Polonium Radiohalos

"Therefore, the existence of polonium halos should no more be questioned." — H. Meier and W. Hecker, Geochemical Journal, 1976¹

"This is, to say the least, extremely disappointing, because so many young-earth creationists (the present authors included) have in the past often used the Po radiohalos as evidence of fiat creation of the rocks containing them." — Andrew Snelling and Mark Armitage, Fifth International Conference on Creationism, 2003²

"As is obvious in the paper, we have proven nothing — simply offered an alternative explanation.  We had included a question mark at the end of our title of the paper, but it was removed apparently by the editor." — A. L. Odom to Robert Gentry, 27 October 1989, as printed by Gentry³

Summary

A radiohalo is a sphere of radiation damage around a tiny radioactive grain in a mineral.  Each ring's radius is the range of one alpha energy, so the pattern names the isotope that made it.  Some halos in biotite show only the rings of three polonium isotopes and none of the uranium rings that should sit inside them.  Robert Gentry measured about 100,000 halos at Oak Ridge National Laboratory, showed that polonium halos are a real and distinct class, and argued from the three-minute half-life of polonium-218 that the granite holding them formed in minutes.  He called them God's fingerprints.

Eight conclusions, from the primary literature on both sides.

1. The halos are real, and Gentry's strongest point is conceded by his critics. An independent laboratory in Bamberg confirmed the halos in Norwegian biotite in 1976 and confirmed, by fission-track analysis sensitive to parts per million, that their centres hold no uranium.¹ Gentry's ring-radius measurements are good physics, and every paper on the subject builds on them, including those that reject him.
2. Gentry's own 1976 paper in Science shows polonium halos forming by fluid transport. In coalified wood he reported thousands of polonium halos, called them "derived from uranium daughters", counted three possible polonium-214 rings and zero polonium-218 rings, and printed the groundwater explanation for his own young-age claim in the same paragraph.⁴
3. The isotope census points the same way in all three published counts, including both creationist ones. Meier and Hecker counted more than 1,000 polonium-210 halos, 90 polonium-218 and 1 polonium-214 in one biotite. Snelling and Armitage counted 1,857 polonium-210, 152 polonium-214 and 2 polonium-218 halos in 895 slides from three plutons.² The population is dominated by the one polonium isotope whose parent, lead-210, lives 22 years. A transport model predicts that. A creation model has no reason to.
4. Gentry's sites are not primordial granite. Richard Wakefield tracked down Gentry's three Bancroft, Ontario localities and read the mines reports on each. Two are calcite vein dikes. The third is a pegmatite in a uranium mine whose ore ran 0.1074 percent uranium oxide. Uranium-bearing betafite sits inside the mica books. All three cut older sedimentary rocks, some with stromatolites.⁵
5. The young-earth movement's own research programme rejected Gentry in 2003 and said so in print. Snelling predicted in 2000 that polonium-218 and polonium-214 halos might be "rare or virtually absent" from Phanerozoic granites, counted, found two polonium-218 halos and 152 polonium-214, and wrote that Gentry's primordial hypothesis "has been falsified".² Gentry answered with an open letter offering the absence of a reply as "scientific proof", and has never withdrawn.
6. RATE's replacement model, hydrothermal transport during accelerated decay, is the critics' mechanism with a faster clock. Wakefield in 1988 and Snelling in 2003 reached the same mechanism from opposite premises. They differ on how long the fluids ran. RATE's version needs the granite to receive 100 million years' worth of uranium decay and cool below 150 °C inside days, on the strength of an annealing paper this project has read only in abstract.
7. RATE's model has unanswered problems of its own, and the sharpest one is Snelling's. In 2000 he argued that fluid transport predicts 67,000 polonium-210 halos for every polonium-218 halo, against 11 to 1 observed.⁶ He adopted fluid transport in 2003 and has never addressed the ratio. A third model, Meier and Hecker's cold groundwater leaching long after the rock formed, dissolves the timescale argument, and no young-earth source engages it.
8. The mainstream side has loose ends too. Its most-cited refutation, Odom and Rink's 1989 Science paper, rests on three giant halos found by accident in quartz, and its first author wrote privately that "we have proven nothing". The photometric test proposed in 1992 that would settle whether a "polonium-218 halo" is really a radon halo has never been run. And nobody has published the supply-rate arithmetic that decides whether a halo took 500 million years or ten days.

Verdict.  Polonium halos do not prove instant creation, and the young-earth movement's own best data say so.  What survives is a real, uranium-free phenomenon that both sides now explain by fluid transport from nearby uranium.  At today's decay rates that takes hundreds of millions of years.  RATE keeps a young age only by adding accelerated decay, a separate claim with its own article.  The halos are evidence for uranium migration in mica, not for the age of the earth.

Part 1: What a polonium halo is, and what nobody disputes

Rings that name their isotope

Snelling's RATE chapter gives the standard description.  "Radiohalos (abbreviated from radioactive halos) are minute circular zones of darkening surrounding tiny central mineral inclusions or crystals within some minerals."⁷  The inclusion is usually a grain of zircon or monazite about a micrometre across, sitting in biotite.  An alpha particle fired from it travels a fixed distance for its energy and dumps most of its damage at the end of its track, so hundreds of millions of decays paint a dark shell at that radius.  Each isotope in a chain emits at a different energy, so a chain paints nested shells, 10 to 40 micrometres across in thin section.

The link from ring radius to isotope is the whole basis of the subject.  "Many years of subsequent investigations have established that the radii of the concentric circles of the radiohalos as observed in thin sections are related to the α-decay energies. This enables the radioisotopes responsible for the α-decays to be identified".⁷  The uranium-238 chain has eight alpha emitters.  Their energies run from 4.19 MeV for uranium-238 up to 7.69 MeV for polonium-214, and a full uranium halo shows their rings.  Three of the eight are polonium isotopes.  Polonium-218 has a half-life of about three minutes and emits at 6.00 MeV.  Polonium-214 lives 164 microseconds and emits at 7.69 MeV.  Polonium-210 lives 138 days and emits at 5.30 MeV.⁸

Snelling's figure caption gives the dose.  "Halo coloration initially develops after about 100 million α-decays, becomes darker after about 500 million, and very dark after about 1 billion (after Gentry [1988])."⁷

The polonium halos

Henderson in 1939 and Gentry from 1968 described halos in biotite that show only the polonium rings.  A "polonium-210 halo" is a single dark disk at the 5.30 MeV radius.  A "polonium-214 halo" shows the polonium-214 and polonium-210 rings.  A "polonium-218 halo" shows all three.  None of them shows the five inner rings that uranium-238, uranium-234, thorium-230, radium-226 and radon-222 should have painted if the polonium had decayed from uranium sitting in the same grain.

Gentry's 1974 paper in Science is the foundation of everything that followed.  Its abstract:  "New photographic evidence, data on halo ring sizes, and x-ray fluorescence analyses provide unambiguous evidence that polonium halos exist as a separate and distinct class apart from uranium halos. Because of the short half-lives of the polonium isotopes involved, it is not clear how polonium halos may be explained on the basis of currently accepted cosmological models of Earth formation."⁸  He had examined about 100,000 halos from Precambrian granites and pegmatites on several continents, and calibrated the ring radii with more than 350 helium-ion irradiations of biotite, fluorite and cordierite.⁸

Two years later Meier and Hecker, at the state geochemistry institute in Bamberg, West Germany, found polonium halos in biotites from several regions.  Gentry's observations, they wrote, "(1968, 1974) concerning these halos could be confirmed.  Therefore, the existence of polonium halos should no more be questioned."¹  They also ran the fission-track test on the halo centres and found no uranium.  "it can be stated that no or only small amounts of uranium are present in polonium halos, a result, which agrees with the lack of uranium ring-structures in these halos".¹  This is Gentry's strongest experimental point, confirmed by people who went on to reject his interpretation.

The three-minute argument

Gentry's case is at its best for polonium-218, and he says why.  "There is no beta progenitor for ²¹⁸Po; so the ²¹⁸Po halo must have originated with this isotope, whose half-life is just three minutes."³  Polonium-218 is born from radon-222 by alpha decay.  No longer-lived atom turns into it by beta decay, so nothing can carry it slowly to a site and release it there.  Whatever put polonium-218 into a radiocentre did so within a few half-lives.

Gentry concedes in the same paragraph that the other two halo types do not carry this constraint.  A polonium-210 halo could have started as lead-210, half-life 22 years, or bismuth-210, 5 days, and "a ²¹⁰Po halo would still have resulted".  A polonium-214 halo could have started as lead-214 or bismuth-214, with half-lives of 27 and 20 minutes.³  So on Gentry's own reckoning only the polonium-218 halo needs a three-minute delivery.  The polonium-210 halo tolerates decades.

For polonium-218 the arithmetic is stark.  A very dark halo needs a billion decays, so "a straightforward explanation of ²¹⁸Po halos implies that the 1-µm radiocenters of very dark halos of this type initially contained as many as 5 × 10⁹ atoms (a concentration of more than 50 percent) of the isotope ²¹⁸Po (half-life, 3 minutes), a problem that almost defies reason."⁸  Half the atoms in the grain would have had to be an isotope that vanishes in an hour.

What Gentry published, and what he said elsewhere

The Science paper does not claim a young earth.  It ends with a question.  "In summary, the combined results of ring structure studies, mass spectrometric analyses, and electron induced x-ray fluorescence present a compelling case for the independent existence of Po halos. The question is, can they be explained by presently accepted cosmological and geological concepts relating to the origin and development of Earth?"⁸  Baillieul, for the National Center for Science Education, is right that "Gentry never presented his hypothesis or conclusions regarding a young earth in these research articles."⁹  Snelling says the same from the other side.  "Of course, Gentry had to be circumspect about these implications of the Po halos when he first published his results in the open scientific literature.  Thus he postulated the possibility of long-lived precursors to the Po isotopes that were now extinct.  Yet at the same time he hinted at fiat creation as the explanation by asking leading questions".¹⁰

The young-earth claim is in Gentry's book, Creation's Tiny Mystery, and even there it is framed as a question.  "Were the polonium halos mute evidence of extinct natural radioactivity?  Was, then, the half-life of ²¹⁸Po—just three brief minutes—the measure of time that elapsed from the creation of the chemical elements to the time that God formed the granites?"³  Retellings that call the 1974 paper published young-earth science overstate it;  what Science printed was a phenomenon and a question.

The measurement the identification hangs on

One more fact is undisputed, and it weakens every position.  The alpha from radon-222 has an energy of 5.49 MeV.  The alpha from polonium-210 has 5.30 MeV.  Moazed, Spector and Ward, who re-measured biotite from Gentry's own Faraday pegmatite in 1973, put it plainly.  "We must point out here that there is a ²²²Rn ring which comes from an alpha decay of energy just 3.6 percent higher than the ²¹⁰Po alpha energy."¹¹  The two rings overlap.

Gentry answered in the 1974 paper that in fluorite the annulus between the polonium-210 and polonium-218 rings measures 3.9 micrometres in a polonium-218 halo and 3.0 micrometres in a uranium halo, because the uranium halo has the radon ring inside that gap and the polonium halo does not.⁸  That is a real measurement.  It is also a 0.9 micrometre difference against a best-case uncertainty he puts at 0.1 micrometre.  Whether a three-ring halo began with polonium-218 or with a radon atom that decayed in place has never been settled by a measurement both sides accept.  It matters because radon is a gas with a half-life of 3.8 days, and needs no miracle to travel.

Part 2: The evidence that the polonium was delivered

Gentry's own transport paper

The most damaging paper to Gentry's granite argument is by Gentry.  In 1976 he and seven Oak Ridge co-authors published on radiohalos in coalified wood from Colorado Plateau uranium deposits in Triassic, Jurassic and Eocene beds.⁴  Nobody disputes that groundwater carried the uranium in.  The abstract names them.  "The discovery of ²¹⁰Po halos derived from uranium daughters, some elliptical in shape, further suggests that uranium-daughter infiltration occurred prior to coalification when the radionuclide transport rate was relatively high and the matrix still plastically deformable."⁴  The radiocentres, the body text says, "are actually secondary sites that preferentially accumulated α-radioactivity during an earlier period of earth history when uranium-bearing solutions infiltrated the logs after they had been uprooted".⁴

Then comes the count.  "Of the thousands of Po halos seen in this matrix, only three show any trace of a ring that could possibly be attributed to ²¹⁴Po α-decay [that is, from the accumulation of the U-daughters ²¹⁴Pb (t½ = 27 minutes), ²¹⁴Bi (t½ = 20 minutes), or ²¹⁴Po (t½ = 164 µsec)], and none has been seen with a ring from ²¹⁸Po α-decay [that is, from the accumulation of short-lived ²¹⁸Po (t½ = 3 minutes)]."⁴  Where transport is not in doubt, the population is pure polonium-210.  That is what a transport model predicts:  polonium-210 is reachable through the 22-year lead-210 bottleneck, and polonium-218 is not.

Gentry gave the timescale for a transported polonium-210 halo.  "a spherical ²¹⁰Po halo could develop in 0.5 to 1 year from the ²¹⁰Po atoms initially present and a second similar ²¹⁰Po halo could develop in 25 to 50 years as the ²¹⁰Pb atoms more slowly β-decayed to produce another crop of ²¹⁰Po atoms."⁴  He accepts a decades-long window when the geology calls for one.

The paper also makes a young-age claim from uranium-lead ratios in the halo centres, and popular retellings drop its second half.  "Such extraordinary values admit the possibility that both the initial U infiltration and coalification could possibly have occurred within the past several thousand years.  At the same time it may be argued that this view is quite improbable for there exists another explanation that could invalidate the association of the U/Pb ratios with the initial introduction of U."⁴  The other explanation is that uranium was added or lead removed "by groundwater circulation after coalification".  Gentry prints the rebuttal of his own result in the same paragraph and does not answer it, so quoting the first sentence without the second misquotes by omission.

One reading of this paper favours Gentry against his own movement.  If years of transport through a wet, plastic log produced zero polonium-218 halos, transport does not easily make them.  The polonium-218 halos that do exist in granite are a real problem for any transport model, including RATE's.

The isotope census

Three groups have counted polonium halos by type, two of them creationist.

Meier and Hecker describe the ordering.  "the most common ones belong to the ²¹⁰Po type;  ²¹⁸Po halos have been found in a less portion whereas ²¹⁴Po halos are very seldom."¹  Snelling and Armitage's abstract says "²¹⁸Po radiohalos were rare", and both of their polonium-218 halos came from one Stone Mountain sample.²  Two out of 2,011 polonium halos is 0.1 percent.

All three counts are dominated by the one polonium isotope with a long-lived beta parent.  A grain created with its polonium in place has no reason to prefer polonium-210.  A grain fed by transported uranium daughters has every reason to, because lead-210 lives 22 years and can travel.

One caveat comes from Snelling.  The same 2000 article that gives the 67,000 to 1 ratio adds that "in other biotites the abundance ratios are ²¹⁸Po>²¹⁰Po>²¹⁴Po, and even ²¹⁴Po>²¹⁸Po or ²¹⁰Po", citing Norman Feather's 1978 review, which this article has not read.⁶  Polonium-210 dominance is established for the three datasets in the table, not for every biotite ever examined.

The geology of Gentry's sites

Gentry's samples were mailed to him, and he published almost nothing about their provenance.  Baillieul's complaint is fair.  "Gentry is a physicist, not a geologist. Contrary to accepted geologic reporting practice, he consistently fails to provide the information that a third party would need to collect comparable samples for testing."⁹

Richard Wakefield, an amateur geologist in Ontario, tracked down the three Canadian sites from Gentry's papers and from Louis Moyd, read the mines reports on each, and in October 1987 took Gentry and the Survey's Bancroft resident geologist Hans Meyn into the field.  He published the result twice in 1988, in Creation/Evolution and in the refereed Journal of Geological Education.⁵¹²  All three sites are in the late Proterozoic Grenville Supergroup, not in Archean basement.

The Fission Mine and the Silver Crater Mine, source of Gentry's "spectacle halo", are calcite vein dikes.  The Ontario Department of Mines report on Silver Crater classes it as a hydrothermal deposit.  Wakefield's verdict:  "the Silver Crater and the Fission Mines are not granites.  The composition and mode of origin is totally wrong for a granite.  In identifying them as granites, Gentry has made a major error."⁵

The Faraday Mine is a granite pegmatite, and Gentry acknowledges his biotite came from it.  It is also a uranium mine.  "This is very difficult to accept since the Faraday pegmatite was mined for uranium."  The ore ran at "0.1074 percent uranium oxide", in uranothorite grains under a tenth of a millimetre scattered through the pegmatite.⁵  At Silver Crater the mines report records that "Small crystals of betafite have been found within the books of mica".  Betafite is a uranium mineral.  Wakefield:  "In my phone conversation with Gentry in February, he admitted the betafite was with his samples.  Why did he leave that observation out of his papers?"⁵  A uranium mineral inside the mica books is exactly the local source Gentry's model requires to be absent.

The dikes cut older rocks.  Wakefield lists contact metamorphism of the country rock, "pieces of sedimentary rocks enclosed within the intrusive rocks, engulfed and surrounded by the melt", and stromatolites in marble just south of Bancroft cut by the Faraday gabbro and pegmatites.⁵  His conclusion:  "the samples of biotite that contain Gentry's polonium halos came from pegmatite dikes and calcite vein dikes which cross-cut metamorphosed volcanic, sedimentary, and igneous rock units.  The dikes are clearly the last to have formed, not the first."⁵  The refereed version states the mechanism in its abstract.  "The geology of the sites shows that the uranium, and most likely the polonium, were deposited via post-magmatic hydrothermal fluids."¹²

Gentry's published reply is one paragraph.  Wakefield's article, he writes, "pictures the region near Bancroft, Ontario … as explainable only by evolutionary processes" by quoting geologists to establish an evolutionary history and then concluding the rocks formed that way, "a clear example of circular reasoning."³  He does not dispute that his micas came from dikes cutting metasediments.  He argues that the geologists' dating is circular.  A cross-cutting relationship is an observation, not a date.  A dike that cuts a stromatolite bed is younger than the bed on any timescale.

Wakefield says what his geology does not show.  "Still, we must give Gentry his due.  Nothing in geology fully explains the apparent occurrence of the polonium halos as described by Gentry.  They do remain a minor mystery in the field of physics."⁵  In the next lines he sketches the model both camps later built.  "The very fact that Gentry's halos at these sites occur in areas of unusually high uranium mineralization and metamorphism suggest that halos may be connected with the migration of uranium-bearing fluids through or within the rocks."⁵

Why only these three isotopes

Lorence Collins, a petrologist at California State University, Northridge, added an argument in 1997 that needs no fieldwork.¹³  Uranium-235 and thorium-232 also decay through four polonium isotopes, with alpha energies that would paint rings of their own.  No halo of any of them has been reported.  Only the three polonium isotopes of the uranium-238 chain make halos.

Radon explains that.  Each chain passes through a radon isotope, a noble gas that moves freely in water.  Radon-222 in the uranium-238 chain lives 3.82 days.  Radon-219 in the uranium-235 chain lives 3.92 seconds.  Radon-220 in the thorium chain lives about 56 seconds.  "The ²²²Rn has about 84,000 or 6,000 times as much time to enter the biotite as does ²¹⁹Rn or ²²⁰Rn!"¹³  The missing halos are exactly the ones a radon-transport model predicts to be missing.  A creation model has no reason to omit them.  Collins adds that polonium halos occur only in granites and granite pegmatites, "never in any other rock types, excepting locally near Bancroft", never in biotite-bearing gabbro, diorite or tonalite, and always in rocks that carry uranium concentrations.¹³

Collins is candid that the argument returns to Gentry's starting point.  "The argument comes full circle when it meets Gentry's initial [truthfully made] observation that magmatically-derived granites cannot contain Po halos: the half-lives of the Po isotopes are simply too short."¹³  His answer is that the host biotite formed by solid-state replacement long after any melt, with radon arriving in fluids over ordinary geological time.  Collins' wider view that most granites are replacement rocks is a minority position among petrologists.  The selectivity argument does not depend on it.

Where the mainstream case is thinner than it looks

Moazed, Spector and Ward measured the central disks of 24 halos of the smaller types and found none as large as the smallest disk of a type B halo, the two-ring kind Gentry reads as polonium-214.  On their reading the type B disk is a darkened uranium halo whose disk and polonium-218 ring have blended.  "The smallest type B disk we found measured 20.9 ± 0.6 µm, whereas the largest type C disk measured 19.8 ± 0.6 µm."¹¹  They read the halo types as stages of one darkening sequence rather than distinct isotope populations, and declined Gentry's no-uranium result because "Gentry (4) reported no direct comparison between the amounts of uranium found in the polonium inclusions and the amounts of uranium in a series of control uranium halos".¹¹  Gentry's reply is that they "concluded that all Po halos are only U halos, without having U halos with normal ring structure available for comparison."⁸  That is fair;  their biotite had no uranium or thorium halos to compare.  Neither side has closed this, though Meier and Hecker's fission-track result three years later answered the uranium question in Gentry's favour.

Odom and Rink's 1989 Science paper, "Giant Radiation-Induced Color Halos in Quartz:  Solution to a Riddle", is the most-cited mainstream solution.  It rests on three giant halos around monazite in quartz, seen by accident.  The mechanism is real physics.  Alpha decay creates charge carriers in quartz;  once local traps saturate, the excess holes migrate outward and the visible halo grows past the alpha range, and at high enough dose the same process destroys the colour centres nearest the inclusion.  "The apparent absence of ring structures readily associated with U in so-called Po halos might be the result of the destruction of color centers by excessive alpha dose near the inclusions."¹⁴  But the extension to mica is labelled a suspicion.  "We strongly suspect, however, that the sizes and structure of giant and Po RICHs in mica also are artifacts of radiation-induced conductivity and that their explanation requires neither unknown radioactivity nor an abandonment of current concepts of geologic time."¹⁴

Gentry sent Science two replies, both rejected for space.³  He prints them in his book's appendix with a letter he says came from Odom on 27 October 1989.  "The Science paper was the result of an accidental finding;  it is not something that we are really working on.  As is obvious in the paper, we have proven nothing — simply offered an alternative explanation.  We had included a question mark at the end of our title of the paper, but it was removed apparently by the editor.  Jack and I would be very happy to see you investigate these halos, and if possible test our model."³  The same letter says why the mica extension is a suspicion.  "We could not do this for Po halos, for we actually do not have a Po halo."  This is Gentry reproducing a private letter, unconfirmed by any independent copy and never confirmed with either author.  If it is genuine, the most-cited refutation of polonium halos is a three-specimen accident whose first author declined to call it a solution.

Then there is the experiment nobody has run.  John Brawley, an amateur working with biotite he collected in the Spruce Pine district and the Black Hills, argued in 1992 that the three-ring "polonium-218 halo" is a radon-222 halo, and proposed a test.  "If it is true that these unambiguous 'Polonium-218' (according to Dr. Gentry) halos are actually Radon-222 halos, then it should be possible to statistically determine the relative density of the three rings.  If this were done with a through-the-microscope photometer, the inner halo should have a density reading roughly twice that of either of the outer two halos, when adjusted for the difference in their diameters."¹⁵  A radon atom that decays in place puts two alphas into the innermost ring, its own and polonium-210's, and one each into the other two.  Brawley is not an authority and his paper is not peer-reviewed.  His test needs a photometer and existing slides.  So far as this project can find, nobody has run it in thirty-four years.

Part 3: The young-earth movement rejects Gentry

2000: the hypothesis is live, and the test is named

In the first RATE volume, in 2000, Snelling reviewed the radiohalo literature and treated Gentry's primordial hypothesis as open.  He was honest about the uranium correlation.  "Though the evidence of Po halos associated with nearby U sources is suggestive, it is by no means yet conclusive."¹⁰  And he named the test.  The prime targets would be Phanerozoic granites that demonstrably intrude fossil-bearing sediments, and "it may even be that ²¹⁸Po and ²¹⁴Po halos are rare or virtually absent from Phanerozoic granitic rocks."¹⁰

The same year, in an ICR Impact article, he argued against fluid transport with a number.  "there should be 67,000 ²¹⁰Po radiohalos for each ²¹⁸Po radiohalo.  Yet in a Norwegian biotite there are more than 1,000 ²¹⁰Po radiohalos, 90 ²¹⁸Po radiohalos and only one ²¹⁴Po radiohalo".⁶  The Norwegian biotite is Meier and Hecker's Kragerø sample.  If polonium arrives in proportion to how long each isotope survives in transit, the observed 11 to 1 is six thousand times too rich in polonium-218.  He also conceded that "16 out of 20 Po-radiohalo-bearing biotite localities have reported U in the rock or in an orebody."⁶  His verdict borrowed Dalrymple's phrase from the 1981 Arkansas trial:  "Thus the Po radiohalos still remain 'a very tiny mystery.'"⁶

2003: the count, and the word "falsified"

Snelling and Mark Armitage then counted.  Nineteen samples from three plutons in Georgia, California and New South Wales, cut into 895 slides.  The totals were 1,857 polonium-210 halos, 152 polonium-214, 2 polonium-218, 559 uranium-238 and 37 thorium-232.²  Half of Snelling's 2000 prediction had come true.  Polonium-218 halos were all but absent, two in 895 slides;  polonium-214 halos were not, at 152.  The half that came true is the half his argument needed.

All three are Phanerozoic plutons that intrude older rocks, so on the authors' own reading they are not primordial.  The abstract draws the consequence:  "neither the Po radiohalos nor the granitic rocks could have been formed by fiat creation."²  And in the body:  "the hypothesis that the three different types of Po radiohalos found in biotites always represent the decay of primordial Po (original Po not derived by U-decay) (Gentry, 1979, 1980, 1982, 1983, 1984, 1986, 1988, 1993) has been falsified".²

They say in the first person what that costs.  "This is, to say the least, extremely disappointing, because so many young-earth creationists (the present authors included) have in the past often used the Po radiohalos as evidence of fiat creation of the rocks containing them.  Nevertheless, the falsifying of this hypothesis does not in any way falsify the general creation hypothesis based on Scripture, that 'in the beginning God created the heaven and the earth.'"²

They also contradict Gentry on the field evidence.  Gentry had maintained that polonium halos do not occur along cracks or conduits in biotite.  "This assertion is emphatically incorrect."²  Biotite is peeled along cleavage planes to make slides, so a halo that looks isolated sits on a surface that is itself a fluid path.  Their model puts hot water in those planes, carrying radon-222 and its daughters from nearby zircons.  It comes with a caveat that popular retellings omit.  "It needs to be stressed that this secondary transport model for the origin of the Po-rich radiocenters and therefore the Po radiohalos is tentative.  Further data collection and analysis is needed."²

This deserves credit.  Snelling named the result that would falsify his movement's most famous argument, measured it, published the word, and named himself among those who had been wrong.

Gentry's reply

Gentry did not accept it.  In November 2002, before the ICC paper, Snelling wrote to him defending the claim that many granites form from melted sediment.  Gentry had demanded location details for "fossiliferous granites".  Snelling's reply:  "Bob, you know as well as we do that granites do not contain fossils."  Gentry was "imputing to us something we did not write or claim."¹⁶  He offered what Baillieul says Gentry never supplied:  "we can provide you with the precise location details (even GPS coordinates in most instances) of every granite sample we have collected."¹⁶  And he acknowledged the debt.  "we appreciate very much your pioneering work on the helium in zircons, and the radiohalos, which we have acknowledged and have merely built upon."¹⁶

Gentry's open letter to ICR of July 2003 runs sixty pages.  Its scientific core is this sentence.  "The failure of any scientist — atheist or agnostic, evolutionist or creationist, including Andrew — to respond to my widely published challenges over the last twenty-five years to refute these results and publish their findings in the same peer-reviewed journals, constitutes scientific proof that this remarkable evidence for Earth's recent fiat creation remains as immovable as the granites themselves."¹⁷  That is an argument from silence, offered as proof.

The challenge it refers to is the one Gentry issued in EOS in 1979 and repeats in his book.  "I would consider my thesis essentially falsified if and when geologists synthesize a hand-sized specimen of a typical biotite-bearing granite and/or a similar size crystal of biotite."  He would give up primordial polonium-218 halos "when coercive evidence (not just plausibility arguments) is provided for a conventional origin".³  Whether that is a fair test is taken up under Objections below.  His website still advertises the halos as unrefuted evidence for instant creation.

After 2003

The creationist organisations followed Snelling, not Gentry.  His 2005 Impact article reports polonium halos in metamorphosed Thunderhead Sandstone along U.S. Highway 441, in biotite that could only have grown during metamorphism, and concludes:  "The hydrothermal fluid transport model for Po radiohalos formation has thus been tested and verified.  Neither the Po nor the biotite flakes were primordial."¹⁸  The current Creation Ministries International article teaches the same.  "Hot water flowing inside the cooling granite was able to carry the polonium short distances and concentrate it into new radiocentres."¹⁹

The split is complete.  Gentry says the granites are Day One creations and the polonium was never derived from uranium.  Snelling says they crystallised from Flood-year magmas and every polonium atom came from uranium in a nearby zircon.  They cannot both be right about the same rocks.

Part 4: What the RATE model needs

The model

RATE's version is the critics' mechanism with a faster clock.  Hot water released by the crystallising granite carries radon-222 and its polonium daughters along biotite cleavage planes from zircons to sites where the polonium is trapped.  The halo registers only once the mica has cooled below the annealing temperature, about 150 °C.  The same granite holds fully formed uranium and thorium halos, 500 million to a billion decays each, so the model requires "at least 100 million years worth (at today's rates) of accelerated radioactive decay".²  Popular versions spell it out.  "In other words, the radioactive decay of uranium was formerly up to a billion times faster than it is today!"  And:  "polonium halos mean that the granites solidified and cooled in just 6 to 10 days!"¹⁹

The 6 to 10 days is not a thermal calculation.  It comes from polonium half-lives set against the annealing ceiling, and granite cooling examines it.  The heat of 100 million years of decay in a few days is treated in RATE and helium in zircons.  This article asks only what the halos require.

The annealing ceiling

Everything short about the model depends on the 150 °C figure.  Its source is Laney and Laughlin's 1981 paper on biotite from the Fenton Hill geothermal boreholes in New Mexico, where the present rock temperature is known at every depth.  The abstract:  "Annealing of the haloes is first observed in samples from a depth of 1850 m (134°C present temperature) and is complete in samples from a depth of 2120 m (151°C)."²⁰  So the number is real, and it is a range:  halos begin to fade at 134 °C and are gone by 151 °C.  This project has read only the abstract.  The full paper and its depth-temperature table are behind a Wiley paywall, and nobody on either side of the halo debate appears to have checked the number against them.  Note also that the abstract reports present temperatures in rock that has been hot for far longer than a few days.  Whether a halo survives a brief excursion above 150 °C during rapid cooling is a kinetic question it does not answer.

Distance is easy, supply is not

Gentry objected that polonium cannot diffuse even a micrometre through a mineral before it decays.  Snelling and Armitage answer that "this argument would be irrelevant if the diffusion were occurring in hydrothermal fluids flowing along the cleavage planes in the biotite flakes."²  They are right about distance.  With their own radon diffusion coefficient in water, 1.14 × 10⁻⁵ cm² per second, a radon atom random-walks about 4.8 cm in one half-life and a polonium-218 atom about 1.1 mm in one of its half-lives.  Both exceed the sub-millimetre separations Snelling reports.  One caution, which no critic has noticed:  for polonium-218 the paper quotes a coefficient measured in humid nitrogen gas, 7.9 × 10⁻² cm² per second, about seven thousand times the aqueous radon figure, which does not apply inside a water-filled cleavage plane.  It does not rescue Gentry, because the aqueous figure is still enough.

Supply is the problem.  A dark halo needs 500 million to a billion alpha decays.¹⁰  Snelling and Armitage note that their model "does not require 5 × 10⁹ atoms of ²¹⁸Po to be delivered to each radiocenter all at the same time", because flow can supply them progressively.²  True, but every one of those atoms is born from a uranium decay in a nearby zircon.  A zircon 10 micrometres across at 1,000 parts per million uranium has a uranium decay rate of roughly 6 × 10⁻⁸ per second.  Each decay yields one atom of each polonium isotope.  At that rate the grain supplies a billion polonium-210 atoms in about 500 million years.  A polonium halo fed from one grain is therefore a record of hundreds of millions of years at today's rates, or of a decay rate some ten billion times faster for a few days.  Both sides accept that framing.  Nobody has published the calculation with real grain sizes and uranium contents.  Until then the arithmetic above is this article's own estimate.

The ratio Snelling raised and never answered

The 67,000 to 1 objection of 2000 was Snelling's argument against fluid transport.  The 2003 paper adopts fluid transport and does not mention it.  Its own count, 1,857 to 2, is about 900 to 1, seventy times short of the prediction;  Meier and Hecker's 11 to 1 is six thousand times short.  The only published transport prediction is Snelling's, and it has never been withdrawn or addressed.

A second problem is mechanical and needs no new data.  If a radon atom reaches a site and decays there, its own alpha lands first, at 5.49 MeV, which is within 4 percent of the polonium-210 radius.  Polonium-218 follows in about three minutes at 6.00 MeV, and polonium-214 within the hour at 7.69 MeV.  Polonium-210 adds a fourth alpha at 5.30 MeV once the 22-year lead-210 step has run.  So a radon atom that decays in place always paints all three ring radii, and paints the inner one about twice as heavily.  That is Brawley's test, stated as arithmetic.  It also means a radon-fed radiocentre cannot build a single-ring polonium-210 halo at all.  That halo requires the chain to be separated at lead-210.  In the creationist plutons single-ring polonium-210 halos outnumber three-ring halos by about 900 to 1.  So the dominant transport step in the observed population is the slow one, and it needs no rapid event.  Neither Snelling nor his critics state this.  The energies are Gentry's Table, p. 63;  the reasoning is this article's own.

Halo darkness is dose, not time

Snelling concedes that halo radii "neither confirms constancy of radioactive decay rates nor precludes the possibility of accelerated decay."¹⁰  The concession cuts both ways.  A mature uranium halo records about a billion decays, which is equally consistent with 500 million years at today's rate and ten days at ten billion times that rate.  Halo maturity is evidence that a large dose occurred, not evidence of a young earth.  The young-earth case must be carried by the short-lived polonium isotopes, and that is where the data are thinnest:  two polonium-218 halos in 2,011.

Accelerated decay also collides with the halos themselves.  "once the idea of variable decay rates is introduced, it becomes impossible to assign discoloration halos to any specific isotope or isotopic series, and Gentry's hypothesis falls completely apart."⁹  Decay rate and alpha energy are set by the same nuclear binding, so a mechanism that changes one changes the other, and a changed alpha energy moves the ring that identifies the isotope.  Gentry's own 1974 paper says halo radii cannot fix the decay constant to better than 35 percent.⁸  The helium article works through the energy shift.  No young-earth source found for this article answers Baillieul's paragraph.

The model nobody engages

Meier and Hecker rejected both the magmatic and the hydrothermal routes in 1976.  The magmatic one "would only be possible in quickly crystallizing systems.  Because of the slow magmatic cooling rates of polonium-halo-bearing rocks such as granites, however, this hypothesis is untenable."¹  The hydrothermal one they rejected on Gentry's own data.  "GENTRY (1968) cites measurements of the alpha-recoil density showing no excess alpha activity near polonium halos which should be expected in the presence of alpha-decay precursors in hydrothermal fluids.  Moreover, polonium halos cannot be ascribed only to biotite samples which have been in contact with hydrothermal solutions during their geological history."¹  That is a published secular objection to the mechanism RATE built on, thirty years before RATE.

Their own answer is supergene:  cold groundwater leaching uranium and radium out of the granite long after it crystallised, with the isotopes separating by their different mobilities in water-filled cracks.  "polonium isotopes are accumulated out of uranium bearing or radium bearing solutions which obtained their uranium or radium from rocks, such as granites, through leaching processes by percolating natural waters."¹  They tie it to a field observation, polonium halos sitting "in the neighbourhood and at the end of these U-bands" of uranium precipitated along biotite fractures.¹  On this model the polonium is supplied continuously in a cold rock, the annealing ceiling is met trivially, and the "cool in days or the polonium decays" argument never arises.  No young-earth source located for this article engages Meier and Hecker.  Neither does Collins or Baillieul.  The one published model that dissolves the timescale argument has been ignored by both sides for fifty years.

Objections and responses

"RATE's model is the live young-earth position.  Refuting Gentry refutes nobody." Agreed.  But RATE's model is fluid transport, which at today's rates is the mainstream answer.  What makes it young-earth is accelerated decay, a separate claim argued in the helium article.  Its authors called it tentative, and it still owes answers to Snelling's 67,000 to 1 ratio, to the 900 to 1 dominance of single-ring polonium-210 halos that no radon route can make, and to Meier and Hecker.  Gentry's version has not gone away either;  his site and much popular material still teach it.

"Polonium-210 still has to be delivered fast.  138 days is short." It does not.  Gentry's own book grants that a polonium-210 halo could have begun as lead-210, half-life 22 years, and his own 1976 paper has a second polonium-210 halo growing over "25 to 50 years" from transported lead.  The isotope that dominates every census tolerates decades.

"Mainstream geology has never explained the polonium-218 halos that do exist." Correct.  Snelling and Armitage found two.  Meier and Hecker found 90 in one biotite.  Gentry found none in wood where transport is certain, which suggests transport does not easily make them.  The mainstream answer, that a three-ring halo is a radon-222 halo, rests on a 3.6 percent energy overlap and has never been tested with Brawley's photometer.  This is open.  But a halo type that is 0.1 to 8 percent of the population cannot carry a claim about the whole granite, and the young-earth model needs it to carry the whole argument.

"Supergene transport has never been demonstrated in granite." Also correct.  Meier and Hecker's model is an inference from field association and from the chemistry of uranium, radium and radon in water.  Nobody has grown a polonium halo in a laboratory by any route.  The hydrothermal model is in the same position, and its authors say so.  The supergene model asks for water and time.  The RATE model asks for the decay of uranium to have been "up to a billion times faster than it is today", on its own popular statement, and about ten times more than that on the supply arithmetic in Part 4.

"Nobody has synthesised a granite with a polonium-218 halo, so Gentry's challenge stands." The challenge asks a geologist to reproduce on a bench a process his model says takes ten thousand to a million years.  Failing confirms his model.  Passing would refute mainstream petrology as thoroughly as Gentry.  A test the opposing model predicts will fail is not a test of the opposing model.

"Wakefield's geology refutes the whole young-earth halo argument." It refutes Gentry's.  It moves the halos out of primordial basement into late hydrothermal dikes and pegmatites, which is where Snelling's model puts them.  What separates Wakefield and Snelling is the clock, not the rock.

"The Odom letter is Gentry's word.  It proves nothing." It is Gentry's reproduction of a private letter, never confirmed with either author.  What does not depend on it is in the paper itself:  three halos, in quartz, around monazite, with the mica extension labelled "We strongly suspect".

Skeptic:  "This is a 1970s curiosity.  Nobody in geology cares." Half right.  The subject has produced almost no mainstream research since 1989, and the refutation everyone cites was an accident its author did not stand behind.  But the halos are real, independently confirmed, and uranium-free to parts per million.  Dalrymple under cross-examination in Arkansas called them "a very tiny mystery which is balanced on the other side by an enormous amount of evidence".³  That is the right proportion.  It is not the same as no mystery.  The experiment that would close it costs a photometer.

Issues and unanswered questions

• Laney and Laughlin 1981 has been read only in abstract. The 150 °C ceiling should be checked against the paper's depth-temperature table.
• Gentry's reply to Wakefield in the Journal of Geological Education, printed beside the article at his request, is his only reply in a refereed venue and has not been read for this article.
• The supply-rate calculation in Part 4 is this article's estimate from round numbers and needs real zircon sizes and uranium contents from RATE's data.
• Brawley's 2 to 1 photometric test has apparently never been run. It would settle the radon question with existing slides.
• Feather's 1978 review is Snelling's authority for biotites in which ²¹⁸Po halos outnumber ²¹⁰Po halos. It has not been obtained, and it is the one published claim that would break the census argument.
• Moazed, Spector and Ward's full ring-radius table could not be recovered; only the two numbers quoted are confirmed.
• The Thunderhead Sandstone test is reported without counts. The strong test is whether the isotope ratio matches the model's prediction, and that comparison has not been published.
• The Snelling and Armitage page numbers are ICR's reprint pages, repaginated 1 to 23 against a printed 243 to 267.
• The Odom letter exists only in Gentry's book. A direct enquiry to W. J. Rink, its surviving co-author, would settle it.

Verdict

Polonium halos are real, uranium-free, and unexplained in detail.  That much Gentry established and an independent laboratory confirmed.  Everything he built on it has failed:  his own 1976 paper, the field geology of his three Canadian sites, the fact that only the three polonium isotopes a radon gas can deliver ever make halos, and his own movement's 2003 count of two polonium-218 halos in 2,011 under the word "falsified".

What replaced it is the mainstream mechanism with accelerated decay added.  Wakefield and Snelling agree on the mechanism and disagree on the clock, and the halos cannot decide the clock, because a halo records dose and not time.  The RATE version still owes answers to Snelling's own 67,000 to 1 prediction, to the single-ring polonium-210 halos no radon route can build, and to Meier and Hecker's cold-water model.

The honest residue is small and real.  Some three-ring halos exist, nobody has proven they are radon halos, and the experiment that would prove it has sat unrun since 1992.  That is a tiny mystery about ring identification in mica.  It is not evidence about the age of the earth.

Sources