass plates (in this case of a glass/air stack) and as a function of the refractive index of. the glass of the plates. Graphs A, B and C represent the theoretical variations of this coefficient as a function of N, for glasses with indexes n = 1.87, n = 1.64, n = 1.46 respectively, such as the glasses referenced 878-385, 648- 462 and 465-657 respectively, in the catalogues of the Corning Incorporated Company. Graph D, in the form of a broken line, corresponds to reflection coefficients measured experimentally and obtained with a stack of glass plates with the same index as that corresponding to graph A. It appears, on these graphs, that a stack of four glass plates with a high refractive index (n = 1.87) ensures a ..- . theoretical overall reflection coefficient R₃ of 0.8, that is to say a reflection of 80% of the light with polarization s received by the stack, which is remarkably high.

Since the manufacturing of a stack of plates is possible at low cost, one sees that the invention makes it possible to produce a converter of natural light into light with planar polarization with a high conversion yield, for a price compatible with an application intended for the general public such as the projection of video images, which would not be the case of a converter equipped with a beam divider made by cutting birefringent crystals, which are very costly, or by deposition under vacuum of stacked interferential dielectric layers. The latter technique is made expensive by the fact that the thickness of the layers formed must be completely adjusted to a low value on extensive surfaces, a restriction which does not burden the plates of the converter of this invention.

The graph of Figure 2 shows that the reflection coefficient r_s of component s of the light can be higher than the Brewster incidence, above θ = θ_b. This can be used to reduce the number of plates of stacks 3 and 5 and therefore their manufacturing cost. However, a non-negligible fraction r_p of the component p is then also reflected. The converter of Figure 1 must then be modified as represented in Figure 4, in which references identical to the references used in

Figure 1 mark identical or similar elements. In Figure 4, it appears that stack 5 is replaced by a simple single mirror 5' and that polarizing filter 12 is installed in front of liquid crystal matrix 8, the plane of polarization of this filter being oriented at 90° with respect to that of filter 9 in order to absorb component p of the light which is reflected by stack 3 and possibly by mirror 5'. The light yield of the converter remains high, nevertheless, because polarizing filter 12, which only has to -absorb a component p of low intensity, can be produced in such a way as not to considerably reduce the illumination of matrix 8.

One should note that mirror 5' can be arranged parallel 5 to stack 3. In all cases, the planes of stack 3 and of mirrors 4, 5, 5' and 7 will advantageously be perpendicular to the plane of Figures 1 and 4 in order not to disturb the polarization of the light beams which are combined coming out of the converter. 0 It is also possible to maximize the homogeneity of the illumination of matrix 8, by ensuring that the two beams are put together and sent back by mirrors 4 and 7, by choosing, for the angle of incidence, θ a value θ₂ for which the intensity of the reflected fraction of component s is equal 5 to the intensity of the transmitted fraction of component p. It is demonstrated that when stack 3 is immersed in a medium with index n_m, the angle θ₂ in consideration is a solution of the following system of equations:

fe)_ s^Min^,,2'¹fe-') ^r _. = „ tg¹(θ₁ -φ) sm +φ) tg + Φ)

φ = arcsin -^{SL •} sin θ, n,

The graph of Figure 6 illustrates the influence of an angular divergence a of the light beam coming from source 1 which illuminates stack 3 in the vicinity of the Brewster incidence, for three wavelengths R, V, B chosen in the red, green and blue respectively. This divergence can result from the fact that the light beam coming from source 1 (not a point source) cannot be parallel. The graph represents the variations of the overall reflection coefficient R_p of the component p which interferes [unconfirmed translation] with the reflection, essentially of component s, on stack 3. This stack has four glass plates with index n = 1.878, giving an overall reflection level R_s of 0.68. It is observed that for the three chosen wavelengths, an angular divergence a = + 0.5° never brings the coefficient R_p above 0.001, which is very low. The Brewster angle θ_b is a function of the wavelength of the light. When one wishes to illuminate a matrix of liquid crystal cells in white light, it is necessary to choose an angle θ, which is obviously unique for all the wavelengths, which results from a compromise. With a stack of plates^' -^" identical to that used to plot the graph of Figure 6, Figure 7 represents the variations of R_p in the visible domain (λ from 0.4-0.8 mm) for three values of θ: θ_x, θ₂, θ₃ respectively, which minimize R_p to 0.4 mm, 0.5 mm and 0.8 mm, respectively. By choosing for θ the angle θ₂ which is the Brewster angle for a radiation with wavelength λ = 0.5 mm, one sees that it is possible to limit R_p, for all the wavelengths, to approximately 2.10^"4 [sic; 2 X 10^"4] . The converter according to the invention, thus defined, then has a very satisfactory achromatism.

Finally, the graph of Figure 8 represents the variations of the level of polarization of the beam leaving the converter according to the invention, as a function of a difference Δθ_p of the real angle of incidence of the beam illuminating stack 3 with respect to the theoretical

Brewster angle θ_b. The theoretical variation is represented by a solid line, the variation taken from experimental measurements is represented by a broken line. For an acceptable error Δθ_b = ±0.5°, one sees that the experimental measurements report a level of polarization higher than 800, with a theoretical level higher that 1,000, which is very satisfactory. Such a level is difficult, if not impossible, to reach with a beam divider with a stack of interferential layers. Various types of polarization rotators can be use to constitute rotator 6 of the converter according to the invention, for example, a half-wave plate made of mica, a film of birefringent polymer such as a drawn polyvinyl alcohol film, a liquid crystal cell or else a plate made of an optically active material as described in the work entitled "Optical Waves in Crystal", authors: A. Yariv, P. Yeh, New York, Wiley International Publication, 1984. One could also use a plate made of a birefringent porous glass, such as one of those described in the work entitled "Phase separation in glass" prepared by O.V. Mazurin and E.A. Porai-Koshits, published in 1984 by North-Holland (Amsterdam-Oxford-New York-Tokyo) . Rotator 6 could also consist of a 90° image rotator such as that described -in

Figure 2 of the article entitled "A polarization converter for High-Brightness Liquid Crystal Light Valve Projection, authors: M. Imai et al . , page 236 of Actes du congres [Transactions of the congress] Japan Display, 1992. Of course, the invention is not limited to the embodiments described and represented which were only given as examples. Thus, the polarizing beam divider can take on forms other than that of a stack of glass plates separated by air; it can also be enclosed between two glass prisms in order to facilitate its operations and the adjustment of its orientation. When the stack is made up of alternating plates with indexes n_x and n_m respectively, let n_x = 1.5, for example in the case of glass plates and n_m = 1, index of the air; the prisms are made of glass and a plate of air must be arranged at both ends of the stack at the interface with the hypotenuse faces of the two prisms which enclose the stack.

The stack could also have only one plate, made of glass, for example, a glass with a high refractive index. Instead of this single plate, it is possible to arrange a pair of prisms made of glass with a high refractive index, their hypotenuse faces being arranged facing one another, while remaining separated by a thin plate of air.

It is also possible to use a stack of plates of glass or of another isotropic material, separated by plates of a birefringent material, such as a liquid crystal, of which one of the indexes is roughly equal to that of the glass of the plates. Thus, for one of the polarized components of the light, the stack behaves as a homogeneous transparent medium. For the other component, one observes reflections on the diopters of the stack, separating the glass from the birefringent material.

Furthermore, the invention is not limited to a converter in which one brings about the turning of the plane of polarization of component p of the light, transmitted by stack 3. It would still be in the scope of the invention to bring about the turning of the plane of polarization of the component s reflected by stack 3, in order to re-align the planes of polarization of the two beams formed by this stack. It would also be in the scope of the invention to eliminate mirrors 4 and 7.

Claims

1. A converter of natural white light into light with planar polarization, of the type which has a) source (1) of a beam of natural white light, b) divider (3) of said beam into a reflected beam and a transmitted beam, these two beams each consisting essentially of one of the two components (p, s) with planar polarization of the light beam coming from source (1), the two planes of polarization being orthogonal with respect to one another, c) rotator (6) in order to turn the plane of polarization of one of the beams coming from the divider to the plane parallel to the plane of polarization of the other beam, and d) some means of combining the two emerging beams with the same polarization, characterized by the fact that beam divider (3) consists of a stack of parallel plates of two distinct optical media and which alternate with refractive index (n and (n_m) respectively, the thickness of each of the these plates being much greater than the wavelengths of coherence of the light coming from the source, the angle of incidence (θ) of the beam coming from source (1) on second plate (10₂) of the stack being such that:

n_t θ ≥ θ_h = arctg— » m

(θ_b) being the Brewster angle of incidence on said second plate (10₂), made up of an optical medium with index (nj.) .

2. A converter according to Claim 1, characterized by the fact that the angle of incidence (θ) of the beam on a second plate (10₂) of the stack is such that: θ_b < θ < θ₂ in which θ₂ is the angle of incidence of the beam coming from source (1) on the second plate (10₂) , for which the intensity of the reflected fraction of the component (s) of the light is equal to the intensity of the transmitted fraction of the component (p) .

3. A converter according to Claim 1, characterized by the fact that the angle of incidence (θ) of the beam emitted by source (1) on second plate (10ι) of stack (3) is such that θ = θ_b.

4. A converter according to any one of Claims 1-3, characterized by the fact that it has planar reflector (5; 5') mounted up stream from rotator (6) in such a way that the beams reflected by stack (3) and reflector (5; 5') combine to form a single beam.

5. A converter according to Claim 4, characterized by the fact that said reflector is made up of a second stack (5) of plates of the same type as the first stack (3) .

6. A converter according to Claim 2, characterized by the fact that it has polarizer (12) passed through by the beam reflected by a stack of plates (3) and by the beam which emerges from the rotator (6), in order to extract any residual interference component from these beams.

7. A converter according to any one of Claims 1-6, characterized by the fact that stack of plates (3) is made up of glass plates (10i) separated by layers of air (Hi) .

8. A converter according to any one of Claims 1-6, characterized by the fact that stack of plates (3) is enclosed between the hypotenuse faces of two identical right angled prisms with refractive index (n_x) each attached to an end plate with index (n_m) of the stack.

9. A converter according to Claim 1, characterized by the fact that stack (3) is made up of a single plate of air produced in a material with a high refractive index.

10. A converter according to Claim 1, characterized by the fact that stack (3) is made up of a single plate of air formed by a layer of air between two prisms made of a material with a high refractive index.

11. A converter according to Claim 1, characterized by the fact that one of the two optical media of stack (3) ..of plates consists of a birefringent material, the other medium being isotropic, one of the refractive indexes of the birefringent material being roughly equal to that of the isotropic medium.

12. A converter according to any one of Claims 1-11, characterized by the fact that polarization rotator (6) acts on the beam transmitted by beam divider (3) .

13. A converter according to any one of Claims 1-12, characterized by the fact that polarization rotator (6) is made up of any one of the elements of the group formed by: a half-wave plate, a 90° image rotator, a plate made of an optically active material, a cell with birefringent liquid crystals.

14. A converter according to Claim 13, characterized by the fact that the half-wave plate (6) is made up of any one of the elements of the group formed by: a plate of mica, a sheet of a birefringent polymer, a cell with birefringent liquid crystals.

15. A converter according to Claim 13, characterized by the fact that the polarization rotator (6) is made up of a plate of porous glass.

16. A video image projector with matrix (8) of liquid crystal cells which is illuminated by source (1) of natural white light, characterized by the fact that it contains a converter according to any one of Claims 1-15, interposed between source (1) and matrix (8) .