(1) Chemistry and Petrology of the Earth and Planets
The goals of this study are to achieve a better understanding of the differentiation and chemical evolution of the Earth and other planets, the geochemical aspects of crust-mantle interaction, the dynamics and evolution of magma systems, the formation of crust and the process of weathering, through the integrated study of isotopic compositions (Sm-Nd, Rb-Sr, Lu-Hf, and U-Th-Pb isotopic systematics) and trace element concentrations and ratios in igneous, sedimentary, and metamorphic rocks.
The main tools in this research are 3 types of mass spectrometers: thermal ionization mass spectrometers, secondary ion mass spectrometers, and inductively coupled plasma mass spectrometers. Mass spectrometry has wide applications in almost all the fields of natural sciences, including earth and space sciences, life sciences (essentially, there is no boundary between different fields of natural sciences; there is only overlap between disciplines. Interdisciplinary research is the way to go).
My recent work has focused on the study of the mantle dynamics and processes using uranium-thorium isotopic disequilibrium, early (Hadean) earth and Moon and other terrestrial planets, and early solar processes, using long-lived isotopes, extinct short-lived isotopes, and equations.
(2) Mathematical Modelling
Just like mass spectrometry, mathematics has broad applications in all fields of natural sciences. To quantify physcial, chemical, geological, and biological processes is always an important goal in natural sciences. The main tools in this part of my research are ordinary and partial differential equations with initial and boundary conditions, and numerical analysis. The facinating aspect of mathematical modeling is a deep understanding of not only mathematics but also the related knowledge of physics, chemistry, and geology. Recently I have been working on models to describe trace element fractionation and uranium-thorium disequilibrium during the processes of melt generation and transport. I also wrote a book "Quantitative Geochemistry" for graduate students and researchers.
(3) Palaeoclimates
Although much of my early research focuses on deep earth processes, my recent research has been extended to earth surface studies of Miocene paleosols (Zou et al., 2004) and Neoproterozoic glaciations (e.g., Xu et al., 2005; Xu et al., submitted), as the powerful tools that I have learned (isotopes, trace elements, clean lab chemistry, mathematics, programming languages) are useful for the studies of not only deep processes but also surface processes.
As a byproduct of the research on continental basalts, we demonstrated that the unusual tridymite-hercynite xenoliths in Niutoushan basalts represent preserved aluminous lateritic paleosols that are not genetically related to host tholeiites. These lateritic paleosols with strongly desilicated minerals were formed by intense chemical weathering under warm and humid tropical conditions (with annual average temperature of >19 degree C and the annual rainfall of >165 cm) in SE China during the interval from 17 to 15 Ma. The formation age of the paleosols corresponds to a period characterized by slow uplift of the Himalayan-Tibetan Plateau region (and thus less consumption of CO2) after 17 Ma, and eruptions of 17-15 Ma Columbia River flood basalts, the Vogelsberg basalts, and East China basalts (and thus more input of CO2 into the atmosphere). The tridymite-hercynite xenoliths in the Niutoushan basalts thus preserve novel evidence of extraordinary climatic greenhouse conditions in middle Miocene that would otherwise have been lost by erosion of paleosols (Zou et al., 2004). In favorable circumstances, hard rocks may contain useful information about paleoclimate conditions!
My approach to research is to combine accurate measurements by mass spectrometry with solid mathematical modeling. I pay special attention to quantitative models, innovative experimental methods, and bold/creative (sometimes a bit crazy) interpretations. A combined experimental and theoretical study can solve scientific problems more thoroughly and elegantly.