阅读说明：本技术 光纤 (Optical fiber ) 是由 铃木雅人 川口雄挥 佐久间洋宇 田村欣章 长谷川健美 于 2020-03-10 设计创作，主要内容包括：根据本发明实施例的光纤具有用于使得能够在预制件阶段确定传输损耗的改善的结构。该光纤包括：包含Cl并且平均折射率低于纯石英玻璃的折射率的芯部；包含F的第一包层；第二包层；以及树脂涂层,其中,在波长为1550nm时的有效面积为135μm<Sup>2</Sup>以上且170μm<Sup>2</Sup>以下,有效面积与截止波长λ<Sub>C</Sub>的比率为85.0μm以上,在波长为1550nm且弯曲半径R为15mm时的LP01模的弯曲损耗为每10匝小于4.9dB,并且树脂涂层包括杨氏模量为0.3MPa以下的初级树脂层。(The optical fiber according to an embodiment of the present invention has an improved structure for enabling transmission loss to be determined at the preform stage. The optical fiber includes: a core containing Cl and having an average refractive index lower than that of pure silica glass; a first cladding layer comprising F; a second cladding layer; and a resin coating layer, wherein the effective area at a wavelength of 1550nm is 135 μm 2 Above 170 μm 2 Effective area and cut-off wavelength λ C Has a bending loss of less than 4.9dB per 10 turns of LP01 mode at a wavelength of 1550nm and a bending radius R of 15mm, and the resin coating includes a primary resin layer having a Young's modulus of 0.3MPa or less.)

1. An optical fiber, comprising:

a core including at least a region containing chlorine, and having an average refractive index lower than that of pure quartz glass;

a first cladding surrounding the core, the first cladding comprising at least fluorine, and the first cladding having a refractive index lower than the average refractive index of the core;

a second cladding surrounding the first cladding, the second cladding having a higher refractive index than the first cladding; and

a resin coating layer surrounding the second clad layer;

wherein the effective area A at a wavelength of 1550nm_effIs 130 μm²Above 170 μm²In the following, the following description is given,

the effective area A_effAnd cutoff wavelength lambda_CRatio (A) of_eff/λ_C) Is more than 85.0 mu m, and the grain size,

the bending loss of LP01 mode at a wavelength of 1550nm and a bending radius of 15mm is less than 4.9dB per 10 turns, and

the resin coating layer includes at least a primary resin layer having a Young's modulus of 0.3MPa or less.

2. The optical fiber according to claim 1, wherein said optical fiber,

wherein the second cladding layer is composed of pure silica glass or silica glass containing at least fluorine.

3. The optical fiber according to claim 1 or 2,

wherein the effective area A_effIs 135 μm²Above and 165 μm²The following.

4. The optical fiber according to any one of claims 1 to 3,

wherein the cutoff wavelength is 1630nm or less.

5. The optical fiber according to any one of claims 1 to 4,

wherein the ratio (A)_eff/λ_C) Is 95 μm or more.

6. The optical fiber according to any one of claims 1 to 5,

wherein the ratio (A)_eff/λ_C) Is 130 μm or less.

7. The optical fiber according to any one of claims 1 to 6,

wherein the mode field diameter of the LP01 mode at a wavelength of 1550nm is 12.5 μm or more and 14.0 μm or less.

8. The optical fiber according to claim 7, wherein said optical fiber,

wherein the bending loss of LP11 mode at a wavelength of 1550nm and a bending radius of 40mm is 0.10dB or more per 2 turns.

9. The optical fiber according to any one of claims 1 to 8,

wherein a difference between a first caustic radius defined as a caustic radius R of the LP01 mode at a wavelength of 1550nm and a bending radius of 25mm is 0.90 [ mu ] m or more and a second caustic radius_CThe second caustic radius is defined as the caustic radius R of the LP01 mode at a wavelength of 1550nm and a bend radius of 15mm_C。

10. The optical fiber according to any one of claims 1 to 8,

wherein R is_C,effAnd Δ D (%) satisfies the following relationship:

R_C,eff>1.46+ΔD(％)×1.93(1/％)，

R_C,effis the caustic radius R of the LP01 mode at a wavelength of 1550nm and a bend radius of 15mm_CAnd the mode field diameter of the LP01 mode at a wavelength of 1550nm, and Δ D (%) is the relative refractive index difference between the average refractive index of the first cladding and the maximum refractive index of the inner region in the second cladding.

11. The optical fiber according to any one of claims 1 to 9,

wherein the optical fiber has a refractive index profile satisfying the following relationship:

0.15≤Δn≤0.29；

0.02≤ΔD≤Δn+0.05；

2.0≤D/d≤3.7；

t is more than or equal to 2.55 and less than or equal to 3.05; and

-0.22≤ΔJ-0.056(μm^-1)×Δn×(D(μm)–d(μm)),

Δ n is a relative refractive index difference between an average refractive index of the core and a refractive index of the first cladding, Δ D is a relative refractive index difference between the refractive index of the first cladding and a maximum refractive index in an inner region of the second cladding, D is a radius of the core, D is an outer diameter of the first cladding, T is a ratio of the outer diameter of the second cladding to the outer diameter of the first cladding, and Δ J is a relative refractive index difference between the refractive index of the first cladding and a minimum refractive index in an outer region of the second cladding.

12. The optical fiber according to any one of claims 1 to 11,

wherein the resin coating further comprises a secondary resin layer surrounding the primary resin layer.

13. The optical fiber according to claim 12, wherein said optical fiber,

wherein the secondary resin layer has a Young's modulus of 800MPa or more.

14. The optical fiber according to claim 12 or 13,

wherein an absolute value of a refractive index difference between the primary resin layer and the secondary resin layer at a wavelength of 546nm is 0.15 or less.

15. The optical fiber according to any one of claims 1 to 15,

wherein an absolute value of a refractive index difference between the outer region of the second clad layer and the primary resin layer at a wavelength of 546nm is 0.08 or less.

Technical Field

The present disclosure relates to optical fibers.

The present application claims priority from japanese patent application No.2019-047245, filed on 3/14/2019, and all the contents described in this japanese patent application are incorporated herein.

Background

Patent document 1 (japanese patent application publication No.2014-238526), patent document 2 (japanese patent application publication No.2015-166853) and patent document 3 (japanese patent application publication No.2017-62486) disclose optical fibers having a W-type refractive index profile. The W-type refractive index distribution is realized by a core, a first cladding layer constituting a depressed cladding structure, and a second cladding layer. The refractive index of the first cladding is lower than that in the core, and the refractive index of the second cladding is lower than that in the core and higher than that in the first cladding.

In the manufacture of a preform for obtaining an optical fiber having such a W-type refractive index profile, a glass region to be a second clad layer is formed on an outer peripheral surface of a glass region to be a core and a first clad layer using a method such as rod-in-collapse method, Vapor Axial Deposition (VAD) method, Outside Vapor Deposition (OVD) method, or the like.

Disclosure of Invention

An optical fiber according to an embodiment of the present disclosure includes a core, a first cladding, a second cladding, and a resin coating. The core includes at least a region containing chlorine (Cl), and an average refractive index of the core is lower than that of pure quartz glass. The first cladding is arranged to surround the core. The first cladding layer contains at least fluorine (F), and the refractive index of the first cladding layer is lower than the average refractive index of the core. The second cladding is arranged to surround the first cladding, and the refractive index of the second cladding is higher than that in the first cladding. The resin coating is arranged to surround the second clad. In particular, the effective area A at a wavelength of 1550nm_effIs 130 μm²Above 170 μm²The following. Effective area A_effAnd cutoff wavelength lambda_CRatio (A) of_eff/λ_C) Is 85.0 μm or more. The bending loss of the LP01 mode at a wavelength of 1550nm and a bending radius R of 15mm is less than 4.9dB per 10 turns. The resin coating layer at least comprises a Young's modulus of 0.3MPa or lessA primary resin layer.

Drawings

Fig. 1 is a diagram showing an example of a cross-sectional structure of an optical fiber;

fig. 2A is a diagram showing an example of the refractive index profile of an optical fiber;

fig. 2B is a diagram showing another example of the refractive index profile of an optical fiber;

fig. 3A is a table summarizing the specifications of optical fibers of samples 1 to 13 according to the present embodiment;

fig. 3B is a table summarizing the bending losses of the optical fibers of samples 1 to 13 according to the present embodiment;

FIG. 4A is a table summarizing the specifications of optical fibers according to comparative examples 1 to 11;

FIG. 4B is a table summarizing the bending losses of optical fibers according to comparative examples 1 to 11;

FIG. 5 is a graph showing the increase in transmission loss at a wavelength of 1550nm (dB/km) and A based on the transmission loss of sample 1_eff/λ_C(μm) a graph of the relationship between;

FIG. 6 is a graph showing a relationship between an increase in transmission loss at a wavelength of 1550nm (dB/km) and Δ D (%) based on the transmission loss of sample 1;

FIG. 7 is a graph showing a relationship between an increase in transmission loss at a wavelength of 1550nm (dB/km) and Δ P (%) based on the transmission loss of sample 1;

FIG. 8 is a graph showing the bending loss (dB/10 turns) of LP01 mode at a wavelength of 1550nm with the bending radius R set to 15mm and A_eff/λ_C(μm) a graph of the relationship between;

FIG. 9 is a graph showing an equivalent refractive index profile of an optical fiber having a certain bending radius;

FIG. 10 is a graph showing each parameter of the optical fiber;

FIG. 11 shows R_C,effA graph of the relationship between (R15 mm, λ 1550nm) and Δ D (%);

FIG. 12 shows R_CA graph of the relationship between (R15 mm, λ 1550nm) (μm) and the outer diameter ratio T (a.u.);

fig. 13 is a graph showing a relationship between Δ J (%) and Δ n × (D-D) (%, μm);

FIG. 14 is a table summarizing preferred ranges and more preferred ranges for each parameter of an optical fiber;

fig. 15 is a graph showing examples of various refractive index profiles suitable for the core 10;

fig. 16 is a graph showing examples of various refractive index profiles suitable for the first cladding layer 20; and

fig. 17 is a diagram showing examples of various refractive index profiles applied to the second cladding layer 30.

Detailed Description

[ problem ] to

The inventors of the present invention have found the following problems by examining the conventional optical fiber.

That is, in the preform manufacturing stage, in order to obtain an optical fiber having a W-type refractive index profile, providing a glass region to be the second cladding outside a glass region to be the first cladding using the VAD method or the OVD method can reduce costs as compared with the rod collapse method. On the other hand, the optical fiber obtained by drawing the preform has an increased refractive index inside the second cladding, resulting in the possibility of deterioration of the transmission loss of the optical fiber at the wavelength of the signal light. In addition, it is difficult to add sufficient fluorine to the inside of the second cladding (in the vicinity of the interface between the first cladding and the second cladding) by the VAD method or the OVD method, and the refractive index distribution inside the second cladding is deformed in the shape of a protrusion. The presence of the protrusions appearing in the refractive index profile promotes the retention of high-order modes in the optical fiber, leading to a problem of deterioration of transmission loss in the obtained optical fiber.

Further, patent document 1 describes that an increase in transmission loss can be suppressed by suppressing an increase in the relative refractive index difference Δ P of the protrusions appearing in the refractive index distribution. However, there is still a higher demand for low transmission loss. Since Δ P may vary in the longitudinal direction of the preform, optical fibers obtained from regions in the preform where Δ P is high will increase transmission loss (high productivity cannot be maintained). In addition, it is difficult to control Δ P with high accuracy by the VAD method or the OVD method. Therefore, in the conventional optical fiber manufacturing technique, Δ P may become large. When Δ P is large, as described above, a high-order mode tends to remain in the inner region (region corresponding to the protruding portion of the refractive index distribution) of the second cladding layer (deteriorating the transmission loss of the optical fiber at the signal light wavelength).

The present disclosure has been made to solve the above-mentioned problems, and an object thereof is to provide an optical fiber having the following structure: which is capable of judging the improvement of transmission loss at the preform stage as compared with the conventional optical fiber.

[ advantageous effects of the invention ]

As described above, according to the embodiments of the present disclosure, an optical fiber substantially improved in transmission loss compared to a conventional optical fiber can be obtained. In addition, since improvement in transmission loss can be judged at the preform stage, improvement in productivity of optical fibers can be expected.

[ description of embodiments of the invention ]

Hereinafter, embodiments of the present disclosure will be separately described.

(1) In one aspect, an optical fiber according to an embodiment of the present disclosure includes a core, a first cladding, and a second cladding constituting a W-type refractive index profile. In addition, the optical fiber further includes a resin coating layer integrally covering the core, the first cladding and the second cladding. The core comprises at least a region doped with Cl, and the average refractive index of the core is lower than the refractive index of pure quartz glass. The first cladding is arranged to surround the core. Further, the first cladding layer contains at least F, and the refractive index of the first cladding layer is lower than the average refractive index of the core. The second cladding is arranged to surround the first cladding, and the refractive index of the second cladding is higher than that in the first cladding. The resin coating is arranged to surround the second clad. In particular, the effective area A at a wavelength of 1550nm_effIs 130 μm²Above 170 μm²The following. Effective area A_effWith a cut-off wavelength (2m cut-off wavelength) lambda_CRatio (A) of_eff/λ_C) Is 85.0 μm or more. The bending loss of the LP01 mode at a wavelength of 1550nm and a bending radius R of 15mm is less than 4.9dB/10 turns. The resin coating layer includes at least a primary resin layer having a Young's modulus of 0.3MPa or less. Note that the above bending loss unit (dB/10 turns)) Refers to a loss value measured in a state where a mandrel having a predetermined bending radius R is wound by a desired number of turns (for example, 10 turns).

(2) In one aspect of the disclosure, the second cladding layer is preferably composed of pure silica glass or silica glass containing at least F. In particular, forming the second cladding layer with a pure quartz cladding can reduce manufacturing costs. In the present specification, in a structure having a second cladding layer composed of silica glass containing at least F, "inner region" and "outer region" of the second cladding layer are defined according to the shape of the refractive index distribution in the second cladding layer. Specifically, the "inner region" of the second cladding is a region including the vicinity of the interface between the first cladding and the second cladding, and is defined as a position having a first local maximum (refractive index peak) in the refractive index distribution in the radial direction of the optical fiber. Further, a local minimum position following the local maximum position of the refractive index distribution is defined as an interface between the "inner region" and the "outer region".

(3) In one aspect of the disclosure, the effective area A_effPreferably 135 μm²Above and 165 μm²The following. Since this case can suppress the nonlinear influence, the span length can be further increased.

(4) In one aspect of the disclosure, the cutoff wavelength is preferably 1630nm or less. In this case, multi-mode transmission in the communication wavelength band of the C-band or the L-band after cable molding can be prevented (single-mode transmission is enabled).

(5) In one aspect of the disclosure, the ratio (A)_eff/λ_C) The lower limit of (B) is preferably 85 μm or 95 μm. Further, the ratio (A)_eff/λ_C) The upper limit of (B) is preferably 120 μm or 130 μm. In this case, the ratio (A) in the optical fiber_eff/λ_C) The suitable range of (b) is preferably 85 μm or more and 120 μm or less, 85 μm or more and 130 μm or less, 95 μm or more and 120 μm or less, and 95 μm or more and 130 μm or less. Further, the ratio (A)_eff/λ_C) The upper limit of (b) may be 120 μm or 130 μm. In particular, in the ratio (A)_eff/λ_C) When the particle diameter is 95 μm or more, the particle diameter may beFurther reducing transmission losses. In addition, in the ratio (A)_eff/λ_C) When the thickness is 120 μm or less, the increase of macrobend loss can be suppressed. In addition, when the ratio (A)_eff/λ_C) 95 μm or more and 130 μm or less, it is possible to achieve suppression of macrobend loss increase, suppression of nonlinear influence, and prevention of multimode transmission in the C-band and L-band communication wavelength bands after cable molding.

(6) In one aspect of the present disclosure, the mode field (hereinafter referred to as "MFD") diameter of the LP01 mode at a wavelength of 1550nm is preferably 12.5 μm or more and 14.0 μm or less. This makes it possible to reduce the connection loss between a standard single mode fiber (hereinafter referred to as "SMF") and the optical fiber of the present disclosure, resulting in a reduction in span loss. Further, in an aspect of the present disclosure, the bending loss of the LP11 mode at a wavelength of 1550nm and a bending radius R of 40mm is preferably 0.10dB or more per 2 turns. In this case, even if the bending radius is likely to allow coupling between the high-order mode and the fundamental mode, the high-order mode is released quickly, resulting in suppressing loss of the fundamental mode due to coupling between the high-order mode and the fundamental mode.

(7) In an aspect of the disclosure, the difference between the first caustic radius and the second caustic radius is 0.90 μm or more. The first caustic radius is defined as the caustic radius R of the LP01 mode at a wavelength of 1550nm and a bend radius R of 25mm_C(25 mm, 1550nm), and a caustic radius Rc of the LP01 mode (15 mm, 1550nm) at a wavelength of 1550nm and a bending radius R of 15mm, of 0.90 μm or more. In this case, the bending loss can be controlled to a practical magnitude at a bending radius in actual use.

(8) In one aspect of the disclosure, R_C,effAnd Δ D (%) satisfies the following relationship:

R_C,eff>1.46+ΔD(％)×1.93(1/％)，

wherein R is_C,effIs a caustic radius R at a wavelength of 1550nm and a bending radius R of 15mm_C(R15 mm, λ 1550nm) (μm) to the mode field diameter of LP01 mode at a wavelength of 1550nm (hereinafter referred to as "MFD"), and Δ D (%) is the average refractive index of the first cladding layerAnd the maximum refractive index of the inner region in the second cladding.

Satisfying the above relationship makes it possible to reduce transmission loss and facilitate the design of the optical fiber regardless of the presence or absence of a refractive index peak in the inner region of the second cladding. In this specification, having a refractive index n₁And has a refractive index n₂Is defined by the following formula: | n₁ ²-n₂ ²|/2n₁ ². Refractive index n as denominator₁The refractive index of pure quartz glass can be approximated to be 1.45.

(9) In an aspect of the present disclosure, as a shape for achieving all of the above aspects, the W-type refractive index profile of the optical fiber preferably satisfies the following relationship:

0.15≤Δn≤0.29；

0.02≤ΔD≤Δn+0.05；

2.0(μm)≤D/d≤3.7；

t is more than or equal to 2.55 and less than or equal to 3.05; and

-0.22≤ΔJ-0.056(μm^-1)×Δn×(D(μm)–d(μm))，

where Δ n is a relative refractive index difference between the average refractive index of the core and the refractive index of the first clad, Δ D is a relative refractive index difference between the refractive index of the first clad and the maximum refractive index in the inner region of the second clad, D is a radius of the core, D is an outer diameter of the first clad, T is a ratio of the outer diameter of the second clad to the outer diameter of the first clad, and Δ J is a relative refractive index difference between the refractive index of the first clad and the minimum refractive index in the outer region of the second clad. From such a refractive index distribution, the above condition can be satisfied: r_C,eff>1.46+ Δ D × 1.93.93 (1/%), and the bending loss of LP01 at a wavelength of 1550nm and a bending radius R of 15mm can be adjusted to be less than 4.9dB/10 turns.

(10) In an aspect of the present disclosure, the resin coating layer may further include a secondary resin layer surrounding the primary resin layer. Specifically, in an aspect of the present disclosure, the young's modulus of the secondary resin layer is preferably 800MPa or more. In this case, microbending loss can be suppressed. In an aspect of the present disclosure, the absolute value of the refractive index difference between the primary resin layer and the secondary resin layer at a wavelength of 546nm is preferably 0.15 or less. In this case, an increase in transmission loss due to reflection at the interface between the primary resin and the secondary resin can be suppressed. Further, in an aspect of the present disclosure, an absolute value of a refractive index difference (average refractive index in the case where the refractive index of the outer region is changed in the radial direction) between the outer region of the second clad layer and the primary resin layer at a wavelength of 546nm is preferably 0.08 or less. In this case, it is also possible to suppress an increase in transmission loss due to reflection at the interface between the second cladding layer and the primary resin.

As described above, each aspect listed in [ description of embodiments of the invention ] applies to all the other aspects or all combinations of these other aspects.

[ details of embodiments of the invention ]

Specific examples of the optical fiber according to the present invention will be described in detail below with reference to the accompanying drawings. The present invention is not limited to these examples but is indicated by the scope of the claims, and is intended to include meanings equivalent to the claims and all modifications within the scope. Further, the same reference numerals are given to the same components, and a repetitive description will be omitted in the description of the drawings.

Fig. 1 is a diagram showing an example of a cross-sectional structure of an optical fiber according to the present embodiment. Namely, the optical fiber 100 includes: a core 10 extending along an optical axis AX (the optical axis AX passing substantially through the center of a cross section of the core 10); a first cladding 20 surrounding the core 10; a second cladding layer 30 surrounding the first cladding layer 20; and a resin coating layer surrounding the second clad layer 30. In the example of fig. 1, the resin coating includes: a primary resin layer 40 surrounding the second clad layer 30; and a secondary resin layer 50 surrounding the primary resin layer 40.

The core 10 is composed of quartz glass doped with a refractive index lowering agent such as F, and the refractive index of the quartz glass is adjusted to be lower than that of pure quartz glass (PS). Specifically, Cl is doped into at least a portion of the core 10. Due to this Cl doping, a tilt in the radial direction r is provided in the refractive index profile of the core 10. The first clad 20 is composed of F-doped quartz glass, and the average refractive index of the first clad 20 is adjusted to be lower than that of the core 10. The second cladding 30 is composed of pure silica glass or silica glass doped with F, and the refractive index of the second cladding 30 is adjusted to be higher than the average refractive index of the first cladding and lower than the average refractive index of the core 10. The first cladding 20 and the second cladding 30 having such a configuration form a depressed cladding structure. The depressed cladding structure enables single-mode propagation at signal light wavelengths and achieves low transmission loss.

Fig. 2A is a diagram showing an example of the refractive index profile of an optical fiber. Fig. 2B is a diagram showing another example of the refractive index profile of the optical fiber. In the refractive index profiles 150 and 160 shown in fig. 2A and 2B, respectively, the second cladding layer 30 is composed of F-doped quartz glass, and the remaining region of the second cladding layer 30 except in the vicinity of the interface between the first cladding layer 20 and the second cladding layer 30 is divided into the inner region 30A and the outer region 30B by the local maximum position and the local minimum position of the refractive index profiles 150 and 160.

In the refractive index distribution 150 shown in FIG. 2A, "Δ n_core(%) "is a relative refractive index difference between the average refractive index of the core 10 and the refractive index of pure silica glass (pure silica level, hereinafter referred to as" PS "). "d" is the radius (μm) of the core 10. "Δ n (%)" is a relative refractive index difference between the average refractive index of the core 10 and the average refractive index of the first cladding 20. "D" is the outer radius (μm) of the first cladding layer 20 (the interface position between the first cladding layer 20 and the second cladding layer 30). "Δ D (%)" is a relative refractive index difference between the average refractive index of the first cladding 20 and the maximum refractive index (refractive index peak) of the inner region 30A. "R-in" is the length (μm) of the inner region 30A in the radial direction R of the optical fiber 100. "Δ P (%)" is a relative refractive index difference (relative refractive index difference at the protrusion in the refractive index distribution) between the maximum refractive index of inner region 30A and the minimum refractive index of outer region 30B (local minimum of refractive index distribution 150). "Δ J (%)" is the average refractive index of the first cladding layer 20 and the minimum refractive index of the outer region 30BRelative refractive index difference between the indices.

As described above, in the refractive index distribution 150 shown in fig. 2A, the second cladding layer 30 is divided into the outer region 30B having a substantially uniform refractive index in the radial direction r, and the inner region 30A which exists inside the outer region 30B and has a refractive index higher than that in the outer region 30B. In the present specification, "substantially uniform" means that the refractive index variation in the radial direction r of the outer region 30B in the second cladding layer 30 is ± 0.01% or less with respect to the average value.

Meanwhile, in the refractive index distribution 160 shown in fig. 2B, the definition of the structural parameters of each portion is similar to the case of the refractive index distribution 150 shown in fig. 2A, whereas the distribution shape at the outer region 30B is different from the case of the refractive index distribution 150 in the refractive index distribution 160. That is, the refractive index distribution 160 has a shape having a concave portion in the radial direction r in the second cladding layer 30. In the refractive index distribution 160, a region inside the peak position of the concave portion (the position where the refractive index distribution 160 takes a local minimum value in the second cladding layer 30) is defined as an inside region 30A, and a side further outside than this region is defined as an outside region 30B. At this time, the relative refractive index difference between the maximum refractive index of the inner region 30A and the minimum refractive index of the outer region 30B is Δ P.

Next, the results of examining the relationship between the structural parameters and the transmission characteristics in the various optical fibers will be described.

Fig. 3A is a table summarizing the specifications of the optical fibers of samples 1 to 13 according to the present embodiment. Fig. 3B is a table summarizing the bending losses of the optical fibers of samples 1 to 13 according to the present embodiment. Fig. 4A is a table summarizing the specifications of the optical fibers according to comparative examples 1 to 11. Fig. 4B is a table summarizing the bending loss of the optical fibers according to comparative examples 1 to 11.

Items shown in fig. 3A and 4A are as follows. That is, the "increase in transmission loss at a wavelength of 1550nm (compared with sample 1)" is an increase in loss in each of the samples or comparative examples based on the transmission loss of sample 1 at a wavelength of 1550 nm. "MFD at wavelength 1550 nm" is the MFD at wavelength 1550 nm. "A at wavelength of 1550nm_eff"is effective at a wavelength of 1550nmArea. 'lambda' is a_C"is the 2m cut-off wavelength defined in ITU-T G.650.1. "MFD (wavelength 1550nm)/λ_CThe MAC value is MFD at 1550nm and a cut-off wavelength λ of 2m_CRatio (MAC value). "A" is_eff(wavelength 1550 nm)/lambda_C"is the effective area A_effAnd a 2m cut-off wavelength lambda_CThe ratio of (a) to (b). 'lambda' is a_CC"is the cable cut-off wavelength (22m cut-off wavelength) defined by ITU-T G.650.1. "MFD (wavelength 1550nm)/λ_CC"is MFD at 1550nm and cable cutoff λ_CCThe ratio of (a) to (b). "A" is_eff(wavelength 1550 nm)/lambda_CC"is the effective area A_effCut-off wavelength lambda of cable_CCThe ratio of (a) to (b). "Δ n" is a relative refractive index difference between the average refractive index of the core 10 and the average refractive index of the first cladding 20. "Δ D" is a relative refractive index difference between the average refractive index of the first cladding layer 20 and the maximum refractive index (refractive index peak) of the inner region 30A. "Δ P" is the relative refractive index difference between the maximum refractive index of inner region 30A and the minimum refractive index of outer region 30B (the local minimum of refractive index profile 150). "Δ J" is the relative refractive index difference between the average refractive index of the first cladding layer 20 and the minimum refractive index of the outer region 30B. "Δ J- Δ n" is the difference between Δ J and Δ n. "d" is the radius of the core 10. "D" is the outer radius of the first cladding 20. "D/D" is the ratio of the outer radius D of the first cladding 20 to the radius D of the core 10. "T" is the ratio of the outer radius of the first cladding layer 20 to the outer radius of the second cladding layer 30. "R-in" is the width of medial region 30A.

Items shown in fig. 3B and 4B are as follows. That is, the "bending loss of LP01 mode (R15 mm, λ 1550 nm)" is the bending loss of LP01 mode at a wavelength of 1550nm and a bending radius of 15 mm. "LP 01 mode bending loss (R25 mm, λ 1550 nm)" is the bending loss of the LP01 mode at a wavelength of 1550nm and a bending radius of 25 mm. "LP 11 mode bending loss (R ═ 40mm, λ ═ 1550 nm)" is the bending loss of the LP11 mode at a wavelength of 1550nm and a bending radius of 40 mm. "LP 01 mode R_C(R ═ 15mm, λ ═ 1550nm) "is the caustic radius of the LP01 mode at a wavelength of 1550nm and a bend radius of 15 mm. "LP 01 mode R_C(R25 mm. lambda. 1550nm) "isThe caustic radius of the LP01 mode at a wavelength of 1550nm and a bend radius of 25 mm. "LP 01 mode R_C(25 mm. lambda. 1550nm) -LP01 mode R_C(R ═ 15mm, λ ═ 1550nm) "is the difference between the caustic radius of the LP01 mode at a wavelength of 1550nm and a bending radius of 25mm and the caustic radius of the LP01 mode at a wavelength of 1550nm and a bending radius of 15 mm. "LP 01 mode R_C,eff(R ═ 15mm, λ ═ 1550nm) "is a value obtained by dividing the caustic radius of the LP01 mode at a wavelength of 1550nm and a bending radius of 15mm by the MFD of the LP01 mode at a wavelength of 1550 nm.

In each of samples 1 to 11 shown in FIGS. 3A and 3B, the effective area A at a wavelength of 1550nm_effIs 135 μm²Above 170 μm²Effective area A below_effAnd cutoff wavelength lambda_CRatio (A) of_eff/λ_C) 85.0 μm or more, and a bending loss of LP01 mode at a wavelength of 1550nm and a bending radius R of 15mm is less than 4.9dB/10 turns. In contrast, in each of comparative examples 1 to 10 shown in FIG. 4A and FIG. 4B, the bending loss in the LP01 mode at a wavelength of 1550nm and a bending radius R of 15mm exceeds 4.98dB/10 turns. In comparative example 11, effective area A_effAnd cutoff wavelength lambda_CRatio (A) of_eff/λ_C) Less than 85.0 μm.

With respect to the optical fiber 100 having the structural parameters and transmission characteristics as described above, the transmission loss at a wavelength of 1550nm and the effective area a by passing the LP01 mode at a wavelength of 1550nm will be described with reference to fig. 5_eff(μm²) Divided by a 2m cutoff wavelength λ_CValue A obtained by (. mu.m)_eff/λ_C(μm). The 2m cutoff wavelength is the fiber cutoff wavelength of the LP01 mode as defined in ITU-T G.650.1. Note that, in fig. 5, the vertical axis represents the increase in transmission loss (dB/km) at a wavelength of 1550nm based on the transmission loss of sample 1. The horizontal axis is A_eff/λ_C(μm) — in addition, the symbol "○" plotted in fig. 5 represents samples 1 to 13 in which the bending loss of the LP01 mode (hereinafter referred to as "LP 01 mode bending loss (R ═ 15mm, wavelength λ ═ 1550 nm)") at a wavelength of 1550nm and a bending radius R of 15mm is less than 4.9dB/10 turns, and the effective area a is_effAnd cuttingStop wavelength lambda_CRatio (A) of_eff/λ_C) 85.0 μm or more, symbol "△" represents comparative example 11 in which the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns, and the ratio (a)_eff/λ_C) Less than 85.0 μm. The symbol "□" represents comparative examples 1 to 10, in which the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is 4.9dB/10 turns or more.

As shown in FIG. 5, when the bending loss of LP01 mode (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns and the ratio A is_eff/λ_CAt 85.0 μm or more (symbol "○"), relative to the ratio A_eff/λ_CThe transmission loss increases more gradually in terms of variation than when the LP01 mode bending loss (R15 mm, λ 1550nm) is 4.9dB/10 turns or more (symbol "□"). Due to less transmission loss due to effective area A_effAnd λ_CDue to structural fluctuation in the longitudinal direction of the optical fiber, and therefore an optical fiber having a small change in transmission loss in the longitudinal direction can be manufactured.

Fig. 6 is a graph showing a relationship between an increase in transmission loss at a wavelength of 1550nm (dB/km) and Δ D (%) based on the transmission loss of sample 1 symbol "○" plotted in fig. 6 represents samples 1 to 7 and samples 10 to 12, in which LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns, and a ratio (a) is_eff/λ_C) 95.0 μm or more, symbol "△" represents comparative example 11 in which the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns, and the ratio (a)_eff/λ_C) Less than 85.0 μm. "◇" (open diamonds) represents samples 8, 9 and 13 where the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns and the ratio (a) is_eff/λ_C) Is 85.0 μm or more and less than 95 μm. The symbol "□" represents comparative examples 1 to 10, in which the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is 4.9dB/10 turns or more.

As shown in FIG. 6, when the bending loss of LP01 mode (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns and the ratio A is_eff/λ_CIs 85.0 μm or more (symbol "○")No. ◇ "), the increase in transmission loss with respect to the change in Δ D is more gradual than that when the LP01 mode bending loss (R15 mm, λ 1550nm) is 4.9dB/10 turns or more (symbol" □ "), that is, even when the F doping amount in the second cladding 30 is small (even when Δ D is large), the increase in transmission loss can be kept within a practically acceptable range (manufacturing cost can be reduced) — in addition, when the LP01 mode bending loss (R15 mm, λ 1550nm) is less than 4.9dB/10 turns and the ratio (a) is a (a) of_eff/λ_C) At 95.0 μm or more (symbol "○"), the increase in transmission loss (as compared with sample 1) can be suppressed to 0.002dB/km or less, regardless of the magnitude of Δ D.

Note that the symbol "○" plotted in fig. 7 represents the cases of samples 1 to 7 and samples 10 to 12, in which the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns, and the ratio (a) is (a) a_eff/λ_C) 95.0 μm or more, symbol "△" represents comparative example 11 in which the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns, and the ratio (a)_eff/λ_C) Less than 85.0 μm. "◇" (open diamonds) represents samples 8, 9 and 13 where the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns and the ratio (a) is_eff/λ_C) Is 85.0 μm or more and less than 95 μm. The symbol "□" represents comparative examples 1 to 10, in which the LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is 4.9dB/10 turns or more. Further, FIG. 8 shows the bending loss (dB/10 turns) and A of the LP01 mode at a wavelength of 1550nm and a bending radius R set to 15mm_eff/λ_C(μm) in the same manner as described above. Note that fig. 8 includes plots of samples 1 to 13 and comparative examples 1 to 11, although they are partially overlapped in the display.

As shown in fig. 7, when the LP01 mode bending loss (R15 mm, λ 1550nm) is less than 4.9dB/10 turns and the ratio (a)_eff/λ_C) At 95.0 μm or more (symbol "○"), the transmission loss can be increased regardless of the Δ P (see the equation:) ((iii))Compared with sample 1) to below 0.002 dB/km. In order to improve the signal-to-noise ratio in an optical transmission system using an optical fiber as a transmission path for transmitting signal light, the optical fiber is required to suppress nonlinearity and realize low loss. Therefore, has a large effective area A of the optical fiber_effSo that the nonlinearity of the optical fiber can be improved. On the other hand, it is known that the effective area A is too large_effMicrobending losses are increased. Therefore, the effective area A is preferably set_effSet to 130 μm²Above 170 μm²The following. More preferably, the effective area A_effSet to 135 μm²Above and 165 μm²The following steps. The 2m cut-off wavelength is preferably 1630nm or less. In this case, when the optical fiber is formed as a cable, it is possible to prevent multimode transmission from occurring in the C-band communication wavelength band and the L-band communication wavelength band.

Ratio (A)_eff/λ_C) Is a physical quantity related to a V parameter (V number) indicating the magnitude of optical confinement in the core, and therefore has a correlation with bending loss. As observed in FIG. 8, the bending loss as a function of the ratio (A)_eff/λ_C) Is increased. Therefore, preferably, the ratio (A)_eff/λ_C) It is preferably set to a value not too large, for example, 120 μm or less. More preferably, the ratio (A)_eff/λ_C) Is set to 110 μm or less, more preferably 105 μm or less. Note that the bending loss of the LP01 mode obtained at a wavelength of 1550nm and a bending radius R of 15mm is about 0.1dB/10 turns. In addition, the effective area A is increased_effDivided by a 22m cutoff wavelength λ_CCValue (. mu.m) obtained_eff/λ_CC) Set to 95 μm or more and 130 μm or less, it is possible to suppress nonlinearity and prevent multimode transmission of a communication wavelength band such as the C-band or the L-band. Here, the 22m cutoff wavelength is the cable cutoff wavelength of LP01 mode defined in ITU-T g.650.1.

By having the ratio (A) predicted in the preform state_eff/λ_C) And LP01 mode bending loss (R15 mm, λ 1550nm) values, so that preforms with increased transmission loss or preforms with a high probability of longitudinal variation of transmission loss can be selected prior to the drawing process. This is achieved bySo that the manufacturing cost can be reduced. It is known that measuring the refractive index distribution in the radial direction from the center of the preform at the completion of the preform and then performing numerical calculation by the Finite Element Method (FEM) based on the refractive index distribution will enable a_effAnd λ_CAnd (6) estimating. That is, the ratio (A) can be easily predicted at the preform stage_eff/λ_C). In addition, in the case where it can be predicted that the LP01 mode bending loss (R ═ 15mm, λ ═ 1550nm) will be 4.9dB/10 turns or more or less than 4.9dB/10 turns, the value of the transmission loss increase (compared to sample 1) or whether the transmission loss is likely to vary in the longitudinal direction of the optical fiber can be predicted using fig. 5. In particular, as described above, when the LP01 mode bending loss (R15 mm, λ 1550nm) is less than 4.9dB/10 turns and the ratio (a)_eff/λ_C) At 95.0 μm or more, the increase in transmission loss (as compared with sample 1) can be suppressed to 0.002dB/km or less, regardless of the magnitude of Δ P. With this configuration, it can be predicted whether or not the transmission loss increase (compared with sample 1) is 0.002dB/km or less before the drawing process even when Δ P is varied in the longitudinal direction of the preform. That is, it is possible to prevent a defective preform, which is expected to have a large increase in transmission loss, from being transferred to the drawing process. As a result, an increase in manufacturing cost can be suppressed.

Note that the ratio (A) is usually used_eff/λ_C) In the bending loss prediction of (a), as shown in fig. 8, the relative ratio (a) is present in the bending loss of LP01 mode (R15 mm, λ 1550nm)_eff/λ_C) A certain correlation of (a) and (b) is also largely changed, and thus it is not easy to predict. Regarding this problem, there is a value called the caustic radius as the ratio (A)_eff/λ_C) Compared to parameters that are physically more closely related to the bending loss of the fiber.

Fig. 9 is a graph showing an equivalent refractive index profile 151 for analyzing light propagation when a bend of a certain radius is applied to an optical fiber (having the refractive index profiles 150 and 160 shown in fig. 2A and 2B, respectively). In the equivalent refractive index distribution 151, the refractive index is high at each position corresponding to the outer side where the optical fiber is bent, and the refractive index is low at each position corresponding to the inner side. By using the equivalent refractive index, the behavior of light propagating in a bent fiber can be replaced with the behavior of light propagating in a straight fiber for analysis. In fig. 9, the effective refractive index level of the LP01 mode at a certain wavelength λ is also indicated by a dashed line. The caustic radius is the distance from the center of the fiber to: at this position, in the equivalent refractive index profile, in the fiber radius direction parallel to the bending radius of the optical fiber to which a certain radius bending has been applied, the equivalent refractive index and the effective refractive index are equal to each other.

Here, the effective refractive index n of the LP01 mode at wavelength λ_eff(λ) is a value obtained by dividing the propagation constant of the LP01 mode at the wavelength λ when the optical fiber is unbent by the wave number at the wavelength λ. Further, the equivalent refractive index profile n of the optical fiber is set_bend(R, λ, R, θ) is defined by the following formula (1):

where n (λ, r) is the refractive index profile in the cross section of the fiber at wavelength λ, and R (mm) is the bend radius.

Further, fig. 10 is a graph showing each parameter of the optical fiber. r (mm) is a distance from the fiber center position (a position intersecting the optical axis AX) to a certain point in the fiber cross section. A straight line connecting the center position of the bend radius and the center position of the optical fiber is defined as an x-axis, the center position of the optical fiber is defined as x 0, and a direction from the center position of the bend radius toward the center position of the optical fiber is defined as a forward direction. In this case, θ is an angle formed by a line segment connecting a certain point in the cross section of the optical fiber to the center position of the optical fiber and a half line defined by a region where x is 0 or more.

Hereinafter, the equivalent refractive index n of the optical fiber is expressed by a plurality of values on the x-axis (where θ is 0 (that is, in a region where x ≧ 0 is satisfied on the axis)_bend(R, λ, R, 0) is equal to the effective refractive index n of the LP01 mode_eff(λ)), some values on the x-axis satisfy the following formula (2):

n_bend(R,λ,0.95x＜r＜0.99x,0)＜n_bend(R,λ,1.01x＜r＜1.05x,0) (2)

these values are defined as the caustic radius R at wavelength λ when the fiber is bent at a bend radius R_C(R, lambda). In the presence of a plurality of such R_C(R, λ), the minimum value thereof will be adopted.

Note that, in the cross section of the optical fiber, light outside the caustic radius is emitted to the outside of the optical fiber, causing bending loss (see patent document 2).

FIG. 11 shows R_C,effA graph of the relationship between (R15 mm, λ 1550nm) and Δ D (%); note that R_C,effIs obtained by dividing a caustic radius R at a wavelength of 1550nm and a bending radius R of 15mm_C(R15 mm, λ 1550nm) divided by the mode field diameter of LP01 at a wavelength of 1550nm, the symbol "○" plotted in fig. 11 represents samples 1 to 13 and comparative example 11, in which LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns, and the symbol "□" represents comparative examples 1 to 10, in which LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is 4.9dB/10 turns or more, the broken line shown in fig. 11 shows R_C,eff(R＝15mm，λ＝1550nm)＝1.46+ΔD×1.93(1/％)。

When R is shown in FIG. 11_C,eff(R＝15mm，λ＝1550nm)>1.46+ Δ D × 1.93.93 (1/%), LP01 mode bending loss (R15 mm, wavelength λ 1550nm) is less than 4.9dB/10 turns_C,effWhen (R15 mm, λ 1550nm) is not more than 1.46+ Δ D × 1.93.93 (1/%), the bending loss of LP01 mode (R15 mm, wavelength λ 1550nm) is not less than 4.9dB/10 turns.

FIG. 12 shows R_C(R15 mm, λ 1550nm) (μm) and an outer diameter ratio T (a.u). Note that R_C(R15 mm, λ 1550nm) is the caustic radius at a wavelength of 1550nm and a bending radius R of 15mm, and the outer diameter ratio is the ratio of the outer radius of the second cladding 30 (the outer radius of the optical fiber 100) to the outer radius of the first cladding 20. Fig. 8 includes plots of samples 1 to 13 and comparative examples 1 to 11, although they are partially overlapped in the illustration.

Such as from the figureObserved at 12, R_CThere is a high correlation between (R15 mm, λ 1550nm) and the ratio T. This ratio T is a parameter that is substantially consistent with the ratio: i.e. the ratio of the outer diameter of the preform (the outer radius of the region corresponding to the second cladding layer 30) and the outer diameter (or outer radius) of the region corresponding to the first cladding layer 20 in the preform state. Thus, R_C(R ═ 15mm, λ ═ 1550nm) can be estimated from the refractive index distribution in the radial direction from the center of the preform at the completion of the preform.

Note that the MFD can be predicted by numerical calculation based on a Finite Element Method (FEM) of refractive index distribution. Therefore, it can be predicted whether the LP01 mode bending loss (R15 mm, λ 1550nm) at the completion of the preform is more than 4.9dB/10 turns or less than 4.9dB/10 turns.

Furthermore, in repeaters in undersea optical cable systems, ITU-T g.652 compliant single mode optical fibers are typically used as feed-throughs (feedthroughs). Therefore, when the MFD of the LP01 mode at a wavelength of 1550nm is 12.5 μm or more and 14.0 μm or less, the fusion loss with the ITU-T G.652-compliant single-mode optical fiber can be reduced, so that the span loss in the submarine optical cable system is reduced.

Further, high-order modes tend to remain in the protrusions of the refractive index profile of the second cladding layer 30 corresponding to the inner region 30A, and therefore, the transmission loss increase is considered to be caused by the interaction between LP01 as a fundamental mode and the high-order modes. The magnitude of the LP01 mode bending loss (R15 mm, λ 1550nm) is believed to be related to the effective refractive index difference between the LP01 mode and the higher order modes. Therefore, reducing the LP01 mode bending loss (R15 mm, λ 1550nm) will increase the effective index difference between the LP01 mode and the higher order modes. This makes it possible to reduce the coupling coefficient from the LP01 mode to a higher-order mode even when the protrusion is large. Accordingly, it is considered that an increase in transmission loss can be suppressed. Further, when the bending loss (R ═ 40mm, λ ═ 1550nm) of the LP11 mode is 0.10dB/2 turns or more, even if light is coupled from the LP01 mode to the high-order mode, the high-order mode light is immediately emitted to the outside of the optical fiber (because of attenuation), so that the interaction between the LP01 mode and the high-order mode can be suppressed. Preferably, the bending loss (R40 mm, λ 1550nm) of the LP11 mode is 0.50dB/2 turns or more, and more preferably 1.00dB/2 turns or more.

When an optical fiber is actually used in an undersea optical fiber system, the bending diameter is set to be small, and is 50mm or more (patent document 2 described above). When R is_C(R＝25mm，λ＝1550nm)-R_CWhen (R ═ 15mm, λ ═ 1550nm) is large, LP01 mode bending loss (R ═ 25mm, λ ═ 1550nm) can be set to withstand practical use. Specifically, when R is_C(R＝25mm，λ＝1550nm)-R_CWhen (R ═ 25mm, λ ═ 1550nm) is 0.90 μm or more and LP01 mode bending loss (R ═ 15mm, λ ═ 1550nm) is less than 4.9dB/10 turns, LP01 mode bending loss (R ═ 25mm, λ ═ 1550nm) can be set to be less than 0.5dB/10 turns. In addition, when R is_C(R＝25mm，λ＝1550nm)-R_CWhen (R ═ 15mm, λ ═ 1550nm) is 1.60 μm or more and LP01 mode bending loss (R ═ 15mm, λ ═ 1550nm) is less than 4.9dB/10 turns, LP01 mode bending loss (R ═ 25mm, λ ═ 1550nm) can be set to be less than 0.2dB/10 turns.

FIG. 13 is a graph showing the relationship between Δ J (%) and Δ n × (D-D) (%. mu.m) Note that the symbol "○" plotted in FIG. 13 represents samples 1, 2, 6 and 7, and comparative examples 3 to 6 and comparative example 10, in which the cutoff wavelength λ is_CIs 1300nm or more and 1490nm or less. The symbol "□" represents samples 3 to 5, samples 8 to 13, comparative examples 7 to 9, and comparative example 11 in which the cutoff wavelength λ is_CIs 1490nm or more and 1630nm or less. The broken line in fig. 13 is represented by Δ J (%) ═ 0.056(μm)^-1) × Δ n × (D (μm) -D (μm)) -0.14, and the solid line indicates a line given by Δ J (%) -0.056(μm)^-1) × Δ n × (D (μm) -D (μm)) -0.22 FIG. 14 is a table summarizing preferred ranges and more preferred ranges for each parameter of the optical fiber.

In fig. 13, the boundary of the drawing region may be defined by having 0.056(μm)^-1) Is approximated by a straight line of slope of, and λ_CThe shorter the intercept tends to be larger. Intercept (i.e., Δ J-0.056(μm)^-1) × Δ n × (D (μm) -D (μm))) is preferably-0.22% or more and-0.14% or less, and more preferably-0.21% or more and-0.15% or less the distribution range shown in FIG. 14 may satisfy R_C,eff(R＝15mm，λ＝1550nm)≥1.46+ΔD(％)×1.93(1/％)。

Next, in the optical fiber state (the state having the cross-sectional structure shown in fig. 1), it is preferable that the young's modulus of the primary resin layer 40 is 0.3MPa or less, and the young's modulus of the secondary resin layer 50 is 800MPa or more. Further, it is preferable that the young's modulus of the primary resin layer is 0.2MPa or less or 0.1MPa or less, and the young's modulus of the secondary resin layer is 1000MPa or more. In this case, it is also possible to have an effect of suppressing optical loss (referred to as microbending loss) caused by random directional bending in the optical fiber, which is mainly generated when the optical fiber is formed into a cable.

In the quality inspection of the manufactured optical fiber, LP01 mode bending loss (R15 mm, λ 1550nm) and an effective area a were measured first_effAnd cutoff wavelength λ_CSo that it can be determined whether the transmission loss increases. Therefore, it is possible to discriminate an optical fiber considered to have an increased transmission loss from an optical fiber without an increased transmission loss without measuring the transmission loss (for ease of manufacturing management). Although it is efficient to wind the fiber on the mandrel when measuring the LP01 mode bend loss, there is a possibility that microbend losses are introduced by lateral pressure when the fiber is wound on the mandrel, resulting in measured values that are larger than actual values. This may lead to an erroneous judgment that an optical fiber without an increase in transmission loss may be determined to have an increase in transmission loss. From this viewpoint, it is also preferable that the young's modulus of the primary resin layer is 0.3MPa or less and the young's modulus of the secondary resin layer is 800MPa or more in the state of the optical fiber. Further, it is preferable that the young's modulus of the primary resin layer is 0.2MPa or less, and the young's modulus of the secondary resin layer is 1000MPa or more.

As described in opt.lett.vol.15,947-949(1990) by r.morgan et al, the difference in refractive index between the second cladding layer 30 and the primary resin layer 40 surrounding the second cladding layer 30 causes Fresnel reflection (Fresnel reflection) to occur at the interface between the second cladding layer 30 and the primary resin layer 40. In this case, it is known that there is a whispering gallery mode phenomenon (whispering gallery mode) in which light coupled to a high-order mode from the LP01 mode is reflected, and the reflected light is coupled to the LP01 mode again. This is one of the reasons why the transmission loss at the wavelength of 1550nm increases. In order to suppress the whispering gallery mode phenomenon, it is important to suppress an increase in the refractive index difference between the outer region 30B of the second clad layer 30 and the primary resin layer 40. Specifically, the absolute value of the refractive index difference between the refractive index of the outer region 30B of the second clad layer 30 and the refractive index of the primary resin layer 40 at a wavelength of 546nm is preferably 0.08 or less. More preferably, a value obtained by subtracting the refractive index of the outer region 30B of the second clad layer 30 (average refractive index when the refractive index of the outer region changes in the radial direction r) from the refractive index of the primary resin layer 40 at a wavelength of 546nm is 0 or more and 0.06 or less.

Further, at the interface of these layers, fresnel reflection due to a refractive index difference between the primary resin layer 40 and the secondary resin layer 50 surrounding the primary resin layer 40 may occur (whispering gallery mode phenomenon may occur). Therefore, it is desirable that the refractive index difference between the primary resin layer 40 and the secondary resin layer 50 is also small. Specifically, the absolute value of the refractive index difference between the primary resin layer 40 and the secondary resin layer 50 at a wavelength of 546nm is preferably 0.15 or less. More preferably, a value obtained by subtracting the refractive index of the primary resin layer 40 from the refractive index of the secondary resin layer 50 at a wavelength of 546nm is 0 or more and 0.10 or less.

Next, the refractive index profile of the region including the core 10 and the cladding portion having the depressed cladding structure surrounding the core 10 is not limited to the stepped form as shown in fig. 2A and 2B. For example, a combination of various shapes as shown in fig. 15 to 17 may be used. Fig. 15 is a diagram showing examples of various refractive index distributions applied to the core 10. Fig. 16 is a diagram showing examples of various refractive index profiles suitable for the first cladding layer 20. Fig. 17 is a diagram showing examples of various refractive index profiles applied to the second cladding layer 30.

As shown in fig. 15, the core 10 may have any distribution shape among the patterns 1 to 3. The pattern 1 has a distribution shape in which the refractive index of the core 10 linearly decreases from the optical axis AX in the radial direction r. The pattern 2 has the following distribution shape: it includes a portion where the refractive index of the core 10 is higher than PS (it is sufficient to have an average refractive index below PS as a whole). The pattern 3 has a distribution shape in which the refractive index of the core 10 increases in the radial direction r from the optical axis AX.

As shown in fig. 16, the first clad layer 20 may have any distribution shape among the patterns 1 to 4. The pattern 1 has a distribution shape in which the first cladding 20 has a uniform refractive index (variation of the relative refractive index difference in the radial direction r from the optical axis AX is ± 0.01% or less). The pattern 2 has a profile shape in which the refractive index of the first clad 20 linearly increases in the radial direction r. The pattern 3 has a profile shape in which the refractive index of the first clad 20 linearly decreases in the radial direction r. The pattern 4 has a distribution shape having a different refractive index between the inner region and the outer region of the first clad 20.

Further, as shown in fig. 17, the second clad layer 30 may have any distribution shape of the patterns 1 to 5. Note that the patterns 1 to 3 have a distribution shape in the case where the second clad layer 30 is composed of F-doped quartz glass. The patterns 4 and 5 have a distribution shape in the case where the second clad layer 30 is composed of pure silica glass. Specifically, the pattern 1 has a distribution shape in which the refractive index peak in the inner region 30A of the second cladding layer 30 is shifted toward the core 10 and the outer region 30B has a uniform refractive index. The pattern 2 has a distribution shape in which the distribution shape of the inner region 30A in the second cladding layer 30 is adjusted to be symmetrical in the radial direction r and the outer region 30B has a distribution shape of a uniform refractive index. Pattern 3 is similar to pattern 2, with the following distribution: the inner region 30A of the second clad layer 30 includes a region having a uniform refractive index in the radial direction r in the vicinity of the interface between the first clad layer 20 and the second clad layer 30. The pattern 4 has a profile shape in which the refractive index is adjusted to a stepwise form in the vicinity of the interface between the first cladding layer 20 and the second cladding layer 30. Pattern 5 shows a profile shape providing a region having a uniform refractive index near the interface between the first cladding layer 20 and the second cladding layer 30.