JP2006343410A - Polariton waveguide - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a polariton waveguide which is more efficiently applicable to a communication waveband. <P>SOLUTION: An n-type semiconductor layer 102 composed of silicon with an n-type impurity of high concentration introduced therein with nearly 50 nm film thickness, a p-type semiconductor layer 103 composed of silicon with a p-type impurity of low concentration introduced therein with nearly 20 nm film thickness, and an insulating layer 104 composed of silicon oxide with nearly 30-50 nm film thickness are laminated on a substrate 101 composed of silicon in this order. Furthermore, an electrode layer 105 composed of a metal with nearly 1 μm width is formed on the insulating layer 104. When a bias voltage is applied to the electrode layer 105 in the positive direction, a band on an interface between the p-type semiconductor layer 103 and the insulating layer 104 is distorted, so as to make the bottom of the band equal to or lower than Fermi energy on the p-type semiconductor layer 103 side and to form a population inversion layer. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an optical waveguide with reduced propagation loss of light propagating on a metal surface.

  With the rapid development of information and communication, there is an increasing expectation for the realization of ultra-fine optical circuits for large-scale integration, and several studies for miniaturization are underway. Application of metal waveguides (polariton waveguides) has been proposed as a technique to innovate optical waveguide technology that confines light using the refractive index difference of a conventional dielectric (see Patent Document 1 and Non-Patent Document 1). ). This utilizes a wave propagating through the interface while being localized at the interface between the metal and the dielectric, and is different in principle from a conventional hollow waveguide using metal reflection. The wave propagating on the surface is a kind of electromagnetic wave and is called a surface plasmon polariton (SPP) wave.

  An electromagnetic wave including light is a wave having a cycle structure in which an electric field and a magnetic field are deduced from Maxwell's equations. An SPP wave is obtained by incorporating elementary excitation of a substance into this cycle structure. The object of incorporation is plasma vibration (plasmon). In a metal waveguide, a conduction electron system is a medium that bears plasmons.

  In general, incorporation of elementary excitation into the electromagnetic wave cycle is limited by various factors, but is possible at interfaces where symmetry is broken. Plasma vibration is a longitudinal wave that forms a dense distribution with respect to the traveling direction. In bulk metal, the polarization due to the vibration of electrons cancels the displacement current and becomes rotH = 0, and is directly incorporated into an electromagnetic wave that is a transverse wave ( Bonding does not occur. However, on the metal surface, due to the electric field of the TM wave existing on the free space side, an electron aggregation region is generated even by the movement of electrons in the vertical direction, and surface electron density waves (surface plasmons) are formed. At this time, the surface plasmon is combined with an electromagnetic field to form an electromagnetic cycle structure (see FIG. 7).

  In order for an SPP wave that propagates in a state where surface plasmons and electromagnetic waves in free space are combined to exist, surface localization of the electromagnetic waves is necessary. For this purpose, it is necessary that there is no outflow of energy from the interface in the vertical direction. Also, the electric and magnetic field components along the surface must be continuous. Considering these, the condition for surface localization can be derived.

FIG. 8 is an explanatory diagram showing the electromagnetic wave field component at the interface between the medium having a dielectric constant of ε _{1 and} the medium having a dielectric constant of ε ₂ divided into two eigenmodes of TM wave and TE wave. In the TM mode having the magnetic field H _x in the surface and the electric field components E _y and E _z along the vertical and traveling directions, “the dielectric constants of the two media have different signs” from the surface localization and continuity conditions. The condition is derived. For example, if ε ₁ > 0, ε ₂ <0 needs to be satisfied. At this time, in order for the energy traveling direction to be the z direction in the two media, the vertical component of the electric field needs to be reversed across the interface, but this is a condition permitted in the electromagnetic field. As described above, it is understood that in order for the SPP wave to be generated on the metal surface, the dielectric constant of the metal needs to be negative. Hereinafter, the conditions under which the dielectric constant of the metal is negative will be considered.

  In metals with low atomic binding properties, metal electrons can be considered as free electron gases in which the interaction between electrons and scattering by the crystal lattice can be ignored. Such an equation of motion of electrons considers the alternating electric field of light as an external force,

m (d ² / dt ² ) r + mγ (d / dt) r = −eE (1)

Can be shown. In equation (1), m is the electron mass, r is the electron position (vector), E is the electric field of the light wave, and γ is a parameter related to the electrical conductivity.

The electric field E of the light having the angular frequency γ includes the vibration term exp (−iωt), and the effective current density J is determined as J = −θN (local electron density N and the velocity of the surrounding electrons dr / dt J = −θN ( In view of the fact that it can be expressed as d / dt) r and that the polarization due to this current is P = iJ / ω, the relative permittivity ε _m of the electron system can be derived as shown in the following equation (2).

ε _m = 1− (ω _p / ω) ² × (1-i / ωτ) / {1 + 1 / (ωτ) ² } (2)

Here, the plasma frequency: ω _p = {Ne ² / (mε ₀ )} ^1/2 and the relaxation time: τ≡1 / γ are used.

  The interaction between the light wave and the free electron gas is characterized by the real part of the dielectric constant. With the plasma frequency as a boundary, the dielectric constant is on the high frequency (short wavelength) side, whereas it is negative on the low frequency (long wavelength) side. Therefore, except for interactions such as photoelectron emission, the electronic system behaves as a transparent medium with respect to light in a high frequency region, as in a normal dielectric. On the other hand, in the low frequency region where the relative permittivity is negative, the wave number becomes an imaginary number, and therefore normal electromagnetic waves cannot exist. At this time, the light to be incident is totally reflected on the surface.

  The metallic condition is that the dielectric constant of the metal is negative. It can be seen from equation (2) that the plasma frequency should be increased in order to enhance this metallicity. For this reason, in a metal waveguide, a metal having a high electron density and a small effective mass (m) is desired. FIG. 9 shows the wavelength dependence of the dielectric constant (real part) of various metal materials. In addition to precious metal materials such as Au, Ag, and Ta except for Cr, Al that exhibits strong metallic properties in a wide wavelength band can be applied to the metal waveguide. However, the conductivity of Al is likely to be greatly deteriorated due to surface oxidation, and this may be a factor that inhibits the propagation of SPP waves.

  Since the surface-localized SPP wave has been found to be a TM wave, a basic equation related to the electromagnetic field is derived from the above equation (1) (Maxell equation). Assuming a special solution that attenuates exponentially with respect to the SPP wave, if the propagation constant is β, the SPP wave can be described as follows.

E = E ₀ f (y) exp [i (βz−ωt)]; H = H ₀ f (y) exp [i (βz−ωt)] (3)
f (y) = exp (−S ₁ y); y ≧ 0, = exp (S ₂ y); y ≦ 0 (4)

  When each component of the electromagnetic field is put into the Maxell equation and solved together with the continuous condition, the propagation constant β and the attenuation parameters (S1, S2) can be determined as follows.

β = k ₀ {(ε ₁ ε _m ) / (ε ₁ + ε _m )} ^1/2 ; S ₁ = {β ² / (ε ₁ ko) ² } ^1/2 ; S ₂ = {β ₂ / (ε _m ko) ² } ^1/2 (5)

As described above, the negative permittivity of the medium is a condition for using the SPP wave. This condition also holds for semiconductors having free carriers in addition to ordinary metals. However, when a semiconductor is used as a waveguide material for SPP waves, the free carrier density in the semiconductor becomes a problem. The plasma frequency ω _p is determined as “ω _p = {Ne ² / (mε ₀ )} ^1/2 ” based on the free carrier density N, and the plasma frequency decreases as the carrier density decreases. This means that the wavelength region having a negative dielectric constant of the semiconductor is limited to the ultra-long wavelength region.

  However, by increasing the density of free carriers in the semiconductor, it is possible to generate SPP waves in the near-infrared to infrared region at the interface between the semiconductor and the dielectric, and to sound the generated SPP waves. It becomes. For this purpose, for example, the semiconductor may be doped with a different number of valence electrons to increase the electron number density to make a highly doped n-type semiconductor. For example, in the case of Si, doping with P results in an n-type semiconductor as shown in FIG.

When a metal is used as a material on the negative dielectric constant side, the electron number density of the metal is about 10 ²³ cm ⁻³ and the wavelength of plasma vibration is about 160 nm. In such short-wavelength plasma oscillation, the wavelength of light (SPP wave) that can be efficiently coupled and confined efficiently is in the ultraviolet region, and high efficiency is achieved at the near-infrared wavelength used in the communication wavelength band. Cannot be obtained. On the other hand, if a semiconductor is used as a material on the negative dielectric constant side as described above, the electron number density can be adjusted by the impurity concentration to be introduced, and light having a wavelength used in the communication wavelength band can be efficiently used. It is possible to efficiently generate (excite) the SPP wave by coupling and efficiently confine the generated SPP wave.

Japanese Patent No. 3391521 Katagiri, et al., “Plasmon Mode Metal Optical Waveguide for Ultrafine Optical Circuits”, Laser Research, Vol. 31, No. 4, pp. 249-256, April 2003.

However, if a semiconductor is doped with a high concentration of impurities in order to realize a high carrier density, propagation loss increases. That is, as shown in FIG. 11, the loss coefficient evaluated from the imaginary part of β in the equation (5) tends to be small in the near infrared region where the metallic property is strong. In particular, in the wavelength region used for optical fiber communication in the vicinity of 1.55 μm, a low loss of 100 cm ⁻¹ or less that can be practically used at the Au—SiO ₂ interface that is not affected by oxidation is obtained. It is expressed as a function of relaxation time like the imaginary part of the dielectric constant shown in the following equation (6), which is the source of the imaginary part of β that can be evaluated by: It has been shown to decay.

ε _m ⁽ⁱ⁾ = (ω _p / ω) ² × [ωτ / {1+ (ωτ) ² }] (6)

  Further, as shown in FIG. 12, the imaginary part of the dielectric constant tends to increase exponentially as the optical frequency becomes lower (as the wavelength becomes longer). This is because as the frequency decreases, the electrons follow the alternating electric field and travel a long distance to receive more scattering. Apparently, the SPP propagation loss decreases on the long wavelength side because the real part of the dielectric constant is negative and its absolute value increases, and the imaginary part of the dielectric constant itself, which is the essence of the loss, decreases. It is not caused by. As described above, a polariton waveguide using a semiconductor as a material on the negative dielectric constant side can be applied to a communication wavelength band, but has a problem that efficiency is not so high.

  The present invention has been made to solve the above-described problems, and an object thereof is to provide a polariton waveguide that can be more efficiently applied to a communication wavelength band.

  The polariton waveguide according to the present invention is formed in contact with the first electrode, the n-type semiconductor layer formed on the first electrode and doped with n-type impurities, and the n-type semiconductor layer. A p-type semiconductor layer doped with p-type impurities, an insulating layer formed on and in contact with the p-type semiconductor layer, and a second electrode formed on the insulating layer. The surface plasmon polariton induced on the surface of the insulating layer side of the type semiconductor layer is guided. When a voltage is applied so that the first electrode is negative and the second electrode is positive, an inversion distribution layer having an increased electron number density is formed in the vicinity of the insulating layer side of the p-type semiconductor layer. A buffer layer made of a semiconductor may be provided between the first electrode and the n-type semiconductor layer.

  Since the surface plasmon polariton is guided through the width in which the second electrode is formed, the second electrode may be formed together with a portion to be a waveguide. In the polariton waveguide, the n-type semiconductor layer is formed around the first electrode, the p-type semiconductor layer is formed to cover the periphery of the n-type semiconductor layer, and the insulating layer is the p-type semiconductor layer. The second electrode may be formed around the insulating layer to form a cylindrical waveguide. The insulating layer is formed around the second electrode, the p-type semiconductor layer is formed to cover the periphery of the insulating layer, and the n-type semiconductor layer is formed to cover the periphery of the p-type semiconductor layer. The first electrode may be formed around the n-type semiconductor layer to form a cylindrical waveguide.

  The polariton waveguide is in contact with another p-type semiconductor layer formed on the opposite side of the second electrode to the p-type semiconductor layer and on the other p-type semiconductor layer opposite to the second electrode. A third electrode formed on the other n-type semiconductor layer formed on the opposite side of the other p-type semiconductor layer, the p-type semiconductor layer, and the other p-type semiconductor. An opening region disposed in a region including a surface facing the layer and formed in the second electrode, and another semiconductor layer formed between the second electrode and another p-type semiconductor layer, The layer and the other semiconductor layer are integrally formed in communication with each other in the opening region, and the interval between the p-type semiconductor layer and the other p-type semiconductor layer is induced on the surface of the p-type semiconductor layer on the insulating layer side. A first surface plasmon polariton and a second surface plasmon polariton induced on the surface of another p-type semiconductor layer on the other semiconductor layer side; It may be formed in a range bound to.

  As described above, according to the present invention, an n-type semiconductor layer into which an n-type impurity has been introduced, a p-type semiconductor layer into which a p-type impurity has been formed formed in contact therewith, Since the inversion distribution layer having a high electron number density is formed in the vicinity of the insulating layer side of the p-type semiconductor layer by the insulating layer formed in contact with the semiconductor layer, it can be applied to the communication wavelength band more efficiently. An excellent effect that a polariton waveguide can be realized is obtained.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view showing a configuration example of a polariton waveguide according to an embodiment of the present invention. In the polariton waveguide shown in FIG. 1, for example, an n-type semiconductor layer 102 having a thickness of about 50 nm made of silicon into which n-type impurities are introduced at a high concentration on a substrate 101 made of silicon, and p-type impurities are made low. A p-type semiconductor layer 103 made of silicon introduced into the concentration and having a thickness of about 20 nm, and an insulating layer 104 made of silicon oxide and having a thickness of about 30 to 50 nm are stacked in this order.

In addition, in the polariton waveguide shown in FIG. 1, an electrode layer 105 made of metal and having a width of about 1 μm is formed on the insulating layer 104. In the n-type semiconductor layer 102, for example, phosphorus (P) is introduced as an impurity at about 10 ¹⁸ cm ⁻³ . The p-type semiconductor layer 103 is doped with, for example, about 10 ¹³ cm ^{−3 of} boron (B). Note that, for example, by introducing impurities into the substrate 101 to impart conductivity, the other electrode to be paired with the electrode layer 105 can be formed. Alternatively, a metal electrode (not shown) may be disposed on the back side of the substrate 101 to serve as the other electrode that forms a pair with the electrode layer 105. In this case, the substrate 101 becomes a semiconductor buffer layer. Incidentally, it goes without saying that not only silicon but also other semiconductor materials may be used.

FIG. 2 shows a band diagram in the layer thickness direction in the polariton waveguide shown in FIG. The junction between the n-type semiconductor layer 102 and the p-type semiconductor layer 103, the force of the chemical potential to be a Fermi energy E _F in the same acts, as shown in FIG. 2, greatly distorted bands at the junction interface Yes. In the n-type semiconductor layer 102, electrons emitted by doping are present in the conductor. On the other hand, in the p-type semiconductor layer 103, since the dopant absorbs electrons in the valence band, a small number of carriers (holes) exist only in the valence band.

  When the electrode layer 105 is disposed outside the p-type semiconductor layer 103 via the insulating layer 104 and a bias voltage is applied in the positive direction, the band at the interface between the p-type semiconductor layer 103 and the insulating layer 104 is distorted. On the p-type semiconductor layer 103 side, the bottom of the band becomes Fermi energy or less and the inversion distribution layer 201 is formed. Since the inversion partial layer 201 formed in this way is formed in a shallow part on the very surface of the p-type semiconductor layer 103 on the insulating layer 104 side, the accumulated electron number density is increased. For this reason, an SPP wave can be generated at the interface between the inversion distribution layer 201 and the insulating layer 104. Further, when the p-type semiconductor layer 103 is thinned, electrons accumulated in the n-type semiconductor layer 102 can be moved to the inversion distribution layer 201 by a tunnel.

  The SPP wave is localized at the interface, and is present in the region of several nanometers at most on the medium (semiconductor layer) side having a negative dielectric constant, and the SPP wave is present on the semiconductor side in the region of the p-type semiconductor layer 103. It is. Therefore, in the band configuration as shown in FIG. 2, the electron supply region (n-type semiconductor layer 102) subjected to high concentration doping and the number of electron scattering sources are reduced by low concentration doping, thereby increasing the relaxation time. It is possible to separate the SPP wave propagation region (p-type semiconductor layer 103). Therefore, according to the polariton waveguide having the configuration shown in FIG. 1, high carrier density and low loss (low resistance) can be achieved simultaneously.

  According to the polariton waveguide shown in FIG. 1 having the above-described configuration, the inversion distribution layer 201 is formed in a region where the electrode layer 105 is disposed. Therefore, in the polariton waveguide shown in FIG. 1, the SPP wave is guided to the region where the electrode layer 105 is disposed. In the polariton waveguide shown in FIG. 1, for example, when light is incident from one end, an SPP wave in which surface plasmon and incident light are coupled is generated on the surface of the p-type semiconductor layer 103 exposed at the end. (Induction) and the generated SPP wave is guided through the waveguide. At the other end, light is excited and output (oscillates) by surface plasmons generated by the guided SPP waves.

When the semiconductor is silicon, the impurity doping limit is about 10 ¹⁹ cm ^−3, which is about four orders of magnitude smaller than the metal electron density of 10 ²³ cm ⁻³ . However, since the inversion distribution layer 201 formed as described above confines electrons in a region of several nanometers from the surface, the electron number density is greatly increased in this region. In particular, the inversion distribution layer 201 has a triangular potential distribution in which the surface potential on the insulating layer 104 side is the lowest and the conduction band edge is pointed, so that electrons tend to concentrate on the surface. Yes.

Here, the electron number density (ρ) is expressed by “ρ = N / L ³ (7)”, and the surface electron number density (ρ _s ) is “ρ _s = N / L ² ”. And there is a relationship of ρ = Lρ _s . Assuming a surface electron number density of about 10 ¹⁵ cm ⁻² on the assumption that one electron is supplied from the dopant, the inversion distribution layer 201 has a layer thickness of about 1 nm. Considering this, the electron number density of the inversion distribution layer 201 is estimated to be 10 ²² cm ⁻² . Therefore, the plasma frequency on the surface of the semiconductor layer on which the inversion distribution layer 201 is formed only decreases to at most about 1/3 times the plasma frequency when metal is used. Since the plasma wavelength of Au, which is a typical metal, is about 160 nm, the plasma wavelength in the inversion distribution layer 201 is about 480 nm, and in the inversion distribution layer 201 on the surface of the semiconductor (p-type semiconductor layer 103), visible to near-infrared. It becomes possible to generate an SPP wave in a region and guide it.

  Next, another polariton waveguide in the embodiment of the present invention will be described. FIG. 3 is a cross-sectional view schematically showing a configuration example of another polariton waveguide in the embodiment of the present invention. The polariton waveguide shown in FIG. 3 includes a center electrode layer 301 disposed at the center, a semiconductor buffer layer 302 disposed around the center electrode layer 301, an n-type semiconductor layer 303 disposed around the center electrode layer 301, and disposed around the periphery. The p-type semiconductor layer 304 is composed of an insulating layer 305 disposed around the p-type semiconductor layer 304 and an outer peripheral electrode layer 306 disposed around the insulating layer 305. The center electrode layer 301 can be made of a metal wire having a diameter of about several μm to several tens of μm. In addition, the center electrode layer 301 may be composed of a thin wire made of a conductive material and having a diameter of about several tens of nanometers. For example, the center electrode layer 301 may be composed of a carbon nanotube.

The semiconductor buffer layer 302, the n-type semiconductor layer 303, and the p-type semiconductor layer 304 are made of, for example, silicon. In the n-type semiconductor layer 303, for example, phosphorus (P) is introduced as about 10 ¹⁸ cm ⁻³ as an impurity, and in the p-type semiconductor layer 304, for example, boron (B) is introduced at about 10 ¹³ cm ⁻³ . The insulating layer 305 may be an oxide such as silicon oxide or an organic resin such as polyimide. Each semiconductor layer of the semiconductor buffer layer 302, the n-type semiconductor layer 303, and the p-type semiconductor layer 304 has a thickness of about several tens of nanometers. On the other hand, in order to suppress the influence of the outer peripheral electrode layer 306, the insulating layer 305 is preferably as thick as several hundred nm to several μm. In addition, the outer peripheral electrode layer 306 should just be comprised from the metal, for example.

  Also in the polariton waveguide shown in FIG. 3, a bias voltage is applied in the positive direction between the central electrode layer 301 and the outer peripheral electrode layer 306 as in the polariton waveguide shown in FIG. Thus, an inversion distribution layer is formed on the surface of the p-type semiconductor layer 304 on the insulating layer 305 side, and a surface plasmon having a plasma wavelength corresponding to the electron number density of the inversion distribution layer is formed on the surface of the p-type semiconductor layer 304. Is induced. Therefore, in the polariton waveguide shown in FIG. 3, the SPP wave is guided on the surface of the p-type semiconductor layer 304 on the insulating layer 305 side. In the polariton waveguide shown in FIG. 3, for example, when light enters from one end, an SPP wave in which surface plasmon and light are coupled is generated at the interface (surface) of the p-type semiconductor layer 304 exposed at the end. (Induction) and the generated SPP wave is guided through the waveguide. The mode shape of the SPP wave is cylindrical. According to the polariton waveguide shown in FIG. 3, the diameter can be significantly reduced to 1/10 or less than the diameter (125 μm) of the conventional silica-based optical fiber.

  Next, another polariton waveguide in the embodiment of the present invention will be described. FIG. 4 is a cross-sectional view schematically showing a configuration example of another polariton waveguide in the embodiment of the present invention. The polariton waveguide shown in FIG. 4 includes a center electrode layer 401 disposed in the center, an insulating layer 402 disposed around the center electrode layer 401, a p-type semiconductor layer 403 disposed around the center electrode layer 401, and a periphery thereof. The n-type semiconductor layer 404 includes a semiconductor buffer layer 405 disposed around the n-type semiconductor layer 404 and an outer peripheral electrode layer 406 disposed around the semiconductor buffer layer 405. The center electrode layer 401 can be composed of a metal wire having a diameter of about several μm to several tens of μm. Similarly, the outer peripheral electrode layer 406 should just be comprised, for example from the metal.

The insulating layer 402 may be an oxide such as silicon oxide or an organic resin such as polyimide. Further, the p-type semiconductor layer 403, the n-type semiconductor layer 404, and the buffer layer 405 are made of, for example, silicon. In the p-type semiconductor layer 403, for example, boron (B) is introduced at about 10 ¹³ cm ⁻³ , and in the n-type semiconductor layer 404, for example, phosphorus (P) is introduced as about 10 ¹⁸ cm ⁻³ as an impurity. Note that each of the p-type semiconductor layer 403, the n-type semiconductor layer 404, and the buffer layer 405 may have a thickness of about several tens of nanometers.

  Also in the polariton waveguide shown in FIG. 4, a bias voltage is applied in the positive direction between the outer peripheral electrode layer 406 and the center electrode layer 401 (the center electrode) as in the polariton waveguide shown in FIGS. The layer 401 is positive), whereby an inversion distribution layer is formed on the surface of the p-type semiconductor layer 403 on the insulating layer 402 side, and the surface of the p-type semiconductor layer 403 has a plasma wavelength corresponding to the electron number density of the inversion distribution layer. Surface plasmons are induced. Therefore, in the polariton waveguide shown in FIG. 4, the SPP wave is guided on the surface of the p-type semiconductor layer 403 on the insulating layer 402 side. In the polariton waveguide shown in FIG. 4, for example, light incident from one end generates an SPP wave in which surface plasmon and light are coupled at the interface (surface) of the p-type semiconductor layer 403 exposed at the end. (Induction) and the generated SPP wave is guided through the waveguide. The polariton waveguide shown in FIGS. 3 and 4 does not have to be a perfect circle in cross section, and may be, for example, an elliptical shape.

  Next, another polariton waveguide in the embodiment of the present invention will be described. FIG. 5 is a cross-sectional view schematically showing a configuration example of another polariton waveguide in the embodiment of the present invention. The polariton waveguide shown in FIG. 5 includes a lower electrode layer 502 made of metal on a substrate 501 made of silicon, for example. Further, on the lower electrode layer 502, a lower n-type semiconductor layer 503, a lower p-type semiconductor layer 504, an insulating layer 505, an upper p-type semiconductor layer 506, an upper n-type semiconductor layer 507, and an upper electrode 508 are arranged in this order. A waveguide structure that is stacked and extends in the normal direction (waveguide direction) of the drawing sheet of FIG. 5 is formed. In addition, a grid electrode 509 having an opening region at the center is provided at an intermediate portion of the insulating layer 505 in the layer thickness direction. The opening region of the grid electrode 509 extends and opens in the waveguide direction of the waveguide structure.

The lower n-type semiconductor layer 503 and the upper n-type semiconductor layer 507 are made of silicon into which an n-type impurity (for example, P) is introduced at a high concentration (for example, about 50 ¹⁸ cm ⁻³ ) and has a thickness of about 50 nm. Is formed. The lower p-type semiconductor layer 504 and the upper p-type semiconductor layer 506 are made of silicon into which p-type impurities are introduced at a low concentration, and are formed to a thickness of about 20 nm. The insulating layer 505 is made of silicon oxide and has a thickness of about 50 nm. Note that the lower n-type semiconductor layer 503, the lower p-type semiconductor layer 504, the insulating layer 505, the upper p-type semiconductor layer 506, the upper n-type semiconductor layer 507, and the upper electrode 508 are formed on the lower electrode layer 502 in a ridge shape. It is not necessary to be formed in the structure. In the waveguide to guide the SSP wave, the opening region of the grid electrode 509 may be formed in the waveguide direction.

  In the polariton waveguide shown in FIG. 5, by applying a positive voltage to the grid electrode 509 and a negative voltage to the lower electrode layer 502 and the upper electrode 508, the insulating layer 505 of the lower p-type semiconductor layer 504 and the upper p-type semiconductor layer 506. An inversion distribution layer is formed from the surface on the side. As a result, SPP waves can be generated on the surface of the lower p-type semiconductor layer 504 on the insulating layer 505 side and on the surface of the upper p-type semiconductor layer 506 on the insulating layer 505 side, respectively.

  Further, in the polariton waveguide shown in FIG. 5, the insulating layer 505 is thin, and both SPP waves are coupled in the same phase. Note that an opening region is provided at the center of the grid electrode 509 so that an electric field can pass therethrough, and both of the SPP waves described above can be coupled. In the polariton waveguide shown in FIG. 5 configured as described above, the distribution of the electromagnetic field in the cross-sectional direction of the guided SPP wave (polariton) has an insulating layer 505 of the lower p-type semiconductor layer 504 as shown in FIG. Both the surface on the side and the surface on the insulating layer 505 side of the upper p-type semiconductor layer 506 have maximum values.

  Here, it is known that there is a coupling mode in which both SPP waves always exist regardless of the distance between the lower electrode layer 502 and the upper electrode 508. Some of these coupling modes are in a state where the phases of both are coupled with a 180 ° shift. In such a case, if the distance between the lower p-type semiconductor layer 504 and the upper p-type semiconductor layer 506 is too small, The SPP wave disappears. On the other hand, in the mode coupled in the same phase as described above, since the SSP wave can exist even in a narrow region below the vacuum wavelength, the polariton waveguide shown in FIG. 5 can have a very small size. It is useful in constructing a micro optical circuit. Note that the size (width) of the opening region may be set as appropriate in accordance with the distance between the lower p-type semiconductor layer 504 and the upper p-type semiconductor layer 506.

  In the opening region of the grid electrode 509, a mesh (mesh) having an interval matching the wavelength of the SSP wave to be guided may be formed. In other words, a plurality of bridging portions spanning the opening region may be formed at intervals matching this wavelength in the direction in which the SSP wave is guided. In other words, a plurality of openings may be formed in the grid electrode 509 at intervals matching this wavelength in the direction in which the SSP wave is guided.

It is a perspective view which shows the structural example of the polariton waveguide in embodiment of this invention. FIG. 2 is a band diagram showing a band diagram in a layer thickness direction in the polariton waveguide shown in FIG. 1. It is sectional drawing which shows typically the structural example of the other polariton waveguide in embodiment of this invention. It is sectional drawing which shows typically the structural example of the other polariton waveguide in embodiment of this invention. It is sectional drawing which shows typically the structural example of the other polariton waveguide in embodiment of this invention. FIG. 6 is a distribution diagram showing a distribution of an electromagnetic field in a cross-sectional direction of an SPP wave (polariton) guided through the polariton waveguide shown in FIG. 5. It is explanatory drawing which shows the state in which the surface plasmon couple | bonds with the electromagnetic field and the cycle structure of electromagnetic waves is formed. It is explanatory drawing which divided and showed the electromagnetic wave field component in the interface of the medium of permittivity (epsilon) ₁ , and the medium of permittivity (epsilon) ₂ in two eigenmodes, TM wave and TE wave. It is a characteristic view which shows the wavelength dependence of the dielectric constant (real part) of various metal materials. It is explanatory drawing for demonstrating the state of Si doped with P and made into n-type. It is a characteristic view for demonstrating the loss coefficient evaluated from the imaginary part of (beta) in (5) Formula in each metal waveguide. It is explanatory drawing which shows the state which the imaginary part of a dielectric constant increases exponentially as an optical frequency becomes low (a wavelength becomes long).

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 ... Substrate, 102 ... N-type semiconductor layer, 103 ... P-type semiconductor layer, 104 ... Insulating layer, 105 ... Electrode layer, 201 ... Inversion distribution layer.

Claims (6)

A first electrode;
An n-type semiconductor layer formed on the first electrode and doped with an n-type impurity;
A p-type semiconductor layer doped with a p-type impurity formed on and in contact with the n-type semiconductor layer;
An insulating layer formed on and in contact with the p-type semiconductor layer;
And at least a second electrode formed on the insulating layer,
A polariton waveguide characterized in that surface plasmon polariton induced on the surface of the p-type semiconductor layer on the insulating layer side is guided. The polariton waveguide according to claim 1, wherein
A polariton waveguide comprising a buffer layer made of a semiconductor and disposed between the first electrode and the n-type semiconductor layer. The polariton waveguide according to claim 1 or 2,
The surface plasmon polariton is guided through a width in which the second electrode is formed. A polariton waveguide. The polariton waveguide according to claim 1 or 2,
The n-type semiconductor layer is formed around the first electrode,
The p-type semiconductor layer is formed to cover the periphery of the n-type semiconductor layer,
The insulating layer is formed so as to cover the periphery of the p-type semiconductor layer,
The polariton waveguide, wherein the second electrode is formed around the insulating layer. The polariton waveguide according to claim 1 or 2,
The insulating layer is formed around the second electrode;
The p-type semiconductor layer is formed to cover the periphery of the insulating layer,
The n-type semiconductor layer is formed to cover the periphery of the p-type semiconductor layer,
The polariton waveguide, wherein the first electrode is formed around the n-type semiconductor layer. The polariton waveguide according to claim 1 or 2,
Another p-type semiconductor layer formed on the opposite side of the second electrode from the p-type semiconductor layer;
Another n-type semiconductor layer formed on and in contact with the other p-type semiconductor layer opposite to the second electrode;
A third electrode formed on the other n-type semiconductor layer opposite to the other p-type semiconductor layer;
An opening region formed in the second electrode and disposed in a region including a facing surface of the p-type semiconductor layer and the other p-type semiconductor layer;
Another semiconductor layer formed between the second electrode and the other p-type semiconductor layer,
The semiconductor layer and the other semiconductor layer are integrally formed in communication with each other in the opening region,
The distance between the p-type semiconductor layer and the other p-type semiconductor layer is the first surface plasmon polariton induced on the surface of the p-type semiconductor layer on the insulating layer side and the other of the other p-type semiconductor layers. A polariton waveguide characterized by being formed in a range in which the second surface plasmon polariton induced on the surface on the semiconductor layer side of the substrate is coupled.

Images