The Fate of PAHs Resulting from In-Situ Oil Burns

PAHs are ubiquitous in the environment and are considered to be of concern for both man and the environment (Lima et al., 2005). There are many sources of PAHs, but primarily come from combustion sources. Natural sources such as forest fires, are a contributor. Crude oil contains PAHs, however oil spills themselves are not a significant contributor of PAHs to the environment. Burning of petroleum could constitute a significant input of PAHs into the environment. In particular, it has been pointed out that the two most significant contributors are diesel engines and home heating using a type of fuel similar to diesel fuel (Lima et al., 2005). Heating with wood also contributes some PAHs. Sootier flames appear to be associated with more PAHs. The amount of PAHs in the starting fuel also influences the amount of PAHs emitted by a burn. Generally low temperature sources such as diesel engines are larger emitters of PAHs than high temperature sources such as fires.

PAHs are compounds consisting of at least two benzene rings. Descriptions of common PAHs are found in the literature (Fingas, 2016). There exists a set of compounds designated by the U.S. EPA as priority PAHs (Wise et al., 2015). The concern with these compounds is that some of them are known to be relatively toxic and some to be carcinogenic. The amounts of PAHs in a typical crude oil varies, but range from 0 to 5%. In crude oils, the alkylated compounds (branched hydrocarbons from the PAH rings) occur more frequently than the parent un-alkylated rings. This can be of use in identifying the source of contamination as many sources of PAH pollution have more abundant parent compounds than alkylated ones.

Diesel contains significant levels of PAHs of smaller molecular size, with 2- 3-ring PAHs predominating. Burning diesel results in more pyrogenic PAHs of larger molecular sizes. Larger PAHs are either created or concentrated by the fire. Larger PAHs, some of which are not even detectable in the Diesel fuel, are found both in the soot and in the residue. The concentrations of these larger PAHs are low and often just above detection limits. The question is, are more PAHs are destroyed by the fires than are created?

The PAHs in oil may have any or several of the following fates: 1) be burned to CO₂ and water, 2) be converted to other chemicals such as other PAHs or to oxygenated PAHs, 3) be transmitted through to the gaseous emissions of a fire, 4) be absorbed on the particulate emissions (soot) from the fire, or 5) accumulate on the residue or unburned portion. These processes are illustrated in Figure 1.

Several early studies noted that the concentrations of PAHs were about the same in the residue as in the starting oil (Fingas et al., 1995; Li et al., 1992). The observations at the burn experiments reported at these trials included: 1) that generally the PAH concentrations in the residue and the starting oil were similar but never identical, 2) that for diesel burns there appeared to be more larger PAHs present in the residue and 3) tests of the volatile emissions showed low concentrations of PAHs during both crude and diesel burns. Several interpretations of these results ensued, however no firm conclusions could be made because the scientists were unsure of the mass balance of the burns between the soot and the residue compared to the starting oil. It is important to recognize that the concentration of PAHs in the starting oil and the residue is only one facet. An important facet of the puzzle is the efficiency of the burn, or how much of the oil (and PAHs contained therein) were burned. To illustrate this, presume that 100 kg of oil were burned with a starting concentration of 1% PAHs - this would mean that there was 1 kg of PAHs. If the burn efficiency was 95% and the concentration of the PAHs was still 1% in the residue, then 95% of the PAHs were destroyed by the fire. In addition, one should consider the PAHs lofted in the soot and emitted as vapour.

A basic study on the fate of PAHs in diesel burns was carried out by Wang et al. (1999). This study used several mesoscale burns conducted in Mobile Bay, AL, to study various aspects of diesel fuel burning. The target PAHs in the diesel, residue, and soot samples collected during each burn were quantitatively characterized by GC/MS (Gas Chromatography/Mass Spectrometry). A simple model based on mass balance of individual petroleum PAHs pre- and post-burn was proposed to estimate the destruction efficiencies of the PAHs. This study demonstrated that distributions of PAHs in the original diesel and soot were actually different. The average destruction efficiencies for the total target diesel PAHs including five alkylated PAH series and other EPA priority unsubstituted PAHs were greater than 99%. Using the model, 27.3 kg of the diesel PAHs were destroyed for each 1000 kg of diesel burned. These were mostly two- and three-ring PAHs and their alkylated homologues. Combustion also generated trace amounts of high-molecular-weight five- and six-ring PAHs as well as the four-ring benz[a]anthracene. But the total mass of these pyrogenic PAHs was found to be extremely low: only 0.016, 0.032, and 0.048 kg of the five- and six-ring PAHs were generated by combustion in the three different scenarios for each 1000 kg of diesel burned. It was concluded that in-situ burning is an effective measure to minimize the impact of an oil spill on the environment, greatly reducing exposure of ecosystems to the PAHs of spilled oils. This study showed several trends related to the destruction of PAHs in diesel burns:

1. It appears that the greater the number of rings, that the poorer the PAH destruction percent. For the priority PAHs, the trend is generally that as the ring numbers increases, the destruction percent decreases.

2. It appears that for the alkylated PAHs, the destruction percent decreases with increasing alkylation.

An examination of the vapour component shows that during three diesel burns, the concentrations of the only measureable PAH, naphthalene, was 0.2 out of 169 μg/m³; 0.4 out of 153 and 0.7 out of 176 total μg/m³ of volatile organic compounds (for three different burns). The background was 0.7 μg/m³, therefore the PAHs in the vapour phase are negligible and often below background levels.

The next step was to examine the fate of PAHs in crude oil burns. The PAHs data set chosen for this study was that of the Newfoundland Offshore Burn Experiment (NOBE) which used a medium Alberta crude oil (Fingas et al., 1995). Table 1 shows the PAHs in the starting oil, soot and residue as measured from one burn at this study. Performing the mass balance calculation using the value of 1% for the soot and the measured oil burn efficiency of 99% (the maximum estimated at this burn), it was found that 99.99% of the PAHs were burned. This also shows that the PAH destruction was higher for crude oil than for diesel fuel. Data for some diesel burns are also shown in Table 1.

It is noted that the destruction of the PAHs in the soot is usually 99% indicating that this is not a strong route of PAH dissemination. This also indicates that PAHs on soot may be ignored in terms of the overall mass balance of PAHs in a burn.

Table 1 also includes data from a diesel fuel burn. The alkylated PAHs burn with approximately the same efficiency with a greater degree of alkylation.

A look at the vapour component shows that during the two NOBE burns, the concentrations of the only measureable PAH, naphthalene was 0 to 7.3 μg/m³ out of 6000 to 24,000 total μg/m³ of volatile organic compounds (for each of the two burns). The background was 0 to 6 μg/m³, therefore the PAHs in the vapour phase are negligible and sometimes below background levels.

From Table 1 it can be seen that the PAHs in the crude oil soot are 2239 μg/g or 2% of that in the residue if considering the alkylated compounds, more if considering the target PAHs. With approximately the same percentages (<0.1%) of residue or soot, it can be concluded that the PAHs on the soot are not as significant fate for PAHs as in the residue. In the case of the crude oil, this is more pronounced than for diesel fuel.

Test burns of diesel fuel were again conducted in 1997 at Mobil, AL (Fingas et al., 1999; Wang et al., 1999). Similar results were obtained as before, namely that the destruction was nearly complete for all burns and all compounds. Similar burn tests were conducted in 1998 at Mobile, AB. (Fingas et al., 2000). Similar results were obtained, namely that the destruction was nearly complete for all burns and all compounds. Again, there does not appear to be differences in destruction percentages among the various alkylated PAHs nor among the priority PAHs. However, there are some higher concentrations of priority PAHs in the residue.

Garrett et al. (2000) carried out burn tests at small scale on Statfjord crude oil and found that the PAHs were reduced through burning. In addition to the 85% burned, an additional 40% of the PAHs were removed.

Burn tests on heavy and residual oils were conducted (Fingas et al., 2005). These burns were special in that they were carried out on heavy oils and bitumen, some mixtures and waste oils. The chosen heavy oils were poorly characterized before the burn. The Bunker C was somewhat characterized, while the test oil was several years old and was used to test skimmers. Its origin was unknown, however it was very viscous and low in PAHs. The Orimulsion and Bitumen were of similar origin and were mixed samples from the supplier of the product in Venezuela. The PAH destruction was variable and low for the two Orimulsion burns and the test oil burns. Burn efficiency was low for the test oils, as low as 2%. The Bitumen burns also had a lower burn efficiency. The PAH destruction percentages were higher compared to what one would expect, but that is because the boiling points of the PAHs are lower than the rest of the components in the oil and would be burned early in the combustion process.

Lin et al. (2005) performed a series of test burns in 30 cm pots containing marsh plants. Both diesel fuel and Louisiana crude oil were used. PAH removals of 99.4 and 98.9 for the diesel and crude were observed. It was also noted that there were increases in the priority PAHs as had been observed by Wang et al. (1999), especially for the diesel fuel. However, overall the PAH destruction was quantitative.

During the Deepwater Horizon spill in the USA, 411 successful burns of weathered and sometimes emulsified oil were carried out. Shigenaka et al. (2014, 2015) carried out a series of test burns as well as collected samples of oil burn residue at the site of the Deepwater Horizon burns. The PAHs in the collected samples and the starting oil was determined. The results show that the destruction percent of the burns are typically well over 99% with some of the priority PAHs being less – typically around 98%. This is consistent with previous data. It should be noted that some of the samples were scraped from the fire-resistant boom and thus they may have a slightly different composition from the residue. Shigenaka et al. (2014) indicated that these samples may not be burned as much as the residue, however from this analysis it appears that they are.

Fritt-Rasmussen et al. (2013) conducted several small scale burns on Troll B crude oil and noted that there was an increase in some of the 16 priority PAH compounds. A mass balance was not calculated. Fritt-Rasmussen et al. (2015) carried out a review of burn residue noting that there was an increase in some of the 16 priority PAH compounds

Shigenaka et al. (2014) also carried out a series of small test burns to assess the chemical changes that burns undergo. These include a very small burn of Macondo oil, pan burns of South Louisiana crude and burns of emulsified Macondo oil. The burn efficiencies were not measured, however can be estimated from comments about the burnability. The Macondo oil burned with a high destruction percent. The Louisiana crude oil similarly. With the estimation of 80% burn efficiency for the Macondo emulsion, PAHs were still largely destroyed.

Gullett et al. (2016) studied the particulate matter emanating from the Deepwater Horizon burns. They sampled PAHs on particulate matter captured by volumetric samplers as well as on the sail of an aerostat flown in the smoke plumes. The mean sum of PAHs detected in the sail PAH extract and PM 2.5 Filters were 80.1 μg/g and 68.2 μg/g of the particulate matter, respectively, accounting for less than 1% of the total particulate matter mass. At the PM collection rate from the fires of 0.088 g particulate matter/g C12, this resulted in an emission factor of 4.5 mg PAH/kg of oil burned. These results are much lower than would be expected by PAH content of the oil, indicating an overall destruction of PAHs.

Stout and Payne (2016) gathered 4 residue samples from the DeepWater Horizon burns both on the surface and sub-surface. They analysed the PAHs and other composition factors and concluded that the burns resulted in 89% destruction of PAHs.

The data summarized in Table 2 show that for the most part PAHs are destroyed in fires. Further this analysis shows that there are slight differences between crude oil and diesel burns, in that crude oil burns appear to be slightly more efficient in terms of destroying PAHs. The analysis of the PAHs going to various compartments clearly shows that most PAHs are destroyed in the fire, with some remaining with the residue. Burn efficiency does not appear to change the PAH distribution to any extent.

The main points concerning PAHs in oil burns can be summarized as:

1. Some diesel burns ended up with minor production of higher molecular weight PAHs. This was not found to be significant in any of the crude oil burns. The concentration of the 16 EPA priority compounds is higher in the burn residue than the starting concentrations and the concentrations of the alkylated PAHs are lower in the residue than that of the starting oil.

2. Diesel burning is somewhat less efficient than crude oil burning in the destruction of PAHs and does result in more soot with its incumbent PAH load.

3. The amount of PAHs emitted as vapour is negligible.

4. The amount of PAHs on the soot is variable, however in most cases is negligible. The amount of priority PAHs on the soot maybe somewhat elevated.

5. Burn conditions vary and result in variable PAH destruction results.

6. PAH destruction is dependent on the total burn efficiency.