O-42 - Fifteen years of searches found no stable strange matter - terrestrial isotope, astrophysical, and AGS-SPS accelerator searches all null
The paper draws two distinct inferences from these nulls: (a) the terrestrial ultraheavy-isotope nulls further reduce the likelihood that strange matter is stable in bulk at zero pressure (Blackman-Jaffe); (b) the accelerator nulls, while not by themselves decisive for RHIC (insufficient luminosity to bound negative-strangelet production there), anchor the production-model fits (coalescence penalty factors) used to argue production is hopeless. A stable light strangelet would masquerade chemically as an ultraheavy isotope of a normal element (e.g. A ≈ 100 with Z = 7 behaving as 100-N nitrogen), which is what the matter searches exploit.
Link to original
Why this is evidence
H-28 (stable bulk strange matter) makes positives plausible on all three fronts: stable strangelets seeded in terrestrial matter would show up as ultraheavy isotopes of normal elements, stable strange matter would permit quark stars with astrophysical signatures (e.g. sub-millisecond pulsars), and heavy-ion collisions could yield detectable strangelets. H-39 predicts every one of these searches comes up null. The uniform nulls are therefore more probable under H-39 — though not decisively, since H-28 is compatible with the nulls if formation or delivery mechanisms are absent (a stable ground state nobody has occasion to reach).
Likelihood
# E-2 (HC-7) — lone edge, one observation O-42 "fifteen years of null strange-matter searches: terrestrial
# ultraheavy-isotope, astrophysical (no sub-ms pulsar/quark star), and AGS-SPS accelerator, all null".
# Members in HC-7.hypotheses order [H-28 stable in bulk, H-39 not stable]; exhaustive, no residual.
# Anchored on H-39 = 1 (rule 7): the null is essentially certain if strange matter is not stable, so it is
# the clean unit; H-28 priced as "how expected is this uniform null if bulk strange matter WERE stable".
# NOTE the no-observation arguments A-56, A-65 are already in HC-7's prior — not re-used here. This edge
# rests on the terrestrial + astrophysical absences (distinct from E-3's RHIC/STAR collision null), so the
# discrimination is leaned on those fronts, not on the collision-production front E-3 carries (rule: don't
# re-use the same underlying absence twice).
lik_strange_H39 = 1.0 # anchor: if strange matter is not the ground state (H-39), every one of these
# searches is predicted null — essentially certain.
lik_strange_H28 = 0.4 # 0.4× vs anchor: the sharpest discriminator is the terrestrial ultraheavy-isotope
# null — under H-28 a stable light strangelet seeded in terrestrial matter would
# masquerade as an anomalously heavy isotope (Blackman-Jaffe), so the strong limits
# are somewhat surprising; the astrophysical null (no quark-star/sub-ms-pulsar
# signature) adds. Not driven low: H-28 survives if no natural formation/delivery
# path exists (a ground state nobody reaches). A-58 (approved) bounds accelerator
# yield ~1e-46/collision, so the AGS-SPS portion of the null is expected even under
# H-28 — that front carries little weight, which is why the ratio stays well above 0.
t_strange = 0.8 # cap = trust_score of O-42's source S-3 = 0.85. Docked slightly: the observation is
# a review's compilation of many heterogeneous searches (terrestrial, astrophysical,
# accelerator) whose individual sensitivity/limit claims were not independently
# audited here — an aggregation step below the reliability of a single dataset.
evidence("HC-7", ["O-42"], [lik_strange_H28, lik_strange_H39], t=t_strange)