Abstract
Provided is an optical modulator which is small in optical loss, is small in a size, and is low in required voltage and is operable to perform high-speed operation. The optical phase modulator 100 comprises a rib-type waveguide structure 110 including: a PN junction 106 which is formed of Si and is formed in a lateral direction on a substrate; and an Si.sub.1-xGe.sub.x layer 108 which is constituted of at least one layer and is doped with an impurity to a p-type and is superposed on the PN junction 106 so as to be electrically connected to the PN junction 106. The rib-type waveguide structure 110 has a substantially uniform structure along a light propagation direction, and in a direction parallel with the substrate and perpendicular to the light propagation direction, a position of a junction interface 106a of the PN junction 106 is offset from a center of the Si.sub.1-xGe.sub.x layer 108.
Claims
1. An optical phase modulator comprising a rib-type waveguide structure, the rib-type waveguide structure comprising: a PN junction or a PIN junction which is formed of Si or Si.sub.1-yGe.sub.y and is formed in a lateral direction on a substrate, the PN junction or the PIN junction comprising a p-type region and an n-type region; and a p-type Si.sub.1-yGe.sub.y layer which is constituted of at least one layer and is doped with a p-type impurity, the p-type Si.sub.1-yGe.sub.y layer being superposed on the PN junction or the PIN junction so as to be electrically connected to the PN junction or the PIN junction, wherein the rib-type waveguide structure has a substantially uniform structure along a light propagation direction, and in a direction parallel with the substrate and perpendicular to the light propagation direction, a position of a junction interface of the PN junction or the PIN junction is offset from a center of the p-type Si.sub.1-yGe.sub.y layer so that a contact area where the p-type Si.sub.1-yGe.sub.y layer contacts the n-type region is larger than a contact area where the p-type Si.sub.1-yGe.sub.y layer contacts the p-type region, so as to increase an electrical capacitance of the PN junction or the PIN junction.
2. The optical phase modulator according to claim 1, wherein the position of the junction interface of the PN junction or the PIN junction is offset to a direction of the p-type region.
3. The optical phase modulator according to claim 1, comprising: a first electrode of a first conductive type; and a second electrode of a second conductive type which neighbor the rib-type waveguide structure, wherein by applying a voltage to each of the first electrode and the second electrode, a carrier density in the rib-type waveguide structure is changed.
4. The optical phase modulator according to claim 1, herein the type Si.sub.1-xGe.sub.x layer being constituted of the at least one layer has lattice strain.
5. The optical phase modulator according to claim 1, wherein the PN junction or the PIN junction being formed in the lateral direction on the substrate is constituted of a layered structure of a PN junction or a PIN junction which is formed of Si and a PN junction or a PIN junction which is formed of Si.sub.1-yGe.sub.y.
6. The optical phase modulator according to claim 5, wherein the PN junction or the PIN junction being formed in the lateral direction on the substrate and being constituted of the layered structure of Si and Si.sub.1-yGe.sub.y includes a rib-type waveguide structure.
7. The optical phase modulator according to claim 1, wherein the type Si.sub.1-xGe.sub.x layer being constituted of the at least one layer includes: an Si.sub.1-xGe.sub.x1 layer being superposed on the PN junction or the PIN junction; and an Si.sub.1-x2Ge.sub.x2 layer being superposed on the Si.sub.1-x1Ge.sub.x1 layer, and x2 is smaller than x1.
8. The optical phase modulator according to claim 1, wherein the type Si.sub.1-xGe.sub.x layer being constituted of the at least one layer includes a strain induction film being formed above the p-type Si.sub.1-xGe.sub.x layer or on a side surface thereof.
9. The optical phase modulator according to claim 1, wherein in the PN junction or the PIN junction being formed in the lateral direction on the substrate or in the type Si.sub.1-xGe.sub.x layer being constituted of the at least one layer and being superposed on the PN junction or the PIN junction so as to be electrically connected to the PN junction or the PIN junction, a doping concentration of a first conductive type is smaller than a doping concentration of a second conductive type.
10. The optical phase modulator according to claim 9, wherein the first conductive type is a p-type and the second conductive type is an n-type.
11. The optical phase modulator according to claim 1, wherein the PN junction is constituted of a single PN junction which is formed of a single p-type region and a single n-type region which extend in the light propagation direction.
12. The optical phase modulator according to claim 1, wherein in the p-type Si.sub.1-xGe.sub.x layer, x is in a range of 0x<0.6.
13. The optical phase modulator according to claim 1, wherein crystal orientation of the substrate is <110>.
14. An optical intensity modulator comprising the optical phase modulator according to claim 1.
15. A method for manufacturing an optical phase modulator, the method comprising: a step of forming a PN junction or a PIN junction which is formed of Si or Si.sub.1-yGe.sub.y and is formed in a lateral direction, the PN junction or the PIN junction comprising a p-type region and an n-type region; a step of forming first conductive type-doping and second conductive type-doping regions which respectively neighbor the PN junction or the PIN junction and of forming a first electrode of a first conductive type and a second electrode of a second conductive type; and a step of, on the PN junction or the PIN junction, forming a p-type Si.sub.1-yGe.sub.y layer which is constituted of at least one layer and is electrically connected to the PN junction or the PIN junction, the p-type Si.sub.1-yGe.sub.y layer being doped with a p-type impurity, wherein by the PN junction or PIN junction and the p-type Si.sub.1-yGe.sub.y layer, a rib-type waveguide structure having a substantially uniform structure along a light propagation direction is formed, and in a direction parallel with a substrate and perpendicular to the light propagation direction, a position of a junction interface of the PN junction or the PIN junction is offset from a center of the p-type Si.sub.1-yGe.sub.y layer so that a contact area where the p-type Si.sub.1-yGe.sub.y layer contacts the n-type region is larger than a contact area where the p-type Si.sub.1-yGe.sub.y layer contacts the p-type region, so as to increase an electrical capacitance of the PN junction or the PIN junction.
Description
BRIEF DESCRIPTION OF DRAWINGS
(1) FIG. 1 is a cross-sectional view schematically illustrating an optical phase modulator 100 according to one embodiment of the present invention.
(2) FIG. 2 is a perspective view schematically illustrating the optical phase modulator 100 illustrated in FIG. 1.
(3) FIG. 3 shows illustrative computation results of relationship between an offset amount of a junction interface 106a of a PN junction 106 and a figure of merit V.sub.L.
(4) FIG. 4 shows illustrative experimental results of relationship between an electrical capacitance C of the PN junction 106 and a modulation efficiency (the figure of merit) V.sub.L.
(5) FIG. 5 is a cross-sectional view schematically illustrating an optical phase modulator 500 according to another embodiment of the present invention.
(6) FIG. 6 is a cross-sectional view schematically illustrating an optical phase modulator 600 according to still another embodiment of the present invention.
(7) FIG. 7 is a cross-sectional view schematically illustrating an optical phase modulator 700 according to yet another embodiment of the present invention.
(8) FIG. 8 is a cross-sectional view schematically illustrating an optical phase modulator 800 according to a further embodiment of the present invention.
(9) FIG. 9 is a cross-sectional view schematically illustrating an optical phase modulator 900 according to a still further embodiment of the present invention.
(10) FIG. 10 is a cross-sectional view schematically illustrating an optical phase modulator 1000 according to a yet further embodiment of the present invention.
(11) FIG. 11A is a diagram explaining a process of manufacturing the optical phase modulator according to the embodiment of the present invention illustrated in FIG. 1.
(12) FIG. 11B is a diagram explaining a process of manufacturing the optical phase modulator according to the embodiment of the present invention illustrated in FIG. 1.
(13) FIG. 11C is a diagram explaining a process of manufacturing the optical phase modulator according to the embodiment of the present invention illustrated in FIG. 1.
(14) FIG. 11D is a diagram explaining a process of manufacturing the optical phase modulator according to the embodiment of the present invention illustrated in FIG. 1.
(15) FIG. 11E is a diagram explaining a process of manufacturing the optical phase modulator according to the embodiment of the present invention illustrated in FIG. 1.
(16) FIG. 11F is a diagram explaining a process of manufacturing the optical phase modulator according to the embodiment of the present invention illustrated in FIG. 1.
(17) FIG. 11G is a diagram explaining a process of manufacturing the optical phase modulator according to the embodiment of the present invention illustrated in FIG. 1.
(18) FIG. 11H is a diagram explaining a process of manufacturing the optical phase modulator according to the embodiment of the present invention illustrated in FIG. 1.
(19) FIG. 12 is a diagram illustrating a typical example of the conventional silicon-based electro-optic phase modulator.
DESCRIPTION OF EMBODIMENTS
(20) Hereinafter, embodiments of the present invention will be explained in detail with reference to the accompanying drawings. FIG. 1 is a cross-sectional view schematically illustrating an optical phase modulator 100 according to one embodiment of the present invention. In the present embodiment, the optical phase modulator 100 is formed on a buried oxide film (BOX) layer 104 formed of silica glass (SiO.sub.2) deposited on a silicon (Si) substrate 102 by employing the silicon photonics technology. The optical phase modulator 100 includes a rib-type waveguide structure 110. The rib-type waveguide structure 110 includes a PN junction 106 which is formed in a lateral direction with respect to the substrate (a horizontal direction with respect to the substrate) and is formed of Si. The PN junction 106 includes a p-type Si region 130 and an n-type Si region 132. As described later, the PN junction 106 may be formed of Si.sub.1-yGe.sub.y (where y is equal to or greater than 0 and equal to or less than 1). The rib-type waveguide structure 110 may include a PIN junction, instead of the PN junction 106. The PIN junction mentioned above may include both a PIN junction intentionally formed and a PIN junction modified from a PN junction and thereby unintentionally formed. The case in which the PIN is unintentionally formed is a case in which in the PN junction 106, electrons and positive holes are recombined with each other due to thermal diffusion and become an I layer. The rib-type waveguide structure 110 also includes an Si.sub.1-xGe.sub.x (where x is equal to or greater than 0 and equal to or less than 1) layer 108 which has conductivity obtained by impurity doping and is constituted of at least one layer superposed on the PN junction 106. In the present embodiment, although a conductive type of the Si.sub.1-xGe.sub.x layer 108 is a p-type, the conductive type of the Si.sub.1-xGe.sub.x layer 108 may be an n-type. As described above, the Si.sub.1-xGe.sub.x layer 108 has the conductivity and accordingly, is electrically connected to the PN junction 106.
(21) As shown in FIG. 1, as one example, a width of the Si.sub.1-xGe.sub.x layer 108 is 0.4 m and a thickness of the Si.sub.1-xGe.sub.x layer 108 is 20 nm to 80 nm. In addition, as one example, a height of the PN junction in the rib-type waveguide structure 110 is 80 nm to 100 nm with respect to a slab layer as shown in FIG. 1. In addition, as one example, a thickness of each of electrodes 112 and 114 is 0.10 m to 0.18 m.
(22) A junction interface 106a of the PN junction 106 is disposed in a position different from a position where a center 108a of the Si.sub.1-xGe.sub.x layer 108 superposed on the PN junction 106 is located. In other words, the position where the junction interface 106a of the PN junction 106 is located is offset from the center 108a of the Si.sub.1-xGe.sub.x layer 108. An offset direction is parallel with the substrate 102 and is in a direction perpendicular to a longitudinal direction (that is, a light propagation direction) of the rib-type waveguide structure 110. In an example shown in FIG. 1, the position of the junction interface 106a of the PN junction 106 is offset from the center 108a of the p-type Si.sub.1-xGe.sub.x layer 108 in a direction toward the p-type Si region 130 (hereinafter, referred to as a minus direction). Therefore, in a portion of the rib-type waveguide structure 110, a contact area where the p-type Si.sub.1-xGe.sub.x layer 108 contacts the n-type Si region 132 is larger than a contact area where the p-type Si.sub.1-xGe.sub.x layer 108 contacts the p-type Si region 130. In other words, in the optical phase modulator 100 according to the present embodiment, as compared with a structure in which the junction interface 106a of the PN junction 106 is disposed in such a way as to coincide with the center 108a of the Si.sub.1-xGe.sub.x layer 108, the contact area where the p-type Si.sub.1-xGe.sub.x layer 108 contacts the n-type Si region 132 is enlarged.
(23) The optical phase modulator 100 also includes a first electrode 112 of a first conductive type (for example, p-type) and a second electrode 114 of a second conductive type (for example, n-type), which neighbor the rib-type waveguide structure 110, a grounding electrode 116, a signal electrode 118, and a clad 120. Each of the grounding electrode 116 and the signal electrode 118 may be formed of, for example, Ti, TiN, AlSiCu, TiN, or the like. Via the grounding electrode 116 and the signal electrode 118, a voltage is applied to each of the first electrode 112 and the second electrode 114, thereby changing a carrier density in the rib-type waveguide structure 110.
(24) FIG. 2 is a perspective view schematically illustrating the optical phase modulator 100 illustrated in FIG. 1. As shown in FIG. 2, the rib-type waveguide structure 110 of the optical phase modulator 100 has a uniform structure along the longitudinal direction (that is, the light propagation direction). More specifically, in the rib-type waveguide structure 110, the p-type Si region 130 integrally extends along the light propagation direction as a single region, and each shape of cross sections of the p-type Si region 130 as a face which is perpendicular to the light propagation direction is invariable, regardless of any cross-sectioned positions. Similarly, the n-type Si region 132 integrally extends along the light propagation direction as a single region, and each shape of cross sections of the n-type Si region 132 as a face which is perpendicular to the light propagation direction is invariable, regardless of any cross-sectioned positions. Accordingly, the junction interface 106a of the PN junction 106 is formed in such a way as to extend over the entire length of the rib-type waveguide structure 110 along the light propagation direction as one planar surface parallel with the light propagation direction. As described above, the PN junction 106 in the rib-type waveguide structure 110 is configured by the single PN junction constituted of the single p-type Si region 130 and the single n-type Si region 132 which extend in the light propagation direction.
(25) The optical phase modulator 100 utilizes the carrier plasma effect to phase-modulate light. A change in refractive index which is attained by plasma dispersion effect of the carrier plasma effect can be represented by the following formula.
(26)
(27) Here, n denotes a change in refractive index, e denotes a unit charge, denotes a light wavelength, c denotes a light speed, .sub.0 denotes a permittivity of vacuum, n denotes a refractive index of Si, N.sub.e denotes a change in electron density, m*.sub.ce denotes an effective mass of electrons, N.sub.h denotes a change in positive hole density, and m*.sub.ch denotes an effective mass of positive holes. As can be understood from the formula (1), when the effective mass of the electrons or the effective mass of the positive holes is decreased, n which is the change in refractive index is increased.
(28) An amount of a change in phase produced in the rib-type waveguide structure 110 of the optical phase modulator 100 can be represented as in the following formula by using the change in refractive index n in the formula (1).
(29)
(30) Here, L denotes a length of the rib-type waveguide structure 110 along the light propagation direction, C denotes an electrical capacitance of the PN junction 106, and V denotes a voltage applied to the PN junction 106. The formula (2) indicates that the applied voltage V required to produce a certain amount of the change in phase is in inverse proportion to the electrical capacitance C of the PN junction 106. In general, as an index of a modulation efficiency in the optical phase modulator, a figure of merit V.sub.L is used. The term V.sub.L is a product of a voltage required to shift the phase by in an optical phase shifter and a length. From the formula (2) and the definition of V.sub.L, it is understood that the figure of merit V.sub.L of the optical phase modulator 100 is in inverse proportion to the electrical capacitance C of the PN junction 106 as in the following formula.
(31)
(32) When the Si.sub.1-xGe.sub.x layer is superposed on the Si layer, due to a difference between a lattice constant of the Si and a lattice constant of the Si.sub.1-yGe.sub.y, strain (lattice strain) is induced in the Si.sub.1-xGe.sub.x layer. When the strain is induced in the Si.sub.1-xGe.sub.x layer, an effective mass of the carriers becomes small. The Si.sub.1-xGe.sub.x layer 108 is used, thereby reducing the effective mass of the free carriers. Accordingly, as can be understood from the formula (1), since the change in the refractive index, which is attained by the plasma dispersion effect, becomes large, the carrier plasma effect can be enhanced. Accordingly, since a phase shift amount required at a shorter distance can be obtained, the modulation efficiency of the optical phase modulator 100 can be ameliorated and a size of the optical phase modulator 100 can be made small, and a loss in the optical phase modulator 100 can be made small. In addition, since the effective mass of the free carriers becomes small, mobility of the free carriers, which is in the relationship of an inverse number to the effective mass, is increased. Accordingly, high-speed operation of the optical phase modulator 100 is enabled.
(33) In addition, as described above, since in the optical phase modulator 100, the contact area where the p-type Si.sub.1-xGe.sub.x layer 108 and the n-type Si region 132 contact each other is large, as compared with a configuration in which the junction interface 106a of the PN junction 106 and the center 108a of the Si.sub.1-xGe.sub.x layer 108 are arranged in such a way as to coincide with each other, the optical phase modulator 100 has the further large electrical capacitance C. Accordingly, as is understood from the formula (3), the modulation efficiency of the optical phase modulator 100 can be further ameliorated.
(34) In the PN junction 106 and/or the Si.sub.1-xGe.sub.x layer 108, a p-type doping concentration may be smaller than an n-type doping concentration. In the SiGe, an enhancement factor attained by the positive holes of the carrier plasma effect is approximately twice as large as that attained by the electrons. Accordingly, the doping concentration of the p-type SiGe layer is made smaller than the doping concentration of the n-type SiGe layer, thereby allowing the trade-off relationship between the refractive index difference and the absorption coefficient to be relaxed. As a result, an increase in the light absorption coefficient is inhibited and speed can be increased.
(35) FIG. 3 shows illustrative computation results of the relationship between the offset amount of the junction interface 106a of the PN junction 106 and the figure of merit V.sub.L as to the above-described optical phase modulator 100. In FIG. 3, each minus sign of a horizontal axis of a graph represents that an offset direction of the junction interface 106a of the PN junction 106 is a minus direction. It can be said that the smaller V.sub.L is, the higher performance of the optical phase modulator is. As shown in FIG. 3, in the optical phase modulator 100 of the present embodiment, the larger the offset amount of the junction interface 106a of the PN junction 106 in the minus direction becomes, the smaller V.sub.L becomes. For example, whereas when the offset amount is zero, V.sub.L is 0.52 Vcm, when the offset amount is 150 nm, V.sub.L is 0.28 Vcm. As described above, the junction interface 106a of the PN junction 106 is offset in the minus direction, thereby reducing V.sub.L and allowing the modulation efficiency of the optical phase modulator 100 to be enhanced. In addition, since a length of the modulator required to obtain a desired change in refractive index is shortened, the size of the optical phase modulator 100 can be made small.
(36) FIG. 4 shows illustrative experimental results of the relationship between the electrical capacitance C of the PN junction and the modulation efficiency (the figure of merit) V.sub.L as to optical phase modulators, which have several different structures, including an optical phase modulator 100 described above. In FIG. 4, a measured value noted with SiGe-MOD (30 nm-offset) shows a result as to an optical phase modulator 100 in which the junction interface 106a of the PN junction 106 is offset by 30 nm in the minus direction, a measured value noted with SiGe-MOD (30 nm-offset) shows a result as to an optical phase modulator 100 in which the junction interface 106a of the PN junction 106 is offset by 30 nm in a plus direction, and a measured value noted with SiGe-MOD (0 nm-offset) shows a result as to an optical phase modulator 100 in which the junction interface 106a of the PN junction 106 is not offset. Note, however, that the plus direction refers to a direction opposite to the minus direction, that is, a direction in which the junction interface 106a of the PN junction 106 is offset to a side of the n-type Si region 132 further than the center 108a of the p-type Si.sub.1-xGe.sub.x layer 108. In addition, in FIG. 4, a measured value noted with Si-MOD shows a result as to an optical phase modulator having a structure in which the Si.sub.1-xGe.sub.x layer 108 is not provided on the PN junction 106, as reference.
(37) Each of the optical phase modulators 100 used to obtain the experimental results shown in FIG. 4 was prepared by the below-described manufacturing method. Specifically, by employing an ultra-high-vacuum CVD method, an Si.sub.0.8Ge.sub.0.2/Si layer was superposed on a PN junction 106. In addition, during film formation of the Si.sub.0.8Ge.sub.0.2/Si layer, diborane (B.sub.2H.sub.6) gas was mixed therewith, thereby doping the Si.sub.0.8Ge.sub.0.2/Si layer with boron (B). The p-type Si.sub.0.8Ge.sub.0.2 layer 108 was thus formed.
(38) When the above-described manufacturing method is employed, hetero-junction of the Si PN junction 106 and the Si.sub.0.8Ge.sub.0.2 layer 108 is comparatively favorably formed. Therefore, a position of the junction interface 106a of the PN junction 106 is offset in the minus direction, and the contact area where the p-type Si.sub.0.8Ge.sub.0.2 layer 108 and the n-type Si region 132 contact each other is made large, thereby allowing the electrical capacitance C of the PN junction 106 to be increased. Accordingly, as described above in regard to the formula (3), the modulation efficiency of the optical phase modulator 100 can be greatly ameliorated. The experimental results shown in FIG. 4 sustain this, and in the case of the configuration of the SiGe-MOD (30 nm-offset), it is shown that the electrical capacitance C is increased and the figure of merit V.sub.L is decreased (in other words, the modulation efficiency is ameliorated).
(39) FIG. 5 is a cross-sectional view schematically illustrating an optical phase modulator 500 according to another embodiment of the present invention. A structure of the optical phase modulator 500 is similar to that of the optical phase modulator 100 shown in FIG. 1 except that two p-type SiGe layers (a p-type Si.sub.1-x1Ge.sub.x1 layer 508 and a p-type Si.sub.1-x2Ge.sub.x2 layer 522) are superposed on a PN junction 506. A rate of Ge in the p-type Si.sub.1-x2Ge.sub.x2 layer 522 may be smaller than a rate of Ge in the p-type Si.sub.1-x1Ge.sub.x1 layer 508 (that is, x1>x2). For example, the p-type Si.sub.1-x1Ge.sub.x1 layer 508 may be a p-type Si.sub.0.7Ge.sub.0.3 layer, and the p-type Si.sub.1-x2Ge.sub.x2 layer 522 may be a p-type Si.sub.0.8Ge.sub.0.2 layer. In addition, the p-type Si.sub.1-x2Gex.sub.2 layer 522 may be an Si layer. As in FIG. 1, in FIG. 5, an Si substrate 502, a BOX layer 504, a PN junction 506, a p-type Si region 530, an n-type Si region 532, a first electrode 512, a second electrode 514, a grounding electrode 516, a signal electrode 518, and a clad 520 are illustrated, and a position of a junction interface 506a of the PN junction 506 is offset in a direction of the p-type Si region 530 (that is, a minus direction) from a center 508a of the p-type Si.sub.1-x1Ge.sub.x1 layer 508. The PN junction 506 may be formed of Si.sub.1-yGe.sub.y. The rib-type waveguide structure 510 may include a PIN junction, instead of the PN junction. A conductive type of each of the Si.sub.1-x1Ge.sub.x1 layer 508 and the Si.sub.1-x2Ge.sub.x2 layer 522 may be an n-type.
(40) In the optical phase modulator 500 shown in FIG. 5, by increasing a thickness of each of the p-type SiGe layers 508 and 522, large strain can be induced. However, when the thickness of each of the SiGe layers superposed on the Si layer has reached a certain film thickness, an absolute amount of the strain of each of the SiGe layers is extremely increased, and bonding in an interface between the Si layer and each of the SiGe layers is broken (which is called lattice relaxation). Accordingly, there is a limit to a thickness of each of the strained SiGe layers which can be superposed thereon (a critical thickness).
(41) By gradually changing composition of each of the SiGe layers in a film thickness direction, lattice defects can be reduced. Accordingly, it is effective to superpose the plurality of SiGe layers which respectively have different kinds of composition. On the other hand, when a rate of the Ge composition is increased, each of the SiGe layers tends to have chemically unstable nature, for example, in which a Ge oxide film dissolves in water. Accordingly, it is useful to use a layer having a small rate of the Ge composition as a topmost layer of the SiGe layers since the topmost layer functions as a protective film. Accordingly, when the plurality of SiGe layers are superposed thereon, it is effective to superpose the layer having the small rate of the Ge composition and the Si layer on a layer having a large rate of the Ge composition.
(42) FIG. 6 is a cross-sectional view schematically illustrating an optical phase modulator 600 according to still another embodiment of the present invention. A structure of the optical phase modulator 600 is similar to that of the optical phase modulator 500 shown in FIG. 5 except that a strain induction film 624 is superposed on a p-type Si.sub.1-x2Ge.sub.x2 layer 622. In the optical phase modulator 600 shown in FIG. 6, instead of superposing the thick SiGe layer, the strain induction film 624 is superposed, thereby enabling large strain. The strain induction film 624 may include, for example, SiN.sub.x or alumina. As in FIG. 5, in FIG. 6, an Si substrate 602, a BOX layer 604, a PN junction 606, a p-type Si region 630, an n-type Si region 632, a p-type Si.sub.1-x1Ge.sub.x1 layer 608, a first electrode 612, a second electrode 614, a grounding electrode 616, a signal electrode 618, and a clad 620 are illustrated, and a position of a junction interface 606a of the PN junction 606 is offset in a direction of the p-type Si region 630 (that is, a minus direction) from a center 608a of the p-type Si.sub.1-x1Ge.sub.x1 layer 608. The PN junction 606 may be formed of Si.sub.1-yGe.sub.y. The rib-type waveguide structure 610 may include a PIN junction, instead of the PN junction. A conductive type of each of the Si.sub.1-x1Ge.sub.x1 layer 608 and the Si.sub.1-x2Ge.sub.x2 layer 622 may be an n-type.
(43) FIG. 7 is a cross-sectional view schematically illustrating an optical phase modulator 700 according to yet another embodiment of the present invention. A structure of the optical phase modulator 700 is similar to that of the optical phase modulator 500 shown in FIG. 5 except that a strain induction film 724 neighboring side surfaces of a p-type Si.sub.1-x2Ge.sub.x2 layer 722 is formed. As shown in FIG. 7, the strain induction film 724 may extend from over a PN junction 706 to over a first electrode 712 and a second electrode 714. The strain induction film 724 may be formed in such a way as to neighbor side surfaces of a p-type Si.sub.1-x1Ge.sub.x1 layer 708. In the optical phase modulator 700 shown in FIG. 7, instead of superposing a thick SiGe layer, the strain induction film layer 724 is formed, thereby enabling large strain. The strain induction film 724 may include, for example, SiN.sub.x or alumina. As in FIG. 5, in FIG. 7, an Si substrate 702, a BOX layer 704, a PN junction 706, a p-type Si region 730, an n-type Si region 732, a p-type Si.sub.1-x1Ge.sub.x1 layer 708, a first electrode 712, a second electrode 714, a grounding electrode 716, a signal electrode 718, and a clad 720 are illustrated, and a position of a junction interface 706a of the PN junction 706 is offset in a direction of the p-type Si region 730 (that is, a minus direction) from a center 708a of the p-type Si.sub.1-x1Ge.sub.x1 layer 708. The PN junction 706 may be formed of Si.sub.1-yGe.sub.y. The rib-type waveguide structure 710 may include a PIN junction, instead of the PN junction. A conductive type of each of the Si.sub.1-x1Ge.sub.x1 layer 708 and the Si.sub.1-x2Ge.sub.x2 layer 722 may be an n-type.
(44) FIG. 8 is a cross-sectional view schematically illustrating an optical phase modulator 800 according to a further embodiment of the present invention. A structure of the optical phase modulator 800 is similar to that of the optical phase modulator 500 shown in FIG. 5 except that a strain induction film 824 on a p-type Si.sub.1-x2Ge.sub.x2 layer 822 and neighboring side surfaces thereof is formed. As with the optical phase modulator 600 shown in FIG. 6, the strain induction film 824 is superposed on the p-type Si.sub.1-x2Ge.sub.x2 layer 822. As with the optical phase modulator 700 shown in FIG. 7, the strain induction film 824 is formed also in such a way as to neighbor the side surfaces of the p-type Si.sub.1-x2Ge.sub.x2 layer 822. The strain induction film 824 may extend from over a PN junction 806 to over a first electrode 812 and a second electrode 814. The strain induction film 824 may be formed in such a way as to neighbor side surfaces of a p-type Si.sub.1-x1Ge.sub.x1 layer 808. In the optical phase modulator 800 shown in FIG. 8, instead of superposing a thick SiGe layer, the strain induction film layer 824 is formed, thereby enabling large strain. The strain induction film 824 may include, for example, SiN or alumina. As in FIG. 5, in FIG. 8, an Si substrate 802, a BOX layer 804, a PN junction 806, a p-type Si region 830, an n-type Si region 832, a first electrode 812, a second electrode 814, a grounding electrode 816, a signal electrode 818, and a clad 820 are illustrated, and a position of a junction interface 806a of the PN junction 806 is offset in a direction of the p-type Si region 830 (that is, a minus direction) from a center 808a of the p-type Si.sub.1-x1Ge.sub.x1 layer 808. The PN junction 806 may be formed of Si.sub.1-yGe.sub.y. The rib-type waveguide structure 810 may include a PIN junction, instead of the PN junction. A conductive type of each of the Si.sub.1-x1Ge.sub.x1 layer 808 and the Si.sub.1-x2Ge.sub.x2 layer 822 may be an n-type.
(45) FIG. 9 is a cross-sectional view schematically illustrating an optical phase modulator 900 according to a still further embodiment of the present invention. A structure of the optical phase modulator 900 is similar to that of the optical phase modulator 100 shown in FIG. 1 except that a PN junction 906 includes a p-type Si.sub.1-yGe.sub.y layer 926 and an n-type Si.sub.1-yGe.sub.y layer 928. Note, however, that a structure of the layers included in the PN junction 906 is not limited to the above-mentioned configuration. In each of the p-type Si.sub.1-yGe.sub.y layer 926 and the n-type Si.sub.1-yGe.sub.y layer 928, a rate of Ge can be any value. In addition, the p-type Si.sub.1-xGe.sub.x layer 908 may include two or more p-type Si.sub.1-xGe.sub.x layers (p-type Si.sub.0.7Ge.sub.0.3 layer, p-type Si.sub.0.8Ge.sub.0.2 layer, and the like). As in FIG. 1, in FIG. 9, an Si substrate 902, a BOX layer 904, a p-type Si region 930, an n-type Si region 932, a first electrode 912, a second electrode 914, a grounding electrode 916, a signal electrode 918, and a clad 920 are illustrated, and a position of a junction interface 906a of the PN junction 906 is offset in a direction of the p-type Si region 930 (that is, a minus direction) from a center 908a of the p-type Si.sub.1-x1Ge.sub.x1 layer 908. The rib-type waveguide structure 910 may include a PIN junction, instead of the PN junction. A conductive type of the Si.sub.1-xGe.sub.x layer 908 may be an n-type. A rate of Ge composition in each of the p-type Si.sub.1-yGe.sub.y layer 926 and the n-type Si.sub.1-yGe.sub.y layer 928 may be smaller than a rate of Ge composition in the p-type Si.sub.1-xGe.sub.x layer 908. For example, the rate of the Ge composition in each of the p-type Si.sub.1-yGe.sub.y layer 926 and the n-type Si.sub.1-yGe.sub.y layer 928 may be 10% to 20%, and the rate of the Ge composition in the p-type Si.sub.1-xGe.sub.x layer 908 may be 30% to 50%. Also in the optical phase modulator 900 shown in FIG. 9, as in FIGS. 6 to 8, on the p-type Si.sub.1-xGe.sub.x layer 908 and/or on side surfaces thereof, a strain induction film may be formed. The PN junction 906 is constituted of a layered structure of a PN junction formed of Si and a PN junction formed of Si.sub.1-yGe.sub.y. In the present embodiment, since the PN junction 906 includes the SiGe layers, a refractive index of the PN junction 906 is increased, as compared with that of the optical phase modulator 100 shown in FIG. 1. Accordingly, according to the present embodiment, light confinement effect is further strengthened, thereby further increasing an optical modulation efficiency.
(46) FIG. 10 is a cross-sectional view schematically illustrating an optical phase modulator 1000 according to a yet further embodiment of the present invention. A structure of the optical phase modulator 1000 is similar to that of the optical phase modulator 900 shown in FIG. 9 except that a p-type Si.sub.1-x2Ge.sub.x2 layer 1022 is superposed on a p-type Si.sub.1-x1Ge.sub.x1 layer 1008. A structure of the layers included in a PN junction 1006 is not limited to a configuration shown in FIG. 10. In each of a p-type Si.sub.1-yGe.sub.y layer 1026 and an n-type Si.sub.1-yGe.sub.y layer 1028, a rate of Ge can be any value. As in FIG. 9, in FIG. 10, an Si substrate 1002, a BOX layer 1004, a p-type Si region 1030, an n-type Si region 1032, a first electrode 1012, a second electrode 1014, a grounding electrode 1016, a signal electrode 1018, and a clad 1020 are illustrated, and a position of a junction interface 1006a of the PN junction 1006 is offset in a direction of the p-type Si region 1030 (that is, a minus direction) from a center 1008a of the p-type Si.sub.1-x1Ge.sub.x1 layer 1008. The rib-type waveguide structure 1010 may include a PIN junction, instead of the PN junction. A conductive type of each of the Si.sub.1-x1Ge.sub.x1 layer 1008 and the Si.sub.1-x2Ge.sub.x2 layer 1022 may be an n-type. Also in the optical phase modulator 1000 shown in FIG. 10, as in FIGS. 6 to 8, on the p-type Si.sub.1-x2Ge.sub.x2 layer 1022 and/or on side surfaces thereof as well as/or on side surfaces of the p-type Si.sub.1-x1Ge.sub.x1 layer 1008, a strain induction film may be formed. In the present embodiment, since the PN junction 1006 includes the SiGe layers, a refractive index of the PN junction 1006 increases, as compared with a refractive index of the PN junction formed only of Si. Accordingly, according to the present embodiment, light confinement effect is further strengthened, thereby further increasing an optical modulation efficiency.
(47) One embodiment of the present invention is an optical intensity modulator which includes the optical phase modulator according to each of the embodiments of the present invention described above. For example, the optical phase modulator 100 shown in FIG. 1 is used at one arm or both arms of the Mach-Zehnder interferometer, thereby allowing the optical intensity modulator to be configured. The optical phase modulator according to each of the other embodiments of the present invention also can be used to configure the optical intensity modulator. The configuration of the optical intensity modulator according to each of the embodiments of the present invention is not limited to each of the above-described configurations. It is understood that by employing a method known by those skilled in the art, the optical phase modulator of the present invention is applied to the optical intensity modulator, thereby allowing the optical intensity modulator including features of the present invention to be obtained.
(48) In one embodiment of the present invention, the rate of Ge of the Si.sub.1-xGe.sub.x layer (108, 508, 608, 708, 808, 908, and 1008) may be set in a range of 0x<0.6. Since when x which is the rate of Ge is equal to or greater than 0.6, light absorption becomes remarkable, the rate of Ge is set in the range of 0x<0.6, thereby allowing an increase in an insertion loss of the optical phase modulator to be inhibited.
(49) In one embodiment of the present invention, crystal orientation of each of the silicon substrates (102, 502, 602, 702, 802, 902, and 1002) may be <110>. The crystal orientation is set to <110>, thereby increasing mobility of free carriers and decreasing the light absorption. Thus, the optical phase modulator whose speed is high and whose loss is low can be realized.
(50) With reference to FIG. 11A to FIG. 11H, processes of manufacturing the optical phase modulator 100, shown in FIG. 1, according to the embodiment of the present invention will be explained.
(51) FIG. 11A illustrates a configuration of the substrate used in the embodiment of the present invention. The substrate is constituted of an SOI substrate which includes the Si layer 105 superposed on the BOX layer (thermal oxide film) 104 superposed on the Si substrate 102.
(52) In FIG. 11B, one part of the Si layer 105 is doped (for example, by boron or the like), thereby forming the p-type Si region 130. In addition, another part of the Si layer 105 is doped (for example, by phosphorus or the like), thereby forming the n-type Si region 132. Thus, the PN junction 106 which is formed of Si in a lateral direction on the substrate is formed. The PN junction 106 may be formed of Si.sub.1-yGe.sub.y. Instead of the PN junction, the PIN junction may be formed.
(53) In FIG. 11C, one part of the p-type Si region 130, which neighbors the PN junction 106, is further p-type-doped to thereby increase a doping concentration thereof, thereby forming the p-type electrode 112. In addition, one part of the n-type Si region 132, which neighbors the PN junction 106, is further n-type-doped to thereby increase a doping concentration thereof, thereby forming the n-type electrode 114.
(54) In FIG. 11D, parts of the PN junction 106, the p-type electrode 112, and the n-type electrode 114 are etched.
(55) In FIG. 11E, an oxide-film mask layer 136 is formed on the etched parts. Furthermore, the oxide-film mask layer on the PN junction 106 is removed, and a recess 138 is formed.
(56) In FIG. 11F, the Si.sub.1-xGe.sub.x layer which is constituted of at least one layer is formed in the recess 138. This layer is p-type-doped, thereby forming the p-type Si.sub.1-xGe.sub.x layer 108. Accordingly, the p-type Si.sub.1-xGe.sub.x layer 108 is electrically connected to the PN junction 106.
(57) In FIG. 11G, the oxide-film mask layer 136 is removed, and the oxide film (for example, SiO.sub.2) clad layer 120 is formed.
(58) In FIG. 11H, the grounding electrode 116 and the signal electrode 118 are formed by use of metal wires in such a way as to contact the p-type electrode 112 and the n-type electrode 114, respectively.
(59) The present invention is described with reference to the specific embodiments. However, it is not intended that the embodiments described in this specification are used to restrictively interpret the present invention, and it is intended that the embodiments described herein are used to illustratively explain the present invention. It is apparent to those skilled in the art that other alternative embodiments can be implemented without departing from the scope of the present invention.
