{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp09_archetype_breadth150_20260804","cell_id":"additive_measure_space_design__chemistry_materials","arm":"BREADTH_PROBE_ONE_SHOT","candidate_id":"additive_measure_space_design__chemistry_materials__P1","proposal_index":1,"version":0,"title":"Partition-Consistent Element Inventory for Battery Black-Mass Leaching","problem":"A battery-recycling development team compares staged leaching flowsheets by summing reported recovery of nickel, cobalt, lithium, and manganese across leachates, washes, recycled liquors, residues, and analytical aliquots. Those categories do not always form disjoint subsets of the original feed: a recycled liquor can appear at multiple stages, an intermediate stream and its downstream products can both enter a total, and unassayed residues or sampling losses can disappear from the denominator. Consequently, apparent elemental recovery depends on how the process is partitioned and can exceed the characterized feed inventory without an explicit reconciliation failure.","actors":["Hydrometallurgy process-development engineer","Analytical chemist","Laboratory technician maintaining sample identities","Process safety and environmental reviewer","Technical lead selecting the candidate flowsheet"],"observable_state":"For at least one element, recovery totals change when the same experiment is summarized by unit operation rather than terminal destination, or the sum of reported recovered, residual, and recorded-loss inventories falls outside the analytical uncertainty interval around the feed inventory. Worksheets contain ancestor-and-descendant streams, recycled material, below-detection results, or unassayed boundary material without distinct status labels.","consequence":"The team can select operating conditions on the basis of partition-dependent recovery, conceal an unclosed inventory as high yield, or overlook where regulated or valuable elements remain concentrated.","affected_objective":"Choose a staged leaching flowsheet using elemental recovery and loss figures that remain coherent under re-partitioning of the same experimental material.","intervention":"Create a nonnegative element-inventory measure for each target element before computing recovery. Declare the universe as all material descended from one uniquely identified black-mass feed charge through a bounded experiment. Admit subsets only when their material membership and lineage are traceable. Assign each measurable subset an inventory in millimoles of the target element, with the empty subset equal to zero. For reconciliation, partition the universe into mutually exclusive terminal destinations such as final product liquor, final wash liquor, terminal solid residue, retained analytical samples, documented spill or handling loss, and unresolved boundary material. Use a Partition Sum Table to test whether disjoint terminal inventories recombine to the feed inventory within a predeclared uncertainty rule. Keep recycled and intermediate streams available for process analysis but prohibit adding them to terminal recovery totals. Route unassayed, ambiguous-lineage, and below-detection cases to a Measurability Boundary Queue rather than silently assigning zero. Attach the resulting scope, units, partition, uncertainty treatment, and exclusions to every downstream recovery claim.","structural_mapping":[{"archetype_element":"Measurable Universe Scope","domain_realization":"All physical material descended from one identified black-mass feed charge during one bounded leaching experiment."},{"archetype_element":"Measurable Subset Family","domain_realization":"Traceable process-material subsets defined by unique sample identity, lineage, and nonoverlapping membership; unions of disjoint terminal samples are admissible."},{"archetype_element":"Empty-Set Zero Rule","domain_realization":"A destination receiving no material from the charge has zero elemental inventory, while a missing assay or below-detection observation is not treated as an empty destination."},{"archetype_element":"Nonnegative Size Assignment Rule","domain_realization":"Each subset receives a nonnegative elemental inventory in millimoles, derived from measured stream quantity and elemental assay with uncertainty retained."},{"archetype_element":"Disjoint Additivity Rule","domain_realization":"Inventories of mutually exclusive terminal destinations must sum to the inventory of their union; intermediate streams cannot be added alongside their descendants."},{"archetype_element":"Normalization or Scale Anchor","domain_realization":"Recovery is normalized only to the characterized elemental inventory of the uniquely identified feed charge, not to stage-specific or changing denominators."},{"archetype_element":"Null-Set and Negligibility Policy","domain_realization":"Below-detection assays, retained films, filter hold-up, and tiny spills remain labeled boundary or censored cases until a declared bound or assay supports their treatment."},{"archetype_element":"Partition Consistency Register","domain_realization":"A versioned table records terminal partitions and compares their recomposed inventory with the feed and with alternative valid groupings."},{"archetype_element":"Measurability Boundary Queue","domain_realization":"Samples with ambiguous lineage, mixed charge identity, missing mass, or unavailable assay are quarantined from recovery totals pending review."},{"archetype_element":"Integration and Downstream Use Contract","domain_realization":"Flowsheet comparisons may use recovery figures only when the associated element inventory states its universe, terminal partition, denominator, uncertainty rule, and unresolved mass."}],"mechanism_mapping":[{"mechanism_slug":"partition_sum_table","role":"Lists each mutually exclusive terminal destination and its elemental inventory, exposes gaps or double-counting, and tests recomposition against the feed inventory.","counterfactual_removal":"Without the table, individually plausible stage recoveries can still be summed across overlapping lineages, leaving partition-dependent totals undetected."},{"mechanism_slug":"measurable_family_closure_check","role":"Checks that combined reporting categories are unions of traceable, disjoint sample subsets and that regrouping does not introduce or discard material.","counterfactual_removal":"Without this check, labels such as total leachate can mix recycled, intermediate, and terminal material without a stable membership rule."},{"mechanism_slug":"null_set_policy_register","role":"Separates true absence from missing assays, censored measurements, retained hold-up, and unresolved losses.","counterfactual_removal":"Without the register, analytically inconvenient material can be converted to zero inventory and inflate apparent closure or recovery."}],"causal_chain":["Stage-based reporting admits overlapping streams and silently omitted boundary material.","Summing those categories violates disjoint additivity and makes elemental recovery depend on the chosen process partition.","A charge-specific universe and traceable measurable subsets establish which material may receive inventory.","A terminal-destination partition makes counted subsets mutually exclusive while preserving unresolved material as an explicit category.","The Partition Sum Table recomposes terminal inventories and compares them with the feed anchor under a declared uncertainty rule.","Only partition-consistent inventories feed recovery comparisons, so the technical lead can distinguish chemical separation performance from bookkeeping artifacts."],"baseline":"The current worksheet multiplies each sampled stream mass or volume by its assay and sums rows labeled recovered. It relies on operator judgment to omit recycled or intermediate streams, uses stage-specific denominators in some columns, and commonly records missing or below-detection assays as blank or zero. Closure is inspected only after recovery metrics have been calculated.","nearest_rivals":["A conventional elemental mass-balance spreadsheet that reconciles inputs and outputs but does not formally restrict which stream groupings are admissible.","A laboratory assay standard operating procedure that improves concentration measurements but does not prevent overlap among the material subsets being summed.","A process-flow simulation that tracks streams numerically but can reproduce double-counting if stream identity, terminality, and recovery-reporting boundaries are specified inconsistently.","An isotope-tracer study that can reveal material pathways but does not by itself define the permitted additive reporting partition."],"remaining_contrastive_claim":"The candidate's distinguishing commitment is that elemental recovery is not accepted merely because assays are accurate or a spreadsheet total is close to 100%; the counted categories must first be demonstrated to be measurable, mutually exclusive descendants of one declared feed universe, with intermediate streams, unresolved material, and below-detection cases governed explicitly.","authority_safety":{"decision_authority":"The hydrometallurgy technical lead may approve the accounting specification and decide whether its results are sufficient for an internal flowsheet comparison; the analytical lead controls assay interpretation and uncertainty treatment, while the safety reviewer retains authority over handling and disposal requirements.","authorized_first_step":"Re-express one already completed bench experiment using existing sample records and assay data; no new chemical processing is authorized.","excluded_actions":["Changing operating conditions or selecting a production flowsheet solely from the pilot accounting exercise","Reclassifying missing, below-detection, or ambiguous samples as zero without analytical-lead approval","Combining material from different feed charges to improve apparent closure","Discarding, reopening, or chemically testing retained samples","Using the inventory reconciliation to waive hazardous-material controls or regulatory characterization"],"halt_rollback":"Stop the pilot if unique sample lineage cannot be reconstructed, units cannot be reconciled, or a proposed terminal category contains both a stream and its descendant. Preserve the original worksheet unchanged, mark the reconstructed result non-decision-grade, and revert downstream comparisons to their prior status until the ambiguity is resolved."},"negative_tests":{"strongest_counterevidence":"A lineage audit shows that the existing reporting categories are already mutually exclusive terminal destinations with one stable feed denominator, and recomposed elemental inventories remain invariant across every valid regrouping within declared analytical uncertainty.","problem_falsifier":"For a preselected completed experiment, stage-based and terminal-destination summaries produce the same elemental inventory for every target element, no ancestor-and-descendant streams enter the same total, and all unassayed or lost material is already explicitly bounded.","intervention_falsifier":"After reconstructing traceable terminal subsets, the proposed measure still yields materially different totals under two logically equivalent disjoint partitions, or it cannot distinguish true zero inventory from unresolved assay and lineage states.","risks":["False confidence if analytical bias affects both feed and terminal assays while arithmetic closure appears satisfactory","Added sample-lineage work may delay experiment comparison","An overly restrictive measurable family may exclude informative but imperfect historical samples","Treating the feed assay as an unquestioned anchor may relocate rather than resolve uncertainty","Explicit unresolved inventory may be misread as evidence of a chemical loss mechanism","Small but safety-relevant residues could be labeled negligible if governance of boundary cases weakens"]},"next_evidence_step":"Using only records from one completed bench leach, select one target element and enumerate every recorded descendant sample from a single feed charge. Construct one terminal-destination Partition Sum Table, separately flag recycled streams, intermediate ancestors, missing assays, and ambiguous lineage, and compare the recomposed terminal inventory with both the original reported recovery and the characterized feed interval. The step ends with one reviewed table and a pass/fail judgment on partition invariance; it authorizes no new experiment or flowsheet decision.","prior_art_status":"UNSEARCHED","diversity_from_prior_proposals":"This is an isolated one-shot breadth-probe candidate; no other experiment candidates or proposals were inspected or used for comparison.","revision_record":{"parent_version":null,"progress_targets_addressed":[],"conceptual_changes":[],"operational_changes":[],"evidence_changes":[],"claim_changes":[]}}