Monday, January 20, 2014

Question from a Reader: Radioactive Dating, Part 2

← Read Part 1 for an explanation of how radiometric dating works.

Critical Assumptions

In order for the math of radiometric dating to work out, two factors need to be known: the half-life of the radioactive isotope and the proportion of the parent isotope remaining in the sample.  At first glance, it seems like these numbers can be measured in a lab and then applied to the mathematical formula, with little room for error or interpretation.  However, there are three important and unproven assumptions that must be made in order to do this.

Initial amounts
The ability of scientists to measure minuscule amounts of isotopes may seem far-fetched to some people, but, in my opinion, isotope-ratio mass spectrometry is founded well enough in operational science that it is not worth criticizing.  The technique uses well-established and tested properties of chemistry and physics to determine the exact isotopic composition of any material or rock sample.

The present amount of parent and daughter isotopes is not enough to calculate an age, however.  Remember, the remaining proportion of the parent isotope is needed, which is calculated by dividing the present amount by the starting amount.  Unfortunately, the starting amount cannot be measured, because it is in the past!  Sometimes it is assumed that all of the daughter isotope formed from the parent, as in the cases of K-Ar (potassium-argon) dating and U-Pb (uranium-lead dating), so that the amount of daughter isotope represents the amount of decayed parent isotope.  While this assumption is loosely based on chemical properties (for example, argon shouldn't be present in forming minerals, as it does not react), it is far from proven.  In fact, scientists appear to recognize this, as seen on TalkOrigins' page refuting creationist claims about radiometric dating.  A study (Baadsgaard et al., 1993, not available without subscription) is cited that supposedly demonstrates the consistency of multiple radiometric methods.  For one of the dated volcanic layers, the U-Pb date was "strongly discordant" (that is, wildly off from what was expected), so the authors dismissed it, claiming that it "inherited" some radioactively-produced (radiogenic) lead from previously-melted older crystals.  All other methods matched the expected dates, so no "inheritance" was claimed for those.

In an attempt to get around this issue, a technique called isochron dating is sometimes used.  This method compares the amounts of radiogenic daughter isotope, nonradiogenic isotope of the same element, and parent isotope across multiple samples of the same age.  In theory, this method does not require the starting amounts, and in fact it can be used to determine them.  However, it still relies on several assumptions.  Apart from the closed system and constant rate assumptions, which will be discussed later, isochron dating assumes that all samples were formed at the same time and that the daughter isotope was uniformly distributed across all samples.  Answers in Genesis also notes that a paper from 1989 demonstrated that false isochron dates can be given when two rock units are mixed.  TalkOrigins responded to this claim, saying that such instances can be identified by creating a "mixing plot" of the data; however, there is no indication of how often this test is actually used in scientific studies.

Carbon dating has a unique problem with the assumption of initial amounts.  The radioactive isotope of carbon, carbon-14, is formed in the upper atmosphere by radiation from the sun.  Over time, it decays into nitrogen-14, which is extremely common.  Instead of comparing the parent and daughter isotopes, scientists using carbon dating compare carbon-14 to its more stable and common isotope, carbon-12.  The C-14/C-12 ratio in atmospheric carbon dioxide can be measured.  Interestingly, because C-14 is slightly heavier than C-12, this ratio is slightly lower in plants that take in carbon dioxide.  The ratio is again reduced in animals that eat the plants, and reduced further in animals that eat those animals.  Nevertheless, the C-14/C-12 ratio in the organisms is based on the atmospheric ratio.  Therefore, when a wood sample is dated, the initial C-14/C-12 ratio is assumed to be a particular factor of the atmospheric ratio at the time the plant died, which in turn is assumed to be the same as the current atmospheric ratio.  However, this assumption is only valid if the atmospheric ratio has remained constant (in equilibrium).  Dr. Willard Libby, inventor of the radiocarbon dating technique, noted that atmospheric carbon-14 should take 20,000–30,000 years to reach equilibrium, yet that it does not appear to have done so.  He attributed this discrepancy to experimental error (I once found a copy of his original work online, but it appears to no longer be available, so I apologize for the lack of source).  It follows, then, that if the earth is significantly less than 20,000 years old, as creationists claim, then carbon dating will not give accurate results, particularly for older specimens that lived under more variable atmospheric C-14 levels.

Closed system
The second critical assumption is that rocks are closed systems; that is, once they solidify, nothing can enter or leave the rock, so that any change in chemistry is the result of radioactivity as long as the rock stays below a certain temperature.  Curiously, this assumption is well-known to be untrue to an extent, yet it is applied to dating rocks anyway, because it is "close enough."  Fossilization, which involves replacing original molecules with surrounding rock, could not occur if rocks and bones were closed systems.  Furthermore, when a radiometric date greatly differs from the expected age, contamination is often blamed, which could not happen if rocks were closed systems.

Evolutionists respond to these criticisms by claiming that repeated and consistent dating results demonstrate that the closed system assumption is valid when care is taken by the scientists, and that isochron dating can correct for contamination.  I could spend weeks relaying all of the examples of consistent dating by evolutionists and inconsistent dating by creationists, so let us be content with leaving the question of consistency up in the air for now.  Perhaps I will address some specific cases later in this series.

Continue to Part 3 to read about the greatest assumption of constant decay rate →

Feel free to leave feedback in the comments section.  Remember, if you have a question you want answered, email it to!

No comments:

Post a Comment