Organic solar cells are a rapidly advancing third-generation solar technology, offering a combination of high efficiency, low production cost, and mechanical flexibility. With reported power conversion efficiencies now exceeding 20%, organic solar cells are beginning to rival traditional silicon-based cells. They also enable new applications, such as integration into building materials and wearable electronics.
A key metric for evaluating solar cell performance is the fill factor, which measures how close a cell comes to delivering its theoretical maximum power. Fill factors are influenced by both the materials used and the design of the device. Historically, recombination losses have been considered the primary limitation to maximising fill factors. This is where electrons and holes recombine before contributing to the electrical current.
Recent research has highlighted another critical variable called transport resistance. In organic semiconductors, which typically have low electrical conductivity, charge carriers move slowly through the material. This slow transport increases the likelihood of recombination before the charges reach the electrodes, leading to significant fill factor losses even in highly efficient devices. If the electrons and holes were runners in a race, recombination losses are runners giving up before they finish. In organic materials, charge transport occurs via hopping: it is like the runners moving through mud, it is much harder – increasing the transport resistance – and they are more likely to give up before the end.
To address this, the authors developed an analytical model that incorporates the nature of this slow transport, energetic disorder, and systematically evaluates transport resistance using experimental data. By analysing current-voltage characteristics and light intensity-dependent open-circuit voltage across a range of temperatures, the model distinguishes between losses due to recombination and those due to charge transport.
This refined approach enables more accurate predictions of fill factors and offers practical strategies to minimise transport-related losses. Improving fill factors not only enhances the performance of organic solar cells but also provides insights applicable to other emerging photovoltaic technologies, helping to guide the future design of third-generation solar power.
Understanding periodically driven quantum systems is currently a major line of research.
These Floquet systems provide versatile platforms to investigate new physical phenomena such as time crystals, and can also be used to create fault-tolerant states for quantum computing.
What’s important here is the ability to precisely control the behaviour of the quantum system by designing its effective Hamiltonian – the mathematical object that governs how the system evolves over time.
When researchers want a system to behave in a very specific way, they engineer the Hamiltonian to match a desired target. This is called Floquet engineering.
Unfortunately, it’s not possible to create a simple (analytical) Floquet Hamiltonian for any given system, and mathematical tools such as the Magnus expansion are usually required to get a Hamiltonian that is sufficiently precise.
However, when you engineer a Hamiltonian using approximations, you get errors – not great for most applications and especially quantum computing.
Mitigating these errors is possible to some degree although up until now it’s been a one system at a time approach. What we really need is a systematic approach for mitigating these errors for any given system.
This is the problem that the latest paper by researchers Xu and Guo tries to address.
They used symmetries (like rotational or mirror symmetry) to simplify the design of these correction terms. This makes the calculations more manageable and the system more predictable.
They also provided a numerical method to calculate these corrections efficiently, which is important for practical implementation
They validated their method by creating Hamiltonians that are directly relevant for quantum computers.
The authors expect to further refine their method in the future, but this represents a big step forward towards practically engineering arbitrary Floquet Hamiltonians.
“Negative time” might sound like science fiction, but an international team of theorists and experimentalists has determined that a photon can, in fact, spend a negative amount of time in an excited atomic state while passing through a cloud of atoms. The finding could have applications in studies of light-matter interactions and quantum sensing – though not, alas, in time travel or other sensational effects.
Quantum mechanics has produced a lot of weird results, and the latest originated in 2022 with an experiment conducted by physicists at the University of Toronto, Canada. Led by Aephraim Steinberg, they found that when a photon passing through a cloud of atoms excites an electron in one of the atoms, it seems to spend a similar amount of time in this atomic excitation as a photon that passes straight through the cloud, apparently without exciting an atom at all.
A theoretical framework
To understand the theory behind this counterintuitive result, Steinberg and colleagues worked with researchers from the Massachusetts Institute of Technology in the US, Griffith University in Australia, and the Indian Institute of Science Education and Research. The framework they developed, which they now describe in APL Quantum, involves a single photon being sent into an atom cloud that is continuously monitored with a so-called “weak probe” that detects the presence of an atomic excitation anywhere in the cloud. Integrating this weak-probe signal over time thus provides a measure of how long the photon spends in an excited atomic state before it leaves the atom.
After crunching the numbers, the researchers made a prediction that surprised even them: the average excitation time can be negative. They also found that this excitation time should be the same as another, more familiar, time known as the group delay.
The team’s lead theorist, Griffith’s Howard Wiseman, says it is important to distinguish between these two times. A negative group delay, he observes, can be explained in a relatively intuitive way. Because the front of the photon pulse exits the atom cloud before the peak of the pulse enters it, and the peak never exits because most of the photons are scattered, there is, he says, an “illusion” that the photons leave the medium before they arrive.
However, he continues, what this framework actually measures is the time a transmitted photon spends in the atom cloud. It says nothing about whether such a photon excites an atom on its way through the cloud. Although it is normally assumed that any photon that excites an atom gets randomly scattered and never reaches the detector, says Wiseman, “We now say that this is not true and forward-scattered photons actually contribute a lot to the average measurement”.
Real-time measurements
To test this theory, Steinberg and colleagues set up a new experiment that sends two counter-propagating laser beams into a cloud of 85Rb atoms that have been cooled to 60–70 µK. The first beam contains the photons that may give rise to atomic excitation and may be either transmitted or scattered. The second beam is used for the weak measurements and detects the presence of an excitation via tiny shifts in its phase. These measurements required a high level of stability and a low level of interference in all parts of the setup.
After refining their system, the researchers measured average atomic excitation times for transmitted photons ranging from (–0.82 ± 0.31)𝜏0 for the most narrowband pulse to (0.54 ± 0.28)𝜏0 for the most broadband pulse. Here, 𝜏0 is the excitation time averaged over both scattered and transmitted photons, which is always positive and ranges from 10–20 ns, depending on various parameters. This result shows that negative excitation times do indeed have a physical reality in quantum measurements.
A matter of time
According to Steinberg, while he and his colleagues previously knew that negative numbers could pop out of the mathematics, they tended to sweep them under the rug and make excuses for them, assuming that while they correctly described the location of a peak, they weren’t physically relevant. “I am now led to revisit this and say: those negative numbers appear to have more physical significance that we would previously have attributed to them,” he tells Physics World. As a result, he hopes to “begin to investigate more deeply what we think the meaning of a ‘negative time’ is”.
Jonte Hance, a quantum physicist at Newcastle University, UK, who was not involved in this research, warns that interpreting negative time too literally can lead to paradoxes that aren’t necessary for the physics to work. Nevertheless, he says, the “anomalous” values recorded in the weak measurement “point to something interesting and quantum happening”.
Hance explains that in his view, a negative value for the mean atomic excitation time for transmitted photons implies contextuality – a property of quantum systems whereby measuring the system in different ways can make it look like it has incompatible properties if we assume that measuring the system does nothing to it. “Contextuality seems to be one of the tell-tale signs a quantum scenario may provide us with an advantage at a certain task over all possible classical ways of doing that task,” he says. “And so it makes me excited for what this could be used for.”
In this episode of the Physics World Stories podcast, Rosemary Coogan offers a glimpse into life as one of the European Space Agency’s newest astronauts. Selected as part of ESA’s 2022 cohort, she received astronaut certification in 2024, and is now in line to visit the International Space Station within the next five years. One day, she may even walk on the Moon as part of the Artemis programme.
Coogan explains what astronaut training really entails: classroom sessions packed with technical knowledge, zero-gravity parabolic flights, and underwater practice in Houston’s neutral buoyancy pool. Born in Northern Ireland, Coogan reflects on her personal journey. From a child dreaming of space, she went on to study physics and astrophysics at Durham University, then completed a PhD on the evolution of distant galaxies.
When not preparing for lift off, Coogan counts sci-fi among her interests – she loves getting lost in the world of possibilities. She’s also candid about the psychological side of astronaut training, and how she’s learned to savour the learning process itself rather than obsess over launch dates. Hosted by Andrew Glester, this episode captures both the challenge and wonder of preparing for an imminent journey to space.
When we started our PhDs in physics at Imperial College London, our paths seemed conventional: a lot of lab work, conferences and a bit of teaching on the side. What we did not expect was that within a couple of years we would be talking with MPs in the House of Commons, civil servants in Whitehall and business leaders in industry. We found ourselves contributing to policy reports and organizing roundtable discussions alongside policy-makers, scientists and investors; focusing on quantum technology and its impact on the economy and society.
Our journey into science policy engagement started almost by chance. Back in 2022 we received an e-mail from Imperial‘s Centre for Quantum Engineering Science and Technology (QuEST) advertising positions for PhD students to support evidence-based policy-making. Seeing it as an opportunity to contribute beyond the lab, we both took up the challenge. It became an integral part of our PhD experience. What started as a part-time role alongside our PhDs turned into something much more than that.
Mixing PhDs and policy
Getting involved From left: Dimitrie Cielecki, Elizabeth Pasatembou and Michael Ho in the UK Houses of Parliament. (Courtesy: Craig Whittall)
Her interest in science policy engagement started out of curiosity and the desire to make a more immediate impact during her PhD. “Research can feel slow,” she says. “Taking up this role and getting involved in policy gave me the chance to use my expertise in a way that felt directly relevant, and develop new skills along the way. I also saw this as an opportunity to challenge myself and try something new.”
Pasatembou also worked on a collaborative project between the Imperial Deep Tech Entrepreneurship and QuEST, conducting interviews with investors to inform the design of a tailored curriculum on quantum technologies for the investors community.
Dimitrie Cielecki
Dimitrie Cielecki joined Imperial’s Complex Nanophotonics group as a PhD candidate in 2021. The opportunity to work in science policy came at a time when his research was evolving in new directions. “The first year of my PhD was not straightforward, with my project taking unexpected, yet exciting, turns in the realm of photonics, but shifting away from quantum,” explains Cielecki, whose PhD topic was spatio-temporal light shaping for metamaterials.
After seeing an advert for a quantum-related policy fellowship, he decided to jump in. “I didn’t even know what supporting policy-making meant at that point,” he says. “But I quickly became driven by the idea that my actions and opinions could have a quick impact in this field.”
Cielecki is now a quantum innovation researcher at the Institute for Deep Tech Entrepreneurship in the Imperial Business School, where he is conducting research on the correlations between technical progress, investors’ confidence and commercial success in the emerging quantum sector.
We joined QuEST and the Imperial Policy Forum – the university’s policy engagement programme – in 2022 and were soon sitting at the table with leading voices in the nascent quantum technology field. We had many productive conversations with senior figures from most quantum technology start-ups in the UK. We also found ourselves talking to leaders of the National Quantum Technology Programme (including its chair, Sir Peter Knight); to civil servants from the Office for Quantum in the Department of Science, Innovation and Technology (DSIT); and to members of both the House of Commons and the House of Lords.
Sometimes we would carry out tasks such as identifying the relevant stakeholders for an event or a roundtable discussion with policy implications. Other times we would do desk research and contribute to reports used in the policy-making process. For example, we responded to the House of Commons written evidence inquiry on Commercialising Quantum Technologies (2023) and provided analysis and insights for the Regulatory Horizons Council report Regulating Quantum Technology Applications (2024). We also moderated a day of roundtable discussions with quantum specialists for the Parliamentary Office of Science and Technology’s briefing note Quantum Computing, Sensing and Communications (2025).
A two-way street
When studying science, we tend to think of it as a purely intellectual exercise, divorced from the real world. But we know that the field is applied to many areas of life, which is why countries, governments and institutions need policies to decide how science should be regulated, taught, governed and so on.
Science policy has two complimentary sides. First, it’s about how governments and institutions support and shape the practice of science through, for example, how funding is allocated. Second, science policy looks at how scientific knowledge informs and guides policy decisions in society, which also links to the increasingly important area of evidence-informed policy-making. These two dimensions are of course linked – science policy connects the science and its applications to regulation, economics, strategy and public value.
Quantum policy specifically focuses on the frameworks, strategies and regulations that shape how governments, industries and research institutions develop and deploy quantum technologies. Many countries have published national quantum strategies, which include technology roadmaps tied to government investments. These outline the infrastructure needed to speed up the adoption of quantum technology – such as facilities, supply chains and a skilled workforce.
In the UK, the National Quantum Technology Programme (NQTP) – a government-led initiative that brings together industry, academia and government – has pioneered the idea of co-ordinated national efforts for the development of quantum technologies. Set up in 2014, the programme has influenced other countries to adopt a similar approach. The NQTP has been immensely successful in bringing together different groups from both the public and private sectors to create a productive environment that advances quantum science and technology. Co-operation and communication have been at the core of this programme, which has led to the UK’s 10-year National Quantum Strategy. Launched in 2023, this details specific projects to help accelerate technological progress and make the country a leading quantum-enabled economy. But that won’t happen unless we have mechanisms to help translate science into innovation, resilient supply chains, industry-led standardization, stable regulatory frameworks and a trained workforce.
Up for discussion Quantum topics being debated as national policy include quantum cryptography and security. (Courtesy: iStock/wavebreakmedia)
Quantum technologies can bring benefits for national security, from advanced sensing to secure communications. But their dual-use nature also poses potential threats as the technology matures, particularly with the prospect of cryptographically relevant quantum computers – machines powerful enough to break encryption. To mitigate these risks in a complex geopolitical landscape, governments need tailored regulations, whether that’s preparing for the transition to post-quantum cryptography (making communication safe from powerful code-cracking quantum computers) or controlling exports of sensitive products that could compromise security.
Like artificial intelligence (AI) and other emerging technologies, there are also ethical considerations to take into account when developing quantum technologies. In particular, we need policies to ensure transparency, inclusivity and equitable access. International organizations such as UNESCO and the World Economic Forum have already started integrating quantum into their policy agendas. But as quantum technology is such a rapidly evolving new field, we need to strike a balance between innovation and regulation. Too many rules can stifle innovation but, on the other hand, policy needs to keep up with innovation to avoid any future serious incidents.
Language barriers
Policy engagement involves collaborating with three sets of stakeholders – academia; industry and investors; and policy-makers. But as we started to work with these groups, we noticed each had a different way of communicating, creating a kind of language barrier. Scientists love throwing around equations, data and figures, often using highly technical terminology. Industry leaders and investors, on the other hand, talk in terms of how innovations could affect business performance and profitability, and what the risk for their investments could be. As for policy-makers, they focus more on how to distinguish between reality and hype, and look at budgets and regulations.
We found ourselves acting as cross-sector translators, seeking to bridge the gap between the three groups. We had to listen to each stakeholder’s requirements and understand what they needed to know. We then had to reframe technical insights and communicate them in a relevant and useful way – without simplifying the science. Once we grasped everyone’s needs and expectations, we offered relevant information, putting it into context for each group so everyone was on the same page.
To help us do this, we considered the stakeholders as “inventor”, “funder”, “innovator” or “regulator”. As quantum technology is such a rapidly growing sector, the groupings of academia, industry and policy-makers are so entangled that the roles are often blurred. This alternative framework helped us to identify the needs and objectives of the people we were working with and to effectively communicate our science or evidence-backed messages.
Finding the right people
During our time as policy fellows, we were lucky to have mentors to teach us how to navigate this quantum landscape. In terms of policy, Craig Whittall from the Imperial Policy Forum was our guide on protocol and policy scoping. We worked closely with QuEST management – Peter Haynes and Jess Wade – to organize discussions, collect evidence from researchers, generate policy leads, and formulate insights or recommendations. We also had the pleasure of working with other PhD students, including Michael Ho, Louis Chen and Victor Lovic, who shared the same passion for bridging quantum research and policy.
Having access to world-leading scientists and a large pool of early-career researchers spread across all departments and faculties, facilitated by the network in QuEST, made it easier for us to respond to policy inquiries. Early on, we mapped out what quantum-related research is going on at Imperial and created a database of the researchers involved. This helped inform the university’s strategy regarding quantum research, and let us identify who should contribute to the various calls for evidence by government or parliament offices.
Getting started Imperial College London encourages its researchers – established and early-career – to get involved in shaping policy. From left: Dimitrie Cielecki, Michael Ho, Louis Chen, Elizabeth Pasatembou. (Courtesy: Elizabeth Pasatembou)
PhD students are often treated as learners rather than contributors. But our experience showed that with the right support and guidance, early-career researchers (ECRs) such as ourselves can make real impact by offering fresh perspectives and expertise. We are the scientists, innovators or funders of the future so there is value in training people like us to understand the bigger picture as we embark on our careers.
To encourage young researchers to get involved in policy, QuEST and DSIT recently organized two policy workshops for ECR quantum tech specialists. Civil servants from the Office for Quantum explained their efforts and priorities, while we answered questions about our experience – the aim being to help ECRs to engage in policy-making, or choose it as a career option.
In April 2025 QuEST also launched an eight-week quantum primer for policy-makers. The course was modelled on a highly successful equivalent for AI, and looked to help policy-makers make more technically informed policy discussions. The first cohort welcomed civil servants from across government, and it was so highly reviewed a second course will be running from October 2025.
Our experience with QuEST has shown us the importance of scientists taking an active role in policy-making. With the quantum sector evolving at a formidable rate, it is vital that a framework is in place to take research from the lab to society. Scientists, industry, investors and policy-makers need to work together to create regulations and policies that will ensure the responsible use of quantum technologies that will benefit us all.
Duality ellipse The amount of coherence is linked to the amount of waveness and particleness of a quantum system through an duality ellipse relation. The coherence, modulated by the object shape, can be found by measuring the waveness and particleness of the photon. The more coherence is present at a specified location in the object, the more circular the duality ellipse relation. Finding the eccentricity of the duality ellipse relation at different points in the object then generates an object image. (Courtesy: Xiaofeng Qian)
The theory of quantum mechanics was born out of the need to explain how an object could behave both as a particle and as a wave. Since then, researchers have been attempting to understand and quantify the degree of “waveness” and “particleness” of quantum systems.
Now, Pawan Khatiwada and Xiaofeng Qian, both based at the Stevens Institute of Technology in the US, have published a paper in Physical Review Research unveiling the missing piece of the puzzle in the unique relationship between the wave and particle nature of a quantum object. The key piece is coherence, which describes the statistical phase relationship between the possible states a quantum system can adopt. If the phase relationship is well-defined, that is, stable and consistent, the system is coherent and therefore has the potential to exhibit interference, a wave-like property. If the system is incoherent, the phase relationship is random and variable. Coherence therefore characterizes the total capacity or potential of a quantum system to exhibit wave-like behaviour. The degree of waveness that is then realised and can be observed is known as the visibility and is characterized by the intensity of the maxima and minima in the interference pattern. The amount of particleness, known as the predictability, is a measure of how well we can predict the path a particle will take and is defined by the degree to which the particle is taking only one of the two paths.
The amount of interference that an object exhibits (the visibility) may only be a fraction of its total capacity to exhibit wave-like behaviour (its coherence) and can change according to how we choose to observe the system. For example, it we decide to track the position of the photon through a double-slit diffraction grating, no interference effects are observed, and the photon will behave as a particle. However, if we remove this tracking, the usual patterns of dark and light bands that are characteristic of interference emerge, and the photon behaves as a wave. In the first situation the visibility is low, but the coherence can nonetheless be high; it is our access to information about the path of the photon that destroys the interference pattern.
New quantum duality relation
Half a century ago, the relationship between a quantum object’s visibility and predictability was characterized by an inequality that restricted the sum of the square of the visibility and predictability to be less than or equal to one. However, this inequality could not have been telling the full story since it failed to capture the exclusivity of wave and particle-like behaviours such that if an object is more wave-like, it should display less particle properties and vice versa. The inequality instead permitted both wave and particle properties to increase or decrease simultaneously.
Qian and Khatiwada recast this inequality into a precise equality relation by introducing coherence as an additional variable. The relation then describes a trade-off between the coherence, visibility, and predictability, where the sum of the square of the predictability and the fraction of visibility exhibited out of the total coherence, must be equal to one. This relationship forms an ellipse equation, with the eccentricity determined by the coherence, where maximum coherence yields a circle and partial coherence an ellipse and is thus known as a duality ellipse relation.
Duality as a resource
Qian applies this formula to a technique for measuring quantum objects called quantum imaging with undetected photons (QIUP) first conceived by Nobel laureate Anton Zeilinger’s research group. One of a pair of entangled photons interacts with the object and depending on its shape, the coherence will change and can be tracked by following the behaviour of the second photon, without ever detecting the first. Information about the object is then given by the second photon even though it has never interacted with it. The more coherence is present at a specified location in the object, the more circular the duality ellipse relation (see figure).
Finding the eccentricity of the duality ellipse relation at each point in the object thus provides a map of its shape, and therefore an image of the object. Since the coherence is related to the visibility and predictability of a quantum object, Qian explains that their formula then becomes a crucial link between ‘fundamental properties of a quantum system like waveness and particleness and operational properties which hold information about an object’.
Of course, most experimental scenarios are not ideal, and Qian accounts for these experimental imperfections in a modified relation. Remarkably, the overall pattern of the elliptical duality relation remains the same and therefore this imaging technique proves to be robust even under such conditions.
Qian explains that ‘in a similar way quantum entanglement is a useful resource in quantum information and quantum computing, quantum duality also proves to be too for certain quantum tasks’. His group are now working on uncovering further avenues through which quantum duality can act as a quantum resource.
Researchers at the University of Warwick in the UK have created an ultrasensitive magnetometer based on nitrogen-vacancy centres in diamond that’s small enough to be used for keyhole surgery. The sensor, which currently measures just 1 cm in diameter and could be made even smaller in the future, is designed to detect small cancer metastases via endoscopy or laparoscopy.
“It’s really bad news when tumour cells spread from their original site, and so it’s very important to detect this metastatic cancer as soon as possible,” says physicist Gavin Morley, who led this research effort together with his doctoral student Alex Newman. “The new cancers are often lodged in the lymph nodes and our device could be used to detect these cancers early when they are still small.”
Existing techniques to detect metastatic tumours include MRI and CT, but these technologies can only detect tumours that are at least 2 mm across. While alternatives like sentinel lymph node biopsy can detect tumours with a volume that is 1000 times smaller, this technique typically involves the use of radioactive tracer fluids that require special safety precautions, or blue dyes, which cause an allergic reaction in one in a hundred people.
Tracer travels to the lymph nodes
Medical device company Endomag recently developed a clinical technique that involves the surgeon injecting a magnetic tracer into a breast cancer tumour, explains Morley. “The tracer fluid travels to the lymph nodes and the surgeon can then identify the metastatic cells there and remove them.”
While this approach is efficient for breast cancer, the magnetometers employed today to detect the tracer are too large for use in keyhole surgery or endoscopy, he explains. “We wanted to create a device that can be used to detect the metastatic tumours and so built a version that’s smaller. The surgeons we’ve spoken to say that colorectal cancer could be the best place for us to focus on first for our magnetometer.”
NV magnetic sensor
Morley’s group has been working on magnetic field sensors using diamonds and lasers for ten years now. The diamonds are grown by the company Element Six in Oxford and they contain quantum defects known as nitrogen-vacancy (NV) centres. These are created when a pair of adjacent carbon atoms in the diamond lattice is replaced by a nitrogen atom, leaving one lattice site vacant. An NV centre is basically an isolated spin that is highly sensitive to an external magnetic field and it emits florescent light in a way that depends on the intensity and direction of this field. Measuring this light allows it to be used as a magnetic sensor.
“Our speciality is using optical fibres to send laser light into the diamond and detect the red light that comes back,” says Morley.
In this work, reported in Physical Review Applied, it was Newman who built the new sensor, Morley tells Physics World. “Alex likes fixing old sports cars and I liked the way he applied that thinking to this new technology. He tries different strategies and has built new types of diamond sensors that no-one has managed to build before.”
The Warwick researchers are now working on a number of applications for their sensors: as well as use within healthcare, they could be employed in space applications and future fusion power plants, says Morley. “Indeed, for Alex’s project, we were working on detecting damage in steel to help the National Nuclear Laboratory who have nuclear waste stored in steel containers. I then met Stuart Robertson, who is a breast cancer surgeon at the University Hospitals Coventry and Warwickshire: he told me how useful the Endomag solution is for breast cancer metastatic cells and I thought we could build a magnetometer that would help.”
Working with several surgeons, Morley, Newman and colleagues are now developing this work as part of the UK Quantum Biomedical Sensing Research Hub (Q-BIOMED). “For example, Jamie Murphy in the Cleveland Clinic in London is an expert on keyhole surgery, with a big interest in colorectal cancer,” says Morley. “And Conor McCann is an expert on gut health at the UCL Great Ormond Street Institute of Child Health. We’re interested in spinning out a company ourselves to take this forward alongside other applications of our diamond sensors.”
The researchers are also busy making the sensor even smaller. “At the moment the probe is 1 cm across, but we think we can get it down to be only 3 mm,” says Morley. “While 1 cm is small enough for keyhole surgery and endoscopy, getting it even smaller would make it useful for even more types of surgeries.”
Flexible thinking, scalable innovation Delft Circuits has established itself as a one-stop shop for scalable cryogenic I/O assemblies in quantum computing. The company’s Cri/oFlex® cabling platform combines fully integrated filtering with a compact footprint and low heatload. (Courtesy: Delft Circuits)
As manufacturers in the nascent quantum supply chain turn their gaze towards at-scale commercial opportunities in quantum computing, the scenic city of Delft in the Netherlands is emerging as a heavyweight player in quantum science, technology and innovation. At the heart of this regional quantum ecosystem is Delft Circuits, a Dutch manufacturer of specialist I/O cabling solutions, which is aligning its product development roadmap to deliver a core enabling technology for the scale-up and industrial deployment of next-generation quantum computing, communications and sensing systems.
Kuitenbrouwer “Cri/oFlex® allows us to increase the I/O cabling density easily – and by a lot.” (Courtesy: Delft Circuits)
In brief, the company’s Cri/oFlex® cryogenic RF cables comprise a stripline (a type of transmission line) based on planar microwave circuitry – essentially a conducting strip encapsulated in dielectric material and sandwiched between two conducting ground planes. The use of the polyimide Kapton® as the dielectric ensures Cri/oFlex® cables remain flexible in cryogenic environments (which are necessary to generate quantum states, manipulate them and read them out), with silver or superconducting NbTi providing the conductive strip and ground layer. The standard product comes as a multichannel flex (eight channels per flex) with a range of I/O channel configurations tailored to the customer’s application needs, including flux bias lines, microwave drive lines, signal lines or read-out lines.
“As quantum computers evolve – think more and more qubits plus increasingly exacting requirements on gate fidelity – system developers will reach a point where current coax cabling technology doesn’t cut it anymore,” explains Daan Kuitenbrouwer, co-founder of Delft Circuits. “The key to our story is that Cri/oFlex® allows us to increase the I/O cabling density easily – and by a lot – to scale the number of channels in a single system while guaranteeing high gate fidelities [minimizing noise and heating] as well as market-leading uptime and reliability.”
Quantum alignment
To put some hard-and-fast performance milestones against that claim, Kuitenbrouwer and colleagues have just published a granular product development roadmap that aligns Cri/oFlex® cabling specifications against the anticipated evolution of quantum computing systems – from 150+ qubits today out to 40,000 qubits and beyond in 2029 (see figure, “Quantum alignment”).
Quantum alignment The new product development roadmap from Delft Circuits starts with the guiding principles, highlighting performance milestones to be achieved by the quantum computing industry over the next five years – specifically, the number of physical qubits per system and gate fidelities. By extension, cabling metrics in the Delft Circuits roadmap focus on “quantity”: the number of I/O channels per loader (i.e. the wiring trees that insert into a cryostat, with typical cryostats having between 6–24 slots for loaders) and the number of channels per cryostat (summing across all loaders); also on “quality” (the crosstalk in the cabling flex). To complete the picture, the roadmap outlines product introductions at a conceptual level to enable both the quantity and quality timelines. (Courtesy: Delft Circuits)
“Our roadmap is all about enabling, from an I/O perspective, the transition of quantum technologies out of the R&D lab into at-scale practical applications,” says Kuitenbrouwer. “As such, we studied the development roadmaps of more than 10 full-stack quantum computing vendors to ensure that our ‘guiding principles’ align versus the aggregate view of quantity and quality of qubits targeted by the system developers over time.”
Notwithstanding the emphasis on technology innovation and continuous product improvement, Delft Circuits is also “coming of age” in line with the wider quantum community. Most notably, the company’s centre of gravity is shifting inexorably from academic end-users to servicing vendors large and small in the quantum supply chain. “What we see are full-stack quantum computing companies starting to embrace horizontal thinking – which, in our case, means a technology partner able to solve their entire I/O cabling challenge,” explains Kuitenbrouwer.
To gain traction, however, systems integrators at the sub-stack level must, as a given, design their product offering with industrial metrics front-and-centre – for example, scalability, manufacturability, reliability, cost per I/O channel and second-sourcing. Equally important is the need to forge long-term vendor-customer relationships that often move beyond the transactional into the realm of co-development and collaboration – though all against a standardized package of cabling options.
“We integrate Cri/oFlex® with cryostats that have relatively standard vacuum feedthroughs and thermalization – more or less the same across the board,” says Kuitenbrouwer. What changes is the type of qubit – superconducting, spin, photonic – which in turn determines the configuration of the I/O line and where to place the attenuators, low-pass filters and IR filters. “This is something we can adjust relatively easily – at high volume and high reliability – with the whole I/O package installed and tested at the customer premises,” he adds.
Timing is key for quantum advantage
Commercially, Delft Circuits is already making real headway, getting “in the door” with many of the leading developers of quantum computing systems in North America and Europe. One of the main reasons for that is the ability to respond to customer requirements in an agile and timely fashion, argues Sal Bosman, a fellow co-founder of Delft Circuits.
Bosman “Currently, we are the only industrial supplier able to deliver flexible circuits of superconducting materials at scale.” (Courtesy: Delft Circuits)
“We work on the basis of a very structured design process, playing to our strengths in superconductor fabrication, integrated microwave components and cryogenic engineering,” Bosman notes. “We have also developed our own in-house software to simulate the performance of Cri/oFlex® cabling in full-stack quantum systems. No other vendor can match this level of customer support and attention to detail.”
Right now, though, it’s all about momentum as Delft Circuits seeks to capitalize on its first-mover advantage and, what Bosman claims, is the unique value proposition of its Cri/oFlex® technology: a complete and inherently scalable I/O solution with integrated flex cables incorporating filters and high-density interconnects to quantum chips or control electronics.
With this in mind, the company is busy constructing a new 750m2 clean-room (with an option to double that footprint) alongside its existing 1000m2 in-house pilot-production and test facility. “Currently, we are the only industrial supplier able to deliver flexible circuits of superconducting materials at scale,” concludes Bosman.
“Over the next two to three years,” he adds, “we have a credible opportunity to grab significant market share when it comes to cabling I/O for quantum. Watch this space: a lot of customers are already coming to us saying ‘we don’t want to buy more coax, we want to work with you.’”
Location, location, location
Cryogenic integration Delft Circuits can supply fully pre-assembled loaders with Cri/oFlex® cabling inside. (Courtesy: Delft Circuits)
Delft Circuits sits within a thriving regional cluster for quantum science and technology called Quantum Delta Delft, which is centred around the canal-ringed city of Delft between The Hague and Rotterdam.
Formed in 2017 and initially located at the Faculty of Applied Sciences at Delft University of Technology (TU Delft), Delft Circuits has since grown as an independent company and is now based in the historic Cable District, where its facilities include a dedicated fabrication, pilot-production and testing area.
TU Delft is itself home to a high-profile interfaculty research institute called QuTech, a collaboration with the Netherlands Organisation for Applied Scientific Research (TNO) that’s tasked with developing full-stack hardware and software layers (including enhanced qubit technologies) for quantum computing and quantum communications systems.
Alongside this academic powerhouse, the Delft region has seen the emergence of other quantum tech start-ups like QuantWare (quantum chips), Qblox (control electronics) and Orange Quantum Systems (test solutions). All three companies work closely with Delft Circuits as part of the ImpaQT UA cooperative, a joint effort to develop open standards and interoperable technologies that enable system integrators to build quantum computing hardware stacks from off-the-shelf components.
“The ImpaQT UA story is ongoing,” explains Kuitenbrouwer. “As partners, we are super-complementary and collaborate closely to shape the future of quantum computing.” That’s why the new development roadmap is so important for Delft Circuits: to communicate a vision from the “component layer” up the value chain to the full-stack quantum computing companies.
As well as the talent pipeline that comes with proximity to TU Delft and QuTech, Quantum Delta Delft is home to TNO’s Quantum Information Technology Testing (QITT) Facility, which enables European companies to evaluate their cryogenic or non-cryogenic quantum devices and software in a full-stack quantum computing set-up.
A few weeks ago, I experienced a classic annoyance of modern life: one of my computer games stopped working. The cause? An “update” to the emulator that translates old games into programs that today’s machines can execute. In my case, this update broke the translation process, and the tenuous thread of hardware and software connecting my laptop to the game’s 30-year-old code was severed.
For individuals, failures like this are irritating. But for the wider digital ecosystem, they’re a real problem – so much so, in fact, that Vint Cerf, who’s known as one of the “fathers of the Internet”, made them the subject of his talk at last week’s Heidelberg Laureate Forum (HLF) in Heidelberg, Germany.
“My big worry is that all this digital stuff won’t be there when we would like it to be there, or when our descendants would like to have it,” Cerf said.
How it used to work
Historically, the best ways of preserving information involved writing it on durable materials such as clay tablets, high-quality paper, or a form of animal skin known as vellum. These media, Cerf observed, “have one thing in common: they don’t require electricity to be stored and preserved.”
Digital media, in contrast, are much less robust. “Many of them are magnetic, and the magnetic material wears away after a while,” Cerf explained. Consequently, some old tapes are now so fragile that attempting to read them can actually lift the magnetic material off the surface: “You read it once and that’s it. It’s now transparent tape,” he said.
Being able to read data is just the beginning, though. As my broken computer game shows, you also need programs and equipment that can persuade those data to do things. “That’s often the thing that goes first,” Cerf told me in a press conference after his talk. For example, when Cerf recently tried to retrieve data from an old three-and-a-half-inch floppy disk, he discovered that doing so would require three additional components: a drive that could read the disk, a program that could open the files stored on the disk and an old computer that could run the program. “I needed a whole lot of software help and several stages in order to make that digital content useful,” Cerf said.
Creating ‘digital vellum’
As for how to fix this problem and create a digital version of vellum, Cerf, who has been the “Chief Internet Evangelist” at Google since 2005, listed three ideas that he finds interesting. The first involves a New Jersey, US-based company called SPhotonix that does research and development work in the UK and Switzerland. It’s using lasers to write bits of data into chunks of quartz crystal, which is a very long-lasting medium. However, each crystal is roughly the size of a hockey puck, and Cerf thinks that “real work” still needs to be done to organize the information the material holds.
The second idea is partly inspired by the clay tablets that proved so successful at preserving cuneiform writing from ancient Mesopotamia. Cerabyte, a start-up with facilities in Austria, Germany and the US, has developed a ceramic material that its founders claim could “store all data virtually forever”.
The third idea, and the one that seems to appeal most to Cerf, is to write digital information into DNA. That might sound like an inherently fragile medium, but as Cerf pointed out, “It’s actually a very robust molecule – otherwise, life wouldn’t have persisted for several billion years.” Provided you dehydrate the DNA first, he added, it lasts for “quite a long time”.
The question of how to read such information is not an easy one, and Cerf doesn’t have an answer to it. He is, however, hopeful that someone will find one. At the HLF, where he is such a revered figure that even the journalists want to take photos with him, he issued a call to arms for the young researchers in the audience. “I want you to appreciate the scope of the work that is required to preserve digital things,” Cerf told them. Without that work, he added, “recreating a digital environment in 100 years is not going to be a trivial matter.”
An unconventional approach to solving the dark energy problem called the cosmologically coupled black hole (CCBH) hypothesis appears to be compatible with the observed masses of neutrinos. This new finding from researchers working at the DESI collaboration suggests that black holes may represent little Big Bangs played in reverse and could be used as a laboratory to study the birth and infancy of our universe. The study also confirms that the strength of dark energy has increased along with the formation rate of stars.
The Dark Energy Spectroscopic Instrument (DESI) is located on the Nicholas U Mayall four-metre Telescope at Kitt Peak National Observatory in Arizona. Its raison d’être is to shed more light on the “dark universe” – the 95% of the mass and energy in the universe that we know very little about. Dark energy is a hypothetical entity invoked to explain why the rate of expansion of the universe is (mysteriously) increasing – something that was discovered at the end of the last century.
According to standard theories of cosmology, matter is thought to comprise cold dark matter (CDM) and normal matter (mostly baryons and neutrinos). DESI can observe fluctuations in the matter density of the universe known as baryonic acoustic oscillations (BAOs), which are density fluctuations that were created after the Big Bang in the hot plasma of baryons and electrons that prevailed then. BAOs expand with the growth of the universe and represent a sort of “standard ruler” that allows cosmologists to map the universe’s expansion by statistically analysing the distance that separates pairs of galaxies and quasars.
Largest 3D map
DESI has produced the largest such 3D map of the universe ever and it recently published the first set of BAO measurements determined from observations of over 14 million extragalactic targets going back 11 billion years in time.
In the new study, the DESI researchers combined measurements from these new data with cosmic microwave background (CMB) datasets (which measure the density of dark matter and baryons from a time when the universe was less than 400,000 years old) to search for evidence of matter converting into dark energy. They did this by focusing on a new hypothesis known as the cosmologically coupled black hole (CCBH), which was put forward five years ago by DESI team member Kevin Croker, who works at Arizona State University (ASU), and his colleague Duncan Farrah at the University of Hawaii. This physical model builds on a mathematical description of black holes as bubbles of dark energy in space that was introduced over 50 years ago. CCBH describes a scenario in which massive stars exhaust their nuclear fuel and collapse to produce black holes filled with dark energy that then grows as the universe expands. The rate of dark energy production is therefore determined by the rate at which stars form.
Neutrino contribution
Previous analyses by DESI scientists suggested that there is less matter in the universe today compared to when it was much younger. When they then added the additional, known, matter source from neutrinos, there appeared to be no “room” and the masses of these particles therefore appeared negative in their calculations. Not only is this unphysical, explains team member Rogier Windhorst of the ASU’s School of Earth and Space Exploration, it also goes against experimental measurements made so far on neutrinos that give them a greater-than-zero mass.
When the researchers re-interpreted the new set of data with the CCBH model, they were able to resolve this issue. Since stars are made of baryons and black holes convert exhausted matter from stars into dark energy, the number of baryons today has decreased in comparison to the CMB measurements. This means that neutrinos can indeed contribute to the universe’s mass, slowing down the expansion of the universe as the dark energy produced sped it up.
“The new data are the most precise measurements of the rate of expansion of the universe going back more than 10 billion years,” says team member Gregory Tarlé at the University of Michigan, “and it results from the hard work of the entire DESI collaboration over more than a decade. We undertook this new study to confront the CCBH hypothesis with these data.”
Black holes as a laboratory
“We found that the standard assumptions currently employed for cosmological analyses simply did not work and we had to carefully revisit and rewrite massive amounts of a lot of cosmological computer code,” adds Croker.
“If dark energy is being sourced by black holes, these structures may be used as a laboratory to study the birth and infancy of our own universe,” he tells Physics World. “The formation of black holes may represent little Big Bangs played in reverse, and to make a biological analogy, they may be the ‘offspring’ of our universe.”
The researchers say they studied the CCBH scenario in its simplest form in this work, and found that it performs very well. “The next big observational test will involve a new layer of complexity, where consistency with the large-scale features of the Big Bang relic radiation, or CMB, and the statistical properties of the distribution of galaxies in space will make or break the model,” says Tarlé.
