Isotope Geochem Notes: Rb-Sr, K-Ar and Ar-Ar Dating

Transcription

1 Isotope Geochem Notes: Rb-Sr, K-Ar and Ar-Ar Dating Reading for this topic: White Lectures 5 and 6 Handout: Guide questions: Which isotope of Rb is unstabe and to which isotope does it decay? What is the genera procedure for age dating rocks via the Rb-Sr method? Which mineras are Rb-rich? Rb-poor? How do Rb and Sr resied in mineras? For an igneous rock, what event does the age we get refer to? How is the cosing temperature different for the Rb-Sr and K-Ar systems? For a sedimentary rock, how can we ages for the source of the minera grains or authigenic phases? After you have potted data on an isochron diagram, what simpe test can hep you eiminate bad age dates? What causes bad age dates? Which naturay occurring isotope of K is unstabe and to which isotopes does it decay? Which mineras are K-rich? Which are K-poor? What are the chemica properties of Ar? What determines the smaest age that can be determined by K-Ar dating? Is this the same for a mineras or does it vary depending on the minera? How does one obtain a date for an igneous rock- what is the procedure for K-Ar and Ar-Ar methods? With the Ar-Ar method, how can better ages be obtained from measurements taken as the crystas are heated stepwise? Rb-Sr age dating Practica procedure: 1) Coect sampes- avoid weathered rock 2) Separate mineras -Rb-rich phases: Biotite, muscovite, or Hornbende (note: easiy weathered) -Rb-poor: Pagiocase- Ca-bearing Fedspar 3) Dissove mineras 4) Purify Sr- ion exchange chromatography 3) Measure Rb/Sr ratio (then cacuate 87 Rb/ 86 Sr) 4) Measure 87 Sr/ 86 Sr - via high-precision TIMS or ICP-MS 87 Sr 87 5) Pot an isochron: Pot 86 Sr vs. Rb Sr, reca that Sr 86 Sr = Ê Á 87 Srˆ 87 Rb + ( e t - 1) Ë 86 Sr 86 0 Sr Ê...fit the data to a ine. Sope = ( 87 Srˆ e t -1), Intercept is Á Ë 86 Sr 0 THIS IS RATHER SLOW AND IS RARELY DONE TODAY. See Ar dating beow for faster methods. See Tabe 8.1 from Faure, 1986 text for ist of Rb and Sr conc s in rocks.

2 Aternative: Use aser-abation and anayze vapor from that- avoids digestion- sti use isochrons. Much faster. Lesser precision, because Sr is anayzed aong with the Rb, etc that abates. Isochrons give age dates of minera soidification: 1) Igneous- reiabe if you know a mineras grew from same iquid May not work for ate-stage putonic rocks that may be hydrotherma- pegmatite Granites stew in their own juices for miions of years 2) Metamorphic- mineras recrystaize, fuids carry Sr around Can have compete resetting, a mineras with same new initia 87/86 See Fig. 8.5 from Faure, Can aso have partia resetting- a mess Coud aso have mineras that grew sowy and/or at different times Detecting probems with isochron dates: If mineras have been cosed (no oss or gain), then a sampes wi fa on a singe isochron (converse not necessariy true!) If mineras have gained or ost Rb or Sr: Assuming the oss/gain is not the same for the various mineras, inearity probaby destroyed Therefore: Poor inearity means a bad date! In some cases, an isochron can be saved by removing one or two mineras that are known to be more easiy reset or may have grown after the others WHOLE-ROCK ISOCHRONS If isochrons for minera separates are destroyed by metamorphism, there is some possibiity that the ength scae for redistribution of Rb + Sr is ony a few cm- in this case, treat whoe rocks as cosed systems In other words: Instead of assuming each minera is a cosed domain, hypothesize that handsampe size pieces of rock have remained cosed overa (but the mineras within may have exchanged Rb and Sr with each other) Procedure: 1) Choose Rb-rich and Rb-poor rock types from same area 2) Grind up and homogenize whoe rock sampes from different rock types presumaby formed from same event 3) Anayze these and pot an isochron see Fig. 8.7 from Faure, 1986 SEDIMENTARY ROCKS: Difficut to date by any method What event are we dating? Coud be... 1) Formation of detrita grains... OR 2) Age of authigenic mineras- grown after buria of the sediments. One approach: Concentrate on authigenic Iite (authigenic means grown in pace) -Iite: KA 4 (Si 7 AO 20 )(OH) 4 [Note muscovite: K 2 A 4 (Si 6 A 2 O 20 )(OH) 4 ] -IMPORTANT: Rb usuay foows K!!! Or...Another minera to try: Gauconitic phases- Fe-rich cays, formed shorty after sedimentation Another approach... remove authigenic mineras, and try for an isochron on the detrita mineras (ony works if they a came from the same source)

3 40 K decay and the K-Ar geochronometer Overa decay const. for 40 K decay = 5.5 *10-10 Penty eft over from nuceosynthesis. One of the important heat sources within the earth. Branched decay: 10.5% to Ar, the rest to Ca. Seems ike Ca woud be good But. 40 Ca is very high in abundance and in most cases, the ingrowth of it is not measurabe. When woud you expect it might work? **Important genera principe: The ratio of ingrowth rate to initia amount present of the daughter isotope is critica. K-Ar Age Dating Very itte Ar taken into crystas. Very high ingrowth/ initia ratio for K-rich crystas, and sti high for K-poor phases or rocks. Probems with eakage of Ar after production: Because Ar is a nobe gas, it is present as Ar 0 in crystas, and resides in defects. It can move rather easiy through crystas because it does not form ionic bonds ike Sr or Rb. 1) At high temperatures for a crystas 2) Even at ow temperatures from some cay mineras 3) At moderate to ow temperatures from very sma crystas of any minera Exampes of Ar dating: DATING TOOL OF CHOICE FOR AGES IN THE 1 TO 200Ma range. Routine dating of vocanic rocks Dating of sediments (e.g., hominid fossi-bearing ayers in Kenya) using vocanic fragments within them Dating of the magnetic reversas- pate tectonics. Tying the magnetic record on and to an actua time scae, then using that to date the ocean foor. Dating of certain cay mineras Practica procedure: 1) Coect sampes - get fresh sampes; weathering causes mineras to give up Ar and take some in from atm. - try to get rocks with high-k phases if you have a choice - vocanic gass is okay if fresh - putonic rocks- may be messed up; post-soidification hydrotherma reactions and inherited Ar 2) Separate mineras OR do whoe rock anayses if you must (ess desirabe) -mineras are better- individua crystas eak ess -vocanic gass more ikey have inherited Ar- from atm or magma 3) Measure [K] and cacuate [ 40 K] 4) Measure [ 40 Ar]- met crystas in a vacuum to reease Ar 5) Often, some inherited Ar present, so measure 40 Ar/ 36 Ar and do the isochron method

4 Specia Cases: a) If you can assume the inherited Argon is from air: 40 Ar Ar = Ê Arˆ Á + 40 e K ( Ë 36 Ar 36 Ar et -1), where the air ratio = air This ooks ike our standard isochron equation, but we know the intercept, and thus, ony one measurement is needed to get an age. e is there to account for the fact that ony 10.5% of K decays go to Ar. If the inherited Ar is not necessariy from air, you must use the fu isochron approach. b) If there s no inherited Ar (as reveaed by ack of 40 Ar ) you can use: ( ) 40 Ar = e 40 K e t -1 What event are we dating? What event starts this chronometer? - Cosure of the crysta w.r.t. Ar. Cosure temps of mineras: Ar moves through crystas more easiy than ions (e.g., Sr 2+ ); not bound into the attice. Reativey ow cosure temps, ~ C (see Faure, 1986, Tabe 7.2) Possibiity for some mineras to reset whie others may not Possibe for some domains of a singe crysta to reset, others not (see beow) Ar-Ar dating: Most common method (K-Ar dating rarey used now) Convenient way to get the amount of 40 K as we measure 40 Ar- ONE ANALYSIS!!! Actuay, we determine 39 K, and we know the 40 K/ 39 K ratio very we. How it works: Bombard sampes with neutrons... - Converts some of the 39 K to 39 Ar (neutron in, proton out, simutaneousy) 39 Ar haf ife is 269 years (sampes are radioactive- a minor hasse) Measure on a mass spectrometer: 40 Ar/ 39 Ar, this is reated to time by: 39 Ar = C ( et -1), where C is a caibration 40 Ar constant that depends on the amount of neutron bombardment. One determines C by anayzing standards aong with the sampes. Step heating of mineras for Ar-Ar dating 20 years ago, they just meted the mineras in one step Now, they use controed heating by asers to extract more information from mineras Heat for a few seconds, then anayze Ar given off. Successive puses at successivey greater power to get successivey higher temperatures At ow temperatures, Ar comes out of more open or eaky domains The ow-t steps are garbage- ignore At high temperatures, Ar comes out of ess open domains- better age dates

5 Comparison: K-Ar (Ar-Ar) dating versus Rb-Sr dating Rb-Sr dating not as common as Ar-Ar- more difficut. 87 Rb decay const = 1.42 * 10-11, 2x sower than 40 K Sr 2+ fits into crysta attice better than Ar, diffuses through attice more sowy o Thus, cosure temp is essentiay the crystaization temp What is the main drawback of Ar dating? o Possibe Ar oss. Rb-Sr is better in some cases. Appendix: Diffusion of Ar and cosure temperatures for Ar dating What is the cosure temp.? Consider the strong temperature dependence of diffusion: Fick s Law; Diffusion Constant is D But we know much about the systematics of diffusion: The diffusion constant, D = D 0 e - E A / RT, E A is the activation energy Background: Activation energy is a term used in Kinetics. E A is the amount of energy needed to get an atom over the hump so it can jump to a neighboring site. The e -E A / RT term is a very common term in kinetics, and is used here to express the statistica probabiity that the activation energy at a given temperature, T Energy barrier easier to cross at higher energy = higher temp. Exampes: 100 C increase in T eads to x100 increase in diffusion. At room temperature, itte Ar oss from sma crysta in 1 Ga At 300 C, itte oss At 400 C, minor oss over tens of Ma At 500 C, major oss over tens of Ma At 600 C, major oss over ess than 1 Ma- useess. Now we aso note that Ar oss from a crysta depends on its size Assuming a rod shape, oss is a function of Dt/a 2 (a = x-sect. Area) AND, crystas are fu of defects o Diffusion of Ar through a crack or even a minor defect is very fast Thus, we can guess that there are arge domains, where Ar retention is good, and others where it is not so good. Exampe: Ar oss from a sma domain, 1/10 the size of arge one, is 100x greater. Appendix: Therma Histories of mineras Therma Histories: Goa is to determine the temperature history of the crysta over miions of years. This then tes us about the depth history of the rock. Exampe: How fast has materia eroded off the top of the Himaaya. Attempts have been made to correate this quantativey to derive therma histories. Ages coming out of ow-t aser puses correspond to domains of the crysta that are more open and were open at the time given by that age

6 Ages coming out of the ast puse (meting puse) correspond to the most cosed domains. If there is a pateau, then we can guess there are some domains that record the age of the puton If the pateau starts abrupty after the ow-t garbage, then the whoe crysta was reset at the same time (except for the very oosest domains) If there is a very broad sope up to the pateau, then the puton must have cooed more sowy The shape of the curve can te you if it was smooth or episodic cooing There are those who think they can cacuate the cooing history, T(t) Here are two papers: Thomson S. N., Stoeckhert B., and Brix M. R. (1998) Thermochronoogy of the high-pressure metamorphic rocks of Crete, Greece; impications for the speed of tectonic processes. Geoogy (Bouder) 26(3), Vance D., Ayres M., Keey S., and Harris N. (1998) The therma response of a metamorphic bet to extension; constraints from aser Ar data on metamorphic micas. Earth and Panetary Science Letters 162(1-4),