JP2013074187A - Mode synchronous semiconductor laser device and control method for mode synchronous semiconductor laser - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make configuration simple while keeping the capability of maintaining optical injection synchronization at the same level as before.SOLUTION: The mode synchronous semiconductor laser device includes a mode synchronous semiconductor laser element 1 which contains a core 30 equipped with an optical gain region 3, an optical modulation region 2, and a passive waveguide region 4 and an optical waveguide 40 containing clads 5 and 6, and in which continuous wave light CW capable of causing optical injection synchronization is injected, for outputting optical pulse train Lcontaining vertical mode whose wavelength is equal to the continuous wave light, separation means 60 which separates first and second optical components L1 and L2 contained in the optical pulse train so that its intensity ratio reflects a ratio of main longitudinal mode relative to total light intensity of the optical pulse train, and control means 50 which controls the main longitudinal mode to be a wavelength capable of maintaining optical injection synchronization by a control index that uses optical intensity of the second optical component.

Description

  The present invention relates to a light injection locking mode-locked semiconductor laser device in which the wavelength of an output light pulse train is variable, and a control method for the device.

  In recent years, the demand for communication is rapidly increasing due to the spread of the Internet and the like, and a large-capacity and high-speed optical communication network using an optical fiber is being developed correspondingly. In general, an optical signal of a long-distance optical communication network gives RZ (Return to) which gives an intensity level corresponding to “1” and “0” to an optical pulse train of a predetermined repetition frequency (hereinafter also referred to as pulse frequency). Zero) format is used. In the RZ format in which the intensity is once zero between pulses corresponding to consecutive “1”, deterioration of the time waveform during long distance transmission can be suppressed.

  In order to perform high-speed communication with an RZ format optical signal (hereinafter also referred to as an RZ optical signal), it is effective to increase the pulse frequency. For this purpose, it is necessary to increase the pulse frequency in the optical pulse light source that generates the optical pulse train that is the source of the RZ optical signal. Here, the optical pulse train is an optical signal in which optical pulses are arranged at equal time intervals on the time axis at time intervals given by the reciprocal of the pulse frequency.

  For high-capacity and high-speed optical communication, optical pulse light sources are required not only to increase the pulse frequency, but also to have the ability to generate high-quality optical pulse trains with low frequency chirping and phase noise. As one of such optical pulse light sources, a mode-locked laser diode (MLLD) that satisfies the above-described capabilities, such as being capable of obtaining a pulse frequency exceeding 10 GHz, and is more compact, robust, and low-cost is proposed. Can be mentioned. Note that MLLD indicates a semiconductor laser having a configuration suitable for mode-locking operation.

  On the other hand, in order to increase the capacity of an optical communication network, an optical multiplexing technique is used in which optical signals of a plurality of channels are transmitted together on a single optical fiber. As one of the optical multiplexing techniques, a wavelength division multiplexing (WDM) that assigns a plurality of different wavelengths for each channel is widely used.

  In order to combine this WDM with an optical pulse light source composed of an MLLD, the MLLD is further required to have a so-called wavelength variable characteristic that changes the wavelength of the optical pulse train as desired.

  Furthermore, in recent WDM, so-called Dense-WDM, in which the wavelength difference between adjacent channels is narrowed, has begun to be used in order to further increase the communication capacity. In Dense-WDM, in addition to the above-described wavelength tunable characteristics, in the optical pulse light source, in order to prevent undesired interaction between adjacent channels, the fluctuation of the center wavelength of the optical signal is suppressed to within a few GHz in terms of frequency. High wavelength stability is required.

  Among the wavelength tunable characteristics and wavelength stability required for MLLD, several techniques have been proposed, particularly focusing on the wavelength tunable characteristics. In one technique, an optical bandpass filter such as a dielectric multilayer filter or a diffraction grating is provided in the laser resonator. That is, the wavelength variable characteristic is obtained by mechanically or electrically changing the center wavelength of the band selected by the optical bandpass filter (see, for example, Non-Patent Document 1). In another technique, the temperature of the distributed Bragg reflector region provided in the integrated MLLD is controlled. That is, the Bragg reflection wavelength of the distributed Bragg reflector is changed by the refractive index change caused by Joule heat when the electric resistance layer provided in the distributed Bragg reflector region is energized to obtain a wavelength variable characteristic (for example, non-patent literature) 2).

  However, the MLLDs of Non-Patent Documents 1 and 2 cannot be said to have sufficient wavelength stability of the output optical pulse train, and are insufficient as optical pulse light sources for Dense-WDM.

  Under such a technical background, the inventor has already proposed MLLD that simultaneously achieves wavelength tunable characteristics and wavelength stability (see, for example, Non-Patent Document 3). Schematically, the technique of Non-Patent Document 3 is a Fabry-Perot (FP) resonator type MLLD that utilizes light injection locking generated by continuous wave light injection. Here, “continuous wave light” is light in which the amplitude intensity of a photoelectric field having a constant magnitude continues without fluctuation of the light intensity over time.

  The technology of Non-Patent Document 3 solves the frequency chirping that is a problem in the conventional FP resonator type MLLD. Further, the technique of Non-Patent Document 3 appropriately adjusts the light intensity of the continuous wave light to be injected for problems such as a decrease in the intensity of the output light pulse train and an increase in the pulse time width that tend to occur when frequency chirping is suppressed. It is solved by that. Furthermore, in the MLLD according to the technique of Non-Patent Document 3, the wavelengths of the output light pulse train and the continuous wave light exactly match. Therefore, if a continuous wave light source having high wavelength stability and variable wavelength is used, wavelength tunable characteristics and wavelength stability can be achieved simultaneously in a wide wavelength range. Thus, according to MLLD of the technique of nonpatent literature 3, the optical pulse light source especially suitable for Dense-WDM is obtained.

Here, the light injection synchronization will be described with reference to FIGS. FIG. 1A is a schematic diagram for explaining the light injection synchronization. FIG. 1 (B), in MLLD that the mode-locking operation in injection without a continuous wave light, laser oscillation light propagating inside (hereinafter, the oscillation light also referred to.) Is an oscillation spectrum of L _A. FIG. 1 (C) is a continuous wave light CW injection of wavelength lambda _CW, an oscillation spectrum of the optical pulse train L _B for optical injection locking is outputted from MLLD that the mode-locked operations when they occur. FIG. 1D is a schematic diagram enlarging the main part of FIG. In FIGS. 1B, 1C, and 1D, the horizontal axis represents wavelength (arbitrary unit), and the vertical axis represents light intensity (arbitrary unit). Further, in FIGS. 1B and 1C, straight lines arranged at intervals of the mode period wavelength Λ each indicate a longitudinal mode of the oscillation spectrum. Since the half width of one longitudinal mode is very narrow, it is drawn as a straight line ignoring the half width in the figure.

According to FIG. 1 (A), the schematically, an optical injection locking and is to MLLD the oscillation light L _A is propagated through the mode-locking operation, when injecting the continuous wave light CW of wavelength lambda _CW, the output light It indicates a phenomenon in which the wavelength of the pulse train L _B is equal to the wavelength lambda _CW.

More particularly, referring to FIG. 1 (B), the case where the optical injection locking does not occur, the oscillation light L _A, the intensity is about the maximum longitudinal mode LM _p, a plurality of longitudinally distributed in wide bell Includes modes. Here, the envelope of all the longitudinal modes oscillating light L _A contains called en _A. The present invention is not limited to the oscillation light L _A, the light including a plurality of longitudinal modes, the envelope, the curve connecting the intensity peaks of the total longitudinal modes.

This MLLD, the wavelength lambda _CW included in the wavelength range of the entire longitudinal mode of oscillation light L _A, and injecting the continuous wave light CW oscillation light L _A and the polarizing direction coincides. The longitudinal mode wavelength nearest to the continuous wave light CW in the longitudinal mode included in the oscillation light L _A (hereinafter, also referred to as the target longitudinal mode.) And LM, appropriately maintain that the wavelength difference δλ the continuous wave light CW As a result, light injection synchronization occurs. Here, the concept of the wavelength difference δλ is shown in FIG. As can be seen from FIG. 1D, when the wavelength of the target longitudinal mode LM is λ _LM , the wavelength difference δλ is expressed as λ _CW −λ _LM .

Thus, as shown in FIG. 1 (C), the optical pulse train _{L B} is outputted from the MLLD, wavelength lambda _LM target longitudinal mode LM is exactly the same with the wavelength lambda _CW continuous wave light CW. Further, the optical injection locking, the optical pulse train L _B, the light intensity is concentrated to longitudinal mode wavelength lambda _CW. That is, the oscillation spectrum of the optical pulse train L _B is narrowed as compared with the oscillation light L _A, significantly smaller than the envelope en _A of FWHM oscillation light L _A of the envelope en _B. Here, in the optical pulse train L _B, referred equally adjusted target longitudinal mode LM to the wavelength lambda _CW main longitudinal mode LM. Also, most strength is symmetrically distributed around the strong main longitudinal mode, a plurality of longitudinal modes other than the main longitudinal mode LM referred従縦mode L _n.

Subsequently, the light injection locking will be described more specifically with reference to the oscillation spectrum of FIG. Figure 2 (A) ~ (F) is a continuous wave light CW with varying wavelength difference δλ above, obtained by injecting the non-patent document 3 MLLD, the oscillation spectrum of the output optical pulse train L _B with envelope FIG. 2A to 2F, the vertical axis represents the light intensity (dBm) based on the total light intensity, and the horizontal axis represents the wavelength (nm).

In more detail, in obtaining FIG. 2, FIG. The experiment was conducted according to No. 10. That is, as the target longitudinal mode, a longitudinal mode having a wavelength λ _LM of 15562.01 nm was selected, and continuous wave light CW in which the wavelength λ _CW was varied between 1561.186 and 1562.16 nm was injected into the MLLD. The wavelength λ _CW of the continuous wave light CW corresponds to a wavelength difference δλ of −0.15 to 0.15 nm. 2A to 2F show the light intensities when δλ = −0.1, 0, 0.05, 0.07, 0.09, and 0.15 nm, respectively. 2A to 2F, the mode period wavelength Λ is approximately 0.31 nm and is substantially constant.

  2A to 2F, when the wavelength difference δλ is 0.05 nm (FIG. 2C) and the wavelength difference δλ is 0.07 nm (FIG. 2D), half of the envelope The value width is narrowed and the oscillation spectrum is narrowed. From this result, it can be seen that the light injection locking is maintained when the wavelength difference δλ is within a range of 0.05 to 0.07 nm, that is, a narrow wavelength range with a width of 0.02 nm. The wavelength width of 0.02 nm corresponds to about 2.5 GHz in terms of optical frequency. An optical pulse train output in a state where optical injection locking is maintained is particularly suitable for dense-WDM because frequency chirping is greatly suppressed.

H. F Liu, et al. , “Generation of Wavelength-Tunable Transform-Limited Pulses from a Monovalent Passive Mode-locked Distributed Bragg ReflectorEducatorEconomical. Technol. Lett. , Vol. 7, no. 10, pp. 1139-1141, 1995. M.M. C. Wu, Y. K. Chen, T.A. Tanbun-Ek, R.A. A. Logan, and M.M. A. Chin, “Tunable monolithic colliding pulse mode-locked quantum-well lasers,” IEEE Photon. Technol. Lett. , Vol. 3, No. 10, pp. 874-976, 1991. S. Arahira, et al. , “Chirp control and broadband-wavelength-tuning of 40-GHz monolithically active mode-locked laser diode with an e-Lit. 11, no. 5, pp. 1103-1111, 2005.

  As is apparent from the description of FIG. 2 and Non-Patent Document 3 (particularly, FIGS. 10 and 11 and the right column on page 1108), the wavelength range allowed for the wavelength difference δλ (0. 02 nm) is very narrow. Therefore, when an optical pulse train is output using light injection locking, the MLLD device is required to have a capability of keeping the wavelength difference δλ within this allowable wavelength range. Here, if this allowable wavelength range is a wavelength synchronization range (WLR), the WLR is schematically represented as shown in FIG.

By the way, from Non-Patent Document 3, FIG. It is suggested that I ₄ / I ₀ (hereinafter also referred to as a master-slave ratio), which is an index described in No. 11, is effective in maintaining light injection locking. I ₀ is the intensity of the main longitudinal mode of the output optical pulse train. I ₄ indicates the intensity of the fourth subordinate mode counted from the main longitudinal mode in the direction of increasing wavelength. The master-slave ratio I ₄ / I ₀ reflects the degree of narrowing, that is, the width of the envelope of the output optical pulse train.

As described above, when the light injection synchronization occurs, the envelope of the output optical pulse train is narrowed. Therefore, the master-slave ratio I ₄ / I ₀ is the degree of the light injection synchronization occurring in the MLLD (hereinafter referred to as the synchronization intensity). May be represented). This point will be further described with reference to FIG. FIG. 3 is a characteristic diagram showing the master-slave ratio obtained from FIGS. The vertical axis in FIG. 3 is the master-slave ratio (I ₊₅ / I ₀ ) (dB), and the horizontal axis is the wavelength difference δλ (nm). In the master-slave ratio in FIG. 3, the light intensity I ₊ 5 of the fifth slave mode, counted from the main longitudinal mode in the direction in which the wavelength increases, is used as the numerator.

Referring to FIG. 3, the master-slave ratio (I ₊₅ / I ₀ ) has a valley in the wavelength difference δλ range (hereinafter referred to as the recess range A1). Particularly, in the range where the wavelength difference δλ is 0.05 to 0.07 nm (hereinafter referred to as the minimum range A2), the master-slave ratio takes a minimum of −27 to −30 dB. When the wavelength difference δλ is in the range of −0.15 to −0.05 nm and 0.09 to 0.15 nm, that is, in a range other than the recess range (hereinafter referred to as a constant value range A3), the master-slave ratio is a constant value ( Take about -3 dB).

2 and 3, it can be seen that the master-slave ratio reflects the synchronization strength very well. That is, when the wavelength difference δλ is (1) within the constant value range A3, the light injection synchronization is not developed, and (2) when within the recess range A1, the master-slave ratio changes according to the synchronization intensity, and (3 ) In the minimum range A2, the light injection synchronization is maintained well. Therefore, as an index master-slave ratio of the output optical pulse train, the main longitudinal mode wavelength lambda _CW output optical pulse train _{L B,} by performing the control to maintain the wavelength difference δλ in WLR of 0.05～0.07Nm, optical injection locking Can be maintained.

  However, in the method of controlling using the master-slave ratio, the apparatus configuration for measuring the strengths of the main longitudinal mode and the slave longitudinal mode is complicated and expensive. For example, if an optical spectrum analyzer is used to measure the intensity of the main longitudinal mode and the subordinate longitudinal mode with sufficient accuracy, an increase in size and cost of the apparatus cannot be avoided. Further, when a wavelength separation filter is used to separate the main longitudinal mode and the slave longitudinal mode, the wavelength filter is required to have very high performance. That is, this wavelength separation filter is required to have a high wavelength resolution less than the mode period wavelength Λ (about 0.31 nm in FIG. 2) and a high spectrum suppression capability that excludes longitudinal modes other than the longitudinal mode to be measured. . As a result, even if a wavelength separation filter is used, the device configuration becomes complicated.

  The present invention has been made in view of such problems. Accordingly, an object of the present invention is to propose a mode-locked semiconductor laser device and a control method for the device that are simpler than the technology while maintaining the capability of maintaining the light injection locking as in the technology of Non-Patent Document 3. It is in.

  As a result of intensive studies, the inventors have conceived that light injection locking can be maintained in the following mode-locked semiconductor laser device as in the technique of Non-Patent Document 3. That is, the mode-locked semiconductor laser device of the present invention partitions an optical pulse train with a predetermined wavelength width centered on the main longitudinal mode wavelength, and uses a control index obtained from the intensity of wavelength components including longitudinal modes other than this wavelength width. Maintain light injection locking.

  Therefore, the mode-locked semiconductor laser device according to the present invention comprises the mode-locked semiconductor laser element, the separating means, and the control means.

  The mode-locked semiconductor laser element has an optical waveguide that includes a core through which oscillation light propagates and a cladding. Here, the core includes an optical gain region that forms an inversion distribution, an optical modulation region that modulates the light intensity, and a passive waveguide region that includes an adjustment unit that changes the wavelength of the oscillation light.

  The mode-locked semiconductor laser element receives continuous wave light and outputs an optical pulse train. Here, the input continuous wave light has a wavelength difference capable of expressing light injection synchronization with the target longitudinal mode which is one of the longitudinal modes included in the oscillation light, and oscillates. The light and the polarization direction coincide. The output optical pulse train includes a main longitudinal mode which is a longitudinal mode having a wavelength equal to that of continuous wave light, and a plurality of subordinate longitudinal modes distributed around the main longitudinal mode wavelength.

  The separating means separates the optical pulse train into first and second light components such that the intensity ratio reflects the ratio of the main longitudinal mode to the total light intensity of the optical pulse train.

  The control means obtains the control index using the light intensity of the second light component, and controls the adjustment unit using the control index, so that the wavelength difference is kept within the wavelength synchronization range in which the light injection synchronization can be maintained. As described above, the wavelength of the main longitudinal mode is changed.

  Then, the first light component is a plurality of longitudinal modes included in the separation wavelength width centered on the main longitudinal mode wavelength, and the second light component is the first light component from all the longitudinal modes included in the optical pulse train. The vertical mode is excluded.

  Since the mode-locked semiconductor laser device of the present invention is configured as described above, the configuration is simpler than the technique of Non-Patent Document 3 while maintaining the capability of maintaining the light injection lock in the same manner as the technique of Non-Patent Document 3. Can be.

(A) is a schematic diagram for explaining light injection locking, and (B) is an oscillation spectrum of laser oscillation light propagating inside in an MLLD operating in mode locking without continuous wave light injection. , (C) is an oscillation spectrum of an optical pulse train output from an MLLD that is mode-locked in a state in which continuous wave light _CW of wavelength λ _CW is injected and light injection locking occurs, and (D) is It is the schematic diagram which expanded the principal part of (B). (A) to (F) are characteristic diagrams respectively showing an oscillation spectrum of an output optical pulse train together with an envelope obtained by injecting continuous wave light CW with a changed wavelength difference δλ into MLLD of Non-Patent Document 3. is there. It is a characteristic view which shows the master-slave ratio calculated | required from FIG. 2 (A)-(F). (A) And (B) is a schematic diagram which shows the envelope of an optical pulse train, respectively. It is a characteristic view which shows the control parameter | index calculated | required from Fig.2 (A)-(F). It is a characteristic view which shows the transmission characteristic of the light in a isolation | separation means. It is a schematic diagram which shows typically the structure of the MLLD apparatus of 1st Example. It is a schematic diagram which shows typically the structure of the MLLD apparatus of 2nd Example. It is a schematic diagram which shows typically the structure of the MLLD apparatus of 3rd Example. It is a schematic diagram which shows typically the structure of the MLLD apparatus of 4th Example. It is a characteristic view for demonstrating the effect of the MLLD apparatus of 4th Example.

  Embodiments of the present invention will be described below with reference to the drawings. In each figure, the shape, size, and arrangement relationship of each component are schematically shown to the extent that the present invention can be understood. Moreover, although the preferable structural example of this invention is demonstrated hereafter, the material of each component, a numerical condition, etc. are only a suitable example. Therefore, the present invention is not limited to the following embodiments. Moreover, in each figure, the same code | symbol is attached | subjected to a common component and the description may be abbreviate | omitted.

<Principle>
First, the principle for maintaining the light injection lock using the MLLD of the present invention will be described with reference to FIGS. FIGS. 4A and 4B are schematic diagrams showing the envelope of the optical pulse train, respectively.

The envelope en _{CW in} FIG. 4A corresponds to the oscillation spectrum of the optical pulse train L _CW output from the MLLD that performs the mode locking operation in a state where the light injection locking is maintained. The envelope en _{NCW in} FIG. 4B corresponds to the oscillation spectrum of the optical pulse train L _NCW output from the MLLD that performs the mode-locking operation in a state where the optical injection locking does not occur. Later, in order to distinguish the optical pulse train _{L NCW} and _{L CW,} an optical pulse train _{L NCW} referred to as asynchronous pulse train _{L NCW,} also referred to as the optical pulse train _{L CW} synchronization light pulse train _{L CW.} Any optical pulse train L _B outputted from MLLD, depending on the synchronization strength, take the oscillation spectrum between asynchronous pulse train L _NCW and synchronization pulse train L _CW.

In order to prevent complication of the drawing, drawing of individual longitudinal modes included in the synchronous optical pulse train L _CW and the asynchronous optical pulse train L _NCW is omitted. Further, it is assumed that both the synchronous optical pulse train L _CW and the asynchronous optical pulse train L _NCW are in a mode-locked state with the main longitudinal mode wavelength λ _CW as the center wavelength. 4A and 4B both indicate the wavelength (arbitrary unit), and the vertical axis indicates the intensity (arbitrary unit).

Inventor, instead of the master-slave ratio described above, for the optical injection maintaining synchronization, the optical pulse train L strength _B in the second wavelength range Rf, i.e. conceive the availability of the strength of the area indicated by reference numeral L2 did. That is, in the state where the light injection locking is maintained, the envelope is narrowed and the intensity is concentrated in the first wavelength range Tr in the vicinity of the main longitudinal mode wavelength λ _CW, so that the portion of the hem of the envelope away from this wavelength λ _CW On the contrary, the strength at is lowered. Therefore, the index using the intensity of the optical pulse train L _B in the second wavelength range Rf corresponding to a portion of the skirt (hereinafter, also referred to as control index.) It is likely to reflect a synchronous intensity occurring in MLLD There is.

  Hereinafter, this point will be described with reference to FIG. FIG. 5 is a characteristic diagram showing the control index obtained from FIGS. The vertical axis in FIG. 5 is the control index (dBm), and the horizontal axis is the wavelength difference δλ (nm).

Prior to the description of the control index used in FIG. 5, first, an amount serving as a basis for deriving the control index will be described. First, as shown in FIG. 4, an optical pulse train L _B, the first light component L1 of the first wavelength range Tr, is separated into a second light component L2 of the second wavelength range Rf. Here, the first wavelength range Tr is set to a wavelength within a separation wavelength width Wλ centered on the main longitudinal mode wavelength λ _CW . Further, the second wavelength range Rf is a wavelength range in which all the longitudinal modes of the optical pulse train L _B is extended, the wavelength range excluding the first wavelength range Tr described above.

First light component L1 is among all longitudinal modes included in the optical pulse train L _B, consisting of longitudinal modes having a wavelength within a first wavelength range Tr. The integrated value of the intensity in the first wavelength range Tr is defined as the first light component intensity IL1. Further, the second light component L2 from the total longitudinal mode included in the optical pulse train L _B, longitudinal mode excluding the first light component L1, that is, a longitudinal mode included in the wavelength range Rf. The integrated value of the intensity in the second wavelength range Rf is set as the second light component intensity IL2. Further, the ratio IL2 / IL1 of the second light component intensity IL2 to the first light component intensity IL1 is defined as an inter-component intensity ratio R.

  At this time, the intensity ratio R between components can be used as a control index. However, the first light component intensity IL1 can be regarded as constant with a practically sufficient accuracy regardless of the presence or absence of light injection synchronization. In this regard, Non-Patent Document 3 discloses that the rate of change of the first light component intensity IL1 remains at about 0.6 dB before and after the light injection locking is lost by increasing the continuous wave light CW to about 6 dBm. It is shown. Therefore, in this case, the second light component intensity IL2 can be used as the control index instead of the inter-component intensity ratio R. By using the second light component intensity IL2 as a control index, the measurement of the first light component intensity L1 is not necessary, and the configuration of the separating means described later can be further simplified.

In FIG. 5 and the following description, the second light component intensity IL2 when the separation wavelength width Wλ is set to 2 nm and 5 _nm is used as the control indices R _{2 nm} and R _{5 nm} . This separation wavelength width Wλ (2 nm and 5 nm) is about twice and five times the full width at half maximum (about 1 nm) of the envelope in the state of maintaining the light injection locking obtained from FIGS. 2 (C) and (D). Each corresponds to a wavelength range. Incidentally, in obtaining the FIG. 5, an optical pulse train L _B it has been assumed to be the strength without loss divided into first and second light components L1 and L2. That is, it is assumed that the total intensity of the optical pulse train matches the sum of the first and second optical component intensities IL1 and IL2.

Comparing FIG. 5 and FIG. 3, it can be seen that the master-slave ratio (I ₊₅ / I ₀ ) and the control indices R _{2 nm} and R _{5 nm} are similar in behavior. In other words, the control indexes R _{2 nm} and R _{5 nm} take the concave portion range A1, the minimum range A2, and the constant value range A3, similarly to the master-slave ratio. As described above, the control indices R _{2 nm} and R _{5 nm} reflect the synchronization intensity very well, similar to the master-slave ratio. Therefore, if the main longitudinal mode wavelength λ _CW is controlled so as to reduce the difference between the control indices R _{2 nm} and R _{5 nm} and a predetermined reference value, that is, the control that keeps the wavelength difference δλ within the WLR described above, the MLLD light Injection locking can be maintained. For the reasons described above, even if the intensity ratio R between components is used as a control index, the behavior is equal to the control indices R _{2 nm} and R _{5 nm} .

Next, the size of the separation wavelength width Wλ will be described. Schematically, separated wavelength width Wλ is preferably set envelope en optical pulse train L _B kurtosis degree, that is, a wavelength width that reflect the degree of concentration of the intensity of the Shutate mode wavelength lambda _CW vicinity . More specifically, the separation wavelength width Wλ, the components between the intensity ratio R (= IL2 / IL1) needs to be set to a wavelength width to reflect the intensity ratio of Shutate mode to the total intensity of the optical pulse train L _B.

More quantitatively, the separation wavelength width Wλ is preferably set to a wavelength range that is less than the asynchronous optical pulse train full width W _NCW in FIG. 4B and exceeds the synchronous optical pulse train full width W _CW in FIG. Here, the asynchronous pulse train full width _{W NCW,} if the peak intensity of the envelope _{en NCW} asynchronous pulse train _{L NCW} was _{P NCW,} this envelope _{en NCW,} strength _{xP NCW} (here, x is 0 <x <1) The wavelength range is as described above. Similarly, the synchronous optical pulse train full width _{W CW,} if the peak intensity of the envelope _{en CW} synchronous optical pulse train _{L CW} was _{P CW,} in the envelope _{en CW,} the wavelength range where the intensity is equal to or greater than _{xP CW} . For example, when the intensity ratio x is 0.5, the asynchronous and synchronous optical pulse train full widths W _NCW and W _CW are equal to the full widths at half maximum of the envelopes en _NCW and en _CW , respectively. By setting the separation wavelength width Wλ under appropriate intensity ratio x, the inter-component intensity ratio R reflects the intensity ratio of Shutate mode to the total intensity of the optical pulse train L _B.

The wavelength width and the duration of individual optical pulses forming the output optical pulse train L _B from MLLD is known that the relation of Fourier conjugates of each other. Therefore, when the first optical component L _CW 1 of the synchronous optical pulse train L _CW is output to the outside as an output optical pulse train, the separation wavelength width Wλ exceeds the wavelength width at which an optical pulse with a sufficiently short time width can be obtained in practice. Preferably it is.

Further, when it excessive separation wavelength width Wramuda, results first light component L1 contains most longitudinal modes of the optical pulse train L _B, in synchronous and asynchronous optical pulse train L _CW and L _NCW components between the intensity ratio R, the light injection The difference due to the presence or absence of synchronization disappears. Therefore, it is preferable that the separation wavelength width Wλ be less than a width that causes a discernable difference in the intensity ratio R between components depending on the presence or absence of light injection synchronization.

Next, a reference value that serves as a reference in the control of the main longitudinal mode wavelength λ _CW using the control indices R _{2 nm} and R _{5 nm} will be described with reference to FIG. _{Taking the} control index R _{5 nm} as an example, generally, the reference value is within a range in which the upper limit is the value of the control index in the constant value range A3 and the lower limit is the value of the control index in the minimum range A2. It is preferable to set by.

More specifically, the asynchronous optical pulse train L _NCW in FIG. 4B is output from the MLLD with the wavelength difference δλ within the constant value range A3, and the wavelength difference δλ within the minimum range A2 is illustrated in FIG. The synchronized optical pulse train _LCW is output from the MLLD. Here, the value of the control index within the constant value range A3 obtained from the asynchronous optical pulse train L _NCW is defined as an asynchronous intensity ratio R _NCW (= IL _NCW 2 / IL _NCW 1). Similarly, the value of the control index within the minimum range A2 obtained from the synchronization optical pulse train L _CW is defined as a synchronization intensity ratio R _CW (= IL _CW 2 / IL _CW 1). In this case, it is preferable to select a suitable value according to the design as the reference value within the range of the synchronous strength ratio R _CW or more and less than the asynchronous strength ratio R _NCW . In particular, from the viewpoint of maintaining good light injection synchronization, the reference value is preferably set to a value in the vicinity of the synchronization intensity ratio _RCW . For example, with the control index R _{5 nm} in the example of FIG. 5, if the control is performed with the reference value set to −31 dBm which is equal to the minimum value, the optical pulse train L _CW in which the optical synchronization state is maintained can always be output from the MLLD.

  As described above, the intensity ratio R between components, which is a control index, has a resolution of, for example, about 2 to 5 nm, rather than a master-slave ratio that requires a high wavelength resolution of, for example, about 0.3 nm for the individual separation of the longitudinal mode. Can be lowered. As a result, the intensity ratio R between components can be obtained by a wavelength separation means having a simpler configuration. Therefore, by using this control index, the structure of the MLLD device, in particular, the wavelength separation means, is simplified compared to the technique of Non-Patent Document 3, while maintaining the capability of maintaining the light injection synchronization as in the technique of Non-Patent Document 3. can do.

In this section, the case where the intensity ratio R between components that is the intensity ratio of the second light component L2 to the first light component L1 is mainly used as the control index has been described. However, control indicator, if that amount to reflect the intensity ratio of Shutate mode to the total intensity of the light pulse train L _B, it is not limited to inter-component intensity ratio R. For example, it can be used as a control index such as the intensity ratio of the second optical component to the total intensity of the intensity itself or of the second light component L2, the optical pulse train of the second optical component L _B, as shown in FIG.

Next, referring to FIG. 6, an optical pulse train L _B, the first light component L1 of the first wavelength range Tr, specific separation means for separating the second light component L2 of the second wavelength range Rf A typical configuration will be described. The vertical axis in FIG. 6 is the transmittance of light transmitted through the separating means, and is a value in the range of 0 to 1. The horizontal axis is the wavelength in arbitrary units. A curve I in FIG. 6 shows the wavelength characteristic of the first light component L1 transmitted through the separating means, and a curve II shows the wavelength characteristic of the second light component L2 reflected by the separating means.

As a separation means for separating the optical pulse train L _B in the first and second light components L1 and L2, as shown in FIG. 6, it is preferable light transmissive reflection filter showing the transmission and reflection characteristics. That is, the light transmission / reflection filter transmits the first light component L1 as transmitted light and reflects the second light component L2 as reflected light using the first wavelength range Tr as a pass band. If such a light transmission / reflection filter is used as the separating means, the first and second light components L1 and L2 can be separated and output as transmitted light and reflected light, respectively. As such a light transmission / reflection filter, a dielectric multilayer film having a high degree of freedom in designing the separation wavelength width Wλ is suitable.

<Example>
Hereinafter, several embodiments of the MLLD apparatus will be illustrated with reference to FIGS. 7 to 10 are schematic views schematically showing the configuration of the MLLD apparatus of the first to fourth embodiments. FIG. 11 is a characteristic diagram for explaining the effect of the MLLD device of the fourth embodiment.

(First embodiment)
Referring to FIG. 7, the MLLD apparatus 102 according to the first embodiment includes a mode synchronization unit 70, a separation unit 60, and a control unit 50. Furthermore, the MLLD device 102 includes an optical coupler 21 as an optional element.

Separating means 60, is transmitted through the output optical pulse train L first light component L1 _B contains from mode synchronization means 70 for mode-locking operation by the optical injection locking state, constituting a second light component L2 as a transmissive reflection filter for reflecting Has been. The configuration of the transmission / reflection filter is as described above.

The optical coupler 21 is an optical branching element interposed in the propagation path of the first light component L1 between the separating means 60 and the intensity detector 51. The optical coupler 21 separates the first light component L1 at a predetermined intensity ratio, outputs one as an optical pulse train L _B ′ outside the system, and outputs the other toward the intensity detector 51.

  The control unit 50 includes an intensity detection unit 51 and a control unit 52. The intensity detector 51 detects the intensity of the first and second light components L1 and L2, respectively. The controller 52 first calculates the control index R from the intensities of the first and second light components L1 and L2. Further, the control unit 52 compares the control index R with a reference value determined in advance by another means, and feedback-controls the adjusting unit 23 described later of the mode synchronization means 70 so as to reduce the difference between the two. To do.

Mode-locking means 70, under the control of the control unit 50 is configured to wavelength of the output optical pulse train L _B so as to keep a range of optical injection locking is maintained. More specifically, the mode synchronization unit 70 includes the MLLD element 1 and the adjustment unit 23. Further, the mode synchronization means 70 includes, as optional elements, a CW light source 19, an input optical system 110, an output optical system 112, a DC power supply 11, a modulation power supply 13, a voltage source 12, and a temperature control unit 32. And.

The MLLD element 1 is generally a semiconductor laser diode that realizes laser oscillation by forming an inversion distribution by injecting a current into an optical gain region 3 including a pn junction. The MLLD element 1 operates in a mode-locked state by applying an appropriate voltage and current. To MLLD device 1 in this state, when injecting the continuous wave light CW of wavelength lambda _CW, optical injection locking occurs, propagates light L _A and interaction with the continuous wave light CW propagating inside the light wavelength is equal to lambda _CW pulse train _{L B} is outputted. The MLLD element 1 is controlled by the control means 50 via the adjustment unit 23. With this control, the main longitudinal mode wavelength λ _LM of the propagation light L _A , that is, the output light pulse train L _B is made to coincide with λ _CW so that the light injection locking is maintained.

  Structurally, the MLLD element 1 includes a core 30 through which propagating light propagates, and an optical waveguide 40 including a p-type cladding 5 and an n-type cladding 6 that sandwich the core 30 from above and below in the thickness direction.

  The core 30 includes an optical modulation region 2, an optical gain region 3, and a passive waveguide region 4, which are arranged in series in the order of description. It should be noted that the division of these three regions is functional until they are tired, and there is no structural difference between the three regions. That is, these three regions integrally form the core 30.

  The optical gain region 3 is a region having an optical amplification function for laser oscillation. More specifically, a constant current is injected into the optical gain region 3 from the DC power supply 11 through the p-side electrode 9 and the n-side common electrode 7 to form an inversion distribution necessary for laser oscillation.

Optical modulation region 2, in order to perform mode-locking operation, a region for modulating light intensity of the propagating light L _A, the transmittance is modulated externally. More specifically, a reverse bias voltage is applied from the voltage source 12 to the light modulation region 2 via the p-side electrode 8 and the n-side common electrode 7. Further, a modulation voltage having a frequency that is a natural number multiple of the resonator circulation frequency of the MLLD element 1 is applied from the modulation power source 13 via the electrodes 7 and 8. The transmittance of the light modulation region 2 is modulated by these applied voltages. The optical modulation region 2 corresponds to an optical modulator such as a saturable absorption band in a passive mode-locked laser or an electroabsorption optical modulator in an active mode-locked laser.

Passive waveguide region 4 is composed of a material transparent to the wavelength of the oscillation light L _A, it receives the adjustment of the refractive index by adjusting unit 23, the wavelength of the oscillation light L _A, or main output optical pulse train L _B This is a region in which the longitudinal mode wavelength λ _LM is changed. More specifically, a pin junction is formed in the optical waveguide 40 extending in a region corresponding to the passive waveguide region 4. In this pin junction, the passive waveguide region 4 is an intrinsic semiconductor layer (i layer), and the p-type and n-type claddings 5 and 6 extending to the region corresponding to the region 4 are a p-layer and an n-layer, respectively. . A control current in accordance with the control of the control means 50 is injected from the adjustment unit 23 into the pin junction via the p-side electrode 10 and the n-side common electrode 7. Thereby, the refractive index of the region 4 is changed by utilizing the plasma effect generated in the passive waveguide region 4. This change in refractive index, the passive waveguide to change the optical path length of the region 4, the core 30 the wavelength of the oscillation light L _A for propagating, that is, the output optical pulse train main longitudinal mode wavelength lambda _LM wavelength tuning range within WLR of L _B Make adjustments so that More specifically, the wavelength difference δλ the continuous wave light CW is adjusted to be kept in optical injection locking a wavelength tuning range WLR sustainable output optical pulse train L _B (see FIG. 1 (C)).

Here, the "adjusting the wavelength of the oscillation light L _A and the output optical pulse train L _B", the main longitudinal mode wavelength λ not only _LM, of the total longitudinal mode included in the oscillation light L _A and the output optical pulse train L _B This means that the wavelengths are simultaneously shifted by the same wavelength width.

Returning to the description of the configuration of the mode synchronization means 70 again. Adjustment unit 23, as described above, under the control of the control unit 50, by adjusting the refractive index of the passive waveguide region 4, by changing the wavelength of the entire oscillation light L _A, the output optical pulse train L _B The main longitudinal mode wavelength λ _LM is kept within the wavelength range in which the light injection locking is maintained. In this embodiment, the adjusting unit 23 is configured as a control current injection device that injects a control current according to the control of the control means 50 into the pin junction of the optical waveguide 40 corresponding to the passive waveguide region 4.

The CW light source 19 is provided outside the MLLD element 1 and is a light source that outputs output light having a single wavelength λ _CW . This output light is converted into continuous wave light CW by the input optical system 110 and then injected into the core 30. Injecting continuous wave light CW into the core 30 indicates that continuous wave light CW is input to the core 30.

  The input optical system 110 is generally an optical system for converting the output light from the CW light source 19 into continuous wave light CW and inputting the continuous wave light CW into the core 30. The input optical system 110 includes a polarization plane adjusting element 20, an optical circulator 18, and a coupling lens 17.

Polarization plane adjustment element 20 is an element for converting the output light from the CW light source 19, the same polarization direction propagating light L _A of the internal MLLD element 1 to the continuous wave light CW. More specifically, the polarization plane adjustment element 20 is configured by, for example, a half-wave plate. In other words, continuous from rotating crystal axis of 1/2-wavelength plate (phase advancing axis or slow axis), which rotates the polarization plane of the output light from the CW light source 19, the polarization direction coincides with the propagation light L _A Wave light CW is obtained.

Optical circulator 18, a continuous wave light CW inputted to the core 30, an element which optical nonreciprocal separate the optical pulse train L _B outputted from the core 30. In other words, the optical circulator 18 includes three input / output ports 18 ₁ , 18 ₂ and 18 ₃ , and the continuous wave light CW input to the input / output port 18 ₁ is output from the input / output port 18 ₂ , It is injected into the core 30 from the input / output end face Q via the coupling lens 17. Further, the optical pulse train L _B outputted from the output end face Q of the core 30 is input to the input-output port 18 _2, is output from the output port 18 ₃ toward the separating means 60.

Coupling lens 17 constituting the input optical system 110, and the coupling efficiency to the core 30 of the continuous wave light CW, and a coupling efficiency to the output optical system 112 of the optical pulse train L _B outputted from the core 30, to enhance each It is an element.

Output optics 112 is schematically an optical system for outputting an output optical pulse train L _B from MLLD element 1 to the outside is interposed between the MLLD element 1 and the separation means 60. The output optical system 112 includes a coupling lens 17 and an optical circulator 18. Here, the output optical system 112 shares the coupling lens 17 and the optical circulator 18 with the input optical system 110.

Temperature control unit 32, to keep the temperature of the MLLD device 1 constant, stabilizing the longitudinal mode wavelength of the output light pulse L _B from MLLD element 1. The temperature control unit 32 includes a heat generation / heat absorption element 14, a temperature monitor 15, and a temperature controller 16. More specifically, the temperature monitor 15 observes the temperature of the MLLD element 1. The temperature controller 16 controls the heat-generating / heat-absorbing element 14 based on the difference between the observed temperature of the temperature monitor 15 and a preset temperature. The heat-generating / heat-absorbing element 14 is composed of a Peltier element, for example, and receives heat or generates heat so as to keep the temperature of the MLLD element 1 constant under the control of the temperature controller 16.

  Next, the control method of the MLLD device will be described together with the operation with reference to FIGS. 1, 4 and 7 as appropriate.

(First step)
In the first step, the injection of the continuous wave light CW, from MLLD device 1 for optical injection locking mode while it is maintained synchronous operation, the optical pulse train L _B is outputted (refer to FIG. 1 (A)).

More specifically, the MLLD element 1 is mode-locked by applying an appropriate voltage and current to the core 30 using the DC power supply 11, the voltage source 12, and the modulation power supply 13. Results of mode-locking operation, the wavelength lambda _p longitudinal mode LM _p centered wavelength, propagating light L _A having an oscillation spectrum as shown in FIG. 1 (B), propagated the core 30.

By injecting continuous wave light CW satisfying a predetermined condition into the MLLD element 1 in this state from the CW light source 19 via the input optical system 110, light injection synchronization appears in the MLLD element 1. Here, the predetermined condition (1) that the wavelength lambda _CW continuous wave light CW is included in the wavelength range in which all longitudinal modes to distribution of the oscillation light L _A, and (2) the polarization direction of the continuous wave light CW oscillation light L Match _A.

From MLLD element 1 in the optical injection locking state, the main longitudinal mode wavelength equal to the wavelength lambda _CW continuous wave light CW, the optical pulse train L _B having an oscillation spectrum as shown in FIG. 1 (C) is output. Optical pulse train L _B via the output optical system 112, is input to the light transmission reflecting filter is a separating means 60.

(Second process)
In the second step, Shutate mode the first and second light components L1 and L2 contained in the optical pulse train L _B, the intensity ratio of the first and second light components L1 and L2 to the total light intensity of the optical pulse train L _B The wavelength separation is performed so as to reflect the intensity ratio.

More specifically, the separation means 60, an optical pulse train _{L B,} the first light component L1 of the first wavelength range Tr, wavelength separated into a second light component L2 of the second wavelength range Rf (FIG 4 (A ) And (B)). More specifically, the separating means 60 transmits the first light component L1 and reflects the second light component L2. As described above, the separation wavelength width Wλ separation means 60, reflects the ratio of Shutate mode intensity ratio of the first and second light components L1 and L2 (IL2 / IL1) is to the total light intensity of the optical pulse train L _B It is decided to do.

(Third process)
In a third step, the intensity ratio of the first and second light components L1 and L2 (IL2 / IL1) or by controlling the wavelength of the oscillation light _{L A} using the light intensity IL2 of the second light component L2, a wavelength difference δλ so it remains in the optical injection locking a wavelength tuning range WLR sustainable, changing the wavelength lambda _LM main longitudinal mode of the output optical pulse train L _B.

  More specifically, the first and second light components L1 and L2 are input to the intensity detector 51 of the control means 50, and the respective intensities are detected. The control unit 52 of the control unit 50 uses the detected intensities of the first and second light components L1 and L2 to calculate a control index R that is an intensity ratio of the second light component L2 to the first light component L1. . As described above, this control index R reflects the synchronization strength of the MLLD element 1.

  Further, the control unit 52 compares the control index R with a predetermined reference value, and feedback-controls the adjustment unit 23 of the mode synchronization means 70 so as to reduce the difference between the two. That is, the control unit 52 controls the adjustment unit 23 so that the light injection synchronization is maintained in the MLLD element 1.

More specifically, the control unit 52 controls the adjusting unit 23 changes the main longitudinal mode wavelength lambda _LM of the propagation light L _A. That is, the longitudinal mode wavelength lambda _LM change of the propagation light L _A as a result of control of the control unit 52 changes the synchronous strength, thereby further longitudinal mode wavelength lambda _LM of change in the optical pulse train L _B is caused. The longitudinal mode wavelength lambda _LM of change in the optical pulse train L _B is observed as a change in the control index R, and returned again to the control unit 52. In this way, the control unit 52 can confirm the suitability of its own control result as the rate of change of the difference between the control index R and the reference value, and performs appropriate control on the adjustment unit 23 based on this control result.

The adjustment unit 23 of the mode synchronization unit 70 that is controlled by the control unit 50 injects a control current corresponding to the control into the passive waveguide region 4. Thereby, a plasma effect is generated in the region 4 and the refractive index of the passive waveguide region 4 is changed. This change in refractive index, the passive waveguide to change the optical path length of the region 4, the core 30 the wavelength of the oscillation light L _A for propagating, that is, the output optical pulse train main longitudinal mode wavelength lambda _LM wavelength tuning range within WLR of L _B Make adjustments so that More specifically, the wavelength difference δλ the continuous wave light CW is adjusted to be kept in optical injection locking a wavelength tuning range WLR sustainable output optical pulse train L _B (see FIG. 1 (C)).

  Thus, according to the control method of the MLLD device of this embodiment, the light injection synchronization can be maintained in the same manner as the technique of Non-Patent Document 3 by a simpler method than the technique of Non-Patent Document 3.

In this embodiment, the case where the coupling lens 17 is used has been described. However, if input and output efficiency of the continuous wave light CW and the optical pulse train L _B is sufficiently large, it may be omitted coupling lens 17.

In this embodiment, the case where the core 30 of the MLLD element 1 includes the optical gain region 3, the optical modulation region 2, and the passive waveguide region 4 has been described. However, the MLLD element 1 does not have to include these three regions, and any MLLD element that satisfies the following three conditions can be used. The conditions are as follows: (1) excitation of the MLLD element 1 by current injection to cause laser oscillation; and (2) optical modulation at a frequency that is a natural number multiple of the round frequency of the resonator of the MLLD element 1. it is realized, and (3) the wavelength of the oscillation light L _a is variable, a.

(Second embodiment)
With reference to FIG. 8, the MLLD apparatus 103 of 2nd Example is demonstrated. The MLLD device 103 of this embodiment is configured in the same manner as the MLLD device 102 of the first embodiment except for the adjustment unit 24 of the mode synchronization means 72. Therefore, this difference will be mainly described.

  The adjustment unit 24 of the MLLD device 103 is a control voltage application device that applies a control reverse bias voltage, which is a reverse bias voltage according to the control of the control unit 50, to the pin junction of the optical waveguide 40 corresponding to the passive waveguide region 4. It is configured. The adjustment unit 24 changes the refractive index of the passive waveguide region 4 using the Pockels effect generated in the passive waveguide region 4 by applying a control reverse bias voltage.

Unlike the MLLD device 102 of the first embodiment, the MLLD device 103 of this embodiment does not inject current into the passive waveguide region 4 when controlling the refractive index. As a result, free carrier absorption does not occur in the passive waveguide region 4 in the MLLD device 103. Therefore, MLLD device 103 of this embodiment can output an output optical pulse train _{L B} of higher strength than MLLD device 102.

  The MLLD device 103 can be modified in the same manner as the MLLD device 102.

(Third embodiment)
With reference to FIG. 9, the MLLD apparatus 104 of 3rd Example is demonstrated. The MLLD device 104 of this embodiment is configured in the same manner as the MLLD device 102 of the first embodiment except for the adjusting unit 27 of the mode synchronization means 74. Therefore, this difference will be mainly described.

  The adjusting unit 27 of the MLLD device 104 is configured as a temperature adjusting device that adjusts the temperature of the passive waveguide region 4 under the control of the control means 50. The adjustment unit 27 adjusts the temperature using Joule heat generated by energization of the resistance film 26 and changes the refractive index of the passive waveguide region 4. More specifically, the adjustment unit 27 includes an insulating layer 25 and a resistance film 26. The insulating layer 25 is provided on the upper surface of the p-type cladding 5 corresponding to the passive waveguide region 4 and ensures electrical insulation between the resistance film 26 and the MLLD element 1. The resistance film 26 is an electrical resistance film such as a platinum thin film provided immediately above the insulating layer 25. In the resistance film 26, Joule heat having a heat amount corresponding to the current value to be energized is generated, and the refractive index of the passive waveguide region 4 changes according to the heat amount.

As in the MLLD device 103 of the second embodiment, the MLLD device 104 of this embodiment does not inject current into the passive waveguide region 4 when controlling the refractive index. Therefore, MLLD device 104, like the MLLD device 103, to prevent the free carrier absorption in the passive waveguide region 4, can output an output optical pulse train _{L B} of higher strength than MLLD device 102.

  Furthermore, the MLLD device 104 that uses temperature changes can provide a wider refractive index change width than the MLLD device 103. Therefore, the MLLD device 104 is particularly suitable when adjustment of the longitudinal mode wavelength in a wider wavelength range is required. For example, it is suitable when the mode-locking frequency is high and the adjustment width of the longitudinal mode wavelength required for stable operation of the MLLD element 1 is on the order of several nm.

(Fourth embodiment)
With reference to FIGS. 10 and 11, the MLLD device 105 of the fourth embodiment will be described. MLLD apparatus of this embodiment 105, the continuous wave light CW and the optical pulse train L _B is, MLLD device of the first embodiment except that it is output from a different end face P and Q of the core 30 of each mode synchronization means 76 The configuration is the same as 102. Therefore, this difference will be mainly described.

That is, in the MLLD device 105, the one end face of the core 30 and the input terminal P of the continuous wave light CW, the other end face of the core 30 and the output terminal Q of the optical pulse train L _B.

Thus, MLLD device 105 may be compared to MLLD devices 102 to 104 of the first to third embodiments, to improve the extinction ratio of the output optical pulse train _{L B.}

Hereinafter, this point will be described with reference to FIG. FIG. 11 is a characteristic diagram for explaining the effect of the MLLD device 105. In FIG. 11, the horizontal axis represents delay time (ps), and the vertical axis represents SHG (Second Harmonic Generation) intensity (dB). Curve I in FIG. 11 is a SHG autocorrelation waveform of the optical pulse train _{L B} outputted from MLLD device 102 of the first embodiment, the curve II is SHG autocorrelation of the optical pulse train _{L B} outputted from MLLD device 105 It is a waveform.

  In the curves I and II, if the ratio between the maximum value and the minimum value is defined as the extinction ratio, the extinction ratio of the curve I corresponding to the first example is only about 18 dB due to the background component. On the other hand, the extinction ratio of the curve II corresponding to this embodiment is increased to about 28 dB, and the extinction ratio is improved as compared with the first embodiment.

This is because in MLLD devices 102 to 104 of the first to third embodiments, the input and output of the continuous wave light CW output optical pulse train _{L B,} derived from the fact that using an optical circulator 18. That is, the return light component of the continuous wave light CW input to the input-output port 18 ₂ of the optical circulator 18, thereby being output from the output port 18 ₃ together with the output optical pulse train L _B. As a result, the return light component of the continuous wave light CW is detected as the background component of the curve I, and the extinction ratio is deteriorated. On the other hand, since the MLLD device 105 of this embodiment does not use an optical circulator, the extinction ratio can be kept good. However, in this embodiment, in order to not using an optical circulator, in order to prevent the reflected return light, the input end P of the continuous wave light CW, optical pulse train L first optical isolator 121 to the output terminal Q of _B, The second optical isolator 122 is preferably connected and used.

  This embodiment can be used in combination with the MLLD devices 102 to 104 of the first to third embodiments.

(Modification)
(1) The first to fourth embodiments can be applied not only to the above-described active mode-locked semiconductor laser, but also to a passive mode-locked semiconductor laser or a hybrid mode-locked semiconductor laser using both.

  (2) Since the first to fourth embodiments do not have the modulation power source 13, they can be applied to passive mode semiconductor lasers that operate at a frequency higher than the repetition frequency of the modulation power source 13.

  (3) In adjusting the refractive index of the passive waveguide region 4, it is possible to use a band filling effect, a Franzkeldish effect which is a quantum size effect, or the like instead of the plasma effect and the Pockels effect described in the embodiments. it can.

1: Mode-locked semiconductor laser element (MLLD element)
2: light modulation region 3: light gain region 4: passive waveguide region 5: p-type cladding layer 6: n-type cladding layer 7: n-side common electrode 8: p-side electrode 9 in the light modulation region 9: p in the light gain region Side electrode 10: p-side electrode in passive waveguide region 11: DC power supply 12: voltage source 13: modulation power supply 14: heat generation / heat absorption element 15: temperature monitor 16: temperature controllers 17, 17-1, 17-2: coupling lens 18: optical circulators 18 ₁ , 18 ₂ , 18 ₃ input / output ports 19, 119: CW light source 20, 120: polarization plane adjusting element 21: optical couplers 23, 24, 27: adjusting unit 25: insulating layer 26: resistive film 30 : Core 32: Temperature control unit 40: Optical waveguide 50: Control unit 51: Intensity detection unit 52: Control unit 60: Separation unit 70, 72, 74, 76: Mode synchronization units 102, 103, 104, 105: MLLD device 110 114: input optics 112, 116: output optics 121: first optical isolator 122: second optical isolator

Claims (11)

An optical gain region that forms an inversion distribution, a light modulation region that modulates the light intensity, and a passive waveguide region in which the wavelength of the oscillation light is changed by the adjustment unit, and includes a core through which the oscillation light propagates, and a cladding Including an optical waveguide
A continuous wave having a wavelength difference capable of realizing light injection locking with a target longitudinal mode which is one of a plurality of longitudinal modes included in the oscillation light and having the polarization direction coincident with the oscillation light A mode-locked semiconductor laser element that receives a light wave and outputs an optical pulse train including a main longitudinal mode that is a longitudinal mode having a wavelength equal to that of the continuous wave light, and a plurality of subordinate longitudinal modes distributed around the main longitudinal mode wavelength;
Separating the first and second light components included in the optical pulse train so that the intensity ratio of the first and second light components reflects the ratio of the main longitudinal mode to the total light intensity of the optical pulse train Means,
A control index is obtained using the light intensity of the second light component, and the adjustment unit is controlled using the control index, so that the wavelength difference is kept within a wavelength synchronization range in which the light injection synchronization can be maintained. Control means for changing the wavelength of the main longitudinal mode,
The first optical component is the longitudinal mode included in a separation wavelength width centered on the main longitudinal mode wavelength, and the second optical component is the first longitudinal component from all longitudinal modes included in the optical pulse train. A mode-locked semiconductor laser device characterized by a longitudinal mode excluding light components. As the control index, an inter-component intensity ratio that is an intensity ratio of the second light component to the first light component is used.
The mode-locked semiconductor laser device according to claim 1, wherein the control unit controls the adjustment unit so as to reduce a difference between the intensity ratio between components and a predetermined reference value. In the envelope of all longitudinal modes included in the asynchronous optical pulse train output from the mode-locked semiconductor laser element in a state where the light injection locking is not manifested, the intensity ratio to the peak of the envelope is x (where x Is a wavelength range in which 0 <x <1) or more is defined as the full width of the asynchronous optical pulse train,
In the envelope of all longitudinal modes included in the optical pulse train output from the mode-locked semiconductor laser element in a state where the light injection locking is maintained, the wavelength at which the intensity ratio with respect to the peak of the envelope is not less than x When the range is the full width of the synchronous optical pulse train,
3. The mode-locked semiconductor laser device according to claim 1, wherein the separation wavelength width of the separation unit is set to a wavelength range that is less than the full width of the asynchronous optical pulse train and exceeds the full width of the synchronous light pulse train. In the first and second light components when the separation means separates the asynchronous light pulse train output from the mode-locked semiconductor laser element in a state where the light injection locking is not manifested, the first light component is compared with the first light component. The intensity ratio of the two light components is the asynchronous intensity ratio,
When the intensity ratio between the components in the state where the light injection synchronization is maintained as a synchronous intensity ratio,
3. The mode-locked semiconductor laser device according to claim 2, wherein the reference value is set to a value between the asynchronous intensity ratio and the synchronous intensity ratio.   5. The light separating / reflecting filter according to claim 1, wherein the separation unit includes a light transmission reflection filter that transmits the first light component and reflects the second light component with the separation wavelength width as a pass band. The mode-locked semiconductor laser device according to one item. The optical waveguide is provided with a pin junction formed by the passive waveguide region as an intrinsic semiconductor layer and the p-type cladding and the n-type cladding as the cladding sandwiching the passive waveguide region from both sides,
The adjusting unit is configured as a control current injection device that changes a refractive index of the passive waveguide region by injecting a control current according to the control of the control unit into the passive waveguide region. A mode-locked semiconductor laser device according to any one of claims 1 to 5. The optical waveguide is provided with a pin junction formed by the passive waveguide region as an intrinsic semiconductor layer and the p-type cladding and the n-type cladding as the cladding sandwiching the passive waveguide region from both sides,
The adjusting unit is configured as a control voltage applying device that changes a refractive index of the passive waveguide region by applying a control reverse bias voltage according to the control of the control unit to the passive waveguide region. The mode-locked semiconductor laser device according to claim 1, wherein   The adjustment unit is configured as a temperature adjustment device that changes a refractive index of the passive waveguide region by adjusting a temperature of the passive waveguide region according to control of the control unit. The mode-locked semiconductor laser device according to any one of 1 to 5.   9. The mode-locked semiconductor laser device according to claim 1, wherein one end face of the core is shared as an input / output end face of the optical pulse train and the continuous wave light.   The mode according to claim 1, wherein one end face of the core is an input end of the continuous wave light, and the other end face of the core is an output end of the optical pulse train. Synchronous semiconductor laser device. An optical waveguide having an optical gain region that forms an inversion distribution, an optical modulation region that modulates light intensity, and a passive waveguide region, and a core through which the oscillation light propagates and a cladding are provided, and the oscillation light includes Continuous wave light having a wavelength difference capable of realizing light injection locking with a target longitudinal mode, which is one of the longitudinal modes included, and having the polarization direction coincident with the oscillation light is input. A first step of outputting an optical pulse train including a main longitudinal mode which is a longitudinal mode having a wavelength equal to the continuous wave light and a plurality of subordinate longitudinal modes distributed around the main longitudinal mode wavelength from the mode-locked semiconductor laser device,
The first and second light components included in the optical pulse train are separated such that the intensity ratio of the first and second light components reflects the intensity ratio of the main longitudinal mode with respect to the total light intensity of the optical pulse train. Two processes,
The wavelength of the oscillation light is controlled using the light intensity of the second light component, and the main vertical length is maintained so that the magnitude of the wavelength difference is maintained within a wavelength synchronization range in which the light injection synchronization can be maintained. A third step of changing the wavelength of the mode,
The first light component is a plurality of the longitudinal modes included in a separation wavelength width centered on the main longitudinal mode wavelength, and the second light component is from all longitudinal modes included in the optical pulse train. A method for controlling a mode-locked semiconductor laser device, wherein a longitudinal mode excluding the first light component is selected.