The Ubatuba Magnesium: Visualisation · Signals from the Periphery

← Intro 中文

View

THE SPECTROMETER

Subject

—

Value

—

Verdict

—

THE UBATUBA MAGNESIUM
An unsigned letter, three lumps of metal, 61 years of laboratories
the only UFO artefact with a continuous analytical record
1957: "purer than Earth could make"
every error-barred ratio: terrestrial, on the line
chain to the original samples: completely lost
run the machine · read the line · price the claims

…

…

…

Evidence

fact ⇄ what kind of claim · live

Cleveland 2018: 79.28 / 9.94 / 10.85, residual 0.4‰, ON the terrestrial line

ICP-MS Services · the record's closing number, CIs published

MEASURED

Condon NAA 1968: Zn 500, Sr 500, Ba 160 ppm — 99.88% pure

National Office Laboratory · "claim of unusual purity ... disproved"

MEASURED

Dow triply sublimed, on the 1957 shelf: ~19 ppm total impurities

the fragment is ~66× dirtier than the metal anyone could order

MEASURED

1957: "absence of any other metallic element" — a detection limit, not a total

Barbosa & Texeira · vs a chempur reference · honest floor ~1,000 ppm

REPORTED

NOLAT 1968: ²⁶Mg "14.3 ± 0.7%" — omitted from Condon, found 2008

wrong sample weight; corrected 23.1%; reproduced by nobody, ever

REPORTED

Real ET magnesium exists: CAI ²⁶Mg* +13 to +100‰, OFF the line

Lee 1976 · Bradley 1978 · routinely measurable since 1976

MEASURED

Pure ²⁶Mg would weigh 1.858 g/cc; the Jolly balance said 1.866

computed here from the masses · AEC flotation 1.7513 ended it

MODELLED

Whose gram was tested: association with the original three "completely lost"

2 pieces never returned · 1 burned · 1 walked out of MIT · Sturrock 2001

READING

View

…

The source

Sector field0.2312 T

r = √(2mV/q)/B. Scan the field and the three masses sweep across the collector slits: that sweep is a mass spectrum. The slits sit where B₀ = 0.2312 T lands each isotope.

Spread aid ×20differences only, labelled on the chips

Labels

The ruler

A relic's Al/MgAl/Mg 35

Canonical ²⁶Al/²⁷Al × Al/Mg predicts the off-line excess a genuine early-solar-system relic would carry. The ghost marker moves; note how fast it leaves the terrestrial band.

The ledger

Readout · live

Field / peak

—

Beam radii

—

Separation

—

Residuals (2018 / 2017)

—

Ghost ²⁶Mg*

—

Impurities vs shelf

—

…

What is measured vs modelled here

Measured: the CIAAW masses and abundances, the Condon NAA impurity table in full, the Oak Ridge / Dow / Elemental purities, every published isotope analysis with its stated precision, and the CAI excesses. Reported: the 1957 Barbosa and Texeira statements (carried as detection-limit bounds), Fontes's 1 ppm sensitivity assertion, the NOLAT 14.3% (a failed measurement, drawn because burying it was the original sin), and the Jolly-balance density. Modelled: the sector geometry (representative, labelled), the fractionation line from exact masses, the ghost-marker arithmetic, the pure-²⁶Mg density. Reading: whose gram survived. The bench declines that one.

The model, in full

The sector solves r = √(2mV/q)/B per isotope at 10 kV; the slits sit at the B₀ landing radii and a 3 mm window decides "on peak". Delta space is δ = (R/R_std − 1)×1000 against 78.99/10.00/11.01; the mass-dependent band spans the kinetic (0.511) and equilibrium (0.521) laws computed from exact masses, and the off-line residual is δ²⁶ − δ²⁵/β. The ghost marker is δ²⁶* = (²⁶Al/²⁷Al)₀ × (Al/Mg) / R₂₆ × 1000 at the canonical 5.23×10⁻⁵. Purity bars are published impurity totals on a log axis; the 1957 claim is drawn as a bound at its detection limit, not as a value. All anchors in scripts/ubatuba-tune.mjs.

Where this drawing takes liberties

The spectrometer is a REPRESENTATIVE mid-century magnetic sector: the laboratories in this file used several different instruments (arc spectrographs, TIMS, SIMS, ICP-MS), and the historic Caltech machine's exact radius and voltage are unpublished, so this bench draws one honest geometry and labels it. Beam brightness maps abundance nonlinearly for visibility; the ion sparks are decoration. The Elemental Research point carries no published uncertainties and is drawn hollow; Caltech 1976 released a bound, not abundances, so it is drawn at the origin with its stated precision. The NOLAT 1968 abundance is renormalised assuming a normal 24/25 split, which the lab never published; the file says so.

…1 / 20

…

…

…