Multi-Scale Investigations in Soot Formation and Chemical Vapor Deposition

Open Access
Jain, Abhishek
Graduate Program:
Mechanical Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
November 29, 2018
Committee Members:
  • Yuan Xuan, Dissertation Advisor
  • Adrianus C Van Duin, Committee Chair
  • Daniel Connell Haworth, Committee Member
  • Richard A Yetter, Committee Member
  • Michael John Janik, Outside Member
  • Multi-scale
  • Soot formation
  • Chemical Vapor Deposition
  • simulation
  • nanomaterials
Progress is made in this thesis in understanding the complex multi-scale chemical and physical processes governing the formation of condensed phase material from gaseous species. The formation of soot through combustion and the synthesis of functional nanomaterial through chemical vapor deposition (CVD) are examined. We first attempt to characterize the sooting tendencies of alternative fuels using different techniques. A new numerical model based on modified flamelet equations is used along with a modified chemical mechanism to predict the effect of fuel molecular structure on soot yield in gasoline surrogates. These simulations provide trends on sooting behavior and are one-dimensional calculations that neglect other phenomenon that govern soot yield and distribution. To determine how other factors influence sooting behavior in laminar flames we carry out experimental and numerical studies to understand how the addition of oxygen to the oxidizer changes soot yield and distribution. Finite-rate chemistry based Direct Numerical Simulations (DNS) are carried out for a series of methane/air flames with increasing Oxygen Index (OI) using an extensively validated, semi-detailed chemical kinetic mechanism, along with an aggregate-based soot model and the results are compared with experimental measurements. It is seen that the effect of variable OI is well captured for major flame characteristics including flame heights, soot yield, and distribution by the numerical simulations when compared to the experimental data. This study is however confined to a small fuel that may not represent behavior seen in real fuels or the constituents that make up these gasoline fuels or their surrogates. Thus, we examine the effects of premixing on soot processes in an iso-octane coflow laminar flame at atmospheric pressure. Iso-octane is chosen as a higher molecular weight fuel as it is an important component of gasoline and its surrogates. Flames at different levels of premixing are investigated ranging from jet equivalence ratios of 1 (non-premixed), 24, 12, and 6. Numerical simulations are compared against experimental measurements and good agreement is seen in soot yield and soot spatial distributions with increasing levels of premixing. While the above studies for soot were carried out for laminar flames combustion devices frequently operate at conditions that lead to turbulent flow. Therefore, to understand how soot is affected by turbulence we computationally study the effects large Polycyclic Atromatic Hydrocarbons species (PAH) have on soot yield and distribution in turbulent non-premixed sooting jet flames using ethylene and and jet fuel surrogate (JP-8). The effects of large PAH on soot are highlighted by comparing the PAH profiles, soot nucleation rate, and soot volume fraction distributions obtained from both simulations for each test flame. Comparisons are also made with experiments when available and further analysis is performed to determine the cause of the observed behavior. Finally, a new multi-scale model is proposed for the computational modeling of the synthesis of functional nanomaterials using CVD. The proposed model is applied to a W(CO)6/H2Se system that has been used by researchers at Penn State to perform WSe2 crystal growth. A force-field for W/C/O/H/Se is developed and favorable agreement is seen when compared to QM data. A reaction mechanism leading from W(CO)6 and H2Se to the crystal precursor is then developed and used in a reacting flow simulation of the custom CVD chamber at Penn State. The bulk reacting flow numerical predictions show promising results for the gas-phase and precursor species, while additional work is still being performed to make the method more robust.