DFG Research Unit FOR 2685 | B4

U-Pb Systematics of Fossilized Biogenic Tissues as a Tracer of Long-term Diagenetic Processes

Fossilization of mineralized (bone, tooth) and non-mineralized organic tissue (wood) is a complex, post-mortem alteration process. Although relative mineralization and replacement sequences have been identified in fossil wood and bones by detailed petrographic, mineral-chemical, and structural investigations [1], the absolute time scales and durations of such processes are not yet known, particular for pre-Pleistocene fossils. If a biogenic template has incorporated radioactive elements during early diagenetic processes, i.e., relatively rapidly after death, and remained a closed system throughout geologic time, then geochronological methods should return a radio-metric age consistent with the near-depositional or stratigraphic age. The available data, however, suggests that fossil tissues do not necessarily record radiometric ages that are concordant with the expected stratigraphic age [2-4]. However, radiometric systems may provide a tool to unravel a sequence of different diagenetic alteration processes that are otherwise not identifiable, particularly when analyzing the radiometric system at the microscale by laser-ablation inductively-coupled plasma mass spectrometry (LA-ICP-MS) or even at the nano-scale using ion microprobes (SIMS, NanoSIMS) or atom probe tomography (APT) [5]. In this project, we will evaluate the potential of the U-Pb system to date distinct fossilization events.

To test the potential of LA-ICP-MS for U-Pb dating of long-term fossilization processes, we analyzed different silica generations of petrified wood from a ~18 Ma old tuff (Fig. 1), and fossil bone from three ~150 Ma old dinosaurs (Fig. 2):

• U-Pb analyses from the fossil wood revealed a protracted sequence of discrete silica minerali-zation and replacement events rather than a continuous maturation of amorphous silica towards a quartz end member (Fig. 1).
• U-Pb analyses from dinosaur bone apatite did not return any information on their deposition age (Fig. 2c,d). However, the alignment of the U-Pb data from two samples along distinct mixing lines indicates that either the U minerali-zation occurred around 80 respectively 90 Ma or the U-Pb systems were reset at these times.

Notably, in both fossil tissues certain domains or distinct diagenetic infillings (calcite) were identi-fied that yielded an age very close to the expected deposition age (Fig. 1e; Fig. 2d). It follows that meaningful U-Pb age information about the timing of distinct fossilization processes may be preserved in fossil bone and wood.

Figure 1 | (Examples and LA-ICP-MS U-Pb isotopic data for silicified wood from Lesvos. (a) Polished specimen under day- and UV-light. Green fluorescence identifies areas with elevated U-content. Many domains (type B) were affected by late stage, uraniferous fluid infiltration, whereas a few domains (type A) have escaped this event. (b) Back-scattered electron image of cross-cutting veins in silicified wood (type B) and (c) Raman phase distribution map, documenting a succession of wood opal matrix, vein filling opal-CT generations, and partially replacement by chalcedony. (d) Tera-Wasserburg concordia diagram with regressions of intercept ages for distinct silica generations (type A). (e) The concordia diagram for one silica sample with low U contents (type B) yields an intercept age close to the published deposition age of the tuff ( ~18 Ma [6]), which is accepted as the chronostratigraphic age of the fossil forest.

Figure 2 | Examples of fossil bone material and results from LA-ICP-MS U-Pb analyses. (a) Half section of a histological drill core (BM01, Camarasaurus, Morrison Formation) with a massive band of agate. (b) Reflected light image of a polished slab of a sauropod bone (BTE1, Giraffatitan, Tendaguru Beds). Inset gives a Raman phase distribution map for the marked area. Note the high amount of goethite inclusions, and texturally distinct calcite populations. (c) Tera-Wasserburg concordia diagram for bone apatite samples from the Morrison Formation. Intercept ages at about 80 and 90 Ma imply a closed system behavior for most parts of the specimen. (d) Concordia diagram for the BTE1 sample. Arrows indicate displacement vectors for major mechanisms of U-Pb disturbance. The main population of near concordant calcite analyses defines an intercept age of 146 ± 3 Ma, which is consistent with the chronostratigraphic age (~150 Ma).

By combining the U-Pb dating of fossilized bone and wood tissue with detailed mineralogical, petrological, and geochemical analyses down to the atomic length scale, we intend to

• study (and quantify) diagenetic processes (e.g., U incorporation, recrystallization/aging) during bone and wood fossilization, which will increase our understanding of the long-term fossilization process that is difficult or even impossible to access experimentally.
• Find geochemical, textural, or mineralogical criteria for screening the most promising petrified wood and fossil bone samples or sample domains for "direct" U-Pb dating of the deposition/stratigraphic age.