60th Annual Report on Research 2015

Carbon particles are formed in combustion environments by complex radical chemical schemes in the gas phase and at the gas–solid interface. Further reactions of gaseous species at the particle surface is likely to be at the origin of the catalytic effect observed during the pyrolysis of hydrocarbons in the presence of carbon particles. During the past funding period we have investigated elementary reactions involved in the formation and reaction of carbon particles in combustion environments. The OH + fulvenallene gas phase reaction was investigated over the 300–450 K temperature range. Over this temperature range, the most likely reaction pathway is the addition of the radical onto the unsaturated molecules. The kinetic results suggest that at higher temperatures (>600 K) the reaction will likely proceed by abstraction to form the resonance-stabilized fulvenallenyl radical, a precursor to polycyclic aromatic hydrocarbons and soot in combustion. A new pump-probe technique was developed in order to investigate radical heterogeneous chemistry at the surface of solid carbon particles. Data have been recorded to demonstrate the feasibility of the technique as well as to study the effect of the presence of particles on the decay rate of OH radicals.

Experimental

Figure 1 Experimental setup for the investigation of heterogeneous pyrolysis using IR laser-heating coupled to laser-induced fluorescence.

The custom-made experimental cell is a 7-way stainless steel cross with 5 optical access ports (Figure 1). Carbon particles are generated by nebulizing a water–particle suspension using a constant output atomizer. The resulting aerosol is dried using a high-flow-rate diffusion dryer. The particle number density and diameter distribution at the cell entrance as well as after the reaction are monitored using a scanning mobility particle sizer (see figure 2). The optics inside the main vacuum chamber are protected from particle deposition using counter-flow aerodynamic windows.

Figure 2 Particle diameter distribution obtained by nebulization of a suspension of 100-nm particle in water using a constant output atomizer.

The aerosol is then carried to the center of the flow cell by a laminar flow. A flow of pure helium concentric to the particle flow minimizes the diffusion of the particles to the optics. The particle flow is crossed by the fundamental output of a Nd:YAG laser with a 8-ns pulse width. The absorption of the laser light by the particles leads to an increase of the particle surface temperature within the time scale of the laser pulse.

Figure 3 displays the laser induced incandescence signal obtained after irradiation of the aerosol by a 1064-nm laser. The black line is the particle surface temperature obtained by fitting the Planck function to the LII signals at 600 nm (blue line) and 800 nm (red line). The laser fluence is set to avoid sublimation of the particle while reaching temperature higher than 1000 K.

Figure 3 Laser induced incandescence at 600 nm (blue line) and 800 nm (red line) together with the inferred particle surface temperature (black line) for 1064-nm laser heated suspended carbon particles.

Heterogeneous reactions of the OH radical

Experiments have been performed to measure the OH decay rate in the presence of suspended particles. The OH radicals are detected by laser-induced fluorescence at pressures ranging from 100 Torr to atmospheric pressure. The decay rate of the radicals is found to be much faster in the presence of particles. This suggests a fast reaction of the gas phase radical at the particle surface. Further experiments will look at the formation and decay of radicals formed at the particle surface. In this case, the gas phase products are detected by Laser Induced Fluorescence (LIF) using the fundamental or second harmonic of a tunable Nd:YAG-pumped dye laser.

Gas phase kinetics of soot precursor formation

The pump-probe laser system for measuring OH radical temporal behavior has been used to measure the bimolecular rate coefficient for the OH + fulvenallene reaction. In this case the OH radicals are formed by 266-nm photodissociation of H₂O₂ in the presence of gaseous fulvenallene. The concentration of the OH radicals is followed in time by exciting the OH radical on the A²S-X²P (1,0) band at 281.954 nm. The following fluorescence from the A²S-X²P (1,1) band is collected through a 310 nm (FWHM ±10nm) band pass filter by a photomultiplier tube.

The fulvenallene reactant was synthetized by flash photolysis of homophtalic anhydride. Juddha thapa, a 4^th year graduate student designed and assembled the pyrolysis set up. He also performed the synthesis and characterized the products using liquid phase NMR and gas phase infrared spectroscopy.

Figure 4 displays the second order reaction rate for the OH + fulvenallene reaction from 300 to 400 K. The data were recorded and analyzed by Juddha Thapa (graduate student) and Michael Spencer (senior undergraduate student).

Figure 4 Temperature dependence of the OH + fulvenallene reaction rate coefficient at 5 Torr.

Together with high-level calculations performed by Michael Spencer (senior undergraduate student), the kinetic data suggest that the most likely reaction mechanism is the association of the radical to the p-electron system of the molecule to form a Van der Waals intermediate. This intermediate may isomerize to the association products or dissociate back to the reactants. The negative temperature dependence of the reaction rate coefficient suggests that the initially formed complex becomes less stable as the temperature increases, thus re-forming the OH radical. At higher temperatures, it is likely that the association mechanism will become less important than the association mechanism. Above 600 K the most likely product is the fulvenallenyl radical formed by loss of a water molecule. The fulvenallenyl radical is the most stable C₇H₅ radical and a precursor to polycyclic aromatic hydrocarbons in combustion flames.