1. New methods for data analysis of complex chemical systems with practical applications for atmospheric studies
- Author
-
de Jesus Medeiros, Diogo and Seakins, Paul W.
- Subjects
540 - Abstract
The hydroxyl radical, OH, is the most important oxidizing agent in the troposphere during daylight periods. This radical, which is predominantly produced from the photolysis of ozone initiates the gas-phase oxidation of the vast majority of volatile organic compounds emitted in the atmosphere. This thesis is focused on the development and exploitation of useful methods of analysis for the study of atmospherically-relevant processes via direct OH measurements. However, the flexibility of the technique makes it also suitable for the study of high temperature combustion-related chemistry. For example, OH recycling can be an important process for the high temperature oxidation of ethers, a class of oxygenated species often used as fuel additives. Such methods provide the necessary tools for the exploration of complex competing processes, stretching the limits of the conventional bimolecular analysis for measuring rate coefficients. Among them, a new method based on Master Equation calculations via global analysis allows the direct analysis of the temporal evolution of OH radicals undergoing temperature and pressure dependent processes. Such unprecedented analysis can not only provide mechanistic information, but also enables a robust evaluation of the thermochemistry of elementary reactions. The method is very useful in situations where a reaction of interest cannot be isolated from competing processes, as required by conventional analysis techniques. For example, for the high temperature oxidation of alkenes which involve both an OH addition and a hydrogen abstraction mechanism, the global technique is capable of discriminating and quantifying the contribution of each channel. The reaction of OH radicals with sulphur dioxide (SO2) was investigated via classical Master Equation analysis. The role of a weakly bound pre-reaction complex formation (~7.2 kJ mol^-1) was tested and a comparison of Leeds experimental data with the literature was undertaken. The results indicated that the pre-reaction complex formation is not significant under atmospheric conditions and much of literature data may have been influenced by secondary chemistry associated with SO2 photolysis. A transition state submerged below the reagents ( 1.0 kJ mol^-1) was required to describe the Leeds measurements, which appear to be more consistent than the rest of the literature. The analysis of high temperature equilibration data (OH + SO2 ⇄ HO-SO2), allowed the enthalpy of reaction to be determined (110.5 ± 6.6 kJ mol^-1). This experimental determination is in excellent agreement with the highest-level theoretical predictions found in the literature (~111.5 kJ mol^-1). The OH + isoprene reaction in the absence of oxygen was explored over a wide range of temperatures (298-794 K) and pressures (~60-1500 Torr). At high enough temperatures (T > 700 K), direct observations of the established isoprene + OH ⇄ isoprene-OH equilibrium were collected. The study also generated unprecedented rate coefficients which were subsequently employed for the study of OH recycling in the presence of O2. The equilibration data were exploited via a bi-exponential analysis of both experimental and Master Equation-simulated traces and used for the determination of the well-depth for OH addition to carbon C1 (153.5 ± 6.2 kJ mol^-1), which is in excellent agreement with our theoretically derived estimate (154.1 kJ mol^-1). This experimental value, however, is dependent on the level of theory at which the vibrational modes of the involved species are treated. The equilibration data also indicated a significant OH loss in the system, incompatible with the reaction of OH with its precursor or diffusion. This process was rationalized as a competing hydrogen abstraction, which interfered in the non-exponential equilibration traces. A global analysis of the data was capable of extracting information about both the OH addition (k_addition_∞(T) = (9.5 ± 1.2) × (10^-11) × (T/298 K)^(-1.33 ± 0.32) cm^3 molecule^-1 s^-1) and the abstraction channel, (k_abstraction_∞(T) = (1.3 ± 0.3) × (10^-11) × exp((-3.61 kJ mol^-1)/RT) cm^3 molecule^-1 s^-1). With respect to the OH addition, a comparison with previous investigations suggests that only our measurements at T > 700 K were in the fall-off region, contradicting some literature studies. A new method of analysis via a global multi-temperature, multi-pressure fitting procedure was developed and used for the study of the ethylene + OH reaction. The method relied on the Master Equation modelling of the OH addition, and a subsequent incorporation of new consumption and formation terms to the rate laws of the involved species. With effect, the simulated traces become comparable to experimental observations and a direct trace analysis is possible. The reaction of OH with ethylene was studied over a range of temperatures (563 - 723 K) and pressures (~60 220 Torr), which included pressure dependent data, to test the limits of this global direct trace analysis. Excellent descriptions of OH traces were obtained when the Master Equations were modified to incorporate a hydrogen abstraction and a unimolecular loss of the adduct. A simultaneous fit of 96 traces where direct ethylene + OH ⇄ adduct equilibration was observed enabled the determination of the well-depth of the adduct (111.8 ± 0.20 kJ mol^-1). This value is in excellent agreement with our theoretical prediction (111.4 kJ mol^-1), calculated at the CCSD(T)/CBS//M06-2X/ aug cc pVTZ level of theory. The high pressure limiting rate coefficient for the OH addition extracted from the experimental traces by this technique (k_addition_∞(T)=(8.13 ± 0.86) × (10^-12) × (T/298 K)^(-0.99 ± 0.18) cm^3 molecule^-1 s^-1), is in very good agreement with the IUPAC recommendation for the 100 - 500 K temperature range (k_1a_∞(T)= 9 × (10^-12) × (T/300 K)^-0.85 cm^3 molecule^-1 s^-1). Furthermore, the experimentally derived abstraction rate coefficients k_abstraction(T) = (3.5 ± 0.75) × (10^-11) × exp((-26.2 ± 1.3 kJ mol^-1)/RT) cm^3 molecule^-1 s^-1 are in excellent agreement with previous investigations. The method proved to be robust enough to discriminate and quantify the competing processes influencing the shapes of the experimental OH profiles. This novel analysis was employed for the study of LIM1, a promising mechanism for the description of OH recycling via isoprene peroxy chemistry. Experiments were undertaken at elevated temperatures (400 < T < 600 K) and large concentrations of O2 were employed ([O2]~10^17 molecules cm^-3) so as to promote the recycling to the millisecond timescale, compatible with direct experimental techniques for OH detection. Rate coefficients measured in the absence of O2 were incorporated to help constrain the analysis of the data. The LIM1 mechanism proved to be capable of accurately describing non-exponential traces collected under such experimental conditions. The OH recycling involves crucial hydrogen shifts, whose barriers were adjusted in unison to provide a good fit to the data. However, very small modifications were necessary for this purpose (~ 2 kJ mol^-1). The incorporation of these findings in an atmospheric model enabled the description of approximately 50% of the OH concentration measured in a field campaign performed in a Borneo rainforest, an environment dominated by biogenic volatile organic compounds emissions. Finally, the thesis concludes in Chapter 7 with a summary of the results of each experimental chapter including suggestions for further areas of study.
- Published
- 2019