The element that arrived fourteen years late

Three things that are already built

Before any of the argument, here is the equipment. Every one of these runs today, and each one is the closest working relative of something in the account that follows.

The account, as told

Robert Scott Lazar, born in Coral Gables, Florida in 1959, first appeared on KLAS-TV in Las Vegas in May 1989, in silhouette, under the name Dennis, interviewed by the investigative reporter George Knapp. He came back unmasked and under his own name that November. A transcript of a KLAS-TV interview carries the date 9 December 1989, and a radio interview on KVEG in Las Vegas is dated 28 December 1989. Two later recordings fill the account out: a video known as The Lazar Tape from mid-1991, and a seminar given at Rachel, Nevada on 1 May 1993.

He described a facility he called S4, dug into the side of a mountain roughly 10 to 15 miles south of Groom Lake and about 125 miles north of Las Vegas, with hangar doors angled and faced to match the slope. Nine discs, he said. He worked on one of them, which he called the Sport Model: about 52 feet 9 inches across, which is 16.1 metres, roughly the frontage of a suburban block, and about 16 feet tall. The skin he described as a metal like unpolished stainless steel.

The part that concerns matter is the reactor. He described a plate about 18 inches across with a sphere on top, a chip of element 115 in the tower, and three gravity amplifiers at the base of the craft that took a wave from the reactor and pointed it. On radio in December 1989 he corrected his own wording: it was not a gravity generator, he said, it was a gravity amplifier.

He said the element could not be synthesised, that it would have to be made in nature, perhaps on the fringes of a supernova. He said the element itself is not antimatter but has the property of producing it. In The Lazar Tape he gave its melting point as 1740 degrees Celsius. Each craft, he said, carried 223 grams of it, good for twenty to thirty years, drawn from a government stockpile of about 500 pounds.

On his own background he claimed master's degrees from MIT in physics and from Caltech in electronics. Neither institution holds a record of them. He attended Pierce Junior College in Los Angeles, and a 1982 piece in the Los Alamos Monitor places him at the Los Alamos Meson Physics Facility as a contractor technician. He said he was hired through the contractor EG&G and employed by the US Navy; EG&G has stated it holds no records on him.

Papoose Lake is a real dry lake bed inside the Nevada Test and Training Range. Commercial satellite imagery of it has been public since the 1990s, and declassified U-2 and CORONA photographs cover the ground back to the 1950s and 1960s. No hangar complex appears on the western Papoose slope in any of them.

The claim, written out as a spec

Strip the story back to numbers and you get a parts list. An engineer handed this sheet would start pricing it immediately, because every line on it is checkable.

Four of those lines can be settled with a bench and a stopwatch: is it stable, can it be made, what is its melting point, and how long does 223 grams last. Two can be settled with arithmetic alone, no laboratory needed: the energy density and the reactor's energy balance. The rest are questions about gravity, and gravity has its own price list.

Moscovium, as it is

Element 115 was made on purpose. From 14 July to 10 August 2003, calcium-48 ions at 248 million electronvolts were fired into a target of americium-243 at Dubna. Three atoms of moscovium-288 appeared, and one of moscovium-287. The result was published in Physical Review C on 2 February 2004.

The rate tells you what kind of work this is. A cross-section is a way of saying how small a target a nucleus presents to an incoming beam, and at three picobarns, a measure of how tiny that target is, the fusion that made moscovium-288 was about as likely as it gets down there. Four weeks of continuous firing returned four atoms. Nothing accumulates. Each atom is born, decays and is gone before the next one arrives.

Dirk Rudolph of Lund University and colleagues at the GSI centre in Darmstadt repeated the work and published in Physical Review Letters on 27 August 2013, with more than thirty decay chains and an X-ray fingerprint of the daughter nucleus. The joint working party of the international chemistry and physics unions, IUPAC and IUPAP, accepted the discovery on 30 December 2015. The name moscovium, after the Moscow region where Dubna sits, was proposed on 8 June 2016, opened for five months of public comment, and approved on 28 November 2016.

Every moscovium isotope known

An isotope is the same element with a different number of neutrons in the nucleus. Five of them have been seen. All five decay by alpha emission, which means the nucleus spits out a small clump of two protons and two neutrons and becomes nihonium, element 113.

Those spreads are worth a moment. A half-life is the time it takes half a sample to decay, and normally you measure it on a sample. Here there is no sample. Each figure comes from timing a handful of individual decay chains, which is why moscovium-290 is quoted as 650 milliseconds with the true value somewhere between about 450 and about 1140. The uncertainty is not sloppiness; it is what counting a dozen atoms gives you.

Put the quantities side by side. A 223 gram charge of moscovium-290 would hold about 4.6 × 10²³ atoms: roughly sixty thousand times the number of grains of sand commonly estimated for every beach and desert on Earth. Around a hundred atoms of moscovium have been made in total. The ratio between those two numbers is about 4.6 × 10²¹, which is close to three and a half times the number of litres of water in all the world's oceans.

The island of stability

Here is the idea in plain words. A nucleus is a crowd of protons and neutrons, and they do not just pile in: they fill up shells, the way seats in a stadium fill row by row. When a shell fills exactly, the nucleus is bound a little tighter and lasts a little longer. The counts where that happens have been read off decades of measurements: 2, 8, 20, 28, 50, 82 and 126.

Lead-208 is the showpiece. Eighty-two protons and 126 neutrons, both shells exactly full, and the result simply sits there for as long as anyone cares to watch. It is the heaviest nucleus nobody has yet caught decaying.

Past lead, every nucleus measured so far decays. Bismuth-209 sat on the stable list for a century until a French team measured it in 2003 at about 2 × 10¹⁹ years, roughly a billion times longer than the universe has been running. That is what stability looks like up the heavy end. Push further and it gets harder fast, because the electrical push between protons grows with every proton added while the glue that holds nucleons together only reaches as far as the next-door neighbour. In the liquid-drop picture, which treats the nucleus as a charged droplet and carries no shell structure at all, the barrier against splitting runs out near a hundred protons. That anything above element 102 exists is what the shells add back.

So the question became obvious: where is the next full shell? For neutrons the answer models keep returning is 184. For protons the answer is not settled: calculations put it at 114, or at 120, or at 126, and they have been arguing about it since the late 1960s. William Myers and Wladyslaw Swiatecki appear to have named the region the island of stability around then, and Glenn Seaborg took up the phrase and carried it. Richard Swinne had guessed at long-lived nuclei up around element 108 as early as 1914, and John Archibald Wheeler proposed superheavy nuclei in 1955.

The island is offshore. Everything made so far sits on the mainland, in the shallows near the coast, and the water in between has not been crossed.

Drive the crosshair up to 184 neutrons and the squares stop. There is nothing there. The chart holds what people have made, and the offshore glow holds what the calculations say should be there if the shells close where they are expected to.

How long would a nucleus on the island last? The answers have swung with the decades. Flerovium-298, with 114 protons and 184 neutrons, is the classic candidate. A 1966 estimate put it above a hundred million years. A 1972 estimate gave it about a year. Current estimates land somewhere over ten days. Those are calculations, not measurements, and the spread between them is the point: the island's height above the water is genuinely unknown.

Against that, look at what has been measured. The longest-lived nucleus above element 103 that anyone has timed is dubnium-268 at around 16 hours. Rutherfordium-267 runs about 48 minutes; copernicium-285 about 28 seconds. Nothing measured has reached even the modest end of the island predictions, and the reason is not effort.

Why 184 neutrons is out of reach

The gap is not vagueness. It is arithmetic you can do on the back of an envelope, and it comes from the ingredients.

Americium-243 brings 148 neutrons. Calcium-48 brings 28. Fuse them and the merged nucleus starts with 176 neutrons, and then it is far too hot to keep them: it boils off three or four before it settles, landing at 172 or 173. That is exactly where the moscovium isotopes sit on the chart. To land at moscovium-299, with 184 neutrons, the merged nucleus would have to start with at least 187, which is eleven to fifteen more neutrons than the best pair of ingredients supplies.

The obvious fix is a heavier target. Californium, element 98, is the heaviest actinide a target can practically be made from. Einsteinium and fermium exist in microgram quantities against the ten milligrams or so a target needs, and the longest-lived einsteinium isotope lasts 470 days, which is not long enough to accumulate that much. Californium plus calcium-48 reaches elements 114 to 118 with neutron counts of 172 to 177. That is the ceiling of the current road.

Across every superheavy element, the most neutron-rich nuclei confirmed are livermorium-293 and tennessine-294, both at 177 neutrons: seven short. Moscovium-290 is nine short.

What a stable 115 would need

Stable in the chemical sense means what a lump of copper means: it sits in a drawer, you can machine it, and it is the same thing next year. For element 115 that would take three things to fail at once, and each of them is a different escape route.

Set that specification against bismuth-209 and its 2 × 10¹⁹ years, because that is what a nearly stable heavy element actually measures. Even the most optimistic island prediction, the hundred million years floated in 1966, sits eleven orders of magnitude below bismuth, that is eleven steps of ten times smaller, and today's ten-day estimates sit more than twenty.

There is one more line in the account worth separating out. Moscovium was predicted long before it was made to behave as a heavy relative of bismuth: a soft metal in the same column of the table as nitrogen and phosphorus. The shell model's variables are proton count, neutron count and binding energy, which is how tightly the glue holds. There is no field term in it to switch on, so an emitted field would have to be a new term rather than a consequence of the count. The element's number sets its chemistry, not its gravity.

See moscovium on the table, alongside the neighbours it was predicted from.

Twenty to thirty years, or seconds

Take the account's own quantities and run them against the measured half-life. A 223 gram charge of moscovium-290, decaying at about 10 million electronvolts per alpha, would put out about 790 gigawatts in its first instant. That is roughly twenty-five times everything Australia's power stations deliver on average, or better than three hundred million kettles at once, a dozen for every person in the country.

And then it would fall off a cliff. Drag the clock.

The comparison bar is the point. Plutonium-238 is the workhorse fuel in the generators aboard Voyager, Cassini and the Perseverance rover: a half-life of 87.7 years and about 0.54 watts per gram. The same 223 grams of it gives roughly 120 watts, which is two old-style light bulbs, and it does that for decades. That is the shape of a long-duration nuclear fuel. A long half-life buys you steady low power. It cannot buy you high power for a long time, because the same atoms cannot be spent twice.

Run the account's stockpile through the same arithmetic. 227 kilograms of moscovium-290, decaying down its full alpha chain at around 55 million electronvolts per atom, carries about 4.2 × 10¹⁵ joules: close to a megatonne, about a fiftieth of the largest device ever set off, and it would deliver essentially all of it inside half a minute.

The account says 223 grams lasts twenty to thirty years. The measured half-life makes it twenty to thirty seconds.

The energy figure, checked with nothing but arithmetic

One line of the spec needs no laboratory at all. The account gives 2.2 pounds, which is one kilogram, as the energy equivalent of 47 ten-megatonne hydrogen bombs. That is 470 megatonnes of TNT, or about 2.0 × 10¹⁸ joules.

The ceiling on what any kilogram of anything can give up is its own mass, converted completely: one kilogram times the speed of light squared, which is 9.0 × 10¹⁶ joules, or 21.5 megatonnes. That ceiling rests on one premise: the fuel is a closed system, so the energy it can give up is bounded by its own rest mass, and relativity fixes the exchange rate at the speed of light squared. Anything above 21.5 megatonnes per kilogram would need energy arriving from somewhere other than the fuel, and the account names no such source.

Megatonnes of TNT, same scale

The 1989 figure sits 21.9 times above the complete conversion of the fuel. Credit the reactor with annihilating a matching kilogram of ordinary matter as well, which is the most generous reading available, and it still sits 10.9 times above. For scale, the largest device ever detonated, Tsar Bomba on 30 October 1961, released about 50 megatonnes: that is the full mass-energy of about 2.3 kilograms, and it took a device the size of a bus to release it.

Run the account's own charge through the same books. 223 grams of fuel annihilating with 223 grams of ordinary matter gives 4.0 × 10¹⁶ joules, which is 9.6 megatonnes, not 470.

The antimatter question

The reactor in the account runs on antimatter, and antimatter is real. It is made every day. The interesting part is how much, and what it costs.

CERN's Antimatter Factory delivers about 400 million antiprotons an hour. That sounds enormous until you convert it. CERN's own figure is that every antiproton the factory makes in a full year of non-stop running would amount to about 500 joules, which would light a 100 watt bulb for five seconds, or lift a ten kilogram bag of spuds onto a five metre shelf. Whole antihydrogen atoms are rarer still: the ALPHA collaboration builds up to about 3,000 an hour, and CERN puts the maximum imaginable in a year, trapped and annihilated in one go, at a few thousandths of a joule, which they compare to the energy of tapping a phone screen.

Across CERN's whole history the total made is under 10 nanograms. That is less than a hundredth of a speck of dust you could see, and annihilating all of it would run a 60 watt bulb for about four hours.

At that rate, the account's single 223 gram charge would take on the order of a trillion years to produce, which is about seventy times as long as the universe has been going. Price is the same story from another angle. NASA put antihydrogen at USD 62.5 trillion a gram in 1999. Gerald Smith at Penn State estimated in 2006 that USD 250 million could produce 10 milligrams of positrons, which works out around USD 25 billion a gram, or about A$35 billion at 71.7 US cents to the Australian dollar, 28 August 2026.

Antimatter still buys something real. It is the densest energy store known, roughly a billion times a chemical fuel by mass, and that is why it keeps coming up in propulsion sketches. The step that has not been found is a way of making it that does not cost more than it returns.

What gravity actually is

In the account, gravity is a wave that the reactor emits and three amplifiers point. In general relativity what feeds gravity, the source term in the equations, is energy and momentum, and the field is the curvature of spacetime itself rather than a substance travelling through it. It is the shape of space, bent by everything with energy in it, and things moving through the bend look to us like things being pulled. There is nothing separate to radiate, so an amplifier would have to work on the curvature, and Einstein's equations name the exchange rate exactly.

That changes the question completely. Amplification needs a quantity to feed into a gain stage, the part of an amplifier that does the lifting. In this picture that quantity is curvature, and curvature is sourced by energy and momentum, so the question turns into how much energy the craft carries and in what volume. To bend spacetime to a radius of one metre, meaning a curve that tight, takes an energy density of about 4.8 × 10⁴² joules in every cubic metre.

That number needs a picture, so here is one. The Sun radiates about 3.8 × 10²⁶ watts. To fill a single cubic metre with 4.8 × 10⁴² joules you would need everything the Sun puts out for around four hundred million years, held in a box a metre on a side.

And here is the scale that makes it land. The Earth, all 6 × 10²⁴ kilograms of it, bends spacetime at its own surface to a curvature radius of about 2.4 × 10¹¹ metres. That is around 1.6 times the distance from here to the Sun. The whole planet, the thing that holds your feet down all day, is barely bending anything at all.

None of this reads the proton number. What Einstein's equations take in is energy and momentum, and nothing else. Iron, hydrogen and element 115 with the same energy in the same volume bend space identically, because the equations have no place to put the element's name.

Where the account puts gravity, and where the objection lands

In The Lazar Tape the account splits gravity in two: Gravity A working at nuclear scale, Gravity B at the everyday scale, with Gravity A identified as what physics calls the strong nuclear force. Morgan's critique goes straight at that. The strong force reaches about a millionth of a billionth of a metre and stops; gravity has no such limit, and the two differ by roughly 20 orders of magnitude in range and 38 in strength. Calling one the other is renaming, not joining. He adds a second objection with a picture in it: a bend in spacetime strong enough to move a craft would not be able to choose what it moved, so everything nearby would go with it.

Morgan grants something too, and it is worth carrying forward. Bending spacetime deliberately is studied seriously. Miguel Alcubierre published a genuine warp-drive solution in Classical and Quantum Gravity in 1994: a shape for space itself that moves a bubble of flat space through curved space, with the passengers never locally exceeding light speed. It requires negative energy density, which has no known bulk source, and even the least demanding modern reworkings call for masses on the order of a planet converted entirely into exotic energy. That is an open research line with a very large number attached, which is a different situation from a claim with no number at all.

The vacuum has its own energy budget, and the Casimir force, the small push two mirrors a whisker apart feel from the empty space between them, is measured in laboratories. That is the nearest thing anyone has to a handle on it.

The instrument that reads a bend in space

This is the calibration that makes every gravity claim measurable.

On 14 September 2015, two detectors 3,000 kilometres apart in the United States recorded the same short chirp, rising from 35 to about 150 cycles a second in a fifth of a second. It was two black holes, about 36 and 29 times the mass of the Sun, falling together into one of about 62. The missing three solar masses, 5.4 × 10⁴⁷ joules, left as ripples in the shape of space. The LIGO teams put the peak power of that radiation above the light of every star in the observable universe combined, briefly.

By the time it reached Earth, 1.3 billion light years later, it stretched things by one part in 10²¹. Across LIGO's 4 kilometre arms that is a length change of 4 × 10^-18 metres: about one four-hundredth of the width of a proton. Stretch the same ruler from here to the Sun and the same wobble would be roughly the width of a single atom.

Two things follow, and they pull in opposite directions, which is why the instrument matters so much.

The first: making gravitational waves costs more than anything on the shelf can supply. The 2015 merger turned three suns' worth of mass, 5.4 × 10⁴⁷ joules, entirely into ripples, and 1.3 billion light years later that moved a four kilometre ruler by 4 × 10^-18 metres. A craft claiming to bend space around a 16 metre disc is claiming a bend big enough to fly on, generated in a hangar, with no energy budget stated.

The second, and the better one: we now have the ruler. Advanced LIGO can pick out a stretch of about one part in 10²³ in a second of listening, across a wide band of notes. Any device on Earth radiating gravitationally above that floor would leave a trace in the data, and the data is public. This is a claim that no longer has to be argued about. It can be looked up.

The 1989 timing, at full strength

Take the strongest version of the account's case, because it deserves to be stated properly.

In 1989 the periodic table stopped, experimentally, at element 109. Meitnerium had been made at Darmstadt in 1982 and nothing heavier had been confirmed. Elements 110 and 111 arrived in 1994, 112 in 1996, 114 in 1998 and 1999, and 115 not until 2003. The account named element 115 specifically, fourteen years and five elements before anyone made it. It put the return of stability at "113 to 116, somewhere in there", which does bracket where the island is now thought to sit, with element 114 the leading candidate for the proton shell closure and confirmed at Berkeley in September 2009. And it tied the element's interest to the strong nuclear force, which is in fact the physical reason nuclei near closed shells hold on longer.

As a 1989 statement about which numbered box on the table would matter, it landed in the right neighbourhood.

Here is what sits alongside it. Element 115 had been public physics for twenty years by then.

So in 1989, anyone with a university library card, a chemistry textbook or a wall chart could name element 115 and say that stability was expected to return around 114. The account's claimed workplace, the Los Alamos Meson Physics Facility, is a nuclear physics site where that literature circulates as a matter of course.

This is the part that cannot be resolved by finding element 115, and it is worth being clear about why. Because the number was in print from 1969 and on periodic tables from 1979, the existence of element 115 cannot separate inside knowledge from library knowledge. Only a claim that was not in the pre-1989 literature, and later turned out to be right, could do that work. The claims the account added beyond the literature are exactly the checkable ones: stable, unsynthesisable, melting at 1740 degrees Celsius, emitting a field.

Said in 1989, and measured since

One more line belongs here, about where the element would come from. The account says a supernova. Supernovae and neutron-star mergers do build heavy nuclei fast, and the neutron-star merger seen in 2017 gave the first direct look at it happening. The catch is distribution: anything a supernova makes and that survives gets spread through the interstellar dust the same way everything else does, and turns up in meteorites, in ores and in starlight. Searches for natural superheavy elements in monazite sand, in cosmic rays and in meteoritic material have run since the 1970s. No confirmed find at element 115 has been reported. Uranium-238 survived from the pre-solar era because its half-life is 4.47 billion years; below about 80 million years, nothing from that era is left in the crust at all.

What would have to be true

None of those is a shrug. Each one is a specific quantity, and each one names an instrument that would return it.

What you would measure

Every one of these uses equipment that exists now, and returns a number rather than an opinion.

The one on that list that would change the most is the last of the physics ones. Fifty years of dedicated work has not put a single nucleus on the island of stability, and until someone does, every half-life quoted for it is a calculation waiting on a measurement. That is a genuinely open question with a machine-shaped answer, and it is being worked on right now at Dubna, at Darmstadt, at RIKEN and at Berkeley.

Element 115 turned out to be real, and to last two thirds of a second. Nine neutrons offshore, the models say, is somewhere nobody has stood yet.

Keep exploring

Sources

• Yu. Ts. Oganessian et al., "Experiments on the synthesis of element 115 in the reaction ²⁴³Am(⁴⁸Ca,xn)^291-x115", Physical Review C 69, 021601(R), 2 February 2004.
• D. Rudolph et al., "Spectroscopy of element 115 decay chains", Physical Review Letters 111, 112502, 27 August 2013 · physics.aps.org/articles/v6/s118
• "Discovery and Assignment of Elements with Atomic Numbers 113, 115, 117 and 118", IUPAC and IUPAP joint working party, 30 December 2015 · iupac.org
• "IUPAC is naming the four new elements nihonium, moscovium, tennessine, and oganesson", IUPAC, 28 November 2016 · iupac.org
• Isotopes of moscovium, half-lives and decay modes · en.wikipedia.org/wiki/Isotopes_of_moscovium
• Island of stability: magic numbers, candidate shell closures and the neutron gap · en.wikipedia.org/wiki/Island_of_stability
• "A new way to make element 116 opens the door to heavier atoms", Lawrence Berkeley National Laboratory, 23 July 2024 · newscenter.lbl.gov
• "Element 114 confirmed", Lawrence Berkeley National Laboratory, 24 September 2009 · newscenter.lbl.gov/2009/09/24/114-confirmed
• O. L. Keller, C. W. Nestor and B. Fricke, "Predicted properties of the superheavy elements. III. Element 115, eka-bismuth", The Journal of Physical Chemistry 78(19), 1945-1949, 1 September 1974 · pubs.acs.org
• "On the Record with George Knapp", KLAS-TV interview transcript dated 9 December 1989, and the KVEG radio interview of 28 December 1989 · papooselake.org
• "The word of Bob" and "Element 115 tidbits", collected primary quotations and the pre-1989 print record, otherhand.org · otherhand.org
• David L. Morgan, "A physicist's critique", first published 26 August 1996, revised April 2001 · otherhand.org
• B. P. Abbott et al., "Observation of Gravitational Waves from a Binary Black Hole Merger", Physical Review Letters 116, 061102, 11 February 2016 · gwosc.org/events/GW150914
• "Properties of the binary black hole merger GW150914" and the Advanced LIGO sensitivity curves · arxiv.org/pdf/1608.01940
• Antimatter production, trapping and quantities, CERN · home.cern/science/physics/antimatter
• The ALPHA collaboration, antihydrogen production and trapping records · alpha.web.cern.ch
• M. Alcubierre, "The warp drive: hyper-fast travel within general relativity", Classical and Quantum Gravity 11, L73, 1994.
• P. de Marcillac et al., "Experimental detection of alpha-particles from the radioactive decay of natural bismuth", Nature 422, 876-878, 2003.