Skip to main content

Reset your password

Please enter the e-mail address you used to register to reset your password

Note: The verification e-mail to change your password should arrive immediately. However, in some cases it takes longer. Don't forget to check your spam folder.

If you haven't received the e-mail in 24 hours, please contact customerservices@ioppublishing.org

Registration complete

Thank you for registering with Physics World
If you'd like to change your details at any time, please visit My account

Close
Home » Page 513

Tandem solar cells break new record

Solar cells made from a combination of silicon and a complex perovskite have reached a new milestone for efficiency. The new tandem devices, made by Steve Albrecht and colleagues at the Helmholtz-Zentrum Berlin, Germany, have a photovoltaic conversion efficiency (PCE) of 29.15%, beating out the previous best reported value of 26.2%. They also retain 95% of their initial efficiency even after 300 hours of operation and have an open-circuit voltage as high as 1.92 V.

Solar cells containing two photoactive semiconducting materials with different but complementary electronic band gaps can reach much higher PCEs when used in a tandem configuration than either material on its own. Perovskites, which have the chemical formula ABX3 (where A is typically caesium, methylammonium or formamidinium; B is lead or tin; and X is iodine, bromine or chlorine), are one of the most promising thin-film solar-cell materials around because they are efficient at converting the visible part of the solar spectrum into electrical energy. Since silicon is an efficient absorber of infrared light, combining silicon with a perovskite helps to make the most of the Sun’s output.

“Perfect bed” for perovskite

Working with Vytautas Getautis and his team at the Kaunas Technical University in Lithuania, Albrecht and colleagues constructed their tandem solar cell by sandwiching a self-assembled monolayer (SAM) of a novel carbazole-based molecule between a complex perovskite with a 1.68 eV band gap and an indium tin oxide electrode connected to the silicon. Electrical charge carriers (electrons and holes) can diffuse through perovskites quickly and over long lengths, and adding the SAM layer facilitates the flow of electrons and holes even further. “We first prepared the perfect bed, so to speak, on which the perovskite lays on,” explains Amran Al-Ashouri, a member of Albrecht’s team.

To understand the various processes at play at the interface of the perovskite and the SAM, the researchers studied the interface using a combination of transient photoluminescence spectroscopy, computational modelling, electrical characterization and time-resolved terahertz photoconductivity measurements. The information gleaned from these and other techniques enabled them to optimize the device’s so-called fill factor – a key parameter for photovoltaic devices, and one where perovskite-based solar cells have long fallen short of better-established solar cell materials.

Accelerating hole transport

In Albrecht and colleagues’ experiments, the fill factor depends on how many charge carriers are “lost” on their way out of the SAM-perovskite interface. These losses occur due to a process known as nonradiative recombination, in which excited electrons and holes recombine without emitting light – an unwanted interaction that lowers the efficiency of power conversion.

In the new tandem device, the electrons flow in the direction of incoming sunlight through the SAM, while the holes move in the opposite direction through the SAM into the electrode. The researchers observed, however, that the speed at which holes are extracted is much lower than the corresponding speed for electrons – something that would normally limit the fill factor. According to Al-Ashouri, the new SAM solves this problem by considerably accelerating hole transport, which improves the fill factor and makes the perovskite cell more efficient.

Members of the team, which also includes researchers from the universities of Potsdam in Germany, Ljubljana in Slovenia and Sheffield in the UK as well as the Physikalisch-Technische Bundesanstalt (PTB), HTW Berlin and the Technische Universität Berlin, say that the maximum PCE possible for their design — 32.4% — is now “within reach”. “To this end, we plan to further reduce resistive losses in the tandem solar cell to explore the full PCE potential well above 30%,” team member Eike Köhnen tells Physics World.

The research is detailed in Science.

Elastic diamond could be used to make LEDs and lasers

Normally brittle diamond has been stretched by almost 10% by an international team of scientists in China. Modelling done by the team suggests that this stretching significantly alters the electronic properties of the material, which could lead to new applications for diamond such as LEDs and variable-colour lasers.

As well as shining brilliantly in jewellery, diamond has long fascinated scientists for its range of extraordinary properties. Diamond is the hardest material on Earth, has very high charge-carrier mobility and is almost totally transparent. It also has a high thermal conductivity and a very high dielectric breakdown strength – a combination that would be useful in high-power electronic devices.

Unfortunately, diamond has several properties that make it unsuitable for electronic applications. It has a very large indirect band gap of 5.47 eV, giving it very few free charge carriers. Moreover, indirect band gap semiconductors do not interact efficiently with light via electronic transitions, so diamond cannot be used as a light source or a solar cell.

Strain engineering

The electronic properties of semiconductors can be altered by adding impurity atoms (doping), but this is very difficult in diamond because of the rigidity of its carbon lattice. An alternative to doping is strain engineering, which involves changing the atomic spacing in a lattice. The processor in every modern PC, for example, contains strained silicon with electron or hole mobility that has been altered by strain engineering.

“In silicon – which we’re also working on – if you apply just a 1% strain you see a huge change in electrical mobility,” explains joint team leader Yang Lu of the City University of Hong Kong. In the new research, the team looked at whether strain engineering could help with diamond.

Theoretical calculations suggest that the diamond lattice should withstand strains up to about 12%. In practice however, when real diamonds are stressed, the strain becomes concentrated at defects in the lattice, causing the diamonds to break rather than stretch.

High elasticity

In 2018, Lu and colleagues etched nanoscale diamonds and, by bending them, demonstrated that diamonds could show high elasticity: “We discovered this, but the straining method can’t be well characterized,” explains Lu. “Since then, a lot of other groups have done follow up work – mostly computational – predicting band gap change, metallization and even superconductivity.” Testing these predictions, however, requires the precise characterization of the strain.

In their new research, Lu and colleagues used advanced fabrication processes to produce micron-scale, T-shaped samples of single-crystal diamond, which they stretched using a custom-made mechanical gripper (see figure). They measured different Young’s moduli along different crystallographic axes, achieving a maximum strain of 9.7% before the diamond fractured. At lower values, the lattice returned to its original configuration when the strain was removed.

“We managed to microfabricate diamond into the shape and geometry that we want,” says Lu. “This is just an illustration of the device concept. We can scale up.”

Direct band gap

Experimental limitations prevented the researchers from directly measuring the effects of stretching on the electronic band structure of their samples. However, calculations done using density functional theory and their experimental data suggest that along one crystalline axis, the band gap falls from above 5 eV to around 3 eV at around 9% strain. Along another axis, the band gap changes from indirect to direct.

Lu’s group is now looking at several possible applications for stretched diamond including making lasers of different colours. “Intrinsically, diamond has a wide band gap so would produce UV light,” he says. “Now imagine if we stretch the diamond, it goes from ultraviolet to violet, and if we stretch it further it goes from violet to red.”

Christoph Nebel, founding director of the German diamond and carbon technology consultancy Diacara, says that tuning the band structure of semiconductors using pressure and tensile stress “is a very old story going to the 1950s”. He adds, “What is surprising is that you can apply such a strong tensile stress without the diamond breaking.”

Mark Newton of the University of Warwick in the UK agrees that the crucial result is the elasticity: “They’ve opened up all of these possibilities by demonstrating this level of strain. The changes that this could have for all the electronic and optoelectronic properties are just a sentence at the end of their paper but that’s where the cool stuff will come.” He is particularly intrigued by the change to a direct band gap which he says, could potentially lead to a diamond-based LED.

The research is described in Science.

 

Sweat evaporator could power fitness trackers, supersonic slap cooks a chicken

If you have resolved to get more exercise in 2021, researchers in Singapore have just the fabric for you. They have invented a new material that allows skin to evaporate sweat six times faster than normal fabrics – and it also stores 15 times more moisture. What is more, the material can generate electrical energy from the sweat it has captured.

It was created by Tan Swee Ching, Ding Jun and colleagues at the National University of Singapore – perhaps not surprising given the city’s hot and humid climate. As well as reducing odours and infections associated with excess sweating, the material could power wearable electronic devices such as watches and fitness trackers.

Above is a finite element analysis simulation that purports to show that you can cook a chicken by slapping it at 3725.95 mph (about 6000 km/h). What puzzles me is that the chicken appears to be vapourized by the slap, whereas the hand only appears to break into several pieces. Surely a human hand is not significantly stronger than a chicken?

Join the next generation of quantum communicators

2020, despite all its multitudinous awfulness, was a good year for quantum science. Here at Physics World, we come across important papers nearly every week on topics from quantum algorithms and improved qubit architectures to better quantum sensors and imaginative new experiments on quantum fundamentals. In fact, there’s so much exciting research coming out in this area that we can’t write about it fast enough – at least, not without some help.

Back in 2017 Physics World set up student contributor networks in materials/nanoscience and medical physics/bioengineering. These contributor networks are made up of PhD students with a passion for science communication, and for the past three and a half years, members of the Physics World editorial team have provided them with training and mentorship to help them write about the most exciting new research in their fields. Some of our contributors are now pursuing careers in science communication, writing for us (and other media outlets) as paid freelancers after “graduating” from the network. Others have remained in academia or entered industry jobs, using their communication skills to write better grant applications, papers and reports. All have benefited from the chance to connect with other members of the network and publish their work on a site read by hundreds of thousands of professional scientists all over the world (they even get their own page on our site – here’s an example).

All of which makes me very pleased to say that we are now expanding this programme to include a new contributor network, this one focused on quantum science and technology. As with our existing networks, we’re looking for PhD students with a talent for writing and a passion for communicating the latest research to a broader scientific audience. We’ll provide initial training on how to craft compelling science news stories, and we’ll also ensure that you get regular feedback from other student contributors as well as our professional journalists, giving you ongoing support to hone your scientific writing skills.

We expect each contributor to write 1–2 articles (of about 500 words each) per quarter about a recent paper in their field, and to provide pre-publication feedback on a similar number of articles written by other contributors in the network. The goal is that after 1–2 years in the network, every contributor should have a body of well-written, published work that they can put on their CVs – either as evidence of strong communication skills when applying for jobs in academia or industry, or as a springboard to a career in science communication. More information is available here.

Get involved

If this sounds like it’s right up your street, and you’d like to apply to join the new network, please send the following information to pwld@ioppublishing.org by 31 January 2021:

  • A short (~300 words or less) explanation of why you’d like to join the network and what you think you’d bring to it. Previous experience in science communication is not required, but if you have some, please say so.
  • A short (~500 words or less) description of what your PhD is about, written so that a physicist in a completely different field (e.g. astronomy) can understand what you’re doing and appreciate why it’s interesting and important.

Materials/nanoscientists and medical physicists/bioengineers wanted too!

Finally, we’re also looking to recruit new students into our existing contributor networks in materials/nanoscience and medical physics/bioengineering. So if you’re a PhD student in one of these fields, and you’d like to be involved, we’d love to hear from you, too.

Biology-guided radiotherapy system spares critical organs

A new, biology-guided radiotherapy (BgRT) system could improve radiation therapy by delivering high dose to the target tumour while avoiding critical organs, according to a study by a radiation oncology research team at the City of Hope National Medical Center. The team simulated intensity-modulated radiotherapy (IMRT) treatments of patients with nasopharyngeal carcinoma (NPC), comparing these plans with those for other IMRT delivery techniques at the hospital.

The researchers created treatment plans for the BgRT system and compared these with helical tomotherapy (HT) and volumetric modulated arc therapy (VMAT) plans for 10 NPC patients who had received prior treatment at the hospital. They hypothesized that the unique delivery pattern of the BgRT system could enable superior beam modulation, which could improve the therapeutic ratio by lowering dose to critical organs. They note that even without utilizing the PET-guidance capabilities of the new system, this type of treatment for NPC patients may offer benefits comparable to those of proton therapy, which is not available to all due to the limited number of proton therapy centres.

The prototype treatment system, developed by RefleXion Medical, combines a compact 6 MV linac and a 16-slice CT with a ring gantry and integrates a PET subsystem to provide real-time tumour tracking. With a continuous rotation speed of 60 rpm, the gantry has much faster rotation than a HT machine. Its 64-leaf multileaf collimator (MLC) transitions 100 times per second to enable synchronized alignment with the linac pulse frequency. The system’s couch remains in a fixed position during radiation delivery, with multiple gantry rotations performed at each couch position, and advances in 2.1 mm steps when the beam is off.

RefleXion biology-guided radiotherapy system
Exploded view of the RefleXion biology-guided radiotherapy system. (Courtesy: RefleXion Medical)

After reviewing the patients’ original HT treatment plans, the team created VMAT and IMRT plans. Principal investigator Chunhui Han generated all the prototype IMRT plans for the BgRT system, which were optimized multiple times to achieve optimal dose homogeneity to all planning target volumes (PTVs) while adequately sparing critical organs. The researchers then compared all treatment plans using dosimetric parameters to PTVs and organs-at-risk (OAR). They report their findings in Medical Dosimetry.

Han and colleagues report that plans for the three modalities had comparable dose coverage, mean dose and dose heterogeneity to the primary PTV. Six of the seven mean dose parameters examined for OARs were lower in the prototype IMRT plans than in the HT or VMAT plans. “The average left and right parotid mean doses in the prototype plans were 10.5 Gy (35.5%) and 10.4 Gy (34.9%) lower than those in the HT plans, respectively, and were 5.1 Gy (21.1%) and 5.2 Gy (21.1%) lower than those in the VMAT plans, respectively,” they write.

However, IMRT plans for the prototype BgRT system had higher dose heterogeneity to non-primary targets. Han tells Physics World that the clinical impact of this is subject to debate and may vary case by case. “Compared to dose increase in non-primary targets, reduction in dose to critical organs is typically more important and desirable,” he explains. “The increased dose to non-primary targets will be more of an issue if they overlap with some critical organs and the dose increase is affecting sparing of critical organs.”

Treatment times were longer on the BgRT system, taking an average of 11.4±1.2 min, compared with 8.3 min for HT and 3.4 min for VMAT. Han believes that this will have limited impact in routine clinical radiotherapy. “A patient typically receives a 3D imaging scan and image-guided positioning corrections prior to the start of treatment,” he points out. “Since this BgRT system has a dedicated kilovoltage CT system with a fast gantry rotation speed, the pre-treatment scan time is a factor of two to three shorter than the time for a MVCT scan of the same length on a helical tomotherapy unit.”

The researchers note that radiotherapy treatment planning for NPC presents unique challenges in plan optimization, because many critical organs are in proximity to target volumes. As such, NPC can serve as a benchmark to evaluate intensity-modulation capabilities of a delivery system. “Improvements in critical organ doses with the prototype plans could have significant clinical impact on patient quality-of-life, such as minimizing the risk of stimulated salivary production,” they write.

Han and colleagues have been investigating the prototype BgRT system for several years. They are currently conducting an ongoing pre-clinical study to explore its use in treating metastases at various sites. They are also performing a retrospective review of historical patient imaging data, including PET data, to evaluate the feasibility and clinical benefits of using the PET feature of the BgRT system.

Quantum cryptography network spans 4600 km in China

A network for quantum key distribution (QKD) spanning thousands of kilometres has been built in China. It links four quantum metropolitan area networks (QMANs) in cities in eastern China with a remote location in the far west of the country. The system comprises a 2000 km fibre optic link between the cities of Shanghai, Hefei, Jinan and Beijing and a satellite link spanning 2600 km between two observatories – one east of Beijing and the other just a few hundred kilometres from China’s border with Kazakhstan.

The network was built by Jian-Wei Pan at the University of Science and Technology of China in Hefei along with colleagues in academia and industry.

QKD uses the principles of quantum mechanics to allow two parties to share a secret cryptography key. A crucial feature of QKD is that the two parties can tell if an eavesdropper has intercepted the key while it is being shared. Once the secrecy of the key is established it can be used to exchange encrypted messages using a conventional telecoms network.

Quantum states of photons

In standard QKD implementations, information is encoded in the quantum states of photons – which are exchanged between the two parties. Photons are used because they can travel several hundred kilometres in optical fibres before their quantum information is lost. Photons can also carry quantum information between ground stations and satellites, allowing QKD to be performed between locations thousands of kilometres apart.

The Chinese network serves about 150 users and comprises more than 700 fibre links and two high-speed satellite-to-ground free-space links – all of which support QKD transmission. The fibre links are supported by 32 “trusted relay nodes” that are capable of forwarding quantum information. The individual QMANS contain trusted relay nodes as well as user nodes and optical switches. The Jinan QMAN is the largest, containing 50 user nodes supporting 95 users.

The satellite portion of the network makes use of the Micius quantum communications satellite, which was launched by China in 2016. Just one year later, Micius was used to make a QKD connection between Beijing and Vienna, which are separated by 7400 km.

To ensure that large numbers of users can access the network, its architecture involves five different layers. These are a quantum physical layer; quantum logical layer; classical physical layer; classical logic layer; and an application layer.

According to the University of Science and Technology of China, Pan and colleagues will further expand the network by working with partners in Austria, Italy, Russia and Canada. The team is also developing low-cost satellites and ground stations for QKD.

The network is described on Nature.

Processing natural language using quantum computers, listening to the oceans’ myriad sounds

Using computers to process natural human language is notoriously difficult, so perhaps its not surprising that researchers are turning to quantum computers. In this episode of the Physics World Weekly podcast, Bob Coecke of Cambridge Quantum Computing explains why natural language processing is “quantum native” – which makes it a perfect candidate for an early practical application of quantum computing.

Also in this episode, Ana Širovic – a marine biologist at Texas A&M University at Galveston – takes us on a sonic journey through the oceans, discussing the many sounds made by marine creatures. She also talks about the threats posed to nature by sounds related to human activity.

Coecke has recently published two preprints on quantum natural language processing on arXiv. They are “Foundations for near-term quantum natural language processing” and “Grammar-aware question-answering on quantum computers”.

Dutch physicist and Nobel laureate Martinus Veltman dies aged 89

The Dutch physicist Martinus Veltman, who won half of the 1999 Nobel Prize for Physics for his theoretical contributions to the Standard Model of particle physics, died on 4 January aged 89. In the 1960s and 1970s, Veltman’s work was instrumental in understanding the weak interaction in particle physics. Veltman shared half the prize with fellow Dutch physicist Gerardus ‘t Hooft, who had been Veltman’s PhD student.

Veltman was born on 27 June 1931 in Waalwijk, the Netherlands. He studied physics and mathematics at the Univeristy of Utrecht, in which he called the teaching “uninspiring” and began to do odd jobs including typing lecture notes as well as selling tools, which he later admitted he was a “complete failure” at. In 1955 Veltman joined the Van Der Waals laboratory at the University of Amsterdam becoming an assistant to Antonius Michels, which involved maintaining the library and preparing talks for Michels.

After two years of military service, Veltman began a PhD in 1959 at Utrecht under the guidance of Léon van Hove. During his doctorate, Veltman spent time at the CERN particle-physics lab near Geneva, which van Hove later served as director general in the late 1970s. Once Veltman’s completed his PhD in 1963, he spent a year at the SLAC National Accelerator Laboratory, which was then known as the Stanford Linear Accelerator Center before heading back to CERN. Veltman remained at CERN until 1966 before returning to Utrecht.

Tackling infinities

It was at Utrecht that Veltman carried out his Nobel-prize-winning work. In the 1960s, Sheldon Glashow, Adbus Salam and Steven Weinberg unified the weak and electromagnetic interaction and predicted the existence of the W and Z bosons, which carry the electroweak force. This theory, which later became known as the Salam-Weinberg theory, was, however, not taken seriously by many in the community because it seemed impossible to subject it to the usual “renomalisation” procedure. This meant it generated infinite and therefore meaningless expressions so it seemed impossible to perform accurate calculations with it.

In the early 1970s, Veltman and ‘t Hooft showed how to carry out this renormalisation and used their theory to make precise calculations of particle properties. These predictions were confirmed when the W and Z particles were detected for the first time in 1983 at the Large Electron-Positron Collider at CERN. This first led Glashow, Salam and Weinberg being awarded the 1979 Nobel Prize of Physics. When later precision calculations agreed with the experimental values of the W and Z boson, ‘t Hooft and Veltman bagged the 1999 Nobel prize “for elucidating the quantum structure of electroweak interactions in physics.”

In 1981 Veltman moved to the University of Michigan before retiring in 1996, when he moved back to the Netherlands.In 2003 Veltman published a popular-science book Facts and Mysteries in Particle Physics and regularly gave lectures on physics. He was also a regular guest at the annual Nobel Lindau Nobel Laureate meeting in Germany.

Relativistic quasiparticles tunnel through barrier with 100% transmission, verifying century-old prediction

A curious effect called “Klein tunnelling” has been observed for the first time in an experiment involving sound waves in a phononic crystal. As well as confirming the century-old prediction that relativistic particles (those travelling at speeds approaching the speed of light) can pass through an energy barrier with 100% transmission, the research done in China and the US could lead to better sonar and ultrasound imaging.

Quantum tunnelling refers to the ability of a particle to pass through a potential-energy barrier, despite having insufficient energy to cross if the system is described by classical physics. Tunnelling is a result of wave–particle duality in quantum mechanics, whereby the wave function of a particle extends into and beyond a barrier.

Normally, the probability that tunnelling will occur is less than 100% and decreases exponentially as the height and width of the barrier increase. However, in 1929 the Swedish physicist Oskar Klein calculated that an electron travelling at near the speed of light will tunnel through a barrier with 100% certainty – regardless of the height and width of the barrier.

Relativistic quasiparticles

Testing this remarkable prediction has proven difficult because of the challenges of accelerating electrons to the required velocity and creating an appropriate barrier for tunnelling. More recently, physicists have discovered that the collective behaviour of electrons in graphene creates massless quasiparticles moving at near to the speed of light. While some indirect features of Klein tunnelling have been seen in graphene, conclusive evidence for 100% transmission has remained elusive.

In this latest work, Xiang Zhang at the University of Hong Kong and colleagues have built an experimental system that uses sound waves to simulate the behaviour of relativistic quasiparticles in graphene. To do this the team created a barrier using two different 2D triangular lattices made from acrylic cylinders (see figure). Just as the 2D atomic lattice of graphene affects the behaviour of electrons in graphene, these phononic crystals affect the behaviour of sound waves – creating quasiparticle acoustic excitations that behave like relativistic electrons.

While the two phononic crystals had the same lattice constant (the separation between cylinders is 28 mm in both), they are each made of cylinders with different radii (12 mm and 7 mm). The tunnelling barrier was created by sandwiching a region of 7 mm lattice between two regions of 12::mm lattice. While the acoustic quasiparticles can move easily in the 12 mm regions, they encounter a potential barrier within the 7 mm region.

Near-perfect transmission

Quasiparticles were created by injecting sound waves into one side of the phononic crystal. Instead of only some of the quasiparticles tunnelling through the barrier, the team measured near 100% transmission. They then tested several systems with different barrier thicknesses and different potential barrier heights – the latter can be adjusted by changing the radii of the cylinders in the barrier lattice. As predicted by Klein, the near-perfect transmission was unaffected by changing these two parameters.

Klein tunnelling was observed for sound over a broad frequency range and the response of the system can be fine tuned by adjusting the size and spacing of the cylinders. As a result, Zhang and colleagues believe that their discovery could be used to boost the transmission of sound waves across interfaces. This could improve sonar systems used for exploring underwater regions and could also lead to the development of new medical ultrasound systems that are better able to see through obstacles within the body.

The research is described in Science.

Visualizing the treatment beam improves radiation therapy delivery

Cancer patients undergoing radiotherapy at Dartmouth-Hitchcock Medical Center are receiving added treatment scrutiny, via real-time monitoring of the radiation dose delivered during their treatments. This treatment verification is enabled by a Cherenkov imaging system, which provides real-time visualization of each radiation beam.

Cherenkov imaging is a novel technique that captures light emissions during radiotherapy. The Cherenkov effect occurs when photon or electron beams interact with tissue, generating light that shows the shape of the treatment beam on the skin’s surface. The intensity of this light is in proportion to the delivered dose. The BeamSite Cherenkov imaging system, developed by DoseOptics, uses time-gating technology to ensure that each linac pulse contributes to the image. Time-integrating software creates images that are overlaid in real time on the patient, providing surrogate maps of surface dose.

These dose maps can be used to visually verify treatment field delivery and patient positioning throughout each treatment fraction. This is not possible using standard quality assurance (QA) measures, according to the research team from the Geisel School of Medicine and Dartmouth Engineering, who developed, validated and commercialized the system.

The Dartmouth team, led by Lesley Jarvis, conducted a study examining the practical utility of Cherenkov imaging in clinical radiotherapy practice. The researchers evaluated the system in 64 patients who received radiotherapy at Dartmouth-Hitchcock’s Norris Cotton Cancer Center. Patients were treated for breast cancer (29), sarcoma/lymphomas (23) and other cancers, using techniques including 3D conformal radiotherapy, total skin electron therapy, arc therapy and total body irradiation. Jarvis and colleagues published their findings in the International Journal of Radiation Oncology, Biology, Physics.

Radiation therapists and radiation oncologists reviewed the Cherenkov images to determine information of clinical benefit and identify treatment incidents. They first noted that the images were useful to monitor the tumour target and adjacent anatomy to ensure appropriate treatment delivery.

To evaluate day-to-day patient positioning accuracy, the researchers evaluated 129 fractions of imaged treatments of 15 breast cancer patients, using the system’s automated image analysis software. The Cherenkov emissions qualitatively matched the expected surface dose distribution predicted by the treatment planning system. Analysing the delivery accuracy by calculating the mean distance to conformity (MDC) revealed that the inter-fraction MDC was within 7 mm compared with the first day of treatment.

To perform absolute dosimetry with Cherenkov imaging, the researchers applied scintillator discs to regions-of-interest (ROIs) on three patients. The system identified the scintillators in the image and converted the intensity to dose, based on known emission response of the scintillator.

Jarvis and colleagues identified six cases among the 64 patients in which Cherenkov images provided particularly beneficial information. This included identification of unintended dose to the contralateral breast, arm or chin in breast cancer treatments. In extremity sarcoma treatments, Cherenkov imaging confirmed that the contralateral leg did not inadvertently receive dose.

Cherenkov imaging
Cherenkov imaging of a leg sarcoma patient (a) shows that the treatment beam remained clear of the contralateral leg. In a breast treatment (b), Cherenkov imaging reveals delivered dose encroaching on the chin. (Courtesy: Int. J. Radiat. Oncol. Biol. Phys. 10.1016/j.ijrobp.2020)

Cherenkov imaging also identified positioning issues, even though these patients had been set up using an optical positioning system that reported positioning within the set tolerance levels. The researchers explain that “typical ROIs for optical surface imaging systems are often focused on a small area and do not detect alignment issues outside of the ROI. This may have the consequence of giving the clinical team a false sense of security.”

In addition to real-time monitoring, Cherenkov images can also be used for post-treatment analysis of accuracy and/or dosimetry. This analysis can be automated for large-scale review of treatment repeatability. Such data can help identify a patient who may require different immobilization due to set-up difficulties.

“In this limited cohort, we found opportunities to improve treatment delivery for individual patients,” write the authors. “Of specific clinical importance, we show that Cherenkov imaging can detect stray radiation dose to tissues. Currently, there is no practical technique available to monitor contralateral breast dose or dose to other adjacent anatomy on a daily basis.”

Co-author Brian Pogue, co-founder and president of DoseOptics, tells Physics World that the research team is now planning a large, retrospective analysis of clinical data of patients receiving radiotherapy at Dartmouth-Hitchcock, to determine the incident rate detection and the value of Cherenkov imaging in daily monitoring of all treatments. He notes that Cherenkov imaging of all radiation oncology patients within the cancer centre began in the autumn of 2020. The BeamSite camera system received 510(k) clearance from the US Food and Drug Administration in December 2020.

“We will compare the value of incident detection at each of our two cancer centres. Only one uses surface guidance tools, so there may be interesting differences in daily patient alignment that Cherenkov imaging may better identify,” explains Pogue. “We are also investigating the ability to match lines between adjacent treatment fields, as well as quantifying absolute dose by using a calibration of the Cherenkov attenuation from a patient’s CT scan. We hope to implement this potentially for quantitative dose imaging.”

Copyright © 2026 by IOP Publishing Ltd and individual contributors