A Consideration of the Potassium-Argon Radiometric Method for

Chemical Education Today

Letters A Consideration of the Potassium–Argon Radiometric Method for Dating Minerals In a recent New York Times editorial castigating intelligent design (1), Paul Krugman laments that “intelligent design doesn’t have to attract significant support from actual researchers to be effective. All it has to do is create confusion, to make it seem as if there really is a controversy about the validity of evolutionary theory.” In a similar vein, William A. Howard’s recent Journal of Chemical Education article “The Relationship between Balancing Reactions and Reaction Lifetimes: A Consideration of the Potassium–Argon Radiometric Method for Dating Minerals” (2) is written to create confusion in his primary audience— undergraduate chemistry students—by implying there is a controversy about the validity of radiometric dating. Instead of its purported objective of “stimulating critical thinking”, the article is itself a caricature of critical thinking, shot through with straw man arguments, false analogies, and red herrings. Howard begins with a straw man argument: There is controversy about the validity of radiometric dating. Though Howard states the “purpose of this paper is not to suggest that mineral ages determined by the potassium–argon method are erroneous”, the entire paper is designed to cast doubt on the accuracy of radiometric dating. Indeed, one of his conclusions is that “…many of those involved with the teaching of biological evolution in public schools erroneously believe that scientists understand radioactive decay reactions very well….The reality is that no one understands these reactions entirely; the present exercise in critical thinking underscores this truth.” Radiometric dating is “controversial” only to young earth creationists, who continually misuse the technique (for examples, see refs 3 and 4). To cast doubt on the accuracy of K–Ar dating, Howard sets up nonsense criteria—red herrings—that inculcate doubt in his audience, but are irrelevant. He also resorts to false analogies. Red herring #1. Howard’s complaint is that “…nobody knows what the actual complete potassium–argon reaction is!” and “If a geochronologist truly knows a mineral’s age, then that geochronologist should be able to describe the chemistry behind the radioactive decay of 40K in that mineral in great detail. If the geochronologist is not able to do so, then the mineral’s age is not really known.” This is nonsense. The only requirement to get a good radiometric K–Ar date is that the system be closed to the isotopes involved in dating. Brent Dalrymple, author of many books and journal articles in this field, offers the following assessment: Like all radiometric methods, the K–Ar method does not work on all rocks and minerals under all geologic conditions. By many experiments over the past three decades, geologists have learned which rocks and minerals act as closed systems and under what conditions they do so (5, p 93). Those of us who develop and use dating techniques to solve scientific problems are well aware that the systems www.JCE.DivCHED.org

•

are not perfect, and we have provided numerous published examples in which the techniques fail. We often test them under controlled conditions to learn when and why they fail so we can avoid using them incorrectly” (6, p 80).

Nowhere in Howard’s article does he reveal anything that affects the 40K:40Ar ratio, the only quantity necessary to determine this age. He implies that there may be “unknown” ways of argon entering or leaving the rock, but provides no specifics, let alone references to scientific research. Dalrymple has already addressed this problem: “Some cases of initial 40Ar remaining in rocks have been documented but they are uncommon, as noted by Dalrymple and Lanphere, who also described studies of historic lava flows showing ‘excess’ argon is rare in these rocks” (5, p 92). Red herring #2. Howard complains that only the argon is quantified by geochronologists; the “identities and quantities of the 40Ca products are unknown.” Dalrymple again, regarding 40Ca, “…it is usually not possible to determine the amount of 40Ca present initially, and the 40K–40Ca method is rarely used for dating.” Red herring, non-problem. Red herring #3. The equations in Scheme 2 and the ensuing discussion. Howard’s two “very important facts” from these equations are: (1) Some compounds with unusual oxidation states are probably produced; and (2) 40Ar is only a “minor product” in the reaction. These “facts” may sound important to undergraduates, especially those who are predisposed by their religion to reject radiometric dating, but what Howard fails to explain is that the small amounts of 40 Ar are completely analyzable by modern analytical techniques! Furthermore, the fact that some unusual oxidation states may be produced in no way affect the 40K:40Ar ratio, the only quantity relevant in the type of radiometric dating. Howard’s own equations demonstrate this. False analogy #1. The corn in the grocery store. Howard states: “Just as the manager must count cans of corn and discarded cartons in order to judge how long the stocker had been working, so must the chemist identify and quantify all products from a reaction in order to determine how long the reaction has been going.” No—this grocery store is locked! K–Ar dating is only performed on rocks that are closed systems. The only way the “corn” gets on the shelf is by the regular decay of 40K, which Howard admits occurs at a known rate. Ideally no one can remove this “corn”, but if the “corn” does get removed, there are ways of detecting this, and the radiometric age may still be determined using the 40Ar–39Ar method detailed by Dalrymple and others. False analogy #2. Corn, carrots, and beans in the grocery store. Howard diminishes the importance of using multiple dating methods (italics mine): “In order to build confidence in their work, geochronologists will occasionally determine mineral ages by more than one dating method…. Many times, the independent dating methods yield very similar ages, usually within experimental error, and so one may think that such results verify the accuracy of the age determination.” This thinking is specious according to Howard, because the “complete chemistries” of uranium–lead and ru-

Vol. 83 No. 4 April 2006

•

Journal of Chemical Education

545


Letters bidium–strontium dating are not known, either! He asserts that one only improves the precision, but not the accuracy, of the radiometric date. This is a worse analogy than the first. The grocery store is still closed (or you don’t use corn, carrots, and beans to date anything) and it ignores the fact that the discarded cartons would at least be labeled “corn”, “carrots”, and ‘beans” (i.e., there would be distinct parent nuclides)! Howard might have a point if exactly the same assumptions about initial conditions were made in all three dating methods (Rb–Sr and U–Pb—the only additional ones Howard lists). In fact, the chemistries of the K–Ar, Rb–Sr, and U–Pb methods are totally different. The fact these disparate methods have led to convergent ages provides powerful support for radiometric dating. Howard claims this paper “provides a detailed examination of a commonly accepted practice in geology” and “offers an example of how to stimulate critical thinking”. Instead, the paper is a caricature of how radiometric dating is performed and how successful it actually is. Rather than stimulating critical thinking, it is itself a great case study in logical fallacies. The Journal of Chemical Education should pull this article from its Web site. You goofed. Literature Cited 1. Krugman, P. Design for Confusion. The New York Times, August 5, 2005, p A15. 2. Howard, W. A. J. Chem. Educ. 2005, 82, 1094–1098. 3. Stassen, C. Criticism of the ICR’s Grand Canyon Dating Project. 1997. Available from http://www.talkorigins.org/faqs/ icr-science.html#sec2 (accessed Feb 2006). 4. Bartelt, K. E. A Visit to the ICR. 1998; Available from http:// www.talkorigins.org/faqs/icr-visit/bartelt1.html#introduction (accessed Feb 2006). 5. Dalrymple, G. B. The Age of the Earth; Stanford University Press: Stanford, CA, 1991. 6. Dalrymple, G. B. Ancient Earth, Ancient Skies; Stanford University Press: Stanford, CA, 2004. Karen E. Bartelt Department of Chemistry Eureka College Eureka, IL 61530

The author replies: I would like to thank Dr. Bartelt for her interest in my work and for taking the time to write her letter. Unfortunately, Bartelt’s letter is emotional, confrontational, and intimidating. Moreover, her statements lead me to wonder if she actually understood the purpose and the science behind my article. My response to her letter has three objectives: (1) I would like to clarify the point of contention between Bartelt and myself for the readers; (2) I would like to address a scientific question inspired by her letter; and (3) I would like to comment briefly on the importance of questioning wellestablished fields of science. 546


•

1. The Point of Contention The real issue is this: Given the task of solving a problem, is it possible to gather too much data or to perform too many experiments? My answer to this question is that scientists should collect as much information about the nature of the problem as possible before drawing conclusions. Chemistry is subtle, and chemists can be tricked into believing results that are not true. The best defense against being deceived is to be thoroughly familiar with the scientific literature, to get as much data as possible, and to try to reproduce results a number of times. There are no unnecessary experiments. It is better to perform an irrelevant experiment than not to perform the experiment and then later discover that the results would have been very important. In stark contrast, Bartelt seems to think that gathering extra data is not important. She has stated that the 40K:40Ar ratio alone is necessary for accurately assessing a mineral’s age. Bartelt does not see the need to learn more about the unknown chemistry resulting from the decay of the 40K nucleus in minerals. She does not comprehend how a knowledge of the stoichiometry of the reaction may aid in the interpretation of the age of the mineral. Therefore, her view is that balancing the potassium–argon reaction in minerals is a red herring and does not need to be done. I respectfully but firmly disagree. If a geologist determines that a mineral is a closed system suitable for potassium–argon dating, then the potassium–argon reaction in that mineral should balance. Identifying and quantifying the products could be used to corroborate the geologist’s claim that the mineral is indeed a closed system.

2. Is the 40K:40Ar Ratio the Only Quantity Necessary for Determining a Mineral’s Age by the Potassium– Argon Method? Bartelt’s assertion that the 40K:40Ar ratio is the only quantity necessary for determining a mineral’s age is correct, as long as the mineral is a closed system and there was not a significant quantity of 40Ar homogeneously distributed in the mineral at time zero. (For a discussion of minerals known to be enriched in 40Ar at time zero, see ref 1.) Consider dating a sanidine mineral (KAlSi3O8) by the potassium–argon method. For the sake of argument, we shall adopt the following conditions for this hypothetical experiment: (1) a team of qualified geologists agree that this sample of sanidine is a perfectly closed system and an excellent candidate for potassium–argon dating; (2) there are no phenocrysts or xenoliths trapped in this sample; (3) there is no air contamination of the sample; (4) there are no redox-active impurities; and (5) this sample of sanidine is protected from all outside environmental influences. A hypothetical balanced reaction is 100 40KAlSi3O8 → 11 40Ar + 78 40Ca(AlII)Si3O8 + 11 40CaAl2Si6O16

This reaction is balanced with the understanding that 40K converts into 40Ar 11% of the time and into 40Ca 89% of the time. It is mathematically impossible to balance the po-


•

www.JCE.DivCHED.org


Letters tassium–argon reaction without forming a reduced product. In the reaction shown above, that reduced product is divalent aluminum in 40Ca(AlII)Si3O8. (Of course, it is not really known that divalent aluminum is a product since the reaction shown above is conjecture, and the real potassium–argon reaction in sanidine is unknown. Divalent aluminum is presented as a possible product because naturally occurring samples of paramagnetic quartz doped with aluminum are known, and aluminumbased electron spin resonance signals have been used to date such materials (2).) The stoichiometric AlII:40Ar ratio is 78:11. If the geochronologist identifies and quantifies the products from the reaction and actually finds an AlII:40Ar ratio reasonably close to 78:11, then the stoichiometry confirms that the 40Ar found within the mineral really is radiogenic. In this case, the geologists were right to certify this sample as an excellent candidate for potassium–argon dating. If the geochronologist finds an AlII:40Ar ratio significantly different from 78:11 however, then there is a problem! There must be an explanation for why the expected 78:11 ratio was not found. Perhaps one or more of the five conditions listed above was not met, despite the approval of the geologists! Or, perhaps the AlII was oxidized to AlIII by some unknown mechanism. (This oxidation does not really solve the problem however, because that odd electron had to go somewhere and the reaction must still be balanced!) In this case, it is not clear that the mineral age can be accurately determined from the 40K:40Ar ratio, and using the 40K:40Ar ratio as the only quantity for determining the mineral’s age may or may not be correct. So, is the 40K:40Ar ratio the only quantity necessary for determining a mineral’s age by the potassium–argon method? Provided that the sample was certified by geologists as appropriate for potassium–argon dating, Bartelt’s answer would be affirmative, and she would strongly object to gathering any additional information. My answer, on the other hand, is that I don’t know. There are too many unknowns, and I would like to learn more about the chemistry resulting from the decay of 40K in sanidine.

3. Comment on Questioning Well-Established Fields of Science Radiometric dating is a well-established field of science, but this fact should not be used to intimidate someone from asking questions. Scientific knowledge advances only when well-established practices are questioned, and questioning radiometric dating is the responsibility of all professional scientists everywhere. When teaching chemistry, I introduce my students to a number of theories, such as Atomic Theory, Quantum Theory, Valence Bond Theory, Molecular Orbital Theory, Crystal Field Theory, and more! These theories are well-established and supported by a great deal of experimental evidence. Nevertheless, I encourage my students to question the experimental support and to think of new experiments that could potentially falsify the theories. This activity is extremely important for producing first rate scientists. The students are well aware www.JCE.DivCHED.org

•

that these theories are not “controversial”, and that we question these ideas simply as an intellectual exercise. In conclusion, Dr. Bartelt’s letter has failed to convince me of any genuine problems. The concepts presented in my article and in this response letter are general chemistry textbook concepts: (1) reactions need to be balanced, and (2) the stoichiometry of the reaction should be considered when trying to determine the reaction lifetime. These ideas are very easy to understand and are commonly taught in general chemistry courses throughout the world. Literature Cited 1. Kelley, S. Chemical Geology 2002, 188, 1–22. 2. Rink, W. J. Radiation Measurements 1997, 27 (5–6), 975– 1025. William A. Howard Department of Chemistry & Biochemistry University of Alaska Fairbanks Fairbanks, AK 99775-6160

A reviewer of the Howard article responds: Karen Bartelt’s letter is not a fair evaluation of the William A. Howard paper. There are educational advantages of getting students involved in the scientific analysis of even controversial topics like the potassium–argon method of dating minerals. The situation is complex. As a general principle of chemical kinetics it is desirable to identify all the reaction products. However this is seldom practical for nuclear reactions, and accurate radiometric dating is possible in spite of this. Howard’s grocery analogy is okay, but if one counts the empty boxes in the trash and gets 2000 plus or minus 40 because there are other sources of boxes present in the dumpster, counting boxes will not be much help. I trust that student discussion of the potassium–argon method of dating minerals can be valuable in teaching the scientific method. I am also sure that radiometric dating of minerals will stand up to open-minded scrutiny. It is true that some individuals are raising questions about dating with the avowed purpose of placing evolution in question, but this is not the thrust of the Howard paper as it appears now in the Journal of Chemical Education. Reed A. Howald Department of Chemistry & Biochemistry Montana State University Bozeman, MT 59717 [email protected]

A reviewer of the Howard article responds: Karen Bartelt’s letter criticizing the paper by William Howard that I reviewed goes too far. I may have been naïve about Howard’s intentions, and I agree with her concerns about intelligent design, but this paper must be considered on its merits, not on perceived intentions. The Howard pa-


•


547


Letters per should be credited for teaching students to be critical in their analysis of data. There is certainly controversy with all dating techniques, not just from the creationist’s point of view, but from the scientific interpretation of events that may be closely spaced in time. Students need to be aware that systematic errors often far exceed statistical ones. Howard is right that a good understanding of the chemistry, or at least the history of the chemistry, is important to interpret a date. Bartelt trivializes this argument while admitting that a closed system is required and downplaying Dalrymple’s conclusion that “the K–Ar method does not work on all rocks and minerals under all geologic conditions”. Bartelt criticizes Howard for not revealing any effects that might change the 40K:40Ar ratio and then provides an example herself that initial 40Ar might be present. In fact, the ratio assumes normal 40K abundance and no significant 40Ar production by cosmic rays. Howard’s mention of problems with the 40K–40Ca method is less useful, but if this method is used in some cases then his criticism is still valid. I do agree with Bartelt that Howard’s concern with oxidation states and 40Ar as a minor product of the reaction are overstated and could have been removed from the paper. Bartelt idealizes the situation in her first false analogy by assuming that the grocery store is locked. This is the crux of the problem with K–Ar dating, and students should be skeptical that this has always been true irrespective of expert arguments. Howard’s argument that using multiple dating methods may lead to similar problems as K–Ar should not be dismissed. All dating methods are subject to error based on the chemical history of the sample. Radiocarbon dating is notoriously incorrect, and if a sample is not always closed, then similar errors may occur in more than one method. Moreover, researchers are more likely to report values that agree and reject values that disagree. Students should be taught to be very critical of established analytical procedures. Results should be analyzed with an eye on whether they are reasonable, not only precise. A Ming vase dated to 700,000 B.C.E. would make no sense, no matter how careful the measurement seems. Howard’s intentions may be clear to Bartelt, but I don’t think that he crossed the line with his arguments. Bartelt’s concerns are shared by many of us, but are we going to reject this paper on the basis of our perception of the author’s intentions? It is not the job of the Journal of Chemical Education to create

548


•

a litmus test for what papers might or might not have hidden intelligent design meanings. I think that Howard did a good job to cast doubt on the accuracy of K–Ar dating, but if he was motivated by intelligent design he did a poor job of dismissing the validity of K–Ar dating. I don’t believe in intelligent design, but neither do I religiously believe in K– Ar dating. This paper will stimulate students, but nobody’s opinion about intelligent design will be changed. Richard Firestone Lawrence Berkeley National Laboratory Berkeley, CA 94720 [email protected]

Getting the Lead Out Every teacher of chemistry and every physical science teacher knows the story of “ethyl gasoline”, gasoline containing tetraethyllead as an antiknock additive. It seems a good assumption that the public must surely know by now the story of the most famous public health hazard, one introduced in the name of better automobile performance. Since 1986 there has been, to my knowledge, no leaded gasoline sold in the United States, and since the 1990s this health hazard has been eliminated in most of the world. Why is it then that television reporters announce the lowest gas prices in our city as “only $2.25 per gallon for unleaded”? Doesn’t this imply that the more expensive grades of gasoline are leaded? If, as is the case, none of the gasoline of any octane rating contains tetraethyl lead why do we continue to say “unleaded”? Why do the signs at gas stations continue to say “unleaded $2.249—diesel $1.949”? I ask these questions of JCE readers because I would like to know if this means that all our efforts to teach science and real life have been a failure. Didn’t we teach this generation of students that chemists (good chemists) took the lead out and made it unnecessary? So why do gasoline companies continue to remind us of chemists’ past mistakes? Do they think this sells more gasoline? Roy W. Clark Department of Chemistry Middle Tennessee State University Murfreesboro, TN 37132 [email protected]


•

www.JCE.DivCHED.org

A Consideration of the Potassium-Argon Radiometric Method for

Recommend Documents