Work

The Edge Zero: What a Single 248 keV Recoil Cannot Show

Year: 2026
Type: preprint
Abstract: In September 2026 the LZ collaboration reported one nuclear-recoil candidate at 248 keV in a search window widened from 55 to about 270 keV, with a global significance of 2.6 sigma and no discovery claimed. What decides the reading of that event is what lies under it: the absence of an excess at low recoil energy. Our series tags zeros by their reason (data, model, family-bounded, blind), a classification that began as a way of sorting the material of jokes between the authors. Here we tag the zeros around this event with a toy instrument built only from public numbers (Helm form factor, Standard Halo Model, approximate LZ efficiency) and three runs sealed with hashes posted publicly before execution (GitHub cds-polytope issue #3).The price tag: an elastic spin-independent signal that puts one event at 200-300 keV brings 665-1155 events at 5.4-55 keV under every halo tested; momentum weights up to q^6 lower this only to about 4.5.The edge zero: inelastic models that pay the price exist only in a narrow band of mass splittings just inside the kinematic edge of xenon, where low-energy recoils are exactly forbidden; the band moves with the escape speed of the halo. Near the edge the exposure another target needs follows an empirical scaling in its margin to the edge, log X = 0.023 - 2.355 log(g_T/g_Xe), tested on unseen target-model pairs; the asymptotic count grows as g^(5/2) (Appendix A).The cliff: just below xenon the rate is exactly zero whatever the exposure; from Ru-102 to Te-130 the share of such models falls from 1 to 0, and the law predicts the width of the cliff (median 4.6 mass units).Of 15 judged claims, 10 came out R, 4 U and 1 F; two forecasts about the next LZ release await their data. Checks after AI comments found a resolution limit of our own halo integral near the edge, recorded without changing any verdict. This is a study of reading, not of physics: it makes no claim about whether the event is dark matter. (1) Text: CC-BY-4.0 | Code: GNU Affero General Public License v3.0 or later (AGPL-3.0-or-later) (2) Data and code. The three seal documents with their SHA-256 files, frozen results, development and post-hoc logs,the forecast ledger, and all instrument, judging and figure scripts (AGPL-3.0-or-later) are inBridge_Paper_The_Edge_Zero_scripts_v1.0.0.zip. The seal and result hashes are posted atgithub.com/watabe-masanori/cds-polytope, issue #3. Only published numbers from the cited LZ material are used;no LZ data files. (3) AI use. Designs, code, analyses and text were produced by Kurado (Claude Opus 5.5, Anthropic)working with M. Watabe, who chose every direction and posted the hashes. Draft comments by ChatGPT andGemini, relayed by M. Watabe, prompted Section 7 and Appendix A and the rewording of Section 6; they are AIcommentary, not peer review.
Language: English
Open Access status: green
OpenAlex ID: W7233579417
Cites: 4
Cited by:
Related to: