Spatio-Temporal Concentration and Temperature Profiles During Periodic Hydrocarbon Oxidation in a Catalytic Monolith

Spatio-temporal concentration and temperature profiles during periodic hydrocarbon oxidation in a catalytic monolith

Hoang Nguyen, Michael P. Harold^*, and Dan Luss^*

 Department of Chemical and Biomolecular Engineering, University of Houston, Houston, TX 77204-4004 United States.

(^*Corresponding authors:  mharold@uh.edu, dluss@uh.edu)

Introduction

Ceria containing catalysts have been successfully used in automotive emission control devices for many years. More recently, new applications of ceria include sulfur trapping, water gas shift conversion, and in hydrocarbon oxidation. In all of these applications, the key is the ability of cerium to cycle between oxidized (Ce⁴⁺) and reduced (Ce³⁺) states. The redox features of ceria are strongly dependent on the operating conditions, understanding ceria oxidation and reduction during reaction is needed in the rational design and improvement of the catalyst and reactor. While most previous studies drew conclusions about the redox features of ceria by monitoring the effluent gas composition, this study simultaneously measures the spatio-temporal temperature and concentration within a monolith reactor. 

In the current study, we measure the spatial-temporal temperature and concentration profiles during a periodic feed of either propylene and/or ethane  (rich phase) and O₂ (lean phase) to a Pt/CeO₂/Al₂O₃ washcoated monolith. Our objective is to study the influence of various operating conditions on the redox properties and how these, in turn, affect the overall reaction conversion and selectivity. We utilize two experimental techniques that allow sampling within the reactor with negligible flow interference (refer to Fig. 1); (i) Spatially resolved mass spectrometry (SpaciMS) which uses a 0.363 mm quartz capillary tube and a quadrupole mass spectrometer. By moving a capillary tip at different axial positions, spatio-temporal concentration can be obtained with 0.3 mm spatial resolution and time resolution of ~ 0.05s. This technique enables us to identify and quantify surface adsorbing and desorbing species. These local concentration measurements provide insight about the underlying chemistry of the adsorption/desorption events, (ii) Coherent Optical Frequency Domain Reflectometry (C-OFDR), using a 0.125 mm single mode optical fiber and an Optical Backscatter Reflectometry, enables a single measurement of the entire axial temperature profile with 0.3 mm spatial resolution and ~1 s time resolution. These spatially-resolved measurements provide detailed data of the spatio-temporal features of the monolith catalyst which is essential for fundamental understanding of the kinetics and model development. Figure 1 is a schematic of the flow reactor system.

Fig 1: Experimental setup for spatially resolved temperature and concentration measurements

We examine the redox activity of the monolith catalyst during cyclic operation; i.e. periodic lean/rich switching between a trapping phase (lean phase) for 60 s with 10 % O₂, and regeneration phase (rich phase) for 10 s with 1.2% C₃H₆.  The spatio-temporal temperature profile at the periodic state is complex (Fig. 2). During the fuel rich phase, C₃H₆ is fed into the reactor where it reacts with oxygen that has accumulated in the reactor. Light off occurs at the upstream of the monolith channels and generates a thermal front. For a typical operation (L = 60 s, R = 10 s) the thermal front propagation velocity is approximately one third of the oxygen adsorption front velocity. During the 10 second rich pulse, the hot spot travels only for a fraction of the reactor length. On the other hand, during the 60 s lean pulse a downstream movement of the upstream temperature front occurs. The experiments show that even under anaerobic condition, sufficient O₂ is stored on the ceria based catalyst so that the reaction  results in a large temperature increase (~60°C ).  Furthermore, the temperature gradient still evolves for many seconds after the main exothermic reaction has been completed. This transient heat generation affects the O₂ trapping efficiency, resulting in multiple temperature maxima at different times. Following the feed switching, two hot spots immediately form.  The feed maximum temperature rise following a rich/lean transition is much larger than that following a lean/rich transition. Similar behavior has been reported during lean/rich cycling of O₂ and H₂ /C₃H₆ [1,2].

We also investigate the redox efficiency of the catalyst, as measured by the conversion or temperature profile.  In these experiments the cycle duty  (the ratio between the lean to the total cycle time) was held constant but the total cycle time was decreased  five-fold. Decreasing the total cycle time leads to a rapid regeneration and increased utilization of O₂ physisorbed on the ceria surface.  During the short cycle time (L = 12s, R = 2s), the 12 s lean pulse only causes a middle-stream  movement of an upstream temperature front. The hot spot develops at the middle of the channel; consequently, the measured periodic spatio-temporal temperature profile using the short cycle time is significantly higher than that at the long cycle time (Fig. 3). Although a high temperature increases the oxidation rate and reduces fuel slip out of the catalyst, it affects the stability of intermediate species, reduces the trapping efficiency, and alternates the product selectivity. This may explain Kabin's finding that a short cycle time results in poor average NO_x cyclic conversion [3]. These and other experiments provide considerable detail about the redox behavior of the catalyst.

We are currently studying the co-oxidation of ethane and propylene on the cerium based catalytic  monolith. The experiments are conducted to determine the influence of the ratio of propylene to ethane of the binary mixture on the ignition-extinction behavior.

Fig 2: The periodic spatio-temporal temperature profile under the cyclic experiment (Lean = 60s, 10% O₂, Ar balance), (Rich = 10s, 1.2 % C₃H₆, Ar balance).

Fig 3: Periodic temperature profiles at the beginning of the lean phases (t=0) for two cycle times (L=60s, R = 10s) & (L=12s, R =2s).

References:

1.J.S. Choi, W.P. Partridge, C.S. Daw, Sulfur impact on NOx storage, oxygen storage, and ammonia breakthrough during cyclic lean/rich operation of a commercial lean NOx trap, Appl Catal B-Environ, 77 (2007) 145-156.

2.W.S. Epling, A. Yezerets, N.W. Currier, The effect of exothermic reactions during regeneration on the NOX trapping efficiency of a NOX storage/reduction catalyst, Catal Lett, 110 (2006) 143-148.

3.K.S. Kabin, R.L. Muncrief, M.P. Harold, Y.J. Li, Dynamics of storage and reaction in a monolith reactor: lean NOx reduction, Chem Eng Sci, 59 (2004) 5319-5327.