Field of the Invention

The present invention relates to a bolometer and to apparatus comprising an array of bolometers.

Background

A bolometer can be used to detect and measure energy (or power) of incident electromagnetic radiation.

Superconductor-based bolometers can exhibit high sensitivity and fast response.

One form of superconductor-based bolometer is a hot-electron bolometer (HEB) which comprises a superconductor and a pair of normal metal contacts. An example of a hot- electron bolometer is described in JP 2004 88852 A.

Another form of superconductor-based bolometer is a cold-electron bolometer (CEB) which comprises a normal metal-insulator-superconductor (NIS) tunnel junction, for example, as described in US 5 634 718 A. The metal region serves as an absorber which is cooled using the tunnel junction.

Yet another form of superconductor-based bolometer is a transition-edge sensor (TES) which is similar to the cold-electron bolometer but which employs direct dc heating. An example of a transition-edge sensor is described in US 2010 304977 Ai.

Although existing forms of bolometers which employ a normal metal and a

superconductor exhibit high sensitivity and fast response, there is scope for improving performance. Summary

According to a first aspect of the present invention there is provided a bolometer. The bolometer comprises a superconductor-insulator-semiconductor-superconductor structure or a superconductor-insulator-semiconductor-insulator-superconduc tor structure. The semiconductor comprises an electron gas in a layer of silicon, germanium or silicon-germanium alloy in which valley degeneracy is at least partially lifted.

Using an electron gas in silicon, germanium or silicon-germanium alloy as the sensing element can help not only to reduce thermal coupling to the sensing element, but also to decrease the thermal capacity of the sensing element. This can help to enhance sensitivity and responsivity.

An insulator may comprise a layer of dielectric material. Additionally or alternatively, an insulator may comprise a layer of non-degenerately doped semiconductor.

In a superconductor-insulator-semiconductor-insulator-superconduc tor structure, the insulators may be of the same type, for example, each comprising a respective layer of dielectric material and/or a respective layer of non-degenerately doped semiconductor. At least part of the insulator may comprise common layer(s).

Using a dielectric layer can help to reduce sub-gap leakage which can improve the sensitivity and cooling power of the junction(s). The layer of silicon, germanium or silicon-germanium may be strained. Thus, valley degeneracy can be at least partially lifted using strain. The layer may be strained by virtue of being formed on a layer of another, different semiconductor, i.e. with a different lattice constant and/or crystal structure. For example, a layer of silicon may be formed on a layer of strain-relaxed silicon-germanium alloy. Strain may be introduced using local strain (or "process-induced" strain). Valley degeneracy may be lifted by confinement of the electron gas. Confinement may occur in a two-dimensional electron gas.

The layer of silicon, germanium or silicon-germanium alloy may comprise a layer of n- type silicon, germanium or silicon-germanium alloy. For example, the layer may be doped with phosphorus (P), arsenic (As) or antinomy (Sb). The layer of silicon, germanium or silicon-germanium maybe doped to a concentration of at least 1 x io ^l8 crrrs or at least 1 x 10 ¹⁹ crrrs, that is, such that the layer is

degenerately doped. The layer may be doped to a concentration of at least 1 x io ²⁰ crrr 3. The layer may be doped to a concentration of up to ι x io ²¹ crrrs or more. The layer may have a thickness no more than 5 nm, no more than 4 nm or no more than 3 nm. Thus, the electron gas maybe a two-dimensional electron gas. However, the layer may be thicker, for example, having a thickness of at least 10 nm or at least 20 nm and may have a thickness up to 50 nm or up to 100 nm or more.

The layer may include a delta-doped layer. The delta-doped layer may be an n-type delta-doped layer, for example doped with phosphorus (P) arsenic (As) or antinomy (Sb). The delta-doped layer may have an areal doping concentration of at least 1 x 10 ¹² cm- ² , at least 5 x 10 ¹² cm- ² , at least 1 x ιο^ cm- ² or at least 2 x ιο^ cm- ² . The delta- doped layer may have a thickness (full width at half maximum) of no more than 3 nm.

The layer of silicon, germanium or silicon-germanium may be undoped. The layer of silicon, germanium or silicon-germanium may be disposed in a structure, for example a heterostructure, which includes a doped layer of semiconductor. Thus, the electron gas may arise as a result of modulation doping. The bolometer may include a gate electrode and a gate dielectric disposed between the gate electrode and the

semiconductor arranged to apply an electric field so as to form the electron gas.

The silicon, germanium or silicon-germanium layer may include a quantum well.

The layer may have an area (i.e. in plan view) of less than 1 μπι ² . The layer may have an area of no more than 1 μπι ² or no more than 0.01 μπι ² . However, the layer may have an area of up to 100 μπι ² or up to 1000 μπι ² or more. The bolometer may include a substrate. The substrate may take the form of a wafer or wafer die of single crystal silicon or single crystal germanium. The bolometer may include one or more buffer layers. The buffer layer(s) may include a graded silicon- germanium layer. The buffer layer(s) may include a silicon-germanium layer. The silicon-germanium layer may be strain relaxed. The insulator may be or include a region of depleted semiconductor. The insulator may be or include a region of low-doped semiconductor which, at temperature of operation of the bolometer, is an insulator. The insulator may comprise a product of a chemical process or deposited material. The dielectric material may comprise or predominantly be an oxide. The dielectric material may comprise or predominantly be a nitride. The dielectric material may comprise or predominantly be an oxynitride. The dielectric material may comprise or predominantly be an oxide of the

semiconductor. Thus, the dielectric material may comprise or predominantly be silicon dioxide or germanium oxide or germanium dioxide.

The dielectric material may comprise or predominantly be an oxide of the

superconductor.

The dielectric material may comprise a mixture of an oxide, nitride, oxynitride or compound of the semiconductor and an oxide, nitride, oxynitride or compound of the superconductor.

The dielectric layer may have a thickness no more than 5 nm, no more than 3 nm or no more than 2 nm.

The superconductor may be an elemental superconductor, such as aluminium (Al), tin (Sn), niobium (Nb), vanadium (V) or tantalum (Ta).

According to a second aspect of the present invention there is provided apparatus comprising a bolometer and circuitry arranged to measure the bolometer comprising a voltage and/ or current bias source and a sensor for measuring for current through the bolometer and/ or for measuring a voltage across the bolometer.

The system may include a system configured to cool the bolometer to a temperature no more than 100 K. The system may include a cryogenic system configured to cool the bolometer to a temperature no more than 4.2 K. The circuitry may be arranged for single photon measurement and/or integrating measurement.

According to a third aspect of the present invention there is provided apparatus comprising an array of bolometers supported on a common substrate.

The system may include a system configured to cool the bolometer to a temperature no more than loo K. The system may include a cryogenic system configured to cool the array of bolometers to a temperature no more than 4.2 K.

The circuitry may be arranged for single photon measurement and/or integrating measurement.

According to a fourth aspect of the present invention there is provided an astronomical detection system (i.e. an imaging system for use in astronomy) comprising the bolometer or the apparatus.

According to a fifth aspect of the present invention there is provided a biomedical imaging system comprising the bolometer or the apparatus.

According to a sixth aspect of the present invention there is provided a security screening system (for example for use in airport for imaging passengers and/or luggage) comprising the bolometer or the apparatus. According to a seventh aspect of the present invention there is provided a remote sensing system (for example, for aerial imaging of terrain) comprising the bolometer or the apparatus.

According to an eighth aspect of the present invention there is provided a quantum information processing system comprising the bolometer or the apparatus. Brief Description of the Drawings

Certain embodiments of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 is schematic block diagram of a detection system which includes a bolometer; Figure 2 is a schematic diagram of a superconductor-insulator-semiconductor- insulator-superconductor (S-I-Sm-I-S) structure;

Figure 3 is a schematic diagram of a superconductor-insulator-semiconductor - superconductor (S-I-Sm-S) structure;

Figure 4 is an isometric view of a bolometer;

Figure 5 is a plan view of the bolometer shown in Figure 4;

Figure 6 is a cross section of the bolometer taken along the line A-A' in Figure 5;

Figures 7a to ηί illustrate a bolometer at different stages during fabrication;

Figure 8 is process flow diagram of a method of fabricating a bolometer;

Figure 9 is a plot of responsivity against voltage;

Figure 10 is a plot of difference in power spectral density against frequency;

Figure 11 is a plot of noise equivalent power against voltage;

Figure 12 illustrates another bolometer;

Figure 13 illustrates a bolometer which employs a lens; and

Figure 14 schematically illustrates apparatus comprising an array of bolometers.

Detailed Description of Certain Embodiments

Referring to Figure 1, a detection system 1 for measuring the energy (or power) of incident radiation in the form of one or more photons 2 in optical, infrared and/or terahertz regions of the electromagnetic spectrum is shown.

The system 1 includes a bolometer 3 (which may be an element in an array) which is cooled to cryogenic temperatures, preferably between 0.5 and 2 K, using a refrigeration system 4, such as a closed-cycle helium-4 system. The bolometer 3 may be provided with a cooling device 5 which may be integrated into the bolometer 3 or which is supported on the same substrate on which the bolometer 3 is formed.

Measurement of photon energy is performed using circuitry 6 which includes a current or voltage bias source 7 which may be a current bias source for driving a constant current through the bolometer 3 or a voltage bias source for applying a constant voltage bias across the bolometer 3, and a voltage or current sensor 8 which may be a voltmeter to measure a voltage across the bolometer 3 or a current meter to measure current flowing through the bolometer 3.

The bolometer 3 may operate in current bias or voltage bias modes.

In current bias mode, a current source 7 drives a constant current through the bolometer 3 and change(s) in voltage across the bolometer 3 is (are) measured using the voltmeter 8 following absorption of photon energy. In voltage bias mode, a voltage source 7 applies a fixed voltage across the bolometer 3 and change(s) in the current flowing through the bolometer 3 is (are) measured arising from photon absorption.

Referring to Figures 2 and 3, the bolometer 3 can employ a superconductor-insulator- semiconductor-insulator-superconductor (S-I-Sm-S) structure 9 ₁ or a superconductor- insulator-semiconductor-insulator-superconductor (S-I-Sm-I-S) structure 9 ₂ .

The structures 91, g ₂ include an electron gas 10 formed in a region (e.g. layer) of semiconductor 11 which is silicon, germanium or silicon-germanium alloy. Valley degeneracy in the semiconductor 11 is at least partially lifted using strain and/ or quantum confinement.

Referring in particular to Figure 2, the first structure 9 ₁ includes a superconductor- insulator-semiconductor junction 12 _! which includes a region of dielectric material 131, which is in direct contact with the semiconductor 11, and a first superconductor contact 141, which is in direct contact with the dielectric 131. The structure 91 includes a superconductor-semiconductor junction 15 which includes a second superconductor contact 142, which is in direct contact with the semiconductor 11. In the first structure g , any depletion region formed at the interface of the

semiconductor 11 dielectric 131 is sufficiently short and/or has a sufficiently low barrier height to be considered not to provide a region of insulator.

Referring also to Figure 3, the second structure 9 ₂ includes first and second

superconductor-insulator-semiconductor junctions 12 _1; 12 ₂ . Each junction 12 _1; 12 ₂ includes a respective region of dielectric material 131, 132 which is in direct contact with the semiconductor 11 and a respective superconductor contact 141, 142 which is in direct contact with the dielectric 131, 132 respectively.

One or both regions of dielectric material 131, 132 maybe omitted. Thus, the insulator may be provided by a depleted region of semiconductor.

As will be explained in more detail hereinafter, the junctions of a structure can be formed on the same surface of the semiconductor 11. As will also be explained in more detail hereinafter, photon energy can be coupled into the electron gas 10 either directly or using via the superconductor contacts 141, 142.

Referring to Figures 4, 5 and 6, a bolometer 3 employing a superconductor-insulator- semiconductor-insulator-superconductor (S-I-Sm-I-S) structure g ₂ will now be described in more detail.

The bolometer 3 includes a substrate 16, a first buffer layer 17 overlying the substrate 16, a second, partially-etched buffer layer 18 overlying the first buffer layer 17 and a semiconductor region 11 supported on the second buffer layer 18. The semiconductor region 11 takes the form of a patterned layer (or "island") of strained silicon having an embedded delta-doped layer 19 which provides the two-dimensional electron gas 10 (Figure 3). The silicon layer 11 has a germanium equivalent strain of about 10 % (which is defined as the strain due to a relaxed silicon germanium alloy layer of 10% Ge content, and is equal to a lattice mismatch of about 0.42 %). However, the silicon layer 11 can have a larger value of equivalent strain, for example, up to 75%. An unetched portion of the second buffer layer 18 and the strained silicon layer 11 generally form a mesa 20 (best shown in Figure 6).

A thick dielectric layer 21 (herein also referred to as a "passivation layer") overlies the partially-etched portion of the second buffer layer 18 and the mesa 20. The thick dielectric layer 21 has first and second windows 22 _1; 22 ₂ on top of the mesa 20 (best shown in Figure 5) on opposite sides of the strained silicon layer 11. In the windows 141, 142 (best shown in Figure 6), first and second thin dielectric layers 131, 132 are formed on the top of the strained silicon layer 11 in the windows 22!, 22 ₂ in the thick dielectric layer 21. The first and/ or second thin dielectric layers 131, 132 may be omitted. In this example, a superconductor antenna 23 in the form of a superconductor twin-slot antenna overlies the dielectric layers 21, 13!, 132. However, other antenna structures can be used. Furthermore, photon energy need not be coupled into the electron gas 10 using an antenna. Other forms of coupling can be used.

Referring in particular to Figure 6, first and second regions of the antenna 23 provide first and second contacts 141, 142 to the silicon layer 11 via the first and second thin dielectric regions 131, 132 respectively. The strained silicon layer 11, the first dielectric regions 131 and the first superconductor region 13! form a first superconducting tunnel junction i2i. The strained silicon layer 11, the second dielectric regions 132 and the second superconductor region 142 form a second superconducting tunnel junction 12 ₂ .

The substrate 16 takes the form of single-crystal (ooi)-orientated silicon wafer or wafer die. However, the silicon wafer or wafer die can have other orientations. The first buffer layer 17 takes the form of a graded layer of silicon-germanium (Sii-xGex) in which germanium content, x, increases along the growth axis, z, and the second buffer layer 12 takes the form of a partially-etched layer of silicon-germanium (Sii-xGex) where germanium content, x, is at least 0.2. Germanium content, x, may be between about 0.2 and 0.5, preferably between 0.2 and 0.3. The strained silicon layer 11 has a thickness of about 30 nm and the delta-doped layer 19 takes the form of a layer of phosphorus having a full width at half maximum thickness no more than 3 nm and having an areal doping density of about 10 ¹³ cm ^-2 . The silicon layer 11 may have a thickness of between 10 and 100 nm. The delta-doped layer 19 is formed a few nanometres from the upper surface of the silicon layer 11. The thick dielectric layer 21 is formed of a layer of silicon dioxide (Si0 ₂ ) having a thickness of about 200 nm. The thin dielectric regions 131, 132 comprise silicon dioxide (Si0 ₂ ) and/or aluminium oxide (AI2O ₃ ) having a thickness less than 3 nm. The superconductor antenna 23 which also provides the superconducting contacts 131, 132, is formed from aluminium (Al).

As shown in Figures 4 and 5, the antenna 23 is patterned with a set of slots including first and second outwardly-facing 'C'-shaped slots 24!, 242 and a cross-connecting slot 25 which defines first and second antenna lead portions 261, 262. The first and second contact regions 141, 142 are disposed close to the distal ends of the lead portions 261, 262. First and second narrow slots 27!, 272 divide the antenna 23 into first and second portions 23!, 232 which are isolated from each other at dc biases. Referring in particular to Figure 5, the silicon layer 11 can be rectangular in plan view having first and second side lengths, s , s ₂ . The sides may be equal, i.e. s = s ₂ = s. The side lengths may lie in a range between 10 nm and 10 μπι, i.e. 10 nm≤ {s , s ₂ }≤ 10 μπι.

The output impedance of the bolometer 3 is about 50 Ω.

Referring to Figures 7a to 7f and 8, a method of fabricating the bolometer 3 will now be described.

Referring in particular to Figure 7a, a heterostructure 30 is prepared by growing epitaxial semiconductors layers 17, 18', 11', 19' on the silicon substrate 16 using chemical vapour deposition or molecular beam epitaxy (step Si). The heterostructure 30 includes a graded layer 17 of silicon germanium alloy (Sii-xGex), a strain-relaxed buffer 18' of silicon germanium alloy (Sii-xGex) followed by a silicon layer 11' containing a thin, degenerately-doped layer 19' in the form of a delta-doped layer (which may also be referred to as a "delta layer").

A layer of photoresist (not shown) is applied to an upper surface 31 of the

heterostructure 30, exposed and developed to provide a mask (not shown) (step S2). The mask (not shown) defines the area to be retained.

In unmasked areas (not shown), the strained silicon layer 11' and a portion of the underlying buffer layer 18' are removed using a dry plasma etching (step S3). In this example, a 10:1 mixture of carbon tetrafluoride (CF ₄ ) and oxygen (0 ₂ ) is used for etching, at a pressure of 30 mTorr (4 Pa) and a power of 100 W.

Referring in particular to Figure 7b, etching defines the mesa 20 which includes the strained silicon island 11 with a fixed volume of electron gas 10 in delta-doped layer 19 and leaves upper and side surfaces 32, 33 of a partially-etched silicon-germanium layer 18 and upper and side surfaces 34, 35 of the strained silicon layer 11.

Referring in particular to Figure 7c, a conformal layer 21' of silicon dioxide (or

"passivation layer") is deposited, for example, by chemical vapour deposition, to protect the upper and side surfaces 34, 35 of the strained silicon layer 11 (step S4). The passivation layer 21' allows contact windows 22 _1; 22 ₂ to be defined and a dry plasma etch to be used when patterning a subsequently-deposited aluminium layer 23' (Figure 7O without damaging the underlying silicon.

A layer of photoresist (not shown) is applied to an upper surface 36 of the passivation layer 21', exposed and developed to provide a mask (not shown) (step S5). The mask (not shown) defines the area to be retained.

Unmasked areas (not shown) of the passivation layer 21' are removed using a wet etch (step S6). In this example, buffered hydrofluoric acid is used as an etchant.

Referring in particular to Figure 7d, etching defines windows 22!, 22 ₂ in the passivation layer 21.

Contacts are made to the island 11 by depositing and patterning an aluminium layer 23' (Figure 5f). A process is used whereby the silicon surface is cleaned by heating to elevated temperatures and is then oxidised in a vacuum chamber (not shown) immediately prior to depositing the aluminium layer 23' (Figure i).

Referring in particular to Figure ye, the work piece 37 is heated to 550 °C in a vacuum (step S7) and then thin layers 131, 132 of silicon dioxide are formed at the surface 34 of the silicon layer in the windows 22, 22 ₂ in the passivation layer 21 by dry oxidation (step S8). However, the oxidation step may be omitted.

Referring in particular to Figure 7f, a layer 23' of aluminium is deposited, for example by sputtering, over the thin oxide layers 131, 132 and the surface 36 of the passivation layer 21 (step S9).

A layer of photoresist (not shown) is applied to the upper surface 37 of the aluminium layer 23', exposed and developed (step S10). Unmasked areas (not shown) of the aluminium layer 23' are removed using a plasma etch (step S11).

The resulting structure is shown in Figure 6.

Referring to Figures 1 to 6, the bolometer 3 which employs an electron gas 10 as an absorber can exhibit one or more advantages over bolometers which using a metal absorber. First, electrons in silicon (or germanium) can have a much weaker thermal link to the lattice compared with electrons in a normal metal. Secondly, using a thin doped layer, particularly a delta-doped layer, can result in a small volume of electrons having small thermal capacity.

Each of these can enhance sensitivity and responsivity. Moreover, formation of thin dielectric layers, for example, layers of silicon dioxide, can enhance cooling power of the bolometer.

Experiments are conducted using a bolometer which is similar to that hereinbefore described but which has an unstrained, undoped silicon absorber and does not include thin, silicon dioxide layers. The antenna is designed to couple 150 GHz (0.15 THz) radiation to the absorber. Radiation at this frequency is generally considered to be low energy and places demands on a radiation detector.

Results for the bolometer at 220 mK, in the current bias mode, are shown in Figures 9, 10 and 11. The results are taken for the bolometer operating in hot-electron mode.

Referring in particular to Figure 9, the bolometer exhibits peak responsivity of order 10 ⁸ VW ^"1 (corresponding to maximum e-cooling) Referring in particular to Figure 10, the bolometer exhibits a response time of less than 1.5 μs c.

Finally, referring to Figure 11, the bolometer exhibits a (dark) noise equivalent power (NEP) of 5 x 10- ¹⁸ WHz- ^o -5.

These results compare favourably with existing metal-based bolometers. The bolometer 3 illustrated in Figures 1 to 6 is expected to exhibit even better

characteristics due to the strained silicon layer, the thin doped layer (particularly a delta-doped layer) and the semiconductor-insulator-superconductor tunnel junction. As mentioned earlier, strain can be used to lift valley degeneracy. Additionally or alternatively, quantum confinement can be used lift valley degeneracy.

Referring to Figure 12, a bolometer 3' is shown which is similar to that described with reference to Figures 3, 4 and 5, but which differs in that it employs an inverted modulation doping heterostructure 41 to form a two-dimensional electron gas 10.

The heterostructure 41 includes p-type silicon substrate 42 and (in order from the substrate 42) a 1 μιη-thick, undoped (N _D = 1 x 10 ¹⁶ crrrs) graded layer 43 of silicon- germanium alloy having germanium content which increases from o to 0.4, a 10 nm- thick undoped layer 44 of silicon-germanium alloy having a germanium content of 0.4, a 50 nm-thick layer 45 of doped (N _D = 6 x 10 ¹⁸ crrrs) silicon-germanium alloy having a germanium content of 0.4, a 3 nm-thick undoped layer 46 of silicon-germanium alloy having a germanium content of 0.4 and a 5 nm-thick layer 47 of undoped (N _D = 1 x 10 ¹⁶ cm-3) silicon. This results in the conduction band edge, ECB, forming a 'V'-shaped potential 48 in the silicon layer 47 in which a two-dimensional electron gas 10 forms close to the Fermi energy, E _F , and which has a sheet carrier concentration of about 1 x 10 ¹¹ crrr ² . In the examples given earlier, an antenna structure is used to couple photon energy into the electron gas 10. However, photon energy can be coupled directly into the electron gas 10.

Referring to Figure 13, the bolometer 3 may be provided, for example, with a lens 51 which is arranged to focus incident radiation 2 onto the electron gas 10. The antenna structure can be omitted, although superconducting contacts 141, 142 are still used.

Referring to Figure 14, an (n x m) array 58 of bolometers 3, 3' is shown. The array 58 can be formed on a common substrate 59 (i.e. silicon substrate) and can include processing circuitry 60 row and column decoders for addressing each bolometer 3, 3' separately, amplifiers and processors. The array 58 may include 1000 x 1000 or more bolometers 3. An aluminium-based bolometer 3 can be operated at different phonon temperatures depending on the required performance for a specific application. A 300 mK phonon temperature can be obtained using a closed-cycle turn-key 4He/ ³ He system and using the S-I-Sm tunnel junction to cool the absorber further, for example, to 30 mK.

A vanadium-based bolometer 3 can be cooled using a simpler 1 K platform and using S- I-Sm tunnel junction to cool the absorber further, for example, to 300 mK or even lower. The bolometer 3 and bolometer array 58 can be used in a variety of different applications including astronomical detection, biomedical imaging, security screening, remote sensing and quantum information processing.

It will be appreciated that many modifications may be made to the embodiments hereinbefore described.

The dielectric layer may be omitted. Thus, the insulator may take the form of a region of non-degenerately-doped semiconductor, for example undoped or low-doped semiconductor which is depleted, interposed between the superconductor and the degenerately-doped semiconductor, e.g. the delta-doped layer.

If strained silicon is used, silicon may have a value of equivalent strain of between 10 % and 75%. A strained layer of silicon need not be used as the semiconductor. Instead, a strained layer of germanium (Ge) or silicon-germanium alloy can be used.

The electron gas need not be a two-dimensional electron gas a thin, three-dimensional electron gas may be used. Other thin, highly-doped doping profiles may be used.

Aluminium need not be used as the superconductor. Instead, other superconducting materials can be used such as, for example, tin (Sn), niobium (Nb), vanadium (V) or tantalum (Ta).

Silicon dioxide need not be used as the thin dielectric layer. - ι ₅ -

The layers can have thicknesses other than those specified. Suitable thicknesses found by routine experiment.