The chemical processing industry certainly doesn’t lack challenges today. Rapidly escalating capital costs and high energy prices are the new reality.
Many studies indicate that separations consume 50% to 70% of the capital and energy of a traditional chemical process ; so, seeking improvements in this area clearly should receive a high priority. One of the more effective methods of reducing both the capital and energy costs of separations units is installation of dividing wall columns (DWC) where appropriate. While results, of course, vary, a DWC typically can cut capital and energy costs by approximately 30% compared to a traditional two-column system.
The idea of the DWC isn’t new. The original patents on the concept were granted to Wright of Standard Oil in 1949  — and that patent cited prior art! Today there’s a large body of literature on the DWC and associated concepts — a recent survey of patents and published articles revealed roughly 400 relevant matches. Schultz et al.  relatively recently authored an excellent overview article. Since that article, other announcements and presentations have indicated a gathering momentum in distillation intensification. For instance, in 2005 at the 7th World Congress of Chemical Engineering, BASF, a leader in DWC implementation, reported more than 40 DWC in commercial service. In a 2006 symposium , ExxonMobil announced the successful revamp of a 380-cm/430-cm-diameter tower that removes xylene from reformate. Dow has operated an experimental DWC facility in Midland, Mich., for about eight years. During the same time, numerical expertise for the modeling of DWC systems has been developed — and bolstered and validated by the experimental units.
Figure 1. Conventional scheme -- When two
Let’s consider the separation of three components, A, B and C, with A being the lightest, C the heaviest, and B the intermediate volatility component. One option is to use a direct sequence, removing A first with B and C going out the bottom. Figure 1 shows this sequence, along with the relative tower concentration profiles for B.
In the first column, the concentration of B reaches a maximum somewhere in that tower but necessarily must be lower in the bottoms. This represents a thermodynamic loss that’s unavoidable in the sequential distillation scheme.
Now, consider an alternative two-column arrangement for the separations where in the first tower, rather than performing the sharp separation between component A and B, we pre-fractionate the components (Figure 2). In the first column, we take the entire amount of A and some fraction of B out of the top of the tower. The remaining fraction of B and the entire amount of the C exits the bottom of the tower.
Figure 2. Alternative arrangement -- In this