JP3783594B2 - Optical fiber, nonlinear optical fiber, optical amplifier using the same, wavelength converter, and optical fiber manufacturing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an optical fiber, a non-linear optical fiber, an optical amplifier using it, a wavelength transforming unit and a method for manufacturing the optical fiber which are provided with a satisfactory non-linear property and also by which a cutoff wavelength is shortened. SOLUTION: A double clad structure in which a first clad regin 20 and a second clad regin 30 are provided on the outer periphery of a core regin 10 is used as the structure of the optical fiber (non-linear optical fiber) having the high non-linear property. An adding density of GeO2 added within the core is raised and whereby a non-linear refractive index is raised in order to enlarge a non-linear coefficient γ by adopting the double clad structure. Also, the cutoff wavelength λc is satisfactorily shortened even in the case of reducing an effective cross section area Aeff by enlarging a difference of a specific refractive index between the core and the clad.

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an optical fiber, a non-linear optical fiber, an optical amplifier using the optical fiber, a wavelength converter, and a method for manufacturing the optical fiber.
[0002]
[Prior art]
In general, when high intensity (high light density) light propagates in a medium, it is known that various nonlinear optical phenomena such as a stimulated Raman effect and four-wave mixing occur in the medium. These non-linear optical phenomena also occur during optical transmission in optical fibers, and such non-linear optical phenomena in optical fibers can be used for optical amplification and wavelength conversion (for example, international (See published WO99 / 10770).
[0003]
[Problems to be solved by the invention]
The nonlinearity of the optical fiber is expressed by the nonlinear coefficient γ
γ = (2π / λ) × (n₂/ A_eff)
Represented by Where λ is the wavelength of light and n₂Is the nonlinear refractive index in the optical fiber, A_effIs the effective area of the optical fiber. From this equation, in order to increase the nonlinear coefficient γ, GeO added in the core of the optical fiber₂Non-linear refractive index n₂The effective area A is increased by increasing the relative refractive index difference between the core and the cladding._effShould be reduced.
[0004]
However, when the nonlinear coefficient γ is increased by applying the above-described configuration conditions, there arises a problem that the cutoff wavelength λc of the optical fiber becomes longer. In particular, if wavelength conversion is to be performed using four-wave mixing generated in an optical fiber, the wavelength of the excitation light needs to be in the vicinity of the zero dispersion wavelength of the optical fiber. On the other hand, in the above configuration, the cutoff wavelength λc is longer than the zero-dispersion wavelength and is not a single mode, so that the wavelength conversion efficiency is lowered.
[0005]
In recent years, in order to expand the wavelength band of signal light used in an optical transmission system, not only the amplification band of an EDFA usually used as an optical amplifier, but also a wavelength of 1.45 μm to 1.53 μm on the shorter wavelength side. The use of the S-band wavelength band is being studied. For this S-band wavelength band, there is no effective optical amplifier because it is out of the amplification wavelength band and EDFA cannot be used. If a Raman amplifier is used, the cutoff wavelength λc is longer than the pumping light wavelength of about 1.3 μm to 1.5 μm in the highly nonlinear optical fiber, and the efficiency of Raman amplification is reduced. End up.
[0006]
The present invention has been made in order to solve the above problems, and has an optical fiber having a sufficient nonlinearity and a shorter cutoff wavelength, a nonlinear optical fiber, an optical amplifier using the optical fiber, An object of the present invention is to provide a wavelength converter and a method for manufacturing an optical fiber.
[0007]
[Means for Solving the Problems]
  In order to achieve such an object, an optical fiber according to the present invention has (1) a maximum value of refractive index n₁The core region is provided on the outer periphery of the core region, and the minimum value of the refractive index is n₂(However, n₂<N₁) And the outer periphery of the first cladding region, and the maximum value of the refractive index is n₃(However, n₂<N₃<N₁) At least a second cladding region, and (2) 11 μm as various characteristics with respect to light having a wavelength of 1.55 μm.²A first cladding region having the following effective area, a cutoff wavelength λc of 0.7 μm or more and 1.6 μm or less at a fiber length of 2 m, and a nonlinear coefficient of 18 / W / km or more: The relative refractive index difference Δ between the two cladding regions⁻But with reference to the second cladding region-0.70% or more-0.15% or lessThe relative refractive index difference Δ between the core region and the second cladding region ⁺ Is 2.7% or more based on the second cladding regionIt is characterized by that.
[0008]
This optical fiber uses not a single clad structure but a double clad structure in which the first and second clad regions are provided on the outer periphery of the core region. Thereby, GeO added in the core in order to increase the nonlinear coefficient γ.₂To increase the non-linear refractive index and increase the relative refractive index difference between the core and the cladding to increase the effective area A_effEven when is made smaller, the cutoff wavelength λc can be made sufficiently short. In this configuration, the dispersion slope can be negative.
[0009]
The cladding structure may be configured such that one or more other cladding regions having a predetermined refractive index and width are further provided between the first cladding region and the second cladding region. .
[0010]
The optical fiber preferably further has a transmission loss of 3.0 dB / km or less and crosstalk between polarized waves of −15 dB or less as various characteristics with respect to light having a wavelength of 1.55 μm. Thereby, a highly nonlinear polarization-maintaining fiber can be obtained.
[0011]
Alternatively, the optical fiber preferably further has transmission loss of 1.0 dB / km or less and polarization mode dispersion of 0.3 ps / √km or less as various characteristics with respect to light having a wavelength of 1.55 μm. As a result, a highly nonlinear optical fiber with low polarization mode dispersion and low transmission loss can be obtained.
[0012]
Further, the relative refractive index difference Δ between the core region and the second cladding region⁺Is 2.7% or more based on the second cladding region. With such a large relative refractive index difference, the cutoff wavelength λc becomes longer in the single cladding structure, but according to the configuration of the optical fiber having the double cladding structure, the cutoff wavelength λc should be sufficiently shortened. Can do.
[0013]
In addition, a hermetic coat is provided on the outer periphery of the second cladding region. In the above optical fiber, GeO at the core₂Although the hydrogen characteristics are likely to deteriorate due to the high concentration, the hydrogen characteristics can be kept good by providing a hermetic coat.
[0014]
In addition, the excess absorption loss due to the OH group with respect to light having a wavelength of 1.38 μm is 0.2 dB / km or less. Thus, by reducing the absorption loss due to the OH group, the transmission loss at the pumping light wavelength of Raman amplification can be reduced, and the transmission loss for light transmitted in the S-band wavelength band can be reduced. it can.
[0015]
Further, the second cladding region is characterized in that fluorine is added. By adding fluorine to the clad, the relative refractive index difference between the core and the clad is increased, and the effective area A_effCan be reduced. Further, since the viscosity of the clad is lowered by the addition of fluorine, the drawing temperature can be lowered, and the formation of glass defects is suppressed. Therefore, the transmission loss in the optical fiber is reduced, and the hydrogen resistance is improved.
[0016]
The optical fiber is characterized in that the outer diameter of the glass portion including the core region, the first cladding region, and the second cladding region is 100 μm or less. Alternatively, the outer diameter of the glass part is further 90 μm or less.
[0017]
As described above, by setting the outer diameter of the glass portion to a small diameter, even when the covering portion provided on the outer periphery of the glass portion is made to have a small diameter, an optical fiber having sufficient strength can be obtained. Further, the strength against bending of the optical fiber is improved.
[0018]
Moreover, the outer diameter of the coating | coated part provided in the outer periphery of the glass part containing a core area | region, a 1st cladding area | region, and a 2nd cladding area | region is characterized by being 150 micrometers or less. Alternatively, the outer diameter of the covering portion is further 120 μm or less.
[0019]
Thus, by making the outer diameter of the covering portion small, when the optical fiber is coiled and used as an optical amplifier module or a wavelength converter module, the module can be reduced in size. Also, longer modules can be accommodated with modules of the same size.
[0020]
Further, in the characteristics with respect to light having a wavelength of 1.00 μm, the transmission loss is 5.0 dB / km or less. Alternatively, in the characteristics with respect to light having a wavelength of 1.00 μm, the transmission loss is further 3.0 dB / km or less.
[0021]
Thus, by reducing the transmission loss on the short wavelength side, the transmission loss at the pumping light wavelength in Raman amplification can be reduced, and the optical fiber having good characteristics can be used.
[0022]
The non-linear optical fiber according to the present invention is the above-described optical fiber, and uses a non-linear optical phenomenon that is manifested by inputting light of a predetermined wavelength. By actively utilizing the high non-linearity of the present optical fiber, a non-linear optical fiber that can be applied to various uses and has good characteristics can be obtained.
[0023]
An optical amplifier according to the present invention has a predetermined wavelength λp with respect to (a) the above-described nonlinear optical fiber having a cutoff wavelength λc, and (b) signal light having a wavelength λs input to the nonlinear optical fiber. (P) a pumping light source that supplies pumping light of λc <λp to the nonlinear optical fiber, and (c) uses the nonlinear optical phenomenon expressed in the nonlinear optical fiber to transmit the signal light. Amplifying.
[0024]
The optical amplifier having such a configuration can be used as a Raman amplifier using the stimulated Raman effect generated in the nonlinear optical fiber. Further, according to the nonlinear optical fiber having the above-described configuration, the cutoff wavelength λc can be made shorter than the wavelength λp of the pumping light (pump light), and optical amplification can be performed with high efficiency in a single mode. .
[0025]
Further, the dispersion value of the nonlinear optical fiber with respect to the signal light having the wavelength λs is +2 ps / km / nm or more, or −2 ps / km / nm or less. In this way, by giving the dispersion value an appropriate value that is not zero, it is possible to prevent four-wave mixing from occurring during the amplification of wavelength division multiplexing (WDM) signal light.
[0026]
Alternatively, the dispersion value of the nonlinear optical fiber with respect to the signal light having the wavelength λs is −10 ps / km / nm or less and the effective area is 10 μm.²It is characterized by the following. Such an optical amplifier can also be used as a dispersion compensator for a transmission line with positive dispersion.
[0027]
In this case, the dispersion slope value for the signal light having the wavelength λs of the nonlinear optical fiber is 0 ps / km / nm.²Is preferably smaller. In the optical fiber of the double clad structure, the dispersion slope can be made negative in this way, so that the dispersion slope can be compensated simultaneously with the dispersion of the transmission line in which the dispersion and the dispersion slope are positive. Become.
[0028]
Further, the wavelength λs of the signal light is 1.45 μm or more and 1.53 μm or less. By making such a wavelength range of signal light an amplification wavelength band, it can be used as an optical amplifier for signal light in the S band wavelength band. Further, since the cut-off wavelength λc can be shortened as described above, optical amplification can be performed with high efficiency in a single mode.
[0029]
The nonlinear optical fiber has an effective area A at the wavelength λp of the excitation light._{eff, p}And effective area A at wavelength λp + 0.1 μm_{eff, s}And the relational expression
(A_{eff, s}-A_{eff, p}) / A_{eff, p}× 100 ≧ 10%
It is characterized by satisfying.
[0030]
Here, the wavelength λp + 0.1 μm obtained by adding 0.1 μm to the wavelength λp of the excitation light corresponds to the wavelength λs of the signal light that is substantially optically amplified. Therefore, according to the above relational expression, the effective sectional area A_{eff, p}By reducing, the non-linearity with respect to the light of the wavelength λp corresponding to the excitation light can be increased, and the efficiency of optical amplification can be improved. Effective area A_{eff, s}By increasing the non-linearity, the non-linearity with respect to the light of the wavelength λp + 0.1 μm, which substantially corresponds to the signal light, can be reduced, and the deterioration of the transmission quality of the signal light can be suppressed.
[0031]
The wavelength converter according to the present invention includes (a) the above-described nonlinear optical fiber having a cutoff wavelength of λc, and (b) signal light having a wavelength λs (provided that λc <λs) input to the nonlinear optical fiber. And a pumping light source for supplying pumping light having a predetermined wavelength λp (where λc <λp) to the nonlinear optical fiber, and (c) using a nonlinear optical phenomenon expressed in the nonlinear optical fiber Then, the wavelength of the signal light is converted, and converted light having a wavelength λs ′ (where λc <λs ′) is output.
[0032]
A wavelength converter having such a configuration can be used as a wavelength converter using four-wave mixing that occurs in a nonlinear optical fiber. Further, according to the nonlinear optical fiber having the above-described configuration, the cutoff wavelength λc can be made shorter than the wavelengths of the signal light, the converted light, and the pumping light, and wavelength conversion can be performed with high efficiency in a single mode. it can. Also, the signal light can maintain good transmission characteristics without being affected by mode dispersion.
[0033]
Further, the optical power of the converted light that is output is greater than the optical power of the input signal light. Such a wavelength converter can also be used as an optical amplifier using parametric amplification.
[0034]
Further, the dispersion value of the nonlinear optical fiber with respect to the excitation light having the wavelength λp is −0.2 ps / km / nm or more and +0.2 ps / km / nm or less. In this way, by setting the dispersion value at the pumping light wavelength of the nonlinear optical fiber to be close to zero dispersion, a condition in which the phases of the signal light, the pumping light, and the converted light are matched is realized, and the four-wave Mixing can occur.
[0035]
The wavelength λs ′ of the converted light is 1.45 μm or more and 1.53 μm or less. By setting the wavelength range of such converted light as the converted wavelength band, it can be used as a wavelength converter that can obtain converted light in the S-band wavelength band. Further, since the cut-off wavelength λc can be shortened as described above, wavelength conversion can be performed with high efficiency in a single mode.
[0036]
An optical fiber manufacturing method according to the present invention includes (1) GeO.₂SiO with a predetermined amount added₂A first step of synthesizing a glass rod for a core to be a core region by a VAD method or an OVD method and stretching the glass rod to have a predetermined outer diameter; and (2) a predetermined amount of F is added. SiO₂A second step of forming a first cladding glass pipe to be a first cladding region by synthesizing by a VAD method or an OVD method and extending to a predetermined inner diameter and outer diameter; (3) A third step of flowing a predetermined gas to the inner surface of the first cladding glass pipe and heating it to perform etching to smooth the inner peripheral surface; and (4) a core in the first cladding glass pipe. A fourth step of inserting a glass rod for heating and air baking at a predetermined temperature of 1300 ° C. or higher, and then heating and integrating to form an intermediate glass rod; and (5) outside the core region and the first cladding region in the intermediate glass rod. After adjusting the ratio of the diameters, a fifth step of forming an optical fiber preform by forming a glass body serving as a second cladding region on the outer periphery of the intermediate glass rod; and (6) an optical fiber preform. It was heated wire drawing Omu, the maximum value of the refractive index n₁The core region is provided on the outer periphery of the core region, and the minimum value of the refractive index is n₂(However, n₂<N₁) And the outer periphery of the first cladding region, and the maximum value of the refractive index is n_Three(However, n₂<N_Three<N₁And a sixth step of creating an optical fiber including at least the second cladding region, and (7) heating integration of the core glass rod and the first cladding glass pipe in the fourth step, The heating temperature is 1800 ° C. or less, the roughness of the outer peripheral surface of the core glass rod is 5 μm or less, the roughness of the inner peripheral surface of the first cladding glass pipe is 5 μm or less, and from the outer peripheral surface of the core glass rod GeO within 2 μm thickness₂(8) In the sixth step, various characteristics with respect to light having a wavelength of 1.55 μm are set to 11 μm.²An optical fiber having the following effective cross-sectional area, a cutoff wavelength λc of 0.7 μm or more and 1.6 μm or less at a fiber length of 2 m, and a nonlinear coefficient of 18 / W / km or more is produced. To do.
[0037]
According to such an optical fiber manufacturing method, an optical fiber having a double-clad structure having high nonlinearity can be produced with good transmission characteristics such as reduced transmission loss.
[0038]
Further, in the sixth step, the optical fiber manufacturing method is characterized in that, in the sixth step, the transmission loss of 1.0 dB / km or less and the polarization mode dispersion of 0.3 ps / √km or less are various characteristics for light having a wavelength of 1.55 μm. The above-mentioned optical fiber further comprising: is produced. As a result, a highly nonlinear optical fiber with low polarization mode dispersion and low transmission loss can be obtained.
[0039]
Alternatively, in the optical fiber manufacturing method, the optical fiber preform obtained in the fifth step is used as the third intermediate glass body between the fifth step and the sixth step, and the third intermediate glass body The method further includes a seventh step of forming an optical fiber preform by forming a hole portion in the first cladding region or the second cladding region and then inserting a glass rod serving as a stress applying portion into the hole portion. In step 6, the optical fiber preform created in the seventh step is drawn by heating, and the core region, the first cladding region, the second cladding region, and a stress applying unit that applies stress to the core region And at least a transmission loss of 3.0 dB / km or less and crosstalk between polarized waves of -15 dB or less as various characteristics with respect to light having a wavelength of 1.55 μm. It is characterized in. Thereby, a highly nonlinear polarization-maintaining fiber can be obtained.
[0040]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, preferred embodiments of an optical fiber, a nonlinear optical fiber, an optical amplifier, a wavelength converter, and an optical fiber manufacturing method according to the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted. Further, the dimensional ratios in the drawings do not necessarily match those described.
[0041]
FIG. 1 is a diagram schematically showing a cross-sectional structure of a first embodiment of an optical fiber according to the present invention and a refractive index profile in the fiber radial direction (the direction indicated by the line L in the figure). The horizontal axis of the refractive index profile shown in FIG. 1 is different in scale, but each section on the cross section perpendicular to the central axis of the optical fiber along the line L shown in the cross sectional structure in the figure. Corresponds to the position. For comparison, the vertical axis of the refractive index profile is pure SiO.₂The refractive index at is indicated by a dotted line. Each region in the refractive index profile is given the same reference numeral as each region in the cross-sectional structure of the optical fiber.
[0042]
This optical fiber is made of SiO₂(Quartz glass) as a main component, an optical waveguide including a core region 10 including the central axis of the optical fiber, a first cladding region 20 provided on the outer periphery of the core region 10, and an outer periphery of the first cladding region 20 The second clad region 30 is provided.
[0043]
The core region 10 has an outer diameter (diameter) of 2r.₁And pure SiO₂GeO as an additive to increase the refractive index of glass₂Is added in a predetermined amount, and the maximum value of the refractive index is n₁(However, n₁> N₀, N₀Is pure SiO₂(Refractive index). Further, as shown in FIG. 1, the core region 10 of the present embodiment has GeO in the vicinity of the central axis of the optical fiber.₂The graded refractive index distribution has a maximum addition amount and refractive index.
[0044]
On the other hand, the first cladding region 20 has an outer diameter of 2r.₂And pure SiO₂A predetermined amount of F is added to the glass as an additive to lower the refractive index, and the minimum value of the refractive index is n.₂(However, n₂<N₀, N₂<N₁). The second cladding region 30 has an outer diameter of 2r._ThreeAnd pure SiO₂Glass or pure SiO₂A predetermined amount of F is added to the glass as an additive to lower the refractive index, and the maximum value of the refractive index is n._Three(However, n_Three≦ n₀, N₂<N_Three<N₁).
[0045]
Here, the relative refractive index difference in each part is expressed as the refractive index n in the second cladding region 30._ThreeIs defined as a standard. At this time, as shown in FIG. 1, the refractive index n in the core region 10.₁The relative refractive index difference corresponding to⁺= (N₁-N_Three) / N_Three× 100 (%) and refractive index n in the first cladding region 20₂The relative refractive index difference corresponding to^-= (N₂-N_Three) / N_ThreeIt is defined as x100 (%).
[0046]
The optical fiber according to the present embodiment uses not a single clad structure but a double clad structure in which the first clad region 20 and the second clad region 30 are provided on the outer periphery of the core region 10. In an optical fiber having a single clad structure, when the nonlinear coefficient γ is increased, there arises a problem that the cutoff wavelength λc becomes longer.
[0047]
On the other hand, GeO added in the core in order to increase the nonlinear coefficient γ by adopting the double clad structure as described above.₂To increase the non-linear refractive index and increase the relative refractive index difference between the core and the cladding to increase the effective area A_effEven when is made smaller, the cutoff wavelength λc can be made sufficiently short. In this configuration, the dispersion slope can be negative.
[0048]
The cladding structure may be configured such that one or more other cladding regions having a predetermined refractive index and width are further provided between the first cladding region and the second cladding region. .
[0049]
The optical fiber according to the present embodiment can be applied to various uses by using a nonlinear optical phenomenon that is expressed by inputting light of a predetermined wavelength (within a predetermined wavelength band). It can be used as a nonlinear optical fiber having various characteristics. In particular, since the nonlinear coefficient γ is increased and the cut-off wavelength λc is sufficiently shortened, a highly efficient optical device using a nonlinear optical phenomenon can be realized. Specific characteristics of the optical fiber will be described later in more detail.
[0050]
An example of an optical fiber manufacturing method for producing an optical fiber (nonlinear optical fiber) having the configuration shown in FIG. 1 will be described. In this manufacturing method, the core region 10 and the first cladding region 20 are not collectively synthesized by sooting by the VAD method or the OVD method, but the core glass rod and the first cladding glass pipe are separately produced. After that, a method of integrating them by heating is used.
[0051]
First, a glass rod for a core that becomes the core region 10 of the optical fiber described above is created (first step). Here, SiO₂As an additive to increase the refractive index with Ge as the main component₂A glass rod to which a predetermined amount is added is synthesized by the VAD method or the OVD method, and then stretched to have a predetermined outer diameter to obtain a core glass rod.
[0052]
Also, a first cladding glass pipe to be the first cladding region 20 of the optical fiber is created (second step). Here, SiO₂A glass pipe to which a predetermined amount of F is added as an additive that lowers the refractive index is synthesized by the VAD method or the OVD method, and then stretched to have a predetermined inner diameter and outer diameter to be used for the first cladding. Glass pipe.
[0053]
Further, vapor phase etching for smoothing the inner peripheral surface of the glass pipe is performed on the obtained first cladding glass pipe (third step). Here, SF₆A predetermined gas such as SF is flowed to the inner surface of the first cladding glass pipe (for example, SF₆+ Cl₂And the inner surface of the pipe is etched.
[0054]
Next, the obtained core glass rod and first clad glass pipe are heated and integrated (fourth step). The core glass rod is inserted into the first clad glass pipe, and is heated and integrated according to the procedure and conditions described below to produce an intermediate glass rod.
[0055]
Subsequently, after adjusting the ratio of the outer diameter of the core region and the first cladding region in the intermediate glass rod to be a predetermined ratio, the glass body that becomes the second cladding region 30 on the outer periphery of the intermediate glass rod Then, an optical fiber preform is formed (fifth step).
[0056]
Here, the adjustment of the ratio of the outer diameter of the intermediate glass rod is performed, for example, by grinding the outer peripheral portion with an HF solution or the like. This grinding uses a flame such as an oxyhydrogen flame as a heat source in the heat integration or stretching process, and when the flame is in contact with the glass surface, OH groups and metals attached to the glass surface, etc. It is necessary to remove impurities in the flame.
[0057]
Moreover, about the synthesis | combination of the glass body used as the 2nd clad area | region 30, you may synthesize | combine, for example by VAD method or OVD method. Alternatively, it may be formed by rod in collapse, or may be further synthesized by VAD method or OVD method after rod in collapse.
[0058]
Then, the obtained optical fiber preform is drawn by heating to produce an optical fiber (sixth step). Through the above steps, the optical fiber having the double clad structure shown in FIG. 1 is obtained.
[0059]
The procedure and conditions of the fourth step of heating and integrating the core glass rod and the first cladding glass pipe will be further described.
[0060]
In the method of manufacturing an optical fiber having a double clad structure, when the core region and the first clad region are collectively synthesized, GeO in the core region is obtained.₂Since the addition concentration is high and the F addition concentration in the first cladding region is also high, each diffuses in the glass fine particle body (soot body) in which the additive easily diffuses. At this time, GeF_FourAs a result, defects such as GeO and GeO occur, and transmission loss is deteriorated. In addition, high concentration GeO by MCVD method₂Addition SiO₂When trying to synthesize glass, there is also a problem that transmission loss is greatly deteriorated.
[0061]
On the other hand, in the above-described manufacturing method, the core region 10 and the first cladding region 20 are separately synthesized (first and second steps), and then heated and integrated (fourth step). However, even in this case, the GeO₂And F react with each other to form a gas such as GeO and may remain as bubbles at the interface between the core region 10 and the first cladding region 20. At this time, the characteristics of the optical fiber deteriorate due to the remaining bubbles.
[0062]
In order to suppress the generation of such bubbles, in this manufacturing method, in the fourth step of performing heat integration, heat integration is performed under any one of the following five conditions or a combination thereof. That is, (1) Integration is performed at a heating temperature of 1800 ° C. or lower. (2) Cl before heating integration₂Empty baking is performed at a predetermined temperature of 1300 ° C. or higher in an atmosphere. (3) The roughness of the inner peripheral surface of the first cladding glass pipe is 5 μm or less. (4) The roughness of the outer peripheral surface of the core glass rod is 5 μm or less. (5) GeO in the region within 2 μm thickness from the outer peripheral surface of the core glass rod₂The maximum concentration is 5 mol% or less. Generation of bubbles can be suppressed by applying heat integration by applying any of the above conditions or a combination thereof.
[0063]
About the effect of said manufacturing conditions, it confirmed by changing conditions and implementing heat integration. Here, for the core glass rod, the refractive index profile in the core is almost parabolic, and GeO₂The addition concentration was 30 mol% at the maximum. Further, the outer diameter of the core glass rod at the time of heat integration (hereinafter, the outer diameter and the inner diameter all indicate the diameter) was 6 mm. On the other hand, for the first cladding glass pipe, the refractive index profile in the first cladding was substantially stepped, and the F addition concentration was 1.5 mol% at the maximum.
[0064]
Moreover, the outer diameter of the glass pipe for 1st clads at the time of heat integration was 32 mm, and the internal diameter was 9 mm. The obtained glass pipe for the first cladding is SF₆300cm^Three/ Min, Cl₂200cm^ThreeThe surface was smoothed by etching at a heating temperature of 1500 ° C./min (maximum temperature of the glass surface measured with a pyroscope). In addition, the atmospheric gas in the pipe during heating integration is 200 cm of chlorine.^Three/ Min, oxygen 300cm^Three/ Min, and the degree of vacuum in the pipe was 1 kPa.
[0065]
First, the effect of suppressing bubble generation was confirmed under the condition that (1) the integration was performed at a heating temperature of 1800 ° C. or less. Here, the heating temperature for heat integration was changed in the range of 1950 ° C. to 1800 ° C., and the heat integration of the core glass rod and the first cladding glass pipe was performed. For other conditions, air baking is performed at 1300 ° C., the roughness of the inner peripheral surface of the first cladding glass pipe is 5 μm, the roughness of the outer peripheral surface of the core glass rod is 5 μm, GeO in the region within 2 μm thickness from the outer peripheral surface₂The maximum concentration was 5 mol%.
[0066]
FIG. 2 shows the number of bubbles generated at the interface between the core glass rod and the first cladding glass pipe. Here, the number of bubbles generated was evaluated by the number of bubbles generated per 10 mm length after collapse (glass rod). As shown in the table of FIG. 2, the number of bubbles generated was reduced by reducing the heating temperature, and bubbles were almost not generated at the heating temperature of 1800 ° C. This is because the progress of the chemical reaction is suppressed by lowering the heating temperature for heat integration.
[0067]
Next, (2) before heating integration, Cl₂The effect of suppressing the generation of bubbles was confirmed under the condition that air baking was performed at a predetermined temperature of 1300 ° C. or higher in the atmosphere. Here, heating integration was performed by changing the baking temperature for baking in the range of 1000 ° C to 1300 ° C. For other conditions, the heating temperature was 1800 ° C., the roughness of the inner peripheral surface of the first cladding glass pipe was 5 μm, the roughness of the outer peripheral surface of the core glass rod was 5 μm, and the outer peripheral surface of the core glass rod was GeO in the region within 2 μm thickness₂The maximum concentration was 5 mol%.
[0068]
FIG. 3 shows the number of bubbles generated at the interface between the core glass rod and the first cladding glass pipe. As shown in the table of FIG. 3, the number of bubbles generated was decreased by increasing the baking temperature, and almost no bubbles were generated at the baking temperature of 1300 ° C. This is because by performing baking at a sufficient temperature, unstable surface Ge compounds and F compounds are removed, and the surface state becomes smooth.
[0069]
Next, the effect of suppressing bubble generation was confirmed under the condition that (3) the roughness of the inner peripheral surface of the first cladding glass pipe was 5 μm or less. Here, the roughness of the inner peripheral surface of the glass pipe was changed in the range of 10 μm to 5 μm, and the heating integration was performed. For other conditions, air baking is performed at 1300 ° C., the heating temperature is 1800 ° C., the roughness of the outer peripheral surface of the core glass rod is 5 μm, and the thickness is within 2 μm from the outer peripheral surface of the core glass rod. GeO₂The maximum concentration was 5 mol%.
[0070]
FIG. 4 shows the number of bubbles generated at the interface between the core glass rod and the first cladding glass pipe. As shown in the table of FIG. 4, the number of bubbles generated was reduced by reducing the roughness of the inner peripheral surface of the first cladding glass pipe, and almost no bubbles were generated when the surface roughness was 5 μm. This is because by making the surface roughness sufficiently smooth, the rough surface portion is prevented from becoming a nucleus of bubble generation.
[0071]
Next, the effect of suppressing bubble generation was confirmed under the condition that (4) the roughness of the outer peripheral surface of the core glass rod was 5 μm or less. Here, the roughness of the outer peripheral surface of the glass rod was changed in the range of 10 μm to 5 μm, and the heating integration was performed. For other conditions, air baking is performed at 1300 ° C., the heating temperature is 1800 ° C., the roughness of the inner peripheral surface of the first cladding glass pipe is 5 μm, and the thickness is within 2 μm from the outer peripheral surface of the core glass rod. GeO in the region₂The maximum concentration was 5 mol%.
[0072]
FIG. 5 shows the number of bubbles generated at the interface between the core glass rod and the first cladding glass pipe. As shown in the table of FIG. 5, the number of bubbles generated was reduced by reducing the roughness of the outer peripheral surface of the core glass rod, and bubbles were almost not generated when the surface roughness was 5 μm. This is because, as in the case of the glass pipe, by sufficiently smoothing the surface roughness, it is possible to prevent the rough surface portion from becoming the core of bubble generation.
[0073]
Next, (5) GeO in a region within 2 μm in thickness from the outer peripheral surface of the glass rod for core.₂Regarding the condition that the maximum value of the concentration was 5 mol% or less, the effect of suppressing the generation of bubbles was confirmed. Here, GeO in the above region₂Heating integration was performed by changing the maximum value of concentration in the range of 10 mol% to 5 mol%. For other conditions, air baking is performed at 1300 ° C., the heating temperature is 1800 ° C., the roughness of the inner peripheral surface of the first cladding glass pipe is 5 μm, and the roughness of the outer peripheral surface of the core glass rod is 5 μm. It was.
[0074]
FIG. 6 shows the number of bubbles generated at the interface between the core glass rod and the first cladding glass pipe. As shown in the table of FIG. 6, the number of bubbles generated is GeO.₂Reduced by reducing the maximum concentration, GeO₂Almost no bubbles were generated at the maximum concentration of 5 mol%. This is because GeO on the surface layer₂This is because the concentration is reduced and bubbles are hardly generated.
[0075]
Under the above conditions, that is, baking is performed at 1300 ° C., the heating temperature is 1800 ° C., the roughness of the inner peripheral surface of the first cladding glass pipe is 5 μm, the roughness of the outer peripheral surface of the core glass rod is 5 μm, GeO in the region within 2 μm thickness from the outer peripheral surface of the core glass rod₂Heat integration was performed under the condition that the maximum value of the concentration was 5 mol%, and an intermediate glass rod (first intermediate glass rod) having an outer diameter of 30 mm without bubbles was obtained.
[0076]
And after extending | stretching the 1st intermediate glass rod to outer diameter 8mm, the outer peripheral part is ground with HF solution to outer diameter 5.4mm, and (core diameter) / (first clad diameter) = 0.30. It was adjusted. Separately from the first intermediate glass rod, a second cladding glass pipe that is the inner peripheral portion of the second cladding region 30 was prepared. This second cladding glass pipe is made of SiO having an F addition concentration of 0.7 mol%, an outer diameter of 32 mm, and an inner diameter of 8 mm.₂Glass pipe was used. And the 1st intermediate glass rod was inserted in the glass pipe for 2nd clads, and it integrated by heating, and obtained the 2nd intermediate glass rod with an outer diameter of 30 mm.
[0077]
Next, on the outer periphery of the obtained second intermediate glass rod, a glass body to be an outer peripheral side portion of the second cladding region 30 is made of SiO having an F addition concentration of 0.7 mol% similar to the second cladding glass pipe.₂An optical fiber preform was prepared by synthesizing the glass by the VAD method or the OVD method. Here, (second cladding diameter) / (first cladding diameter) = 7.8.
[0078]
In the above-described method of synthesizing the second cladding region 30, the inner peripheral side portion is formed by heat integration of the glass pipe. This is to reduce the amount of OH group mixed in the optical fiber. Further, the outer peripheral side portion is formed by the soot method of the VAD method or the OVD method. This is for increasing the size of the optical fiber preform.
[0079]
Various methods may be used for synthesizing the second cladding region 30 according to individual conditions. For example, in the case where the power field distribution of light is not so wide and the influence of OH groups mixed in the second cladding synthesis by the soot method can be ignored, it is not necessary to perform heating integration of the glass pipe. Or you may synthesize | combine a 2nd clad only by heating integration of a glass pipe, without performing the synthesis | combination by a soot method.
[0080]
The optical fiber preform produced by the above manufacturing method and manufacturing conditions was drawn by heating to obtain an optical fiber having a double clad structure shown in FIG. Its configuration is the outer diameter 2r of the core region 10₁= 4.8 μm, relative refractive index difference Δ⁺= 3.3%, outer diameter 2r of the first cladding region 20₂= 16 μm, relative refractive index difference Δ^-= -0.25%, outer diameter 2r of second cladding region 30_Three= 125 μm.
[0081]
Various characteristics for light with a wavelength of 1.55 μm are as follows:
Dispersion = + 0.22 ps / km / nm,
Dispersion slope = +0.045 ps / km / nm²,
Effective area A_eff= 10.4 μm²,
Cut-off wavelength λc = 1510 nm,
Zero dispersion wavelength = 1545 nm,
Transmission loss = 0.46 dB / km,
Mode field diameter = 3.69 μm,
Non-linear coefficient γ = 20.8 / W / km,
Polarization mode dispersion PMD = 0.05 ps / √km
Thus, an optical fiber (non-linear optical fiber) with good characteristics was obtained.
[0082]
The characteristics of the optical fiber described above are as follows for the light with a wavelength of 1.55 μm.
11 μm²Effective sectional area A_eff,
Cutoff wavelength λc of 0.7 μm or more and 1.6 μm or less with a fiber length of 2 m,
A nonlinear coefficient γ of 18 / W / km or more,
Meet. Further, the transmission loss for light having a wavelength of 1.55 μm satisfies the characteristic condition of 3.0 dB / km or less, or 1.0 dB / km or less.
[0083]
Thus, by adopting the double clad structure, the GeO of the core₂Effective area A with increasing concentration_effEven when the non-linear coefficient γ is increased by reducing the non-linear coefficient γ, a highly non-linear optical fiber having a suitable cutoff wavelength λc can be obtained.
[0084]
Note that the relative refractive index difference Δ between the core region 10 and the second cladding region 30.⁺Is effective area A_effTo make it sufficiently small, Δ⁺Is preferably 2.7% or more. In the case of such a large relative refractive index difference, the cutoff wavelength λc becomes long in the single cladding structure, but the cutoff wavelength λc can be sufficiently shortened as described above according to the double cladding structure.
[0085]
FIG. 7 is a diagram schematically showing a cross-sectional structure of a second embodiment of an optical fiber according to the present invention and a refractive index profile in the fiber radial direction (direction indicated by line L in the drawing).
[0086]
This optical fiber is made of SiO₂(Quartz glass) as a main component, an optical waveguide including a core region 10 including the central axis of the optical fiber, a first cladding region 20 provided on the outer periphery of the core region 10, and an outer periphery of the first cladding region 20 The second clad region 30 is provided. Here, the configurations of the first cladding region 20 and the second cladding region 30 are the same as those in the first embodiment.
[0087]
On the other hand, the core region 10 has an outer diameter (diameter) of 2r.₁And pure SiO₂GeO as an additive to increase the refractive index of glass₂Is added in a predetermined amount, and the maximum value of the refractive index is n₁(However, n₁> N₀). Further, as shown in FIG. 7, the core region 10 of the present embodiment has a GeO region in the vicinity of the central axis of the optical fiber.₂The graded refractive index distribution has a maximum addition amount and refractive index.
[0088]
Further, an intermediate region 15 is provided in a predetermined range on the outer peripheral side in the core region 10 at a position between the core region 10 and the first cladding region 20. This intermediate region 15 has a slightly higher concentration of GeO so as to have a refractive index distribution (addition concentration distribution) protruding in an angular shape as shown in FIG.₂Is added. Here, the maximum value of the refractive index of the intermediate region 15 is n_Five(However, n_Five> N₀), The relative refractive index difference is Δ_Five= (N_Five-N_Three) / N_ThreeAnd
[0089]
Similar to the optical fiber according to the first embodiment, the optical fiber according to the present embodiment is not a single clad structure, but a double clad structure in which the first cladding region 20 and the second cladding region 30 are provided on the outer periphery of the core region 10. Used. Thereby, GeO added in the core in order to increase the nonlinear coefficient γ.₂To increase the non-linear refractive index and increase the relative refractive index difference between the core and the cladding to increase the effective area A_effEven when is made smaller, the cutoff wavelength λc can be made sufficiently short. In this configuration, the dispersion slope can be negative. The effect of the intermediate region 15 will be described later together with an optical fiber manufacturing method.
[0090]
The optical fiber according to the present embodiment can be applied to various applications by using a non-linear optical phenomenon expressed by inputting light of a predetermined wavelength (within a predetermined wavelength band) and is good. It can be used as a nonlinear optical fiber having various characteristics.
[0091]
An example of an optical fiber manufacturing method for producing an optical fiber (non-linear optical fiber) having the configuration shown in FIG. 7 will be described.
[0092]
First, a glass fine particle body (soot body) composed of a region to be the core region 10 including the intermediate region 15 and a precursor region to be the first cladding region 20 is synthesized. Here, the region to be the core region 10 is GeO.₂SiO with up to 30 mol% added₂The region of the outer periphery corresponding to the intermediate region 15 is made of GeO.₂Is added in a square shape as described above so that the addition concentration at the peak value is 5 mol%.₂Glass was used. The precursor region of the first cladding region 20 is pure SiO on its outer periphery.₂Synthesized as glass.
[0093]
The obtained glass fine particles (glass porous body) are put in a sintering furnace, heated at a heating temperature of 1300 ° C. in a mixed atmosphere of chlorine and helium, and then dehydrated, and then heated at 1400 ° C. in a helium atmosphere. The region to be the core region 10 and the intermediate region 15 was selectively densified (transparent).
[0094]
At this time, the regions to be the core region 10 and the intermediate region 15 are GeO at a high concentration.₂Is added to lower the densification temperature, so that the effect of densification by heating can be sufficiently obtained. On the other hand, the precursor region of the first cladding region 20 is pure SiO.₂Since it is glass, the densification temperature is high, and heating at 1400 ° C. does not increase the densification but remains as a glass particulate.
[0095]
In this state, C is added with helium and F at a heating temperature of 1400 ° C.₂F₆, SiF_Four, CF_FourThe glass body is heated in a mixed atmosphere of gas such as F, and F is added at a concentration of 1 mol% to the precursor region of the first cladding region 20 that has not been densified to form the first cladding region 20.
[0096]
Here, in the case where F is added at the time of heating and sintering the glass fine particle body in this way, in a normal method, F added to the clad also enters the core region. At this time, the refractive index of the core region is lowered, and impurities such as GeO and Ge—F compounds are generated, resulting in a problem that transmission loss is deteriorated. On the other hand, in this manufacturing method, GeO is formed on the outer peripheral portion of the core region 10.₂Is formed at a high concentration, and the regions are selectively densified by heating at a slightly low temperature. By subsequently adding F, F can be selectively added only to the precursor region of the first cladding region 20.
[0097]
On the outer periphery of the obtained glass body, a glass body to be the second cladding region 30 was formed to prepare an optical fiber preform. Here, for the second cladding region 30, SiO with F added at an addition concentration of 0.3 mol%.₂Glass was used. The ratio of the outer diameters of the respective regions was (core diameter) / (first cladding diameter) = 0.40, (second cladding diameter) / (first cladding diameter) = 11.6.
[0098]
The optical fiber preform prepared by the above manufacturing method and manufacturing conditions was drawn by heating to obtain an optical fiber having a double clad structure shown in FIG. Its configuration is the outer diameter 2r of the core region 10₁= 4.3 μm, relative refractive index difference Δ⁺= 3.1%, relative refractive index difference Δ of the intermediate region 15_Five= 1.0%, outer diameter 2r of the first cladding region 20₂= 10.8 μm, relative refractive index difference Δ^-= −0.26%, outer diameter 2r of second cladding region 30_Three= 125 μm. In addition, the refractive index profile (GeO of the core region 10)₂(Addition concentration distribution) was approximately α to 3.0 power distribution.
[0099]
Various characteristics for light with a wavelength of 1.55 μm are as follows:
Dispersion = +0.98 ps / km / nm,
Dispersion slope = +0.035 ps / km / nm²,
Effective area A_eff= 10.2 μm²,
Cut-off wavelength λc = 1465 nm,
Zero dispersion wavelength = 1520 nm,
Transmission loss = 0.49 dB / km,
Mode field diameter = 3.64 μm,
Non-linear coefficient γ = 21.5 / W / km,
Thus, an optical fiber (non-linear optical fiber) with good characteristics was obtained.
[0100]
The characteristics of the optical fiber described above are as follows for the light with a wavelength of 1.55 μm.
11 μm²Effective sectional area A_eff,
Cutoff wavelength λc of 0.7 μm or more and 1.6 μm or less with a fiber length of 2 m,
A nonlinear coefficient γ of 18 / W / km or more,
Meet. Further, the transmission loss for light having a wavelength of 1.55 μm satisfies the characteristic condition of 3.0 dB / km or less, or 1.0 dB / km or less.
[0101]
Thus, by adopting the double clad structure, the GeO of the core₂Effective area A with increasing concentration_effEven when the non-linear coefficient γ is increased by reducing the non-linear coefficient γ, a highly non-linear optical fiber having a suitable cutoff wavelength λc can be obtained.
[0102]
Note that the relative refractive index difference Δ between the core region 10 and the second cladding region 30.⁺Is effective area A_effTo make it sufficiently small, Δ⁺Is preferably 2.7% or more. In the case of such a large relative refractive index difference, the cutoff wavelength λc becomes long in the single cladding structure, but the cutoff wavelength λc can be sufficiently shortened as described above according to the double cladding structure.
[0103]
The preferred configuration conditions and various characteristics of the optical fiber (nonlinear optical fiber) according to the present invention will be further examined. Of the various characteristics of the optical fiber shown below, those depending on the wavelength show characteristics for light having a wavelength of 1.55 μm unless otherwise specified.
[0104]
First, the transmission loss in the optical fiber having the above configuration will be examined. In an optical fiber having high nonlinearity, GeO is highly concentrated in the core in order to increase nonlinear refractive index and increase nonlinearity.₂Is added. At this time, transmission loss is likely to deteriorate due to heating during drawing. Such deterioration of transmission loss can be suppressed by lowering the heating temperature at the time of drawing, but in drawing at a low temperature, excessive tension is applied during drawing of the optical fiber. Therefore, there is a problem that the optical fiber is broken.
[0105]
On the other hand, in the optical fiber having the double clad structure shown in FIGS. 1 and 7, it is preferable to add F (fluorine) to the second clad region 30 occupying most of the volume of the optical fiber. As a result, the viscosity in the second cladding region 30 can be reduced, so that the drawing temperature can be lowered, and deterioration of transmission loss is suppressed.
[0106]
To reduce the transmission loss, two types of optical fibers A1 and A2 having a structure shown in the refractive index profile of FIG.
[0107]
The optical fiber A1 uses the refractive index profile shown in FIG. 8A, and the core region 10 has a parabolic distribution shape of GeO.₂Addition SiO₂(Maximum addition concentration 30 mol%), the first cladding region 20 is made of F-added SiO.₂(Addition concentration 1.6 mol%), the second cladding region 30 is made of F-added SiO.₂It was prepared as (addition concentration 0.9 mol%).
[0108]
Further, the optical fiber A2 uses the refractive index profile shown in FIG. 8B, and the core region 10 has a parabolic distribution shape of GeO.₂Addition SiO₂(Maximum addition concentration 30 mol%), the first cladding region 20 is made of F-added SiO.₂(Addition concentration 1.6 mol%), the second cladding region 30 is pure SiO₂Created as.
[0109]
Both optical fibers A1 and A2 were drawn with a drawing speed of 300 m / min and a tension of 4 N (400 gw). Here, the maximum temperature on the glass surface of the optical fiber A1 was 1900 ° C., the maximum temperature on the glass surface of the optical fiber A2 was 2000 ° C., and the optical fiber A1 could be drawn at a lower temperature.
[0110]
Various characteristics of the obtained optical fibers A1 and A2 are shown in FIG. From the table in FIG. 9, it can be seen that the optical fiber A1 in which F is added to the second cladding region 30 has a smaller transmission loss and a larger nonlinear coefficient γ than the optical fiber A2.
[0111]
Next, the cutoff wavelength λc and the effective cross-sectional area A in the optical fiber_effAnd the nonlinear coefficient γ. In an optical fiber having high nonlinearity, as described above, GeO is highly concentrated in the core.₂Is added to increase the nonlinear refractive index, and the effective area A_effIs preferably reduced. At this time, the nonlinear coefficient γ increases, while the cut-off wavelength λc increases. On the other hand, if an optical fiber having a double clad structure is used, the nonlinear coefficient γ can be increased and the cutoff wavelength λc can be sufficiently shortened.
[0112]
In addition, when a nonlinear optical fiber is applied to wavelength conversion using four-wave mixing, the phase must be matched, so the dispersion value at the wavelength λp of the wavelength conversion excitation light needs to be almost zero. There is. Therefore, it is desirable that λp is in the vicinity of the zero dispersion wavelength. The wavelength λs ′ of the converted light that has undergone wavelength conversion with respect to the signal light of wavelength λs is:
λs ′ = λp− (λs−λp)
It becomes. For example, when the wavelength of WDM signal light having a wavelength of 1530 nm to 1565 nm is collectively converted by excitation light having a wavelength of 1525 nm, the wavelength of the converted light is in the range of wavelengths from 1520 nm to 1490 nm. The cut-off wavelength λc needs to be a suitable value in consideration of the wavelength of these signal light, converted light, pump light, amplified light, and the like.
[0113]
This cutoff wavelength λc, effective area A_effFor the non-linear coefficient γ, four types of optical fibers B1, B2, C1, and C2 having the structure shown in the refractive index profile of FIG.
[0114]
Each of the optical fibers B1 and B2 uses the refractive index profile shown in FIG.₂Addition SiO₂The first cladding region 20 is made of F-added SiO.₂(Addition concentration 2.1 mol%), the second cladding region 30 is made of F-added SiO.₂It was prepared as (addition concentration 0.9 mol%). GeO in the core region 10₂The addition concentrations were different values.
[0115]
In addition, the optical fibers C1 and C2 each have a refractive index profile shown in FIG.₂Addition SiO₂The first cladding region 20 is made of F-added SiO.₂(Addition concentration 2.1 mol%), the second cladding region 30 is pure SiO₂Created as. GeO in the core region 10₂The addition concentrations were different values.
[0116]
For comparison, optical fibers D1 to D5 having a single clad structure were prepared. Each of these optical fibers D1 to D5 was created by the refractive index profile shown in FIG. Here, reference numeral 60 denotes a core region, and reference numeral 70 denotes a clad region having a single clad structure.
[0117]
Each of the optical fibers D1 to D5 has a refractive index profile shown in FIG.₂Addition SiO₂The cladding region 70 is made of F-added SiO.₂It was prepared as (addition concentration 0.9 mol%). GeO in the core region 60₂The addition concentrations were different values. The relative refractive index difference Δ of the core region 60⁺Is based on the cladding region 70.
[0118]
The relative refractive index difference Δ of the obtained optical fibers B1, B2, C1, C2⁺, Δ^-FIG. 12 shows various characteristics, and the relative refractive index difference Δ of the optical fibers D1 to D5 for comparison.⁺And various characteristics are shown in FIG. From the table of FIG. 13, in the optical fibers D1 to D5 having the single clad structure, the GeO in the core₂Low additive concentration Δ⁺Is small, the effective area A_effIs large, and the value of the nonlinear coefficient γ is also small. Δ⁺Is 2.7% or more, the wavelength of the WDM signal light having a cutoff wavelength of 1530 nm to 1565 nm is longer than the wavelength of the converted light in the collective wavelength conversion using the excitation light having a wavelength of 1525 nm.
[0119]
On the other hand, from the table of FIG. 12, the effective cross-sectional area A is obtained for the optical fibers B1, B2, C1, C2 having the double clad structure._effIs small and a large nonlinear coefficient γ is obtained. For example, Δ⁺The effective cross-sectional area A is such that the cut-off wavelength is 1469 nm (optical fiber C2) even if the ratio is 4.5%._effIs 11μm²A sufficiently short cut-off wavelength is realized even when the value is small and the nonlinear coefficient γ is 18 / W / km or more.
[0120]
Next, hydrogen resistance characteristics in optical fibers will be examined. GeO in the core₂If the concentration of is high, its hydrogen resistance tends to deteriorate. On the other hand, a hermetic coating mainly composed of a substance having a shielding property against water molecules and hydrogen molecules such as amorphous carbon, silicon carbide, etc. on the outer peripheral portion of the second cladding region 30 which is the outermost layer of the optical fiber (see FIG. 1 and the hermetic coat 50 shown in FIG. 7).
[0121]
At this time, diffusion of hydrogen into the core region and the cladding region of the optical fiber can be blocked. Further, the static fatigue coefficient becomes 100 to 160, and the fracture probability is extremely reduced. Thereby, the long-term reliability of the optical fiber can be improved.
[0122]
As described above, eight types of optical fibers E1 to E8 related to the optical fiber of the present invention were manufactured on the basis of the configuration, manufacturing method, and preferable manufacturing conditions of the optical fiber (non-linear optical fiber) examined.
[0123]
These optical fibers E1 to E8 are formed of GeO having a refractive index profile approximately in the power of α to 3.0 in the core region 10.₂Addition SiO₂The first cladding region 20 is made of F-added SiO.₂The second cladding region 30 is made of F-added SiO.₂Or pure SiO₂Created as. The relative refractive index difference Δ of the obtained optical fibers E1 to E8⁺, Δ^-F concentration in the second cladding region 30, outer diameter 2r of the core region 10 and the first cladding region 20₁2r₂FIG. 14 is a table showing characteristics thereof. Of the characteristics shown, OH absorption transmission loss indicates an increase in transmission loss (excess absorption loss) at a wavelength of 1.38 μm due to OH group absorption.
[0124]
The characteristics of the optical fibers E1 to E8 shown in the table of FIG. 14 are all the following characteristic conditions for light having a wavelength of 1.55 μm.
11 μm²Effective sectional area A_eff,
Cutoff wavelength λc of 0.7 μm or more and 1.6 μm or less with a fiber length of 2 m,
Transmission loss of 1.0 dB / km or less,
Polarization mode dispersion PMD of 0.3 ps / √km or less,
A nonlinear coefficient γ of 18 / W / km or more,
Meet. Thus, by adopting the double clad structure, the GeO of the core₂Effective area A with increasing concentration_effEven when the non-linear coefficient γ is increased by reducing the non-linear coefficient γ, a highly non-linear optical fiber having a suitable cutoff wavelength λc can be obtained. In addition, a highly nonlinear optical fiber with small polarization mode dispersion and low transmission loss can be obtained.
[0125]
Here, the excess absorption loss due to the OH group with respect to light having a wavelength of 1.38 μm is preferably 0.2 dB / km or less. The optical fibers E1 to E8 shown in FIG. 14 all satisfy this characteristic condition.
[0126]
When a stress applying portion is provided at a predetermined site in the optical fiber, a polarization maintaining optical fiber is obtained. FIG. 15 shows a cross-sectional structure of another embodiment of an optical fiber that is such a polarization-maintaining optical fiber. In this optical fiber, B on both the left and right sides sandwiching the core region 10₂O_ThreeAddition SiO₂Each of the stress applying portions 40 is formed. In the polarization-maintaining optical fiber having such a configuration, although the transmission loss may be deteriorated due to the stress applying unit 40, a random coupling between orthogonal polarizations can be suppressed. As a result, the quality of the transmitted signal light can be kept good.
[0127]
The manufacturing method of the optical fiber having such a configuration is substantially the same as the manufacturing method described above for the optical fiber having the configuration shown in FIG. 1, but the second cladding is formed on the outer periphery of the intermediate glass rod in the fifth step. What formed the glass body used as the area | region 30 is not used as an optical fiber preform as it is, but this is made into a 3rd intermediate glass body, and is further processed.
[0128]
That is, a hole is formed in the first cladding region or the second cladding region of the obtained third intermediate glass body. And the glass rod used as the stress provision part 40 is inserted in the opening part, and an optical fiber preform is created. By drawing the optical fiber preform by heating, an optical fiber having a configuration having the stress applying portion 40 is obtained.
[0129]
An example of the above manufacturing method will be described. Here, for the core glass rod, the refractive index profile in the core is almost parabolic, and GeO₂The addition concentration was 30 mol% at the maximum. Moreover, the outer diameter of the core glass rod at the time of heat integration was 8 mm. On the other hand, for the first cladding glass pipe, the refractive index profile in the first cladding was substantially stepped, and the F addition concentration was 1.5 mol% at the maximum.
[0130]
Moreover, the outer diameter of the glass pipe for 1st clads at the time of heat integration was 32 mm, and the internal diameter was 9 mm. The obtained glass pipe for the first cladding is SF₆300cm^Three/ Min, Cl₂200cm^ThreeThe surface was smoothed by etching at a heating temperature of 1500 ° C./min (maximum temperature of the glass surface measured with a pyroscope).
[0131]
For grilling before heating integration, Cl₂500cm^Three/ Min, baking was performed at a heating temperature of 1500 ° C. Atmosphere gas in the pipe at the time of heating integration is chlorine 200cm^Three/ Min, oxygen 300cm^Three/ Min, and the degree of vacuum in the pipe was 1 kPa.
[0132]
For heating integration, the heating temperature is 1700 ° C., the roughness of the inner peripheral surface of the first cladding glass pipe is 3 μm or less, the roughness of the outer peripheral surface of the core glass rod is 2 μm or less, GeO in the region within 2 μm thickness from the outer peripheral surface₂Heat integration was performed under the condition that the maximum value of the concentration was 3 mol%, and an intermediate glass rod (first intermediate glass rod) having an outer diameter of 30 mm without bubbles was obtained.
[0133]
And after extending the first intermediate glass rod to an outer diameter of 9 mm, the outer periphery thereof was ground to an outer diameter of 6 mm with an HF solution to adjust (core diameter) / (first cladding diameter) = 0.40. . Separately from the first intermediate glass rod, a second cladding glass pipe that is the inner peripheral portion of the second cladding region 30 was prepared. This second cladding glass pipe is made of almost pure SiO 2.₂SiO with outer diameter 32mm and inner diameter 9mm₂Glass pipe was used. And the 1st intermediate glass rod was inserted in the glass pipe for 2nd clads, and it integrated by heating, and obtained the 2nd intermediate glass rod with an outer diameter of 30 mm.
[0134]
Next, on the outer periphery of the obtained second intermediate glass rod, a glass body that becomes the outer peripheral side portion of the second cladding region 30 is made of substantially pure SiO as in the second cladding glass pipe.₂SiO₂The glass was synthesized by the VAD method or the OVD method to produce a third intermediate glass body. Here, (second cladding diameter) / (first cladding diameter) = 10.8.
[0135]
Furthermore, this third intermediate glass body was stretched to an outer diameter of 36 mm. At this time, the outer diameter of the core region 10 portion of the stretched third intermediate glass body was 1.3 mm, and the outer diameter of the first cladding region 20 portion was 3.3 mm. In the second clad region 30 portion of the third intermediate glass body, two apertures serving as the stress applying portion 40 shown in FIG. 15 were formed. In these apertures, the distance between the centers of the two apertures was 15.2 mm, and the outer diameter of each aperture was 10 mm. In addition, the centers of the two opening portions, the centers of the core region 10 and the first cladding region 20 were made to be substantially in a straight line.
[0136]
Polishing was performed until the roughness of the inner peripheral surface of the formed aperture became 2 μm or less, and the product was washed with water, alcohol, or aqua regia so as to remove foreign substances such as abrasives and grinding scraps. And as a glass rod used as the stress provision part 40, B with an outer diameter of 9 mm₂O_ThreeAddition SiO₂A glass rod was inserted into the aperture and sealed to create an optical fiber preform.
[0137]
An optical fiber preform having the structure shown in FIG. 15 was obtained by drawing the optical fiber preform produced by the above manufacturing method and manufacturing conditions. Here, the glass rod inserted into the opening portion is integrated with the cladding region by heating at the time of drawing, and becomes the stress applying portion 40. The configuration of the obtained optical fiber is the outer diameter 2r of the core region 10.₁= 4.6 μm, relative refractive index difference Δ⁺= 3.0%, outer diameter 2r of the first cladding region 20₂= 11.6 μm, relative refractive index difference Δ^-= -0.5%, outer diameter 2r of second cladding region 30_Three= 125 μm.
[0138]
Various characteristics for light with a wavelength of 1.55 μm are as follows:
Dispersion = + 0.01 ps / km / nm,
Dispersion slope = + 0.042ps / km / nm²,
Effective area A_eff= 10.6 μm²,
Cut-off wavelength λc = 1349 nm,
Zero dispersion wavelength = 1550 nm,
Transmission loss = 1.5 dB / km,
Mode field diameter = 3.75 μm,
Non-linear coefficient γ = 20.2 / W / km,
Crosstalk between polarized waves = -20 dB (fiber length 1 km)
Thus, an optical fiber (non-linear optical fiber) with good characteristics was obtained.
[0139]
As optical fibers having such a configuration, three types of optical fibers F1 to F3 according to the optical fiber of the present invention were further prototyped.
[0140]
These optical fibers F1 to F3 are similar to the optical fibers E1 to E8, and the GeO having a refractive index distribution of approximately α to 3.0 power in the core region 10.₂Addition SiO₂The first cladding region 20 is made of F-added SiO.₂The second cladding region 30 is made of F-added SiO.₂Or pure SiO₂Created as. The relative refractive index difference Δ of the obtained optical fibers F1 to F3⁺, Δ^-F concentration in the second cladding region 30, outer diameter 2r of the core region 10 and the first cladding region 20₁2r₂FIG. 16 is a table showing various characteristics according to the above. Of the characteristics shown, OH absorption transmission loss indicates an increase in transmission loss (excess absorption loss) at a wavelength of 1.38 μm due to OH group absorption.
[0141]
The characteristics of the optical fibers F1 to F3 shown in the table of FIG. 16 are all the following characteristic conditions for light having a wavelength of 1.55 μm.
11 μm²Effective sectional area A_eff,
Cutoff wavelength λc of 0.7 μm or more and 1.6 μm or less with a fiber length of 2 m,
Transmission loss of 3.0 dB / km or less,
Crosstalk between polarizations of -15 dB or less,
A nonlinear coefficient γ of 18 / W / km or more,
Meet. Thus, by adopting the double clad structure, the GeO of the core₂Effective area A with increasing concentration_effEven when the non-linear coefficient γ is increased by reducing the non-linear coefficient γ, a highly non-linear optical fiber having a suitable cutoff wavelength λc can be obtained. In addition, a highly nonlinear polarization-maintaining optical fiber can be obtained.
[0142]
Here, the excess absorption loss due to the OH group with respect to light having a wavelength of 1.38 μm is preferably 0.2 dB / km or less. The optical fibers F1 to F3 shown in FIG. 16 all satisfy this characteristic condition.
[0143]
In the optical fiber having the above-described configuration and various characteristics, by utilizing the nonlinear optical phenomenon expressed by inputting light of a predetermined wavelength, high nonlinearity is actively used, and the cutoff wavelength A nonlinear optical fiber having good characteristics with respect to λc and the like can be obtained. Such a non-linear optical fiber can be applied to various optical devices using a non-linear optical phenomenon.
[0144]
Here, in an optical device such as an optical amplifier or a wavelength converter that uses an optical fiber having the above-described configuration as a nonlinear optical fiber, an optical module (for example, an optical device) is obtained by forming an optical device into a coil by housing the optical fiber in a coil. An amplifier module or a wavelength converter module) may be used. In such a case, it is necessary to maintain the characteristics of the optical fiber so as to be suitable for modularization, such as bending characteristics including changes in strength and bending loss of the optical fiber.
[0145]
On the other hand, as a configuration of the optical fiber, the outer diameter of the glass portion of the optical fiber is preferably set to 100 μm or less. Alternatively, it is preferable that the outer diameter of the glass portion be 90 μm or less. In this way, by making the outer diameter of the glass portion small, even when the coating portion provided on the outer periphery of the glass portion is thin, an optical fiber having sufficient strength including the strength against bending is obtained. be able to.
[0146]
For example, considering the strength against bending of the optical fiber, when the optical fiber is bent to be coiled and accommodated in the optical module, bending stress is generated in each part in the glass portion of the optical fiber. This bending stress causes the optical fiber to break due to the strength of the optical fiber against bending.
[0147]
Specifically, when the optical fiber is bent and wound in a coil shape, the bending stress generated is substantially zero at the central portion (near the central axis) of the glass portion of the optical fiber. On the other hand, a compressive stress is generated in the glass portion because the bending diameter is smaller at the inner portion in the radial direction of the optical fiber coil than at the central portion. On the other hand, a tensile stress is generated in the glass portion because the bending diameter is larger at the outer portion in the radial direction of the optical fiber coil than at the central portion. And the magnitude | size of these compressive stress and tensile stress becomes large as the distance from the center site | part of a glass part becomes large.
[0148]
On the other hand, according to the optical fiber having the above-described configuration in which the outer diameter of the glass portion is small, the distance from the central portion of the portion of the glass portion located on the innermost side or the outer side in the radial direction of the coil is reduced, The magnitude of stress generated in the glass portion of the optical fiber is reduced. Thereby, the strength with respect to the bending of the optical fiber is improved, and breakage due to the stress of the optical fiber when coiled is prevented.
[0149]
Further, in the above-described optical fiber having a highly non-linearity double clad structure, the effective area A_effThe spread of the electromagnetic field distribution of the light transmitted through the glass portion is small due to, for example, the reduction in the size. Also, in such an optical fiber, the numerical aperture NA is generally large. For this reason, in the above-described optical fiber, the bending loss is small, and the influence on the transmission loss due to the small outer diameter of the glass portion is small. Therefore, it is possible to obtain an optical fiber having a sufficient bending property and bending loss is reduced, and having good bending characteristics.
[0150]
In addition, the glass part of an optical fiber means the part containing a core area | region, a 1st clad area | region, and a 2nd clad area | region except the resin-made coating | cover parts provided in the outer periphery of an optical fiber. For example, in the optical fiber shown in FIGS. 1 and 7, a portion including the core region 10, the first cladding region 20, and the second cladding region 30 is a glass portion. Further, when another cladding region made of glass is further provided on the outer periphery of the second cladding region, a portion including the cladding region becomes a glass portion.
[0151]
Moreover, about the coating | coated part provided in the outer periphery of a glass part, it is preferable that the outer diameter of a coating | coated part shall be 150 micrometers or less. Or it is preferable that the outer diameter of a coating | coated part shall be 120 micrometers or less further. Thus, by making the outer diameter of the covering portion small, the optical module can be reduced in size when the optical fiber is coiled and accommodated in the optical module as described above. Further, if the optical modules have the same size, a longer optical fiber can be coiled and accommodated in the optical module.
[0152]
Considering the characteristics of the optical fiber when applied to an optical device as a nonlinear optical fiber, it is preferable that the transmission loss is 5.0 dB / km or less in the characteristics with respect to light having a wavelength of 1.00 μm. Alternatively, it is preferable that the transmission loss is further 3.0 dB / km or less.
[0153]
By reducing the transmission loss on the short wavelength side in this way, the transmission loss at the pumping light wavelength in Raman amplification is reduced. Thus, it has good characteristics when applied to an optical device as a nonlinear optical fiber. It can be an optical fiber.
[0154]
For example, high concentration GeO by MCVD₂Addition SiO₂When an optical fiber having a large relative refractive index difference Δn is prepared by synthesizing glass, transmission loss is deteriorated due to many glass defects. Such a tendency becomes remarkable especially on the short wavelength side. On the other hand, according to the configuration of the optical fiber and the manufacturing method thereof described above, it is possible to obtain an optical fiber in which the transmission loss on the short wavelength side is sufficiently reduced. Moreover, in such an optical fiber, since the Rayleigh scattering coefficient is low, it is possible to suppress signal noise due to double Rayleigh scattering that occurs during Raman amplification.
[0155]
Considering the above conditions, an optical fiber having the double clad structure shown in FIG. 1 was produced. Its configuration is the outer diameter 2r of the core region 10₁= 4.6 μm, relative refractive index difference Δ⁺= 3.2%, outer diameter 2r of first cladding region 20₂= 13.1 μm, relative refractive index difference Δ^-= −0.50%, outer diameter of second cladding region 30 (outer diameter of glass portion of optical fiber) 2r_Three= 110 μm. Here, the concentration of F added to the second cladding region 30 was 0.6 mol%. Further, the outer diameter of the covering portion that covers the optical fiber from the outer periphery was 150 μm.
[0156]
Various characteristics for light with a wavelength of 1.55 μm are as follows:
Dispersion = −0.64 ps / km / nm,
Dispersion slope = + 0.042ps / km / nm²,
Effective area A_eff= 10.0 μm²,
Cut-off wavelength λc = 1396 nm,
Zero dispersion wavelength = 1565 nm,
Transmission loss = 0.70 dB / km,
Non-linear coefficient γ = 22.2 / W / km,
Polarization mode dispersion PMD = 0.05 ps / √km
Thus, an optical fiber (non-linear optical fiber) with good characteristics was obtained.
[0157]
The optical fiber of this example was coiled by being wound around a bobbin having a fiber length of 1.0 km and a diameter of 60 mm. FIG. 17 shows the wavelength dependence of transmission loss in such an optical fiber. Here, in the graph of FIG. 17, the horizontal axis indicates the wavelength λ (nm) of light transmitted through the optical fiber, and the vertical axis indicates the transmission loss (dB / km) at each wavelength.
[0158]
As shown in this graph, by using this optical fiber, it is possible to create a good optical module in which transmission loss does not deteriorate even in a long wavelength region. Such an optical fiber, for example, supplies pumping light having a wavelength of 1565 nm, and converts the wavelength of signal light having a wavelength band of C band into the L band, or converts the wavelength of signal light having the wavelength band of L band into the C band. It can be used in a converter module. Alternatively, it can be used in a Raman amplifier module that optically amplifies signal light by supplying pump light having a shorter wavelength than the signal light.
[0159]
Further, this optical fiber is produced based on the optical fiber manufacturing method described above with reference to FIG. 1, and has a transmission loss of 3.4 dB / km for light having a wavelength of 1.00 μm. This is a low value that satisfies the condition of 5.0 dB / km or less. Thus, according to the optical fiber having a low transmission loss on the short wavelength side, the transmission loss at the pumping light wavelength in the Raman amplification on the short wavelength side than the signal light is reduced. Further, in such an optical fiber, the Rayleigh scattering coefficient is low, so that noise generated by double Rayleigh scattering can be suppressed.
[0160]
Moreover, the optical fiber which has the double clad structure shown in FIG. 1 as another optical fiber was created. Its configuration is the outer diameter 2r of the core region 10₁= 2.5 μm, relative refractive index difference Δ⁺= 2.9%, outer diameter 2r of the first cladding region 20₂= 10.0 μm, relative refractive index difference Δ^-= −0.50%, outer diameter of second cladding region 30 (outer diameter of glass portion of optical fiber) 2r_Three= 89 μm. Here, the concentration of F added to the second cladding region 30 was 0.6 mol%. Further, the outer diameter of the covering portion that covers the optical fiber from the outer periphery was 115 μm.
[0161]
Various characteristics for light with a wavelength of 1.55 μm are as follows:
Dispersion = -110.6 ps / km / nm,
Dispersion slope = -0.408 ps / km / nm²,
Effective area A_eff= 10.6 μm²,
Cut-off wavelength λc = 729 nm,
Transmission loss = 0.52 dB / km,
Non-linear coefficient γ = 20.0 / W / km,
Polarization mode dispersion PMD = 0.03 ps / √km
Thus, an optical fiber (non-linear optical fiber) with good characteristics was obtained.
[0162]
This optical fiber has a negative dispersion and a dispersion slope. As a result, the present optical fiber is a highly nonlinear optical fiber capable of compensating both the dispersion and dispersion slope of a single mode optical fiber having a zero dispersion wavelength in the 1.3 μm band in the 1.55 μm band. Yes.
[0163]
Further, in this optical fiber, the outer diameter of the glass portion is set to a small outer diameter value of 89 μm that satisfies the condition of 100 μm or less, or even 90 μm or less. In addition, the outer diameter of the covering portion is set to a small outer diameter value of 115 μm that satisfies the condition of 150 μm or less, or even 120 μm or less. As a result, the optical fiber has good bending characteristics when coiled.
[0164]
The optical fiber of this example was coiled with a fiber length of 7.7 km to make a module. However, in the optical fiber coiling, the optical fiber is not wound around the bobbin, but the optical fiber F is coiled without being wound around the bobbin as shown in FIG. A configuration in which the fiber bundle was covered with the coating resin R was used.
[0165]
According to such a configuration, since there is no bobbin for winding the optical fiber, no winding tension is generated, and since the entire fiber bundle is covered with the resin, there is no problem of distortion due to the weight of the optical fiber. . Therefore, it is possible to greatly suppress the deterioration of transmission loss due to microbending.
[0166]
The wavelength dependence of transmission loss in such an optical fiber is shown in FIG. Here, in the graph of FIG. 19, the horizontal axis indicates the wavelength λ (nm) of light transmitted through the optical fiber, and the vertical axis indicates the transmission loss (dB / km) at each wavelength.
[0167]
As shown in this graph, by using the configuration of the present optical fiber and the above-described optical fiber coil, it is possible to create a good optical module in which transmission loss does not deteriorate even in a long wavelength region. As for the temperature characteristics of the optical fiber, the variation in transmission loss is ± 0.01 dB / km in the temperature range of −40 ° C. to + 80 ° C. with respect to light having a wavelength of 1620 nm whose transmission characteristics are most affected by the temperature variation. As a result, favorable temperature characteristics were obtained. On the other hand, in the shape wound around the conventional bobbin, the winding tension applied to the optical fiber changes due to the thermal expansion of the bobbin, so that the temperature characteristic on the long wavelength side is likely to be defective.
[0168]
Further, this optical fiber is prepared based on the optical fiber manufacturing method described above with reference to FIG. 1, and the transmission loss for light with a wavelength of 1.00 μm is 2.1 dB / km. This is a low value that satisfies the condition of 5.0 dB / km or less, or even 3.0 dB / km or less. Thus, according to the optical fiber having a low transmission loss on the short wavelength side, the transmission loss at the pumping light wavelength in the Raman amplification on the short wavelength side than the signal light is reduced. Further, in such an optical fiber, the Rayleigh scattering coefficient is low, so that noise generated by double Rayleigh scattering can be suppressed.
[0169]
Further, according to the present optical fiber, the dispersion and dispersion slope of a single-mode optical fiber having a fiber length of 50 km having a zero dispersion wavelength in the 1.3 μm band can be compensated in the 1.55 μm band.
[0170]
Furthermore, an optical fiber having the double clad structure shown in FIG. 1 was prepared as another optical fiber. Its configuration is the outer diameter 2r of the core region 10₁= 2.2 μm, relative refractive index difference Δ⁺= 3.2%, outer diameter 2r of first cladding region 20₂= 8.8 μm, relative refractive index difference Δ^-= -0.60%. Here, the concentration of F added to the second cladding region 30 was 0.6 mol%.
[0171]
Various characteristics for light with a wavelength of 1.55 μm are as follows:
Dispersion = −205.7 ps / km / nm,
Dispersion slope = -1.35ps / km / nm²,
Effective area A_eff= 10.1 μm²,
Cut-off wavelength λc = 707 nm,
Transmission loss = 0.51 dB / km,
Non-linear coefficient γ = 21.7 / W / km,
Polarization mode dispersion PMD = 0.01 ps / √km
Thus, an optical fiber (non-linear optical fiber) with good characteristics was obtained.
[0172]
This optical fiber has the following characteristics for light with a wavelength of 1.50 μm.
Dispersion = -147.4 ps / km / nm,
Dispersion slope = -0.696 ps / km / nm²,
Effective area A_eff= 8.6 μm²,
Transmission loss = 0.58 dB / km,
Non-linear coefficient γ = 24.0 / W / km,
Polarization mode dispersion PMD = 0.01 ps / √km
Have
[0173]
This optical fiber has a negative dispersion and a dispersion slope. As a result, the present optical fiber is a highly nonlinear optical fiber capable of compensating both the dispersion and dispersion slope of a single mode optical fiber having a zero dispersion wavelength in the 1.3 μm band in the 1.50 μm band. Yes. Therefore, for example, it is possible to supply pumping light having a wavelength of 1.40 μm band and use it as an optical fiber for Raman amplification.
[0174]
Next, as an example of an optical device (or an optical module obtained by modularizing the optical device) that can apply the optical fiber having the above-described configuration and various characteristics as a nonlinear optical fiber, a Raman amplifier that is an optical amplifier, and The wavelength converter will be described.
[0175]
FIG. 20 is a block diagram showing an embodiment of a Raman amplifier according to the present invention. This Raman amplifier 100 optically amplifies input signal light having a wavelength λs, a Raman amplification optical fiber 110 (cut-off wavelength λc) in which the above-described optical fiber is applied as a nonlinear optical fiber, and a predetermined wavelength. A pumping light source 150 that supplies pumping light of λp to the Raman amplification optical fiber 110 is provided.
[0176]
The excitation light source 150 is connected to an optical transmission line in the Raman amplifier 100 via an optical multiplexing unit 160 on the downstream side of the Raman amplification optical fiber 110. Accordingly, the Raman amplifier 100 is configured as an optical amplifier of backward pumping (reverse pumping). Thus, the input signal light is optically amplified using the stimulated Raman effect, which is a nonlinear optical phenomenon expressed in the Raman amplification optical fiber 110, and is output as amplified light.
[0177]
Such an Raman amplifier is different from an optical amplifier such as an EDFA in that the wavelength band to be amplified is not limited, and SiO₂In the case of a system optical fiber, the amplification wavelength band is as wide as about 100 nm, which is suitable for optical amplification in broadband WDM transmission. Further, as the wavelength λp of the excitation light, a wavelength shorter than the wavelength λs of the signal light is used. For example, if signal light having a wavelength of 1.55 μm is optically amplified, excitation light having a wavelength of about 1.45 μm is used.
[0178]
In the Raman amplification optical fiber 110 applied to the Raman amplifier 100, when optically amplifying WDM signals all together, the dispersion value for the signal light of wavelength λs is +2 ps / km / nm so that four-wave mixing does not occur. It is preferable to set it above or below -2 ps / km / nm. For example, the optical fibers E1 and E2 in FIG. 14 are suitable for signal light having a wavelength of 1.55 μm.
[0179]
When the dispersion value is positive, the outer diameter 2r of the core region 10₁Therefore, the cutoff wavelength λc is slightly longer. On the other hand, in the above optical fiber having a double clad structure, the cutoff wavelength λc can be made shorter (λc <λp) than the pumping light wavelength λp of about 1.45 μm. In this way, by setting λc <λp, optical amplification can be performed with high efficiency in a single mode.
[0180]
If a nonlinear optical fiber having positive and negative dispersion values is combined, a Raman amplifier can be configured so that the dispersion becomes zero as a whole. An example of the configuration of such a Raman amplifier is shown in FIG.
[0181]
The Raman amplifier 200 has the same configuration as that of the Raman amplifier 100 shown in FIG. 20, except that the Raman amplification optical fiber 110 is a non-linear optical fiber having a negative dispersion value (for example, −2 ps / km / nm or less). In addition, a Raman amplification optical fiber 120 having a positive dispersion value (for example, +2 ps / km / nm or more) is connected in series between the Raman amplification optical fiber 110 and the optical multiplexing unit 160. According to such a configuration, the dispersion of the output amplified light can be made substantially zero.
[0182]
In addition, signal light having a wavelength of 1.45 μm to 1.53 μm called S band cannot be optically amplified by the EDFA, but if the Raman amplifier does not select the wavelength band to be excited, the wavelength λs is 1.45 μm. Optical amplification is possible even for signal light of 1.53 μm or less. Further, as described above, in the double clad structure, the cut-off wavelength λc can be shortened, for example, as in the optical fiber E5 in FIG. 14, and therefore, it is preferably applied to the optical amplification of S-band signal light. Is possible. The dispersion value of the optical fiber E5 at a wavelength of 1.40 μm is a preferable range of −6.1 ps / km / nm.
[0183]
Also, if the dispersion value of the optical transmission line is positive within the signal light wavelength band to be used, if the dispersion value of the Raman amplification optical fiber used for the Raman amplifier is set to be negative, the positive dispersion value is set simultaneously with the optical amplifier. It can also be used as a dispersion compensator for a transmission line. At this time, if the dispersion value with respect to the signal light having the wavelength λs is −10 ps / km / nm or less, the dispersion compensation amount is large, and the dispersion compensator can be used particularly suitably. At this time, the effective area A_eff10 μm²The following is preferable.
[0184]
Further, in the nonlinear optical fiber having the double clad structure, the dispersion slope is set to a negative value (0 ps / km) at the wavelength of the signal light as in the optical fibers E3 and E4 in FIG. 14 and the optical fiber F1 in FIG. / Nm²Smaller value). In this case, the dispersion slope can be compensated simultaneously with the dispersion of the transmission line having the positive dispersion and the positive dispersion slope. Therefore, it is suitable for WDM transmission.
[0185]
Here, in order to realize high-efficiency Raman amplification, it is preferable that the nonlinear optical fiber used in the Raman amplifier has high nonlinearity at the wavelength λp of the pumping light. Moreover, in order to prevent the deterioration of transmission quality due to the nonlinear effect, it is preferable that the nonlinearity at the wavelength λs of the signal light is low.
[0186]
In order to realize such a characteristic condition for nonlinearity, the nonlinear optical fiber used in the optical amplifier has an effective area A at the wavelength λp of the pumping light._{eff, p}And effective area A at wavelength λp + 0.1 μm_{eff, s}And the relational expression
(A_{eff, s}-A_{eff, p}) / A_{eff, p}× 100 ≧ 10%
And effective area A_{eff, s}Is effective sectional area A_{eff, p}It is preferable that it is 10% or more larger than
[0187]
A wavelength λp + 0.1 μm obtained by adding 0.1 μm to the wavelength λp of the excitation light corresponds to the wavelength λs of the signal light that is optically amplified in the Raman amplifier. Therefore, according to the characteristic condition satisfying the above relational expression, the effective area A_{eff, p}By reducing the above, it is possible to improve the nonlinearity at the wavelength λp with respect to the excitation light, and to improve the efficiency of optical amplification. Effective area A_{eff, s}By increasing the non-linearity at the wavelength λp + 0.1 μm with respect to the signal light, the transmission quality of the signal light can be prevented from deteriorating.
[0188]
For example, the amount of phase shift due to self-phase modulation is proportional to the reciprocal of the effective area. Therefore, the effective cross-sectional area A at the wavelength λs to λp + 0.1 μm of the signal light._{eff, s}Is the effective area A at the wavelength λp of the excitation light_{eff, p}If it is 10% larger than that, the phase shift amount is 10% smaller.
[0189]
This effective area A_effThe optical fiber having the double clad structure shown in FIG. Its configuration is the outer diameter 2r of the core region 10₁= 3.1 μm, relative refractive index difference Δ⁺= 3.4%, outer diameter 2r of the first cladding region 20₂= 8.8 μm, relative refractive index difference Δ^-= -0.15%. Here, the concentration of F added to the second cladding region 30 was 1.1 mol%.
[0190]
Various characteristics for light with a wavelength of 1.55 μm are as follows:
Dispersion = −49.0 ps / km / nm,
Dispersion slope = +0.005 ps / km / nm²,
Effective area A_eff= 8.4μm²,
Cut-off wavelength λc = 1060 nm,
Transmission loss = 0.54 dB / km,
Non-linear coefficient γ = 23.4 / W / km,
Polarization mode dispersion PMD = 0.02 ps / √km
Met.
[0191]
FIG. 22 shows the effective area A in the optical fiber of this example._effThe wavelength dependence of is shown. Here, in the graph of FIG. 22, the horizontal axis represents the wavelength λ (nm) of the light transmitted through the optical fiber, and the vertical axis represents the effective area A at each wavelength._eff(Μm²). As shown in this graph, in this optical fiber, the effective area A_effIncreases as the wavelength λ increases.
[0192]
For example, when the excitation light having the wavelength λp = 1.40 μm is used for the wavelength λs = 1.50 μm of the signal light, the effective cross-sectional areas for the signal light and the excitation light are respectively
Signal light: A_{eff, s}= 7.85 μm²
Excitation light: A_{eff, p}= 6.93 μm²
It has become. At this time, the difference in effective area at wavelengths λs and λp is
(A_{eff, s}-A_{eff, p}) / A_{eff, p}× 100 = 13.3%
It is.
[0193]
Further, when the excitation light having the wavelength λp = 1.45 μm is used for the wavelength λs = 1.55 μm of the signal light, the effective cross-sectional areas for the signal light and the excitation light are respectively
Signal light: A_{eff, s}= 8.37 μm²
Excitation light: A_{eff, p}= 7.37 μm²
It has become. At this time, the difference in effective area at wavelengths λs and λp is
(A_{eff, s}-A_{eff, p}) / A_{eff, p}× 100 = 13.6%
It is.
[0194]
In addition, when the excitation light having the wavelength λp = 1.50 μm is used for the wavelength λs = 1.60 μm of the signal light, the effective cross-sectional areas for the signal light and the excitation light are respectively
Signal light: A_{eff, s}= 8.93 μm²
Excitation light: A_{eff, p}= 7.85 μm²
It has become. At this time, the difference in effective area at wavelengths λs and λp is
(A_{eff, s}-A_{eff, p}) / A_{eff, p}× 100 = 13.8%
It is.
[0195]
As described above, in the present optical fiber, suitable characteristic conditions for any of signal light with wavelengths λs = 1.50 μm, 1.55 μm, and 1.60 μm.
(A_{eff, s}-A_{eff, p}) / A_{eff, p}× 100 ≧ 10%
Is satisfied. Therefore, a nonlinear optical fiber and a Raman amplifier in which the efficiency of optical amplification is improved and the deterioration of the transmission quality of the signal light is suppressed with respect to light in a wavelength range including these wavelengths is realized.
[0196]
FIG. 23 is a block diagram showing an embodiment of a wavelength converter according to the present invention. The wavelength converter 300 converts the wavelength of the input signal light having the wavelength λs, and includes a wavelength conversion optical fiber 310 (cutoff wavelength λc) in which the above-described optical fiber is applied as a nonlinear optical fiber; A pumping light source 350 that supplies pumping light having a predetermined wavelength λp to the wavelength conversion optical fiber 310 is provided.
[0197]
The excitation light source 350 is connected to an optical transmission line in the wavelength converter 300 via an optical multiplexing unit 360 on the upstream side of the wavelength conversion optical fiber 310. As a result, the input signal light having the wavelength λs is wavelength-converted using four-wave mixing, which is a nonlinear optical phenomenon expressed in the wavelength conversion optical fiber 310, and the wavelength λs ′ is transmitted via the wavelength selection unit 370.
λs ′ = λp− (λs−λp)
Is output as converted light (see FIG. 24A).
[0198]
Such a wavelength converter can individually or collectively convert WDM signals having a high transmission rate per channel. Further, in the nonlinear optical fiber having the double clad structure, the nonlinear coefficient γ is set with the cut-off wavelength λc kept short as in the optical fibers E6 and E8 in FIG. 14 and the optical fiber F3 in FIG. It is possible to perform wavelength conversion with high efficiency by making it sufficiently large. In particular, if the cut-off wavelength λc is shorter than the wavelengths λs, λs ′, and λp of the signal light, converted light, and pump light (λc <λs, λs ′, λp), the wavelength conversion is highly efficient in a single mode. It can be performed.
[0199]
Here, four-wave mixing is likely to occur when the phases of the signal light, the excitation light, and the converted light are matched, so that the dispersion value in the excitation light of the wavelength λp is −0.2 ps / km / nm or more +0. The range of 2 ps / km / nm or less is preferable, and it is particularly preferable that the excitation light wavelength λp substantially coincides with the zero dispersion wavelength. If the power of the pumping light is increased, the optical power of the output converted light can be made larger than the optical power of the input signal light. In this case, the wavelength converter is also used as a parametric amplifier. be able to.
[0200]
In the wavelength conversion from the C band to the S band, it is desirable that the zero-dispersion wavelength is around 1.53 μm and the cutoff wavelength λc is shorter than the wavelength λs ′ of the converted light. In the linear optical fiber, such a characteristic condition can be realized as in an optical fiber E7 in FIG. 14, for example.
[0201]
Further, if the excitation light source 350 is a variable wavelength light source and the wavelength λp of the excitation light is changed, arbitrary wavelength conversion is possible. For example, in the example of FIG. 24B, the pumping light wavelength is λp1 with respect to the signal light having the wavelength λs, and the wavelength λs1 ′.
λs1 ′ = λp1− (λs−λp1)
The converted light is obtained. On the other hand, as shown in FIG. 24C, when the excitation light wavelength is changed to λp2, the wavelength λs2 'different from the wavelength λs1' is obtained.
λs2 ′ = λp2− (λs−λp2)
The converted light can be obtained. Also in this case, in order to match the phase, the dispersion value with respect to the excitation light wavelength is preferably in the range of −0.2 ps / km / nm to +0.2 ps / km / nm.
[0202]
Also, when performing C-band Raman amplification, the excitation light has a wavelength of around 1.45 μm, and when performing S-band Raman amplification, the excitation light has a wavelength of 1.3 to 1.4 μm and the signal light has a wavelength of 1 .45 to 1.53 μm. Further, when performing wavelength conversion to the S band or wavelength conversion from the S band to the C and L bands, the signal light and the converted light have a wavelength of 1.45 to 1.53 μm. In these cases, it is susceptible to absorption loss at a wavelength of 1.38 μm due to OH groups. On the other hand, as described above, the optical fibers E1 to E8 in FIG. 14 and the optical fibers F1 to F3 in FIG. 16 increase the transmission loss (excess absorption loss) due to the OH group absorption at the wavelength of 1.38 μm. Are all 0.2 dB / km or less, and can be suitably used in such a case.
[0203]
【The invention's effect】
As described in detail above, the optical fiber, the nonlinear optical fiber, the optical amplifier, the wavelength converter, and the optical fiber manufacturing method according to the present invention have the following effects. That is, according to the optical fiber and the non-linear optical fiber having the above-described configuration using a double clad structure instead of a single clad structure, GeO added to the core in order to increase the non-linear coefficient γ.₂To increase the non-linear refractive index and increase the relative refractive index difference between the core and the cladding to increase the effective area A_effEven when is made smaller, the cutoff wavelength λc can be made sufficiently short. In this configuration, the dispersion slope can be negative. Further, a highly nonlinear polarization-maintaining fiber and a highly nonlinear optical fiber with low polarization mode dispersion and low transmission loss can be obtained. Moreover, if the outer diameter of the glass part or the covering part of the optical fiber is made small, an optical fiber suitable for modularization in an optical device can be obtained.
[0204]
In addition, according to the above-described optical fiber manufacturing method in which the core glass rod and the first cladding glass pipe are heated and integrated under a predetermined condition, an optical fiber having a highly nonlinear double-clad structure can be transmitted at a low transmission rate. It can be created with good characteristics such as loss.
[0205]
Such an optical fiber has high nonlinearity, and as a nonlinear optical fiber having suitable characteristics with respect to a cutoff wavelength λc and the like, an optical device using nonlinear optical phenomena such as an optical amplifier and a wavelength converter It is possible to apply to. In particular, when the cut-off wavelength λc is short, optical amplification and wavelength conversion can be performed with high efficiency in a single mode.
[Brief description of the drawings]
FIG. 1 is a diagram schematically showing a cross-sectional configuration and a refractive index profile of a first embodiment of an optical fiber.
FIG. 2 is a table showing the dependence of the number of bubbles generated on the heating temperature.
FIG. 3 is a table showing the dependence of the number of bubbles generated on the baking temperature.
FIG. 4 is a table showing the dependence of the number of bubbles generated on the surface roughness of the first cladding glass pipe.
FIG. 5 is a table showing the dependence of the number of bubbles generated on the surface roughness of the core glass rod.
FIG. 6 shows GeO in a region within 2 μm in thickness from the outer peripheral surface of the core glass rod with the number of bubbles generated.₂It is a table | surface which shows the dependence to a density | concentration.
FIG. 7 is a diagram schematically showing a cross-sectional structure and a refractive index profile of an optical fiber according to a second embodiment.
FIG. 8 is a diagram showing a refractive index profile of optical fibers A1 and A2.
9 is a table showing various characteristics of the optical fiber shown in FIG.
FIG. 10 is a diagram showing refractive index profiles of optical fibers B1, B2, C1, and C2.
FIG. 11 is a diagram showing a refractive index profile of optical fibers D1 to D5.
12 is a table showing various characteristics of the optical fiber shown in FIG.
13 is a table showing various characteristics of the optical fiber shown in FIG.
FIG. 14 is a table showing various characteristics of optical fibers E1 to E8.
FIG. 15 is a diagram schematically showing a cross-sectional structure of another embodiment of an optical fiber.
FIG. 16 is a table showing various characteristics of optical fibers F1 to F3.
FIG. 17 is a graph showing the wavelength dependence of transmission loss of an optical fiber.
FIG. 18 is a diagram schematically showing a configuration of an optical fiber coil.
FIG. 19 is a graph showing the wavelength dependence of transmission loss of an optical fiber.
FIG. 20 is a configuration diagram illustrating an embodiment of a Raman amplifier.
FIG. 21 is a configuration diagram showing another embodiment of a Raman amplifier.
FIG. 22 is a graph showing the wavelength dependence of the effective cross-sectional area of an optical fiber.
FIG. 23 is a block diagram showing an embodiment of a wavelength converter.
24 is a diagram schematically showing wavelength conversion by the wavelength converter shown in FIG. 23. FIG.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 ... Core area | region, 15 ... Middle area | region, 20 ... 1st clad area | region, 30 ... 2nd clad area | region, 40 ... Stress application part, 50 ... Hermetic coat,
DESCRIPTION OF SYMBOLS 100, 200 ... Raman amplifier, 110, 120 ... Raman amplification optical fiber, 150 ... Excitation light source, 160 ... Optical multiplexing part,
DESCRIPTION OF SYMBOLS 300 ... Wavelength converter, 310 ... Optical fiber for wavelength conversion, 350 ... Excitation light source, 360 ... Optical multiplexing part, 370 ... Wavelength selection part.

Claims (23)

A core region having a maximum refractive index of n ₁ , a first cladding region provided on an outer periphery of the core region and having a minimum refractive index of n ₂ (where n ₂ <n ₁ ), and the first Provided at the outer periphery of the cladding region and having at least a second cladding region having a maximum refractive index of n ₃ (where n ₂ <n ₃ <n ₁ ), and various characteristics with respect to light having a wavelength of 1.55 μm,
An effective area of 11 μm ² or less;
A cutoff wavelength λc of 0.7 μm or more and 1.6 μm or less at a fiber length of 2 m;
A nonlinear coefficient of 18 / W / km or more;
Have
Wherein the first cladding region, the relative refractive index difference between the second cladding region delta ^- is, Ri said second -0.15% der less -0.70% or more cladding region as a reference,
Said core region, said second relative refractive index difference between the cladding region delta ⁺ is, optical fibers, wherein said second der Rukoto 2.7% or more cladding region as a reference. As various characteristics for light with a wavelength of 1.55 μm,
A transmission loss of 3.0 dB / km or less;
Crosstalk between polarizations of -15 dB or less;
The optical fiber according to claim 1, further comprising: As various characteristics for light with a wavelength of 1.55 μm,
A transmission loss of 1.0 dB / km or less;
Polarization mode dispersion of 0.3 ps / √km or less,
The optical fiber according to claim 1, further comprising:   The optical fiber according to claim 1, wherein a hermetic coat is provided on an outer periphery of the second cladding region.   2. The optical fiber according to claim 1, wherein excess absorption loss due to OH groups with respect to light having a wavelength of 1.38 μm is 0.2 dB / km or less.   The optical fiber according to claim 1, wherein fluorine is added to the second cladding region.   The optical fiber according to claim 1, wherein an outer diameter of a glass portion including the core region, the first cladding region, and the second cladding region is 100 μm or less. The optical fiber according to claim 7, wherein an outer diameter of the glass portion is 90 μm or less.   2. The optical fiber according to claim 1, wherein an outer diameter of a covering portion provided on an outer periphery of a glass portion including the core region, the first cladding region, and the second cladding region is 150 μm or less. The optical fiber according to claim 9, wherein an outer diameter of the covering portion is further 120 μm or less.   2. The optical fiber according to claim 1, wherein a transmission loss is 5.0 dB / km or less in the characteristics with respect to light having a wavelength of 1.00 [mu] m. The optical fiber according to claim 11 , wherein the transmission loss is further 3.0 dB / km or less in characteristics with respect to light having a wavelength of 1.00 μm.   The non-linear optical fiber according to claim 1, wherein a non-linear optical phenomenon expressed by inputting light having a predetermined wavelength is used. The nonlinear optical fiber according to claim 13 , wherein the cutoff wavelength is λc;
A pumping light source for supplying pumping light of a predetermined wavelength λp (where λc <λp) to the nonlinear optical fiber with respect to the signal light of wavelength λs input to the nonlinear optical fiber;
An optical amplifier characterized by optically amplifying the signal light by utilizing a non-linear optical phenomenon expressed in the non-linear optical fiber. The optical amplifier according to claim 14 , wherein a dispersion value of the nonlinear optical fiber with respect to the signal light having a wavelength λs is +2 ps / km / nm or more or −2 ps / km / nm or less. The optical amplifier according to claim 14 , wherein a dispersion value of the nonlinear optical fiber with respect to the signal light having a wavelength λs is -10 ps / km / nm or less and an effective area is 10 μm ² or less. . The optical amplifier according to claim 16 , wherein a dispersion slope value of the nonlinear optical fiber with respect to the signal light is smaller than 0 ps / km / nm ² . The optical amplifier according to claim 14, wherein the wavelength λs of the signal light is 1.45 μm or more and 1.53 μm or less. In the nonlinear optical fiber, the effective area A _{eff, p at} the wavelength λp of the excitation light and the effective area A _{eff, s at} the wavelength λp + 0.1 μm are expressed by a relational expression (A _{eff, s} −A _{eff, p} )
/ A _{eff, p} × 100 ≧ 10%
The optical amplifier according to claim 14, wherein: The nonlinear optical fiber according to claim 13 , wherein the cutoff wavelength is λc;
A pumping light source for supplying pumping light having a predetermined wavelength λp (where λc <λp) to the nonlinear optical fiber with respect to signal light having a wavelength λs (where λc <λs) input to the nonlinear optical fiber; With
Wavelength conversion characterized in that the signal light is wavelength-converted by using a nonlinear optical phenomenon expressed in the nonlinear optical fiber, and converted light having a wavelength λs ′ (where λc <λs ′) is output. vessel. 21. The wavelength converter according to claim 20, wherein the optical power of the output converted light is larger than the optical power of the input signal light. 21. The wavelength converter according to claim 20 , wherein a dispersion value of the nonlinear optical fiber with respect to the excitation light having a wavelength [lambda] p is -0.2 ps / km / nm or more and +0.2 ps / km / nm or less. 21. The wavelength converter according to claim 20, wherein a wavelength λs ′ of the converted light is 1.45 μm or more and 1.53 μm or less.

Images