Title: High sensitivity, directional dc-SQUID magnetometer
Abstract: A solid state dc-SQUID includes a superconducting loop containing a plurality of Josephson junctions, wherein an intrinsic phase shift is accumulated through the loop. In an embodiment of the invention, the current-phase response of the dc-SQUID sits in a linear regime where directional sensitivity to flux through the loop occurs. Changes in the flux passing through the superconducting loop stimulates current which can be quantified, thus providing a means of measuring the magnetic field. Given the linear and directional response regime of the embodied device, an inherent current to phase sensitivity is achieved that would otherwise be unobtainable in common dc-SQUID devices without extrinsic intervention.
Patent Number: 6,905,887 Issued on 06/14/2005 to Amin, et al.
| Inventors:
|
Amin; Mohammad H. S. (Vancouver, CA);
Duty; Timothy (Vancouver, CA);
Omelyanchouk; Alexander (Kharkov, UA);
Rose; Geordie (Vancouver, CA);
Zagoskin; Alexandre (Vancouver, CA);
Hilton; Jeremy P. (Vancouver, CA)
|
| Assignee:
|
D-Wave Systems, Inc. (Vancouver, CA)
|
| Appl. No.:
|
192623 |
| Filed:
|
July 9, 2002 |
| Current U.S. Class: |
438/3; 505/874; 505/874 |
| Intern'l Class: |
H01L 021/00 |
| Field of Search: |
438/3,2
505/1,874
257/9
|
References Cited [Referenced By]
U.S. Patent Documents
| 5157466 | Oct., 1992 | Char et al.
| |
| 5767043 | Jun., 1998 | Cantor et al.
| |
| 5880069 | Mar., 1999 | Nakao et al.
| |
| 5962866 | Oct., 1999 | DiIorio et al.
| |
| 6476413 | Nov., 2002 | Jia et al.
| |
| 6624122 | Sep., 2003 | Holesinger et al.
| |
| Foreign Patent Documents |
| WO02/0694/11 | Sep., 2002 | WO.
| |
Other References
D. Dimos, P. Chaudhari, J. Mannhart, and F.K. LeGoues, "Oreintation Dependence
of Grain-Boundary Critical Currents in YBa2Cu3O7 Bicrystals" Physcal Review Letters,
vol. 61, No. 2, pp 219-224.
A. Di Chiara et al., "YBa2Cu3O7-x grain boundary Josephson junctions with a MgO
seed layer", IEEE Transactions on Applied Superconductivity, vol. 7, No. 2, Jun.
1997, pp 3327-3330.
C. Bruder, A. van Otterlo, and G. T. Zimanyi, "Tunnel junctions of Unconventional
Superconductors", Physical Review B 51, p. 12904 (1995).
H. Hilgenkamp, J. Mannhart, and B. Mayer, "Implications of dx2-y2 symmetry and
faceting for the transport properties of grain boundaries in high-TC
superconductors", Physical Review B 53, pp. 14586-14593 (1996).
E. ll'ichev, V. Zakosarenko, R. P. J. Ijsselsteijn, V. Schultze, H.-G. Meyer,
H. E. Hoenig, H. Hilgenkamp, and J. Mannhart, "Nonsinusoidal Current-Phase Relationship
of Grain Boundary Josephson junctions in High-Tc Superconductors", Physical Review
Letters 81, pp. 894-897 (1998).
E. ll'ichev, M. Grajcar, R. Hlubina, R.P.J. IJsselsteijn, H.E. Hoenig, H.G. Meyer,
A. Golubov, M.H.S. Amin, A.M. Zagoskin, A.N. Omelyanchouk, and M.Yu. Kupriyanov,
"Degenerate ground state in a mesoscopic YBa2Cu3O7-x
grain boundary Josephson junction", Los Alamos National Laboratory preprint server
cond-mat/0102404 v2, (2001), last accessed Dec. 18, 2002.
S. Nicoletti, H. Moriceau, J.C. Villegier, D. Chateigner, B. Bourgeaux, C. Cabanel,
and J. Y. Laval, "Bi-epitaxial YBCO grain boundary Josephson junctions on SrTiO3
and sapphire substrates", Physica C 269, pp. 255-267 (1996).
R.R. Schulz B. Chesca, B. Goetz, C.W. Schneider, A. Schmehi, H. Bielefeldt, H.
Hilgenkamp, J. Mannhart, and C.C. Tsuei, "Design and realization of an all d-wave
dc-superconducting quantum interference device", Applied Physics Letters 76, pp.
912-914 (2000).
F. Tafuri, F. Carillo, F. Lombardi, F. Miletto Granozio, F. Ricci, U. Scotti
di Uccio, A. Barone, G. Testa, E. Sarnelli, J.R. Kirtley, "Feasibility of Biepitaxial
YbaCuO Josephson Junctions for Fundamental Studies and Potential Circuit Implementation",
Los Alamos National Laboratory preprint server condmat/0010128 (2000), accepted
for publication Phys. Rev. B, last accessed Dec. 18, 2002.
Y. Tanaka and S. Kashiwaya, "Theory of Josephson effects in anisotropic superconductor",
Physical Review B 56, pp. 892-912 (1997).
C. C. Tsuei and J. R. Kirtley, "Pairing symmetry in cuprate superconductors",
Reviews of Modern Physics 72, Issue 4, pp. 969-1016 (2000).
D. J. Van Harlingen , "Phase-sensitive tests of the symmetry of the pairing state
in the high-temperature superconductors-Evidence for dx2-y2
symmetry", Reviews of Modern Physics 67, 515-535 (1995).
D. Koelle, R. Kleiner, F. Ludwig, E. Dantsker and John Clarke, "High-transition-temperature
superconducting quantum interference devices", Reviews of Modern Physics 71, pp.
631-686 (1999).
H. Hilgenkamp, "Josephson devices with intrinsic π-phase shifts" abstract
from "Structure and Arrangement of Vortices in Superconductors", Conference Prague
Czech Republic, Apr. 3-7, 2002.
|
Primary Examiner: Coleman; W. David
Attorney, Agent or Firm: Jones Day, Lovejoy; Brett
Parent Case Text
CROSS REFERENCE TO RELATED APPLICATION
This application is a divisional of U.S. patent application Ser. No. 09/823,895,
filed Mar. 31, 2000 now U.S. Pat. No. 6,627,916 , entitled "High sensitivity, Directional
DC-Squid Magnetometer".
Claims
1. A method of producing a SQUID magnetometer, comprising:
depositing a seed layer on a substrate;
depositing a first buffer layer on the seed layer;
etching the seed layer and the first buffer layer to form a boundary separating
(i) a first portion having the seed layer and the first buffer and, (ii) a second
portion where the seed layer and the first buffer are substantially removed, wherein
the boundary includes a first part and a second part and wherein the first part
of the boundary is not collinear with the second part of the boundary;
depositing a second buffer layer on the second portion;
depositing a superconducting material on the first buffer layer and the second
buffer layer; and
etching the superconducting material to form a SQUID loop across said boundary
thereby forming said SQUID magnetometer, the SQUID loop comprising a plurality
of Josephson junctions, wherein
a first Josephson junction in the plurality of Josephson junctions is defined
by the first part of the boundary, and
a second Josephson junction in the plurality of Josephson junctions is defined
by the second part of the boundary.
2. The method of claim 1, wherein the substrate is of sapphire.
3. The method of claim 1, wherein the substrate is of SrTiO
3.
4. The method of claim 1, wherein the seed layer is of MgO.
5. The method of claim 1, wherein the first buffer layer is of CeO
2.
6. The method of claim 1, wherein the second buffer layer is of CeO
2.
7. The method of claim 1, wherein the superconducting material is YBa
2Cu
3O
7-x
where x is greater than or equal to 0 and less than about 0.6.
8. The method of claim 1, wherein the superconducting material is Bi
2Sr
2Ca
n-1Cu
nO
2n-4,
where n is an integer.
9. The method of claim 1, wherein the superconducting material is Sr
2RuO
4.
10. The method of claim 1, wherein depositing the seed layer is performed bi-epitaxially.
11. The method of claim 1, wherein depositing the first buffer layer is performed bi-epitaxially.
12. The method of claim 1, wherein depositing the second layer is performed bi-epitaxially.
13. The method of claim 1, wherein depositing the superconducting material is
performed bi-epitaxially.
14. The method of claim 1, wherein etching the seed layer and the first buffer
layer is accomplished with a milling process.
15. The method of claim 1, wherein etching the superconducting material is accomplished
with a milling process.
16. The method of claim 1 wherein a superconducting current in the SQUID loop
experiences an accumulated intrinsic phase shift over the path of the SQUID loop.
17. The method of claim 1 wherein,
the first part of the boundary is perpendicular to a current path in the SQUID
loop, and
the second part of the boundary is not perpendicular to a current path in the
SQUID loop.
18. The method of claim 17 wherein the first Josephson is a π/2-phase shift
grain boundary Josephson junction and the second Josephson junction is a 0-junction.
Description
BACKGROUND
1. Field of the Invention
This invention relates to dc-SQUID magnetometry and superconducting electronics
and, in particular, to a magnetometer including a superconducting SQUID having
an inherent phase shift without application of external magnetic fields.
2. Discussion of Related Art
Very precise measurements of small magnetic fields can be accomplished with
a dc-SQUID magnetometer device. A conventional dc-SQUID magnetometer includes a
superconducting loop containing a plurality of Josephson junctions, coupled to
terminals. Any change in the magnetic field which penetrates the superconducting
loop disturbs the current through the device, which is detectable at the terminals.
Thus, the dc-SQUID can be used as a device for measuring changes in a magnetic field.
Conventional dc-SQUID magnetometers lack inherent sensitivity. Furthermore,
a conventional dc-SQUID magnetometer can only determine the magnitude of the change
in a magnetic field, but cannot distinguish the direction of the change. In order
to hyper-sensitize a standard dc-SQUID, flux-biasing can be used to shift the latent
flux position in the SQUID loop into a linear response regime. A standard dc-SQUID
loop behaves in accordance with a well defined current-phase relationship. The
equilibrium position of the current-phase relation of a standard dc-SQUID lies
in a region of sensitivity where the induced superconducting current is proportional
to a small perturbation in the flux squared (I∝Φ
2), and
resultingly there is no directional sensitivity. By biasing the SQUID loop with
an applied flux, the equilibrium position can be shifted into a more sensitive
linear response regime, thus introducing directional sensitivity into the current
response as well. This can be accomplished by introducing a phase shift of the
equilibrium position in the current-phase relation. The phase shift is realized
in conventional dc-SQUID devices by application of an external magnetic field to
the dc-SQUID device, a technique called flux biasing. In other words, with an externally
applied magnetic flux on the SQUID device, a small perturbation in the flux induced
by the magnetic field that is being measured will result in a linear response in
the superconducting current from the SQUID device.
Furthermore, by coupling multiple SQUID loops, it is possible to enhance
the sensitivity of the dc-SQUID magnetometer. See U.S. Pat. No. 5,767,043, entitled
"Multiple Squid Direct Signal Injection Device Formed On a Single Layer Substrate,"
to Cantor et al., herein incorporated by reference in its entirety. One application
of dc-SQUID magnetometry is as a non-destructive testing device in the field of
semi-conductor electronics. In the electronics industry, each circuit that is manufactured
must be non-destructively tested for correct operating parameters. This is accomplished
by running current through the circuit to be tested and measuring the resulting
magnetic fields. However, in order to detect flaws in the magnetic field a high
degree of resolution is required, which cannot be achieved without flux biasing
or coupling the dc-SQUIDs that make up the magnetometer. Thus, there is a necessity
for increasing the latent sensitivity of the SQUID magnetometer.
Further applications for dc-SQUID magnetometers range in practical uses.
For example, dc-SQUID magnetometers are used in Magnetic Resonance Imaging, microscopic
metal defect detection, mine detection, and submarine detection. Additional examples
of uses for dc-SQUID magnetometers include analogue-to-digital converters and optical
switches. Given the broad range of applications of dc-SQUID Magnetometers, there
is a need for devices with increased sensitivity, including directional sensitivity,
wherein the overall size and cost of a device is reduced.
There are, however, practical limitations to current methods of dc-SQUID sensitizing.
Biasing the loop introduces magnetic fields that may interfere with the fields
or system being measured. Similarly, coupling dc-SQUIDs can lead to bulky measurement
tools that increase the obtainable distance from the sample, thereby also decreasing
the ability to measure magnetic fields in the sample.
The use of a phase shifter in order to sensitize the current-phase behavior in
a superconducting loop is known; however, the inherent sensitization has been restricted
to a π-phase shift. Thus there is a need for a device that can be used in
dc-SQUID magnetometry with a high level of latent sensitivity, as well as directional
sensitivity without the application of external magnetic fields.
SUMMARY
In accordance with the present invention, a dc-SQUID magnetometer is presented
which provides an inherent phase shift in a superconducting loop, i.e. a phase
shift in the absence of an external magnetic field. Some embodiments of a dc-SQUID
magnetometer according to the present invention include a high sensitivity, directional,
superconducting Josephson device formed of a superconductive loop having a π/2-Josephson
junction and a 0-Josephson junction. The superconductive loop is further coupled
to at least two terminals by which a current may flow through the loop.
The superconducting materials forming the superconducting loop and terminals
can have dominant order pairing symmetry with non-zero angular momentum. In some
embodiments, the superconducting material can be a high temperature, d-wave superconductor
such as YBa
2Cu
3O
7-x, where x has values less than
0.4 and greater than 0.05, or Bi
2Sr
2Ca
n-1Cu
nO
2n+4.
In some other embodiments, a dc-SQUID magnetometer according to the present invention
can include a p-wave superconducting material forming 0-junctions and π/2-junctions.
An example of a p-wave superconducting material includes Sr
2RuO
4.
Junctions having a π/2 phase shift or a 0 phase shift, for example,
can be fabricated at the grain boundary of two d-wave superconducting materials.
For example, in a junction formed at the grain boundary between two d-wave superconducting
materials with a 45° misalignment in their crystal lattice structures, a π/2
phase shift results in a junction that is perpendicular to the terminals of the
junction. Similarly, a 0° phase shift can be achieved in a grain boundary
Josephson junction in which the misalignment in the crystal orientation between
the superconductors on either side of the grain boundary is zero (in the trivial
case) or, the 0° phase shift can be achieved in the case of a symmetric 22.5°
grain boundary junction, where the a-axis of the order parameter of the two superconductors
are rotated ±22.5° from parallel to the junction interface, respectively.
The combination of a 0-junction and a π/2-junction induces an overall π/2-phase
shift in the current as the superconducting loop is traversed, thus shifting the
equilibrium position of the current-phase relation. Resultingly, a π/2 dc-SQUID
loop according to the present invention has a linear current-phase response with
small changes in externally applied magnetic flux. The measured current is also
sensitive to the direction of the flux through the loop. Further, the π/2
dc-SQUID loop does not require any externally applied flux biasing. This inherent
phase shift allows for an order of 100 fold increase over the sensitivity of standard
embodiments of dc-SQUID loops without the use of external means. Additionally,
no external circuitry is required to bias the SQUID loop.
An embodiment of a SQUID magnetometer according to the present invention can
be
fabricated by bi-epitaxial methods, although other deposition methods can also
be utilized. For example, in the fabrication of a d-wave superconducting SQUID
magnetometer according to the present invention, a seed layer may be deposited
on a substrate and a first buffer layer may be deposited on the seed layer. In
some embodiments, the seed layer may be MgO, the substrate SrTiO
3 or
Sapphire, and the first buffer layer CeO
2. The first buffer layer and
the seed layer may be etched, for example by Xe-ion milling although any appropriate
etching method can be used, to form a boundary. The boundary separates a first
area having the seed layer and the first buffer layer from a second area where
the seed layer and the first buffer layer have been removed. A second buffer layer
can be deposited on the second area and the first buffer layer. A superconducting
material may then be deposited on the second buffer layer and etched, for example
by Xe-ion milling although any appropriate etching method can be used, to form
a loop and terminals. Resultingly, Josephson junctions are formed along the boundary.
The boundary can be shaped so that both a 0-junction and a π/2 junction are
achieved at the boundary. The second buffer layer may also be of CeO
2.
The first buffer layer and the second buffer layer are deposited so that there
is a lattice mismatch between the superconducting material deposited over the first
area where the seed layer has been deposited and the superconducting material deposited
over the second area, resulting in a grain boundary in the superconducting material
at the boundary.
A magnetometer according to the present invention may exhibit inherent phase
shifts
of any value. Phase shifts other than a π/2 phase shift can also result in
a magnetometer operating in a linear region. These and other embodiments are further
discussed below with respect to the following figures.
BRIEF DESCRIPTION OF THE FIGURES
FIG. 1 shows a plan view of an embodiment of a SQUID Magnetometer according
to the present invention.
FIG. 2 shows a plan view of an embodiment of a π/2-Josephson junction.
FIG. 3 shows a plan view of a symmetric 22.5° Josephson junction.
FIG. 4 shows a plan view of an embodiment of a 0 Josephson junction.
FIGS. 5
a, 5
b, and 5
c show critical current-flux
relations for a standard 0-phase shift dc-SQUID, a π-phase shift dc-SQUID,
and a π/2-phase shift dc-SQUID, respectively.
FIGS. 6
a through 6
c illustrate an example bi-epitaxial
fabrication method.
FIG. 7 shows a cross sectional view of an exemplary bi-epitaxial fabrication method.
FIG. 8 shows a plan view of a 2-dimensional array of SQUID magnetometers according
to the present invention.
FIG. 9 shows a dc SQUID magnetometer according to the present invention utilized
to measure a magnetic field.
In the figures, elements with the same designation have similar or identical functions.
DETAILED DESCRIPTION
FIG. 1 shows a plan view of an embodiment of a SQUID magnetometer 100
according to the present invention. SQUID magnetometer 100 of FIG. 1 includes
loop 150, junctions 110 and 120, and terminals 140
and 141. Loop 150 and terminals 140 and 141 can be
formed of a d-wave superconducting material or a p-wave superconducting material
deposited on areas 101 and 102. Portion 151 of loop 150
is formed in area 102 and portion 152 of loop 150 is formed
in area 101.
Loop 150 and junctions 110 and 120 provide an intrinsic
phase shift to the current in magnetometer 100. Intrinsic phase shifters
are described in M. H. S. Amin, T. Duty, A. Omelyanchouk, G. Rose and A. Zagoskin,
U.S. Provisional Application Ser. No. 60/257624, "Intrinsic Phase Shifter as an
Element of a Superconducting Phase Quantum Bit", filed Dec. 22, 2000, and the references
therein, which is herein incorporated by reference in its entirety. A phase shifting
structure with 0 and π-phase shifts in a two-terminal DC SQUID is described
in R. R. Schulz, B. Chesca, B. Goetz, C. W. Schneider, A. Schmehl, H. Bielefeldt,
H. Hilgenkamp, J. Mannhart and C. C. Tsuei, "Design and realization of an all d-wave
dc pi-superconducting quantum interference device", Appl. Phys. Lett. 76, 7 p.912
(2000), and the references therein, which is incorporated herein by reference in
its entirety.
Areas 101 and 102 indicate the surface of two misaligned crystal
lattice structures. In some embodiments, the crystal lattice structure of the surface
of area 101 is rotated about 45° with respect to the crystal lattice
structure of the surface of area 102. The areas 101 and 102
form grain boundaries 103 and 104 at their intersection. As a result
of this misalignment, when the superconductor materials of terminals 140
and 141 and of SQUID loop 150 are deposited, the crystal orientation
is determined by that of the material of areas 101 and 102 respectively.
In other words, terminal 140 and portion 151 of loop 150 have
a crystal lattice structure determined by the material of area 102 whereas
terminal 141 and portion 152 of loop 150 have a crystal lattice
structure determined by the material of area 101. Examples of superconducting
materials which may be included in a d-wave superconducting Josephson device 100
are Yba
2Cu
3O
7-x and Bi
2Sr
2Ca
n-1Cu
nO
2n+4,
which both have d-wave order pairing symmetry. An example of a p-wave superconducting
material which can be utilized to form device 100 includes Sr
2RuO
4.
To achieve the π/2-phase shift in the p-wave superconductor case, the grain
boundary rotation angle should be about 45°, and the order parameter should
change to having a vertical alignment on one side of the grain boundary, and a
horizontal alignment on the other side of the grain boundary. Any superconducting
material with a dominant order pairing symmetry having a non-zero angular momentum
can be used to form device 100.
Josephson junctions 110 and 120 are formed at grain boundaries
103 and 104, respectively. In the embodiment shown in FIG. 1, grain
boundary 103 (and hence junction 110) is formed parallel with a horizontal
axis of the magnetometer. A vertical axis can be defined in device 100 along
the directions of terminals 140 and 141 as shown in FIG. 1.
Grain boundary 104, however, is angled with respect to grain boundary 103
by an angle of Θ, such that the resulting junction on the SQUID loop is a
zero or pi-phase shift Josephson junction, which in some embodiments can be a symmetric
22.5° Josephson junction.
The current passing through junction 110 behaves according to the relationship
I=I
c Sin 2 (θ-θ′), where θ is the phase of
the superconducting region 151, and θ′ is the phase in the
superconducting region 152. Furthermore, the current passing through junction
120 behaves according to the relationship I=I
c Sin (ψ-ψ′),
where again, ψ represents the phase of the superconductor in region 151,
and ψ′ represents the phase of the superconductor in region 152.
The total current through the terminals 140 and 141 is just the sum
of the currents through the junctions, which is dependent upon the embodiment of
the invention. Specifically, variation in the width of the junctions or branches,
the width of the overall loop, and the roughness of the junctions are the key factors
involved in calculating the ratio of current in each branch of the loop. In an
exemplary embodiment, the ratio is I
1=2I
2=I
c.
Thus, the total current is given by I=I
c (Sin(φ
2+Φ
e)-½Sin
2φ
2), where φ
2=θ-θ′, φ
e=φ
1-φ
2,
and the dependence of φ
2 on φ
e can be easily
calculated by taking the derivative of I with respect to φ
2 and
finding the maximums and minimums. Furthermore, it is possible to derive from this
the critical current-flux relationship (see FIG. 5
c). In some embodiments,
the angle of symmetry, Θ, can be about π/8 radians, such that junction
120 is a symmetric 22.5°-phase shift Josephson junction.
FIG. 2 shows an example of a π/2-phase shift grain boundary Josephson
junction 110. Junction 110 is formed at grain boundary 103
(FIG. 1) by portion 152 and portion 151 separated by a junction boundary
210. Boundary 210 can be a small gap between the two superconductors,
or it can include an insulating material. As described above, portion 152
and portion 151 are each formed of a superconducting material having a dominant
order pairing symmetry with non-zero angular moment. A lattice mismatch between
the two superconducting materials in regions 151 and 152 introduces
a phase shift in the quantum order parameter ψ as the junction is traversed.
In some embodiments, the a-crystallographic direction of the superconducting material
of portion 151 is rotated by π/4 from the a-crystallographic direction
of the superconducting material of portion 151. The order parameter of the
superconducting material is directly related to the crystallographic orientation.
Therefore, the phase shift of current that has traversed the grain boundary of
FIG. 2, is θ-φ, where θ represents the phase of the supercurrent
before crossing the grain boundary junction, and φ represents the phase incurred
by the supercurrent in crossing the grain boundry junction. In FIG. 2, φ
is π/2, and the overall phase shift of the junction 210 is π/2.
The embodiment of junction 110 shown in FIG. 2 is an asymmetric junction.
In that case, the a-crystallographic axis of portion 152 is rotated from
boundary 212 by about 0° and the a-crystallographic axis of portion
151 is rotated from boundary 211 by about 45°, yielding a total
mismatch between the a-crystallographic axis between the superconducting materials
of portions 151 and 152 of 45° (π/4). The a-crystallographic
axis of the superconducting materials of portions 151 and 152 is
indicated by arrows in FIG. 2. Additionally, the phase θ, and θ′
of the supercurrent is shown in each of portion 151 and 152.
A 0 phase shift grain boundary Josephson junction, embodied here as a symmetric
22.5° grain boundary Josephson junction, 300 is illustrated in FIG.
3. Junction 300 is symmetric because order parameter 149 in
portion 152 is a mirror image of order parameter 149 in portion 151.
The supercurrent I
1 crossing grain boundary 310 from one direction
has a dominant path that is the same regardless of the direction from which junction
300 is approached. Thus, the total phase incurred across grain boundary
310 is 0. Symmetric 22.5° Josephson junctions are further discussed
in E. Il'ichev, M. Grajcar, R. Hlubina, R. P. J. Ijsselsteijn, H. E. Hoenig, H.
G. Meyer, A. Golubov, M. H. S. Amin, A. M. Zagoskin, A. N. Omelyanchouk, and M.
Yu. Kupriyanov, "Degenerate ground state in a mesoscopic YBa
2Cu
3O
7-x
grain boundary Josephson junction", LANL, cond-mat/0102404 v2; 23 Feb., 2001,
and the references therein, which is herein included by reference in its entirety.
FIG. 4 shows an example of a 0-junction 120. Again, the lattice mismatch
between the a-crystallographic directions of the superconducting materials of portions
151 and 152 is the same as the lattice mismatch indicated in the
discussion of FIG. 2 above. The a-crystallographic directions and the superconducting
order parameter ψ and ψ′ are shown in FIG. 4 for each of the
superconducting materials of portions 151 and 152, which shows a
π/2 phase shift in the order parameter ψ′ with respect to the
order parameter orientation of ψ. Portions 151 and 152 are
separated by junction boundary 320. In junction 120 of FIG. 4, however,
boundary 320 is not perpendicular to the direction of the superconducting
current I. Instead, boundary 320 follows the grain boundary 104 and
therefore is angled from the perpendicular direction by the angle of symmetry θ.
The angle of symmetry θ can vary, but in an exemplary embodiment of the invention,
θ=22.5°, and the resulting junction is a symmetric 22.5° grain
boundary Josephson junction, which, as shown in FIG. 4, results in a 0 phase shift
in the supercurrent I across junction 120.
The widths of the junctions, 110 and 120, L
1 and L
2,
respectively, are chosen to maximize the device sensitivity to flux threading in
loop 150 by controlling the amount of supercurrent that travels along each
branch of dc-SQUID 100. In some embodiments where junction 110 is
a π/2 junction and junction 120 is a 0-junction, then widths L
1
and L
2 can be on the order of 1 μm. One skilled in the art
will recognize that SQUID magnetometer 100 according to the present invention
can have any combination of junctions 110 and 120 such that total
intrinsic phase shift of loop 150 is realized. Sensitization of the dc-SQUID
is realized for a range of phase shifts, where the phase shift can vary from 0<φ<π.
Phase shifts of around π/2, as described above, further lead to directional
sensitivity. Therefore, there is a hardy tolerance in the fabrication of the invention
in terms of junction behavior, as well as allowing for a plurality of Josephson
junctions that total a phase shift within the desired range.
FIGS. 5
a through 5
c illustrate a comparison of the Critical
Current-Flux relation for a 0-phase shift SQUID magnetometer, a π-phase shift
SQUID magnetometer and a π/2-phase shift SQUID magnetometer according to
the present invention. FIGS. 5
a through 5
c show that the critical
current in the superconducting loop can be directly correlated with the flux through
the loop. Characterization of the relationship can be found experimentally by controlling
the supercurrent through dc-SQUID magnetometer 100 and measuring the flux
in the loop with an instrument such as a magnetometer, or, in a contrary fashion,
by applying a magnetic field through SQUID loop 100 and measuring the resulting
supercurrent. In another experimental procedure, a constant current is applied
across the terminals and the potential drop across the SQUID loop is measured.
FIG. 5
a shows the current-phase relationship for a 0 phase-shift SQUID
loop. At 0 current and a normalized phase shift of 1, the response curve is at
a peak. Therefore, a change in the flux through the loop results in a decrease
in the supercurrent, that is independent of the direction of the flux in the loop.
The response of a π dc-SQUID as shown in FIG. 5
b is a useful variation
of the response shown in FIG. 5
a of the dc-SQUID as the sensitivity of the
π dc-SQUID is linear with overall phase. As is clear from FIG. 5
b,
a change in the flux in the loop results in a steep change in the current, but
again, the change is positive in both cases and thus is independent of the direction
of the flux in the loop. The response of a π/2 dc-SQUID is shown in FIG.
5
c. In the π/2 dc-SQUID, the equilibrium position (at 0 current) is
also at a point where the sensitivity is linear with phase. However, a small perturbation
in the flux through the loop causes a positive change in the current for one direction,
and a negative change in the current for the other direction, thus allowing for
directional sensitivity. An advantage of embodiments of a dc-SQUID magnetometer
100 in accordance with the present invention is a 2 fold increase in sensitivity
over a conventional π dc-SQUID due to the sensitivity to direction of flux
through the loop.
FIGS. 6
a through 6
c illustrate an example of fabrication
of a SQUID magnetometer according to the present invention. In this example, a
bi-epitaxial fabrication method is employed, although one skilled in the art will
recognize that other deposition methods can also be employed. Exemplary methods
of biepitaxial fabrication are described in S. Nicoletti, H. Moriceau, J. C. Villegier,
D. Chateigner, B. Bourgeaux, C. Cabanel, and J. Y. Laval, "Bi-epitaxial YBCO grain
boundary Josephson junctions on SrTiO
3 and sapphire substrates," Physica
C, 269, p.255-267, 1996, and the references therein, which is hereby included by
reference in its entirety. Further discussions regarding fabrication of a grain
boundary between two d-wave superconductor materials is further discussed in F.
Tafuri, F. Carillo, F. Lombardi, F. Miletto Granozio, F. Ricci, U. Scotti di Uccio,
A. Barone, G. Testa, E. Sarnelli, J. R. Kirtley, "Feasibility of Biepitaxial YBaCuO
Josephson Junctions for Fundamental Studies and Potential Circuit Implementation",
Los Alamos preprint server condmat/0010128, accepted for publication Phy. Rev.
B (2000), which is herein incorporated by reference in its entirety. The behavior
of such junctions on the phase shift of the order parameter is discussed in C.
Bruder, A. van Otterlo, and G. T. Zimanyi, "Tunnel junctions of Unconventional
Superconductors," PRB 51, 12904 (1995); and C. C. Tsuei, "Design and realization
of an all d-wave dc π-superconducting quantum interference device," Applied
Physics Letters, 76, p.912 (2000), each of which is included herein by reference
in its entirety.
In some embodiments, a seed layer followed by a buffer layer is deposited onto
a substrate such as SrTiO
3 or Sapphire by means of pulsed laser deposition.
The buffer layer forms on the substrate with a rotated crystallographic orientation
(π/2 in some embodiments). Different crystallographic orientations can be
achieved through the use of different buffer materials. A section of the seed and
buffer layer can then be removed by a process such as Xe ion milling. The milling
creates the weak link boundary (i.e., boundaries 103 and 104 of FIG.
1) for the device. Next, a second buffer layer followed by the high-T
c superconducting
film are deposited, and the final structure is realized through further etching
of the film. The buffer layers can be formed of CeO
2 for example.
FIG. 6
a shows an intermediate structure having a substrate layer 500
as a base for the bi-epitaxial fabrication process. Substrate layer 500
can be, for example, a layer of SrTiO
3 or Sapphire. A seed layer 510
is formed on substrate layer 500. Seed layer 510, for example, can
be of MgO. In some embodiments, seed layer 510 can be about 5 nm thick.
A buffer layer 515 can then be deposited on seed layer 510. In some
embodiments, buffer layer 515 can be of CeO
2 and can have a thickness
of approximately 11 nm. Seed layer 510 and buffer layer 515 can then
be milled away (for example by Xe-ion milling) from section 590, as illustrated
in FIG. 5A, such that a boundary 550 is formed. Angle Θ illustrates
a bend in boundary 550, corresponding to the bend between boundary 103
and boundary 104 of FIG. 1. In some embodiments, the angle Θ
can be chosen as 22.5°. Boundary 550, with a bend angle Θ, can
be fabricated with smooth and sharp features using any of a number of etching techniques
such as Xe ion milling.
In some embodiments, boundary 550, with bend angle Θ, is oriented
with respect to the crystal structure of buffer layer 515 such that, once
a d-wave crystal structure is deposited on top of buffer layer 515, the
d-wave crystal structure is oriented with it's a-axis at a 22.5° angle with
respect to boundary 550 in one portion and at 0° in a second portion
where the portions are separated by the bend at bend angle Θ. Reversly, boundary
550 can be arranged such that the a-axis of the superconducting crystal
structure is oriented at 45° with respect to boundary 550 in one portion
and 22.5° with respect to boundary 550 in a second portion.
FIG. 6
b shows deposition of a second buffer layer 530 on top of
section 590. Buffer layer 530 can have a thickness of approximately
18 nm and, again, can be of CeO
2. The crystal orientation of buffer
layer 515 differs from the crystal orientation of buffer layer 530
such that the intersection of layers 530 and 515 at boundary 550
creates a grain boundary.
A superconducting layer 532 can then be deposited on buffer layer 515
and a superconducting layer 534 can be deposited on buffer layer 530
in such a way that boundary 550 remains clean and sharp. Superconducting
layers 532 and 534 can be about 200 nm in thickness. The crystal
orientations of the superconducting materials of layers 532 and 534
are determined by the crystal orientation of underlying seed layer 510 or
substrate 500, respectively. In some embodiments, a 45° lattice mismatch
is arranged between the superconducting material of layer 532 and the superconducting
material of layer 534. In some embodiments the high-T
c superconducting
material of layers 532 and 534 is a material such as YBa
2Cu
3O
7-x,
where x is some value greater than 0 and less than 0.6.
FIG. 6
c shows SQUID magnetometer 100 completely fabricated. As
illustrated in FIG. 6
c, superconducting layers 532 and 534
are etched into dc-SQUID 100 using, for example, a process such as Xe ion
milling, with the resulting junctions 110 and 120 being a π/2-junction
and 0-junction, respectively. The width of each of junctions 110 and 120
helps define the operation parameters of the SQUID by effecting the capacitance
and critical current values. For example, having the width of junction 110
as twice that of junction 120 doubles the current in the 110 branch
and halves the current in the 120 branch. In an exemplary embodiment, the
width of branches 581 and 582, in superconducting loop 150,
are on the order of 1 micrometer, which is much smaller than the width if the overall
SQUID loop 150. Furthermore, by tuning the widths of branches 581
and 582, dc-SQUID 100 can be made substantially impervious with respect
to the degree of cleanliness of junctions 110 and 120. In one particular
embodiment formed with superconducting material YBa
2Cu
3O
7-x,
where x has values less than 0.6 and greater than 0.05, the widths of junctions
110 and 120 (L
1 and L
2, respectively) are both
about 1 μm respectively, the width of branches 581 and 582
are also approximately 1 μm as well, and the inside separation between branches
581 and 582 is approximately 10 μm.
FIG. 7 shows a cross-sectional view of an example of a junction fabricated as
described above with FIGS. 6
a through 6
c. Each of the contributing
layers are shown approximately to scale before any ion milling. Substrate 500
can be SrTiO
3, and can be approximately 50 nm in thickness. Seed layer
510 can be of MgO, and can be approximately 5 nm thick. Seed layer 510
has been milled away from the right hand side of the sample (under section 102)
shown in FIG. 7. Buffer layer 515 can be of CeO
2 and can
be approximately 11 nm thick. First buffer layer has also been milled away from
the side under section 102. Second buffer layer 530 can be of the
same material as the first buffer layer, but with a thickness of approximately
18 nm. Further, second buffer layer 530 can be deposited over first buffer
layer 515. Finally, the superconductive material 532 and 534
is deposited onto the sample with a thickness of approximately 200 nm. The thickness
values and materials given in this example are exemplary, and in no way limit the
scope of the fabrication of the invention.
FIG. 8 illustrates an embodiment of a 2-dimensional array 600 of dc-SQUID
magnetometers according to the present invention. Array 600 includes SQUID
magnetometers 620-1,1 through 620-M,Q as shown in FIG.
6. In the embodiment of FIG. 8, each row can include a different number
of magnetometers. Each of SQUID magnetometers 620-1,1 through
620-1,N is formed along a grain boundary 630-1 between
regions 630-1 and 630-2. Grain boundary 610-1
is shaped to provide positions for the formation of both 0-junctions and π/2-junctions
and conversely π-junctions and π/2 junctions. Each of grain boundaries
610-1 through 610-M is shaped to provide for the fabrication
of at least one 0-junction or π-junction and at least one π/2-junction
for each SQUID magnetometer 620-1,1 through 620-M,Q
formed along that grain boundary.
Array 600 includes regions 610, 612 and 614. In
regions 610 and 614, d-wave superconducting material can be deposited
with a first crystallographic orientation and in region 612 d-wave superconducting
material can be deposited with a second crystallographic orientation such that
the lattice mismatch at the grain boundaries allows for creation of 0-junctions
and π/2 junctions along the grain boundaries between regions 610 and
612 and between regions 612 and 614. Furthermore, the shape
of the grain boundary between regions 610 and 612 and regions 612
and 614 can be set in order to facilitate the production of a 0-junction
and a π/2 junction in each SQUID magnetometer. SQUID magnetometers 620-1,1
through 620-M,Q include 0-junctions 640-1,1 through
640-M,Q, respectively, and π/2 junctions 642-1,1
through 642-M,Q, respectively, as shown in FIG. 6. SQUID magnetometers
620-1,1 through 620-M,Q of array 600 can be
fabricated as described above with respect to FIGS. 6
a through 6
c.
One skilled in the art will recognize that array 600 can be extended to
an array including a plurality of dimensions and a plurality of π/2 dc-SQUID
magnetometers. For example, a series of array 600 can be coupled to create
a multi-dimensional array. In some embodiments, a series of array 600 magnetometers
is stacked to form a three dimensional magnetometer.
FIG. 9 illustrates the utilization of a π/2 SQUID magnetometer 702
according to the present invention. Magnetometer 702 can include a single
SQUID loop or may be an array of SQUID loops such as array 600 illustrated
in FIG. 8. Magnetometer 702 is positioned in proximity to a magnetic
field source 701.
In some applications, source 701 can be a quantum qubit. In some other
applications, source 701 can be an electronic circuit. In some further applications,
source 701 can be a superconducting circuit. Source 701 can also
be a magnetic resonance imaging system, a metallic sample being tested for defects,
a mine, or a submarine. In general, source 701 can be any source of a magnetic field.
Magnetometer 701 is coupled to a current device 703 in
series and a voltmeter 704 in parallel. Current device 703 applies
a constant current to the dc-SQUID Magnetometer. Any magnetic fields can be detected
by the voltmeter. Embodiments of current device 703 are well known in the
art. In an exemplary embodiment, the voltmeter 704 can be a radio-frequency
single electron transistor. In operation, presence of an external magnetic field
induces a superconducting current in magnetometer 701 which exceeds the
critical current of the superconducting loop. Resultingly, a the junctions in the
loop become resistive, and the voltmeter 704 registers a voltage. Since
magnetometer 701 is direction sensitive, voltmeter 704 is capable
of determining both the strength of the magnetic field {right arrow over (B)},
and any variation in the magnetic field {right arrow over (B)}, and the direction
of the magnetic field {right arrow over (B)}.
The above described embodiments are exemplary only and are not intended to be
limiting. One skilled in the art will recognize variations from the particular
embodiments described above that are intended to be within the spirit and scope
of this invention. As such, the invention is limited only by the following claims.
*
