(279f) From Quantum Chemistry to Diesel Injector Deposits: Revisiting Liquid Phase Oxidation

Liquid phase oxidation of hydrocarbons has been the subject of experimental and theoretical investigations for over 75 years owing to its applications in many areas of practical importance.¹Although the chemical identity of primary and secondary products of liquid phase oxidation are well established, elementary reactions leading to their formation and consumption are known for only a few. Understandably, kinetic modeling of these processes is challenging and requires tracking thousands of intermediates and reactions. The commonly accepted paradigm in the literature has been to build models manually, often lumping isomers and skipping reaction intermediates to keep them tractable. While these methods may help reproduce experimental observations, their predictive capability is unclear. Moreover, with this approach, mechanistic insight is not easily transferable from one system to another.

The experiments performed by Korcek and co-workers^2-6 over 30 years ago remain the most detailed analysis of primary and secondary oxidation products resulting from the liquid phase oxidation of long chain alkanes. Through the use of complex analytical techniques, Korcek et al.quantified the yields of primary products like mono-, keto- and di-hydroperoxides. Through thermal decomposition experiments they also hypothesized that ketohydroperoxides were precursors to the formation of carboxylic acids and methyl ketones which are the main secondary products. No detailed kinetic models have been reported to quantitatively explain these experiments. Furthermore, the new pathways for carboxylic acids and methyl ketones that were hypothesized have remained largely ignored in the literature.

In this work, we combine computational chemistry and automated kinetic modeling tools to model the chemistry of liquid phase oxidation. We provide theoretical evidence for pathways leading from the ketohydroperoxide to carboxylic acids and carbonyl compounds via cyclic peroxide intermediates, confirming Korcek’s hypothesis. Next, we incorporate this and other relevant pathways in a detailed kinetic model built automatically using the Reaction Mechanism Generator (RMG) software⁷ to explain product yields observed by Korcek et al. The RMG approach significantly improves the efficiency of kinetic modeling for such large molecules but may require user-guidance in certain cases. Recent efforts at incorporating solvent effects into RMG will also be briefly discussed.⁸Finally, we demonstrate the application of these chemistry and kinetic modeling tools to understand the origin of oxidative deposits formed in diesel injectors using a multi-scale approach incorporating kinetics, transport and phase separation.

References

(1)        Denisov, E. T.; Afanas'ev, I. B. Oxidation and Antioxidants in Organic Chemistry and Biology; CRC Press, 2005.

(2)        Hamilton, E. J.; Korcek, S.; Mahoney, L. R.; Zinbo, M. Int J Chem Kinet 1980, 12, 577.

(3)        Jensen, R. K.; Korcek, S.; Mahoney, L. R.; Zinbo, M. J Am Chem Soc 1979, 101, 7574.

(4)        Jensen, R. K.; Korcek, S.; Mahoney, L. R.; Zinbo, M. J Am Chem Soc 1981, 103, 1742.

(5)        Jensen, R. K.; Korcek, S.; Zinbo, M.; Johnson, M. D. Int J Chem Kinet 1990, 22, 1095.

(6)        Jensen, R. K.; Zinbo, M.; Korcek, S. Journal of Chromatographic Science 1983, 21, 394.

(7)        William H. Green, Joshua W. Allen, Robert W. Ashcraft, Gregory J. Beran, Caleb A. Class, Connie Gao, C. Franklin Goldsmith, Michael R. Harper, Amrit Jalan, Gregory R. Magoon, David M. Matheu, Shamel S. Merchant, Jeffrey D. Mo, Sarah Petway, Sumathy Raman, Sandeep Sharma, Jing Song, Kevin M. Van Geem, John Wen, Richard H. West, Andrew Wong, Hsi-Wu Wong, Paul E. Yelvington, Joanna Yu; “RMG - Reaction Mechanism Generator v3.3”, 2011, http://rmg.sourceforge.net/

(8)        Jalan, A.; West, R. H.; Green, W. H. In AIChE Annual MeetingMinneapolis, MN, USA, 2011.