Apr-2014

HCN and NOx control strategies in the FCC

Results of research into emissions from the FCC, plus guidance on how to obtain the minimum level of NOx.

XUNHUA MO, BART DE GRAAF, CHARLES RADCLIFFE and PAUL DIDDAMS
Johnson Matthey, Process Technologies, IntercatJM Additives

Article Summary

Increasing environmental awareness at a global level is driving enforcement of more stringent regulations to limit emissions of sulphur and nitrogen containing gases to the atmosphere.

NOx (mixed nitrogen oxides) and SOx (mixed sulphur oxides) emissions have been under scrutiny for some time now, especially in North America and Europe. More recently, however, hydrogen cyanide (HCN) emissions (present in FCC flue gas at levels up to about 150ppm) have come under scrutiny in the US and some parts of Europe. Controlling flue gas emissions of this new species is an emerging challenge for refiners operating FCC units.

The origin of HCN in FCC flue gas has been poorly understood and public domain information very limited. In this article we explain our recent research on nitrogen chemistry as it relates to the FCC regenerator, covering the origin of HCN and its relationship with other nitrogen containing gases. We also explain how modern FCC unit designs geared for low NOx emissions may unintentionally have been trading these off against HCN and ammonia emissions. Continuing development of catalytic additives for FCC regenerator flue gas emissions control will help refiners to meet the emerging regulatory emission limits. We explain how to use a combination of operating variables, additives and regenerator design to reduce emissions.

FCC NOx and SOx emissions
The FCC unit is a major conversion unit present in many refineries throughout the world. FCC units are highly flexible and able to upgrade feeds comprising many components ranging from light-sweet hydrotreated VGO to heavy-sour residues, and may include additional heavy streams from other refinery units, such as coker gas oils. Because of the resilience of the process and catalyst system, the majority of FCC feedstocks contain many contaminants including metals (Ni, V, Cu, Fe, Ca, Na, K) and heteroatoms (sulphur, S and nitrogen, N).

The FCC converts low value, high molecular weight (high boiling point) hydrocarbon feeds to lighter, more valuable products via cleavage of C-C bonds (cracking). Typically, 5-6wt% of the feed is converted to coke as a byproduct 
of cracking. This is burned in 
the FCC regenerator where the 
heat of combustion is used to provide the energy required to vaporise and crack the feed (hence, the FCC unit is said to be in ‘heat balance’).

Metals accumulate on the catalyst where they deactivate/poison the catalyst or cause undesired side reactions such as dehydrogenation and additional coke and gas formation. Some contaminants in the feed are transferred to the regenerator in the coke. In this way, a portion of the sulphur and nitrogen in the feed is combusted in the regenerator.

The products of combustion of sulphur and nitrogen in the FCC regenerator include the gases SO2, SO3, COS, H2S, N2, NO, N2O, NO2, NH3 and HCN, all which may contribute to stack emissions at various concentrations in the FCC flue gas. The exact composition of these gases in the flue gas depends upon the detailed reaction conditions. A simplified reaction pathway for nitrogen compounds is shown in Figure 1.

In full burn FCC units (where coke is combusted in an excess of oxygen) the main species formed are the most highly oxidised species: SO2, SO3, N2 and NO. However, in partial burn FCC units (where the coke is combusted under sub-stoichiometric oxygen conditions) much higher levels of the ‘reduced’ S and N species (COS, H2S, NH3, HCN) are present in the flue gas as it leaves the regenerator. In partial burn units a controlled amount of the carbon in the coke burned is combusted to CO rather than CO2 in order to decrease the heat produced in the regenerator via coke combustion. This allows the processing of heavier, higher coke making feeds within regenerator temperature constraints. Most partial burn units have a CO boiler downstream of the regenerator in which CO is converted to CO2 to control CO emissions and recover additional heat of combustion for steam production. The CO boiler also converts reduced S and N species to more highly oxidised forms, so the result in both full burn and partial burn with a CO boiler is that the flue gas contaminants reaching the stack are predominantly SOx (SO2, SO3) and NOx (NO, N2O, NO2). Other species are generally only present at much lower concentrations. Note that a partial burn unit without a CO boiler will emit substantial levels of the CO and reduced S and N gaseous species – and there are actually still a number of such FCC units being operated in this way.

Nitrogen chemistry in the FCC regenerator
FCC feedstocks contain many different nitrogen compounds, generally measured as total and basic nitrogen. Normally, 30-50% of the feed nitrogen compounds are basic nitrogen species that strongly adsorb on acid sites on the catalyst. This tightly bound nitrogen is not removed in the stripper and therefore carried with the catalyst into the regenerator where it is combusted as part of the coke together with the carbon, hydrogen and sulphur. As a rule of thumb, about half of the feed total nitrogen is combusted in the FCC regenerator. See Table 1 for a typical FCC nitrogen balance.

Coke composition primarily depends on feed properties and stripper efficiency. Coke is comprised of carbon-rich polycyclic aromatic structures containing heteroatoms and contaminant metals as well as unstripped hydrocarbon products (10-30% of coke can be gasoline, diesel and fuel oil range products that are not stripped from the pores of the catalyst). Typically, coke has a hydrogen content of 5-8 wt%, largely present in the unstripped products. The concentration of nitrogen in coke is an order of magnitude higher than in the feed (about 50% of feed nitrogen goes to coke compared to about 5% of feed carbon).

Combustion of carbonaceous coke and unstripped hydrocarbons in the regenerator forms CO2, CO and H2O. Sulphur in coke forms SO2, SO3, COS and H2S, but nitrogen in coke behaves very differently. When oxygen reacts with carbon-rich coke, much of the nitrogen is initially converted to HCN; the same chemistry is observed in coal combustion. At typical steady-state FCC regenerator bed temperatures (680-755°C, 1255-1390°F), HCN is thermodynamically unstable and, given sufficient time, all of the HCN would be converted. Nitrogen followed by NO are the most thermodynamically stable nitrogen species under FCC regenerator conditions, with the thermodynamic equilibrium concentration of NO being about 10 ppm in nitrogen. However, the much higher levels of HCN found in commercial FCC unit flue gases clearly illustrates that nitrogen reactions are kinetically controlled and do not reach thermodynamic equilibrium. HCN present as a reactive intermediate can be hydrolysed by steam in the regenerator to form NH3. Both HCN and NH3 can be readily oxidised to form N2 or NO, depending on regenerator conditions and the presence of combustion promoters which 
catalyse these reactions. Previously, the reaction of CO + NO was thought to be one of the main driving forces for reducing NO. Our work shows this reaction to play a very minor role under FCC conditions.