My research involves combining high-pressure experiments with interdisciplinary collaboration to study the physics of deep Earth materials, structures, and processes. My recent experimental work has involved developing a multi-technique approach for measuring the melting of iron-bearing materials at high-pressures. I have applied this new approach to an Fe-Ni-Si alloy compatible with recent estimates of core compositions for Earth and Mercury, clarified the effect of silicon on core temperatures, and addressed discrepancies among melt detection techniques. In addition, I am active in interdisciplinary efforts to quantify the compositions of ultralow velocity zones, the most extreme seismically-detected structures at Earth’s core-mantle boundary region. My contributions have involved constructing mineralogical models of a solid ULVZ compatible with constraints from seismology, mineral physics, and geodynamics. I am also involved in several additional experimental projects investigating the high-pressure properties of water-bearing phases relevant to the interiors of Earth, Mars, and icy satellites.
Please scroll below for more details, related publications, and upcoming presentations!
Melting of iron-bearing materials at high pressures
High-pressure experiments provide essential constraints on thermal profiles of planetary interiors by measuring, for example, the melting temperatures of candidate core compositions and extrapolating to known pressures of a solid-liquid interface like the inner core boundary. However, several major challenges continue to present obstacles for precise measurements of melting temperatures: ongoing disagreements among melt detection techniques even for end-member phases like pure iron, uncertainties in how sample pressures evolve during heating in a laser-heated diamond-anvil cell (LHDAC), and a sparsity of studies investigating viable core compositions that include nickel and one or more proposed light elements.
As the bulk of my PhD research, I have developed a multi-technique approach for precisely measuring the melting temperatures and subsolidus phase transitions of iron-bearing materials in an LHDAC. The experimental approach combines results from two in-situ atomic-level techniques that access different time and length scales: synchrotron Mössbauer spectroscopy (SMS), sensitive to the dynamics of 57Fe nuclei (e.g., Jackson et al. 2013, Zhang et al. 2016), and synchrotron x-ray diffraction (XRD), sensitive to long-range crystalline order. Melting is detected independently by the loss of the Mössbauer signal, produced exclusively by solid-bound 57Fe atoms, and by the onset of an x-ray liquid diffuse signal, produced by molten sample material. For the XRD measurements, we have implemented a burst heating and background updating method to reduce noisy background fluctuations and improve detectability of small melt volumes. XRD data additionally constrain subsolidus phase transitions, provide in-situ access to sample pressures at high temperature, and allow for monitoring of any unexpected chemical reactions during heating.
In collaboration with researchers at Sectors 3-ID-B and 13-ID-D of the Advanced Photon Source in Chicago, Illinois, we applied this technique to Fe0.8Ni0.1Si0.1, a material compatible with recent estimates for the core compositions of Earth and Mercury. We found excellent agreement in melting temperatures measured independently by the two techniques, giving confidence in the results and improving our understanding of how silicon affects the temperatures of Fe-Ni cores. We also presented an updated model of sample pressure evolution during heating, using in-situ pressure information from this and previous studies. With this model, we re-analyzed published melting data, used multiple independent data sets to fit updated melting curves for fcc-Fe and fcc-Fe0.9Ni0.1, and discussed experimental factors that can explain disagreements among melt detection techniques.
Moving forward, I am currently working on applying this multi-technique approach to other iron-bearing phases relevant to Earth’s deep interior. In the meantime, I invite you to take a look at our recent publication on the melting of Fe-Ni-Si for more of the details. You can also watch a recent virtual talk I gave about these experiments here.
Publications:
Dobrosavljevic, V. V., Zhang, D., Sturhahn, W., Zhao, J., Toellner, T. S., Chariton, S., Prakapenka, V. B., Pardo, O. S., Jackson, J. M. (2022) Melting and Phase Relations of Fe-Ni-Si Determined by a Multi-Technique Approach. Earth and Planetary Science Letters, 584, 117358, doi.org/10.1016/j.epsl.2021.117358.
Upcoming Presentations:
Vasilije will be presenting at the upcoming AGU 2022 Fall Meeting in Chicago, IL.
Quantifying compositions of ultralow velocity zones at Earth’s core-mantle boundary
Increasing evidence supports a holistic view of Earth’s rocky mantle as a unified dynamic complex system featuring chemical cycling from the planet’s surface on one end to the core-mantle boundary (CMB) at the other. While the relatively more accessible surface has been extensively studied and characterized, observation of the CMB has become possible only in the last several decades with advances in the analysis of seismic waves emitted by earthquakes that travel through Earth’s interior and back to the surface. Recent seismic observations have been revealing a complex landscape of multi-scale structures whose physical properties remain major open questions. Through regulation of spatially heterogeneous heat flow across the CMB, this complex landscape controls both mantle dynamics and the generation of Earth’s magnetic field, making it an essential area of study for understanding interactions between our planet’s surface and deep interior.
The most extreme class of detected heterogeneous structures are the ultralow velocity zones (ULVZs), with early pioneering observational discoveries led by the late Don Helmberger at Caltech. These structures are tens of kilometers tall, hundreds of kilometers wide, and exhibit seismic wave speeds 10%-50% slower than the surrounding mantle. ULVZs may be associated with the remnants of primordial mantle material, the generation of deep-rooted mantle plumes like under Hawai’i and Iceland, and interactions between subducted slabs and larger thermochemical piles. Various compositional hypotheses have been proposed to explain the observed extreme velocity reductions, each with important consequences for ULVZ formation histories and dynamics. However, little work has been done to quantify viable compositions of specific ULVZs given the latest constraints with uncertainties from both seismology and mineral physics.
Throughout my PhD, I worked on quantitative approaches for investigating and comparing proposed compositional hypotheses, with a focus on iron-rich (Mg,Fe)O, a material recently shown to have extremely low velocities at CMB conditions (Wicks et al. 2017), large seismic anisotropies (Finkelstein et al. 2018), very low viscosities (Reali et al. 2019), and possible origins in the highly iron-enriched remnants of a crystallizing magma ocean. In 2019, I implemented a method, now included in the MINUTI open-source software, to invert for best-fitting ULVZ compositions with correlated uncertainties given seismic constraints on velocities and densities and mineral physics data on phase elasticity, including measurements I performed on (Mg0.06Fe0.94)O using x-ray diffraction and synchrotron Mössbauer spectroscopy . In our paper (Dobrosavljevic et al. 2019), we showed that the presence of solid iron-rich (Mg,Fe)O can explain seismic observations of many different ULVZs. More recently, I have been part of an interdisciplinary effort with seismology and geodynamics researchers at Caltech to measure and interpret the ULVZ under the Hawai’i hotspot. As part of this recently published study, I have constructed a compositional model of a ULVZ containing solid iron-rich (Mg,Fe)O that is compatible with seismic constraints on the observed ULVZ velocity and height, with mineral physics constraints on phase elasticity and viscosity, and with morphological constraints from geodynamic simulations of a dense, solid ULVZ.
Moving forward, I am looking to further develop this type of multidisciplinary approach, apply it to other study areas along the CMB, and use this quantitative framework to directly compare various proposed compositional hypotheses. In the meantime, I invite you to check out our recently published work on this topics below.
Publications:
Lai, V. H., Helmberger, D. V., Dobrosavljevic, V. V., Wu, W., Sun, D., Jackson, J. M., Gurnis, M. (2022) Strong ULVZ and Slab Interaction at the Northeastern Edge of the Pacific LLSVP Favors Plume Generation. Geochemistry, Geophysics, Geosystems, 23(2), doi.org/10.1029/2021GC010020.
Dobrosavljevic, V. V., Sturhahn, W., Jackson, J. M. (2019) Evaluating the Role of Iron-Rich (Mg,Fe)O in Ultralow Velocity Zones. Minerals, 9 (12), 762, doi:10.3390/min9120762.
Additional project collaborations
In addition to my main research projects, I have had the opportunity to collaborate on several other studies. I am involved in a series of projects using a suite of experimental techniques to measure the high-pressure and low-temperature physical properties of an iron-bearing hydrated sulfate (szomolnokite), with applications for icy planetary bodies and satellites. For more details, I invite you to be on the lookout for a study in press at American Mineralogist from lead author Olivia Pardo. I have also been involved in a study measuring the thermal equation of state of 𝛿-(Al,Fe)OOH, with consequences for the properties of hydrous rocks in the lower mantle and the cycling of water through Earth’s deep interior more generally, led by Johannes Buchen.
Vasilije V. Dobrosavljevic 