TERAHERTZ TRANSISTOR

Abstract

Superconducting Meissner effect transistors, methods of modulating, and systems are disclosed. In one aspect a disclosed transistor includes a superconducting bridge between a first and a second current probe, the first and second current probe being electrically connected to a source and a drain electrical connection, respectively and a control line configured to emit a magnetic field signal having signal strength Hsig at the superconducting bridge. In one aspect the emitted magnetic field is configured to break Cooper pairs in the superconducting bridge.

Description

BACKGROUND

This application is a continuation of U.S. patent application Ser. No. 15/600,036, filed May 19, 2017, which claims priority under 35 U.S.C. § 119(e) to US provisional patent application Ser. No. 62/339,384 filed on May 20, 2016, both of which are hereby incorporated by reference in their entirety herein.

Terahertz radiation covers a range of frequencies between the infrared and microwave portions of the spectrum. Like other frequency ranges, there are applications suited to the use of Terahertz radiation. The small size of Terahertz waves (sub-millimeter) coupled with bio-safety enables viewing objects in fine detail for security, radar, and medical applications. The related high frequency provides high bandwidth for telecommunications. And the Terahertz band enables sensing elements that emit radiation in that range. The ability to more effectively generate and sense Terahertz waves is an ongoing technical challenge to developing products for these applications. Transistors that operate in the Terahertz frequency make that possible.

Furthermore, computer performance is currently lacking improved performance. Because of the heat produced by running at the higher frequencies needed for improved performance, traditional silicon computer chips long ago reached their limit in the Gigahertz range (1,000 times slower than Terahertz). Therefore, there is a need for improved transistors capable of efficiently operating at higher frequencies and/or lower power.

SUMMARY

The described embodiments alleviate the problems of the prior art and provide improved transistors and other applications capable of efficiently operating at higher frequencies and/or lower power.

In one aspect a disclosed superconducting Meissner effect transistor (MET) includes a superconducting bridge between a first and a second current probe, the first and second current probe being electrically connected to a source and a drain electrical connection, respectively. In another aspect a control line is configured to emit a magnetic field signal having signal strength H_sig at a superconducting bridge, wherein the emitted magnetic field is configured to break Cooper pairs in a superconducting bridge. In another aspect breaking Cooper pairs in a superconducting bridge decreases conductivity of a superconducting bridge. And in yet another aspect of the disclosed examples, a magnetic field bias is provided, where a strength of the magnetic field bias at a superconducting bridge (H_a) is less than or equal to a critical field value (H_c) for the superconducting bridge.

In one aspect of the disclosure H_a is less than or equal to the magnitude of H_sig subtracted from H_c. In another aspect of the disclosure, H_sig is a time varying magnetic field having a maximum strength of H_sig-max and a minimum strength of H_sig-min. In another aspect of the disclosure H_a is less than or equal to the magnitude of H_sig subtracted from H_c. And in yet another aspect of the disclosure, a superconducting bridge is formed including a type I superconductor. In one particular aspect of the disclosure, a superconducting bridge is formed including a type II superconductor and H_c is equal to H_c1 and H_c1 is a strength of the magnetic field at the superconducting bridge at an onset of a mixed state of superconductivity for the superconducting bridge.

In one aspect of the disclosure, a superconducting bridge has a temperature that is less than or equal to its critical temperature. In another aspect of the disclosure, a superconducting bridge includes niobium. In yet another aspect a superconducting bridge is a temperature that is less than or equal to its critical temperature (T_c) and greater than or equal to about 0.2K, is less than or equal to its critical temperature (T_c) and greater than or equal to about 2.2K, is less than or equal to its critical temperature (T_c) and greater than or equal to about 3K, or is less than or equal to its critical temperature (T_c) and greater than or equal to about 5K.

In one aspect of the disclosure, a MET has a frequency response about equal to the recombination of Cooper pairs for the superconducting bridge material being utilized. In another aspect of the disclosure a MET has a frequency response between about 0.7 THz and about 1.25 THz. In another aspect of the disclosure a logic gat includes one or more MET transistors. In yet another aspect of the disclosure, a first disclosed transistor is configured to emit a photon and a second disclosed transistor is configured to vary its conductivity based on the emitted photon.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic of the operating principle of a FET.

FIG. 2 is a schematic of an example MET, in accordance with embodiments described herein.

FIG. 3 is a schematic of a bridge coordinate system, in accordance with embodiments described herein.

FIG. 4 is a schematic of an example portion of a MET, in accordance with embodiments described herein and describes an example MET geometry: A time varying magnetic field (H_a) applied at location G modulates the conductivity of a superconducting bridge with width W, length L, and thickness δ; D and S refer to the drain and source connections.

FIG. 5 is a schematic of an equivalent circuit, in accordance with embodiments described herein.

FIG. 6 is a graph of example source drain characteristics, in accordance with embodiments described herein.

FIG. 7 is a graph of example MET magnetic drain resistance as a function of applied field, in accordance with embodiments described herein.

FIG. 8 is a graph of example MET amplification factor versus H_a/H_c for various values of V_ds, in accordance with embodiments described herein.

FIGS. 9A-9B include a schematic and an example implementation of MET amplifier circuit, in accordance with embodiments described herein.

FIG. 10 is a schematic of an equivalent MET amplifier implemented in a waveguide mount circuit, in accordance with embodiments described herein.

FIGS. 11A-C are schematics of MET configurations, in accordance with embodiments described herein.

FIG. 12 is a schematic of example MET operation of a MET oscillator and/or power detector, in accordance with embodiments described herein.

FIG. 13 is a graph of current vs. voltage and showing a negative resistance region, in accordance with embodiments described herein.

FIG. 14 is a schematic of a MET logic gate, in accordance with embodiments described herein.

FIG. 15 is a schematic of a MET logic gate, in accordance with embodiments described herein.

DETAILED DESCRIPTION OF THE INVENTION

A standard Field Effect Transistor (FET) 100 has 3 ports formed by metallic electrodes 102 deposited on a semiconductor substrate 104 (see FIG. 1). These ports 3 are the gate (G), source (S), and drain (D). During fabrication impurities are added to the substrate 104 to create a p or n-type conducting channel 106 between S and D. The application of a DC voltage across S and D, V_DS, establishes an electric field {right arrow over (E)}_DS and allows current to flow between the two ports D and S. A time varying electric field applied to G modulates the conductivity in the underlying channel exponentially; i.e. small changes in the time varying gate voltage can cause large variations in the channel conductivity σ. Since the current density {right arrow over (J)} in the channel depends directly on the product of the channel's conductivity and the relatively large value of {right arrow over (E)}_DS set by V_DS, the time varying input signal on G appears as an amplified signal on S. The optimum operating point for the FET is obtained by applying the appropriate DC voltages to each port.

Disclosed embodiments, discussed herein collectively as various Meissner Effect Transistors (MET), are analogous only in concept to an FET, but differ in both theory of operation and construction. In one example disclosed MET 200, with reference to FIG. 2, the “channel” is a thin superconducting bridge 202 and the gate electrode G is replaced by an oscillating magnetic field 204 that modulates the bridge's (202) conductivity also in a nonlinear manner. The greatest gain (or sensitivity) occurs when the bridge 202 is magnetically or thermally biased near the point of going normal. The Operating principle of an example MET: A time varying magnetic field 204 applied at the gate electrode G modulates the conductivity of the superconducting bridge 202.

MET Superconductor Theory of Operation

Superconductors are made by forming Cooper pairs between electrons. The electrons in a Cooper pair have opposite (antiparallel) spins. Magnetic fields work to align the electron spins. Therefore, as a material makes the transition from a normal to superconducting state, it excludes magnetic fields from its interior—the Meissner effect. Similar to the skin depth effect of an electric field on the surface of a conductor, the strength of the magnetic field decays exponentially as it enters the superconductor. The London equation relates the curl of the current density {right arrow over (J)} to the magnetic field:

$\begin{array}{cc}\nabla \times \overrightarrow{J}=-\frac{1}{{\mu }_0{\lambda }_L^2}\overrightarrow{B}& \left(1\right)\\ \mathrm{where}{\lambda }_L={\left(\frac{{\varepsilon }_0mc^2}{ne^2}\right)}^{1/2}& \left(2\right)\end{array}$

λ_L is the London penetration depth, the distance required for the incident magnetic field to decay by 1/e. n is the superconducting electron density, ε₀ is the material permittivity, m is the atomic mass, e is the charge of an electron.

The coherence length ξ₀ of a superconductor is the binding distance between Copper pairs and (to first order) marks the thickness over which the transition between a superconducting and nonsuperconducting state can be made. ξ₀ is inversely related to the superconducting bandgap energy E_g and critical temperature T_c of the material. The higher the T_c, the more tightly bound are the electrons in a Cooper pair.

In a Type I (where ξ₀>λ) superconductor magnetic fields applied parallel to the surface are excluded completely up to a critical field value of H_c. Above this value the field has sufficient strength to parallelize electron spin vectors, thereby breaking Cooper pairs and driving the material out of the superconducting state. Fields applied perpendicular to a Type I superconductor can have effective internal values, H_c1, far higher than the applied field. The extent of the field enhancement is a strong function of the shape of the superconductor, with values ranging from 1.5 for a sphere to orders of magnitude for a thin plate with a magnetic field normal to the surface. For fields between H_c1 and H_c the superconductor is in an “intermediate state,” with alternating superconducting (S) and normal (N) domains. Magnetic fields pass through the N domains and are excluded (by the Meissner effect) from the S domains. As the applied field approaches H_c, a higher percentage of the material is in the N domain, until above H_c no material is left in the S domain. See additional discussion, for example at page 101 in Kadin, A., 1999, Introduction to Superconducting Circuits, pub. John Wiley & Sons (New York) (“Kadin”), which is incorporated by reference herein in its entirety.

The value of H_c is a function of temperature.

$\begin{array}{cc}{H}_c=H_c\left(0\right)\left[1-{\left(\frac{T}{T_c}\right)}^2\right]& \left(3a\right)\end{array}$

Type I superconductors have only one H_c. Type II superconductors (where ξ₀<λ) have two critical field values: H_c1, at the onset of a “mixed state” of superconductivity, and H_c2, where the B-field is strong enough to destroy all superconductivity.

$\begin{array}{cc}{H}_{c1}=\left(\frac{H_c}{\sqrt{2}\kappa }\right)\ln \left(\kappa \right)& \left(3b\right)\\ H_{c2}=\sqrt{2}\kappa H_c\mathrm{where}\kappa ={\lambda }_L/{\xi }_0.& \left(3c\right)\end{array}$

The mixed state of superconductivity is different than the intermediate state described above. In a Type II superconductor the surface energy is negative and favors breaking up of the S and N domains into microscopic dimensions. The magnetic field penetrates the superconductor in the form of magnetized vortices and the N domains take the form of tubes of radius ξ₀. Each vortex contains one flux quantum ϕ₀=h/2e=2.07×10⁻¹⁵ Wb On a macroscopic scale the magnetic field appears to penetrate the entire superconductor (Kadin 1999). However, for applied field values below H_c1, Type II material behaves much like Type I material and the magnetic field is excluded through the Meissner effect.

Modulating the Conductivity of a MET Superconducting Bridge

The formation of Cooper pairs causes the superconducting material to become more ordered, thereby causing its entropy to decrease rapidly as the temperature drops below T_c. The reduction in entropy is reflected in the drop of free energy density F_S in the superconductor as it continues to cool. As discussed above, Cooper pairs can be broken by the application of a magnetic field. The greater the value of the applied magnetic field, B_a, the more Cooper pairs are broken and the more F_S is restored. As the value of B_a increases to the critical value B_c for the superconductor, the electrical conductivity, σ_e, of the material decreases and F_S becomes F_N, the normal free energy density of the material.

σ_e∝(1/F_S)ⁿ, for B_a<B_c   (4)

In a thin film superconducting bridge of thickness δ (FIGS. 3-4) at T=0 K with δ<<λ_L (Kittel 2005),

F _s(x,B_a)=U_s(0)+(δ²−4x²)B_a²/64πλ_L²   (5)

See additional discussion, for example, at p. 295 of Kittel, C., 2005, Introduction to Solid State Physics, pub. John Wiley& Son (USA) (“Kittel”), which is incorporated by reference herein in its entirety.

With reference to FIG. 3, in disclosed embodiments, the x axis is perpendicular to the film plane, x=0 is at the center of the film, B_a is parallel to the film, and U_S (0) is the film internal energy at zero temperature and field.

The magnetic contribution to F_s when averaged over the film thickness is

$\begin{array}{cc}{F}_S=\frac{1}{96\pi }{B_a^2\left(\delta /{\lambda }_L\right)}^2& \left(6\right)\end{array}$

substitution into Eq. (4) yields

$\begin{array}{cc}{\sigma }_e={\left(\frac{96\pi {\lambda }_L^2}{{\delta }^2}\left(\frac{1}{B_a^2}\right)\right)}^n& \left(7\right)\end{array}$

In the thin superconducting bridge the value of B_c is scaled by the factor (λ_L/δ).

$\begin{array}{cc}{B}_c^B=\frac{{\lambda }_L}{\delta }B_c& \left(8\right)\end{array}$

When B_a≈B_c^B the superconducting bridge goes normal and behaves as a resistor. In terms of FET operation this would be analogous to the “pinch-off” condition. Under this condition, σ_e→σ_N a and the expression for the bridge conductivity is

$\begin{array}{cc}{\sigma }_N={\left(\frac{96{\mathrm{\pi \lambda }}_L^2}{{\delta }^2}\left(\frac{1}{{B_c^2\left({\lambda }_L/\delta \right)}^2}\right)\right)}^n& \left(9\right)\end{array}$

Taking the ratio of (6) to (8) and solving for σ^e we find

$\begin{array}{cc}{\sigma }_e={{\sigma }_N\left({\left(\frac{{\lambda }_L}{\delta }\right)}^2{\left(\frac{B_c}{B_a}\right)}^2\right)}^n& \left(10a\right)\end{array}$

Since in MKS units B=μ₀H, the above expression can also be written in terms of the magnetic field intensity (H). Assuming the relationship between σ_e and FS is linear (n=1) in the vicinity of H_c, Eq. (10a) reduces to

$\begin{array}{cc}{\sigma }_e={\sigma }_N{\left(\frac{{\lambda }_L}{\delta }\right)}^2{\left(\frac{H_c}{H_a}\right)}^2& \left(10b\right)\end{array}$

The above expression describes the conductivity in a Type I superconducting bridge with H_a<H_c. By substituting H_c1 for H_c, the above expression can also be used for a Type II superconducting bridge.

Electrical conductivity can have both a real, σ₁, and imaginary, σ₂, component; σ_e=σ₁+jσ₂. As a function of temperature, T, the imaginary component varies as

$\begin{array}{cc}{\sigma }_2\propto \left(1-\frac{T}{T_c}\right),\mathrm{for}T<T_c& \left(11a\right)\\ {\sigma }_2=0,\mathrm{for}T\ge T_c& \left(11b\right)\end{array}$

See, for additional discussion, Ohashi, T., Kitano, H., Maeda, A., Akaike, H., and Fujimaki, A., 2006, “Dynamic fluctuations in the superconductivity of NbN films from microwave conductivity measurements”, Phys. Rev. B, vol 73, 174522, which is incorporated by reference herein in its entirety.

For example embodiments disclosed here, the bridge is electrically, thermally, and/or magnetically biased close to the transition region of the superconductor, therefore we assume σ₂ is negligible.

Example MET Current-Voltage Characteristics

The current density through the bridge J^B is equal to the product of its conductivity σ^e and the value of the electric field E^B across it.

J^B=σ_eE^B   (12)

Assuming the geometry shown in FIG. 4, the drain-source current I_DS in the bridge is given by

$\begin{array}{cc}{l}_{DS}=J^BA={\sigma }_eE^BA={\sigma }_eA\left(\frac{V_{DS}}{L}\right),\mathrm{where},A=\delta W.& \left(13\right)\end{array}$

Substituting (10b) into (13) yields the following expression for I_DS.

$\begin{array}{cc}{I}_{DS}={\sigma }_N{\frac{{\lambda }_L^2W}{\delta L}\left[\frac{H_c}{H_a}\right]}^2V_{DS},\mathrm{where},H_c>H_a>0.& \left(14\right)\end{array}$

In the large field limit (H_a>H_c), σ_e→σ_N and (12) reduces to Ohm's Law,

$\begin{array}{cc}{I}_{DS}={\sigma }_N\left(\frac{A}{L}\right)V_{DS}={\rho }^{-1}\left(\frac{A}{L}\right)V_{DS}=\frac{V_{DS}}{R}& \left(15\right)\end{array}$

where, ρ is the bridge resistivity (Ω−m) and R is the bridge resistance (Ω).

Equation (14) shows that the relationship between I_DS and H_a is parabolic, analogous to the relationship between I_DS and the gate voltage V_G in a conventional FET. This similarity suggests the gain characteristics of an MET will also be analogous to those of a FET.

MET DC Small Signal Model

Because of the applications discussed above, a linear small signal model for an MET is derived using the same approach as is used for a FET, but with improved MET features. For the approach used for a FET, see for example at page 318, Millman, J. and Halkias, C., 1972, Integrated Electronics: Analog and Digital Circuits and Systems, pub. McGraw-Hill (New York), which is incorporated by reference herein in its entirety.

With respect to disclosed METs, from (13) we find that he small signal drain current, i_D, is a function of both the small signal drain voltage, v_DS, and the magnetic field applied to the bridge, H_a.

i _D =f(H_a,v_DS)   (16)

The magnetic drain resistance can be defined as

$\begin{array}{cc}{r_d=\frac{\partial v_{\mathrm{DS}}}{\partial i_D}}_{h_a}=\frac{1}{{\sigma }_N}\frac{\delta L}{W{\lambda }_L^2}{\left(\frac{H_a}{H_c}\right)}^2& \left(17\right)\end{array}$

with units of Ω.

The small signal drain current can then be expressed as

$\begin{array}{cc}{i}_D=\frac{A}{L}h_a+\frac{v_{\mathrm{ds}}}{r_d}& \left(18\right)\end{array}$

Analogous to a FET, a magnetic amplification factor for the MET can be defined as

$\begin{array}{cc}{{\mu }_m=\frac{L}{A}\frac{\partial i_D}{\partial H_a}}_{{\nu }_D}=1+{\sigma }_N\frac{{\lambda }_L^2}{\delta L}\frac{H_c^2}{H_a^3}V_D& \left(19\right)\end{array}$

A non-limiting example MET equivalent circuit 500 with the features of (18) is shown, for example, in FIG. 5.

In a simple example implementation of an example MET discussed here, the input signal, H_sig, is applied directly to the superconducting bridge from the waveguide or control line; there is no gate electrode. Therefore, there is no gate source or drain capacitance to limit the high frequency response. The MET works by modulating the conductivity of the superconducting bridge through applying time varying and static magnetic fields. The superconducting bridge has been biased under a magnetic field separate from the input signal under an applied magnetic field H_a Together these applied fields destroy Cooper pairs decreasing the conductivity of the bridge. The response time of the MET is therefore set by the relaxation time τ₀ to of the Cooper pairs, i.e. the time it takes for them to recombine. To first order τ₀ can be estimated from the uncertainty principle,

Δt≥/ΔE   (20)

where Δt is identified with τ₀ and ΔE with kT_c, the thermal energy of the superconductor at its critical temperature. For additional discussion see, for example, Ramallo, M. V., Carballiera, C., Vina, J., Veira, J. A., Mishonov, T., Pavuna, D., and Vidal, F., 1999, “Relaxation time in Cooper pairs near T_c in cuprate superconductors”, Europhys. Lett, 48 (1), p. 79, which is incorporated by reference herein in its entirety.

$\begin{array}{cc}{\tau }_0\text{∼}\frac{\hslash }{{\mathrm{kT}}_c|}.& \left(21\right)\end{array}$

In one non-limiting example embodiment, the superconductor bridge includes niobium. The bridge material may include any material configured to operate in the Meissner effect regime. That is, in one example embodiment a type I superconductor material at or below its respective critical temperature T_c and in the presence of an applied magnetic field H_a at or below its respective H_c or a type II superconductor at or below its respective critical temperature T_c and in the presence of an applied magnetic field H_a at or below its respective H_c1. For example, niobium (Nb) is a suitable type II superconductor and, for the purposes of the following discussion niobium, will be used. However, it should be understood that other superconductors may also be used. For example, a non-exhaustive list of bridge materials may include lead, tantalum, and alloys thereof including niobium titanium and niobium nitride alloys, as well as high temperature superconductors, such as bismuth strontium calcium copper oxide (BSCCO) and yttrium barium copper oxide (YBCO).

Substituting into the above expression (21) for niobium (T_c=9.5K) gives a value of τ₀=8×10⁻¹³ sec, yielding an upper frequency limit for the corresponding operating MET

v _max≤1/τ₀≤1.25 THz.   (22)

v_max is closely related to the gap frequency of the superconductor v_g. Above v_g incident photons have sufficient energy to break Cooper pairs and drive the superconductor normal.

A superconductor with energy gap E_g has a gap frequency v_g,

$\begin{array}{cc}{v}_g=E_g/h\mathrm{where},E_g\approx \frac{7}{2}kT_c.& \left(23\right)\end{array}$

(for additional discussion see, for example, Karecki, D., Pena, R., and Perkowitz, S., 1982, “Far-infrared transmission of superconducting homogeneous NbN films: Scattering time effects”, Phys. Rev. B., vol. 25, no. 3, p. 1565), which is herein incorporated by reference in its entirety).

For niobium, substitution into Eq. (23) gives a value of v_g≃0.7 THz, consistent with the estimated value for v_max. Since v_g≤v_max, in certain MET embodiments, we will adopt v_g as the maximum operating frequency for a MET.

Example Parameters

In one example embodiment, we want the magnetic field of the input signal to have the maximum effect on the conductivity of the bridge. The strength of the field drops with distance from the input probe, so λ_L should be small, but not so small that only an exceptionally thin layer of the bridge is affected. A compromise is to use a superconducting material where λ_L≈8. we want the magnetic field of the input signal to have the maximum effect on the conductivity of the bridge. As an example, if we assume the bridge is made of Nb. Typical parameters for Nb bridges used in a hot-electron bolometer (“HEB”) work are:

Normal State Conductivity=σ_N=6.93×10⁶ (Ω−m)⁻¹

Critical Field Density at 0 K=B_c(0)=0.198 T

London Penetration Depth=λ_L=100 nm

Coherence Length=ξ₀=40 nm

Bridge Thickness=δ=100 nm

Bridge Width=W=1×10⁻⁴ cm

Bridge Length=L=2×10⁻⁴ cm

Critical Temperature with no magnetic field=T_c=9.5 K

Physical temperature=T=6.5K

Permeability of Free Space=μ₀=4π×10⁻⁷ H/m

Permittivity of Free Space=ε₀=8.85×10⁻¹² F/m

Niobium is an example of an elemental superconductor that is Type II (where ξ<λ_L). Since it is Type II, should be replaced by H_c1 in the above equations. Before substitution, all parameters should be converted into MKS units. Substitution of the parameters for Nb into (12) allows a graphical representation (see FIG. 6) of the common source drain characteristics,

$I_D\left(V_D,\frac{H_c}{H_a}\right),$

of example METs operating in various conditions.

As a comparison, the drain characteristics of a FET have an “ohmic region” where I_D is proportional to V_D and a “constant current region” where I_D rolls off due to an ohmic voltage drop across the channel. This voltage drop causes the channel to “pinch-off” when V_D gets sufficiently high. In FIG. 6 we can see that an MET also has an ohmic region where I_D is proportional to V_D. The bridge of an MET does not experience a pinch-off as does the doped channel of a FET. However, at sufficiently high values of V_D (or for H_a≥H_c) the bridge will go normal, causing the family of curves in FIG. 6 to all break to a common slope. The value of V_D at which the break will occur depends on the geometry of the bridge (13), the temperature of the bridge (15), and the magnitude of the magnetic bias H_a. The drain resistance, r_d is found to be <3.Ω for values of H_a/H_c<1, see for example FIG. 7.

Example METs are capable of providing a high amplification factor μ_m. Substituting the device parameters listed above into (19) yields values of μ_m>10⁹ for V_DS=0.25 mV and H_a/H_c<1. The ratio of conductivity of the bridge in superconducting state to normal state

$\left(\frac{{\sigma }_s}{{\sigma }_n}\right)$

sets the lower limit for amplification factor for a given value of H_a/H_c in the absence of VDS. (See FIG. 8)

Additional MET Operations

In example METs an increase in applied magnetic field H_a causes the conductivity and the current flowing through the superconducting bridge to decrease (see FIG. 13). By increasing applied magnetic field and applying a DV voltage, the MET bridge can be biased into a negative resistance regime, from which it can operate as an amplifier or oscillator. METs can exhibit a “magnetic” negative resistance. This behavior is apparent from FIG. 6, where higher values of H_c/H_a result in higher current levels and from FIG. 8 where lower values of H_a/H_c yield higher amplification factors. As with other negative resistance devices, the MET can be configured into any number of devices, non-limiting examples include an amplifiers, oscillators, or any number of logic gates for which known transistors may be used (see FIGS. 14 and 15).

Example MET Amplifier

An example schematic representation and an example embodiment of a MET amplifier circuit 900 with stabilizing magnetic feedback is shown in FIGS. 9A and 9B, respectively. However, other configurations known to a skilled person may also be used. As ID increases, the voltage drop across R_s increases as does the drive current on the solenoid 902 providing the negative feedback gate magnetic bias, H_fb. Larger values of H_fb reduce the gain of the amplifier (see FIGS. 9A and 9B) and thereby reduces I_D. Similarly, a drop in I_D reduces H_fb, increasing gain and thereby increasing I_D to the desired value. H_sig (also notated as B_sig) is the time varying, input signal to be amplified.

An example configuration for using an MET as an amplifier 1000 is shown in one example configuration in FIG. 10. This example shows the input signal coming in via a control line formed from a waveguide (shown as a horn 1002) coupled to the resonant cavity 1004). Thus, the MET is immersed in the signal waveguide. Other control lines may also be used that produce the emitted magnetic field. For example, the control line may include any radiating structure, for example, a current loop, exposed feed line connected to ground (or, for example, the substrate), a microstrip close to the superconducting bridge, a radiating antenna (such as a planar antenna, dipole, or the like). The control line may be fed by any transmission line known in the art, for example, parallel cable, coaxial cable, or the like in accordance with frequency, current, and loss requirements. The distance of the control line to the superconducting bridge may be any effective distance for the emitted magnetic field to interact with the superconducting bridge. However, a more sensitive MET will require closer distances. The distance, in one example, will be at or greater than the coherence length of the superconductor, i.e. the binding distance of the respective Cooper pairs. The maximum distance will vary depending on the strength of the magnetic field in the control line.

In one example, as shown in FIG. 10, the input and output signals care coupled to a full height, 1λ resonant waveguide cavity 1004 via feedhorns 1002, 1006 and small (≤10 dB) coupling apertures 1008. The dimensions of the resonant cavity 1004, feedhorns 1002, 1006, and coupling apertures α (1008) may be varied and tuned with respect to the particular MET 1010 in use. In one example, the depth of the resonant cavity 1004 (the dimension of the resonant cavity normal to the FIG. 10 paper) is an integer multiple of λ/2. In one example, the height 1012 of the resonant cavity (the dimension of the resonant cavity from one aperture to the other aperture in FIG. 10) is an integer multiple of λ/2. In one example, the width 1014 of the resonant cavity (as shown in FIG. 10 and discussed further below) is an integer multiple of λ/2. With reference to FIGS. 11A-11Cm the MET detail 1100 example shown in FIG. 11A includes a niobium bridge 1102 located at the apex of a “bow-tie” current probe 1104 (FIG. 11A). The superconductor bridge 1100, shown in FIG. 11(B) is oriented such that the width w (FIG. 11A) is substantially parallel to the plane of the page. However, in other example embodiments, the superconductor bridge may be oriented different, for example such that the width w (FIG. 11A) is perpendicular to the plane of the page.

As discussed above, other materials besides Niobium appropriate to a particular desired MET performance may also be used. The current probe 1104 is shown in FIGS. 11A-11C as an Au current probe. However, other highly conductive materials known in the art may also be used. The bridge 1102 and probe 1104 are fabricated on a thin (δ<<λ) dielectric substrate, for example, a substrate 1106 comprising crystalline quartz and/or silicon oxide, or any other known substrate material which is suspended across the waveguide. To either side of the probe 1104 (FIG. 11B) are a series of high-low microstrip sections 1108 which serve as an radiofrequency (RF)-choke to keep the time varying field within the cavity while providing a path for the DC bias current. In one amplifier configuration, unlike the situation with typical voltage-controlled devices, the MET should be located at the current (not voltage) maximum. The field geometry within a 1λ cavity will put a current maximum along the MET when the device is located across the cavity center. In order to magnetically bias the MET, the waveguide mount is immersed in a field, H_a , specified by the devices drain characteristics, see for example, FIG. 6.

Example MET Oscillator

For the MET to operate as an amplifier the value of H_a will often approach H_c. However, there are many applications where a signal source may be required. For these applications, the negative feedback provided by the ambient magnetic field is reduced, in one example configuration, until the MET operates as a negative resistance oscillator. For these applications the resonant cavity (e.g. 1004 of FIG. 10) is sealed at one end and only an output feedhorn 1006 is required.

As discussed above, when operated near the normal-superconducting transition, the impedance of a MET will have little, if any reactive component (see Eq. 10b). Therefore the oscillator frequency will be set largely by the physical dimensions of the waveguide cavity. The size of the cavity determines the time delay/phase per reflection and therefore the cavity resonant frequency. For a cavity of length z the round-trip time is

$t=\frac{2z}{c}.$

The oscillator can operate at frequencies

$f=\frac{cm}{2z},$

where m is a multiple of

$\frac{\lambda }{2}.$

The upper frequency limit, v_max, is given by Eq. (22).

The maximum power P_out from the oscillator is a function of the drain current, I_D, and the MET drain resistance, r_d.

P_out=I_D²r_d   (24)

For a value of H_a/H_c=0.8, FIG. 7 indicates a value of r_d≈1.7Ω. Similarly, for an H_c/H_a=1.25 and a drain voltage, V_D, of 1.0 mV, FIG. 6 gives a value of I_D≈2 mA. Substitution into (24) yields a value of P_out≈1.7 μW. This power level is sufficient to serve as a local oscillator for superconductor-insulator-superconductor (“SIS”) or HEB mixers. These mixers operate at physical temperatures comparable to that of the MET. Therefore, in one configuration an MET oscillator and/or amplifier can be integrated into an SIS or HEB mixer block. FIG. 12 illustrates the operation of one embodiment of an MET oscillator. A super conducting bridge, in one example configuration can be used as an oscillator 1202, mixer, and/or as a detector 1204.

Advantages of a MET as Compared to a Flux Flow Transistor

A superconducting flux flow transistor (SFFT) also works by modulating the conductivity of a superconducting bridge by applying a time varying magnetic field. However, the conductivity in a SFFT is controlled by injecting magnetic vortices into a Type II superconductor operating in the mixed state. Thus it does not operate in a Meissner effect regime. See, for example, Nordman, J., 1995, “Superconductive amplifying devices using fluxon dynamics”, Supercond. Sci. Technol., 8, p. 881, which is incorporated by reference herein in its entirety. In contrast, the conductivity in a MET is controlled by way of the Meissner effect operating on a Type I superconductor where H_a<H_c or on a Type II with H_a<H_c1. For a SFFT the upper operating frequency is limited by the vortex traversal time to frequencies ≤1 GHz (See discussion in Kadin). In contrast, the MET does not rely on vortex diffusion. As discussed above, its upper frequency limit is determined by the Cooper pair relaxation time and is of order 1 THz for commonly used superconductors (e.g. Nb). The MET is therefore better suited for high frequency applications.

We have introduced the concept of the Meissner Effect Transistor and presented its theory of operation. We have also derived performance curves for a MET that utilizes a niobium bridge typical of those used in superconducting devices. The application of the MET in both a THz amplifier and oscillator were described as just several examples of uses of a MET. Others will be apparent to a person of ordinary skill after reading this disclosure. The fast switching speed and low power dissipation of the MET make it a candidate for high speed computer applications.

Example MET Gates

With reference to FIGS. 14 and 15, MET's, as described throughout this disclosure, can be configured to create any classic logic gate known in the art. Shown in FIG. 14 is an example “OR” gate 1400. In addition to classical logic gates, METs, in one example, can be configured to as “multi-state” logic gates as shown, for example, in FIG. 15. Photons 1502 generated in METs 1504 can, in one example, entangle the “conducting states” of neighboring devices 1504 providing for multiple states.

Appendix I. Input Signal Power Level

An MET amplifier is sensitive to the time variations in the magnetic field component of the signal to be amplified. The relationship between the electric and magnetic field components of an electromagnetic wave is provided by the Poynting vector. (Griffiths, D., 1999, Introduction to Electrodynamics, pub. Prentice Hall (New Jersey))

The energy flux density transported by the electric and magnetic fields is

$\begin{array}{cc}\overrightarrow{S}=\frac{1}{{\mu }_0}\left(\overrightarrow{E}\times \overrightarrow{B}\right).& \left(A\text{-}1\right)\end{array}$

The time average over many cycles is

$\begin{array}{cc}〈\overrightarrow{S}〉=\frac{1}{2}c{\varepsilon }_0E_G^2\hat{z}& \left(A\text{-}2\right)\end{array}$

The average power per unit area (W/m²) of an EM wave is called the intensity (or flux).

F=1/2cε₀E_G²   (A-3)

Rearranging,

$\begin{array}{cc}{E}_G={\left[\frac{2F}{c{\varepsilon }_0}\right]}^{1/2},B_G={\left[\frac{2F}{c^3{\varepsilon }_0}\right]}^{1/2}& \left(A\text{-}4\right)\end{array}$

E_G and B_G are the amplitudes or peak values of the fields. If rms values are used, then the factor of ½ in the Poynting vector is dropped.Equation (A-4) then becomes

$E_G^{\mathrm{rms}}={\left[\frac{F}{c{\varepsilon }_0}\right]}^{1/2}\mathrm{and}B_G^{\mathrm{rms}}={\left[\frac{F}{c^3{\varepsilon }_0}\right]}^{1/2}.$

Claims

1. A superconducting Meissner effect transistor comprising: a superconducting bridge between a first and a second current probe, the first and second current probe being electrically connected to a source and a drain electrical connection, respectively, wherein the superconducting bridge is adapted to include a plurality of Cooper pairs having anti-parallel spins when the superconducting bridge is subjected to a magnetic field bias; anda control line connected to a magnetic field signal generator, wherein the magnetic field signal generator is adapted to generate a magnetic field signal emitted from the control line, the magnetic field signal having a signal strength Hsig at the superconducting bridge sufficient to break at least a portion of the Cooper pairs in the superconducting bridge into forced parallel spins.

2. The transistor of claim 1, wherein the superconducting bridge is adapted such that breaking Cooper pairs in the superconducting bridge decreases conductivity of the superconducting bridge.

3. The transistor of claim 1, wherein the superconducting bridge is adapted to include the plurality of Cooper pairs having anti-parallel spins when the superconducting bridge is subjected to the magnetic field bias and a strength of the magnetic field bias at the superconducting bridge (Ha) is less than or equal to a critical field value (Hc) for the superconducting bridge.

4. The transistor of claim 3, wherein Ha is less than or equal to the magnitude of Hsig subtracted from Hc.

5. The transistor of claim 3, wherein Hsig is a time varying magnetic field having a maximum strength of Hsig-max and a minimum strength of Hsig-min.

6. The transistor of claim 5, herein Ha is less than or equal to the magnitude of Hsig-max subtracted from Hc.

7. The transistor of claim 3, wherein the superconducting bridge is a type I superconductor.

8. The transistor of claim 3, wherein the superconducting bridge is a type II superconductor and Hc is equal to Hc1 and Hc1 is a strength of the magnetic field at the superconducting bridge at an onset of a mixed state of superconductivity for the superconducting bridge.

9. The transistor of claim 1, wherein the superconducting bridge has a temperature that is less than or equal to its critical temperature (Tc).

10. The transistor of claim 1, wherein the superconducting bridge comprises niobium and/or niobium alloys.

11. The transistor of claim 10, wherein the superconducting bridge has a temperature that is less than or equal to its critical temperature (Tc) and greater than or equal to about 0.2K.

12. The transistor of claim 9, wherein the superconducting bridge has a temperature that is less than or equal to its critical temperature (Tc) and greater than or equal to about 2.2K.

13. The transistor of claim 9, wherein the superconducting bridge has a temperature that is less than or equal to its critical temperature (Tc) and greater than or equal to about 3K.

14. (canceled)

15. The transistor of claim 1 having a frequency response about equal to the recombination of Cooper pairs for the superconducting bridge material being utilized.

16. The transistor of claim 10, wherein the frequency response is between about 0.7 THz and about 1.25 THz.

17. A logic gate comprised using one or more transistors of claim 1.

18. A first transistor of claim 1 and a second transistor of claim 1, wherein the first transistor is configured to emit a photon in the presence of the magnetic field bias and the magnetic field signal, and the second transistor is configured to vary its conductivity based on the emitted photon.

19. (canceled)

20. (canceled)

21. The first and second transistors of claim 18, wherein the first and second transistors are configured such that a quantum state between the first and second transistors is entangled when the first transistor is in the presence of the magnetic field bias and the magnetic field signal.

22. The first and second transistors of claim 21, where the first and second transistors form a logic gate when the first transistor is in the presence of the magnetic field bias and the magnetic field signal.

23. The transistor of claim 4, wherein the magnitude of Ha and Hsig, combined, is equal to Hc and the superconductor bridge is normal.