Page 73 - Gas Purification 5E

AIRanoImines for Hvdmgen Sulfide and Carbon Dioxide Removal 63

mous. Although significant data gaps still exist, numerous WE studis have been conducted
and reported in literature. Particularly notewarthy are the extensive publications of Dr. A. E.
Mather and coworkers at the University of Alberta, Edmonton, Canada.
The acid gas solubility data presented in the following sections are generally limited to condi-
tions near those most commonly encountered in commercial systems. References are provided
to sources of additional data and to correlations that have been developed for predicthg VLE
relationships in the absence of specific data. Charts and tables of experimental data are useful for
preliminary studies; however, correlations are needed for interpolating and extrapolating data to
specific conditions, and are required for computer-based amine system design programs.

Since MEA was one of the first ethanolamines used for gas mating and is still widely
used a large amount of VLX data has been published covering MEA solutions of CO, and
H.S. Thc magnitude of this effort can be appreciated by inspection of Table 2-5, which lists
most of the papers presenting experimental data on the subject.
Much of the early work was conducted with dilute (-15 wt%) MEA solutions because
such solutions were commonly used in commercial plants at the time since higher concentra-
tions were considered too corrosive. With the advent of corrosion inhibited solutions and a
better understanding of corrosion mechanisms, more concentrated solutions have become
popular. This is reflected in the recent VLE data, which typically covers both 15 and 30 wt%
solutions. More data are provided for MEA than for the other amines because of its wide
spread and long time commercial use. Also, many of the conclusions for MEA, such as the
general effects of temperature, amine concentration, and the presence of other acid gases are
also applicable to other amines.
Figures 2-15 through 2-28 and Tables 2-6 and 2-7 present data on the solubiliiy of CO?,
H2S, and mixtures of the two acid gases in MEA solutions. Most of the data are for 2.5N
(approximately 15 wt%) and 5.0N (approximately 30 wt%) solutions. Figure 2-17, which
gives data for H2S in 15.3 wt% MEA, includes curves calculated by the Kent-Eisenburg cor-
relation, which is discussed later.
Figures 2-26 and 2-27 show the effect of temperature on the vapor pressures of CO? and
H2S respectively for various acid-gas/amine mole ratios. The curves are nearly straight lines
on the log P versus 1/T coordinates, which aids in extrapolation to other temperatures. The
plots also provide an indication of the heat of reaction, which, in accordance with the Clau-
sius-clapeyvn equation. is proportional to the slope of the Iines. The decreasing slope with
increasing concentration of acid gas in the solution indicates that the heat of reaction
decreases as more acid gas is absorbed.
Figure 2-28 shows the effect of increasing the concentration of MEA on the vapor pes-
sure of CO, at various acid gas to amk mole ratios and at a temperature of 77°F. Increasing
the &EA concentration increases the C02 vapor pressure at the same mole ratio. From e
prarical standpoint, this means that the quantity of solution required does not decrease in
inverse proportion to the amine concentration. For example, with 10 psia C02 partial pres-
sure in the feed gas, and equilibrium at 77"F, the maximum solution capacity is about 0.8
moles C02/mole amine in a 10 wt% MEA solution and 0.6 moledmole in a 40 wt% solution.

Sources of data for the solubility of acid gases in DEA solutions are given in Table 2-8.
Typical data are given in Figures 2-29 through 2-35. The principal charts cov&g the indi-