JP6222734B2 - Variable frequency filter - Google Patents

Description

  The present invention relates to a frequency variable filter using a graphene structure.

  In the current communication system, a terahertz (THz) frequency region ranging from 100 GHz to 10 THz is expected to be applied to measurement and communication as a new frequency region. Prepare two waves of laser light that continuously oscillates in the infrared frequency range used in optical communications (CW, Continuous Wave), shift the frequencies slightly to about terahertz, mix them, and irradiate the photodiodes. A photocurrent component corresponding to the difference frequency terahertz component is generated from the diode. If a load resistor is loaded at the output end of the photodiode, a terahertz electrical signal can be obtained and radiated as a terahertz electromagnetic wave via the antenna. When one CW laser beam carries data by amplitude modulation, an output signal in which the data signal is carried in a terahertz wave corresponding to the difference frequency between the two laser beams is obtained as the output of the photodiode. be able to. That is, the data signal in the optical communication band can be down-converted (down-converted) to the terahertz carrier frequency band. A photomixer that generates terahertz electromagnetic waves from a laser light source for optical communication using this principle has been developed as a generation source of millimeter waves and terahertz waves.

  Until now, a single-running carrier photodiode (UTC-PD) has been studied as an optical signal processing device typified by this photomixer.

  Although UTC-PD uses only electrons with a small effective mass as carriers, its operating speed is fast, but it is still limited by the electron transit time effect and remains at about 10 microwatts at 1 THz output.

  In future ultra-high-speed information communication systems, fusion of optical communication and radio is being studied. In the fusion of optical communication and wireless communication, a data signal is independently modulated into a plurality of optical carrier signals having different wavelengths, and any one of optical division multiplexing (WDM) signals obtained by multiplexing the data signals. The function of filtering the signal of one wavelength component and down-converting it to the terahertz radio frequency band is important. This method includes filtering the optical WDM signal to extract one wavelength line segment, and then downconverting the optical signal component to the terahertz band, or lowering the optical WDM signal as it is to the terahertz radio frequency band. There is a method of filtering any one carrier frequency component of a data signal component that has been converted and multiplexed on a plurality of carrier frequencies converted into a terahertz frequency band and modulated. The former can be functionally realized by the above-described UTC-PD photomixing function, but is difficult to realize in the terahertz frequency band due to the frequency response limit of the above-described UTC-PD. On the other hand, in the latter case, it is important to realize a frequency variable filter function for a terahertz signal, but a filter device operating in the THz region has not been realized. Therefore, various devices to which graphene is applied as a signal processing device that covers this higher frequency region are being studied.

One of the related technologies is a light intensity modulator using a graphene electric double layer (capacitor) using interband light absorption of graphene (for example, see Non-Patent Document 1). The structure of the light intensity modulator in the related art is shown in FIGS. In FIG. 2, 31 is a first graphene layer, 32 is a second graphene layer, 33 is an insulating film, 34 is a first metal electrode, 35 is a second metal electrode, 36 is an optical waveguide, and 37 is a substrate. is there. As shown in FIG. 2, the first graphene layer 31, the second graphene layer 32, and the insulating film 33 constitute a double-layer graphene for the optical waveguide 36 made of Si. The insulating film 33 that insulates between the first graphene layer 31 and the second graphene layer 32 is made of Al ₂ O ₃ , and a capacitor is made of double layer graphene.

When a voltage is applied from the outside between the first graphene layer 31 and the second graphene layer 32, each graphene layer becomes p-type or n-type. At that time, _if ΔE _{f is} the amount of change on the Fermi surface and E _ph is the energy of the photons transmitted through the optical waveguide 36, no light absorption occurs in each graphene layer when ΔE _f > E _ph / 2, When ΔE _f ≦ E _ph / 2, light absorption occurs in each graphene layer. As described above, each graphene layer can function independently as a light absorption layer, and absorption / non-absorption can be controlled in accordance with a potential difference. As a result, a light absorption type light intensity modulator using graphene can be configured.

  In the second related technology, there is a graphene heterojunction structure in which a planar gate electrode is disposed on the outer surface of both graphenes of the same graphene double layer capacitor structure via separate thin insulating layers. (For example, refer nonpatent literature 2.). In FIG. 3, 11 is a first graphene layer, 12 is a second graphene layer, 13 is an insulating film, 14 is a first metal electrode, 15 is a second metal electrode, 17 is a first insulator layer, 18 is a second insulator layer, 22 is a first gate electrode, and 23 is a second gate electrode. As illustrated in FIG. 3, the graphene double layer capacitor includes a graphene double layer in which an insulating film 13 is sandwiched between a first graphene layer 11 and a second graphene layer 12. Since the resonance electron emission transition energy between the two graphene layers 12 and the spectrum of the emitted photon energy can be adjusted according to the bias applied to the first gate electrode 22, terahertz for absorption and modulation is performed. Since the frequency of the signal can be variably controlled, the terahertz signal can be used more effectively.

  4 and 5 are explanatory diagrams when a bias is applied to the graphene double layer capacitor according to the related art. In FIG. 4, when a bias is applied to a graphene double layer capacitor in which an insulating film 13 is sandwiched between a first graphene layer 11 and a second graphene layer 12 to form a pin junction, the tunnel barrier layer is sufficiently thin. However, since the momentum conservation of the start state and the end state of the tunnel is not established, the tunnel is blocked.

  In FIG. 5, a gate stack structure in which the first gate electrode 22 and the first insulator layer 17 are connected to the outside of the graphene double layer capacitor structure is added, and the first graphene layer 11 and the second graphene layer are added. A related technique in which the resonance of the tunnel is established by shifting the gate bias to apply the potential of the layer 12 to the first gate electrode 22 and canceling the band offset, thereby conserving the momentum of the initial state and the final state of the tunnel. (For example, refer nonpatent literature 3).

In the third related technology, there is a light intensity modulator in which a planar gate electrode is arranged on the outer surface of both graphenes of a similar graphene double layer capacitor structure via separate thin insulator layers (for example, (Refer nonpatent literature 4.). 6, the first graphene layer 11, the second graphene layer 12, 13 denotes an insulating film, the first metal electrode 14, 15 second metal electrodes, 1 7 the first insulator Reference numeral 18 denotes a second insulator layer, 22 denotes a first gate electrode, and 23 denotes a second gate electrode. The light intensity modulator shown in FIG. 6 has GL (Graphene Layer) conductivity composed of a first graphene layer 11 and a second graphene layer 12 between a double graphene layer resonant tunnel and a negative differential. . As shown in FIG. 6, the light intensity modulator can have a stable electron-hole plasma with respect to the plasma oscillation and the self-excited oscillation of the non-periodic perturbation in the graphene double layer capacitor structure. Excitation of plasma oscillation can result in a sharp resonance dependence of frequency response on the radiation detector and a bias voltage.

M.M. Liu et al., “Double-layer graphene optical modulator,” Nano Lett. , 12, 1482, (2012) V. Ryzhi et al., “Injection terahertz laser using the resonant inter-layer radial transitions in double-grafene-layer structure,” Appl. Phys. Lett. 103, 163507 (2013) L. Britnell et al. , “Resonant tunneling and negative differential conductance in graphene transistors,” Nature Comm. 4, 1794, 2013. V Ryzhi et al. “Dynamic effects in double graphene-layer structures with inter-layer resonant-tunneling conductivities”, J. et al. Phys. D: Appl. Phys. 46 (2013) 315107 (6pp)

  Here, in order to realize a future ultra-high-speed information communication system, fusion of optical communication and radio is indispensable. The terahertz (THz) frequency range from 100 GHz to 10 THz is expected to be applied to measurement and communication as a new frequency range. As a signal processing device that covers this wavelength region, various devices to which graphene is applied are being studied. Further, in future ultra-high-speed information communication systems, fusion of optical communication and radio is being studied, and any one carrier frequency component of a data signal component that is multiplexed and modulated on a plurality of carrier frequencies approaching the THz frequency band. It is important to realize a variable frequency filter function for filtering the frequency.

  For example, a transmission rate of 40 Gbit / s or more per video signal I channel is required for non-compressed transmission of next-generation ultra-high-definition (super high-definition) television video signals having two or more digits than the current high-definition television video. In order to transmit this with a simple amplitude shift keying (ASK) modulation scheme, a carrier frequency that is approximately 10 times higher than the transmission rate is required. For example, when a 40-Gbit / s ASK signal is modulated to a carrier frequency of 1 THz, and a large number of signals modulated independently at different carrier frequencies at intervals of about 100 GHz are multiplexed and transmitted, a passband of 80 GHz In addition, a “frequency variable filter” function capable of changing the frequency of the pass band while at the same time achieving a steep cutoff characteristic in which the transition frequency band from the pass band to the cutoff band is less than 20 GHz is required.

  It is important to realize a “frequency variable filter” function that filters any one carrier frequency component of a data signal component multiplexed and modulated on a plurality of carrier frequencies approaching the terahertz frequency band. However, a filter device that operates with such frequency selection performance in the THz region has not been proposed. In addition, the related art described above cannot sufficiently utilize the graphene layer effectively as an optical device used for optical communication.

  In order to solve the above-described problems, the present invention provides a thin insulator layer functioning as a tunnel barrier sandwiched between graphene, a source electrode is provided on one graphene layer, and a drain electrode is provided on the other graphene layer. In addition, the graphene heterojunction structure in which lattice-shaped gate electrodes are arranged on the outer surfaces of both graphenes via separate thin insulator layers, and multiplexed and modulated to a plurality of carrier frequencies approaching the THz frequency band. It is an object of the present invention to realize a frequency variable filter function for filtering any one carrier frequency component of the data signal component that has been generated.

  In the variable frequency filter constituted by the graphene heterojunction structure according to the present invention, a thin insulator layer functioning as a tunnel barrier is sandwiched between graphenes, a source electrode is provided in one graphene layer, and the other graphene layer is provided. Has a drain electrode, and further includes a grid-like gate electrode disposed on the outer surface of both graphenes via a separate thin insulating layer.

Specifically, the variable frequency filter according to the present invention includes a graphene double layer in which an insulating film is sandwiched between a first graphene layer and a second graphene layer,
A first metal electrode connected to one end of the first graphene layer;
A second metal electrode connected to one end of the second graphene layer;
A first insulator layer in surface contact with the first graphene layer ;
A second insulator layer in surface contact with the second graphene layer;
The first insulator layer is disposed on a surface not in surface contact with the first graphene layer, and incident light input from the outside to the first insulator layer is a first gate electrode of the insulator layer permeable to the lattice shape you form an exposed portion of the order to reach the first graphene layer,
The second being disposed on the surface of the second graphene layer surface contact with non side of the insulator layer, a second of the lattice shape you form an exposed portion to said first insulating layer A gate electrode;
Equipped with a,
By applying a bias between the first gate electrode and the second gate electrode, the energy difference between the first graphene layer and the second graphene layer is relatively shifted.
It shall be the features a.

In the frequency variable filter according to the present invention,
The graphene bilayer is
In the first graphene layer, incident light may pass through and reach the exposed portion of the first insulator layer to generate surface plasmons at the interface of the first graphene layer.

In the frequency variable filter according to the present invention,
The first gate electrode is
Receives electromagnetic waves as a receiving antenna,
The second gate electrode is
The electromagnetic wave may be selectively radiated through the graphene bilayer as a radiating antenna.

In the frequency variable filter according to the present invention,
The graphene bilayer is
It may have a resonance frequency of the surface plasmon set in advance so as to coincide with the resonance tunnel frequency of the reception signal in the first gate electrode.

In the frequency variable filter according to the present invention,
The second gate electrode is
If the resonant tunneling frequency of the received signal does not match the resonant frequency of the surface plasmon, the optical component of the received signal may be transmitted.

In the frequency variable filter according to the present invention,
The graphene bilayer is
A photovoltaic force may be generated for the first metal electrode and the second metal electrode by utilizing nonlinear rectification of the surface plasmon.

In the frequency variable filter according to the present invention,
The graphene bilayer is
The baseband signal may be extracted by separating the intensity-modulated data signal included in the frequency signal component of the received signal in which the resonant tunneling frequency of the received signal matches the resonant frequency of the surface plasmon.

In the frequency variable filter according to the present invention,
The graphene bilayer is
In the first graphene layer, the carrier concentration may be periodically modulated in a lattice shape in accordance with the bias amount applied to the first gate electrode.

  The above inventions can be combined as much as possible.

  According to the present invention, it is possible to realize a “frequency variable filter” function that filters any one carrier frequency component of a data signal component that is multiplexed and modulated on a plurality of carrier frequencies approaching the terahertz frequency band. Furthermore, it is possible to simultaneously realize the function of separating the intensity modulation data signal included in the frequency signal component extracted by the frequency variable filter function from the carrier frequency and extracting it as a baseband signal.

It is an example which shows the structure figure of the light intensity modulator which concerns on a 1st related technique. It is an example which shows the structure figure of the light intensity modulator which concerns on a 1st related technique. It is an example which shows the structure figure of the graphene double layer capacitor concerning the 2nd related technology. It is an example which shows the 1st explanatory view at the time of impressing a bias to a graphene double layer capacitor concerning related technology. It is an example which shows the 2nd explanatory view at the time of impressing a bias to a graphene double layer capacitor concerning related technology. It is an example which shows the structure figure of the light intensity modulator which concerns on a 3rd related technique. It is an example which shows the structure figure of the frequency variable filter which concerns on this embodiment. It is an example which shows explanatory drawing at the time of applying a bias to the frequency variable filter which concerns on this embodiment. It is an example which shows explanatory drawing at the time of applying a bias to the frequency variable filter which concerns on this embodiment. It is an example which shows the graph which shows the absorption efficiency of light energy.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In addition, this invention is not limited to embodiment shown below. These embodiments are merely examples, and the present invention can be implemented in various modifications and improvements based on the knowledge of those skilled in the art. In the present specification and drawings, the same reference numerals denote the same components.

FIG. 7 shows the structure of the frequency variable filter described in this embodiment. The first graphene layer 11, 12 is a second graphene layer, 13 denotes an insulating film, the first metal electrode 14, 15 is a second metal electrodes. The first graphene layer 11, the second graphene layer 12, and the insulating film 13 that insulates between the two form a graphene double layer. The first graphene layer 11 and the second graphene layer 12 are connected to the first metal electrode 14 and the second metal electrode 15, respectively. Reference numeral 24 denotes a lattice-shaped first gate electrode for forming an exposed portion in the first insulator layer 17 in surface contact with the first graphene layer 11. Reference numeral 25 denotes a lattice-shaped second gate electrode for forming an exposed portion in the second insulator layer 18 in surface contact with the second graphene layer 12.

In the present embodiment, in the frequency tunable filter shown in FIG. 7, when light of a certain energy enters, the first graphene layer 11 is passed through the lattice-shaped exposed portion provided in the first insulator layer 17. Resonant tunneling is assisted by absorption of the reached photons. Here, since the first gate electrode has a lattice shape, the first gate electrode is locally formed on the surface of the first graphene layer 11 by photons reaching from the exposed portion of the first insulator layer 17 exposed according to the lattice shape. By alternately exchanging existing electromagnetic waves and polarization, surface plasmons that are plasma surface waves can be excited on the surface of the graphene layer. 8 and 9 show the state of the frequency variable filter to which a bias is applied according to the present embodiment. FIG. 8 shows resonance tunneling generated by assisting photon absorption in this embodiment, and FIG. 9 shows resonance tunneling generated by assisting photon emission . Specifically, when a bias is applied between the first metal electrode 14 and the second metal electrode 15 of the variable frequency filter having a graphene double layer capacitor structure to form a pin junction, a tunnel barrier is formed. Even if the layer is thin enough, the tunnel is blocked because the momentum conservation of the initial and final states of the tunnel is not established. However, here, if a bias is applied to the lattice-shaped first gate electrode 24 and the lattice-shaped second gate electrode 25, as shown in FIGS. The energy difference between the two graphene layers 12: Δ can be relatively shifted according to the bias. As a result, as shown in FIG. 8, when a photon having an energy equal to the momentum conservation law is incident, the resonance tunnel is assisted by absorption of the photon of the incident light. In addition, as shown in FIG. 9, the resonant tunneling is also assisted by emitting and emitting photons of energy equal to the momentum conservation law.

The first gate electrode 24 and the second gate electrode 25 having a frequency variable filter according to the present embodiment is constructed in the form of a lattice. The first gate electrode 24 and the second gate electrode 25 are the conductors of the grid shape, it can function as an antenna element of the lattice shape. In a related technology having a simple planar antenna element widely used as a patch antenna, the patch antenna couples only an electromagnetic wave whose length of one side matches a half wavelength of the electromagnetic wave or an integral multiple thereof (reception or reception). It is a “narrowband antenna” that can radiate. For this reason, a “wideband” response that can realize the frequency variability intended by the present invention is not possible. On the other hand, in the variable frequency filter according to this embodiment, the lattice-shaped metal planar array structure functions as a broadband antenna element, and can efficiently combine electromagnetic waves and plasmons over a wide frequency band.

  Specifically, in the variable frequency filter according to the present embodiment, the periodicity of the resonance frequency of the excited plasma wave changes the dispersion relationship (relationship between frequency and wave number) of the plasma wave. The dispersion relationship of the plasma wave under gate control as in this embodiment is linear, and the plasma wave velocity is generally about two orders of magnitude lower than the light velocity (corresponding to the slope of the dispersion line). The dispersion line of the plasma and the dispersion line of the plasma wave do not intersect, and no energy exchange can occur between them. However, as in this embodiment, the carrier concentration in graphene is periodically modulated by the lattice-shaped metal planar array gate electrodes 24 and 25, and as a result, plasmon regions are periodically formed, and the lattice shape of the gate In accordance with the periodicity, a large number of plasma wave dispersion lines are periodically folded with the fundamental wave number as a period. When the dispersion line of the plasma wave is folded, the dispersion line of the light intersects densely, and at each intersection, the conservation law of energy and momentum is satisfied between the light wave and the plasma wave. Therefore, light waves and plasma waves are combined, and there are a large number of dense frequencies at which energy conversion can be performed between the two. Thus, the alternately arranged first gate electrode 24 and second gate electrode 25 having the same period function as a broadband antenna, and mode conversion between terahertz electromagnetic waves and plasma waves can be performed.

  With the above-described mechanism, in the variable frequency filter according to this embodiment, the first gate electrode 24 and the second gate electrode 25 convert the energy of the terahertz electromagnetic wave into a plasma wave and excite it with a periodic lattice-shaped structure. Can do. On the other hand, when the plasma wave is excited by some external factor, the energy of the plasma wave can be converted into a terahertz electromagnetic wave. The conversion from the terahertz electromagnetic wave to the plasma wave can be used as an antenna for receiving the terahertz electromagnetic wave, while the conversion from the plasma wave to the terahertz electromagnetic wave can be used as an antenna for emitting the terahertz electromagnetic wave.

  In this embodiment, in the variable frequency filter, the dependence of the tunnel probability on the incident photon energy has resonance characteristics depending on the thickness of the tunnel barrier layer and the doping concentrations of the first graphene layer 11 and the second graphene layer 12. Depending on the tunnel probability, the tunneling current at the resonant frequency, that is, the absorption rate of photons increases compared to the non-resonant frequency.

  The doping concentrations of the first graphene layer 11 and the second graphene layer 12 are given on average by the source-drain voltage V, and are further periodically modulated by the gate voltage Vg. The resonance absorption photon energy depending on the band offset can be controlled by the gate voltage Vg. Here, the lattice-shaped structure of the first gate electrode 24 and the second gate electrode 25 may function as a broadband antenna that performs coupling between the in-graphene two-dimensional plasmon and the incident electromagnetic wave.

  Since the structure of the first gate electrode 24 and the second gate electrode 25 has a lattice shape, the carrier concentration induced in the graphene plane by the gate bias Vg is periodically modulated in a stripe shape corresponding to the gate lattice. Is done. Along with the periodic modulation of the carrier concentration, a resonance mode is determined in the two-dimensional plasmon in the graphene plane (resonance frequency ∝ (carrier concentration) ¼, ∝ (reciprocal of stripe width)). By matching the plasmon resonance frequency to the resonance tunnel frequency, the tunnel current at the resonance frequency, that is, the absorption rate of photons, as shown in FIG. 10, increases by three orders of magnitude or more compared to the non-resonance frequency.

  In FIG. 10, the dependence of the photodetection sensitivity Rω on the incident photon energy when the source-drain voltage V is changed is plotted with a solid line assuming a structural dimension condition of d = 4 nm. At the same time, the characteristic of the condition of V = 87 mV is plotted with a broken line for different structural dimension conditions of d = 2 nm, and compared with the case of d = 4 nm. In any of the characteristics, the photodetection sensitivity reaches a peak (maximum) at the resonance frequency determined by the structure dimension and the carrier concentration, and an increase in sensitivity of 3 digits or more is recognized as compared with the detection sensitivity at the nonresonant frequency. (The characteristics of V = 149 mV and 87 mV are such that the photodetection sensitivity value in the non-resonant frequency range is small outside the vertical axis in the figure and cannot be confirmed in the figure. (Increased by more than 3 digits) in the calculation results).

  Non-resonant photons are transmitted as they are and are radiated again by the antenna function of the lattice gate on the back surface. As a result, it functions as a frequency variable stop band filter. Photovoltaic power can be excited between the first metal electrode 14 functioning as a drain electrode and the second metal electrode 15 functioning as a source electrode by the non-linear rectification action of graphene plasmon.

  The response speed of the photovoltaic power is about several tens of times the momentum relaxation time of graphene plasmon (from picoseconds to subpicoseconds at room temperature), so if the incident terahertz wave is the carrier frequency and the intensity modulation modulates the data signal In this case, only the data signal is extracted and output between the first metal electrode 14 functioning as the drain electrode and the second metal electrode 15 functioning as the source electrode.

  According to the present embodiment described above, it is possible to simultaneously realize a frequency variable filter function and a function of separating an intensity modulated data signal included in a frequency signal component extracted by a filter from a carrier frequency and extracting it as a baseband signal. .

  The present invention can be applied to the information communication industry.

11: first graphene layer 12: second graphene layer 13: insulating film 14: first metal electrode 15: second metal electrodes
1 7: first insulating layer 18: second insulator layer 22, 24: first gate electrode 23, 25: second gate electrode 31: first graphene layer 32: second graphene layer 33 : Insulating film 34: first metal electrode 35: second metal electrode 36: optical waveguide 37: substrate

Claims (8)

A graphene double layer in which an insulating film is sandwiched between a first graphene layer and a second graphene layer;
A first metal electrode connected to one end of the first graphene layer;
A second metal electrode connected to one end of the second graphene layer;
A first insulator layer in surface contact with the first graphene layer ;
A second insulator layer in surface contact with the second graphene layer;
The first insulator layer is disposed on a surface not in surface contact with the first graphene layer, and incident light input from the outside to the first insulator layer is a first gate electrode of the insulator layer permeable to the lattice shape you form an exposed portion of the order to reach the first graphene layer,
The second being disposed on the surface of the second graphene layer surface contact with non side of the insulator layer, the second of the second lattice shape form an exposed portion relative to the insulator layer A gate electrode;
Equipped with a,
An energy difference between the first graphene layer and the second graphene layer is relatively shifted by applying a bias between the first gate electrode and the second gate electrode. A variable frequency filter. The graphene bilayer is
2. The incident light in the first graphene layer passes through and reaches the exposed portion of the first insulator layer to generate surface plasmons at the interface of the first graphene layer. Frequency variable filter. The first gate electrode is
Receives electromagnetic waves as a receiving antenna,
The second gate electrode is
The frequency variable filter according to claim 1, wherein the electromagnetic wave is selectively radiated through the graphene double layer as a radiating antenna. The graphene bilayer is
4. The variable frequency filter according to claim 1, wherein the frequency resonance filter has a resonance frequency of the surface plasmon set in advance so as to coincide with a resonance tunnel frequency of a reception signal in the first gate electrode. 5. . The second gate electrode is
5. The variable frequency filter according to claim 4, wherein when a resonance tunnel frequency of the reception signal does not match a resonance frequency of the surface plasmon, the optical component of the reception signal is transmitted. The graphene bilayer is
The frequency tunable filter according to claim 4, wherein a photovoltaic force is generated for the first metal electrode and the second metal electrode using nonlinear rectification of the surface plasmon. The graphene bilayer is
5. The baseband signal is extracted by separating an intensity modulation data signal included in a frequency signal component of the reception signal in which a resonance tunnel frequency of the reception signal and a resonance frequency of the surface plasmon coincide with each other. The frequency variable filter described in 1. The graphene bilayer is
2. The variable frequency filter according to claim 1, wherein in the first graphene layer, the carrier concentration is periodically modulated in a lattice shape in accordance with a bias amount applied to the first gate electrode.

Images