You are currently viewing: Articles


Mercaptans removal from gases by absorption into amines and caustic

A mass transfer rate-based model of the absorption process is used to discuss quantitatively the ease and extent of removal of methyl through n-butyl mercaptans from gases using MDEA and caustic soda of various strengths.

Clayton E Jones, Nathan A Hatcher and Ralph H Weiland
Optimized Gas Treating Inc
Viewed : 5167
Article Summary
A direct positive relationship is found between the absorption rate into amines and the molecular weight of the mercaptan. The relationship is inverted when the solvent is caustic soda.

Sulphur emissions from refinery combustion sources are traditionally reported by measuring H2S in the fuel gas before combustion and converting the H2S to an equivalent amount of post-combustion SO2. In reality, all sulphur species present in the fuel gas (not just H2S) contribute to SO2 emissions, and some operating facilities have to report the total sulphur emissions. Mercaptans are commonly present in refinery gases in small amounts. As the regulatory limits on emissions are progressively lowered, knowledge of the behaviour of mercaptans in amine and caustic treating systems becomes increasingly important for designers and operators both to track accurately and to comply with total sulphur emissions restrictions.

The ProTreat® gas treating process simulator makes exclusive use of the mass transfer rate approach to predicting the separation that will actually be achieved in a given tower containing specific internals under actual operating conditions. Absorption (and stripping) rates are calculated directly from knowledge of the mass transfer characteristics of the internals, the hydraulic conditions within the tower, and the driving force for the absorption process.

The driving force for absorption relates directly to the difference between actual and equilibrium concentrations of the mercaptan dissolved in the solvent. Indeed, even if we choose to use an equilibrium stage model of the column and to step off ideal stages, equilibrium mercaptan solubility is still an important parameter. Thus, an essential factor to consider is the solubility of the mercaptans in treating solvents. Because mercaptans hydrolyse as weak acids in water, proper accounting of the chemistry is vital to predicting accurately their solubility behaviour.

To provide perspective to the substance of this article, we begin with a general description of what constitutes a mass transfer rate model, and what enables it to be truly predictive, as opposed to just descriptive. This leads to an account of the equilibrium solubility of the C1 through nC4 mercaptans in amine and inorganic caustic treating solutions, together with an explanation for the differences amongst the mercaptans and between the solvents. The mass transfer rate model is then used to show how mercaptan concentrations can be expected to vary within a real absorber when MDEA and a caustic soda solution are used for their removal. The focus is on how process variables affect mercaptan removal. Stripping of the absorbed mercaptans from a spent treating solution is left for a future article.

Mass Transfer Rates for Column Calculations
An ideal stage is a theoretical concept that, in principle at least, allows one to replace a certain number of real trays or a certain depth of real packing in a column with a hypothetical equivalent in which all the streams leaving each ideal stage are in perfect equilibrium. The origin of the concept is obscure, but a simple physical example is a flash tank in which feed is flashed into vapour and liquid, assumed to be in mutual equilibrium. The first method for calculating a multistage column separation can probably be attributed to Ponchon1 and Savarit2, although the method of McCabe and Thiele3 has seen the most widespread use. The methodology is elegant in its simplicity, requiring only phase equilibrium data and the flow, composition and thermal condition of the tower feeds. It has one great weakness, however: it has no direct connection to real equipment. The way this kind of theory is connected to equipment is usually through stage efficiency or height equivalent to a theoretical plate. Unfortunately, the weakness continues: efficiencies and height equivalents are not easily related to the construction and functioning of equipment, and in multicomponent mixtures, their behaviour can be extremely confusing (see Kooijman and Taylor4). A true mass transfer rate model suffers from no such drawbacks.

At its core, a mass transfer rate model focuses on the rate of interchange of chemical components between two phases as driven by the extent to which temperatures and species concentrations between phases depart from equilibrium. This is the very antithesis of the ideal stage model with its assumption of perfect equilibrium, not disequilibrium. Figure 1 shows schematically two phases being contacted in some portion of the biphase on a tray or in a segment of packing within a column.

Transferring components (H2S, RSH and H2O as shown in the diagram) first diffuse through the vapour to the interface where they then dissolve into the liquid and diffuse away, perhaps accompanied by chemical reaction and the liberation of associated heat. The diffusion process is by both molecular and convective means. It depends on the size of molecules (molecular means) and on the state of agitation of the individual phases (convective means). The central understanding, however, is that diffusion by whatever means is a rate process. It takes place at a rate that can be calculated, and the separation actually achieved is governed by that rate. The separation really reached (on a given tray, for example) does not depend on the efficiency of the tray — it depends on the rate at which the mass transfer process takes place.

The mass transport process is characterised by mass transfer coefficients. In heat transfer these coefficients are known as heat transfer coefficients corresponding to either side of the heat transfer surface. The overall heat transfer coefficient for a given exchanger is obtained by combining the film coefficients via the law of additivity of resistances. The film coefficients themselves are found from charts (or internally by software) where very well-established generalised coefficients have been plotted or tabulated against flow parameters and physical properties (c.f. Kern5). This is the way heat exchangers have been designed for nearly a century, so the concept of a transfer rate-based process is anything but new. Indeed, heat exchangers are probably the best analog of the rate model for mass transfer. Mass transfer equipment has been designed this way only since the mid-1980s, not because the mass transfer rate-based concept was new, but because the computing power needed to perform the calculations on large numbers of components over multiple trays only became available then.

Mass transfer coefficients and interfacial areas are in no sense adjustable parameters any more than heat transfer coefficients are. The flow of high and low viscosity fluids across trays and over packing agitates the fluid and results in unique values for mass transfer coefficients. Engineers are not free to pick and choose coefficient values to obtain the preferred result or some preconceived notion of what the separation should be. On the contrary, values of the coefficients are a result of fluid flows, fluid properties, and the mechanical design of the specific tower internals using well-established correlations. The parallel with heat transfer is extremely close. Tower performance is predetermined and predictable in exactly the same way and to exactly the same extent as heat exchanger performance.
Current Rating :  4

Add your rating:

Your rate: 1 2 3 4 5