(431e) Partial Oxidation of Heavy Oil Model Compounds in Supercritical Water

Background

Asphaltenes are the heaviest and most complex molecules found in heavy oil. They are composed of polycyclic aromatic hydrocarbons (PAH) and normally contain considerable amounts of heteroatoms (S and N) and metals (Ni and V) [1]. The presence of these compounds represent a great challenge in oil processing as they cause temporary or permanent catalyst deactivation, high coke deposition and low yields to valuable products [2]. Heavy oil upgrading is normally performed through processes based on increasing the hydrogen to carbon ratio in the feedstock as catalytic hydrocracking. Studies with PAHs have shown that reduction of the aromatic rings is effectively achieved and preferentially proceeds starting at the peripheral rings rather than central rings [3]. The previous suggest that cracking proceeds from the peripheral rings to the central cluster which causes a great mass loss throughout the upgrading process. In order to effectively process heavy crude oils, it is necessary to develop processes in which PAH clusters can be cracked into smaller compounds of greater value keeping a high atom economy within the process. In nature, PAHs are effectively degraded by enzymes found in fungi through oxidation of the central aromatic rings [4]. A process combining the concept of fungi degradation of PAHs and the properties of supercritical water (SCW) to upgrade heavy oil feedstocks is being proposed. It takes advantage of the reactive nature of the medium and the high solubility of gas and organic compounds in SCW to partially oxidize heavy oil molecules. Studies with model compounds have shown that partial oxidation of PAHs at SCW conditions readily proceed through oxidation of the middle ring and give rise to aromatic systems of a reduced size with low yields to coke [5]. In this work, partial oxidation of PAHs in SCW at early reaction stages has been studied in a continuous flow reaction system. Conversion, yields and selectivities to products were determined and an initial reaction pathway has been proposed.  The effect of the presence of a solvent was also studied.

Methodology

Experiments were performed in a 1.5 m SS316 (1/4”) coil type flow reactor with Sigma methyl naphthalene 98% and Sigma phenanthrene 98% as PAH and Sigma heptane 99%, Sigma toluene 99.8% and Sigma 99.5% as organic solvents. Reaction conditions studied were temperatures between 400°C and 450°C, pressure of 230 bar, oxygen to organic ratio of 5% and residence times between 15 sec and 1 min.  Organic reactants were fed into the system with a Varian 9002 HPLC pump. Water and H₂O₂ 30% w/v as oxygen source were initially fed into a 2m SS316 (1/8”) coil with a Varian 9010 HPLC pump. The peroxide fully decomposes into oxygen and water before mixing with the organic stream. The H₂O₂ decomposition coil and the coil reactor were heated in a TECHNE SBL-2D fluidized sand bath. After reaction the product is cooled down in a heat exchanger to fully stop the reaction. Pressure in the system is sustained by means of a Swagelok back pressure regulator (BPR). Solids formed are filtered before reaching the BPR. Water soluble, gas and organic soluble products are depressurized and collected in a product vessel. Gas generated was analysed in a Perkin Elmer Clarus GC with TCD equipped with a Carboxen plot 1010 capillary column. The organic soluble fraction was then analyzed in a Perkin Elmer Clarus GC with FID and in a Varian Star 3400/Saturn 2000 GC/MS to identify the amount and nature of the products obtained. Both instruments are equipped with a non-polar HT-5 (25m × 0.32mm) column. The solid fraction obtained were measured and analyzed in a Perkin Elmer Pyris 1 TGA to determine the amount of coke produced.

Results and Discussion

Experimental results showed that partial oxidation of polycyclic aromatic hydrocarbons proceeds with high conversions at short residence times. Reactions in the presence of a solvent showed that PAH conversion, yield and selectivity to products greatly depend on the nature of the organic solvent used. In reactions at 1 min residence time phenanthrene conversion were 47%, 37% and 7% with heptane, toluene or cyclohexane respectively. The previous suggest that heptane and toluene react with phenanthrene which increases conversion and will also have an impact in the product selectivity. Low yields to coke below 1% were obtained in the presence of all of the solvents. Yields to gas of 3% were obtained when toluene was used as solvent while the yields to gas remained below 1% in the presence of heptane and cyclohexane. In all cases a H₂ reach gas was obtained. Blank experiments with each of the solvents showed that toluene was considerably more reactive than other solvents producing an important amount of oxygenated products and polymerization products. The previous show that the nature of the solvent used plays an important role in the reaction pathway, overall conversion and selectivity to different products. Experiments in the absence of a solvent were performed with mehyl naphthalene in order to avoid the solvent influence. Residence times as low as 15 sec were studied in order to analyse the partial oxidation at early reaction stages. It was observed that in the absence of a solvent a narrow product distribution is obtained where the main products identified were naphthol, naphtoquinone, and organic acids. Similarly as when solvent was present the amount of coke produced was negligible in the range of temperatures and residence times studied. Yields to gas varied between 2 and 4 % and were strongly dependant on the reaction temperature. From experimental results a reaction pathway for the partial oxidation of methyl naphthalene will be proposed.

References

1. Ancheyta J, Speight JG. Hydroprocessing of heavy oils and residua. CRC Press; 2007. 366 p.
2. Ancheyta J, et al. Hydroprocessing of Maya heavy crude oil in two reaction stages. Applied Catalysis A: General. 2002 Jul 10;233(1–2):159–70.
3. Korre S.C, et al. Polynuclear Aromatic Hydrocarbons Hydrogenation. 1. Experimental Reaction Pathways and Kinetics. Ind Eng Chem Res. 1995;34(1):101–17
4. Hammel KE. Mechanisms for polycyclic aromatic hydrocarbon degradation by ligninolytic fungi. Environ Health Perspect. 1995 Jun;103(Suppl 5):41–3.
5. Daud ARM, et al. Heavy oil upgrading in subcritical and supercritical water: studies on model compounds. Prepr. Pap.-Am. Chem. Soc., Div. Energy Fuels 2012, 57 (2), xxxx