JPH11295544A - Manufacture of buried planar optical wave circuit element - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce propagation loss and to provide an optical wave circuit element having characteristics more corresponding to design in a buried planar optical wave circuit element provided with plural cores provided with a fine structure. SOLUTION: At the time of forming an upper clad layer 34 so as to cover the cores 48 having the fine structure, a film forming process and a film etching process are successively repeatedly performed. That is, a film being a part of the upper clad layer formed by the film forming process is anisotropically etched by the film etching process so as to widen the opening of the part pertinent to the upper side (viewing from the side of a lower clad layer 12) of the film. By repeating the film forming process and the film etching process for the plural number of times, the fine clearance of the cores 48 is covered and filled with the upper clad layer 34. Thus, the confinement of light to the cores 48 is improved, or desired propagation constant and coupling length are obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a method of manufacturing a buried planar lightwave circuit device.

[0002]

2. Description of the Related Art Conventionally, embedded planar lightwave circuit elements such as directional couplers and optical switches have been developed and put to practical use. The buried planar lightwave circuit element has a structure in which cores serving as optical waveguides are arranged in a plane, and a cladding layer (a lower cladding layer and an upper cladding layer) is provided around these cores. The directional coupler is a lightwave circuit element utilizing a phenomenon that power is transferred between the optical waveguides when two optical waveguides are arranged at a minute interval within several times the wavelength.

On the other hand, an optical switch is an element that changes the degree of light wave power transfer by intentionally changing the propagation constant of an optical waveguide, thereby changing the coupling length of the optical waveguide. By controlling the bond length, 100% of light is transferred or 50% of light is transferred.

A structural feature common to such light wave circuit elements using light coupling or branching is that a plurality of cores have a structure very close to each other, that is, two cores are separated at a fine interval. This is a point that it has a structure provided.

Prior to the description of the present invention, a reference (NTT)
R & D Vol. 40, no. 2,1991, 199th
To 204), a conventional method of manufacturing a buried planar lightwave circuit device will be described. Here, FIG.
, Description will be made focusing on the constituent part of the optical coupling part in which two cores are provided at a fine interval.

FIG. 6 is a process chart for explaining a method for manufacturing a quartz-based embedded planar lightwave circuit device. According to this conventional manufacturing method, a flame deposition method (Flame Hydrolysis De), which is an application of an optical fiber manufacturing technique, is used.
(hereinafter, may be abbreviated as the FHD method).

First, glass fine particles (particle diameter: 0.1 μm) obtained by heating and hydrolyzing silicon tetrachloride gas or the like in an oxyhydrogen flame of an oxyhydrogen burner 30 are converted into a glass fine particle layer 24 serving as a lower cladding layer. Further, the glass fine particle layer 22 serving as a core layer is deposited on a substrate 26 such as a silicon wafer. At this time, since the refractive index of the core needs to be higher than that of the lower cladding layer, for example, germanium oxide is contained in the glass fine particle layer 22 serving as the core layer, and The glass fine particle layer 24 to be formed may contain impurities for lowering the refractive index, and may be deposited. By this step, a glass fine particle layer 24 serving as a lower clad layer and a glass fine particle layer 22 serving as a core layer are formed.
(A)).

[0008] Next, two layers of glass fine particles 2
4 and 22 are vitrified transparently. Therefore, the substrate on which the glass fine particle layer is formed is placed in an electric furnace or the like, and is subjected to high temperature (about
(250 ° C.). By this heat treatment, each fine particle layer is changed into the lower clad layer 12 and the core layer 28 (FIG. 6).
(B)).

Subsequently, after forming an etching mask by photolithography to process the core layer 28,
Etching is performed by a reactive ion etching apparatus or the like. This etching forms two spaced, ridge-shaped (projecting lines) cores 48.
(C)).

Using the FHD method again, the glass fine particle layer 32 serving as the upper cladding layer is deposited so as to cover the core 48 and the lower cladding layer 12 (FIG. 6D).

Finally, in order to make the upper clad layer transparent vitrified, it is placed in an electric furnace or the like and heated to a high temperature as described above. By this step, an upper clad layer 34 having the same refractive index as the lower clad layer 12 is formed (FIG. 6E).

Through the above steps, a buried planar lightwave circuit element having a structure using light coupling or branching is formed.

As a general rule of thumb for a single mode optical waveguide, the width of the core is 4 to 10 μm and the height of the core is 4 μm.
〜１０10 μm. Further, the minimum interval between the cores of the Mach-Zehnder interferometer optical switch must be such that light can be coupled or branched, that is, 4 μm or less.

[0014]

In the lightwave circuit element, the propagation constant has an important meaning. It is necessary to produce a propagation constant determined by the wavelength of light passing through the core, the refractive index of the core, the refractive index of the cladding layer, and the like to a value according to the design. On the other hand, as described above, as a property common to lightwave circuit elements in general, it is necessary to have a structure in which a plurality of cores come close to or intersect at minute intervals.

However, according to the above-mentioned FHD method, it has been difficult to satisfy this requirement and produce a lightwave circuit device. The reason is that glass particles do not wrap around as much as gas, especially at corners with steps, so even if glass particles are melted and vitrified, molten glass does not flow into the corners and pores This is because Such a phenomenon of "vacancy" is particularly remarkable when the aspect ratio is about 4 or more.

As described above, if the gap between the cores is not filled with the cladding layer, a propagation constant according to the design cannot be obtained, that is, the coupling length cannot be controlled according to the design. In addition, the confinement of the light in the core becomes worse, causing a propagation loss.

On the other hand, when the glass particles are melted and vitrified at a high temperature, the gap between the cores can be filled. However, when heat treatment is performed at such a high temperature (about 1250 ° C.), impurities doped for adjusting the refractive index of the core or the cladding layer diffuse. As described above, when the impurities in the core or the cladding layer diffuse into the adjacent cladding layer or the core, a steep refractive index distribution cannot be obtained between the core and the cladding layer. This means that a propagation constant according to the design cannot be obtained, that is, the characteristics of the intended lightwave circuit element cannot be obtained. Also, the propagation loss increases.

As described above, according to the FHD method, it is difficult to obtain a lightwave circuit element having characteristics according to the design. Also, the FHD method has a ceramic aspect, such as glasswork at high temperatures, making it more difficult to implement.

Further, a chemical vapor deposition method was also studied for forming the core and the cladding layer. However, as far as the inventor of the present application knows, formation of a good-quality cladding layer on an underlayer having an aspect ratio of 2.5 or more is considered. There was no chemical vapor deposition apparatus to guarantee the film.

Therefore, there has been a demand for a method of manufacturing a light-wave circuit device capable of filling a fine gap (or gap) between cores with a cladding layer.

There has been a demand for a method of manufacturing a lightwave circuit device that does not include processing that adversely affects the characteristics of the lightwave circuit device during the manufacturing process.

[0022]

According to the present invention, there is provided a first step of forming a lower cladding layer on a substrate,
A second step of forming a core layer on the lower cladding layer, a third step of forming cores adjacent to each other at a fine interval by etching the core layer, and forming a ridge shape; A fourth step of forming an upper clad layer so as to cover the clad layer, the fourth step being a film forming step of forming a film covering the core and the lower clad layer, and a film etching step of etching the film The above steps are sequentially repeated to form the obtained laminated film as the above-mentioned upper clad layer.

According to this configuration, since the fourth step of filling the fine gap between the cores with the upper cladding layer includes the film forming step and the film etching step, the fine gap between the cores is formed by the upper cladding layer. Can be satisfied.

That is, in the first film forming step,
The minute gap between the cores may be closed by the growth of the film from the wall surface of the core. The first film formation is terminated when it is expected to be occluded or when occlusion occurs. Next, a film etching step is performed to partially remove the film from the upper surface side in the thickness direction. This etching is anisotropic etching performed in the direction perpendicular to the substrate surface and from the core side. At this time, by performing etching with high anisotropy, the portion exposed to the upper surface of the film is more likely to be etched. For this reason, the film remaining by this etching substantially remains on the wall surface of the core with an equal thickness, or remains so as to become thinner toward the upper side. Therefore, the opening at the top of the recess between the cores is approximately the same as or larger than the opening at the bottom of the recess. Therefore, in the next film forming step, it is easy to fill the minute gap between the cores with the second film. If the minute gaps in the core still cannot be filled with the second layer film and there is a possibility that voids may be formed, the etching and film forming steps are repeated.

According to the steps of this configuration, the minute gaps between the cores can be filled without gaps by the finally laminated film. Then, this laminated film becomes an upper clad layer. Therefore, since the upper cladding layer does not have a hole between the cores, the lightwave circuit device including the upper cladding layer can obtain a propagation constant according to the design. It is possible to control the bond length.

This is particularly suitable for those having an aspect ratio of 3 or more, which has been difficult with conventional forming techniques.

It is more preferable to use a plasma enhanced chemical vapor deposition method in at least one of the first step, the second step, and the film forming step. In this case, since the deposition gas is excited by the plasma, the film is formed by setting the substrate temperature to a temperature as low as the temperature at which the film is formed, that is, a temperature in the range of about 300 to 500 ° C. It becomes possible. That is, since diffusion of impurities doped for adjusting the refractive index of the core or the cladding layer can be reduced, a sharp refractive index can be obtained between the core and the cladding layer.

The film etching step in the fourth step is performed by a dry etching method using an inert gas-containing gas, and the number of times of the repetition of the film forming step and the film etching step is increased. It is preferable to reduce the content of the inert gas in the active gas-containing gas. This content reduction adjustment can be performed by reducing the flow rate of the inert gas into the chamber in which these processes are performed. By decreasing the proportion of the inert gas, the etching is changed from a physical (anisotropic) one by inert gas ions to a chemical (isotropic) one by neutral active species. That is, the etching can be made more isotropic by reducing the content of the inert gas in the inert gas containing gas.

By using an inert gas-containing gas, the surface of the formed film is bombarded with inert gas ions. Therefore, a high-density film having a low density of dangling bonds and a low polycyclic structure can be formed.

It is preferable that the above-mentioned film etching step is performed by a dry etching method. In this dry etching, it is preferable to increase the pressure of the etching gas in the chamber in accordance with the number of repetitions of the film forming step and the film etching step using a widely used etching gas. Increasing the pressure enables more isotropic etching.

Further, the plasma chemical vapor deposition method is used in the film forming step and the dry etching method is used in the film etching step, and the plasma chemical vapor deposition method and the dry etching method are used in the same apparatus and the same apparatus. It is good to do in a chamber. By performing the chemical vapor deposition method using plasma and the dry etching with the same apparatus, the movement between chambers can be omitted, so that the manufacturing is further simplified.

Preferably, one or both of the core layer and the cladding layer (the lower cladding layer and the upper cladding layer) are layers containing silicon oxide. This is because silicon oxide glass has excellent physical and chemical properties, such as low light propagation loss, excellent processing characteristics, and research on processing technology is in progress.

When a layer containing silicon oxide is formed in such a manner, a chemical vapor deposition method is used, and triethoxysilane (TR) is used.
It may be preferable to use a gas containing IES. This is because particles are less likely to be mixed when TRIES is used than when monosilane or the like is used.

Preferably, when the laminated film forming the lower clad layer and the upper clad layer is a film containing silicon oxide, the first step and the film forming step are performed using an organic fluoride gas. It is preferable to perform the plasma chemical vapor deposition method and to perform the film etching step by a dry etching method using the same organic fluoride gas. By using the same gas for a plurality of processes, gas piping, a facility for removing harmful gas and the like can be simplified.

Alternatively, when the laminated film forming the lower clad layer and the upper clad layer is a film containing silicon oxide, the first step and the film forming step may be performed by a plasma chemical method using an organic fluoride gas. It is preferable to carry out the vapor phase growth method and to carry out the third step by a dry etching method using the same organic fluoride gas. By using the same gas for a plurality of processes, gas piping, a facility for removing harmful gas and the like can be simplified.

[0036]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. Note that the drawings merely schematically show the shapes, sizes, and arrangements of the components to the extent that the present invention can be understood, and therefore, the present invention is not limited to the illustrated examples.

FIGS. 1, 2 and 3 to 5 are views for explaining embodiments of each invention of this application. FIG.
2A is a diagram showing a cross-section cut along the line AA in FIG. 2 and FIG. 1B is a cross-section cut along the line BB in FIG. FIG. FIG.
Is a configuration of a Mach-Zehnder interferometer optical switch according to the manufacturing method of the present invention, which is an element having a fine gap between cores arranged in parallel and which is an example of a lightwave circuit element using light coupling or branching. It is a top view showing an example. FIG. 3-
FIG. 5 is a manufacturing process chart for explaining a method of manufacturing a buried planar lightwave circuit element.

In the following description, the Mach-Zehnder interferometer optical switch (hereinafter abbreviated as MZ optical switch) 38 shown in FIG. 2 will be described as an example. This MZ optical switch 3
8 has a core 48 (48a, 48a,
48b), and two optical coupling portions 50 are provided in the middle of the first core 48a and the second core 48b which are the optical waveguides.

As is well known, the two optical couplers 50 are couplers for equally dividing the optical power. For example, from the optical signal input / output end 40 to the first core 48a which is an optical waveguide
It is assumed that an optical signal has been input to. The optical signal is transmitted to the first core 48a and the second core 48 by one optical coupling unit 50.
Branch so that the optical power becomes equal to b. First core 4
In the case where there is no phase difference between the light in the second core 48a and the second core 48b, that is, when the propagation constant is not controlled by the thin film heater 52, the first core 48a to the second core 48b, which are the optical waveguides into which the optical signal is input, are provided. And output from the other optical signal input / output terminal 46. Further, the first core 48a, which is two optical waveguides, is formed by the thin film heater 52.
When the phase difference π is given to the second core 48b and the second core 48b, the optical signal is output from the optical signal input / output terminal 44 without being transferred from the first core 48a which is the optical waveguide to which the optical signal is input. The same applies when an optical signal is input from the optical signal input / output terminal 42 to the second core 48b, which is an optical waveguide.

FIGS. 3 to 5 show a process of forming the upper clad layer, which is a feature of the present invention, along the line AA in FIG. It is the figure shown by the part corresponding to the cut of the cross section shown.

Here, the material for forming the substrate, the core and the cladding layer is not particularly limited as long as it is a material capable of forming the lightwave circuit element intended in the present invention. However, it is sufficient that the thickness of the lower cladding layer between the core and the substrate is so large that the interaction between the core and the substrate can be ignored. However, when the lower clad layer does not have such a thickness, the material of the substrate may be determined in consideration of the absorption coefficient and the refractive index of the substrate with respect to the wavelength of light to be used.

As a material for forming the core or the cladding layer, quartz glass containing silicon oxide, polymethyl methacrylate (hereinafter sometimes abbreviated as PMMA), organic glass containing polyurethane, epoxy, or the like; Other multi-component glass or crystallized glass. Further, the material may be the same or different between the core and the cladding layer.

Preferably, one or both of the core layer and the cladding layer (the lower cladding layer and the upper cladding layer) are formed as a layer containing silicon oxide. Because silicon oxide has excellent processing characteristics,
Finer processing becomes possible, and it is therefore possible to improve the general characteristics of the lightwave circuit device.

It is particularly preferable that at least the core layer is formed as a layer containing silicon oxide. When the core that conducts optical waveguides includes silicon oxide, the core is formed of the same material as the silica-based optical fiber, which has high light transmittance of silicon oxide and widely used, so that the propagation loss is further reduced. It is possible to obtain a small number of lightwave circuit elements having better matching with the optical fiber.

When forming a layer containing silicon oxide,
It is preferable to use a chemical vapor deposition method using a gas containing TRIES (triethoxysilane) as a source gas. T
The gas containing RIES is, for example, O ₂ , O ₃ , N
_A gas containing ₂ O and the like, and also containing an impurity gas for adjusting the refractive index. In TRIES, a reaction proceeds near the surface of an object to be formed such as a substrate as compared with the case of using monosilane or the like, and thus particles such as intermediate products are less likely to be mixed into a layer or a film to be formed. Therefore, the propagation loss is reduced, and a desired propagation constant is easily obtained. When a layer mainly composed of PMMA is formed, a conventionally known method such as a spin coating method or a dipping method may be used.

First, the substrate 26 is cleaned as a pretreatment for manufacturing (FIG. 3A).

As a first step, the lower cladding layer 12 is formed on the substrate 26 (FIG. 3B). Subsequently, the core layer 28 is formed (FIG. 3C). The formation method used in the first step and the second step is not particularly limited. For example, a vapor deposition method, a sputtering method, a chemical vapor deposition method (hereinafter sometimes abbreviated as a CVD method), an FHD method, a sol-gel method, a spin coating method, a dip method, and the like. The lower clad layer 12 and the core layer 28 are formed according to the respective materials.

However, when doping the core layer or the cladding layer with an impurity, it is desirable to adopt a forming method that does not cause undesired diffusion of the impurity. In the formation, impurities or the like that absorb light to be used may be mixed. In such a case, heat treatment or the like may be performed.

Subsequently, by photolithography or the like,
A resist pattern 36 for processing the core layer 28 is formed on the surface of the core layer 28 (FIG. 3D). Even if the resist pattern 36 is formed using only the resist,
A resist pattern 36 having a multilayer resist structure may be formed using a film of amorphous silicon, tungsten silicon, aluminum, chromium, or the like and a resist under the resist.

Next, as a third step, a region of the core layer 28 exposed between the resist patterns 36 is etched to form a core 48 (FIG. 4A). The etching method used in the third step is not particularly limited. Wet etching or dry etching can be used. However, any method may be used as long as it can be etched so that the side wall of the core 48 is perpendicular or almost perpendicular to the upper surface of the substrate 26. Preferably, dry etching using plasma, which can perform anisotropic etching, such as reactive ion etching or reactive ion beam etching, is preferably used. Through this etching step, two parallel, stripe-shaped cores 48 are obtained. Therefore, the opposing side walls of these two adjacent cores 48 are also substantially parallel.

Next, as a film forming step of a fourth step, a film 13 to be a part of the upper clad layer is formed (FIG. 4).
(B)). In this film forming step, a CVD method, a sputtering method, or any other suitable method depending on the material can be used. However, in other cases, the material of the film or the material constituting the film is supplied from outside. Accordingly, any method may be used as long as the film is formed gradually from the surface so that the film thickness is substantially uniform. In addition, the film forming step may be completed when the fine gap between the cores is closed or predicted to be closed by the growth of the film from the wall surface of the core.

When a film is formed by these forming methods, the material of the film or the material constituting the film is supplied from an external source to a certain site depending on the shape of the base or the difference in the speed of film formation at each site on the surface. And an unintended void is formed in that part.
More specifically, as schematically shown in FIG. 4B, the film growth from the corners of the core 48 protruding opposite to each other causes the substrate 2 on the flat side wall of the core 48 to twist.
The growth rate may be higher than that of the film growth from the lower part close to 6. If the formation of the film is continued as it is, the fine gap of the core is first closed by the film in the direction parallel to the substrate from the corner, and the film is formed in a portion below it. As a result, an undesired gap is formed in the gap between the two cores 48.

Therefore, as a fourth film etching step, the film 13 which is to be a part of the upper clad layer is etched to form a film 14 (FIG. 4C). The etching used in the film etching step is not particularly limited. Wet etching or dry etching can be used. However, since it is desirable to perform etching with high anisotropy, it is desirable to use reactive ion etching or reactive ion beam etching.

It is desirable that the anisotropy of etching can be controlled. This is because the anisotropy of etching can be reduced, that is, more isotropic etching can be performed in accordance with the number of repetitions of the film forming step and the film etching step. In this case, the direction of etching of the film becomes more isotropic since the etching direction is superior to the direction perpendicular to the substrate, so that unnecessary etching in the direction perpendicular to the substrate can be reduced. Good etching becomes possible.

In order to reduce the etching anisotropy, the pressure in the chamber may be increased every time the film forming step and the film etching step are repeated.

Alternatively, in order to reduce the anisotropy of the etching by using an inert gas for the etching, the flow rate of the inert gas into the chamber is reduced every time the film forming step and the film etching step are repeated. Good to go. As the inert gas, an argon gas is preferable because it is economical.

When the laminated films (the films 14, 16, 18,... Forming a part of the upper clad layer) constituting the lower clad layer 12 and the upper clad layer 34 are made of a film containing silicon oxide, Is performed by plasma enhanced chemical vapor deposition, and the refractive indices of the lower cladding layer 12 and the upper cladding layer
It is preferable to use an organic fluoride gas as an impurity-adding gas to be added to make the refractive index lower than the refractive index, and to carry out the above-mentioned film etching step by a dry etching method using the same organic fluoride gas. Alternatively, a laminated film constituting the lower clad layer 12 and the upper clad layer 34 (films 14, 16, 18,... Which are part of the upper clad layer)
Is a film containing silicon oxide, the first step and the film forming step are performed by a plasma enhanced chemical vapor deposition method, and the refractive indices of the lower cladding layer 12 and the upper cladding layer 34 at that time are set to the core. It is preferable to use an organic fluoride gas as an impurity-adding gas to be added to make the refractive index lower than 28, and to carry out the third step by a dry etching method using the same organic fluoride gas. .

For example, CF ₄ , C ₂ F ₆ , C ₄ F ₈ , and CHF ₃ can be used as such an organic fluoride gas. Among them, C ₂ F ₆ is preferable. Thus, the same gas is used as a gas for adding impurities.
In addition, by using the etching gas in a plurality of steps, a gas pipe, a harmful gas and other abatement equipment, a leak detector, and the like can be simplified. In addition, it is more preferable to perform the plasma chemical vapor deposition method and the dry etching method with the same apparatus because the process can be simplified.

Next, a second film forming step is performed. A film 15 to be a part of the upper clad layer is formed so as to cover the film 14 to be a part of the upper clad layer formed by performing the first film forming step and the film etching step (FIG. 5).
(A)). In the second film forming step, the same forming method as in the above-described first film forming step can be employed.

If the gap between the cores is not yet filled with the upper cladding layer by the second film forming step, the second film etching step is performed. That is, the film 15 that becomes a part of the upper clad layer is etched to form the film 16 that becomes a part of the upper clad layer. At that time, as described above, it is preferable to lower the anisotropy of the etching. In the second film etching step, a formation method similar to the first film etching step described above, more preferably under the condition of increasing the pressure or decreasing the flow rate of the inert gas is added. good.

If the core gap is not yet filled with the upper cladding layer even by the third film forming step,
Through the above-described film etching step and film forming step, a film to be a part of the upper clad layer is formed. In the figure, the gap between the cores is filled with the upper cladding layer 34 (the film 14, the film 16, and the film 18) by the third film forming process, and the forming process is completed (FIG. 5B).

The film forming step and the film etching step are repeated until the fine gap of the core is filled with the upper clad layer and the upper clad layer from the upper surface of the core has a sufficient thickness. However, as is clear from the gist of the present invention,
The fourth step is completed by the film forming step.

As described above, by repeating the film forming step and the film etching step, the fine gap between the cores can be filled with the upper cladding layer without any gap.

As described above, when at least the film forming step and the film etching step are performed using the plasma CVD method and the dry etching method, respectively, the plasma CVD method and the dry etching method are performed in the same chamber of the same apparatus. Is more preferable. This is because there is no movement between the chambers, so that the manufacturing process can be shortened. For example, a cathode-coupled parallel-plate plasma CVD apparatus (or a cathode-coupled parallel-plate plasma reactive ion etching apparatus) or an electron cyclotron resonance CVD apparatus (reactive ion beam etching apparatus using electron cyclotron resonance) is used. Good. Note that the cathode-coupled type refers to a structure in which the substrate stage and the chamber are electrically insulated so that a plasma sheath is formed on the substrate side, and a self-bias is applied to the substrate stage.

The optical waveguide portion of the MZ optical switch shown in FIG. 2 is formed as described above. After that, if a control part such as a thin film heater made of chromium is formed thereon, an MZ optical switch can be obtained. It is apparent that a directional coupler, a Y-branch element, an arrayed waveguide grating (AWG), and the like can be similarly manufactured.

As can be understood from the above description, it is more possible to cover the fine gap between the juxtaposed cores for coupling or branching light with the cladding layer.
Light confinement in the core is improved. That is, the propagation loss is reduced.

Further, since it is possible to fill the fine gap between the cores arranged in parallel for coupling or branching light with the cladding layer, the propagation of the optical waveguide constituting the lightwave circuit as designed. It becomes easier to obtain a constant. That is,
It becomes easy to obtain characteristics of the lightwave circuit element according to the design such as control of the coupling length.

In addition to the lightwave circuit device using the light coupling or branching, the above-mentioned film forming step and film etching step are used for forming the upper cladding layer of the embedded lightwave circuit having a crossing core structure. It is clear that by repeating the above, at least the coverage of the cladding layer with respect to the core is improved. Thereby, the confinement of light in the core is improved, so that a reduction in propagation loss can be expected.

It is preferable to form the core layer or the clad layer by using a plasma enhanced chemical vapor deposition method, that is, to adopt a method capable of forming at a low temperature. This is because impurities for adjusting the refractive index do not diffuse into the cladding layer or the core layer, respectively, so that a steep refractive index can be obtained between the core and the cladding layer. Become.

[0070]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a specific manufacturing method will be described as an embodiment with reference to FIGS. 1 and 2 and FIGS.

However, the materials used, devices, numerical conditions, and the like described in this embodiment are merely examples within the scope of the present invention. Accordingly, the present invention provides the following materials, devices,
It is not limited to numerical conditions.

Due to the nature of the lightwave circuit element, a core having a low propagation loss is required. Therefore, in the following examples, a lightwave circuit element serving as a core and a cladding layer containing silicon oxide as a main component was formed by a plasma CVD method.

In this embodiment, a cathode-coupled parallel plate type plasma CVD apparatus was used. This device uses TR
IES, oxygen, C ₂ F ₆ and argon, and can introduce into the chamber. In addition, a heater for heating the substrate is provided. The pressure in the chamber can be reduced to 1 × 10 ⁻⁵ Pa by a rotary pump and a turbo molecular pump. Many holes are formed in the surface of the upper electrode, so that gas is supplied to the substrate surface in a shower shape. The gas is controlled to a desired flow rate by a flow controller. The pressure in the chamber can be controlled to a desired constant value by a pressure gauge and a butterfly valve installed in the chamber. 13.56 MHz between the substrate stage and the upper electrode
By applying the high frequency, plasma is generated in the chamber. The formation is performed using such an apparatus. In addition, a silicon wafer is used as the substrate 26.

First, the substrate 26 was washed with a solution of sulfuric acid and hydrogen peroxide mixed at a ratio of 3: 1 at a temperature of 85 ° C. for 5 minutes. Subsequently, the substrate was washed with pure water. Next, the substrate was washed with a hydrogen fluoride solution diluted to 1% with pure water for 20 minutes. The substrate was washed again with pure water (FIG. 3A).

Next, the substrate 26 was set in the chamber of the above-mentioned plasma CVD apparatus. The substrate 26 was heated to 400 ° C. 1Pa inside the chamber by rotary pump
After evacuating to 1 x 1 by a turbo molecular pump
It was evacuated to 0 ^-5 Pa. TRIES is introduced into the chamber from the upper electrode at a flow rate of 12 sccm, oxygen at a flow rate of 400 sccm, and C ₂ F ₆ at a flow rate of 10 sccm. TRIES is vaporized at 80 ° C. in advance. C ₂ F ₆ is doped for the purpose of reducing the refractive index of the cladding layer with respect to the core by about 0.3%. The pressure in the chamber is 30Pa
, And a high frequency of 13.56 MHz was applied between the upper electrode and the substrate stage at a power density of 1.6 W / cm ² . Plasma is generated in the chamber, and the substrate 26
A lower clad layer 12 made of a quartz film having a thickness of about 25 μm and a refractive index of 1.452 was formed thereon (FIG. 3B).

Subsequently, in the same apparatus, the pressure in the chamber was increased to 1 × by a rotary pump and a turbo molecular pump.
Evacuation was performed to 10 ⁻³ Pa. TRIES 12sccm,
Oxygen is introduced into the chamber at a flow rate of 400 sccm. The substrate 26 is heated up to 400.degree. T
RIES was previously vaporized at a temperature of 80 ° C. The pressure in the chamber was maintained at 30 Pa and 13.5
A high frequency of 6 MHz is applied between the upper electrode and the substrate stage.
The voltage was applied at 6 W / cm ² . Plasma is generated in the chamber, and a film thickness of about 8 μm is formed on the lower clad layer 12 in 1 hour and 10 minutes.
m, the core layer 28 having a refractive index of 1.456 was formed.
(C)). Thereafter, the high-frequency discharge and the introduction of gas were stopped, and the pressure in the chamber was evacuated to 1 × 10 ⁻³ Pa using a rotary pump and a turbo molecular pump in this order.
After evacuation, nitrogen was introduced into the chamber to atmospheric pressure, and the substrate 2
6 was taken out.

Next, using a mask on which a pattern of an optical waveguide, which is a light wave circuit element, which couples or branches light, is formed for processing the core layer 28, a light exposure generally used widely is used. A resist pattern 36 is formed on the core layer 28 by a method. Specifically, the core layer 28
The resist is spin-coated to a thickness of about 1
μm was applied. Next, the substrate 26 coated with the resist is
After being kept in a thermostat at 0 ° C. for 10 minutes, the mercury lamp i
The mask and the substrate 26 were set in a reduction projection exposure device using lines. Next, the resist on the core layer 28 was irradiated with i-line at a predetermined exposure dose. After the development, the resist pattern 36 of the optical waveguide of the lightwave circuit element was formed on the core layer 28 by holding the substrate in a constant temperature bath at a temperature of 80 ° C. for 10 minutes (FIG. 3).
(D)).

Next, the substrate 26 on which the resist pattern 36 was formed was placed in a chamber of a reactive ion etching apparatus generally used widely. After the chamber was evacuated to a pressure of 1 Pa by a rotary pump, the gas was evacuated to 1 × 10 ⁻³ Pa using a turbo molecular pump. Next, CHF ₃ was introduced into the chamber at a flow rate of 100 sccm. Hold the pressure in the chamber at 2Pa,
A high frequency of 13.56 MHz was applied at a power density of 1 W / cm ² . The core layer 28 is etched using the resist pattern 36 as a mask. After 40 minutes, the high-frequency discharge and the introduction of CHF ₃ were stopped, and the pressure in the chamber was reduced to 1 × using a rotary pump and a turbo molecular pump.
^Evacuated to 10 ^-3 . After evacuation, the pressure in the chamber was adjusted to atmospheric pressure by introducing nitrogen, and then the base including the substrate 26 was taken out. Next, the resist pattern 36 was peeled off by an oxygen plasma asher method which is generally widely used, followed by washing with a solution of sulfuric acid and hydrogen peroxide mixed at a temperature of 85 ° C. in a ratio of 3: 1. Subsequently, the substrate was washed with pure water. By this step, the core has a line width of 8 μm and a height of 8 μm.
m, a set of cores 48 having a minimum core distance of 1 μm was formed (FIG. 4A).

Next, the core-processed substrate 26 is set in the plasma enhanced chemical vapor deposition apparatus, and a film forming step is performed. The substrate 26 was heated until the temperature reached 400 ° C., and the inside of the chamber was evacuated to 1 × 10 ⁻⁵ by a rotary pump and a turbo molecular pump. TRIES 12sccm,
Oxygen was introduced into the chamber at a flow rate of 400 sccm and C ₂ F ₆ at a flow rate of 10 sccm from the upper electrode. TR
The IES was vaporized at a temperature of 80 ° C. in advance. The pressure in the chamber was maintained at 30 Pa and 13.56
A high frequency of MHz was applied between the upper electrode and the substrate stage at a power density of 1.6 W / cm ² . Plasma was generated in the chamber, and a film 13 which was a part of an upper clad layer having a thickness of about 5 μm and a refractive index of 1.452 was formed on the upper surface of the core 48 in 52 minutes (FIG. 4B). After the formation, the application of the high frequency was stopped.

Next, in order to anisotropically etch the formed film 13 which becomes a part of the upper clad layer so as to leave the film on the side surface of the core 48, the same chamber as that used in the film forming process is used. Subsequently, a film etching process is performed. First, the chamber was evacuated using a rotary pump and a turbo molecular pump until the pressure in the chamber became 1 × 10 ⁻³ . Next, 50 sccm of C ₂ F ₆ and 100 sccc of oxygen
m and argon were introduced at a flow rate of 50 sccm. The pressure in the chamber is maintained at 2 Pa and the frequency is 13.56 MHz.
Is applied at a power density of 1.6 W / cm ² ,
The film 13 to be a part of the upper clad layer was anisotropically etched to form a film 14 to be a part of the upper clad layer. Since the etching rate of the film on the upper surface of the core 48 and the upper surface of the lower clad is three times or more higher than the etching rate of the film on the side surface of the core, the film thickness other than the side surface of the core decreases (FIG. )). After 10 minutes, the application of the high frequency was stopped, the introduction of C ₂ F ₆ , oxygen and argon was stopped, and the pressure in the chamber was evacuated to 1 × 10 ⁻³ using a rotary pump and a turbo molecular pump.

Subsequently, in the same chamber, a film 15 to be a part of the upper clad layer is formed by a second film forming step. This was performed in the same manner as in the first film forming step (FIG. 5A).

In the second film forming step, the core 48
Was not filled by the upper cladding layer,
A second film etching step was performed in the same manner as the first film etching step, and a film 16 to be a part of the upper clad layer was formed.

However, as the film forming step and the film etching step are repeated, it is preferable that the anisotropy of the etching is moderated. Therefore, the flow rate of argon was set to 30 sccm in the second film etching step. By changing the etching from the anisotropic one to the isotropic one, the embedding between the cores is further facilitated.

Next, in the same chamber, a film 18 to be a part of the upper clad layer is formed by a third film forming step. By this third film forming step, the core 48
Was filled with the upper cladding layer (FIG. 5B).

Finally, the pressure in the chamber was evacuated to 1 × 10 ^-3 using a rotary pump and a turbo molecular pump. After evacuation, nitrogen is introduced into the chamber to atmospheric pressure and the substrate 2
6 was taken out.

As described above, the juxtaposed core portions of the buried planar lightwave circuit element for coupling or branching light are formed. Since the gap between the cores having the aspect ratio of 8 is filled with the cladding glass, a propagation constant according to the design can be obtained, so that the coupling length can be controlled as designed. Further, by filling the fine gaps between the cores with the clad glass, the propagation loss is also reduced. Further, since the core layer and the cladding layer are formed at 400 ° C., the impurity doped in the cladding layer does not diffuse, and a sharp refractive index is obtained in the core-cladding layer, and the propagation constant according to the design is obtained. And the propagation loss is reduced. In addition, for forming a core and a clad layer mainly composed of silicon oxide, TR
Since IES is used, the reaction in the vicinity of the surface of the object to be deposited is centered compared to the case where monosilane or the like is used, and particles such as intermediate products are less likely to be mixed into the deposited film. Therefore, the absorption of light by the particles is reduced.

As described above, the gap between the fine cores can be filled with the clad glass, so that the characteristics of the lightwave circuit element are improved.

[0088]

As is apparent from the above description, according to the present invention, it is possible to fill the small gaps between the engaging cores with the cladding glass.

When a plasma enhanced chemical vapor deposition apparatus is used, a clad layer or a core layer can be formed at a low temperature.

As described above, it is possible to obtain a lightwave circuit device having characteristics according to the design. In addition, propagation loss can be reduced.

Further, by performing the film formation and the etching of the film by the same apparatus, or by using the same gas in a plurality of processes, the manufacturing process can be simplified and the cost can be reduced.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a cross-section of the Mach-Zehnder interferometer optical switch taken along the line AA and BB of FIG. 2 for explaining the embodiment;

FIG. 2 is a schematic diagram of a Mach-Zehnder interferometer optical switch, which is an example of a buried planar lightwave circuit element, for describing the embodiment.

FIG. 3 shows a method of manufacturing a buried planar lightwave circuit element.
It is a process sectional view showing a manufacturing process of an optical coupling part.

FIG. 4 is a process cross-sectional view showing a manufacturing process of an optical coupling portion in the method of manufacturing the buried planar lightwave circuit device following FIG. 3;

FIG. 5 is a process cross-sectional view showing a manufacturing process of an optical coupling portion in the method of manufacturing the buried planar lightwave circuit device following FIG. 4;

FIG. 6 is a conventional manufacturing process diagram.

[Explanation of symbols]

12: Lower clad layer 13: Film to be a part of upper clad layer (after the first film forming step) 14: Film to be part of the upper clad layer (after the first film etching step) 15: Upper clad layer 16: A film that becomes a part of the upper cladding layer (after the second film etching step) 18: A film that becomes a part of the upper cladding layer (after the third film forming step) 22: glass fine particle layer to be a core layer 24: glass fine particle layer to be a lower clad layer 26: substrate 28: core layer 30: oxyhydrogen burner 32: glass fine particle layer to be an upper clad layer 34: upper clad Layer 36: resist pattern 38: Mach-Zehnder interferometer optical switch 40, 42, 44, 46: optical signal input / output terminal 48: core 48a: first core 48: b second core 50: optical coupling unit 52: Thin film heater

Claims (5)

[Claims] In manufacturing a buried planar lightwave circuit device having a lower cladding layer, a core and an upper cladding layer, a first step of forming a lower cladding layer on a substrate; A second step of forming a core layer, a third step of forming the cores adjacent to each other at a fine interval in a ridge shape by etching the core layer, and an upper part covering the core and the lower clad layer. A fourth step of forming a clad layer, wherein the fourth step sequentially repeats a film forming step of forming a film covering the core and the lower clad layer, and a film etching step of etching the film. And forming the obtained laminated film as the upper cladding layer. 2. The method of manufacturing a buried planar lightwave circuit device according to claim 1, wherein the film etching step is performed by a dry etching method using an inert gas-containing gas, and the film forming step and the film etching are performed. A method of manufacturing a buried planar lightwave circuit device, wherein the content of the inert gas in the inert gas-containing gas is reduced every time the process is repeated. 3. The method for manufacturing a buried planar lightwave circuit device according to claim 1, wherein the film etching step is performed by a dry etching method using an arbitrary etching gas, and the film forming step and the film etching step are performed. Wherein the pressure of the etching gas is increased each time the process is repeated. 4. The method of manufacturing a buried planar lightwave circuit device according to claim 1, wherein said film forming step uses a plasma chemical vapor deposition method and said film etching step uses a dry etching method. A method for manufacturing a buried planar lightwave circuit device, wherein the chemical vapor deposition method and the dry etching method are performed in the same chamber. 5. The method for manufacturing a buried planar lightwave circuit device according to claim 1, wherein one or both of the core layer and the cladding layer (the lower cladding layer and the upper cladding layer) are oxidized. A method for manufacturing a buried planar lightwave circuit element, wherein the method is formed as a layer containing silicon.