Switch for use under cryogenic conditions, and method of manufacturing such a switch

Abstract

A switch is used to control propagation of electric signals under cryogenically cooled conditions. The switch comprises a substrate (301) forming mechanical support and a thermally and electrically insulating membrane section (305) separated from the substrate (301) by a void. Supported by the membrane section (305) are a signal line (306) comprising first superconductive material and a heater (307) comprising heater material (308) that is resistively conductive under the cryogenically cooled conditions. Mechanically supported by the substrate (301) are signal connections (309, 310) for coupling a signal to the signal line (306) and heater connections (311, 312) for coupling an electric current to the heater (307).

Description

SWITCH FOR USE UNDER CRYOGENIC CONDITIONS , AND METHODOF MANUFACTURING SUCH A SWITCHTECHNICAL FIELD

[0001] The following disclosure is generally related to hardware solutions for information processing . In particular, the disclosure is related to hardware and methods that can be used to control the propagation of signals under cryogenically cooled conditions .BACKGROUND

[0002] In the following, quantum processing is used as an example of a discipline that benefits from the ability to control the propagation of signals in cryogenically cooled environments . However, the following disclosure is also applicable to other kinds of information processing .

[0003] For quantum processing to be possible , the quantum processing circuits must be cooled to very low temperatures , such as close to or even below 10 millikelvins . This is accomplished by thermally coupling the quantum processing circuits to a high-performance cooling apparatus inside a cryostat . While some types of quantum processing, such as the use of spin qubits , may allow operating at temperatures at 1 K or above , the temperatures involved are still cold enough to require the use of cryostats , which then leads to certain challenges related to the hardware used .

[0004] Switches are used for a large variety of purposes in association with quantum processing . As an example , for device testing purposes it is common practice to use switches to selectively route signals to / fromdifferent input / output channels to reduce the number of control / readout lines from room temperature down to the cryogenic stage hosting the devices . This can significantly speed up mass testing of such devices by enabling a larger number of devices per single cooldown of the cryostat . Such routing can be used to address in- puts / outputs of different individual devices under test simultaneously installed in a cryostat , or to address separate input / output channels in a single device , such as a large quantum processing unit with several channels corresponding to different qubits or groups of qubits . The tests may be for example screening tests of several qubit test structures or small QPUs , or pre-screening test of larger QPUs enabling addressing, for example , one or a few qubits / couplers alleviating the connectivity needs of the cryostat . This may be very beneficial as for large superconducting quantum computing systems connectivity and related cryogenic infrastructure can be a substantial cost-driving effect .

[0005] Fig . 1 illustrates a prior art system, some parts of which are located in the cryogenically cooled domain 101 while others are in the room-temperature domain 102 . A number of devices to be tested, li ke device 103 for example , are in the cryogenically cooled domain 101 . Signal processing electronics 104 , which produce the signals used in the test and analyses the results , are in the room temperature domain 102 . A first array of switches 105 for controlling the propagation of input signals to the devices is also located in the room temperature domain, as is a second array of switches 106 for controlling the propagation of output signals from the devices . As a part of its operation, the signalprocessing electronics part 104 controls also the switches in the first and second arrays of switches 105 and 106 .

[0006] While the solution of fig . 1 has advantages in the form of avoiding excessive heat loads to the cooling mechanisms of the cryostat , namely those that would result from operating the switches , it may prove costly if the number of signal lines , particularly RF signal lines , to and / or from the cryogenically cooled domain is large .SUMMARY

[0007] This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description . This summary is not intended to identify key features or essential features of the claimed subj ect matter, nor is it intended to be used to limit the scope of the claimed subj ect matter .

[0008] It is an obj ective to provide a switch that can be used to control the propagation of electric signals under cryogenically cooled conditions with improved characteristics compared to the prior art solutions described above . Another obj ective is to provide a quantum computing system utilising one or a plurality of such switches . A further obj ective is to provide a method for manufacturing such switches .

[0009] According to a first aspect , there is provided a switch for use to control propagation of electric signals under cryogenically cooled conditions . The switch comprises a substrate forming mechanical support and a thermally and electrically insulating membrane section separated from the substrate by a void .Supported by the membrane section are a signal line comprising first superconductive material and a heater comprising heater material that is resistively conductive under the cryogenically cooled conditions . Mechanically supported by the substrate are signal connections for coupling a signal to the signal line and heater connections for coupling an electric current to the heater .

[0010] According to an embodiment , the membrane section spans the void unpatterned . This involves at least the advantage that the structure is relatively easy to manufacture .

[0011] According to an embodiment , the membrane section is patterned to comprise a first edge portion, a second edge portion, and a middle portion between the first and second edge portions . The cross section for heat conduction of the first edge portion and the cross section for heat conduction of the second edge portion may then both be smaller than the cross section for heat conduction of the middle portion, all said cross sections being taken perpendicular to a longitudinal direction along the membrane section between the first and second edge portions . The signal line may be coupled to the signal connections along respective parts of the first and / or second edge portions . The heater may be coupled to the heater connections along respective parts of the first and / or second edge portions . This involves at least the advantage that the conduction of heat from the heater to surroundings of the switch can be reduced .

[0012] According to an embodiment , the membrane section has the form of a letter H, so that the first edge portion is formed by the vertical branches of the H atone extremity, the second edge portion is formed by the vertical branches of the H at the other extremity, and the middle portion is formed by the cross line of the H . This involves at least the advantage that necessary parts of the membrane section may be utili zed for practical purposes that serve the structure and operation of the switch .

[0013] According to an embodiment , the first superconductive material has a first critical temperature and the switch comprises ( i ) couplings of the signal line to the signal connections through sections of second superconductive material that has a second critical temperature , higher than the first critical temperature , and / or ( ii ) couplings of the heater material to the heater connections through sections of the second superconductive material . This involves at least the advantage that important parts of the switch may remain superconductive even when the first superconductive material is made ohmically conductive by heating .

[0014] According to an embodiment , a superconductive ground layer covers at least a part of a surface of the substrate . One or more parts of the superconductive ground layer may then extend on the membrane section, following the path of the signal line . This involves at least the advantage that the characteristics of the signal line with respect to the propagation of radio frequency signals can be tailored in many ways .

[0015] According to an embodiment , at least those parts of the superconductive ground layer that extend on the membrane section comprise superconductive material the critical temperature of which is higher than the first critical temperature . This involves at leastthe advantage that the tailored characteristics of the signal line with respect to the propagation of radio frequency signals may remain practically unaffected even when the first superconductive material is made ohmically conductive by heating .

[0016] According to an embodiment , the signal line and those parts of the superconductive ground layer that extend on the membrane section constitute a coplanar waveguide . This involves at least the advantage that the switch may be used to control the propagation of radio frequency signals while maintaining easily tuneable characteristics of the switch and its signal line .

[0017] According to an embodiment , the void is a cavity through the substrate , opening onto an opposite side of the substrate than the side on which the membrane section is . This involves at least the advantage that certain well-known and widely available manufacturing methods can be used .

[0018] According to an embodiment , the void is a hollow in the substrate below the membrane section . This involves at least the advantage that the switch does not reserve any substrate surface from the other side of the substrate .

[0019] According to an embodiment , the switch comprises a sacrificial layer on a surface of the substrate and the void is a hollow formed in the sacrificial layer . This involves at least the advantage that the substrate can be left whole at the location of the switch, and certain well-known and widely available manufacturing methods may be used to produce the void .

[0020] According to an embodiment , the void is a first void, the membrane section is a first membrane sectionseparated from the substrate by the first void, the signal line is a first signal line , and the heater is a first heater . The switch may then comprise a thermally and electrically insulating second membrane section separated from the substrate by a second void . Additionally, the switch may comprise , supported by the second membrane section, a second signal line , comprising the same first superconductive material , and a second heater comprising the same heater material . The switch may further comprise , mechanically supported by the substrate , one or more serially connecting signal connections for coupling signals through the first and second signal lines in series . The switch may comprise , mechanically supported by the substrate , one or more serially connecting heater connections for coupling electric currents through the first and second heaters in series . This involves at least the advantage that an improved on / off ratio can be obtained .

[0021] According to an embodiment , the signal line is a first signal line and the heater is a first heater . The switch may then comprise a thermally and electrically insulating second membrane section separated from the substrate by the void . Additionally, the switch may comprise , supported by the second membrane section, a second signal line , comprising the same first superconductive material , and a second heater comprising the same heater material . The switch may further comprise one or more serially connecting signal connections for coupling signals through the first and second signal lines in series . The switch may comprise one or more serially connecting heater connections for coupling electric currents through the first and second heatersin series . This involves at least the advantage that an improved on / off ratio can be obtained without making further voids in the substrate .

[0022] According to an embodiment , the switch may comprise , supported by the membrane section, at least two overlapping layers comprising any of the superconductive materials electrically insulated from each other by one or more dielectric layers to form a capacitor . This involves at least the advantage that the electrical characteristics of the switch structure can be tuned to include more than j ust switching .

[0023] According to a second aspect , there is provided a quantum computing system comprising a cryogenically cooled domain, a room-temperature domain, and at least one switch of a kind described above in the cryogenically cooled domain .

[0024] According to a third aspect , there is provided a method for manufacturing a switch for use to control propagation of electric signals under cryogenically cooled conditions . The method comprises producing a membrane section on a surface of a substrate , the membrane section comprising material that is thermally and electrically insulating under the cryogenically cooled conditions , and producing a signal line and a heater on the membrane section, of which the signal line comprises first superconductive material that has a first critical temperature and the heater comprises heater material that is resistively conductive under the cryogenically cooled conditions . The method comprises producing signal connections for coupling a signal to the signal line and heater connections for coupling an electric current tothe heater, and arranging for a void to separate the membrane section from the substrate .BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings , which are included to provide a further understanding of the disclosure and constitute a part of this specification, illustrate practical embodiments and together with the description help to explain the principles of the disclosure . In the drawings : figure 1 i llustrates a prior art approach to controlling the propagation of signals , figure 2 illustrates an approach to controlling the propagation of signals , figure 3 illustrates a switch for use in cryogenically cooled environments , figure 4 illustrates an embodiment of a switch, figure 5 illustrates an embodiment of a switch, figure 6 illustrates an embodiment of a switch, figure 7 illustrates an embodiment of a switch, figure 8 illustrates an embodiment of a switch, figure 9 illustrates an embodiment of a switch, figure 10 illustrates an embodiment of a switch, figure 11 illustrates an embodiment of a switch, figure 12 illustrates an embodiment of a switch, figure 13 illustrates an embodiment of a switch, figure 14 illustrates an embodiment of a switch, figure 15 illustrates an embodiment of a switch, figure 16 illustrates an embodiment of a switchfigure 17 illustrates a circuit in which switches of the described kind may be used, figure 18 illustrates a circuit in which switches of the described kind may be used . figure 19 illustrates a circuit in which switches of the described kind may be used, figure 20 illustrates a circuit in which switches of the described kind may be used . figure 21 illustrates a circuit in which switches of the described kind may be used, figure 22 illustrates a method, figure 23 illustrates a method, and figure 24 illustrates a method .DETAILED DESCRIPTION

[0026] In the following description, reference is made to the accompanying drawings , which form part of the disclosure , and in which are shown, by way of illustration, specific aspects in which the present disclosure may be placed . It is understood that other aspects may be utilised, and structural or logical changes may be made without departing from the scope of the present disclosure . The following detailed description, therefore , is not to be taken in a limiting sense , as the scope of the present disclosure is defined in the appended claims .

[0027] For instance , it is understood that a disclosure in connection with a described method may also hold true for a corresponding device or system configured to perform the method and vice versa . For example , if a specific method step is described, a corresponding device may include a unit to perform the described method step, even if such unit is not explicitly described or illustrated in the figures . On the other hand, forexample , if a specific apparatus is described based on functional units , a corresponding method may include a step performing the described functionality, even if such step is not explicitly described or illustrated in the figures . Further, it is understood that the features of the various example aspects described herein may be combined with each other, unless specifically noted otherwise .

[0028] Fig . 2 illustrates a quantum processing system, some parts of which are located in a cryogenically cooled domain 201 while other parts are in a surrounding or adj acent room-temperature domain 202 .

[0029] According to a definition accepted by the 13th HR International Congress of Refrigeration, held in Washington DC in 1971 , a controlled environment equipped for being held at a temperature below 120 K under operation may be considered cryogenically cooled . When quantum processing is involved, the cryogenically cooled domain 201 in fig . 2 refers to a controlled environment , at least some parts of which are built and equipped for being held at a temperature below 120 K, preferably below 4 K, more preferably below 1 K, and most preferably below 100 mK under operation .

[0030] The cryogenically cooled domain 201 may consist of different parts built and equipped for being held at different cryogenically cooled temperatures under operation . As a non-limiting example , one or more mechanical coolers such as pulse tubes may be provided for establishing and maintaining higher temperature levels , if any, of the cryogenically cooled domain 201 . As another non-limiting example , one or more dilution refrigerators may be provided for establishing and maintaining lowertemperature levels , if any, of the cryogenically cooled domain 201 . Assuming these non-limiting examples , there may be a first part of the cryogenically cooled domain 201 built and equipped for being held at 50 - 70 K; a second part built and equipped for being held at about 4 K; a third part built and equipped for being held at about 1 K; a fourth part built and equipped for being held at about 100 mK; and a fifth part built and equipped for being held at about 10 mK under operation .

[0031] Referring to the room-temperature domain 202 does not mean that it should be built and equipped for operating at temperatures that a human user would consider comfortable in a room . In the relevant field of technology, the term "room-temperature" is commonly used to designate other than cryogenically cooled environments . As an illustrative but non-limiting example , the distinction between the cryogenically cooled domain 201 and the room-temperature domain 202 may be made so that in the former it is relatively straightforward to cool various superconductor materials under their critical temperatures , while in the latter all electric conductor materials are in their state of ohmic conduction . It is not excluded, though, that at least some rare high- temperature superconductor materials could be in their superconductive state even in the room-temperature environment 202 .

[0032] In the arrangement shown in fig . 2 , one or more devices are located in the cryogenically cooled domain 201 . The N : th device 203 is singled out as an example . The number, structure , and purpose of devices is immaterial to the present description, except for there being one or more such signal lines to and / or from atleast one of the devices along which controllable switching is desirable . An example of three such signal lines is shown with reference designator 204 in fig . 2 .

[0033] Signal lines of the kind meant here are meant for the routing of electric signals . One or more of the signal lines may comprise sections that under operation are in a state of ohmic conduction . One or more of the signal lines may comprise sections that under operation are in a superconductive state . One or more of the signal lines may be designed at least partly for the routing of DC electric signals . One or more of the signal lines may be designed at least partly for the routing of high- frequency electric signals , such as signals with their frequency in several megahertz , several gigahertz , or dozens of gigahertz .

[0034] Controllable switching along the signal lines means the ability to use one or more control signals to affect the extent to which the electric signals may propagate through the signal lines . A simple example of controllable switching is single-channel on-off switching, in which a control signal determines whether an electric signal may pass through an individual signal line or not . Arrays of on-off switches may be used to implement more complicated switching functions , such as multi-channel on-off switching, signal branching, signal multiplexing, signal summing, and the like .

[0035] In fig . 2 , one or more switches for implementing such switching are located in the cryogenically cooled domain 201 . As a non-limiting example , two switching blocks 205 and 206 are shown in fig . 2 . Assuming a generally counterclockwise direction of signal flows in fig . 2 , the first switching block 205 may beprimarily for performing controllable switching of signals directed from signal-processing electronics 207 in the room-temperature domain 202 towards one or more of the devices 203 in the cryogenically cooled domain 201 . Correspondingly, the second switching block 206 may be primarily for performing controllable switching of signals directed from one or more of the devices 203 in the cryogenically cooled domain 201 towards the signal-pro- cessing electronics 207 in the room-temperature domain202 . In addition to , or in place of , switches at locations shown by the two switching blocks 205 and 206 there may be one or more switches along signal lines within one or more of the devices 203 themselves , and / or along signal lines between at least some of the devices203 .

[0036] According to a principle schematically shown in fig . 2 , control signals for controlling the switches mentioned above come from the signal-processing electronics 207 in the room-temperature domain 202 . This is only an assumption shown in fig . 2 for illustrative purposes . For the detailed description later, it is not significant where the control signals come from . As an example , one or more control signals for one or more switches may come from one or more of the devices 203 in the cryogenically cooled domain 201 . As another example , one or more control signals for one or more switches may come from one or more other switches included in any of the switching blocks 205 and 206 .

[0037] Placing switches in the cryogenically cooled domain 201 like in fig . 2 , and not ( or at least not exclusively) to the room-temperature domain as in fig . 1 , involves at least the advantage of allowing areduction in the number of signal lines that must be built between the room-temperature domain and the cryogenically cooled domain . In the simplified examples of figs . 1 and 2 , there are four signal lines for input signals produced by the signal-processing electronics . These must be controllably conveyed to the devices in the cryogenically cooled domain through any desired combination of a total of 3 *N input signal lines . In the solution of fig . 1 , all such 3 *N input signal lines must pass from the room-temperature domain 102 to the cryogenically cooled domain 101 . In the solution of fig . 2 , only the original four signal lines plus a control line (or control bus ) must pass from the room-temperature domain 202 to the cryogenically cooled domain 201 .

[0038] According to an example , switches in the cryogenically cooled domain 201 ( such as those included in any of the switching blocks 205 and 206 ) may comprise mechanical switches utili zing magnetic actuation . However, such switches involve a disadvantage in the form of generating a relatively large amount of dissipated heat at the moment of performing the switching action . With cryostats , cooling technologies , and magnetically actuated mechanical switches of the kind known at the time of writing this description, it is not uncommon that performing a switching action may raise the temperature of the coldest parts of the cryostat enough to require a subsequent re-cooling phase of several tens of minutes before all operations related to quantum processing can continue .

[0039] According to an example , switches in the cryogenically cooled domain 201 ( such as those included in any of the switching blocks 205 and 206 ) may compriseCMOS-based switches . Taken the technologies known at the time of writing this description, CMOS-based switches may produce less waste heat than magnetically actuated mechanical switches . However, not even CMOS-based switches are completely free of the waste heat problem . Additionally, CMOS-based switches may involve a disadvantage in the form of not allowing the switching of large currents when used in cryogenically cooled environments , at least not with any reasonable levels of dissipated heat . In applications meant here , currents are large if they are in the mill iampere region, like between 0 . 1 and 10 mA . Such large currents may be needed in superconducting qubit systems for magnetic flux biasing of qubit and / or coupler structures , for example . In some superconducting quantum computing systems dynamical magnetic flux control of qubit couplings needs signals with a frequency band from DC to a maximum frequency, which last-mentioned may typically be in the order of 1 GHz .

[0040] Fig . 3 illustrates schematically a switch for use to control propagation of electric signals under cryogenically cooled conditions . According to an embodiment , at least some of the switches in the switching blocks of a system as the one shown in fig . 2 could be built according to this principle .

[0041] The switch in fig . 3 comprises a substrate 301 that forms mechanical support . The substrate 301 may comprise any solid material suitable for use as a structural base on which circuit elements may be manufactured with known methods , such as photolithographic methods and physio-chemical processes . The substrate 301 may comprise a piece of crystalline silicon or sapphire cutfrom a silicon or sapphire wafer, for example. The substrate 301 may consist of a pure substance and or it may comprise layers and / or intentionally implanted impurities. For the purposes of the present description, it is sufficient to consider the substrate 301 as something that offers an essentially planar surface on which circuit elements can be built in a way that makes them withstand vacuum and cryogenically cooled conditions.

[0042] The switch comprises a thermally and electrically insulating membrane section 305 that is separated from the substrate 301 by a void. Fig. 3 illustrates an example of such a void schematically as a cavity 302 in the substrate 301, so that the membrane section 305 spans the cavity 302. Other ways of making a structure in which a void separates a membrane section from the substrate are discussed later in this text.

[0043] As the switch is meant for use under cryogenically cooled conditions, the membrane section 305 should have thermally and electrically insulating characteristics at least under such conditions. As a comparison, the characteristic thermal conductivity of a material under cryogenically cooled conditions may be considered high if it is, for example, at least 100 W / (m*K) at or above 10 K, at least 10 W / (m*K) at 1 K, at least 1 W / (m*K) at 0.1 K, or at least 0.1 W / (m*K) at 0.01 K. A characteristic thermal conductivity of a material under cryogenically cooled conditions may be considered low but finite if it is, for example, between 2 and 50 W / (m*K) at 100 K, between 0.2 and 5 W / (m*K) at 10 K, between 0.03 and 0.75 W / (m*K) at 1 K, between 0.003 and 0.075 W / (m*K) at 0.1 K, and between 0.0003 and 0.0075 W / (m*K) at 0.01 K. A material may be consideredthermally insulating under cryogenically cooled conditions at least if its thermal conductivity is below any of the values listed above . Thermal conductivity of insulator materials such as silicon dioxide , silicon nitride , or aluminium oxide may be as low as 0 . 0001 W / (m*K) or even lower below 0 . 1 K .

[0044] Supported by the membrane section 305 is a signal line 306 that comprises superconductive material . For purposes of unambiguous reference later in this text , the material comprised in the signal line 306 may be called the first superconductive material . Again, as the switch is meant for use under cryogenically cooled conditions , designating a material as superconductive means that the material has a critical temperature in the temperature range considered to occur only under the effect of cryogenic cooling . Under the critical temperature the material becomes superconductive . Concerning the first superconductive material , its critical temperature may be called the first critical temperature for unambiguous reference .

[0045] Also supported by the membrane section 305 is a heater 307 that comprises heater material 308 . The heater material 308 has the characteristic of being resistively conductive under the cryogenically cooled conditions where the switch is to be used . In particular, the heater material 308 should be resistively conductive sufficiently far below the first critical temperature . This may mean, for example , that the heater material 308 would not become superconductive at any practically reachable low temperature . Another possibility is that the heater material 308 would only become superconductive at a temperature that is significantly lower thanthe first critical temperature , like at least a few kelvins below the first critical temperature . As the name implies , one should be able to selectively use the heater material 308 as a heater by making an electric current flow therethrough, generating a desired amount of heat through resistive dissipation . For this reason, if there is a critical temperature below which the heater material 308 becomes superconductive , it should most advantageously be lower than all typical temperatures at which the switch of fig . 3 is meant to be used .

[0046] The switch comprises signal connections 309 and 310 for coupling a signal to the signal line 306 , as well as heater connections 311 and 312 for coupling an electric current to the heater 307 . The substrate 301 acts as a mechanical support for the signal and heater connections 309 , 310 , 311 , and 312 , for example so that the same membrane that forms the membrane section 305 continues onto the surface of the substrate , the signal and heater connections being conductor lines on top of the membrane .

[0047] In the schematic representation in fig . 3 , the signal connections 309 and 310 are mere extremities of the signal line 306 , extending beyond the hori zontal dimension of the cavity 302 . Correspondingly, the heater connections 311 and 312 are pieces of electrically conductive material extending from the heater 308 beyond the hori zontal dimension of the cavity 302 . The implementation shown here is only an example , as at least one of the signal connections and / or heater connections could be implemented otherwise , for example by using wire bonding or some other coupling technique that canbe used in circuits used to build parts of quantum processing systems .

[0048] A switch, the general structure and operating principle of which follow those shown in fig . 3 , may be used as follows . An electric signal , the propagation of which is to be controlled, is coupled to flow through the signal line 306 . I f the switch is under cryogenically cooled conditions and hence cooled below the first critical temperature , the signal line 306 is superconductive and the electric signal may flow therethrough as such, without any loss . In other words , concerning the electric signal the switch is closed or in an on- state . I f then an electric current of sufficient magnitude is coupled to flow through the heater 307 , resistive losses therein produce a local anomaly in temperature , heating at least a part of the signal line 306 above the first critical temperature . As a result , the signal line 306 becomes resistively conductive , which means essentially non-conductive compared to its previous superconductive state . When that happens , concerning the electric signal the switch is open or in an off- state . The switching speed is affected by the heat capacity C of the membrane , causing a switching transient with a length of the order C / G, where G is the thermal conductance from the membrane section 305 to the substrate 301 .

[0049] The electric current that is made to flow through the heater 307 to open the switch may be called a control signal . It may be a DC (direct current ) signal , an AC ( alternating current ) signal , an RF ( radio frequency) signal , or a microwave signal with a frequency above the thermal cutoff frequency 1 / ( 27tC / G) .

[0050] Even if the appropriate operation of the switch depends on generating heat, the amount of heat generated is minimal and will only create a negligible heat load to the mechanism (s) responsible for maintaining the temperature (s) of the coldest parts in the cryogenically cooled environment. As the membrane section 305 is made of thermally insulating material and has a very small heat-conducting cross section, the heat generated through resistive loss in the heater 307 remains quite localized and only warms up its very immediate surroundings, including a part of the signal line 306. The task of properly maintaining the cryogenically cooled environment cold necessitates, in any case, establishing and maintaining a vacuum at and around the cryogenically cooled circuits and components, so the cavity 302 ensures that there will be no vertical conduction of heat from the heater 307 downwards to the substrate 301.

[0051] Some rough calculations can be made concerning the estimated amount of the heat load. Examples of materials of the membrane section 305 include but are not limited to silicon nitrides SiNx, silicon dioxide SiCy, aluminium oxides AlOx, and cadmium germanium di-arsenide CdGeAs2- At temperatures below 100 mK, all of these may be assumed to have a characteristic thermal conductivity in the order of K = 10“6W / (cm*K) or less. If the membrane section 305 is 50 micrometres wide (width w = 50*10-6m) and 300 nanometres thick (thickness d = 0.3*10-6m) and there are 500 micrometres long "end" portions of membrane (length L = 500*10-6m) between the middle portion (the one with the heater 307) and each edge of the cavity 302, we get a thermal conductivity 2* (wd / L)K = 6 pW / K where the factor 2 comes from the fact that there aretwo 500 micrometres long "end" portions of membrane . Assuming that the first critical temperature , to which the superconductive signal line 306 must be heated to open the switch, is at most 4 K, the switching power is about 24 pW in the off-state ( and 0 in the on-state ) of the switch . The cooling power at temperatures below 50 mK of a cryostat may be several microwatts or even tens of microwatts , and typical qubit driving signals may generate heat loads in the order of some nanowatts or more . This means that the heat load generated by a switch of a kind described here is on a quite acceptable level for many purposes .

[0052] The thermal conductivity between the middle portion of the membrane section can be made even smaller by making parts of the membrane section smaller in cross section . Fig . 4 is a top view of a switch according to an embodiment , in which the membrane section 305 comprises a first edge portion 401 , a second edge portion402 , and a middle portion 403 between these two . The cross section for heat conduction of the first edge portion 401 and the cross section for heat conduction of the second edge portion 402 are both smaller than the cross section for heat conduction of the middle portion403 . All cross sections are here taken perpendicular to a longitudinal direction along the membrane section 305 between the first and second edge portions .

[0053] In the embodiment of fig . 4 , the signal line 306 is coupled to the signal connections 309 and 310 along respective parts of the first 401 and / or second 402 edge portions . The heater 307 is correspondingly coupled to the heater connections 311 and 312 along respective parts of the first 401 and / or second 402 edgeportions . In other words , at the first and second edge portions 401 and 402 there is only j ust enough of the membrane material (measured in the transverse direction) to support the signal connections and heater connections .

[0054] Edge portions , the cross section of which for heat conduction are small , can be made with various geometric forms . In the embodiment of fig . 4 , the membrane section 305 has the general shape of a letter H : the first edge portion 401 is formed by the vertical branches of the H at one extremity, the second edge portion 402 is formed by the vertical branches of the H at the other extremity, and the middle portion 403 is formed by the cross line of the H . Each narrow bridge in the first and second edge portions 401 and 402 may be for example 5 micrometres wide (w = 5 * 10“6m) . In such a case the thermal conductivity out of the middle portion 403 is 4 * (wd / L) K = 1 . 2 pW / K where the factor 4 comes from the fact that there are four narrow bridges .

[0055] To keep the generated heat load small , it should most advantageously be ensured that the heating required for the switching action takes place only very locally . To this purpose , in the embodiments of both figs . 3 and 4 the heater 307 has the non-superconductive heater material 308 only as a delimited strip portion in the middle . Couplings of the heater material 308 to the heater connections 311 and 312 go through respective sections of superconductive material . Most advantageously the material of such couplings is not the same as the first superconductive material comprised in the signal line 306 but another, second superconductive material that has a second critical temperature , higherthan the first critical temperature . This way the couplings may remain in their superconductive state even when the switch is open, i . e . when the first superconductive material is not superconductive any more .

[0056] This same principle can be applied also with respect to the signal line . In a switch according to the embodiment shown in fig . 5 , couplings of the signal line 306 to the signal connections 309 and 310 go through sections 501 and 502 of a superconductive material other than the first superconductive material . To keep the number of different materials in the structure at minimum, it is advantageous to select the same ( second) superconductive material both for sections 501 and 502 and for the corresponding superconductive sections 503 and 504 of the heater 307 .

[0057] The first superconductive material may be for example titanium ( Ti ) , titanium nitride ( TiN) , or aluminium (Al ) . The second superconductive material , if used, may be for example niobium nitride (NbN) which, as a disordered superconductor, has relatively low thermal conductivity but can have a high critical temperature , substantially above 10 K .

[0058] It is not necessary to use two separate superconductor materials to achieve the obj ectives described above . With compound superconductors such as NbN, TiN, or TaN it may be possible to tune both the critical temperature and the normal state resistivity through changes in the deposition parameters . Consequently, such materials can be potentially tuned to be optimal for both two purposes for which the first and second superconductive materials were described above .

[0059] In the embodiments of figs . 4 and 5 , a further difference to the embodiment of fig . 3 is the meandering form of the signal line 306 . Such a meandering form can be used to achieve at least two obj ectives . First , as the operation of the switch dictates that at least a portion of the signal line must become resistively conductive when heated, the longer such a portion is the higher its resistance can be made , implying more complete switching . The second obj ective is related to avoiding thermal noise caused by the "hot" part of the signal line ( the word hot being relative here , as the part may still be at a temperature only slightly over 4 K) . The longer is the noise-conducting path along any available route from the "hot" part to any noise-sensitive element such as a qubit , the smaller the disadvantageous effect of thermal noise can be made . The particular meandering form shown in figs 4 and 5 is only an example , as numerous other meandering forms could be presented . In addition to , or in place of , a meandering form of ( a part of ) the signal line , also the heater or any part thereof could have one or more meander-formed sections .

[0060] Fig . 6 illustrates a switch according to an embodiment . In fig . 6 , a superconductive ground layer 601 covers at least a part of the surface of the substrate . One or more parts of the superconductive ground layer 601 extend on the membrane section 305 , following the path of the signal line 306 . In fig . 6 , this refers to the strip-formed parts 602 and 603 of superconductive material that flank the signal line 306 across the membrane section 305 , separated from the signal line 306 by a respective gap .

[0061] Following the principle shown in fig . 6 , it is possible to make a switch in which the ( superconductive ) grounding effect follows closely along the signal line even through the part where the signal line forms part of a switch . It may be advantageous to make the parts of the superconductive ground layer that extend on the membrane section of a superconductive material the critical temperature of which is higher than the above- mentioned first critical temperature . This way it can be ensured that those parts of the superconductive ground layer that extend on the membrane section remain superconductive even when the signal line becomes "hot" and, consequently, the switch assumes an off-state .

[0062] Following the principle shown in fig . 6 , it is possible to make at least a section of the signal line appear as a transmission line for conveying high- frequency signals . For example , in fig . 6 the signal line 306 and the parts 602 and 603 of the superconductive ground layer 601 that extend on the membrane section 305 constitute a coplanar waveguide . By using different geometries , different kinds of transmission lines may be formed . With proper design of the geometry, it is also possible to maintain appropriate impedance matching despite the effective dielectric constant along the membrane section 305 being different than on the surface of the substrate . It is also possible to design and manufacture dedicated capacitive and / or inductive elements on the membrane section 305 and / or on the surface of the substrate in the vicinity of the switch to ensure impedance matching .

[0063] A further advantageous detail shown in fig . 6 are the superconductive connections between otherwiseisolated parts of the superconductive ground layer 601 . In fig . 6 , these consist of the superconductive bridges 604 , 605 , 606 , and 607 across the respective signal connections 309 and 310 and heater connections 311 and 312 . Other known solutions than superconductive bridges that utili ze the certain degree of three-dimensionality in designing and building microcircuits could be used in place of or in addition to superconductive bridges . Superconductive connections like these can be used to ensure uniformity of ground potential in different parts of the structure .

[0064] Figs . 7 to 16 illustrate some possible modifications and further developed embodiments that utilise the same principles as described above . For simplicity, figs . 7 to 16 illustrate switches in which no parts of a superconductive ground layer extend on the membrane section . It should be noted, however, that such a choice is made here only to maintain graphical clarity . All aspects described above in association with the embodiment of fig . 6 can be generali zed to all other embodiments .

[0065] In fi gs . 7 to 10 and 12 to 16 , each roundcornered rectangle in the background represents an area where a membrane section is separated from a mechanically supporting substrate by a void . In fig . 11 the corresponding area is round, as a reminder that there are no detailed requirements as to what form such an area should have .

[0066] In fig . 7 , a common heater 701 is used to control two separate signal lines 702 and 703 , while in fig . 8 , two heaters 801 and 802 are used to control a common signal line 803 . These serve to remind that thereare no restrictions concerning the relative numbers of heaters and signal lines in the switch . Using two or more heaters in the switch may provide redundance for more reliable operation of the switching function, and / or allow using smaller currents per heater . Using one heater to control two or more signal lines may enable minimi zing the total heat load in applications where multiple switches need to be set to their off-states simultaneously .

[0067] The embodiments shown in figs . 9 and 10 serve as a reminder that the previously described form of a letter H, or indeed any particular form in which the membrane section is patterned at edges to decrease its cross section for heat conduction, is not limiting . In figs . 9 and 10 , the membrane section is supported with patterned portions on four sides : pairs of narrow necks or bridges in fig . 9 , and single necks or bridges in fig . 10 . Earlier in the embodiment of fig . 3 , the membrane section 305 could have extended onto the substrate 301 also at its sides , not having to leave any of the cavity 302 visible on the top side of the substrate 301 . In other words , the membrane can be left completely unpatterned, spanning the whole void that separates it from the substrate . Such an unpatterned membrane section may not be preferable with respect to thermal conductivity but may be preferable with respect to the simplicity of the manufacturing process . In contrast , membrane sections that only cover a part of the void may be described as "patterned" .

[0068] The switch of fig . 9 also follows the principle described above with reference to fig . 7 , having a common heater 901 to control a plurality of (here : three )signal lines 902 , 903 , and 904 . Fig . 10 shows how, if any bent or meandering forms are used to design the heater 1001 and / or any signal line 1002 , one does not need to be limited to rectangular forms . Also , any signal line and / or any heater may have its respective signal connections or heater connections extending in any direction with respect to the middle portion of the membrane section .

[0069] Fig . 12 illustrates an embodiment in which a first membrane section 1203 is separated from a substrate 301 by a first void 1201 and a second membrane section 1204 is separated from the substrate 301 by a second void 1202 . Both the first 1203 and second 1204 membrane sections are thermally and electrically insulating in the sense described above with reference to fig . 3 . A first signal line 1205 and a first heater 1206 are supported by the first membrane section 1203 , while a second signal line 1207 and a second heater 1208 are supported by the second membrane section 1204 . The second signal line 1207 comprises the same ( first ) superconductive material as the first signal line 1205 , and the second heater 1208 comprises the same heater material as the first heater 1206 . The switch of fig . 12 comprises , mechanically supported by the substrate 301 , one or more serially connecting signal connections 1209 for coupling signals through the first 1205 and second 1207 signal lines in series . Similarly, the switch comprises , mechanically supported by the substrate 301 , one or more serially connecting heater connections 1210 for coupling electric currents through the first 1206 and second ( 1208 ) heaters in series .

[0070] In essence , a switch according to the principle shown in fig . 12 comprises two mutually similar component switches in series , controlled with a common control signal . The number two is not limiting here , as the switch could comprise three or more similar component switches in series . The principle of serially connected component switches can be used to increase the total off-state resistance of the switch, for example if the resistance integrable on a single membrane is limited by implementation aspects affected by membrane si ze , lithography limits in patterning, material parameters ( resistivity) , and / or - in case of microwave switches - limitations set by the microwave design in conductor geometry . The increased total off-state resistance is beneficial for enabling high on-off ratio . A switch based on several cascaded component switches , each on its own individual small membrane , may also be faster compared to a single switch integrated on a larger membrane as the heat capacity of the small membranes is lower leading to a smaller thermal time constant .

[0071] Fig . 13 illustrates an embodiment that resembles that of fig . 12 , the same reference designators being used for similar parts . As a difference to fig . 12 , in fig . 13 the first membrane section 1203 and the second membrane section 1204 are separated from the substrate 301 by the same void 1301 . Such an embodiment may help to save substrate area and possibly simplify some manufacturing steps , as only one void needs to be made .

[0072] Fig . 14 illustrates an embodiment otherwise similar to the embodiment of fig . 5 but with the membrane section patterned for support on all four sides . This serves as a reminder that in case there are portions ofthe membrane section extending onto the substrate surface for support , not all such portions need to simultaneously serve as supports for some signal connection or heater connection . Some such portions may be there simply for enhanced mechanical support .

[0073] All previously described embodiments have had the signal line and heater in essentially the same horizontal plane , separated from each other in the horizontal direction . This is not a limitation, as the switch may comprise , supported by a membrane section, at least two overlapping layers , one of which belongs to a signal line and the other to a heater . Fig . 15 illustrates such an embodiment , in which the signal line 1501 passes through the switch in a direction that goes hori zontally across fig . 15 and the heater 1502 passes through the switch in the up-down direction of fig . 15 . In the middle of the membrane section 1503 , the signal line 1501 and the heater 1502 briefly overlap, separated from each other by an electrically insulating but thermally conductive layer 1504 . Such an embodiment allows very effective and very locali zed heating of the signal line , with advantageous consequences in the form of minimi zing both the power needed for the heating and unwanted dissipation of heat to any surrounding components .

[0074] The possible use of overlapping layers on the membrane section is not limited to embodiments like those in fig . 15 , as similar principles can be used to implement e . g . electronic components . Fig . 16 illustrates an embodiment in which there are , supported by a membrane section 1601 , two overlapping layers comprising superconductive material . The lower layer, shown partlycovered by an intermediate dielectric ( i . e . electrically insulating) layer 1602 and the upper layer, is the one with a signal line 1603 . The upper layer forms a superconductive patch 1604 on top of the dielectric layer 1602 . The superconductive patch 1604 and the coupling line 1605 connecting it with some other circuit element on the substrate may comprise at least one same superconductive material as the signal line 1603 and / or some other superconductive material . The overlapping portions of the superconductive patch 1604 and the signal line 1603 , with the dielectric layer 1602 therebetween, form a capacitor .

[0075] A further embodiment , not shown in the drawings , could combine principles from figs . 6 and 16 . In other words , a switch according to such an embodiment could have the signal line configured as a transmission line in accordance with the principle shown in fig . 6 , and additionally have one or more built-in capacitors in accordance with the principle shown in fig . 16 . The one or more built-in capacitors could be used to ensure proper impedance characteristics of the transmission line . Additionally or alternatively, built-in capacitors and possibly some further circuit elements could be used to combine a filter to the switch .

[0076] Fig . 17 shows an example of how switches of any kind described above could be used in a quantum processing system . In the circuit arrangement of fig . 17 , a signal on an input line 1701 should be selectively allowed to propagate to any of the N devices # 1 to #N that operate under cryogenically cooled conditions . To this end, the input line 1701 is branched to N switches 1702 to 1704 , each used to control the propagation ofthe signal to a corresponding device . Control signals on control lines 1705 are used to selectively set each of the switches 1702 to 1704 to either an on-state or an off-state . I f the input line 1701 and the control lines 1705 all come from a surrounding room temperature environment to the cryogenically cooled environment , a solution like that in fig . 17 does not reduce the number of connections that need to be built between the environments compared to a solution in which the signals from the room temperature environments to the devices # 1 to #N would each come along its own input line . However, as the control lines 1705 can be DC lines , which are much cheaper than RF lines in cryogenically cooled environments , significant advantages may be gained even in such a case .

[0077] Fig . 18 shows a further development , in which a control line 1801 is used for feeding, from a surrounding room temperature environment to the cryogenically cooled environment , a control signal that should further be used to control one or more of the switches 1702 to 1704 . Within the cryogenically cooled environment , there is a control multiplexer 1802 that selecta- bly directs such a control signal to one or more of the switches 1702 to 1704 . The operation of the control multiplexer 1802 is in turn controlled by a multiplexer control signal on a multiplexer control line 1803 . This way the number of connections that need to be built between the room temperature and cryogenically cooled environments can be significantly reduced . Assuming an N-bit multiplexer ( i . e . a multiplexer controllable with an N-bit binary signal brought in through themultiplexer control line 1803 ) , one may control the operation of 2AN switches .

[0078] In a quantum processing system, it may be possible to place only the most temperature-sensitive parts to the coldest part of the cryostat and use the other cooled levels of the cryostats for other components . In figs . 17 and 18 this distinction is schematically shown so that only the devices # 1 to #N are in the part cooled to ( an exemplary coldest temperature of ) 10 mK, while the switches 1702 to 1704 as well as the multiplexer 1802 of fig . 18 are , for example , on a 100 mK level , a 1 K level , or a 4 K level . The cooling system of a cryostat can typically offer much higher cooling powers on other than the very coldest level , advocating the use of such other levels to accommodate heat-dissipating components that do not need the very coldest temperatures for their appropriate operation . As further example , in an embodiment otherwise like that of fig . 18 , at least the multiplexer 1802 may be located on other than the coldest level while the switches 1702 to 1704 , or at least some of them, may be on the coldest level .

[0079] Fig . 19 illustrates the principle of possibly cascading individual switches to obtain an improved on- off ratio in resistance . Structural solutions in accordance with for example those explained earlier with reference to figs . 12 or 13 could be used .

[0080] Fig . 20 illustrates the possibility of including a shunting resistor 2001 coupled between the output of the switch 2002 and the ground potential . The resistance of the shunting resistor 2001 is designated as R1 . The impedance of whatever further component , to which the propagation of signals is to be controlled,is called a load impedance Zload . The off-state resistance of the signal line in the switch 2002 is RN . The role of the shunting resistor 2001 is to ensure a high switching ratio when the load impedance Zload is very high . By imposing a condition R1 << Zload, RN the off-state signal suppression is significantly suppressed by the voltage division . The resistance of the shunt resistor 2001 may also be chosen such that R1 > Z 0 , where Z 0 is the impedance of the signal-feeding transmission line (e . g . Z 0 = 50 Ohm) such that it does not significantly affect the signal passed through the switch in the on-state .

[0081] A further benefit is that the presence of a shunting resistor 2001 suppresses thermal noise that could otherwise be coupled from RN to the large-impedance load when the switch is in its off-state . This may be important in some embodiments such as driving a multi-qubit system where noise even at an idling qubit may have harmful effects for the measurement of other (coupled) qubits .

[0082] The large-impedance load assumption is typically valid in qubit driving configurations where Zload typically represents a large capacitive reactance . For other functionalities such as magnetic flux driving and readout channels Zload can be very small whence R1 is not needed or can even be harmful creating excess dissipation which can have effects in performance .

[0083] Fig . 21 illustrates a variant of the principle shown in fig . 20 . In the embodiment of fig . 21 , an attenuator is formed by the resistors 2001 , 2101 , and 2102 , their resistances being designated as Rl , R2 , and R3 respectively . When the switch 2002 is in its on-state , the resistors form a standard attenuator having the Pi configuration . When the switch 2002 is in its off-state , the attenuation is increased roughly by factor ( (RN + R2 ) / R2 ) . Using the examples in figs . 20 and 21 as an instruction, the skilled person is capable of presenting also other types of configurations of resistors , capacitors , and / or inductors in the circuit arrangement to optimi ze for different loading configurations .

[0084] Figs . 22 to 24 illustrate embodiments of a method for manufacturing a switch for use to control propagation of electric signals under cryogenically cooled conditions . A substrate 2201 , the main purpose of which is to provide mechanical support , is shown simplified as a uniform layer of material . In fig . 22 , a membrane section 2202 is produced on a surface of the substrate 2201 , for example by first producing a larger, uniform membrane and thereafter patterning it to get the desired si ze and form of the membrane section . As already explained above , the membrane section 2202 should comprise material that is thermally and electrically insulating under the intended cryogenically cooled conditions .

[0085] The steps of producing a signal line and a heater on the membrane section appear in the cross- sectional view selected for figs . 22 to 24 as adding a further layer 2203 on top of the membrane section 2202 . As already explained above , the signal line should comprise ( first ) superconductive material that has a ( first ) critical temperature and the heater should comprise heater material that is resistively conductive under the intended cryogenically cooled conditions . Alsoappearing as j ust adding the further layer 2203 is the step of producing signal connections for coupling a signal to the signal line and heater connections for coupling an electric current to the heater . As with the membrane section, thin-film-depositing and patterning techniques generally known from the production of microcircuits can be used to produce the signal line , the heater, the signal connections , and the heater connections .

[0086] To provide the intended thermal insulation from the substrate , the manufacturing method comprises a step of arranging for a void to separate the membrane section 2202 from the substrate 2201 . In fig . 22 this is done by producing a cavity 2204 through the substrate 2201 , which cavity 2204 opens onto an opposite side of the substrate 2201 than the side on which the membrane section 2202 is . This may be accomplished by using, for example , deep anisotropic reactive ion etching from the back side of the substrate wafer, also called the "Bosch process" . In the method of fig . 23 , the arranging for a void comprises forming a hollow 2301 in the substrate 2201 below the membrane section 2202 . This may be accomplished by using, for example , isotropic etching from the front side .

[0087] The method of fig . 24 differs from those of figs . 22 and 23 in that a sacrificial layer 2401 is produced on the surface of the substrate 2201 before producing the membrane section 2202 . At a later stage of the method shown in fig . 24 , the step of arranging for a void to separate the membrane section 2202 from the substrate 2201 comprises selectively removing a part of the sacrificial layer 2401 so that a void 2402 isformed between the membrane section 2202 and the original surface of the substrate 2201 .

Claims

CLAIMS1. A switch for use to control propagation of electric signals under cryogenically cooled conditions, the switch comprising:- a substrate (301) forming mechanical support,- a thermally and electrically insulating membrane section (305) separated from the substrate (301) by a void,- supported by the membrane section (305) , a signal line (306) comprising first superconductive material,- supported by the membrane section (305) , a heater (307) comprising heater material (308) that is resistively conductive under the cryogenically cooled conditions,- mechanically supported by the substrate (301) , signal connections (309, 310) for coupling a signal to the signal line (306) and heater connections (311, 312) for coupling an electric current to the heater (307) .

2. A switch according to claim 1, wherein the membrane section (305) spans the void unpatterned.

3. A switch according to claim 1, wherein the membrane section (305) is patterned to comprise at least :- a first edge portion (401) ,- a second edge portion (402) , and- a middle portion (403) between the first and second edge portions (401, 402) ; wherein :- the cross section for heat conduction of the first edge portion (401) and the cross section for heat conduction of the second edge portion (402) are both smaller than the cross section for heat conduction of the middle portion (403) , all said cross sections being taken perpendicular to a longitudinal directionalong the membrane section (305) between the first and second edge portions,- the signal line (306) is coupled to the signal connections (309, 310) along respective parts of the first (401) and / or second (402) edge portions, and- the heater (307) is coupled to the heater connections (311, 312) along respective parts of the first (401) and / or second (402) edge portions.

4. A switch according to claim 3, wherein the membrane section has the form of a letter H, so that- the first edge portion (401) is formed by the vertical branches of the H at one extremity,- the second edge portion (402) is formed by the vertical branches of the H at the other extremity, and- the middle portion (403) is formed by the cross line of the H.

5. A switch according to any of the preceding claims, wherein the first superconductive material has a first critical temperature and the switch comprises at least one of the following:- couplings of the signal line (306) to the signal connections (309, 310) through sections (501, 502) of second superconductive material that has a second critical temperature, higher than the first critical temperature,- couplings of the heater material (308) to the heater connections (311, 312) through sections (503, 504) of the second superconductive material.

6. A switch according to any of the preceding claims, wherein:- a superconductive ground layer (601) covers at least a part of a surface of the substrate,- one or more parts of the superconductive ground layer extend on the membrane section (305) , following the path of the signal line (306) .

7. A switch according to claim 6, wherein at least those parts of the superconductive ground layer that extend on the membrane section comprise superconductive material the critical temperature of which is higher than the first critical temperature.

8. A switch according to any of claims 6 or7, wherein the signal line (306) and those parts of the superconductive ground layer that extend on the membrane section constitute a coplanar waveguide.

9. A switch according to any of the preceding claims, wherein the void is a cavity through the substrate, opening onto an opposite side of the substrate than the side on which the membrane section is.

10. A switch according to any of claims 1 to8, wherein the void is a hollow in the substrate below the membrane section.

11. A switch according to any of claims 1 to 8, wherein:- the switch comprises a sacrificial layer on a surface of the substrate, and- the void is a hollow formed in the sacrificial layer .

12. A switch according to any of the preceding claims, wherein:- the void is a first void (1201) , the membrane section is a first membrane section (1203) separated from the substrate (301) by the first void (1201) , the signal line is a first signal line (1205) and the heater is a first heater (1206) ,- the switch comprises a thermally and electrically insulating second membrane section (1204) separated from the substrate (301) by a second void (1202) ,- the switch comprises, supported by the second membrane section (1204) , a second signal line (1207) ,comprising the same first superconductive material, and a second heater (1208) comprising the same heater material ,- the switch comprises, mechanically supported by the substrate (301) , one or more serially connecting signal connections (1209) for coupling signals through the first (1205) and second (1207) signal lines in series, and- the switch comprises, mechanically supported by the substrate (301) , one or more serially connecting heater connections (1210) for coupling electric currents through the first (1206) and second (1208) heaters in series.

13. A switch according to any of claims 1 to 12, wherein:- the signal line is a first signal line (1205) and the heater is a first heater (1206) ,- the switch comprises a thermally and electrically insulating second membrane section (1204) separated from the substrate (301) by the void (1301) ,- the switch comprises, supported by the second membrane section (1204) , a second signal line (1207) , comprising the same first superconductive material, and a second heater (1208) comprising the same heater material ,- the switch comprises one or more serially connecting signal connections (1209) for coupling signals through the first (1205) and second (1207) signal lines in series, and- the switch comprises one or more serially connecting heater connections (1210) for coupling electric currents through the first (1206) and second (1208) heaters in series.

14. A switch according to any of the preceding claims, comprising, supported by the membrane section, at least two overlapping layers (1603, 1604)comprising any of the superconductive materials electrically insulated from each other by one or more dielectric layers ( 1602 ) to form a capacitor .15 . A quantum computing system comprising :- a cryogenically cooled domain ( 201 ) ,- a room-temperature domain ( 202 ) , and- at least one switch according to any of claims 1 to 14 in the cryogenically cooled domain ( 201 ) .16 . A method for manufacturing a switch for use to control propagation of electric signals under cryogenically cooled conditions , the method comprising :- producing a membrane section on a surface of a substrate , the membrane section comprising material that is thermally and electrically insulating under the cryogenically cooled conditions ,- producing a signal line and a heater on the membrane section, of which the signal line comprises first superconductive material that has a first critical temperature and the heater comprises heater material that is resistively conductive under the cryogenically cooled conditions ,- producing signal connections for coupling a signal to the signal line and heater connections for coupling an electric current to the heater, and- arranging for a void to separate the membrane section from the substrate .

Images