WO2000052506A1

WO2000052506A1 - An apparatus and a method for changing refractive index

Info

Publication number: WO2000052506A1
Application number: PCT/DK2000/000082
Authority: WO
Inventors: Mikael Svalgaard
Original assignee: Ionas A/S
Priority date: 1999-02-26
Filing date: 2000-02-28
Publication date: 2000-09-08
Also published as: CA2363719A1; AU2794400A; IL145009A0; EP1180247A1

Abstract

In an apparatus and a method for changing the refractive index of at least part of a sample the degree with which the photosensitivity of the sample is decreased during UV exposure, the temperature of the sample is controlled at least during part of the exposure process in order to keep the temperature of the sample within a predetermined range. Preferably the temperature is kept low by a cooling device keeping the temperature within the range of -100 - +10°C. The sample to be exposed comprises a glass material, which is first subjected in a so-called loading process to pressures above 1 bar by H2 or D2, or other materials that act to enhance the UV-sensitivity of the glass. In an embodiment of the invention the method and the apparatus further comprises control means for controlling the relative movement between mounting means for a glass sample and one or more optical light beams.

Description

AN APPARATUS AND A METHOD FOR CHANGING REFRACTIVE INDEX

FIELD OF INVENTION

The present invention relates to apparatus and a method for changing the refactive index of a least part of a sample. In particular, the present invention relates to the use of ultraviolet electromagnetic radiation so as to change the refractive index of a sample so as to fabricate optical waveguide structures capable of guiding electromagnetic radiation.

BACKGROUND OF THE INVENTION

In the late 1970^'s and early 1980^'s it was discovered that UV radiation around 240 nm could permanently increase the refractive index of germanium doped silica. Typical index changes were in the 10^"6 to 10^"4 range. These findings quickly lead to the development of fiber Bragg grating technology which has found widespread industrial use in the late 1990's.

In the early 1990's several methods were discovered which permitted the photosensitivity of a given sample containing germanosilica to be increased. Most notably, a research group at AT&T Bell Labs (NJ, USA) showed in 1993 that loading the glass with molecular hydrogen or deuterium prior to UV exposure could enhance the photosensitivity by several orders of magnitude, so that index changes in the 10^"2 range could be induced with relatively low fluences. Absorption measurements show large changes occurring around 1.4 μm due to induced OH centers and a growing short wavelength tail caused by increased numbers of absorbers in the deep UV.

The photolytic reactions are currently not understood in detail, but it is evident that the physical mechanism involves some reaction between indiffused H₂ molecules with the glass matrix leading to the formation of several defect species. These defect species can then react further with the UV field, leading to refractive index changes. The dissociation energy of H₂ is 4.5 eV, so intense UV irradiation (hv « 5 - 6.5 eV) must evidently cause a large fraction of the indiffused molecules to photodissociate quickly. By hydrogen or H₂ we mean herein hydrogen and its isotope deuterium, D .

The photon energy is also large enough to excite and ionise several types of Ge related defect centres which adds to the complexity of the possible reactions. It is currently believed that several processes are occurring, of which one is a reaction known from earlier studies of thermally driven reactions in fibers with indiffused hydrogen. In this process dissociated H₂ reacts with oxygen atoms in the neighbourhood of Ge0₂ sites, while no reaction occurs if the dissociated H₂ is surrounded by Si0₂ sites:

H₂ - 2H (thermal or photolytic) (a)

I I

≡Si-0 ^{• • •}Ge- -0-Sb + 2H → 2≡Si-OH + ^"Ge= + H₂ (i.4 μm abs., uvabs.) (b)

≡Si-O-Si≡ + 2H - sSi-O-Sb + H₂ (recombination) (c)

The reaction with the glass matrix occurs only at Ge co-dopant sites because the oxygen bonds here are weaker than in the normal Si0₂ network (weaker bonds at co-dopant sites is believed to be the explanation why hydrogen loading also enhances the photosensitivity of P₂O₅ doped glass). Heating during UV exposure causes this reaction to be triggered both thermally and photolytically. The reaction results in the formation of Ge²⁺ defects, i.e. a Ge atom in the 2+ oxidation state rather than the 4+ oxidation state. The Ge²⁺ defect is one of several types of Ge related oxygen deficient defects which are also naturally present in germanosilica, accounting for about 0.1-1% of the total number of Ge atoms. These defects absorb strongly in the hv « 3.6 - 6.9 eV range and can thus react further with the UV field in a manner similar to that occurring for intrinsic Ge²⁺ defects when H₂ loading is not applied. It is this type of reaction and consequences thereof which is believed to be responsible for the observed UV induced index change at longer wavelengths. The presence of H₂ facilitates a reaction at every Ge site in the glass; without H₂ loading UV induced reactions only occur at a very limited number of Ge sites, namely those intrinsically incorporated in lower oxidation states. Thus, H₂ loading can dramatically increase the photosensitivity of germanosilica.

When large index changes were within reach it was quickly realised that not only could UV radiation be used to modify the cores of existing waveguides (as in Bragg gratings); it could also be used to induce the waveguiding structures themselves in planar glass samples. Using UV radiation to induce waveguides could potentially yield more simple, reliable and cheap planar waveguide fabrication methods than the currently applied photolithography and etching based techniques. In 1993 and 1994 it was demonstrated by several groups that channel waveguides could be photo-induced with UV radiation in thin film waveguides having a photosensitive core layer. In chronological order:

1) Meltz et al. ("UV-induced Bragg gratings in optical fibers and thin-film waveguides", SPIE vol. 2044, 1993, 236-245) created straight waveguides using a cylindrical lens focussing pulsed UV radiation into a thin line. Besides demonstrating that it is possible to induce waveguides, this technique had little further development potential since it is only possible to make straight waveguides.

2) Mizrahi et al. ("Ultraviolet laser fabrication of ultrastrong optical fiber gratings and of germania-doped channel waveguides", Appl. Phys. Lett., Vol. 63, 13, 1993, 1727-1729 demonstrated a much more versatile technique in which the desired waveguide structure is photo-induced with UV radiation from an excimer laser through a metal mask deposited on the top cladding layer.

3) Svalgaard et al. ("Direct UV-writing of buried single-mode channel waveguides in Ge-doped silica films", Elec. Lett., Vol. 30, 1994, 1401-1402) has developed a fabrication method where the desired waveguide pattern is written directly into the glass by scanning the sample beneath a continuous wave UV beam focussed to a small spot size (hereafter denoted the "direct writing technique"). Loading a glass sample with hydrogen to enhance the photosensitivity is carried out at low temperatures (<250 °C) and high partial pressure (>1 bar, typically a few hundred bar) and occurs by means of a diffusion process where hydrogen inertly occupies voids in the glass matrix. The glass should be in this loaded state when UV exposure is carried out. After completion of the UV exposure, remaining hydrogen will diffuse inertly out of the glass. The diffusion process may be described using simple one-dimensional diffusion equations with a diffusivity, DH2_> given by the Arrhenius equation:

D_H2 = 2.8x10^"4 Exp[- E/RT] (1)

Where E = 40 kJ/mol for silica, R = 8.31 J/Kxmol and T is the absolute temperature. The equilibrium H₂ concentration at room temperature (T = 21 °C) has been determined to be approximately:

c_eq = 166 ppm/bar (2)

where 1 ppm is defined as 10^"6 moles of H₂ per mole of Si0₂. Thus the indiffused concentration is proportional to the applied pressure. For one dimensional diffusion the characteristic time to traverse a length I is given by:

tι = I74D (3)

Consequently, for a glass sample of a given geometry (i.e. fiber or planar), it follows that the characteristic diffusion time is proportional to the square of the smallest sample dimension, i.e. the diameter of a fiber or the layer thickness of a planar waveguide.

For silica-based optical fibers with an outer cladding diameter of 125 μm the time required to reach, say 95%, of the equilibrium concentration at room temperature is approximately τ₉₅ = 100 hours. For a 30 μm thick silica thin film on a silicon wafer the equivalent time is τ_g5 = 20 hours. The characteristic times for indiffusion and for outdiffusion are very similar since the process is governed by the same diffusivity.

For practical reasons, the UV exposure cannot be carried out while the sample is confined in a high pressure hydrogen atmosphere; instead the sample is removed from the loading chamber and mounted in the set-up used for UV exposure. Hence, the hydrogen concentration will continually decrease during a UV exposure.

It is a major problem, that outdiffusion will lead to a continually decreasing degree of photosensitivity. This phenomenon can be of especially large importance for applications where a number of identical structures are to be induced sequentially on a single sample, such as with the direct UV writing technique.

To illustrate the significance of this effect consider the following example where a number of straight waveguides were directly written with a focussed UV beam. The sample consisted of a three layer silica on silicon structure with a photosensitive core layer. The sample was loaded with D₂ at a pressure of 190 bars for 10 days at room temperature. The experiment was carried out at room temperature (24 °C) for which the D₂ concentration decreased with a 1/e time constant of 13 hours. A number of 2 cm long, identical waveguides were written over a time period, t, being up to several hours.

The waveguides were evaluated by measuring the total insertion loss and polarization dependent loss (PDL) using butt-coupled standard telecom fibers and a polarized light source with a wavelength of 1557 nm. The measured insertion loss and PDL is shown in figure 1 against the time of fabrication, measured from the start of UV writing. Initially, the insertion loss is -0.5 dB while the PDL is ~0.15 dB. However, the insertion loss increased rapidlywith time reaching 3 dB for t = 2 hours. Over the same time interval the PDL increased to 2 dB. These changes in waveguide performance are most probably caused by the guided mode becoming less confined as the D₂ concentration and thus also the UV induced index change decreases with time. Changes of this magnitude in waveguide performance with the time of fabricationare not acceptable for commercial applications of the direct UV writing technique.

US patent 5,287,427 discloses a method where large refractive index changes can be obtained in oxide glass-based optical waveguides by a treatment that comprises exposing at least a portion of the waveguide to H₂ or D₂ and irradiating at least part of the exposed portion with UV radiation. US patent 5,287,427 also mentions that storage of H₂ or D₂ loaded samples should be carried out at low temperature so that the rate of outdiffusion is reduced. US patent 5,287,427 reveals nothing about how to stabilize the photosensitivity during UV exposure.

US Patent 5,235,659 also discloses a method where large refractive index changes can be obtained in oxide glass-based optical waveguides using H₂ or D₂ loading and exposing the loaded part to UV radiation. US patent 5,235,659 also mentions that storage of H₂ or D₂ loaded samples should be carried out at low temperature so that the rate of outdiffusion is reduced. Again, no emphasis has been put on stabilizing the photosensitivity during UV exposure.

It is a disadvantage of the method disclosed in US patent 5,287,427 and US Patent 5,235,659, that the performance of the waveguides changes over a few hours due to outdiffusion of H₂ or D₂.

It is a further disadvantage of the method disclosed in US patent 5,287,427 and US Patent 5,235,659, that due to outdiffusion of H₂ or D₂ the method disclosed is not acceptable for commercial applications where several identical structures are to be induced sequentially in a sample.

It is an object of the present invention to provide an apparatus and a method for reducing the degree with which the photosensitivity of a sample decreases during UV exposure. It is a further object of the present invention to provide an apparatus and a method for reducing the degree with which the photosensitivity of a hydrogen and/or deuterium loaded sample decreases during UV exposure.

It is a still further object of the present invention to provide an apparatus and a method for controlling the temperature of a hydrogen and/or deuterium loaded sample during UV exposure so as to reduce the degree with which the photosensitivity of a sample decreases during UV exposure.

Finally, it is an object of the present invention to provide an apparatus and a method for fixing the photosensitivity of a hydrogen and/or deuterium loaded sample prior to fabricating refractive index structures by UV exposure.

SUMMARY OF THE INVENTION

The above-mentioned objects are complied with by providing an apparatus for changing the refractive index of at least part of a sample while, simultaneously, stabilising the photosensitivity of said same sample, said apparatus comprising:

- means for emitting one or more optical light beams,

- mounting means for holding the sample,

- means for transforming the one or more optical light beams into a predetermined intensity pattern, and for projecting said predetermined intensity pattern onto at least part of the sample so as to change the refractive index of the sample in accordance with said projected intensity pattern,

- means for appointing a temperature to at least part of the sample at least during part of the exposure process, and - a control means for controlling the temperature of at least part of the sample at least during part of the exposure process so as to keep the temperature of at least part of the sample within a predetermined range.

Further, the above-mentioned objects are complied with by providing an apparatus for fixing the photosensitivity of at least part of a sample, said apparatus comprising:

means for emitting one or more optical light beams,

mounting means for holding the sample,

- means for transforming the one or more optical light beams into a predetermined intensity pattern, and for projecting said predetermined intensity pattern onto at least part of the sample in a so-called pre-exposure process so as to cause indiffused H₂ or D₂ to react with the glass material in accordance with said projected intensity pattern,

- means for appointing a temperature to at least part of the sample at least during part of the pre-exposure process, and

- a control means for controlling the temperature of at least part of the sample at least during part of the pre-exposure process so as to keep the temperature of at least part of the sample within a predetermined range.

The emitting means may comprise one or more laser light sources, such as an argon-ion laser or an excimer laser. The argon-ion lasers may be frequency doubled argon-ion lasers emitting laser light in the range 220-270 nm, preferably in the range 244-257 nm, such as approximately 244 nm or 257 nm. The emitting means may also be one or more excimer lasers. The excimer lasers may be emitting light at a wavelength of approximately 193 nm or approximately 248 nm. The appointing means may comprise a cooling device, such as at least one Peltier element adjoining a cooling element, such as a heat sink having an inlet and an outlet for a coolant. The mounting means may comprise a vacuum chuck so as to affix the sample. Preferably, the vacuum chuck is in contact with a thermocoupler, said thermocoupler being positioned between the sample and the cooling element.

The temperature of at least part of the sample at least during part of the exposure process should preferably be within the range -100° - +10° C, more preferably in the range -50° - -20° C, even more preferably in the range -45° - -25° C, even more preferably in the range -40° - -30° C, even more preferably in the range -38° - -32° C, even more preferably the range -36° - -34° C, such as approximately - 35° C. The temperature may be controlled using a microprocessor.

The projecting means may comprise a transparent material comprising a one- dimensional surface-relief grating on one side of the transparent material so as to vary the thickness of the transparent material.

The sample to be exposed to the predetermined intensity pattern may comprise a glass material which is first loaded at pressures above 1 bar by H₂ or D₂ or other materials that act to enhance the UV-sensitivity of the glass.

Further, the above-mentioned objects are complied with by providing an apparatus for changing the refractive index of at least part of a sample, while, simultaneously, stabilising the photosensitivity of said same sample, said apparatus comprising:

- means for emitting one or more optical light beams,

- mounting means for holding the sample,

- means for preparing the one or more optical light beams so as to expose at least part of the sample to the prepared one or more optical light beams, - means for moving the mounting means and the prepared one or more optical light beams relative to each other at least during part of the exposure process,

- means for appointing a temperature to at least part of the sample at least during part of the exposure process,

- a first control means for controlling the relative movement between the mounting means and the prepared one or more optical light beams, said relative movement being determined by a set of moving parameters, and

- a second control means for controlling the temperature of at least part of the sample at least during part of the exposure process so as to keep the temperature of at least part of the sample within a predetermined range.

The emitting means may comprise one or more laser light sources of the type previously mentioned. The cooling device, mounting means and the preferred temperature ranges are also similar to what has been previously mentioned.

The moving means must be capable of moving the mounting means and the prepared one or more optical light beams relative each other in at least one- dimension. The moving means typically translate the sample in a computer controlled trajectory (two-dimensional) scanning process, which is carried out with sub-micron accuracy to ensure a high degree of reproducibility. Thus, the moving parameters comprise scan direction and scan speed.

At least one of the moving parameters may be varied at least during part of the exposure process. In particular the scan direction and/or scan speed may be varied at least during part of the exposure process. Also the power and/or the beam diameter of the prepared one or more optical light beams exposing at least part of the sample may be varied at least during part of the exposure process. The preparing means may comprise reflective elements, such as mirrors and/or gratings and/or holograms. The preparing means may also comprise diffractive elements, such as gratings and/or holograms. Finally, the preparing means may comprise refractive elements, such as one or more lenses.

The first and second control means may comprise a microprocessor for controlling the moving means and the temperature of the sample.

Finally, the above-mentioned objects are complied with by providing a method for changing the refractive index of at least part of a sample, said method comprising the steps of:

- preparing one or more emitted optical light beams so as to obtain a predetermined intensity pattern,

exposing at least part of the sample to the predetermined intensity pattern,

The method may further comprise the steps of:

- appointing a temperature to at least part of the sample at least during part of the exposure process, and

- controlling the temperature of at least part of the sample at least during part of the exposure process so as to keep the temperature of at least part of the sample within a predetermined range.

The refractive index change according to the above-mentioned method is provided using any of, or a combination of, the apparatus previously described. BRIEF DESCRIPTION OF THE DRAWINGS

In the following, the invention is further disclosed with reference being made to the drawings.

Figure 1 shows measurements of the insertion loss and PDL of directly UV written waveguides as a function of the time of fabrication, as measured from the start of UV writing (sample temperature: 24 °C). The increasing insertion loss and PDL is a consequence of D₂ outdiffusion during UV writing.

Figure 2 shows the calculated time available for UV processing as a function of the sample temperature for various values of the smallest tolerable relative hydrogen concentration, α.

Figure 3 shows a cooling system for samples of a planar geometry.

Figure 4 shows measured insertion loss and PDL of directly UV written waveguides as a function of the time of fabrication, as measured from the start of UV writing.

The sample used here was cooled to a temperature of -33 °C during the UV writing.

Due to the low sample temperature no significant D₂ outdiffusion occurs and hence the insertion loss and PDL does not increase with the rate that it would otherwise do if no cooling was applied (fig. 1).

DETAILED DESCRIPTION OF THE INVENTION

According to a first aspect of the present invention a hydrogen and/or deuterium loaded sample is affixed to an actively cooled mount, thereby lowering the temperature of the sample. From the Arrhenius equation (1) it is seen that the diffusivity is proportional to Exp[-E/RT], hence by lowering the temperature during a UV exposure the diffusivity is reduced and consequently, the outdiffusion occurs at a lower rate. Since the outdiffusion occurs at a lower rate the decline in photosensitivity with time is reduced.

For most practical applications the decreasing hydrogen concentration, c, in a sample as a function of the time, t, since it was removed from the loading facility can be approximated by an exponential decay with a characteristic time constant, τ:

c(t) = Co Exp[-t/τ] (4)

where Co is the initial hydrogen concentration. The time constant, τ, is dependent on the sample geometry and dimensions. As a simple example, consider one dimensional diffusion for which it is seen from eq. 3 that the time required to traverse a length I, and thus also the characteristic decay time, is inversely proportional to the diffusivity. Hence; in this example we have:

τ = τ₀ Exp[E/RT - E/RT₀] (5)

where τ₀ is the decay time at the temperature T₀. By cooling the sample during UV exposure and hence lowering T is it seen that τ increases, i.e. the outdiffusion occurs more slowly. Assume that a maximum relative degree of outdiffusion, α, exists for which the performance of a UV induced structure remains acceptable. No outdiffusion corresponds to α = 1 and full outdiffusion corresponds to = 0. Hence, there exists some upper time limit, τuv, above which it is no longer possible to obtain the desired performance. From eq. (4) and (5) it is seen that:

α = Exp[- τuv/τ] α = Exp[- τuv/τo Exp[E/RT₀- E/RT]] τuv = -Ln[α] τ₀ Exp[E/RT - E/RT₀] (6)

Hence, it is an advantage of the first aspect of the present invention that by lowering the sample temperature during UV exposure the exposure time available is significantly increased. This is illustrated in fig. 2 for τ₀ = 13 hours and T₀ = 20 °C, as measured for a typical three layer silica on silicon planar sample used for direct UV writing of waveguides. For α = 0.95 the available time for UV exposure is ~40 minutes at T = 20 °C. By cooling the sample to T = -30 °C the available time for UV exposure is increased to almost 20 hours.

According to a second aspect of the present invention a UV pre-exposure of the hydrogen loaded sample is performed prior to the UV exposure (fabrication exposure) that would normally be performed to induce the desired refractive index change. The purpose of the pre-exposure is to photolytically trigger a reaction between indiffused H₂ and the glass material. One such reaction is where H₂ is photodissociated, subsequently reacting with Ge-O bonds, thereby creating OH centres and Ge related oxygen deficient centres. This may lead to an ordinary Ge0₂ site being transformed into a Ge²⁺ centre, as described in eq. (b). When the density of Ge related oxygen deficient centres has been increased in this way the pre-exposure can be stopped and remaining H₂ be allowed to outdiffuse. At this time the photosensitivity of the sample has been increased, however the sample is no longer loaded with H₂.

Hence, it is an advantage of the second aspect of the present invention that problems associated with H₂ outdiffusion during the fabrication exposure are avoided and the photosensitivity remains stable. There are the following requirements to the pre-exposure:

1) the photon energy should be greater than or comparable to the H₂ photodissociation energy of 4.5 eV,

2) the total fluence should cause a fraction of the H₂ molecules to react with the glass material, and 3) the pre-exposure should partially or fully cover the region of the sample which is to be processed in the fabrication exposure.

When H₂ diffuses through silica based glass it inertly occupies voids capable of holding typically just one single molecule. If H₂ is brought to react in a void which does not have a Ge-O bond in sufficient proximity no Ge atoms will be reduced.

Thus, it is possible that only a fraction of the indiffused H₂ will be brought to react with GeO₂ sites in the duration of the pre-exposure. However, if the pre-exposure is repeated one or more times with a time interval sufficiently large for widespread diffusion of H₂ to occur it is possible to increase the fraction of H₂ which reacts with Ge0₂ sites. Hence, multiple pre-exposures can lead to larger concentrations of Ge related oxygen deficient centres and thus also to higher degrees of photosensitivity.

The fluence required to satisfactorily complete the pre-exposure may be reduced by increasing the number density of indiffused H2. Hence, there will be a larger number og H₂ molecules in voids where reactions with Ge-O bonds are possible and therefore a smaller fluence is required in the pre-exposure. The number density of indiffused H₂ may be increased by increasing pressure in the loading chamber from the usual several hundred bar to several thousand bar.

The present invention comprises active cooling of glass samples during fabrication of optically induced refractive index structures using UV exposure. It further comprises a pre-exposure of H₂ or D₂ loaded glass samples prior to fabrication of optically induced refractive index structures using UV exposure

In a first embodiment of the first aspect of the present invention and referring to figure 3, a planar glass sample is affixed to a chuck made of a material with a good degree of thermal conductivity, such as copper. The chuck should have lateral dimensions greater than or equal to those of the sample area which is to be cooled. The chuck contains a number of small holes through which a vacuum is applied so that the sample can be affixed firmly to the upper side of the chuck. The temperature of the chuck can be monitored with a thermocouple placed in a small hole through the side of the chuck.

A Peltier element with approximately the same lateral dimensions as the vacuum chuck is affixed to the lower side of the vacuum chuck. The other side of this Peltier element is affixed to a heat sink. For most applications the heat sink will need to be actively cooled. For example, the heat sink cooling may be performed with circulating air or water. For the cooling system to perform most efficiently it is important to ensure that there is a good degree of thermal contact between adjacent components. By applying a voltage of the correct polarity to the leads of the Peltier element heat will be transported from the vacuum chuck to the heat-sink and the mounted sample will be cooled. To prevent frost from forming on the sample the mount may be enclosed in a container with a dry atmosphere, such as pure N₂. Obviously this container should have a transparent window through which the radiation used for material processing can enter.

A total of two cooling systems have been implemented. The first system consists of a 4x4 cm² vacuum chuck, a 4x4 cm² single-stage Peltier element with a capacity of 69 W and an air-cooled heat sink. This system can cool a sample approximately 40 K below the ambient temperature and it is mainly limited by the limited efficiency of the air-cooled heat-sink. The second system consists of a 6x6 cm² vacuum chuck, a 6x6 cm² single-stage Peltier element with a capacity of 120 W and a water-cooled heat-sink. This system can cool a sample approximately 60 K below ambient temperature when a voltage of 10 V is applied.

A second embodiment of the first aspect of the present invention comprises a cooling system for optical fibers. The cooling system is identical to that described in Example 1 except in that the vacuum chuck is replaced with a metal plate having one or more V-grooves machined into the top side. An optical fiber may then be placed in each V-groove and by affixing it in at least two points the system can cool the fiber efficiently. The preferred embodiment of the present invention comprises active cooling during fabrication of optical waveguides using UV radiation. The cooling system according to the first embodiment of the present invention is very useful for reducing the effects of D₂ outdiffusion when fabricating optical waveguides or components in planar glass samples with UV radiation.

In a preferred embodiment of the second aspect of the present invention a planar glass sample or a fiber is affixed to a mount of the type described in the previous sections. Prior to the UV expsoure used to fabricate the desired structure in the sample the sample is subjected to a pre-exposure of the type previously described. During the pre-exposure the sample is translated so that a pre-determined area of the sample is covered homogeneously by the pre-exposure. After the pre-exposure is completed residual H₂ or D₂ is allowed to diffuse inertly out of the sample.

The primary requirement when fabricating Bragg gratings is to create a UV interference pattern with a period resulting in a Bragg resonance at the desired wavelength. This involves splitting a beam to create two coherent sources and subsequently recombining the two parts to set up the interference pattern. Two commonly applied methods for realising interference patterns are:

1) The free space interferometric method uses a two beam interference pattern produced in a Mach Zender interferometer, hence applying amplitude splitting.

2) The phase mask method employs a diffractive optical element, a phase mask, which is much more robust and simpler to apply than the free space interferometric method. The phase mask consists of a high quality optical flat made of UV transparent material (typically fuzed silica) with a one-dimensional surface-relief grating on one side. A quasimonochromatic plane wave phase mask results in an emergent field consisting of a set of diffracted plane waves, the interference of which produces the desired interference pattern. WO 00/52506 _n PCT/DKOO/00082

18

The phase mask method can easily be combined with other waveguide fabrication techniques such as patterning or direct writing. For example, after a waveguide has been photo-induced by e.g. direct writing a phase mask can be placed in contact with the sample and another UV exposure can be carried out, wherein the previously fabricated waveguide is exposed to a UV interference pattern thereby creating a Bragg grating in the waveguide.

Hence, it is an advantage of the present invention that it provides the possibility to combine the fabrication of waveguides by direct writing with the fabrication of Bragg gratings using the phase mask technique.

The direct writing set-up may employ a 244 nm, frequency doubled Ar⁺ laser producing -50 mW of power incident on a sample in a 1/e² spot size of 5-6 μm. The waveguide pattern is defined by translating the sample in a computer controlled trajectory scanning process, which is carried out with sub-micron accuracy to ensure a high degree of reproducibility. The fabrication time is determined by the sample photosensitivity which sets an upper limit on the applicable scan speed. With the samples currently in use, scan speeds in the range from 100-400 μm/sec are typically applied. The resulting fabrication time of basic components such as 1x2 splitters or 2x2 directional couplers is roughly one minute per device.

Since no masking is required direct writing requires no cleanroom processing besides the basic glass deposition process. Waveguide parameters such as core width and index step are determined by the spot size, incident power and the scan speed. Since all these parameters can be controlled independently during the scanning process it is possible to fabricate waveguides with a precisely defined, longitudinally varying refractive index step. This is useful for a variety of applications, such as minimizing bending loss, optimizing fiber-waveguide coupling, introduction of phase delays and fabrication of long period gratings. For example, by varying the scan speed it is possible to control the magnitude of the UV induced index change. Then one will typically apply a low scan speed when writing a curved waveguide where a large index step is required to minimize bend losses. When writing a straight waveguide a higher scan speed can be used since the required index step generally is smaller for straight waveguides as compared to curved waveguides. In another example, the waveguide width can be varied by changing the UV spot size on the sample. This is most easily achieved by changing the distance between the sample and the optical element used for focussing.

In yet another example, the waveguide index step can be controlled by varying the amount of UV power incident on the sample. In the example of curved and straight waveguides mentioned above, one may increase the UV power when making curved waveguides requirering a large index step and decrease the UV power when making straight waveguides where the required index step is lower.

Hence, it is an advantage of the present invention that it offers a high degree of control over the waveguide parameters, more than any other silica-based waveguide fabrication technique.

Typical samples are deposited on silicon wafers using Plasma Enhanced Chemical Vapour Deposition, and consist of a three layer buffer-core-cladding glass structure. Such samples can also be fabricated by other means, such as flame hydrolysis or the like. The buffer/cladding thickness is -15 μm while the core layer thickness typically is in range from -2 μm to -6 μm.

The chemical composition of the core layer is highly critical for the optical characteristics of UV written waveguides and the speed with which they can be fabricated. Besides doping the core with germanosilica we have worked with a variety of additional co-dopants, including silica-oxynitride and boron. Both enhance the photosensitivity; however adding silica-oxynitride increases the refractive index while boron lowers it. Thus, boron co-doping permits realization of so-called 'index-matched' structures where the refractive index of the core layer differs insignificantly from that of the buffer/cladding, even though it contains a significant concentration of germanosilica. Unless index-matching is used UV generated waveguides will exhibit elliptical mode profiles (and thus high coupling loss to standard telecom fiber) due to the asymmetry imposed by the presence of a high-index core layer.

To demonstrate the usefulness of active cooling consider the following example where a number of straight waveguides were directly written with a focussed UV beam. The sample was loaded with 2.2 mole% of D₂ and consisted of a three layer, index matched silica on silicon structure with a photosensitive core layer.

Two experiments were carried out where the only difference was whether or not active cooling was applied. Without cooling the sample temperature was 24 °C and with cooling the sample temperature was -33 °C. The time from removal of the sample from the D₂ pressure chamber to the start of UV writing was 50 minutes.

A number of 2 cm long, identical waveguides were written over a time period ranging up to several hours. After the UV fabrication was completed, the samples were annealed at 80 °C for 24 hours to remove residual D₂. The waveguides were evaluated by measuring the total insertion loss and polarization dependent loss (PDL) using butt-coupled standard telecom fibers and a polarized light source with a wavelength of 1557 nm. Without cooling waveguides fabricated shortly after UV writing was started exhibited an insertion loss of 0.5 dB and a PDL of 0.15 dB. However, the insertion loss and PDL increased steadily with time reaching 3 dB and 2 dB, respectively, for waveguides made two hours after the start of UV writing (figure 1).

These changes in waveguide performance are most probably caused by the guided mode becoming less confined as the D₂ concentration and thus also the UV induced index change decreases with time. When cooling was applied the insertion loss remained at -0.5 dB while the PDL remained at less than -0.2 dB (figure 4) for up to 10 hours of UV writing.

Hence, it is an advantage of the present invention that active cooling of a D₂ loaded planar glass sample used for UV writing of waveguides can prevent degradation of the waveguide performance due to outdiffusion.

Claims

1. An apparatus for changing the refractive index of at least part of a sample while, simultaneously, stabilising the photosensitivity of said same sample, said apparatus comprising:

means for emitting one or more optical light beams,

mounting means for holding the sample,

- means for appointing a temperature to at least part of the sample at least during part of the exposure process, and

- a control means for controlling the temperature of at least part of the sample at least during part of the exposure process so as to keep the temperature of at least part of the sample within a predetermined range.

2. An apparatus for fixing the photosensitivity of at least part of a sample, said apparatus comprising:

- means for emitting one or more optical light beams,

- mounting means for holding the sample,

means for transforming the one or more optical light beams into a predetermined intensity pattern, and for projecting said predetermined intensity pattern onto at least part of the sample in a so-called pre-exposure process so as to cause indiffused H₂ or D₂ to react with the glass material in accordance with said projected intensity pattern,

3. An apparatus according to claims 1 or 2, wherein the emitting means comprises at least one laser light source, such as an argon-ion laser or an excimer laser.

4. An apparatus according to claim 3, wherein the emitting means comprises at least one frequency doubled argon-ion laser emitting laser light in the range 220-270 nm, preferably in the range 244-257 nm, such as approximately 244 nm or 257 nm.

5. An apparatus according to claim 3, wherein the emitting means comprises at least one excimer laser emitting light at a wavelength of approximately 193 nm or approximately 248 nm.

6. An apparatus according to any of the preceding claims, wherein the appointing means comprises a cooling device.

7. An apparatus according to claim 6, wherein the cooling device comprises at least one Peltier element, said at least one Peltier element adjoining a cooling element, such as a heat sink.

8. An apparatus according to any of the preceding claims, wherein the temperature of at least part of the sample at least during part of the exposure process is within the range -100° - +10° C, preferably in the range -50° - -20° C, even more preferably in the range -45° - -25° C, even more preferably in the range -40° - -30° C, even more preferably in the range -38° - -32° C, even more preferably the range -36° - -34° C, such as approximately - 35° C.

9. An apparatus according to any of claims 1 , 3-8, wherein the projecting means comprises a transparent material comprising a one-dimensional surface-relief grating on one side of the transparent material so as to vary the thickness of the transparent material.

10. An apparatus according to any of the preceding claims, wherein the control means comprises a microprocessor.

11.An apparatus according to any of the preceding claims, wherein the mounting means comprises a vacuum chuck.

12. An apparatus according to any of the preceding claims, wherein the sample to be exposed to the predetermined intensity pattern comprises a glass material.

13. An apparatus according to any of the preceding claims, wherein the sample to be exposed to the predetermined intensity pattern comprises a glass material which is first subjected in a so-called loading process to pressures above 1 bar by H₂ or D₂ or other materials that act to enhance the UV-sensitivity of the glass.

14. An apparatus according to any of the preceding claims, wherein the sample to be exposed to the predetermined intensity pattern comprises a glass material which is first subjected in a so-called loading process to pressures between 1 bar and 500 bar by H₂ or D₂ or other materials that act to enhance the UV- sensitivity of the glass.

15. An apparatus according to any of the preceding claims, wherein the sample to be exposed to the predetermined intensity pattern comprises a glass material which is first subjected in a so-called loading process to pressures between 500 bar and 5000 bar by H₂ or D₂ or other materials that act to enhance the UV- sensitivity of the glass.

16. An apparatus for changing the refractive index of at least part of a sample, while, simultaneously, stabilising the photosensitivity of said same sample, said apparatus comprising:

- means for emitting one or more optical light beams,

- mounting means for holding the sample,

- means for preparing the one or more optical light beams so as to expose at least part of the sample to the prepared one or more optical light beams,

- means for moving the mounting means and the prepared one or more optical light beams relative to each other at least during part of the exposure process,

17. An apparatus according to claim 16, wherein the emitting means comprises at least one laser light source, such as an argon-ion laser or an excimer laser.

18. An apparatus according to claim 17, wherein the emitting means comprises at 5 least one frequency doubled argon-ion laser emitting laser light in the range

220-270 nm, preferably in the range 244-257 nm, such as approximately 244 nm or 257 nm.

19. An apparatus according to claim 17, wherein the emitting means comprises at 10 least one excimer laser emitting light at a wavelength of approximately 193 nm or approximately 248 nm.

20. An apparatus according to any of claims 16-19, wherein the appointing means comprises a cooling device.

15

21. An apparatus according to claim 20, wherein the cooling device comprises at least one Peltier element, said at least one Peltier element adjoining a cooling element, such as a heat sink.

20 22. An apparatus according to any of claims 16-21 , wherein the temperature is within the range -100° - +10° C, preferably in the range -50° - -20° C, even more preferably in the range -45° - -25° C, even more preferably in the range -40° - - 30° C, even more preferably in the range -38° - -32° C, even more preferably the range -36° - -34° C, such as approximately - 35° C.

25

23. An apparatus according to any of claims 16-12, wherein the moving parameters comprise scan direction and scan speed.

24. An apparatus according to any of claims 16-23, wherein at least one of the 30 moving parameters is/are varied at least during part of the exposure process.

25. An apparatus according to any of claims 16-24, wherein the scan direction and/or scan speed are/is varied at least during part of the exposure process.

26. An apparatus according to any of claims 16-25, wherein the power of the prepared one or more optical light beams exposing at least part of the sample is varied at least during part of the exposure process.

27. An apparatus according to any of claims 16-26, wherein the beam diameter of the prepared one or more optical light beams exposing at least part of the sample is varied at least during part of the exposure process.

28. An apparatus according to any of claims 16-27, wherein the preparing means comprises reflective elements, such as mirrors and/or gratings and/or holograms.

29. An apparatus according to any of claims 16-28, wherein the preparing means comprises diffractive elements, such as gratings and/or holograms.

30. An apparatus according to any of claims 16-29, wherein the preparing means comprises refractive elements, such as one or more lenses.

31. An apparatus according to any of claims 16-30, wherein the first and second control means comprises a microprocessor.

32. An apparatus according to any of claims 16-31 , wherein the mounting means comprises a vacuum chuck.

33. An apparatus according to any of claims 16-32, wherein the sample to be exposed to the prepared one or more optical light beams comprises a glass material.

34. An apparatus according to any of claims 16-33, wherein the sample to be exposed to the prepared one or more optical light beams comprises a glass material which is first subjected in a so-called loading process to pressures above 1 bar by H₂ or D₂ or other materials that act to enhance the UV-sensitivity of the glass.

35. An apparatus according to any of claims 16-34, wherein the sample to be exposed to the prepared one or more optical light beams comprises a glass material which is first subjected in a so-called loading process to pressures between 1 bar and 500 bar by H₂ or D₂ or other materials that act to enhance the UV-sensitivity of the glass.

36. An apparatus according to claims 16-34, wherein the sample to be exposed to the prepared one or more optical light beams comprises a glass material which is first subjected in a so-called loading process to pressures between 500 bar and 5000 bar by H₂ or D₂ or other materials that act to enhance the UV- sensitivity of the glass.

37. A method for changing the refractive index of at least part of a sample, said method comprising the steps of:

- preparing one or more emitted optical light beams so as to obtain a predetermined intensity pattern, and

- exposing at least part of the sample to the predetermined intensity pattern.

38. A method according to claim 37, wherein said method further comprises the steps of:

- appointing a temperature to at least part of the sample at least during part of the exposure process, and - controlling the temperature of at least part of the sample at least during part of the exposure process so as to keep the temperature of at least part of the sample within a predetermined range.

5 39. A method according to claims 37-38, wherein the refractive index change is provided using an apparatus according to any of claims 1 , 3-15.

40. A method according to claims 37-38, wherein the refractive index change is provided using an apparatus according to any of claims 16-36.

10

41. A method according to claims 37-38, wherein the refractive index change is provided using an apparatus according to any of claims 1 , 3-15 and subsequently using an apparatus according to any of claims 16-36.

15 42. A method according to claims 37-38, wherein the refractive index change is provided using an apparatus according to any of claims 16-36 and subsequently using an apparatus according to any of claims 1 , 3-15.

43. A method according to claim 37 for changing the refractive index of at least part 20 of a sample, wherein the photosensitivity is first fixed using an apparatus according to claim 2.

44. A method according to claims 37-38 for changing the refractive index of at least part of a sample, wherein the photosensitivity is first fixed using an apparatus

25 according to claim 2 and the refractive index change is subsequently provided using an apparatus according to any of claims 1 , 3-15.

45. A method according to claims 37-38 for changing the refractive index of at least part of a sample, wherein the photosensitivity is first fixed using an apparatus

30 according to claim 2 and the refractive index change is subsequently provided using an apparatus according to any of claims 16-36.

46. A method according to claims 37-38 for changing the refractive index of at least part of a sample, wherein the photosensitivity is first fixed using an apparatus according to claim 2 and the refractive index change is subsequently provided using an apparatus according to any of claims 1 , 3-15 and subsequently using an apparatus according to any of claims 16-36.

47. A method according to claims 37-38 for changing the refractive index of at least part of a sample, wherein the photosensitivity is first fixed using an apparatus according to claim 2 and the refractive index change is subsequently provided using an apparatus according to any of claims 16-36 and subsequently using an apparatus according to any of claims 1 , 3-15.