Jun-2016

The role of tail gas treating unit quench towers

A quench tower cools gas entering a tail gas amine unit but also protects the amine against SO2 breakthrough.

RALPH WEILAND, CLAYTON JONES and NATHAN HATCHER
Optimized Gas Treating, Inc

Article Summary

Tail gas from most sulphur recovery units (SRU) is routed to a reduction quench amine tail gas treating unit (TGTU). A hydrogenation reactor first converts (reduces) residual sulphur species such as SO2, COS, CS2, and elemental sulphur into H2S for eventual recycling back to the SRU. Before its sulphur content is recovered, however, the hot gas from the hydrogenation reactor is quenched to make it cool enough for feeding to the TGTU amine system. Here, the hydrogen sulphide content is captured by a H2S selective solvent, usually N-methyldiethanolamine (MDEA). This article focuses on the quench tower, which tends to receive little or no engineering attention, despite its very important role in the overall sulphur recovery process.

The primary purpose of the quench tower is to cool the hot gas (500–600°F, 260-315°C) from the hydrogenation reactor to around 100°F (38°C) by direct contact with cooling water. In addition to lowering the temperature, generally about 85% of the water content is removed as well. This water would otherwise need to be purged from the amine system downstream to maintain amine strength. A sometimes poorly appreciated secondary role of the quench tower is to afford some measure of protection of the TGTU from harmful contaminants that would otherwise enter with the quenched tail gas. In particular, any small amounts of ammonia and SO2 in the gas can be removed in the quench tower. Unless the amine is protected, over a period of time, even small levels of SO2 contamination can generate heat stable salts and other more reactive amine degradation products of the MDEA solvent, harming the selectivity for hydrogen sulphide.

Trays are rarely used any more in quench towers because of the fast volume decrease that accompanies the rapid drop in temperature, which has a tendency to buckle trays. Today, quench towers are almost always packed, and since high efficiency is not needed to achieve adequate cooling, large diameter random packings are quite suitable. As will be seen later, although cooling itself occurs very rapidly, water removal requires a little more extended contact, and the removal of any ammonia or sulphur dioxide requires even more contact with the cooling water. In other words, very little packed height is needed to achieve cooling, but quite a bit more is needed to remove contaminants. Thus, accurate simulation of the quench tower can benefit operations by predicting how much sulphur dioxide from an SO2 breakthrough will actually reach the TGTU amine section, how much will be removed in the quench water, and how much ammonia will be captured as well.

How a quench tower performs in terms of heat transfer and protection of the TGTU is discussed in the context of a case study using conditions from an operating refinery unit. In particular, SO2 breakthroughs and controlling these events using ammonia or caustic injection are addressed.

Case study
For this study, the TGTU quench column and quench water circuit shown in Figure 1 have been isolated from the overall SRU-TGTU flowsheet The column contains 20 ft (6.1 m) of IMTP 50 random packing. The gas enters at 15 psia (1.03 bara) and 555°F (290°C)1 with the component molar flows shown in Table 1. The ammonia concentration of 50 ppmv is not atypical for the gas going to a quench column in a tail gas treating unit. For the case study the SO2 to NH3 molar ratio was varied from 0.1 to 3.0. Quench water was maintained at 93°F (33.9°C). Condensed water was drawn from the circuit via Stream 7 at whatever rate was necessary to keep the circulating water flow to the quench column constant at 300, 450 or 683 bbl/h (209, 314, 476 std. USgpm or 47.5, 71.2, 108.1 std. m3/h).

More than the overall performance of the quench system, the study (using the ProTreat mass and heat transfer rate based simulator) was done to expose the inner workings of the column itself, and how it might handle various levels of SO2 breakthrough from the SRU. In other words, when can the quench column be expected to protect the downstream amine system from an SO2 breakthrough, what level of protection can be provided, and what can be done operationally to mitigate a higher level breakthrough? First, however, how does the quench fulfill its primary function of cooling the gas?

Gas cooling
Figure 2a shows temperature profiles across the column while Figure 2b shows the changing flow rate of water in the gas. The parameter in these plots is the circulation rate of the quench water. As soon as the gas enters the column, it is immediately quenched by about 400°F.

However, at the lowest quench water rate, much of the initial cooling at the bottom of the packing is by evaporation of water (that is, transpiration, or swamp cooling), and the gas remains very high in water until it approaches the upper half of the bed where it meets cooler water. Cooling is by sensible heat transfer as well as by transpiration cooling. At low circulation rates, much of the 20ft packed bed is necessary to cool the gas — the heat transfer process is not fast. None of this is obvious from inlet and outlet stream measurements, because the outlet gas temperatures in these extreme cases differ by only a fraction of a degree. Furthermore, as will become apparent, the outlet SO2 concentrations differ by at most 2 ppmv in a total value of 65 ppmv. ProTreat simulation shows that virtually all ammonia is removed, but the fraction of SO2 removed depends strongly on the SO2 to NH3 ratio in the inlet gas.

NH3 and SO2 removal in the quench column
Ammonia and sulphur dioxide, being alkaline and acidic respectively, react not just with water, but strongly associate with each other in the water phase:

Ammonia is highly soluble in water, whereas sulphur dioxide is only sparingly soluble. Any dissolved ammonia will tend to drag an equal amount of SO2 into the water phase with it through the aqueous acid base reactions above. Thus, if the SO2 to NH3 ratio in the gas is less than one, the cooled gas can be expected to be virtually SO2 free. Conversely, if this ratio is greater than about one, the gas phase should be ammonia free but will contain whatever SO2 cannot dissolve physically in the quench water. If the gas has more SO2 than ammonia, the water will be acidic but if there is more ammonia than SO2 it will be alkaline. This expected behaviour is why monitoring the pH of quench tower pumparound water is a good practice.

First, it is worth noting that the SO2 and to a lesser extent the NH3 concentrations predicted to remain in the quenched gas are relatively insensitive to the quench water flow rate (as long as this is high enough to leave the quenched gas temperature relatively unaffected). Therefore, there are almost no external observations that would give any indication of behaviour inside the quench tower. Table 2 shows these predictions.

The most striking predictions of molar flow rate2 profiles of SO2 and ammonia are at the 300 bbl/h quench water rate shown in Figure 3. Profiles at the highest quench water flow (683 bbl/h) are shown in Figure 4 for comparison. When the SO2 and ammonia levels are equal in the inlet gas (at 300 bbl/h), th SO2 flow profile through the quench column (blue line in Figure 3a) follows the temperature profile (blue line) in Figure 2a. The ammonia profile in the lower half of the column remains fairly steady until it falls through two orders of magnitude in the top half.