Secondary Emission Calorimetry

In high-radiation environments, electromagnetic calorimetry is particularly challenging. To address this, a feasible approach involves constructing a sampling calorimeter that employs radiation-hard active media, albeit at the expense of high energy resolution. In response, we developed an innovative technique, secondary emission calorimetry, which offers radiation resistance, rapid response, robustness, and cost-effectiveness. Our efforts involve the creation of prototype secondary emission sensors, subjected to comprehensive testing within test beams. In the secondary emission detector module, incident charged hadrons or electromagnetic shower particles trigger the generation of secondary emission electrons from a cathode. These generated electrons are subsequently amplified in a manner similar to the process within photomultiplier tubes. This report provides an insight into the principles underlying secondary emission calorimetry, presents findings from beam tests, and outlines Monte Carlo simulations that project towards the potential application of large-scale secondary emission electromagnetic calorimeters.


I. INTRODUCTION
T HE future collider experiments impose unprecedented radiation conditions for the calorimeter systems, particularly in the forward region.The calorimeters envisaged for these operating conditions must be sufficiently radiation-hard and robust in order to perform as expected for the entire lifetime of the experiments.The most feasible option is a sampling calorimeter with radiation-hard active media and a common absorber material such as iron or tungsten.For such implementations, we developed secondary emission (SE) calorimetry and investigated its performance in the beam tests with prototype secondary emission modules.
In this report, we detail the foundational principles of secondary emission calorimetry and present the outcomes derived from beam tests conducted on a dedicated secondary emission module built on these fundamental principles.Additionally, we offer insights into Monte Carlo simulations concerning the potential implementation of large-scale secondary emission electromagnetic calorimeters.

II. DESCRIPTION OF SECONDARY EMISSION
CALORIMETRY A secondary emission calorimeter consists of SE sensor modules in a sampling calorimeter arrangement or as a homogeneous calorimeter consisting entirely of SE sensors as the absorbers.In an SE sensor module, SE electrons are generated from a surface which works as a cathode and are then multiplied in a dynode chain.The sensors can be placed between absorber materials (Fe, Cu, Pb, W etc.) in a sampling calorimeter setup [1], [2].
An SE cathode takes the form of a thin film, resembling the dynodes seen in photomultiplier tubes (PMTs), which are typically composed of simple metal-oxide materials such as Al 2 O 3 , MgO, CuO/BeO, or other materials with higher secondary electron emission yields.These materials are recognized for their exceptional radiation resistance, proven by their use in PMTs enduring doses of up to 50 Grad and in accelerator beam monitors subjected to particle fluxes exceeding 10 20 mip/cm 2 (refer to sources such as [3] or [4]).In terms of SE electron generation, the SE sensor cathode is analogous to the photocathode of a PMT.The SE electrons generated in the top SE surface by the traversing shower particles and the those produced at the dynodes, are similar to photoelectrons.The electron multiplication chain can be metal-meshes or other planar dynode structures.Fig. 1 shows a picture of a Hamamatsu PMT with a mesh dynode structure, the sketch of the electron multiplication in the dynode chain, and a microscope image of the mesh structure.The secondary electron yield exhibits a robust dependence on momentum of the impinging particle.This dependence follows dE/dx similar to the Sternglass formula [5].Consequently, substantial SE electron yields can be achieved for particles with low energy that are produced in hadronic showers.As an example, yields from a MIP on robust materials like alumina or titania films are only 1.1-1.2,requiring many dynodes for a MIP signal.On the other hand, as the shower is fully absorbed, those yields rise to as much as 50 for low energy large dE/dx shower particles.
The statistical characteristics of photoelectrons and SE electrons exhibit similarities [6].In a scintillation calorimeter, numerous photons are produced per GeV, but typically only a fraction of approximately 1% to 0.1% are detected and converted into photoelectrons.In the case of an SE calorimeter, a relatively few number of SE electrons is generated from the shower particles as they traverse the cathode/dynodes.However, the downstream dynodes amplify essentially all of these SE electrons.
Compared to the PMTs, the construction of the SE sensors offers many simplifications.The entire assembly of the SE sensors can be done in air and there are no critically controlled thin film vacuum depositions.Bake-out and vacuum sealing can be performed under dramatically relaxed conditions.
The SE sensor modules envisaged are exceptionally radiation damage resistant, compact, high gain, high speed, rugged, and cost effective, and can be fabricated in arbitrary tileable shapes.

III. CONSTRUCTION AND TESTING OF SECONDARY EMISSION SENSOR MODULE PROTOTYPE
A feasible and cost-effective SE sensor module prototype can be constructed with an array of PMTs after making electronics modifications so that the photoelectron generation is disabled and only the dynode chain of the PMTs is used.Utilizing seven Hamamatsu single-anode R7761 PMTs, the initial prototype of the SE sensor module underwent comprehensive testing at the Fermilab Test Beam Facility [8], using electron beams with energies of 8 and 16 GeV.The characterization of the PMTs associated with this first SE sensor module iteration is detailed in [7].Fig. 2 shows a picture of the SE sensor module prototype.The prototype for the SE sensor module is constructed to accommodate the seven SE sensors in a densely-packed arrangement.The custom design baseboards provided three different modes of operation: • Normal divider mode: The voltage divider chain for the photomultiplier remains unaltered, adhering to the original design provided by Hamamatsu.• Cathode-first dynode shorted mode: The cathode and the first dynode are kept at the identical potential so that the multiplication probability for the photoelectrons is minimal.
• Floating cathode mode: The cathode is isolated from the remainder of the divider chain and can be independently powered using a separate high voltage source.Additionally, the option to apply a reverse bias to the cathode has been confirmed not to result in any discernible operational differences when compared to the cathode-first dynode shorted mode.Following the laboratory tests, the cathode-first dynode shorted mode was selected as the default operation mode for beam tests.In order to mimic a sampling calorimeter setup, absorber plates made of steel and tungsten were positioned upstream of the SE module, with varying thicknesses, to assess the progression of particle showers.Employing steel absorbers measuring 20 cm × 20 cm × 1.9 cm, data from all seven SE detectors were recorded.Conversely, when 3 cm × 3 cm × 0.35 cm tungsten absorbers were utilized, only the central module was involved in data acquisition.
The extent of lateral coverage provided by the SE module did not yield a shower signal that demonstrated consistent scaling with shower depth when steel absorbers were used.Consequently, the tests conducted with tungsten absorbers were established as the baseline.
Fig. 3 illustrates the module's response to 8 GeV (top) and 16 GeV (bottom) positrons when employing the tungsten absorbers.The size of the trigger counters was chosen in compliance with the sensor size and the event selection was performed based on the position of the positrons as measured by the upstream wire chambers.Hence, precise electromagnetic shower profiles are generated.The measurements (depicted in black) are further corroborated through comparison with Monte Carlo (MC) simulations (shown in red).Fig. 3 confirms the viability of secondary emission sensors that incorporate dynode chains.
In order to investigate the effect of different multiplication geometries, we repeated the electromagnetic shower sampling test with the 16 anode square Hamamatsu R5900-00-M16 PMT which has the same photocathode and window materials as the 7761 PMT, but metal channel dynode structure as the multiplication chain.A baseboard to combine the signals of the 16 anodes and to apply the cathode-first dynode shorted condition was designed and produced in order to make an SE sensor.Prior to the beam tests, the response uniformity across the anodes was measured in the laboratory as being less than 3% [7].In addition, the gain of the particular SE detector that was used in the electromagnetic shower sampling was measured as 2.3 (0.4) ×10 6 .Fig. 4 shows the electromagnetic shower shapes for 16 GeV (red squares) and 8 GeV (black circles) positrons measured with the square-shaped SE sensor with metal channel dynodes and with tungsten absorbers.A combined fit to the shower shapes was performed with the function A where X is the shower depth in units of radiation lengths, X U 0 is the amount of the upstream material in units of radiation lengths, a and c are the shape parameters and A is the scaling parameter.The showers shapes can be represented with the simultaneous fit with the prediction

IV. ENHANCEMENT OF SECONDARY ELECTRON EMISSION
The electron multiplication in the SE cathode and dynodes can be enhanced by applying thin film coatings of high secondary electron emission yield materials such as Al 2 O 3 , SnO 2 , TiO 2 or ZrO 2 .Fig. 5 illustrates the simulated efficiencies corresponding to varying thicknesses of Al 2 O 3 , SnO 2 , TiO 2 , and ZrO 2 .Optimal performance is achieved with a 100 nm thickness of Al 2 O 3 .The secondary electron emission efficiency spans from 2% to 5% across all coatings with secondary electron emission capability.The simulations also predict a secondary electron yield around 68 for 100-nm Al 2 O 3 -coated copper foil.

V. ANTICIPATED PERFORMANCE OF A LARGE-SCALE SECONDARY EMISSION CALORIMETER
A simulation study was conducted for an SE calorimeter system on a large scale.The SE sensor modules were modeled as 9-stage dynode chains.The dynodes are separated by a distance of 150 µm and contain holes with diameters ranging from 10 µm to 100 µm, spaced at intervals of 50 µm to 100 µm.For the events with minimum ionizing particles producing an SE electron at the cathode, the average charge per SE sensor module was obtained as 300 fC.We simulated the electromagnetic response of a sampling calorimeter comprising of 16 SE modules, with 1 X 0 tungsten absorbers interleaved, using electron energies ranging from 1 GeV to 32 GeV, the typical energies at FTBF.Within this energy range, the detector response exhibits linearity, and the electromagnetic energy resolution is determined as (16.7%)/ √ E, where the constant term is negligible.The simulations predict sufficient calorimetric performance for a radiation-resistant secondary emission electromagnetic calorimeter.Fig. 6 depicts the linearity of the response (top) and the energy resolution (bottom) of the secondary emission calorimeter.

VI. CONCLUSIONS
Calorimetry in high radiation environments is particularly challenging.There are also other implementations such as Compton polarimeters and beam loss monitors which pose similar challenges in terms of instrumentation.Secondary emission calorimetry offers a feasible solution for these challenging conditions with its radiation-hard sensor modules, high speed, high performance and cost effectiveness.
The secondary emission sensors resemble the dynode chains of photomultiplier tubes.Therefore, the first principles can be obtained by appropriate modifications on the powering and readout of the photomultiplier tubes.By deactivating the photocathodes, the first secondary emission sensor prototypes were built and tested.The preliminary findings validate the concept and propose the development of a complete secondary emission calorimeter prototype.Simulations of a 16stage tungsten-secondary emission calorimeter indicate a good response linearity and an energy resolution of (16.7%)/ √ E, projected for electron energies up to 32 GeV.Special surface coatings of high secondary electron emission yield materials on the dynodes would enhance the overall gain and sensitivity of the secondary emission modules.With the flexibility of practically any anode geometry, secondary emission calorimetry also provides options of highly segmented readout for imaging calorimetry.

Fig. 1 .
Fig. 1.Picture of a Hamamatsu PMT with a mesh dynode structure, the sketch of the electron multiplication in the dynode chain, and a microscope image of the mesh structure

Fig. 2 .
Fig. 2.An image of the first SE sensor module is presented.Each individual sensor possessed a diameter of 39 mm, with an active window diameter measuring 27 mm.The sensors were 50 mm in length.

Fig. 4 .
Fig. 4. Electromagnetic shower shapes for 16 GeV (red squares) and 8 GeV (black circles) positrons measured with the metal channel SE sensor.

Fig. 6 .
Fig. 6.The predictions of the response linearity (top) and the energy resolution (bottom) of the 16-layer SE calorimeter prototype.