CHEMICAL ENGINEERING TRANSACTIONS VOL. 52, 2016 A publication of The Italian Association of Chemical Engineering Online at www.aidic.it/cet Guest Editors: Petar Sabev Varbanov, Peng-Yen Liew, Jun-Yow Yong, Jiří Jaromír Klemeš, Hon Loong Lam Copyright © 2016, AIDIC Servizi S.r.l., ISBN 978-88-95608-42-6; ISSN 2283-9216 Carbon Dioxide Capture from Model Marine Diesel Engine Exhaust by means of K2CO3-Based Sorbents Marco Balsamo*, Alessandro Erto, Amedeo Lancia, Francesco Di Natale Dipartimento di Ingegneria Chimica, dei Materiali e della Produzione Industriale, Università degli Studi di Napoli Federico II, Piazzale Vincenzo Tecchio 80, 80125 Napoli, Italy marco.balsamo@unina.it The on-board ship installation of a high-efficiency and low-cost CO2 separation unit seems to be an attractive option to comply with recent regulations aimed at reducing the carbon footprint derived from the maritime sector. Our research group is currently developing a capture and storage scheme based on the use of potassium carbonate for CO2 conversion into solid potassium bicarbonate. In this paper, we summarize preliminary CO2 capture tests on both raw and alumina-supported K2CO3 carried out in a fixed bed apparatus under typical model marine diesel engine exhaust composition (CO2 5 %, H2O 5 %, N2 90 % by vol.) and temperatures (60 - 105 °C). Carbonation data for the parent bulk K2CO3 showed a maximum capture capacity of 0.138 mmol g-1 corresponding to a nearly 2 % sorbent utilization factor. An increase in the operating temperature produced a reduction of the carbonation capacity and faster capture kinetics. The alumina-supported sorbent tested at 60 °C displayed enhanced CO2 capture capacity with a maximum conversion degree of 43 %. This testifies the positive effect derived by the dispersion of the active phase onto a substrate with large surface area in the CO2 capture process, likely for a remarkable reduction of diffusion limitations during the carbonation reaction in the case of nano-sized potassium carbonate with respect to granular sorbents. 1. Introduction The reduction of CO2 emissions deriving from human activities is nowadays required to face global climate change (Figueroa et al., 2008). Besides electricity and heat generation systems that produce nearly two-thirds of global CO2 emissions, about 23 % of carbon footprint derives from the transport sector (International Energy Agency, 2014). Contextually, the European Union set the target of 40 % cut in maritime shipping emissions to be achieved in 2050 (European Commission, 2011). In order to cut CO2 emissions, the International Maritime Organization (IMO) introduced different measures, among which the Energy Efficient Design Index, EEDI (International Maritime Organization, 2014). Ship design architecture, design of auxiliary systems, alternative propulsion systems and routing can reduce energy consumption and consequently CO2 emissions (Di Natale and Carotenuto, 2015). In the framework of the carbon capture and storage (CCS) approach, post-combustion purification systems for CO2 capture from flue-gas have the greatest near-term potential to mitigate CO2 environmental impacts, as they can be retrofitted to existing emission sources (Erto et al., 2015). Chemical scrubbing of CO2 in aqueous amine solutions (mainly monoethanolamine, MEA) represents the most widely investigated technology in the CCS field, but the process suffers drawbacks related to the considerable amounts of thermal energy required for absorbent regeneration, solvent degradation and equipment corrosion (Boot-Handford et al., 2014). In 2012, Det Norske Veritas (DNV) and Process Systems Enterprise Ltd. (PSE) developed a design concept for the application of CCS on-board ships based on CO2 chemical absorption into amine solutions (DNV-GL, 2013). Our research group is currently developing a CCS scheme for on-board naval installation based on the use of potassium carbonate that converts CO2 into potassium bicarbonate. The process can be operated at low temperature, about 50 °C, as a tail-end unit after a SO2 scrubbing system. The exhaust sorbent can be regenerated in the temperature range 100 - 200 °C (Zhao et al., 2013) to recover CO2 as an almost pure gas, DOI: 10.3303/CET1652070 Please cite this article as: Balsamo M., Erto A., Lancia A., Di Natale F., 2016, Carbon dioxide capture from model marine diesel engine exhaust by means of k2co3-based sorbents, Chemical Engineering Transactions, 52, 415-420 DOI:10.3303/CET1652070 415 stored as a liquid in cryogenic tank and disposed, or sold, at docks. The use of alkali metal-based sorbents (e.g. K2CO3) is considered a cost-effective and an energy-efficient technology for CO2 capture when compared with the conventional MEA process (Zhao et al., 2013). In this paper, we report experimental results on CO2 removal from a model marine diesel engine exhaust by means of K2CO3 both raw and supported onto porous alumina. CO2 carbonation tests were carried out in a fixed bed column integrated in a lab-scale unit in the temperature range 60-105 °C. The obtained results allowed a preliminary assessment of the potentiality of this purification technology for CO2 capture in the maritime sector. This research activity aims at the development of a CO2 capture unit to be integrated into depuration systems that we are developing to cut gaseous pollutants and particulate matter emissions deriving from marine diesel engine exhaust (Di Natale et al., 2013). 2. Materials and methods 2.1 Sorbents The sorbent material used in carbonation tests is a commercially available potassium carbonate (Carlo Erba Reagents). The raw K2CO3 (termed K2CO3-raw) was mechanically sieved in the range 300 - 630 m so to obtain sorbent particles ensuring low pressure drops (10-3 bar) across the fixed bed reactor. The as-received sorbent textural properties were obtained by N2 adsorption at -196 °C in a volumetric apparatus (Sorptomatic 1990). The porosimetric analysis highlighted a prevailing macroporous nature with a mean pore diameter (dpore) equal to 119 nm. Moreover, the total pore volume (Vp) and the surface area (SBET) are 0.17 cm3 g-1 and 5.9 m2 g-1, respectively. An alumina-supported K2CO3 sorbent (termed K2CO3-sup) was also synthesized to verify the effectiveness of dispersing nano-sized K2CO3 onto a large surface area substrate in the CO2 capture process. To this aim, a commercial γ-Al2O3 (1 mm diameter spheres, supplied by SASOL) was adopted as substrate. The nitrogen porosimetric analysis of the bare support revealed a mesoporous structure with dpore = 10 nm, Vp = 0.47 cm3 g-1 and SBET = 166 m2 g-1. A functionalized sorbent with 20 %wt. K2CO3 loading was prepared via incipient wetness impregnation: an aqueous solution of anhydrous potassium carbonate (active phase concentration and total solution volume equal to 0.625 g mL-1 and 4 mL, respectively) was added dropwise under stirring to 10 g of γ-Al2O3. Finally, the sorbent was dried in a fixed bed reactor at 115 °C under a N2 flow (0.45 L min-1, evaluated at 25 °C and 1 bar). 2.2 Carbonation tests and data analysis Figure 1 depicts a schematic representation of the lab-scale apparatus adopted for CO2 capture runs. Figure 1: Layout of the experimental apparatus 416 The carbonation reactor is a stainless steel fixed bed column (length = 11 cm; inner diameter = 2 cm), equipped with a 35 m porous septum. The fixed bed temperature was controlled by means of a heating system arranged coaxially with the carbonation unit. It is made up of two 320 W cylindrical shell band heaters (Watlow) enveloped in a thermal insulating layer of stone wool and connected to PID controller (i/16 series Omega). The fixed bed temperature was measured by means of a K-type thermocouple. Pure carbon dioxide was fed to the reactor by means of a mass flow controller (series El Flow Bronkhorst 201-CV), and mixed with a pre-humified nitrogen stream (N2 flow rate controlled by a flow meter) to generate a gas mixture of desired CO2 concentration. The humidity level of the feed gas was set by fluxing the N2 stream in a thermostatically- controlled water saturator set at 33 °C. Carbon dioxide concentration was measured by a continuous NDIR gas analyzer (AO2020 Uras 26 provided by ABB); data acquisition were performed by interfacing the gas analyzer with a PC via LabView™ software. A CaCl2 trap was placed at the fixed bed outlet to remove water prior to CO2 concentration measurements. Continuous carbonation tests were performed by feeding the column with a 1.2 L min-1 gas mixture with 5 % CO2, 5 % H2O and 90 % N2 to simulate the typical exhaust composition of a marine diesel engine (Environmental Protection Agency, 2000). The column was loaded with a known sorbent amount (20 g for K2CO3-raw and 12.5 g at 20 %wt. potassium carbonate loading for K2CO3-sup). The effect of operating temperature on the K2CO3-raw capture performances was investigated at 60, 75, 90 and 105 °C. A preliminary carbonation test for the supported sorbent was carried out at 60 °C, which resulted the best operating temperature (among those investigated) for CO2 capture, as obtained from tests conducted on the raw K2CO3. Dynamic carbonation tests allowed to calculate the amount of CO2 captured per unit mass of K2CO3 at any time t, (t) [mmol g-1], via the following material balance over the fixed bed reactor:            t 0 in CO out CO CO CO in CO dt Q (t)Q -1 mM ρQ ω(t) 2 2 2 22 (1) where: (t)Q out 2CO and in 2CO Q [L s-1] represent the CO2 volumetric flow rates at the bed outlet and inlet, respectively; m [g] is the K2CO3 mass loaded into the column; CO2 [mg L-1] represents the CO2 density (at 20°C and 1 bar); MCO2 is CO2 molecular weight [mg mmol-1]. If t=t*, where t* [s] is the saturation time for which the CO2 outlet concentration is approximately equal to 99 % of its inlet value, (t) coincides with the saturation capture capacity s. The carbonation kinetics was conveniently expressed in terms of time evolution of the K2CO3 conversion degree x(t) [-] corresponding to the molar fraction of active phase reacted with CO2. Considering a 1:1 stoichiometry for the reaction between CO2 and K2CO3 (Zhao et al., 2013), x(t) can be expressed as: 3CO2K Mtωtx )()(  (2) where MK2CO3 is the molecular weight of K2CO3 [g mmol-1]. The experimental value of x(t) corresponding to saturation conditions is named xmax. Kinetic differences in the CO2 capture process for the sorbents tested under different operating conditions were also interpreted in light of a simple pseudo-first order model (termed PFO) applied to x(t) vs t patterns:  )ktexp(1x)t(x max  (3) where k [s-1] is a pseudo-first order kinetic constant and it was obtained as best-fitting parameter via the least square method. Finally, an initial carbonation rate  [s-1] can be computed from the PFO model as: max 0t kx dt )t(dx    (4) 3. Results and discussion Table 1 summarizes the main CO2 capture data obtained from the experimental carbonation tests performed onto both raw (K2CO3-raw) and alumina-supported potassium carbonate (K2CO3-sup) together with the kinetic parameters derived from the PFO model applied to dynamic removal data. Figure 2(a)-(b) depicts the time dependence of the carbonation degree x(t) in the different operating conditions. 417 Table 1: Main parameters obtained from the carbonation process of K2CO3-raw and K2CO3-sup sorbents Sorbent Carbonation temperature [°C] s [mmol g-1] xmax [-] t* [s] k [s-1]  [s-1] K2CO3-raw 60 1.38 × 10-1 1.9 × 10-2 500 7.7 × 10-3 1.5 × 10-4 75 1.01 × 10-1 1.4 × 10-2 480 8.2 × 10-3 1.1 × 10-4 90 2.90 × 10-2 4.0 × 10-3 40 1.0 × 10-1 4.0 × 10-4 105 2.60 × 10-2 3.6 × 10-3 40 1.2 × 10-1 4.3 × 10-4 K2CO3-sup 60 3.1 4.3 × 10-1 900 3.5 × 10-3 1.5 × 10-3 time, s 0 100 200 300 400 500 600 x (t ), - 0.000 0.005 0.010 0.015 0.020 T=60 °C T=75 °C T=90 °C T=105 °C time, s 0 200 400 600 800 1000 1200 1400 x (t ), - 0.0 0.1 0.2 0.3 0.4 0.5 K2CO3-raw K2CO3-sup a) b) Figure 2: Time evolution of the carbonation degree x(t) for a) raw K2CO3 at different temperatures and b) raw and alumina-supported K2CO3 at 60 °C Data obtained for the raw K2CO3 testify the significant effect exerted by the process temperature on the sorbent capture. In fact, the experimental value of the carbonation degree obtained under saturation conditions xmax monotonically decreases as the process temperature is raised: at 60 °C xmax is about 5-times the values retrieved at 90 and 105 °C. This behavior can be related to the exothermic nature of the carbonation process for the tested sorbent (Zhao et al., 2013). On contrary, the dynamic carbonation profiles obtained for K2CO3-raw generally highlight a faster CO2 capture process when the temperature increases. Tests carried out at 60 and 75 °C show that saturation conditions are attained in times shorter than 500 s, whereas a plateau value of the carbonation degree is reached for t<50 s for higher temperatures (Figure 2 (a)). These observations are also corroborated by the kinetic parameters retrieved from the PFO model (Table 1). Contextually, the pseudo-first order kinetic constant k derived at 105 °C is approximately 16- and 15-times greater than the values obtained at 60 and 75 °C, respectively. Similarly, the initial adsorption rate both at 90 and 105 °C increases about 3-times with respect to the values obtained at lower temperatures. Coherently with experimentally observed x(t) patterns, negligible differences in k and values can be inferred for tests performed at 90 and 105 °C. Faster carbonation kinetics at higher temperatures can be likely related both to a quicker diffusion process of CO2 molecules in the pore network of K2CO3 and to the already described decrease of xmax (smaller number of reactive sites to be converted) (Ruthven, 1984). As a general comment, carbonation data obtained for the raw potassium carbonate highlight a poor exploitation of the active phase, with a maximum conversion degree equal to nearly 2 % in the best case (i.e. at 60 °C). The experimental evidences suggest that K2CO3 characterized by large particle sizes experiences strong diffusion limitations during the conversion process: the formation of a less porous product layer (KHCO3) hinders the CO2 molecules to access to the potassium carbonate active sites (Guo et al, 2015). Therefore, a large particle volume fraction is left as an unreacted core of potassium carbonate, which results in a very low carbonation degree. In turn, this makes unpractical the use of raw K2CO3 for reducing CO2 emissions on-board ships, because large-volume reactors would be required with associated problems of space allocation and high additional fuel consumption. Consequently, we investigated the use of K2CO3 supported onto a large surface area alumina to overcome the aforementioned drawbacks. The sorbent obtained at 20 %wt. loading of the active phase was tested under the same feed composition adopted for K2CO3-raw at 60 °C, the latter chosen as the best T-level on the basis of the previously analyzed carbonation data. 418 A comparison of xmax and s data for K2CO3-raw and K2CO3-sup (see Table 1) remarks the effectiveness of the dispersion of K2CO3 on alumina, determining a significant increase in the CO2 capture capacity. In fact, s for the supported sorbent is 3.1 mmol g-1, a value about 23-times greater than the one obtained for the unsupported sorbent. Moreover, a maximum conversion degree of 0.43 is reached. It should be highlighted that the contribution of the raw substrate in the CO2 capture process was practically negligible under the tested experimental conditions (5×10-3 mmol g-1) with respect to the K2CO3-sup capture capacity. The dynamic CO2 capture data reported for K2CO3-raw and K2CO3-sup (cf. Figure 2 (b)) show that the supported sorbent requires a longer time to reach saturation conditions with respect to the parent K2CO3 (t*900 and 500 s for K2CO3-sup and K2CO3-raw, respectively). This is mirrored by the k value derived for K2CO3-sup that is almost halved when compared to the value obtained for K2CO3-raw. The generally slower capture process observed for the supported sorbent, mainly in the late stages of the process, could be related to a greater number of active sites of potassium carbonate to be converted into potassium bicarbonate. On the other hand, in the early stages of the process a higher number of reactive centers for K2CO3-sup with respect to K2CO3-raw is likely to determine a higher driving force for the carbonation reaction, thus producing a faster capture rate (cf. the initial slopes of x(t) curves in Figure 2 (b)). This pattern is also witnessed by the initial carbonation rate , being one order of magnitude greater in the case of K2CO3-sup with respect to K2CO3-raw. 4. Conclusions This work belongs to a wider research framework aimed at developing high-efficiency depuration processes of marine diesel engine exhaust, which includes the abatement of NOX, SO2 and PM (Di Natale et al., 2015). The depuration train developing at our research facilities includes a final stage for reducing CO2 emissions, based on the use of potassium carbonate as a selective reactant. In this paper, we reported the results of experiments on CO2 capture from a mimicking marine diesel engine effluent on both raw and alumina-supported potassium carbonate. Carbonation tests for the raw K2CO3 indicated a higher reactivity of the sorbent at 60 °C, due to the exothermic nature of the carbonation process. On the other hand, a low maximum conversion degree was obtained (2 %), likely due to strong diffusion limitations occurring for coarse particles, thus making unsupported K2CO3 not suitable for CO2 emissions reduction in the maritime sector. The dispersion of the active phase onto a large surface area and porous alumina support allowed to significantly enhance the sorbent exploitation, with a maximum conversion percentage equal to 43 %. The preliminary results of CO2 capture performances obtained for the supported sorbent are very encouraging in the light of on-board ships installation of a carbonation reactor, based on this class of sorbents, after a SO2 scrubbing unit. Future research efforts should be devoted at investigating different classes of substrates and functionalization conditions (i.e. active phase loading) allowing a maximization of the reactant conversion degree, also to reduce the carbonator size (at fixed removal efficiency), the latter being a critical aspect for on-board integration of this purification unit. Regeneration studies in steam will be also performed, in order to ensure a proper re-usability of the sorbent under multiple carbonation-decarbonation cycles and for recovering almost pure carbon dioxide for storage purposes. References Boot-Handford M.E., Abanades J.C., Anthony E.J., Blunt M.J., Brandani S., Mac Dowell N., Fernández J.R., Ferrari M.C., Gross R., Hallett J.P., Haszeldine R.S., Heptonstall P., Lyngfelt A., Makuch Z., Mangano E., Porter R.T.J., Pourkashanian M., Rochelle G.T., Shah N., Yao J.G., Fennell P.S., 2014, Carbon capture and storage update, Energy Environ. Sci. 7, 130-189. Di Natale F., Carotenuto C., D’Addio L., Lancia A., Antes T., Szudyga M., Jaworek A., Gregory D., Jackson M., Volpe P., Beleca R., Manivannan N., Abbod M., Balachandran W., 2013, New technologies for marine diesel engine emission control, Chem. Eng. Trans. 32, 361-366. Di Natale F., Carotenuto C., 2015, Particulate matter in marine diesel engines exhausts: emissions and control strategies. Transport. Res. Part D-Transport., 40, 166-191. Di Natale F., Carotenuto C., D’Addio L., Jaworek A., Krupa A., Szudyga M., Lancia A., 2015, Capture of fine and ultrafine particles in a wet electrostatic scrubber, J. Environ. Chem. Eng. 3, 349-356. DNV-GL, 2013, DNV and PSE report on ship carbon capture and storage accessed 05.04.2016. Environmental Protection Agency, 2000, Analysis of commercial marine vessels emissions and fuel consumption data1 accessed 05.04.2016. 419 Erto A., Silvestre-Albero A., Silvestre-Albero J., Rodriguez-Reinoso F., Balsamo M., Lancia A., Montagnaro F., 2015, Carbon-supported ionic liquids as innovative adsorbents for CO2 separation from synthetic flue-gas, J. Coll. Interf. Sci. 448, 41-50. European Commission, 2011, White paper, roadmap to a single European transport area-towards a competitive and resource efficient transport system accessed 05.04.2016. Figueroa J.D., Fout T., Plasynski S., McIlvried H., Srivastava R.D., 2008, Advances in CO2 capture technology-the U.S. Department of Energy's Carbon Sequestration Program, Int. J. Greenhouse Gas Control 2, 9-20. Guo Y., Zhao C., Li C., Wu Y., 2015, CO2 sorption and reaction kinetic performance of K2CO3/AC in low temperature and CO2 concentration, Chem. Eng. J. 260, 596-604. International Energy Agency, 2014, CO2 emissions from fuel combustion accessed 05.04.2016. International Maritime Organization, 2014, Reduction of GHG emissions from ships-third IMO GHG study 2014-final report MEPC 67th session accessed 05.04.2016. Ruthven D.M., Ed., 1984, Principles of Adsorption and Adsorption Processes. John Wiley & Sons, New York, United States. Zhao C., Chen X., Anthony E.J., Jiang X., Duan L., Wu Y., Dong W., Zhao C., 2013, Capturing CO2 in flue gas from fossil fuel-fired power plants using dry regenerable alkali metal-based sorbent, Prog. Energy Combust. Sci. 39, 515-534. 420