DUNE at Sheffield

The Deep Underground Neutrino Experiment (DUNE) is a next-generation long-baseline neutrino experiment in the USA, scheduled to start data-taking in the mid-2020s.



DUNE has similar physics goals to Hyper-Kamiokande but uses a completely different strategy to achieve them. As a result, the DUNE and Hyper-Kamiokande projects are highly complementary: the combination of the two will be much more effective than either one alone.

Like all long-baseline neutrino experiments, DUNE has three principal components: a beamline, a near detector, and a far detector. The far detector will be located at the Sanford Underground Research Facility (SURF)—better known to neutrino physicists as the Homestake Mine, former home of the original Davis solar neutrino experiment.

The dark matter experiment LZ, in which the Sheffield Dark Matter group is participating, is also housed at SURF. The near detector and neutrino beamline are at Fermilab, giving a baseline of 1300 km from the target to the far detector. 

Components of DUNE

Unlike T2K and Hyper-K, DUNE is an on-axis experiment. It therefore sees a broad range of neutrino energies, peaking between 1 and 5 GeV as shown in the figure, as opposed to Hyper-K's relatively narrow energy range peaking around 600 MeV.

However, the key variable for neutrino oscillation experiments is the ratio of baseline to neutrino energy, L/E: Hyper-K has an L/E value of 500 km/GeV, and DUNE's range of ~260–1300 km/GeV covers this value, showing that both are optimised for the same squared mass difference Δm213.

The choice of an on-axis geometry has advantages and disadvantages: the absolute sensitivity to oscillation is reduced, because a smaller fraction of the beam is at the L/E value corresponding to peak oscillation probability (this is why T2K and NOνA, designed when we had only an upper limit to the value of the mixing angle θ13, both chose an off-axis geometry), but the ability to observe the oscillation over a range of L/E values helps to distinguish between different phenomena, e.g. the effects of CP violation and those of matter-enhanced oscillation. It does mean that accurate reconstruction of the neutrino energy is crucial to realising the physics potential of the experiment.

The DUNE far detector, as currently planned, consists of four independent LAr TPCs, each containing about 17 ktons of liquid argon. There are two possible technologies: single-phase, as used in all the large-scale LAr TPCs that have been built to date, and dual-phase, as used (with liquid xenon rather than liquid argon) in the LZ experiment.

In dual-phase designs, the electrons are drifted out of the liquid and into an overlying layer of gaseous argon, which allows higher electric fields to be used and therefore greatly increases the gain. 

Both technologies are set to be tested over the next couple of years in the ProtoDUNE project at CERN, and both may eventually be used in DUNE—however, as the single-phase design is the one that has been proven in large LAr detectors, it is agreed that the first of the four modules will use this technology.

The DUNE near detector will be located a few hundred metres downstream of the target. Its design is not yet finalised, but the current reference design, loosely based on the NOMAD experiment (1995–1998) which ran in the CERN neutrino beam, calls for a straw-tube tracker and electromagnetic calorimeter running in a 0.4 T dipole magnetic field for momentum measurement and charge determination—an important capability since, as can be seen from the figure, the principal contaminant in the Fermilab νμ beam is ν̅μ (and vice versa).

Alternative designs include a high-pressure gaseous argon TPC and a liquid argon TPC: these would have the advantage of using the same target nucleus as the far detector, thereby reducing near-far systematics.

The current plan is that the four modules of the far detector will be installed over the period 2024–2027, with the beam operational in 2026.

DUNE at Sheffield

The proposed DUNE far detectors will be by far the largest LAr TPCs ever constructed: at 760 tons, the ICARUS LAr TPC is less than one-twentieth the size of a single DUNE module. 

Consequently, an extensive R&D programme is needed to ensure that the detector design is appropriate and that testing procedures and assembly protocols are well thought out. This has been the main focus of DUNE work in Sheffield to date. 

For Sheffield involvement in SBND, which—though it has a well-motivated physics programme of its own—is in many respects a testbed for DUNE, and the ProtoDUNE-SP experiment at CERN, see the links menu. Here we will focus primarily on the 35-ton prototype and on physics studies.

The 35-ton prototype


Although there are several single-phase LAr TPCs in operation (ICARUS, MicroBooNE, ArgoNeuT) or under construction (SBND), the current design for the DUNE far detector includes various specific features that have not been field-tested. 

The 35-ton prototype was designed to incorporate these features so that they could be tested in a realistic environment before the construction of ProtoDUNE-SP. The 35-ton working group lists these as "the FR4 field cage construction, the gaps between modules, the wrapped wire planes, the light guide style photon detection system, cold ADC ASIC in the readout electronics, and the development of triggerless DAQ operation." 

It was also the first LAr TPC for neutrino physics to use a membrane cryostat—a system designed for bulk shipment of liquid natural gas, in which the support structure is warm and the cryogenic liner is not load-bearing. 

Some of these, such as the segmented anode plane assemblies and the wrapped wire planes, are also being implemented in SBND and ProtoDUNE-SP.

Construction and operation of the 35-ton prototype

A schematic diagram of the 35-ton prototype is shown in the figure. Like the reference DUNE single phase design, but unlike SBND and ProtoDUNE-SP, it has a double-sided anode plane assembly (APA) within the liquid argon volume. 

The DUNE design is symmetrical, but this proof-of-principle prototype has the APA close to one end, with the main drift region on one side and a nominal 20 cm drift region on the other.

As described in the introduction to LAr TPCs, 3D track reconstruction is achieved by using the electron drift time for the third coordinate. The reference time, corresponding to the time at which the particle crossed the detector volume, is determined by reading out the scintillation light from the liquid argon, since the light travel time is negligible compared to the electron drift time. 

Liquid argon is a good scintillator, but it emits in the far ultraviolet (~128 nm), which is not optimal for most photodetectors. Consequently, the photon detection is done using lightguides coated with a wavelength shifter which absorbs the UV light and re-emits in the visible. 

This secondary emission is then detected by "silicon photomultipliers" (SiPMs)—a form of avalanche photodiode.  The wavelength shifting lightguides were installed between the wire planes of the APA.

The 35-ton prototype operated in two phases: a test phase from December 2013 to February 2014, which was a test of the cryostat and purification system with no detector installed, and a fully instrumented run in March 2016 (cooldown and filling December 2015 – January 2016; detector commissioning February 2016; data run started March 7 2016). 

Unfortunately, the latter phase came to an abrupt end on March 19 when a pipe broke as a result of vibration in the pump to which it was connected, resulting in contamination of the liquid argon with air. 

Including data taken during commissioning, this yielded a total of 22 days of good-quality data for analysis. The data consist of through-going cosmics triggered using the cosmic ray counters surrounding the detector.

The cold camera

One of Sheffield's contributions to the 35-ton prototype was a camera system designed to operate in the liquid argon volume and monitor the detector during and after cooldown and filling. 

The components of the camera are shown in the photo, both separately and assembled in the housing. A total of 8 cameras were installed to monitor various parts of the detector.

The principal purpose of the camera system was to observe the interior of the detector during cooldown, filling and HV switch-on, with the aim of identifying problems such as breakdown in the high-voltage system.

This was successfully accomplished: the picture is a still from a movie showing the liquid argon filling. It proved possible to power cycle the cameras in the cold: in particular, after a 9-day power outage they all came back on without problems.

The cameras used were not, of course, designed to operate at LAr temperatures, and it was in fact necessary to provide a small amount of local heating, using resistors as labelled in the photo. This did not cause any problems with detector operation (in particular, worries that they might cause the formation of bubbles of gaseous argon proved unfounded). 

The images did tend to degrade with longer time spent in the cold, but this is a relatively minor disadvantage: the main value of the cameras is to help diagnose any problems that become apparent at or shortly after the start of detector operation.

Event reconstruction

Data from the 35-ton prototype turned out to be unexpectedly noisy, with a signal to noise ratio of ~10 on the collection plane (vertical wires, y) and only ~2 on the induction planes (diagonal wires, u and v). Worse, the detector sometimes entered a "High Noise State" involving coupled oscillation of all detector components. 

This obviously adversely affected event reconstruction and data analysis. Subsequent investigations have identified most of the noise sources and proposed modifications to the electronics that would eliminate them. There is also a proposed, though as yet not verified, explanation for the High Noise State.

As a result of this work, it is unlikely that the ProtoDUNE and DUNE electronics will suffer from the same pathologies.  However, this does not help those trying to analyse the 35-ton data, in which the noise is unavoidably present. 

High noise levels cause problems in track reconstruction by making the initial hit-finding procedure more difficult.  Typically, hit-finding algorithms look for a peak in the signal from each wire. 

If there is a lot of electronic noise, a low threshold for defining a peak will result in a large number of spurious hits, whereas a high threshold will cut out the noise but will reduce hit-finding efficiency, as some real hits are also likely to be lost. Either of these outcomes is likely to reduce the efficiency and precision of the track reconstruction.

Sheffield students are among those trying to mitigate the effects of the noise by developing more sophisticated hit-finding algorithms. One of these, developed by Matthew Thiesse, is RobustHitFinder, which uses a variable threshold optimised for each individual wire.

Another feature of hit and track reconstruction which can be studied in the 35-ton prototype, but not in detectors such as SBND where the APAs are at the edges of the LAr volume, is the behaviour of a track that crosses the APA. 

Because of the geometry of the electric field, ionisation electrons produced between the wire planes of the APA and the grounded mesh that separates the two sides of the double-sided APA actually drift back on to the plane that the track has just passed through.

 If the resulting drift time is interpreted as being in the usual direction, this results in a hook-like feature in the track, which can be seen in the data (see figure).  Clearly it will be possible to correct for this effect, but only after hits have been assigned to tracks.

Lifetime measurements

A key goal of the 35-ton prototype was to measure the lifetime of the ionisation electrons, which is a function of the purity of the liquid argon (argon itself being a noble gas with a full outer electron shell, electrons are much more likely to be captured on impurities). 

A standard method of measuring this is to reconstruct through-going cosmic muon tracks—which, since muons are minimally ionising particles, should lose, on average, a constant amount of energy to ionisation per unit track length—and look at the amount of charge collected as a function of drift distance. 

The result should be a negative exponential, q(t) = q0 exp(–t/τ), where q =dQ/dis the mean charge deposited per unit track length and the time t is derived from the drift distance.

This measurement is also affected by high noise levels. The reason for this is as follows:

  • The most probable value of dQ/dx (a more stable proxy for the mean value) is found by fitting a Landau-Gaussian convolution to the distribution of deposited charge values.
  • However, if the threshold for identifying hits has to be raised because of noise, hits with lower dQ/dx are systematically missed, and the most probable value is overestimated. Because the threshold is not sharp (since hits are Gaussian peaks rather than delta functions) and the rising edge of a Landau-Gaussian convolution is quite steep, this does not produce a noticeably worse fit to the Landau-Gaussian, so it is not obvious that this has happened.
  • This effect becomes more pronounced as the true most probable value of dQ/dx becomes smaller, so the result is that the exponential fall-off appears slower than it really is, and the lifetime is overestimated.

This effect can be simulated by adding real-data noise (from events with no tracks) to simulated hit waveforms with known electron lifetimes. The results of such studies suggest that the true electron lifetime is ~3.2 ms, which is in agreement with the value of ~3.5 ms implied by direct measurements of the purity of the argon.

Physics studies


Fundamentally, DUNE is a long-baseline neutrino oscillation experiment, with L/E value optimised for θ23 and θ13.  The primary physics goal of DUNE is to study neutrino oscillations, specifically:

CP violation—the difference in oscillatory behaviour between neutrinos and antineutrinos, caused by the e factor in the PMNS matrix and matter-enhanced oscillations, which should be significant at DUNE's 1300 km baseline (and not for Hyper-Kamiokande's 300 km) - which are sensitive to the mass hierarchy, i.e. whether m3 is the lightest or the heaviest neutrino mass eigenstate.

(It is assumed that the actual oscillation parameters θ23, θ13, Δm213 and Δm223 will be well measured by the time DUNE is operational, although reducing the systematic error budget to the levels anticipated in the DUNE and Hyper-Kamiokande design reports will also permit improvements in these.)

These are important measurements, especially the measurement of CP violation which, it could reasonably be argued, is on a par with the discovery of the Higgs boson: the Higgs boson tells us about the origin of mass, while CP violation in the neutrino sector may tell us about the origin of matter, in the sense of the matter-antimatter asymmetry in the Universe (why is our Universe almost entirely matter, when creating particles in the laboratory always yields particle-antiparticle pairs?). 

However, the DUNE and Hyper-Kamiokande physics programmes both extend well beyond neutrino oscillations. In particular, the duty cycle of neutrino beams is usually poor (which is actually a good thing, assuming that you don't want to vaporise the target—it simply means that the duration of the intense bunches of protons on target is much shorter than the time between bunches), so both experiments have a wide programme of non-accelerator physics to make use of the data that the far detector collects between bunches. 

This is especially relevant for DUNE, since the first module of the DUNE far detector is scheduled to be up and running two years before the beamline.  Sheffield physicists are active in several aspects of the non-accelerator programme.

Proton decay

It is not always remembered that searching for proton decay was the original motivation for the first generation of large water Cherenkov detectors: KamiokaNDE originally stood for Kamioka Nucleon Decay Experiment not Neutrino Detection Experiment. 

The motivation for looking for proton decay was, and remains, the study of theories of particle physics which go beyond the Standard Model. Such theories include supersymmetry—which predicts an entire spectrum of new particles—and Grand Unified Theories, i.e. theories which attempt to unify the strong force with the electroweak force. 

By their nature, GUTs include interactions which couple hadrons to leptons, allowing the violation of baryon number and the possibility of proton decay (albeit with an extremely long lifetime!).

At that time, the favoured decay mode was p → e+ π0: a particularly suitable signature for water Cherenkovs since the positron and the two photons from the π0 decay will produce clear Cherenkov rings. Indeed, limits on proton decay from water Cherenkov detectors sounded the death knell of the first popular Grand Unified Theory, SU(5). 

SU(5) predicted this decay mode for the proton, with a lifetime of < 1032 years: Kamiokande and IMB set lower limits approaching 1033 years, and SU(5) in its original form was therefore disfavoured.

 It's amusing to note that another prediction of SU(5) is massless neutrinos, so the large water Cherenkovs definitely seem to have a dislike for this particular GUT!  (A somewhat different GUT called "flipped SU(5)" is still with us: theorists are inventive people and not easily deterred.)

Supersymmetry-based GUTs tend to predict much longer lifetimes (moving from 1028–32 years to 1035–38 years): this is because the GUT-mediated proton lifetime is of order M42m5, where M is the unification mass scale, α is the fine structure constant and m is the proton mass, and including supersymmetry pushes the unification scale up by a bit more than a factor of 10. 

However, SUSY itself introduces additional mechanisms for proton decay: the most straightforward one, d + u → s̃ → e+ + u̅, predicting p →e+ π0 with a lifetime of 1 second(!), is suppressed by R-parity (decays such as this are the reason for introducing R-parity in the first place), but loop diagrams such as the one in the figure are still allowed. 

This shifts the dominant decay mode from p → e+ π0 to p → K+ ν̅τ and predicts a proton lifetime of 1029–35 years depending on the SUSY mass scale.

This decay mode is much harder for water Cherenkovs: the neutrino is invisible and to reduce background you really need to identify the kaon (i.e. separate kaons from pions at kinetic energies of a hundred MeV or so). With its good tracking of low-energy particles and potentially good particle identification by dE/dx, DUNE should be well suited to investigating decay modes like this.

The critical issue in all searches for rare events is controlling the background. In this case, the decay signature is an isolated, fully contained kaon track. The kaon energy for the decay of a stationary free proton would be 599 MeV (momentum 339 MeV), but some variation will be introduced by the decaying proton being bound in a nucleus. 

One possible source of background would be cosmic-ray interactions just outside the detector volume: if such a decay produced a K0, the K0 could propagate into the detector volume unseen and then undergo a charge-exchange scatter, K0 + p → K+ + n (the d quark in the K0 being exchanged for one of the u quarks in the proton), which could in principle give the isolated-kaon signature.

Sheffield's Vitaly Kudryavtsev is an expert in the simulation of cosmic-ray muons and is co-convenor of the DUNE cosmogenic backgrounds working group, and has been working on assessing this background and developing cuts to suppress it.

For more information on this topic, please see Martin Richardson's PhD thesis.

Supernova neutrinos and supernova relic neutrinos

The dominant interaction channel for low-energy neutrinos in DUNE is quasi-elastic scattering off argon, νe + 40Ar → e + 40K*.  Note that this is sensitive to electron neutrinos, unlike the dominant channel in water Cherenkovs, inverse beta decay on hydrogen, which sees electron antineutrinos. 

The signature is a single low-energy electron, with the possibility of tagging the gamma ray from the de-excitation of the potassium as a further identification.

The unique property of electron neutrinos from a core-collapse supernova is that the initial neutronisation spike from the formation of the neutron star produces only νe, from p + e → n + νe.  As can be seen from the figure, this feature is considerably suppressed by neutrino oscillations in the case of inverted ordering of neutrino masses (i.e. m3 < m1 < m2)—but it's totally eliminated in the case of normal ordering (m1 < m2 < m3).  Therefore, observation of the neutronisation spike in a Galactic core-collapse supernova should immediately resolve the mass hierarchy.

Recent 3D simulations of core-collapse supernovae exhibit a striking and unexpected feature known as lepton-emission self-sustained asymmetry (LESA).  This is a dipole anisotropy between νe and ν̅e fluxes, which can have an amplitude of up to 20%. 

Involvement in both DUNE (sensitive mainly to νe) and Hyper-Kamiokande (mainly ν̅e) will offer the opportunity to investigate this effect: with several thousand events expected in DUNE and tens of thousands in Hyper-K, an asymmetry of a few percent should be clearly visible.

The principal problem for supernova neutrino detection with DUNE is that the detector is optimised for much higher energy neutrinos (the LBNF neutrino beam is in the several-GeV range), and so supernova neutrinos in DUNE are "crummy little stubs" (Kate Scholberg, 2016) and will present a challenge to trigger designers: although the burst itself is probably unmistakable (several thousand events in a 10-second window), it will be necessary to trigger on such events without deluging the DAQ with triggers from background. 

It is likely that this makes studying supernova relic neutrinos in DUNE extremely difficult—they have the same low energies (even lower if redshifted) without the large flux.

In order to realise the full physics potential of a core-collapse supernova observation in DUNE, it is necessary not only to count the events but also to reconstruct them and measure the neutrino energy. 

Collective neutrino oscillations, driven by neutrino-neutrino interactions in the interior of the supernova, may cause changes in the observed neutrino energy spectrum as a result of essentially complete transformation of one neutrino flavour into another above a certain threshold. 

Such effects are more dramatic in the νe case than ν̅e, because the νe spectrum is more different from νμ,τ than is the case for antineutrinos, but clearly in order to investigate this it is necessary to reconstruct the neutrino energy.

 This is the advantage of both DUNE and Hyper-K over IceCube—with its much larger effective volume, IceCube will see far more events, but it will be unable to do anything other than simply count them.

In summary, observing supernova neutrinos in DUNE provides a unique window on the supernova—both water Cherenkovs and large liquid scintillator detectors see ν̅e rather than νe. However, reconstructing such low-energy events is a non-trivial task, and the comparatively small size of DUNE compared to Hyper-K and IceCube will make it difficult to explore some of the features of modern supernova simulations (e.g. rapid oscillations in flux) owing to lack of statistics. 

Combining data from all experiments (IceCube for statistics, Hyper-K for lower statistics but with energy measurements, DUNE for comparatively poor statistics but νe sensitivity) will be the best way to maximise the physics potential of a Galactic core-collapse supernova, should we be fortunate enough to observe one while these experiments are operational.  



Liquid argon TPCs are a relatively new technology: although the idea has been around for decades, the number of currently operating LAr TPCs is quite small, and the TPCs planned for the DUNE far detector will be more than ten times the volume of the largest currently operating (ICARUS). 

Therefore, it is worth exploring design ideas in detail before committing to a final DUNE design. The ProtoDUNE project is designed to test both the "standard" single-phase design and an alternative dual-phase setup in the CERN H4 testbeam, which provides a mixture of electrons, muons, pions, kaons and protons at a selected (within 5–7%) momentum in the range 0.5 to 7 GeV.

Either positive or negative particles can be selected: for negatively charged particles, the beam is mostly electrons and pions, with 1–2% π, ~1% antiprotons, and a K fraction that rises from 0 at low momentum to 3.4% at 7 GeV, while for positive charges there is (unsurprisingly) a much higher proton fraction, 10–20% except at the very lowest momenta.

Clearly ProtoDUNE is not a neutrino experiment in itself. The main aims of the ProtoDune project, as stated in the ProtoDUNE-SP Technical Design Report, are to:

  • prototype the production and installation procedures for the single-phase far detector design
  • validate the design from the perspective of basic detector performance—this can be achieved with cosmic-ray data
  • accumulate large samples of test-beam data to understand/calibrate the response of the detector to different particle species
  • demonstrate the long-term operational stability of the detector as part of the risk mitigation program ahead of the construction of the first 10-kt far detector module.

The Sheffield group is working on the single-phase detector, ProtoDUNE-SP (CERN experiment NP04). There is also a dual-phase prototype, ProtoDUNE-DP (CERN NP02) which will pursue similar goals for the dual-phase design, in which the ionisation electrons are drifted out of the liquid into an overlying layer of gaseous argon, in which a more intense electric field can be used to amplify the produced charge by avalanching.

This technique is also used by the liquid xenon dark matter detector LZ, in which Sheffield is also involved; however, Sheffield is not participating in ProtoDUNE-DP. 

Note that as the DUNE far detector will consist of four independent modules, the choice of single vs dual phase is not either/or: it would be entirely possible to have, say, two of each. As the single-phase technology is more mature, the current plan is for the first module to be single-phase.

ProtoDUNE-SP at Sheffield

ProtoDUNE-SP is similar in design to SBND, with a central cathode plane assembly (CPA) and two sets of anode plane assemblies (APAs) at the ends of the cryostats.

 As this is a larger detector, roughly 11 m × 11 m × 11 m, there are three APAs per side instead of two. As with SBND, Sheffield has undertaken to deliver the steel APA frames, which are being manufactured by Portobello-RMF Ltd.

Obviously, the physics goals of both ProtoDUNE detectors involve the reconstruction and analysis of charged particle tracks and showers from the testbeam. 

The mixture of electrons and pions should give both showers (from the electrons) and tracks (from the pions), and as the beam momentum is pre-selected it will be possible to test methods for momentum/energy estimation, such as the multiple scattering method described in the SBND section. 

Sheffield has existing expertise from SBND and the DUNE 35-ton prototype, and will contribute to this team effort.

In addition, the ProtoDUNE detectors will not be deep underground and will collect a substantial sample of cosmic-ray data. Vitaly Kudryavtsev of Sheffield is a specialist in this field, and therefore the Sheffield group will focus in particular on understanding this cosmic sample through simulation, analysis and tuning reconstruction algorithms, with the goal of improving reconstruction and particle identification for cosmic-ray events in the DUNE far detector. 

DUNE itself will of course have proportionately fewer cosmic events, because of its deep underground site, but cosmic-ray muons are useful for calibration purposes, and some classes of muon-induced events may be significant backgrounds for some DUNE physics analyses, in particular the search for proton decay.