JP4189404B2 - Optical device - Google Patents

Description

  The present invention relates to an optical device including an optical fiber array having one or more optical fibers, and more particularly to an optical device suitable for monitoring signal light propagating through an optical fiber.

Recently, with the development of wavelength division multiplexing using fiber amplifiers, in order to secure amplifier characteristics, the amount of light at each wavelength is monitored with a photodiode (PD), and the amount of light is adjusted and then amplified with an amplifier. It has come to be adopted.
Various methods are known for this monitor, but since a monitor device is mounted on each optical fiber, a considerable size is required only by the monitor device.
Therefore, miniaturization and high density of the monitor device are desired. Further, when monitoring, a part of the signal light is taken out, but what can be monitored without greatly attenuating the signal light is desired.
Conventionally, for example, a technique as disclosed in JP 2001-264594 A has been proposed. In this technique, an optical fiber is disposed in a V-groove of a glass substrate, and then a parallel groove is formed so as to obliquely cross the optical fiber (with respect to its optical axis) with respect to the glass substrate. Then, a light reflecting substrate (optical member) is inserted into the parallel groove, and an ultraviolet curable resin (adhesive) is filled in the gap.
Thereby, out of the signal light propagating through the optical fiber, the light component reflected by the light reflecting substrate (reflected light) is extracted out of the cladding. Therefore, the signal light can be monitored by detecting the reflected light with a light receiving element, for example.
By the way, when PDs are arranged on an optical fiber, most of them are single-core, and metal package type PDs are often arranged (for example, JP-A-10-300936 and JP-A-11-133255). Gazette, International Publication No. 97/06458 pamphlet). This is because there are few spatial restrictions because it is a single core, and many metal package type PDs are on the market and have a great track record in price and reliability.
On the other hand, in the case of many cores, it is difficult to use a metal package type PD. In particular, when high-density mounting is required such that the arrangement pitch of optical fibers is 250 μm, it is necessary to install a photodiode array (PD array) in which a plurality of bare photodiodes are arranged. .
In this case, considering the derivation of electrode signals from the PD array, it can be considered that the PD array is mounted on a wiring board (submount) and the submount is mounted on a multi-fiber optical fiber array.
However, of the reflected light from a certain channel, the light that has gone out of the active layer of the PD array or the light that has passed through the active layer is reflected on the mounting surface (the surface on which the PD array is mounted) of the PD array or submount, and stray light And entering into other channels may cause crosstalk.
This is caused by the presence of an electrode pattern or a solder layer for mounting on the mounting surface of the PD array and / or submount, which is a material having high reflectivity.
The above-described problem occurs particularly when reflected light is incident on the PD array from an oblique direction. That is, in the case of normal incidence, since the reflected light is perpendicularly incident on the active layer of the PD array, it is recombined with the same channel (desired channel) and does not cause a problem of crosstalk. When this is obliquely incident, the reflected light hits a material having a high reflectance, so that it becomes stray light thereafter, resulting in a problem of crosstalk.
The present invention has been made in consideration of such problems, and an object thereof is to provide an optical device capable of effectively reducing crosstalk even when reflected light is incident obliquely.

An optical device according to the present invention includes a first substrate having a V-groove, one or more optical fibers fixed to the V-groove of the first substrate and provided with a reflection function, and the light Out of the fiber cladding, at least an optical element fixed on an optical path of reflected light generated by the reflection function via an adhesive layer, and a second substrate for mounting the optical element on a main surface are provided. A light reflecting material is not present in at least a gap between the active layer of the optical element and the main surface of the second substrate in the gap between the main surface of the second substrate and the optical element. Features.
That is, of the gap between the main surface of the second substrate and the optical element, the gap (first gap) between the active layer of the optical element and the main surface of the second substrate, and the vicinity of the first gap By adopting a configuration in which a light reflecting material such as a metal does not exist in the peripheral gap (second gap), it corresponds to the first gap and the second gap among the main surfaces of the optical element and the second substrate. Reflection at the part can be suppressed. Thereby, even if the reflected light is incident obliquely, that is, even if the incident angle of the reflected light to the optical element is 10 ° or more with respect to the vertical direction, it is possible to effectively reduce the crosstalk.
And in the said structure, it is preferable that the refractive index of the gap | interval of the active layer of the said optical element and the said main surface of the said 2nd substrate is lower than the refractive index of the said optical element. As a result, first, the reflected light spreads or scatters by diffraction when entering the first gap and the second gap. If the reflected light spreads, the reflected light density decreases, and even if it becomes stray light and enters the other channel, its component is small, so it does not reach the point where it is detected as crosstalk.
In order to obtain a large diffraction in the first gap and the second gap, it is only necessary to be filled with a material having a low refractive index, and the most preferable is air. However, in the case of a low-refractive index material such as air, the transmitted light can surely obtain the spread of light due to diffraction, but a reflection component is generated due to the refractive index difference between the optical element and air. Although this reflected light is considerably smaller than the reflected light with respect to a metal substance, it becomes a problem in some cases.
Therefore, it is preferable that the refractive indexes of the first gap and the second gap are adjusted to the surface of the optical element to some extent, and the light spread due to scattering is obtained in the first gap and the second gap. The refractive index of the gap is preferably ± 30% or less of the refractive index of the optical element surface.
As a result, reflection at the interface between the optical element and the first gap and the second gap can be suppressed, and the spread of reflected light in the first gap and the second gap can be obtained. The characteristic of being very small can be obtained.
As a means for obtaining scattering in the first gap and the second gap, for example, if the particles are scattered in the material, the reflected light is scattered by the particles, and desired scattering can be obtained. If the particles are in such a high-density state that the particles are in contact with each other, the reflection will increase. Therefore, even if they are partially in contact with each other, a state where they are scattered in a macro view is a preferable state. .
An anisotropic conductive material is mentioned as such a material. An anisotropic conductive material, for example, an anisotropic conductive paste, is dotted with spherical polymers having a diameter of about 3 μm coated with a conductive material such as Au (gold), and thus scatters in the first gap and the second gap. It can be said that it is a material suitable for obtaining.
Further, an electrode layer is interposed in a portion of the gap between the main surface of the second substrate and the optical element excluding the first gap and the second gap (a portion that does not affect the reflection optical path). By conducting thermocompression bonding of the anisotropic conductive material, the electrodes having a raised electrode pattern such as Au are electrically connected by a spherical polymer coated with the conductive material. On the other hand, in the part far from the electrode layer, the polymer is not in contact with each other, and the density is low and scattered. That is, a more preferable state can be realized by arranging the electrode layers around the first gap and the second gap.
As a result, the anisotropic conductive material makes it possible to obtain both conduction of the electrode and scattering of reflected light between the optical element and the submount.
As described above, it is an object to obtain the spread of light in the first gap and the second gap, whether diffraction or scattering. That is, if the distance between the first gap and the second gap is long, the effect can be obtained more. From this viewpoint, the gap is preferably 20 μm or more.
In order to obtain this distance, for example, a spacer or the like may be disposed. However, from this viewpoint, it is preferable that the electrode layer is disposed as described above because the electrode layer also serves as a spacer.

FIG. 1 is a cross-sectional view showing a main part of the optical device according to the first embodiment cut along a plane orthogonal to the optical axis of the optical fiber.
FIG. 2 is a cross-sectional view showing the main part of the optical device according to the first embodiment cut along the optical axis of the optical fiber.
FIG. 3 is a cross-sectional view showing a mounting form of the PD array on the submount in the optical device according to the first embodiment.
FIG. 4 is a cross-sectional view showing a mounting form of a PD array on a submount in the optical device according to the second embodiment.
FIG. 5 is a cross-sectional view showing a mounting form of a PD array on a submount in the optical device according to the third embodiment.
FIG. 6 is a cross-sectional view illustrating a main part of the optical device according to the example.
FIG. 7 is an explanatory diagram illustrating a back-illuminated PD array used in the optical device according to the example.
FIG. 8 is a cross-sectional view showing a state in which the PD array is mounted on the lower surface of the submount.
FIG. 9 is a perspective view showing a state where an optical head is formed by mounting a submount on which a PD array is mounted on an optical fiber array.
FIG. 10 is a perspective view showing a state where the electrode pads of the submount and the pins of the package are wire-bonded.
FIG. 11 is a perspective view showing a state in which a ring and a boot are fixed to the optical head, and the optical head is sealed with resin.
FIG. 12 is a perspective view showing a state where an optical device is completed by fixing a lid to the upper surface opening of the ring.

Hereinafter, several embodiments in which an optical device according to the present invention is applied to, for example, an in-line type power monitor module will be described with reference to FIGS.
The optical device 10A according to the first embodiment is basically fixed to a glass substrate 12 and a plurality of V grooves 14 provided on the glass substrate 12, as shown in FIGS. On the optical path of the light (reflected light) 24 generated at least in the optical fiber array 16 composed of a plurality of optical fibers 15 each provided with a reflective part 33 and the light 22 transmitted through each optical fiber 15. And a PD (photodiode) array 28 that detects the reflected light 24, and a wiring board (submount) 30 for mounting the PD array 28. The submount 30 is configured such that the mounting surface 30 a of the PD array 28 is placed facing the glass substrate 12.
Here, since an example in which the optical fiber array 16 is configured by a plurality of optical fibers 15 is shown, “each optical fiber 15” means “each of four optical fibers”. However, since the optical fiber array 16 can be configured with a single optical fiber 15, in this case, “each optical fiber” or “a plurality of optical fibers” may be read as “one optical fiber”. .
The PD array 28 has a front-illuminated configuration as shown in FIG. That is, the PD array 28 includes a substrate 29 and a number of active layers 26 corresponding to the number of channels formed in the substrate 29, and the plurality of active layers 26 include the optical fiber array 16 ( (See FIG. 1). Of the upper surface (surface facing the submount 30) and the lower surface (surface facing the optical fiber array 16) of the PD array 28, antireflection films (AR coatings) 100 and 102 are provided on the portions corresponding to the active layer 26, respectively. Is formed.
On the submount 30 side of the PD array 28, an anode electrode and a cathode electrode are formed by an Au electrode pattern 104, and a common cathode electrode and an anode electrode of each channel are formed by an Au electrode pattern 64 on the mounting surface 30a of the submount 30. Is formed.
The Au electrode pattern 104 formed on the PD array 28 and the Au electrode pattern 64 of the submount 30 are formed by a conductive layer 106 such as a solder layer, ACP (anisotropic conductive paste), or ACF (anisotropic conductive film). Electrically connected. In addition, although the above-mentioned PD array 28 uses the cathode common type, the same effect can be obtained even if the anode common type is used.
The optical device 10A according to the first embodiment includes a gap (first gap) 112 corresponding to the active layer 26 in the gap 110 between the PD array 28 and the mounting surface 30a of the submount 30. A gap (second gap) 114 around the first gap 112 is air 116.
With this configuration, of the reflected light 24 from a certain channel, the light that has passed through the active layer 26 of the PD array 28 or the light that has passed through the active layer 26 is reflected by the air 116 and the PD array in the first gap 112 and the second gap 114. The submount 30 passes through the interface with the substrate 29 at 28, the interface between the air 116 and the antireflection film 102 in the first gap 112 and the second gap 114, and the air 116 in the first gap 112 and the second gap 114. The mounting surface 30a is reached.
Since the mounting surface 30a of the submount 30 is not a surface having a high reflectance like the Au electrode pattern 64, the first gap 112 and the first gap among the mounting surface 30a of the PD array 28 and the submount 30 are included. The reflection at the portion corresponding to the two gaps 114 can be suppressed. Thereby, even if the reflected light 24 is incident obliquely, that is, even if the incident angle of the reflected light 24 to the PD array 28 is 10 ° or more with respect to the vertical direction, it is possible to effectively reduce the crosstalk. it can.
By the way, for example, when the substrate 29 of the PD array 28 is made of InP, the refractive index of the substrate 29 is about 3.2. When SiN is formed on the surface of the substrate 29 as the antireflection film 102, the antireflection is performed. The refractive index of the film 102 is about 1.9.
The reflected light 24 passes through the substrate 29 and the antireflection film 102 of the PD array 28, but it is preferable that the reflected light 24 spreads by diffraction or by scattering in the first gap 112 and the second gap 114. . This is because when the reflected light 24 spreads, the linear density of the reflected light 24 is reduced, and even if it becomes stray light and enters other channels, its component is small.
In order to obtain a large diffraction in the first gap 112 and the second gap 114, it is sufficient that a material having a low refractive index is filled. In the first embodiment, the substrate 29 and the antireflection film of the PD array 28 are used. Since air 116 having a refractive index lower than the respective refractive indexes of 102 is filled, large diffraction can be obtained in the first gap 112 and the second gap 114.
The submount 30 is preferably a ceramic polycrystalline material. In the case of a metal, the reflectance is high, but since it is a metal, scattering of reflected light is small, and stray light is likely to occur. On the other hand, if the submount 30 is a ceramic polycrystalline material, even if the mounting surface 30a of the submount 30 is a surface close to a mirror surface for forming the Au electrode pattern 64, light is transmitted by the crystal grain boundary. Scattering occurs. For this reason, the amount of reflected light incident on other channels as stray light is reduced, and the problem of crosstalk can be reduced.
In the case of a low-refractive index material such as air 116, the reflected light 24 can surely obtain a light spread due to diffraction, but a reflected component due to a difference in refractive index is generated. This reflection is much less than that of metallic materials, but in some cases it is problematic.
Therefore, the refractive index of the first gap 112 and the second gap 114 is adjusted to some extent to the refractive index of the substrate 29 in the PD array 28, and the light spread due to scattering is obtained in the first gap 112 and the second gap 114. It is preferable to keep it. The refractive index of the first gap 112 and the second gap 114 is preferably ± 30% or less of the refractive index of the substrate 29 in the PD array 28. In the first embodiment, since the antireflection film 102 such as SiN is formed on the surface of the PD array 28 (the surface facing the submount 30), a preferable refractive index difference can be realized.
Therefore, reflection at the interface between the PD array 28 and the first gap 112 and the second gap 114 can be suppressed, and the spread of the reflected light 24 in the first gap 112 and the second gap 114 can be obtained. As a result, a characteristic that crosstalk is very small can be obtained.
Next, an optical device 10B according to a second embodiment will be described with reference to FIG.
The optical device 10B according to the second embodiment has substantially the same configuration as the optical device 10A according to the first embodiment described above. However, as shown in FIG. The difference is that the gap 114 is filled with an anisotropic conductive material 120 such as ACP (anisotropic conductive paste) or ACF (anisotropic conductive film).
In this case, the anisotropic conductive material 120 is in a state in which a spherical polymer coated with a conductive material such as gold is scattered in the material 120, and the reflected light 24 is scattered by these particles, and the desired scattering is achieved. Can be obtained. In addition, since a reflection will become large conversely if it is a high density state so that particles may contact, even if it contacts partially, the state which is scattered macroscopically is preferable.
Thus, also in the optical device 10B according to the second embodiment, similarly to the optical device 10A according to the first embodiment, at the interface between the PD array 28, the first gap 112, and the second gap 114. In addition, the spread of the reflected light 24 in the first gap 112 and the second gap 114 can be obtained, and as a result, the characteristic that the crosstalk is very small can be obtained.
Next, an optical device 10C according to a third embodiment will be described with reference to FIG. In addition, the same code | symbol is attached | subjected about the part corresponding to FIG. 4, and the duplication description is abbreviate | omitted.
The optical device 10C according to the third embodiment is different in that the PD array 28 has a back-illuminated configuration as shown in FIG.
That is, the PD array 28 has a substrate 29 and a number of active layers 26 corresponding to the number of channels formed in the substrate 29, and the plurality of active layers 26 are close to the submount 30 in the substrate 29. The submount 30 is formed in a portion facing the mounting surface 30a.
In the optical device 10C, the Au electrode pattern 104 formed on the PD array 28 and the Au electrode pattern 64 of the submount 30 are, for example, ACP (anisotropic conductive paste), ACF (anisotropic conductive film), etc. The conductive layer 106 and the Au bump 122 are electrically connected.
Further, in the optical device 10C according to the third embodiment, the ACP (anisotropic conductive paste) is provided in the first gap 112 and the second gap 114, similarly to the optical device 10B according to the second embodiment. ) Or an anisotropic conductive material 120 such as ACF (anisotropic conductive film).
Therefore, in this optical device 10C, similarly to the optical device 10B according to the second embodiment, reflection at the interface between the PD array 28 and the first gap 112 and the second gap 114 can be suppressed. The spread of the reflected light 24 in the first gap 112 and the second gap 114 can be obtained, and as a result, the characteristic that the crosstalk is very small can be obtained.
Further, in this optical device 10C, a portion of the gap 110 between the mounting surface 30a of the submount 30 and the PD array 28 excluding the first gap 112 and the second gap 114 (which has an effect on the optical path of the reflected light 24). The Au bump 122 is interposed in the non-existing portion. Accordingly, the spherical polymer in the anisotropic conductive material 120 filled in the first gap 112 and the second gap 114 is attracted to the Au bump 122.
Therefore, in the anisotropic conductive material 120, the density around the Au bump 122 is high enough to make contact with the spherical polymer, and the portion far from the Au bump 122 (for example, the portion of the first gap 112) is the density of the spherical polymer. Is low and moderately scattered. This state is suitable for scattering of the reflected light 24, and further spread of the reflected light 24 in the first gap 112 and the second gap 114 can be obtained.
As a result, the anisotropic conductive material 120 makes it possible to obtain both electrical conduction between the PD array 28 and the submount 30 and scattering of the reflected light 24.
As described above, the object is to obtain the spread of the reflected light 24 in the first gap 112 and the second gap 114, whether it is diffraction or scattering. That is, if the distance L between the first gap 112 and the second gap 114 is long, the effect can be obtained more. From this viewpoint, the gap L is preferably 20 μm or more.
In order to obtain this distance L, for example, a spacer or the like may be provided. From this viewpoint, if the Au bump 122 is provided between the submount 30 and the PD array 28 as described above, the Au bump 122 is provided. Since this also serves as a spacer, the interval L between the first gap 112 and the second gap 114 can be easily adjusted by the Au bump 122, which is preferable.
Next, an example of the optical device 10C according to the third embodiment will be described with reference to FIGS.
As shown in FIG. 6, an optical device 10Ca according to this embodiment includes an optical fiber array 16 including a glass substrate 12 and a plurality of optical fibers 15 fixed to a plurality of V grooves 14 provided on the glass substrate 12. A slit 18 provided from each upper surface of each optical fiber 15 to the glass substrate 12, a branch member 20 inserted into the slit 18, and light 22 transmitted through each optical fiber 15. A PD array 28 that is fixed on the optical path of the reflected light 24 generated at 33 via an adhesive layer 60 and in which a plurality of active layers 26 that detect the reflected light 24 are arranged; and the PD array 28 is mounted, Further, a submount 30 for fixing the PD array 28 toward the optical fiber array 16 and a spacer 32 for stably fixing at least the PD array 28 are provided. To. The two end surfaces of the slit 18 and the front and back surfaces of the branching member 20 function as a reflecting portion 33 that reflects a part of the light 22 that passes through the optical fiber 15.
Here, a manufacturing method of the optical device 10Ca according to this embodiment will be described. First, the glass substrate 12 used for the in-line optical fiber array 16 was produced by grinding.
As the material of the glass substrate 12, a borosilicate glass material (here, in particular Pyrex (registered trademark) glass) was used. The glass substrate 12 had a length of 16 mm and a thickness of 1 mm. The V-grooves 14 for aligning the optical fiber array 16 were formed by grinding 12 pieces at a pitch of 250 μm and a depth of about 90 μm.
Next, the optical fiber array 16 was assembled. As the optical fiber array 16, a 12-core fiber core 46 (see FIG. 9) having a pitch of 250 μm was used. The 12-core tape core wire 46 is peeled in the middle so that the coating removal portion (striped portion) in the middle is 12 mm, placed on the V-groove 14 of the glass substrate 12, and the ultraviolet curable adhesive 52 (see FIG. 9). ).
Next, the slit 18 for the optical fiber array 16 was processed. The slit 18 had a thickness of 30 μm, a depth of 200 μm, and an inclination angle α of 20 ° (see FIG. 6).
Next, the branch member 20 was manufactured. The substrate 124 of the branch member 20 is made of quartz glass. A branching multilayer film 126 was formed on the quartz glass substrate 124. The inclination design was 20 °, the branching ratio was 93% transmission, and 7% reflection. For example, a multilayer film 126 of SiO ₂ , TiO ₂ , and Al ₂ O ₃ was formed on the quartz glass substrate 124 by an evaporation method. The quartz glass substrate 124 provided with the multilayer film 126 was cut into a size of 6 mm × 2 mm to form a chip. The chip-formed substrate 124 was polished to 25 μm and thin plate processing was performed.
Thereafter, the branch member 20 was inserted into the slit 18, an ultraviolet curable adhesive was applied, and the branch member 20 was mounted by being cured by ultraviolet irradiation.
The design of the PD array 28 was as follows. First, the number of channels of the PD array 28 was 12 ch, and the dimensions were a height of 150 μm, a width of 420 μm, and a length of 3 mm.
The structure of the PD array 28 is a back-illuminated type as in the optical device 10C according to the third embodiment. The upper part (on the submount 30 side) of the active layer 26 was filled with the anisotropic conductive material 120.
The reason why the back-illuminated PD array 28 is used is as follows. That is, for example, when a front-illuminated PD array 28 as shown in FIG. 3 is used, even if reflection on the back surface is suppressed, a slight amount of reflected light 24 reaches the front surface and reaches the active layer 26 of another channel. May recombine. On the other hand, in the case of the back-illuminated type, a part of the reflected light 24 on the front surface is absorbed by the active layer 26 of a predetermined channel in the immediate vicinity, so that the reflected light 24 is considerably reduced.
As shown in FIG. 7, a metal mask structure provided with a metal film 58 having a plurality of φ70 μm entrance windows 56 is employed on the incident surface side (front surface) of the back-illuminated PD array 28. The diameter of the entrance window 56 is preferably φ40 to 80 μm. Since crosstalk becomes a problem in the dimension of −30 dB, even a very small amount of light is incident on other channels. From this viewpoint, when the reflected light 24 is viewed, the adhesive in the slit 18, the end face of the slit 18, the core of the optical fiber 15, the cladding of the optical fiber 15, the adhesive layer 60 for fixing the PD array 28, etc. Since the PD array 28 is reached after passing through various states, it is affected by diffraction and scattering. Furthermore, slight reflections on the two end surfaces of the slit 18 and the back surface of the branch member 20 are combined therewith and reach the PD array 28. When such a slight component is taken into consideration, the light incident on the PD array 28 enters with a spread of about φ80 to 100 μm. If the light enters the PD array 28 with this spread, a part of the light protrudes from the active layer 26, causing stray light. Therefore, it is possible to improve crosstalk characteristics by restricting the opening on the incident surface side of the PD array 28. When the size of the incident window 56 is less than 40 μm, the aperture is excessively narrowed, so that the PD light receiving efficiency (light receiving efficiency in the active layer 26) is lowered. If a window of 80 μm or more is opened, the crosstalk characteristics deteriorate.
The reflected light 24 from the reflecting portion 33 is incident on the PD array 28 at an angle (an angle of 10 ° or more with respect to the vertical direction), so that the incident window is located at a position where the reflected light 24 is optimally incident on the active layer 26. Needless to say, 56 is provided.
The light receiving portion (active layer 26) of the back-illuminated PD array 28 was set to about 60 μm. The size of the light receiving portion (active layer 26) is preferably φ40 to 80 μm. If the thickness is less than 40 μm, the size of the light receiving portion (active layer 26) is too small, and there is a concern that the PD light receiving efficiency may decrease. This is because, when the thickness is 80 μm or more, stray light is easily picked up, so that the crosstalk characteristic may be deteriorated.
Further, as shown in FIG. 7, physical separation is performed by forming a physical groove 62 in the block between channels in the back-illuminated PD array 28. In general, a method of forming a p-type area between channels of an n-type substrate and forming a barrier between the channels is conceivable for such a block between channels. However, in this embodiment, the channel between the channels is more reliably separated by forming the physical groove 62 having a depth not less than the thickness of the active layer 26 having a depth of 2 to 20 μm around the light receiving area.
Further, as the mounting configuration of the submount 30, the configuration of the optical fiber 15-PD array 28-submount 30 was adopted. Although the configuration of the optical fiber 15-the submount 30-the PD array 28 can be taken, in this case, since the submount 30 exists between the optical fiber 15 and the PD array 28, the optical path length of the reflected light 24 becomes long. The spread of the reflected light 24 becomes large, which is not preferable from the viewpoint of PD light receiving efficiency and crosstalk.
In the case of the configuration of the optical fiber 15 -PD array 28 -submount 30, if the PD array 28 is in a surface incident state, wire bonding is required for conduction from the surface to the submount 30. In this case, a space of about 100 μm is required for wire bonding. This space needs to be filled with the adhesive layer 60 in terms of refractive index matching and reliability with the optical fiber 15 (quartz). That is, in the case of surface incidence, an adhesive layer 60 of 100 μm exists in the optical path, and this adhesive layer 60 causes instability in characteristics such as PDL and wavelength dependency. In addition, since a wire usually uses a metal such as Au, when light hits the wire, the light is scattered and becomes stray light, which causes deterioration of crosstalk.
In the case of back incidence, theoretically, the PD array 28 can be in contact with the optical fiber 15. Since the contact between the PD array 28 and the optical fiber 15 may cause a physical defect, the space of about 10 μm may be used as the adhesive layer 60 for safety.
The optical path lengths of these two are compared. Assuming that the active layer 26 exists on the substrate surface of the PD array 28 (the surface facing the optical fiber 15), the distance between the surface of the optical fiber 15 and the active layer 26 is 100 μm in the case of surface incidence. When the refractive index of the adhesive layer 60 is 1.45, which is the same as that of quartz, 100 / 1.45≈69 μm. In the case of backside incidence, if the thickness of the adhesive layer 60 is 10 μm and the thickness of a general PD array 28 is 150 μm, the optical path length is 10 / 1.45 + 150 / 3.5≈50 μm. It can be said that this is preferable from this point of view.
Furthermore, in the case of front-side incidence and back-side incidence, the incident angle of light to the active layer 26 is greatly different. In the case of surface incidence, even when the surface is coated with silicon nitride (refractive index 1.94), if the inclination angle α of the slit 18 is 20 °, the incident angle to the PD array 28 is about 35 °. . On the other hand, the back incident is 18.5 °, which is very small compared to the front incident, which is preferable from the viewpoint of PD light receiving efficiency and the like.
Next, the PD array 28 was mounted on the submount 30. As will be described later (see FIG. 10), the optical fiber array 16 side is mounted on the package 72 and the pin 74 of the package 72 and the electrode pad of the submount 30 are secured by wire bonding 76. As shown in FIG. 3, an Au electrode pattern 64 was formed on the lower surface of the submount 30. As a mounting form of the PD array 28, the PD array 28 is arranged on the lower surface of the submount 30, and the electrode pattern 64 is routed to the upper surface of the submount 30 through the through hole 66. Accordingly, the electrode pad 65 is formed on the upper surface of the submount 30 in accordance with each electrode pattern 64. The constituent material of the submount 30 was Al ₂ O ₃ .
The anisotropic conductive material 120 is thermocompression bonded so that a spherical polymer having a diameter of about 3 μm coated with a conductive substance such as Au in the material 120 is sandwiched only between the raised Au bump 122 and the electrode pattern 64, This provides conduction between the electrode pattern 64 and the Au bump 122. Further, although the spherical polymer in the anisotropic conductive material 122 remains under the entrance window 56 (see FIG. 7) formed in the PD array 28, the spherical polymer has a surface coated with a conductive material. Since the light emitted from the window 56 is scattered, it can be expected to prevent the light from returning to the PD array 28.
In addition, the antireflection film (SiN) 100 was coated on the lower surface of the submount 30 for the purpose of suppressing reflection due to the difference in refractive index on the portion corresponding to the active layer 26.
Next, alignment of the PD array 28 was performed. Specifically, first, as shown in FIG. 6, a spacer 32 for determining a gap between the optical fiber array 16 and the PD array 28 was attached to the submount 30.
The constituent material of the spacer 32 is borosilicate glass, in this case, in particular, Pyrex (registered trademark) glass material. The gap length was set to 10 μm. That is, since the thickness of the PD array 28 including the Au bump 122 is 190 μm, the spacer 32 is set to 200 μm.
Then, a necessary amount of the adhesive layer 60 was applied to the upper part of the optical fiber 15 that becomes the optical path of the reflected light 24. The alignment with the PD array 28 is such that light enters the channels at both ends of the optical fiber array 16 and the PD light receiving power of the reflected light 24 (light receiving power at the active layer 26 corresponding to the both end channels) is maximized. Active alignment was performed. At this time, the PD light receiving power was monitored while the output from the active layer 26 corresponding to the both-ends channel was applied to the submount 30 while the current value was observed. Thereafter, the PD array 28 was fixed to the optical fiber array 16 with ultraviolet rays. At this stage, the optical head 70 shown in FIG. 9 is completed.
Next, as shown in FIG. 10, so-called die bonding was performed in which the optical head 70 is fixed to the central portion of the package 72.
As the package 72, a 14-pin metal package was used. The external dimensions were 20 mm long x 12.5 mm wide. The optical head 70 was fixed to the package 72 with a thermosetting adhesive.
Next, wire bonding was performed between the plurality of pins 74 fixed on both sides of the package 72 and the electrode pads 65 (see FIG. 8) of the submount 30. The wire 76 was an Al-Si wire having a diameter of 20 μm. As for the wire condition, the 1st side was the pin 74 of the package 72, and the 2nd side was the electrode pad 65 of the submount 30.
Next, as shown in FIG. 11, a ring 78 is fixed so as to surround the optical head 70, and a boot 80 is fixed to a lead-out portion of the optical fiber array 16 in the optical head 70. Sealing with 82 was performed.
The ring 78 serves as a dam during resin sealing. The constituent material of the ring 78 was a stainless steel material. A resin molded product may be used from the viewpoint of cost reduction. The height of the ring 78 was about 4 mm. A thermosetting adhesive was used to fix the ring 78. The boot 80 is a rubber general-purpose boot fixed to the ring 78.
Si gel material was used for the sealing resin 82. This was potted so that the wire 74 was completely covered, and cured by ultraviolet irradiation and heat curing. In addition, if a purple plug in which the contact resistance between the electrode pad 65 (Au) and the wire 76 (Al) is increased, if the process temperature does not rise to about 150 ° C., the sealing resin 82 is It does not have to be. In order to reduce the cost, when reducing the resin 82, considering the moisture resistance, an amount in the range from the foot (crimped portion) of the wire 76 to about 1/3 of the height of the wire 76 is desirable.
Next, as shown in FIG. 12, the upper surface opening of the ring 78 was covered with a lid 84 and fixed. The lid 84 was a stainless steel plate. Of course, a resin molded product may be used from the viewpoint of cost reduction. The lid 84 was fixed with a thermosetting adhesive to obtain a finished product.
A completion inspection was performed on the inline-type power monitor module (the optical device 10Ca according to this example) completed through the above steps.
Each item was inspected for transmission side characteristics and branch side characteristics. For the transmission side characteristics, the characteristics of each channel were measured for insertion loss, polarization dependent loss, and wavelength dependence. As a result, the insertion loss <0.8 dB, the polarization dependent loss <0.05 dB, and the wavelength dependency <0.1 dB were obtained so that there was no problem in use.
Regarding the branching side characteristics, the characteristics of each channel were measured with respect to the PD light receiving efficiency, the polarization dependence of the PD light receiving efficiency, the wavelength dependence, and the crosstalk between channels on the PD array side. As a result, it was confirmed that the PD light receiving efficiency was 50 to 70 mA / W, the polarization dependency of the PD light receiving efficiency was <0.3 dB, and the wavelength dependency was <0.5 dB. Moreover, about crosstalk, it inspected as total crosstalk. That is, the sum of how much current has flown through the other channels in the state where any one of the 12 channels is lit, and the ratio of the current in this input channel to the sum of the currents in the other channels is expressed in 10 logs. did. As a result, it was confirmed that all the channels were -34 dB or less and showed extremely excellent characteristics.
The optical device according to the present invention is not limited to the above-described embodiment, but can of course have various configurations without departing from the gist of the present invention.

Claims (4)

A first substrate (12) having a V-groove (14) formed thereon;
One or more optical fibers (15) fixed to the V-groove (14) of the first substrate (12) and provided with a reflection function;
An optical element (28) fixed through an adhesive layer (60) on at least an optical path of reflected light (24) generated by the reflection function out of the cladding of the optical fiber (15);
A second substrate (30) for mounting the optical element (28) on the main surface (30a);
Of the first gap (110) between the main surface (30a) of the second substrate (30) and the optical element (28), at least the active layer (26) of the optical element (28) and the second element. There is no light reflecting material in the second gap (112) between the substrate (30) and the main surface (30a), and the refractive index of the second gap (112) is the refractive index of the surface of the optical element (28). And the second gap (112) is filled with an anisotropic conductive material (120) . The optical device according to claim 1 .
An optical device, wherein an electrode layer (106) is interposed in at least a portion of the first gap (110) excluding the second gap (112). The optical device according to claim 1 or 2 ,
An optical device characterized in that at least the second gap (112) is 20 μm. The optical device according to any one of claims 1 to 3 ,
An optical device, wherein an incident angle of the reflected light (24) to the optical element (28) is 10 ° or more with respect to a vertical direction.

Images