https://doi.org/10.1016/j.clet.2021.100249
“Using amine-based solvents, which was initially developed for gas processing and has more recently been customized for carbon capture from flue gases, has several challenges including:
- i.
-
A low CO2 molar fraction: This leads to small driving forces for absorption, resulting in large absorber size and, thereby, a higher CAPEX compared to gas processing with amines.
- ii.
-
The large flow rate of flue gas is not easy and energy-intensive to compress. This also dictates a large absorber size due to low operating pressures.
- iii.
-
The thermal and oxidative degradation of amines due to the presence of oxygen in flue gas while going through plenty of absorption/regeneration cycles. Amines are susceptible to thermal degradation and this adverse effect happens even faster where oxygen is present. Amines degrade to heat-stable salts, some of which can be converted back in reclaimers. Makeup streams of amines even with anti-degradation chemicals, are often unavoidable to compensate for the loss caused by degradation.
- iv.
-
The corrosion rate is higher due to the presence of oxygen in the flue gas. This implies either selecting corrosion-resistant materials for critical equipment or adding inhibitors to the amine solvent formulations. Again, both of these solutions result in a higher CAPEX or OPEX.
As stated before, the focus of this study is the impact of process modifications on the performance of amine-based PCC units. However, cost estimations are also necessary to rank the selected modifications from an economic standpoint, but this is rarely presented in the literature, where energy aspects are most often presented. Therefore, the CAPEX and OPEX of the studied cases for PCC units are estimated and compared. The analysis of CAPEX and OPEX are based on the cost breakdown approach proposed by (Peters et al., 2003) and later advanced by (Towler and Sinnott, 2013). This was done by transferring simulator data from Aspen HYSYS® to the Aspen Process Economic Analyzer (APEA), mapping, sizing, and costing/evaluating based on the cost functions embedded in the Icarus databanks of Aspen (IcarusReferenceGuide, 2004). It is noteworthy that such procedures for the design and economic evaluation are reportedly claimed to be valid by a recently published book (Madeddu et al., 2019).
2.4.1. Equipment sizing
The major process equipment in the PCC unit includes a blower, a DCC, absorber and stripper columns, a reboiler and a lean-rich heat exchanger. A resizing procedure was applied to specify the process equipment relative to the specifications mentioned in the reference Table 4. Absorber and stripper columns were simply resized to maintain the same vapor superficial velocity in both columns. For the MEA base case design, the dimensions and flow rates were scaled according to Table 5.(1)V2V1=D22D12×H2H1=F2F1:H2≅H1→D2D1=F2F1
Table 4. Process specifications for the simulation cases using MEA and a-MDEA.
| PCC Configuration | DCC Dimensions/Internals Type and Size | Absorber Column Dimensions/Internals Type and Size | Stripper Column Dimensions/Internals Type and Size |
|---|---|---|---|
| MEA cases Base case AIC AIC-PEA AIC-LVR AIC-LVR-PEA |
8.5 m × 23 m/11.5 m Sulzer MELLAPAK+ 752Y | 8 m × 44 m 6 m of 3.5″ (90 mm) washing pall ring, 1 bed 72′ (22 m) MELLAPAK+ 252Y, 2 beds |
3 m–2 m × 43 m 3 m × 29 m & 2 m × 14.5 m 92′ (28 m) of 3.5″ (90 mm) pall ring, 3 beds |
| a-MDEA cases Base case AIC AIC-PEA AIC-LVR AIC-LVR-PEA |
8.5 × 23 m/11.5 m Sulzer MELLAPAK+ 752Y | 11 m × 44 m 6 m of 3.5″ (90 mm) washing pall ring, 1 bed 22 m MELLAPAK+ 252Y, 2 beds |
3 m–2 m × 43 m 3 m × 29 m & 2 m × 14.5 m 28 m of 3.5″ (90 mm) pall ring, 3 beds |
Table 5. Process design parameters.
| Scaled down operating variable (dependent) | Adjusted Parameter (independent) | Remarks |
|---|---|---|
| Circulating water flow rate in DCC | Ratios of water molar flow rate in flue gas | Pump around: return temperature fixed at 32.8 °C Duty calculated by the software |
| MEA recirculation rate (USGPM) | Ratios of component molar flow rates of CO2 (directly proportional) | To maintain a similar lean CO2 loading in the optimal range of 0.2–0.25 for MEA and ~0.1 for a-MDEA (Amann and Bouallou, 2009) |
| Qreb | Ratios of component molar flow rates of CO2 (directly proportional) | Crosscheck was performed for the reboiler energy (~3.76 MJ/kg CO2) |
| Qcond | – | Adjusted automatically to condense the full reflux at 37.8 °C |
| Vessel volumes | Diameters adjusted using equation (1) obtained from the study by (Lee et al., 2016) | Height identifies the separation efficiency, and diameter identifies the hydrodynamics |
The results from mass and energy balances including composition, flow rate, temperature, pressure, and the enthalpy of different streams were solved in the simulation and adopted to size the process equipment. The sizing and cost estimation of all heat exchangers was performed in Aspen Exchanger Design and Rating (Aspen EDR).
2.4.2. Economic model
The results of the process simulations were used to estimate the total capital investment (TCI) and OPEX of the PCC unit. A delivery allowance of 10% was considered for the purchased equipment (PE) (Peters et al., 2003). This study used the US Gulf Coast as the reference location that assumes the delivered PE cost is equal to 115% of the PE cost. Further details are provided in a recent publication (Hu et al., 2017).
The economic results presented in this study are based on the following assumptions:
- ●
-
Plant capacity: 2000 TPD of CO2 captured at a capture rate of 90%
- ●
-
Cost reference year: 2018, 1st quarter
- ●
-
Capital escalation: 2.5%
- ●
-
Equipment materials: stainless steel (SS304) for all of the process equipment (Nexant Inc., 2016)
- ●
-
Plant life: 30 years
- ●
-
Annual operation: 8150 h (~93% on-stream factor or operational year)
- ●
-
Corporate tax rate: 30%
The first step of the economic analysis is to estimate the CAPEX of all major process equipment (Towler and Sinnott, 2013). In the absence of vendor quotes and/or invoices for previously purchased equipment, the cost curves/functions that are available in the literature for various types of common equipment are used. Overall, Aspen Process Economic Analyzer (more particularly Aspen Capital Cost Estimator) has been claimed to be a valid source for a consistent cost estimation (Madeddu et al., 2019).
In practice, APEA uses cost models from Icarus Evaluation Engine, which is adequately accurate for preliminary studies.
A capital cost scaling formula is applied, like seven tenth rule of thumb. It is of paramount importance to select the correct scaling factor, e.g., volumetric flow rate, area, or duty, and exponent for each piece of equipment (Hu et al., 2017). Apart from process equipment cost, the total plant cost (TPC) consists of engineering, procurement and construction (EPC); contractor services; and project contingency costs. The cost for EPC is considered to be approximately 8.4% of the bare erected cost (Giglio et al., 2015). The estimated cost is presumably for base equipment made from carbon steel and operating at atmospheric pressure as the reference condition. Material and pressure factors can be used to estimate the price of the purchased equipment with different materials and operating pressure.
Another simple technique to estimate the capital cost of a plant is the Lang factor method. The total cost is determined by multiplying the total purchase cost for all major equipment by a constant. A Lang factor in the range of 2.7–3 was back calculated in this study. It is worth mentioning that the compressor/intercooler capital costs were not priced correctly in Aspen’s Icarus databanks embedded in the Aspen Capital Cost Estimator (ACCE). This would cause an underestimation of the cost of the module and therefore an overly optimistic CAPEX estimation. The TPC of the compression package was overwritten manually in APEA based on a quoted cost. The purchase cost of columns with packing estimated by ACCE was reasonably close to a quoted price from the same year.
Total capital cost was calculated from annualized CAPEX and adding it to OPEX, according to the following formula, equation (2) (Towler and Sinnott, 2013):
(2)TOTEX=[i(1+i)n][(1+i)n−1]×CAPEX+OPEX
“