Buschow K.H.J. (Ed.) Concise Encyclopedia of Magnetic and Superconducting Materials

Подождите немного. Документ загружается.

YBCO (Weinstein et al. 1996), i.e., the trapped ﬁeld

reduces by about 5% per time decade. Hence, ﬂux

creep seems to be a serious problem for supercon-

ducting permanent magnets. However, successful re-

duction of the relaxation rate was achieved by

cooling the YBCO sample by several degrees during

or after activation. Investigating a YBCO mini-mag-

net with B

¼ 4.3 T at 65 K, the creep rate was found

to lower by factors of 6, 197, or X 1000 by decreasing

the temperature for DT ¼2 K, 4 K, and 6 K, respec-

tively, after activation (Weinstein et al. 1996).

4. Limiting Fac tors for Superconducting

Permanent Magnets in Different Temperature

Ranges

The simplest way to enhance trapped ﬁelds in YBCO

is to work at temperatures below 77 K using the

higher critical current density and pinning force at

lower temperatures. As the trapped ﬁeld increases,

the mechanical properties of the material also become

important, because large forces act on the supercon-

ductor during the magnetizing procedure. Since the

ﬂux lines are pinned on defects within the supercon-

ductor, a force is transferred to the crystal lattice

where the pinning defects are embedded. As a result

of these body forces, the material undergoes macro-

scopic mechanical deformations during the magnet-

izing procedure. Magnetoelastic effects in HTS have

been reviewed by Johansen (2000).

A trapped ﬁeld in the superconductor is connected

to a tensile stress, which causes cracking of the disk if

the tensile strength of the material is reached. The

different limitations of the trapped ﬁeld of a melt-

textured YBCO sample are illustrated in Fig. 5. In the

temperature range between about 50 K and T

, the

trapped ﬁeld is determined by the pinning properties

of the material and the size of the superconducting

domains. At temperatures around and below about

50 K, the trapped ﬁeld is limited by the mechanical

properties of the superconductor. Thermomagnetic

instabilities become predominant at temperatures be-

low 20 K. In this region the trapped ﬁeld is strongly

reduced by ﬂux jumps (Wenger 2000).

The limitations of the trapped ﬁeld in the ranges of

high and low temperatures have been discussed in the

previous chapters. In the following, the limitation of

the trapped ﬁeld by mechanical properties and several

possibilities to overcome this problem will be con-

sidered in more detail.

5. Very High Trapped Fields in Bulk YBCO

Below 77 K

The temperature dependence of the maximum

trapped ﬁeld is usually measured in superconducting

magnets. For magnetizing the bulk YBCO sample, a

magnetic ﬁeld is applied at a temperature above T

and the sample is cooled to the measuring tempera-

ture. Then, the magnetic ﬁeld is reduced to zero and

is recorded by means of a Hall probe. However, at

high trapped ﬁelds, large magnetic tensile stresses act

on the superconductor during this procedure result-

ing in cracking of the material which is indicated by a

sudden decrease of the signal of the Hall probe dur-

ing the magnetizing procedure. A visual inspection of

damaged samples shows in some cases one crack on

the surface of the sample, but sometimes no crack is

visible.

The crack within the superconductor can be easily

detected by a ﬁeld proﬁle measurement at 77 K

(Fuchs et al. 1997). A typical example is shown in

Fig. 6. The crack divides the disk into two supercon-

ducting domains. Currents can ﬂow only within the

superconducting domains, but not across the bound-

ary between the two domains.

The distribution of the tensile stress within a bulk

superconductor during the activation process was

calculated by Ren et al. (1995) for long cylindrical

samples. Later, this model was extended to arbitrary

lengths of cylindrical samples (Johansen 1999). Ac-

cording to these models, the maximum tensile stress

occurs at the center of the sample and increases

with the square of the trapped ﬁeld B

The tensile strength s

max

of a damaged sample can

be estimated from the applied magnetic ﬁeld at which

cracking was observed using the theoretical s

)

dependence predicted by the stress model. A typical

value for the tensile strength of melt-textured YBCO

is s

max

B40 MPa (Ren et al. 1995), which is more

than one order of magnitude lower than the tensile

strength of metallic low-temperature superconductors

originating from small cracks in melt-textured

Figure 5

Schematic representation of the temperature dependence

of the maximum trapped ﬁeld B

of a YBCO disk

showing the limitations of B

in different temperature

ranges.

1170

Superconducting Permanent Magnets: Principles and Results

YBCO, limiting the tensile strength of the material.

The cracks start to propagate when the external stress

reaches the fracture toughness K

of the material. The

tensile strength depends in this case on K

and the

initial size a of the largest crack according to the re-

lation (Paker 1981)

max

ðpaÞ

0:5

ð2Þ

Typical values of the fracture toughness of melt-

textured YBCO are in the range of K

(1–2) MPa m

1/2

(Scha

tzle et al. 1999). The fracture

toughness was found to increase by the incorporation

of ductile particles impeding the propagation of the

cracks. In YBCO containing ductile silver precipi-

tates, an enhancement up to K

B2.4 MPa m

1/2

was

reported compared with K

B1.9 MPa m

1/2

in samples

without silver inclusions (Scha

tzle et al. 1999). It

was also reported that YBCO samples can be rein-

forced against the tensile stress by placing them into

steel tubes. Steel has a larger coefﬁcient of thermal

expansion than YBCO in the a, b plane resulting in a

compressive stress after cooling the sample from

300 K to the measuring temperature p77 K. Using a

steel tube, the trapped ﬁeld of YBCO disks with silver

inclusions could be enhanced to values of B

¼11.4 T

(at 17 K) and of B

¼14.3 T (at 22 K) on the surface

of a single YBCO sample, and in the gap between two

YBCO samples, respectively (Fig. 7) (Fuchs et al.

2000). In both cases, the YBCO disks had a diameter

of 25 mm.

For comparison, trapped ﬁelds up to 10.1 T at 42 K

have been reported by Weinstein et al. (1996) for a

YBCO mini-magnet without silver additions and

without mechanical reinforcement. This mini-magnet

consisted of stacked irradiated YBCO tiles.

6. Levitation Forces

One of the most fascinating features associated with

HTS is the stable levitation of a permanent magnet

over a bulk HTS. This levitation is completely pas-

sive, because the magnetic ﬂux surrounding the per-

manent magnet is pinned by defects within the

superconductor. On the other hand, stable levitation

is not possible for two permanent magnets and,

therefore, conventional magnetic bearings require

an active feedback control in order to stabilize the

levitation. Passive superconducting bearings consist-

ing of a superconductor and a permanent magnet are

one of the most promising applications of bulk

Figure 6

Typical ﬁeld proﬁle of a damaged YBCO disk after

cracking (top panel) and schematic view of the

corresponding two superconducting domains and the

circulating supercurrents (bottom panel).

Figure 7

Temperature dependence of the maximum trapped ﬁeld

for YBCO disks containing 10 wt.% Ag: J, single

YBCO disk, ’, YBCO disk-pair. In both cases a

bandage made of a steel tube ﬁlled with epoxy was used

(after Fuchs et al. 2000).

1171

Superconducting Permanent Magnets: Principles and Results

YBCO. They can be used in ﬂywheels for energy

storage, in motors for liquid gas pumps and for

transportation systems (see Superconducting Perma-

nent Magnets: Potential Applications). The possibility

of linear transportation is nicely demonstrated by a

model railway, where an old-fashioned steam loco-

motive equipped with superconducting permanent

magnets and cooled with liquid nitrogen runs con-

tactless and frictionless on a magnetic track (Fig. 8).

The levitation force between a permanent magnet

and a superconductor cooled in the ﬁeld of the mag-

net and containing a trapped magnetic ﬂux can be

written as

Fpj

RdB=dx ð3Þ

where R is the size of the current loops in the super-

conductor and dB/dx is the ﬁeld gradient produced

by the permanent magnet within the superconductor.

As the superconductor approaches the permanent

magnet after zero ﬁeld cooling, the repulsive levitation

force increases and reaches a maximum value for a

small distance. Typical values of the maximum levi-

tation force F between a samarium–cobalt perma-

nent magnet (diameter 25 mm) and YBCO disks are

between 50 and 90 N (Gruss et al. 1999) at 77 K. The

corresponding levitation pressure F/A (A ¼area of the

permanent magnet) varies between 10 and 18 N cm

2

The levitation pressure of this type of supercon-

ducting magnetic bearings is mainly limited by the

surface ﬁeld of the available permanent magnets,

which is lower than the maximum trapped ﬁeld of

YBCO at 77 K. Therefore, the bulk YBCO can not be

magnetized up to the maximum value of the trapped

ﬁeld, especially in the temperature range below 77 K,

where the trapped ﬁeld of the superconductor be-

comes yet higher.

The upper limit of the levitation force achievable

for a given permanent magnet would be obtained for

an ideal superconductor of inﬁnite size and inﬁnitely

large critical current density perfectly shielding the

magnetic ﬁeld of the permanent magnet. The result-

ing levitation force corresponds to the force between

two permanent magnets and is given by (Teshima

et al. 1996)

max

¼ AB

=2m

ð4Þ

where B

is the surface ﬁeld of the permanent mag-

net. For a permanent magnet with B

¼0.4 T, the

Figure 8

Model railway with steam locomotive equipped with two superconducting permanent magnets and cooled with liquid

nitrogen as demonstration of stable levitation.

1172

Superconducting Permanent Magnets: Principles and Results

maximum achievable levitation pressure is F

max

/A ¼

/2m

¼25 N cm

2

. The experimental values for

YBCO samples mentioned above achieve 50–72%

of this upper limit. Much larger levitation pressures

up to 5000 N cm

2

are expected if a superconducting

permanent magnet with a trapped ﬁeld as high as

11 T were to be used instead of a conventional per-

manent magnet.

The most important problem which has to be

solved in order to realize YBCO magnets for super-

conducting permanent magnets or for other applica-

tions at temperatures lower than 77 K is to magnetize

the superconductor. Pulsed magnetic ﬁelds can be

used for magnetizing YBCO disks, however, large

viscous forces act on the penetrating magnetic ﬂux,

especially in the case of short rise times (B1 ms) of

the pulsed ﬁeld. Therefore, (i) higher pulsed ﬁelds are

needed in order to trap the same magnetic ﬁeld in the

superconductor as by applying a static ﬁeld and (ii)

heat generation due to the rapid motion of magnetic

ﬂux strongly reduces the trapped ﬁeld achievable at

temperatures lower than 77 K (Yanagi et al. 1998). At

77 K, promising results were achieved by using pulsed

ﬁelds with a relatively long rise time of 16 ms (Gruss

et al. 1999).

7. Conclusions

These new superconducting permanent magnets form

a new class of ultrahigh performance permanent

magnets. The trapped ﬂux density can be higher by

an order of magnitude compared to the remanence of

the best ferromagnetic permanent magnets (see Rare

Earth Magnets: Materials). Since stored magnetic

energies and forces scale quadratically with the

trapped ﬂux density, these new ‘‘cold magnets’’ of-

fer magnetic energies and forces two orders of mag-

nitude higher than conventional magnets, reaching

completely new ﬁelds of applications accessible for

permanent magnets. Two examples are the rotational

frictionless support, i.e., superconducting magnetic

bearings, and frictionless linear transportation. The

latter should allow transportation of extremely heavy

goods which can not easily be moved by wheeled

systems, like concrete ports or moving bridges. Ad-

ditionally, contactless transportation, which does not

create any wear debris, can be advantageously used in

clean rooms in the microelectronics industry (for fur-

ther reading, see Superconducting Permanent Mag-

nets: Potential Applications).

See also: Superconducting Machines: Energy Stor-

age; Magnetic Levitation: Superconducting Bearings

Bibliography

Bean C P 1962 Magnetization of hard superconductors. Phys.

Rev. Lett. 8, 250

Fuchs G, Krabbes G, Scha

tzle P, Gruss S, Stoye P, Staiger T,

ller K-H, Fink J, Schultz L 1997 High trapped ﬁelds in

bulk YBa

7d

samples at temperatures around 50 K.

Appl. Phys. Lett. 70, 117–9

Fuchs G, Scha

tzle P, Krabbes G, Gruss S, Verges P, Mu

ller

K-H, Fink J, Schultz L 2000 Trapped magnetic ﬁeld larger than

14 Tesla in bulk YBa

7x

. Appl. Phys. Lett. 76, 2107–9

Garwin E L, Rabinowitz M, Frankel D J 1973 Permanent

multipole magnetic ﬁelds stored in superconductors. Appl.

Phys. Lett. 22, 599–600

Gruss S, Fuchs G, Krabbes G, Scha

tzle P, Fink J, Mu

ller K-H,

Schultz L 1999 High trapped ﬁelds in melt-textured

YBa

7d

. IEEE Trans. Appl. Supercond. 9, 2070–3

Ikuta H, Mase A, Mizutani U, Yanagi Y, Yoshikawa M, Itoh

Y, Oka T 1999 Very high trapped ﬁeld in melt-processed

SmBaCuO. IEEE Trans. Appl. Supercond. 9, 2219–22

Johansen T H 1999 Flux-pinning-induced stress and strain in

superconductors: Case of a long circular cylinder. Phys. Rev.

B 60, 1–14

Johansen T H 2000 Flux-pinning-induced stress and magneto-

striction in bulk superconductors. Supercond. Sci. Technol.

13, R121–37

Klupsch T, Strasser T, Habisreuther T, Gawalek W, Gruss S,

May H, Palka R, Mora Serrano F J 1997 Field mapping

characterization for axially magnetized superconducting cyl-

inders in the remanent critical state: theory and experiment.

J. Appl. Phys. 82, 3035–41

Krabbes G, Scha

tzle P, Bieger W, Fuchs G, Wiesner U, Sto

ver G

1997 Thermodynamically controlled melt processing to im-

prove bulk materials. IEEE Trans. Appl. Supercond. 7,1735–8

Krabbes G, Fuchs G, Scha

tzle P, Gruss S, Park J W, Hard-

inghaus F, Sto

ver G, Hayn R, Drechsler S-L, Fahr T 2000

Zinc doping of YBa

in melt-textured materials: peak

effect and high trapped ﬁelds. Physica C 330, 181–90

McGuinn P 1995 Bulk processing and charactrization of

YBa

6 þ x

. In: Donglu S (ed.) High Temperature Super-

conducting Materials Science and Engineering. Elsevier,

Amsterdam, pp. 345–82, Chap. 8

Murakami M 1992 Melt processing. In: Murakami M (ed.)

Melt Processed High-Temperature Superconductors. World

Scientific, Singapore, pp. 21–44, Chap. 3

Murakami M 1993 Melt processing, ﬂux pinning and levitation.

In: Jin S (ed.) Processing and Properties of High T

Super-

conductors. World Scientific, Singapore, Vol. 1, pp. 215–68,

Chap. 6

Paker A P 1981 The Mechanics of Fracture and Fatigue, An

Introduction. E & F N Spon, London

Rabinowitz M, Arrowsmith H W, Dahlgren S D 1977

Dependence of maximum trappable ﬁeld on superconducting

Sn cylinder wall thickness. Appl. Phys. Lett. 30, 607–9

Ren Y R, Weinstein R, Liu J, Sawh R P, Foster C 1995 Dam-

age caused by magnetic pressure at high trapped ﬁeld in

quasi-permanent magnets composed of melt-textured YBa-

CuO superconductors. Physica C 251,15

Scha

tzle P, Krabbes G, Gruss S, Fuchs G 1999 YBCO/Ag bulk

material by melt crystallization for cryomagnetic applica-

tions. IEEE Trans. Appl. Supercond. 9, 2022–5

Teshima H, Morita M, Hashimoto M 1996 Comparison of the

levitation forces of melt-processed YBaCuO superconductors

for different magnets. Physica C 269, 15–21

Walter H, Delemare M P, Bringmann B, Leenders A, Freyhardt

H C 2000 Melt-textured YBaCuO with high trapped ﬁelds up

to 1.3 T at 77 K. J. Mater. Res. 15, 1231–4

Weinstein R, Liu J, Ren Y, Sawh R, Parks D, Foster C, Obot V

1996 Very high trapped ﬁelds: cracking, creep, and pinning

1173

Superconducting Permanent Magnets: Principles and Results

centers. In: Batlogg B, Chu C W, Chu W K, Gubser D U,

ller K A (eds.) Proc. 10th Anniversary HTS Workshop on

Physics, Materials and Applications. World Scientific, Singa-

pore, pp. 625–8

Weinstein R, Ren Y, Liu J, Chen I G, Sawh R, Parks D, Foster

C, Obot V 1994 Progress in J

, pinning and grain size for

trapped ﬁeld magnets. In: Fujita T, Shiohara Y (eds.) Ad-

vances in Superconductivity IV. Springer, Tokyo, pp. 855

Wenger C. 2000 Untersuchung der thermodynamischen Stabi-

lita

t des schmelztexturierten YBa

7x

—Materials (In-

vestigation of the thermodynamic stability of melt-textured

YBa

7x

material), Ph.D. thesis. Technical University

Dresden, Germany

Yanagi Y, Itoh Y, Yoshikawa M, Oka T, Terasaki A, Ikuta H,

Mizutani U 1998 Pulsed-ﬁeld magnetization of YBaCuO su-

perconducting bulk magnet cooled by refrigerator. In:

Osamura K, Hirabayashi I (eds.) Advances in Superconduc-

tivity X. Springer, Tokyo, pp. 941–4

G. Fuchs and L. Schultz

Institut fu

r Metallische Werkstoffe, Dresden, Germany

Superconducting Radiation Sensor

Applications: Detectors and Mixers

Superconducting detectors and mixers belong to the

wide class of small-scale applications of supercon-

ductivity. They are used in measurements of weak

radiating electromagnetic ﬁelds and nuclear radia-

tion. Superconductivity offers various basic princi-

ples for effective radiation detection. Several devices

have been developed, exploiting the different features

of the condensed matter in the superconducting state:

(i) heating by the absorption of radiation, which is

sensitively detected by a superconducting device;

(ii) creation of nonthermal excitations by the ab-

sorption of radiation and their collection by a super-

conducting device;

(iii) capability of active superconducting devices of

processing high-frequency electromagnetic waves.

The most relevant types of superconducting sen-

sors for radiation detection are microcalorimeters,

superconducting tunnel junction detectors, and mix-

ers. Superconducting detectors are developed for vis-

ible, ultraviolet, and x-ray single-photon detection.

They can also be applied in detecting large-mass

molecules or nuclear radiation, like electrons, a-par-

ticles, and neutrons. Mixers are typically used in

measurements of electromagnetic radiation in the in-

frared and millimeter wavelength regions.

The major motivation for using superconducting

sensors is their superior sensitivity. The detection of

very small energy deposition (high sensitivity) with

high energy resolution (i.e., the capability of distin-

guishing slightly different energies) is achieved by

operating the sensors at low temperatures, where lat-

tice vibrations (phonons) and excitation energy are

more than one hundred times smaller.

The need for detectors with advanced performance

is a continuous request from frontier experiments in

elementary particle physics and astronomy. The

progress in the development of both sensors and

complementary technologies (cryogenics, in particu-

lar) opens the possibility of using these sensors in a

novel generation of spectrometers for microanalysis

of materials in civil applications (laboratory and in-

dustrial measurements).

1. Superconducting Microcalorimeters

Microcalorimeters have been developed for x-ray as-

tronomy to detect single-photon events from faint

sources in the energy range 0.1–10 keV (1 eV ¼

1:602  10

19

J). A microcalorimeter consists of an

absorber in thermal contact with a thermometer (see

Fig. 1). The energy (E) of the incident radiation,

typically one photon, is converted to heat in the

absorber, leading to a temperature rise DT ¼ E=C,

where C is the total heat capacity of the absorber and

the thermometer. The temperature rise is read by the

thermometer and indicates the radiation energy. Any

microcalorimeter must have a low heat capacity mass

to absorb incident radiation, a weak link to a low-

temperature heat sink which provides the thermal

isolation needed for a temperature rise to occur, and

a thermometer with a high sensitivity to measure

change in temperature. In a microcalorimeter, one

exploits the fact that the heat capacity of supercon-

ductors or pure dielectric crystals becomes small at

low temperatures (50–100 mK). Thus, the tempera-

ture rise can be measured by resistive thermometers

(thermistors), for which a change in the electrical re-

sistance accompanies a change in temperature.

Thermistors are characterized by their sensitivity,

a ¼ dðlog RÞ=dðlog TÞ, where R is the electrical re-

sistance and T is the temperature. In an ideal therm-

istor-based calorimeter, the energy resolution, DE,

scales as T( C /|a|)

1/2

so that low operating tempera-

tures and heat capacities contribute to achieve good

energy resolution.

Two types of thermistors are used in microcalo-

rimeters: doped semiconductor thermistors and tran-

sition edge sensors (TES). Microcalorimeters of

neutron transmutation doped (NTD) germanium

thermistors paired with superconducting tin absorb-

ers have provided an energy resolution of DE ¼

5:2eVatE ¼ 5:9 keV (Alessandrello et al. 1999). TES

are thin superconducting ﬁlms deposited on the ab-

sorber. The temperature rise is measured by the TES

which works in the region of its normal-to-supercon-

ductor phase transition (see Superconducting Materi-

als, Types of), where the temperature dependence of

the resistance is steep. TES are made from supercon-

ductors (aluminum, titanium, tungsten, or proximity

bilayers, like Al/Ag, Ir/Au, Mo/Au), with transition

temperatures in the range 50–100 mK (see Fig. 2).

1174

Superconducting Radiation Sensor Applications: Detectors and Mixers

Because the transition from the normal to the super-

conducting state occurs over a very narrow temper-

ature range, a TES is a much more sensitive

thermometer than a semiconductor thermistor. The

advantage of TES-based devices is that a large vol-

ume of normal metals, off-limits to semiconductor-

based microcalorimeters, can be used. The rapid and

efﬁcient thermalization of metal absorbers allows

better energy resolution and faster detector response.

TES microcalorimeters achieved energy resolution

DEo4:7eV at E ¼ 6 keV and DE ¼ 2:38 eV at

E ¼ 1:5 keV, with count rates in excess of 500 counts

per second.

2. Superconducting Tunnel Junction Detectors

A superconducting tunnel junction (STJ) consists of

two superconducting thin ﬁlms (B200 nm in thick-

ness) separated by a very thin insulating barrier

(B1 nm in thickness) (see the sketch in Fig. 3). Typ-

ically used superconductors are niobium, tantalum,

or aluminum. Normal electrons (so-called quasipar-

ticles) can ﬂow across the barrier via quantum tun-

neling processes. In STJ detectors, one exploits the

fact that the energy gap of a superconductor is of the

order of millielectronvolts, so that millions of elec-

trons can be created after a complex energy cascade

following the absorption of a single x-ray photon.

The amount of created charge is proportional to the

energy of the absorbed radiation. The absorber of an

STJ detector can be either one of the two junction

electrodes or a separate superconductor which injects

the STJ with the excitations in the form of electrons

or high-energy phonons.

The nonequilibrium electronic charges produced by

the radiation diffuse and ﬂow across the tunneling

barrier of the STJ, where they are measured as an extra

tunneling current. The tunneling barrier is made trans-

parent to improve tunneling so that STJs are physi-

cally the same as the more widely known Josephson

tunnel junctions (see Josephson Junctions: High-T

;

Josephson Junctions: Low-T

), but the Josephson su-

percurrent is suppressed by applying a weak magnetic

ﬁeld. STJs are biased in the subgap region of the cur-

rent–voltage characteristics and are operated at very

low temperature (TBT

=10) in order to reduce the

thermal population of electrons and the recombination

Figure 1

Sketch of a microcalorimeter with the resistance vs. temperature curve of a TES thermometer.

Figure 2

A Mo–Cu TES microcalorimeter on a Si

membrane

patterned as a ‘‘ﬂy-swatter.’’ The TES has dimensions

400 400 mm. The large bars are 50 mm wide, and the

small bars are 5 mm wide on 50 mm centers. The device

was fabricated at NIST, Boulder, CO, USA (from Irwin

et al. 2000, reprinted with permission from Elsevier

Science Ltd.).

1175

Superconducting Radiation Sensor Applications: Detectors and Mixers

of the extra electrons. Compared with semiconductor

detectors, the very high number of charges generated

gives rise to improved energy resolution and detection

threshold. Intrinsic energy resolution of few electron-

volts at 6 keV is possible and 12 eV has been measured

in Al-STJ (Angeloher et al. 2001).

STJs are nonthermal, single-particle detectors. In

nonthermal detectors, the nonequilibrium distribu-

tion of excitations is detected before it relaxes. This

feature makes STJs particularly suitable in high-

count-rate applications (5000 counts per second can

be achieved). Similarly to TES, STJ detectors were

developed to answer the demanding request of x-ray

astronomy for high-energy resolution detectors.

However, they are also capable of detecting single

photons with lower energy, like ultraviolet and visible

photons. At these short wavelengths, thousands of

electrons are still created in a superconductor, sufﬁ-

cient to generate an appreciable pulse signal. Imaging

can be obtained by arrays of STJ pixels (see Fig. 4).

3. Superconducting Mixers for Infrared Detection

Infrared and millimeter wave radiation detectors

are realized by several superconducting devices, like

STJs (in this context often called superconductor in-

sulator superconductor (SIS) junctions), Josephson

junctions, or hot electron bolometers. Applications in

radioastronomy, in passive molecular spectroscopy

for environmental monitoring and air quality control,

have been developed. Infrared and millimeter wave-

lengths correspond to energy values too low (BmeV)

for single photon detection and to frequencies too

high (B0.1 GHz–10 THz) where ampliﬁcation and

processing of electronic signals are difﬁcult. In this

case a useful concept is the mixing to down-convert

high-frequency (HF) signals. In heterodyne mixers

Figure 3

Sketch of a STJ with the corresponding current–voltage

(I–V) curves with and without radiation. The red curve

shows the increased tunneling current with respect to the

dark condition (black curve).

Figure 4

(a) A 10 10 array of STJs. Each STJ has dimensions

100 100 mm. (b) A single STJ detector on a silicon

substrate patterned as an ‘‘island.’’ The STJ has

dimensions 50 50 mm, and the narrow wirings are 3 mm

wide. (The devices were fabricated at CNR-Istituto di

Cibernetica, Pozzuoli, Naples, Italy.)

1176

Superconducting Radiation Sensor Applications: Detectors and Mixers

the HF signal, which has to be detected, is mixed with

a laboratory local oscillator (LO), at a slightly dif-

ferent frequency, to give a low difference frequency

signal (IF) that can be ampliﬁed and studied by con-

ventional devices. In the output signal the informa-

tion about the frequency and the phase of the

incoming signal is preserved. The spectral resolution

is limited only by the line width of the LO. The con-

version efﬁciency (ratio between the power of incom-

ing signal to the power at IF signal) and the noise

temperature (equivalent noise level added by the

mixer during the conversion process) characterize the

performance of the mixer. Superconducting mixers

have one common advantage over conventional

(Schottky) mixers, namely extremely low LO power

requirements (of the order of nanowatts).

SIS mixers are the technology of choice in the range

40–800 GHz. These receivers are now in daily use in

several millimeter-wavelength telescopes in observa-

tories. SIS mixers are based on the photon-

assisted tunneling process and exploit the strong

nonlinearity of the SIS current–voltage characteris-

tics. SIS mixers exhibit the best performance (conver-

sion efﬁciency close to unity and noise temperature

close to the quantum limit, T

¼ 502170 K). How-

ever, their operation is limited to the frequency range

below the gap frequency of superconductor. This is

about 700 GHz for niobium and slightly above 1 THz

for NbN.

Josephson effect mixers, with photon-assisted tun-

neling of Cooper pairs of electrons as the underlying

operational mechanism, show moderate performance

¼ 120022000 K), but they can be fabricated us-

ing the high transition temperature superconductors

(HTS). HTS allow the use of liquid nitrogen

(T ¼ 77 K) as coolant or cryogen-free cryocoolers,

so that they are specifically indicated for space

applications.

Hot-electron bolometers (HEBs) can also be oper-

ated as mixers in heterodyne receivers. Intrinsically,

frequency-independent performance and no gap lim-

itation predestinates these type of mixers for opera-

tion at very high frequencies, around and well above

1 THz. Two concepts of HEB mixer materials are

exploited: phonon-cooled bolometers (relying typi-

cally on NbN, T

¼ 40022000 K) and diffusion-

cooled HEB (fabricated mostly of Nb, T

¼ 650 K).

4. Conclusions

Superconducting sensors are undergoing con tinuous

evolution. The above overviewed sensors are not ex-

haustive of devices under development. The reader is

referred to the literature for a complete review of the

subject (Booth et al. 1996, De Korte and Peacock 2000)

An interesting perspective for applications is

represented by spectrometry for high-sensitivity

microanalysis. This is due to the fact that both

superconducting microcalorimeters and STJ detec-

tors have surpassed the ultimate energy resolution

of semiconductor detectors (DE ¼ 120 eV at E ¼

6 keV), which are typically used in x-ray ﬂuorescence

energy-dispersive spectrometers (EDS). The im-

proved energy resolution of superconducting sensors

combined with their fast count rate makes possible

the realization of spectrometers having the energy

resolution of Bragg crystals used in wavelength-dis-

persive spectrometers as well as the fast operation

mode and the simplicity of use of EDS-like instru-

ments (Wollman et al. 1997). Furthermore, the de-

tection principle of TES and STJ detectors is heat

transfer rather than ionization. This opens the pos-

sibility of an efﬁcient detection of very slow massive

ions like macromolecules (biopolymers, proteins,

DNA fragments) and neutral particles (neutrons).

The sensitivity allows high throughput, even in the

presence of dilute samples with traces of significant

elements. To exploit this property, time-of-ﬂight mass

spectrometers equipped with superconducting detec-

tors have been proposed (Hilton et al. 1998).

See also: Radiation and Particle Detectors; SQUIDs:

The Instrument; Superconducting Thin Films: Mate-

rials, Preparation, and Properties; Superconducting

Thin Films: Multilayers

Bibliography

Alessandrello A, Beeman J W, Brofﬁero C, Cremonesi O,

Fiorini E, Giuliani A, Haller E E, Monfardini A, Nucciotti

A, Pavan M, Pessina G, Previstali E, Canotti L 1999 High

energy resolution bolometer for nuclear physics and x-ray

spectroscopy. Phys. Rev. Lett. 82, 513–6

Angeloher G, Hettl P, Huber M, Jochum J, Feilitzch F v,

Mossbauer R L 2001 Energy resolution of 12 eV at 5.9 keV

from AI-superconducting tunnel junction detectors. J. Appl.

Phys. 89, 1425–9

Booth N E, Cabrera B, Fiorini E 1996 Low temperature par-

ticle detectors. Annu. Rev. Nucl. Part. Sci. 46, 471–532

De Korte P, Peacock T (eds.) 2000 Proc. 8th Int. Workshop on

Low Temperature Detectors. Nucl. Instr. Methods A 444

Hilton G C, Martinis J M, Wollman D A, Irwin K D, Dulcie L

L, Gerber D, Gillevet P M, Twerenbold D 1998 Impact en-

ergy measurement in time-of-ﬂight mass spectrometry with

cryogenic microcalorimeters. Nature. 391, 672–5

Irwin K D, Hilton G C, Martinis J M, Deiker S, Bergren N,

Nam S W, Rudman D A, Wollman D A 2000 Proc. 8th Int.

Workshop on Low Temperature Detectors. Nucl. Instr. Meth-

ods A 444, 184–7

Qing Hu, Richards P L 1990 Quasiparticle mixers and detec-

tors. In: Ruggiero S T, Rudman D A (eds.) Superconducting

Devices. Academic Press, London, Chap. 5, pp. 169–96

Wollman D A, Irwin K D, Hilton G C, Dulcie L L, Newbury D

E, Martinis J M 1997 High-resolution, energy-dispersive mi-

crocalorimeter spectrometer for x-ray microanalysis. J.

Microscopy 188, 196–222

R. Cristiano

CNR-Institute of Cybernetics, Naples, Italy

1177

Superconducting Radiation Sensor Applications: Detectors and Mixers

Superconducting Thin Films: Materials,

Preparation, and Properties

The crystal structures of high T

superconductors are

described in this article together with the deposition

methods used to produce high-quality ﬁlms of these

materials. This article complements High-temperature

Superconductors: Thin Films and Multilayers.Itis

impossible to include all of the important work con-

tributed by the large number of groups active in this

rapidly progressing ﬁeld. Additional information can

be found in reviews of high T

materials and depo-

sition methods (Humphreys et al . 1990, Eckstein and

Bozovic 1995, Schlom and Harris 1995, Matijasevic

and Bozovic 1996, Fisk and Sarrao 1997, Wo

rden-

weber 1999).

1. Materials

1.1 Crystal Structures and Types of Building Layers

All high T

superconductors have perovskite-related

structures. The crystal structure of perovskite

(CaTiO

) and some of the better-known high T

superconductors are shown in Fig. 1. All high T

Figure 1

Crystal structure of (a) perovskite (CaTiO

) and some of the better known high T

superconductors: (b) (La,Sr)

CuO

, (d) YBa

7d

, (e) Bi

CaCu

, (f) Tl

, and (g) TlBa

. Two

equivalent representations of these crystal structures are shown: the atomic positions (above) and the copper

coordination polyhedrons (below). The oxygen atoms occupy the vertices of the copper coordination polyhedrons.

The tetragonal subcells of the (La,Sr)

CuO

and Bi

CaCu

8 þd

structures are shown for clarity and to illustrate

the similarities between these perovskite-related phases. The approximate superconducting transition temperatures are

also shown.

1178

Superconducting Thin Films: Materials, Preparation, and Properties

structures can be assembled from a relatively small

number of more fundamental layer types (Tokura and

Arima 1990), referred to as building layers (Schlom

and Harris 1995). A common feature of the crystal

structures of all known copper-containing high T

su-

perconductors is the presence of CuO

layers. The

CuO

layers are separated from one another along the

c-axis of the crystal structure by various intervening

layers, the known types of which are shown in Fig. 2.

By stacking the building layers shown in Fig. 2 on top

of one another along the c-axis direction, all known

copper-containing high T

structures can be assembled.

There is only one other known oxide supercon-

ductor with a relatively high T

that does not contain

CuO

layers, (Ba,K)BiO

, which has a T

of about

30 K (Cava et al. 1988). Like the structurally related

solid solution Ba(Pb,Bi)O

with T

of approximately

13 K (Sleight et al. 1975), (Ba,K)BiO

has a simple

cubic perovskite structure, shown in Fig. 1, with bis-

muth occupying the octahedrally coordinated titani-

um position and barium and potassium occupying

the larger calcium site. From a building layer stand-

point, (Ba,K)BiO

consists of alternating BiO

and

(Ba,K)O building layers.

Although high T

phases are commonly referred to

by formulas implying that they are stoichiometric,

e.g., Bi

CaCu

, this is an oversimpliﬁcation. In

reality, significant cation mixing, anion vacancies, or

cation vacancies often occur within the building lay-

ers, with several percent mixing quite common in

some layers, although not within the CuO

layers.

For example, site-sensitive structural reﬁnement of a

single crystal of the high T

superconductor Bi

CaCu

has determined that its overall composition

Figure 2

Building layers of all known layered CuO

structures: (a) [(La,Sr)O–(La,Sr)O] or [(Sr,Ca)(Br,Cl)–(Sr,Ca)(Br,Cl)]*, (b)

[BaO–CuO

1d

–BaO], (c) [BaO–CuO–CuO–BaO], (d) [BaO–TlO–BaO] or [BaO–HgO

–BaO] or [SrO–(Bi,Cu)O

–SrO]

or [SrO–(Bi,Cd)O–SrO] or [SrO–(Pb,Sr)O–SrO] or [SrO–(Pb,Cu)O

–SrO] or [SrO–(Pb,Cd)O–SrO] or [SrO–

(Ce,Cu)O

–SrO]* or [SrO–(Ce,Cd)O–SrO], (e) [BaO–TlO–TlO–BaO] or [BaO–HgO

–HgO

–BaO] or [SrO–BiO–BiO–

SrO] or [(Ba,Sr)O–(Pb,Cu)O

–(Pb,Cu)O

–(Ba,Sr)O], (f) [SrO–PbO–CuO

–PbO–SrO], (g) [LaO–SnO

–LaO]* or

[BaO–TaO

–BaO]* or [SrO–TaO

–SrO] or [SrO–NbO

–SrO], (h) [SrO–GaO–SrO] or [SrO-CoO–SrO]* or [SrO–AlO–

SrO]*, (i) [Sr–CO

–Sr] or [(Re,Ba)–BO

–(Re,Ba)] where Re is a rare earth ion, (j) [(Re,Ca)], (k) [(Ce,Re)–O

–(Ce,Re)],

and (l) [(Sr,Pb)–Cl–(Sr,Pb)]*. The tetragonal subcell of each building layer is outlined. CuO

layers are shown above

and below each building layer to illustrate their attachment positions. Partial substitution for the constituents of these

basic building layers is frequently possible and may be used for doping purposes (e.g., Bi 2 Pb, Tl 2 Pb, Sr 2 Ba,

Sr 2 La, Sr 2 Ca, Ca 2 Re). Most of these building layers are constituents of the copper-containing high T

superconductors discovered to date. Those intervening layers known to occur between CuO

layers, but so far not

constituents of superconducting structures, are marked with an asterisk (reproduced by permission of Noyes

Publications from ‘‘Molecular Beam Epitaxy: Applications to Key Materials,’’ 1995, pp. 505–622).

1179

Superconducting Thin Films: Materials, Preparation, and Properties