Apr-2008

Reducing FCC SOx in 
partial-burn regenerators

Latest developments in SOx-reduction additives are discussed along with their relative merits and appropriate applications

Martin Evans
Intercat (Johnson Matthey)

Article Summary

Refinery emissions of pollutants such as SOx, NOx, CO and particulate matter are coming under increased scrutiny in all parts of the world. In most refineries, the FCC unit is the major source of these airborne pollutants and is thus the primary target for emissions reductions. Of these pollutants, the emissions of SOx (SO2 and SO3) are of particular concern because they are known to be a major contributor to acid rain and, as they promote the formation of PM10 particulates, they are also a cause of respiratory health problems in major cities and surrounding areas.
There are several proven technologies available to reduce SOx emissions from the FCC process. However, the use of catalyst additives has proven to be preferred over hardware technologies, as it has been repeatedly shown in commercial trials that additives provide a more cost-effective means of controlling SOx emissions.

SOx additives used to be treated as a “one additive fits all” type of product. Indeed, some suppliers of additives still provide only a single SOx-reduction additive for all applications. However, as the understanding of SOx chemistry has evolved, several new additives have been developed for specific SOx-reduction applications. In particular, a new SOx reduction additive, Intercat’s Lo-SOx PB, has recently been commercialised for use in partial-burn applications. This additive has had several successful trials in commercial FCCUs with both single- and two-stage partial-burn regenerators, succeeding where other additives have failed to perform.

SOx-reduction additive developments
The first generation of SOx-reduction additives began commercial use in the mid-1980s. Research and development work conducted at Arco and then at Katalistiks resulted in the commercialis-ation of DeSOx, a separate particle additive based on the MgAl2O4 spinel structure. This additive was effective in reducing SOx emissions from FCC flue gas in units that operated in a full combustion mode. However, the efficiency of this additive was too low to enable most refiners to meet the ultra-low SOx emission levels that the best available hardware-control technologies were able to achieve.

Additional research work showed that the efficiency of SOx reduction additives was a function of a number of variables; primary among them was the MgO content. Unfortunately, spinel-based additives have a limited capability for incorporating additional MgO into the stabilised structure. Calcination of DeSOx to remove the water of hydration led to the development of Super DeSOx and resulted in a slight increase in the MgO content, and thus an improvement in additive efficiency. However, the performance of this additive was still limited by the capacity of the spinel structure to incorporate additional MgO.

At the same time, Intercat was developing a new hydrothermally 
stable hydrotalcite-based SOx-reduction additive. The significance of the hydrotalcite structure is that it allows for the incorporation of far more MgO into the additive’s structure, which results in significantly higher SOx-reduction activity and efficiency. Intercat’s SOxGetter was the result of this work and gained quick acceptance in the industry as a result of its superior efficiency relative to DeSOx.

Further refinements to the controlled production of the hydrotalcite material yielded the first second-generation SOx-reduction additive, Super SOxGetter.  It is the highest MgO content additive on the market and is also the most active SOx-reduction additive available. Tables 1 and 2 give some typical properties and commercial unit efficiencies for these additives.

As Table 2 shows, Super SOxGetter with its high MgO content always outperformed the lower MgO content additives. However, MgO content alone is not the only variable that determines additive performance. Other variables such as the concentration of oxidants, surface area, pore size distribution and accessibility can all have an impact.

As SOx additive use really began to increase in the late 1990s due to increased governmental regulatory limits on SOx emissions, it became apparent that there were large differences in SOx additive performance from unit to unit. By investigating these differences, it became possible to modify the SOx additive formulations to take advantage of some of these differences. As a result, the first of the third- generation SOx-reduction additives, Ultra Lo-SOx, was developed. This third-generation SOx-reduction additive was introduced to provide an alternative for those refiners who needed a faster response to keep SOx emissions below an imposed limit. It has proven itself to be especially effective for those refiners involved in peak shaving operations, where the additive use is initiated or increased during specific periods of high sulphur feed operations.

What makes Ultra Lo-SOx unique is that it incorporates the same high MgO content as Super SOxGetter, yet does not have a conventional hydrotalcite or spinel structure. The unique mixed metal oxide structure, along with its higher pore volume and surface area, gives the additive its higher initial activity. This bias towards improved front-end pickup efficiency enables a faster response to SOx emissions excursions. When compared in back-to-back commercial trials, the overall efficiency of Ultra Lo-SOx is, however, similar to Super SOxGetter, as shown in Table 3. Both of these additives far surpass the performance of Super DeSOx.

Additive development 
and optimisation for 
partial-burn units
The focus on SOx additive development has recently been on improving performance in partial-burn FCCUs. It has been apparent for some time that oxygen availability is the factor limiting SOx-reduction additive performance in partial-burn units. Oxygen availability is critical for the SOx additive to oxidise SO2 to SO3 in order to capture the SO3 generated in the regenerator, and release it as H2S in the riser and reactor. In partial-burn units, it is well known that O2 availability is extremely limited. Recent detailed flue gas analyses have shown that in deep partial-burn regenerators only about 30% of the sulphur species are present in an oxidised form (Figure 1). The majority of the sulphur species are in reduced forms such as CS2 
and H2S.

This discovery has led to increased research into the oxidation function of these additives. This has resulted in the development and commercialisation of the first SOx-reduction additive specifically designed for partial-burn units.