Release of Ni from birnessite during transformation of birnessite to todorokite: Implications for Ni cycling in marine sediments

The phyllomanganate birnessite is the main Mn-bearing phase in oxic marine sediments where it exerts a primary control on the concentration of micronutrient trace metals in seawater. However, during sediment diagenesis and under mild hydrothermal conditions birnessite transforms into the tectomanganate todorokite. We have recently shown that the transformation of birnessite to todorokite proceeds via a four-stage nucleation and growth mechanism, beginning with todorokite nucleation, then crystal growth from solution to form todorokite primary particles, followed by their self-assembly and oriented growth via oriented attachment to form crystalline todorokite laths, culminating in traditional crystal ripening (Atkins et al., 2014). Here we determine the fate and mobility of Ni sorbed by birnessite during this transformation process. Specifically, in our recent work we predict that the presence of Ni within the phyllomanganate matrix will disrupt the formation of todorokite primary particles. As such, contrary to current understanding, we suggest that Ni sorbed by birnessite will slow the transformation of birnessite to todorokite and/or be released to marine porewaters during sediment diagenesis. Here we transform a synthetic, poorly crystalline, Ni-sorbed (~1 wt% Ni) hexagonal birnessite, analogous to marine birnessite, into todorokite under a mild reflux procedure, developed to mimic marine diagenesis and mild hydrothermal conditions. We characterise our birnessite and reflux products as a time series, including X-ray diffraction (XRD), high-resolution transmission electron microscopy (HR-TEM) and extended X-ray absorption fine structure (EXAFS) spectroscopy. In addition we determine Ni speciation and mineral phase associations in a suite of natural marine ferromanganese precipitates, containing intermixed phyllomanganate and todorokite. Our work shows for the first time that Ni significantly slows the transformation of birnessite to todorokite and reduces the crystallinity of the neo-formed todorokite phase, but does not alter the mechanism and pathway of todorokite formation, compared to a Ni-free system. Furthermore, in systems tending towards todorokite as the final diagenetic product, we see that up to 50% of the Ni originally sequestered by birnessite is released to solution during the transformation. Our work indicates that the transformation of birnessite to todorokite in oxic marine sediments likely provides a significant source of Ni to marine sedimentary porewaters and potentially a hitherto unrecognized benthic flux of Ni to seawater.