JP4764964B2 - Method for forming minute periodic polarization reversal structure and minute periodic polarization reversal structure - Google Patents

Description

  The present invention relates to a method for forming a micro periodic polarization inversion structure of a ferroelectric nonlinear optical crystal substrate, a micro periodic polarization inversion structure formed by the method, and an optical waveguide device provided with the micro periodic polarization inversion structure.

Quasi-Phase-Matching (QPM) Second Harmonic Generation (SHG) devices using ferroelectric nonlinear optical crystal materials are most suitable for improving the efficiency of wavelength conversion devices. Since then, research has been conducted energetically at major internal and external research institutions.
In recent years, the application of QPM-SHG blue light source devices using LiNbO3 (Lithium Niobate: LN) crystals and LiTaO3 (Lithium Tantalate: LT) crystals, which are ferroelectric nonlinear optical crystals, has been spreading, and high-intensity three-primary lasers Since it is regarded as important as a scanning laser display using light generation, ultraviolet laser light processing technology, and multifunctional applied light measurement from mid-infrared to ultraviolet, high output of 10W class is desired.

  In order to realize such a QPM-SHG device, as shown in FIG. 11, a periodically poled structure 110 having a period of coherent length (lc) is formed in the nonlinear optical crystal 100 in the element, and an optical waveguide is formed. 120 was formed, incident light (pump light) 130 was converted, and outgoing light 114 as a blue light source was created. As a method for forming such a periodically poled structure, a so-called voltage application method is known. One of the most important processes in this electric field application method is the production of a “periodic electrode”. Conventionally, as the “periodic electrode”, a rectangular electrode-shaped lift-off method and a polarization inversion method using an FCE method have been used.

In the lift-off method, metal is deposited on the entire surface of the substrate including the periodic resist pattern, the organic solvent is infiltrated into the direst pattern, the resist is dissolved, the metal is removed together with the dissolved resist pattern, and the substrate is directly contacted. The process of leaving only the part that was made. However, it is difficult to manufacture a lift-off electrode that works satisfactorily as a polarization reversal electrode as the period decreases to about 2 μm or less.
In order to realize a QPM-SHG blue light source, the coherent length (lc) in the LN crystal or LT crystal is about 1 to 2 μm, so the period of the polarization inversion layer, that is, the line and space width (hereinafter abbreviated as L & S). ) Is about 2 to 4 μm. Considering the addition of DBR (distributed Bragg reflector) as a resonator, periodic polarization inversion with an L & S width of 0.3 to 0.6 μm is required. In particular, when a device as described above is manufactured using the lift-off method, it is difficult to manufacture a very fine periodic electrode.

As a method for overcoming these problems, there is a proposal of polarization inversion by applying an electric field by covering a resist pattern formed on a crystal substrate with a liquid or a metal electrode (see Non-Patent Documents 1 and 2). Hereinafter, this method is referred to as FCE (Full Cover Electrode Method) method.
The FCE method is an abbreviation for Full-Cover Electrode Method, which is a method in which a resist pattern formed on a crystal substrate is covered with a liquid or metal electrode, and polarization inversion is performed by applying an electric field. Since this FCE method has many parameters such as resist pattern shapes of various cross sections and their intervals, period, resist thickness, resist dielectric constant, etc., there is a degree of design freedom that does not exist in the lift-off method.
However, in the above reports, only one point of experimental data is shown, and it is unclear whether the best state is realized. It is also unclear to what extent a minute domain-inverted structure can be produced. As described above, the whole picture is completely unknown, and the technique of minute periodic polarization reversal indispensable for a QPM-SHG device having a high aspect ratio (crystal thickness d / polarization reversal layer width lc) has not yet been established. Currently.
LE Myers, RC Eckardt, MM Fejer, RL Byer, WR Bosenberg, and JW Pierce, "Quasi-phase-matched optical parametric oscillators in bulk periodically poled LiNbO3", J. Opt. Soc. Am. B, 12 (1995), 2102. Kenji Kinatak, Masatoshi Fujimura, Toshiaki Suhara, and Hiroshi Nishihara, "High-efficiency LiNbO3 waveguide second-harmonic generation devices with ferroelectric-domain-inverted gratings fabricated by applyin voltage", IEEE J. Lightwave Technol., 14 (1996), 462 .

Whether performing polarization inversion using any of the methods of the lift-off method or FCE method, as long as the rectangular electrode shape of the electrode is used as a "periodic electrodes", the rectangular electrode, due to the presence of the strongest electric field across the electrodes Inverted nuclei are generated in the non-electrode portion outside the electrode end . As illustrated in FIG. 4 as an experimental example performed by the inventor, the inversion nucleus grows downward (−z direction) with a minute size, and then grows and spreads in the width direction. When the integral value of the polarization reversal current increases and the polarization reversal is completed, the polarization reversal width becomes considerably wider than the electrode width (see FIG. 4 (f)). For this reason, there is a problem that it is difficult to produce minute periodic polarization reversal.

Therefore, the present invention proposes a new method for forming a micro periodic polarization reversal in a ferroelectric nonlinear optical crystal, which overcomes the disadvantages of the conventional periodic polarization reversal method using a rectangular electrode and is easy and versatile. It is intended to provide a polarization reversal technique having In order to solve this problem, it is necessary to produce an extremely minute periodic domain-inverted structure deeply and uniformly in a larger crystal (to realize a high aspect ratio).
Another object of the present invention is to provide a method for forming a micro periodic polarization inversion for realizing a high output device having a high aspect ratio (d / lc).

According to the method for forming a minute periodic polarization reversal structure of the present invention, a polarization reversal structure having a very minute period can be formed in a ferroelectric nonlinear optical crystal substrate. For example, in an LN crystal having a thickness of 500 μm, a domain-inverted structure with a period of 2 μm or less can be formed.
Furthermore, it is possible to form an optical waveguide element having a minute periodic polarization inversion structure having a high aspect ratio (d / lc).

The present invention has the following features.
(1) The method of forming a micro periodic polarization inversion structure in the ferroelectric nonlinear optical crystal of the present invention is as follows:
A first electrode provided on the + z plane which is one surface of the z-cut substrate of the ferroelectric nonlinear optical crystal substrate;
By applying a voltage between the second electrode provided on the -z plane which is the other surface of the z-cut substrate so as to face the first electrode,
A method of forming a micro periodic polarization reversal structure in the ferroelectric nonlinear optical crystal substrate,
The first electrode is brought into point contact with the + z plane of the z-cut substrate in a yz cross-sectional view perpendicular to the polarization reversal period direction in the substrate ,
So as to continuously contact linearly in the x-axis direction parallel to the polarization inversion period direction in the substrate,
The second electrode covers the electrode area of the first electrode in contact with the + z plane,
Contacting the -z surface of the z-cut substrate;
Applying a pulse voltage between the first electrode and the second electrode;
A domain-inverted nucleus is generated immediately below the first electrode and the domain-inverted region is advanced in the -z plane direction of the z-cut substrate .
(2) The method of forming a micro periodic polarization inversion structure in the ferroelectric nonlinear optical crystal of the present invention is as follows:
In claim 1,
The first electrode has a triangular shape in the yz cross-sectional view of the z-cut substrate, and one end thereof is in contact with the + z plane of the z-cut substrate.
(3) The method of forming a micro periodic polarization inversion structure in the ferroelectric nonlinear optical crystal of the present invention is as follows:
In claim 1,
The shape of the first electrode is an obtuse triangle in the yz cross-sectional view of the z-cut substrate, and the obtuse angle tip is brought into contact with the + z plane of the z-cut substrate.
(4) The method of forming a micro periodic polarization inversion structure in the ferroelectric nonlinear optical crystal of the present invention is as follows:
In claim 1,
The first electrode has a circular shape in the yz cross-sectional view of the z-cut substrate, and a part of the circle is brought into contact with the + z plane of the z-cut substrate.
(5) The minute periodic polarization inversion structure in the ferroelectric nonlinear optical crystal of the present invention is:
It formed by the method in any one of Claims 1-4 .
(6) The minute periodic polarization inversion structure in the ferroelectric nonlinear optical crystal of the present invention is:
5. A micro periodic polarization reversal structure formed by the method according to claim 1 , wherein the inversion region is narrower than the non-inversion region.
(7) The optical waveguide device of the present invention is
An optical waveguide device comprising an optical waveguide, wherein the periodic polarization reversal structure according to claim 5 or 6 is provided in the optical waveguide.
(8) The optical waveguide device of the present invention, in claim 7,
It functions as a wavelength conversion element of a quasi phase matching system.

Hereinafter, the present invention will be described in detail with reference to the drawings. FIG. 1 is a schematic perspective view of an optical waveguide device having a minute periodic polarization reversal structure formed by the method of the present invention. In FIG. 1, in the optical waveguide element 10 formed on the ferroelectric nonlinear optical crystal substrate 11, a minute polarization inversion structure 12 is formed for each coherent length (lc) in order to achieve phase matching. In order to obtain a blue-violet light source, it is desirable that the minute domain inversion structure 12 has a very narrow period of domain inversion of about 2 μm.
As the crystal used for the substrate 11 used in the present invention, a ferroelectric nonlinear optical crystal is used. However, the LN crystal and the LT crystal are uniaxial crystals belonging to the trigonal system, have a large nonlinear optical constant, and have a large optical size. It is widely used in QPM-SHG devices because it has the advantage that crystals can be obtained easily. The LN crystal is a promising crystal as a nonlinear optical material because a large wafer having a diameter of 5 inches can be easily obtained and the nonlinear optical constant is the largest among materials that do not absorb near the ultraviolet region. However, this crystal has a problem of optical damage, and it is necessary to devise such as heating to obtain a high output.
LT has a slightly smaller nonlinear optical constant than LN, but is said to be a crystal resistant to optical damage.Since the absorption edge on the short wavelength side is at 280 nm, conversion to ultraviolet is possible. Like LN crystals, LT is considered to be a promising crystal.
Since the LN crystal and the LT crystal have a 180 domain structure and a periodic polarization inversion layer can be produced by inverting the direction of spontaneous polarization, phase matching can be achieved by the QPM method. Fabrication of a highly efficient QPM-SHG device requires LN (Periodically-Poled-Lithium-Niobate: PPLN) and LT (Periodically-Poled-Lithium-Tantalate: PPLT) having a periodically poled structure. It can be said that LN crystal and LT crystal are suitable materials for fabrication of high-efficiency QPM-SHG devices.

In the following, the method for forming a minute periodic polarization reversal structure of the present invention will be described in detail step by step.
(Preparation of periodic resist pattern using laser beam lithography system)
One of the most important things in the fabrication of QPM-SHG devices is the formation of a periodically poled structure. To form the fine periodically poled ferroelectric nonlinear optical crystal substrate, it is necessary to perform a group of the periodic electrode production "Preparation of periodic fine resist pattern", firstly, the minute period using a laser drawing apparatus The preparation of the resist pattern will be described.
FIG. 2 shows a procedure for producing a minute periodic resist pattern in the present invention. On the thickness 500 mm in z-cut substrate 21 of the LN crystal wafers, the waveform resist pattern formed by the following steps by FCE method. First, the ferroelectric nonlinear optical crystal substrate 11 to which the resist is applied is cleaned. After the crystal substrate 11 is washed, a resist 22 is applied (1.3 μm thickness), and a pattern having a sin-like light intensity is exposed by a two-beam interference exposure method . In the two-beam interference exposure, an Ar + laser (wavelength: 488 nm, 1 mmφ) was divided into two beams by a beam splitter. Each irradiation light intensity was 1.2 mW, and the exposure time was 15 to 60 seconds. The incident angle of the laser beam was made to interfere at q = 14 ° so that the period was 2 μm. Thereafter, development was performed, and an Al electrode (FCE) as the first electrode 23 was produced from the top by vapor deposition. On the −z plane, a circular electrode, which is the second electrode 24 that can cover the entire size of the + z plane electrode, was produced by sputtering.

Next, in room temperature Si oil, a high voltage pulse generator and a DC high voltage power source were connected in series as a bias voltage between the first and second electrodes, and an appropriate number of high voltage pulses for polarization inversion were applied. . Since the polarization reversal threshold of the crystal used was Vth = 10.7 kV (reversal threshold electric field 21.4 kV / mm), the potential difference applied to the crystal was 11.1 kV (22.0 kV / mm) and the pulse width was 1 ms. The XY recorder was used to confirm the state of polarization reversal during electric field application and the current flowing. After the electric field was applied, the electrode was removed by aqua regia etching and etched with a hydrofluoric acid mixed solution . For LN, the mixing ratio was HF: HNO3 = 1: 2, the liquid temperature was 60 ° C., and the etching time was 2 minutes. The -z face of the LN crystal is selectively etched and the + z face is not etched. Therefore, in the region where the polarization inversion is performed, the + z plane is concave and the −z plane is convex. The formed etching pattern was observed with an electron microscope (SEM).

FIG. 3 shows an SEM observation image of a 2 μm periodic polarization reversal shape. In the etching shape in the vicinity of the point contact in the yz cross-sectional shape (Line B in FIG. 2), the inversion region width L and the non-inversion region width S are 0.84 μm and 1.16 μm, respectively. An etching shape narrower than the non-inversion region was obtained . In the present invention, since polarization inversion is performed by a point contact electrode in the yz cross-sectional shape, a very small inversion region width can be obtained in this way. The reason for this will be described below.

In the model of the growth process of domain inversion that takes into account the electrostatic field distribution using the conventional electric field application method, the domain inversion is perpendicular to the -z plane from the inverted nucleus generated at the electrode end where the electric field is most concentrated. It has been thought that it grows so as to penetrate and proceeds in the lateral direction with a strong electric field based on the inversion region. In addition, since the strongest electric field exists outside the electrode width, it has been considered that the domain-inverted region extends considerably beyond the electrode width . In order to demonstrate the superiority of the method for forming the inversion structure of the present invention, FIG. 4 shows an example of the results of experiments conducted by the inventor to verify the conventional polarization inversion mechanism. The growth process in which very small inversion regions (inversion nuclei) expand with the electric field application time was examined in detail.

First, using a LT crystal, 500μm thick, harmonic composition z-cut substrate, resist with 12μm period (electrode width: resist portion = 1: 1) on the + z surface of the crystal by electron beam drawing method The pattern was made to be parallel to the x axis. The EB resist film thickness was 0.69 μm.
Next, an Al electrode (FCE method) 43 covering the entire surface of the produced EB resist pattern was formed on the + z plane, and an Au electrode 44 having the same size as the + z plane was formed on the -z plane. A high voltage pulse for polarization reversal was applied in Si oil at room temperature. The pulse voltage applied between the electrodes was 10.6 to 11.1 kV (DC bias voltage (7.5 kV to 8.0 kV) + pulse voltage (−3.1 kV)), and an appropriate number of pulses was applied with a pulse interval of 5 sec. The pulse width was 100 μs to 1 ms.
After confirming the polarization reversal current, cross-section polishing was performed, and etching (fluid temperature 60 ° C., etching time 15 minutes) was performed with a hydrofluoric acid mixed solution (HF: HNO3 = 2: 1).
In the same manner, a plurality of samples with the number of applied pulses increased little by little were prepared, and the + z plane and the cross section were observed. The etched shape was observed with an optical microscope (OM) and an electron microscope (SEM).

As a result, when the short pulse electric field application is performed and the integrated current value is increased, the cross-sectional etching shape is such that the center position of the inversion nucleus 41 protrudes from the end of the first electrode 43 by about 0.7 μm. Occurred and a depth of 290 nm was observed, and a triangular shape was observed (see FIG. 4A). Next, the inverted nuclei grew in a diagonally downward (inside electrode) direction (see FIG. 4B). In the middle of these, it was observed that an inversion region appeared in the vicinity of the depth of 30 to 150 μm from the + z plane independently of the polarization inversion of the crystal surface, and the inversion shape grew in the −z plane direction (FIG. 4 ( c) and (d)). In this inversion region, an inversion shape that reached the + z plane at the center of the electrode while avoiding the area immediately below the end of the electrode 43 was obtained (see FIG. 4E). These inversion regions grew further as a whole, and the inversion dimension when the inversion shape of the + z plane became uniform was 11.4 μm , and a polarization inversion region extending almost twice as large as the electrode width of 6 μm was observed. (See FIG. 4 (f)).

There is a slight difference between these states depending on the crystal. From a large number of experimental examples, these states are schematically shown in FIG . Thus, in the polarization inversion by the voltage application method using the conventional rectangular electrode, the generation position of the inversion nuclei is outside the electrode end, and as a result, the inversion region is wider than the width of the electrode. It was found that polarization reversal using a rectangular electrode is disadvantageous for minute periodic polarization reversal.

In contrast, a method for forming a fine periodic polarization inversion structure according to the present invention we describe below.
FIG. 6 is a schematic diagram showing an electric field distribution (maximum electric field intensity *) and a polarization inversion shape when the circular electrode of the present invention (the actual shape is an example of a corrugated electrode) is used. As shown in FIG. 6 , in the polarization inversion using the point contact electrode (first electrode) of the present invention, the polarization inversion starts immediately below the electrode end, and the inversion region width is narrower than the non-inversion region width. can get. The reason for this will be described using the first embodiment of the present invention.

(First embodiment)
(Applied electric field distribution in circular electrode)
FIG. 7 shows the first embodiment of the forming method of the present invention, and shows the calculation result of the electric field distribution in the z direction when FCE is formed on a waveform resist pattern. The relative dielectric constant of the resist is 3.0. FIG. 7A shows the calculation result of the applied electric field distribution corresponding to the case where the diameter of the corrugated circle of the first electrode is large (the radius of curvature of the electrode at the contact point is large). The electric field strength of the electrode part of the first electrode was 8 (8.36), the resist part was 0 (0.92), and the E / ER value was 9.08. This value is 1.7 times the E / ER value of LN crystal lift-off.

Here, the field intensity ratio A (E / ER), the z component of the electric field intensity in the central portion of the resist just under the crystalline substrate surface and ER, the electric field strength of the electrode directly below the central portion when the E, and the electrode portion This is the ratio of the electric field strength to the resist part. Polarization inversion occurs where the electric field strength is high and does not occur where the electric field strength is low. There is a threshold for polarization reversal, and the larger the value of (E / ER), the easier it is to form a polarization reversal with a micro period. The reason for this is that the inventor separately observed the inversion formation in the 12 μm period of the LT crystal composed of the lift-off electrode and the FCE method by OM and investigated the electric field distribution of the z component corresponding to the electrode configuration as follows. Because it can be said. That is, when E / ER was as small as 3.46, the widths of the inversion region and the non-inversion region were 11.4 μm and 0.6 μm, respectively, and most of the regions were inverted. On the other hand, when E / ER was as large as 7.96, the widths of the inversion region and the non-inversion region were 9.0 μm and 3.0 μm, respectively, and the non-inversion region appeared clearly.

FIGS. 7A, 7B and 7C show the diameters of the corrugated circles of the first electrode (twice the radius of curvature of the electrode at the contact point) of about 4 μm, 2 μm and 1 μm, respectively. It is a measurement result of the applied electric field distribution corresponding to the case of having. When the diameter was 2 μm and 1 μm, the values of E / ER were 16.2 and 23.0. This value is 3.1 times and 4.4 times the value of the lift-off E / ER of the LN crystal, respectively. Thus, when the diameter of the contact circle of the first electrode at the contact point is reduced, the value of E / ER is increased .

Figure 8 is the substrate y-z cross section is an explanatory diagram showing the electric field distribution in the case of changing the diameter of the first electrode circle contact point that contacts the optical crystal substrate. FIG. 8 shows the calculation result when the boundary region between the crystal part at the center of the applied electric field distribution and the resist part is enlarged and the aspect ratio is 1. If the diameter of the circle is 1 μm , the middle and large are about 2 μm and 4 μm , respectively. The strongest electric field (indicated by * in the figure) is present in the resist portion when the diameter is 1 to 4. Also, the strong electric field in the crystal substrate that existed when the diameter was 1 μm gradually leaked to the resist part as the diameter increased from 2 μm to 4 μm , and the E / ER value decreased. became. Further, the difference in the electric field at the boundary between the electrode portion on the crystal substrate surface and the resist portion became smaller as the diameter increased.
Thus, by making the shape of the first electrode circular, a strong electric field on the surface of the crystal substrate does not exist in the crystal, and the maximum electric field can exist in the resist portion, reducing the influence of the piezoelectric effect. is, Ru can be generated inverted nuclei from a relatively strong electric field area of the electrode directly below the crystalline substrate. In the present embodiment, the circular diameter of the first electrode in contact with the crystal substrate has been described as 1 to 4 μm. However, in the present invention, if the cross-sectional shape is a point contact, the diameter is less than this, or more Obviously, it includes those of different diameters.

(Second Embodiment)
(Applied electric field distribution in triangular electrode)
FIG. 9 shows a second embodiment of the forming method of the present invention, and shows the calculation result of the electric field distribution in the z direction when FCE is fabricated on a triangular resist pattern. The dielectric constant of the resist was 3.0. That is, when the first electrode and a triangular cross section, the substrate y-z cross section is a description view showing an electric field distribution in the case of changing the angle of the first electrode at the point of contact with the substrate . The electric field strengths in the crystal and in air are indicated by scales and numbers. In addition, the relative magnitude of the Ez component is expressed in 11 steps with the maximum intensity Ezmax being 10 (maximum value 10).

FIG. 9A shows the calculation result of the applied electric field distribution when the triangular angle (electrode contact angle) is 150 °. The electric field strength of the electrode part was 8 (8.37), the resist part was 0 (0.48), and the E / ER value was 17.4. This value is 3.3 times the E / ER value (5.22) of the lift-off of the LN crystal. The maximum electric field of the electrode part in contact with the crystal substrate surface was 8, and the electric field inside the crystal was 1 and uniform. FIGS. 9B, 9C, and 9D are calculation results of the applied electric field distribution when the electrode contact angles are 120 °, 90 °, and 40 °, respectively. The values of the electric field strength ratio E / ER at 120 °, 90 °, and 40 ° were 26.9, 37.8, and 48.2, respectively. In particular, at 40 °, the value was 9.2 times higher than the E / ER value of the lift-off of the LN crystal. The smaller the apex angle of the first electrode in contact with the crystal substrate and the electrode at the contact point, the greater the value of the electric field strength ratio E / ER, if not ∞. Regardless of the apex angle of the electrode at the contact point, the electric field strength on the surface of the crystal substrate in point contact appears greatly, and the electric field is concentrated. This strong electric field region exists only in the vicinity of the surface of the crystal substrate, and the electric field strength inside the crystal suddenly dropped to 1, indicating a uniform value with almost no difference in electric field.
It was found that the strongest electric field existing on the crystal substrate surface gradually leaked to the resist portion as the angle became gentle (the apex angle of the electrode increased). Further, the difference in electric field at the boundary between the electrode portion and the resist portion on the surface of the crystal substrate became smaller as the angle became milder. Thus, by making the shape of the first electrode triangular , the maximum electric field on the surface of the crystal substrate does not exist in the crystal, and as with the circular electrode, the polarization inversion is performed with the first inversion current density being large. It can be generated from directly under the electrode.

(Dependence of electric field distribution on triangular electrode)
FIG. 10 is an explanatory diagram showing angle dependency when the shape of the first electrode is triangular. That is, the calculation result is shown when the boundary region between the crystal part and the resist part (ε = 3.0) in the center of the applied electric field distribution shown in FIG. FIG. 10 shows the dependence of the applied electric field distribution on the apex angle of the triangular electrode (first electrode) when the respective angles are 40 °, 90 °, 120 °, and 150 °. When the electrode contact angle is 40 ° and 90 °, the strongest electric field exists only at the portion where the electrode is in contact with the crystal substrate. When the electrode contact angle is 120 °, the strongest electric field appears in the resist portion and the crystal portion, and in the case of 150 °, the strongest electric field existing in the crystal moves to the resist portion. As the angle becomes gentler, the strongest electric field present on the crystal substrate surface gradually leaks to the resist portion, and the value of E / ER decreases. It can also be seen that the difference in electric field at the boundary between the electrode portion on the crystal substrate surface and the resist portion becomes smaller as the angle becomes gentler . In the present embodiment, the first electrode in contact with the crystal substrate is described as having a triangular shape.However, in the present invention, the cross-sectional shape may be in a point contact state with the crystal substrate, and the first electrode is not limited to the triangular shape. There is no broad concept.

  In the above embodiment, the case of point contact in the yz section has been described as an example. However, as shown in FIG. 2, the first electrode has a linear shape in the x-axis direction of the substrate and the substrate. It can also have continuous contact points.

According to the present invention, it is possible to easily form a minute periodic polarization reversal structure, which has been difficult with a polarization reversal method using a conventional rectangular electrode shape. In addition, the present invention extends the FCE method and provides a new method necessary for fabrication of minute periodic polarization inversion. That is, by changing the rectangular electrode used conventionally to a triangular electrode and a circular electrode that are in point contact, the value of the electric field strength ratio E / ER can be increased, and the maximum electric field strength (* By moving the mark) from the inside of the crystal to the inside of the resist , a minute periodic polarization reversal structure having an applied electric field distribution that can suppress the piezoelectric effect can be obtained.

It is a schematic perspective view of the optical waveguide element which has the micro periodic polarization inversion structure formed by the method of this invention. It is a preparation procedure of the micro periodic resist pattern in this invention. The SEM observation image of a 2 micrometer period polarization inversion shape by this invention circular electrode (waveform shape electrode using a laser interference exposure method) is shown. This shows a growth process in which a very small inversion region (inversion nucleus) spreads with the electric field application time. It is a figure which shows typically a mode that the generation | occurrence | production position of the inversion nucleus has arisen in the non-electrode part avoiding the area | region where the integrated electric field value seen from the crystal substrate surface changes rapidly. The circular electrode of the present invention (whole shaped electrode as a whole), the maximum electric field strength (* mark), the generation and growth of inversion nuclei when using the FCE method, and a schematic diagram showing the polarization inversion width are shown. The width can be controlled by the applied pulse voltage and the number of pulses. FIG. 4 shows a first embodiment of the forming method of the present invention, and is a calculation result of an electric field distribution in the z direction when an FCE is formed on a corrugated resist pattern. It is explanatory drawing which shows electric field distribution at the time of changing the diameter of the 1st electrode circle of the contact point which contacts an optical crystal substrate in a board | substrate yz cross section. FIG. 6 shows a second embodiment of the forming method of the present invention, and is a calculation result of an electric field distribution in the z direction when FCE is produced on a triangular resist pattern. It is explanatory drawing which shows angle dependence when the shape of a 1st electrode is made into a triangular shape. It is a schematic perspective view which shows the conventional nonlinear optical crystal.

Explanation of symbols

10: Optical waveguide element 11: Ferroelectric nonlinear optical crystal substrate 12: Minute polarization inversion structure 21: z-cut Substrate 22: Resist 23: First electrode 24: Second electrode 43: Al electrode 44: Au electrode 100 : Nonlinear optical crystal 110: Periodic polarization inversion structure 120: Optical waveguide 130: Incident light (pump light)

Claims (8)

A first electrode provided on the + z plane which is one surface of the z-cut substrate of the ferroelectric nonlinear optical crystal substrate;
By applying a voltage between the second electrode provided on the -z plane which is the other surface of the z-cut substrate so as to face the first electrode,
A method of forming a micro periodic polarization reversal structure in the ferroelectric nonlinear optical crystal substrate,
The first electrode is brought into point contact with the + z plane of the z-cut substrate in a yz cross-sectional view perpendicular to the polarization reversal period direction in the substrate ,
So as to continuously contact linearly in the x-axis direction parallel to the polarization inversion period direction in the substrate,
The second electrode covers the electrode area of the first electrode in contact with the + z plane,
Contacting the -z surface of the z-cut substrate;
Applying a pulse voltage between the first electrode and the second electrode;
Microperiodic polarization in a ferroelectric nonlinear optical crystal characterized in that a domain-inverted nucleus is generated immediately below the first electrode and the domain-inverted region is advanced in the -z plane direction of the z-cut substrate. Inversion structure forming method. The shape of the first electrode is a triangle shape in the yz cross-sectional view of the z-cut substrate, and one end thereof is brought into contact with the + z plane of the z-cut substrate. A method for forming a minute periodic domain-inverted structure in the ferroelectric nonlinear optical crystal as described. The shape of the first electrode is an obtuse triangular shape in the yz cross-sectional view of the z-cut substrate, and the obtuse angle tip is in contact with the + z plane of the z-cut substrate. A method for forming a minute periodic domain-inverted structure in a ferroelectric nonlinear optical crystal according to 1. The shape of the first electrode is circular in the yz cross-sectional view of the z-cut substrate, and a part of the circle is brought into contact with the + z plane of the z-cut substrate. 2. A method for forming a minute periodic domain-inverted structure in a ferroelectric nonlinear optical crystal according to 1. Either, characterized in that it is formed by the method of claim, strong fine periodically poled in the dielectric nonlinear optical crystal according to claim 1 to 4. A ferroelectric nonlinear optical crystal having a microperiodic domain-inverted structure formed by the method according to any one of claims 1 to 4 , wherein the inversion region is narrower than the non-inversion region. Micro periodic polarization reversal structure. An optical waveguide device comprising an optical waveguide, wherein the periodic polarization reversal structure according to claim 5 or 6 is provided in the optical waveguide. 8. The optical waveguide device according to claim 7 , wherein the optical waveguide device functions as a quasi-phase matching type wavelength conversion device.