Choose timezone
Your profile timezone:
R&D activities
→
Europe/Zurich
32/1-A24 (CERN)
Gas R&D activities
Gas R&D activities 2026-09-03
Key Outcomes
The team reviewed process simulation results for the Atlas distillation column, confirming 96% validation accuracy and identifying 2 bar as the optimal operating pressure. Physical commissioning is near-ready, pending fluid verification.
uRWELL Detector alignment work improved efficiency from ~74% to ~79–80%.
A quantum chemistry study (ORCA/DFT) characterized electron attachment behavior across candidate gas mixtures.
Distillation Simulation Findings
· Validation: Vapor-liquid equilibrium simulation validated against experimental literature data at 96% accuracy.
· Optimal pressure: Operating at 2 bar yields >99% product purity and highest recovery (~86%).
· CO2 feed trade-off:
· Higher CO2 (up to ~30%) improves R134A recovery by reducing losses in the top exhaust stream.
· Above 30% CO2, product purity drops and condenser temperature decreases, widening the top-to-bottom temperature gradient and making column control harder.
· Feed flow: Tested at 300 L/hr baseline; varying to 100–200 L/hr showed minimal effect on performance.
· Next simulation step: Dynamic simulation in development to model control loops, valve behavior, and fault scenarios (e.g., low carryover pressure, loss of feed) for potential integration into the ROG schema.
Physical Commissioning Status
ATLAS R134a distillation
· Humidity analyzer installed; flexible pipes connected to chiller.
· Pressure regulator was incorrect size — tanks not yet connected; target: switch-on next week.
· No refrigerant currently in the Atlas distillation circuit — to be verified with Fabio this afternoon.
LHCb C4F10 distillation
· C4F10 system: Filled to 1.6 bar; CO2 introduction procedure completed; recovered C4F10 from detector emptying with ~300 mbar reserved for distillation testing.
· Reflux liquid issue: No liquid observed in reflux, likely because sufficient pressure/temperature not yet reached for C4F10 condensation; further testing planned.
HF/HCl Sensor Testing
· Lab baseline: No HF/HCl signal in standard mixture; small HF signal appeared with CO2 mixture.
· GIF results: HF peaks correlated with dose and voltage scans; HCl showed a falling background with CO2.
· CO2 interference concern: Sensor datasheet certifies no CO2 interference only up to 10%; behavior at 100% CO2 is unverified.
· Temperature dependency suspected: Tests at 28°C (gas room) vs. 17–18°C (lab) may explain signal differences; temperature effect on electrochemical cell reaction rate is a key open question.
· Chlorine/fluorine cross-interference: Chlorine gives false signal on HF sensor; fluorine interference on HCl sensor not yet confirmed.
uRWELL Detector Alignment & Efficiency
· Efficiency gain: Millepede II alignment improved efficiency from 74–75% to 79–80%.
· Tracker residual shape anomaly: Trackers 2 and 3 (different micro-MEGAS type) do not show expected Gaussian residual shape — suspected geometric effect; under investigation.
· DOT alignment issue: Post-alignment chi-squared values are poor; orientation output shows implausible 24.85° Y-axis shift — likely fit underestimation due to noise.
· Suggested fix: Subtract noise from signal before fitting, or use Millepede directly; try ROOT for easier data handling.
RPC Detector Frame & Strip Profile Issue
· Detector 32 (frame C): Very low efficiency and abnormal waveforms; strips mounted on thin mylar + copper instead of standard polystyrene backing — changes strip impedance from nominal 550 Ω.
· Detectors 33 & 35: Comparable efficiency to February baseline; minor issues suspected in 33 (possible solder joint).
· Planned fix: Swap strips from EcoVast chamber (currently only used for current scans) into frame C on Monday to confirm impedance as root cause.
Quantum Chemistry Study (ORCA/DFT)
· SF6 is the only candidate that binds an electron without bond dissociation, releasing ~1.09 eV — consistent with its known electron-trapping quenching mechanism.
· HFOs (YD, ZD) and R134A require bond breaking to interact with electrons; C–Cl bonds are weaker than C–F bonds, making HFOs easier to dissociate.
· C4F7N behaves similarly to SF6 (electron binding without breaking) but has high GWP (~3,000–4,000) and faces EU restrictions.
· CO2 substitution implication: CO2 has fewer vibrational modes than R134A; replacing R134A with CO2 reduces quenching capacity, raising questions about where excess electron energy is dissipated.
· Limitation acknowledged: Simulation covers isolated two-body interactions only; does not model detector operating conditions, electric field effects, or three-body collisions.
Action Items
· Team (today): Check Atlas circuit for existing refrigerant with Fabio; verify chiller operation.
· Emmanuel Ryan + team (tomorrow if available): Fill distillation column and observe liquid formation.
· Simulation team: Test 1.5 bar operating point; assess whether a compressor upstream is needed.
· Bjorn: Subtract noise from detector fit; retry alignment with cleaner signal.
· Max: Swap strip profile in frame C; confirm impedance fix for detector 32.
· Albin: Add CF3I and C4F10 molecules to the simulation table.
· Max: Finalize plots for ECOGAS contribution from May test beam and ZD scan results.
There are minutes attached to this event.
Show them.