WO2007010074A2

WO2007010074A2 - Optical demultiplexer

Info

Publication number: WO2007010074A2
Application number: PCT/ES2006/000431
Authority: WO
Inventors: José SÁNCHEZ DEHESA; Andreas Hakansson
Original assignee: Universidad Politecnica De Valencia
Priority date: 2005-07-22
Filing date: 2006-07-21
Publication date: 2007-01-25
Also published as: WO2007010074A3; ES2297981B2; ES2297981A1

Abstract

The invention relates to a demultiplexer comprising an ultra-compact device which spatially separates an incident beam (4) into at least two beams (5, 6) having wavelengths (λ-1, λ2), forming a pre-fixed angle (α) therebetween. The device, which is represented by two grilles (2) comprising bars (1), consists of a plurality of layers of bars (1) which are made from a dielectric material. The invention also comprises a plane (3) which defines the area at which such wavelengths (λ1, K2) are collected for analysis and which is located at a distance (Xf) from the device. The dispersed beams (5, 6) have a determined width (δ) and are separated from the axis of the incident beam (4) by a separation distance (d). According to the invention, the thickness of the device, the cross-section of the bars and the material used are determined by the required separation angle (α) and are optimised in order to minimise crosstalk between optical channels at said wavelengths (λ1, K2).

Description

OPTICAL DEMULTIPLEXER

D E S C R I P C I Ó N

OBJECT OF THE INVENTION

The present invention applies to the field of telecommunications, particularly in optical communication networks, extending its application to industrial sectors that use them for high-speed data transmission, such as mobile telephony, telemedicine, communications in space systems, etc.

The object of the invention is to provide an ultra-compact device, whose structure is based on multiple layers that form a network composed of bars of suitable dielectric material and conveniently configured to separate, with a predetermined angle, a beam of light incident in several of the wavelengths that constitute it, allowing a dispersion of the beams corresponding to said wavelengths or demultiplexing of optical channels, with minimal interference between channels, or what is the same, a maximum attenuation of the distortion for all channels.

BACKGROUND OF THE INVENTION

At present, the use of optical systems to transmit large amounts of information, representing voice, video and / or data, at a very high speed is increasing. The demand for bandwidth for optical communications is increasing, as they are used to support a high load of channels; for example, in high definition television, in UMTS third generation telephone services, etc.

Within optical communication technologies, it is known that more than one wavelength can be used to transmit the information through different channels. In particular, each wavelength can be a carrier for signals corresponding to analog or digital systems.

One of the techniques widely used to multiplex a number of different signals in an optical transmission system is the Wavelength Division Multiplexing (WDIVI), whereby the information, transported with digital signaling, is transmitted by modulating each group of signals digital with a different wavelength, traveling all wavelengths simultaneously through the optical fiber.

With the WDM multiplexing, the bandwidth is increased thanks to the number of independent channels due to the diversity of wavelengths used within the same optical transmission path, such as a fiber or a waveguide.

Obviously, when a number of wavelengths is multiplexed and transmitted through a single guide or optical fiber, it is subsequently necessary for the multiple channels to be demultiplexed into separate wavelengths. Obviously, it is desirable that this demultiplexing process be carried out at low cost and with minimal signal loss. The losses that may exist due to interference between the different wavelengths should be similar in magnitude for all channels that are demultiplexed.

A problem associated with the wavelength division is that if the physical dispersion between each wavelength that crosses the optical fiber is too narrow, several wavelengths can be joined in the same channel, instead of a single carrier, which would cause noise and distortion about the information contained in that channel. Therefore, a demultiplexer capable of performing a physical separation of wavelengths in a sufficiently wide area, so that the multiple information channels are divided with the least possible interference.

A solution to said problem Ia constitutes the diffractive optical elements, described by the authors J. L. Homer and P.D. Gianino in the Applied Optics publication, volume 23, of the year 1984, which can model the beam from an optical source, such as a laser, according to virtually any pattern, by controlling the phases of the different dispersed waves. This property is basically achieved by mounting in a layer small optical elements that each manipulate a part of the incident beam. The implementation of these elements is carried out by varying their index of refraction and their dimensions in thickness, parameters that are generated by computer, looking for the optimum with which a beam is obtained by propagating through the layer of such elements in the desired pattern.

However, diffractive optical elements do not allow to control the reflection of a ray of light from total opacity to complete transparency.

Other devices that come to solve the separation of channels by wavelengths are those collected, citing some examples, in EP 0938205 and US 2004/0051868.

The optical demultiplexer described in EP 0938205 is of a plurality of dichroic filters, interconnected sequentially and optically coupled to each other, which separate each of the channels in the order of the sequence of their respective central wavelengths, beginning with the channel of Shorter wavelength until reaching the one with the largest central wavelength, eliminating one channel from the remaining ones, finally obtaining all the channels separately at the output of said filters.

In US 2004/0051868, a holographic demultiplexer is described that uses the spectroscopy technique to reflect the different wavelengths that composes a light signal, according to different spatial positions on a detection device of one or more of the reflected wavelengths. This device comprises the detector, a light scattering element, such as a prism or a diffraction grating, plus several holograms, each hologram in turn constituting a scattering element that redirects a specific range of wavelengths or a concrete wavelength, of those contained in the ray of light.

DESCRIPTION OF THE INVENTION

The invention described herein consists of a photonic device that performs the spatial separation between at least a couple of wavelengths, acting as an optical element of dispersion of light in multiple beams corresponding to the different central wavelengths, which they are divided with a certain width for each beam and forming a specific angle to each other.

More specifically, the present invention is conceived for its integration into flat waveguides or optical fibers in order to effect the demultiplexing of the signal that travels through said waveguide or fiber.

The proposed demultiplexer is composed of a grid or grid of dielectric bars, arranged in at least one pair of photonic sheets, implemented in two dimensions by a processing procedure on a single integrated circuit and followed by a micromanipulation for assembly, of according to a reverse design technique, which solves Maxwell's equations for a two-dimensional system using the Multiple Dispersion Theory (MST).

This optical demultiplexer provides substantial novelty characteristics and notable advantages in terms of its dimensions, since in its implementation is achieved an ultra-compact device, of the order of about 2 μm thick, while representing a significant improvement in the way of separating the beams in space, since it achieves greater efficiency with respect to known light dispersers, assuming an efficiency measured in terms of crosstalk with a value below -25 dB for each optical channel.

Each of the sheets or layers of dielectric bars, manufactured with a semiconductor such as Gallium Arsenide (AsGa), can preferably be grown by usual lithographic techniques.

The method of manufacturing the structure of the photonic device in three dimensions consists essentially in dividing the structure prepared by a semiconductor nanofabrication technique into several, at least two, sheets in two dimensions. Next, these photonic sheets are mounted in a three-dimensional structure by means of micromanipulation, preferably applying the procedures defined in the document "Three-dimensional photonic crystals for optical wavelengths assembled by micromanipulation" of volume 81 of Applied Physics Letters, year 2002, as well as in the article "Micro assembly of semiconductor three-dimensional photonic crystals" of volume 2 of the publication Nature Materials of 2003, both of the authors K. Aoki, HT. Miyazaki, H. Hirayama, K. Inoshita, T. Baba, N. Shinya and Y. Aoyagi.

Depending on the spatial separation that you want to achieve from the optical channels, that is, depending on the wavelengths that you want to disperse and the desired angle of separation between beams, in the physical realization of the photonic device certain design parameters are selected, which are the thickness of the laminar structure, the section of the dielectric bars and their construction material, chosen after applying an optimization process in a computer, for example, the Genetic Algorithm (GA), which assumes that all the elements of dispersion have a fixed position, in combination with the MST Theory. In general, the optimization method could be combined with other alternative calculation algorithms that define a procedure capable of simulating the dispersion of light through networks of dispersing centers. If you want to go into more detail about the combined implementation technique MST-GA for photonic devices, you can refer to the article written by A. Hákansson, J. Sánchez-Dehesa and L. Sanchis, in the 2005 publication in the IEEE Journal on selected In Communications areas, or, signed by the same authors together with D. López-Zanón and J. Bravo-Abad, in Applied Physics Letters, volume 84, year 2004.

With respect to the background, the proposed optical demultiplexer shapes the light flow, allowing control of the parameters that determine the form of dispersion of the beams with a greater degree of freedom, according to the arrangement of the dielectric bars that make up the structure of the device and that differs from the implementations of the demultiplexers existing in the state of the art.

The photonic dispersion device object of the invention is based on a plurality of layers formed by a network of individual optical dispersers, instead of a single layer of components that control the phase of the dispersed beams as is the case of the diffractive optical elements that are known and mentioned above. In addition, the grid or porous structure of the device of the invention makes it possible to control the reflection of the light from absolute opacity, similar to what occurs in a Bragg reflector, to almost total transparency. However, this possibility of controlling the light field in such a wide way cannot be carried out using conventional diffractive optical elements.

This photonic dispersion device thus described thus constitutes an ultra-compact demultiplexer applicable in WDM systems, where the bandwidth of the optical communications in local area networks (LAN) is increased, working in collaboration with the routing optical multiplexers. Simultaneously different information channels through a fiber optic network. The effect of crossing lines through the expected coupling between fiber optic cabling signals can be reduced below -25 dB in each channel, thanks to the incorporation of the optical demultiplexer in the communication network with WDM multiplexing. In short, the main advantage of the demultiplexer described is that, in addition to being a passive optical device, it has a smaller size, a characteristic that makes it a practical solution for various applications of optical communications, where other devices, active or passive, they would not take place or imply greater dimensions of the system to which they were integrated.

DESCRIPTION OF THE DRAWINGS

To complement the description that is being made and in order to help a better understanding of the characteristics of the invention, according to a preferred example of practical implementation thereof, a set of figures is attached as an integral part of said description. Illustrative and not limiting, the following has been represented:

Figure 1 shows a schematic representation of the structure of the device object of the invention and its functionality, indicating the parameters involved in the reverse design process for its implementation: Xf defines the distance between the light scattering device and the plane where the divided beams are collected, with wavelengths λi and K ₂ respectively, adopting predetermined values over a beam width δ and an angle α between beams.

Figure 2 - Shows a cut in the plane orthogonal to the axes of the dielectric bars that make up the device, according to a preferred embodiment of the invention, representing in its two coordinate axes its spatial arrangement and the dimensions of the bars in μm.

Figure 3.- Shows a graphical representation of the operation of the optical disperser / demultiplexer device, in the form of a two-dimensional map generated by the simulation of the inverse design algorithm that implements the device and gives the value of the 20log function (| E (x, y) λi | / | E (x, y) λ ₂ |) in the spatial area of the device and its surroundings, according to a possible embodiment of Ia invention, which predicts crosstalk between channels or wavelengths λi and λ ₂ .

PREFERRED EMBODIMENT OF THE INVENTION

It can be described as one of the possible embodiments of the invention, an optical device composed of a plurality of layers, in the preferred example are five layers, of dielectric bars (1) constructed with the semiconductor AsGa. Preferably, only five layers of dielectric bars (1) are used to ensure their perfect alignment by the micromanipulation technique.

In view of Figure 1, the function of the device as a light scattering or channel demultiplexer can be described as follows: The light beam (4) is the photon beam that contains a certain frequency range and perpendicularly affects the surface of the device, represented by two gratings (2) of bars (1). The disperser / demultiplexer device divides the incident beam (4) into two beams (5, 6) with two wavelengths (λ-i, A ₂ ), centered at 1.50 μm and 1.55 μm respectively, following different paths so that they form an angle (α) approximately 28 ° apart. Having a plane (3), which determines the region where such wavelengths (λi, λ ₂ ) are collected for analysis, located at a distance (X _f ) of about 20 μm, the photon beams (5, 6 ) dispersed have a width (δ) of about 1 μm in said plane (3).

In a practical example, the light beam (4) can be generated by a polarized laser so that its electric field oscillates in a direction parallel to the transverse axis of the bars (1), represented as an ordinate axis (and) in the Figure 2.

Figure 2 shows a cut in the plane perpendicular to the axes of the dielectric bars that make up the device, representing two dimensions its spatial arrangement. The material of manufacture of the bars (1), their number, measurements of the section and the spatial arrangement in layers can vary, according to the lithographic technique that is used in the implementation of the device, at the same time it depends on the parameters of distance (X _f ), width (δ) and angle (α), relative to the dispersion of the beams (5, 6), as set as optimal operating values of the demultiplexer. Such bars (1), in this embodiment using AsGa as a dielectric material, have a dielectric constant equal to approximately 11, 36, for the wavelengths (λ-i, λ ₂ ) of interest, that is, in this case , of the order of 1.50 μm.

As seen in the example of Figure 2, each of the dielectric bars (1) has a square section with dimensions of 0.4 μm x 0.4 μm. Since there are five layers, represented on the abscissa axis of Figure 2, the total thickness of the device is 2 μm, which makes it an ultra-compact demultiplexer.

To avoid the crosstalk between the two optical channels constituted by the pair of wavelengths (λ-ι, λ ₂ ), the separation (d) between the axis of incidence of the light beam (4) and each of the beams' (5, 6) in which it is divided by the demultiplexer, as shown in Figure 1, is set at 5 μm, said separation (d) being defined from the axis of the incident beam (4), transverse to the device, with respect to The position where said beams (5, 6) are captured within the plane (3). Therefore, the spatial separation (2d) between the two beams (5, 6) is worth 10 μm, in the plane (3) located at a distance (X _f ) of 20 μm.

Furthermore, when capturing the beams (5, 6) on said plane (3), a minimum area or width (δ) of 1 μm is provided, as stated above. All these parameters (δ, d, x _f ) introduced in the reverse design procedure result in a separation between beams (5, 6) with an angle equal to 28 ° and a crosstalk in both channels reduced to less than -25 dB , as can be seen in Figure 3.

Figure 3 illustrates the two-dimensional mapping of the 20log function (| E (x, y) λi | / 1 E (x, y) λ ₂ |), expressing on the abscissa axis (X) the distance between Ia first layer of the device and the plane (3) at the distance (X _f ) of up to approximately 20 μm, while the ordinate axis (Y) takes values of the cross section of the beams (5, 6) in said plane ( 3); that is, between the interval (d + δ 12, d- δ / 2) for the beam (5) measured in the positive ordinate (Y) values and the interval (-d + δ 12, -d- δ / 2) for the beam (6) considered in the quadrant of the negative (Y) ordinate space.

It is about maximizing the attenuation of the sum of crosstalk in the two established optical channels, that is, guaranteeing a minimum crosstalk between the two wavelengths (λ-ι, K ₂ ) for the coordinates (X _f , d) and ( X _f , -d), with a resolution equal to the width (δ) of the respective beams (5, 6). In Figure 3, the attenuation of about 25 dB achieved at the desired coordinates is indicated with a pair of rectangles (7), at 20 μm distance (x _f ) to the analysis plane (3) and a separation (d) of 5 μm ± 1 μm, which is the value of the width (δ) of the beams (5, 6). The dielectric bars (1) of the device are drawn, in the aforementioned Figure 3, as squares of 0.4 μm x 0.4 μm, according to a possible spatial arrangement in five layers, between 0 μm and 2 μm total thickness of the demultiplexer The light beam (1) is assumed to affect the first layer of the device, at X = 0, according to the direction of positive abscissa (X).

The terms with which this report has been written must always be taken in a broad and non-limiting sense.

The invention has been described according to this text and set of figures for some preferred embodiments thereof, but the person skilled in the art will be able to understand that multiple variations can be introduced in said embodiments and combined in various ways giving rise to more possible variants, without leave the scope defined by the claims included below.

Claims

1.- Optical demultiplexer, which can be integrated into a flat waveguide or an optical fiber, which spatially separates the wavelengths (λ-ι, λ ₂ ) contained in a beam of light (4) incident on the demultiplexer, to give its output a plurality of beams (5, 6) corresponding to the different wavelengths (λi, λ ₂ ) separated by an angle (α), characterized in that it consists of a plurality of layers of bars (1), manufactured in a dielectric material, configured with an optimum section to obtain a certain angle (α) of separation between the output beams (5, 6), which disperse the light beam (4) into its multiple wavelengths (λ- ι, λ ₂ ).

2. Optical demultiplexer according to claim 1, characterized in that the dielectric material with which the bars (1) are manufactured is AsGa.

3. Optical demultiplexer according to claim 1, characterized in that the section of the bars (1) is square and with a side equal to 0.4 μm.

4. Optical demultiplexer according to claim 1, characterized in that the number of layers of dielectric bars (1) that make up the demultiplexer is five.

5. Optical demultiplexer according to claim 1, characterized in that the angle (α) of separation between the beams (5, 6) is 28 °.

6. Optical demultiplexer according to claim 1, characterized in that the output beams (5, 6) correspond to the wavelengths (λi, A ₂ ) of 1.50 μm and 1.55 μm respectively.