DsTau is a new project which has been proposed at the CERN SPS to study tau-neutrino (ντ ) production  with the aim of providing important data for future ντ measurements.
Indications of possible lepton non-universality have been recently reported by several experiments, including studies of B meson decays , that showed enhancements of phenomena involving τ and ντ . These results may indicate new physics effects between heavy flavour quarks and leptons. One approach to investigate these anomalies is to study neutrino scattering. If there are additional particles mediating the interactions between heavy quarks and heavy leptons, an increase in the ντ cross section would be suggestive evidence of new physics. However, to date, experimental results on the ντ cross section are too poor to probe these questions . A new precise measurement of the ντ cross section is necessary to test new physics effects in ντ –nucleon CC interactions. This measurement also has practical importance for neutrino oscillation experiments and astrophysical ντ observations
The concept of ντ cross section measurement in a nonoscillated ντ beam is shown in Fig. 1. The dominant source of ντ is the sequential decay of Ds mesons, Ds+ → τ +ντ → Xντ ντ and Ds- → τ −ντ → Xντ ντ , produced in high-energy proton interactions. However, there is no experimental measurement of the Ds differential production cross section in fixed target experiments using proton beams, which leads to a large systematic uncertainty on the ντ flux estimation. This was the main source of error in the ντ cross section measurement  at >50%, larger than the 33% relative statistical uncertainty due to the limited number of detected events (nine in total). While the statistical uncertainty is expected to be reduced down to the 2% level in future ντ programs as already incorporated in the SHiP project , it is vital to reduce the uncertainty of the ντ flux prior to such high statistics experiments.
The DsTau project aims to reduce the systematic uncertainty in the cross section measurement from >50% to 10%. This will be achieved by detecting 1000 Ds → τ → X events and thus measuring the Ds differential production cross section in 400 GeV proton interactions. This double decay occurs at a distance of ∼5 mm. The challenge of this measurement is the detection of the 1 tiny kink angle of the Ds → τ decay, which has a mean kink angle of 7 mrad. For this purpose, emulsion detectors with nanometric precision readout will be used. The emulsion detector has a position resolution of 50 nm , which leads to an intrinsic angular resolution of 0.35 mrad with a 200-µm-thick plastic base layer (Fig. 2 left). As shown in Fig. 2 right, each detector unit comprises a 500 µm-thick tungsten target, followed by 10 emulsion films interleaved with 200 µm-thick plastic sheets acting as high-precision particle trackers as well as decay volumes for short-lived particles
A module is made of ten such units followed by an ECC to measure the momenta of the daughter particles. A total of 370 modules will be exposed to the 400 GeV proton beam from SPS at a density of 105protons/cm2 uniformly on the module surface. 4.6 × 109 protons on target will be collected, yielding to 2.3×108 proton interactions in the tungsten plates, and 1000 detected Ds → τ decays.
The data analysis will require the full area scanning of the 1000 m2 emulsion surface by the world’s fastest readout system, the Hyper Track Selector (HTS) . After detecting τ decay topologies, events will be analysed by dedicated high-precision systems using a piezo-based high-precision Z-axis, allowing emulsion hits to be measured with nanometric resolution. To study the differential production cross section of Ds mesons, the momentum of the Ds meson (PDs ) must be measured. Because Ds mesons decay quickly and the invisible ντ ’s escape measurement, the direct measurement of PDs is not possible. However, the peculiar event topology gives us indications of PDs. For example, Ds → τντ is a two-body decay with well-defined decay momentum. Therefore, the kink angle of Ds→ τ is a good indicator of PDs . Because the Ds→ τ → X decay topology has two kink angles (θDs→τ , θτ→X) and two flight lengths (FLDs→τ , FLτ→X), the combination of these four variables effectively provides an estimate of PDs . A machine-learning algorithm was trained with a simulated sample (τ → 1 prong) using the four variables to estimate PDs , the result of which is shown in Figure 3. The momentum resolution is estimated to be 18%.
In addition to the primary aim of measuring Ds production, analysing 2.3×108 proton interactions, combined with the high yield of 105 charmed decays produced as by-products, will enable the extraction of additional physical quantities such as the interaction length of charmed hadrons, the Λc production rates and the search of super-nuclei.
Two test beam campaigns were performed in November 2016 and May 2017 at the CERN SPS. The upper left panel of Figure 4 shows the detector setup at the H4 beamline. To analyse the data, a new tracking algorithm has been developed to reconstruct tracks in the extremely high track density of O(105–106 ) protons/cm2 , which is 1000 times higher than that of OPERA. An example of the reconstructed data from the detector is shown in the upper right panel of Figure 4. A systematic search of the decay topologies of charmed particles was applied and the first double charm event is shown in the lower panel of Figure 4, proving that analyses of short-lived particles in actual experimental conditions are possible.
The CERN-SPSC approved a pilot run in 2018, and recommended beam time for a physics run in 2021. We are currently producing emulsion films for 30 detector modules for the pilot run in August 2018. This is primarily intended to provide a test of large data taking and an estimation of the background, but which also already allows us to re-evaluate the ντ cross section measured by DONUT by significantly reducing the overall systematic uncertainty. With the outcome of the physics run, DsTau will provide essential inputs for future tau neutrino experiments and pave a way for the search of new physics effects in ντ –nucleon interactions.
 S. Aoki et al. [DsTau collaboration], CERN-SPSC-2017-029, SPSC-P-354 (2017).
 G. Ciezarek et al., Nature 546 (2017) 227-233.
 K. Kodama et al., Phys. Rev. D 78 (2008) 052002.
 M. Anelli et al., CERN-SPSC-2015-016, SPSC-P-350 (2015).
 C. Amsler et al., JINST 8 (2013) P02015.
 M. Yoshimoto, T. Nakano, R. Komatani and H. Kawahara, PTEP 10 (2017) 103.