The DHCAL – An overview José Repond Argonne National Laboratory Presented by Katja Krüger Deutsches Elektronen Synchrotron CALICE Meeting Santander, Spain June 1, 2016 Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

The DHCAL Description 54 active layers Resistive Plate Chambers with 1 x 1 cm2 pads → ~500,000 readout channels Main stack and tail catcher (TCMT) Electronic readout 1 – bit (digital) Digitization embedded into calorimeter Tests at FNAL with Iron absorber in 2010 – 2011 without absorber plates 2011 Tests at CERN with Tungsten absorber 2012 1st time in calorimetry

Current situation Funding from DOE-HEP completely stopped on September 30, 2014 Directed to cease any activity immediately

The Electron – Ion Colliders EICs eRHIC 250 GeV/nucleon protons/ions 1.3 – 21.2 GeV electrons √sep= 36 – 146 GeV Luminosity ~ 4 x 1033 cm-2s-1 JLEIC 8 - 100 GeV/nucleon protons/ions 3 – 12 GeV electrons √sep= 10 – 69 GeV Luminosity ~ 5.6 x 1033 cm-2s-1 For comparison HERA √sep= 318 GeV Luminosity (peak) ~ 1032 cm-2s-1

The Electron – Ion Collider or EIC Nuclear Science Advisory Committee (NSAC) Long Range Plan In this context, submission of proposals for support for DHCAL studies A) Laboratory Directed Research and Development (LDRD) → as part of larger proposal to develop a detector concept for the EIC B) EIC detector R&D → specifically for DHCAL studies Recommendation III “We recommend a high-energy high-luminosity polarized EIC as the highest priority for new facility construction following the completion of FRIB.” Initiative for Detector and Accelerator R&D “We recommend vigorous detector and accelerator R&D in support of the neutrinoless double beta decay program and the EIC.”

Proposed studies for the EIC I Long-term tests of 1-glass RPCs Novel RPC design: readout pads in gas volume Advantages: thinner, pad multiplicity constant and around 1 (easy to calibrate!), higher rate capability Design validated: JINST 10 (2015) P05003 Chemical stability of anode plane in gas volume? Long-term tests with cosmic rays needed Long-tem tests of fast RPCs Use of semi-conductive glass as resistive plates Rate capability improved by 2 orders of magnitude: JINST 10 (2015) P10037 Chemical stability of glass plates over time, with prolonged periods of HV? Long-term tests with cosmic rays needed

Proposed studies for the EIC II 3. Development of a gas recycling system Typical RPC avalanche gas harmful for the environment (green house gas) Challenge: capture gas without changing the gas pressure in the chambers Designed system which captures exhaust gas: ‘Zero Pressure Containment’ system Assembled system and tested successfully Implementation of filtering system and test with RPCs 4. Development of a High Voltage distribution system An RPC based HCAL needs approximately 3,000 chambers with individual HV Requires a HV distribution system Each channel: option to turn on/off, monitor current and voltage, adjust voltage to within ±100 V First prototype designed, built (by University of Iowa) and tested successfully 1 channel, no monitoring, no voltage adjustment yet Implement all features and test

The Min-DHCAL Special tests Fermilab test beam Analysis Use of DHCAL layers (0.29 X0) No absorber plates between layers 50 layers in total Fermilab test beam Mixture of e+, μ+, π+ Momentum range of 1 – 10 GeV Data with extremely fine spatial segmentation Analysis Positron data published Pion data being analyzed Analysis performed by Benjamin Freund (McGill) MC samples by Burak Bilki (Iowa) 8 GeV e+ shower in the Min-DHCAL

Test Beam and Data Samples Min-DHCAL placed in secondary beam at Fermilab’s FTBF Secondary beam mixture of electrons, muons, pions in 1 – 66 GeV/c momentum range Particles arrive every minute in 4 s spills Cerenkov counter to tag electrons Data collected in November 2011 Momentum range limited to 1 – 10 GeV/c (due to shallow depth of Min-DHCAL)

Analysis of Min-DHCAL Events Quality cuts Reject multiple particles, cosmic rays, fake triggers, out of time events… Equalization of the response of RPCs Use through-going muons to equalize response of RPC i (εi = efficiency, μi = average pad multiplicity of RPC i) MC Simulation of the set-up GEANT4 and digitizer (RPC_sim) Digitizer tuned to muon and e+ data Complete study of experimental systematic errors Equalization (dominant), rate limitation, contamination from other particles (negligible), noise hits (negligible) Systematic errors of the simulation No meaningful systematic error for e+ simulation (used in the tuning process) Systematic error of π+ simulation calculated from the deviations between data and simulation observed with e+

Selection of Min-DHCAL Positron Events Entries = percentage of surviving events Data Momentum [GeV/c] 1 2 3 4 6 8 10 Timing cuts 99.9 99.8 99.95 99.96 Requirements on first layer 88.5 87.0 80.3 88.1 86.6 88.2 At least 6 active layers 86.4 80.0 79.8 88.0 86.5 Čerenkov signal 60.3 31.7 40.0 30.7 53.9 41.7 33.0 Simulation 100.0 98.3 97.9 97.6 97.2 97.1 96.8 Cleaning cuts PID provided by Cerenkov (not simulated) Only 2 – 3 % of events rejected

Min-DHCAL Positron Results I Published in JINST (2016) 11 P05008 Main messages from paper Default simulation of electromagnetic interaction not adequate Need detailed simulation of electromagnetic interactions (Option 3 or _EMY) High precision data → Well reproduced by simulation Response in number of hits Non-linear (saturation) Fit to power law m = 0.76 ± 0.02 (m = 1 would be perfectly linear) Invert power law to reconstruct energies

Min-DHCAL Positron Results II Energy spectra Fit to Gaussian within ± 2σ No evidence of contamination from muons/pions Response Linear by construction Resolution Well reproduced by _EMY Significantly worse with default simulation Fit to standard formula c [%] α[%] Data 6.3±0.2 14.3±0.4 FTFP_BERT_EMY 6.2±0.1 13.2±0.2

Min-DHCAL Positron Results III Precision shower shapes Longitudinal shape Fit to Gamma distribution Small leakage observed Radial shower shape Covering 6 orders of magnitude in hit density Hit density Distribution of number of neighbors to a given hit Some differences between data and simulation

Selection of Min-DHCAL Pion Events Cleaning cuts Positrons rejected by Cerenkov (not simulated) Muons rejected by requesting interaction Rejects positrons and minimizes leakage Final efficiency between 8 and 18% Positrons in data of course rejected by Cerenkov (not simulated) Muons efficiently rejected

GEANT4 Physics Lists Hadronic Shower Models Physics list combine various hadronic shower models Models cover different energy ranges Linear transition between ranges

Min-DHCAL Pion Results I These results are not yet preliminary Distributions of number of hits Not well reproduced by simulation Fit to the Novosibirsk function (works well for data, not so much for simulation) Response in number of hits In average more hits (~15%) in simulation Discontinuities for some physics lists Very linear response Fit to power law m = 1.00 ± 0.04 (m = 1 is perfectly linear) Invert power law to reconstruct energies

Min-DHCAL Pion Results II Energy spectra Distinct differences for different physics lists Simulated spectra different from data

Min-DHCAL Pion Results III Radial shower shape High precision data Results based on different physics lists differ from each other None of the simulations agree with the data

Min-DHCAL Pion Results IV Longitudinal shower shape High precision data Results based on different physics lists more or less agree with each other None of the simulations agree with the data Simulated showers deeper into the calorimeter (remember the longitudinal shower shapes for positrons were well simulated!) Notice the agreement between data and simulation before the shower start

Min-DHCAL Pion Results V Distribution of density of hits High precision data Results based on different physics lists more or less agree with each other None of the simulations agree with the data (but not too bad)

The Fe-DHCAL Data set Collected at Fermilab Covering 2 – 60 GeV/c About 13M events Both main stack and TCMT with RPCs Data analysis Performed by Coralie Neubüser (Hamburg) Tuning digitizer Different operating conditions from Min-DHCAL 6 Parameters to tune Spread of charge on anode plane: σ1, σ2, and relative weight R Scale of charge distribution: Q0 Threshold for hit definition: T Radius of inefficiency for additional avalanches: dcut

Tuning of Fe-DHCAL Digitizer Data sets 10 GeV muons 10 and 20 GeV positrons Tuning Quite good agreement already Further improvements to come

Fe-DHCAL Positron Results Distribution of hits Nice agreement at low energies Increasing difference at higher energies Response in number of hits Very non-linear (as expected) Simulation significantly fewer hits at high energies Further improvements to the simulation to come

Summary Overall situation Not easy to make strident progress without support from the funding agencies Possibility to revive DHCAL effort with support from Nuclear Physics (EIC) Min-DHCAL positron paper High precision measurement of electromagnetic showers Results published in JINST (2016) 11 P05008 Min-DHCAL pion analysis Analysis well advanced Paper draft in preparation Fe-DHCAL analysis Tuning of digitizer almost complete Results soon