Constraining dolomitization by Mg isotopes: A case study from partially dolomitized limestones of the middle Cambrian Xuzhuang Formation, North China

The "dolomite problem" refers to the rare dolomite formation in modern oceans that is in sharp contrast to the widespread ancient dolostone in rock record, as well as failure of laboratory inorganic dolomite precipitation at near Earth-surface temperature. Novel Mg isotope systematics provides a promising tool in resolving the "dolomite problem". Here, we develop a protocol to place constraints on the dolomitization process by using Mg isotopes. In this study, we measured Mg isotopic compositions (δ²⁶Mg) of two batches of partially dolomitized limestone samples from the middle Cambrian Xuzhuang Formation in North China. δ²⁶Mg varies between -0.55‰ and -3.18‰, and shows a negative linear correlation with 1/[Mg], suggesting that δ²⁶Mg can be described by a binary mixing between the calcite and dolomite components. Mg isotopic composition of the dolomite component (δ²⁶Mg_dol) for the lower sample set that is collected from a 4 m stratigraphic interval containing three high-frequency ribbon rock-packstone cycles is -1.6‰, while δ²⁶Mg_dol for the upper sample set (from a thick sequence of ribbon rock) is significantly higher (-0.3‰). However, neither mineralogical and elemental compositions, carbon and oxygen isotopes, nor crystal morphologies of dolomite provides diagnostic criteria to differentiate these two batches of samples. δ²⁶Mg_dol of the Xuzhuang limestone is simulated by the Advective Flow (AF) and the Diffusion-Advection-Reaction (DAR) models. The AF model assumes that Mg is transported by advective fluid flows, while the DAR model simulates a contemporaneous seawater dolomitization process, in which Mg is delivered by diffusion. The AF modeling result indicates that δ²⁶Mg of the dolomitization fluid is +0.4‰ and +1.7‰ for the lower and upper sample sets, respectively. These values are significantly higher than modern and Cenozoic seawater Mg isotopic composition, suggesting that the dolomitization fluid is not contemporaneous seawater. The AF model also predicts spatially heterogeneous δ²⁶Mg_dol with progressive enrichment in ²⁶Mg along the fluid flow pathway. In the DAR model, both dolomite content and δ²⁶Mg_dol of the lower sample set can be simulated by using seawater Mg isotopic composition of -0.75‰, thus contemporaneous seawater dolomitization may explain δ²⁶Mg_dol of the Xuzhuang limestone. Furthermore, the DAR model demonstrates spatially homogeneous δ²⁶Mg_dol. To differentiate the AF and DAR models, samples from multiple sections are required. Nevertheless, this study implies that Mg isotope might be a useful tool in the study of dolomitization.