Flue Gas Drying
The drying of flue gas for post-combustion CO2 capture
process incurs high energy demands and is thus often considered
impractical.22-24 If flue gas can be dried, however,
the options available for CO2 capture broaden
considerably. In our process, flue gas drying is carried out in three
stages, first using cooling water from the plant, then taking advantage
of Joule-Thomson cooling from the sub-ambient process, and finally, an
adsorbent dryer bed. We expect the water from the first two drying
stages to be recovered and sent to a cooling tower where it can be
returned to the desired conditions. We assume that cooling water at 30oC is used to remove the water before compression
through the application of a direct contact chiller (DCC). The removal
of as much water as possible before compression decreases the total
volume of gas that must be compressed, which ultimately reduces the
volume of gas to be compressed by ~10%. Removal of
water is also vital when using axial compressors, as the condensation of
water inside the compressor may damage the airfoils.
Importantly, the sub-ambient process we considered integrates water
removal with feed compression and cooling via boiler feed water
preheating. This decreases the cost of dehydration that would otherwise
be expected if the equivalent energy needed to come from an external
source. After the direct contact chiller, compression, and cooling the
mole fraction of water in the flue gas stream is estimated in the Aspen
model to be ~0.5%, as shown in Table S1. After the flue
gas has been compressed to the target pressure, it is cooled to 2oC using available cold N2 enriched
products as mentioned previously. This additional cooling followed by a
knock out decanter results in 528 ppm residual water left in the stream
to be removed via the adsorbent. With such a low concentration of water
in the flue gas, it is assumed an adsorption dryer will remain
essentially isothermal during drying. If the heat integration was
unavailable, the economic and energy penalty of the drying would be
associated with the cost of drying 10-12 times as much water. Upon
completion of its expansion to near atmospheric pressure, the
N2-rich product stream is used as a dry sweep gas for
the regeneration of the dryer bed and then sent to the stack.
Sub-Ambient Heat
Exchange
Extensive heat integration throughout the process is critical to avoid
an expensive external refrigeration utility. The compression of the flue
gas upstream is primarily responsible for the process’ ability to
operate at a steady state with no external refrigeration, as
conceptualized in Figure 1. The more the flue gas is compressed, the
more high-quality heat removal becomes available from the multistage
expansion of the N2 rich byproduct. The base case
considers two-stage expansion to produce the required heat removal so
the N2-rich product passes through the heat exchanger
network three times as seen in Figure 2. This expansion-cooling train is
responsible for the majority of the following steps: cooling the inlet
gas entering the dryer unit, knocking out additional water, cooling the
inlet gas to the separation unit, cooling the compressed
CO2 enriched product after recompression to the
liquefaction condition, and liquefying the CO2 enriched
product. In all cases considered, the minimum temperature approach of
the process heat balance was determined by the combination of the
primary separation final purity and recovery and the liquefaction
temperature and pressure. All cases discussed in the following
discussion had a fixed recovery of 92% in the primary separation system
to enable an overall system recovery of 90%.