JP2003248142A - Two-dimensional optical member array and two- dimensional waveguide unit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a two-dimensional optical member array in which aligning accuracy of optical members (optical fibers, lenses, etc.), on a board is high and a man-hour of packaging or connecting is reduced, and which has excellent reliability for a long period, a high density, and a large capacity. <P>SOLUTION: The two-dimensional optical member array 10 comprises the optical fibers 1, grooves 21 corresponding to a profile of the fiber 1 on one surface, so that a plurality of optical fiber array units 5 having the board 2 in which the fibers 1 are aligned and fixed are hierarchically laminated on the grooves 21. The plurality of the units 5 are laminated in a layer form in a state where opposite surfaces of the boards 2 constituting the units 5 (e.g. an upper surface 22 of the board 2a and a lower surface 23 of the board 2b in the figure) are not brought into direct contact with each other, and do not directly apply mechanical influence to each other. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a two-dimensional optical member array and a two-dimensional waveguide device. For more details,
A two-dimensional optical member array with high alignment accuracy of optical members (for example, optical fibers, lenses, etc.) on a substrate and excellent long-term reliability, high density and large capacity, and man-hours for packaging and connection. The present invention relates to a two-dimensional waveguide device capable of reducing the number.

[0002]

2. Description of the Related Art In recent years, with the increase in communication data capacity,
There is an increasing demand for optical cross-connect switch technology with excellent processing capacity for communication data capacity. As one of such techniques, an optical switch using a MEMS (micro-electro-mechanical system) that performs fine processing in a semiconductor process such as silicon etching, which is a technique used for micromachining, is used. It is like this. In addition to the above capacity processing capacity, the demand for ensuring reliability has increased,
Surface emitting lasers, which enable high-definition and stable communication, have also come into general use.

Optical member arrays (eg, optical fiber arrays, lens arrays, waveguide (PLC) arrays, semiconductor laser (LD) arrays, photodiode (PD) arrays, etc. used in such optical switches and surface emitting lasers, The "optical member array" will be described below by taking an "optical fiber array" as an example.) Is cut at a cross section perpendicular to the central axis of the aligned optical fibers in order to increase the processing capacity and save space. A so-called two-dimensional optical fiber array (hereinafter sometimes referred to as “2DFA”) having a two-dimensional (hierarchical) structure in cross section is used.

As shown in FIG. 16, in such a conventional two-dimensional optical fiber array 100, for example, the thickness of the V-groove substrates 102 is controlled with high accuracy, and the space between the V-groove substrates 102 and the uppermost portion thereof are controlled. By arranging the optical fibers 101 between the V-groove substrate 102 and the fixing member 103, and stacking them with the front surface of the V-groove substrate 102 and the back surface of the adjacent V-groove substrate 102 in contact with each other, A device for determining the pitch in the thickness direction has been proposed (see, for example, JP-A-56-56).
113114).

On the other hand, as shown in FIG. 17, a waveguide substrate (unit) 205 in which one or more waveguides 201 are patterned near the surface thereof is, for example, a splitter or an AW.
It is used in G, waveguide type modulators and the like. FIG. 17 (a)
17B is a plan view schematically showing a splitter having one channel on the input side and eight channels on the output side, and FIG.
FIG. 18 is a cross-sectional view taken along line XX of FIG.

[0006]

However, the conventional two-dimensional optical fiber array has the following problems.

It is difficult to control the thickness accuracy of a substrate having a V-shaped groove on its surface (V-groove substrate, hereinafter sometimes simply referred to as “substrate”) in submicrons, and when viewed from the entire surface of the substrate, The industrial limit is about ± 1 μm. For example,
When 8 layers of +1 μm substrates are stacked in a layered manner, an error of maximum +7 μm occurs after 2DFA, and the alignment accuracy of the optical fibers must be low.

Since the substrates come into contact with each other,
The thickness of the adhesive layer during this period is substantially zero. This state is not preferable for adhesion with most adhesives, and particularly when the substrate is in contact with the entire surface, it is not always sufficient in terms of long-term reliability.

When the substrate arranged on the upper layer serves as a fixing member including the lid on the substrate arranged on the lower layer, the substrates are laminated in advance and ferrules are formed, and then the optical fiber is inserted. Array (FA) ",
Alternatively, after arranging the optical fibers in the V-shaped groove (V groove) of the lowermost layer, the optical fiber is arranged while aligning the second-layer substrate from the bottom,
Next, after arranging the optical fibers in the V-shaped groove of the second layer from the bottom, the substrates of the third layer from the bottom are aligned and arranged,
This is a system that repeats this in sequence "FA at the same time as stacking"
I had no choice but to adopt. In the former case of “fabrication after lamination,” the hole into which the optical fiber should be inserted must be designed with a minimum clearance with the optical fiber in order to ensure accuracy, so it is extremely small, and the optical fiber can be assembled without disconnection. Was extremely difficult. For example, 2D in which eight layers of substrates in which eight optical fibers are aligned in one layer are laminated in a hierarchical manner
In the case of FA, the number of optical fibers to be inserted is 64. Also, in the latter case of "fabrication at the same time as stacking", the work is complicated and it is difficult to assemble the optical fibers without disconnection, and it is extremely difficult to perform positioning and optical axis alignment such as parallel alignment in each layer at the same time. It was difficult.

In order to solve the above-mentioned problems, between the V-groove substrates that are stacked on each other, the head of the optical fiber is slightly out of the V-groove, and the optical fiber pressing member (the fixing member in the present invention is There is proposed an optical fiber array provided with an accommodating portion for accommodating an optical fiber pressing member in a state where (corresponding) does not contact another V-groove substrate (Patent No. 310).
8241). This optical fiber array is excellent in that workability is improved and array accuracy is improved because the procedure of stacking after FA conversion is performed, but the above-mentioned alignment accuracy and the above-mentioned long-term reliability are related. The problem remains unresolved.

As another approach to solve the above problem, an optical fiber multicore connector having a groove for an optical fiber and a rod for aligning the connector terminals with each other has been proposed (JP-A-55-55). 45051). This optical fiber multi-fiber connector is FA-fabricated at the same time as the procedure, but since the positioning is naturally determined by the V groove and the optical fiber, the workability is improved and the long-term reliability is improved. Although the problem can be solved by setting the depth of the groove, the above-mentioned problem regarding the alignment accuracy remains, and further, it is difficult to match the relative positions in the width direction of the V grooves on both surfaces of the substrate. Was to cause a new problem.

In the optical communication network using the two-dimensional optical fiber array as described above, there are various connection points, the light passing through each connection point is reflected there, and the reflected light is re-input to the original fiber. Then, when returning, the laser and the like are adversely affected (noise is generated, etc.), which is inconvenient. Especially in the case of 2DFA, which is mainly used for MEMS switches, etc., it is often coupled with a lens.
Immediately after A, since it is a space, the reflection becomes large, so the influence of re-input of reflected light was great.

In order to solve the above-mentioned inconvenience, conventionally, an AR coat (generally, a SiO ₂ film and a TiO ₂ film are provided with 1/2) on end faces of a substrate and an optical member on a side where light is emitted.
By laminating with a thickness of 4λ and forming a total thickness of about a wavelength (λ)) to improve the reflection characteristics at the end face, the reflection at the end face was suppressed. However, the coating film formed by AR coating is
There is a problem that it is easily deteriorated by the influence of temperature, humidity, and other environments, and the reflection characteristics are adversely affected. In particular, in recent years, with the development of wavelength division multiplexing (WDM) communication, the amount of light transmitted by one fiber has increased. Opportunities for characteristic changes and deterioration are increasing. Furthermore, since the AR coating on the end face of the fiber array is performed in a state where the fibers are installed, it is difficult to carry out a vacuum process when the AR coat is vapor-deposited, and a large number cannot be collectively processed at a time, resulting in cost reduction. There was also the problem of raising it.

Further, the above-mentioned waveguide substrate has the following problems. That is, when the waveguide substrate and the optical fiber array are connected, it is necessary to optically align (align) one optical fiber array with one waveguide substrate. Since the alignment at the micron level is performed between the waveguide substrate and the optical fiber array substrate, there is a problem that the work is extremely sophisticated and requires a lot of man-hours.

The present invention has been made in view of the above problems, and is a two-dimensional optical member having high alignment accuracy of optical members (eg, optical fibers, lenses, etc.) on a substrate and excellent long-term reliability. It is an object of the present invention to provide an array and a two-dimensional waveguide device that has a high density and a large capacity, and that can reduce the number of steps in packaging and connection.

[0016]

Means for Solving the Problems As a result of intensive research to solve the above problems, the present inventor has made an optical member array which is a set of substrates in which one or more optical members are aligned and fixed on a groove. The optical member array in which the opposing surfaces of the substrates forming the optical member array unit adjacent to each other of the unit are not in direct contact with each other and do not directly exert a mechanical influence on each other By stacking a plurality of units (the same applies to a plurality of waveguide substrate units each having one or more planarly patterned waveguides), the above object can be achieved. The inventors have found out what can be done and have completed the present invention.

That is, the present invention provides the following two-dimensional optical member array and two-dimensional waveguide substrate device.

[1] An optical member, and a substrate having on one surface thereof one or more grooves corresponding to the outer shape of the optical member, on which one or more optical members are aligned and fixed. A two-dimensional optical member array formed by stacking a plurality of optical member array units in a hierarchical manner, each of the substrates constituting the adjacent optical member array units among the plurality of optical member array units. It is characterized in that a plurality of the optical member array units are layered in a layered manner in a state where the facing surfaces do not come into direct contact with each other and do not directly exert a mechanical influence on each other. Two-dimensional optical member array. The "state in which mechanical influences do not directly influence each other" specifically means
"A state where forces, vibrations, etc. are not directly transmitted to each other"
Means The same applies hereinafter.

[2] The two-dimensional optical member array according to [1], wherein the optical member is an optical fiber or a lens.

[3] The apex of the optical members arranged on the substrate forming the optical member array unit and the surface of the substrate facing the optical member array unit facing the apex are in contact with each other to form the optical member array unit. A plurality of the optical member array units are layered in a layered manner in such a manner that the opposing surfaces of the constituent substrates do not come into direct contact with each other and the mechanical influences do not directly affect each other. The two-dimensional optical member array according to the above [1] or [2].

[4] An adhesive layer is interposed between the apexes of the optical members arranged on the substrate forming the optical member array unit and the surface of the substrate facing the optical member array unit facing it. With the above, the apex of the optical member and the surface of the substrate are brought into contact with the adhesive layer, respectively, in a state in which the respective facing surfaces of the substrates constituting the optical member array unit do not come into direct contact with each other, The two-dimensional optical member array according to [1] or [2], wherein a plurality of the optical member array units are layered in a layered manner without directly exerting mechanical influences on each other.

[5] The optical member is provided on the one surface of the substrate forming the optical member array unit of the uppermost layer and between the substrates forming the optical member array unit adjacent to each other. The two-dimensional optical member array according to [1] or [2], further including a fixing member that presses or mounts on one surface side having the groove to align and fix.

[6] With the surface of the fixing member and the surface of the substrate constituting the optical member array unit facing the surface of the fixing member not in direct contact,
The fixing member presses or places the optical member on one surface side of the substrate having the groove to align and fix the optical member without directly exerting mechanical influences on each other. ] The two-dimensional optical member array of statement.

[7] With the optical member in contact with the surface of the fixing member and the side wall forming the groove, the optical member is pressed or placed on the substrate to be aligned and fixed. The two-dimensional optical member array according to the above [5] or [6].

[8] An adhesive layer is further provided between the surface of the fixing member and the surface of the substrate constituting the optical member array unit facing the surface of the fixing member. [5] ~ The two-dimensional optical member array according to any one of [7].

[9] The thickness of the adhesive layer is 2 to 10
The two-dimensional optical member array according to the above [8], which is 0 μm.

[10] The above [1] to [1], wherein a positioning guide is formed at a predetermined position on one surface of the substrate having the groove which constitutes the optical member array unit.
The two-dimensional optical member array according to any one of [9].

[11] The two-dimensional optical member array according to any one of [1] to [10], wherein the groove is a V-shaped groove.

[12] An end face of the optical member forming the optical member array unit, the end face on the light emitting side and the end face on the light incident side, or any one of them is perpendicular to the central axis of the optical member. The two-dimensional optical member array according to any one of [1] to [11], wherein the two-dimensional optical member array has an inclination that forms a predetermined angle (θ) with respect to a surface.

[13] The end surface of the optical member on the side from which light is emitted and the end surface on the side from which light is incident, or any one of them is disposed on a surface perpendicular to the central axis of the optical member. The two-dimensional optical member array according to the above [12].

[14] An end surface of the optical member on the side where light is emitted and an end surface on the side where light is incident, or any one of them is at a predetermined angle with respect to a plane perpendicular to the central axis of the optical member. The two-dimensional optical member array according to the above [12], which is arranged on a surface forming θ).

[15] The end face of the optical member on the side of emitting light and the end face on the side of entering light, or any one of them is perpendicular to the emitted light and the incident light or any of their optical axes. The two-dimensional optical member array according to the above [12], which is arranged on a surface.

[16] A method for measuring the core position of the optical member constituting the two-dimensional optical member array according to any one of [1] to [15], wherein the optical member array unit is hierarchical. When each of the optical member array units has n channels (where the optical members are aligned in m rows and n columns), the cores of each row of the m optical members are stacked in m stages. While measuring the respective positions, the core positions of at least two rows of optical members among the n rows of optical members are respectively measured, and further, one optical member is arbitrarily selected from each row of the at least two rows of optical members. In particular, the distance between the core positions of the specified optical member (specific optical member) is measured, and four squares formed by diagonally connecting the line segments connecting the core positions of the specific optical members. An optical member forming a two-dimensional optical member array, wherein the positional relationship between the core positions of the optical members corresponding to the corners of the matrix is calculated, and the core positions of the entire optical member are calculated. Core position measurement method.

[17] A two-dimensional waveguide device in which a plurality of waveguide substrate units each having one or more planarly patterned waveguides are layered in a layered manner, wherein a plurality of the waveguide substrates are provided. A plurality of the waveguide substrate units are arranged in a state in which the surfaces of the waveguide substrate units that are adjacent to each other among the units are not in direct contact with each other and do not directly exert mechanical influences on each other. A two-dimensional waveguide device comprising a plurality of layers stacked in layers.

[18] The two-dimensional waveguide device according to the above [17], wherein an adhesive layer is provided between the surfaces of the waveguide substrate units adjacent to each other among the plurality of the waveguide substrate units that face each other. .

[19] The thickness of the adhesive layer is 2 to 1
The two-dimensional waveguide device according to the above [17] or [18], which is 00 μm.

[20] The two-dimensional waveguide device according to any one of [17] to [19], wherein a positioning guide is formed at a predetermined position on the surface of the waveguide substrate unit.

[21] Each of the end faces on the light emitting side of the waveguide forming the waveguide substrate unit is provided with an inclination forming a predetermined angle (θ) with respect to a plane perpendicular to the optical axis. The two-dimensional waveguide device according to any one of [17] to [20] above.

[22] An end face on the side of emitting light and an end face on the side of entering light of the waveguides constituting the waveguide substrate unit are perpendicular to the central axis of the waveguide. The two-dimensional waveguide device according to the above [21], which is arranged on a flat surface.

[23] An end face on the side of emitting light and an end face on the side of entering light of the waveguides constituting the waveguide substrate unit are perpendicular to the central axis of the waveguides. The two-dimensional waveguide device according to the above [21], which is arranged on a surface that forms a predetermined angle (θ) with respect to another surface.

[24] The end face on the side of emitting light and the end face on the side of entering light of the waveguides constituting the waveguide substrate unit, or any one of them, respectively, is the emitted light and the incident light or their The two-dimensional waveguide device according to the above [21], which is arranged on a surface perpendicular to any optical axis.

[0042]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, an embodiment of the two-dimensional optical member array and the manufacturing method thereof according to the present invention will be referred to as an "optical member".
As an example, a case of using an "optical fiber" will be specifically described with reference to the drawings.

FIG. 1 is a sectional view schematically showing a two-dimensional optical fiber array which is a first embodiment of a two-dimensional optical member array of the present invention. As shown in FIG. 1, the two-dimensional optical fiber array 10 according to the present embodiment includes an optical fiber 1
And an optical fiber having one or more grooves 21 corresponding to the outer shape of the optical fiber 1 on one surface, and a substrate 2 on which the one or more optical fibers 1 are aligned and fixed. A two-dimensional optical fiber array 10 is formed by stacking a plurality of array units 5 in a hierarchical manner, and the substrates 2 constituting the adjacent optical fiber array units 5 of the plurality of optical fiber array units 5 face each other. In a state where the surfaces are not in direct contact with each other and do not directly exert a mechanical influence on each other (in FIG. 1, for example, the upper surface of the substrate 2a which constitutes the lowermost optical fiber array unit 5a). (The surface having the groove 21) 22 and the lower surface (back surface) 23 of the substrate 2b forming the optical fiber array unit 5b, which is the second layer from the bottom, are not in direct contact with each other, and mechanical influences are exerted on each other. In the absence of the mutually exerted directly), characterized by comprising hierarchically stacking a plurality of optical fiber array units 5.

With such a configuration, the alignment accuracy of the optical fiber on the substrate does not depend on the thickness accuracy of the substrate, so that the alignment accuracy of the optical fiber on the substrate can be improved. That is, unlike the conventional case where the optical fiber array units 5 are stacked by bringing the substrates 2 into contact with each other, the stacking accuracy is not determined by the thickness accuracy of the substrate 2, which is extremely difficult. By avoiding the control of the thickness accuracy of the optical fiber, the alignment accuracy of the optical fiber on the substrate can be easily increased without complication.

In this embodiment, in addition to the fact that the alignment accuracy of the optical fibers does not depend on the thickness accuracy of the substrate, as will be described later, the position is determined by a positioning jig and the optical fibers are By aligning and stacking the array units 5 with each other, the alignment accuracy of the optical fibers on the substrate can be easily increased.

Further, as shown in FIG. 1, the optical fiber array 10 according to the present embodiment has an optical fiber array adjacent to each other on one surface of the substrate 2d constituting the uppermost optical fiber array unit 5d. The optical fiber 1 is provided between the substrates 2 (for example, the substrates 2a and 2b) forming the unit 5 (for example, the optical fiber array units 5a and 5b).
(For example, the substrates 2d, 2a, 2b) are preferably further provided with a fixing member 3 for aligning and fixing by pressing or placing on one surface side having the groove 21.

In this case, the material of the substrate 2 and the fixing member 3 is not particularly limited, but for example, borosilicate glass or the like which is light transmissive can be mentioned as a preferable example.

With this structure, the surfaces of the substrates 2 forming the adjacent optical fiber array units 5 of the plurality of optical fiber array units 5 are not in direct contact with each other, and A state in which mechanical influences do not directly affect each other can be easily realized.

Further, as shown in FIG. 1, the two-dimensional optical fiber array 10 of the present embodiment is a substrate constituting the surface of the fixing member 3 and the optical fiber array unit 5 facing the surface of the fixing member 3. Of the optical fiber array unit 5a, which is the lowest layer, in FIG. Fixing member 3a
Of the two surfaces of the fixing member 3a are not in direct contact with the surfaces 22 and 23 of the substrates 2a and 2b constituting the optical fiber array units 5a and 5b, which face the two surfaces of the fixing member 3a. It is preferable that the fixing member 3 aligns and fixes the optical fiber 1 on one surface side having the groove 21 of the substrate 2 without directly affecting each other). In this case, the fixing member 3a
The lower surface (back surface) of the fixing member 3a does not contact the front surface 22 by contacting the top of the optical fiber 1 on the substrate 2a, as will be described later, and the upper surface (front surface) of the fixing member 3a.
When they come into contact with the adhesive layer 4 described later, they do not come into direct contact with the surface 23 and do not directly exert mechanical influences on each other.

With this structure, it is possible to easily realize a structure in which the alignment accuracy of the optical fibers does not depend on the thickness accuracy of the substrate.

Further, as shown in FIGS. 1 and 2, the two-dimensional optical fiber array 10 according to the present embodiment has the fixing member 3
The optical fiber 1 is pressed or placed on the substrate 2 in a state where the optical fiber 1 is brought into contact with the surface 3s of the above and the side walls 21a and 21b forming the groove 21, and the optical fiber 1 is aligned and fixed. It is preferable. In this manner, the optical fiber 1 has the top portion abutting the surface 3 s of the fixing member 3 with its head protruding from the surface of the substrate 2, and the side wall 21 forming the groove 21.
It abuts on each of a and 21b.

With this configuration, it is possible to easily realize a configuration in which the alignment accuracy of the optical fibers does not depend on the thickness accuracy of the substrate.

Further, as shown in FIG. 1, the two-dimensional optical fiber array 10 of the present embodiment is a substrate constituting the surface of the fixing member 3 and the optical fiber array unit 5 facing the surface of the fixing member 3. Between the other surface (rear surface) of the two surfaces (for example, between the upper surface (front surface) of the fixing member 3a and the other surface (rear surface) 23 of the substrate 2b constituting the optical fiber array unit 5b). ), And an adhesive layer 4 is preferably further provided.

The adhesive layer 4 used in the present embodiment is not particularly limited, but as the substrate 2 and the fixing member 3,
When the borosilicate glass or the like that is light-transmissive is used, for example, an ultraviolet curable adhesive can be given as a suitable example.

The thickness of the adhesive layer 4 depends on the type of adhesive used, but is preferably 2 to 100 μm, and 3 to
20 μm is more preferable. If the thickness is less than 2 μm, the adhesive properties may be insufficient, or if the substrate precision is poor, the optical fiber array units may partially contact each other, resulting in a deterioration in precision. If the thickness exceeds 100 μm, the thermal expansion may be large. Sometimes the effect of cure shrinkage cannot be ignored.

With such a configuration, the FA
Since there will be an adhesive layer of appropriate thickness between them,
It is possible to fully bring out the adhesive properties of the adhesive,
It is possible to ensure good long-term reliability.

It should be noted that Japanese Patent No. 310824 proposes an optical fiber array provided with a storage portion for storing the above-mentioned optical fiber pressing member (corresponding to the fixing member in the present invention).
In the case of Japanese Patent Laid-Open No. 1, it is possible to partially secure the adhesive layer between the optical fiber array units, as in the portion between the optical fiber pressing member and the other V-groove substrate. On a substrate having a three-dimensional shape, various complicated stresses are generated due to curing shrinkage of the adhesive, shrinkage due to heat fluctuation after curing, tension, etc., and there is a risk of cracking or the like in some cases. Further, peeling may occur from the outside of the abutting surface, and long-term reliability may be reduced. On the contrary,
Since the two-dimensional optical fiber array of the present embodiment has a simple bonding structure of flat plates as a bonding structure of the substrate, such a complicated stress does not occur and long-term reliability is high. It will be expensive.

By the way, in the two-dimensional optical member array of the present invention, for example, an optical fiber 30 as shown in FIG.
Those including one or more cylindrical members 302 each having a through hole 302a capable of penetrating and containing 1a are also included. In this case, as the substrate 303, a substrate having at least one surface on which one or more grooves 303a corresponding to the outer shape of the cylindrical member 302 are formed is used. For example, the cylindrical member 302 is temporarily fixed to the groove 303a formed in the substrate 303 with an adhesive, and this is used as a ferrule. Then, the optical fiber 30 is inserted into the through hole 302a of the cylindrical member 302 of the ferrule.
1a is inserted, and finally the substrate 303 and the cylindrical member 30 are inserted.
2. The cylindrical member 302 accommodating the optical fibers 301a is two-dimensionally aligned on the plurality of hierarchically stacked substrates 303 by a method of adhering the fixing members 304 and the fixing member 304 with an adhesive and finally fixing them. Two-dimensional optical member array 3
10 can be configured.

Since the optical fiber is usually made of quartz, it has a property of easily breaking due to concentration of stress when the outer surface is damaged. The two-dimensional optical member array 310 configured as shown in FIG.
Since a is aligned and fixed in a state of being housed in the cylindrical member 302, the outer surface of the core is effectively prevented from being damaged and broken due to contact with the substrate 303 or the like. The reliability can be improved. Also,
The work of aligning and fixing the optical fibers 301a is simple without the need for complicated handling, the work time can be shortened, and the productivity can be improved.
Due to such shortening of time, the chance of breakage due to contact with the substrate 303 or the like can be reduced, and reliability can be improved. Further, when the optical fiber is a polarization maintaining optical fiber, as shown in FIG. 19, the polarization maintaining optical fiber 30 is inserted in the through hole 302a of the cylindrical member 302.
Since 1c is inserted, the adhesive layer 305 around the polarization-maintaining optical fiber 301c becomes substantially uniform, and the stress caused by the adhesive layer 305 on the polarization-maintaining optical fiber 301c becomes substantially the same, so that extra stress is generated. It is preferable in that it does not occur and the ability to hold the polarized wave is not deteriorated. In FIG. 19, reference numeral 306 indicates a stress applying portion, and reference numeral 307 indicates a core.

In the two-dimensional optical member array including the cylindrical member, as shown in FIG. 20B, the through hole 302a of the cylindrical member 302 accommodates the optical fiber 301a from which the outer coating 301b is peeled off. For this reason, the taper portion 302b having a diverging shape toward the opening on the insertion side can be smoothly inserted into the through hole 302a of the cylindrical member 302 of the optical fiber 301a. It is preferable in terms. Also, the taper portion 302b
The opening-side end 302c of the corresponding substrate end 303b
In consideration of the surface roughness of the corresponding fixing member end 304a, the substrate end of the optical fiber 301a is projected to the outer side by 1 μm or more than the substrate end 303b and the fixing member end 304a. It is preferable in order to effectively prevent disconnection due to contact with 303b and the end portion 304a of the fixing member. The protrusion is preferably 5 mm or less.

As a material of the cylindrical member 302 used in the present invention, high-definition processing is possible, and the through hole 30
There is no particular limitation as long as it is a material that is sufficient to prevent the optical fiber 301a housed in 2a from being damaged by contact with the outside, but when fixing to the substrate 303 using an ultraviolet curable adhesive Is preferably light transmissive. For example, borosilicate glass can be mentioned as a suitable example.

The outer diameter of the cylindrical member 302 varies depending on the thickness of the optical fiber 301a and so on, and therefore its optimum value cannot be uniquely determined, but normally φ0.25 to 0.25
2.7 mm can be preferably used. on the other hand,
The inner diameter of the through hole 302a is preferably set to be slightly larger than the outer diameter thereof in order to facilitate the insertion of the optical fiber 301a. Specifically, φ0.126 to 0.13
It is preferably mm.

The substrate in the two-dimensional optical member array including the cylindrical member has one or more grooves 303 a corresponding to the outer shape of the cylindrical member 302, as shown in FIG. The shape is not particularly limited as long as it can be smoothly aligned and securely fixed, and for example, a V-shaped groove or the like that can reliably support the cylindrical member 302 at three points is preferably used. . Groove 30
Regarding the installation interval of 3a, as shown in FIG. 20 (a), the cylindrical members 302 may be arranged so as to have a constant interval therebetween, and as shown in FIG.
The cylindrical members 302 may be densely arranged so as to contact each other.

Further, the substrate in the two-dimensional optical member array including the cylindrical member can have the cylindrical members 302 aligned on both surfaces thereof, for example, as shown in FIG.
It may have grooves 303a having substantially the same shape and installation intervals corresponding to the outer shape of the cylindrical member 302. When such a substrate is used, the two substrates 303 are hierarchically arranged so that the respective grooves 303 a face each other, and the cylindrical member 302 is provided between the two substrates 303.
It is possible to form a two-dimensional optical member array 310 in which the other cylindrical members 302 are aligned and fixed on the substrate 303 located above by sandwiching and aligning.

In the substrate 303 having grooves on both sides as shown in FIG. 22, the substrate of the cylindrical member 302 is formed by processing the grooves 303a on both sides with substantially the same shape and installation intervals with high precision. Since the positioning on 303 can be performed accurately and easily, the upper and lower cylindrical members 302 that are two-dimensionally and hierarchically arranged on it.
Can accurately set the distance between the centers of the opposed and adjacent cylindrical members 302, and the substrate 3 shown in FIG.
The cylindrical member 302 having a higher definition than the case of No. 03.
Can have a positional relationship of. Further, like the two-dimensional optical member array 310 shown in FIG. 23, when the cylindrical members 302 are aligned and fixed also under the lower substrate 303, the cylindrical members 30 arranged in three rows hierarchically.
The distance between the centers of the two opposing and adjacent cylindrical members 302 can be set accurately.

FIG. 3 is a sectional view schematically showing a two-dimensional optical fiber array which is a second embodiment of the two-dimensional optical member array of the present invention. As shown in FIG. 3, it is preferable that the positioning guide 6 is formed at a predetermined position on one surface of the substrate 2 forming the optical fiber array unit 5 having the groove 21. Since the positioning guide 6 shown in FIG. 3 is formed on the same surface as the surface of the substrate 2 on which the groove 21 is formed, the positioning guide 6 can be processed simultaneously with the groove 21 and is efficient.

With this configuration, it is possible to easily determine the arrangement position of the optical fiber array units 5 by using an appropriate positioning jig (not shown) or stack the optical fiber array units 5 by aligning them. Therefore, the alignment accuracy of the optical fiber on the substrate can be improved.

As the shape of the groove used in each of the above-described embodiments, as described above, the optical fiber 1 can be smoothly aligned and reliably fixed in accordance with the outer shape of the optical fiber 1. If there is no particular limitation,
It is preferably a V-shaped groove that can reliably support the optical fiber 1 at three points.

When manufacturing the two-dimensional optical fiber array of the above-described embodiment, first, an optical fiber array unit, which is a set of substrates in which one or more optical fibers are aligned and fixed on the groove, is manufactured. Since this is a normal alignment and fixing of the optical fiber on one surface of the substrate, there is little risk of the optical fiber breaking. In addition, since the optical fiber array units are laminated, there is almost no risk of disconnection because the optical fiber array units are laminated. Further, since the laminating work is independent as the laminating work, the positioning work becomes simple. Currently, two-dimensional optical fiber array (2DFA)
Is also required to have a 1000-core scale (for example, 32 × 32), and the greater the number of cores, the more remarkable the usefulness of being able to assemble without causing this wire breakage. .

FIG. 4 is a cross-sectional view schematically showing a two-dimensional optical fiber array (a mode without the fixing member 3 (see FIG. 1)) which is a third embodiment of the two-dimensional optical member array of the present invention. Is. As shown in FIG. 4, the two-dimensional optical fiber array 10 according to the present embodiment includes an optical fiber array unit 5
The apexes of the optical fibers 1 arranged on the substrate 2 forming (5a, 5b) and the optical fiber array unit 5 facing the apexes.
Between the surface of the substrate 2 (2a, 2b) constituting the
The adhesive layer 4 (4a, 4b) is interposed, and the apex of the optical fiber 1 and the surface of the substrate 2 (2a, 2b) are brought into contact with the adhesive layer (4a, 4b), respectively, and the optical fiber array unit 5 Substrate 2 (2a, 5b) constituting (5a, 5b)
2b) a plurality of optical fiber array units 5 are layered in a layered manner in a state where they do not come into direct contact with each other and do not directly exert a mechanical influence on each other. May be The adhesive layer 4a connects the optical fiber array units 5 to each other,
Fix the adhesive layer 4b on the substrate 2 and the optical fiber array 1
They are connected and fixed to each other. In addition, the substrate 2 (2a, 2
As long as it is possible to realize a state in which the surfaces facing each other in b) do not come into direct contact with each other and the mechanical influences do not directly affect each other, the adhesive layer 4 is not interposed. May be.

FIGS. 5 and 6 are sectional views schematically showing the two-dimensional optical fiber arrays that are the fourth and fifth embodiments of the two-dimensional optical member array of the present invention. As shown in FIGS. 5 and 6, in the two-dimensional optical fiber array 10 of the present embodiment, the end surface S of the optical fiber 1 forming the optical fiber array unit 5 on the light emitting side is the center of the optical fiber. It is provided with an inclination forming a predetermined angle (θ) with respect to a plane V perpendicular to the axis L. Although FIGS. 5 and 6 show the case where only the end surface S on the light emitting side has an inclination, the end surface on the light incident side may have an inclination, and Both the end surface S on the light emitting side and the end surface on the light incident side may be provided with an inclination.

With such a configuration, it is possible to improve the reflection characteristic of the end surface of the optical member (optical fiber) on the side where the light enters and exits the substrate, and to maintain the reflection characteristic for a long period of time. It is possible to prevent the loss of light quantity and the adverse effects on other devices. That is,
Since the reflected light is reflected to the outside of the core of the optical fiber, it does not return to the original fiber after being re-input, and excellent reflection characteristics can be obtained. In addition, a means for improving the reflection characteristics is to directly process the optical fiber,
Since the end face on the light emitting side is inclined, peeling or deterioration does not occur unlike the AR coating film. Further, even with respect to strong light, it only results in the problem of durability of the optical fiber itself, and the provision of the inclination itself does not become a negative factor. Further, since the inclination can be formed on the end face only by obliquely polishing the end face, the cost is excellent.

In this case, as shown in FIG. 5, the end surface S of the optical fiber 1 on the light emitting side and the end surface on the light incident side or one of them is perpendicular to the central axis of the optical fiber 1. The optical fiber 1 may be disposed on the surface V, and as shown in FIG. 6, the end surface S of the optical fiber 1 on the side where light is emitted and the end surface S on the side where light is incident, or those end surfaces. Either one may be arranged on a surface U forming a predetermined angle (θ) with respect to a surface V perpendicular to the central axis of the optical fiber 1. 5 and 6, the inclination is formed at a predetermined angle in the column direction (thickness direction), but the inclination may be formed at an angle in the row direction (width direction).

As described above, 2DFA is often lens-coupled, and in this case, it is necessary to examine the optical system. For example, when light is obliquely incident on the lens,
Although it depends on the characteristics of the lens, the angle deviation (deviation between the optical axis of the light emitted from the optical fiber (the incident light to the flat plate microlens) and the optical axis of the flat plate microlens) (Δθ) is 15 °.
The degree is the allowable limit (Δθ =
0 °). If the angle deviation (Δθ) exceeds 15 °, the tolerance of coupling becomes severe, and a loss actually occurs. Therefore, from the viewpoint of easy handling as an optical system,
The angle deviation (Δθ) is preferably 10 ° or less.

As shown in FIG. 7, in the case of the fourth embodiment shown in FIG. 5, the inclination angle θ of the end face S of the optical fiber 1
When the angle is 8 °, lens coupling is performed by a space system using a general SM fiber made of quartz (refractive index = 1.45) and the flat plate microlens 7, and the end face S of the optical fiber 1 is Since it is arranged on the plane V perpendicular to the central axis L, the flat microlenses 7 can have the same focal length without being tilted. In this case, under the same conditions as before,
The angle deviation (Δθ) is as small as 3.6 °, and this angle facilitates coupling with high efficiency.

Here, the angle deviation (Δθ) can be calculated by the following equation (1).

[0077]

## EQU1 ## Δθ = −sin ⁻¹ (1.45 × sin θ) + θ (1)

When θ is 8 °, the angle deviation (Δθ) between the optical axis P of the light emitted from the optical fiber 1 and the optical axis Q of the flat plate microlens is 3.6 ° from the above formula (1). It is calculated.

Further, as shown in FIG. 8, in the case of the fifth embodiment shown in FIG. 6, similarly, in the case of θ = 8 °, a general silica SM fiber (refractive index = 1.45) is used. ) Is used to perform lens coupling in a spatial system, the angle deviation (Δθ) between the optical axis P of the emitted light from the optical fiber 1 and the optical axis Q of the flat plate microlens is 11.6 ° from the same calculation. Is calculated. In this case, in order to make the focal lengths equal, the flat plate microlens 7 is coupled in parallel with the end surface S of the optical fiber 1, that is, inclined by 8 °.

End face S of optical fiber 1 in 2DFA
In order to eliminate the reflection return (re-input), the angle θ of inclination at the end face of a general quartz fiber may be 8 ° or more. Further, in order to reduce the angle deviation (Δθ) to 15 ° or less, when a general quartz fiber is used, the inclination angle θ of the end face of the fiber is set to 28 ° or less in the case of the fourth embodiment shown in FIG. It suffices to do so, and in the case of the fifth embodiment shown in FIG. 6, it may be set to 15 ° or less. Furthermore, in order to reduce the angle deviation (Δθ) to 10 ° or less, the tilt angle θ may be set to 20 ° or less in the case of the fourth embodiment shown in FIG.

FIG. 9 is a sectional view schematically showing a two-dimensional optical fiber array which is a sixth embodiment of the two-dimensional optical member array of the present invention. As shown in FIG. 9, in the two-dimensional optical fiber array 10 according to the sixth embodiment, the end face S of the optical fiber 1 on the side where light is emitted and the end face S on the side where light is incident, or one of them. Each of them is arranged on a surface W perpendicular to the optical axis P of the incoming and outgoing light. In addition,
FIG. 9 shows a case where the end surface S on the light emitting side is disposed on the surface W perpendicular to the optical axis P of the emitted light.

With this configuration, for example, light is incident at a right angle on the flat plate microlens 7, so that the optical system is simple and easy to handle, and the optical length is not uneven. It can be resolved.

In any of the above-described embodiments, the lens array in the flat microlens 7 is
Usually, they are arranged at equal pitches, and if the layers are not uniform in the angle direction and angle, the incoming and outgoing light will not have the same pitch.
The angle direction and the angle are preferably unified in each layer.

On the other hand, the cross-connect switch of recent years is required to have a very high speed, and if the switching optical path lengths are not uniform, there is a difference in switching time, which poses a problem. As described above, it is important to make the switching optical path lengths uniform, and in the conventional configuration, the switching optical path lengths become inconsistent, but it is possible to make them uniform in the present embodiment. It can be said that it is more advantageous in an application in which the optical path lengths (optical path lengths) need to be the same as in the case shown in FIG.

FIG. 10 schematically shows the configuration of a simple in-line switch. Since the switching element depends on the configuration and method, the optical path length within the element is ignored here. The optical path length in the configuration shown in FIG. 10 is from the input side fiber array 24 to the in-line type optical switch 26.
Is L1, and from the in-line optical switch 26 to the output side fiber array 25 are L2 and L3, L1 + L2 is for switching, and L1 + L3 is for switching. In the case of the configurations shown in FIGS. 10A and 10B, L2 and L3 are different from each other, so that the switching (or) causes the optical path lengths in the element to be nonuniform. On the other hand, in the case of the configuration shown in FIG. 10C, since L2 and L3 are the same, the optical path lengths in the elements do not become uneven due to switching. Therefore, it is necessary to make the optical path lengths the same length. Can be said to be more advantageous in certain applications.

For example, a MEMS switch has a large element pitch because of low loss and low crosstalk.
Generally, it is about 3 mm. 2 DFA is 10 × 1
When the inclination angle of the end face of the optical fiber is 0 ° on a scale of 0, the pitch between the first layer and the tenth layer is 3 × 9 = 27 mm, so | L−l | = 27 × tan8 = 3.8 mm, This difference in optical path length is at a level that cannot be ignored. The difference in optical path length cannot be uniquely determined because it depends on the inclination angle of the end face of the optical fiber, the fiber pitch, etc., but the difference in optical path length (| L-1 |) is 1 mm or more. In terms of form, it can be said that this embodiment is more preferable.

Hereinafter, a method for manufacturing the two-dimensional optical member array (optical fiber array) of the present invention will be described with reference to FIGS. 1 to 3.

As a method of manufacturing the two-dimensional optical fiber array of the present invention, for example, one or more optical fibers are provided on the substrate 2 having one or more grooves 21 corresponding to the outer shape of the optical fiber 1 on one surface. 1 is aligned and fixed to form an optical fiber array unit 5, and a plurality of formed optical fiber array units 5 are laminated in a hierarchical manner. Among the fiber array units 5, in a state where the surfaces of the substrates 2 forming the optical fiber array units 5 adjacent to each other are not in direct contact with each other and in which the mechanical influences do not directly affect each other. (For example, in FIG. 1, for example, the upper surface (the surface having the groove 21) 22 of the substrate 2a forming the lowermost optical fiber array unit 5a and the optical fiber array of the second layer from the bottom) In a state where the lower surface (back surface) 23 of the substrate 2b forming the unit 5b does not come into direct contact with each other and the mechanical influences do not directly affect each other (the same applies to other substrates)). , A plurality of the optical fiber array units 5 are laminated in a hierarchical manner.

In the case of manufacturing a two-dimensional optical fiber array with such a structure, as described above, first, as described above, an optical fiber which is a set of substrates in which one or more fibers are aligned and fixed on the groove. An array unit is prepared, but since this is a normal alignment and fixing of optical fibers on one surface of a substrate, there is little risk of disconnection of the optical fibers. In addition, since the optical fiber array units are laminated, there is almost no risk of disconnection because the optical fiber array units are laminated. Further, since the laminating work is independent as the laminating work, the positioning work is simplified. At present, a two-dimensional optical fiber array (2DFA) is also required to have a scale of 1000 cores (for example, 32 × 32), and the larger the number of cores, the more assembly is possible without causing this disconnection. The usefulness of the thing becomes remarkable.

In this case, when one or more optical fibers 1 are aligned and fixed to form the optical fiber array unit 5, or after the formation, the uppermost optical fiber array unit 5d is formed on one surface of the substrate 2d. Of the optical fiber array units 5 adjacent to each other (for example, the optical fiber array units 5a and 5b) (for example, the substrates 2a and 2).
b) between the optical fiber 1 and the substrate 2 (eg 2d, 2
a, 2b) a plurality of fixing members 3 for aligning and fixing by pressing or placing on one surface side having the groove 21 are provided, and a plurality of optical fiber array units 5 in which the fixing members 3 are arranged are arranged. It is preferable to stack them in a hierarchical manner.

In addition, the surface of the fixing member 3 and the surface of the substrate 2 constituting the optical fiber array unit 5 facing the surface of the fixing member 3 do not come into direct contact with each other, and a mechanical influence is exerted. In a state where they do not directly influence each other (in FIG. 1, for example, the two surfaces of the fixing member 3a arranged on the lowermost optical fiber array unit 5a face the two surfaces of the fixing member 3a, respectively). Substrates 2a, 2b constituting the optical fiber array units 5a, 5b
The surfaces 22 and 23 of the
And the fixing member 3 on one surface of the substrate 2 and the substrate 2 without directly exerting mechanical influences on each other.
It is preferable to dispose between the two.

Further, while the optical fiber 1 is in contact with the surface of the fixing member 3 and the side walls 21 a and 21 b forming the groove 21, the optical fiber 1 is pressed or placed on the substrate 2,
It is preferable to align and fix.

Further, when a plurality of optical fiber array units 5 provided with the fixing member 3 are layered in layers, the fixing member 3
Between the surface of the fixing member 3 and the other surface (back surface) of the substrate 2 constituting the optical fiber array unit 5 facing the surface of the fixing member 3 (for example, the upper surface (front surface) of the fixing member 3a). It is preferable to form an adhesive layer between the other surface (back surface) 23 of the substrate 2b constituting the optical fiber array unit 5b.

In addition, the thickness of the adhesive layer 4 is set to 2 to 100.
It is preferably formed to have a thickness of μm.

Further, as the substrate 2 constituting the optical fiber array unit 5, it is preferable to use one in which the positioning guide 6 is formed at a predetermined position on one surface having the groove 21.

Further, it is preferable to form the groove 21 in the shape of a V-shaped groove.

As shown in FIG. 4, the fixing member 3
When manufacturing the optical fiber array 10 that does not include (see FIG. 1), the vertices of the optical fibers arranged on the substrate forming the optical fiber array unit and the substrate forming the optical fiber array unit facing the apex Contact the surface,
A plurality of optical fiber array units are hierarchically arranged in a state in which the respective surfaces of the substrates constituting the optical fiber array unit do not directly contact each other and do not directly exert a mechanical influence on each other. May be stacked,
In addition, an adhesive layer is provided between the apex of the optical fibers arranged on the substrate forming the optical fiber array unit and the surface of the substrate forming the optical fiber array unit facing the apex of the optical fibers. The apex and the surface of the substrate are brought into contact with the adhesive layer, respectively, and the opposing surfaces of the substrates constituting the optical fiber array unit do not come into direct contact with each other, and exert a mechanical influence directly on each other. In such a state, a plurality of optical fiber array units may be layered in layers.

In this case, at the stage of manufacturing each of the optical fiber array units, the temporary fixing member (not shown) is used to bring the optical fibers into contact with the V-grooves to assemble the optical fiber array units. , Remove this temporary fixing member,
This can be dealt with by hierarchically stacking the optical fiber array units. In this case, the temporary fixing member can be easily removed by changing the material of the temporary fixing member to a fluororesin such as polytetrafluoroethylene or by coating a release agent on the substrate.

By thus adopting the configuration in which the fixing member is not provided, the thickness of the entire optical fiber array can be easily reduced. That is, normally, the pitch in the thickness direction (stacking direction) depends on the thickness of each optical fiber array unit, and cannot be made smaller than that. In order to realize a narrow pitch, it is necessary to make each optical fiber array unit itself thin, but there is a limit in terms of strength and the like. When each optical fiber array unit is composed of a substrate and a fixing member, the thickness limit is the sum of the substrate limit and the fixing member limit. However, by omitting the fixing member, the thickness limit of each fiber array unit becomes the limit of the substrate, and the pitch in the thickness direction (stacking direction) can be narrowed. Specifically, the limit of the substrate is about 0.5 mm, and the limit of the fixing member is 0.4 mm.
Since the thickness is about mm, the thickness limit when each fiber array unit is composed of the substrate and the fixing member is about 0.9 mm, but it can be about 0.5 mm by omitting the fixing member.

Another advantage of omitting the fixing member (each fiber array unit is composed of the substrate and the optical fiber) is to avoid adverse effects due to the magnitude of the thermal expansion coefficient (α) of the adhesive layer. You can list what you can do. That is, when the fixing member is used, it is necessary to fix the substrate and the fixing member with the adhesive layer. Empirically, the thickness of the adhesive layer is preferably about 30 μm. Since the layer thickness is about 10 μm, the total layer thickness is as large as about 40 μm, and the effect of the thermal expansion coefficient (α) of the adhesive layer cannot be ignored. Specifically, the thermal expansion coefficient (α) of the adhesive layer used in the present embodiment is about 10 × 10 ⁻⁶ , and the substrate is made of borosilicate glass (trade name, manufactured by Corning Incorporated). : Pyrex) and the pitch is 1.5 mm (in the case of Example 1 described later), the coefficient of thermal expansion of the entire laminated two-dimensional fiber array including an adhesive layer between the substrate and the fixing member ( In the case of α), the width direction is 33 × 10 ⁻⁷ , whereas the thickness direction is as large as 58 × 10 ⁻⁷ , which causes a difference. Since the MEMS optical switch or the like is formed on the surface of silicon (Si) or the like, the direction dependence of the thermal expansion should not be originally present, and thus the direction dependence of the thermal expansion may be a problem. .

In order to solve this problem, the adhesive layer may be thinned, but by omitting the fixing member, the adhesive layer can be thinned without any special measures. . One reason for setting the adhesive layer between each fiber array unit to about 10 μm is that the thickness variation of the substrate and the fixing member is about ± 3 μm when a general manufacturing method is used. This is because it is necessary to consider about ± 6 μm for each fiber array unit. That is, the adhesive layer between each fiber array unit is 10
The thickness of about μm means about 10 ± 6 μm, and the thickness of about 4 μm at the thinnest. When omitting the fixing member, it is only necessary to consider the variation of about ± 3 μm of the substrate, so it is 4 μm in the thinnest case.
In order to secure m, it may be about 7 ± 3 μm,
The thickness of the adhesive layer should be about 7 μm.
That is, the thickness of the adhesive layer per layer in this case is 7 μm.
It will be about. In this case, the coefficient of thermal expansion in the thickness direction (α)
Is 37 × 10 ⁻⁷ , which is a level at which the direction dependence of thermal expansion can be ignored.

As shown in FIGS. 5 and 6, when manufacturing the two-dimensional optical fiber array of the present invention, the end face S on the light emitting side of the optical fiber 1 constituting the optical fiber array unit 5 and the light are An end face on the incident side or any one of them may be provided with an inclination forming a predetermined angle (θ) with respect to a plane V perpendicular to the central axis L of the optical fiber 1. In this case, as shown in FIG. 7, the end surface S of the optical fiber 1 on the side where light is emitted and the end surface on the side where light is incident, or any one of them, is a surface V perpendicular to the central axis L of the optical fiber. Alternatively, as shown in FIG. 8, the end surface S of the optical fiber 1 on the side where light is emitted and the end surface on the side where light is incident, or any one of them, is respectively disposed on the central axis of the optical fiber. It may be provided on the surface U forming a predetermined angle (θ) with respect to the surface V perpendicular to L. Further, as shown in FIG. 9, the end surface S of the optical fiber 1 on the side where light is emitted and the end surface on the side where light is incident, or any one of them, is used as the emitted light and the incident light or any one of them. It may be provided on the surface W perpendicular to the axis P.

With such a configuration, the reflection characteristics of the end surface of the optical member on the side where the light enters and exits the substrate are excellent, and the reflection characteristics can be maintained for a long period of time, and the loss of the light amount and the A two-dimensional optical member array capable of preventing adverse effects on other devices can be efficiently manufactured at low cost.

The end surface S of the optical fiber on the side where light enters and exits
As the formation of the inclination to, for example, the following can be mentioned. That is, in each fiber array unit, after assembling the fibers and adhering and fixing them, the end surface S is polished by using a lap polisher or the like as in the case of a normal one-dimensional fiber array. In this case, the end surface S is polished by inclining the surface plate of the lapping machine at a desired angle.
In this way, a desired angle can be formed on the end surface S.

[0105] Hereinafter, in the method for producing a two-dimensional optical fiber array of the present invention, the two-dimensional optical fiber array unit fabrication (2DFA conversion) after the fabrication of the two-dimensional optical fiber array unit will be specifically described.

As the first two-dimensional optical fiber array, while actively adjusting the optical fiber array unit,
Mention may be made of laminating and fixing to each other. For example, white light is incident from the end surface on the side of incidence and emission on the side opposite to the end surface on the side of incidence and emission of the two-dimensional optical fiber array unit,
While observing the emitted light with a CCD camera, the position of the optical fiber of each optical fiber array unit is grasped, the relative position of the optical fiber array unit of each layer is adjusted, and the optical fiber array units are laminated with an adhesive or the like, Fixing can be mentioned. Since a large-scale apparatus is required for simultaneously stacking and fixing a plurality of optical fiber array units, it is preferable to sequentially carry out layer by layer.

Further, when the adhesive is cured, it is preferable to use a device capable of reliably gripping the optical fiber array unit in order to prevent the determined position from being displaced due to curing and contraction.

For example, a method using the guide pin jig shown in FIG. 11 can be mentioned as a suitable example. Figure 11
First, as shown in FIG.
a is inserted between the first guide pin 6a and the second guide pin 6b provided on the vertical beam jigs on the two sides of the guide pin jig 11, respectively. Next, in order to secure the contact between the guide pin and the guide groove provided on the V-groove substrate of the FA, a load G1 is applied by pulling the lowermost optical fiber array unit 5a from below. Next, the optical fiber array unit 5b of the second layer from the bottom is attached to the second guide pin 6b and the third guide pin 6c.
, And a load G2 is applied so as to press the optical fiber array unit 5b from above in order to secure the contact between the guide pin and the guide groove of the FA V-groove substrate. In this state, an ultraviolet curable adhesive is poured into the gap between the optical fiber array unit 5a and the optical fiber array unit 5b, and ultraviolet rays are irradiated to cure the adhesive. At this time,
Since the guide pin is fixed when the adhesive flows into the guide groove portion, the adhesive is allowed to flow only between the V-groove substrate of the optical fiber array unit 5a and the upper lid substrate of the optical fiber array unit 5b. As with the optical fiber array unit 5b, the optical fiber array unit 5 on the third layer from the bottom is bonded and fixed in a state in which the optical fiber array unit 5 is inserted between the guide pins and a load G2 is applied from above in the same manner as the optical fiber array unit 5b. Is repeated, and for example, the eighth layer is laminated in the same manner.

Position adjustment is performed in the X-axis and Y-axis directions, where the optical axis is the Z-axis, the stacking direction is the Y-axis, and the adjacent arrangement direction of optical fibers perpendicular to the Z-axis and the Y-axis is the X-axis. It is preferable that the center of the beam is grasped and adjusted by image recognition, and the optical axis parallelism θy of each layer is grasped and adjusted by grasping the distance in the Z-axis direction by an autofocus function or a method of searching the beam waist. . The optical axis parallelism θ
Regarding x, it is preferable to observe the optical fiber array units from the side surface and adjust so that the bottom surfaces of the V-groove substrates of the upper and lower optical fiber array units are parallel and have a desired spacing.
Since the bottom surface of the V-groove substrate and the V-groove are parallel, θx, θ
It is also possible to adjust z. This method requires a two-dimensional optical fiber array (2DFA)
Can be reliably realized.

As the second two-dimensional optical fiber array, as shown in FIG. 3, a guide groove (positioning guide) 6 is provided in the optical fiber array unit 5, and positioning is performed by a positioning guide pin that matches this. A method can be mentioned. In this case, a large-scale positioning device is unnecessary, and only a highly accurate guide pin jig is required. Further, since the optical fiber V-groove 21 and the guide V-groove 6 can be formed on the surface of one V-groove substrate 2, the optical fiber V-groove 21 and the guide V-groove 6 can be arranged only in position. It is also possible to secure a very high degree of parallelism. That is, with this method, the X and Y positions and θx, θy, and θz can be adjusted at the same time, and the workability is extremely high. Here, the guide groove used for positioning may be used as a polishing reference after lamination, or may be used for coupling with a mating optical component as a two-dimensional optical fiber array (2DFA). For example, a user of another existing optical component can use this 2DFA guide groove as a reference parallel to the optical axis when connecting the other existing optical component and the new 2DFA. If it is not necessary to use it, the guide groove may be cut off for downsizing.

One embodiment of the method for measuring the core position of an optical member that constitutes the two-dimensional optical member array of the present invention will be described below with reference to FIG. The present embodiment is a method for measuring the core position of the optical fiber 1 constituting the above-mentioned two-dimensional optical fiber array 10, in which the optical fiber array unit 5 is hierarchically m stages (4 stages in FIG. 1).
When the optical fiber array units 5 are stacked and each have n channels (8 in FIG. 1), the optical fibers 1 are arranged in m rows (4 rows) and n columns (8 columns). Case), the core position of each row of the optical fibers 1 of m rows (4 rows) is measured, and the core position of at least two rows of the optical fibers 1 of n rows (8 rows) is measured. Further, one optical fiber 1 is arbitrarily specified from each row of at least two rows of optical fibers 1, and the specified optical fiber 1 (in FIG. 1,
Shown as a specific optical fiber 1a. ), The distance D between the core positions is measured, and the core positions of the optical fiber 1 corresponding to the four corners of the quadrangle formed by diagonally connecting the core positions of the specific optical fiber 1a It is characterized in that the positional relationship of the matrix is calculated and the core position of the entire optical fiber 1 is calculated.

Specifically, in FIG. 1, the optical fiber array unit 5 is arranged in rows 1 to 4 in order from the bottom row, and columns A to H are arranged in order from the left column, and the core positions ( 8 columns (8 channels) each are measured, and core positions of columns A and H (4 rows (4 channels) each)
To measure. Furthermore, for example, the optical fibers 1 corresponding to the matrix (1, A) and the matrix (4, H) are set to the specific optical fiber 1
The diagonal distance D between the two points is measured as a. This diagonal line (1, A)-(4, H), and the straight line (1,
A)-(1, H), and the straight line (1, H)-(4,
H) consisting of three sides: Δ (1, A) (4, H) (1, H)
From this, the relative position between (4, H) on the 5a and 5d rows can be grasped. Similarly, the diagonal line (4, A)-(1, H)
And the distance of (1, A)-(4, A) in row A are measured in advance, and 5a and 5 are obtained from Δ (1, A) (4, A) (1, H).
The relative position with respect to (4, A) on the d-th row can be grasped. The positional relationship between row 5a and row 5d thus identified is 5.
The core position matrix can be formed by combining the position data of each row with the positional relationship between the row a and the columns A and H. In this case, the core positions may be calculated by measuring the core positions in all columns and the diagonal distance measurement between the specific optical fibers specified in arbitrary two rows.

Further, the columns to be measured are not limited to the above-mentioned columns A and H, but if the distance between them is larger, the influence of each measurement error on the whole can be suppressed to a small level. It is preferable to measure the rows at both ends. Also, the specific optical fiber 1a is not limited to the matrix (1, A) and the matrix (4, H), but the specific optical fiber 1a is also specified with a greater distance. Since the influence can be suppressed to a small level, it is preferable to select those located at both corners.

As an apparatus used for carrying out the method for measuring the core position of the optical member constituting the two-dimensional optical member array of the present invention, an image processing or length measurement may be performed in the measuring system to
There is no particular limitation as long as it is possible to measure and calculate the coordinates of the core position of the optical members that form the two-dimensional optical member array, and it is not necessary to use a dedicated measuring device.

FIG. 12 schematically shows one embodiment of the two-dimensional waveguide device of the present invention, FIG. 12 (a) is a plan view, and FIG. 12 (b) is that of FIG. 12 (a). It is sectional drawing in the YY line. As shown in FIG. 12, the two-dimensional waveguide device 200 according to the present embodiment has a structure in which a plurality of waveguide substrate units 205 each including one or more planarly patterned waveguides 201 are layered in a hierarchical manner. Of the plurality of waveguide substrate units 205, the mutually opposing surfaces of the mutually adjacent waveguide substrate units 5 do not come into direct contact with each other, and the mechanical influences are exerted on each other. It is characterized in that a plurality of the waveguide substrate units 5 are layered in a hierarchical manner without directly affecting each other.

With such a configuration, it is possible to achieve high density and large capacity, and to reduce man-hours for packaging and connection.

In this case, the plurality of waveguide substrate units 205
It is preferable that the adhesive layer 204 is provided between the mutually facing surfaces of the waveguide substrate units 205 adjacent to each other. As the adhesive layer 204, the same one as that used in the above-mentioned two-dimensional optical fiber array can be used.

The thickness of the adhesive layer 204 is 2 to 100 μm as in the case of the above-mentioned two-dimensional optical fiber array.
It is preferably m.

FIG. 13A shows the ZZ of FIG. 12A.
FIG. 13B is a cross-sectional view (one aspect) taken along the line of FIG.
FIG. 12 is a cross-sectional view (another aspect) taken along line ZZ in FIG. As shown in FIG. 13, the two-dimensional waveguide device 200 of the present embodiment emits light from the waveguide 201 that constitutes the waveguide substrate unit 205, as in the case of the above-mentioned two-dimensional optical fiber array. The end surface S ′ on one side and the end surface on the side on which light is incident, or any one of them has an inclination forming a predetermined angle (θ) with respect to a surface V ′ perpendicular to the central axis L ′ of the waveguide. Is.

In this case, the end surface S ′ of the waveguide 201 constituting the waveguide substrate unit 205 on the light emitting side and / or the end surface of the light incident side is the central axis L ′ of the waveguide. May be disposed on a plane V'perpendicular to the plane V ', and may be a plane V'perpendicular to the central axis L'of the waveguide.
It may be arranged on a surface U ′ forming a predetermined angle (θ) with respect to. Further, as in the case of the two-dimensional optical fiber array shown in FIG. 9, an end face S ′ on the light emitting side and an end face on the light incident side of the waveguide constituting the waveguide substrate unit, or one of them However, they may be arranged on the surface perpendicular to the emitted light and the incident light or any one of their optical axes.

As a method of manufacturing the two-dimensional waveguide device of the present invention, for example, a two-dimensional method in which a plurality of waveguide substrate units 205 each including one or more planarly patterned waveguides 201 are stacked in a hierarchical manner A method of manufacturing a waveguide device, wherein a plurality of waveguide substrate units 205, which are adjacent to each other, are not in direct contact with each other, and the mechanical influences of the waveguide substrate units 205 are not directly contacted with each other. Waveguide substrate unit 2 without directly affecting each other
One of them is characterized by stacking a plurality of layers 05 in a hierarchical manner (see FIG. 12).

In this case, the plurality of waveguide substrate units 205
Among them, the adhesive layer 204 may be formed between the surfaces of the waveguide substrate units 205 adjacent to each other, which are opposed to each other.

The thickness of the adhesive layer 204 is 2 to 100 μm.
It is preferable to form m.

Further, each of the end faces S ′ of the waveguide 201 constituting the waveguide substrate unit 205 on the light input / output side has a predetermined angle with respect to the face V ′ perpendicular to the central axis L ′ of the waveguide. An inclination that forms (θ) may be provided (see FIG. 13).

In this case, the end surface S ′ of the waveguide 201 constituting the waveguide substrate unit 205 on the light emitting side and the end surface on the light incident side, or any one of them is defined as the central axis L ′ of the waveguide. May be disposed on a plane V ′ perpendicular to
An end surface S ′ of the waveguide 201 constituting the waveguide substrate unit 205 on the side of emitting light and an end surface on the side of entering light, or any one of them, is a plane V ′ perpendicular to the central axis L ′ of the waveguide.
It may be arranged on the surface U ′ forming a predetermined angle (θ) with respect to (see FIG. 13). Further, similarly to the case of the two-dimensional optical fiber array shown in FIG. 9, the end face on the side of emitting light and the end face on the side of entering light of the waveguide forming the waveguide substrate unit, or one of them, They may be arranged on the surface perpendicular to the emitted light and the incident light or any one of their optical axes.

As a method of stacking the waveguide substrate units 205 to make it two-dimensional, the same method as in the case of the above-mentioned two-dimensional optical fiber array can be used. In this case, for example, a positioning guide may be formed at a predetermined position on the surface of the waveguide substrate unit. FIG. 14 shows 4
The case where the waveguides 201 for the chips are formed on one wafer in a state where the center positions are aligned and the positioning guides 206 are formed on both sides of the waveguides 201 are shown. The waveguide 201 can be formed by using, for example, a photolithography technique. The waveguide substrate to be laminated is cut out and secured from the one wafer thus formed. By using such a method, the positions of the positioning guide 206 with respect to the waveguide 201 in the lateral direction and the depth direction are relatively matched between the stacked waveguide substrates even if they are not absolutely secured. Therefore, the stacking accuracy can be secured.

FIG. 15 schematically shows a state in which a two-dimensional optical fiber array and a one-dimensional optical fiber array are connected via a two-dimensional waveguide device. (A) is a plan view and (b) is a side view thereof. It is a figure. In FIG. 15, the four-channel one-dimensional optical fiber array 110 rotated by 90 ° is connected to the eight-channel two-dimensional optical fiber array 10 via the two-dimensional waveguide device 200. The waveguide device 200 is configured by laminating four waveguide substrate units 201 and the fixing member 3, and the two-dimensional optical fiber array 10 is configured by laminating four optical fiber array units 5 and the fixing member 3. The case is shown.

As described above, when connecting the waveguide substrate and the optical fiber array, it is necessary to optically align (align) one optical fiber array with one waveguide substrate. Since the alignment work is performed at the submicron level between the waveguide substrate and the optical fiber array substrate, there is a problem that the work is extremely sophisticated and requires many man-hours. With this configuration, the alignment of four substrates can be realized by one alignment, and the alignment man-hours and the connection man-hours can be significantly reduced.

[0129]

EXAMPLES Hereinafter, the present invention will be described in more detail with reference to Examples, but the present invention is not limited to these Examples.

Example 1 An 8 × 8-fiber two-dimensional optical fiber array (2DFA) having lateral and thickness-direction pitches of 1.5 mm was produced. The optical fiber array unit had a width of 13 mm and a thickness of 1.49 mm (total thickness of the V-groove substrate and the fixing member), and the thickness of the adhesive layer between the optical fiber array units was 10 μm. In order to obtain high reliability, it is preferable to set the thickness of the adhesive layer to about 10 to 20 μm,
The adhesive layer has a large coefficient of thermal expansion, and 10 μm was adopted in order to prevent the overall coefficient of thermal expansion from becoming too large.

Low curing shrinkage (2%) for manufacturing optical fiber array units, stacking and fixing optical fiber array units
Epoxy adhesive was used. At the lamination stage, the guide pin jig is held until the resin hardens so that the optical fiber array unit does not move due to hardening shrinkage. An adhesive was used.

As the stacking and fixing method, the above-mentioned second two-dimensional optical fiber array was adopted, and the guide groove pitch was set to 15 mm. Each optical fiber array unit was not subjected to end face polishing but was collectively polished after lamination. For this polishing, as shown in FIG. 3, the V-groove substrate has a structure in which it projects outward in the width direction from the fixing member, and the side surface of the V-groove substrate is used as the polishing reference plane. It is easy to cut the side surface of the V-groove substrate parallel to the V-groove at the stage of processing the substrate. It should be noted that, in this way, when the reference surface is projected outward, when the above-mentioned first two-dimensional optical fiber array is used, if the side surface of the V-groove substrate is gripped, the optical axis θy is always constant with a certain accuracy. This is preferable when it is necessary to know the relationship with the optical axis, not limited to the polishing reference, such that the adjustment of θy becomes easy because it can be secured.

For the core position measurement as a two-dimensional optical fiber array after lamination, a device capable of measuring the core position only one dimension at a time was used. Therefore, the core position of each layer was measured column by column, and the first row and the eighth row were measured. Eyes were measured in the thickness direction, and only one side was measured diagonally, and by combining these, the accurate core position was grasped.

As a result of the core position measurement, even with respect to the ideal matrix coordinates, even the optical fiber having the largest deviation amount (the optical fiber of the channel having the largest deviation amount among the 64 optical fibers) is 2 μm or less, Good results have been obtained. In this embodiment, since the guide groove is unnecessary, the side surface is cut off to have a width of 13 mm.

Further, the obtained two-dimensional optical fiber array was tested for heat cycle (-40 to 85 ° C. × 70 cycles) and high temperature and high humidity (85 ° C./85%×2 weeks). Good results were obtained with no peeling of the adhesive layer between the units and no abnormality such as disconnection.

Example 2 As an optical fiber, each end face on the light emitting side was polished so as to have an inclination forming a predetermined angle (θ = 8 °) with respect to a plane perpendicular to the optical axis. The optical fiber array unit using the product is laminated in multiple layers, and the reflection attenuation flow is small, and it is 8 × 1 at a pitch of 1.5 mm.
A two-dimensional optical fiber array with 0 channel specifications was manufactured. In this example, the production of the optical fiber array units, the lamination of the optical fiber array units, and the evaluation of the produced two-dimensional optical fiber array were performed in this order.

[Production of Optical Fiber Array Unit] As a substrate material, borosilicate glass (coefficient of thermal expansion: 32 × 1)
0 ⁻⁷ ), and this is 50 mm × 55 mm × 1.49
The wafer was processed into a dimension of 5 mm (thickness). A V-shaped groove and a positioning guide (guide V-groove) for aligning and fixing the optical fibers by grinding this.
Was formed. Eight V-grooves were formed at a pitch of 1.5 mm, and positioning guides (guide V-grooves) were formed one each at a position 2 mm outside the first V-groove and the eighth V-groove. .

Next, the wafer was cut into a predetermined size to obtain a substrate chip used for each optical fiber array unit. Since the two-dimensional optical fiber array to be produced has 8 × 10 channels, ten substrate chips (optical fiber array unit) were produced.

Next, the optical fiber array unit was assembled. To assemble, first place the optical fiber in the V-shaped groove of the substrate, and then apply a fixing member for temporary fixing (a SUS material having a thickness of 1 mm with a fluororesin such as polytetrafluoroethylene). Therefore, the fixing member for temporary fixing does not adhere itself.), The optical fiber is pressed, a fixing adhesive (epoxy resin) is applied, and ultraviolet light is irradiated to fix the optical fiber. After the fixing adhesive was cured, the fixing member for temporary fixing was peeled from the substrate to obtain an optical fiber array unit.

Next, in order to prevent the coated portion of the optical fiber from becoming free and causing a disconnection, a coating fixing adhesive (urethane acrylate resin) is used at the rear end portion of the optical fiber array unit. The covering portion of was fixed.

Then, polishing processing was performed to form an inclination on the end face of the optical fiber array unit. A dedicated jig was prepared to give the end face of the optical fiber array unit an angle of θ = 8 °. In this case, when the optical axis direction is 0 °, the inclination angle at the end face of the optical fiber array unit is 8
It becomes 2 °. Using this angled jig, rough lapping, precise lapping,
Polishing was performed in the order of polishing. As a result, the angle accuracy was ± 0.3 ° with respect to 82 °.

[Lamination of Optical Fiber Array Units] There are various methods for laminating the optical fiber array units as described above, but in this example, a guide pin stand jig was used. Here, the guide pin standing jig means a jig in which guide pins are erected with high accuracy for positioning the optical fiber array unit in the stacking direction. A zirconia material was used for both the guide pin and the jig substrate. For the beam that erects the guide pin, V at 1.5 mm pitch
A V-shaped groove is formed, and a guide pin (diameter: 0.7 mm) is inserted into this V-shaped groove and the guide pin is inverted by being pressed by a fixing member. In addition, the relative positions of the two beams on which the guide pins are erected must be defined with high accuracy. As described above, in order to determine the relative position with high accuracy, the relative positions of the pin-standing beams are accurately defined by using the lateral beam and the diagonal beam.

The optical fiber array units were sequentially inserted into the guide pins set on the guide pin setting jig described above. In this case, it is necessary that the guide pin and the positioning guide (guide V groove) formed in the optical fiber array unit are in accurate contact with each other. Therefore, by applying a load to the optical fiber array unit, it is possible to make sure contact between the two. In this case, the load applied to the optical fiber array unit was about 30 g. The relative positions of the optical fiber array units in the optical axis direction also need to be aligned with high accuracy. In this embodiment, the relative position is adjusted by bringing the end of each optical fiber array unit into contact with the substrate of the guide pin jig. In this way, the alignment variation of the end faces of each optical fiber array unit is about 10 μm.
Became. In this way, the optical fiber array unit was fixed to the guide pin jig, and the optical fiber array unit in the next stage was also fixed to the guide pin jig in the same manner.

Next, an adhesive (ultraviolet curing epoxy resin) was applied and fixed between the optical fiber array units. The thickness of the adhesive layer formed between the optical fiber array units was about 10 μm. By repeating this work,
A 10-dimensional two-dimensional optical fiber array was produced.

[Evaluation of Two-Dimensional Optical Fiber Array] First, the accuracy of the two-dimensional optical fiber array produced in this example was evaluated. Optical fiber array unit from the bottom row 1
10, the columns from the left column to columns A to H, the core positions of rows 1 to 10 (8 channels each) are measured, and the core positions of columns A and H (10 channels each).
Was measured. Further, the optical fibers of the matrix (1, A) and the optical fibers of the matrix (10, H) were specified as specific optical fibers, and the distance between the core positions was measured. In this way, the mutual positions of the lowermost optical fiber array unit corresponding to row 1 and the uppermost optical fiber array unit corresponding to row 10 are grasped, and the two-dimensional optical fibers are calculated from the core positions of columns A and H. The overall core position as an array was measured. The amount of deviation between this core position matrix and the ideal core position matrix was calculated.

As a result, it was confirmed that the core positions of the two-dimensional optical fiber array manufactured in this example were within ± 2 μm from the ideal matrix in all channels.

Next, the return loss on the end face of the two-dimensional optical fiber array produced in this example was evaluated. An interferometric reflectometer (OCD) is used to measure return loss.
R) was used. The length of the optical fiber from the connector of the two-dimensional optical fiber array to the end face of the array is about 1.8 m.
And

As a result, it was confirmed that the return loss on the end face of the two-dimensional optical fiber array manufactured in this example was 60 dB or more in all channels.

Next, the reliability of the two-dimensional optical fiber array manufactured in this example was evaluated. Two-dimensional optical fiber array heat cycle test (-40 to 85 ° C x 70 cycles), high temperature and high humidity test (85 ° C, 85% x 2 weeks)
Exposed to.

As a result, it was confirmed that the two-dimensional optical fiber array manufactured in this example did not change in its core position accuracy and return loss before and after the test. The variation of the core position was 0.3 μm or less, which was sufficiently small, and there was no change in the return loss, and good results were obtained.

Example 3 A two-dimensional waveguide device in which four 1 × 8 channel splitters were stacked was connected to a two-dimensional optical fiber array. Si on a 1 mm thick Si wafer
A waveguide core is located at a waveguide pitch (8 channel side) of 250 μm at a position of 1.03 mm from the bottom surface of the wafer, and a cladding is formed on the same with a thickness of 0.025 mm, and a total thickness of 1.05
A 5 mm splitter unit was formed. As shown in FIG. 14, a 4-chip splitter is formed on one wafer,
Positioning guides (grooves) were formed on this wafer by grinding at a pitch of 5 mm. In order to secure the stacking accuracy, the center position when the stacking guide pin is placed in the guide groove and the depth and pitch relative to the waveguide are the same,
Processing was performed as follows. The wafer is
On a processing jig as shown in FIG. 3 of Japanese Patent No. 273442, it is attached so as to be parallel to the workpiece reference surface (both the side surface side and the bottom surface side) of the jig, and the processing machine Positioning guide processing was performed so as to be parallel to the side reference surfaces (both the side surface side and the bottom surface side) so that the relative positions were exactly the same. In this case, a necessary relative relationship such as a parallel relationship or a right angle relationship between the workpiece side reference surface of the processing jig and the processing machine side reference surface is ensured. The wafer was cut and processed into splitter chips, which were stacked and fixed by a stacking method shown in FIG. 13B so that the core pitch in the stacking direction was 1.06 mm. That is, the thickness of the adhesive layer between the splitter units was 5 μm. The end face of this two-dimensionalized splitter was polished by 8 ° so as to have the shape shown in FIG. 13 (b). Next, an 8-fiber × 4-stage two-dimensional optical fiber array having the shape shown in FIG. 1 was produced by the same method as in Example 1.
In this case, the 8 ° polishing process on the end face was performed after the lamination. At this time, the thickness of the optical fiber array unit (distance from the bottom surface of the V-groove substrate to the top of the fiber) was 1.055 mm together with the splitter unit, and the stacking pitch was 1.06 mm. That is, the thickness of the adhesive layer between the optical fiber array units was 5 μm. Further, a one-stage (one-dimensional) 4-fiber optical fiber array with an optical fiber pitch of 1.06 mm was produced.
The end face was polished by 8 ° (8 ° in the direction of the fiber pitch of the one-dimensional optical fiber array) in the direction in which the end face becomes parallel when connected to the two-dimensional splitter. Two-dimensional splitter modules were produced by aligning, connecting, and fixing the three members in the same manner as the ordinary waveguide splitter module.

As a result, the two-dimensional splitter module manufactured in this example has a connection loss of 0.5d.
B, the reflection was 60 dB, which was slightly larger than the loss of the one-dimensional splitter module, but it was a two-dimensional splitter module having characteristics that it could be used sufficiently.

[0153]

As described above, according to the present invention, a two-dimensional optical member array and a two-dimensional optical member array having high alignment accuracy of optical members (for example, optical fibers, lenses, etc.) on a substrate and excellent long-term reliability are provided. It is possible to provide a two-dimensional waveguide device that has a large capacity in terms of density and can reduce the number of steps in packaging and connection.

[Brief description of drawings]

FIG. 1 is a sectional view schematically showing a two-dimensional optical fiber array which is a first embodiment of a two-dimensional optical member array of the present invention.

FIG. 2 is an enlarged sectional view showing a groove portion of the substrate in FIG.

FIG. 3 is a cross-sectional view schematically showing a two-dimensional optical fiber array which is a second embodiment of the two-dimensional optical member array of the present invention.

4A and 4B are a cross-sectional view and an enlarged view of part A schematically showing a two-dimensional optical fiber array which is a third embodiment of the two-dimensional optical member array of the present invention.

FIG. 5 is a sectional view schematically showing a two-dimensional optical fiber array which is a fourth embodiment of the two-dimensional optical member array of the present invention.

FIG. 6 is a sectional view schematically showing a two-dimensional optical fiber array that is a fifth embodiment of the two-dimensional optical member array of the present invention.

FIG. 7 is a cross-sectional view schematically showing the relationship between the reflection characteristic and the angular deviation (Δθ) in the two-dimensional optical fiber array according to the fourth embodiment shown in FIG.

FIG. 8 is a cross-sectional view schematically showing the relationship between the reflection characteristic and the angular deviation (Δθ) in the two-dimensional optical fiber array according to the fifth embodiment shown in FIG.

FIG. 9 is a sectional view schematically showing a two-dimensional optical fiber array that is a sixth embodiment of the two-dimensional optical member array of the present invention.

FIG. 10 is explanatory diagrams (a) to (c) schematically showing the configuration of a simple in-line switch.

FIG. 11 is a cross-sectional view schematically showing a guide pin jig used for making a two-dimensional optical fiber array which is an embodiment of a two-dimensional optical member array of the present invention two-dimensional.

FIG. 12 schematically shows an embodiment of a two-dimensional waveguide device of the present invention, (a) is a plan view and (b) is a sectional view taken along line YY.

FIG. 13 is a cross-sectional view (two aspects), (a), and (b) taken along line ZZ in FIG.

FIG. 14 is a plan view schematically showing a method of stacking waveguide substrate units to make them two-dimensional in one embodiment of the two-dimensional waveguide device of the present invention.

FIG. 15 schematically shows a state in which a two-dimensional optical fiber array and a one-dimensional optical fiber array are connected via a two-dimensional waveguide device, (a) is a plan view, and (b) is a side view thereof. .

FIG. 16 is a sectional view schematically showing an example of a conventional two-dimensional optical member array (optical fiber array).

FIG. 17 schematically shows an example of a conventional waveguide device,
(A) is a top view, (b) is sectional drawing in the XX line of (a).

FIG. 18 shows a two-dimensional optical member array according to the present invention,
It is sectional drawing which shows typically one embodiment of the state which accommodated the optical fiber in the through-hole of a cylindrical member.

FIG. 19 shows a two-dimensional optical member array according to the present invention,
It is sectional drawing which shows typically the case where a polarization maintaining optical fiber is used as an optical fiber.

FIG. 20 shows a two-dimensional optical member array according to the present invention,
It is sectional drawing which shows typically another embodiment of the state which accommodated the optical fiber in the through-hole of a cylindrical member, (a) is a vertical sectional view with respect to the axial direction of an optical fiber, (b) is (a). )
6 is a sectional view taken along line XX of FIG.

FIG. 21 shows a two-dimensional optical member array according to the present invention,
It is sectional drawing which shows other embodiment of the state which accommodated the optical fiber in the through-hole of a cylindrical member typically.

FIG. 22 shows a two-dimensional optical member array of the present invention,
It is sectional drawing which shows other embodiment of the state which accommodated the optical fiber in the through-hole of a cylindrical member typically.

FIG. 23 shows a two-dimensional optical member array according to the present invention,
It is sectional drawing which shows other embodiment of the state which accommodated the optical fiber in the through-hole of a cylindrical member typically.

[Explanation of symbols]

1 ... Optical fiber, 1a ... Specific optical fiber, 2 ... Substrate, 2
a ... the bottom substrate, 2b ... the second lowest substrate, 2d ...
The uppermost substrate, 3 ... Fixing member, 3a ... Fixing member arranged on the lowermost optical fiber array unit, 3s ... Surface of fixing member, 4 ... Adhesive layer, 5 ... Optical fiber array unit,
5a ... bottommost optical fiber array unit, 5b ... 2 from the bottom
Optical fiber array unit of layer 5d ... Optical fiber array unit of top layer, 6 ... Positioning guide (guide V
Grooves), 6a to 6d ... Guide pins, 7 ... Flat plate microlens, 10 ... Two-dimensional optical fiber array, 21 ... Grooves (V-shaped grooves), 21a, 21b ... Groove sidewalls, 22 ... Bottom optical fiber array unit The upper surface of the substrate constituting 23, the lower surface of the substrate constituting the second optical fiber array unit from the bottom,
24 ... Input side fiber array, 25 ... Output side fiber array, 26 ... In-line type optical switch, 100 ... Conventional two-dimensional optical fiber array, 101 ... Optical fiber, 102
... V-groove substrate, 103 ... Fixing member, 110 ... One-dimensional optical fiber array, 200 ... Two-dimensional waveguide device, 201 ... Waveguide, 202 ... Substrate, 204 ... Adhesive layer, 205 ... Waveguide substrate unit, 206 ... Positioning guide, 301a ... Optical fiber, 301b ... Outer coating, 301c ... Polarization maintaining optical fiber, 302 ... Cylindrical member, 302a ... Through hole, 30
2b ... Tapered portion, 302c ... Opening side end portion, 303 ... Substrate, 303a ... Groove, 303b ... Substrate end portion, 304 ... Fixing member, 304a ... Fixing member end portion, 305 ... Adhesive layer, 3
06 ... stress applying part, 307 ... core, 310 ... two-dimensional optical member array, θ ... angle of inclination, D ... distance between core positions of specific optical fiber, G1, G2 ... load, S ... light of optical fiber Input / output end faces, S '... End face of waveguide of light input / output, L ... Center axis of optical fiber, L' ... Center axis of waveguide, P ... Optical axis of input / output light from optical fiber , Q
... an optical axis of the flat plate microlens, V ... a surface perpendicular to the central axis of the optical fiber, V '... a surface perpendicular to the central axis of the waveguide, U ... a predetermined angle with respect to a surface perpendicular to the central axis of the optical fiber ( θ), a surface that forms a predetermined angle (θ) with respect to a surface that is perpendicular to the central axis of the waveguide, and a surface that is perpendicular to the optical axis P of the incoming and outgoing light.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Akihiro Ide             2-56, Sudacho, Mizuho-ku, Nagoya-shi, Aichi             Inside Hon insulator Co., Ltd. F term (reference) 2H036 JA01 LA03 LA07                 2H047 KA03 KB08 TA05

Claims (24)

[Claims] 1. An optical member and a substrate having, on one surface thereof, one or more grooves corresponding to the outer shape of the optical member, on which one or more optical members are aligned and fixed. A two-dimensional optical member array formed by stacking a plurality of the optical member array units in a hierarchical manner, each of which opposes a substrate constituting the adjacent optical member array unit among the plurality of optical member array units. The plurality of optical member array units are laminated in a layered manner in such a manner that the surfaces to be contacted with each other are not in direct contact with each other and the mechanical influences are not directly exerted on each other. -Dimensional optical member array. 2. The two-dimensional optical member array according to claim 1, wherein the optical member is an optical fiber or a lens. 3. The optical member array unit is formed by contacting a vertex of an optical member arranged on a substrate forming the optical member array unit with a surface of a substrate facing the optical member array unit facing the vertex. A plurality of the optical member array units are layered in a layered manner in such a manner that the opposing surfaces of the substrates are not in direct contact with each other and do not directly exert a mechanical influence on each other. Item 2. The two-dimensional optical member array according to Item 1 or 2. 4. An apex of optical members arranged on a substrate forming the optical member array unit and a surface of a substrate facing the optical member array unit facing the apex,
With an adhesive layer interposed, the apex of the optical member and the surface of the substrate are respectively brought into contact with the adhesive layer, and the respective surfaces of the substrates constituting the optical member array unit are directly in contact with each other. The two-dimensional optical member array according to claim 1 or 2, wherein a plurality of the optical member array units are layered in a layered manner in a state in which the optical members do not directly affect each other without mechanical influence. 5. The optical member of the substrate is provided on one surface of the substrate forming the optical member array unit in the uppermost layer and between the substrates forming the optical member array unit adjacent to each other. The two-dimensional optical member array according to claim 1 or 2, further comprising a fixing member that presses or mounts on one surface side having the groove to align and fix. 6. The surface of the fixing member and the surface of the substrate constituting the optical member array unit facing the surface of the fixing member are not in direct contact with each other, and
Without directly affecting each other mechanically,
The two-dimensional optical member array according to claim 5, wherein the fixing member presses or mounts the optical member on one surface side of the substrate having the groove to align and fix the optical member. 7. The optical member is aligned or fixed by pressing or placing it on the substrate in a state where the optical member is in contact with the surface of the fixing member and the side wall forming the groove. The two-dimensional optical member array according to claim 5 or 6. 8. The adhesive layer is further provided between the surface of the fixing member and the surface of the substrate constituting the optical member array unit facing the surface of the fixing member. The two-dimensional optical member array described in any one of 1. 9. The thickness of the adhesive layer is 2 to 100 μm.
The two-dimensional optical member array according to claim 8. 10. The two-dimensional structure according to claim 1, wherein a positioning guide is formed at a predetermined position on one surface of the substrate forming the optical member array unit having the groove. Optical member array. 11. The groove is a V-shaped groove.
The two-dimensional optical member array according to any one of 0. 12. An end surface on the side of emitting light and an end surface on the side of entering light of the optical members constituting the optical member array unit, or one of them is a surface perpendicular to the central axis of the optical member. The two-dimensional optical member array according to any one of claims 1 to 11, comprising an inclination that forms a predetermined angle (?) With respect to. 13. An end surface of the optical member on the side where light is emitted and an end surface on the side where light is incident, or any one of them.
The two-dimensional optical member array according to claim 12, wherein the two-dimensional optical member array is arranged on a surface perpendicular to the central axis of each of the optical members. 14. An end surface of the optical member on the side where light is emitted and an end surface on the side where light is incident, or one of them,
13. Each of the optical members is arranged on a surface forming a predetermined angle (θ) with respect to a surface perpendicular to the central axis of the optical member.
The two-dimensional optical member array described in 1. 15. The end surface of the optical member on the side from which light is emitted and the end surface on the side from which light is incident, or any one of them,
The two-dimensional optical member array according to claim 12, wherein the two-dimensional optical member array is provided on each of the outgoing light and the incident light or a surface perpendicular to the optical axis of either one of them. 16. A method for measuring a core position of the optical member constituting the two-dimensional optical member array according to claim 1, wherein the optical member array unit is layered in m layers in a hierarchical manner. And each one of the optical member array units has n channels (when the optical members are aligned in m rows and n columns), the core position of each row of the m optical members is measured. At the same time, the core positions of at least two rows of the optical members among the n rows of optical members are respectively measured,
Further, one optical member is arbitrarily specified from each row of the at least two rows of optical members, and a distance between core positions of the specified optical member (specific optical member) is measured, The core positions of the optical members corresponding to the four corners of the quadrangle formed by diagonally connecting the core positions of the optical members are calculated, and the positional relationship of the mutual matrix is calculated, and the cores of the entire optical members are calculated. A method of measuring a core position of an optical member that constitutes a two-dimensional optical member array, characterized by calculating a position. 17. A two-dimensional waveguide device in which a plurality of waveguide substrate units each having one or more planarly patterned waveguides are laminated in a layered manner, wherein a plurality of the waveguide substrate units are provided. Among the plurality of the waveguide substrate units, the surfaces of the waveguide substrate units adjacent to each other, which are opposed to each other, are not in direct contact with each other, and the mechanical influences are not directly exerted on each other. A two-dimensional waveguide device characterized by being laminated in a hierarchical manner. 18. The two-dimensional waveguide device according to claim 17, wherein an adhesive layer is provided between the surfaces of the waveguide substrate units adjacent to each other among the plurality of waveguide substrate units that face each other. 19. The adhesive layer has a thickness of 2 to 100 μm.
The two-dimensional waveguide device according to claim 17 or 18, wherein m is m. 20. A positioning guide is formed at a predetermined position on the surface of the waveguide substrate unit.
The two-dimensional waveguide device according to any one of 19 above. 21. Each of the end faces on the light emitting side of the waveguide forming the waveguide substrate unit is provided with an inclination forming a predetermined angle (θ) with respect to a plane perpendicular to the optical axis. The two-dimensional waveguide device according to any one of claims 17 to 20. 22. An end face on the side of emitting light and an end face on the side of entering light of the waveguides constituting the waveguide substrate unit, or any one of them is perpendicular to the central axis of the waveguide. The two-dimensional waveguide device according to claim 21, wherein the two-dimensional waveguide device is arranged on a surface. 23. An end face on the side of emitting light and an end face on the side of entering light of the waveguides constituting the waveguide substrate unit, or any one of them is perpendicular to the central axis of the waveguide. The two-dimensional waveguide device according to claim 21, wherein the two-dimensional waveguide device is arranged on a surface forming a predetermined angle (θ) with respect to the surface. 24. An end face on the side of emitting light and an end face on the side of entering light of the waveguides constituting the waveguide substrate unit, or any one of them is an emitted light and / or an incident light, respectively. The two-dimensional waveguide device according to claim 21, wherein the two-dimensional waveguide device is arranged on a surface perpendicular to the optical axis.