Geological Survey of Denmark and Greenland. Bulletin 10, 25-28 Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) was developed in 1985 and the first commer- cial laser ablation systems were introduced in the mid 1990s. Since then, LA-ICP-MS has become an important analytical tool in the earth sciences. Initially, the main interest for geol- ogists was in its ability to quantitatively determine the con- tents of a wide range of elements in many minerals at very low concentrations (a few ppm and below) with relatively high spa- tial resolution (spot diameters of typically 30–100 μm). The potential of LA-ICP-MS for rapid in situ U–Th–Pb geo- chronology was already realised in the early to mid 1990s. However, the full potential of LA-ICP-MS as the low-cost alternative to ion-microprobe techniques for highly precise and accurate in situ U–Th–Pb age dating was not realised until the relatively recent advances in laser technologies and the introduction of magnetic sectorfield ICP-MS (SF-ICP- MS) instruments. In March 2005, the Geological Survey of Denmark and Greenland (GEUS) commissioned a new laser ablation magnetic sectorfield inductively coupled plasma mass spectrometry (LA-SF-ICP-MS) facility employing a ThermoFinnigan Element2 high resolution magnetic sector- field ICP-MS and a Merchantek New Wave 213 nm UV laser ablation system. The new GEUS LA-SF-ICP-MS facility is widely used on Survey research projects in Denmark and Greenland, as well as in collaborative research and contract projects conducted with partners from academia and indus- try worldwide. Here, we present examples from some of the these ongoing studies that highlight the application of the new facility for advanced geochronological and trace element in situ microanalysis of geomaterials. The application of LA- SF-ICP-MS based in situ zircon geochronology to regional studies addressing the Archaean geology of southern West Greenland is presented by Hollis et al. (2006, this volume). Zircon U–Pb geochronology using LA-SF-ICP-MS In situ U–Th–Pb geochronology was developed in the mid- 80s with the introduction of ion-microprobe techniques, most commonly referred to as secondary ion mass spectrometry (SIMS) and sensitive high resolution ion microprobe (SHRIMP). The advantage of in situ U–Th–Pb geochronol- ogy over conventional chemical dating by isotope dilution thermal ionisation mass spectrometry (ID-TIMS) is the capa- bility to analyse different domains in heterogeneous single zircons with high spatial resolution (spot diameters of typi- cally 10–30 μm). This allows resolution of igneous and meta- morphic events separated by intervals of only a few tens of million years from polychronic zircons. The disadvantages of ion-microprobe techniques are the very high purchasing and operating costs for the instrument. The rapid improvements in laser based U–Th–Pb geochronology makes it now possi- ble to obtain in situ U–Th–Pb geochronological data with comparable spatial resolution as well as analytical precision and accuracy at only a fraction of the costs of ion-microprobe techniques (e.g. Jackson et al. 2004; Janoušek et al. 2006). Dating of magmatic and metamorphic events The capabilities of LA-SF-ICP-MS for the precise and accu- rate U–Pb age dating of relatively young igneous zircons are demonstrated by the analysis of a population of 40 zircons extracted from a gabbro from the Coastal Cordillera at Tre- gualemu, central Southern Chile. The gabbro is believed to have been formed by regional extension during the late Trias- sic to early Jurassic (Charrier 1979). Three zircons proved to be too small for analysis (< 30 μm). The results for the remain- ing 37 zircons (Fig. 1) define a highly precise igneous con- cordia age of 203 ± 2 Ma (2σ; MSWD = 1.7) and indicate a Late Triassic (Rhaetian) intrusion age of the gabbro. An even Advanced in situ geochronological and trace element microanalysis by laser ablation techniques Dirk Frei, Julie A. Hollis,Axel Gerdes, Dan Harlov, Christine Karlsson, Paulina Vasquez, Gerhard Franz, Leif Johansson and Christian Knudsen © GEUS, 2006. Geological Survey of Denmark and Greenland Bulletin 10, 25–28. Available at: www.geus.dk/publications/bull 25 240 220 200 180 1600.025 0.027 0.029 0.031 0.033 0.035 0.037 0.17 0.19 0.21 0.23 0.25 0.27 207 Pb/ 235 U 2 06 P b /2 38 U GEUS LA-SF-ICP-MS: Igneous Concordia age 203 ± 2 Ma (95% conf.) MSWD: 1.7 Gabbro, central southern Chile (Sample PV 4-22) Fig. 1. Concordia diagram for igneous zircons from a gabbro from the Coastal Cordillera in central Southern Chile. younger igneous age of 158 ± 2 Ma was recently obtained for a zircon xeno- or phenocryst derived from a newly discovered carbonatite in southern West Greenland (Steenfelt et al. 2006, this volume). The high sensitivity of the Element2 SF-ICP-MS allows U–Pb zircon age dating with a laser spotsize of 30 μm or less, depending on the Pb content of the zircons. This makes it feasible to analyse different age domains in polychronic zir- cons, e.g. igneous cores and metamorphic rims. For example, zircons from an orthogneiss from the Nuuk region, southern West Greenland, display characteristic textures in back-scat- tered electron (BSE) and cathodoluminescence (CL) pictures that are interpreted as igneous cores surrounded by rims grown during a metamorphic event (see inset in Fig. 2). The U–Pb age data of the cores suggest an emplacement of the igneous protolith at c. 3660 Ma, while the rim data indicate meta- morphism close to 2700 Ma (Fig. 2). Dating of detrital zircons Analyses of the crystallisation ages of detrital zircons in clas- tic sediments are a powerful tool in sedimentary provenance analysis. Accurate and precise U–Pb ages of >100 detrital zir- con grains in a sample are needed to detect all major sedi- mentary source components with statistical confidence (cf. Vermeesch 2004; and references therein). The relatively high costs and the limited capacities of ion microprobe techniques (c. 75 zircon age analyses per day) impose restrictions on the number of samples that can be studied. Because LA-SF-ICP- MS provides very high capacities (in excess of 300 zircon age analyses per day) without compromising accuracy and preci- sion, it constitutes the economic method of choice for prove- nance studies based on detrital zircon U–Pb ages. An example for detrital zircon age data obtained by LA- SF-ICP-MS is shown in Fig. 3, where the 207Pb–206Pb age distribution for a population of 100 zircon grains separated from a Cambrian sandstone from Torekov, southern Sweden, are shown in a combined histogram and probability density distribution (PPD) diagram. The concordance filtered zircons (dark shaded area; 90–110% concordance, defined as 100* [206Pb–238U age / 207Pb–206Pb age]) show a polymodal age distribution with a minor peak at ~1000 Ma and two major peaks at c. 1150 Ma and 1650 Ma. The presence of two older sedimentary sources (c. 2150 Ma and c. 3050 Ma) is indica- ted by discordant grains (lighter shaded grey areas) that most likely suffered lead loss during their petrogenetic evolution. Figures of merit The short and long term precision and accuracy of LA-SF- ICP-MS for U–Pb zircon age dating has been assessed using two zircon reference materials, Plesovice (with an ID-TIMS age of 338 ± 1 Ma; Aftalion et al. 1989; provided by Jan Kosler, University of Bergen) and 91500 (ID-TIMS age = 1065 ± 0.4 Ma; Wiedenbeck et al. 1995). The PL zircon is routinely analysed as unknown for quality control purposes in every analytical session in the GEUS laboratory. The results for 16 analyses of the zircon from a typical single ana- lytical session are shown in Fig. 4A. They define a concordia age that is in excellent agreement with the ID-TIMS age reported by Aftalion et al. (1989). Long-term precision (2σ) based on 109 analyses of the Plesovice zircon by two differ- ent operators was 2%, 2.3% and 1.1% for the 206Pb/238U, 207Pb/235U and 207Pb/206Pb ratios, respectively. The widely used 91500 zircon has so far only been analysed during one analytical session. The results for all seven analyses carried out 26 3900 3700 3500 3300 3100 2900 2700 2500 3800 3400 3000 2600 2200 c. 2700 Ma n=8 c. 3660 Ma n=4 2 0 6 P b /2 3 8 U 207 Pb/ 235 U 0.9 0.8 0.7 0.6 0.5 0.4 0.3 5 15 25 35 45 Fig. 2. Concordia diagram and inset of 207Pb/206Pb age plot for poly- chronic zircons from an orthogneiss in the Nuuk region, southern West Greenland. Inset CL image shows laser ablation pits (pit diameter = 30 μm) in igneous cores and metamorphic rims of polychronic zircons. Note the shallow depth (usually < 20 μm) of the laser ablation pits. 0 40 0 8 00 1 20 0 16 00 2 00 0 2 40 0 28 00 3 20 0 3 60 0 40 00 Age (Ma) P ro b ab ili ty 0 5 10 15 20 25 30 35 F re q u en cy Fig. 3. Combined display of histogram and probability density distribu- tion (PPD) diagram for zircons from a Cambrian sandstone from Torekov, south-western Sweden. See text for explanations. during this session (Fig. 4B) define a concordia age which is in excellent agreement with the ID-TIMS age reported by Wiedenbeck et al. (1995). Trace element analysis using LA-SF-ICP-MS Due mainly to the introduction of LA-ICP-MS techniques, the trace element signatures of individual minerals (e.g. gar- net, clinopyroxene, epidote, rutile, calcite) are now frequently being used to deduce the petrogenetic evolution of magmatic rocks, unravel water–rock interactions, identify geotectonic settings and sediment sources, track down the pathways of potentially health-damaging pollutants, and to unravel the change of seasurface and atmospheric temperatures. Further- more, LA-ICP-MS is an important tool for the characterisa- tion of synthetically produced geomaterials. Water–rock interaction Fluid-mediated mass transfer during metamorphism and metasomatism is a hotly debated issue. Mobility of geochem- ically important trace elements during fluid flow has far- reaching implications for e.g. ore-forming processes and mass transfer during subduction. Detailed investigation of the mecha- nisms of element mobility during fluid–rock interaction on a grain-scale are pivotal for an understanding of the processes that lead to the characteristic element enrichments and deple- tions observed in nature. In the Söndrum stone quarry in Sweden, a localised dehy- dration zone of 2.5 to 3 m width occurs in garnet-bearing granitic gneiss around an approximately 1 m wide pegmatoid dyke. Whole-rock chemistry suggests that the solid-state dehydration of the granitic gneiss to charnokite via low H2O activity fluids consisting principally of CO2 and a minor brine component was predominantly isochemical (Harlov et al. 2006). Exceptions include Y and the heavy rare earth ele- ments (HREE), which are markedly depleted throughout the dehydration zone. In order to assess the mechanism of Y and HREE depletion, the trace element geochemistry of garnet was studied in a traverse across the dehydration zone. Garnets from the pristine, unaltered granitic gneiss are characterised by a strong negative Eu-anomaly and the steep, HREE en- riched pattern typical for garnet (sample SD45-600 in Fig. 5). In contrast, garnets from within the dehydration zone show a less pronounced Eu-anomaly and are characterised by dra- matic Y and HREE depletions, leading to almost flat REE pattern (sample SD9-120 in Fig. 5). This observation pro- vides direct evidence for massive release of Y and HREE from garnets, the principal hosts of these elements in the granitic gneiss, via solid-state fluid–rock interaction which was also accompanied by dehydration of hornblende and biotite to ortho- and clinopyroxene. Analysis of synthetic geomaterials Another application of LA-ICP-MS is the trace element analysis of synthetic geomaterials, e.g. products of experi- ments carried out for petrogenetical purposes. Since the exper- imentally produced mineral phases are usually very small, in situ microanalysis with high spatial resolution is needed. This was traditionally achieved by SIMS techniques. Because LA- ICP-MS analyses are much cheaper and facilities are much 380 360 340 320 300 0.047 0.049 0.051 0.053 0.055 0.057 0.059 0.061 0.34 0.36 0.38 0.40 0.42 0.44 0.46 GEUS LA-SF-ICP-MS: concordia age 339.0 ± 2.7 Ma (95% conf.) MSWD: 2.1 A: Plesovice Zircon Standard ID-TIMS Age: 338 ± 1 Ma 980 1020 1060 1100 1140 1180 0.16 0.17 0.18 0.19 0.20 1.6 1.7 1.8 1.9 2.0 2.1 2.2 GEUS LA-SF-ICP-MS concordia age 1070 ± 10 Ma (95% conf.) MSWD: 0.27 B: 91500 Zircon Standard ID-TIMS Age: 1065 ± 0.4 Ma 207 Pb/ 235 U 207 Pb/ 235 U 2 06 P b /2 38 U 2 06 P b /2 38 U Fig. 4. Concordia diagrams for 16 analysis of the Plesovice zircon reference material obtained during a single analytical ses- sion (A); and 7 analysis of the 91500 zir- con reference material obtained during a single analytical session (B). 27 Sm Eu Gd Tb Dy Ho Er Tm Yb Lu C 1 n o rm al is ed SD45-600 SD9-120 1 10 100 1000 10000 Fig. 5. Average chondrite-normalised REE-pattern of garnets from the unaffected granitic gneiss (SD45-600) and the dehydration zone (SD9- 120). Note the depletion in HREE of garnets from the dehydration zone. more widely available, LA-ICP-MS is increasingly used for the trace element analysis of synthetic geomaterials. The inset in Fig. 6, for example, shows a back-scattered elec- tron photomicrograph of an experimental charge containing euhedral clinopyroxene and a coexisting anhydrous silicate melt. The experimental charge was synthesised in order to determine the clinopyroxene-melt trace element partition coef- ficients Di (defined as Di = [concentration of i]cpx/[concen- tration of i]melt), knowledge of which is important for geochemical melt modelling purposes (Landwehr et al. 2001). In order to test the ability of laser ablation techniques to cor- rectly determine partition coefficients from such experimen- tal charges, we have analysed clinopyroxenes and coexisting silicate melts from a number of experiments with both SIMS (at the NERC ion-microprobe facility in Edinburgh) and LA-SF-ICP-MS (at GEUS). The resulting partition coeffi- cients determined by SIMS and LA-SF-ICP-MS are graphi- cally compared in Fig. 6 for a representative sample. For all investigated samples, the partition coefficients determined by SIMS and LA-SF-ICP-MS are in excellent agreement, clearly demonstrating the reliability of laser ablation techniques for the characterisation of synthetic geomaterials. Acknowledgements The establishment of the new LA-SF-ICP-MS facility was funded with a grant of the Danish Ministry of Education and Technology to the Geocenter Copenhagen (‘Geocenterbevilling’) and financial support by GEUS. PV thanks the DAAD for awarding a PhD scholarship. References Aftalion, M., Bowes, D.R. & Vrána, S. 1989: Early Carboniferous U–Pb zircon age of garnetiferous, perpotassic granulites, Blanský les massif, Czechos- lovakia. Neues Jahrbuch für Mineralogie, Monatsheft 4, 145–152. Charrier, R. 1979: El Triásico en Chile y regiones adyacentes de Argentina: Una reconstrucción paleogeográfica y paleoclimática. Comunicaciones 26, 1–37. Harlov, D.E., Johansson, L., Van Den Kerkhof, A. & Förster, H.-J. 2006: The role of fluid flow and diffusion during localised, solid-state dehy- dration: Söndrum Stenhuggeriet, Halmstad, SW Sweden. Journal of Petrology 47, 3–33. Hollis, J.A., Frei, D., van Gool, J.A.M., Garde, A.A. & Persson, M. 2006: Using zircon geochronology to resolve the Achaean geology of south- ern West Greenland. Geological Survey of Denmark and Greenland Bulletin 10, 49–52. Jackson, S., Pearson, N.J., Griffin, W.L. & Belousova, E.A. 2004: The application of laser ablation – inductively coupled plasma – mass spectrometry to in situ U–Pb zircon geochronology. Chemical Geology 211, 47–69. Janoušek, V., Gerdes, A., Vrána, S., Finger, F., Erban, V., Friedl, G. & Braithwaite, C.J.R. 2006: Low-pressure granulites of the Lišov massif, Southern Bohemia: Viséan metamorphism of late Devonian plutonic arc rocks. Journal of Petrology 47, 705–744. Landwehr, D., Blundy, J.D., Chamorro-Perez, E.M., Hill, E. & Wood, B.J. 2001: U-Series disequilibria generated by partial melting of spinel lher- zolite. Earth and Planetary Science Letters 188, 329–348. Steenfelt, A., Hollis, J.A. & Secher, K. 2006: The Tikiusaaq carbonatite and associated kimberlites: a new alkaline magmatic province in the Nuuk region, southern West Greenland. Geological Survey of Denmark and Greenland Bulletin 10, 41–44. Wiedenbeck, M., Allé, P., Corfu, F., Griffin, W.L., Meier, M., Oberli, F., von Quadt, A., Roddick, J.C. & Spiegel, W. 1995: Three natural zircon stan- dards for U–Th–Pb, Lu–Hf, trace element and REE analysis. Geo- standards Newsletters 19, 1–23. Vermeesch, P. 2004: How many grains are needed for a provenance study? Earth and Planetary Science Letters 224, 441–451. Authors’ addresses D.F., J.H. & C.Kn., Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark. E-mail: df@geus.dk A.G., Institute of Mineralogy, Johan-Wolfgang-Goethe University, Senckenberganlage 28, D-60054 Frankfurt, Germany. D.H., GeoForschungsZentrum Potsdam, Section 4.1 Experimental Geochemistry and Mineral Physics, Telegrafenberg, D-14473 Potsdam, Germany. C.Ka. & L.F., Department of Geology, University of Lund, Sölvegatan 12, S-22362 Lund, Sweden. P.V. & G.F., Institut für Angewandte Geowissenschaften, Technische Universität Berlin, Ernst-Reuter-Platz 1, D-10587 Berlin, Germany. 28 10 -3 10 -2 10 -1 10 0 10 1 10 -3 10 -2 10 -1 10 0 10 1 La Ce Nd Sm Lu Ti Zr Hf Th U D SI M S D LA-SF-ICP-MS 10 -3 10 -2 10 -1 10 0 10 1 10 -3 10 -2 10 -1 10 0 10 1 Experimentally determined Cpx-melt partition coefficients La Ce Nd Sm Lu Ti Zr Hf Th U D SI M S D LA-SF-ICP-MS 500 μm Fig. 6. Comparison of clinopyroxene-melt trace element partition coeffi- cients (open symbols: REE; solid symbols: HFSE) determined from synthetic run products by SIMS and LA-SF-ICP-MS. Inset shows a BSE photomicrograph of synthetic, euhedral clinopyroxene coexisting with a glassy melt produced experimentally at high pressure and temperature. Shallow, bright spots are SIMS ablation pits (c. 20 μm diameter), while deeper, dark spots are laser ablation pits (c. 30 μm in diameter).