1

Project1 Number: 690835 Project Acronym: MUSE

Project title: Muon Campus in US and Europe Contribution

Periodic Technical Report Part B

Period covered by the report: from 01/01/2016 to 31/12/2017 Periodic report: 1st

1 The term ‘project’ used in this template equates to an ‘action’ in certain other Horizon 2020 documentation

2

1. Explanation of the work carried out by the beneficiaries and Overview of the progress

The MUSE project coordinates the activities of about 70 researchers from four European research institutes [HZDR (DE), INFN (IT), UCL (UK), University of Liverpool (UK)] and three small/medium-sized enterprises [AdvanSiD (IT), CAEN (IT), Prisma Electronics (GR)] which participate in experimental research at the Fermilab (USA) Muon Campus. MUSE promotes international and intersectoral collaboration by means of secondments of personnel, thus enhancing the European contribution and visibility in this field. The duration of the project is four years, and is well adapted to the Muon Campus schedule that will host in the same period two world class experiments dedicated to the search of new physics: Muon (g-2), for a four-fold improvement on the measurement of the muon magnetic anomaly, and Mu2e, for the search of the as-yet unobserved Charged Lepton Flavour Violation (CLFV) process. European research institutions have a leading role in these activities in both detector development and construction and in calibration and analysis of the data.

Detector development and construction are progressing very well. In Muon (g-2), MUSE researchers are involved in the straw-tracking system, which measures the muon beam vertical profile with an accuracy better than 10 mrad and identifies pileup and lost-muon events, and in the laser calibration system which monitors the photosensor gain at the sub-per mil level. Both detectors components were installed inside the experiment and commissioned with the first muon beam in June 2017. Calibration techniques have been outlined and are being refined. The first physics run is planned for the beginning of February 2018. Mu2e detectors of interest to MUSE are the crystal calorimeter, which provides unprecedented timing performance for low energy electrons in the presence of a strong magnetic field, exploiting solid state photosensors, and the Mu2e high-purity germanium detector, reconstructing X-rays at rates surpassing previous experiments. Both detectors operate in a very hostile environment due to the high radiation levels. Mu2e detectors are currently completing the prototyping stage and are moving to the production phase, in line with the scheduled plan.

Networking among participants, training of personnel, dissemination and outreach are also important aspects of the project. A lot of effort has been dedicated to these activities. This has produced an intense transfer of knowledge among participants and a high visibility of the project both towards the scientific community and the general public. Overall, MUSE is advancing as planned, with almost all tasks progressing as expected and all deliverables and milestones foreseen in this reporting period completed. The list is reported in Tab. 1.

1.1 Objectives

The following list summarizes the MUSE objectives and goals achieved in the first two years:

1. Establish new collaborations among European groups participating in the Muon Campus experiments, increasing their presence and visibility at Fermilab and strengthening the already existing partnership with the laboratory.

The European research groups participating in the Muon Campus experiments successfully profited from the mutual exchange of skills triggered by MUSE existence, sharing and consolidating different area of expertise in advanced technologies. Notable examples are the common work on irradiation damage, which involves all the parties, the participation of Muon (g-2) laser experts in the design of the Mu2e laser system and the transfer of

3

knowledge on latest generation silicon photosensors for the development of detectors able to monitor online high-intensity laser plasma experiments.

Nr. Name WP Type / Diss.Lev.

Lead Beneficiary

Expected deliv. date

Submitted Achieved

D1.1 Report on laser integration 1 Report/Public INFN 31-12-2017 12-01-2018

D1.2 Report on g-2 tracker 1 Report/Public LIVERPOOL 31-12-2017 31-12-2017

MS1 Tracker-DAQ integration 1 UCL 31-05-2017 10-04-2017

D2.1 Calorimeter TDR 2 Report/Public INFN 31-12-2016 28-12-2016

D3.1 g-2 laser

calibration system

3 Other/Public INFN 31-10-2016 30-12-2016

D3.2 g-2 tracker tools 3 Report/Public UCL 30-06-2017 04-07-2017

D3.3 Mu2e laser system 3 Report/Public INFN 30-06-2017 13-10-2017

D4.2 g-2 full muon simulation 4 Report/Public UCL 31-12-2016 30-12-2016

MS6 New release of the g-2 software 4 UCL 31-12-2016 22-11-2016

D5.1 MUSE @

HZDR Open Day

5 Other/Public HZDR 30-09-2016 30-09-2016

D5.2 Annual Physics Meetings 5 Other/Public INFN 31-10-2017 17-10-2017

D7.1 First Progress Report 7 Report/Confid. INFN 31-12-2016 30-12-2016

D7.3 MUSE website 7 Website/Public INFN 31-05-2016 31-05-2016

D7.4 1st MUSE General Meeting

7 Other/Public INFN 30-09-2016 30-09-2016

D7.5 2nd MUSE

General Meeting

7 Other/Public INFN 30-09-2017 01-08-2017

D8.1 NEC – Req. 1 8 Ethics/Confid. INFN 31-01-2016 20-01-2016 D8.2 NEC – Req. 2 8 Ethics/Confid. INFN 31-01-2016 20-01-2016 D8.3 NEC – Req. 3 8 Ethics/Confid. INFN 31-01-2016 20-01-2016

MS8 Management structure in

place 8 INFN 31-01-2016 11-01-2016

Table 1: List of deliverables and milestones due in the Reporting Period, with the expected date of delivery and the actual date of completion.

4

All EU partners significantly increased the presence at Fermilab of their personnel, with a grand total of 130 person-months of secondments. This has granted a larger involvement in the day-to-day life of the experiments and a higher visibility. This is demonstrated by the new institutional roles assigned to EU researchers since the start of the project: chair of the Institutional Board, DAQ & online coordinator, detector operation coordinator, conveners of three analysis groups, three members of the Executive Board and the Speaker Committee. The increased presence at Fermilab has allowed us to carry out these leadership roles effectively. 2. Design, construction and commissioning of state-of-the-art detectors for the Muon (g-2)

and Mu2e experiments

Detector related activities are progressing very well, with all tasks in line with the expected progress. The meticulous work planning and the application, when needed, of the foreseen mitigations allowed the submission of all deliverables and the achievement of all due milestones in this period connected with Work Packages 1-4: “g-2 detectors”, “Mu2e detectors”, “Calibration” and “Software Tools”. The description of this work is reported in Sections 1.2.1-1.2.4.

3. Exploit the existing European infrastructure to create a network of radiation hardness tests

and characterization of detector components.

In the framework of the project, it is important to underline the new-born collaboration among MUSE partners for the irradiation tests. Dedicated campaigns took place at the ELBE facility in HZDR since the start of the action. MUSE researchers from all EU research institute (HZDR, INFN, LIVERPOOL and UCL) – together with US colleagues – tested the radiation hardness of calorimeter components and the functionality of the Stopping Target Monitor of the Mu2e experiment. More details on irradiation tests are reported in Sections 1.2.2 and 1.2.6.

4. Transfer of knowledge among partners

Besides the activities among research institutes already discussed in point 1., transfer of knowledge is enhanced by the presence of industrial partners, thus enforcing inter-sectoral exchanges. As an example, during the collaboration between INFN researchers and PRISMA engineers for the construction of the Quality Assurance (QA) test stations for the Mu2e calorimeter (Sec. 1.2.6), the experience in the commissioning of complex electronic systems of a private company was applied to develop a complete list of procedures for crystal traceability during the production phase, from delivery to QA tests and storage, including non-conformance. On the other hand, irradiation tests of crystals with thermal neutrons have been performed in Frascati during the secondment of PRISMA engineers, which acquired expertise also in this sector. Another example is the recent INFN-CAEN collaboration for the development of a robust waveform digitizer board able to survive in a harsh radiation environment (Sec. 1.2.6).

1.2 Explanation of the work carried per WP

MUSE activities are organized in seven Work Packages. WPs 1-4 include R&D activities, design and construction of Mu2e and Muon (g-2) detectors where European groups are directly involved and have leading roles. Dissemination and promotion of these results towards the general public are coordinated by WP5. WP6 manages the transfer of knowledge among

5

partners to exploit as much as possible the network, including contributions from different experiments, institutions and competences. WP7 coordinates and supervises activities across all the WPs.

1.2.1 Work Package 1: g-2 Detectors

The objectives of this Work Package are the development, construction, test and commissioning of key elements of the Muon (g-2) experiment: the tracker system with its readout electronics and DAQ, the laser DAQ interface boards and the Bias Voltage Supply (BVS) for the calorimeter Silicon Photo-Multipliers (SiPMs). Three of the four tasks have been carried out as expected, with all related deliverables and milestones completed. Task 1.2, carried out by CAEN in collaboration with INFN, had a delay in the completion of the R&D phase. The tight schedule of the experiment required a different solution for the BVS system, so that the contribution of the CAEN company within the project has been reorganized. Details are reported in Section 5.1.

Task 1.1, “Development, construction and commissioning of Laser-DAQ boards” has been carried out by INFN, in collaboration with FNAL for the final phases of installation and commissioning. Deliverable 1.1 “Report on laser integration” was uploaded on time on 12/1/2018 and all the goals of the laser DAQ interface boards have been achieved. The DAQ interface boards were tested during the engineering run (15/5/2017 – 7/7/2017) and completed during the fall. The laser Control Board (LCB) is at present installed and working (Fig. 1.a, b), along with six Source Monitor (SM) controllers, both embedded in a crate (Fig. 1.e). The DAQ interface boards are ready for the physics running that is planned for 01/02/18. The Laser Control Board (LCB) is a key element of the Laser Calibration system which acts as an interface between the beam cycle and the calibration system, triggers the laser pulses and distributes the time reference signals to the monitoring electronics. The LCB can be operated in three different modes:

a) Pulse train generation. In this mode triggers to the lasers are provided at a fixed, programmable frequency, either while muons are stored in the ring (“in-fill” pulses) or in absence of muons in the ring (“out-of-fill” pulses). In order to homogeneously sample all time intervals, the pulse train is regularly shifted by a fraction of the pulse period.

b) Physics event simulation, or “flight simulator” mode (Fig. 1.c). In this mode, the LCB generates a time sequence and a mean number of pulses according to an exponential function A×e −t/τ with a characteristic time of 64.4 µs and a normalization factor A, reflecting the muon decay time distribution.

c) Synchronization mode. The lasers send a reference signal to the calorimeters, for reset, synchronization and initialization of the detectors and their electronics.

The six Source Monitor (SM) controllers filter and digitize the signal coming from the source monitor (Fig. 1.d). In a source monitor light enters into an integrating sphere, viewed by 3 photo-detectors, namely 2 PIN diodes and one photomultiplier. The PMT is also coupled to a radioactive Americium-241 source embedded in a sealed NaI crystal. The α particles traversing the scintillator provide an absolute energy reference. Custom electronics has been designed to read, process and digitize the signals coming from the SM photo-detectors. The SM controller also reads environment and operating parameters, like temperature and bias/HV voltage, provides the operating voltage to the photo-detectors and self-calibrates the electronic readout

6

Figure 1: (a) Block diagram of the LCB. (b) The LCB implemented by a Spartan6 FPGA board, a BeagleBone Black CPU and a custom board which acts as an interface for signal exchange and component communication. (c) Time distribution of pulses generated according to an exponentially decreasing function “flight simulator” for the two implemented methods. (d) Preamplifier and SM board scheme for a single readout channel. (e) The Source Monitor crate in operation at the Fermilab Muon (g-2) experiment (E989). It contains, from left to right, the Laser Control board, a fast FANOUT unit (100 ps uncertainty) providing triggers to the 6 lasers, 6 Source Monitor Board and the Controller board.

preamp board

(a)

(b)

(c) (d)

(e)

7

channels. All these modules are controlled by a Crate Controller Board, which acts as data collector and sends out data over a TCP/IP connection.

Tasks 1.3 (“Design and construction of the straw trackers”) and 1.4 (“Readout system for the straw trackers”) have been carried out by LIVERPOOL and UCL, respectively, in collaboration with FNAL. PRISMA company partially contributed to the electronics and to the tests of Quality Assurance in Liverpool. Deliverable 1.2 “Report on installation and initial commissioning of all trackers” was uploaded, on time, on 31/12/2017 and all associated Muon (g-2) detector milestones have been achieved. M1: “DAQ Integration” was completed 10/04/2017, 12 weeks ahead of the milestone date. All straw tracking detectors have been installed and they took their first muon beam data during the Muon (g-2) commissioning run in June 2017. The physics running with the detectors begins on 01/02/18. The detectors underwent a very successful test-beam in June 2015 where a radial position resolution of 100 microns was demonstrated: well below the design-requirement of 200 microns. A picture of the straw trackers in-situ in the storage ring's vacuum chamber is shown in Fig. 2 along with the detector infrastructure (gas, cooling, off detector electronics), and a module in the test-stand in Lab-3 at FNAL, where all the modules were leak tested and characterised with radioactive sources prior to installation in the Muon (g-2) ring.

Figure 2: (LEFT) The straw trackers in situ inside the vacuum chamber of the Muon (g-2) ring. (MIDDLE) The off-detector electronics, cooling and gas infrastructure for the straw trackers. (RIGHT) A single straw tracker being characterised in the 'source-stand' in Lab-3 at Fermilab.

The performance of the tracking detectors has been excellent, far better than their equivalent predecessors in the BNL experiment. The flash at beam injection has not caused any issues with the HV or the frontend electronics and high quality data has been accumulated by the DAQ system with very little downtime. Tracks have been reconstructed in real-time as part of the online Data Quality Monitor system and subsequently in the offline framework. The impact on the storage-ring's vacuum has also been well within the required 10-6 Torr.

The trackers are used to make six key measurements, all of which are critical in the determination of the systematic uncertainty for the aµ measurement:

• The profile of the beam which is then convoluted with the B-field map to determine the average B-field experienced by the decaying muons.

• The momentum of the decay positrons (matched to energy deposits in the calorimeter) so that E/p measurements can be used for an in-situ energy calibration of the calorimeter and to monitor time dependent response changes.

• The rate of muons lost from the storage region.

8

• The level of pileup in the calorimeter, determined by counting the number of calorimeter clusters matched to multiple tracks.

• The vertical asymmetry in the e+ decays which would be a signature of a non-zero muon EDM.

• The amplitude and period of the coherent betatron oscillations (CBO) of the beam both vertically and radially which are used to optimise and monitor the beam injection and which must be minimised since they are a source of known bias to the aµ determination.

Fig. 3 shows that the trackers have already largely met their performance goals in terms of being able to reconstruct these important distributions.

Figure 3: Charged particle and beam distributions measured by the straw trackers. On the top row: the momentum of the tracks (LEFT) and the radial and vertical beam position (RIGHT). On the bottom row: the time distribution of the tracks, showing the characteristic g-2 oscillation (LEFT) and the Fourier-Transform of the time distribution of tracks (RIGHT) showing the g-2 frequency (∼ 230 kHz) and the CBO frequency (∼ 430 kHz).

The DAQ/readout has been demonstrated to work at instantaneous data rates and volumes a factor of 10 higher than anticipated and has run without issues with beam since June 2017.

1.2.2 Work package 2: Mu2e detectors

Work Package 2 refers to the activities carried out for the Mu2e detector. The package is focused on two detector components: the Electromagnetic Calorimeter (EMC) and the Stopping

9

Target Monitor (STM). The EMC is a large size crystal calorimeter whose work has reached full steam in these first two years of the network. The STM is a high-resolution photon detector that has started its prototyping phase and is now completing its design phase.

The only deliverable of this WP due in the first two years of the project, D2.1: “Calorimeter TDR: Technical Design Report of the calorimeter, including engineering design”, was uploaded on time on 28/12/2016. The related Task 2.1, “Technological choice and calorimeter design”, carried out by INFN in collaboration with FNAL and CALTECH, has been completed. The other tasks are still in progress, in line with expectations at this stage of the project. INFN is the main contributor to Tasks 2.2 to 2.5, while LIVERPOOL and UCL to Task 2.6. HZDR is contributing to the irradiation tests of calorimeter crystals (Task 2.2, “Tests and characterization of crystals”) and photosensors (Task 2.3, “Test and characterization of silicon photosensors”) and of the STM prototype (Task 2.6, “Prototyping and design of the HPGe monitor”). The PRISMA company was involved in Tasks 2.2 and 2.3, contributing to the development of the procedures for Quality Assurance, a database to store test parameters and the electronics needed to operate the test stations. More details on the contribution of HZDR and PRISMA are described in Sec. 1.2.6. CAEN contributed to the development of a prototype for the high-current low voltage power supply of the calorimeter electronics (Task 2.4, “Design, construction, test of calorimeter FEE”).

In the first six months of 2016, the calorimeter underwent a long list of changes to complete its technical choice and its final design. These changes were caused by the technical delay on reaching full functionality with the baseline solution of BaF2 crystals readout with solar blind Avalanche Photodiodes (APDs). Indeed, the development of these APDs was slower than expected and was impacting the overall calorimeter preparation schedule. The foreseen mitigation of using a backup option, based on un-doped CsI crystals readout with UV extended Silicon Photomultipliers (SiPMs), worked very well. A test beam on a small size prototype demonstrated the viability of this detector choice and proved to satisfy all calorimeter requirements. After this down-select, a lot of effort was carried out on adjusting the electronics design especially in the front-end part. The calorimeter mechanics scheme remained substantially unchanged since the density of CsI and BaF2 crystals are very similar. On July 14th, the USA Department of Energy (DOE) granted its approval (CD3) for the calorimeter thus providing the green light for the construction.

The second half of 2016 was dedicated to completing the design by adjusting the mechanical (Task 2.5) and electronics (Task 2.4) engineering and starting the pre-production phase for crystals (Task 2.2) and sensors (Task 2.3). Two international bids were carried out for the vendors’ selection. In the first tender, 72 CsI crystals were procured from St.Gobain (France), Siccas (China), Amcrys (Ukraine) and their characteristics measured at the semi-manual QA stations at LNF. Tests of mechanical precision and radiation hardness were also performed. Results of the optical parameters are shown in Fig. 4 (left). Most of the crystals satisfied the technical specifications although crystals from Siccas and St.Gobain were much better performing than Amcrys and have been selected as final vendors for the calorimeter production. For the photosensors’ bid, a custom Mu2e sensor was designed as an array of (6´6) mm2 UV SiPMs and 150 pieces were procured from Hamamatsu (Japan), AdvanSiD (Italy) and Sensl (Ireland). Tests of their properties and functionality were carried out at the semi-manual QA station in Pisa. Examples of the measurements are shown in Fig. 4 (right). Radiation hardness tests were performed with a strong neutron flux at the pELBE facility at HZDR. At the moment of writing, the whole production is starting and a set of automatic test stations are being assembled to precisely measure the properties for the 1450 (3350) production crystals (SiPMs). A database prototype has been tested on the pre-production and is being optimized in view of

10

the production phase. The upcoming deliverable due on 31/12/2018, D2.2: “Production DB”, is expected to be completed on time.

Figure 4: Distribution of technical parameters for pre-production of crystals and sensors. (LEFT) Light Yield, Longitudinal Response Uniformity, Energy Resolution and ratio between Fast/Total components. (RIGHT) Gain and Photon Detection Efficiency for the Mu2e SiPMs.

At the end of pre-production, in 2017, the construction and test of a large size calorimeter prototype with all final components (Module-0) is important to validate the engineering model. Module 0 consisted of an array of 51 pre-production crystals, 102 pre-production Mu2e SiPMs and 102 pre-production preamplifiers. In Fig. 5 (left), the picture of the rear side of the Module-0 is shown. While assembling this prototype, we have tested the procedure for gluing of Mu2e SiPMs on the SiPM holders (Fig. 5 (right-top)), the functionality of the Front End electronics, the realization of the aluminum support disk for the crystals, the procedure of crystal assembly, the realization of the rear-disk in insulating material (Zedex, PEEK) and the installation of the copper cooling lines and of the SiPM holders. The Module-0 was completed in April 2017 and tested at the Frascati Beam Test Facility (BTF) for one week in May 2017. The detector was exposed to an electron beam of energy between 60 and 120 MeV in order to test the achievable resolution and tune the electronics readout. The integration of the readout was completed up to the preamplifier stage. For the digital readout, a commercial device from CAEN was used since the prototype of the digitizer was not ready in time for the beam test and was completed in the fall of 2017. In Fig. 5 (right-bottom), the picture of the first prototype of the Mu2e calorimeter digitizer (DIRAC) is shown. The analysis of the test beam data demonstrated that Module-0 comfortably satisfies the calorimeter requirements. Indeed, despite an unexpected noise related to the commercial boards, we were able to achieve an energy resolution of 5.4% at 100 MeV, with a clear understanding of the energy resolution terms, and a timing resolution better than 120 ps at 100 MeV (Fig. 6).

11

Figure 5: (LEFT) Rear view of the Module-0; details of the cooling lines and of the FEE chips are visible. (RIGHT) One holder with two Mu2e SiPMs glued on its front face and two preamplifiers (top) and a DIRAC board (bottom).

Figure 6: (LEFT) Energy resolution as a function of the beam energy. Black dots are the best fit, red dots are for the average of fits performed on many integration ranges. The insert is an example of a fit at 100 MeV. (RIGHT) timing resolution as a function of the beam energy.

The completion of the mechanical, thermal and electronics executive drawings is underway. The construction of a full-size mockup, loaded with dummy iron crystals wrapped as in the final configuration, will guide our assembly procedure and determine the tolerances needed for the construction of the final pieces, towards the fulfillment of milestone MS2, “First EMC disk assembled”, due on 30/06/2019. A set of tests for radiation hardness of the electronics components is being planned: neutron and dose irradiation will be carried out at HZDR

12

(Dresden) and in Calliope (Enea, Italy). The level of dose expected in the experiment has been evaluated with precision by inserting the mechanical design in the simulation and, although some shielding has been added, it is still very high for some of the components such as the FPGA, DC-DC converters and optical links. A revision of the design for the electronics is underway at the moment of writing.

Deliverable 2.3: “Design of Mu2e HPGe completed”, related to Task 2.6, is due with a report on 31/03/2018 and is on track for completion. A prototype HPGe detector underwent a beamtest at HZDR/Elbe in August 2017 to refine the design and ensure that the detector could run at the required rates with the appropriate resolution and to determine any changes in resolution as a function of the irradiated dose. Bremsstrahlung photons with a mean energy of 5 MeV were derived from a 15 MeV electron beam. The energy spectrum is shown in Fig. 7 (left). The mean energy and occupancy (20%) were configured to be the same as expected in Mu2e, albeit with a larger pulse separation (2.4 µs vs 1.8 µs in Mu2e). At the same time, as being irradiated by 54-72 kHz of photons, the detectors were exposed to 137Cs and 60Co sources at a rate of 100 Hz to determine whether these lines could be discerned amongst the background at a rate comparable to that required in Mu2e.

Figure 7: (LEFT) The spectrum of photons at HZDR/Elbe used to irradiate the prototype HPGe detector. (RIGHT) The reconstructed Cs, Co X-ray lines at two different photon rates (54 and 72 kHz).

Fig. 7 (right) shows the X-ray lines can clearly be reconstructed (at 100 Hz) and the resolutions at the two photon rates are within one sigma of each other. The detector resolution was also determined prior and subsequent to the irradiation (3×1010 MeV) and no degradation in resolution was observed, although the total dose was no more than the equivalent dose from a day of Mu2e running. This test will be repeated in subsequent source and beamtests with a higher dose rate. The beamtest data has also been used to begin the development of the pulse finding algorithms that have to run at high speed (and high efficiency) on an FPGA, although to date these have been developed solely in software. Different alignments of the detector with respect to the beam demonstrate that the detector’s high-energy (E > 2 MeV) performance can be degraded without affecting the more critical, low-energy (E < 2 MeV) performance. A HPGe detector very similar to the prototype should be able to fulfil the requirements of the STM. The detector will be procured before the end of FY17/18. A first design of the DAQ/readout using an FPGA-based TDC and a FPGA readout board has already been reviewed at FNAL and a test-stand has been established at UCL and also at FNAL where it has been interfaced to a

13

prototype clock and control distribution system. The version of the Moving-Window-Deconvolution (MWD) pulse finding algorithm applied to the beamtest data is presently being translated into firmware. The design of the collimation system which reduces the radiation damage to the STM, while ensuring a sufficient number of X-rays to be recorded, has begun. It comprises of two components: a field-of-view collimator and a spot-size collimator. The former, a 15 cm long, 45 cm thick lead collimator with an aperture of approximately 7 cm, has jaws that are movable in the x and y directions, and ensures the STM has a line of sight to the aluminium stopping target only and not to other components that would affect the X-ray rate. This is located 34 m upstream of the STM after the main detector systems and a so-called sweeper magnet used to remove low energy electrons. The spot-size collimator is a 15 cm thick tungsten disk with a 5 mm opening that is rotated across the face of the detector such that the detector can be uniformly irradiated and operated for several months without annealing. The collimators must be controllable remotely and reliably and interface to the FNAL-wide interlock and control system (iFix). The Milestone associated with the STM: MS3, “HPGe detector installed” due on 31/12/2019 is expected to be satisfied without any issues.

1.2.3 Work package 3: Calibration

The objectives of WP3 are the development, construction and assembling of the calibration systems and procedures of both the Muon (g-2) and Mu2e detectors. All tasks are progressing as expected, with all the related deliverables and milestones due in the first two years of the project completed.

Task 3.1, “Development and assembly of the g-2 laser system” was carried out by INFN in collaboration with FNAL, and it is now completed. The report for deliverable D3.1 “g-2 laser calibration system” was uploaded, on time, on 30/10/2016. After the completion of the baseline design, some improvements have been applied in 2017. The laser calibration system is now ready for the physics run planned for 01/02/18. The Muon (g-2) laser calibration system consists of the following main components:

• high frequency pulsed laser (405 nm); • optical components and collimators; • 24, 25 m-long quartz fibers for light distribution (one per calorimeter); • in the vicinity of each calorimeter a beam expander incorporating an engineered diffuser

for the uniform distribution of the laser light to the 54 transport fibers; • 24 bundles of 54 fibers (+ spare + monitor) that transport the light from the diffusers to

the calorimeters; • 24 light distribution panels, in front of the calorimeters’ crystals made of Delrin, which

host 54 right-angle prisms to deflect by 90° the output of the optical fibers into the PbF2 crystals;

• 6 source monitors to measure the laser intensity stability and furnish a fast pulse to the local monitor PMTs;

• 24, 25 m-long quartz fibers for light monitor, embedded in the 24 fiber bundles; • 24 local monitors to control the stability of the light distribution system.

The source monitor (SM) measures the fluctuations of the laser light source and provides corrections for these fluctuations (Fig. 8.c). The laser light is mixed in the SM and measured by a redundant system of two large-area PIN diodes (Hamamatsu S3590-18) and a PMT (Hamamatsu H5783). The PMT is also illuminated by an absolute reference signal provided by

14

a low-activity (about 6 Hz) Americium radioactive source coupled to a NaI crystal. In order to ensure a uniform distribution of light on the detectors and an insensitivity to “beam pointing” fluctuations, a commercial diffusing sphere was used (Thorlabs, mod. IS200).

Figure 8: (a) The optical table in the laser hut with lasers and optical components. (b) The rack of the laser calibration system. (c) The correction made using the laser calibration system. (d) The on-line monitor for the laser calibration system showing the 24 signals from the local monitors and the 18 signals from the source monitors. (e) The double pulse optical setup. (f) The measured short-term gain variation of SiPM on crystal 23, calo 17, as a function of the delay between the two laser pulses.

Time(ns)(nsec)

Relativ

egain

(e)

(f)

(a)

(b)

(c)

(d)

15

The local monitor consists of a PMT (Hamamatsu R1924A-100) that receives two optical signals. The first signal is the reference from the source monitor, collected from a port of the integrating sphere, and is used to calibrate the gain of the PMT. The second signal comes from the fiber bundle in the vicinity of the calorimeter and is representative of the calibration signal sent to the SiPM. The two optical signals are separated in time by 250 ns, since the first signal travels a distance of approximately 2 meters, while the second one travels about 50 m (25 m one way in the launching fiber, 25 m return in the local monitor fiber).

In July 2016, the assembly of the laser calibration system started at the Fermilab Dzero building. The hardware installation was completed in June 2017 and was used during the first engineering run (Figs. 8.a-b-d). Indeed, in June, for the first time, it was possible to calibrate the gain of all 1296 SiPMs using laser pulses of different intensity by varying the filters in the filter wheels placed in front of the six lasers (energy calibration). During the second half of 2017 the hardware was updated in order to perform the so-called “double pulse calibration” (Fig. 8-e), which measures the short and long term (50 ns and 50 µs respectively) gain curve drop for the SiPMs, due to high rate signals (Fig. 8-f). Movable and fixed mirrors have been added in order to send two laser pulses, generated by a Digital Delay Generator (SRS mod. DG645) and separated in time by a suitable delay, to the same calorimeter. This work towards milestone MS5 (“g-2 calibration commissioned”, due on 31/12/2018) is part of Task 3.3, “Commissioning of the calibration systems in g-2”, which is a common effort of INFN (for the laser system) and LIVERPOOL/UCL groups (for the tracking detector).

Tasks 3.2, “g-2 straw calibration system”, and 3.3, “Commissioning of the calibration systems in g-2”, for the tracker has been carried out by UCL and LIVERPOOL in collaboration with FNAL. The report for deliverable D3.2 “g-2 tracker tools” was uploaded, on time, on 04/07/2017 and the associated milestone MS5 “g-2 calibration commissioned” is complete as far as the Muon (g-2) tracking detectors are concerned. The straw tracker calibration system comprises three test-stands: one measuring the leak rate (vacuum test-stand), one measuring the gain and wire positions (source test-stand) and one determining the time to distance calibration (cosmic test-stand). The results of these are used to calibrate the detectors and set their optimum operating conditions (high-voltage, gas mixture, threshold) and to optimise the simulation so that it reliably matches the observed data. This optimisation is critical to ensure that the performance of the trackers meets the design goals. The Muon (g-2) muon beam is vertically focussed by 4 sets of quadrupoles that pulse for O(1ms) at 20-35 kV. To avoid sparking, they must operate in a vacuum of at least 1 µTorr. The trackers flow Ar/Ethane through mylar straws in the vicinity of the quadrupoles and the capacity of the vacuum pumps requires a leak rate of less than 6 µTorr l/s per tracker module to maintain the appropriate vacuum for operation of the quadrupoles. A vacuum/leak calibration test-stand has thus been established at FNAL. At the start of the module production prior to wiring and mounting, straws are initially selected from the large batch available to have a leak rate at least a factor of two less than the specification. Once the module has been constructed and shipped to FNAL it then undergoes a rigorous leak/vacuum calibration in the test-stand at MTEST. The leak rate of the straws is determined using the “rate-of-rise” (ROR): the module is inserted into the vacuum tank and pumped down until the pressure is low and stable, the vacuum pump is then turned off and the pressure ROR is measured which is then converted to a leak rate. The time it takes to attain a pressure of 30 µTorr, the pressure attained after 24 hours of pumping and the ROR can all be used to determine whether the module has a leak rate that is sufficiently low for the module to be put into the Muon (g-2) storage ring. The results of this calibration for eight modules are shown in Fig. 9 (left). The leak/vacuum calibration test-stand has also been used to compare the leak rates with different gasses: N2, Ar/CO2, Ar/Ethane and has been used to measure the dependence of the

16

leak rate (primarily through permeation through the mylar straw wall) on temperature. A test-stand to calibrate the straw tracking modules using cosmic rays was also established in Lab-3 at MTEST. The stand comprised of two scintillator paddles and aluminium plates one of which has been milled to be the same as the vacuum plates in the storage region such that the trackers can be placed with identical spacing and mounted in the same way but rotated by 90 degrees. A custom PCB was developed so that a TTL signal from the scintillators PMT NIM electronics could be readout by the same TDC used by the straw tracker, thus ensuring that the same readout/DAQ software could be utilised and that the clocks would be synchronous across the two detector systems. A dedicated MC simulation was also developed so that the results from the cosmic-stand could be used to improve the simulation, and hence calibration of the detectors. Hits recorded in a straw tracker from the cosmic stand are shown in Fig. 9 (right).

Figure 9: (LEFT) Measured pressure as a function of time recorded in the vacuum/leak calibration test-stand in FNAL MTEST. The dotted lines indicate the lowest pressure ultimately reached by a given module when flushed with N2. (RIGHT) Hits recorded by a tracker in the cosmic test-stand showing the uniform distribution of hits in both straw planes except at the edge where the scintillator acceptance drops off.

The final test-stand established at FNAL was one using radioactive sources where each module was exposed to both a 90Sr source: to measure the wire positions and determine the optimum HV setting and a 55Fe source to measure the gain. The sources were mounted in a collimated holder and scanned across the straws in X-Y using a computer-controlled motor stepping in 6 micron increments. The trajectory of the 90Sr electrons is determined by the source collimation and a scintillator/Si PMT detector that is readout using the same bespoke PCB developed for the cosmic test-stand. The number of hits from a 90Sr source is recorded as a function of HV to determine the so-called “plateau” region, where the gain does not vary significantly with HV and where the rate of “re-firing” is not significant. A plot of this for two different gases is show in Fig. 10 and defined the operating voltages of 1625 V (Ar/Ethane) and 1500 V (Ar/CO2) used in the June-2017 commissioning run.

A 55Fe source deposits a known charge (at two energies) in the straws and can be used to calibrate the gain of the detectors. The hit rate is recorded as a function of threshold at thresholds above the range at which the ASDQ saturates. This rate is fitted with two Gauss error functions and an exponentially falling noise distribution as a function of threshold to determine the gain at several HV values.

17

The above systems calibrate the detectors in terms of gain (hit efficiency), initial alignment, resolution and the time-to-distance calibration. The final aspect of the calibration and the one critical to attaining the best possible performance is the implementation of these results in the Monte Carlo simulation: the detector measures hit times and these must be converted to a hit position which are then fitted to form tracks. The time-to-distance (so called “r-t”) calibration is thus particularly important and the inverse of this must be implemented in the Monte-Carlo i.e. from GEANT hit positions, hit times are computed. The GARFIELD++ programme is used to simulate the generation of hits from the ionisations and avalanches created by charged particles passing through the gas of the straw trackers. A new C++ version of the programme was developed and we have implemented several improvements to the code so that it now reproduces all the features of the well-established Fortran version of the code and matches well the data recorded in the test-stands at FNAL.

Figure 10: (LEFT) Measured number of hits from a single straw tracker module recorded in 10ms as a function of HV for two different gases: Ar/CO2 and Ar/Ethane from a 90Sr source. (RIGHT) The hit rate (Hz) recorded from a 90Sr source as it is moved across the detector. The peak can be used to accurately determine the straw (and wire) positions as a constraint in the final alignment procedure. The motor moves in 6 µm steps.

Task 3.4, “Development of the laser calibration system for Mu2e” and Task 3.5, “Installation/test of laser calibration system in Mu2e”, are being carried out by INFN in collaboration with FNAL for the integration in the Mu2e DAQ system. The deliverable D3.3 “Mu2e Laser system”, due on 30/06/2017, had a report uploaded on 13/10/2017. The major reason for the delay was the change of technology that occurred in 2016, when BaF2 crystals, emitting at 220 nm but blind above 280 nm, were replaced with CsI crystals, emitting at 315 nm, so that many system components, such as the Laser head, had to be substituted.

The laser system in Mu2e has a very important role and is part of a set of calibration tools needed to reach the timing and energy resolution goals. A sophisticated source calibration system will perform weekly crystal by crystal equalization whilst offline reconstruction of in-situ cosmic rays and decay in orbit events will keep track of slow changes in the timing and energy scales. However, calorimeter performance will be quickly changing due to the high neutron and ionizing dose irradiation that could provide: (i) a combined reduction of light yield and increase of radiation induce noise of the crystals; (ii) an increase of leakage currents of the sensors. This harsh environment requires the sensors to be cooled to 0 °C and a fast and

18

continuous monitoring of the detector gains, the timing offsets and the energy and timing resolutions to be performed. These tasks will be carried out by the laser system that, in order to do so, should satisfy these technical specifications:

1. have enough power to illuminate all 1374 crystals with an optical distribution system; 2. track (with 1% precision) the variation of the light at the source, and at the last

distribution step, by means of a monitoring system based on photodiodes; 3. emit on a green wavelength to be in a region far away from the CsI emission peak (310

nm), where the transmittance changes due to irradiation are small; 4. send pulses at a constant frequency by means of an external trigger and get them

synchronized with the machine Clock in order to be acquired in the “beam-off” region.

The design was completed thanks to a fruitful transfer of knowledge between the Mu2e and Muon (g-2) collaborators, relying on the expertise and progresses already achieved by the latter group and on the existence of the MUSE network. An overall scheme is shown in Fig. 11(left). Similarities with the Muon (g-2) laser system are: the need to get an optical distribution system to divide the beam in many sectors and a photodiode monitoring system, while large differences exist on the needs to: (i) send light with launching fibers inside the vacuum region with dedicated feed-throughs, (ii) favor a uniform distribution on the secondary bundles by means of diffusion spheres, (iii) select radiation hard fiber bundles with low outgassing values and (iv) design of a custom fiber-needle to bring light to the crystal inside the custom Mu2e SiPM/FEE holders.

An example of the monitor capability of the Mu2e laser system was obtained by the first test of a small size laser prototype with Module-0. The distribution of pulse height for one crystal fired with the laser is shown in Fig. 11 (right-top) for one minute running. The laser was fired at 7 Hz, so as to acquire around 400-440 events/minute. The relative precision on the mean, reached in 1 minute, was of 0.4%, thus allowing variations of 1.5% on the mean to be quickly determined. Changes on the response along time (see Fig. 11 (right-bottom)) was consistent

LASER noise and resolution @ BTF

� LASER resolution/pulse 7%, 4.5% resolution due to 500 pe received at the end of the diffusing system,  5% due to the laser  jitterÆ 25 MeV equivalent signal.

� Each 400 events, error on mean is equivalent to 0.4%  Æ as expected 7%/sqrt(400)� Drift of the system up to 1.5 % in 35’ .. Consistent with variation inside +/‐ 1 C (+/‐ 3 %)

First 1’of run# 260

last 1’of run# 260

QL(pC) QL(pC)

QL/QR Nev

<QL>

Example ofLASER runningfor run # 260.

35 ’ run50 Hz trigger

7/25/2017F. Porter | DOE IPR Review 43

Figure 11: (LEFT) Overall scheme of the Mu2e laser distribution system. The black line in the middle indicates the Bulkhead of the Mu2e cryostat that keeps the detector in an evacuated region. The diffusing spheres are in vacuum. (RIGHT) Top: precision of light distribution in one minute of run. Bottom: gain drift in one hour running observed in the Module-0 SiPMs even in presence of a ± 1 ° C stable cooling system.

19

with the temperature stability of the cooling system that presented a maximum variation of ±1 °C. The precision obtained is related to the number of photoelectrons (p.e.) reaching the sensor; this has been estimated from the width of the ratio (L/R) distribution between the Left and Right SiPMs reading out the central crystal, and corresponded to around 600 p.e./pulse. The intrinsic laser precision achieved with this light yield is approximately 4% per event. At the moment of writing, work is ongoing on procuring the final components to meet the upcoming milestone MS4, “Mu2e laser assembled”, due on 30/30/2018, of assembling and commissioning a large size prototype with all final pieces and test them under vacuum.

1.2.4 Work package 4: Software tools

Work Package 4 refers to the development and commissioning of the software packages needed to simulate the detectors related to the project and to perform data reconstruction. Since Muon (g-2) is in a more advanced stage, data analysis is also planned. All the four tasks are progressing on track, and the milestone and deliverable due in this reporting period, both on Muon (g-2) software tools, were fulfilled.

Tasks 4.1, “Development of g-2 simulation and reconstruction code”, and 4.3, “Commissioning of g-2 analysis code”, are a common effort of INFN, LIVERPOOL, UCL and FNAL. The report for deliverable D4.2: “g-2 full muon simulation” was uploaded on time on 30/12/2016 and the subsequent milestone MS6: “New release of the g-2 software” was reached on schedule. Deliverable D4.2 was entirely MC based and incorporated a full simulation of the Muon (g-2) ring in GEANT4 within the ART software framework. Output events of were used in a series of mock data challenges and stored using the SAM data handling system. An example event from this sample, visualised using the Paraview package, is shown in Fig. 12.

Figure 12: A Paraview visualisation of a simulated event in the Muon (g-2) storage ring.

Since this milestone, the Muon (g-2) software tools have become significantly more sophisticated to reconstruct and analyse the data from the June 2017 commissioning run. Raw data events have been unpacked, reconstructed, calibrated, analysed and persistently stored and catalogued using the complete offline ART framework and FIFE data handling tools. The tools

20

have allowed the reconstruction of more than 10 million charged particle tracks in the straw tracking detectors and over 100 million clusters in the calorimeters. This has allowed the first analysis of the rate of precession of the decay positrons to be made using the June 2017 data. Fig. 13 shows the arrival times of the positrons recorded in the 24 calorimeters. It exhibits the characteristic oscillation due to the g-2 precession and illustrates that Muon (g-2) has the software tools ready for an end-to-end analysis of the data.

Figure 13: The arrival times (modulo 80 µs) of the 1 million positrons with energy > 1.8 GeV from the June 2017 commissioning run. aµ is extracted from a 5-parameter fit to this distribution (along with a 0.1 ppm measurement of the B-field).

The work on Task 4.2, “Development of Mu2e simulation and reconstruction code”, is carried out by INFN, HZDR, LIVERPOOL and FNAL, with the goal of completing the corresponding deliverable D4.1, “Mu2e code”, on time before 31/08/2018.

On the calorimeter side, the digitized signal shapes have been included in the simulation of the Mu2e calorimeter. Rise time, fast and long decay time components can be tuned to study the detector response and resolution. An algorithm to identify signal peaks in case of hits in pile-up has been developed. Fig. 14 shows an example of three overlapped signals. The fit is able to correctly identify the contributing hits. This procedure improves the energy resolution from 5% to 3.5% for conversion electrons of 105 MeV momentum when inserting a full background simulation. The extracted shape of the energy resolution as a function of the Longitudinal Response Uniformity (LRU) and Light Output (LO) was used to set the technical specifications of the combined calorimeter crystals and sensors as follows: LRU < 5% and LO > 20 photo-electrons/MeV.

Further technical details have been added to the calorimeter simulation: mechanical structure, DAQ crates and digitization boards (Fig. 15). This improved Monte Carlo version has been used to evaluate with high precision the ionizing dose and neutron flux reaching the electronics. The observed dose turned out to be too high for the selected FPGA chips; and a study to optimize the shielding with W/Cu plates is underway.

21

Figure 14: Signal shapes of simulated hits in pileup and the fit used to extract single hit contributions.

Figure 15: (LEFT) Simulated geometry of the Mu2e calorimeter. (RIGHT) Ionizing dose (top) and 1 MeV equivalent neutron total flux (bottom) in the SiPM region as a function of the calorimeter radius. Initial work has begun for milestone MS7: “Mu2e HPGe reconstruction code” which is due on 30/04/2019. Data from both the Mu2e simulation and the HZDR/Elbe beam-test is being analysed to develop the pulse finding algorithms that must run in real time to reconstruct the energy of the X-rays incident on the STM. These algorithms must run at high rate and be able to recover the degradation in resolution incurred due to irradiation. A moving-window-

22

deconvolution (MWD) algorithm has been developed and is in the first stages of being implemented on an FPGA where ultimately it must run to satisfy the rate and dead-time requirements.

The work for Task 4.4, “Commissioning of Mu2e reconstruction code with cosmic ray events”, involves INFN and LIVERPOOL in collaboration with FNAL, and will cover the last phase of the project. The related milestone, MS7 – “Mu2e software running on cosmic ray data” is due on 30/04/2019. First simulation studies for the calorimeter started using a Monte Carlo simulation of cosmic ray events (CRs) to determine the time offset of each calorimeter channel. As a first step, CRs crossing several calorimeter cells have been selected. An iterative procedure is then applied, fitting the calorimeter timing as a function of the relative distance of the selected cells imposing the muon speed as the velocity of light. A fit to the residual of each calorimeter channel provides a calibration set that is applied before repeating the procedure. This method allows to calibrate the calorimeter time offsets at a level better than 20 ps after few iteration steps, thus matching well the calibration needs of the detector.

1.2.5 Work package 5: Dissemination and outreach

The MUSE web site http://muse.lnf.infn.it/ documents all the dissemination and outreach activities of the project. All the tasks and deliverables of WP5 have been completed for the part due in the period reported in this document (2016-2017). Results have been disseminated to the scientific community through the participations to international conferences/workshops (55 events) and the publication in professional journals (28 peer-reviewed publications, more than 30 contributions to conference proceedings). Outreach events involved all major MUSE participants. Both activities promoting communication towards general public and training sessions for university students were organized, for a total of 14 events.

Task 5.1 (“MUSE Workshop day”) activities were performed during the MUSE general meetings of September 2016 in Pisa and of May 2017 in Frascati. In the Pisa meeting, we scheduled four seminars (http://muse.lnf.infn.it/event/597/) for students and for general attendance: one 1h seminar on the Fermilab Muon (g-2) experiment, one 1h seminar on the Mu2e experiment and two 15m reports by Fermilab summer students that worked in 2016 on MUSE activities. In connection with the Frascati MUSE meeting, a two-day outreach event has been delivered for the students of the Rome Tor Vergata university (http://muse.lnf.infn.it/event/2017-muse-general-meeting-outreach-event/), focused on the Mu2e experiment electromagnetic calorimeter. In the first day, three training lessons were delivered in the University, and on the second day a group of students has been involved in two sessions of hands-on experiences in the research laboratory of the local MUSE collaborators. Each session was introduced by a short presentation.

Task 5.2 (“MUSE open day”) has included the participation of MUSE collaborators to the open laboratories days in Dresden (May and June 2016), Pisa (September 2016) and Fermilab (September 2017). The HZDR MUSE collaborators have published a presentation of the MUSE project on the laboratory magazine (http://muse.lnf.infn.it/event/insider-issue-20-april-2016-in-german/), have attended the May 2016 HZDR open day with two posters (http://muse.lnf.infn.it/event/hzdr-tag-des-offenes-tur/) and the June 2016 “Long Night of the Science” in Dresden again with two posters (http://muse.lnf.infn.it/event/hzdr-muse-dresden-long-night-of-the-sciences/). HZDR activities were reported in the deliverable D5.1, “MUSE @ HZDR Open Day”, submitted on 30/09/2016. MUSE collaborators attended, with posters on MUSE Mu2e activities, the visitors of the September 2016 “Researchers' Night” event in Pisa

23

(http://muse.lnf.infn.it/event/muse-european-researchers-night/), in the context of the laboratories open day scheduled by the INFN Pisa branch (http://www.bright-toscana.it/2016/pisa/laboratoriaperti/). Several MUSE collaborators have participated to the Fermilab 50th anniversary Open House in September 2017 (http://muse.lnf.infn.it/event/fermilab-50th-anniversary-open-day/). They have produced and attended two exhibits for the Muon (g-2) experiment (small scale demonstrative prototypes of the laser calibration system and of the tracker) and the Mu2e slideshow. The Muon (g-2) area was the most visited place of the event, and MUSE collaborators produced two of the three major exhibits there.

Figure 16: Poster for the illustration of the Muon (g-2) laser calibration system exhibit at the Fermilab 50th anniversary Open House outreach event.

Figure 17: Poster for the illustration of the Muon (g-2) tracker exhibit at the Fermilab 50th anniversary Open House outreach event.

24

Task 5.3 (“Annual Physics Meeting at LNF”) has been fulfilled on the "Incontri di Fisica" events consisting in three-day training courses for high school teachers and scientific journalists, which INFN Frascati National Laboratory organized in October 2016 and October 2017. In both cases, MUSE collaborators acted as tutors of a working group providing introductive seminars followed by hands-on laboratory experiences conducted under their supervision. In 2016, MUSE researchers organized a working group on “Particle Detection with Scintillating Materials” (http://muse.lnf.infn.it/event/muse-idf-2016/), presenting the innovative Silicon Photo-Multipliers photosensors of the Mu2e experiment calorimeter and comparing their performance with traditional photomultipliers. In 2017, MUSE researchers organized a working group on “Particle Detection with Scintillating Materials” (http://muse.lnf.infn.it/event/muse-incontri-di-fisica-2017/) focused on the properties of scintillating crystals and Silicon Photo-Multipliers (SiPMs) of the Mu2e calorimeter. A MUSE collaborator was also involved in the organization of the whole training event. The above activities have been reported in the deliverable D5.3, “Annual Physics Meeting”, submitted on 17/10/2017.

Task 5.4 (“Coordination of UK outreach activities”) included public presentations of research activities, “masterclasses” and work-experience placements of students. Muon (g-2) and Mu2e research activities were presented in public in the overview of Physics research at UCL in February 2016 and March 2017. Liverpool MUSE researchers organized masterclasses in February 2016, June 2016 and March 2017. Each masterclass was provided to 80-100 students and included lectures, guided tours of the laboratory clean rooms and a demonstration of the straw chamber production process, over about seven hours. In the period from June to August 2016, two students did work-experience placements with the UCL group working on the production of the Muon (g-2) tracker, including a one-month stay at Fermilab. Two groups of four students each did work-experience placements on the production of the Muon (g-2) tracker with the Liverpool group in June and in August 2016.

Task 5.5 (“Summer School at Fermi National Accelerator Laboratory”) was performed scheduling seminars and guided tour of the MUSE collaborators laboratories for the Fermilab summer students. In August 2016 (http://muse.lnf.infn.it/event/training-lectures-fnal-summer-school-students/), we provided four seminars on the Muon (g-2) experiment, corresponding to a total of 3 hours and 10 minutes, and three guided tours to the UK MUSE collaborator laboratories for ten students each. In August 2017 (http://muse.lnf.infn.it/event/training-lectures-fnal-summer-school-students-2/), we provided five seminars on the Mu2e experiment, corresponding to a total of 3 hours and 40 minutes, and guided tours to the Muon (g-2) and Mu2e experimental halls.

Task 5.6 (“Outreach web site”) was fulfilled with getting the MUSE web site http://muse.lnf.infn.it/ online by May 2016 and collecting there information on the MUSE outreach events, publications, conference talks, posters and proceedings.

As an additional contribution to outreach, a Muon (g-2) Pisa collaborator has published an educational article on the Muong (g-2) experiment on the popularized Italian magazine Ithaca (http://ithaca.unisalento.it/nr-10_2017/), titled “La ricerca di nuova fisica nel vuoto quantistico”.

25

1.2.6 Work package 6: Transfer of knowledge

The objective of this work package is to coordinate all the activities dedicated to the training of research and industry personnel, to achieve a substantial transfer of knowledge among network participants and to increase the quality of the research and the competitiveness of the partners. Task 6.1 is shared between INFN, HZDR, LIVERPOOL, CAEN and PRISMA and is mainly related to the transfer of knowledge between institutions working on fundamental science and industry, Task 6.2 is more related to medical applications, while Task 6.3, coordinating trainings activities, involves all institutes.

Task 6.1, “Research-industry transfer of knowledge”, spans all the 48 months of the project. The first months of activity were mainly devoted to understanding the needs and expectations of the project partners. Both CAEN and PRISMA have their core business in designing and building electronic systems and were very interested in gaining experience in testing and validating electronics systems operated in harsh environments. On the other hand, scientific institutions are involved in the acceptance and quality assurance for both crystals and photosensors and were very interested on the QA techniques which are commonly used by industry. PRISMA seconded four engineers to INFN, one in Pisa and three in Frascati. Pisa researchers were designing a quite complex test station for the Mu2e calorimeter photo detectors mass production. The station measures all the photo detectors relevant parameters and is operated in vacuum, freezing the components under test up to -20 degrees Celsius. PRISMA engineer collaborated with the INFN researchers in the design of the electronic cards needed to operate the station. He was already experienced in the use of the microcontrollers, extensively used in the design, while the INFN Pisa researchers are expert in measuring the tiny analog parameters needed to qualify a photodetector. The completely assembled test station is shown in Fig. 18.

Figure 18: Test station for the QA tests of Mu2e photosensors.

The SiPM test station is currently fully operational and will be moved and operated in Fermilab to test and characterize the full production of about 4000 Mu2e SiPMs. This will be a crucial step before the calorimeter assembly because some parameters, relative to the single SiPM cell, can be measured only with the test station. This huge amount of sensors and measurements

26

implies strict QA methodologies which are more the field of expertise of the industrial partners than of the research personnel. To help defining these QA procedures PRISMA engineers were seconded at the INFN National Laboratory of Frascati where they collaborated with INFN researchers to define a relational database to store all the data related to the crystals and SiPM to be used in the experiment. The database will be updated regularly and has been designed, following Fermilab specifications, to be easily interfaced to the Mu2e offline software data analysis tools. Two PRISMA researchers were also involved in QA tests of the straws and in FEE test and the design of the QA for the Muon (g-2) tracker in Liverpool.

Another important activity included in Task 6.1 is related to qualifying electronic components and systems to be used in highly ionizing and non-ionizing radiation (neutron) fields. Research institutions working in the high energy physics have to deal frequently with these problems while industrial partners are very happy to gain experience in it. Only a few electronic components are already qualified as radiation tolerant by the producer and are extremely expensive and so presently the methodology, even in sensitive sectors like avionics or aerospace, is to qualify standard commercial components for use in extreme environments. Mu2e will be operated in vacuum, in a high magnetic field and in the presence of high levels of both ionizing radiation and neutron (non-ionizing) so INFN researchers qualified several electronics components, producing a list of “good to use” components. Some of the qualifications tests were performed at the ENEA facilities Calliope (Casaccia) and FNG (Frascati). PRISMA engineers participated in irradiation tests with thermal neutrons performed during their secondment in Frascati. Three irradiation campaign were performed at the ELBE facility at HZDR, involving both the local institutes and INFN, LIVERPOOL and UCL. Example of irradiation hardness results are reported in Figs. 7 and 19. Further tests will be performed in 2018. Deliverable D6.1, “Irradiation tests”, due on 30/11/2018 is expected to be completed on time.

Figure 19: Dark current as a function of the neutron flux for pre-production SiPMS of the Mu2e calorimeter. Performances of three different vendors (Hamamatsu (red), AdvanSiD (blue) and SensL (green)) are compared. The list of qualified components was shared between the MUSE participants, so that industrial partners could profit from it in their internal designs. Qualified components were used in the design of the Mu2e main acquisition card, called DIRAC. This is a 20 channels waveform digitizer, with sampling capability of 200 Msamples/sec at 12 bits. Waveform digitizers are the core business of CAEN who has several models in his catalogue, covering many combinations

27

of channels number and sampling rates, but it presently does not have a model qualified to be used in a harsh radiation environment. INFN Pisa researchers and engineers discussed deeply the design of the board with CAEN colleagues, merging the experience of both groups. The result is a prototype board that was successfully tested and qualified. The board, shown in Fig. 5 (right-bottom), includes fast data transmission capabilities that are custom designed for the Mu2e use but follows also a CAEN requirement to include a Gigabit Ethernet interface so that it could also be used commercially. Task 6.2, “Medical applications”, covers studies on SiPM applications for particle therapy. It is carried out by HZDR in collaboration with AdvanSiD and INFN, which have a high level of expertize on these photosensors. The design of detectors in this field is strongly constrained from their usage in medical therapies. Indeed, the working conditions are a significant neutron background, short measurement times and the presence of magnetic fields from the equipment. This requires a high bandwidth, background insensitive system design with fast, efficient and magnetic insensitive sensors. SiPMs are good candidates for their characteristics: they are small, inexpensive, efficient and fast. Therefore, a multi-channel device, reducing pileup, with good spatial and time resolution can be designed. The compatibility with the medical environment is assured by their insensitivity to magnetic fields. The only limitation is the limited radiation hardness. A first on going application is the optimization of the Prompt Gamma Slit Camera of Ion Beam Applications used in proton beam therapy. A crystal array prototype with SiPM readout has been tested to measure how deeply the protons penetrate the patient’s body in real time, detecting prompt gamma rays emitted when the primary proton beam interacts with tissues. This improved technique makes proton beam therapy more effective and less damaging to healthy tissues. Studies on In-Beam Positron Emission Tomography (PET) are also in progress. b+ emitters created by primary beam are used for range and dose verification. Very challenging background conditions are present compared to standard PET. The goal is to improve detector performances by changing the current technology, based on BGO crystals with PMT readout, to LYSO-SiPMs. In parallel, SiPMs irradiation tests are in progress. This is an R&D issue in common with the calorimeter Mu2e readout system. For this reason, a training session on latest generation SiPM was held at HZDR by a MUSE researcher from INFN, responsible of the Mu2e SiPMs R&D. Contacts with AdvanSiD are also crucial to understand the effects limiting radiation hardness, to mitigate them and to identify the most performing SiPMs for medical applications.

Task 6.3, “Training courses”, takes care of the organization of special training courses on advanced topics from research and industry during MUSE General Meetings. During the 2016 GM, a half day session for MUSE researchers was devoted to the following topics: “Writing for general public”, “Application of SiPMs in Positron Emission Tomography”, “Medical imaging processing using brain emulator”, “Neutron applications”. In 2017, the subjects of trainings were: “Laser systems and their applications”, “Novel proposal for a low emittance muon collider”, “Nanotechnologies and their applications”, “Frascati Scienza Experience: 12 years of data and results in outreach”.

1.2.7 Work package 7: Management

Work Package 7 coordinates and supervises activities across all the WPs. It ensures an efficient, transparent and productive organization of the project while supervising secondments and monitoring the scientific activities and the fulfilment of the deliverables. Several mailing lists were created to distribute information among participants, both for scientific and administrative

28

activities. Maximization of the knowledge sharing among the involved institutions, equal opportunity for all participants and visibility of the project are also key elements of this Work Package. These objectives are shared among four different tasks, whose deliverables and milestones due in this reporting period were all fulfilled.

Task 7.1, “Project supervision”, is a common effort of all institutes, coordinated by INFN. Two boards are in charge for running the project: the Management Board (MB) and the Scientific Board (SB). They have been appointed on January 2016, fulfilling the first milestone of the project (MS8, “Management structure in place”). The MB is constituted by one member of each main department, plus the coordinator acting as the chair. It is responsible for the revision and authorization of the secondments, the organization of General Meetings and the monitoring of the progresses towards the completion of milestones and deliverables. MB meetings took place 4 times per year as planned. The first two meetings were mainly devoted to inform MUSE partners about general rules and obligations of RISE projects. Procedures on secondment management within the consortium have been defined. Continuous revision and authorization of secondments guarantees steady progress towards the completion of the action. This is done by modifying the starting date of the planned secondments and distributing them to different researchers with the same role and an equivalent professional experience, when needed. The goal is to follow as much as possible the secondments planned for the different WPs along the project, by taking also into account variations in the schedule of the planned activities. Following these criteria, the overall fraction of completion for secondments exceeds 70%. This high percentage value demonstrates that the project is progressing well, although the delay in the implementation of some secondments has needed to be monitored.

The Management Board monitors the gender balance within the workforce and the working conditions, increases gender awareness and promotes women in science. Starting from a 15% of female with a permanent or temporary research position, the fraction of women in the MUSE management has increased to 24%. The fraction of female speakers at conferences, training events and tutoring activities is 25%. Participation in outreach events aiming to attract girls into STEM careers is being organized.

The SB coordinates the scientific activities of the network, and it is constituted by the conveners of the seven Work Packages. Each WP has two conveners, one from the “Lead Beneficiary” and the other one selected among the experienced researchers. SB meetings have been scheduled and have taken place every two months, as planned. Reports from the Work Package conveners monitor the progresses of tasks and the status of deliverables and milestones. The SB members are also in charge of Task 7.3, “Preparation of general and periodic reports”, which involves all institutes. Reports summarizing the status of tasks and deliverables are drafted each six months. List of trainings attended by MUSE personnel and of all dissemination and outreach activities are also included. Reports are made available to all participants through the MUSE web site. Deliverables connected to this task are D7.1, “First periodic report”, due on 31/12/2016, and D7.2, “Second periodic report”, due on 31/12/2018. The first one was submitted on time on 30/12/2016 while the second one will be drafted by the end of this year.

Task 7.2, “Organization of meetings”, takes care of the planning and organization of General Meetings, planned as one per year and included as deliverables of the project. The “First MUSE General Meeting”, deliverable D7.4, was held in Pisa, on 28-30 September 2016. More than 30 researchers from different institutions attended the meeting. The complete timetable of the meeting, with slides of presentations, is available at the following link: http://agenda.infn.it/event/muse2016. The first two days were devoted to the review of the activities connected to the project. Some specific talks of general interest were also included in

29

the agenda. A joint MB/SB meeting took place, mainly focused on an in-depth discussion on the running of the first months of the project. The last day was dedicated to a training session for MUSE researchers and outreach activities for University students, with two seminars on the experiments of the Muon Campus and two talks by two participants of the "Summer Students at Fermilab and other US Laboratories" program, describing their 9-week internship experience at FNAL. The “Second MUSE General Meeting”, deliverable D7.5, took place in Frascati, on 10-12 May 2017. About 40 researchers from different institutions attended the General Meeting in person. MUSE researchers from PRISMA and AdvanSiD could not come to Frascati for other concurrent commitments and connected remotely from their institutes. The same applied to some researchers from UCL, Liverpool and INFN which in those days were seconded to Fermilab for the commissioning of the Muon (g-2) detectors. The complete timetable of the meeting, with slides of presentations, is available at the following address: http://agenda.infn.it/event/muse2017. The meeting started with a public seminar on the status of the Mu2e experiment delivered by the Mu2e co-spokesperson, D. Glenzinski. The seminar was then followed by the joint MB-SB meeting. On May 11th, the General Meeting hosted the MUSE Mid-Term Meeting. The MUSE Project Officer, assisted by an external expert, reviewed the status and the progress of the project. The agenda included management issues, scientific achievements and the status of objectives, milestones and deliverables of the project. Activities connected with training, transfer of knowledge and networking were also addressed, together with a presentation of all dissemination and outreach activities carried on. The Project Officer had the possibility to interview several MUSE seconded personnel to get a feedback on their experience. The last day of the meeting, May 12th, was devoted to the training session and to the Young Scientists Forum, where MUSE young researchers presented their activities in the framework of the Mu2e and Muon (g-2) experiments.

The “MUSE web site”, Task 7.4 and deliverable D7.3, due on 31/05/2016, was established on 31/05/2016: http://muse.lnf.infn.it/. It includes a private section to share documents and information among participants and a wide section for the general public to maximize the visibility of the project. Besides general information on the project and its organization, dissemination events are being kept updated and outreach events highlighted. In the private section, minutes and slides of all meetings, including MB and SB ones, are available to all participants. Moreover, Grant Agreement, status of deliverables and milestones, biannual reports from the SB and useful tools (logo, acknowledgements, templates) are also included.

1.3 Impact

The Muon Campus is a unique world-leading facility providing the most intense pulsed-muon beams which will be exploited as one of the cleanest probes for new physics: both Muon (g-2) and Mu2e experiments will search for the evidence of new fundamental interactions beyond those predicted by the Standard Model. Fermilab is progressing on schedule on the construction of the infrastructures for the accelerator complex and for the experiments. Therefore, the impact of the project is unchanged since its beginning. The first Muon (g-2) physics run and the Mu2e commissioning are expected to be completed within the timeline of the project, as planned.

Technological challenges adopted to have state-of-the-art detectors, able to operate in the unique environment of the FNAL Muon Campus, find application in different fields. Remarkable examples are the development of electronics components able to survive in high radiation environments, which is an R&D common to space applications, and EMC compliant equipment that are used in medical instrumentation. Another field of interest for medical

30

imaging is the development of large-area Silicon photosensors able to work in the deep-UV region, opening the possibility of adding timing information to the standard PET imaging. 2. Update of the plan for exploitation and dissemination of result

The original plan for exploitation and dissemination of results covered the dissemination in the scientific community and the increase of training and skill development of researchers. It has been put in place, in line with Annex 1, as follows:

1. Results have been disseminated to the scientific community through the participations to international conferences/workshops (55 events) and the publication in professional journals (28 peer-reviewed publications and more than 30 contributions to conference proceedings, all in gold/green open access).

2. MUSE researchers participated to 38 training courses to increase their scientific and soft skills. Moreover, half-day of specific training sessions on advanced topics from research development or industry were organized during annual General Meetings.

3. The large number of secondments to Fermilab, equivalent to 130 person-months, fostered the opportunity both for young and senior MUSE researchers to come in contact with colleagues from more that 50 research institutes, in a high-quality training environment.

4. Personnel from industry and research institutes had the opportunity to acquire new competences and to improve their skills during secondments.

4. Follow-up of recommendations and comments from previous review

After the Mid-Term Review, recommendations and comments both for the mid-term period and for the future were received on December 18th 2017. They are reported below in italic, followed by the reply for each pointed inserted as normal text.

Recommendations concerning the mid-term period:

The careful management of the collaboration is reflected in the mid-term review (dated in December 2016) and the follow-up mid-term meeting in Frascati (May 2017), which I attended. I had a number of questions related to the review document (including the status of the Mu2e’s design of lasers and SiMPs, as well as the decision to move to CsI), and they were positively answered by the MUSE members present at the meeting.

My main recommendation concerning the period under review is related to perhaps the attitude towards the physics impact of the project. Both in the mid-term review and during the meeting, it seemed clear that the scientists in charge focus on enabling top performance for the experiments, on training and diversification of expertise, and on the dissemination of the results. All these activities will have a long-lasting impact in the experimental area. But a potentially huge impact of the activity carried by the MUSE collaboration could come from the significance of the discoveries ahead. Discoveries, when communicated properly, inspire the general public to appreciate scientific endeavors, and drive junior people towards degrees on STEM subjects.

31

MUSE organizes excellent outreach activities, but I wonder if it is ready to capitalize the stir that the results from g-2 will likely cause in the media. Have the Scientific and Management Boards discussed a contingency plan when (if) g-2 makes a discovery? is enough emphasis made on the measurement of the EDM of the muon, which would be the first measurement of this kind ever made? Theoretical biases should not prevent the researchers from pointing out the novelty of this achievement. Also, good communication of results may impact the next stages of g-2, for example by obtaining more funding for extended running and improvement of the experimental set-up.

Reply: Muon (g-2) presently has monthly meetings with the FNAL management and the communications team and there is a well-defined strategy in place for communicating a discovery (or even non-discovery) both for aµ and for the EDM. We are keeping them informed on the progress of the analysis and when we expect to announce results such that there is enough time to generate and coordinate the appropriate publicity in both the US and Europe. The EDM is very much part of this strategy as highlighted by the fact that at the recent EPS where there were two Muon (g-2) presentations (both by MUSE members) there was one on aµ and one on the EDM.

One also has to keep in mind that MUSE is a collaborative effort with a strong European component and European funding, whereas the main platforms of the g-2 and Mu2e experiments are in the USA. This geographical separation, and the fact that the host USA institution (FNAL) has a strong tradition in disseminating scientific results may overshadow the European contribution, unless MUSE takes this task on their own hands.

Reply: The UK (STFC) and Italian funding (INFN) clearly also require the European contributions to be recognized and highly visible. This has not been overlooked and the MUSE network will allow an added level of coordination on highlighting the results. MUSE members are well represented within the Muon (g-2) management to ensure the European contribution is fully recognized: MUSE members provide the chair of the Institutional board, the analysis conveners for the aµ and the EDM analyses and the chair of the publications committee and a member on the speakers’ committee.

Related to this point, I would like to note that the FNAL web presence of the g-2 experiment is rather disappointing, with barely any new content since 2015. I find this surprising for an experiment which will commence this summer and could already give interesting physics results next year. Also, I did not find a link between MUSE’s web portal and the g-2 or Mu2e

FNAL’s pages. Clearly there is room for improvement to showcase MUSE’s contribution to the experiments.

Reply: This is being updated in coordination with the Fermilab communications team such that on the timescale of a result being announced the web pages will have the appropriate content and we will seek to ensure that the MUSE contribution is also highlighted. A link to the MUSE website is now present on the Mu2e public home page.

Recommendations for the future:

During the period under review, the collaboration has developed an excellent track-record of delivery of objectives, and demonstrated a robust set of mechanisms to react to unexpected challenges. I expect the collaboration will continuing doing so during the next period, and would re-structure their activities if major changes occur, e.g. substantial changes in the funding situation of any of the partner institutions.

32

Regarding future work, I have few recommendations. First, I would like to re-emphasize the point made before regarding the impact of discoveries. Second, I would like to draw attention to the activities regarding the development of a European network of irradiation experts. I find this aspect of the proposal quite important, as establishing this network is a goal which transcends the MUSE project, and hence it has an impact beyond the science focus on the muon campus.

The development of this network is carried out in association with HZDR. MUSE members have been developing the use of the ELBE irradiation facility to advance the muon science. During the mid-term review, we learned of the interest from ELBE’s Scientific Advisory Committee in developing this new activity involving High Energy Physics and, generally speaking, detectors.

MUSE activities are beneficial to HZDR as they extend the breadth of ELBE's scientific scope, which could create technological spin-offs, and enlarge the international subscription to the facility, enhancing ELBE’s relevance in the scientific domain. I understand improvements in terms of access are being discussed, and I believe this common effort should be pushed forward. Moreover, ELBE with MUSE as a facilitator, could be the platform for a formal framework of the European irradiation network.

So far, the development of the network has been focused on developing the necessary expertise. But for this network to go beyond MUSE, one would need to provide a more comprehensive structure with ELBE as the focal point, for example by supporting an up-to-date public list of member experts, enabling specialized meetings, maintaining (or linking to) a database of results relevant to irradiation, and hosting training events. There could be many benefits of hosting such network: bringing together expertise from different domains tends to create new technological/scientific opportunities, and HZDR could seek additional public funding to support the European network after MUSE is finished.

Reply: The interest of the ELBE Scientific Committee on this new activity involving the High Energy Physics community is demonstrated by the additional beamtime assignment to our proposals for the first half of 2018. The usage of ELBE as the formal framework of the European irradiation network is being discussed with the coordinator of the facility. To start this process, we have proposed to organize ELBE user meetings and a database with publications related to irradiation.

Finally, I would like to remind the collaboration of the importance of reporting results as widely as possible, e.g. by placing in public repositories such as arXiV their results like the recent TDR report.

Reply: The Calorimeter “Technical Design Report” is now available on a public repository (arXiv:1802.06341).

5. Deviations from Annex 1 and Annex 2

Almost all the activities planned in the Description of the Action are evolving as expected and on track with respect to the schedule. Nevertheless, two specific tasks (1.2 and 6.2) did not progress as planned, due to reasons not directly connected to the project. To minimize the impact, in the first case partners have been involved in other activities while in the other one the work has been reorganized in order to complete, although with a reduced impact, the activity. Details are given in the next section.

33

5.1 Tasks

Two tasks from Work Packages 1 and 6 were not implemented as planned. Being independent activities, this had no impact on other tasks and on the overall planning of the project.

1. Task 1.2 – Development, commissioning and maintenance of SiPM bias voltage supply – WP1, “g-2 detectors” – Involved institutes: CAEN, INFN, FNAL

The goal of this task is the design and construction of a high precision Bias Voltage Supply for the SiPM of the Muon (g-2) electromagnetic calorimeter. Stability at the mV level, corresponding to a precision of 10-5 with respect to the set value, is requested in the entire 700 µs time window used for data acquisition, thus requiring a fast bias recovery for the power supplies. Two power supplies have been tested in Pisa (Italy) using a test board and a SiPM exposed to a laser beam. Although these supplies were demonstrated to work well for an average fixed current they did not meet the requirement of a fast bias recover for a rapidly changing current. This delayed the completion of the R&D phase and, consequently, the start of the related CAEN secondment to Fermilab for the commissioning of the system. The tight schedule of the experiment required a different solution for the BVS system, so that the contribution of the CAEN company within the project has been reorganized. A twofold collaboration with the INFN group in charge of the electronics of the Mu2e calorimeter, WP2, has been put in place. CAEN is currently developing a prototype for the high-current low voltage power supply of the calorimeter electronics. Secondments to Fermilab connected to this activity started in June 2017. Moreover, INFN and CAEN are collaborating to develop a waveform digitizer board able to survive in a high radiation environment.

2. Task 6.2 – Medical Applications – WP6, “Transfer of knowledge” – Involved institutes: HZDR, ADVANSID, INFN

The goal of this task is the transfer of the INFN and AdvanSiD expertize on Silicon photo-multipliers to the HZDR group working on medical applications, both in PET nuclear imaging and in-vivo-dosimetry at proton and ion beams.

The transfer of knowledge on silicon photosensors among INFN, HZDR and AdvanSiD started with a training session on latest generation SiPM, held at HZDR on 21 April 2016. The Mu2e researcher responsible of the photosensor R&D trained the HZDR colleagues involved in medical physics and detector technologies. Besides a general introduction on SiPM technology, the training provided a review of the studies performed by the Mu2e group on several types of SiPMs (UV extended, large active area, with different protection covers). The studies were based on cosmic rays and test beam data, on illumination with light pulsers and on the determination of the SiPM radiation hardness to dose and neutron irradiation tests. The development of custom SiPMs for the Mu2e experiment had also been discussed.

A delay in the collaborating activities between HZDR and AdvanSiD has been accumulated, due to the parental leave of two key researchers of the HZDR medical group. First secondments to AdvanSiD started on May 2017. Few months after, the Radiation Physics Division at HZDR was closed. MUSE researchers working there have been redistributed in other HZDR departments. Although they will be involved in other projects, they will continue their activities on medical applications of silicon photo-multipliers, in

34

collaboration with other Dresden research institutes. This will allow to complete the corresponding MUSE task and to produce the deliverable as planned. Nevertheless, the number of secondments between the two institutes will be significantly reduced.