Quantum Device Physics
Welcome to the Quantum Device Physics page
The University of Birmingham Quantum Device Group
This area of condensed matter physics deals with exploring the fundamental physics of
structures with potential for applications. The quantum device physics programme is a collaboration
between Dr C M Muirhead,
Dr Mark Colclough
, who helped design the World's first commercial HTS SQUID for Conductus in the USA, and
Dr Ed Tarte.
The research students working on device design, fabrication and testing are Georgina Klemencic and Bindu Malini Gunupudi.
In 1987 following the first demonstration of flux quantisation in HTS, Birmingham were the first to realise a rudimentary HTS SQUID, making use of the naturally occurring weak-links between grains in a ceramic sample. Subsequently the group has worked with a number of groups world-wide investigating the latest state-of-art structures with a view to understanding the physics underpinning device performance. In 1996 a new clean room facility was established, enabling the group to make its own device structures. In the light of new international developments in superconductivity and the interests of new staff, this work has been broadened to include studies of the microscopic processes which limit the performance of quantum devices and, more recently the interaction of microwave resonators with nanobar mechanical resonators.
The last few years have seen rapid international development in a range of sophisticated experimental
and theoretical techniques for studies of superconducting devices at the quantum level, in particular quantum
bits (qubits) for future quantum computers. Here at Birmingham we have instrumented one of our dilution
refrigerators for the ultralow noise measurements of devices called phase-qubits (see below), and performed a
range of studies on these devices. All such devices are limited by a process called 'quantum decoherence'.
A particular interest here is the role played in quantum decoherence by atomic scale defects in the structure,
called two-level systems.
Another exciting areas we are working in is the coupling of low-loss microwave resonators to nanometer size mechanical resonators, in collaboration with the University of Nottingham. Reduction of the mechanical vibration (cooling) can be obtained by careful tuning of the microwave resonant frequency. Again, these experiments require, not only very low temperatures, but also ultralow noise electronic measurement. A second dilution refrigerator has been instrumented for these experiments. In the early stages, we will be studying the factors which affect loss in these devices and, in later stages, will be incorporating mechanical resonators into qubit devices.
The topics listed below give examples of projects currently under investigation. As a group we welcome people who are interested in the field and would like to suggest similar related projects.If you are interested in the work that we do and are considering applying for a PhD or PostDoctoral position in the group, please visit the Opportunities page.
Our main areas of interest are as follows:
These topics are discussed in more detail below:
Intrinsic Josephson Junctions
Initially it was believed that High Temperature Superconductors had no possible applications in quantum electronics. This belief was mostly due to the prescence of 'quasi-particle' (Electron-like and hole-like excitations) down to very low temperatures, characteristic of the d-wave nature of the materials. These excitations would normally destroy the delicate quantum coherence. Recently however, there has been a renewed interest following several experiments apparently demonstrating quantum effects at relatively high temperatures.
BSCCO is an important HTS material as its properties are highly anisotropic. This means that the conductivity in the a-b planes is much greater than along the c-axis. The layered nature of BSCCO gives rise to Josephson Junctions between the layers, these are known as Intrisic Josephson Junctions (IJJ) and are of very high quality in single crystal superconductors. Some of our research focusses on the Josephson junctions in such environments.
Left: A whisker sample with 4-terminal contacts using FIB Platinum deposition
A 3D structure etched into a BSCCO whisker
We are currently developing a measurement system to work in conjunction with our dilution refrigerator to measure the properties of such Josephson Junctions at temperatures below 100mK. The fridge includes custom made filtering modules to prevent electronic interference masking the signals from the sample. This is tremendously important as the signals from the Junctions can look similar to noise at low temperatures. If these measurements are sucessful we will use the same apparatus to make measurements on our qubit structures which are also under development (See Superconducting Quantum Devices below)
Left: Photo of the lower (<4K) stage of the dilution frige
A JJ sample ready for measurements in the RF sheilded box
Superconducting samples of High Tc YBCO films have successfully been patterned into tracks as thin as 50nm. We are currently investigating various properties of these sub-micron tracks, including the effect of Gallium poisoning due to FIB milling procedures, which may provide a mechanism to reduce the tracks' effective dimensions to below the ultimate patterning linewidth Interesting effects have already been observed in such tracks, including evidence for phase-slip centres.
Josephson junction and Qubit devices
We are currently performing feasibility studies into the design and application of qubit (Quantum Bit) structures, and the potential for combinination with other quantum devices. The field of research into qubits and quantum computing is still in the early stages, and vital steps have been made towards the understanding of such phenomena, however there have of yet been no real demonstrations of fully functional solid-state systems. It is important to understand the physics behind the operation of devices in the quantum regime; superconductors allow us to probe this regime with relative ease due to the macroscopic nature of the superconducting wavefunction. The structures that are currently under investigation mainly comprise of 'pi-junction' flux qubits, although we are also considering other designs. Our design of Pi-qubits utilise both High Temperature and Low Temperature Superconductors in their operation and have many advantages over qubit designs based on a single material type.
The dilution fridge measurement system allows temperatures down to 50mK to be explored
For a short video introduction to quantum computing, please click here.
A superconducting qubit in a microwave resonant cavity provides the opportunity to construct a solid state analogue of an atom in an optical trap, but because the qubit is a circuit, we now have a quantum system whose properties can be dramatically altered using control lines, weak magnetic fields, etc. This not only gives rise to new physical phenomena, but also opens the way to a range of novel microwave quantum devices, such as sources of single photons on demand, a new class of microwave laser and quantum encryption. The development of such quantum microwave systems presents a considerable challenge, because the quantum states are fragile. We are currently investigating ways of coupling persistent current qubits with microwave cavities in collaboration with colleagues at the National Physical Laboratory. The project combines the very low temperature and low noise device expertise in the department of Physics with the microwave design expertise at Birmingham and Cardiff. This will put us into a very strong position internationally to construct a range of sophisticated microwave circuits to drive and manipulate qubits and microwave photons, and to investigate the exciting physics of such systems. Support from our colleagues at NPL will continue.
The cavity and resonator for the QED experiments
The resonators are made from a submircon film of the superconductor Niobium (TC = 9.2K) on sapphire or silicon substrates. These are patterned into resonant structures in our class 100 cleanroom, using very similar methods to those used for making integrated circuits. The use of superconducting niobium means that resistive losses are very low (but not zero) and this leads to a very sharply defined resonance (Q=resonant frequency/width of resonance). We can obtained Q's up to 106 at the lowest temperatures.
These very high Qs are limited at the lowest temperatures by the very same two-level systems (TLS) that cause decoherence in qubits, so studies of microwave resonators alone provide a powerful tool for studies of TLS in substrates and interfaces with superconductors.
Left: Resonator response at low temperatures. The Q of the resonaor (blue data) increases as the temperature
is reduced and the number density of Copper pairs approaches a maximum. The maximum in the resonant frequency (red
data) is not predicted by superconductivity theory and is due to TLS effects.
Right: Resonant response versus applied
magnetic field at a series of temperatures. The response is accurately dependent on the square of the magnetic field.
The small offset on the horizontal axis is due to trapped magnetic flux. There is no perceptible decrease in the Q of the resonance.
An interesting spin-off of this work has been the observation that the resonant frequency of a resonator can be changed by up to 200 linewidths by the application of magnetic fields only a few times larger than that from the earth. We are pursuing this phenomenon because of its potential application to the generation of single photons 'on-demand'.
Coupling of mechanical nanobars to microwave resonators
It has been known for a few years that the vibrational modes of a mechnical resonator can be coupled to the oscillation of the electromagnetic field in a microwave resonator. This coupling relies on the force which the electric field component of the EM wave in the resonator applies to the electrically conducting mechanical resonance. This is very closely analogous to a phenomena well known to the optics community, where there is an interaction between the photon field in an optical cavity and the vibrational modes of the end mirrors. This interaction places a serious limitation on the use of such cavities as gravitational wave detectors. The sort of experiments that we are engaged in are, in some ways, much easier than the optical counterpart, particularly because we can cool the thermal vibrations of the nanobar down to mK temperatures.
The temperature of such bars can be further cooled by driving the system at a microwave frequency slightly lower than the electromagnetic resonance frequency. If the bars can be cooled to the quantum mechanical ground state of the mechanical resonance (E = 0.5hf where h is Planck's constant), then they may be incorporated into qubit structures.
Niobium nanobar. The nanobar is the long thing strip near the top of the photo. It is 10um long and about 100nm in cross section. It is constrained at its ends, but is free of the substrate underneath so it can vibrate like a stretched string.
These studies are being made in collaboration with the University of Nottingham who made the structure shown above using advanced e-beam lithography techniques. Measurements are being made in our second dilution refrigerator at temperatures down to 15mK. Our first experiments are aimed at determining the role TLS in limiting the Q of the mechanical response.
We are currently collaborating with the School of Electronic, Electrical and Computer Engineering at the University of Birmingham to fabricate YBCO-Au-Nb ramp-type Josephson junctions in order to investigate their Current-Phase Relation (CPR), and to further the potential of microwave applications stemming from the cavity QED work. We also have strong links with other departments within the University, such as the Department of Materials Science and the Micro Engineering group (located in the School of Mechanical Engineering). We are also working in collaboration with the Device Materials Group in Cambridge, in order to assess new types of Josephson junctions as potential qubit candidates. We are also working with the National Physical Laborotory to characterise qubits in cavities
Class 1000 clean Room with extensive Lithography facilities
It is essential for micron and sub micron work that the patterning of samples is carried out in a clean environment. Even a particle of less than a micron in diameter can be detrimental to sample production. A clean room minimizes the number of dust particles in the atmosphere by an air conditioning system and laminar flow hoods under which the work is carried out. The class of a clean room is determined by the remaining number of dust particles after filtration. The device physics custom built clean room contains equipment used for photolithography and ion beam milling. The photolithography tools used include a programmable variable speed resist spinner, a Karl Suss mask aligner and a developing and wet-etch area.
The device physics cleanroom lithography facilities
Deposition and milling systems
Wet etching can often be damaging to samples, and for some materials there are no appropriate chemicals. In these cases Ion Beam milling can be used, which produces a cleaner etch profile and is more controllable and less damaging to delicate samples. The group possess a Millatron IV Argon Ion beam milling machine, a picture of which is shown below:
The Ion Beam Milling system
The custom built sputter deposition system
Reactive Ion Etching (RIE) uses oxygen as the active species to clean and etch samples. We own an automated PLASMALAB RIE Plasma etching system.
Our magnetron sputtering system can deposit 2 different materials in-situ. The system is currently set up to deposit thin films of Gold and Niobium, two materials in common use in our devices. The sputtering system has the advantage of being quick and easy to operate, with depositions possible within a couple of hours.
Ultra Low Temperature experimental apparatus
A large amount of experimental apparatus is available to measure fabricated devices. The apparatus is designed to accomodate as many variables as possible so that the devices can be investigated in a number of different ways. For High Tc superconductors, masurements are generally made above 4.2K and therefore Liquid Helium cryostat apparatus is used. For lower Tc superconductors, dilution fridges are available, which can lower the temperature down to 10's of mK. Apparatus contains electrical feedthroughs for the determination of IV characteristics, and multi-axis magnet set-up for investigation into field dependence of samples.
A typical device measurement cryostat setup
The SQUID magnetometer is used to make accurate measurements of magnetic field, and can be used to 'image' a magnetic sample. This technique is extremely useful when combining superconducitng materials with ferromagnetic or magneto-resistive materials, as the physics of the interaction between the materials can be established.
Thin Film Laser Ablation facility
Laser ablation is one of the primary methods of thin film deposition used for High Temperature superconducting materials. The group has access to an excimer laser, allowing us to deposit thin films of many materials, including YBCO, STO, LCMO, LSMO, PBCCO. The film growth methods and quality are intensively studied by members of the condensed matter group, resulting in high quality films for use in our device production. To learn more about the laser ablation facility visit the Pulsed Laser deposition website.
Focused Ion Beam Milling apparatus with integrated SEM
The FIB has been one of the most useful tools in recent research. Under vacuum, an energetic beam of Gallium ions is focussed upon the sample, removing material in its path. Very small, precisely milled structures can be acheived down to dimensions of below 100nm. The dual beam system also contains an SEM for high quality imaging of the structures during fabrication. An additional gas assisted deposition system present on the chamber allows structures to be built upon the surface in addition to the milling away of material.
The Dual FIB SEM system
Veeco AFM/STM imaging/lithography system
The AFM is principally used for imaging the surfaces of films grown in the laser deposition system, although it is also used to image samples made by the device physcis group. The AFM is of use when accurate measurements of device geometries are required, especially in the z (depth) direction. It is difficult to judge the depth of a milled structure by SEM observation, but the stylus operating principle of the AFM accurately traces out the contours of the sample in 3D. The AFM also has Scanning Tunneling microscopy (STM), Magnetic Force microscopy (MFM) and hi-voltage lithography modes.