One of the most prominent process-intensified separation processes in the chemical industry is the dividing wall
column (DWC). A DWC has at least two conventional distillation columns integrated into a single shell, with
a wall partitioning the feed side from the product collection side (Gorak and Olujic, 2014). Results from open
literature claimed that the energy and capital costs of DWC are about 15% and 30% cheaper, respectively, when
compared to its conventional counterpart with two columns in series (Kaibel, 2014). The savings in energy are
achieved due to the wall in DWCs that allows the middle boiling component to be freely distributed, so the
composition profile in a DWC is well-developed, leading to minimized mixing and remixing effects (i.e., more
energy-efficient). Despite being energy-efficient, the industrial applications of DWCs are limited owing to their
structural complexities (e.g., DWC and DWC variations with multiple dividing walls installed). To reduce
the structural complexities, Agrawal (2000) proposed alternative DWC structures (a.k.a. Agrawal structures)
that replace some or all thermal coupling (TC) streams in the conventional DWC structure with liquid (L)
transfer streams. The Agrawal structures were claimed to have better controllability whilst being economically
comparable to regular DWCs (Agrawal, 2000). Considering a regular DWC, there are two TC streams, one at
the top of the wall and the other at the bottom of the wall. Depending on the TC stream replaced, the Agrawal
structure are coined (1) LTC structure, where the top TC stream is replaced with a liquid stream and the wall
extends to the top so an additional condenser is required; (2) TCL structure, where the bottom TC stream is
replaced and the wall extends to the bottom so an additional reboiler is required; and (3) LL structure, where
both top and bottom TC streams are replaced and the wall spans from the top to the bottom of the column shell
so an additional condenser and an additional reboiler are required. These structures are referred to as reduced
vapor transfer dividing wall column (RVT-DWCs) in this work.
With the structural complexities tackled, DWCs can be made more competitive by enhancing their energy
efficiency. Energy efficiency can be improved either through heat integration (e.g., heat pump or double-effect
column arrangement) or by replacing the typical non-renewable energy sources (e.g., fossil fuel) with renewable
electric power (Kooijman and Sorensen, 2022). The heat integration of DWCs has been studied extensively,
where one of the most popular heat integration methods is using a heat pump to upgrade heat from a lower
temperature source to a higher temperature one (Bruinsma and Spoelstra, 2010). However, there are limited
studies on the heat integration of RVT-DWCs. Therefore, this work will focus on heat integrated designs for
RVT-DWCs and compare them to the base DWC and vapor recompression assisted DWC. The LL structure
is chosen as a case study as it is the most unique structure with a wall spanning throughout the column, thus
opening more possibilities for heat integration.
Two base designs (DWC and LL structure) together with their improved structures, including vapour recompression
etc, will be optimized to minimize the total annualized cost, and the separation and economic performances
of each model will be compared. The optimization problems involve discrete variables such as the
number of stages and feed locations, and the heat integrated design is even more complex due to the increased
degree of freedom, requiring the initial designs to be robust for a successful optimization. Therefore, a shortcut
design for each model is carefully prepared and considered as an initial design for the optimization. An
optimization-based shortcut method from our previous study (Duanmu et al., 2022) is applied for the standard
DWC and LL structure in this work.
This work explored the benefits and challenges of heat integration of the DWC and LL structures in terms
of separation performance, energy consumption, and economics. The optimal results and additional sensitivity
analysis will reveal that vapor recompression assisted structures have great potential for energy and cost savings,
but the degree of saving is affected by the electricity and utility costs.