JP4107354B2 - Millimeter-wave / far-infrared photodetector - Google Patents

Description

  The present invention relates to a millimeter-wave / far-infrared light detector for use in a millimeter-wave / far-infrared light measuring instrument, and in particular, for controlling a semiconductor quantum dot to detect a video signal in the millimeter-wave / far-infrared region. Is.

  In general, electromagnetic wave detectors include a frequency mixer (mixer) that performs phase-sensitive detection and a video signal detector that performs incoherent detection. However, the latter video signal detector is used to detect weak light. The sensitivity is better.

  Among the conventional video detectors in the millimeter-wave / far-infrared region, the most sensitive is the germanium composite bolometer used at an extremely low temperature of 0.3 K or less with a wavelength range of 0.1 mm to 1 mm. The photoconductivity detector by doped germanium used at a low temperature of about 2K having a wavelength range of 0.06 mm to 0.1 mm.

The noise equivalent output (Noise Equivalent Power, hereinafter referred to as “NEP”) reaches from 10 ⁻¹⁶ WHz ^−1/2 to 10 ⁻¹⁸ WHz ^−1/2 .
When viewed in terms of energy quanta of electromagnetic waves, that is, photons, this sensitivity cannot be detected as a signal over noise unless a million photon flux is incident on the detector, considering a measurement time of 1 second. Means.

  Furthermore, such a detector has a very slow response speed of about 100 milliseconds. A superconducting bolometer, a superconducting tunnel junction, a hot electron in a semiconductor (InSb), or the like is used as a detector having a high response speed, but the sensitivity is inferior to that of a germanium composite bolometer.

In addition to the above-described detector, it is known that when a normal single-electron transistor is irradiated with microwaves, a signal due to photon assisted tunneling is obtained. Since only one electron moves between the electrodes by absorption, the sensitivity as a detector is low.
JP-A-11-004017 Japanese Patent Laid-Open No. 08-274298 RJ Scoelkopf et al., "A Concept for a Submillimeter-Wave Single-Photon Counter", IEEE Transactions on Applied Superconductivity, Vol.9, No.2, June 1999, pp.2935-2939

  As described above, none of the conventional detectors has excellent sensitivity and response speed. The reason is that all detectors have a short lifetime in the state excited by electromagnetic waves because conduction electrons are in a continuous energy level band structure, and the change in average conductivity due to all electrons in the detector. Because the effect of excitation of a small number of electrons is overwhelmingly diminished by a large number of other electrons, the efficiency is low, and one electron is absorbed by the absorption of a single electromagnetic photon like the photon-assisted tunneling effect. However, it was because it moved between the electrodes.

  Therefore, the present invention fundamentally eliminates the problems to be solved related to the conventional detector based on a mechanism that is completely different from all the above-described mechanisms, so that the sensitivity is extremely superior and the response speed is high. It aims at providing a millimeter-wave / far-infrared photodetector.

  In order to achieve the above object, the millimeter-wave / far-infrared light detector of the present invention causes an electromagnetic wave coupling means for concentrating electromagnetic waves in a submicron-sized minute space region, and ionizes by absorbing the concentrated electromagnetic waves. A single electron transistor including a first quantum dot and a second quantum dot electrostatically coupled to the first quantum dot, the single electron transistor being a single electron forming a two-dimensional electron system A first quantum dot having a heterostructure, a gate electrode for controlling an electrostatic potential of a quantum dot tunnel-coupled in the two-dimensional electron system, and a source electrode and a drain electrode tunnel-coupled to the quantum dot. Electromagnetic waves are detected based on the change in the electrical conductivity of the single-electron transistor due to the change in the electrostatic state of the second quantum dot accompanying the ionization of.

In the above-described configuration, the ionization of the first quantum dot is caused by exciting electrons in a quantization bound state of the first quantum dot to a free electronic state of an electron system outside the first quantum dot. To do.
The ionization energy of the first quantum dot can be controlled by the magnitude of the bias voltage applied to the gate of the first quantum dot.
The first quantum dots and the second quantum dots are formed on the same semiconductor structure substrate and are electrostatically separated by a bias voltage applied to each gate.
The first quantum dot and the second quantum dot are formed adjacent to each other through a gap in a semiconductor.
The second quantum dot is a metal dot formed on the first quantum dot, and a single electron transistor is formed by tunnel junction with a metal lead wire formed on the metal dot.
The second quantum dot is an aluminum metal dot, and the tunnel junction is made of aluminum oxide.
The electromagnetic wave coupling means is an antenna that electrically couples the first quantum dot and the electromagnetic wave.
The electromagnetic wave coupling means also serves as a gate for applying a bias voltage for forming the first quantum dot and the second quantum dot.
The length direction of the lead portion of the electromagnetic wave coupling means is formed perpendicular to the polarization axis direction of the electromagnetic wave coupling means.
The size of the nodal point of the electromagnetic wave coupling means and the maximum size of the quantum dot are approximately the same.
The electrode passing length of the electromagnetic wave coupling means is about ½ of the wavelength of the electromagnetic wave.
The distance between the source electrode and the drain electrode of the single electron transistor is not less than the length in the polarization axis direction of the electromagnetic wave coupling means.
The single electron transistor is a compound semiconductor. The single electron transistor is a group III-V compound semiconductor.
The single electron transistor is characterized by having a selectively doped single heterostructure of a III-V compound semiconductor superlattice. The single-electron transistor has an aluminum gallium arsenide / gallium arsenide selectively doped single heterostructure. The single electron transistor is a group IV semiconductor.
A single electron transistor is formed symmetrically with respect to the quantum dot.
In addition to the above-described configuration, a light introducing portion for guiding the electromagnetic wave to the electromagnetic wave coupling means is provided.

  In the above configuration, the first quantum dot that absorbs the electromagnetic wave and the conductive second quantum dot that detects the first quantum dot are separated, so that the holes and electrons excited by the electromagnetic wave absorption are the first and second, respectively. Since the quantum dots are generated separately inside and outside the quantum dots, an extremely long excited state, that is, an ionization lifetime can be realized without applying a magnetic field. Therefore, the sensitivity can be further improved without applying a magnetic field, and single photon detection can be easily performed.

  Furthermore, since the electron system forming the first quantum dot uses excitation from the discrete level to the continuous band level, it has a good detection sensitivity in a continuous wavelength band having energy equal to or higher than the threshold, that is, the ionization energy. . In addition, the threshold wavelength of detection, that is, ionization energy, can be directly controlled through adjustment of the potential barrier height by the gate voltage. Furthermore, the operating temperature can be increased up to 2K by making the second quantum dots minute.

  The millimeter-wave / far-infrared light detector of the present invention has an effect that the sensitivity is orders of magnitude higher than that of the conventional millimeter-wave / far-infrared light detector and operates at a high speed.

The millimeter-wave / far-infrared photodetector of the present invention uses a single electron transistor (hereinafter referred to as “SET”) using semiconductor quantum dots. SET, for example, has a single heterostructure of a semiconductor superlattice that forms a two-dimensional electron gas, and is a very small isolated conductive region dot that is weakly tunnel-coupled to the source and drain regions by the source and drain electrodes. And a control gate electrode for controlling the electrostatic potential of the dot.
The SET may have a compound semiconductor, particularly a III-V group compound semiconductor, a III-V group compound semiconductor superlattice selectively doped single heterostructure. Furthermore, in the case of a millimeter wave / far infrared photodetector having a plurality of quantum dots, a group IV semiconductor may be used.

When the bias voltage of the control gate electrode is changed, the electrochemical potential of conduction electrons in the dot changes, and the source / drain current _ISD flows only under the condition that it becomes equal to the Fermi energy of the source and drain electrodes.
The conductivity of SET in this conducting state, G = I _SD / V _SD is approximately [200 to 400 kΩ] ⁻¹ .
However, V _SD is a source-drain voltage and must be 100 μV or less in the SET used in the present invention.

When a semiconductor quantum dot having an effective size of 0.02 μm to 0.6 μm is used as the conductive dot, the energy level of the internal electron system is quantized according to the size effect or the magnetic field applied from the outside. The energy interval corresponds to the photon in the millimeter wave / far infrared region. The energy interval can be controlled by changing the size of the quantum dots, applying an external magnetic field, or applying a bias voltage.
Therefore, electrons can be resonantly excited inside the quantum dot by irradiation with millimeter wave / far infrared light.
However, the excited state differs depending on the excitation method and whether or not a magnetic field is applied, as will be described later.

In either case, the wave function of the excited electron changes in its spatial symmetry and spatial distribution compared to that in the ground state, so the quantum potential of the quantum dot and the tunnel coupling strength with the source / drain region It changes a lot.
As a result, only by the excitation of one electron in the semiconductor quantum dot, the SET conductivity is greatly changed to about 20% to 99%, and the state in which the conductivity is changed becomes the ground state because the excited state disappears. Until the return to, i.e., the lifetime of the excited state and its relaxed state.

On the other hand, the state of the excited quantum dot is a highly sensitive detector because the lifetime until returning to the ground state is as long as 10 to 1000 seconds due to the discrete level structure of the energy. The change in the number of electrons sent from the source electrode to the drain electrode, N = GV _SD T (X / 100) / e, when the conductivity G changes by X% and lasts for T seconds is a typical condition. Below, G = 1/300 kΩ, X = 50%, T = 1 msec, V _SD = 0.05 mV, 10 ^6, and so many.
That is, about one million or more electrons can be transferred between the electrodes by absorption of one photon.

Furthermore, the fundamental time constant of the SET operation, C _SD / G, is as short as several tens of p seconds. Here, C _SD is a source-drain capacitance.
Therefore, a single millimeter wave / far infrared photon can be detected through fast time-resolved measurement of current.

Hereinafter, a millimeter wave / far infrared photodetector according to the present invention will be described in detail based on a specific preferred embodiment with reference to the drawings.
First, the configuration of the millimeter wave / far infrared photodetector of the present invention will be described.
FIG. 1 is a schematic cross-sectional view of a millimeter wave / far infrared photodetector according to the present invention including a condensing system.

  As shown in FIG. 1, the millimeter-wave / far-infrared light detector according to the present invention includes a millimeter-wave / far-infrared light introducing unit 1 for guiding incident millimeter-wave / far-infrared light to an antenna of the detector. And a semiconductor substrate 4 on which a single electron transistor for controlling a current passing through the semiconductor quantum dot is formed, and a semiconductor quantum dot formed in the single electron transistor, which is a sub-micron-sized microspace region, A bowtie antenna 6 for concentrating infrared light is provided, and the semiconductor substrate 4 is mounted on a package 7 for an IC chip.

The millimeter wave / far infrared light introducing section 1 includes a light introducing tube 3 that guides the millimeter wave / far infrared light 2, a dielectric lens 5 that collects the millimeter wave / far infrared light 2, and condensing light. And an auxiliary dielectric objective lens 9.
A silicon hemisphere lens is used as the dielectric objective lens 9, and this dielectric objective lens 9 leaves a gap of about 10 μm with respect to the bow tie antenna 6 and the semiconductor quantum dot surface described later so as not to directly touch them. Is fixed. In FIG. 1, reference numeral 8 denotes a back surface gate electrode on the back surface side of the single electron transistor formed on the semiconductor substrate 4.

  The millimeter-wave / far-infrared photodetector 10 including the light introducing section 1 shown in FIG. 1 is cooled to 0.3 K or less, and generates a magnetic field B perpendicular to the semiconductor substrate 4 (quantum dots) as necessary. Apply.

2A and 2B are diagrams showing a millimeter-wave / far-infrared photodetector according to the present invention. FIG. 2A is a plan view of a single-electron transistor using a bowtie antenna and quantum dots, and FIG. FIG.
As shown in FIG. 2 (a), the millimeter-wave / far-infrared photodetector 10 of the present invention includes a bowtie antenna 6, a semiconductor quantum dot 12, and a single electron transistor 14 including the semiconductor quantum dot. The single-electron transistor 14 is integrally formed on the semiconductor substrate 4 so that a source / drain current flows through the ohmic electrodes 16 and 17 under a predetermined condition.
As described above, a metal thin film is deposited on the back surface of the semiconductor substrate 4 to form a back gate electrode.

The single electron transistor 14 has a structure in which a modulation-doped GaAs / Al _0.3 Ga _0.7 As single heterostructure is stacked on a semi-insulating GaAs single crystal semiconductor substrate 4 as shown in FIG. The mesa structure of the single electron transistor 14 shown in (a) is formed using a lithography technique.

The GaAs / Al _0.3 Ga _0.7 As a single heterostructure crystal, two-dimensional electron mobility at 4.2K is 60 m ² / Vs or more, the electron concentration those from 2x10 ¹⁵ / m ² of 4.5 × 10 ¹⁵ / m ² Is used.

As shown in the cross-sectional view of FIG. 2B, the heterostructure includes a GaAs layer 22 having a thickness of about 10 nm doped with Si of about 10 ^{18 / cm 3} from the crystal surface, and a thickness of about 1 × 10 ¹⁸ / cm ³ doped with Si. An Al _0.3 Ga _0.7 As layer 24 having a thickness of about 70 nm, a pure Al _0.3 Ga _0.7 As layer 26 which is a spacer layer having a thickness of 20 nm or more, and a GaAs layer 28 having a non-doped thickness of about 100 nm. 4 is selectively doped by molecular beam epitaxy or the like and laminated.
2B shows that a two-dimensional electron system is formed, and the thickness is about 10 nm.
The semiconductor substrate 4 is a standard semi-insulating GaAs single crystal and has an overall thickness of about 0.5 mm and a planar size of about 1 mm to 3 mm square.

Each component of the millimeter wave / far infrared photodetector according to the present invention will be described in more detail.
As shown in FIG. 2A, the single electron transistor 14 including the semiconductor quantum dots 12 has a two-dimensional elongate mesa structure formed on a GaAs semiconductor substrate 4 and is near the center of the mesa structure. However, in order to prevent excessive absorption of millimeter-wave / far-infrared light by the external two-dimensional electron system of the semiconductor quantum dots 12 (details will be described later), the semiconductor quantum dots 12 are formed thinly to a width of about 4 μm over a length of about 200 μm. Yes. That is, the central part where the quantum dots are formed is thinner than the mesa structure parts at both ends.
The single-electron transistor using quantum dots is preferably formed symmetrically with respect to the quantum dot formed at the center.

  At both ends of the mesa structure, a standard ohmic electrode source electrode 16 and a drain electrode 17 formed by alloying with Au / Ge are formed. The distance between the two electrodes is separated from each other by a distance not less than the length H of the bowtie antenna 6 so as not to interfere with the collection of electromagnetic waves on the semiconductor quantum dots 12.

The bowtie antenna 6 is formed of a vapor-deposited metal thin film, and is formed of, for example, Ti having a thickness of 20 nm and Au having a thickness of 60 nm.
Further, as shown in FIG. 2 (a), the bowtie antenna 6 has two mesa structures facing each other or straddling each other with a mesa structure formed so as to have a width of about 4 μm of the single-electron transistor 14, and forming two equilateral triangles in opposite directions. It spreads and creates a knot at the center of the mesa structure.

The length of the bowtie antenna 6, that is, the length H of electrode passing, is about half of the wavelength of the millimeter wave / far infrared light to be measured.
However, since light is incident at various incident angles during the light collection process, the bowtie antenna 6 can actually detect not only wavelength = 2H but also broadband millimeter-wave / far-infrared light.

As will be described later, one of the blades of the bowtie antenna 6 is divided into three so as to serve as the gate electrodes 32 and 34 for forming the semiconductor quantum dots 12 and the control gate electrode 36 as will be described later. Yes.
In order to apply a bias voltage to each of the divided gate electrodes 32, 34, and 36, each of the divided parts is separated from the metal pad portions 43, 45, 47 (20 nm: Ti, 150 nm: Au).
The other wing serves as a gate electrode 30 and is connected to the metal pad portion 41 through a lead portion 31 having a width of 5 to 10 μm.

  The length directions of the lead portions 31, 33, 35, 37 are perpendicular to the polarization axis direction of the bowtie antenna 6 in order to reduce the influence on the electromagnetic waves. The ohmic electrodes 16 and 17 and the gate electrodes 30, 32, 34, and 36 are wired by bonding to terminals of a standard IC chip package using a gold wire.

Next, the node of the bowtie antenna will be described.
FIG. 3 is a schematic view showing a planar structure in the nodal point region of the bowtie antenna according to the present invention. FIG. 3A shows a millimeter wave / far infrared light having a wavelength of 0.5 to 10 mm, which is used without applying a magnetic field. (B) is used in a magnetic field of 1 to 7T, and is used for a millimeter wave / far-infrared light detector having a wavelength of 0.1 to 0.4 mm, and (c) is 4 to 13T. It is a figure which shows what is used with a millimeter wave and a far-infrared photodetector with a wavelength of 3-10 mm used under a magnetic field. T is a unit symbol of magnetic flux density and indicates Tesla.

  As shown in FIGS. 3A, 3B, and 3C, quantum dots 12a, 12b, and 12c are formed at the nodal points of the bowtie antennas 6a, 6b, and 6c, and a short circuit occurs at the nodal point of the bowtie antenna. The presence or absence and the size of the quantum dots should be the following three different ones depending on the use conditions and the wavelength range of the electromagnetic wave to be measured. In FIG. 3, reference numerals 14a, 14b, and 14c denote two-dimensional electron system mesa structures.

  The first is a use condition without a magnetic field, the detection wavelength range is 0.5 to 10 mm, the bowtie antenna type is standard electrical coupling, and the quantum dot electrode size (effective diameter of the quantum dot two-dimensional electron system) is 0. .2 to 0.4 μm (0.02 to 0.2 μm).

  The second is a use condition of 1T to 7T magnetic field application, the detection wavelength range is 0.1 to 0.4 mm, the bowtie antenna type is standard electrical coupling, quantum dot electrode size (two-dimensional electron system of quantum dots) Is an effective diameter of 0.6 to 0.8 μm (0.4 to 0.6 μm).

  The third is the use condition of 4T-13T magnetic field application, the detection wavelength range is 3-10mm, the bowtie antenna type is the magnetic coupling of short-circuited nodal point, quantum dot electrode size (effective of quantum dot two-dimensional electron system) In this case, the diameter is 0.6 to 0.8 μm (0.4 to 0.6 μm).

FIGS. 3A, 3B, and 3C are diagrams showing a planar structure in the nodal point region of the bowtie antenna corresponding to each of the first, second, and third cases described above.
In addition, the node area | region which forms a quantum dot is the quantum dot electrode size mentioned above.
In the first case shown in FIG. 3A, the quantum dot 12a is electrically coupled to the electromagnetic wave by the bowtie antenna 6a in a configuration for use without applying a magnetic field.
The wavelength range of the electromagnetic wave to be measured is 0.5 mm to 10 mm. When a magnetic field is not applied, the life of the excited state of the quantum dots 12a is relatively short, 10 ns to 1 μs, but a current amplification circuit combining a HEMT amplifier cooled to a helium temperature and an LC tank circuit as described above. Single photon detection is used.

One wing of the bow tie antenna 6a is divided into three to be gate electrodes 32a, 34a, and 36a, and the other wing is a gate electrode 30a.
The widths of the two protrusions 52a and 52a at the tip of the gate electrode 30a are 0.15 μm, and the widths of the tip portions 54a and 54a of the gate electrodes 32a and 34a are 0.15 μm. Further, the gap 55a facing each protrusion is 0.15 μm.
In FIG. 3A, Wa is 2 μm, La is 0.4 μm, and Ma is 0.35 μm.

By biasing the three gate electrodes 32a, 34a, 30a to about -0.6V and the gate electrode 36a to a negative voltage of -0.2V to -3V, the two-dimensional electron system under the gate electrode is driven away ( depletion), a two-dimensional electron system is confined inside the central 0.3 μm square portion to form a quantum dot 12a.
However, the bias voltage of the gate electrode 34a and the gate electrode 30a is finely adjusted so that the quantum dot is weakly tunnel-coupled to the outer two-dimensional electron system.

The gate electrode 36a functions as a control gate electrode, and a single electron transistor is formed by quantum dots.
The bias voltage V _CG of the control gate electrode can vary between -0.2V of -3 V, the effective diameter of the two-dimensional electron system in the quantum dots changes from about 0.2μm to 0.02 [mu] m.

Next, FIGS. 3B and 3C are configurations for applying the magnetic fields 1T to 7T and 4T to 13T, respectively.
When a magnetic field is applied, the lifetime of the quantum dots in the excited state reaches 1 to 1000 seconds depending on the value of the magnetic field and the electron concentration in the quantum dots, and it is extremely easy to use without using a high-speed current amplification circuit. Single photon detection can be performed.

In the second case shown in FIG. 3B, the quantum dot 12b is electrically coupled to the electromagnetic wave by the bowtie antenna 6b, and the wavelength range of the electromagnetic wave to be measured is 0.05 mm to 0.4 mm. It is.
The geometric configuration and the role of each gate electrode forming the bowtie antenna 6b are the same as those in FIG. 3A, but the size of each part is different as follows. That is, the widths of the two protrusions 52b and 52b of the gate electrode 30b are 0.3 μm, and the widths of the tip portions 54b and 54b of the gate electrodes 32b and 34b are 0.3 μm. Further, the gap 55b facing each protrusion is 0.3 μm.
In FIG. 3B, Wb is 4 μm, Lb is 0.7 μm, and Mb is 0.7 μm.

Thus, the two-dimensional electron system is confined inside the 0.7 μm square in the center, and quantum dots having an effective diameter of 0.4 μm to 0.6 μm are formed.
Functions gate electrode 36b as a control gate electrode, is formed a single-electron transistor according to the quantum dots, the bias voltage V _CG of the control gate electrode is varied between -1.5V from -0.3 V.

In the above third case shown in FIG. 3C, the quantum dot 12c is magnetically coupled to the electromagnetic wave by the bowtie antenna 6c, and the wavelength range of the electromagnetic wave to be measured is 3 mm to 10 mm. The width of the mesa structure portion of the two-dimensional electron system, Lc and Mc is 0.7 μm, and constricted portions 56 and 56 having a width of 0.4 μm are formed at two locations separated by 0.7 μm.
One wing of the bowtie antenna 6c is divided into three to form gate electrodes 32c, 34c, and 36c. Of these, the gate electrode 36c is short-circuited to the gate electrode 30c of the other wing of the bowtie antenna 6c by a bridge having a width of 0.2 μm.

By biasing the gate electrode 32c and the gate electrode 34c to a negative voltage, the two-dimensional electron system is confined inside a region of about 0.8 μm square sandwiched between the constricted portions 56 and 56 and the gate electrodes 32c and 34c. The quantum dots 12c having an effective diameter of 0.4 μm to 0.6 μm are formed.
The gate electrode 36c functions as a control gate electrode, and a single electron transistor using quantum dots is formed. The bias voltage V _CG of the control gate electrode is changed between +0.1 V and −0.1 V so as not to give a large change to the electron concentration in the quantum dots.

Next, the operation of the millimeter wave / far infrared photodetector of the present invention will be described. Details and operations of the embodiment differ depending on the first, second, and third cases described above.
The operation in the first case shown in FIG.
In the first case, since the quantum dot is small in size and contains only about 10 to 50 conduction electrons, the electron level is split into discrete energy levels ε _n by the size effect and exchange interaction. .

First, the control gate voltage V _CG is adjusted so that energy splitting in the vicinity of the Fermi level, Δε _nm = ε _n −ε _m , satisfies the following resonance condition for the measured millimeter wave / far infrared light.

ω = 2πΔε _nm / h (1)

In equation (1), ω is the angular frequency of the millimeter wave / far infrared light to be measured, and h is the Planck constant. In general, ε _nm is inversely proportional to the square of the effective quantum dot diameter. For example, V _CG = −3 V to −2 V (quantum dot effective diameter of about 0.02 μm with respect to the wavelength of millimeter wave to be measured / far infrared light of 0.5 mm. ) V _CG = −0.5 V to −0.2 V (quantum dot effective diameter 0.2 μm) with respect to a wavelength of 10 mm of measured millimeter wave / far infrared light.

Next, SET is placed in the state of conductivity peak. That is, even if a source / drain voltage V _SD (100 μV or less) is applied between the two ohmic electrodes in FIG. 2, normally, Coulomb blockage occurs, and the current I _SD does not flow between the source / drain electrodes. When the bias voltage V _CG of the control gate electrode 36a in (a) is slightly changed, Coulomb oscillation is generated in which a finite I _SD appears with a sharp peak every time the value of V _CG changes by 3 mV to 20 mV.

The V _CG and the fine fixed in accordance with the one of the peak position of the I _SD. The minute _VCG adjustment at this time does not substantially affect the above-described resonance condition (1). When the measured millimeter wave / far infrared light is incident in the peak state of this Coulomb vibration, the incident millimeter wave / far infrared light creates a vibrating electric field in the quantum dot portion by the bowtie antenna, and resonance excitation of electrons ε Causes _n → ε _m .

Since the excited state generally has a different spatial symmetry of the electron wave function compared to the ground state, both the tunnel coupling strength and the electrochemical potential of the quantum dot change, and the SET conductivity G changes greatly by about 10% to 90%. Will occur.
This change in conductivity is held for approximately 10 ns to 1 μs until excitation is extinguished by phonon emission, which is measured by a high speed current amplifier.

Next, the operation in the second case shown in FIG.
FIG. 4 is a conceptual diagram of an electrical transition showing electronic excitation between levels due to millimeter-wave / far-infrared single photon absorption inside a quantum dot under a magnetic field, and (a) is an electron between Landau levels. Excitation, (b) relaxation of excited electrons and holes to a stable state, (c) polarization within the quantum dot, (d) electrostatic potential change ΔU and electrochemical potential change Δμ ₀ ↑ FIG.

  In FIG. 4, 0 and 1 indicate electron energy levels, ↑ indicates up spin, and ↓ indicates down spin.

In this second case, the quantum dots are large in size and contain 200 to 400 conduction electrons, so the size effect Δε _nm on the electron level is small, but by applying a magnetic field as shown in FIG. energy structure is split into intervals _{(h / 2π) ω C ≒} (h / 2π) eB / m * of the Landau level.

Here, ω _C is the angular frequency when energy splitting in the vicinity of the Fermi level satisfies the resonance condition for the measured millimeter-wave / far-infrared light, and e is the elementary quantity of 1.6 × 10 ^−19. Coulomb, B is magnetic flux density, m * is an effective mass of 0.068 m, and m is an electron mass.

  In the case of this second configuration, first, a magnetic field is applied so that the angular frequency ω of the millimeter wave / far infrared light to be measured satisfies the resonance condition including the influence of the next plasma vibration.

ω = [ω _P ² + (ω _C / 2) ² ] ^1/2 + (ω _C / 2) (2)

Here, ω _P is the plasma angular frequency of the quantum dot, which is about 0.43 mm for the quantum dot in FIG. 3B when expressed by the plasma wavelength λ _P = 2πc / ω _P.
In addition, c shows the speed of the light in a vacuum.

  The magnetic field that specifically satisfies Equation (2) is approximately B = 6T to 7T with respect to the measurement wavelength of 0.1 mm, and B = 1T to 1.5T with respect to the measurement wavelength of 0.4 mm. T is a unit symbol of magnetic flux density and indicates Tesla.

  Next, in the same manner as in the first case described above, SET is placed in the peak state of Coulomb oscillation, and the measured millimeter wave / far infrared light is incident. The incident millimeter-wave / far-infrared light creates an oscillating electric field in the quantum dot portion by means of a bowtie antenna, and as shown by the arrow in FIG. 4 (a), resonance excitation of electrons between Landau levels, that is, magnetoplasma resonance is performed. Wake up.

Excited electrons (black circles in FIG. 4 (a)) and holes (white circles in FIG. 4 (a)) lose 10 s by losing excess energy to the lattice system as shown in FIG. 4 (b). Relax in about time.
At that time, as shown in FIG. 4 (c), electrons and holes move to the inside and outside of the quantum dot due to the influence of the electrostatic potential forming the quantum dot, and are separated spatially. Annular polarization occurs inside.

Therefore, the electrochemical potential of the electron level of the outermost shell of the quantum dot changes by the amount of change in electrostatic potential due to polarization, ΔU = 30 to 60 μV.
As a result, the SET operation changes from a conduction peak, ie, a state where the conductivity G is maximum, to a Coulomb blockage state, ie, G≈0.
The conduction blocking state lasts until the electrons and holes recombine in the quantum dot, but the recombination lifetime is long due to the spatial separation between the electrons and holes, making it easy to detect single photon absorption. It can be carried out.

At this time, the average electron concentration N _d in the quantum dot is adjusted by the control gate voltage V _CG and the bias voltage of the back gate (FIGS. 1 and 3), and the Landau level occupation index ν in the dot is set to an even number, for example, By setting ν = 2.4 to 1.9, 4.6 to 4.0, 6.6 to 6.0, etc., a particularly long life state can be realized.
Here, the Landau level occupation index ν can be expressed by the following equation (3).

ν = N _d / (2πeB / h) (3)

Next, the operation in the third case shown in FIG.
FIG. 5 is a conceptual diagram of magnetic transitions showing electronic excitation between levels due to millimeter-wave / far-infrared single-photon absorption inside a quantum dot under a magnetic field, and (a) shows electronic excitation between spin states. (B) is relaxation of excited electrons and holes to a stable state, (c) is polarization in a quantum dot, and (d) is a diagram showing a change ΔU in electrostatic potential.

In the third case, the size effect is small as in the case of the quantum dots in the second case. In addition to the splitting to the Landau level by the application of a magnetic field, as shown in FIG. Energy separation, Δε _M = g * μ _B B.
Here, g * represents an effective g factor, and μ _B represents a Bohr magneton.

  First, a magnetic field is applied so as to satisfy the following resonance condition with respect to the angular frequency ω of the millimeter wave / far infrared light to be measured.

ω = 2πΔε _M / h (4)

  The effective g factor is about g * = 0.44.

Next, the SET is placed in the peak state of the Coulomb vibration in the same manner as in the first case described above, and the measured millimeter wave / far infrared light is incident.
The incident millimeter-wave / far-infrared light generates an oscillating current at the short-circuited node of the bowtie antenna and creates an oscillating magnetic field in the quantum dot area. As a result, as shown by the arrow in FIG. 5A, magnetic resonance excitation of electrons occurs.

As shown in FIG. 5B, the excited electrons and holes are alleviated within about 10 ns by losing excess energy to the lattice system.
At this time, as shown in FIG. 5C, electrons and holes are spatially separated inside the quantum dot due to the influence of the electrostatic potential forming the quantum dot, and annular polarization occurs inside the quantum dot.

Therefore, the electrochemical potential of the electron level of the outermost shell of the quantum dot changes by an electrostatic potential change due to polarization, ΔU = 10 to 50 μeV.
As a result, the SET operation changes from a conduction peak, ie, a state where the conductivity G is maximum, to a Coulomb blockage state, ie, G≈0.
The state of conduction interruption continues until electrons and holes recombine in the quantum dot, but the recombination lifetime is long due to the low scattering probability of spin inversion, and single photon absorption can be easily detected. Can do.

Next, another preferred embodiment having a plurality of quantum dots will be described.
In the case of having one quantum dot as described above, unless the magnetic field is applied, the lifetime of the excited state is as short as 1 μsec or less. Therefore, in order to perform single photon detection, it is necessary to use a current amplifier that combines a HEMT amplifier cooled to a helium temperature and an LC circuit, or to apply a magnetic field so that the lifetime of the excited state is 1 ms or more. .
However, in other embodiments having a plurality of quantum dots, the sensitivity can be further improved without applying a magnetic field.

  In the millimeter-wave / far-infrared photodetector according to another embodiment, the first and second sub-micron conductive quantum dots that are electrostatically coupled to each other are used as constituent elements. The first quantum dot absorbs electromagnetic waves, the second quantum dot operates as a single electron transistor (SET), and this SET detects electromagnetic wave absorption in the first quantum dots.

Hereinafter, the operation principle will be described with reference to the conceptual diagram of FIG.
FIG. 6 is a conceptual diagram showing an operation principle of another embodiment, and is an energy structure conceptual diagram in which two adjacent quantum dots are formed in an electron system.
When a semiconductor quantum dot having an effective size of 0.02 μm to 0.3 μm is used as the first quantum dot 61 characterized by the electrostatic potential Ua, it is determined by the applied gate voltage as shown in FIG. The electronic state lower than the energy threshold, that is, the ionization energy is quantized to form discrete bound levels 59 bound to the first quantum dots 61.

On the other hand, an electron level whose energy is larger than the threshold value forms a continuous free level 58 of the electron system 63 that spreads outside the first quantum dot 61.
This ionization energy takes a value of 100 μV to 20 mV in terms of voltage depending on the height of the potential barrier 57 that forms the first quantum dots 61.
The height of the potential barrier 57 can be controlled by a bias voltage applied to the gate electrode of the first quantum dot 61.
Therefore, as shown by the arrows in FIG. 6, by irradiating millimeter-wave / far-infrared light having electromagnetic photon energy equal to or higher than the ionization energy, electrons in the first quantum dots 61 are separated from the discrete binding levels 59. It is possible to excite the continuous free level 58 of the electron system 63 outside the potential barrier 57. In FIG. 6, black circles indicate excited electrons, and white circles indicate holes from which electrons have been removed, that is, holes.

The electrons excited by the continuous free level 58 quickly escape from the potential barrier 57 of the first quantum dot 61 to the external electron system 63 within 1 ns, and the first quantum dot 61 is divided by the amount of elementary charge + e. Only positively charged, ie ionized.
On the other hand, electrons that have escaped to the outside of the potential barrier 57 lose extra energy within about 10 ns due to electron-electron interaction and electron-lattice interaction, and quickly relax to the Fermi level of the electron system 63. Therefore, the first quantum dot 61 cannot return to the first quantum dot 61 due to the potential barrier 57.
Therefore, the ionization state of the first quantum dots 61 is held for a long time, for example, 10 μs to 1000 seconds.

The second quantum dot 62 adjacent to the first quantum dot 61 has an electrostatic potential Ub in which electrons are confined at discrete levels, and forms a SET.
The second quantum dots 62 are formed of either semiconductor quantum dots or metal quantum dots.
The second quantum dots 62 do not have electrical continuity with the first quantum dots 61, but are electrostatically coupled because they are adjacent via a potential barrier.

Therefore, the electrostatic potential of the second quantum dot 62 is changed by charging of the first quantum dot 61, that is, ionization, and as a result, the conductivity of the SET is increased, for example, about 20% to 99%.
At this time, the state in which the conductivity is changed is maintained until the ionization of the first quantum dot 61 disappears and the state returns to the neutral state.
On the other hand, as described above, the lifetime of the ionization state of the first quantum dots 61 is as long as 10 μs to 1000 seconds, so that it becomes a very sensitive detector.
In particular, single millimeter-wave / far-infrared photons can be detected through time-resolved measurement of current.

Next, the configuration of another embodiment will be described.
Basically, two types of preferred embodiments are possible for the millimeter-wave / far-infrared photodetectors according to other embodiments. In other embodiments, the structure is basically the same as the structure shown in FIGS. 1 and 2, but in another embodiment, the bipolar electrode has two quantum dots that are electrostatically separated and serves also as a gate electrode. The difference is that a first quantum dot that is formed at a node of the child antenna and absorbs electromagnetic waves and a second quantum dot that detects the electromagnetic wave absorption of the first quantum dots and forms a SET are formed.

7A and 7B are diagrams showing a millimeter wave / far-infrared light detector according to another embodiment, where FIG. 7A is a plan view of an A-type structure, and FIG. 7B is a view of a B-type structure. A plan view is shown.
In any structure, the first quantum dot is electrically coupled to the electromagnetic wave by the dipole antenna.

Referring to FIGS. 7A and 7B, the millimeter-wave / far-infrared photodetector according to another embodiment is similar to that shown in FIG. Both structures are fabricated using lithographic techniques on a modulation-doped GaAs / Al _0.3 Ga _0.7 As single heterostructure substrate. Further, both the A-type structure and the B-type structure are formed in a symmetric predetermined shape by mesa etching the electronic heterostructure.
An equivalent structure can be similarly manufactured on a group IV semiconductor, for example, a Si substrate by using a lithography technique.

  Further, one wing 67a, 67b and the other wing 68a, 68b of the dipole antenna 65a, 65b are applied with a bias voltage by the metal lead wires 69a, 69b and the metal lead wires 69a ′, 69b ′, respectively. The metal pads 71a and 71b are connected to the metal pads 72a and 72b. In this embodiment, the metal lead wire and the metal pad are formed by alloying with a deposited film of Ti of 20 nm and Au of 150 nm.

  As shown in FIG. 7A, in the A-type structure, the electronic mesa structure 63a is formed so as to have a constriction in the nodal point region 70a of the dipole antenna 65a, and is divided in two from the nodal point region 70a. Electronic mesa structures 76a and 77a are formed. An ohmic electrode 66a is formed at one end of the electronic mesa structure 63a, and ohmic electrodes 81a and 82a are formed at the other end divided into two. The ohmic electrode 81a is a source electrode of SET 64a described later, and the ohmic electrode 82a is a drain electrode of SET 64a.

First quantum dots 61a and second quantum dots 62a are formed in the nodal point region 70a of the dipole antenna 65a. The dipole antenna 65a couples the first quantum dot 61a and the electromagnetic wave in the nodal point region 70a.
Further, the first quantum dot 61a is an electrostatic potential barrier formed by the tip of the gate electrode 67a that doubles as one wing of the dipole antenna 65a and the gate electrode 68a that doubles as the other wing, and is separated from the electron system outside the quantum dot. It has been.

  The second quantum dot 62a is formed adjacent to the first quantum dot 61a in the electronic system. The second quantum dots 62a are formed by applying a bias voltage from the gate electrodes 78a, 79a, 80a by the metal lead wires 73a, 74a, 75a, and are weakly tunnel-junctioned to the electron system of the electron system mesa structures 76a, 77a. .

The second quantum dots 62a, the electronic mesa structures 76a and 77a, and the ohmic electrodes 81a and 82a form a SET 64a.
The metal lead wires 73a, 74a, and 75a are connected to the gate electrodes 78a, 79a, and 80a, respectively. In the present embodiment, the ohmic electrodes 66a, 81a and 82a are formed by alloying Au / Ge.
Furthermore, the metal lead wires 69a, 69a ′, 73a, 74a, and 75a for applying a bias voltage to the electronic mesa structure 63a (including the electronic mesa structures 76a and 77a) and the gate are used to avoid absorption of electromagnetic waves. Each is formed to have a width of 5 μm or less, and its length direction is formed perpendicular to the axial direction of the dipole antenna 65a.

Next, in the millimeter-wave / far-infrared light detector having the B-type structure shown in FIG. 7B, the end of the constricted portion of the electronic mesa structure 63b is a nodal point region 70b of the dipole antenna 65b. A first quantum dot 61b is formed in the nodal point region 70b.
Here, the configuration of the dipole antenna 65b and the first quantum dot 61b is the same as that of the A-type structure.
Further, the second quantum dot 62b in the B-wave structure millimeter-wave / far-infrared photodetector is formed of a metal film provided on the upper surface of the first quantum dot 61b, and the first quantum dot 61b. Are electrically coupled, but electrical conduction (tunnel junction) is interrupted.
The second quantum dots 62b formed of this metal film are weakly tunnel-bonded to the metal lead wire 76b and the metal lead wire 77b connected to the source electrode 81b and the drain electrode 82b, respectively, and a SET 64b is formed.

Next, the dipole antenna, the first quantum dot, and the second quantum dot will be described in detail.
FIG. 8 is an enlarged view of the main part of the nodal region of the dipole antenna. FIG. 8A is a structure in which the second quantum dot having the A-type structure is separated from the first quantum dot by the gate electrode, and FIG. A structure in which an electronic mesa structure in which a first quantum dot and a second quantum dot having a structure are formed is formed separately, and FIG.

Referring to FIG. 8A, the first quantum dot 61a is formed in the gap between the protrusion 83a of the gate electrode 67a and the protrusion 84a of the gate electrode 68a by applying a bias voltage.
This gap L ₁ a is about 0.5 μm. The width Ma of the node region 70a of the electronic mesa structure 63a is about 0.4 μm to 0.5 μm.

The second quantum dot 62a is formed in a gap formed by the protrusion 84a of the gate electrode 68a and the tips of the metal lead wires 73a, 74a, 75a extending from the gate electrode by applying a gate voltage, and this gap L ₂ a Is about 0.3 μm to 0.5 μm.
The gate electrodes 67a and 68a also serve as the wings of the dipole antenna 65a, and also serve to electrically couple electromagnetic waves to the first quantum dots 61a.
The protrusion 83a has a width of about 0.3 μm and a length of about 0.7 μm and is formed across the nodal point region 70a of the electronic mesa structure. The protrusion 84a has a width of 0.1 μm and a length of 0.3 μm. It is formed so as not to cross the nodal point region 70a and overlap with a part thereof.

This is to ensure a sufficiently large capacitance coupling between the first quantum dot 61a and the second quantum dot 62a.
However, the electrical continuity (tunnel junction) between the first quantum dot 61a and the second quantum dot 62a is cut off by applying a negative bias voltage to the gate electrode 68a (projection 84a).

  In the present embodiment, the tip portions of the metal lead wires 73a, 74a, 75a extending from the gate electrode are each about 0.1 μm wide, and the gap between them is formed about 0.1 μm.

By biasing the gate electrode 67a to a negative voltage of about −0.3V to −2V and the gate electrode 68a to a negative voltage of about −0.7V, the first quantum dot 61a is formed.
By biasing the metal lead wire 73a and metal lead wire 75a of the gate electrode to a negative voltage of about -0.7V, and further, the metal lead wire 74a of the control gate electrode to a negative voltage of about -0.3V to -0.7V. A quantum dot 62a is formed.
Here, the bias voltage of the gate electrode 67a (projection 83a) determines the ionization energy when the first quantum dot 61a absorbs electromagnetic waves.

  For example, the ionization energy is about 0.2 meV at −0.3 V, the threshold detection wavelength corresponds to 5 mm, and the value continuously changes with a negative voltage change, about 30 meV at −2 V, about the threshold detection wavelength. It reaches about 30 μm.

In order to operate the second quantum dot 62a as the SET 64a, the bias voltage of the metal lead wire 73a and the metal lead wire 75a of the gate electrode is finely adjusted, so that the second quantum dot 62a has the electronic mesa structures 76a and 77a. It is weakly tunnel-coupled to the electronic system.
The bias voltage of the gate electrode 68a (protrusion 84a) is selected in the vicinity of the threshold voltage at which the tunnel coupling between the first quantum dot 61a and the second quantum dot 62a disappears so that the electrostatic coupling is maximized. Is done.

Next, FIG. 8A ′ shows that the second quantum dot 62a ′ is formed by being separated from the first quantum dot 61a ′ by mesa etching, and no protrusion is formed on the gate electrode 68a ′. Different from the structure of (a).
The corresponding dimensions Ma ′, L ₁ a ′, L ₂ a ′, the bias voltage and the like are the same as those in FIG. Note that members corresponding to those in FIG. 8A are distinguished from each other by adding “dash” to the same reference numerals.
In this embodiment, the gap between the first quantum dot 61a ′ and the second quantum dot 62a ′ is formed to be about 0.1 μm.

Next, in the millimeter-wave / far-infrared light detector having the B-type structure shown in FIG. 8B, the end of the constricted portion of the electronic mesa structure 63b is in the nodal point region 70b, and the first nodal point is located in this region. A quantum dot 61b is formed. The first quantum dots 61b, the external electronic mesa structure 63b, and the gate electrodes 67b and 68b are the same in size and shape as the corresponding members shown in FIG.
The SET by the second quantum dot 62b is produced on the first quantum dot 61b by the Doran bridge method. “TA Fulton and GJ Dolan, Phys. Rev. Lett. 59, p. 109, 1987” can be cited as a document showing the Doran bridge method.

Specifically, first, after the second quantum dots 62b having a thickness of 0.06 μm, a width of 0.1 μm, and a length of 0.3 μm are formed by vapor deposition of aluminum, an oxygen gas of about 10 mTorr is formed on the surface thereof. Oxidize in atmosphere and cover with aluminum oxide film.
In oxidizing the surface, the oxidation time is adjusted so that the electrical resistance at normal temperature between the metal lead wire 76b and the metal lead wire 77b, which will be described later, is within 100 kΩ to 400 kΩ.

Next, the metal lead wires 76b and 77b are formed by vapor deposition of aluminum by superimposing on the first quantum dots 61b by oblique vapor deposition. The metal lead wires 76b and 77b have a thickness of about 0.06 μm and a tip width of about 0.1 μm.
However, in this embodiment, the gap 85b at the tip of the metal lead wire is set to 0.1 μm.

  At this time, the tip portions of the metal lead wires 76b and 77b are formed so as to overlap the second quantum dots 62b by about 0.05 μm. At this time, the metal lead wire 76b is connected to the source electrode 81b, the metal lead wire 77b is connected to the drain electrode 82b, and the aluminum oxide interposed between the metal lead wires 76b and 77b and the second quantum dot 62b. The layer becomes a tunnel junction and a SET is formed.

Next, the operation of the millimeter wave / far infrared photodetector according to another embodiment will be described.
The operation of the millimeter-wave / far-infrared photodetector having the structure shown in FIG. Referring to FIGS. 7A and 8A, first, a bias voltage is applied to the gate electrode 67a, that is, the protrusion 83a and the gate electrode 68a, that is, the protrusion 84a, to form the first quantum dot 61a. To do.

Next, in this state, the gate electrodes 78a, 79a, and 80a are further biased, that is, a bias voltage is applied to the metal lead wires 73a, 74a, and 75a to form the second quantum dots 62a. Works as.
That is, the current flowing by applying a source / drain voltage of V _SD = 100 μV or less between the electronic mesa structure 76a and the electronic mesa structure 77a is measured. The bias voltage of the control gate electrode 79a, that is, the metal lead wire 74a is finely adjusted so that the SET conductivity is maximized when no electromagnetic wave is incident on the dipole antenna 65a. Further, the potential of the electronic mesa structure 63a and the electronic mesa structure 76a are kept at the same potential.

  When the electromagnetic wave captured by the dipole antenna 65a is absorbed by the first quantum dot 61a, the first quantum dot 61a is ionized to + e, and the electrostatic potential of the second quantum dot 62a is changed to change the SET. The conductivity of is greatly reduced. The absorption of one electromagnetic wave photon can be detected by detecting it with a current amplifier. The electrons that have escaped from the first quantum dots 61a to the electron system of the external electronic mesa structure 63a by ionization are absorbed there.

Next, the operation of the millimeter wave / far infrared photodetector having the structure shown in FIG.
Referring to FIGS. 7A and 8A ′, first, a first quantum dot 61a ′ is formed by applying a bias voltage to the gate electrode 67a ′, that is, the protrusion 83a ′. Further, the gate electrode 68a ′ has the same potential as the electronic mesa structure 63a ′.
Next, in this state, the gate electrodes 78a ′, 79a ′, and 80a ′ are further biased, that is, a bias voltage is applied to the metal lead wires 73a ′, 74a ′, and 75a ′ to form the second quantum dots 62a ′. The second quantum dot 62a ′ operates as a SET. The other operations are the same as in the structure of FIG.

Finally, the operation of the millimeter-wave / far-infrared photodetector having the structure shown in FIG. 8B will be described.
The formation of the first quantum dots 61b is the same as the formation of the first quantum dots 61a ′ in the structure shown in FIG.
The second quantum dot 62b is already formed on the first quantum dot 61b as a SET. Referring to FIGS. 7B and 8B, a source-drain voltage of V _SD = 100 μV or less is applied between source electrode 81b, that is, metal lead wire 76b and drain electrode 82b, that is, metal lead wire 77b. Measure the current that flows when applied.

  However, by setting the electronic mesa structure 63b and the gate electrode 68b to the same potential and finely adjusting the potential within ± 1 mV with respect to the metal lead wire 76b formed of aluminum, the SET of when there is no electromagnetic wave incidence is obtained. Maximize conductivity.

  When the first quantum dot 61b absorbs the electromagnetic wave and is ionized to + e, the electrostatic potential of the second quantum dot 62b is changed and the SET conductivity is greatly reduced as described above. The absorption of one electromagnetic wave photon can be detected by detecting it with a current amplifier. The electrons that have escaped from the first quantum dots 61b to the electron system of the external electronic mesa structure 63b by ionization are absorbed there.

In the three types of structures shown above, holes and electrons excited by electromagnetic wave absorption are generated separately inside and outside the first quantum dot, respectively, so that an extremely long excitation state without applying a magnetic field, That is, an ionization lifetime can be realized. The lifetime of the ionized state of the first quantum dots 61a, 61a ′, 61b is 10 μsec or longer, and the electromagnetic wave single photon can be detected very easily.
Therefore, in the millimeter wave / far-infrared light detector according to another embodiment, the sensitivity is further increased and a detector operating at a higher temperature can be realized without applying a magnetic field.

  Furthermore, when using excitation between discrete levels, wavelength selectivity occurs in the detection of electromagnetic waves, but in this other embodiment, a threshold, that is, ionization energy, is used in order to use excitation from discrete levels to continuous band levels. Sensitivity can be detected in a continuous wavelength band having the above energy.

  The upper limit of the operating temperature is determined by the charging energy of the second quantum dots 62a, 62a ′, 62b forming the SET, and is about 1 K in the structure of FIG. 8B, and in the structure of FIG. 8A ′. Up to about 1.3K, and up to about 2K in the structure of FIG. Therefore, the operating temperature can be increased up to 2K by making the second quantum dots minute.

Furthermore, since the ionization energy can be directly controlled through the adjustment of the potential barrier height by the gate voltage of the first quantum dot, the ionization energy, that is, the detection threshold wavelength can be controlled.
Therefore, the long wavelength limit of the detectable electromagnetic wave can be determined by the bias voltage of the gate electrodes 67a, 67a ′, 67b forming the first quantum dots in all structures.

  The present invention is not limited to the above-described embodiments, and various modifications are possible based on the spirit of the present invention, and these are not excluded from the scope of the present invention.

Next, the measurement results of actual single photon detection using the millimeter-wave / far-infrared single photon detector of the present invention under the use conditions in the second case described above are shown in FIGS. 11 and FIG.
9 to 12 show the shape and the shape of FIG. 3B using a GaAs / Al _0.3 Ga _0.7 As heterostructure crystal having a two-dimensional electron mobility of 80 m ² / Vs and an electron concentration of 2.3 × 10 ¹⁵ / m ² . A measurement example of a millimeter-wave / far-infrared single photon detector created in size is shown.

FIGS. 9 and 10 show extremely weak far-infrared emission from the quantum Hall effect element, that is, the wavelength under the conditions of measurement temperature 0.07K, V _SD = 25 μV, V _CG = 0V, B = 3.67T. This is a measurement example of SET conductivity when a far-infrared light having a power of about 10 ⁻¹⁸ W at the bowtie antenna position is incident when the current of the light emitting element is 0.19 mm and I _emit = 4 μA. However, the measurement time constant is 3 milliseconds.

FIGS. 9A to 9C show the dependence of the SET conductivity on the control gate voltage V _CG . The sharp Coulomb oscillation near V _CG = −0.6881 V when there is no irradiation with far-infrared light. The peak (FIG. 9A) is disturbed by extremely weak far-infrared light irradiation (FIG. 9B), and the peak near V _CG = −0.6886 V corresponding to the excited state as the light intensity increases. Migrate to

FIGS. 10D to 10F show temporal changes in the conductivity G when the control gate voltage V _CG is fixed to the peak position without irradiation, V _CG = −0.6881V. Each time absorption of a single photon occurs, the state is switched to a conduction cut-off state, indicating that the frequency of switching (the frequency of photons flying) increases as the far-infrared light intensity increases.

FIG. 11 shows the magnetic field dependence in which the lifetime of the excited state is strong. When the magnetic field B is 3.8 T, the maximum value is taken in the vicinity of ν = 2, and the value reaches the order of 1000 seconds.
The structure of the details depending on the magnetic field occurs when the number of electrons existing in the upper Landau level changes one by one due to the magnetic field change.
FIG. 12 shows that single photons can be detected even if the temperature is increased to 0.37 K under the same conditions as in FIGS.

The millimeter wave / far-infrared single photon detector of this embodiment measures resonance excitation between electron levels in a semiconductor quantum dot through the amplification action of a single electron transistor.
For this reason, about 1 extremely weak photon flux can be detected in 100 seconds. Considering a measurement time of 100 seconds, this sensitivity corresponds to NEP = 10 ⁻²³ W / Hz ^1/2 , which is about 10 million times better than the conventional highest sensitivity detector.
Further, high-speed measurement can be performed with a time constant of about 3 n seconds without losing sensitivity.

  In addition, this invention is not limited to the said embodiment, A various deformation | transformation is possible based on the meaning of this invention, and these are not excluded from the scope of the present invention.

1 is a schematic sectional view of a millimeter-wave / far-infrared photodetector according to the present invention including a condensing system. It is a figure which shows the millimeter wave far-infrared photodetector based on this invention, (a) is a top view of the single electron transistor by a bowtie antenna and a quantum dot, (b) is a partial schematic cross section of a mesa structure part FIG. BRIEF DESCRIPTION OF THE DRAWINGS It is the schematic which shows the planar structure in the nodal point area | region of the bowtie antenna which concerns on this invention, (a) is used for the millimeter wave far-infrared light detector of wavelength 0.5-10mm used by no magnetic field application. Used, (b) used in a magnetic field of 1 to 7T, used for a millimeter-wave / far-infrared photodetector with a wavelength of 0.1 to 0.4 mm, and (c) used in a magnetic field of 1 to 13T. It is a figure which shows what is used and is used with the millimeter wave and far-infrared photodetector of wavelength 0.35-10mm. It is a conceptual diagram of the electrical transition which shows the electronic excitation between the levels by the millimeter wave far-infrared single photon absorption inside the quantum dot in the magnetic field in this invention, (a) is an electrical transition (magnetoplasma resonance). ) Excitation between Landau levels, (b) relaxation of excited electrons and holes to a stable state, (c) polarization in quantum dots, (d) electrostatic potential change ΔU and electrochemical It is a figure showing change of potential Δμ0 ↑. It is a conceptual diagram of the magnetic transition which shows the electronic excitation between the levels by the millimeter wave far-infrared single photon absorption inside the quantum dot in the magnetic field in this invention, (a) is a magnetic transition (magnetic resonance). (B) shows relaxation of excited electrons and holes to a stable state, (c) shows polarization in the quantum dot, and (d) shows a change ΔU in electrostatic potential. is there. It is a conceptual diagram which shows the operation principle of other embodiment which concerns on this invention. It is a figure which shows the millimeter wave far-infrared photodetector which concerns on other embodiment, (a) is a top view of A type structure, (b) shows the top view of B type structure. It is the principal part enlarged view of the dipole antenna nodal point area | region which concerns on this invention, (a) is the structure which isolate | separates the 2nd quantum dot of A-type structure from a 1st quantum dot by a gate electrode, (a ') is A structure in which an electronic mesa structure in which a first quantum dot and a second quantum dot having an A-type structure are formed is formed separately, and (b) is an enlarged view of a main part of the B-type structure. It is a figure which shows the example of a measurement of the millimeter wave and far-infrared single detection in this invention, (a) when there is no irradiation of far-infrared light, (b) when the electric current of a light emitting element is 2 microamperes, (c ) Is a graph showing the dependence of the SET conductivity on the control gate voltage when the current of the light emitting element is 3.5 μA. It is a figure which shows the example of a measurement of the millimeter wave and the far-infrared single detection in this invention, and shows the switching of SET operation | movement which occurs by single photon absorption. (D) is when the current of the light emitting element is 2 μA, (e) is when the current of the light emitting element is 3 μA, and (f) is when the current of the light emitting element is 4 μA. Further, (g) shows the dependence of the excited state probability on the light emitting element current. It is a figure which shows the example of a measurement of the millimeter wave and far-infrared single detection in this invention, and shows the magnetic field strength dependence of the lifetime of an excited state. It is a figure which shows the example of a measurement of the millimeter wave and far-infrared single detection in this invention, and shows the temperature dependence of switching of SET operation | movement which occurs by single photon absorption.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Millimeter wave / far infrared light introduction part 2 Millimeter wave / far infrared light 3 Optical introduction tube 4 Semiconductor substrate 6, 6a, 6b, 6c Bowtie antenna 7 Package 8 Back surface gate electrode 9 Dielectric objective lens 10 Millimeter wave / Far-infrared photodetector 12 Quantum dot 14 Single electron transistor 16, 17 Ohmic electrodes 30a, 32a, 34a, 36a Gate electrodes 30b, 32b, 34b, 36b Gate electrodes 30c, 32c, 34c, 36c Gate electrodes 31, 33, 35, 37 Lead portion 41, 43, 45, 47 Metal pad portion 57 Potential barrier 58 Continuous free level 59 Discrete bound levels 61, 61a, 61a ′, 61b First quantum dots 62, 62a, 62a ′, 62b Second quantum dot 63 Electronic system 63a, 63a ', 63b Electronic mesa structure 76a, 77a, 76a', 77a 'Electronic system mesa structure 64a, 64b SET
65a, 65a ', 65b Dipole antennas 67a, 67a', 67b Gate electrodes 68a, 68a ', 68b Gate electrodes 69a, 69a', 69b, 69b 'Metal lead wires 70a, 70b Node regions 72a, 72b Metal pads 73a, 74a, 75a Metal lead wires 73a ', 74a', 75a 'Metal lead wires 78a, 79a, 80a Gate electrodes 81a, 81b Source electrodes (ohmic electrodes)
82a, 82b Drain electrode (ohmic electrode)
83a, 83a ', 83b, 84a Projection

Claims (20)

An electromagnetic wave coupling means for concentrating the electromagnetic wave in a sub-micron-size micro space region, a first quantum dot that absorbs the concentrated electromagnetic wave to cause ionization, and a second electrostatically coupled to the first quantum dot A single-electron transistor comprising a quantum dot of
The single electron transistor includes a single heterostructure forming a two-dimensional electron system, a gate electrode for controlling the electrostatic potential of the quantum dot tunnel-coupled in the two-dimensional electron system, the quantum dot and the tunnel A source electrode and a drain electrode to be coupled,
The electromagnetic wave is detected based on a change in electrical conductivity of the single-electron transistor due to a change in electrostatic state of the second quantum dot accompanying ionization of the first quantum dot. , Millimeter wave / far infrared light detector.   The ionization of the first quantum dot is caused by exciting electrons in a quantization bound state of the first quantum dot to a free electron state of an electron system outside the first quantum dot, The millimeter wave / far infrared photodetector according to claim 1.   3. The millimeter wave / far field according to claim 1, wherein the ionization energy of the first quantum dot is controllable by a magnitude of a bias voltage applied to a gate of the first quantum dot. Infrared light detector.   The first quantum dots and the second quantum dots are formed on the same semiconductor structure substrate and electrostatically separated by a bias voltage applied to each gate, The millimeter-wave / far-infrared light detector according to claim 1.   5. The millimeter wave / far field according to claim 1, wherein the first quantum dot and the second quantum dot are formed adjacent to each other through a gap in a semiconductor. Infrared light detector.   The second quantum dot is a metal dot formed on the first quantum dot, and the single electron transistor is formed by tunnel junction with a metal lead formed on the metal dot. The millimeter-wave / far-infrared light detector according to any one of claims 1 to 5.   The millimeter-wave / far-infrared photodetector according to claim 6, wherein the second quantum dot is an aluminum metal dot, and a portion where the tunnel junction is made is aluminum oxide.   The millimeter-wave / far-infrared photodetector according to claim 7, wherein the electromagnetic wave coupling means is an antenna that electrically couples the first quantum dot and the electromagnetic wave.   9. The millimeter wave / far red according to claim 8, wherein the electromagnetic wave coupling means also serves as a gate for applying a bias voltage to form the first quantum dot and the second quantum dot. Outside light detector.   10. The millimeter wave according to claim 1, wherein a length direction of a lead portion of the electromagnetic wave coupling means is formed perpendicular to a polarization axis direction of the electromagnetic wave coupling means. Far-infrared light detector.   10. The millimeter-wave / far-infrared light detection according to claim 1, wherein a size of a nodal point of the electromagnetic wave coupling means and a maximum size of the quantum dot are approximately the same. vessel.   10. The millimeter-wave / far-infrared photodetector according to claim 1, wherein an electrode passing length of the electromagnetic wave coupling means is about ½ of a wavelength of the electromagnetic wave.   The millimeter-wave / far-infrared light detection according to claim 1, wherein a distance between a source electrode and a drain electrode of the single electron transistor is equal to or longer than a length in a polarization axis direction of the electromagnetic wave coupling means. vessel.   The millimeter-wave / far-infrared photodetector according to claim 1, wherein the single electron transistor is a compound semiconductor.   The millimeter-wave / far-infrared photodetector according to claim 1, wherein the single electron transistor is a III-V compound semiconductor.   The millimeter-wave / far-infrared photodetector according to claim 1, wherein the single-electron transistor has a selectively doped single heterostructure of a III-V compound semiconductor superlattice.   2. The millimeter-wave / far-infrared photodetector according to claim 1, wherein the single-electron transistor has an aluminum gallium arsenide / gallium arsenide selective-doped single heterostructure.   The millimeter wave / far infrared photodetector according to claim 1, wherein the single electron transistor is a group IV semiconductor.   2. The millimeter wave / far infrared photodetector according to claim 1, wherein the single electron transistor is formed symmetrically with respect to the quantum dot.   The millimeter-wave / far-infrared light detector according to claim 1, further comprising a light introducing portion that guides an electromagnetic wave to the electromagnetic wave coupling means.

Images