{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp09_archetype_breadth150_20260804","cell_id":"bulkhead_isolation__chemistry_materials","arm":"BREADTH_PROBE_ONE_SHOT","candidate_id":"bulkhead_isolation__chemistry_materials__P1","proposal_index":1,"version":0,"title":"Utility-Bulkheaded Parallel Polymerization Screening","problem":"In a parallel polymer-formulation screening station, independently controlled reaction vessels share one recirculating cooling loop and one vent header. An unexpectedly rapid exotherm in one vessel can consume shared cooling capacity, warm the common coolant return, and drive vapor or pressure into the shared vent path, perturbing otherwise healthy experiments.","actors":["polymer and materials researchers","laboratory process-safety officer","parallel-reactor operator","reaction vessels and their formulations","cooling and vent-service maintainers"],"observable_state":"During a single-vessel upset, its temperature and vent flow rise while the shared coolant-return temperature or header pressure also changes; neighboring vessels then show temperature deviations, pressure deviations, or detectable tracer carryover despite having normal local control settings.","consequence":"A local formulation upset can invalidate neighboring experiments, obscure which material recipe caused each observed result, and expose multiple vessels to a common thermal or pressure disturbance.","affected_objective":"Preserve valid, independently controlled reaction trajectories and minimum viable temperature and pressure control in unaffected vessels during a local upset.","intervention":"Divide the array into small reaction pods with physically separate, capacity-bounded coolant reservoirs and pumps, thermal barriers between pods, and pod-specific one-way vent paths terminating in separate compatible capture vessels. Permit only monitored data and command signals across pod boundaries. Provide each pod with local isolation, cooldown, inspection, and restart procedures so an upset pod can be taken offline without stopping healthy pods.","structural_mapping":[{"archetype_element":"Coupled system vulnerable to propagation","domain_realization":"Parallel reaction vessels coupled through a common coolant inventory, pump capacity, return temperature, and vent header."},{"archetype_element":"Identifiable failure domain","domain_realization":"A reaction pod containing a deliberately limited number of vessels, its own coolant circuit, and its own vent capture path."},{"archetype_element":"Resource partitioning","domain_realization":"Cooling inventory, pumping capacity, and vent-handling volume are allocated per pod rather than drawn from one unrestricted pool."},{"archetype_element":"Containment boundary","domain_realization":"Thermal barriers, non-communicating coolant circuits, and one-way pod-specific vents block heat, fluid, and vapor propagation."},{"archetype_element":"Selective coupling","domain_realization":"Only monitored control data and authorized operating commands cross pod boundaries; process fluids, coolant returns, and vent streams do not."},{"archetype_element":"Minimum viable function","domain_realization":"Healthy pods retain local circulation, sensing, shutdown capability, and controlled reaction conditions when another pod is isolated."},{"archetype_element":"Local recoverability","domain_realization":"The affected pod can be cooled, disconnected, inspected, cleaned, and restarted without opening or rebuilding the entire array."}],"mechanism_mapping":[{"mechanism_slug":"resource_partitioning","role":"Bounds how much cooling and vent-handling capacity one upset can consume and prevents it from altering the service state of every vessel.","counterfactual_removal":"If all pods still draw from one unrestricted coolant reservoir or vent volume, the main exhaustion and propagation paths remain global."},{"mechanism_slug":"fault_domain_isolation","role":"Makes each pod an independently observable and isolatable thermal-hydraulic failure domain.","counterfactual_removal":"Without independently isolatable pods, operators must treat the entire array as affected by a single-vessel upset."},{"mechanism_slug":"selective_coupling","role":"Preserves necessary supervision while excluding process-fluid and thermal cross-flow between pods.","counterfactual_removal":"Either uncontrolled physical coupling reappears or total separation prevents coordinated monitoring and emergency response."}],"causal_chain":["A formulation in one vessel enters an unexpectedly fast exothermic reaction.","That vessel demands cooling and vent capacity beyond its normal allocation.","In the shared baseline, the upset changes common coolant or vent conditions and thereby perturbs neighboring vessels.","Separate coolant inventories, pumps, barriers, and vent paths force the disturbance to encounter an enforced pod boundary.","The affected pod can exhaust only its bounded local capacity and can be isolated through an explicit interface.","Unaffected pods retain their own service capacity and continue under monitored control.","Operators cool, inspect, and recover the affected pod locally before authorized reintegration."],"baseline":"A parallel array with per-vessel sensors and controllers but a single recirculating bath, common coolant return, and common vent header; an array-wide emergency stop is available, but service capacity and propagation paths are not partitioned.","nearest_rivals":["Increase the capacity of the shared chiller and vent header: this adds headroom but preserves a single failure domain and shared-state pathway.","Add a high-temperature shutdown interlock: this can stop heating or feeds after detection but does not structurally reserve cooling capacity or prevent stored heat and vapor from perturbing neighbors.","Run formulations sequentially: this avoids simultaneous cross-effects but abandons unaffected parallel operation instead of containing failure while preserving it.","Add more spacing or insulation between vessels: this reduces direct heat transfer but leaves shared coolant and vent coupling intact."],"remaining_contrastive_claim":"The candidate's distinguishing claim is structural: separately bounded cooling and vent-service domains, joined only by explicit monitored interfaces, should contain a simulated single-pod disturbance while healthy pods remain controllable. It is not merely added capacity, earlier shutdown, physical modularity, or duplicate backup equipment.","authority_safety":{"decision_authority":"The laboratory principal investigator and designated process-safety officer jointly authorize any test; qualified facilities or pressure-systems personnel authorize changes to cooling, vent, or pressure-rated hardware.","authorized_first_step":"Construct and test only a nonreactive benchtop mock-up using electrically heated dummy vessels, a compatible water-glycol coolant, and a harmless detectable tracer, all within documented equipment ratings.","excluded_actions":["Introducing energetic, toxic, pressurized, or polymerizing reagents during the first test","Intentionally overpressurizing any vessel, line, header, or capture container","Bypassing existing temperature, pressure, flow, ventilation, or emergency-stop safeguards","Modifying pressure-rated vessels, relief devices, or building exhaust without qualified engineering review","Routing incompatible vent streams into a shared or unverified capture medium","Continuing a test after loss of sensing, circulation, containment, or operator supervision"],"halt_rollback":"Stop heater input and tracer release, use the existing safe shutdown, and drain or disconnect the mock-up only after it reaches its predefined safe state if any rating is approached, leakage occurs, a sensor fails, coolant circulation is lost, tracer escapes containment, or an unaffected compartment leaves its allowed operating band. Revert to the unchanged shared mock-loop configuration for diagnosis; do not advance to live chemistry without a separate hazard review and authorization."},"negative_tests":{"strongest_counterevidence":"The proposed boundaries would be undermined if propagation is dominated by an unavoidable shared dependency—such as room ventilation, electrical supply, control hardware, structural conduction, or a mandatory common relief path—or if the separated pods cannot individually sustain safe temperature control.","problem_falsifier":"Across predefined credible dummy-vessel disturbances, the shared-loop baseline shows no measurable neighbor temperature, pressure, flow, or tracer response beyond sensor uncertainty and normal control variation.","intervention_falsifier":"Under the same disturbance, the partitioned mock-up does not reduce cross-boundary temperature, pressure, flow, or tracer transmission relative to baseline, or an unaffected pod loses minimum viable circulation, sensing, or temperature control.","risks":["False isolation through an overlooked shared return, relief path, power source, controller, ventilation path, or conductive frame","Insufficient local coolant or vent capacity causing severe degradation inside the upset pod","Capacity stranding while a neighboring pod needs unused cooling capacity","One-way valves or isolation hardware introducing blockage, maintenance burden, or new pressure hazards","Thermal barriers impairing normal heat removal or concealing hot spots","Separate capture vessels being mismatched to later chemical compatibility requirements","Operators bypassing boundaries for convenience and restoring uncontrolled coupling","Additional interfaces increasing leak points, calibration burden, and procedural complexity"]},"next_evidence_step":"Run one bounded comparison on a two-pod, four-dummy-vessel benchtop rig: first configure a shared coolant return and shared tracer header, then configure separate pod coolant reservoirs, pumps, barriers, and capture paths. Apply the same predefined electrical heat pulse and harmless tracer release to one dummy vessel while logging temperatures, flows, pressures, and tracer signals in every pod. Confirm equipment ratings and stop thresholds beforehand; use the result only to test propagation and controllability, not to infer performance with reactive chemistry.","prior_art_status":"UNSEARCHED","diversity_from_prior_proposals":"One-shot isolated proposal; no other experiment candidates or prior proposals were inspected or used for comparison.","revision_record":{"parent_version":null,"progress_targets_addressed":[],"conceptual_changes":[],"operational_changes":[],"evidence_changes":[],"claim_changes":[]}}