Effect pressure INTODUCTION. a swirl-type. in this study. marine diesel. engines. The. agent, urea is. (UWS), con- NH 3 slip, and 1 G.

Transcription

1 SNAK, 014 Int. J. Nav. N Arhit. Oean Eng. (014) 6:7~38 pissn: , eissn: Taewha Park 1, Yonmo Sung 1, Taekyung Kim 1, Inwon Lee, Gyungmin Choi 3 and Dukjooll Kim 3 1 G Effet of stati mixerr geometry on flow mixing and pressure drop in marine SCR appliations Graduate Program, Shool of Mehanial Engineering, Pusan National University, Busan, Korea Global Core Researh Center for Ships andd Offshore Plants, Pusan National University, Busan, Korea 3 Pusan Clean Coal Center, Pusan National University, Busan, Korea ABSTRACT: Flow mixing and pressure drop harateristis for marine seletive atalyti redution appliations weree investigated numerially to develop an effiient stati mixer. Two different mixers, line- and swirl-type, were onsidered. The effet of vane angles on the relative intensity, uniformity index, and pressure drop was investigated in a swirl-type mixer; these parameters are dramatially affeted by the mixer geometry. Thee presene of f a mixer, regardless of the mixer type, led to an improvement of approximately 0% in the mixing performane behind the mixer in omparison to not having a mixer. In partiular, there was a tradeoff relationship between thee uniformity and the pressure drop. Con- sidering the mixing performane and the pressure drop, the swirl-type Uniformity index; Pressure drop; Marine diesel mixer was more suitable than the line-type mixer in this study. KEY WORDS: Seletive atalyti redution; Stati mixer; engine. INTODUCTION In order to meet stringent future emission regulations from the environmental protetion ageny (EPA), espeially e with respet to reduing nitrogenn oxides (NOx), x various tehnologies suh as basi internal engine modifiations, fuel f swithing, diret water injetion, exhaust gas reirulation (EGR), and seletivee atalyti redution (SCR) have been reommended for marine diesel engines. The internationall maritime organization (IMO) has regulated NO x emissions from marine vessels. The IMO s Tier III standard requires thatt marine vessels must redue NO x emissions by 80% between 010 and a 016 (Die- selnet, 008) ). SCR is one of the most promising tehnologies for aomplishing this aggressive regulation. As a reduing agent, urea is preferred in marine SCR appliations beause of its safety and low toxiity. t Urea water solution (UWS), on- taining 3.5wt% urea, is injeted into thee hot gas streamm from exhaust manifolds, and a then, NH 3 is generated by dewatering, thermolysis, and hydrolysiss proesses. The urea-scr reator should be ontrolledd to ensure high de-no x performane, low NH 3 slip, and low urea onsumption. To maximize thee de-no x effiieny and minimize the NHH 3 slip, a ontrolled turbulentt mixing proess for two-phase flow, suh as UWS with the exhaust gas stream, and a a highly uniform flow in i front of the SCR reator must be obtained. Flow mixing is a ommon devie unit operation in a large number of proesses, and a it is used inn many different appliations Corresponding author: Gyungmin Choi, This is an Open-Aess artile distributed under u the termss of the Creative Commons Attribution Non-Commerial Liense ( whih permits unrestrited non-ommerial use, distribution, and reprodution in any medium, provided the original work is properly ited.

2 8 Int. J. J Nav. Arhit. Oean Eng. (014) 6:7~38 where a defined degree of homogeneity of a fluid is desired (Regner et al., 006). In I partiular, a mixing devie, e.g., a stati mixer is usually installed to improve the rate of deomposition for urea to NH 3 and to enhane the uniformity of the spatial distribution of NH 3 and isoyani aid (HNCO) (Munnannur and Liu, 010). However, the inreasee in the system pressure drop beause of mixing must be minimized. There have been many attempts to develop stati mixers for mobile SCR appliations, and omputational fluid f dynamis (CFD) has been widely used in the design optimization of spray nozzles, flow mixing m harateristis, NO x redution r pro- esses, and urea deomposition (Thakur et al., 003; Zheng et al., 009; Zheng ett al., 010; Zhang et al., 006; Zhang and Romzek, 007; Birkhold et al., 007; Nguyen et al., 010; Larmi and Tiainen, 003; Chen and Williams, 005; Battoei et al.., 006). Thakur et al. (003) provided an extensive review of stati mixers in the proessing p industry, presenting guidelines for the seletion of stati mixers. Zheng et e al. (009) developed several types of mixers, m inluding one, -stage, and butter- fly mixers. They also investigated the effet of in-pipe mixing devies on urea deposits with respet to mixer onfigurations and various exhaust gas temperatures. Zhang et al. (006) introdued a simple flow mixer with h twisted blades based on an original delta wing mixer for the purpose of reating both swirling and turbulent flows. Turbulent flow has a dominant effet on the flow mixing index or uniformity index i in the short distane immediately behind the floww mixer (Zhang et al., 006; Zhang and Romzek, 007). In allowing a suffiient mixing length by reduing the oupied spae, it is neessary to developp a proper stati mixer with a high mixing performane as well as a low pressure drop. However, there is insuffiient researh on the relationship between flow mixing harateristis and pressure drops in the marine engine fields. In this study, both line- and swirl-type mixers were onsidered; eah mixer was divided d into three ases of vane angles: 30º,, 45º, and 60º. The effets of a mixer s geometri struture on the flow mixing harateristis andd the resulting pressure drop were investigated numerially using a ommerial finite volume, three-dimensio onal (3-D) CFD ode; FLUENT (version 6.3.6). The purpose of this study was to evaluate the effet of mixer geometry on the t relative intensity, uniformity index, and pressure drop with the objetive of enhaning the de-no x effiieny. Additionally, information pertaining to the seletion of proper stati mixers was provided based on the orrelation between the uniformity index and the pressure drop. NUMERICAL METHODS AND CONDITIONS Analysis model The geometry of the SCR system, whih inludes the position of the spray injetor, mixer, SCR reator, and a measuring points, is shown in Fig. 1. Line-type and swirl-type mixers, as shown in Fig., have 36 vanes eah. A swirl-type mixer was developed in this study with several unique features: a simple design for prodution; a variable vane angle to generate different swirl flows; and the flexibility for installation and to ontrol the mixer volume in the pipe. Both types of mixers with various vane angles were simulated in 3-D to investigate the flow pattern, turbulene harateristis, and uniformity of water, whih is assumed to be UWS at the SCR atalyst entrane. Fig. 1 Computational domain for the SCR system.

3 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~38 9 Fig. Configuration of stati mixers: (a) line-type mixer, andd (b) swirl-type mixer. The test onditions are listed in Table 1. The omputational grid is omposed of approximatelyy 650,000 hexahedral and te- trahedral ells using FLUENT/GAMBIT (ANSYS, In., USA). The mesh struture e beame denser toward the spray injetor and the mixer beause of the high gradients of veloity, temperature, and speies onentrations. Position of mixer and spray 7 D and 3 D from inlet boundary Table 1 Computational onditions with various mixerr types and positions. Case Mixer type 1 W/O Swirl, 30º 3 Swirl, 45º 4 Swirl, 60º 5 Line (up-down), 45º Table Summary of initial and boundary onditions. Item Setionn Inlet Boundary ondition of Outlet omputational domain Wall Catalyst Spray material Veloity Initial ondition of Angle spray injetion SMD: (Rosin-Rammler) Injetion type Conditionsns and value Exhaust gas g veloity: 0m/s, 573K Pressure outlet: atmosphere Adiabati / No-slip Porous media: m 1/α = , C = H O (liquid water), 300K 5m/s 70º Mean diameter: 35µm,, Spread parameter: hole, solid one type e Boundary onditions The model assumes thatt the exhaust gas is fully developed and it onsists of 77% % N and 3% % O by mass. A time step of 1ms was determined to be suffiient to produe results independent of the hoie off time step. The injetor onsists of 6 holes with 0.45mm diameter. The initial and boundary onditions used in the alulation are a shown in Table. The initial onditions of the exhaust gases were set based on full f load and 900rpm on a 90kW diesel engine (HYUNDAI HiMSEN). The tem- perature of the exhaust gas was set to 573K, and the inlett veloity was 0m/s.

4 30 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~38 Numerial proedure The flow region in the SCR system is divided into three regions: the 1 st region is the turbulent flow region in the upstream and downstream of the SCR reator; the nd region is the laminar flow region in the SCR reator; and the 3 rd region is the region that ontains the dispersed two-phase flow in the surrounding spray injetor. The Lagrangian disrete phase model is used whih ontains sub-models for droplet dispersion, drag, and evaporation. An injetion type of solid one for primary breakup model is used to injet water liquid. To minimize omputational time and stable onvergene strategy, seondary breakup model for droplet breakup and ollision are not onsidered in this study. The injeted drop-size was found to follow a Rossin- Rammler distribution with a mean diameter of 35 mirons and spread parameter of 3.5. For an inompressible, unsteady twophase turbulent flow, the 3-D Reynolds-averaged Navier-Stokes (RANS) governing equations for mass, momentum, speies onentration, and energy were solved. The standard κ-ε turbulent model was used to alulate the turbulent quantities. The ontinuity, momentum, and energy equations are expressed as follows in Eqs. (1)-(3), respetively. u i t 0 (1) u uu i i j p u j u i t t xj xi x j xi x j () T ut i t T t xi xi x j Pr Prt x i (3) The turbulent kineti energy κ and the rate of energy dissipation ε are omputed from a standard two-layer κ-ε turbulent model. u t xi x j x i i t G (4) u t x x x i t C1GC i j i (5) G denotes the prodution rate of κ and is given below: G uu u u u u i i j i i j t x j xj x i xj, and t C (6) In the above equations, the oeffiients are as follows: C 1 = 1.44, C = 1.9, C = 0.09, = 1.0, and = 1.3. The atalyst is the ore of SCR reator. In this study, a honeyomb type SCR atalyst filter was adopted. If the atalyst filter is onstruted physially from a numerial simulation without any simplifiations, the grid of the model will reah a level that is beyond the alulation apabilities of most omputing systems. Therefore, an approah of a porous media model was adopted

5 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~38 31 to simulate the flow in the atalyst filter. The mass and momentum transfer in the radial diretion was ignored beause the axial veloity is dominant in the SCR filter (Jeong et al., 005). In this simple model, the t pressure hange (drop) iss defined by a ombination of Dary s Law and an additional inertial loss term along the SCR filter (Fluent, 007) ): 1 p v C v m (7) p.73v 199.7v (8) where p is the pressure drop, µ is the laminar fluid visosity, α is the permeability of the medium, C is the pressure-jumpp oeffiient, v is the veloity normal to the porous fae,, and m is the thikness of the medium. To obtain the two unknownn values, α and C, a quadrati equation off Eq. (8) is derived from the alulation of the simple part ell analysis as a shown in Fig. 3. Appropriate values for α and C an be alulated using Eqs. (7) and (8) as shown in Table. Fig. 3 Pressure drop distribution of the part ell in the SCR filter withh respet to inlet bulk veloity. RESULTS AND DISCUSSION Distributionss of veloity and water masss fration in SCR system Fig. 4 illustrates the ontours of the veloity and water onentration of partss of the SCR system, inluding the spray injetor, mixer, and reator. Case 1 represents the alulation onditions of no mixer.. Cases, 3, and 4 represent the alulation onditions of swirl-type mixers with vane angles of 30º,, 45º, and 60º, respetively. Case C 5 represents the alulation onditions of a line-typee mixer with a vane angle off 45º. Considerable disturbanes our surround the spray injetor for all ases. The flow patterns behind the mixers differ with the mixer type. In the swirl l-type mixers of o ases, 3, and 4, the entral reirulation zone (CRZ) is generated near the mixer fields, and the CRZ inreases with inreasingg vane angle. The line-type mixer, however, does not exhibit a CRZ, beause the main flow near the mixer moves up and down. d The veloity distribution and water onentrationn of ase 1 in front of the SCR reator are onentrated in the enter region. This s phenomenon may promote serious problems, suh as atalyst filter damage and NHH 3 slip (NMRI, 011). For the t remainder of the ases (ases -5), the veloity and onentration of water are relatively well distributed in omparison to ase 1. In partiular, Fig. 4(b) ) demonstratess that the mass fration of water for ase 3 appears a well mixed with input gases in frontt of the SCR reator. Based on these results, NH 3 slip, low mixing of UWS with hot exhaust gases, and atalyst filer damage in front of the SCRR reator have the t potential to be mitigated without using any guide vanes, suh as baffles.

6 3 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~38 Fig. 4 Contours of (a) veloity and (b) water onentration for different ases with a alulation time 1.5s. Effet of mixer geometry on turbulent flow harateristis Turbulent flow ours when instabilities in a flow are not suffiiently damped by visous ation and the fluid veloity at eah point in the flow exhibits random flutuations (Turns, 000). Turbulene an be depited as flutuations in a fluid flow. When working with hemials as in an SCR reator, typially a high level of flow flutuations is preferable for the mixing of UWS with exhaust gases. It is ommon to define the relative turbulent intensity for the veloity as follows: u TI (9) U where u is the root-mean-square (RMS) of the turbulent veloity flutuations (U U ) at a partiular loation over a speified period of time, and U is the average of the veloity for the instantaneous value (U) at the same loation over the time period. For the same theory, the standard deviation of the temporal variation of Reynolds averaged veloity ( ) and its relative intensity ( RI / U) are adopted. In this RANS simulation, U U is the temporal variation of Reynolds averaged veloity, not the turbulent veloity flutuations beause U represents the Reynolds averaged veloity. Fig. 5 shows for 0.3s for different measuring positions and mixer geometries. For ase 1 (no mixer), there were few hanges in veloity at positions 1,, and 3. However, signifiantly inreased near the SCR filter beause a large reirulation zone formed in the diffuser region. When the flow enters the porous zone, it aligns with the hannel diretion and the higher pressure is loated around the atalyst entrane and the enter line (Chen et al., 004). The results of all ases at positions 4 and 5 exhibit a similar tendeny. At position 5, for all ases is relatively lower than that at positions, 3, and 4. Case 4 has the highest after the mixer owing to a large CRZ. This behavior an be explained as a result of the vortiity magnitude generated by the indued flow diretion of the mixers. Vortiity is a measure of the rotation of a fluid element as it moves in a field, and is defined as the url of the veloity vetor: V (10)

7 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~ Fig. 5 History of the standard deviation (σ) up to 0.3s for different measuring points and different ases. Fig. 6 Contours of different ases of (a) vortiity magnitude and the iso-surfae of vortiity att 500 1/s and (b)) pressure, both obtained with a alulation time of 1.5s. Fig. 6 shows the ontours of the vortiity magnitudee and the iso-surfae of the vortiity magnitude for all ases. Exept for ase 1, the other ases exhibit a high vortiity magnitude regardless of the mixer type. This is beause the introdued flow is onentrated at the enter region for the ase 1, while the line-type and swirl-type models disperse the flow to the wall. For ase 5 (line-type mixer), the vortiity magnitude was relatively high by the turbulent flow immediately behind the mixer. A small-sale vortex may be generated diretly behind the mixer, and vanishes rapidly after passing through it. For the swirl- type mixers, the vortiity magnitude is longer with stream-wise axis than that for the line-type mixer. In addition, the volume of the vortiity as shown in Fig. 6 inreases beause thee CRZ gradually inreases with w the vane angle. The results above indi- ate that the size and length of the CRZ are ontrolledd by the mixer shape. The swirl-type mixer reates both turbulent and

8 34 Int. J. J Nav. Arhit. Oean Eng. (014) 6:7~38 swirling flows. Therefore, it is important to selet the appropriate mixer in order to maximize floww mixing and minimize NH 3 slip. The reation for NO x redution in SCR atalysts requires a suffiient reationn time within a ertain temperature range. The NO and the NH 3 do not reat at lower temperatures and insuffiient flow mixing. NH 3 slipp is a means of NH 3 passing through the SCR reator un-reated. Fig. 7 Distributions of relativee intensity (RI)) for different measuring points and a ases using a alulation time of 1.5s. Fig. 7 presents relative intensity (RI) distributions at 1.5s for different measuring positions and mixer geometries. In alll mixer ases, there are relatively small hanges in the RI at position 1. RI substantially inreases at position after passing through the mixers. After this point, RI gradually dereases, and exhibits uniform distributions as it approahes the t SCR filter. The maximumm RI values were as follows: 1.54% at position 1 for ase 1, 1.96% at position 5 for ase,.75% att position for ase 3, 4.67% % at position for ase 4, and.99% at position for ase 5. Inorporating the mixerr is most effet tive for impro- ving mixing performane immediately behind the mixerr when the vane angle is greater than 45º. Although ase 4 possesses the highest RI for all measuring positions, the levels of the mean RI for ase 3 are onstantly maintained along the stream-wise axis from measuring positions to 3 as shown in Fig. 8. From an operational point of view v with respet to the durability and the likelihood of system failure, ase 3, a swirl-type mixer with a 45º vane angle, is more favorable thann other ases. Fig. 8 Distributions of mean RI with respet to measuring points for different ases using a alulation time of 1.5s.

9 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~38 35 Relationship between pres ssure drop and uniform floww A uniform flow and UWS distribution aross the front of the atalyst filter are neessary to obtain a maximum per- formane for de-no x and a minimum NH 3 slip in the SCR system. The flow uniformity index iss a ommonly used indiator of the flow distribution degree for after-treatment appliations (Weltens et al., 1993; Bressler et al., 1996; Girard et al., 006) ). In this study, a similar definition is adopted to evaluate the distribution degree off water onentration at a ertain loation along the pipe: UI 1 1 n n i 1 C i C C (11) where, UI is the uniformity index for the t water onentration, Ci is the loal onentration of water, C is the averagee onentrationn of water, and n is the number of ells. Generally, higher UI indiates better b fluid mixing. Fig. 9 shows the UI with respet to the alulation time (from 0.3 to 1.5s) for the stream-w wise positions 1 to 5 and for ases 1 to 5. The effets of the mixer on the UI are quite strong, regardless of the type of mixer. Compared to ase 1, the asess with mixers improve the UI by approximately 0% inn positions and 3, approximately 30% in n the position 4, and approxi- mately 5% in position 5. For all ases, the UI inreased along the stream-wise diretion. The maximum values of the UI were 98.5% for ase 4, 96% % for ase 3, 94% for ase, 93.6% for ase 5, and 90% % for ase 1. The sequenes for f the UI at different position are as follows: ase 3 > ase 5 > ase 1 > ase 4 > ase before the mixer at the position 1; ase 4 > ase 5 > ase 3 > ase > ase 1 behind the mixerr at the positionn ; ase 4 > ase 3 > ase 5 > ase > ase 1 at positions 3 and 4; and ase 4 > ase 3 > ase > ase 5 > ase 1 in the right front of the atalyst filter at the position 5. These results indiate that the UI inreasess with inreasee in the vane angle a of the swirl-type mixer, and the ases for the swirl-type mixer generally show better mixing performane than the line-type mixer, evenn though the line-type mixer outperforms itt behind the mixer and in the region of the diffuser. This superior performane is due to a small-sale vortex generated by thee up and down indued flow behind the line-type mixer that vanishes rapidly r in a short region. Conversely, for the swirl-type mixer, a large-sale vortex in terms of CRZ is generated in a long region n. Therefore, obtaining a ertain distane too mix of the UWS with the exhaust gases is important, and the swirl-type mixer is better than the line-type mixer with respet too uniform floww and the higher mixing per- formane in pratial SCR appliations. Fig. 9 History of uniformity index ( UI ) for H O from 0.3 to 1.5s for different measuring points and ases.

10 36 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~38 Mixer development generally fouses on the high effiieny of NO x redution beause of its diret impat on the design target; however, it is diffiult to determine the total performane of mixing devies based on only the UI. A more thorough understanding of the flow pattern harateristis and pressure drop is helpful in designing an SCR system with optimal performane for both NO x redution and system durability. Fig. 10 depits the relationship between the UI and pressure drop for different ases. As expeted, the ase 1 has the lowest pressure loss and UI, whih are 647 Pa and 90%, respetively. The presene of any type of mixer resulted in an improved flow mixing, but a penalty of additional pressure loss. Case 4 had the best mixing performane although the pressure drop was the worst. There is a tradeoff relationship between the UI and the pressure drop with inreasing vane angle for swirl-type mixers. Case 5 has a higher UI and a pressure drop than ase 1. However, the UI is lower and pressure drop is higher than those for the swirl-type mixers (ases -4) at the position 5, as shown in Fig. 9. Furthermore, in a omparison of ase 3 and 5, whih ontain the same angle of 45, ase 3 shows higher UI and a lower pressure drop than ase 5, as shown in Fig. 10. Therefore, it is onluded that the swirl-type mixer is more effetive than the line-type mixer with respet to the enhanement of UI and RI. When seleting a stati mixer in SCR appliations, it is neessary to onsider the mixing and the uniform distribution of the UWS in front of the SCR reator as well as the pressure drop, whih is not desirable for the optimization of engine power throughout the system. Finally, it an be onluded that the swirl-type mixer with a vane angle of 45 is more suitable model in this study based on the results of mean RI, UI and pressure drop. Further researh foused on the effet of the vane size beoming larger or smaller may be needed under line- and swirltype mixers. The effetiveness of swirl-type mixers is arefully guessed to be initiated from the larger-sale swirl than those of line-type mixers. It is beause the line-type mixer is able to generate the vane-sale swirls whereas the dut-sale swirl is produed by the swirl mixer. Nevertheless, this study may provide useful information for seleting a stati mixer in SCR appliations. Fig. 10 Relationship between uniformity index and pressure drop for different ases with a alulation time of 1.5s. CONCLUSIONS 3-D numerial simulations were performed to investigate the effets of mixer geometry on the mixing performane and the pressure drop. Flow mixing and uniformity an be greatly improved by using a stati mixer in the SCR system. Turbulent and swirling flows an also ahieve a great improvement for flow mixing with respet to flow reirulation phenomena through a longer distane. In this study, information regarding the seletion of proper stati mixers was provided based on the orrelation between the uniformity index and the pressure drop. The results show that the mixer for SCR appliations an be effetively optimized by using a well-designed mixing devie. The main results are summarized as follows:

11 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~ ) In omparison to the ase without a mixer, the ases with a mixer improve the uniformity index by approximately 0% diretly behind the mixer, approximately 30% in the diffuser, and approximately 5% immediately in front of the mixer. ) In the swirl-type mixer, the CRZ is generated diretly behind the mixer, and inreases with the vane angles of the mixer. Therefore, it is expeted that the mixing region is longer with stream-wise axis beause the swirl-type mixer reates turbulent and swirling flows. 3) The swirl-type mixer is more effetive than the line-type mixer with respet to the enhanement of mixing performane, even though there is a tradeoff relationship between the uniformity index and the pressure drop. Therefore, the swirl-type mixer with a vane angle of 45º is the most suitable model in this study based on the results of both parameters. ACKNOWLEDGEMENTS This work was supported by the National Researh Foundation of Korea (NRF) grant funded by the Korea government (MEST) through GCRC-SOP and this work was supported by the Human Resoures Development program (No ) of the Korea Institute of Energy Tehnology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Trade, Industry and Energy. REFERENCES Battoei, M., Doroudian, M. and Lain, F., 006. Optimization of exhaust gas distribution on the atalyti onverter inletone diffuser using advaned shape deformation tehnology, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Birkhold, F., Meingast, U., Wassermann, P. and Deutshmann, O., 007. Modeling and simulation of the injetion of ureawater-solution for automotive SCR DeNO x -systems. Applied Catalysis B : Environmental, 70, pp Bressler, H., Rammoser, D., Neumaier, H. and Terres, F., Experimental and preditive investigation of a lose oupled atalyti onverter with pulsating flow, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Chen, M. and Williams, S., 005. Modelling and optimization of SCR-exhaust aftertreatment systems. SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Chen, M., Aleixo, J., Williams, S., Leprine, T. and Yong, Y., 004. CFD modeling of 3-way atalyti onverters with detailed atalyti surfae reation mehanism. SAE Tehnial Paper, , Pennsylvania: 4 SAE International. DieselNet, 008. Emission Standards, IMO Marine Engine Regulations. [online] Available at: < standards/inter/imo.php > [Aessed Marh 013]. Fluent, 007. Fluent 6.3 User s Guide. Lebanon: ANSYS In. Girard, J. W., Lain, F., Hass, C.J. and Hodonsky, J., 006. Flow uniformity optimization for diesel aftertreatment systems, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Jeong, S.J., Lee, S.J., Kim, W.S. and Lee, C.B., 005. Simulation on the optimization shape and loation of urea injettion for urea-scr system of heavy-duty diesel engine to prevent NH 3 slip, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Larmi, M. and Tiainen, J., 003. Diesel spray simulation and KH-RT wave model, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Munnannur, A. and Liu, Z. G., 010. Development and validation of a preditive model for DEF injetion and urea deomposition in model SCR deno x systems, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Nguyen, T.D.B., Lim, Y.I., Eom, W.H., Kim, S.J. and Yoo, K.S., 010. Experimental and CFD simulation of hybrid SNCR-SCR using urea solution in a pilot-sale reator. Computers and Chemial Engineering, 34(10), pp NMRI, 011. National Maritime Researh Institute 011. [online] Available at: < ontent/pamphlet011_e.pdf > [Aessed November 013]. Regner, M., Östergren, K. and Trägårdh, C., 006. Effet of geometry and flow rate on seondary flow and the mixing proess in stati mixers-a numerial study. Chemial Engineering Siene, 61(18), pp

12 38 Int. J. Nav. Arhit. Oean Eng. (014) 6:7~38 Thakur, R.K., Vial, Ch., Nigam, K.D.P., Nauman, E.B. and Djelveh, G., 003. Stati mixers in the proess industries-a review. Chemial Engineering Researh and Design, 81(7), pp Turns, S.R., 000. An Introdution to ombustion: onept and appliations. nd ed. Boston: MGraw-Hill. Weltens, H., Bressler, H., Terres, F., Neumaier, H. and Rammoser, D., Optimization of atalyti onverter gas flow distributions by CFD predition, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Zhang, X. and Romzek, M., D numerial study of flow mixing in front of SCR for different injetion systems, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Zhang X., Romzek, M. and Morgan, C., D numerial study of mixing harateristis of NH 3 in front of SCR, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Zheng, G., Fila, A., Kotrba A. and Floyd, R., 010. Investigation of urea deposits in urea SCR systems for medium and heavy duty truks, SAE Tehnial Paper, , Pennsylvania: 4 SAE International. Zheng, G., Palmer, G., Salanta G. and Kotrba, A., 009. Mixer development for urea SCR appliation, SAE Tehnial Paper, , Pennsylvania: 4 SAE International.