Speaker
Description
The future Electron–Ion Collider (EIC) will offer a unique opportunity to explore parton distributions
inside nucleons and nuclei thanks to unprecedented luminosity, a wide range of energies,
a large choice of nuclear species, and polarization of both beams (in the case of light hadrons).
The electron–Proton–Ion Collider (ePIC) detector will enable, among other performance goals, precise determination of the positions
of primary and secondary vertices, which is essential, for example, for the identification of charm hadrons
with typical decay lengths of the order of 100 microns. This capability, achieved via topological selections,
provides access to the gluon distribution inside nucleons.
This measurement performance is ensured by the Silicon Vertex Tracker (SVT), the innermost detector of the ePIC experiment.
The SVT Inner Barrel (IB) adopts 65 nm MAPS (Monolithic Active Pixel Sensor) technology with stitching, pioneered by the ALICE
collaboration for the ITS3 upgrade, resulting in the MOSAIX sensor. This technology is employed to build three layers
of wafer-scale sensors, bent into a cylindrical shape thanks to their reduced thickness.
The low power density of the MOSAIX sensor allows cooling by airflow, which represents a significant advantage
in reducing the material contribution of service components within the detector and, consequently,
limiting multiple scattering that degrades tracking and vertexing resolution.
The stringent space and material budget constraints require careful design of the support structures and cooling system components,
which must ensure uniform temperature distribution over the sensor surface and adequate redundancy.
An additional challenge is posed by the presence of the Left End Cap (LEC), hosting the front-end electronics
for power distribution and data transmission, which introduces a localized region of higher power density at one edge of the sensor.
A dedicated test campaign was carried out using dummy heat loads in a closed setup, reproducing the MOSAIX power dissipation,
including the LEC contribution. The setup allows tunable airflow and power consumption and includes multiple
temperature sensors placed at critical locations along the sensor. The main goal of the tests was the optimization
of airflow and heat removal from regions characterized by different power densities.
A possible solution based on 3D-printed aluminum heat sinks produced via additive manufacturing was investigated.
Key design aspects of the dissipators include minimization of the material budget—targeting an impact comparable
to or lower than that of carbon foam, currently the baseline solution—and control of the airflow, which must transition
to a turbulent regime shortly after the air outlet and ensure uniform mass flow over the sensor surface.
The effect of thermal grease on the thermal coupling of the dissipators was also studied.
Several shapes of air distributors and aluminum heat sinks were tested, and first conclusions on their effectiveness were drawn.
The current status of the cooling system design and initial results from tests
performed on thermal mock-ups of silicon sensors are presented.
| Speaker confirmation | No |
|---|