Unlocking the economic potential of Direct Air Capture technology: Insights from a component-based learning curve

Abstract

Direct air capture (DAC) technology is gaining increased attention for its flexibility and effectiveness in carbon removal. However, the high cost of DAC hinders the potential for emission reductions. We provide an approach to help evaluate these costs. To overcome the limitations of poor data quality, high technical complexity, and uncertainty in the cost forecasting of DAC techniques, we develop component-based learning curves based on four available DAC technologies (potassium hydroxide, monoethanolamine, solid amine, and bipolar membrane electrodialysis adsorbents). The results indicate that the capital cost learning rate ranges from 4.87 % to 11.02 % and is influenced by components like contactors and scrubbing towers. In contrast, the operational and maintenance cost learning rate ranges from 13.70 % to 20.61 %, with the key components being contactors and adsorbers. Upon reaching the “learning saturation point”, the levelized (US dollar) cost per ton of carbon dioxide (CO₂) capture of the four techniques is projected to decline significantly to 56 % ($120/t CO₂), 28 % ($253/t CO₂), 23 % ($412/t CO₂), and 25 % ($356/t CO₂) of their initial values, respectively. Bayesian methods enhance learning rate reliability, and sensitivity analysis reveals energy price fluctuations significantly impact DAC costs. These insights support techno-economic modeling, climate assessments, and strategic DAC deployment.