Project Overview

Summary:

Overview

In January 2016, Dr. Ryan Rodgers at the Florida State University, National High Magnetic Field Laboratory , was awarded an RFP-V grant totaling $2,943,781 to lead the GoMRI project entitled “The State-of-the-Art Unraveling of the Biotic and Abiotic Chemical Evolution of Macondo Oil: 2010-2018”, which consisted of 3 collaborative institutions and approximately 12 research team members (including students). This project aimed to apply ultra-high resolution mass spectrometric analyses in order to understand how weathering processes affect oil, to quantify rate(s) of oxygenated oil weathering product formation and degradation, and characterize toxicological effects on the ecosystem. More specifically, this project aimed to answer the questions: (1) How does the molecular composition of MWO oil change over time? (2) Which compositional changes are caused by photo-oxidation? Biodegradation? How does the structural / chemical composition of the oil influence oxidation? (3) How does this compositional change influence toxicity of weathered MWO? (4) What is the overall fate of MWO on a time scale of 8 years.

This project tracked the continued weathering of MWO and focused on early sampling dates (0-10 months) immediately after the spill, where a rapid formation of oxygenated products was hypothesized, as well as highly weathered samples (to be collected up to eight years after the spill). The proposed analytical methodologies captured bulk and molecular level, biotic/abiotic temporal compositional changes in the MWO as it weathers in the environment. The efforts generated a compositional database of the quantitative and qualitative weathering of MWO. Second, analysis of field samples was combined with controlled laboratory experiments of MWO photo-oxidation and biodegradation. Third, MWO and other oils, their structurally defined fractions, and all weathering products for each, were screened for toxicity (narcosis), and observed effects were linked (correlated) to the molecular compositional change in MWO during weathering. Finally, since the structural dependence of weathering captured herein, along with each fractions toxicity (and water soluble fractions), a simple model was constructed based on the quantitative yields of each structural fraction, its associated weathering products, and rate of formation. Thus, simple quantitative fractionation of any future contaminant could potentially be used to predict the rate, mass, and type of weathering products formed.

Research Highlights

As of December 31, 2019, this project’s research resulted in 7 peer-reviewed publications, 60 scientific presentations and 11 datasets being submitted to the GoMRI Information and Data Cooperative (GRIIDC), which are/will be made available to the public. The project also engaged 2 PhD students over its award period. Significant outcomes of this project’s research are highlighted below.

Prior to the DWH incident, the majority of field/laboratory studies monitored molecular changes to oil via gas chromatography (GC) based methods.  In fact, 2 of the 3 PI’s on this grant are experts in field, and continue to use the methods to understand oil weathering.  However, the significant findings of this grant have contributed to our increased understanding of what species are not captured by GC methods, and developed methods/techniques to expand our “analytical window” to capture a wider range of chemistry and structural information on oil weathering products.  Specific, noteworthy findings are as follows:

Oxidation is much more important than previously thought.  Bulk analytical techniques (thin layer chromatography – flame ionization detection (TLC-FID) and Fourier Transform Infrared spectroscopy (FT-IR)) applied to field samples over an 8 year period revealed that oxidation of the oil was rapid (days – week(s)) and extensive (> 50 wt.% of remnant oil were oxidized species).  Thus, the current results were used to correct previous weathering models to underscore the importance of photo-oxidation (rate and mass of material formed).  GC analyses of the same samples were used to highlight classic weathering patterns such as loss of alkanes and aromatics with a concomitant increase in the unresolved complex mixture (UCM).  Unsurprisingly, the GC-based methods did not reveal/identify any of the newly formed, oxidized transformation products.  Conversely, High resolution Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (FT-ICR MS) revealed tens-of-thousands of newly formed, oxidized species that changed with weathering time period.  Notably, the mass spectral resolution and mass accuracy required for successful identification of these species can be only currently be attained with high field (9.4T or greater) FT-ICR mass spectrometers.  Lab-based photo-oxidation experiments also highlighted the utility of FT-ICR MS in the identification of water soluble species formed by the photo-oxidation of crude oil.  The process generates a continuum of oxidized species that span both oil and water solubles.  Comparison of the oil-soluble photooxidation transformation products from the microcosm experiments to that of the oxidized species identified in field samples confirmed that they were formed by photooxidation.  Separate bio-oxidation microcosms were shown to generate very small amounts of oil soluble Ox species that were very different than those formed by photooxidation.

Photo-oxidation generates interfacial material.  The microcosm experiments facilitated the discovery of interfacially active material in photooxidized crude oils.  A previously developed method allowed for the isolation of this material and subsequent characterization.  The results confirmed previous theories that sunlight generates surfactants in situ, and these surfactants subsequently lead to emulsion formation.  We also confirmed that these photooxidation generated surfactants are what limit the effectiveness of dispersants applied to weathered surface slicks.  Simply, the application of a surfactant does not disperse the oil effectively because it already contains photo-generated surfactants.

Different oil, different result.  Comparison of microcosm results for light, medium, and heavy crude oils, as well as refined products such as residual fuel oil, demonstrated that the photooxidation behaviors were sample dependent.  Thus, future spills will require rapid, lab-based experiments to determine how they will be affected by weathering processes.  MicroTox toxicity assays revealed that MWO water soluble species generated by photooxidation were more toxic than the original oil.  However, the increased toxicity was only evident in early time points (24-48 hours of irradiation) and dropped below that of the oil for later time points.  The exact reason behind this observation is currently unknown. Other oils tested did not exhibit increased toxicity at any irradiation time period.

Photooxidation mechanisms are more complicated than previously thought.  Surprisingly, the analysis of field and lab microcosm samples yielded abundant oxidized alkanes and cycloalkanes.  This is noteworthy, as they contain no chromophore (aromatic rings) and absorb weakly in the visible light spectrum.  The results were consisted across multiple oil samples and irradiation periods, and therefore suggest that both direct and indirect oxidation mechanisms must be considered.  Furthermore, the isolation of structurally-defined (saturates, 1-, 2-, 3-, 4-, and 5+aromatic ring) fractions followed by photoirradiation yielded oxidized transformation products that were all similar.  This result was not expected, and subsequent experiments were performed to reveal the process that led to the photo-products.  The results demonstrated that all three proposed photooxidation processes (oxidation, polymerization, and cracking) occur simultaneously.  Thus, low molecular weight species can polymerize to form higher molecular weight species, and high molecular weight species can crack to form low molecular weight species; all of these products can then oxidize further.  This greatly complicates the analysis of the structurally-defined fractions, but explains the similarity of the photo-oxidation products.  In separate studies, we also demonstrated the ability to isolate and characterize photogenerated acids and ketones from both field samples and microcosms. 

In Summary.  Photooxidation is now known to be a dominant weathering process in oil spills.  It generates both oil soluble and water soluble transformation products, and a fraction of the oil soluble species (those with the higher number of oxygen atoms) are interfacially active and hamper the effective use of dispersants.  Those that contain the highest number of oxygen atoms are water soluble.  Solubility (water vs. oil) is a function of both oxygen content and carbon number.  The water soluble species may exhibit increased toxicity relative to the parent oil, but it is sample dependent.  Photooxidation, photopolymerization, and photocracking occur simultaneously and generate a continuum of oxidized products that contain acid, ketone, and alcohol functional groups.  Most importantly, analytical methods now exist to study weathering products that cannot be analyzed by GC-based methods.  Thus, future research can advance beyond what can be eluted and detected from a gas chromatograph.

 Proposal Abstract - RFP-V PI Ryan Rodgers