The Four Corner Pillars of Quantum Technology

Plus the fifth one, supporting them all. The entire world is studying quantum technology. All the main universities and many billion-dollar companies are trying to develop something in the quantum realm. Swedish universities are at the forefront, thanks to formidable support from the fifth corner pillar Myfab, allowing them to manufacture their building blocks for quantum devices in a set of high-quality, high-reliability machines. Economical resources are also required, and fortunately Swedish universities have forward-looking sponsors.

Image: Myfab

The Wallenberg Center for Quantum Technology (WACQT) is a 12-year project with a budget of SEK 1 billion. The project began in 2018 and intends to bring Swedish universities and industry to the forefront of quantum technology. Quantum technology has four cornerstones: quantum computing, quantum simulation, quantum communication and quantum sensing.

The work is supported by the Knut and Alice Wallenberg Foundation (KAW) and is managed from Chalmers, and other Swedish universities are also involved. For example, KTH is involved in the work with quantum communication, while Lund University deals with quantum sensors. In addition, researchers from Stockholm, Linköping and Gothenburg universities are involved in the project, which is led by nine researchers, five of whom are based at Chalmers in Gothenburg.

The WACQT project is extensive, spreading out into many subprojects. However, the work is so extensive it is only possible to present a few examples from each corner pillar in this short article. Quantum research has four corner pillars and for the sake of readability, only a few sub-projects have been selected from each.

Quantum Computers

Sweden's foremost quantum computer developers are housed at Chalmers University of Technology in Gothenburg, manufacturing the components of the computer, the quantum bits.

Quantum computers have no processors, but are built up of quantum bits that act as both calculating elements and memory. The quantum bit or qubit is a microcircuit with quantum mechanical behaviour. The qubits are made of thinly vaporized aluminium on a silicon substrate. Everything is about 5x5 millimetres square.

The Chalmers researchers use microwave resonators for qubits, resonant circuits at about 6 gigahertz, which have had the inductance replaced by a superconducting Josephson junction that basically acts as a non-linear, controllable inductance, whose resonant frequency can be adjusted.

The resonant circuit has the special property that it can only contain a single microwave photon. A single photon is a single particle. Chalmers calls it an “artificial atom”. Microwaves, light and other electromagnetic radiation consist of photons, which are both waves and particles, and here the researchers have managed to capture and retain a single microwave photon.

During the first three project years, quantum computer researchers within WACQT have focused on making quantum bits work really well, on a small scale. A milestone was when in 2020 they managed to solve a small part of a real optimization problem with their very well-functioning two-qubit quantum computer.

The goal is to create a quantum computer with a hundred qubits, but right now they are involved in producing qubits that work for an extended period of time. Calculations have shown that it pays more for the quantum computer’s performance to increase the quality of the quantum bits than to add additional quantum bits. The better the quality, the more useful tasks the finished quantum computer will be able to perform. The qubits manufactured at Chalmers are already among the best in the world, so it is a matter of pushing the entire research front forward and eliminating the extremely small disturbances, but just understanding what the disturbances are and which ones dominate, requires more research.

The goal right now is not to create a large quantum computer, but to create a smaller system that works without too many errors. It is no use trying to build large systems if you do not have high quality qubits, which have a lifespan longer than the time it takes to actually solve the problem.

Let's take a closer look at how a quantum computer is built.

The picture shows a Josephson junction (the “T”) whose 100 nm wide electrodes connect ground (top) with the qubit electrode (bottom). The component (light gray shades) is made of aluminum with a thin layer of insulating aluminium oxide that the superconducting electron pairs can tunnel through. The substrate (dark gray) is silicon. Image: Chalmers University of Technology.

A superconducting transmon-qubit (the plus sign). The Josephson junction is located at the upper end (zoomed in in the previous image). At the lower end, the qubit is connected to the end of a resonator, for reading out the quantum state. The ground plane is perforated with holes to capture magnetic flux that can otherwise jump around, as a superconducting plane would not otherwise let magnetic fields through due to the Meissner effect. The arms of the “plus” are about 10 µm wide. Image: Chalmers University of Technology.

A test chip with eight microwave resonators around 4-8 GHz, made out of aluminum on silicon. All resonators are connected to the same supply line. The contacts are in the southwest and northeast in the picture. The chip is bonded with aluminum wires and you can see all the wires connecting the ground plane to the holder. Image: Chalmers University of Technology.

Quantum computer prototype with two qubits (the grey chip). The box (microwave package) is made of copper and there are four microwave contacts for input/output. Image: Johan Bodell, Chalmers University of Technology.

A dilution cryostat. The quantum processor is mounted on the lowest flange, which is coldest (temperature 10 mK). All cables, filters, attenuators and amplifiers on the way down are needed for input/output signals. Several layers of cryo vessels are used, including radiation shields for EM waves (heat radiation, light) and magnetic fields, and vacuum is pumped between them. Image: Johan Bodell, Chalmers University of Technology.

Until now, it has not been possible to solve any real world problems with quantum computers, only those that are designed to be easy to solve with a quantum computer, but difficult for ordinary, physical computers. The next milestone in quantum computing is finding a real-world problem that is beyond the reach of ordinary computers, but may be solved with a quantum computer with fifty to one hundred quantum bits. The group works intensively on this. The problem will probably be in logistics or simulation of small molecules.

Quantum Simulation

When developing new materials or drugs, it can be a good idea to simulate their properties first. These materials constitute extensive quantum mechanical systems that are difficult to simulate even on today’s supercomputers. Therefore, it would be better for the simulation if one could use a quantum computer, which might be easier to control. This is what quantum simulation is all about. The strength of the quantum simulator is being able to imitate a natural quantum system with a quantum system that can be controlled in the lab, where one has coded e.g. the properties of the molecule into the quantum hardware.

Quantum simulation is not about building hardware, but more about thinking in quantum terms when simulating molecules or atoms and their behaviour. One could either build a quantum computer that mimics the molecule, or one could build a programmable quantum computer and run an algorithm that calculates something, such as the basic energy of the molecule, or the like. This is done by encoding the properties of the system in a quantum mechanical, mathematical expression (a Hamiltonian) rather than having the circuit itself mimicking the properties of the quantum system.

In addition to help identifying new materials and drug substances, quantum simulation promises for example to solve routing and scheduling problems and to be very useful in advancing research in many fields of physics, quantum chemistry, and cosmology.

Quantum Communication

Today’s digital society depends on secure information transfer, but the development of quantum computers that could potentially break today’s encryption increases the risks rapidly. Quantum communication, or quantum encryption, is secure against eavesdropping. Encryption is dependent on cryptographic keys – usually strings of ones and zeros – that are used to encrypt and decrypt information. If the recipient of an encrypted message has the key, he can decrypt and read the information. The problem is generally to transfer the key without an opponent getting hold of it.

In quantum communication, the encryption key is transmitted with quantum particles, so-called Quantum Key Distribution (QKD). According to the laws of quantum physics, it is impossible to measure or copy the unknown state of a quantum particle without changing it. Therefore, one can always be sure to detect eavesdropping.

The quantum particles commonly used in quantum communication are photons sent over optical fibre. Quantum-secure information is encoded into the properties of single or entangled photons, providing “flying qubits” travelling across existing optical fibre networks. Such systems are already on the market, but the problem is that they are not able to transmit data over distances longer than 200-300 kilometres, due to the properties of optical fibres.

Photonic nanotechnology is at the core of the developments which might overcome such challenges and increase the capacity and reach of QKD transmissions in large-scale optical fibre networks. Nanophotonic devices are also expected to provide key components enabling the ultimate vision of a global quantum Internet, connecting multiple quantum processors using optical transmission channels.

Transmission capacity and reach of quantum information over existing optical fibre networks can be boosted by optical transceivers implemented in integrated nanophotonic chips, which enable efficient and ultra fast quantum state generation, manipulation and analysis and use miniaturized, stable and scalable optical circuitry in emerging nanophotonic platforms such as LiNbO3 on insulator (LNOI).

To reach even further in optical fibre links, the message must be decrypted and re-encrypted in one or more intermediate nodes, carrying the risk of intrusion. Alternatively, one could use satellites as intermediate nodes, but this is an expensive and difficult solution. If one wants to create a secure large-scale quantum Internet, one would need simple and cheap quantum repeaters. The latter requires optical memories and units capable of entangling condensed-matter qubits with flying optical qubits (photons), which is again most efficiently done in nanoscale devices.

Nanotechnology will also play a crucial role in enabling quantum processor interconnectivity via optical transmission channels, a functionality which relies on the development of a key component capable of seamlessly converting qubits from the microwave to the optical domain without degradation of their quantum features, a so-called quantum transducer.

The state-of-the-art nanofabrication and imaging facilities of the Myfab node located at Albanova, i.e. the Albanova Nanofab Lab (ANL), are supporting WACQT researchers working in Stockholm to develop the next-generation of nanophotonic devices for quantum communications.

Examples include the development of photon sources in periodically poled materials and nanophotonic platforms such as LNOI, for multi-photon entanglement and high-brightness devices, as well as heterogeneous integration of different nanophotonic materials providing multiple functionalities (photon generation, switching and detection) on chip.

Here is a closer look at on of the photonic chip technologies on the lithium niobate platform made by the Nonlinear and Quantum Photonics team at Albanova NanoLab.

LNOI (Lithium Niobate on Insulator) nanophotonic chips

This is an example of an electrically tunable optical cavity in sidewall modulated LN photonic nanowires.

The cavity traps photons at a given wavelength (colour) which can be changed by the application of a voltage of a few volts across the nanowire. This allows for tunable and ultra-narrowband (10 pm) filtering of the photons generated e.g. by spontaneous parametric down-conversion. It also enables to trap, release or bounce-off photons from the nanocavity.

Image: Katia Gallo, NQP team, KTH.

Sketch of an electrically tunable photonic nanocavity implemented in a 1D photonic wire in Lithium Niobate (LN). The cavity trapping the photon is created by inserting a defect in an otherwise perfectly period grating, made by modulating the width of the nanowire in which the photon propagates. The feature size of the sidewall ”teeth” is ~250 nm. The photon propagates in the nanowire and is trapped at the defect. An electric field applied by the side electrodes (in orange) can change the colour of the photon trapped at the defect.

Image: Katia Gallo, NQP team, KTH.

Atomic force microscopy (AFM) image of the sidewall modulated nanowire and its central defect, creating the cavity trapping the photon. The structure has been fabricated with the electron beam lithography (EBL-Raith Voyager) and ion milling facilities at the ANL.

Image: Katia Gallo, NQP team, KTH.

Scanning electron microscope (SEM) image of the gold electrodes used to tune a set of 1D photonic cavities implemented with sidewall modulated photonic wires of the kind shown in the previous image. The scale of the image does not allow resolving the sub-µm nanowires, located between the gold electrodes.

Image: Katia Gallo, NQP team, KTH.

Picture of the final device tested in the optical lab after fabrication in the ANL. Photons are coupled in and out of the photonic chip via optical fibres, almost vertically (which allows packing many components on a single chip). The electrical pads of the chip are bonded to the external, macroscopic controlling circuit board.

Quantum Sensors

In fact, sensors that exploit the quantum state of particles have been around for a long time. The magnetic resonance imager (MRI) senses the spin direction of protons, a quantum physics property. It was awarded the Nobel Prize in 2003. In rubidium-type atomic clocks, the outer electron of the rubidium atom is excited to a higher energy state, but it soon drops back to a hyperfine energy level. The hyperfine energy level is part of the interaction between the quantum state of the electron and the atomic nucleus. This can be detected by examining how the rubidium atoms absorb light.

But research continues, aiming to use individual atoms and photons in various sensors. If researchers succeed, one could for example make gravity sensors with a sensitivity one could previously only dream about, for use in mining exploration, for instance.

There are other great potential opportunities with quantum sensors. New positioning systems, clocks, gravitational, electric and magnetic field sensors. Also, optical microscopy with a resolution beyond the wavelength limit could be achieved.

Take for example the LIGO gravity telescope, which will use this principle in future detectors. The quantum improvement will enable the observation of gravitational waves from sources ten times farther than presently achievable.

Let’s just look at two projects.

Gravity and Magnetic Field Sensors

A novel type of sensor that can sense gravity, acceleration or external magnetic fields is made up of a superconducting lead micro particle that is pushed between opposing magnetic fields from two electromagnetic superconducting niobium coils. Thanks to the Meissner effect, the particle will position itself at a local minimum between the magnetic fields and remain there.

The particle’s position will be disturbed if gravity changes, for example if the sensor is passed over a mineral deposit, if the speed of the sensor changes or if the external magnetic field changes. As the particle moves, it generates a change in the magnetic flux, which can be detected by a SQUID (superconducting quantum interference device). The output signal from the SQUID is purely electric and can be measured with standard electronics.

Image: Martí Gutierrez Latorre, Chalmers

The niobium windings of the chip-based coil are created by etching. A cross section is shown here. The niobium layer resides on a silicon substrate and is about 1 µm thick, and the etched superconductor about 500 nm wide.

Image: Martí Gutierrez Latorre, Chalmers

Two coils are needed, stacked on top of each other. This is done by making two similar coils and sawing the substrate into 7x7 mm pieces, etching a hole (through-silicon via, TSV) through the upper piece with Bosch process etching and then stacking the etched piece on top of the other (a two-chip stack).

Finally, a small lead particle is dropped into the hole (at the arrow).

Image: Matthias Rudolph, Chalmers

The set-up is placed in a cryostat and cooled to 4K so that everything becomes superconducting. 0.9 amps is passed through the coils, the lead particle levitates in the magnetic field and one can start measuring the SQUID output voltage.

Sensor for Microwave Photons

Another group is studying microwave photodetectors. The purpose of such sensors is to be able to detect individual photons, as well as constituting a readout method for other devices in the WACQT project. The sensitive microwave photodetector may additionally revolutionize the field of radio astronomy, detecting very weak radio sources in the universe.

Microwave photodetector based on semiconductor heterostructure integrated into a superconducting resonator. Image: Ville Maisi, LTH

The image shows a typical sensor that can sense microwave photons at 6 gigahertz and convert them into an electric current measured with standard electronics.

The schematic shown in Figure a) consists of a superconducting resonator circuit that captures an individual photon. Photons give rise to electrons that tunnel (photon-assisted tunnelling) through the two quantum dots on the nanowire between terminals S and D, generating an electrical current. A single microwave photon with a frequency of for example 6 gigahertz, has an energy of about 4 yoctojoule. A stream of photons with such a miniscule energy gives rise to a measurable electrical current in the photodetector.

The physical circuit in Figure b) is based on two quantum dots located on the nanowire between the S and D terminals visible in the lower electron microscope image. The microwave photons enter the detector from the port on the left hand side of the main image next to the long snaking resonant circuit. The quantum dots have a size of about 100 nm.

The device is fabricated at Myfab, combining several areas of expertise:

  • Group III-V semiconductor (such as InP, InAs, GaAs, GaN, and InSb) heterostructure growth.
  • Electron beam lithography and device processing for semiconductor nanostructures.
  • Superconducting resonator fabrication including optical lithograpy and high purity aluminum deposition.

The Fifth Corner Pillar

None of the other four corner pillars could have been realised without the fifth pillar: Myfab.

Myfab is absolutely indispensable for WACQT quantum research. Without Myfab, there would be no WACQT. Myfab provides access to almost everything the project needs to develop materials and components for quantum computers and quantum technology in general. In fact, the Wallenbergs play a very important role in the research and have supplied many of the machines from Myfab.

Here is a brief overview of the various machines used in the production of quantum circuits at Myfab’s various departments in the country, such as Chalmers’ clean rooms, KTH’s Nanolab and Lund Nanolab (LNL).

It should be made clear that superconducting quantum bits do not consist of semiconductors, but of artfully designed conductors in aluminium, niobium, etc., which in the case of Chalmers’ quantum computer has been designed as resonant circuits for the microwaves around 6 gigahertz that make up the qubit itself. But semiconductor materials also provide important input for quantum technology. Photo detectors are a good example of this, combining superconducting circuits with semiconductor nanostructures.

Methods such as electron beam lithography, photolithography, metal deposition (aluminium, titanium, niobium) and plasma etching on undoped silicon are used. The Myfab organisation maintains a very high quality of machines and processes, which means that the quantum bits are also of the highest possible quality.

Image: Nanofabrication Laboratory, Chalmers

The masks for the metal patterns on the silicon chips are created using both optical and electron beam lithography in machines like this, a Raith EBPG 5200, capable of a resolution below 5 nm. It also features large vertical stage travel (Z-Lift) for 3D applications.

Image: Nanofabrication Laboratory, Chalmers

The unwanted metal can be etched away with an Oxford Plasmalab 100 reactive ion etching machine. It can etch into polymer, gallium arsenide, gallium nitride and aluminum, leaving only the conductor pattern.

Image: Anders Liljeborg, Vladislav Korenivski, KTH

Two machines at Albanova Nanolab, the RAITH Voyager electron beam lithography unit capable of a line width of less than 8 nm and a writing speed of more than 1 cm2/h, and the FEI Quanta FIB-SEM high-resolution dual beam SEM for 2D and 3D characterization and analysis, imaging, deposition and ablation of materials. The FIB acts as a ‘nano scalpel’ enabling high precision cutting and slicing (up to 10 nm/slice) into samples to reveal their 3D internal structure. These are essential facilities for all the nanophotonic devices made at Albanova.

Image: Anders Liljeborg, Vladislav Korenivski, KTH

The AJA2 sputter system by AJA International Inc. utilizes ionized gas (Ar, O2, N2) to sputter Nb and Ti from source targets onto substrates, depositing a thin film. It is used for depositing films for e.g. superconducting single photon detectors.

Image: Sandoko Kosen, Martí Gutierrez Latorre, Chalmers

A happy PhD student at Chalmers who is pleased with the high quality of the machines, which means that he can get on with his work, instead of wasting time troubleshooting.

More Reading

This article was not about quantum phenomena as such, but about WACQT’s cooperation with Myfab. Read about quantum computers to your heart’s delight at: www.chalmers.se/en/centres/wacqt/about%20us/Pages/default.aspx

The four pillars of quantum technology: www.chalmers.se/en/centres/wacqt/discover/Pages/default.aspx

Press Release: https://news.cision.com/se/chalmers/r/sveriges-kvantdatorbygge-vaxlar-upp,c3305292

More about quantum sensors: https://wieczorek-lab.com/, https://www.nano.lu.se/research/quantum-physics, https://www.nano.lu.se/ville-maisi

Article about quantum sensing of microwave photons: https://arxiv.org/abs/2011.05736

More about quantum communication work done at Albanova Nanolab: https://albanova-nanolab.org/

Of quantum computing, in Swedish: https://techworld.idg.se/2.2524/1.483680/datorn-som-ingen-forstar

Of quantum key distribution, in Swedish: https://techworld.idg.se/2.2524/1.620188/kvantkryptering-gor-all-annan-kryptering-overflodig-men-fungerar-det-verkligen