Process Design of Natural Gas Sweetening Plants
Raw NG varies substantially in composition, depending on the source reservoir. Methane is always the major component, typically 75%-90% of the total, but NG also contains undesirable impurities such as water, carbon dioxide, nitrogen and hydrogen sulfide that need to be removed before it can be stored as compressed gas or delivered to the pipelines. C02 content varies from 3% to 10% and needs to be removed in order to avoid pipeline corrosion and minimize atmospheric pollution [159]. Owing to the very low concentration of C02, a single-stage membrane, as shown by modeling, cannot produce high purity permeate or residue even at very high inlet pressures and/or large membrane areas. The solution is a combination of a multiple membrane stages, in parallel or in series. Such combinations, however, result in higher capital costs, due to high membrane area, and operating expenses, due to high compression costs. In such cases membranes are less competitive than conventional gas separation technologies. The viability of membrane systems will be dependent on process synthesis, configuration and design [26].
The cascade model (shown in Figure 6.6) is the most general multistage membrane design. The downstream membranes enrich the desired components in the permeate and upstream membranes strip remaining traces of desired components. In the case of NG sweetening, an upstream stripping section is required in order to minimize the concentration of sour gas
FIGURE 6.6
Schematic of cascade membrane system with recycles [160].
(C02 and H2S) in retentate. Furthermore, in order to minimize NG loss in the permeate stream, a downstream stripping section is also needed.
Pettersen and Lien [161] and Avgidou et al. [162] studied the behavior of several single-stage and multi-stage permeator systems. It is proven that at very low feed concentrations or with low-efficiency membranes, a two- or three-stage membrane system is the most technoeconomically optimal configuration. Similar results were reported by Carapellucci and Milazzo, who highlighted in 2003 that the two-stage design is the best option for enriching the CO, stream [163]. Ho et al. [164] investigated cascade systems with the introduction of vacuum instead of pressurized permeation. According to their results, the vacuum two-stage system with retentate recycling could achieve the highest C02 purity. A slight modification examined by Merkel et al. [165] also introduced a two-stage and two-step membrane system considering countercurrent flow at vacuum mode and using air as a sweep gas to generate driving force. Alshehri et al. [160] introduced a multi-stage membrane network superstructure of possible flowsheet configurations. An optimization formulation was then developed and solved using an objective function that minimizes the costs associated with operating and capital expenses. Their main conclusion is that developing membranes with higher permeance rather than high selectivity will help membrane technology's competitiveness.
The membrane NG sweetening systems are either single or two-stage processes [166]. The single-stage design is suitable for remote and unmanned facilities, where maintenance and troubleshooting are more difficult. Furthermore, the absence of a compression station in the single-stage design diminishes the additional capital and operating cost. However, due to membranes' non-perfect selectivity the single-stage design is less advantageous than conventional technologies (e.g., amines). It is therefore necessary that a two-stage process (single recovery stage) be introduced. With this scheme, the overall separation outcome and economical indexes are more attractive than those of conventional technologies [166].
Finally, several studies have recommended the use of hybrid units consisting of a membrane unit followed by an amine unit [167]. The hybrid units can accomplish targeted separation at a lower cost but such units are not widely utilized in industry and they have increased operational and troubleshooting complications.
