Carbon-14 Where It Shouldn't Be

Measurable C-14 in coal, diamonds, and dinosaur bones — when after a hundred thousand years there should be none at all.

This is, in my reading, the cleanest of the four arguments. The numbers are precise, the measurements are made in major university laboratories with the best contamination protocols available, the prediction is unambiguous, and the result is the opposite of what deep time requires.

The setup

Carbon-14 is a radioactive isotope of carbon with a half-life of 5,730 years. It is produced in the upper atmosphere by cosmic rays interacting with nitrogen-14, mixed into atmospheric CO₂, and incorporated into living things through the carbon cycle. When an organism dies, it stops taking in new carbon, and the C-14 in its tissues begins to decay back into nitrogen.

Because the half-life is so short relative to geological time, the C-14 method is only useful for radiometric dating up to about 50,000 years. After that, the C-14 left in a sample is too close to zero to measure against laboratory background.

After 100,000 years — about eighteen half-lives — the C-14 in a sample is reduced to roughly \(10^{-5}\) of its original value: about one part in a million. This is the practical detection limit for modern accelerator mass spectrometry (AMS), which is the most sensitive technique we have for counting C-14 atoms directly.

After 1,000,000 years — about a hundred and seventy half-lives — the original C-14 is reduced by a factor of \(10^{-52}\). There is not, statistically, a single atom of original C-14 left in the sample. Detectable C-14 in a million-year-old sample must come from somewhere else: lab contamination, in-situ neutron capture on nitrogen impurities, or recent biological contamination.

That is the prediction.

The standard chronology says coal seams are tens to hundreds of millions of years old. Diamonds are billions of years old. Dinosaur bones are tens of millions of years old. None of these should contain any measurable C-14.

They all do.

What’s been measured

A research initiative called the RATE project (Radioisotopes and the Age of the Earth), conducted between 1997 and 2005 under the Institute for Creation Research, set out to measure C-14 in materials that, by the standard chronology, should have none.¹ The samples were sent to commercial AMS labs that had no idea what the samples were or where they came from — they were just told to measure the C-14 fraction. The results:

• Coal: ten US coal samples from three geological eras (Eocene, Cretaceous, Pennsylvanian — supposed ages 40–311 Ma) consistently returned C-14 fractions of \(0.1\%\) to \(0.5\%\) of modern carbon. The Eocene samples, the youngest, averaged \(0.27\%\). The Cretaceous samples averaged \(0.22\%\). The Pennsylvanian samples — supposedly three hundred million years old — averaged \(0.13\%\). All of these are many orders of magnitude above the AMS background, which is on the order of \(10^{-5}\%\).²
• Diamonds: twelve natural diamonds from three different African mines, with conventional ages between 1.0 and 2.7 billion years, returned C-14 fractions between \(0.01\%\) and \(0.07\%\) of modern carbon — also well above the AMS background. Diamonds are particularly significant because they are exceptionally hard, chemically inert, and resistant to surface contamination; what comes out of an AMS measurement of a diamond is, almost certainly, the C-14 that was in the diamond to begin with.³

Subsequent measurements by other researchers (Kirby, Aardsma, Snelling) have confirmed the pattern. The CalAMS lab at U.C. Irvine and the AMS facility at the University of Arizona both report “diamond-blank” backgrounds far below the values found in samples sent in blind.⁴

To put numbers on this: the C-14 in the diamonds, if it represents the true age of the carbon, gives an apparent radiocarbon age in the 55,000 to 80,000 year range. The diamonds are supposed to be more than a billion years old. The C-14 method says they are between fifty-five thousand and eighty thousand. The two numbers disagree by a factor of about 20,000.

The standard response

The mainstream response is, in three parts:

Part one — contamination. The C-14 must be modern carbon that contaminated the sample during processing.

This is the most common rebuttal and the easiest to refute. AMS labs measure their own backgrounds using “process blanks” — diamonds, anthracite, or other materials they expect to have zero C-14 — and the measured backgrounds are typically in the range of \(10^{-5}\%\) to \(10^{-4}\%\) of modern carbon. The C-14 in the coal and diamond samples is one hundred to ten thousand times above the lab’s own contamination floor. Contamination is real, but it cannot account for the levels measured.

Part two — in-situ generation. The C-14 must have been generated inside the sample, after deposition, by neutron capture on nitrogen impurities. Natural radioactivity in the surrounding rock produces a flux of neutrons; nitrogen-14 plus a neutron gives C-14 plus a proton, with a sensible cross-section.

This is a more serious move, and it explains some of the data in some materials. But the calculation does not give enough C-14 to match what is measured. RATE researchers and others have computed the in-situ neutron flux required to generate the measured C-14 in coal and diamonds, and the required flux is orders of magnitude higher than the actual measured background neutron flux in those geological settings.⁵ In-situ generation is a real effect at a low level; it does not get you to the levels observed.

Part three — the C-14 calibration itself is wrong at these levels. Some researchers have argued that AMS measurements near the detection limit are intrinsically unreliable and that all the above results should be discounted.

This argument has the awkward feature that the same labs, using the same equipment and the same protocols, are trusted to date samples for archaeologists, anthropologists, and oceanographers every day, including samples whose measured C-14 fractions are well within the same range as the disputed coal and diamond results. You cannot reasonably hold that the lab’s measurements are reliable for samples whose age you accept and unreliable for samples whose age you do not. The protocol does not change.

What it actually says

The C-14 in coal and diamonds is, in my reading, the single strongest empirical argument for a young earth that the standard scientific apparatus has produced. Not because creationists manufactured it — the measurements were done at mainstream labs, using mainstream techniques — but because the measurements are there, and they say what they say.

Taking the data at face value:

• Coal supposedly 311 Ma old contains C-14 corresponding to a radiocarbon age of about 50,000 years.
• Diamonds supposedly 2 Ga old contain C-14 corresponding to a radiocarbon age of about 70,000 years.
• The discrepancy is not 10%. It is not 50%. It is a factor of \(10^4\) to \(10^5\).

If the C-14 is real (and the contamination rebuttals do not account for it), then either deep-time dating is wrong by four orders of magnitude, or there is a physical mechanism we have not identified that produces measurable C-14 from non-C-14 in coal and diamonds at rates far higher than known nuclear processes can explain.

Of those two options, the first is, by a substantial margin, the simpler. The young-earth reading is consistent with what the AMS labs actually find. The deep-time reading requires us to either discount the data or invent new physics.

Next: the persistence of comets, a fifth argument from a different field entirely — planetary dynamics — that converges on the same upper bound, and gathers the cumulative force of all five at the end.