The patents expired in 2012. One inventor died of cancer in 2014. The second inventor and the general who funded the program vanished in 2025–2026, inside a cluster of five unresolved disappearances at named US defense institutions. This is the documented record.
Five senior cleared defense personnel have vanished on foot, without a trace, in a ten-month window — roughly 10× the expected rate for a population this size. Two of them held irreplaceable knowledge of Mondaloy, a rocket-engine alloy the United States currently cannot reproduce.
An inventor. A qualifier. A funder. Between them they held every step of how Mondaloy moved from a laboratory composition to a flight-qualified rocket engine component.
High-performance rocket engines push hot oxygen through metal parts at extreme pressure. Hot oxygen makes most metals catch fire. Russia solved this in the 1960s. For decades the US military launched its own spy satellites on Russian-built RD-180 engines because America could not make an equivalent.
In the mid-1990s two metallurgists at Rockwell International invented a nickel-based superalloy that could survive that environment without igniting or cracking. They called it Mondaloy, a portmanteau of Monica Jacinto and Dallis Hardwick.
After Russia cut off RD-180 servicing in March 2022 over Ukraine sanctions, Mondaloy became the material standing between the United States and a gap in its ability to put national security assets into orbit.Documented
Mondaloy is used in roughly twelve components of the AR1 engine: preburner, turbine rotor, turbine housing, ducts, lines, and hot-gas manifold. Everything that touches oxygen-rich combustion. It can be cast, forged, or 3D-printed. Two variants exist, Mondaloy 100 and Mondaloy 200, optimised for different temperature and pressure ranges.
The AR1 program was an $804 million public-private partnership between the US Air Force, Aerojet Rocketdyne, and United Launch Alliance, designed to be the domestic replacement for the RD-180.
All three US patent applications for Mondaloy were filed between 2001 and 2009. All three were abandoned. The last expired in December 2012. After that date the composition was technically public from expired filings, but composition is not the recipe. The recipe is twenty years of process development — how to cast it, heat-treat it, machine it, print it, integrate it — and that lived in human heads.Inferred
Key events from the invention of Mondaloy through the current disappearance cluster — patents, corporate transfers, institutional milestones, and the five cases.
Four of the five disappearances fall inside a ninety-mile triangle anchored by Los Alamos, Taos County, and Albuquerque — three sites whose defense facilities share classified networks and the same material-science mission. The fifth is Reza, last seen in California.
| Name / Age | Date | Institution | Location | Behavioural Signature | Status |
|---|---|---|---|---|---|
Anthony Chavez |
May 4, 2025 | LANL | Los Alamos, NM | on footno phonewallet leftdaylight | Missing |
Monica Reza |
Jun 22, 2025 | Aerojet / JPL | Angeles NF, CA | on foot · hikingcompanion presentno trace | Missing |
Melissa Casias |
Jun 26, 2025 | LANL | Taos County, NM | on footphones leftphones factory-reset | Missing |
Steven Garcia |
Aug 28, 2025 | KCNSC | Albuquerque, NM | on footno phonewallet leftfirearm | Missing |
W. Neil McCasland |
Feb 27, 2026 | AFRL | Albuquerque, NM | on footno phonefirearmdaylighthiking gear | Missing |
A broader "ten dead/missing scientists" narrative is currently cirulating. This file uses a tighter filter: unresolved disappearance of cleared personnel at a named US defense institution, with the on-foot / no-phone / no-body signature. That filter excludes the following cases — not because they don't matter, but because they don't fit the pattern documented here.
| Name | Incident | In cluster? | Filter reason |
|---|---|---|---|
Suzanne Grillmair |
Shot, Oct 2025 | Out | Homicide with an identified suspect (Freddy Snyder). A known cause of death, not an unresolved disappearance. Included in the base-rate footnote for adjacent facility ties, not in the core cluster. |
Felipe Loureiro |
Murdered, 2025 | Out | Homicide with forensic evidence and a suspect pool. A different profile: a body, a scene, an investigation with a trajectory. The anomaly this file documents is disappearances without traces. |
Dallis Hardwick |
Died, Jan 2014 | Out of disappearance cluster; in Mondaloy triangle | Natural death from Stage IV breast cancer, confirmed by death certificate and hospice records. Appears in the triangle as one of the three who held the knowledge, not as a disappearance. |
Various scientists in circulating lists |
Various | Out | Includes deaths with known causes (accident, illness), professional profiles outside defense (academic biotech, unrelated industries), and incidents where the individual was later located alive. Including these inflates the denominator and weakens the signal. |
The Texas-sharpshooter critique (that the cluster was noticed first and the boundary was drawn around it) is a legitimate one. The response is not to widen the net but to define it publicly and hold to it: a specific profile (cleared, named institutions) and a specific signature (on foot, no phone, no trace, usually leaving a wallet behind). Under that filter the observed rate is roughly ten times expected at p = 0.0002. Widen the filter and the signal collapses into noise. Narrow it further and every case inside still fits.
A Poisson test against the expected rate of disappearances for senior cleared personnel at named US defense institutions, using the filter defined above.
A Poisson test asks a simple question. Given how often something usually happens, how surprising is it that we saw this many in this window? The output, a p-value, is the probability that a run this extreme or worse could have come from ordinary background rates alone. A p-value of 0.0002 means there is roughly a one-in-five-thousand chance of seeing five disappearances in ten months if nothing unusual is happening. That does not prove anything unusual is happening. It says the null hypothesis — "this is just the normal rate" — is a poor fit to the data.
The Mondaloy trade secrets have passed through seven corporate owners in thirty years. The 2026 transaction raises specific regulatory questions.
The L3Harris 10-K filed February 12, 2026 does not mention CFIUS — the Committee on Foreign Investment in the United States — anywhere in relation to the AE Industrial transaction. Neither does any other SEC filing between December 2025 and March 2026.Documented
CFIUS is the federal body that must review any foreign investment in critical American defense technology. Rocket propulsion alloys are regulated under ITAR. AE Industrial Partners is backed by unnamed sovereign wealth fund investors. Three possibilities exhaust the space:
One of these must be true. All three are findings.
Content discussing the Reza memorial or the phone evidence was removed across four platforms by four apparent actors. Deletions are not evidence of coordination. The pattern is on the record regardless.
Every question below is resolvable through public records, a FOIA request, or subpoena power. Status updates as responses come in. If you file one or receive a response, DM @tomnowa on 𝕏.