4.4.2 LIF OH Detection using Bidirectional Fiber Probe in Harsh Environment
In Section 4.2.1 conventional measurements of the OH concentration and temperature profiles in a stoichiometric methane-air flame were outlined. OH LIF was also per-formed using the fiber probe technique described above by traversing the fiber endface perpendicular to the flames’s symmetry axis. Figure 4.6 shows the OH LIF spectra (A2Σ- X2Π1 ←1 and 0 ←1 band) for the two LIF arrangements, as well as the non-resonant fiber scattering background (dashed lines). Three conclusions may be drawn: first, the OH LIF spectrum is unaffected by the presence of the fiber probe, meaning that quenching effects are not observed in the spectrum; second, the inelastic fiber scattering, which was the same when the fiber was inside or outside the flame, is exceeded by the OH LIF signal; third, one finds a comparable LIF signal intensity - the conventional LIF spectrum being scaled by factor of 1/2 in the figure - for both optical arrangements as predicted by the above collection efficiency considerations.
The OH concentration gradient was scanned using the fiber probe, the outcome of which is included in Figure 4.2. The OH profile is resolved and in good agreement with the reference experiment. The relatively large error bars of the LIF measurements shown for the fiber in Figure 4.2 (conventional measurement error bars of similar order but omitted for clarity) are due to flame fluctuations and shot-to-shot fluctuations of the laser which could not be corrected for by the averaging power-meter used.
At first glance, it might be surprising that the fiber withstands the excessive flame temperatures. This may be explained by assuming an equilibrium between convective heat transferQ˙conv = α A(T −Tgas) from the hot product gases to the fiber, and heat losses due to conductionQ˙condand radiative coolingQ˙rad=ε σ A(T4−T04)(σ is the Stefan-Boltzmann constant). An emissivity of 0.9 has been assumed for fused silica, and the heat transfer coefficientαhas been evaluated from a Nusselt number correlation of the flow field (cylinder in cross flow). It turns out that conduction losses are negligible and an equilibrium temperature ofT ∼1400 K is found. Nevertheless the physical intrusion of the fiber close to the reaction zone disturbs the flow field and thus the flame cone which causes slightly biased results at low heights.
4.5 Conclusion
Detection of OH LIF has been demonstrated using a bidirectional fiber probe technique with the potential for minimally-invasive measurement under harsh conditions. Refer-ence experiments have been performed (OH-LIF and Raman) in order to characterize a stoichiometric methane-air flame serving as a reference experiment. An optical setup has been presented which permits simultaneous coupling into the fiber and detection of the backscattered fluorescence signal. The detection efficiency and the dimensions of the sample volume have been characterized numerically using geometrical consid-erations. All-silica optical fibers may be found which are suitable for simultaneous UV-radiation transmission and detection of fluorescence, which is red-shifted beyond the vibrational Raman signature of the fiber (i.e.>1000 cm−1). OH LIF was detected and the OH concentration profile in a laminar premixed flame could be reproduced.
310 315 320 325 0
100 200 300 400
Wavelength (nm)
Fiber LIF Counts
310 315 320 325 −100
0 100 200 300
Conv. LIF Counts / 2
Fiber LIF (0.7 mJ) Conv. LIF (1.1 mJ)
Figure 4.6: OH LIF spectrum of fiber probe and conventional lens-type optical setup LIF spectrum (scaled by a factor of 1/2 and shifted for clarity). The dashed line is the background including chemiluminescence and fiber scattering. For fiber probe LIF
measurements a low-OH fiber was used.
Chapter 5
Fuel-Rich Methane Oxidation in a High-Pressure Flow Reactor studied by Optical-Fiber
Laser-Induced Fluorescence, Multi-Species Sampling Profile Measurements and Microkinetic Simulations
Abstract
A versatile flow-reactor design is presented that permits multi-species profile measure-ments under industrially relevant temperatures and pressures. The reactor combines a capillary sampling technique with a novel fiber-optic Laser-Induced Fluorescence (LIF) method. The gas sampling provides quantitative analysis of stable species by means of gas chromatography (i.e. CH4, O2, CO, CO2, H2O, H2, C2H6, C2H4), and the fiber-optic probe enables in situ detection of transient LIF-active species, demon-strated here for CH2O. A thorough analysis of the LIF correction terms for the temperature-dependent Boltzmann fraction and collisional quenching are presented. The laminar flow reactor is modeled by solving the two-dimensional Navier-Stokes equations in conjunction with a detailed kinetic mechanism. Experimental and simulated profiles are compared. The experimental profiles provide much needed data for the continued
This chapter is adapted from the author’s version of a manuscript by Heiner Schwarz, Michael Geske, C. Franklin Goldsmith, Robert Schlögl, Raimund Horn, which was accepted for publication in Combustion and Flame, Elsevier (http://www.journals.elsevier.com/combustion-and-flame/). Changes resulting from the publishing process may not be reflected in this document. A definitive version will be subsequently published in the journal.
validation of the kinetic mechanism with respect to C1and C2chemistry; additionally, the results provide mechanistic insight into the reaction network of fuel-rich gas-phase methane oxidation, thus allowing optimization of the industrial process.
5.1 Introduction
Owing to the predicted depletion of petroleum reserves, the transformation of natural gas (i.e. methane) into value-added chemical products is of growing interest for the chemical industry. Conventional approaches rely on indirect conversion via synthesis gas production (from steam reforming, CO2reforming or partial oxidation), followed by a gas-to-liquid process, but these multi-step processes are particularly capital inten-sive. Therefore, the direct conversion of methane to ethylene, methanol or formalde-hyde is economically more favorable [87, 11].
Oxidative Coupling of Methane (OCM) could be a desirable direct conversion route in which methane is transformed into ethylene under fuel-rich conditions (CH4/O2= 2 - 8, or an equivalence ratioφ= 4 - 16) at temperatures around 1000 K and pressures up to 30 bar. It has been suggested in the literature that the OCM reaction proceeds via a homogeneous/heterogeneous coupled mechanism [88, 22]. According to this model, methane is first activated on the catalyst, and the resulting methyl radical des-orbs. Two gas-phase methyl radicals combine to form ethane, which is subsequently dehydrogenated to ethylene. In fact, OCM can occur even without a catalysts, albeit with very low selectivity [89, 90, 23, 91]. Although the exact role of oxygen in the homogeneous/heterogeneous mechanism is unclear, it is known that small concentra-tions of oxygen are necessary for OCM. If the concentration of oxygen is too high, however, the C2 products will be oxidized, thereby decreasing the yield. Computa-tional engineering will play a key role in the optimization of the reactor design, catalyst choice, and operating conditions. An essential component of this approach includes detailed models that describe the coupling between fluid mechanics and the kinetics of elementary surface and gas phase reactions. The predictive utility of these models depends upon the accuracy of the underlying rate coefficients for the elementary re-actions. These kinetic1mechanisms are often tested against experimental data taken under low-pressure and/or highly dilute conditions. A more desirable approach would be to validate the mechanisms against data taken under industrially relevant conditions, since it requires less extrapolation, but this approach can work only if the flow field and chemistry can be modeled simultaneously in a rigorous yet computationally efficient manner. This manuscript presents an experimental apparatus designed precisely for this purpose and the accompanying kinetic simulations.
Gas-phase Oxidative Coupling of Methane was studied in a novel, versatile flow reactor designed for spatially resolved kinetic profile measurements under homoge-neous (and/or catalytic) conditions, with temperatures up to 1300 K and pressures up to 45 bar [92]. The reactor features a sampling capillary through which a small frac-tion of the reacting gas mixture is transferred to quantitative gas analytics, e.g. a mass spectrometer (MS) or a gas chromatograph (GC). Complementarily, a recently devel-oped fiber-optic Laser-Induced Fluorescence (LIF) method [93] was applied for in situ detection of CH2O, which is an important intermediate in the oxidation process. It is worth emphasizing that the reactor does not require optical viewports; optical access
1Mechanisms of elementary rate constants are commonly referred to as “microkinetic” in the catalysis community, whereas the preferred term in gas-phase chemistry is “detailed kinetics” or “elementary kinet-ics”.