Essay:Radiometric dating - a non-Creationist Perspective

From Conservapedia
Jump to: navigation, search

See also Counterexamples to an Old Earth.

Radiometric dating is a method of determining the approximate age of an artifact by measuring the amount of radioactive decay that has occurred.[1] Radiometric dating is mostly used to determine the age of rocks, though a particular form of radiometric dating—called Radiocarbon dating—can date wood, cloth, skeletons, and other organic material. Radiometric dating requires careful analysis and control over the isotopic mix of atoms in the original sample, as well as careful analysis and control of factors (e.g. chemical changes) that might have changed the amounts of the various elements in the sample during the decay interval. These difficulties are considerable, and are discussed below. It also requires knowledge of the rates at which various isotopes decay. These rates are known to great accuracy.

Because analysis of the various control variables that could affect the chemical composition of the sample during the decay period often depends on shrewd guesswork, radiometric dating as a whole could be said to fail the standards of testability and falsifiability, and so claims based on radiometric dating may fail to qualify under the Daubert standard for court-admissible scientific evidence. The underlying decay rates, on the other hand, are completely testable and falsifiable.

Radiometric dating is more accurate for shorter time periods (e.g., hundreds of years), during which control variables are less likely to change.

Due to so many different kinds or radiometric dating in use (i.e Radiocarbon dating, also called carbon dating, Potassium-argon dating, Uranium-lead dating, Uranium-thorium and Rubidium-strontium dating), all of which concur with each other when dating objects, it is extremely unlikely that all of them are wrong in the exact same way. Therefore, support for radiometric dating is virtually universal in the scientific community.

Key assumptions

There are a number of assumptions involved in radiometric dating with respect to long time periods.

Initial quantities

One key assumption is that the initial quantity of the parent element can be determined. With uranium-lead dating, for example, the process assumes the original proportion of uranium in the sample is known with reasonable accuracy. One assumption that can be made is that all the lead in the sample was once uranium, but if there was lead there to start with, this assumption is not valid, and any date based on that assumption will be incorrect (too old). So care is required.

In the case of carbon dating, it is not the initial quantity that is important, but the initial ratio of C14 to C12, but the same principle otherwise applies.

Recognizing this problem, scientists try to focus on rocks that do not contain the decay product originally. For example, in uranium-lead dating, they use rocks containing zircon (ZrSiO4), though it can be used on other materials, such as baddeleyite.[2] Zircon and baddeleyite incorporate uranium atoms into their crystalline structure as substitutes for zirconium, but strongly reject lead. Zircon has a very high closure temperature, is very chemically inert, and is resistant to mechanical weathering. For these reasons, if a rock strata contains zircon, running a uranium-lead test on a zircon sample will produce a radiometric dating result that is less dependent on the initial quantity problem.

Rate of decay

Another assumption is that the rate of decay is constant over long periods of time. Radiometric dating requires that the decay rates of the isotopes involved be accurately known, and that there is confidence that these decay rates are constant. Fortunately, this is the case. The physical constants (nucleon masses, fine structure constant) involved in radioactive decay are well characterized, and the processes are well understood. Careful astronomical observations show that the constants have not changed significantly in billions of years—spectral lines from distant galaxies would have shifted perceptibly if these constants had changed. In some cases radioactive decay itself can be observed and measured in distant galaxies when a supernova explodes and ejects unstable nuclei. Indirect observations can allow us to infer radioactive decay rates over time scales that are quite long. For example, we can measure gamma radiation rates at specific frequencies from distant supernovae and compare this to the rate expected for the mass of the star. This has given rates for supernovae as distant as 169,000 light years which are consistent with those measured today. Thus it would seem decay rates have been the same for at least the past 169,000 years.[3][4] This gives considerable confidence that the decay rates, and the physical constants that determine them, have not changed perceptibly.

Some people, perhaps in support of a Creationist viewpoint, have suggested that decay rates have changed significantly because "energy levels" have changed significantly. Whether electron energy levels or nucleon energy levels are being referred to, this is simply not true.

There are a few effects that can alter radioactive half-lives, but they are mostly well understood, and in any case would not materially affect the radiometric dating results. That is, the analysis of the isotopic and chemical composition of the sample has far greater uncertainty than any uncertainty in the decay rate itself.

The major reason that decay rates can change is that the electric field, from the atom's electron cloud, can change due to chemical changes. That is, electrons can move closer to or farther away from the nucleus depending on the chemical bonds. This affects the coulomb barrier involved in Alpha decay, and therefore changes the height and width of the barrier through which the alpha particle must tunnel. The effect of this on alpha decay, which is the most common decay mode in radiometric dating, is utterly insignificant.

There is another effect that takes place in the "electron capture" type of Beta decay. This is an example of the Weak force, and is fairly rare. Electron capture requires that there be an electron in the vicinity of the nucleus, so its activity depends strongly on the configuration of the electron cloud, which depends on the chemical state. In fact, it is possible to shut down electron capture completely—simply ionize the substance so that there are no electrons nearby.

There is a fairly well-known example of chemical state affecting electron capture activity. The 7Be nucleus (Beryllium-7) is an electron capturer with a half-life of about 53 days, turning into Lithium-7. The variation is about 1.5%. While this half-life is way too short to be useful for radiometric dating, the effect of the chemical state is noticeable. The reason is that, because the atomic number is only four, the 2s valence electrons are very close to the 1s electrons involved in capture.[5]

There is another effect, that is not understood. It appears that some radioactive decays are affected the Sun, and fluctuate over a period of about 33 days as the Sun rotates. The variation is about one tenth of a percent. It has been observed in silicon-32 and chlorine-36. While it is not understood why this happens, it would average out over long time periods and therefore not affect the final result.

Creationists cast doubt on this analysis, and some apparently believe that decay rates could be so far off (by a factor of millions or more) that it could explain observations pointing to an old-Earth time-frame while young-Earth cosmology was actually taking place. They cite large numbers of articles on Creationist web sites.[6] While most of these articles accept that the possible discrepancies are nowhere near large enough to refute old-Earth cosmology, the Woodmorappe paper does make that claim. It uses the phenomenon of ionization, described above, to shut down electron capture decay, which could indeed cause a billion-fold discrepancy. However, this ionization would not have taken place under real-world circumstances. The Walker and Knapp articles refer to the noticeable discrepancies in Beryllium-7, described above.

Creationists also suggest that decay rates were almost certainly not constant near the creation or beginning of the universe. However, the billions of years of Uranium decay, for example, did not take place near the beginning of the universe.

See Half-life for an explanation of the exponential decay involved in radioactivity, and the meaning of the term "half-life". The exponential decay pattern is the same for all kinds of nuclear radiation—alpha decay, beta decay, and gamma decay.

This governs what is known as the "decay rate." The rate is unique to different particles and to different atomic elements. This makes different elements useful for different time scales of dating; an element with too short an average lifetime will have too few particles left to reveal much one way or another of potentially longer time scales. Hence, elements such as potassium, which has an average lifetime of nearly 2 billion years before decaying into argon, are useful for very long time scales, with geological applications such as dating ancient lava flows or Martian rocks. Carbon, on the other hand, with a shorter mean lifetime of over 8000 years, is more useful for dating human artifacts.

Outside influences

It is important that the sample not have had any outside influences. One example of this can be found in metamorphic rocks.[7] This does not mean that all rock samples are unreliable, but it is possible to account for a process which throws off the data for metamorphic rocks.

For example, with Uranium-lead dating with the crystallization of magma, this remains a closed system until the uranium decays. As it decays, it disrupts the crystal and allows the lead atom to move. Likewise, heating the rock such as granite forms gneiss or basalt forms schist. This can also disrupt the ratios of lead and uranium in the sample.


In order to calibrate radiometric dating methods, the methods need to be checked for accuracy against items with independently-known dates.

Carbon dating, with its much lower maximum theoretical range, is often used for dating items only hundreds and thousands of years old, so can be calibrated in its lower ranges by comparing results with artifacts whose ages are known from historical records.

Scientists have also attempted to extend the calibration range by comparing results to timber which has its age calculated by dendrochronology, but this has also been questioned because carbon dating is used to assist with working out dendrochronological ages.

Otherwise, calibration consists of comparing results with ages determined by other radiometric dating methods.

However, tests of radiometric dating methods have often shown that they do not agree with known ages of rocks that have been seen to form from volcanic eruptions in recent and historic times, and there are also examples of radiometric dating methods not agreeing with each other.

Young earth creationists therefore claim that radiometric dating methods are not reliable and can therefore not be used to disprove Biblical chronology.

Acceptance and reliability

Radiometric dating methods are widely quoted by scientists, giving, for example a 13.7 billion year age for the universe and a 4.5 billion year age for the Earth. Creationists suggest an age for the universe and the Earth of about 6-10 thousand based on the Bible[8][9].

As with all scientific endeavors, one needs to be careful in interpreting the data. Carbon-14 dating is particularly subject to misinterpretation, because biological processes are involved.

C14 dating was being discussed at a symposium on the prehistory of the Nile Valley. A famous American colleague, Professor Brew, briefly summarized a common attitude among archaeologists towards it, as follows:
"If a C14 date supports our theories, we put it in the main text. If it does not entirely contradict them, we put it in a footnote. And if it is completely 'out of date', we just drop it."
Few archaeologists who have concerned themselves with absolute chronology are innocent of having sometimes applied this method...[10]

A geological guidebook published by the Queensland government acknowledges that the dates are not absolute, but must be interpreted:

Also, the relative ages [of the radiometric dating results] must always be consistent with the geological evidence. ... if a contradiction occurs, then the cause of the error needs to be established or the radiometric results are unacceptable[11]

One example of scientists not accepting radiometric dates is that of Mungo Man, a human fossil from New South Wales. When originally found, it was dated by radiocarbon dating at around 30,000 years old. This was later revised to 40,000 years. Another scientist later used other methods to derive a date of 62,000 years. The original discoverer, unconvinced by this result, used a different method again, and again came up with a date of 40,000 years. So radiocarbon dating is really not very precise.

The fallibility of dating methods is also illustrated by the fact that dating laboratories are known to improve the likelihood of getting a "correct" date by asking for the expected date of the item. For example, the Sample Record Sheet for the University of Waikato Radiocarbon Dating Laboratory asks for the estimated age, the basis for the estimate, and the maximum and minimum acceptable ages.[12]

Some major methods of radiometric dating

There are several major types of radiometric dating in use:[13][14]

  1. Radiocarbon dating, also called carbon dating
  2. Potassium-argon dating
  3. Uranium-lead dating
  4. Uranium-thorium
  5. Rubidium-strontium dating

Explanation by Analogy

No method exists for measuring time, except by measuring it as it is passing. Therefore, the age of an artifact must be calculated.

The basic principle in any dating method is to find a process that is occurring at a measurable rate and which is causing a change, measure the rate of that process, work out what state the artifact was in at the beginning of the process, observe what state it is in now, and to calculate how long the process at the measured rate would need to occur to effect that change.

For example, to work out how long a candle has been burning, the following steps would be needed:

  1. Measure how long it takes the candle to burn down a given amount.
  2. Find out how long the candle was when it started burning.
  3. Measure how long the candle is now.
  4. Calculate the difference between the two lengths.
  5. Calculate how long it would need to burn in order to burn that length.

For most radiometric dating methods, one radioactive element changes by a process of nuclear decay into another element (often through a number of intermediate steps). For example, uranium will eventually decay into lead. So to measure how old a specimen containing some uranium and some lead is, the following steps are required:

  1. Measure the decay rate of uranium.
  2. Find out how much uranium was in the specimen to start with (this might be done by assuming that all the lead was originally uranium).
  3. Find out how much uranium is in the specimen now.
  4. Calculate how much uranium has turned into lead.
  5. Calculate how long it would take that much uranium to turn into lead, given the measured rate. Calculations involve the well-established function for exponential decay: .


Notes and references

  1. Radiometric Time Scale USGS
  2. Wingate, M.T.D. (2001). "SHRIMP baddeleyite and zircon ages for an Umkondo dolerite sill, Nyanga Mountains, Eastern Zimbabwe". South African Journal of Geology 104 (1): 13–22. doi:10.2113/104.1.13. 
  5. Huh, C.-A., Dependence of the decay rate of 7Be on chemical forms, Earth and Planetary Science Letters 171:325–328, 1999.
  6. Multiple references:
  7. Radiometric Dating Course notes for EENS 211 at Tulane University
  10. T. Säve-Söderbergh and I. U. Olsson, ‘C14 dating and Egyptian chronology’, in Radiocarbon Variations and Absolute Chronology, Proceedings of the Twelfth Nobel Symposium, Ingrid U. Olsson (editor), Almqvist & Wiksell, Stockholm, and John Wiley & Sons, Inc., New York, p. 35, 1970. Quoted in Lamb, 2007.
  11. Walker, 2002
  12. University of Waikato Radiocarbon Dating Laboratory Sample Record Sheet
  13. Walker, 2005
  14. Faure and Mensing, 2005

See also