Pre-combustion CO2 capture: Analysis of integrated reforming combined cycle

Abstract

This thesis presents processes for reducing CO2 emissions from natural gas (NG) power plants, which could help attenuate the rise in atmospheric temperature. Objectives for the thesis work were process design and integration of NG pre-combustion CO2 capture plants, and evalution, through process simulation, of the concepts. An important aspect of the evaluation was investigation of plant flexibility, specifically off-design analysis. Contributions comprised detailed heat recovery steam generator (HRSG) design for NG pre-combustion cycles and off-design behavior of the integrated reforming combined cycle (IRCC). Additional contributions were quantification of the efficiency potential of a process, subsystem efficiency losses, and model uncertainties. The IRCC with pre-combustion CO2 capture is a process for generating power with very low CO2 emissions, typically below 100 g CO2/net kWh electricity. This should be compared to a state-of-the-art natural gas combined cycle (NGCC) with CO2 emissions around 365 g CO2/net kWh electricity. The IRCC process reforms natural gas to a syngas, converts the CO to CO2 in the shift reactors, separates the CO2 in the capture subsystem, and the resulting hydrogen-rich fuel is used for the gas turbine (GT) in a combined cycle setup. For the reforming of natural gas, an air-blown autothermal reformer was selected for the processes studied. Included in the study of the IRCC were process design and integration, reliability analysis, thermodynamic analyses through process simulation, analysis of efficiency losses and efficiency potential, and uncertainty analysis. As part of the design process, HRSG design proved important. The design of an HRSG for an IRCC plant requires the ability to operate on both a hydrogen-rich fuel and on NG. Also, since a significant amount of steam is produced from the heat generated in the autoreforming process, the HRSG design differs from a design in an NGCC plant. For an IRCC with a lot of high-pressure saturated steam generated in the process, a single-pressure steam cycle can actually perform in parity with a dual- or triple-pressure system (with or without reheat). Preheating of process streams further add to the complexity. The complexity of selecting an HRSG design increased when also considering that steam could be superheated and low-pressure and intermediate-pressure steam could be generated in the reforming process heat exchangers. For the concept studied it was also of importance to maintain a high net plant efficiency when operating on NG. Therefore the selection of HRSG design had to be a compromise between NGCC and IRCC operating modes. Duct burning proved positive for plant flexibility and the option to switch between a hydrogen-rich fuel and NG for the GT. Functional analysis and FMECA are important steps in a system reliability analysis, as they can serve as a platform and basis for further analysis. Also, the results from the FMECA can be interesting for determining how the failures propagate through the system and their failure effects on the operation of the process. From the FMECA performed in this work, it is clear that the gas turbine is the most critical equipment in an IRCC plant. One of the reasons for this is the process integration between the power island and the pre-combustion process. For example, the gas turbine feeds air to the ATR and receives fuel from the pre-combustion process. This integration has an effect on the overall reliabilityof the system. In addition to the integration issues, the gas turbine technology is less mature for hydrogen fuels than for natural gas fuels. It should also be mentioned that even in an NG-fired combined cycle plant the gas turbine is the most critical equipment. The need for part load analysis and consideration to dual fuel capability were important conclusions from the reliability analysis since many of the failures resulted in IRCC plant shutdown (if no backup fuel) or operation at reduced load. Thermodynamic analyses through process simulation were conducted as part of the thesis work. By combining simulation tools for chemical engineering and power plant engineering analyses respectively, a representation of the overall system could be accomplished for an IRCC process. The reforming and CO2 capture processes were simulated in Aspen Plus; the power island was simulated in GT PRO/GT MASTER. The IRCC process involved process integration between the power cycle and the reforming process meaning an efficient way of linking the softwares were important. The Aspen SimulationWorkbook and Thermoflow’s E-link proved capable of performing this task for an IRCC process. Design simulations showed net plant efficiencies between 41.9% and 45.3% with net plant power output in the range of 350–420 MW. The CO2 capture rate ranged between 85.1% and 93.4% for the IRCC processes studied. The off-design simulations, as part of the plant flexibility analysis, showed the possibility to operate an IRCC plant at part load conditions down to approximately 60% gas turbine load with capture efficiency penalties at part load similar to full load operation. Also, it can be concluded that considering off-design conditions, such as part load steam turbine extraction pressures and air booster compressor pressure ratio, are important during the design stage of a plant. Analysis of the contribution to efficiency losses in the IRCC process showed that the reforming losses were almost twice as high as the CO2 capture losses. From the analysis,it was evident that to decrease the efficiency losses in an IRCC process, efforts should be concentrated towards improving (1) the reforming process to decrease fuel conversion losses and needed steam mass flow, (2) the CO2 capture process to decrease the reboiler duty, (3) the gas turbine technology to allow for a higher firing temperature, and (4) the CO2 compression process. When investigating the efficiency potential of the type of IRCC concepts studied in the thesis work, net plant efficiencies of 49% was achieved and based on these results it is conceivable that efficiencies up towards 50% could be realistic in a 5–10 years time horizon. Challenges to overcome to reach these high efficiencies include attenuating or eliminating process limitations due to metal dusting and reduced GT turbine inlet temperature. For the IRCC setup studied in the uncertainty analysis, results showed that there was considerable uncertainty in the predicted net power output whereas net plant efficiency, CO2 capture rate, and CO2 emitted were less affected by input uncertainties. Parameters with the largest impact on uncertainties of power output and efficiency predictions proved to be gas turbine inlet temperature, and compressor and turbine efficiencies. For the CO2 emissions, the equipment pressure drop and the steam-to-carbon ratio proved important. Therefore, the focus of future work should be to reduce uncertainties in these parameters in order to improve the confidence in the IRCC model.