Project Overview

     Dr. Ryan Rodgers at Florida State University’s National High Magnetic Field Laboratory, was awarded an RFP-VI grant at $876,444 to conduct the RFP-VI project titled, “Molecular Understanding of Emulsion Formation in Environmentally Photo- Oxidized Crude Oils: In Situ Generation of Interfacially Active Species and Their Impact on Emulsification”. The project consisted of 1 institution (University of New Orleans); principal investigator (Rodgers); 2 co-PIs (Drs. Matthew Tarr, Yuri Corilo); 1 post-doc working on the project (Martha Chacon); 4 PhD students (Mohamed Elsheref, Chaka Brown, Sydney Niles, Shaina Patil), 4 undergraduate students (Lena Messina, Jennifer Zeron, Malaurie Giraudier, Camille Infante), and 1 high school teacher (Alexandria Cluen-Brown).

 

     Once released, petroleum undergoes transformation (weathering) that includes physical processes that modify its native composition (water washing and evaporative losses) and chemical processes (largely oxidative, photo-oxidation and biodegradation) that generate new, increased oxygen-containing chemical functionalities in the released oil. The transformation alters both the structure and chemistry of the petroleum, and thus significantly changes the native oil’s physical properties (density, viscosity, surface tension, etc.). Combined, these changes are responsible for the formation of oil mats, thick emulsions, and differing sheen morphologies observed in oil spills that affect oil transport, fate, and spatial distribution. Weathering processes that most influence these changes are evaporative loss of the light ends (topping) and generation of surface active (interfacially active) oxygen-containing species through photo-oxidation. Recent advances in analytical instrumentation have exposed tens-of-thousands of oxygen-containing transformation products, and established methodologies for their detailed molecular-level characterization. Specifically, a novel method to isolate interfacially active species from hydrocarbon matrices allows for the identification of species responsible for emulsion formation. For the first time, emulsion-causing species can be isolated and analyzed prior to emulsion formation. Interestingly, unaltered crude oils that form stable emulsions in oil production/refining are shown to contain oxygen-containing species similar to those molecular-level compositions identified in photooxidized crude oils. Thus, photo-oxidation appears to generate interfacially active species in situ from the largely hydrocarbon matrix of crude oil. We hypothesize that these species are largely responsible for thick emulsions/mousses observed on the gulf surface after the Deepwater Horizon oil spill.

 

     In response to research theme ii, we aim to combine the photo-chemistry expertise of Dr. Matthew Tarr (University of New Orleans) with the analytical, petroleum emulsion, and data analysis capabilities of Dr. Ryan Rodgers and Dr. Yuri E. Corilo (both Florida State University) to address the following: (1) what interfacially active species are generated during photo-oxidation? (2) what is their effect on stable emulsion formation? (3) How does this composition change with increased irradiation times? (4) What is the predominate oxygen-containing functionality (aldehyde, ketone, hydroxyl, carboxylic acid) that contributes to stable emulsion formation, and how does that change with the initial oil composition?

 

     The current proposal will irradiate a light (Macondo surrogate), medium, and heavy crude oil for different time periods. To these photo-microcosm samples, we will add a collection of previously collected field samples (sheen, mousse, sediment extracts, and tar balls). The FSU patented “Wet Silica Method” will be used to isolate interfacially active species from photo-irradiated crude oils and field samples. Previous results reveal that although these isolated species account for less than 1 wt.% of the crude oil, they alone are responsible for stable emulsion formation, and are largely composed of oxygen-containing species similar to those identified in photo-irradiated crude oils. Since the method allows clean isolation of interfacially active species, their ability to form stable emulsions in seawater can be directly measured with simple bottle tests. Detailed molecular-level information is provided by Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (FT-ICR MS). Attribution of chemical functionality within the interfacial materials and their importance in stable emulsion formation is afforded by previous methods developed for DEEP-C and a carboxylic acid fractionation technique developed by FSU, called MAPS. Thus, the molecular composition, chemical functionality, and importance to emulsion formation of all transformation products are revealed. Since these highly oxygenated species are not native to the parent crude oil, confident assignment of the mass spectral data by previously developed algorithms can be difficult. Dr. Corilo is a world expert in complex mixture software development, and creator of PetroOrg, a commercial software platform for petroleum data analysis/visualization, used by most of the world’s largest oil companies. We will deliver a software platform specifically aimed at processing mass spectral data from environmentally transformed crude oils that will be provided to the scientific community free of charge. It will contain a photo-oxidation database that will be populated with all data obtained in this project. Given the importance of emulsification in the fate, transport, and cleanup of spilled oil, it is essential to comprehensively understand the emulsification process, its temporal change, and relation to field samples/observations, including previously unaddressed/critical photochemical inputs.

 

Research Highlights

 

     Dr. Rodgers’ research, which included 1 outreach activity, resulted in 3 peer-reviewed publications, 26 scientific conference presentations to date, and 4 datasets submitted to the GoMRI Information and Data Cooperative (GRIIDC), which are available to the public.  Significant outcomes of their research (all related to GoMRI Research Theme 2) are highlighted below.

 

University of New Orleans:

Results to date have directly demonstrated the emulsification of oil as the result of exposure to simulated sunlight.  We separately exposed neat oil from the Macondo well, surrogate oil, and a heavy fuel oil (NIST N2717a) to simulated sunlight and assessed the mass loss and ability to emulsify.  A commonly used emulsification bottle test was applied to dark and irradiated samples (times ranging from 1-48 hours, equivalent to approximately ¼ day to 12 days of average solar exposure in the Northern Gulf of Mexico).  Dark controls were heated at 50 ?C for up to 24 hours to assess evaporative effects.  Dark controls, heated or not, showed no emulsification.  By contrast, irradiated samples showed emulsification after one hour of exposure, and the degree of emulsification increased with exposure time.  Table 1 shows crude oil emulsion capacity vs. time (heated in dark or irradiated) for various oil samples neat or on water.  A semi-quantitative scale of emulsification capacity was defined as: 0 = no emulsion, 1 = minor emulsion, 2 = moderate emulsion, 3 = stable emulsion, 4 = very stable emulsion.  The three oils behaved differently under simulated solar exposure (Table 1).  In addition, irradiated neat oil yielded higher emulsion capacity compared to oil irradiated over water.  This effect was more dramatic when the oil was irradiated over seawater. 

 

Table 1. Crude oil emulsion stability vs. time (heated in dark or irradiated) for various oil samples neat or on water.  N2717a is a high sulfur heavy fuel oil standard from NIST.  Degree of emulsification scale: 0 = no emulsion, 1 = minor emulsion, 2 = moderate emulsion, 3 = stable emulsion, 4 = very stable emulsion.

 

 

Derivatization of oil photoproducts with dansyl chloride revealed the formation of products.  Through the use of o-phthalaldehyde derivatization, it was confirmed that no amine products were observable, indicating that the dansyl chloride derivatives were phenolic.  Mass spectrometric analysis confirmed this conclusion.  Time dependent studies revealed that these photoproducts increased in concentration with increased simulated solar exposure, further confirming that photochemistry was responsible for the formation of these products.  Figure 1 compares dark (top) and irradiated (bottom) samples, showing formation of new derivatized compounds in the irradiated sample.

 

 

 

National High Magnetic Field Laboratory (FSU):

Recent results have highlighted the importance of carboxylic acid functionalities on emulsion formation, although ketone and hydroxyl functionalities are also created and detected.  However, molecular structure was also determined to greatly contribute to the generation of oil soluble and water soluble photo-products.  Single aromatic core (island) dominated crude oil asphaltenes have been shown to generate a small amount of water soluble species, but the photo-products largely remain oil soluble (Figure 2, right).  Interestingly, crude oil asphaltene fractions enriched in multi-core (archipelago) structures generate much higher levels of water soluble species, and depending on the crude oil, abundant surfactants that can cause very stable emulsions (Figure 2, left).  Future research will highlight the compositional similarities and differences between the photo-products. 

 

 

Figure 2. Extrography separation enables the isolation of asphaltene fractions enriched in either single-core (island) structures (right) or multi-core (archipelago) structures (left).  Representative structures of each are shown in black.  Fractionation of a Kuwait asphaltene yielded island dominant and archipelago dominant fractions that were each photo-irradiated.  After photo-irradiation, the two fractions exhibit very different behavior.  The island dominant (right) yields a small amount of water soluble species, but most of the photo-products remain oil soluble.  The archipelago fraction yielded a very stable oil in water emulsion (left).

 

The results led to a follow-up experiment to test for the potential generation of asphaltenes from the maltene fraction (in samples where asphaltenes have been removed).  The maltene fraction has been previously shown to be mostly island (single-core), but smaller in size than asphaltenes.  Most importantly, it has been suggested that asphaltene content can be used to estimate the amount of spilled oil in an accidental release.  This hypothesis would be invalid if sunlight produced asphaltenes from non-asphaltene (maltene) species.  Figure 3 shows that maltenes do indeed form asphaltenes upon solar irradiation.

 

 

Figure 3. Photo-irradiation of a maltene film on seawater generates abundant water-soluble species, but more importantly, 17% wt asphaltenes.  Thus, photo-oxidation of non-asphaltenic species generates asphaltenes.

 

This behavior has been shown to be crude oil dependent (Figure 4).

 

 

Figure 4. Maltenes isolated from Wyoming, South American, and Middle Eastern crude oil reveal different photo-chemical production yields of asphaltenes. 

 

The molecular analysis of the original maltene fraction, the photo-irradiated maltenes, and photo-generated asphaltenes reveals that solubility is correlated with solubility; the species that contain the greatest number of oxygen atoms are the least soluble in heptane (i.e. asphaltenic).  The FT-ICR mass spectrometry results are shown in Figure 5.

 

 

Figure 5. FT-ICR mass spectral analysis of maltenes isolated from Wyoming crude oil (blue), after photo-irradiation (red), and asphaltene fraction (green) reveals that the most photo-oxygenated species are asphaltenic. 

 

Collectively, the results underscore the importance of in situ generated surfactants in surface slicks due to photo-oxidation.  These surfactants contribute to increased water content of surface slicks, increased viscosity, and decreased effectiveness of dispersants. Carboxylic acids are largely responsible for emulsion formation, but the mechanism and rate of formation remain undetermined.  However, the low aromaticity of the carboxylic acids identified in this, and previous studies on microcosm and field samples suggest that indirect photo-oxidation must be considered. Asphaltenes, previously thought to be unreactive and recalcitrant, have been shown to photofragment in simulated surface slicks to generate both water soluble, oil soluble, and oil soluble interfacially active species.  However, their behavior is structure dependent.  Single-core asphaltenes overwhelmingly generate oil soluble photoproducts that lead to a “tar-like” water insoluble material.  Conversely, multi-core asphaltenes lead to abundant oil and water-soluble photo-oxidized species that can contribute to very stable emulsions. Finally, the utility of asphaltene content for field samples to estimate the amount of spilled oil is not recommended.  Deasphalted (asphaltene free) oil samples were shown to contain asphaltenes (up to 17% wt) after photo-irradiation.  Thus, the current results suggest that a fraction of asphaltenes are photofragmented to water and oil soluble species (“consumed”) concurrent with the oxidation of maltenic species (non-asphaltenes) to produce asphaltenes (“produced”).  This process is oil dependent and cannot currently be predicted from the analysis of the parent oil.  Thus, asphaltene content of field samples does not appear to be a reliable indicator of the amount of spilled oil, as it would depend on the initial oil composition and photo-history of the sample.