Predicting CO combustor performance

Detailed description of a CO to CO2 reaction kinetic model for estimating performance of an RFCCU CO combustor

Tek Sutikno
Fluor Corporation

Viewed : 4161

Article Summary

Predicting the performance of a CO combustor often becomes necessary when a residual fluid catalytic cracking unit (RFCCU) is revamped for capacity expansion or installation of a flue gas power recovery system. Relative to a conventional FCCU, which typically converts gas oil streams in a refinery to lighter products, mainly gasoline, an RFCCU is a processing option to catalytically crack heavier than gas oil feed such as atmospheric and vacuum bottoms to gasoline. An RFCCU may involve a two-stage catalyst regeneration system to control the combustion heat from the high carbon content of the feed stream. The first-stage regenerator reduces the coke content of the spent catalyst in a partial combustion environment, and catalysts from the first-stage regenerator enter the second-stage regenerator, where the catalyst coke content is further reduced through complete combustion.

The flue gas originating from the first-stage regenerator operating in partial combustion contains high concentrations (about 6 mol%) of carbon monoxide (CO), which is converted to carbon dioxide (CO2) in a downstream CO combustor. Fuel gas and air are fed to the combustor for raising the flue gas temperature to around 1800°F. This is the typical design target temperature for the reaction of CO with the excess oxygen to CO2 to reduce the flue gas CO content below 50 ppmv. Figure 1 shows a schematic of the RFCCU flue gas system.

As shown in Figure 1, flue gas from the CO combustor combines with the flue gas from the second-stage regenerator. The combined stream enters a flue gas cooler (FGC), generating steam typically at 600 psig, and discharges to atmosphere through a wet gas scrubber, removing mainly particulates and sulphur oxides (SOx).

High conversion of CO to CO2 in the CO combustor is essential to comply with the CO emission limit, recover the CO combustion heat for steam generation and prevent CO after-burning, which potentially leads to excessively high operating temp-eratures in the downstream equipment. As such, CO combustor vendors are generally required to guarantee the CO conversion performance of the combustor.

An RFCCU capacity expansion project applying the latest reactor and/or regenerator technology generally leads to higher flue gas rates, exceeding the original design flue gas flow rate of an existing CO combustor. As higher flue gas flow rates directly reduce the residence time of an existing CO combustor, the CO conversion level can be expected to decrease at higher RFCCU throughput capacities. In addition to RFCCU capacity expansion projects, a power recovery system (PRS) installation project can improve energy recovery of the flue gas system. A PRS includes an expander, recovering power from the first-stage regenerator flue gas stream and discharging to the CO combustor, and the flue gas temperature entering an existing CO combustor can be 300–400°F less than the original design inlet temperature. A lower flue gas inlet temperature will require higher fuel gas and air flow rates to reach the design target CO combustion temperature and the required conversion level. These higher air and fuel gas flow rates increase the required flow capacity of the existing CO combustor air blower and result in a reaction residence time reduction, which decreases the conversion performance of the existing CO combustor.

During the scope definition stage of an RFCCU capacity expansion or a PRS installation project, the unit engineer often needs to determine if the existing CO combustor system can still meet the CO reduction requirement of the flue gas stream at higher flow rates or lower temperatures at the combustor inlet. One obvious option for the unit engineer is to ask the original vendor to evaluate if the existing CO combustor can meet the required conversion level at the new process conditions. As performance guarantees typically require certain capacity margins to cover design uncertainties, the original vendor may propose modification or replacement of the CO combustor or decline to continue the conversion guarantees for the new process conditions. Modifying or replacing the existing CO combustor will significantly increase project capital cost along with turnaround complexity and potentially leads to new expansion scope for the downstream flue gas cooler. These additional modifications make the capacity expansion or PRS installation project less attractive financially.

Dependent upon the extent of capacity expansion, the original design capacity margin of the existing CO combustor may, in some cases, allow satisfactory operation at the new process conditions. These cases are sometimes evident when the operating performance of an existing CO combustor exceeds the guaranteed performance. Facing this situation, the unit engineer needs a performance prediction model to identify if any new scope is required for the existing CO combustor.

Reaction kinetic model
The reaction kinetic model (Equation 1) is based on a plug flow model, as the flue gas flow velocity through the CO combustor is generally in the turbulent region:

FAO dXA = R dV (1)

where
FAO = Molar flow rate of CO at the CO combustor inlet, lb mol/hr
XA = Fraction of CO converted to CO2 â€¨or XA = (CAO - CA ) / CAO
CAO = initial flue gas CO concentration at CO combustor inlet after mixing with the combustion flame products, lb mol/ft3
CA = CO concentration at axial direction (x) of the CO combustor, or CA = CAO (1 - XA)
R = CO reaction rate, lb mol/ft3/hr
dV = differential CO combustor volume or = πr 2 d x, and r, x - combustor radius and axial distance respectively, ft

The following Equation 2 shows the stoichiometric reaction of CO with oxygen to CO2:

CO + ½ O2 → CO2 (2)

While other reaction mechanisms or paths leading to the formation of CO2 have been suggested in the literature, Equation 2 reaction is the primary mechanism considered in reaction kinetic modelling studies, which give reasonable agreements with the experimental data. Moreover, other compounds such as water, which is present in significant amounts in the flue gas from the RFCCU first-stage regenerator (R1), have been shown to affect the reaction rate of CO to CO2. As this reaction involves a second order reaction between CO and O2, the reaction rate expression mathematically becomes more complex if water vapour is included.