An Investigation On The Solvent Based Carbon Capt
An Investigation On The Solvent Based Carbon Capt
An Investigation On The Solvent Based Carbon Capt
and Comparisons with Another Carbon Control Methods for Different Ships
OrcID: 0000-0002-1553-4553
2 Istanbul Technical University, Faculty of Naval Architecture and Ocean Engineering, Maslak, 34469
Istanbul, Turkey. E-mail: ergin@itu.edu.tr
OrcID: 0000-0001-8343-2455
* Corresponding author
1
1 Abstract
2 This study investigates the effects of hydraulic design parameters of separation columns on the
3 effectiveness of the solvent-based carbon capture and storage (CCS) systems and performs the cost
4 analyses of CCS systems for different ships. The CCS and power systems of ships are simulated by
5 using process modeling with Aspen HYSYS. A very large crude carrier (VLCC) tanker and three
6 different sizes of liquefied natural gas (LNG) carriers (Q-Max, Q-Flex, and conventional LNG carrier)
7 are considered. The study also employs different carbon control methods such as speed reduction and
8 LNG usage. These methods are compared in terms of their CO2 emission reduction effectiveness and
9 costs. Comparisons between the speed reduction, LNG usage, and the CCS system show that the speed
10 reduction is the best cost-oriented and easiest solution for ships with low freight values like crude oil
11 tankers. However, for ships with high speed and freight values like Q-Max and Q-Flex LNG carriers, it
12 is shown that the CCS is more cost-effective than other compared CO2 control methods. The results
13 show that the CCS system is a promising and feasible method for reducing CO2 emissions of ships.
14
15 Keywords: Carbon control methods; CO2 emissions; Ship emissions; Solvent-based carbon capture
17
2
18 1. Introduction
19 The International Maritime Organization (IMO) has dramatically accelerated its work since 2008
20 and has adopted a set of mandatory rules and regulations to limit the CO2 emissions from international
21 shipping. As a result of the 72nd Marine Environmental Protection Committee meeting held in London
22 in April 2018, it was accepted that CO2 emissions by 2050 would be reduced by 50% compared to 2008
24 The decisions taken to reduce CO2 emissions are favorable for the environment and world climates.
25 Still, they have negative consequences for maritime and world trade due to the economic obligations it
26 imposes. Methods like using LNG or hydrogen as a fuel, using fuel cells and renewable energy sources
27 can be employed to reduce CO2 emissions. However, these methods are not low-cost solutions (Trivyza
28 et al., 2019). On the other hand, methods such as increasing energy efficiency, improving hull and form
29 design, improving power, and propulsion systems are economic, but they have a limited impact on
30 reducing emissions (Buhaug et al., 2009). None of these methods except the speed reduction and using
31 hydrogen as a fuel can meet the IMO's 2050 emission target alone.
32 Carbon capture and storage (CCS) system is one of the methods to reduce CO2 emissions. According
33 to the Intergovernmental Panel on Climate Change (IPCC), CCS involves taking CO2 from industry and
34 energy sources and then transporting it to the storage area. Thus, CO2 is removed from the atmosphere
35 in the long term (IPCC, 2001). The system's main advantage is providing between 85–95% capture rate
36 in practice for plants onshore. The most promising technology for the marine CCS system is the solvent-
37 based post-combustion capture according to many studies (Rochelle, 2009; Wang et al., 2011; Boot-
38 Handford et al., 2014; Jinyue, 2015). Because this system has been commercialized and very well known
39 for a long time and it is an “end-of-pipe” technology. It should be noted that the precombustion capture
40 system is not applicable for use in ships and the oxy-combustion system may be expensive nowadays.
41 For marine applications, different studies show that up to 60–80% capture rate is more feasible (Luo
43 The changes in the oil and freight prices are the critical factors to determine the cruising speed which
44 is also an important parameter for the high profitability of ships. When the operation and design speed
3
45 reduce, the fuel and operating costs will be lower, but the revenue also will be lower (Ronen, 1982).
46 When the oil prices rise or ship freight costs decrease, reduced ship speed increases profitability and
47 decreases emissions. For example, Maersk Tankers reduce their cruising speeds from 16 knots to less
48 than 10 knots and their ballast-laden speeds to around 11-13 knots. A tanker which only reduced the
49 ballast-laden speed from 16 knots to 8.5 knots increased the expedition time from 42 days to 55 days
50 but earned $ 400,000 in fuel, which operates between Basra Gulf to South Asia (“Slow spur for Maersk
51 VLCCs,” 2010). These results show that the speed reduction could be used to reduce CO2 emissions.
52 According to the literature one of the first studies for the investigation of the application of CCS
53 systems is Zhou and Wang's study, which examined the solidification and storage of CO2 in ships (Zhou
54 and Wang, 2014). In their later study, numerical analysis of computational fluid dynamics results and
55 experimental results were discussed (Wang et al., 2016). The first systematic analysis of solvent-based
56 CCS systems on ships belongs to Luo and Wang's study (Luo and Wang, 2017). In 2017, Iliuta and
57 Larachi studied the enzymatic CO2 hydration process for oscillating structured packed-bed columns
58 (2017). Then, in 2019, Awoyomi et al. designed a cycle in which CO2 and sulfur oxide were captured
59 together in the CCS system on ships. In this study, ammonia was used to capture these emissions, the
60 ammonia concentration, and the different engine loads were investigated (Awoyomi et al., 2019). In the
61 same year, Feenstra et al. studied two different engines and two different solvents for a vessel, using
62 two different fuels; diesel and LNG (Feenstra et al., 2019). In 2018, Trivyza et al. worked on the
63 selection and optimization of energy, and emission reduction methods such as using LNG, diesel fuel,
64 exhaust gas recirculation, selective catalytic reduction, scrubber, fuel cell applications, and CCS system
65 for an Aframax tanker by taking into account the environmental conditions. In this study, the non-
66 dominated sorting genetic algorithm (NSGA) - II was used as an optimization technique (Trivyza et al.,
67 2018). In their next study, published in 2019, they examined the life cycle cost (LCC) based on four
68 different carbon tax policies for a cruise ship. With Pareto analysis, it is concluded that the best solution
69 to achieve IMO's 2050 target is using LNG as a fuel, WHR system, and CCS system together (Trivyza
70 et al., 2019).
71 In this study, the application of solvent-based CCS systems to different sizes and types of ships will
72 be examined. It is aimed to investigate the effects of hydraulic design parameters of separation columns
4
73 on the effectiveness of the CCS systems for ships, to reduce the energy efficiency design index (EEDI)
74 or energy efficiency existing ship index (EEXI) value by 50% compared to the energy efficiency design
75 index (EEDI) phase 0. Besides, it is aimed to determine the most feasible CO2 control method for
76 different types and sizes of ships by comparing the CCS system and other CO2 control methods in terms
78 For this purpose, the solvent-based CCS and power systems of a very large crude carrier (VLCC)
79 tanker and different sizes of LNG carriers (Q-Max, Q-Flex, conventional LNG carrier) are simulated by
80 using process modeling with Aspen HYSYS. The waste heat recovery systems (WHR) of the ships are
81 also modeled. Then, the CCS systems’ costs as per CO2 capture are examined for different dimensional
82 parameters that affected the hydraulic designs of the systems. The costs of the CCS system are
83 investigated by considering MEA makeup, freight loss, WHR loss, liquefaction equipment, liquefaction
84 power, capture equipment, capture power, and heat costs for the VLCC tanker. After then, the costs of
85 the solvent-based CCS systems are compared with the costs of the LNG usage and the speed reduction
87 Boil-off gas occurs in cargo tanks of LNG carrier ships if the temperature rises above the designed
88 LNG storage temperature. This boiling gas must be removed from the tanks. New types of gas injection
89 main engines use this boil-off gas as fuel and this situation is considered in this study.
90 The storage time of CO2 on ships depends on the route of the ship. Large size ships, such as VLCC
91 tankers, cannot pass from some canals like the Suez-canal in their routes due to canal size limitations.
92 Thus, the route becomes longer, and the storage time of carbon dioxide on ships increases. This
94 The effects of separation columns sizing and hydraulic design parameters on the carbon capture rate
95 and carbon capture and storage cost have not been previously examined for ships. In addition, there are
96 no studies comparing the costs of CCS systems with the speed reduction method which is recently used
97 to reduce the CO2 emissions for ships. Trivyza et al. compared the LNG usage and CCS system for ships
98 (Trivyza et al., 2019). However, they do not consider the LNG prices at ports correctly. According to
99 the Danish Maritime Authority (DMA), the price of the LNG at ports should be higher than the import
5
100 price which is used in Trivyza et al. study (Danish Maritime Authority, 2012). This study considers the
102 The IMO has a strategy that the total annual GHG emissions from international shipping should be
103 reduced by at least 50% by 2050 compared to 2008. In this study, different from the other studies, the
104 CCS system is modeled to satisfy the IMO 2050 emission criteria rather than to maximize the captured
105 CO2 emissions. As a result of this, the total life cycle cost of the CCS system is reduced and cost-
106 effective solutions for ships are obtained. Furthermore, this makes it possible to compare the CCS
107 system with other CO2 reduction methods such as speed reduction and LNG usage.
108 When the electrical power of the ship is produced by the generators and waste heat recovery systems,
109 the required energy for the CCS system is generally supplied by the waste heat recovery system. In this
110 case, extra power is required for the ship's electrical power requirements. This results in an extra cost
111 and it is considered as an indirect cost. This cost is named as the WHR loss cost throughout this study.
112 A comprehensive cost analysis of the marine CCS systems different than the previous studies has
113 been carried out in this study (Feenstra et al., 2019; Luo and Wang, 2017; Trivyza et al., 2018; Zhou
114 and Wang, 2014). The costs are investigated as solvent supply cost, freight loss cost, power and heat
115 loss costs for capture and liquefaction, the indirect cost of using waste heat for CO2 regeneration, and
117
120 There are some challenges in using the CCS systems for ships. One of the biggest challenges is
121 limited space on the ship. On land facilities, the absorber and stripper columns heights of the CCS
122 systems could be up to 20–30 meters (Luo and Wang, 2016; Mores et al., 2014; Agbonghae et al., 2014).
123 On the other hand, these heights would be shorter onboard because the CO2 emission rate is less
124 compared to the land facilities. However, the height of the columns can still be critical for ships due to
125 the stability requirements. Besides, this system contains many pieces of equipment. The CCS system is
126 also heavy, and it causes freight loss costs. Det Norske Veritas proposes using the CCS systems on
127 VLCC tankers because of their size and relative abundance of deck space (Ship Technology, 2013).
6
128 Therefore, VLCC tanker and other big size ships Q-Max, Q-Flex, and conventional LNG carriers are
130 The main dimensions of ships and their power requirements are defined using MAN’s study on the
131 selection of propulsion systems for LNG carrier ships and tankers (MAN Diesel & Turbo, 2014; MAN
132 Diesel & Turbo, 2013). Table 1 shows the main dimensions and properties of ships.
133
135 The power systems of the ships are modeled by using Aspen HYSYS. The main part of the model
136 development of a marine diesel engine is the modeling of the thermal processes that occur inside the
137 cylinders of the diesel engine. A typical diesel cycle is employed for the thermal processes, including
138 adiabatic compression, fuel injection, combustion, and power stroke. The thermodynamic properties of
139 exhaust gas are predicted by the Peng-Robinson equation (Ma et al., 2015). The compressors and
140 turbines are assumed to be isentropic. The combustion room is assumed as adiabatic and Gibbs free
142
Large
Description VLCC conventional Q-Flex Q-Max
LNG carrier
Scantling ship size (DWT for
300000 150000 210000 265000
VLCC/m3 for LNG Carriers):
Scantling draught (m): 22 12.3 12.7 12.7
Length overall (m): 333 288 315 345
Length between perpendicular (m): 320 275 303 332
Breadth (m): 58 44.2 50 54
Design draught (m): 19 11.6 12 12
Design ship speed (kn): 15.5 20 20 20
Estimated maximum power (kW): 25900 31400 39300 45200
144
145
146 According to the estimated maximum power, as given in Table 1, the MAN Ceas program has been
147 used to select the engine of each ship (“Man Energy Solutions,” 2018). The engine characteristics and
7
148 specifications including International Organization for Standardization (ISO) specific fuel oil
149 consumption (SFOC), specific pilot oil consumption (SPOC), and specific gas consumption (SGC) are
150 shown in Table 2. For the VLCC tanker, the lower heating value (LHV) of the fuel is given for the
151 marine diesel oil (MDO) by the engine manufacturer. However, the heavy fuel oil (HFO) 380 is used as
152 a fuel for the VLCC tanker because of its low cost. The fuel was initially modeled as MDO with an LHV
153 of 42700 kJ/kg to validate the specifications of the engine in the power system model of the VLCC
154 tanker. Then, the fuel is changed to HFO 380 with an LHV of 40200 kJ/kg. It is assumed that the fuel
155 contains sulfur 3.5% by weight. It is accepted that exhaust pressure loss is 3 kPa at scrubber and
156 pipeline. The main engine and equipment are modeled as in Figure 1.
157 In the model, the combustion air at atmospheric pressure is pressurized in the compressor and passed
158 through a two-stage cooler. The process water is used in the first stage of the cooler. For the second
159 stage, the seawater is used as a coolant. The cooled air enters the cylinder and is re-pressurized by the
160 piston, then reacts with the fuel injected from the nozzle in the combustion chamber. The exhaust gas
161 generates work after the combustion process and exits from the manifold at 435 °C for the VLCC tanker.
162 Then, the exhaust gas enters the Exhaust Gas Boiler (EGB). The exit temperature of the exhaust gas
163 from the EGB is 180 °C, and the process water is used in the EGB as a cooler again.
164
Large
Description VLCC Q-Flex Q-Max
conventional
8S70ME- 9S50ME- 7G60ME- 8G60ME-
Main engine
C10.5 C9.7 C10.5 C10.5
Total main engine power at 100% MCR
27440 2x16020 2x19880 2x22720
(kW)
NOx emission standards Tier III Tier III Tier III Tier III
8
Outlet pressure of air compressor at 75%
3.33 3.03 3.07 3.45
MCR (bar)
168 Chemical solvents are used to absorb CO2 from flue gas for the solvent-based carbon capture
169 process. Although MEA requires high enthalpy of reaction, has thermal degradation and corrosive
170 effects, it provides favored kinetics and high chemical reactivity with CO2. Besides, MEA is a well-
171 known solvent and its cost is low. Because of these properties, an aqueous solution containing 35%
172 MEA by weight is used as the solvent in this study (Conway et al., 2015; Gupta et al., 2013). When the
173 absorption of CO2 in the MEA solution is examined, the following reversible reactions occur:
𝐾𝑐𝑎𝑟𝑏 (1)
𝑀𝐸𝐴𝐶𝑂𝑂− + 𝐻2 𝑂 ↔ 𝑀𝐸𝐴 + 𝐻𝐶𝑂3−
𝐾𝑎,𝐶𝑂2 (2)
𝐶𝑂2 + 2𝐻2 𝑂 ↔ 𝐻𝐶𝑂3− + 𝐻3 𝑂+
𝐾𝑎,𝐻𝐶𝑂− (3)
3
𝐻𝐶𝑂3− + 𝐻2 𝑂 ↔ 𝐶𝑂3−− + 𝐻3 𝑂+
𝐾𝑎,𝑀𝐸𝐴𝐻+ (4)
𝑀𝐸𝐴𝐻+ + 𝐻2 𝑂 ↔ 𝑀𝐸𝐴 + 𝐻3 𝑂+
𝐾𝑤 (5)
2𝐻2 𝑂 ↔ 𝐻3 𝑂+ + 𝑂𝐻 −
174
9
175
177
178 The exhaust gas-cooled in the EGB is transferred to the carbon capture system, which is the primary
179 concern of this study. The solvent-based CCS system is modeled by using Aspen HYSYS and it is
180 presented in Figure 2. The exhaust gas passing through the EGB and SOx scrubber is cooled by a further
181 stage in the absorber cooler without entering the absorption column. The seawater cooling is used here.
182 In accordance with the Bureau Veritas rules, the seawater pump inlet temperature is modeled as 32 °C
183 and the seawater's maximum overboard temperature is 42 °C (Veritas, 2014). The cooled exhaust gas
184 needs to be pressurized before entering the absorption column because it loses pressure in the column.
185 Therefore, an exhaust fan is added to the model, and pressurization is performed according to the need.
186 The exhaust gas and the solvent enter the absorber column where the carbon capture is taken place.
10
187
189
190 After the capture, the carbon-rich solution passes from the absorption column is first pressurized by
191 the pump and then heated through the heat exchanger. The heating of the solution is crucial because it
192 reduces the stripping column's reboiler's heat requirement. The heated carbon-rich solution is reheated
193 to the beginning of boiling, and the CO2 is separated from the MEA solvent in the stripping column
194 reboiler at the bottom. However, it is not the only CO2 that evaporates, some amount of water vapor also
195 evaporates along with CO2 in the stripping column. The evaporated water is liquefied by the separating
196 column's cooler and it is recycled. In the cooler of the column at the top, the CO2 vapor mixture is cooled
197 by process water. The heat rejected from the process water is transferred to the reboiler of the stripper
198 column. There is an extra heater which is used if needed. During the whole process, all the heat
199 exchangers except the reboiler and condenser of the stripper column are modeled with a pinch point of
200 at least 10 degrees Celsius. For the reboiler and condenser of the stripper column, the pinch point
201 temperatures are reduced to 5 degrees Celsius. The lean carbon solution remaining in the boiler is
202 pressurized with a pump and gives a large amount of its heat to the heat exchanger for the carbon-rich
203 solution. The lean carbon mixture is then further cooled by using the process water to obtain the
204 temperature at which it enters the absorption column. The water and MEA which are evaporated at the
11
205 exhaust gas outlet of the absorber column are recovered and recycled. The stripped CO2 is liquefied by
207 The equilibrium based method is applied for the calculation of the mass transfer in the absorber and
208 stripper columns. The plate efficiencies are obtained by using the American Institute of Chemical
209 Engineers (AIChE) method (Sinnott, 2014). The Scheffe method is used to obtain the interface area
210 within the columns (Scheffe and Weiland, 1987). The HYSIM inside-outside method is used for the
211 solution of the MESH (mass, equilibrium, summary, heat) equations (Øi, 2007). The solution of this set
212 of equations forms the basis of the solution methods for each stage of absorber and stripper column
213 operations. Flows entering and leaving the stage are schematized in Figure 3. The modified MESH and
214 additional equations for the inside-out method algorithm are given as follows:
𝑆𝑛 = 𝑊𝑛 + 𝑈𝑛 (10)
𝑙𝑖,𝑛−1 − (𝑅𝐿𝑗 + 𝛼𝑖,𝑛 𝑆𝑡𝑏,𝑛 𝑅𝑉𝑛 )𝑙𝑖,𝑛 + 𝛼𝑖,𝑛+1 𝑆𝑡𝑏,𝑛+1 𝑙𝑖,𝑛+1 = −𝑓𝑖,𝑛 (14)
𝐻𝑛 = ℎ𝐿,𝑛 𝑅𝐿,𝑛 𝐿𝑛 + ℎ𝑉,𝑛 𝑅𝑉,𝑛 𝑉𝑛 − ℎ𝐿,𝑛−1 𝐿𝑛−1 − ℎ𝑉,𝑛+1 𝐿𝑛+1 − ℎ𝐹,𝑛 𝐹𝑛 (15)
− 𝑞𝑛 = 0
12
215 Where α represents the relative volatility, index b is the non-volatile hypothetical reference component,
216 index i is any component, index n is any stage, K is the liquid-vapor phase equilibrium constant, St is
217 the stripping factor, V is the total vapor amount, L is the total liquid amount, R L is the liquid-phase
218 withdrawal factor, R V is the vapor-phase withdrawal factor, Un is the liquid-phase withdrawal amount,
219 Wn is the vapor-phase withdrawal amount, Sn is the total withdrawal amount, x is the liquid-phase mole
220 fraction, y is the vapor-phase mole fraction, li is the liquid amount of component i, vi is the vapor amount
221 of component i, F is the total feed amount, fi is the feed amount of component i, H is the energy and h
222 is the enthalpy. Equations (12) - (15) represent mass balance, modified phase equilibrium, modified
223 component material balance, and modified energy balance. For estimating the thermodynamic properties
224 of the liquid phase, the electrolyte non-random two-liquid thermodynamic model and for the gas phase,
226
227
230 For validation purposes of the carbon capture system, a similar model of the carbon capture system
231 from the literature is modeled and rich loading (mol CO2/mol MEA) performances are compared (Luo
13
232 and Wang, 2017). The absolute percentage error is 0.08%. It can be said that the results are in good
233 agreement.
234 In this study, a model of the closed two-stage liquefaction system is used to store the CO2 as shown
235 in Figure 4. The working pressure and temperature of the system are assumed as 17.5 bar and -24 °C,
236 respectively (Seo et al., 2016). According to Figure 4, the CO2 from the stripper column is pressurized
237 in two stages and intercooled. Then, it is separated from the water and other contaminations. Finally,
238 the CO2 is cooled to the storage temperature by using the two-stage closed cooling system with
239 intercooling. The NH3 is used as a refrigerant in this system. Before entering the first Joule-Thomson
240 (JT) valve, the NH3 is at 16.5 bar pressure and 42 °C temperature. After passing the JT valve, the
241 evaporated NH3 is recycled and pressurized. Finally, the NH3 enters the main cooler at a temperature of
242 -29.4 °C and a pressure of 121.3 kPa. The pressure loss of the NH3 in the main cooler is assumed to be
243 20 kPa. The pinch point temperature of the main cooler is 5.4 °C.
244
245
247
249 On the specified CCS system model, the equipment with installation cost (EWIC) and the total
250 weight of the equipment are calculated using the Aspen HYSYS. The other capital costs (OCC) which
251 include engineering services, construction management, commissioning, soft costs for contractors, and
14
252 owner costs are estimated as 45.2% of the equipment with installation cost (United States Energy
253 Information Administration, 2016). The annualized capital cost (ACAPEX) is calculated by multiplying
254 capital cost with the capital recovery factor (CRF). For the capital recovery factor, it is assumed that the
255 ships’ life is 30 years (n) and 8% of interest rate (i) without inflation. The inflation effect is ignored. The
256 system's cost as per captured CO2 (CPCC) is a key parameter of economic evaluation. The CPCC is
257 calculated as dividing annual life cycle cost (ALCC) by CO2 captured annually (ACCO2). The ALCC
258 is obtained by summation of ACAPEX, annual operational expenditure (OPEX), and annual waste heat
261 fan, solution pumps, condenser water pumps, process water pumps, water supply pumps, MEA supply
262 pumps, seawater cooling pumps, and CO2 and NH3 compressors. In some cases, an extra heater is
263 required. Because waste heat may not be sufficient to achieve the desired level of carbon dioxide
264 regeneration. During the absorption process, some of the MEA evaporates (MMEA) and flows through
265 the funnel to the atmosphere. Therefore, the costs of the extra heater and make-up MEA (AMEA) are
267 The CCS system in a ship causes a significant weight increase, so freight rate and revenue of ships
268 decrease. Therefore, this cost is added to the calculation as a freight loss cost. The freight loss cost
269 (AFLC) is calculated through multiplying the daily freighting rate (DFR) of the ship by total extra weight
270 (TEW) and dividing it by the ship capacity (DWT). The total extra weight (TEW) is calculated by
271 subtracting the fuel weight from the sum of the equipment’s weight (EQW), MEA make-up weight, and
272 stored maximum CO2 weight. Besides, the power requirement of the capture and liquefaction equipment
15
273 causes extra fuel consumption and cost (AFUEL). This cost is calculated by multiplying SFOC/SPOC
274 with main engine power, and annual working hours. As a result, the MEA make-up cost, freight loss
275 cost, capture, and liquefaction equipment power requirement cost constitute OPEX.
276 For the calculation of the OPEX, the annual working hours (AWH) of the main engines should be
277 estimated. The annual working hours of the main engines are assumed as 7500 hours (Euronav, 2013;
278 Fevre, 2018). The thermal efficiency for the extra heater (boiler) is assumed to be 0.9 (Rackley, 2017).
279 The HFO 380 fuel price is taken to be 265 $ per ton according to Rotterdam/Antwerp ports in September
280 2019 (Petrol Bunkering, 2019). The MEA price is taken as 1325 $ per ton (Intratec Solutions, 2019).
281 The average daily freight rate of VLCC tankers that operates between 1996 and 2018, and the total
282 contract price of VLCC tankers are presented in Figure 5. According to Figure 5, the average daily
283 freight rate from 1996 to 2018 is obtained as about 38000 $/day for the VLCC tanker.
284 The filling capacity of the CO2 tank is determined by considering the distances between the ports
285 on the important crude oil trade routes and the Antwerp Port of Belgium based on the speed of 12 knots
286 for the VLCC tanker. According to this data, it takes 931 hours when passing through Cape of Good
287 Hope to Ras Tanura in Saudi Arabia, 980 hours passing through Cape of Good Hope to Singapore, 366
288 hours from Bonny in Nigeria, and 418 hours from Texas in the United States (“Sea-distances.org,”
289 2019). Therefore, the CO2 storage capacity (CO2SC) is assumed as 1000 hours for the VLCC tanker.
16
80 18
70 16
60 14
12
50
USDk/day
billion $
10
40
8
30
6
20
4
10 2
0 0
1996 2000 2004 2008 2012 2016
Years
Total contract price Daily average freight rate
292
293 Figure 5: Daily average freight rate and the total contract price of VLCC tankers (Clarkson Research
296 For the calculation of OPEX for LNG carriers, the relation between the LNG market and LNG
297 carriers should be known. The development of the LNG market indicates a significant change in the
298 perspective regarding the nature of the global gas trade. Table 3 shows the capacity statistics of the
299 average LNG carrier vessels over a period of time. One of the biggest problems in LNG transportation
300 is the loss of gas by evaporation. This problem causes freight loss and extra cost. Vapor loss rates at
301 16.5 knots cruising speed are about 0.15% for a steam turbine on a ship with a carrying capacity of
302 145000 m3 LNG, 0.10% for a three-fuel diesel-electric (TFDE) engine on a ship with a carrying capacity
303 of 160000 m3 LNG, and 0.06% for a MAN gas injected engine (ME-GI) on a ship with carrying capacity
304 of 174000 m3 LNG per day. In order to prevent this situation, the use of evaporating gas has gained
305 importance over time. With increasing energy efficiency and the use of evaporating gas, the maximum
306 economical speed has increased from 14.5 knots to 19.5 knots over time. The current LNG tanker fleet
307 with newly ordered ships, according to the propulsion systems, is shown in Figure 6. In line with these
308 data, it is assumed that the ME-GI propulsion system is used in this study for LNG carriers to examine
309 the costs of the methods used to reduce CO2 emissions. The vapor loss rate of this propulsion system is
310 taken into account. In Figure 7, the average daily freight rates for the 160000 m3 capacity LNG carrier
311 are shown in years. According to the figure, the daily average freight rate is about 74500 $ for the 160000
17
312 m3 capacity TFDE. If the cost/volume of the LNG is assumed as constant at about 0.466 $/m3, the daily
313 average freight rate will be about 70000 $/day for the 150000 m3 LNG carrier, 98000 $/day for the
314 210000 m3 Q-Flex LNG carrier, and 125000 $/day for the 266000 m3 Q-Max LNG carrier.
315
316 Table 3: Predominant propulsion type and the average capacity of LNG carriers in operation (Novotony,
317 2019).
60
50
Number of builded ships
40
30
20
10
0
<1990 1994 1998 2002 2006 2010 2014 2018
Years
320 Figure 6: LNG fleet by propulsion type in years (Dynagas LNG Partners LP, 2018).
321
18
200
180
160
140
USDk/day
120
100
80
60
40
20
0
2012 2013 2014 2015 2016 2017 2018 2019
Years
322
323 Figure 7: The daily average freight rates of 160,000 m3 TFDE LNG carriers from 2012 to 2018 (Dynagas
325
326 LNG price is assumed as the export price of 220 $/ton because of boiling gas usage. Boiling gas of
327 LNG from cargo is cheaper than the port price since it does not includes transportation, port fee, storage,
328 and bunkering expenses. Besides, as mentioned earlier, VLCC tankers have to operate around the Cape
329 of Good Hope instead of passing through the Suez Canal due to draft limitations of the Canal in some
330 trade routes. However, investigated LNG carriers can pass through the Suez Canal. So, the storage time
331 of the CO2 is reduced considerably. In addition, the speed of LNG carriers is higher than the speed of
332 VLCC tankers, therefore the CO2 storage time is reduced further. As a result, the storage time is taken
334 If the waste heat is not used for the stripper column, it would be possible to generate energy by using
335 the WHR system with a steam turbine. Therefore, the cost resulting from the absence of steam turbine
336 should also be considered. The equipment and installation costs of the steam turbine are considered to
337 be 103 $/kW (Livanos et al., 2014). Like the CCS system, the interest rate is also considered for the
338 steam turbine capital cost estimation. If the waste heat is used to generate power, the WHR system with
339 the steam turbine can be modeled as Figure 8 to determine the provided amount of power. According to
340 the model, the steam temperature entering the steam turbine is selected so that the pinch point
341 temperature in the exhaust gas boiler is 30 K (Ghaffari et al., 2020; Nordin and Majid, 2016). The design
19
342 pressure of the steam that enters the steam turbine is taken as 14.7 bar (Senary et al., 2016).
343 Consequently, the net power of the WHR system is calculated for each ship by assuming the isentropic
344 efficiency of the steam turbine as 65% (U.S. EPA, 2015) . The thermal efficiency of the WHR system
345 is calculated as 22.8%. The relative error of WHR net electric power is found 0.77% with this isentropic
346 efficiency assumption of the turbine when the WHR model validation is performed (Luo and Wang,
347 2017). The results are in good agreement with the literature.
348 The scrubber is employed to reduce the SOx emissions to meet the IMO MARPOL Annex VI criteria
349 for the VLCC tanker (International Maritime Organization, 2020). Therefore, the cost of using the
350 scrubber is considered with the CCS system cost for the VLCC tanker. The initial investment cost of the
352
353 Figure 8: Model of the waste heat recovery system with the steam turbine.
354
355 5. The Effects of Different Design Parameters on the Carbon Capture and Storage System
356 The effects of the different design parameters on the captured CO2 emission from the main engine
357 and the cost of the CCS system per captured CO2 are investigated for the absorber column of the CCS
358 system of the VLCC tanker. The studied design parameters are the absorber diameter, absorber
359 downcomer (DC) width, the distance between the absorber plates, absorber weir height, absorber type,
20
360 and absorber packed height as shown in Figure 9. The total cost of the system includes MEA makeup,
361 freight loss, WHR loss, liquefaction equipment, liquefaction power, capture equipment, capture power,
362 and heat costs. The design of the stripping column of the CCS system is also carried out using the results
364
365
366 Figure 9: Different design parameters of the absorber and stripping columns for the plate type column.
367
369 The effects of the absorber column diameter on the total captured CO2 and different costs are shown
370 in Figure 10. Increasing the diameter of the absorber column enhances the contact surface area so that
371 the capture rate increases. The capture equipment cost also increases with diameter. However, the
372 capture equipment cost per captured CO2 decreases until the diameter of 9 meters. The installed cost of
373 the absorber increases approximately with the 2nd power of the absorber diameter as seen in Figure 11.
374 Other equipment costs almost unchanged with the absorber diameter. On the other hand, Figure 10
375 shows the amount of captured CO2 increases with the diameter of the absorber column. It should be
21
376 noted that the slope of this graph decreases with the diameter increasing. Therefore, the capture
377 equipment cost per unit captured CO2 first decreases up to a minimum value, then increases. When the
378 absorber diameter reaches 9 meters, the waste heat is not enough to regenerate the CO2 from the solvent
379 in the stripping column due to the increased capture rate. Therefore, an extra heater should be used but
380 the extra heater increases both the capture equipment and heat costs. Increasing the amount of captured
381 CO2 increases the cost of MEA makeup. Besides, the freight loss cost increase with the absorber column
382 diameter. Since the weight of the equipment increases with an increase in the diameter of the absorber.
383 Also, the make-up MEA amount per captured CO2 increases slightly when the capture rate and
384 accordingly vaporized MEA increases. The WHR loss cost is gradually decreased with increasing
385 captured CO2. The cost of the power required by the liquefaction is almost unchanged per unit of CO2
387
80 60
58 MEA make-up cost
The costs of the CCS system ($/ton)
70
56
Captured rate of CO2 (%)
389 Figure 10: The changes in the total captured CO2 and different costs with the absorber column
390 diameters.
391
22
Installed Cost of Absorber [USD]
4
3.5 y = 0.03x1.971
R² = 0.9744
3
Million $
2.5
1.5
1
5.5 6.5 7.5 8.5 9.5 10.5 11.5
Absorber Diameter (m)
392
393 Figure 11: The installed cost of the absorber column for different absorber diameter.
394
396 In plate type columns, the flowing liquid is transferred by the vertical channels that name as the
397 downcomer (DC). The DC is used to keep the liquid pool on the plate, so it enhances the contact times
398 of the gas and solvent. When the DC width increases, the performance of the system decreases. The
399 effects of the downcomer (DC) width of the absorber column on the total captured CO2 and different
400 costs are shown in Figure 12. As the width of the DC increases, the surface contact area decreases, so
401 the amount of capture rate decreases. However, the contact area change is small at the considered DC
402 widths. While the equipment cost remains unchanged, the reduction in CO2 capture rate increases the
403 equipment cost per captured CO2, slightly. On the other hand, increasing the width of DC reduces the
404 pressure loss, therefore the power cost of the exhaust fan reduces. The changes in the other costs with
406
23
70 50
The costs of the CCS system ($/ton) MEA make-up cost
60
Freight loss cost
408 Figure 12: The changes in the total captured CO2 and different costs with the absorber column DC width.
409
411 Figure 13 presents the effects of the distance between plates of the absorber column on the total
412 captured CO2 and different costs. The most important criterion determining the distance between the
413 absorber column trays is the maximum DC backup ratio. According to the literature, this ratio is not
414 desired to be above 50% (Sinnott, 2014). This ratio increases as the distance between the plates
415 decreases. In order to meet the DC back-up ratio criteria, the distance between the plates starting from
416 300 mm is studied. At this distance, the maximum DC backup ratio is obtained as 43.5%. The distance
417 between the plates does not affect the capture rate. Therefore, the costs except the equipment cost do
418 not change with the distance between column plates. At the design stage, the best choice for the distance
419 between column plates will be the smallest distance between the plates, that meets the DC back-up ratio
420 criteria. This results in a compact system and good ship stability.
24
80 50
The costs of the CCS system ($/ton) MEA make-up cost
70
Freight loss cost
422 Figure 13: The changes in the total captured CO2 and different costs with the distance between the
424
426 The changes of total captured CO2 and different costs with the weir height of the absorber column
427 are shown in Figure 14. Increasing the height of the absorber column weir increases the amount of liquid
428 deposited on the plate. Therefore, the contact time between the flue gas and the liquid solvent increases.
429 The weir height is one of the most effective parameters on the absorption. Increasing the height of the
430 weir does not directly affect the equipment costs, but it decreases the equipment cost per unit of captured
431 CO2, since the amount of captured CO2 increases. The MEA makeup and freight loss costs increase as
432 the captured CO2 rate increases. On the other hand, the weir height has the most significant effect on the
433 pressure loss of the exhaust gas. Besides, when the capture rate exceeds a certain level, the need for
434 heating increases considerably. Therefore, the power and heat costs required for these capture rates
435 increase significantly. As a result, the amount of waste heat is the restrictive parameter here. Figure 14
436 shows that the absorber weir height of about 70 mm is the best solution in terms of the total cost of the
437 CCS system. In the literature, the absorber column weir height is recommended between 40 - 90 mm
438 (Sinnott, 2014). Therefore, it can be said that the results are in good agreement with the literature.
439
25
90 65
MEA make-up cost
80
The costs of the CCS system ($/ton)
60
70 Freight loss cost
441 Figure 14: The changes in the total captured CO2 and different costs with the absorber column weir
442 height.
443
445 The effects of different absorber column types on the total captured CO2 and different costs are
446 presented in Figure 15. In this study, the investigated absorber column types are the Glitsch ballast valve,
447 Koch flexitray valve, nutter float valve, sieve as trayed types, and pall rings as packed type. Figure 15
448 presents the effects of different absorber column types on the total captured CO2 and different costs.
449 According to the literature, the sieve type plate is the cheapest one (Sinnott, 2014). However, our results
450 show that the low capture rate and high-pressure loss make the sieve type plate the most costly type in
451 terms of cost per unit captured CO2 for ships considered in this study. The packed type column provides
452 the best capture rate and the minimum exhaust gas fan power due to the smallest pressure lost at the
453 absorber column. Therefore, it can be said that the packed type with pall rings is the best choice for our
454 study.
455
26
90
The costs of the CCS system [$/ton] 80 53 MEA make-up cost
457 Figure 15: The changes in the total captured CO2 and different costs with the absorber column type.
458
460 The changes in the total captured CO2 and different costs with the absorber packed height are shown
461 in Figure 16. If the absorber packed height increases, the capture rate of the CO2 increases as expected.
462 Increasing the packed height is a quite reasonable solution for the high-rate captures. As the column
463 packed height increases, the cost of capturing equipment increases. However, the significant increase in
464 the capture rate reduces the cost per unit of captured CO2. The capture heat costs increase with the
466 The low-cost CCS system can be defined by analyzing the results. Table 4 shows the main
467 characteristics and dimensions of the CCS system for the VLCC tanker. The results show that the
468 absorber and stripper heights are respectively 64% and 69.2% is less than that of given in the literature
469 although the captured CO2 emission rate is less than only 22.2% (Luo and Wang, 2017). So, absorber
470 height is reduced by 34.4% and stripper height is reduced by 38.4% per captured CO2. As mentioned
471 earlier the maximum flooding rates for both columns are below 80%, as recommended in Sinnot (2014).
472
27
60
100 MEA make-up cost
The costs of the CCS system ($/ton)
55
Freight loss cost
474 Figure 16: The changes in the total captured CO2 and different costs with the absorber packed height.
475
476 Table 4: The main characteristics and dimensions of the CCS system for the VLCC tanker.
478 The calculation results of the power, weight, and installation costs of capture and liquefaction
479 equipment, and different costs of the CCS system which used pall ring packed type columns with
480 scrubber are presented in Table 5. The power requirement of the CCS system is obtained to be
481 approximately 4.9% of the total power which is almost the same value given in the previous study (Luo
482 and Wang, 2017). On the other hand, the process water at the exit of the stripper column reboiler has an
483 enormous amount of heat at 110.1 °C temperature and 182.7 ton/hour flow rate. Therefore, a waste heat
484 recovery system based on the organic Rankine cycle (ORC) can be used. Taking the thermal efficiency
485 of the ORC as 9.24%, the 791.4 kW net power can be obtained (Akman and Ergin, 2020, 2016). The
28
486 CCS system power will be reduced from 4.9% to 1.1% of the total power. As can be seen from Table 5,
487 the total life cycle cost of the CCS system is about 60.04 $/ton-captured CO2. This cost is a minimum
488 of 14.5% lower when compared to a study in the literature that has an 18.2% higher CO 2 capture rate
489 (Luo and Wang, 2017). This is mainly due to reduced fuel costs by using HFO 380 instead of MDO and
490 equipment costs by considering the design parameters of the columns of the CCS system in this study.
491 Besides, this cost is at least a minimum of 38.4% lower than another study in the literature (Feenstra et
492 al., 2019). However, it should be noted that the amount ratio of CO2 captured in the literature study to
494
495 Table 5: Main cost items and different costs of the CCS system with SOx scrubber for the VLCC tanker.
29
Total Installed Capturing Equipments Cost 5.538 M$
Total Installed Liquefaction Equipment Cost 5.907 M$
Total Installed Scrubber Cost 1.921 M$
Total Installed Equipment Cost 13.36 M$
Captured CO2 6.104 ton/h
MEA Requirements 10.35 kg/h
Capture Equipment Cost 15.63 $/ton CO2 Capt.
Capture Power & Heat Req. Cost 2.03 $/ton CO2 Capt.
MEA Makeup Cost 2.25 $/ton CO2 Capt.
Freight Loss Cost 3.57 $/ton CO2 Capt.
Liquefaction Equipment Cost 16.68 $/ton CO2 Capt.
Liquefaction Power Req. Cost 5.93 $/ton CO2 Capt.
WHR Loss Cost 13.95 $/ton CO2 Capt.
Life cycle cost of the CCS system 60.04 $/ton CO2 Capt.
Total SOx scrubber cost 5.42 $/ton CO2 Capt.
Total life cycle cost 65.47 $/ton CO2 Capt.
496
497 6. Comparisons of Solvent Based Carbon Capture and Storage System with Alternative
499 The carbon tax price is 23.8 €/ton according to the European Emission Allowances (EUA) for
500 October 2019 (Markets Insider, 2019). The carbon tax price means that the factory or foundation has to
501 pay a tax per ton of CO2 emitted in Europe and the European Economic Community Area regions.
502 According to the same source, from 2009 to 2019, the carbon tax price has been increased by 66.6%
503 which can be seen from Figure 17. The inflation rate of the European Union from 2009 to 2019 is
504 13.84% (Inflation Tool, 2019). It can be seen that the carbon tax price will increase regardless of
505 inflation in the future. For 2030, the carbon tax price is estimated as 50 €/ton approximately (Schjolset,
506 2014). In another study, it is predicted that the carbon tax price is expected to be around 50 $/ton, but
507 this price will not encourage innovative technologies unless this price goes above 250 $/ton (Lloyd’s
508 Register, 2018). According to these data, it is obvious that this price will significantly increase by 2050.
509 Therefore, it would be more advantageous to use the CCS system instead of paying the carbon tax.
510 However, the CCS system is not the only method that can be used to reduce CO2 emissions for
511 ships. The CO2 emissions from the ships can be reduced by using renewable energy sources, LNG, fuel
30
512 cells, speed reduction, propulsion system, hull and form optimization, etc. In this study, the LNG usage
513 and speed reduction methods are considered and compared with the CCS system for different ships.
514
30
25
20
$/ton CO2
15
10
0
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Years
515
516 Figure 17: Changes in the value of CO2 in the euro according to EUA data over the years (Markets
518
521 The speed of the VLCC tanker is obtained by using the Holtrop method to investigate the effects of
522 the speed reduction on CO2 emissions (Holtrop, 1984). Accordingly, the speed at 100% MCR with the
523 assumption of 60% overall propulsion efficiency is found as 15.82 knots (Taskar et al., 2019). Also, the
524 service speed at 75% MCR is calculated as 14.42 knots. The service speed that meets the IMO’s 2050
525 CO2 emission criteria for the EEDI is calculated as 10.25 knots at 60% overall propulsion efficiency
526 with 7650 kW operation power. The SFOC of the machine engine is about 166.8 g/kWh, and the engine
527 rate is between 25-30% MCR for the VLCC tanker (“Man Energy Solutions,” 2018). The EEDI
528 calculation is carried out according to IMO MARPOL Annex VI (International Maritime Organization,
529 2020). According to these results, the required EEDI value of phase 0 which indicates the EEDI baseline
530 for ships built between 2013-2015 is calculated as 2.589 g CO2/ton-mile for the tanker, and the attained
531 EEDI value is 1.292 g CO2/ton-mile which is about half of the required EEDI value.
31
532 The profitability and CO2 reduction costs for the VLCC tanker are analyzed according to operational
533 speed change, and the results are shown in Table 6. The minus sign for the cost indicates that it is a
534 profit. It can be seen that form Table 6, the total fuel cost of the VLCC tanker with the speed reduction
535 method is reduced by about 62.0%. On the other hand, the carried load per unit time is reduced so the
536 revenue of the ship decreased almost linearly with speed. The freight loss cost due to the speed reduction
537 is over 4 M$ yearly and this cost is about 97.5% of the fuel revenue. According to these results, the
538 speed reduction method is not only a cheaper solution than using the CCS system but also very cost-
539 efficient. However, operating the engine at low loads continuously has some negative effects and
540 problems on the engine’s working process. To solve these problems, the engine and equipment require
541 some retrofits and precautions like using slide-type fuel injectors, changing cylinder lubricators,
542 controlling exhaust gas temperature after EGB, etc (Le et al., 2019; MAN Primeserv, 2012). Also, these
543 retrofits and solution of problems have some costs, but they are not considered in this study.
544
545 Table 6: VLCC tanker profitability and CO2 reduction costs for the speed reduction method
32
CO2 reduction and SOx scrubber cost: 1.35 $/ton CO2
546
548 The use of natural gas is not sufficient alone to meet the IMO 2050 criteria for the CO2 emissions
549 criteria. Therefore, it is used with the speed reduction method. In this study, the 8S70ME-C10.5 gas
550 injection MAN engine is selected for the VLCC tanker (“Man Energy Solutions,” 2018). The engine
551 power is 20580 kW, with a SGC of 136.3 g/kWh and SPOC of 3.02 g/kWh at 75% MCR. The service
552 speed that meets the IMO’s 2050 CO2 emission criteria for the EEDI is calculated as 12.26 knots at 60%
553 overall propulsion efficiency with 12800 kW operation power. The SGC of the engine is about 134.6
554 g/kWh, and the engine rate is between 45-50% MCR for the VLCC tanker (“Man Energy Solutions,”
555 2018). According to these results, The required EEDI value of phase 0 is calculated as 2.589 g CO2/ton-
556 mile, and the attained EEDI value is 1.287 g CO2/ton-mile which is about half of the required EEDI
557 value.
558 Diesel is used as pilot fuel, and the diesel price for Antwerp port is 560 $/ton in September 2019
559 (Petrol Bunkering, 2019). The price of LNG is 4620 $/MMBtu, according to the export figures of the
560 European Union in December 2019 (Ycharts, 2019). For a lower heating value of 50000 kJ/kg, the price
561 of LNG is about 220 $/ton. However, this price is the only export price, and when the LNG arrives at a
562 port, it is assumed that the price increases by 50 $/ton (Balcombe et al., 2019). The selected engine is
563 dual fuel and its cost is about 280 $/kW (Danish Maritime Authority, 2012). On the other hand, the cost
564 of the HFO fueled engine is about 235 $/kW (Livanos et al., 2014).
565 The biggest problem of using LNG as a fuel in ships is that the system requires much equipment
566 and cooling for the LNG storage. The LNG tanks and equipment costs are assumed as 365 $/kW
567 (Livanos et al., 2014). On the other hand, LNG combustion emissions contain a quite low amount of
568 NOx and SOx emissions that meet the IMO limits for these emissions. This means that the scrubber is
569 not required for this case. Therefore, the cost of the scrubber is not considered.
570 Table 7 presents the CO2 reduction costs obtained for using the LNG as a fuel and the speed
571 reduction method together for the VLCC tanker. The power of the LNG fueled engine is about 62.2%
572 of the HFO fueled engine and LNG has more calorific value than HFO 380, and the total fuel cost of
33
573 the VLCC tanker is reduced about 44.3% by the LNG usage as shown in Table 7. For the LNG usage
574 with speed reduction, the freight loss due to the speed change is over 2 M$ yearly and this cost is about
575 70.6% of the fuel revenue. Also, there is an initial investment cost that reaches almost 1 M$ yearly
576 because of the main engine refits, LNG tanks, and equipment costs. According to these results, the cost
577 of the LNG usage method is lower than the CCS system cost and higher than the cost of the speed
581 The effects of the fuel and freight prices on the costs of the CO2 reducing methods are significant,
582 especially for the speed reduction method. When the freight rates and ships' economic speeds increase,
583 the best solution to reduce CO2 emissions may not be to reduce the speed. Therefore, in this part of the
584 study, the CCS system, LNG usage, and speed reduction methods are employed for different sizes of
585 LNG carriers and their CO2 reduction costs are determined and compared.
586 Table 7: VLCC tanker profitability and CO2 reduction costs for the combined usage of the LNG and
34
Total fuel cost: 3711398 $/year
Freight loss due to speed change: 2077614 $/year
Main engine and cost due to power change: 41160 $/year
System and equipment change cost: 333853 $/year
Initial investment cost with interest rate: 999036 $/year
Net revenue/loss: -133171 $/year
Reduced carbon amount: 6.104 tons/h
Carbon reducing cost: 2.91 $/ton CO2
588
589 For estimating speeds of the LNG carries, the resistance of the ships is obtained using the Series 60
590 method (Todd, 1963). The results and key cost parameters are shown in Table 8. The CO2 emission rates
591 of the Q-Flex and Q-Max main engines are respectively 24.1% and 41.8% over than that of the
592 conventional LNG carrier. The life cycle costs of the CCS system for Q-Flex and Q-Max are respectively
593 11.4% and 19.4% lower by comparing them with that of the conventional LNG carrier. It is shown that
594 the CCS system cost per unit CO2 decreases with an increase in fuel consumption and CO2 emissions.
595 On the other hand, the life cycle costs of the speed reduction method for the Q-Flex and Q-Max are
596 respectively 78.8% and 38.8% higher than that of the conventional LNG carrier. These changes are
597 almost linear with ship capacities. According to the results for the speed reduction method, when the
598 ship capacity increases, the total life cycle cost of CO2 reduction increases. Also, it is found that for the
599 high capacity ships such as Q-Flex and Q-Max, the use of the CCS system is more cost-effective than
601
602 Table 8: Economic evaluation keys and results for different size of LNG carriers.
Large
Description Q-Flex Q-Max
conventional
Ship capacity (m3) 150000 210000 265000
DWT 72800 103838 131529
Max. power of ships’ main engine (kW) 31400 39300 45200
Required EEDI for phase 0 (g CO2/ ton-mile) 6.793 5.778 5.187
Attained EEDI (g CO2/ ton-mile) 4.864 4.305 3.885
Daily freighting value ($/day) 70000 98000 125000
35
MAN MAN MAN
Selected main engine 9S50ME- 7G60ME- 8G60ME-
C9.7 x 2 C10.5 x 2 C10.5 x 2
Max. power of selected main engines (kW) 32040 39760 45440
Max. estimated speed (kn) 20.0 19.7 19.7
75% MCR estimated speed (kn) 18.4 17.9 17.9
CCS System
ISO SFOC at 75% MCR (g/kWh) 135.0 135.0 135.0
Captured CO2 amount (%) 31.9 34.5 34.9
Attained EEDI (g CO2/ ton-mile) 3.363 2.876 2.583
Life cycle cost of the CCS system ($/ton) 73.74 65.35 59.46
Speed Reduction Method
Required EEDI for phase 0 (g CO2/ ton-mile) 10.18 8.572 7.703
Attained EEDI (g CO2/ ton-mile) 5.061 4.249 3.824
Operation power of main engines (kW) 16400 18600 21000
ISO SFOC at operation power of engines (g/kWh) 132.5 132.7 132.8
Estimated speeds of ships (kn) 16.22 15.35 15.25
Total life cycle cost of CO2 Rreduction ($/ton) 54.85 76.17 98.10
603
604 7. Conclusions
605 In this paper, an investigation was presented, for different ship types and sizes, on a solvent-based
606 CCS system for reducing the CO2 emissions by using process modeling with Aspen HYSYS. A VLCC
607 tanker and three different sizes of LNG carriers (Q-Max, Q-Flex, conventional LNG carrier) were
608 considered. The cost of the CCS system as cost per CO2 capture was also analyzed by considering
609 different design parameters for the columns of the CCS system. After that, the solvent-based CCS
610 system was compared with the speed reduction and LNG usage methods for their CO2 emission
611 reduction capability and costs. It was found that the CCS system for ships is capable of meeting the
612 IMO’s 2050 goal of reducing CO2 emissions below 50% compared to 2008. Therefore, the application
613 of the CCS system to ships is promising in terms of reducing CO2 emissions.
614 The CCS system requires much equipment, a wide area, and a large volume. Also, the weight of the
615 system is another problem in the application of the system to the ships. The weight of the CCS system
616 causes the reduction of cargo carried. Therefore, it reduces profitability. In In this study, in order to
36
617 avoid these the large volume requirements and weight problems, it was proposed to apply the CCS
618 systems to ships that are large and capable of providing enough area and volume on the deck.
619 The hydraulic design principles were taken into account to design the absorber and stripping towers
620 of the CCS systems on ships. This consideration leads to the sizing of the absorber and stripping columns
621 on the ship to meet the stability and arrangement criteria of the ship. An essential result of studying the
622 effects of different parameters on the CCS system was the heights of the absorber, and stripper columns
623 could be quite shorter compare to the land facilities. The absorber height was found 4.5 meters and the
624 stripper height 2.0 meters for the CCS system of the VLCC tanker in this study. The absorber and
625 stripper heights were respectively 64% and 69.2% was less than that of given in the literature although
626 the captured CO2 emission rate was less than only 28.9%. So, tThe absorber height was reduced by
627 34.4% and stripper height was reduced by 38.4% per captured CO2 when compared to the literature.
628 When there was enough reservoir backup for the downcomer, the heights of the plate type columns
629 could be further reduced for ships. Besides, the effect of column height on the cost was negligible for
630 the plate type columns. The maximum carbon capture level provided by the waste heat and the
631 area/volume that the ship could provide were the critical factors when determining the absorber and
633 The carbon capture rate of the CCS system increased with increasing the contact surface area,
634 column diameter and number of plates and with decreasing the DC width. The absorber column's weir
635 height increased with the deposited amount of the liquid on the plate. The contact time of the gas and
636 liquid phases increased, and as a result of this the carbon capture rate increased.
637 The waste heat usage for stripping CO2 from solution caused an indirect cost, because, in this case,
638 it would not be possible to produce energy from the waste heat using a steam turbine. This cost was
639 named as the WHR lost cost. It was found as 13.95 $/ton-captured CO2 which is 21.3% of the total life
640 cycle cost of the CCS and SOx scrubber system for the VLCC tanker. On the other hand, the process
641 water passed from the stripper columns reboiler had a it is possible to gain significant amount of heat at
642 above 100 °C from stripper columns reboiler heat exchanger. In a further study, the Organic Rankine
643 Cycle (ORC) or Stirling Cycle could be considered to use this waste heat for reducing both the energy
37
644 requirements and the cost of the CCS system. The CCS system power requirements could be reduced
645 from 4.9% to 1.1% of the total power by using the ORC for the VLCC tanker.
646 In this study, the costs were considered as the solvent supply cost, freight loss cost, WHR loss cost,
647 carbon capture equipment cost, liquefaction equipment cost, power and heat loss cost for capture, power
648 and heat loss cost for liquefaction. Therefore, it was possible to define the importance of cost types. The
649 cost analyses of the CCS system showed that the equipment and the capital costs had more effects on
650 the total life cycle cost than the operating cost of the CCS system for ships. The capture equipment cost
651 was obtained as 21.6% and liquefaction equipment costs 27.2% of the total life cycle cost for the VLCC
652 tanker. As the amount of CO2 emission increased, the total life cycle cost per captured CO2 of the CCS
653 system decreased. For the conventional LNG carrier with a 150000 m3 capacity, the freight loss cost
654 was found as 8.56 $/ton-captured CO2 while it was found as 6.13 $/ton-captured CO2 for Q-Max with a
655 265000 m3 capacity. Therefore, it would be more feasible to use CCS systems on ships with relatively
656 big size and high power. On the other hand, as the amount of captured carbon increased, the heat energy
657 requirement of the system increased. When the required heat energy exceeded the amount of available
659 When the cost of the CCS system was examined by its components, it could be said that the highest
660 cost, in general, was the liquefaction cost. It was found that the liquefaction power cost was 9.1% and
661 liquefaction equipment costs 25.5% of the total life cycle cost for the VLCC tanker. In this case, to
662 reduce this cost, the options of solidification rather than liquefaction or storing CO2 under different
663 operating conditions would be considered for suitable type of ships like bulk carriers. However, the
664 solidification process would require handling systems, increase the freight loss due to the heavier
665 product than the liquefied CO2, and require larger volume than the liquefied CO2. However, for bulk
666 carriers, the application of solidification might be more suitable than liquefaction. The storage of the
667 liquefied CO2 under different conditions by changing the liquefaction pressure and temperature could
669 Comparing the cost of the CCS system with the costs of LNG usage and speed reduction method
670 showed that the speed reduction was the best cost-oriented solution for ships with low freight values
671 like crude oil tankers. The total life cycle cost of the CCS system was found to be 65.47 $/ton-captured
38
672 CO2 and the speed reduction cost 1.35 $/ton-captured CO2 for the VLCC tanker. However, for high
673 speed and having high freight values ships like Q-Max and Q-Flex LNG carriers, it was found that the
674 CCS system was more cost-effective than the LNG usage and speed reduction method. The total life
675 cycle costs of the CCS system were respectively 14.2% and 39.4% lower than that of the speed reduction
676 costs for the Q-Flex and Q-Max. These contributions would make it easier for shipowners and operators
677 to determine the most appropriate solution to meet the EEDI and EEXI criteria for their ships.
678 Future studies could be on the benefits of selling captured CO2, different solvent usage to improve
679 the CO2 absorption rate and reducing the energy demand of the CCS system, and the design of stripping
680 columns to lower the energy demand by considering different pressure and temperature conditions. Also,
681 the benefits of combining different CO2 capture methods and different CCS systems could be further
39
683 Declaration of Competing Interest
684 The authors declare that they have no known competing financial interests or personal relationships that
685 could have appeared to influence the work reported in this paper.
686
687 References
688 Agbonghae, E.O., Hughes, K.J., Ingham, D.B., Ma, L., Pourkashanian, M., 2014. Optimal process
689 design of commercial-scale amine-based CO2 capture plants. Ind. Eng. Chem. Res. 53 (38),
690 14815–14829. https://doi.org/10.1021/ie5023767
691 Akman, M., Ergin, S., 2020. Thermo-environmental analysis and performance optimisation of
692 transcritical organic Rankine cycle system for waste heat recovery of a marine diesel engine.
693 Ships Offshore Struct. 0 (0), 1–10. https://doi.org/10.1080/17445302.2020.1816744
694 Akman, M., Ergin, S., 2016. Thermodynamic analysis of organic rankine cycle for waste heat
695 recovery system of a ship, in: 1st International Congress on Ship and Marine Technology Green
696 Technologies. pp. 717–728.
697 Awoyomi, A., Patchigolla, K., Anthony, E.J., 2019. CO2/SO2 emission reduction in CO2 shipping
698 infrastructure. Int. J. Greenh. Gas Control 88, 57–70.
699 Balcombe, P., Brierley, J., Lewis, C., Skatvedt, L., Speirs, J., Hawkes, A., Staffell, I., 2019. How to
700 decarbonise international shipping: Options for fuels, technologies and policies. Energy Convers.
701 Manag. 182 (January), 72–88. https://doi.org/10.1016/j.enconman.2018.12.080
702 Boot-Handford, M.E., Abanades, J.C., Anthony, E.J., Blunt, M.J., Brandani, S., Mac Dowell, N.,
703 Fernández, J.R., Ferrari, M.-C., Gross, R., Hallett, J.P., 2014. Carbon capture and storage update.
704 Environ. Energy Sci. 7 (1), 130–189.
705 Buhaug, Ø., Corbett, J.J., Endresen, O., Eyring, V., Faber, J., Hanayama, S., Lee, D.S., Lindstad, H.,
706 Markowska, A.Z., Mjelde, A., Nelissen, D., Nilsen, J., Pålsson, C., Winebrake, J.J., Wu, W.,
707 Yoshida, K., 2009. Second IMO GHG Study 2009 220.
708 Bureau Veritas, 2014. Rules for the Classification of Steel Ships, Part C-Machinery, Electricity,
709 Automation and Fire Protection.
710 Clarkson Research Services Limited, 2019. No Title.
711 Conway, W., Bruggink, S., Beyad, Y., Luo, W., Melián-Cabrera, I., Puxty, G., Feron, P., 2015. CO2
712 absorption into aqueous amine blended solutions containing monoethanolamine (MEA), N,N-
713 dimethylethanolamine (DMEA), N,N-diethylethanolamine (DEEA) and 2-amino-2-methyl-1-
714 propanol (AMP) for post-combustion capture processes. Chem. Eng. Sci. 126, 446–454.
715 https://doi.org/10.1016/j.ces.2014.12.053
716 Danish Maritime Authority, 2012. North European LNG infrastructure project. A feasibility study for
717 an LNG filling station infrastructure and test of recommendations.
718 Dynagas LNG Partners LP, 2018. Investor Presentation.
719 Euronav, 2013. New building VLCCs and/or suezmax are neither eco nor good value, Euronav
720 website.
721 Feenstra, M., Monteiro, J., van den Akker, J.T., Abu-Zahra, M.R.M., Gilling, E., Goetheer, E., 2019.
722 Ship-based carbon capture onboard of diesel or LNG-fuelled ships. Int. J. Greenh. Gas Control
723 85, 1–10.
724 Fevre, C. Le, 2018. A review of demand prospects for LNG as a marine transport fuel.
40
725 https://doi.org/10.26889/9781784671143
726 Ghaffari, A., Ahmadi, R., Eyvazkhani, M., 2020. Modeling and optimization of finless and finned tube
727 heat recovery steam generators for cogeneration plants. Eng. Reports 2 (11), 1–22.
728 https://doi.org/10.1002/eng2.12262
729 Gupta, M., Da Silva, E.F., Hartono, A., Svendsen, H.F., 2013. Theoretical study of differential
730 enthalpy of absorption of CO2 with MEA and MDEA as a function of temperature. J. Phys.
731 Chem. B 117 (32), 9457–9468. https://doi.org/10.1021/jp404356e
732 Holtrop, J., 1984. Data, a statistical re-analysis of resistance and propulsion. Int. Shipbuild. Prog. 31
733 (363).
734 Iliuta, I., Larachi, F., 2017. Enzyme-mediated CO2 capture in oscillating structured packed-bed
735 columns-Hydrodynamics and process performance for offshore applications. Ocean Eng. 144,
736 157–174.
737 Inflation Tool, 2019. No Title.
738 International Maritime Organization, 2020. MARPOL Annex VI.
739 International Maritime Organization, 2018. Adoption of the initial IMO strategy on reduction of GHG
740 emissions from ships and existing IMO activity related to reducing GHG emissons in the
741 shipping sector.
742 Intratec Solutions, 2019. Intratec chemicals pricing data monoethanolamine prices.
743 IPCC, 2001. Climate change 2001: the scientific basis. Contribution of working group I to the 3rd
744 assessment report of the Intergovernmental Panel on Climate Change (IPCC).
745 Jinyue, Y., 2015. Carbon capture and storage (CCS). Appl. Energy 148, A1-6.
746 Le, V.V., Truong, T.H., Dong, T.M.H., 2019. Efficient operation solutions for marine engine series of
747 Man B&W S46Me-B8.5 in low steaming. J. Mech. Eng. Res. Dev. 42 (5), 51–56.
748 https://doi.org/10.26480/jmerd.05.2019.51.56
749 Livanos, G.A., Theotokatos, G., Pagonis, D.-N., 2014. Techno-economic investigation of alternative
750 propulsion plants for Ferries and RoRo ships. Energy Convers. Manag. 79, 640–651.
751 Lloyd’s Register, 2018. Zero-Emission Vessels 2030 . How do we get there?
752 Luo, X., Wang, M., 2017. Study of solvent-based carbon capture for cargo ships through process
753 modelling and simulation. Appl. Energy 195, 402–413.
754 Luo, X., Wang, M., 2016. Optimal operation of MEA-based post-combustion carbon capture for
755 natural gas combined cycle power plants under different market conditions. Int. J. Greenh. Gas
756 Control 48, 312–320.
757 Ma, P.C., Bravo, L., Ihme, M., 2015. Supercritical and transcritical real-fluid mixing in diesel engine
758 applications.
759 MAN Diesel & Turbo, 2014. Propulsion trends in LNG carriers. Two-stroke engines.
760 https://doi.org/10.1016/j.jbiotec.2014.06.034
761 MAN Diesel & Turbo, 2013. Propulsion Trends in Tankers, MAN Diesel.
762 Man Energy Solutions, 2018.
763 MAN Primeserv, 2012. Slow steaming.
764 Markets Insider, 2019. No Title.
765 Mores, P.L., Godoy, E., Mussati, S.F., Scenna, N.J., 2014. A NGCC power plant with a CO2 post-
41
766 combustion capture option. Optimal economics for different generation/capture goals. Chem.
767 Eng. Res. Des. 92 (7), 1329–1353.
768 Nordin, A., Majid, M.A.A., 2016. Parametric study on the effects of pinch and approach points on heat
769 recovery steam generator performance at a district cooling system. J. Mech. Eng. Sci. 10 (2),
770 2134–2144. https://doi.org/10.15282/jmes.10.2.2016.17.0201
771 Novotony, T., 2019. LNG Shipping Economics.
772 Øi, L.E., 2007. Aspen HYSYS simulation of CO2 removal by amine absorption from a gas based
773 power plant, in: The 48th Scandinavian Conference on Simulation and Modeling (SIMS 2007);
774 30-31 October; 2007; Göteborg (Särö). Linköping University Electronic Press, pp. 73–81.
775 Petrol Bunkering, 2019. No Title.
776 Rackley, S.A., 2017. Power generation fundamentals, in: Carbon Capture and Storage. pp. 37–71.
777 https://doi.org/10.1016/b978-0-12-812041-5.00003-9
778 Rochelle, G.T., 2009. Amine scrubbing for CO2 capture. Science (80-. ). 325 (5948), 1652–1654.
779 Ronen, D. %J J. of the O.R.S., 1982. The effect of oil price on the optimal speed of ships 33 (11),
780 1035–1040.
781 Scheffe, R.D., Weiland, R.H., 1987. Mass-transfer characteristics of valve trays. Ind. Eng. Chem. Res.
782 26 (2), 228–236.
783 Schjolset, S., 2014. The MSR: Impact on market balance and prices, in: Point Carbon. Thomson
784 Reuters.
785 Sea-distances.org, 2019.
786 Senary, K., Tawfik, A., Hegazy, E., Ali, A., 2016. Development of a waste heat recovery system
787 onboard LNG carrier to meet IMO regulations. Alexandria Eng. J. 55 (3), 1951–1960.
788 https://doi.org/10.1016/j.aej.2016.07.027
789 Seo, Y., Huh, C., Lee, S., Chang, D., 2016. Comparison of CO2 liquefaction pressures for ship-based
790 carbon capture and storage (CCS) chain. Int. J. Greenh. Gas Control 52, 1–12.
791 Ship Technology, 2013. Onboard carbon capture: dream or reality?
792 Sinnott, R.A.Y., 2014. Chemical engineering design. Elsevier.
793 Slow spur for Maersk VLCCs, 2010. TradeWinds.
794 Taskar, B., Regener, B., Andersen, P., 2019. The impact of propulsion factors on vessel performance
795 in waves, in: Sixth International Symposium on Marine Propulsors.
796 Todd, F.H., 1963. Series 60 methodical experiments with models of single-screw merchant ships.
797 Trivyza, N.L., Rentizelas, A., Theotokatos, G., 2019. Impact of carbon pricing on the cruise ship
798 energy systems optimal configuration. Energy 175, 952–966.
799 Trivyza, N.L., Rentizelas, A., Theotokatos, G., 2018. A novel multi-objective decision support method
800 for ship energy systems synthesis to enhance sustainability. Energy Convers. Manag. 168, 128–
801 149.
802 U. S. Environmental Protection Agency Combined Heat and Power Partnership, 2015. Catalog of CHP
803 Technologies: Section 4. Technology Characterization – Steam Turbines.
804 United States Energy Information Administration, 2016. Capital Cost Estimates for Utility Scale
805 Electricity Generating Plants, United States Department of Energy.
806 Wang, H., Zhou, P., Wang, Z., 2016. Experimental and numerical analysis on impacts of significant
42
807 factors on carbon dioxide absorption efficiency in the carbon solidification process. Ocean Eng.
808 113, 133–143.
809 Wang, M., Lawal, A., Stephenson, P., Sidders, J., Ramshaw, C., 2011. Post-combustion CO2 capture
810 with chemical absorption: a state-of-the-art review. Chem. Eng. Res. Des. 89 (9), 1609–1624.
811 Ycharts, 2019. European union natural gas import price.
812 Zhou, P., Wang, H., 2014. Carbon capture and storage-Solidification and storage of carbon dioxide
813 captured on ships. Ocean Eng. 91, 172–180.
814
43