(345f) Process Optimization of Calcium-Looping for Concentrating Solar Power Plants | AIChE

(345f) Process Optimization of Calcium-Looping for Concentrating Solar Power Plants


Rodrigues, D. - Presenter, Instituto Superior Tecnico
Pinheiro, C., Instituto Superior Técnico
Mendes, L. F., Instituto Superior Técnico
Energy storage is essential to deal with the variability of renewable energy. Concentrating solar power (CSP) systems rely on the concentration of solar energy in a receiver and allow high-temperature thermal energy storage (TES). The most common solution for TES relies on nitrate-based molten salts, the operational limitations of which do not allow high efficiencies. An alternative method for TES in CSP systems is thermochemical energy storage (TCES), which is based on reversible reactions. For example, the calcium-looping (CaL) process for TCES relies on a cycle that comprises (i) an endothermic calcination reaction, in which CaO and CO2 are generated from CaCO3 in a solar reactor, and (ii) an exothermic carbonation reaction, in which CaCO3 is formed from CaO and CO2. This process is advantageous since CaO precursors such as limestone are cheap, abundant, and harmless, the energy density is large (>3 GJ m‑3), and carbonation can occur at high temperatures, which improves the efficiency in CSP plants [1]. A main disadvantage is CaO deactivation after several cycles, which decreases the conversion in the carbonator [2].

Several versions of the CaL-TCES process have been proposed. For example, [2] proposed a closed and direct CO2 Brayton cycle, where CO2 is fed to the carbonator in excess and the non-reacting CO2 transfers the energy released in the carbonator to a gas turbine for electrical power generation. However, continuous solid-solid heat exchangers and gas-solid heat exchangers with counter-current flow are assumed, which may be unrealistic. For this reason, [1] proposed another version of the CaL-TCES process without these types of heat exchangers. Efficiencies above 38% were achieved for specific carbonator and main turbine outlet pressures, but their combined effect was not fully evaluated. In the CaL process for CO2 capture, a make-up stream with fresh CaCO3 is typically added to the calciner and a purge stream is used to prevent excessive CaO deactivation [3]. The purge stream can then be used as a raw material for cement production. However, the previous studies found in the literature about the CaL-TCES process do not consider CaCO3/CaO make-up and purge streams.

Hence, in this work, the CaL-TCES process in [1] is used as a starting point for a simulation and optimization study. The flowsheeting tool gPROMS Process is used to simulate and optimize the CaL-TCES process, which includes: a solar calciner where solar energy heats up CaCO3 and CaO and fully converts CaCO3 to CaO and CO2; a carbonator where the stored energy is released by converting CaO and CO2 to CaCO3; turbines for electrical power generation; tanks modelled as sources and sinks, where the reaction products are stored before their use as reactants; and a heat exchanger network around the reactors to enable heat integration.

When solar radiation is available, the CO2 that leaves the calciner (i) heats up the solids that enter the calciner and (ii) evaporates water in a heat recovery steam generator, which is then used in a Rankine cycle; a part of the CO2 passes through a high-pressure compressor before entering the CO2 storage tank when solar radiation is available and passes through a high-pressure turbine after leaving the storage tank when no solar radiation is available; the CO2 that enters the carbonator either passes through the main compressor or comes from the calciner when solar radiation is available or from the CO2 storage tank when no solar radiation is available; this CO2 is pre-heated by the solids that leave the carbonator and the CO2 that leaves the carbonator and passes through the main turbine. Intercoolers and pre-coolers are used for all the compressors, while a pre-heater and two reheaters are used for the high-pressure turbine. Several process parameters are given in [1].

The optimization goal is the maximization of the thermal-to-electrical efficiency of this process subject to the constraints mentioned above. The thermal power is the heat supplied in the calciner, while the electrical power is the difference between power generation in turbines and power consumption in pumps, compressors, coolers, and solids transport. However, the innovative procedure in this work differs from [1] in two relevant ways:

1) The two most relevant cases in [1] are: main turbine outlet pressure at 1 bar and pressure ratio between the carbonator and main turbine outlet pressures varying around the nominal value of 3; carbonator outlet pressure at 1 bar and the same pressure ratio varying around the nominal value of 3. The nominal cases lead to efficiencies of 38.1%, while an efficiency of 38.7% is achieved by increasing the pressure ratio in the first case. However, in fact both pressures can be modified independently. Hence, in this work, these pressures are decision variables determined via numerical optimization.

2) In [1], the conversion in the carbonator is a parameter with nominal value of 15%, and a sensitivity analysis with respect to this parameter is performed, from which 38.7% of efficiency is obtained for a conversion of 40%. Moreover, the process in [1] does not include make-up and purge streams, thus the conversion corresponds to the residual conversion of CaO after many cycles, which is lower than the conversion for a small number of cycles. In addition, the absence of make-up and purge streams does not allow accounting for the energy savings owing to the CaO that leaves the process, which avoids the energy consumption for calcination of CaCO3 in cement production plants. Hence, in this work, (i) make-up and purge streams are included, (ii) the conversion in the carbonator is computed according to the number of cycles experienced by the particles, and (iii) the CaO that leaves the process is considered for the efficiency. A make-up stream (FRESH) with 100% of fresh CaCO3 is added to the recycle stream, which corresponds to the solids from the storage tank of carbonation products that do not leave the process in the solids purge stream (SPURGE). The heat duty required to cool down SPURGE heats up FRESH, although this solid-solid heat exchange does not need to be continuous. In addition, a CO2 purge stream (GPURGE) is included, and the molar flow rates of GPURGE and FRESH depend on the molar flow rates of CaCO3 and CaO in SPURGE to ensure that the mass balance is satisfied.

The fraction of particles that have experienced a given number of cycles depends on the ratio between make-up and recycle molar flow rates, and the conversion decreases with the number of cycles. Then, the average conversion in the carbonator is obtained as an explicit function of the solids purge split fraction f_p that leaves the process in SPURGE. The efficiency depends not only on the generation and consumption of electrical power, but also on the power savings owing to the CaO in SPURGE. From the perspective of cement production plants, for each mole of CaO in SPURGE that replaces a mole of CaCO3 in FRESH, one can avoid the supply of high-temperature thermal energy that would be required for calcination of one mole of CaCO3, which would be enabled by a burner. Since that consumption is avoided, electrical power can be produced elsewhere in a combined cycle.

Regarding the optimization of the carbonator and main turbine outlet pressures without purge, which considers the residual conversion of 15%, the efficiency can be improved from 38.1% in the nominal conditions of [1] to 39.2%. This is achieved by changing the carbonator and main turbine outlet pressures from 3 bar and 1 bar to 1.485 bar and 0.293 bar, respectively. This shows the importance of optimizing the most relevant decision variables.

Concerning the simulation results with varying f_p between 0 and 1, which considers the carbonator and main turbine outlet pressures of 3 bar and 1 bar, respectively, the efficiency can be improved by increasing f_p from 0 to 1. The efficiency without savings owing to the CaO decreases from 38.1% to 32.0% when f_p increases from 0 to 1. On the other hand, the savings owing to the CaO increase from 0% to 17.1% of the solar input. For an efficiency of a combined cycle equal to 0.54 and a burner efficiency equal to 0.90, the efficiency increases from 38.1% to 42.2%.

Future work will consider: (i) energy consumption required to transport the fresh CaCO3 to the process and the solids purge from the process; (ii) energy consumption required for gas separation in a fluidized bed reactor with a fluidizing gas other than CO2 for calcination; (iii) time-varying solar power input throughout the day and the year.

[1] Ortiz C., Romano M.C., Valverde J.M., Binotti M., Chacartegui R., 2018, Process integration of Calcium-Looping thermochemical energy storage system in concentrating solar power plants, Energy, 155, 535–551.

[2] Chacartegui R., Alovisio A., Ortiz C., Valverde J.M., Verda V., Becerra J.A., 2016, Thermochemical energy storage of concentrated solar power by integration of the calcium looping process and a CO2 power cycle, Applied Energy, 173, 589–605.

[3] Li Z., Cai N., Croiset E., 2008, Process Analysis of CO2 Capture from Flue Gas Using Carbonation/Calcination Cycles, Environmental and Energy Engineering, 54, 1912–1925.