Jobs – Page 92 of 530

Partnership yields big wins for the EIC

**The EIC in outline** Polarised protons, light ions and (unpolarised) heavy ions are injected via the existing source and pre-injection stages (green) and are accelerated and stored in an upgraded RHIC superconducting accelerator ring (yellow). A new suite of components will accelerate polarised electrons in a rapid-cycling synchrotron (red ring) and inject them into the electron storage ring (light blue). The low-energy beam of cooling electrons is shown in purple. The stored beams of ions and electrons cross at two possible collision points, where detectors may be located. Credit: BNL

The international nuclear-physics community will be front-and-centre as a unique research facility called the Electron–Ion Collider (EIC) moves from concept to reality through the 2020s – the latest progression in the line of large-scale accelerator programmes designed to probe the fundamental forces and particles that underpin the structure of matter.

Decades of research in particle and nuclear physics have shown that protons and neutrons, once thought to be elementary, have a rich, dynamically complex internal structure of quarks, anti-quarks and gluons, the understanding of which is fundamental to the nature of matter as we experience it. By colliding high-energy beams of electrons with high-energy beams of protons and heavy ions, the EIC is designed to explore this hidden subatomic landscape with the resolving power to image its behaviour directly. Put another way: the EIC will provide the world’s most powerful microscope for studying the “glue” that binds the building blocks of matter.

Luminous performance

When the EIC comes online in the early 2030s, the facility will perform precision “nuclear femtography” by zeroing in on the substructure of quarks and gluons in a manner comparable to the seminal studies of the proton using electron–proton collisions at DESY’s HERA accelerator in Germany between 1992 and 2007 (see “Nuclear femtography to delve deep into nuclear matter” panel). However, the EIC will produce a luminosity (collision rate) 100 times greater than the highest achieved by HERA and, for the first time in such a collider, will provide spin-polarised beams of both protons and electrons, as well as high-energy collisions of electrons with heavy ions. All of which will require unprecedented performance in terms of the power, intensity and spatial precision of the colliding beams, with the EIC expected to provide not only transformational advances in nuclear science, but also transferable technology innovations to shape the next generation of particle accelerators and detectors.

The US Department of Energy (DOE) formally initiated the EIC project in December 2019 with the approval of a “mission need”. That was followed in June of this year with the next “critical decision” to proceed with funding for engineering and design prior to construction (with the estimated cost of the build about $2 billion). The new facility will be sited at Brookhaven National Laboratory (BNL) in Long Island, New York, utilising components and infrastructure from BNL’s Relativistic Heavy Ion Collider (RHIC), including the polarised proton and ion-beam capability and the 3.8 km underground tunnel. Construction will be carried out as a partnership between BNL and Thomas Jefferson National Accelerator Facility (JLab) in Newport News, Virginia, home of the Continuous Electron Beam Accelerator Facility (CEBAF), which has pioneered many of the enabling technologies needed for the EIC’s new electron rings.

Beyond the BNL–JLab partnership, the EIC is very much a global research endeavour. While the facility is not scheduled to become operational until early in the next decade, an international community of scientists is already hard at work within the EIC User Group. Formed in 2016, the group now has around 1300 members – representing 265 universities and laboratories from 35 countries – engaged collectively on detector R&D, design and simulation as well as initial planning for the EIC’s experimental programme.

A cutting-edge accelerator facility

Being the latest addition to the line of particle colliders, the EIC represents a fundamental link in the chain of continuous R&D, knowledge transfer and innovation underpinning all manner of accelerator-related technologies and applications – from advanced particle therapy systems for the treatment of cancer to ion implantation in semiconductor manufacturing.

The images “The EIC in outline” and “Going underground” show the planned layout of the EIC, where the primary beams circulate inside the existing RHIC tunnel to enable the collisions of high-energy (5–18 GeV) electrons (and possibly positrons) with high-energy ion beams of up to 275 GeV/nucleon. One thing is certain: the operating parameters of the EIC, with luminosities of up to 10³⁴ cm^–2 s^–1 and up to 85% beam polarisation, will push the design of the facility beyond the limits set by previous accelerator projects in a number of core technology areas.

For starters, the EIC will require significant advances in the field of superconducting radiofrequency (SRF) systems operating under high current conditions, including control of higher-order modes, beam RF stability and crab cavities. A major challenge is the achievement of strong cooling of intense proton and light-ion beams to manage emittance growth owing to intrabeam scattering. Such a capability will require unprecedented control of low-energy electron-beam quality with the help of ultrasensitive and precise photon detection technologies – innovations that will likely yield transferable benefits for other areas of research reliant on electron-beam technology (e.g. free-electron lasers).

The EIC design for strong cooling of the ion beams specifies a superconducting energy-recovery linac with a virtual beam power of 15 MW, an order-of-magnitude increase versus existing machines. With this environmentally friendly new technology, the rapidly cycling beam of low-energy electrons (150 MeV) is accelerated within the linac and passes through a cooling channel where it co-propagates with the ions. The cooling electron beam is then returned to the linac, timed to see the decelerating phase of the RF field, and the beam power is thus recovered for the next accelerating cycle – i.e. beam power is literally recycled after each cooling pass.

The EIC will also require complex operating schemes. A case in point: fresh, highly polarised electron bunches will need to be frequently injected into the electron storage ring without disturbing the collision operation of previously injected bunches. Further complexity comes in maximising the luminosity and polarisation over a large range of centre-of-mass energies and for the entire spectrum of ion beams. With a control system that can monitor hundreds of beam parameters in real-time, and with hundreds of points where the guiding magnetic fields can be tuned on the fly, there is a vast array of “knobs-to-be-turned” to optimise overall performance. Inevitably, this is a facility that will benefit from the use of artificial intelligence and machine-learning technologies to maximise its scientific output.

Prototype bunched-beam polarised electron source — **Test case** The prototype bunched-beam polarised electron source, shown here in a recent test beam, achieved a peak current of 4 A with a bunch charge of 8 nC, meeting the EIC performance specifications. Credit: BNL

At the same time, the EIC and CERN’s High-Luminosity LHC user communities are working in tandem to realise more capable technologies for particle detection as well as innovative electronics for large-scale data read-out and processing. Exploiting advances in chip technology, with feature sizes as small as 65 nm, multipixel silicon sensors are in the works for charged-particle tracking, offering single-point spatial resolution better than 5 µm, very low mass and on-chip, individual-pixel readout. These R&D efforts open the way to compact arrays of thin solid-state detectors with broad angular coverage to replace large-volume gaseous detectors.

Coupled with leading-edge computing capabilities, such detectors will allow experiments to stream data continuously, rather than selecting small samples of collisions for readout. Taken together, these innovations will yield no shortage of downstream commercial opportunities, feeding into next-generation medical imaging systems, for example, as well as enhancing industrial R&D capacity at synchrotron light-source facilities.

The BNL–JLab partnership

As the lead project partners, BNL and JLab have a deep and long-standing interest in the EIC programme and its wider scientific mission. In 2019, BNL and JLab each submitted their own preconceptual designs to DOE for a future high-energy and high-luminosity polarised EIC based around existing accelerator infrastructure and facilities. In January 2020, DOE subsequently selected BNL as the preferred site for the EIC, after which the two labs immediately committed to a full partnership between their respective teams (and other collaborators) in the construction and operation of the facility.

Nuclear femtography to delve deep into nuclear matter

Internal quark and gluon substructure of the proton — **Looking closer** The internal quark and gluon substructure of the proton grows more complex when probed at increasing centre-of-mass energies. Credit: BNL

Nuclear matter is inherently complex because the interactions and structures therein are inextricably mixed up: its constituent quarks are bound by gluons that also bind themselves. Consequently, the observed properties of nucleons and nuclei, such as their mass and spin, emerge from a dynamical system governed by quantum chromodynamics (QCD). The quark masses, generated via the Higgs mechanism, only account for a tiny fraction of the mass of a proton, leaving fundamental questions about the role of gluons in the structure of nucleons and nuclei still unanswered.

The underlying nonlinear dynamics of the gluon’s self-interaction is key to understanding QCD and fundamental features of the strong interactions such as dynamical chiral symmetry-breaking and confinement. Yet despite the central role of gluons, and the many successes in our understanding of QCD, the properties and dynamics of gluons remain largely unexplored.

If that’s the back-story, the future is there to be written by the EIC, a unique machine that will enable physicists to shed light on the many open questions in modern nuclear physics.

Back to basics

At the fundamental level, the way in which a nucleon or nucleus reveals itself in an experiment depends on the kinematic regime being probed. A dynamic structure of quarks and gluons is revealed when probing nucleons and nuclei at higher energies, or with higher resolutions. Here, the nucleon transforms from a few-body system, with its structure dominated by three valance quarks, to a regime where it is increasingly dominated by gluons generated through gluon radiation, as discovered at the former HERA electron–proton collider at DESY. Eventually, the gluon density becomes so large that the gluon radiation is balanced by gluon recombination, leading to nonlinear features of the strong interaction.

The LHC and RHIC have shown that neutrons and protons bound inside nuclei already exhibit the collective behaviour that reveals QCD substructure under extreme conditions, as initially seen with high-energy heavy-ion collisions. This has triggered widespread interest in the study of the strong force in the context of condensed-matter physics, and the understanding that the formation and evolution of the extreme phase of QCD matter is dominated by the properties of gluons at high density.

The subnuclear genetic code

The EIC will enable researchers to go far beyond the present one-dimensional picture of nuclei and nucleons, where the composite nucleon appears as a bunch of fast-moving (anti-)quarks and gluons whose transverse momenta or spatial extent are not resolved. Specifically, by correlating the information of the quark and gluon longitudinal momentum component with their transverse momentum and spatial distribution inside the nucleon, the EIC will enable nuclear femtography.

Such femtographic images will provide, for the first time, insight into the QCD dynamics inside hadrons, such as the interplay between sea quarks and gluons. The ultimate goal is to experimentally reconstruct and constrain the so-called Wigner functions – the quantities that encode the complete tomographic information and constitute a QCD “genetic map” of nucleons and nuclei.

• Adapted from “Electron–ion collider on the horizon” by Elke-Caroline Aschenauer, BNL, and Rolf Ent, JLab.

The construction project is led by a joint BNL–JLab management team that integrates the scientific, engineering and management capabilities of JLab into the BNL design effort. JLab, for its part, leads on the design and construction of SRF and cryogenics systems, the energy-recovery linac and several of the electron injector and storage-ring subsystems within the EIC accelerator complex.

More broadly, BNL and JLab are gearing up to work with US and international partners to meet the technical challenges of the EIC in a cost-effective, environmentally responsible manner. The goal: to deliver a leading-edge research facility that will build upon the current CEBAF and RHIC user base to ensure engagement – at scale – from the US and international nuclear-physics communities.

As such, the labs are jointly hosting the EIC experiments in the spirit of a DOE user facility for fundamental research, while the BNL–JLab management team coordinates the engagement of other US and international laboratories into a multi-institutional partnership for EIC construction. Work is also under way with prospective partners to define appropriate governance and operating structures to enhance the engagement of the user community with the EIC experimental programme.

With international collaboration hard-wired into the EIC’s working model, the EIC User Group has been in the vanguard of a global effort to develop the science goals for the facility – as well as the experimental programme to realise those goals. Most importantly, the group has carried out intensive studies over the past two years to document the measurements required to deliver EIC’s physics objectives and the resulting detector requirements. This work also included an exposition of evolving detector concepts and a detailed compendium of candidate technologies for the EIC experimental programme.

Cornerstone collaborations

The resulting Yellow Report, released in March 2021, provides the basis for the ongoing discussion of the most effective implementation of detectors, including the potential for complementary detectors in the two possible collision points as a means of maximising the scientific output of the EIC facility (see “Detectors deconstructed”). Operationally, the report also provides the cornerstone on which EIC detector proposals are currently being developed by three international “proto-collaborations”, with significant components of the detector instrumentation being sourced from non-US partners.

The EIC represents a fundamental link in the chain of continuous R&D and knowledge transfer

Along every coordinate, it’s clear that the EIC project profits enormously from its synergies with accelerator and detector R&D efforts worldwide. To reinforce those benefits, a three-day international workshop was held in October 2020, focusing on EIC partnership opportunities across R&D and construction of accelerator components. This first Accelerator Partnership Workshop, hosted by the Cockcroft Institute in the UK, attracted more than 250 online participants from 26 countries for a broad overview of EIC and related accelerator-technology projects. A follow-up workshop, scheduled for October 2021 and hosted by the TRIUMF Laboratory in Canada, will focus primarily on areas where advanced “scope of work” discussions are already under way between the EIC project and potential partners.

Nurturing talent

While discussion and collaboration between the BNL and JLab communities were prioritised from the start of the EIC planning process, a related goal is to get early-career scientists engaged in the EIC physics programme. To this end, two centres were created independently: the Center for Frontiers in Nuclear Science (CFNS) at Stony Brook University, New York, and the Electron-Ion Collider Center (EIC²) at JLab.

The CFNS, established jointly by BNL and Stony Brook University in 2017, was funded by a generous donation from the Simons Foundation (a not-for-profit organisation that supports basic science) and a grant from the State of New York. As a focal point for EIC scientific discourse, the CFNS mentors early-career researchers seeking long-term opportunities in nuclear science while simultaneously supporting the formation of the EIC’s experimental collaborations.

Conceptual general-purpose detector — **Detectors deconstructed** The layout of a conceptual general-purpose detector from the EIC User Group’s Yellow Report. The design exploits multiple detector technologies for particle momentum and energy measurement, as well as particle identification, at the track level over the full solid-angle range. The report includes an exposition of evolving detector concepts and a detailed compendium of candidate technologies for the EIC experimental programme. Credit: JLab

Core CFNS activities include EIC science workshops, short ad-hoc meetings (proposed and organised by members of the EIC User Group), alongside a robust postdoctoral fellow programme to guide young scientists in EIC-related theory and experimental disciplines. An annual summer school series on high-energy QCD also kicked off in 2019, with most of the presentations and resources from the wide-ranging CFNS events programme available online to participants around the world.

In a separate development, the CFNS recently initiated a dedicated programme for under-represented minorities (URMs). The Edward Bouchet Initiative provides a broad portfolio of support to URM students at BNL, including grants to pursue masters or doctoral degrees at Stony Brook on EIC-related research.

Meanwhile, the EIC² was established at JLab with funding from the State of Virginia to involve outstanding JLab students and postdocs in EIC physics. Recognising that there are many complementary overlaps between JLab’s current physics programme and the physics of the future EIC, the EIC² provides financial support to three PhD students and three postdocs each year to expand their current research to include the physics that will become possible once the new collider comes online.

Beyond their primary research projects, this year’s cohort of six EIC² fellows worked together to organise and establish the first EIC User Group Early Career workshop. The event, designed specifically to highlight the research of young scientists, was attended by more than 100 delegates and is expected to become an annual part of the EIC User Group meeting.

The future, it seems, is bright, with CFNS and EIC² playing their part in ensuring that a diverse cadre of next-generation scientists and research leaders is in place to maximise the impact of EIC science over the decades to come.

Strongly unbalanced photon pairs

**Fig. 1.** Measurements of the differential diphoton cross-section as a function of the transverse momentum of the photon pairs (black points), compared to theoretical predictions including those at next-to-leading order (orange) and next-to-next-to leading order (purple). Credit: CERN

Most processes resulting from proton–proton collisions at the LHC are affected by the strong force – a difficult-to-model part of the Standard Model involving non-perturbative effects. This can be problematic when measuring rare processes not mediated by strong interactions, such as those involving the Higgs boson, and when searching for new particles or interactions. To ensure such processes are not obscured, precise knowledge of the more dominant strong-interaction effects, including those caused by the initial-state partons, is a prerequisite to LHC physics analyses.

The electromagnetic production of a photon pair is the dominant background to the H → γγ decay channel – a process that is instrumental to the study of the Higgs boson. Despite its electromagnetic nature, diphoton production is affected by surprisingly large strong-interaction effects. Thanks to precise ATLAS measurements of diphoton processes using the full Run-2 dataset, the collaboration is able to probe these effects and scrutinise state-of-the-art theoretical calculations.

Measurements studying strong interactions typically employ final states that include jets produced from the showering and hadronisation of quarks and gluons. However, the latest ATLAS analysis instead uses photons, which can be very precisely measured by the detector. Although photons do not carry a colour charge, they interact with quarks as the latter carry electric charge. As a result, strong-interaction effects on the quarks can alter the characteristics of the measured photons. The conservation of momentum allows us to quantify this effect: the LHC’s proton beams collide head-on, so the net momentum transverse to the beam axis must be zero for the final-state particles. Any signs to the contrary indicate additional activity in the event with equivalent but opposite transverse momentum, usually arising from quarks and gluons radiated from the initial-state partons. Therefore, by measuring the transverse momentum of photon pairs, and related observables, the strong interaction may be indirectly probed.

A surprising role of the strong interaction in electro-magnetic diphoton production is revealed

Comparing the measured values to predictions reveals the surprising role of the strong interaction in electromagnetic diphoton production. In a simple picture without the strong interaction, the momentum of each photon should perfectly balance in the transverse plane. However, this simplistic expectation does not match the measurements (see figure 1). Measuring the differential cross-section as a function of the transverse momentum of the photon pair, ATLAS finds that most of the measured photon pairs (black points) have low but non-zero transverse momenta, with a peak at approximately 10 GeV, followed by a smoothly falling distribution towards higher values.

Extending calculations to encompass next-to-next-to-leading order corrections in the strong-interaction coupling constant (purple line), the impact of the strong interaction becomes manifest. The measured values at high transverse momenta are well described by these predictions, including the bump observed at 70 GeV, which is another manifestation of higher-order strong-interaction effects. Monte Carlo event generators like Sherpa (red line), which combine similar calculations with approximate simulations of arbitrarily many-quark and gluon emissions – especially relevant at low energies – properly describe the entire measured distribution.

The results of this analysis, which also include measurements of other distributions such as angular variables between the two photons, don’t just viscerally probe the strong interaction – they also provide a benchmark for this important background process.

Arthur M Poskanzer 1931–2021

Arthur M (Art) Poskanzer, distinguished senior scientist emeritus at Lawrence Berkeley National Laboratory (LBNL), passed away peacefully on 30 June 2021, two days after his 90th birthday. Art had a distinguished career in nuclear physics and chemistry. He made important discoveries of the properties of unstable nuclei and was a pioneer in the study of nuclear collisions at very high energies.

Born in New York City, Art received his degree in physics and chemistry from Harvard in 1953, an MA from Columbia in 1954, and a PhD in Chemistry from MIT in 1957 under Charles D Coryell. He spent the first part of his career studying the properties of nuclei far from stability produced in high-energy proton collisions. After graduating from MIT, he joined Gerhard Friedlander’s group at Brookhaven National Laboratory (BNL), which was using the Cosmotron to produce beta-delayed proton emitters and neutron-rich light nuclei. In 1966 he moved to the Lawrence Radiation Laboratory (now LBNL) and continued to study nuclei far from stability at the Bevatron in collaboration with Earl Hyde, Joe Cerny and others. He also began his long connection to research in Europe as a Guggenheim fellow at Orsay in 1970–1971, during which he worked with Robert Klapisch’s group on a ground-breaking experiment at the CERN Proton Synchrotron measuring the masses of sodium isotopes.

Soon after Art’s return to Berkeley, beams from the SuperHILAC were injected into the Bevatron, creating the Bevalac, the world’s first high-energy nuclear accelerator. Together with Hans Gutbrod he led the Plastic Ball Project. Analysis of its data in 1984 by Art and Hans Georg Ritter identified directed flow, the first definitive demonstration of the collective behaviour of nuclear matter in nuclear collisions. In 1986 the experiment was moved to CERN and the collaboration with GSI continued with a series of experiments at the Super Proton Synchrotron. During these years, Art made two more extended visits to CERN as a Senior Alexander von Humbold Fellow: first in 1986–1987 working on the WA80 experiment, and then in 1995–1996 on NA49.

From 1990 to 1995 Art was the founding head of LBNL’s relativistic nuclear collisions programme, bringing together local groups to plan an experiment at the Relativistic Heavy Ion Collider (RHIC) under construction at BNL. This resulted in the proposal for STAR, one of the two large multi-purpose RHIC detectors. Art stepped down as programme head in 1995 and returned to research, authoring a seminal paper with Sergey Voloshin on methods for flow analysis and leading the measurement of elliptic flow by STAR. After his retirement in 2002, he remained active for a further decade, leading the successful search for higher order flow components at STAR, and enthusiastically mentoring many postdocs and young scientists.

Art was a well-known and well-loved member of the heavy-ion community. For his work on nuclei far from stability, he was awarded the Nuclear Chemistry Prize of the American Chemical Society in 1980. For the discovery of collective flow, he was awarded the Tom Bonner Prize of the American Physical Society in 2008. This rare “double” is a lasting tribute to his half-century career at the frontiers of nuclear science.

Emergence

A murmuration of starlings — **Emergent simplicity** The evolution of a murmuration of starlings cannot be described by following the motion of any individual bird. Credit: iStock/georgeclerk

Particle physics is at its heart a reductionistic endeavour that tries to reduce reality to its most basic building blocks. This view of nature is most evident in the search for a theory of everything – an idea that is nowadays more common in popularisations of physics than among physicists themselves. If discovered, all physical phenomena would follow from the application of its fundamental laws.

A complementary perspective to reductionism is that of emergence. Emergence says that new and different kinds of phenomena arise in large and complex systems, and that these phenomena may be impossible, or at least very hard, to derive from the laws that govern their basic constituents. It deals with properties of a macroscopic system that have no meaning at the level of its microscopic building blocks. Good examples are the wetness of water and the superconductivity of an alloy. These concepts don’t exist at the level of individual atoms or molecules, and are very difficult to derive from the microscopic laws.

As physicists continue to search for cracks in the Standard Model (SM) and Einstein’s general theory of relativity, could these natural laws in fact be emergent from a deeper reality? And emergence is not limited to the world of the very small, but by its very nature skips across orders of magnitude in scale. It is even evident, often mesmerisingly so, at scales much larger than atoms or elementary particles, for example in the murmurations of a flock of birds – a phenomenon that is impossible to describe by following the motion of an individual bird. Another striking example may be intelligence. The mechanism by which artificial intelligence is beginning to emerge from the complexity of underlying computing codes shows similarities with emergent phenomena in physics. One can argue that intelligence, whether it occurs naturally, as in humans, or artificially, should also be viewed as an emergent phenomenon.

Data compression

Renormalisable quantum field theory, the foundation of the SM, works extraordinarily well. The same is true of general relativity. How can our best theories of nature be so successful, while at the same time being merely emergent? Perhaps these theories are so successful precisely because they are emergent.

As a warm up, let’s consider the laws of thermodynamics, which emerge from the microscopic motion of many molecules. These laws are not fundamental but are derived by statistical averaging – a huge data compression in which the individual motions of the microscopic particles are compressed into just a few macroscopic quantities such as temperature. As a result, the laws of thermodynamics are universal and independent of the details of the microscopic theory. This is true of all the most successful emergent theories; they describe universal macroscopic phenomena whose underlying microscopic descriptions may be very different. For instance, two physical systems that undergo a second-order phase transition, while being very different microscopically, often obey exactly the same scaling laws, and are at the critical point described by the same emergent theory. In other words, an emergent theory can often be derived from a large universality class of many underlying microscopic theories.

Successful emergent theories describe universal macroscopic phenomena whose underlying microscopic descriptions may be very different

Entropy is a key concept here. Suppose that you try to store the microscopic data associated with the motion of some particles on a computer. If we need N bits to store all that information, we have 2^N possible microscopic states. The entropy equals the logarithm of this number, and essentially counts the number of bits of information. Entropy is therefore a measure of the total amount of data that has been compressed. In deriving the laws of thermodynamics, you throw away a large amount of microscopic data, but you at least keep count of how much information has been removed in the data-compression procedure.

Emergent quantum field theory

One of the great theoretical-physics paradigm shifts of the 20th century occurred when Kenneth Wilson explained the emergence of quantum field theory through the application of the renormalisation group. As with thermodynamics, renormalisation compresses microscopic data into a few relevant parameters – in this case, the fields and interactions of the emergent quantum field theory. Wilson demonstrated that quantum field theories appear naturally as an effective long-distance and low-energy description of systems whose microscopic definition is given in terms of a quantum system living on a discretised spacetime. As a concrete example, consider quantum spins on a lattice. Here, renormalisation amounts to replacing the lattice by a coarser lattice with fewer points, and redefining the spins to be the average of the original spins. One then rescales the coarser lattice so that the distance between lattice points takes the old value, and repeats this step many times. A key insight was that, for quantum statistical systems that are close to a phase transition, you can take a continuum limit in which the expectation values of the spins turn into the local quantum fields on the continuum spacetime.

This procedure is analogous to the compression algorithms used in machine learning. Each renormalisation step creates a new layer, and the algorithm that is applied between two layers amounts to a form of data compression. The goal is similar: you only keep the information that is required to describe the long-distance and low-energy behaviour of the system in the most efficient way.

A neural network — **Gauge symmetry** To ensure that a neural network recognises a pixelated lizard, its algorithm should be invariant under rotations. Credit: M Weiler/arXiv:2106.06020

So quantum field theory can be seen as an effective emergent description of one of a large universality class of many possible underlying microscopic theories. But what about the SM specifically, and its possible supersymmetric extensions? Gauge fields are central ingredients of the SM and its extensions. Could gauge symmetries and their associated forces emerge from a microscopic description in which there are no gauge fields? Similar questions can also be asked about the gravitational force. Could the curvature of spacetime be explained from an emergent perspective?

String theory seems to indicate that this is indeed possible, at least theoretically. While initially formulated in terms of vibrating strings moving in space and time, it became clear in the 1990s that string theory also contains many more extended objects, known as “branes”. By studying the interplay between branes and strings, an even more microscopic theoretical description was found in which the coordinates of space and time themselves start to dissolve: instead of being described by real numbers, our familiar (x, y, z) coordinates are replaced by non-commuting matrices. At low energies, these matrices begin to commute, and give rise to the normal spacetime with which we are familiar. In these theoretical models it was found that both gauge forces and gravitational forces appear at low energies, while not existing at the microscopic level.

While these models show that it is theoretically possible for gauge forces to emerge, there is at present no emergent theory of the SM. Such a theory seems to be well beyond us. Gravity, however, being universal, has been more amenable to emergence.

Emergent gravity

In the early 1970s, a group of physicists became interested in the question: what happens to the entropy of a thermodynamic system that is dropped into a black hole? The surprising conclusion was that black holes have a temperature and an entropy, and behave exactly like thermodynamic systems. In particular, they obey the first law of thermodynamics: when the mass of a black hole increases, its (Bekenstein–Hawking) entropy also increases.

The correspondence between the gravitational laws and the laws of thermodynamics does not only hold near black holes. You can artificially create a gravitational field by accelerating. For an observer who continues to accelerate, even empty space develops a horizon, from behind which light rays will not be able to catch up. These horizons also carry a temperature and entropy, and obey the same thermodynamic laws as black-hole horizons.

It was shown by Stephen Hawking that the thermal radiation emitted from a black hole originates from pair creation near the black-hole horizon. The properties of the pair of particles, such as spin and charge, are undetermined due to quantum uncertainty, but if one particle has spin up (or positive charge), then the other particle must have spin down (or negative charge). This means that the particles are quantum entangled. Quantum entangled pairs can also be found in flat space by considering accelerated observers.

Crucially, even the vacuum can be entangled. By separating spacetime into two parts, you can ask how much entanglement there is between the two sides. The answer to this was found in the last decade, through the work of many theorists, and turns out to be rather surprising. If you consider two regions of space that are separated by a two-dimensional surface, the amount of quantum entanglement between the two sides turns out to be precisely given by the Bekenstein–Hawking entropy formula: it is equal to a quarter of the area of the surface measured in Planck units.

Holographic renormalisation

The area of the event horizon — **Black-hole maths** Jacob Bekenstein and Stephen Hawking related entropy to the area of the event horizon. Credit: J Bekenstein

The AdS/CFT correspondence incorporates a principle called “holography”: the gravitational physics inside a region of space emerges from a microscopic description that, just like a hologram, lives on a space with one less dimension and thus can be viewed as living on the boundary of the spacetime region. The extra dimension of space emerges together with the gravitational force through a process called “holographic renormalisation”. One successively adds new layers of spacetime. Each layer is obtained from the previous layer through “coarse-graining”, in a similar way to both renormalisation in quantum field theory and data-compression algorithms in machine learning.

Unfortunately, our universe is not described by a negatively curved spacetime. It is much closer to a so-called de Sitter spacetime, which has a positive curvature. The main difference between de Sitter space and the negatively curved anti-de Sitter space is that de Sitter space does not have a boundary. Instead, it has a cosmological horizon whose size is determined by the rate of the Hubble expansion. One proposed explanation for this qualitative difference is that, unlike for negatively curved spacetimes, the microscopic quantum state of our universe is not unique, but secretly carries a lot of quantum information. The amount of this quantum information can once again be counted by an entropy: the Bekenstein–Hawking entropy associated with the cosmological horizon.

This raises an interesting prospect: if the microscopic quantum data of our universe may be thought of as many entangled qubits, could our current theories of spacetime, particles and forces emerge via data compression? Space, for example, could emerge by forgetting the precise way in which all the individual qubits are entangled, but only preserving the information about the amount of quantum entanglement present in the microscopic quantum state. This compressed information would then be stored in the form of the areas of certain surfaces inside the emergent curved spacetime.

In this description, gravity would follow for free, expressed in the curvature of this emergent spacetime. What is not immediately clear is why the curved spacetime would obey the Einstein equations. As Einstein showed, the amount of curvature in spacetime is determined by the amount of energy (or mass) that is present. It can be shown that his equations are precisely equivalent to an application of the first law of thermodynamics. The presence of mass or energy changes the amount of entanglement, and hence the area of the surfaces in spacetime. This change in area can be computed and precisely leads to the same spacetime curvature that follows from the Einstein equations.

The idea that gravity emerges from quantum entanglement goes back to the 1990s, and was first proposed by Ted Jacobson. Not long afterwards, Juan Maldacena discovered that general relativity can be derived from an underlying microscopic quantum theory without a gravitational force. His description only works for infinite spacetimes with negative curvature called anti-de Sitter (or AdS–) space, as opposed to the positive curvature we measure. The microscopic description then takes the form of a scale-invariant quantum field theory – a so-called conformal field theory (CFT) – that lives on the boundary of the AdS–space (see “Holographic renormalisation” panel). It is in this context that the connection between vacuum entanglement and the Bekenstein–Hawking entropy, and the derivation of the Einstein equations from entanglement, are best understood. I have also contributed to these developments in a paper in 2010 that emphasised the role of entropy and information for the emergence of the gravitational force. Over the last decade a lot of progress has been made in our understanding of these connections, in particular the deep connection between gravity and quantum entanglement. Quantum information has taken centre stage in the most recent theoretical developments.

Emergent intelligence

But what about viewing the even more complex problem of human intelligence as an emergent phenomenon? Since scientific knowledge is condensed and stored in our current theories of nature, the process of theory formation can itself be viewed as a very efficient form of data compression: it only keeps the information needed to make predictions about reproducible events. Our theories provide us with a way to make predictions with the fewest possible number of free parameters.

The same principles apply in machine learning. The way an artificial-intelligence machine is able to predict whether an image represents a dog or a cat is by compressing the microscopic data stored in individual pixels in the most efficient way. This decision cannot be made at the level of individual pixels. Only after the data has been compressed and reduced to its essence does it becomes clear what the picture represents. In this sense, the dog/cat-ness of a picture is an emergent property. This is even true for the way humans process the data collected by our senses. It seems easy to tell whether we are seeing or hearing a dog or a cat, but underneath, and hidden from our conscious mind, our brains perform a very complicated task that turns all the neural data that come from our eyes and ears into a signal that is compressed into a single outcome: it is a dog or a cat.

Emergence is often summarised with the slogan “the whole is more than the sum of its parts”

Can intelligence, whether artificial or human, be explained from a reductionist point of view? Or is it an emergent concept that only appears when we consider a complex system built out of many basic constituents? There are arguments in favour of both sides. As human beings, our brains are hard-wired to observe, learn, analyse and solve problems. To achieve these goals the brain takes the large amount of complex data received via our senses and reduces it to a very small set of information that is most relevant for our purposes. This capacity for efficient data compression may indeed be a good definition for intelligence, when it is linked to making decisions towards reaching a certain goal. Intelligence defined in this way is exhibited in humans, but can also be achieved artificially.

Artificially intelligent computers beat us at problem solving, pattern recognition and sometimes even in what appears to be “generating new ideas”. A striking example is DeepMind’s AlphaZero, whose chess rating far exceeds that of any human player. Just four hours after learning the rules of chess, AlphaZero was able to beat the strongest conventional “brute force” chess program by coming up with smarter ideas and showing a deeper understanding of the game. Top grandmasters use its ideas in their own games at the highest level.

In its basic material design, an artificial-intelligence machine looks like an ordinary computer. On the other hand, it is practically impossible to explain all aspects of human intelligence by starting at the microscopic level of the neurons in our brain, let alone in terms of the elementary particles that make up those neurons. Furthermore, the intellectual capability of humans is closely connected to the sense of consciousness, which most scientists would agree does not allow for a simple reductionist explanation.

Emergence is often summarised with the slogan “the whole is more than the sum of its parts” – or as condensed-matter theorist Phil Anderson put it, “more is different”. It counters the reductionist point of view, reminding us that the laws that we think to be fundamental today may in fact emerge from a deeper underlying reality. While this deeper layer may remain inaccessible to experiment, it is an essential tool for theorists of the mind and the laws of physics alike.

Building the future of LHCb

**Fast track** Planes of LHCb’s SciFi tracker being assembled at CERN in late 2019. Credit: M Brice

It was once questioned whether it would be possible to successfully operate an asymmetric “forward” detector at a hadron collider. In such a high-occupancy environment, it is much harder to reconstruct decay vertices and tracks than it is at a lepton collider. Following its successes during LHC Run 1 and Run 2, however, LHCb has rewritten the forward-physics rulebook, and is now preparing to take on bigger challenges.

During Long Shutdown 2, which comes to an end early next year, the LHCb detector is being almost entirely rebuilt to allow data to be collected at a rate up to 10 times higher during Run 3 and Run 4. This will improve the precision of numerous world-best results, such as constraints on the angles of the CKM triangle, while further scrutinising intriguing results in B-meson decays, which hint at departures from the Standard Model.

**Asymmetric anatomy** LHCb’s successive detector layers located downstream from the interaction point in the heart of the VELO. Credit: LHCb

At the core of the LHCb upgrade project are new detectors capable of sustaining an instantaneous luminosity up to five times that seen at Run 2, and which enable a pioneering software-only trigger that will enable LHCb to process signal data in an upgraded computing farm at the frenetic rate of 40 MHz. The vertex locator (VELO) will be replaced with a pixel version, the upstream silicon-strip tracker will be replaced with a lighter version (the UT) located closer to the beamline, and the electronics for LHCb’s muon stations and calorimeters are being upgraded for 40 MHz readout.

Recently, three further detector systems key to dealing with the higher occupancies ahead were lowered into the LHCb cavern for installation: the upgraded ring-imaging Cherenkov detectors RICH1 and RICH2 for sharper particle identification, and the brand new “SciFi” (scintillating fibre) tracker.

SciFi tracking

The components of LHCb’s SciFi tracker may not seem futuristic at first glance. Its core elements are constructed from what is essentially paper, plastic, some carbon fibre and glue. However, its materials components conceal advanced technologies which, when coupled together, produce a very light and uniform, high-performance detector that is needed to cope with the higher number of particle tracks expected during Run 3.

Located behind the LHCb magnet (see “Asymmetric anatomy” image), the SciFi represents a challenge, not only due to its complexity, but also because the technology – plastic scintillating fibres and silicon photomultiplier arrays – has never been used for such a large area in such a harsh radiation environment. Many of the underlying technologies have been pushed to the extreme during the past decade to allow the SciFi to successfully operate under LHC conditions in an affordable and effective way.

**SciFi spools** Scintillating-fibre mat production at EPFL, Lausanne, in 2017. Credit: P Hopchev/M Pietrzyk

More than 11,000 km of 0.25 mm-diameter polystyrene fibre was delivered to CERN before undergoing meticulous quality checks. Excessive diameter variations were removed to prevent disruptions of the closely packed fibre matrix produced during the winding procedure, and clear improvements from the early batches to the production phase were made by working closely with the industrial manufacturer. From the raw fibres, nearly 1400 multi-layered fibre mats were wound in four of the LHCb collaboration’s institutes (see “SciFi spools” image), before being cut and bonded in modules, tested, and shipped to CERN where they were assembled with the cold boxes. The SciFi tracker contains 128 stiff and robust 5 × 0.5 m² modules made of eight mats bonded with two fire-resistant honeycomb and carbon-fibre panels, along with some mechanics and a light-injection system. In total, the design produces nearly 320 m² of detector surface over the 12 layers of the tracking stations.

The scintillating fibres emit photons at blue-green wavelengths when a particle interacts with them. Secondary scintillator dyes added to the polystyrene amplify the light and shift it to longer wavelengths so it can be read out by custom-made silicon photomultipliers (SiPMs). SiPMs have become a strong alternative to conventional photomultiplier tubes in recent years, due to their smaller channel sizes, easier operation and insensitivity to magnetic fields. This makes them ideal to read out the higher number of channels necessary to identify separate but nearby tracks in LHCb during Run 3.

The width of the SiPM channels, 0.25 mm, is designed to match that of the fibres. Though they need not align perfectly, this provides a better separation power for tracking than the previously used 5 mm gas straw tubes in the outer regions of the detector, while providing a similar performance to the silicon-strip tracker. The tiny channel size results in over 524,288 SiPM channels to collect light from 130 m of fibre-mat edges. A custom ASIC, called the PACIFIC, outputs two bits per channel based on three signal-amplitude thresholds. A field-programmable gate array (FPGA) assigned to each SiPM then groups these signals into clusters, where the location of each cluster is sent to the computing farm. Despite clustering and noise suppression, this still results in an enormous data rate of 20 Tb/s – nearly half of the total data bandwidth of the upgraded LHCb detector.

One of the key factors in the success of LHCb’s flavour-physics programme is its ability to identify charged particles

LHCb’s SciFi tracker is the first large-scale use of SiPMs for tracking, and takes advantage of improvements in the technology in the 10 years since the SciFi was proposed. The photon-detection efficiency of SiPMs has nearly doubled thanks to improvements in the design and production of the underlying pixel structures, while the probability of crosstalk between the pixels (which creates multiple fake signals by causing a single pixel to randomly fire without incident light following radiation damage) has been reduced from more than 20% to a few percent by the introduction of microscopic trenches between the pixels. The dark-single-pixel firing rate can also be reduced by cooling the SiPM. Together, these two methods greatly reduce the number of fake-signal clusters such that the tracker can effectively function after several years of operation in the LHCb cavern.

RICH2 photon detector plane — **RICH2 to go** One of the two photon detector planes of RICH2, fully assembled before installation. Credit: S Gambetta

The LHCb collaboration assembled commercial SiPMs on flex cables and bonded them in groups of 16 to a 0.5 m-long 3D-printed titanium cooling bar to form precisely assembled photodetection units for the SciFi modules. By circulating a coolant at a temperature of –50 °C through the cold bar, the dark-noise rate was reduced by a factor of 60. Furthermore, in a first for a CERN experiment, it was decided to use a new single-phase liquid coolant called Novec-649 from 3M for its non-toxic properties and low greenhouse warming potential (GWP = 1). Historically, C₆F₁₄ – which has a GWP = 7400 – was the thermo-transfer fluid of choice. Although several challenges had to be faced in learning how to work with the new fluid, wider use of Novec-649 and similar products could contribute significantly to the reduction of CERN’s carbon footprint. Additionally, since the narrow envelope of the tracking stations precludes the use of standard foam insulation of the coolant lines, a significant engineering effort has been required to vacuum insulate the 48 transfer lines from the 24 rows of SiPMs and 256 cold-bars where leaks are possible at every connection.

To date, LHCb collaborators have tirelessly assembled and tested nearly half of the SciFi tracker above ground, where only two defective channels out of the 262,144 tested in the full signal chain were unrecoverable. Four out of 12 “C-frames” containing the fibre modules (see “Tracking tall” image) are now installed and waiting to be connected and commissioned, with a further two installed in mid-July. The remaining six will be completed and installed before the start of operations early next year.

New riches

One of the key factors in the success of LHCb’s flavour-physics programme is its ability to identify charged particles, which reduces the background in selected final states and assists in the flavour tagging of b quarks. Two ring-imaging Cherenkov (RICH) detectors, RICH1 and RICH2, located upstream and downstream of the LHCb magnet 1 and 10 m away from the collision point, provide excellent particle identification over a very wide momentum range. They comprise a large volume of fluorocarbon gas (the radiator), in which photons are emitted by charged particles travelling at speeds higher than the speed of light in the gas; spherical and flat mirrors to focus and reflect this Cherenkov light; and two photon-detector planes where the Cherenkov rings are detected and read out by the front-end electronics.

The original RICH detectors are currently being refurbished to cope with the more challenging data-taking conditions of Run 3, requiring a variety of technological challenges to be overcome. The photon detection system, for example, has been redesigned to adapt to the highly non-uniform occupancy expected in the RICH system, running from an unprecedented peak occupancy of ~35% in the central region of RICH1 down to 5% in the peripheral region of RICH2. Two types of 64-channel multi-anode photomultiplier tubes (MaPMTs) have been selected for the task which, thanks to their exceptional quantum efficiency in the relevant wavelength range, are capable of detecting single photons while providing excellent spatial resolution and very low background noise. These are key requirements to allow pattern-recognition algorithms to reconstruct Cherenkov rings even in the high-occupancy region.

**Tracking tall** Completed SciFi C-frames in the assembly hall at Point 8. Credit: B Leverington

More than 3000 MaPMT units, for a total of 196,608 channels, are needed to fully instrument both upgraded RICH detectors. The already large active area (83%) of the devices has been maximised by arranging the units in a compact and modular “elementary cell” containing a custom-developed, radiation-hard eight-channel ASIC called the Claro chip, which is able to digitise the MaPMT signal at a rate of 40 MHz. The readout is controlled by FPGAs connected to around 170 channels each. The prompt nature of Cherenkov radiation combined with the performance of the new opto-electronics chain will allow the RICH systems to operate within the LHC’s 25 ns time window, dictated by the bunch-crossing period, while applying a time-gate of less than 6 ns to provide background rejection.

To keep the new RICHes as compact as possible, the hosting mechanics has been designed to provide both structural support and active cooling. Recent manufacturing techniques have enabled us to drill two 6 mm-diameter ducts over a length of 1.5 m into the spine of the support, through which a coolant (the more environmentally friendly Novec649, as in the SciFi tracker) is circulated. Each element of the opto-electronics chain has been produced and fully validated within a dedicated quality-assurance programme, allowing the position of the photon detectors and their operating conditions to be fine-tuned across the RICH detectors. In February, the first photon-detector plane of RICH2 (see “RICH2 to go” image) became the first active element of the LHCb upgrade to be installed in the cavern. The two planes of RICH2, located at the sides of the beampipe, were commissioned in early summer and will see first Cherenkov light during an LHC beam test in October.

RICH1 spherical mirrors — **Cherenkov curves** The spherical mirrors of the new RICH1. Credit: C Frei

RICH1 presents an even bigger challenge. To reduce the number of photons in the hottest region, its optics have been redesigned to spread the Cherenkov rings over a larger surface. The spatial envelope of RICH1 is also constrained by its magnetic shield, demanding even more compact mechanics for the photon-detector planes. To accommodate the new design of RICH1, a new gas enclosure for the radiator is needed. A volume of 3.8 m³ of C₄F₁₀ is enclosed in an aluminium structure directly fastened to the VELO tank on one side and sealed with a low-mass window on the other, with particular effort placed on building a leak-less system to limit potential environmental impact. Installing these fragile components in a very limited space has been a delicate process, and the last element to complete the gas-enclosure sealing was installed at the beginning of June.

The optical system is the final element of the RICH1 mechanics. The ~2 m² spherical mirrors placed inside the gas enclosure are made of carbon fibre composite to limit the material budget (see “Cherenkov curves” image), while the two 1.3 m² planes of flat mirrors are made of borosilicate glass for high optical quality. All the mirror segments are individually coated, glued on supports and finally aligned before installation in the detector. The full RICH1 installation is expected to be completed in the autumn, followed by the challenging commissioning phase to tune the operating parameters to be ready for Run 3.

Surpassing expectations

In its first 10 years of operations, the LHCb experiment has already surpassed expectations. It has enabled physicists to make numerous important measurements in the heavy-flavour sector, including the first observation of the rare decay B⁰_s → µ⁺µ^–, precise measurements of quark-mixing parameters, the discovery of CP violation in the charm sector, and the observation of more than 50 new hadrons including tetraquark and pentaquark states. However, many crucial measurements are currently statistically limited, including those underpinning the so-called flavour anomalies (see B_s decays remain anomalous). Together with the tracker, trigger and other upgrades taking place during LS2, the new SciFi and revamped RICH detectors will put LHCb in prime position to explore these and other searches for new physics for the next 10 years and beyond.

Science Gateway under construction

Science Gateway foundation stone — **First stone** (from left) Architect Renzo Piano, CERN Director-General Fabiola Gianotti, Geneva state councillor Antonio Hodgers and chairman of Stellantis and the FCA Foundation (the main donor) John Elkann with the symbolic foundation stone, adorned with the newly unveiled Science Gateway logo. Credit: M Brice/CERN

On 21 June, officials and journalists gathered at CERN to mark “first stone” for Science Gateway, CERN’s new flagship project for science education and outreach. Due to open in 2023, Science Gateway will increase CERN’s capacity to welcome visitors of all ages from near and afar. Hundreds of thousands of people per year will have the opportunity to engage with CERN’s discoveries and technology, guided by the people who make it possible.

The project has environmental sustainability at its core. Designed by renowned architect Renzo Piano, the carbon-neutral building will bridge the Route de Meyrin and be surrounded by a freshly planted 400-tree forest. Its five linked pavilions will feature a 900-seat auditorium, immersive spaces, laboratories for hands-on activities for visitors from age five upwards, and many other interactive learning opportunities.

“I would like to express my deepest gratitude to the many partners in our Member and Associate Member States and beyond who are making the CERN Science Gateway possible, in particular to our generous donors,” said CERN Director-General Fabiola Gianotti during her opening speech. “We want the CERN Science Gateway to inspire all those who come to visit with the beauty and the values of science.”

Surveyors eye up a future collider

Levelling measurements — **Level headed** CERN’s Pierre Valentin performing levelling measurements between Certoux and Saint-Julien in France. Credit: J Ordan/CERN

CERN surveyors have performed the first geodetic measurements for a possible Future Circular Collider (FCC), a prerequisite for high-precision alignment of the accelerator’s components. The millimetre-precision measurements are one of the first activities undertaken by the FCC feasibility study, which was launched last year following the recommendation of the 2020 update of the European strategy for particle physics. During the next three years, the study will explore the technical and financial viability of a 100 km collider at CERN, for which the tunnel is a top priority. Geology, topography and surface infrastructure are the key constraints on the FCC tunnel’s position, around which civil engineers will design the optimal route, should the project be approved.

The FCC would cover an area about 10 times larger than the LHC, in which every geographical reference must be pinpointed with unprecedented precision. To provide a reference coordinate system, in May the CERN surveyors, in conjunction with ETH Zürich, the Federal Office of Topography Swisstopo, and the School of Engineering and Management Vaud, performed geodetic levelling measurements along an 8 km profile across the Swiss–French border south of Geneva.

Such measurements have two main purposes. The first is to determine a high-precision surface model, or “geoid”, to map the height above sea level in the FCC region. The second purpose is to improve the present reference system, whose measurements date back to the 1980s when the tunnel housing the LHC was built.

“The results will help to evaluate if an extrapolation of the current LHC geodetic reference systems and infrastructure is precise enough, or if a new design is needed over the whole FCC area,” says Hélène Mainaud Durand, group leader of CERN’s geodetic metrology group.

The FCC feasibility study, which involves more than 140 universities and research institutions from 34 countries, also comprises technological, environmental, engineering, political and economic considerations. It is due to be completed by the time the next strategy update gets under way in the middle of the decade. Should the outcome be positive, and the project receive the approval of CERN’s member states, civil-engineering works could start as early as the 2030s.

Web code auctioned as crypto asset

The web’s original source code — **Unique** A portion of the web’s original source code. Credit: NFT ERC-721

Time-stamped files stated by Tim Berners-Lee to contain the original source code for the web and digitally signed by him, have sold for US$5.4 million at auction. The files were sold as a non-fungible token (NFT), a form of a crypto asset that uses blockchain technology to confer uniqueness.

The web was originally conceived at CERN to meet the demand for automated information-sharing between physicists spread across universities and institutes worldwide. Berners-Lee wrote his first project proposal in March 1989, and the first website, which was dedicated to the World Wide Web project itself and hosted on Berners-Lee’s NeXT computer, went live in the summer of 1991. Less than two years later, on 30 April 1993, and after several iterations in development, CERN placed version three of the software in the public domain. It deliberately did so on a royalty-free, “no-strings-attached” basis, addressing the memo simply “To whom it may concern.”

The seed that led CERN to relinquish ownership of the web was planted 70 years ago, in the CERN Convention, which states that results of its work were to be “published or otherwise made generally available” – a culture of openness that continues to this day.

The auction offer describes the NFT as containing approximately 9555 lines of code, including implementations of the three languages and protocols that remain fundamental to the web today: HTML (Hypertext Markup Language), HTTP (Hypertext Transfer Protocol) and URIs (Uniform Resource Identifiers). The lot also includes an animated visualisation of the code, a letter written by Berners-Lee reflecting on the process of creating it, and a Scalable Vector Graphics representation of the full code created from the original files.

Bidding for the NFT, which auction- house Sotheby’s claims is its first-ever sale of a digital-born artefact, opened on 23 June and attracted a total of 51 bids. The sale will benefit initiatives that Berners-Lee and his wife Rosemary Leith support, stated a Sotheby’s press release.

Climate Change and Energy Options for a Sustainable Future

In Climate Change and Energy Options for a Sustainable Future, nuclear physicists Dinesh Kumar Srivastava and V S Ramamurthy explore global policies for an eco-friendly future. Facing the world’s increasing demand for energy, the authors argue for the replacement of fossil fuels with a new mixture of green energy sources including wind energy, solar photovoltaics, geothermal energy and nuclear energy. Srivastava is a theoretical physicist and Ramamurthy is an experimental physicist with research interests in heavy-ion physics and the quark–gluon plasma. Together, they analyse solutions offered by science and technology with a clarity that will likely surpass the expectations of non-expert readers. Following a pedagogical approach with vivid illustrations, the book offers an in-depth description of how each green-energy option could be integrated into a global-energy strategy.

In the first part of the book, the authors provide a wealth of evidence demonstrating the pressing reality of climate change and the fragility of the environment. Srivastava and Ramamurthy then examine unequal access to energy across the globe. There should be no doubt that human wellbeing is decided by the rate at which power is consumed, they write, and providing enough energy to everyone on the planet to reach a human-development index of 0.8, which is defined by the UN as high human development, calls for about 30 trillion kWh per year – roughly double the present global capacity.

Human wellbeing is decided by the rate at which power is consumed

Srivastava and Ramamurthy present the basic principles of alternative renewable sources, and offer many examples, including agrivoltaics in Africa, a floating solar-panel station in California and wind-turbines in the Netherlands and India. Drawing on their own expertise, they discuss nuclear energy and waste-management, accelerator-driven subcritical systems, and the use of high-current electron accelerators for water purification. The book then finally turns to sustainability, showing by means of a wealth of scientific data that increasing the supply of renewable energy, and reducing carbon-intensive energy sources, can lead to sustainable power across the globe, both reducing global-warming emissions and stabilising energy prices for a fairer economy. The authors stress that any solution should not compromise quality of life or development opportunities in developing countries.

This book could not be more timely. It is an invaluable resource for scientists, policymakers and educators.

Designing an AI physicist

Merging the insights from AI and physics intelligence — **Best of both worlds** We need to merge the insights gained from artificial intelligence and physics intelligence. Credit: D Dominguez/CERN

Can we trust physics decisions made by machines? In recent applications of artificial intelligence (AI) to particle physics, we have partially sidestepped this question by using machine learning to augment analyses, rather than replace them. We have gained trust in AI decisions through careful studies of “control regions” and painstaking numerical simulations. As our physics ambitions grow, however, we are using “deeper” networks with more layers and more complicated architectures, which are difficult to validate in the traditional way. And to mitigate 10 to 100-fold increases in computing costs, we are planning to fully integrate AI into data collection, simulation and analysis at the high-luminosity LHC.

To build trust in AI, I believe we need to teach it to think like a physicist.

I am the director of the US National Science Foundation’s new Institute for Artificial Intelligence and Fundamental Interactions, which was founded last year. Our goal is to fuse advances in deep learning with time-tested strategies for “deep thinking” in the physical sciences. Many promising opportunities are open to us. Core principles of fundamental physics such as causality and spacetime symmetries can be directly incorporated into the structure of neural networks. Symbolic regression can often translate solutions learned by AI into compact, human-interpretable equations. In experimental physics, it is becoming possible to estimate and mitigate systematic uncertainties using AI, even when there are a large number of nuisance parameters. In theoretical physics, we are finding ways to merge AI with traditional numerical tools to satisfy stringent requirements that calculations be exact and reproducible. High-energy physicists are well positioned to develop trustworthy AI that can be scrutinised, verified and interpreted, since the five-sigma standard of discovery in our field necessitates it.

It is equally important, however, that we physicists teach ourselves how to think like a machine.

**Jesse Thaler** is a professor at MIT and director of the US National Science Foundation’s Institute for Artiﬁcial Intelligence and Fundamental Interactions. Credit: J Charney

Modern AI tools yield results that are often surprisingly accurate and insightful, but sometimes unstable or biased. This can happen if the problem to be solved is “underspecified”, meaning that we have not provided the machine with a complete list of desired behaviours, such as insensitivity to noise, sensible ways to extrapolate and awareness of uncertainties. An even more challenging situation arises when the machine can identify multiple solutions to a problem, but lacks a guiding principle to decide which is most robust. By thinking like a machine, and recognising that modern AI solves problems through numerical optimisation, we can better understand the intrinsic limitations of training neural networks with finite and imperfect datasets, and develop improved optimisation strategies. By thinking like a machine, we can better translate first principles, best practices and domain knowledge from fundamental physics into the computational language of AI.

Beyond these innovations, which echo the logical and algorithmic AI that preceded the deep-learning revolution of the past decade, we are also finding surprising connections between thinking like a machine and thinking like a physicist. Recently, computer scientists and physicists have begun to discover that the apparent complexity of deep learning may mask an emergent simplicity. This idea is familiar from statistical physics, where the interactions of many atoms or molecules can often be summarised in terms of simpler emergent properties of materials. In the case of deep learning, as the width and depth of a neural network grows, its behaviour seems to be describable in terms of a small number of emergent parameters, sometimes just a handful. This suggests that tools from statistical physics and quantum field theory can be used to understand AI dynamics, and yield deeper insights into their power and limitations.

If we don’t exploit the full power of AI, we will not maximise the discovery potential of the LHC and other experiments

Ultimately, we need to merge the insights gained from artificial intelligence and physics intelligence. If we don’t exploit the full power of AI, we will not maximise the discovery potential of the LHC and other experiments. But if we don’t build trustable AI, we will lack scientific rigour. Machines may never think like human physicists, and human physicists will certainly never match the computational ability of AI, but together we have enormous potential to learn about the fundamental structure of the universe.