WO2018105712A1

WO2018105712A1 - Spot size converter and manufacturing method therefor

Info

Publication number: WO2018105712A1
Application number: PCT/JP2017/044102
Authority: WO
Inventors: 石榑　崇明; 和貴安原; 帆志彦戸田; 豊於
Original assignee: 学校法人慶應義塾
Priority date: 2016-12-07
Filing date: 2017-12-07
Publication date: 2018-06-14
Also published as: JP2018097012A

Abstract

The present invention provides a spot size converter that couples optical waveguides of different sizes with high efficiency and low loss, and has high axial misalignment resistance and design flexibility. A spot size converter coverts a beam diameter between optical wires of different core sizes, and comprises: a first surface optically connected to a first optical wire of a first size; a second surface optically connected to a second optical wire of a second size larger than the first size; and a tapered polymer waveguide the diameter of which conically decreases from the first surface toward the second surface.

Description

Spot size converter and manufacturing method thereof

The present invention relates to a spot size converter and a manufacturing method thereof, and more particularly to a polymer optical waveguide type spot size converter and a manufacturing method thereof.

In recent years, with the rapid development of cloud computing services, the amount of traffic in the data center has increased, and a data rate of 40 Gbps or more per lane is being demanded. In response to such demands, the use of optical interconnect technology, particularly silicon photonics technology to which a CMOS process is applied, is expected to further increase the bandwidth, density, and power consumption of the entire system.

In silicon photonics technology, which is expected to be used mainly in the on-chip region, a nanoscale optical waveguide can be fabricated by using silicon (Si) with a high refractive index for the core. On the other hand, in the off-chip region, it is assumed that an existing single mode optical fiber (SMF: Single Mode Fiber) is connected to a Si photonics chip. While the Si optical waveguide has a core size of 300 to 500 nm, the core size of general-purpose SMF is generally 8 to 10 μm. When both are directly butt-connected, coupling loss occurs due to the difference in mode field diameter (MFD: Mode Field Diameter). As a solution, a spot size converter (SSC: SpotＳＣSizeSConverter) is used. The SSC is suitable for space saving because in-plane connection is possible, and has an advantage of low wavelength dependency.

A configuration using a polymer waveguide has been proposed as a low-cost interface between an optical fiber and a Si wire waveguide (see Non-Patent Document 1, for example). In this document, the polymer waveguide is planar processed into a shape in which the waveguide width gradually decreases from the Si waveguide toward the SMF by photolithography.

In addition, a configuration in which a Si waveguide and a multicore fiber (MCF) are connected by photonic wire bonding has been proposed (for example, see Non-Patent Document 2). Photonic wire bonding is manufactured using a laser direct drawing method, and has a tapered structure with a diameter that expands toward the core of the MCF at the tip.

On the other hand, a technique has been proposed in which a discharge needle is inserted into an uncured clad, moved inside the clad while discharging the core material from the tip of the needle, and the core material is cured to form a polymer optical waveguide (for example, Patent Document 1). This method is also called “mosquito method”.

International Publication No. WO2013 / 002013

Although many SSC fabrications using photolithography and photonic wire bonding have been reported, SSCs that can be combined with general-purpose SMFs with high efficiency and low loss, yet have high tolerance for axial misalignment and high design freedom are still being realized. Absent.

It is an object of the present invention to provide a spot size converter that couples optical transmission lines of different sizes with high efficiency and low loss, and has high tolerance for misalignment and design freedom.

In order to solve the above-described problems, the present invention realizes a polymer optical waveguide type SSC having a truncated cone shape or a three-dimensional taper shape in which the radial size changes symmetrically with respect to the optical axis. In the first aspect of the present invention, a spot size converter for converting a beam diameter between optical wires having different core sizes is provided.
A first surface optically connected to a first optical wiring of a first size;
A second surface optically connected to a second optical wiring of a second size larger than the first size;
A tapered polymer waveguide whose diameter decreases conically from the first surface toward the second surface;
Have

In a second aspect of the present invention, a method for manufacturing a spot size converter is provided. This manufacturing method is
Forming an uncured cladding layer on the substrate on which the first optical wiring is formed;
While injecting an injection needle into the cladding layer and injecting an uncured core material from the injection needle into the cladding layer, the injection needle is changed in at least one of a moving speed and a discharge amount in a predetermined direction. A moving step,
Curing the core material after extracting the injection needle from the cladding layer at a predetermined position to form a tapered polymer waveguide having one end connected to the first optical wiring;
Optically connecting an end of the polymer waveguide opposite to the first optical wiring to a second optical wiring having a larger core size than the first optical wiring;
including.

The above configuration and method realizes a spot size converter that can couple optical waveguides of different sizes with high efficiency and low loss, and has high tolerance for axis deviation and design freedom.

It is the schematic of SSC of embodiment. It is a diagram showing a core diameter different cladding materials and (SSC diameter) LP ₀₁ mode of the MFD relationship. It is a figure which shows the simulation result of the optical coupling using the polymer optical waveguide type SSC of embodiment. It is a schematic diagram which shows the preparation methods of polymer optical waveguide type SSC of embodiment. It is a figure which shows the relationship between a needle scanning speed and a core diameter. It is a cross-sectional microscope picture of the side (a) and side (b) of the SSC sample of Example 1. It is a figure which shows the intensity | strength profile and MFD in the both sides of the SSC sample of Example 1. FIG. It is a figure which shows the result of FIG. 7 typically. It is a figure which shows the relationship between the core diameter of SSC sample of Example 1, and waveguide length. It is a figure which shows the normalized intensity distribution in each SSC length. It is a figure which shows the insertion loss of the SSC sample of Example 1 with a comparative example. It is a figure which shows the measurement result of the axial deviation tolerance of the SSC sample of Example 1. It is a figure which shows MFD corresponding to the cross-sectional microscope picture of the both sides of the SSC sample of Example 2. FIG. It is a figure which shows the insertion loss measurement result of the SSC sample of Example 2. It is a figure which shows the intensity | strength profile on both sides of the SSC sample of Example 2, and NFP and MFD. It is a figure which shows the insertion loss reduction effect of Example 2. FIG. It is a schematic diagram which shows the difference in a structure of solid-state taper-shaped SSC of embodiment, and the planar taper-shaped SSC by the photolithographic method as a comparative example. It is a figure which shows the simulation result of the intensity | strength profile of the solid taper-shaped SSC of embodiment, and the planar taper-shaped SSC of a comparative example, and a coupling loss.

FIG. 1 is a schematic diagram of a spot size converter (SSC) 10 according to an embodiment. The SSC 10 is a polymer optical waveguide type SSC that optically connects optical waveguides having different sizes. In the example of FIG. 1, the SSC 10 couples between a silicon (Si) optical waveguide 101 and a general-purpose SMF 102. As a feature of the present invention, the SSC 10 has a truncated cone shape or a tapered shape whose diameter decreases from the output end of the Si optical waveguide 101 toward the core of the SMF 102. The core size of the Si optical waveguide 101 is generally 300 to 500 nm, and the core diameter of the general-purpose SMF 102 is 8 to 10 μm. In the SSC 10, the diameter D1 of the cross section 11 at the emission position of the Si optical waveguide 101 is larger than the diameter of the cross section 12 on the connection side with the SMF 102. The shape of the SSC 10 may be referred to as “reverse taper” in the sense that the diameter of the SSC 10 increases on the Si optical waveguide 102 side having the smaller diameter.

As an example, the diameter of the SSC 10 at the output end of the Si optical waveguide 101 is 4 to 5 μm, and the diameter at the connection surface with the SMF 102 is 1.5 to 2.5 μm. The length L of the SSC 10 from the cross section 11 to the cross section 12 is 4.0 cm or less, preferably 1.0 to 3.5 cm. The SSC 10 may cover the end of the Si optical waveguide 101, but the MFD conversion function is performed between the cross section 11 and the cross section 12. In FIG. 1, only the lower clad 105 is shown for convenience of illustration, but the entire periphery of the Si optical waveguide 101 and the SSC 10 may be covered with the clad.

In the SSC 10 having the shape of FIG. 1, the mode field diameter (MFD) of the propagation light greatly expands toward the connection surface 12 with the SMF 102. For example, for light with a wavelength of 1.3 μm to 1.5 μm, the MFD of the SSC 10 at the exit position of the Si waveguide 101 is 3.8 to 3.9 μm, whereas the MFD at the connection surface with the SMF 102 The MFD extends to 5.6 to 8.0 μm. As will be described later, this is considered to be due to the effect of leakage of evanescent light. Since the MFD is widened on the connection surface with the SMF 102 by the truncated cone-shaped SSC 10 whose diameter changes symmetrically with respect to the optical axis, it has a high resistance to misalignment.

Before describing the characteristics and effects of the SSC 10 in detail, the design and fabrication of the polymer optical waveguide type SSC 10 will be described.
<Design and fabrication of polymer optical waveguide type SSC>
Optimal design is performed for the production of a reverse-tapered polymer optical waveguide type SSC having different MFDs at both ends. The MFD depends on the wavelength, the refractive index difference between the core and the clad, the core diameter, and the like. Therefore, first, a theoretical design of a polymer optical waveguide type SSC is performed using a mode solver tool (FIMMWAVE, FIMMPROP). A conceptual diagram of the SSC fabricated in practice is as shown in FIG.

SSC10 has a conical taper shape, and its cross section is substantially circular. The change in diameter in the optical axis direction is symmetric with respect to the optical axis. By gradually reducing the diameter toward the SMF 102 side, it is possible to suppress the generation of higher order modes.

As a polymer waveguide material constituting the SSC 10, an organic-inorganic hybrid resin SUNCONNECT (registered trademark) series manufactured by Nissan Chemical Industries, Ltd. is used. This material has high heat resistance and high compatibility with the solder reflow process. Further, the absorption loss at a wavelength of 1550 nm, which is assumed to be used in Si photonics technology, is low. Specifically, SUNCONNECT (registered trademark) series NP-005 (relative refractive index n _d = 1.60) is used as the core material. On the other hand, for the cladding material, the optimum cladding material is determined by simulation.

FIG. 2 shows the relationship between the core diameter and the MFD of the LP ₀₁ waveguide mode at a wavelength of 1550 nm. As a clad material, a material having a refractive index of 1.59 (plotted by a square) and a material having a refractive index of 1.52 (plotted by a circle) are compared. The cladding material with a refractive index of 1.59 is NP-211 manufactured by Nissan Chemical Industries, Ltd., and the cladding material with a relative refractive index of 1.52 is ORMOCLAD (registered trademark) manufactured by Micro Resist Technology GmbH. A material having a refractive index of 1.59 has a small difference in refractive index from the core, so that the MFD cannot be reduced to 7 μm or less regardless of how much the core diameter (SSC diameter) is reduced. On the other hand, since a material having a refractive index of 1.52 can secure a difference in refractive index from the core, when the core diameter (SSC diameter) is reduced to 3 μm, the MFD is reduced to 3.9 μm. When the core diameter (SSC diameter) is further reduced to about 2 μm, the MFD expands to 7 μm, which is equivalent to the general-purpose SMF. This is presumably because light is not sufficiently confined in the core (SSC) due to a significant reduction in the core diameter (SSC diameter) and leaks to the cladding as evanescent light.

From this simulation result, the diameter of the SSC 10 at the output end (side (a)) of the Si optical waveguide 101 is 4 to 5 μm, and the diameter of the SSC 10 at the connection end (side (b)) with the SMF 102 is 1.75 μm. By adopting the taper shape, the MFD can be converted from 3.8 μm to 8 μm.

FIG. 3 shows a result of a propagation simulation assuming that light is actually coupled from the Si optical waveguide 101 to the SSC 10 and further connected to a general-purpose SMF 102. The upper part is a top view, the middle part is NFP (Near Field Field Pattern), and the lower part is a side view. The tapered shape of the SSC is indicated by a white broken line.

The Si optical waveguide (WG) extends in parallel from the positions P1 to P2, and only the light confined in the Si optical waveguide is observed. From the positions P2 to P3, the Si optical waveguide is tapered, and light oozes out. The position P3 is a cross-sectional position of the SSC at the output end of the Si optical waveguide. This position is defined as side (a). The MFD on side (a) is 3.9 μm. Position 4 is the cross-sectional position of the SSC on the connection surface with the SMF. This position is defined as side (b). The MFD at side (b) is 6.2 μm. It can be seen that the LP ₀₁ mode MFD changes adiabatically on side (a) and side (b). As will be described later, the MFD can be further expanded by adjusting the manufacturing conditions of the SSC 10.

FIG. 4 is a schematic diagram for explaining a production process of a sample for evaluation of a polymer optical waveguide type SSC 10. First, as shown in FIG. 4A, a support body 41 having a removable frame 43 on a base 42 is prepared, and a clad material 44 is arranged in the frame 43. As the clad material 44, a material having a paste-like resin precursor having viscosity as a main component and having a large refractive index difference from the core material can be appropriately selected. As an example, the above-described ORMOCLAD (registered trademark) manufactured by Micro Resist Technology GmbH is applied in the frame 43 using a dispenser or the like.

Next, as shown in FIG. 4B, the core material (precursor or monomer before polymerization) is inserted from the tip of the needle 31 while the main body 32 is moved by inserting the needle 31 of the discharge device 30 into the clad material 44. The core layer 45 is formed in the cladding material 44 by implantation. The diameter of the core layer 45 formed by accelerating the moving speed of the main body 32 can be continuously reduced. The degree of change in the diameter of the core layer 45 can be controlled by adjusting the acceleration. When the core layer 45 having the desired length and diameter change is formed, the needle 31 is extracted from the cladding material 44. The direction of the needle speed change and the direction of movement are arbitrary, and the taper shape can be produced by accelerating or decelerating the needle movement speed. Further, the scanning direction of the needle is not limited to a plane parallel to the substrate, and scanning may be performed in the vertical direction. Furthermore, the taper shape may be realized by changing the discharge amount (or discharge pressure) of the core while keeping the scanning speed constant, or by combining the change of the moving speed and the discharge amount.

Next, as shown in FIG. 4C, the core layer 45 and the clad material 44 are cured. In this example, an ultraviolet curable resin is used for the cladding material 44 and the core material and is cured by irradiation with ultraviolet rays, but a thermosetting resin can also be used.

Finally, as shown in FIG. 4D, the frame 43 is removed, and the cured layer is peeled off from the base 42 to obtain a sample of the truncated cone in the clad 105 or the three-dimensional tapered SSC 10. It is done. In actual production of the SSC 10, it is not necessary to use the support 41. For example, a slit or a recess is formed in an optical wiring substrate on which an Si optical waveguide is formed, and a clad material is applied in the slit (or recess). The needle is inserted into the clad material before polymerization in the vicinity of the end of the Si optical waveguide, and the needle is moved while being accelerated in a predetermined direction while injecting the core material. Thereafter, the SSC can be formed by removing the needle, curing the core material and the clad material, and polishing the end face.

In this method, when the core material is injected into the uncured clad material 44, the core material isotropically diffuses into the clad material, and a GI (Graded Index) type refractive index distribution having a Gaussian distribution shape is obtained. The core layer 45 can be formed. A general GI-type refractive index profile is connected to a clad having a low refractive index and a uniform refractive index while the refractive index decreases from the core center and the profile remains convex upward. On the other hand, the refractive index profile of the waveguide according to the mosquito method of the present invention changes from convex upward to convex (having an inflection point), and has a Gaussian distribution connected to the clad by pulling the tail. Profile. Since the speed of light is inversely proportional to the refractive index, the optical path length of the light passing through the outside of the cured core while meandering (refracting) and the optical path length of the light passing through the center of the core cancel each other, and incident light is emitted almost simultaneously. can do.

FIG. 5 is a graph showing the relationship between the scanning speed of the needle 31 and the core diameter. Prior to fabrication of the truncated cone or tapered SSC 10 of FIG. 4, a linear core having a constant diameter was fabricated, and core diameter control was examined. Using a needle having an inner diameter of 80 μm, the discharge pressure of the core material was fixed at 50 kPa, and the results of measuring the core diameter while changing the scanning speed of the needle 31 each time were plotted. From FIG. 5, the core diameter tends to be inversely proportional to the 1/2 power of the scanning speed of the needle. It can be confirmed that the core diameter can be reduced to about 2 μm when the scanning speed of the needle 31 is 100 mm / s.

Based on the results of FIGS. 2 and 5, the SSC sample having a truncated cone shape or a three-dimensional taper shape was produced by accelerating the needle scanning speed from 8 mm / s to 40 mm / s within 100 milliseconds (ms). The length of the SSC sample in the optical axis direction is 3.5 cm. The needle inner diameter is 80 μm, and the discharge pressure of the core material is 50 kPa.

FIG. 6 is a cross-sectional micrograph of side (a) and side (b) of the produced SSC sample. The scanning speed of the needle 31 is accelerated from the side (a) to the side (b). The core diameter (SSC diameter) at side (a) is 4.6 μm, whereas the diameter is reduced to 2.8 μm at side (b). By arranging the side (a) on the Si optical waveguide side and the side (b) on the SMF side, the MFD can be enlarged toward the side (b). Optical characteristics are evaluated using this SSC sample.

FIG. 7 shows the intensity profile, NFP, and MFD of the SSC sample of Example 1 at a wavelength of 1550 nm. The solid line in FIG. 7A is an intensity profile at side (a) of the SSC sample, and shows the corresponding NFP and MFD. A broken line in FIG. 7A is an intensity profile of SMF (UHNA) having an ultra-high NA as an alternative to the Si optical waveguide, and shows the corresponding NFP and MFD. The solid line in FIG. 7B is an intensity profile on the side (b) of the SSC sample, and shows the corresponding NFP and MFD. The broken line in FIG. 7B is a general-purpose SMF intensity profile, and shows the corresponding NFP and MFD.

FIG. 7 shows that the MFD changes greatly from 3.8 μm to 5.6 μm at both ends of the SSC without the occurrence of higher-order modes. In FIG. 7A, the SSC side (a) and the intensity profile of UHNA are very close, and the MFD is also close. Similarly, in FIG. 7B, the intensity profiles of the SSC side (b) and the general-purpose SMF are very similar, and the MFD is also approaching.

FIG. 8 is a diagram schematically showing the evaluation result of FIG. It can be seen that the MFD is well matched on both the large-diameter side and the small-diameter side of the solid tapered SSC sample, and insertion loss (including coupling loss and propagation loss) is reduced.

FIG. 9 is a diagram showing the evaluation results of the light propagation characteristics with respect to the optical axis direction of the SSC sample of Example 1. First, NFP (Near Field Field Pattern) at a wavelength of 1550 nm at both ends of an SSC sample having a length of 3.5 cm is measured, and MFD is calculated. Thereafter, the SSC sample is shortened by 1 cm, and the MFD is calculated in each case where the length of the sample is 3.5 cm, 2.5 cm, 1.5 cm, and 0 cm.

As the length increases from the large diameter side (a) to the small diameter side (b) of the SSC sample, the core diameter in the cross section becomes 4.7 μm. It decreases to 3.9 μm, 3.2 μm, and 2.9 μm. In contrast, NFP at 1550 nm gradually increases. Similarly, MFD increases to 3.7 μm, 4.5 μm, 5.3 μm, and 5.7 μm.

This is considered to be because, as described above, as the SSC diameter decreases, the light that is no longer confined in the SSC leaks into the cladding as evanescent light. Thus, it can be seen that the SSC sample of Example 1 manufactured by the method of FIG. 4 can realize the spot size conversion function as designed.

FIG. 10 is a diagram showing an intensity profile at each SSC length. The tendency of the intensity profile is the same for each SSC length, and the MFD at the position where the intensity falls from the peak to e ⁻² increases as the SSC length increases.

FIG. 11 shows the evaluation results of the loss / loss characteristics using the SSC sample of Example 1. The insertion loss at a wavelength of 1550 nm is measured using the produced 3.5 cm long SSC sample. Two types of arrangements (configuration 1 and configuration 2) of the excitation probe on the LD side and the light receiving probe on the power meter side are prepared. In both arrangements, a butt connection is made between SMF and SSC, and between SSC and UHNA, and no matching oil is used.

In the configuration 1 of FIG. 11A, the UHNA serving as the excitation probe is connected to the side (a) having a small MFD, and the light from the side (b) having a large MFD is received by the general-purpose SMF. Configuration 1 assumes the propagation of signal light from the Si optical waveguide to the SMF in the arrangement configuration of the present invention.

In the configuration 2 in FIG. 11B, the general SMF is connected to the side (a) having a small MFD, the UHNA is connected to the side (b) having a large MFD, and the connection relationship is reversed from that of the configuration 1. Receive light on the side. The configuration 2 is an arrangement generally employed when optically connecting optical wirings having different diameters.

As shown in FIG. 11C, the insertion loss of the SSC in configuration 1 is sufficiently low. The coupling loss between the SM optical waveguide and the SMF is theoretically determined by the overlap integral of both electromagnetic field distributions. As can be seen from the intensity profile in FIG. 7, it is considered that the intensity profiles at both ends of the SSC greatly overlap with the intensity profiles of the excitation probe and the light receiving probe. Propagation loss at a wavelength of 1.55 μm of the polymer optical waveguide material used in the example (“SUNCONNECT (registered trademark)”) is 0.45 dB / cm, so that the propagation of a 3.5 cm long optical waveguide at 1550 nm is performed. The loss is estimated at 1.58 dB. The coupling loss at both ends of the SSC sample is estimated to be 5.36 dB including Fresnel reflection.

In contrast, in the configuration 2 in which the side (a) having a small MFD is connected to the large-diameter SMF and the side (b) having a large MFD are connected to the small-diameter UHNA, the insertion loss is large.

FIG. 12 shows the evaluation results of misalignment tolerance using the SSC sample of Example 1. The solid taper-shaped SSC of the embodiment has a high MFD on the connection side with the SMF, and thus is considered to have a high resistance to misalignment when connected to the SMF. Therefore, the produced SSC sample is used to evaluate the resistance to misalignment with SMF.

12 (A) and 12 (B) show the configuration of the measurement system, and FIG. 12 (C) shows the misalignment measurement result. A laser diode (LD) having a wavelength of 1550 nm is used as the light source, and a general-purpose SMF is used for each of the excitation probe and the light receiving probe. In FIG. 12 (A), the side (a) having a small MFD is connected to the SMF of the light receiving probe, and the loss change amount when the position of the light receiving side SMF is changed in the radial direction of the SSC is measured as a misalignment tolerance. . In FIG. 12B, the side with the large MFD (b) is connected to the SMF of the light receiving probe, and the amount of loss change when the position of the light receiving side SMF is changed in the radial direction of the SSC is measured as misalignment tolerance. .

As shown in FIG. 12C, the -0.5 dB misalignment tolerance on the side (a) was ± 1.3 μm, whereas the side (b) has a wide tolerance width of ± 2.1 μm. It is done. This is a value equivalent to the misalignment tolerance between the general-purpose SMF and the SM optical waveguide. The reason why such a wide tolerance width is obtained is considered that the MFD on the side (b) of the three-dimensionally tapered SSC sample can be brought close to the MFD of the SMF. As described above, the solid tapered SSC sample of Example 1 has a high resistance to misalignment.

FIG. 13 shows a schematic configuration and a cross-sectional micrograph of the SSC sample of Example 2, and NFP and MFD of propagating light. In the second embodiment, the MFD is optimized by increasing the rate of change in the scanning speed of the needle 31. The inner diameter of the needle is 80 μm, the discharge pressure of the core material is fixed at 50 kPa, the needle scanning speed is accelerated from 8 mm / s to 100 mm / s in 100 milliseconds (ms), and the length in the optical axis direction is 1. A SSC sample having a frustoconical shape of 5 cm or a solid taper shape was prepared.

The core diameter on the side (a) of the SSC sample of Example 2 is 4.64 μm, and the core diameter on the side (b) is 2.78 μm. The MFD on the side (a) is enlarged to 3.8 μm, and the MFD on the side (b) is enlarged to 7.2 μm. Since the MFD of the general-purpose SMF is 7.4 μm, it was confirmed that the desired MFD can be achieved by changing the needle scanning speed.

FIG. 14 shows the evaluation results of the loss loss characteristics using the SSC sample of Example 2. The insertion loss at a wavelength of 1550 nm is measured using the produced 1.5 cm long SSC sample. Two types of arrangements of an excitation probe on the LD side and a light receiving probe on the power meter side are prepared.

In configuration 1 in FIG. 14A, the UHNA of the excitation probe is connected to the side (a) having a small MFD, and light from the side (b) having a large MFD is received by the general-purpose SMF. Configuration 1 assumes the propagation of signal light from the Si optical waveguide to the SMF in the arrangement configuration of the embodiment.

In the configuration 2 in FIG. 14B, the general SMF is connected to the side (a), the UHNA is connected to the side (b), and the UHNA side receives light by reversing the connection relationship with the configuration 1. The configuration 2 is a configuration generally employed when connecting optical wirings having different diameters with SSC.

As shown in FIG. 14C, the insertion loss of the SSC in the configuration 1 is further reduced as compared with the first embodiment. On the other hand, in the configuration 2 in which the large diameter SMF is connected to the side (a) having a small MFD and the small diameter UHNA is connected to the side (b) having a large MFD, the insertion loss is large.

FIG. 15 shows intensity profiles, NFP and MFD on both sides of the SSC sample of Example 2. The solid line in FIG. 15A is the intensity profile at side (a) of the SSC sample, and shows both the corresponding NFP and MFD. The broken line in FIG. 15A is an intensity profile of UHNA as an alternative to the Si optical waveguide, and shows the corresponding NFP and MFD. The solid line in FIG. 15B is the intensity profile on the side (b) of the SSC sample, and shows the corresponding NFP and MFD. The broken line in FIG. 15B is a general-purpose SMF intensity profile, and shows the corresponding NFP and MFD.

From FIG. 15, it can be seen that in the SSC of Example 2, the MFD greatly changes from 3.9 μm to 7.2 μm between both ends without the occurrence of the higher-order mode. In FIG. 15A, the SSC side (a) and the UHNA intensity profile are very close, and the MFD is also close. In FIG. 15B, the intensity profiles of the SSC side (b) and the general-purpose SMF are almost the same, and the MFD is also very similar.

FIG. 16 is a diagram illustrating the insertion loss reduction effect of the configuration of the second embodiment. The chart on the left side of the figure shows the insertion loss of the 3.5 cm long SSC sample of Example 1, and the chart on the right side shows the insertion loss of the 1.5 cm long SSC sample of Example 2. When the insertion loss is represented by the sum of the propagation loss, the Fresnel reflection loss, and the coupling loss, the Fresnel reflection loss is the same in the first and second embodiments. The propagation loss is reduced to about half by shortening the length of the SSC in the optical axis direction.

The greatest reduction effect is coupling loss. In the second embodiment, a coupling loss of 1 dB or less is realized. This is considered to be due to optimization of the overlap of intensity profiles on both side (a) and side (b).

In addition, the SSC sample of Example 2 has a high tolerance against misalignment in the same manner as in Example 1 in which the change in diameter in the optical axis direction is symmetric with respect to the optical axis in a plane perpendicular to the optical axis. Further, due to the coincidence between the intensity profile on the side (b) on the SMF side and the MFD, a wider tolerance width than that of the first embodiment can be obtained on the side (b) side.

FIG. 17 is a schematic diagram showing a difference in configuration between the solid tapered SSC of the embodiment and a planar tapered SSC by a photolithography method as a comparative example. As shown in FIG. 17A, in the embodiment, an SSC having a Gaussian refractive index distribution is realized by the method of FIG. 4 (mosquito method).

On the other hand, when a tapered polymer optical waveguide type SSC is manufactured by a general photolithography method as shown in FIG. 17B, only the width direction is reduced, and a flat taper type SSC having a constant height is obtained. Become. In the photolithography method, a core having a uniform refractive index is obtained, and it is difficult to obtain a Gaussian refractive index distribution.

FIG. 18 is a diagram showing a comparison result between the solid tapered SSC of the embodiment and the planar tapered SSC according to the conventional method. FIG. 18A shows a simulation result of the intensity profile and coupling loss of the SSC produced by the method of the embodiment. FIG. 18B shows a simulation result of an intensity profile and coupling loss of an SSC manufactured by a general photolithography method.

In FIG. 18 (A) and FIG. 18 (B), focusing on the MFD, the SFD side (b) and the SMF have the same MFD size. Looking only at this result, the effect of aligning the size of the MFD to the SMF on the side (b) side of the SSC appears to be the same.

However, when the entire intensity profiles are compared, in the embodiment, the tendency of the intensity distribution is almost the same between SSC and SMF, whereas in the conventional method, the tendency of the intensity distribution is reversed at the MFD. This difference appears as a difference in the effect of reducing the coupling loss. According to the method of the embodiment, a GI type SSC (in particular, having a Gaussian type refractive index distribution) is obtained, and the area where the light intensity distributions overlap is increased, so that the coupling loss can be reduced. On the other hand, according to the conventional photolithography method, an SI (Step Index) type distribution having a constant refractive index in the radial direction of the SSC is obtained. The SSC intensity distribution and the SMF intensity distribution characteristic are reversed with the MFD as a boundary, and the area where the light intensity distribution overlaps is reduced, resulting in an increase in coupling loss. Thus, the superiority of SSC produced by the mosquito method of FIG. 4 was confirmed.

In summary, in the solid taper (or truncated cone) SSC produced by the method of the embodiment, the MFD greatly changes in adiabatic manner between both ends. In particular, in Example 2, the MFD is changed from 3.9 μm to 7. Enlarged to 2 μm. In a non-cured clad, the needle is accelerated from a speed of 8 mm / s or less to 40 to 100 mm / s within a predetermined time, thereby forming a desired three-dimensional taper-shaped polymer waveguide to leak evanescent light. The effect of taking out can be achieved.

The misalignment tolerance between the SSC of the embodiment and the general-purpose SMF is ± 2.1 μm or more, and a value equivalent to the axis deviation characteristic of the SM optical waveguide and the general-purpose SMF can be obtained.

The insertion loss of the SSC of the embodiment is sufficiently low, and particularly in Example 2, it can be suppressed to a low value of 2.34 dB at a wavelength of 1550 nm.

The SSC of the embodiment and the manufacturing method thereof can couple optical transmission lines of different sizes with high efficiency and low loss, and have high tolerance for misalignment and high design freedom. Such SSC can also be applied to multi-channel optical communication. For example, the SSC formed of the polymer waveguide of the embodiment is arranged corresponding to each channel on the edge of a silicon chip where Si optical waveguides such as 4 channels and 8 channels are formed at the transmission / reception front end of the optical transceiver. It can be connected to SMF which is an external optical wiring. The SSC of the embodiment can realize optical coupling with very high efficiency and low loss, particularly when applied to an optical transmission front end.

This application is based on Patent Application No. 2016-238038 filed with the Japan Patent Office on December 7, 2016, and includes the entire contents thereof.

10 SSC (spot size converter)
11 Cross section (first surface)
12 Cross section (second surface)
14 Polymer waveguide 30 Discharge device 31 Needle 44 Uncured clad material 45 Core layer

Claims

In a spot size converter that converts the beam diameter between optical wires of different core sizes,
A first surface optically connected to a first optical wiring of a first size;
A second surface optically connected to a second optical wiring of a second size larger than the first size;
A tapered polymer waveguide whose diameter decreases conically from the first surface toward the second surface;
A spot size converter characterized by comprising:
2. The diameter of the polymer waveguide on the first surface is 4 to 5 μm, and the diameter of the polymer waveguide on the second surface is 1.8 to 2.8 μm. The listed spot size converter.
The mode field diameter at the first surface at a wavelength of 1550 nm is 3.8 to 3.9 μm, and the mode field diameter at the second surface is 5.6 to 8.0 μm. The spot size converter according to 1 or 2.
The spot according to any one of claims 1 to 3, wherein the first surface of the polymer waveguide is connected to a silicon fine wire waveguide, and the second surface is connected to an optical fiber. Size converter.
5. The spot size converter according to claim 1, wherein the core portion of the polymer waveguide has a refractive index distribution having a Gaussian distribution shape.
Forming an uncured cladding layer on the substrate on which the first optical wiring is formed;
While injecting an injection needle into the cladding layer and injecting an uncured core material from the injection needle into the cladding layer, the injection needle is changed in at least one of a moving speed and a discharge amount in a predetermined direction. A moving step,
Curing the core material after extracting the injection needle from the cladding layer at a predetermined position to form a tapered polymer waveguide having one end connected to the first optical wiring;
Optically connecting an end of the polymer waveguide opposite to the first optical wiring to a second optical wiring having a larger core size than the first optical wiring;
A method for manufacturing a spot size converter.
The moving speed of the injection needle and / or the diameter of the polymer waveguide at the connection surface with the second optical wiring is smaller than the diameter of the polymer waveguide at the light exit position of the first optical wiring. Or the discharge amount is changed, The manufacturing method of the spot size converter of Claim 6 characterized by the above-mentioned.
8. The method of manufacturing a spot size converter according to claim 6, wherein the moving speed of the injection needle is accelerated from a speed of 8 mm / second or less to a speed of 40 to 100 mm / second within a predetermined time. .
The uncured cladding layer is formed on a substrate on which a silicon fine wire waveguide is formed,
The method for manufacturing a spot size converter according to any one of claims 6 to 8, wherein an end of the polymer waveguide opposite to the silicon fine wire waveguide is connected to an optical fiber.