Fan Ejector
Fan Ejector
Fan Ejector
A R T I C L E I N F O A B S T R A C T
Keywords: In the present study, an integrated micro-ejector was proposed to provide a mixture of n-butane and air for
Miniature ejector micro-combustors. The effects of fuel jet velocity and geometrical parameters of the ejector on the entrainment
Micro-combustor ratio were numerically investigated and theoretically analyzed. The results show that the incoming air mixes
Entrainment ratio
with the fuel jet flow through two routes. The entrainment ratio increases with an increasing fuel jet velocity due
Negative pressure
Geometrical parameters
to the increased magnitude of negative pressure. Moreover, increasing the nozzle diameter leads to a sharp
decrease in the entrainment ratio. However, the entrainment ratio can be drastically augmented by increasing
the diameter of the mixing section. In addition, an increase in the length of mixing section will slightly decrease
the entrainment ratio of the ejector. Furthermore, increasing the total area of air inlets gives rise to a drastic
decrease in the entrainment ratio. These variation tendencies have a close relationship with the negative pressure
contours. The findings of this study provide a guidance to the optimal design of this kind of micro-ejectors.
1. Introduction surface area of micro combustors. Karagiannidis et al. [7] and Kaisare
et al. [8] numerically studied the catalytic (Pt) combustion of CH4 and
Nowadays, various micro-electro-mechanical-systems are constantly C3H8 in micro-reactors, respectively. They confirmed that the flame
emerging. It is urgent to develop more reliable and convenient portable stability limits of catalytic micro-reactors were wider than the coun
energy sources with high power density. As is well known that con terparts of for non-catalytic systems. Moreover, they found that catalytic
ventional electrochemical batteries have short duration, long recharging micro-burners can operate with much more insulating materials and can
time, low specific energy and environmental pollution. In contrast, hy operate with lower wall temperatures. Sitzki et al. [9] and Ahn et al.
drocarbon fuels such as liquid butane possess a much higher energy [10] applied catalytic combustion into micro Swiss-roll combustors,
density and the fuel tank is easy and fast to be recharged [1,2]. Hence, which revealed that flame can be sustained at very small flow rate of
combustion-based power generators are one of the promising alterna fuel/air mixture. Norton and Vlachos [11] reported that H2 addition can
tives to batteries in the near future. For this, micro- and meso-scale assist the self-ignition of C3H8/air mixtures in catalytic micro-burners.
combustion has gained extensive attention in past decades. Chen et al. [12] demonstrated that hydrogen reaction in a
To date, premixed combustion in micro- and meso-scale combustors micro-channel can be enhanced by catalyst segmentation. Li et al. [13]
have been widely investigated. A variety of flame instabilities were improved the flame stability in a small-scale reactor with catalyst seg
found through both experimental study and numerical simulation, such mentation and cavities.
as flashback, extinction, blow-out, and repetitive flame dynamics [3–5]. Using liquid fuels for micro- and meso-scale power generators is
It is pointed out that the drastically increased heat loss ratio and more preferable since they have higher specific energy (kW/kg) than
shortened residence time of gaseous mixture are the main reasons for the gaseous fuels. Some pioneer works have been conducted by the groups
frequent occurrence of flame instabilities. Fortunately, many ap led by Dunn-Rankin [14,15] and Kyritsis [16,17]. For example, William
proaches were proposed to improve the flame stability for micro- and et al. [14] developed a miniature liquid-fuel-film (methanol) burner in
meso-scale premixed combustion, which have been summarized in a 2002. Differing from spray combustion that occurs in small volumes, the
recent review article [6]. One of the most effective flame stabilization fuel is flowed in a wall film so that it can reduce heat losses, inhibit
technologies is applying catalyst, which can make full use of the large quenching and make full advantage of the large surface-to-volume ratio
https://doi.org/10.1016/j.cep.2020.108295
Received 24 November 2020; Received in revised form 20 December 2020; Accepted 27 December 2020
Available online 2 January 2021
0255-2701/© 2020 Elsevier B.V. All rights reserved.
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
2
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
Fig. 2. Mesh generation of the prototype model (Note: the wall thickness is
not considered).
Fig. 4. Comparison of calculated entrainment ratios and experimental
data [32].
∂p
div(ρvu) = − − g + div(μgradv) (3)
∂y
∂p
div(ρwu) = − + div(μgradw) (4)
∂z
Species:
div(ρYi u) = div(Dim gradYi ) (5)
Fig. 5. Effect of fuel jet velocity on the entrainment ratio and outlet velocity.
3
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
Fig. 6. Negative pressure contours in the X-Y cross section under different fuel jet velocities.
with each other. Therefore, 580,641 grids are sufficient to capture the predicted entrainment ratios is in consistent with that of experimental
flow characteristics. The convergence of numerical simulation was results. Moreover, the relative error is less than 12.5 %. Therefore, the
judged based on the residuals of all governing equations to be less than numerical results have an acceptable accuracy.
1.0 × 10− 6.
3. Results and discussion
2.4. Model validation 3.1. Effect of fuel jet velocity on the entrainment ratio
To verify the accuracy of the numerical scheme, we compared the Fig. 5 shows the entrainment ratio and outlet velocity of the ejector
calculated entrainment ratio (defined as the ratio of inhaled air flow rate under different fuel jet velocities. It can be seen that the entrainment
to the fuel jet flow rate) of the ejector presented in [30] with their ratio exhibits a monotonic increase trend with an increasing fuel jet
experimental data, as shown in Fig. 4. It can be seen that the tendency of
Fig. 7. Air streamlines in the X-Y cross section under different fuel jet velocities.
4
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
negative regions existing before the mixing section, which are created by
the fuel jet flow. Moreover, the magnitude of negative pressure grows
larger as the fuel jet velocity is increased. This implies that the air
entrainment ability of fuel jet becomes stronger at high fuel jet velocity,
as demonstrated in Fig. 5.
Fig. 7 depicts the air streamlines under the fuel jet velocity of
V = 10 m/s, 30 m/s, 60 m/s and 90 m/s. It is interesting to observe that
the incoming air mixes with the fuel flow through two routes, as indi
cated by the orange and blue streamlines, respectively. The orange and
blue streamlines are divided by whether a streamline will be inhaled
into the front cavity of the suction section. Meanwhile, it is evident that
the orange streamlines are denser than the blue ones, which means that
the air is inhaled largely through the orange routine. This is because the
magnitude of negative pressure in the vicinity of the nozzle exit is much
larger than that in other areas. However, it is also noted that at
V = 10 m/s, there are two large vortexes located at the right-hand side of
the air inlets, where a large amount of air will leak out again from the
ejector. Obviously, this is one of the main reasons for the small air
entrainment ratio under low fuel jet velocity (Note: Another reason is
Fig. 8. Variation of the entrainment ratio with the nozzle diameter under the small magnitude of negative pressure near the nozzle exit). On the
different fuel jet velocities. other hand, no vortex is formed in the corners on both sides of the nozzle
in this case. As the fuel jet velocity is increased, the size of the vortexes
velocity. Specifically, the entrainment ratio increases more rapidly when that near the air inlets decreases significantly and disappears at
the fuel jet velocity is relatively low, and slows down at high fuel jet V = 90 m/s. At the meantime, two small vortexes appear in the corners
velocity. This means that the entrainment ratio can be significantly that near the nozzle, which indicates that the ability of air entrainment
enhanced by increasing the fuel jet velocity when it is below 50 m/s. On becomes stronger at large fuel jet velocity. Consequently, the air
the other hand, it is noted that the outlet velocity increases almost lin entrainment ratio increases monotonically with an increasing fuel jet
early with the fuel jet velocity. Moreover, the outlet velocity are much velocity, as have been demonstrated in Fig. 5.
larger than the laminar burning velocity of stoichiometric n-butane-air
mixture (⁓0.4 m/s). Therefore, in order to sustain a stable combustion, 3.2. The impacts of geometrical parameters of the ejector on the
the diameter of downstream combustor should be several times greater entrainment ratio
than that of the ejector. In the following, we will analyze the pressure
field and velocity field to discuss the effect of fuel jet velocity on the The entrainment ratio not only depends on the fuel jet velocity, but
entrainment ratio. also is associated with the geometrical parameters of the ejector. In the
Fig. 6 illustrates the negative pressure contours in the X–Y cross following sub-sections, we will examine the impact of each geometrical
section under the fuel jet velocity of V = 10 m/s, 30 m/s, 60 m/s and parameter on the entrainment ratio, including the nozzle diameter, the
90 m/s. It is clearly seen from Fig. 6 that there are two symmetric diameter and length of the mixing section, and the number and total
area of air inlets. In each case, except for the variable parameter, all the
Fig. 9. Negative pressure contours in the X-Y cross section under different nozzle diameters at V = 60 m/s.
5
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
Fig. 10. Entrainment ratio calculated by Eq. (6), η, versus that by numerical
simulation, ηNum.
Fig. 11. Variation of the entrainment ratio and equivalence ratio at the ejector
exit with the diameter of mixing section under V = 60 m/s.
3.2.1. Effect of nozzle diameter on the entrainment ratio Fig. 12. Negative pressure contours in the X-Y cross section under different
To investigate the influence of nozzle diameter on the entrainment mixing section diameters at V = 60 m/s.
ratio, we chose seven values for the nozzle diameter, i.e., D1 = 0.1, 0.13,
0.15, 0.17, 0.19, 0.21 and 0.25 mm, respectively. The variation of through the second routine and also slow down the flow of fuel-air
entrainment ratio with the nozzle diameter under different fuel jet ve mixture. Hence, the air entrainment ratio will be decreased as the
locities are shown in Fig. 8. First, it is seen that for the same nozzle nozzle is enlarged.
diameter, the entrainment ratio increases with the increase of fuel jet In order to provide a guidance for the ejector design, we obtained an
velocity, which is consistent with that demonstrated in Fig. 5. However, empirical correlation for the entrainment ratio, which is a function of
it can also be seen from Fig. 8 that for a fixed jet velocity, the entrain nozzle diameter and fuel jet velocity, as shown in Eq. (6):
ment ratio declines sharply with the increase of nozzle diameter. For
example, when V = 100 m/s, as the nozzle diameter is increased from η = D−1 0.951 (0.0051V + 2.068) (6)
0.1 mm to 0.25 mm, the entrainment ratio decreases by 56 %.
The prediction accuracy of Eq. (6) has been verified by a comparison
In order to analyze the influence of nozzle diameter on entrainment
with the numerical results, as shown in Fig. 10. It is evident that all the
ratio, the negative pressure contours under D1 = 0.1 mm, 0.15 mm,
data fall within the two error lines of +10 % and -10 %, which confirms
0.19 mm and 0.25 mm at V = 60 m/s are shown in Fig. 9. Because the
that the prediction by Eq. (6) has an acceptable accuracy.
volumetric fuel flow rate is proportional to the square of nozzle diam
eter, both the magnitude and area of the negative pressure contours will
3.2.2. Effect of the diameter of mixing section on the entrainment ratio
be drastically increased, extending to the mixing section when
In this section, the influence of the mixing section diameter, D3, on
D1>0.1 mm. More importantly, the pressure gradient in the mixing
the entrainment ratio is studied. Nine values of D3 are selected for
section is positive under D1>0.1 mm, which will retard the air suction
comparison, i.e., D3 = 1.38 mm, 1.5 mm, 1.6 mm, 1.7 mm, 1.8 mm,
6
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
Fig. 14. Variation of the entrainment ratio with the length of mixing section.
Fig. 13. Air streamlines in the X-Y cross section under different mixing sec
tion diameters.
1.9 mm, 2.0 mm, 2.2 mm and 2.8 mm. The numerical results of the
entrainment ratio and equivalence ratio at the ejector exit for various
mixing section diameters under V = 60 m/s are shown in Fig. 11. It is
obvious that the entrainment ratio increases almost linearly with the
increase of the mixing section diameter. The empirical correlation be
tween the entrainment ratio and the mixing section diameter can be
expressed through Eq. (7). Moreover, it is noted that when D3 = 2.8 mm,
the entrainment ratio reaches up to 32.05, which can satisfy the
requirement of entrainment ratio for a complete combustion of n-butane
in air. Accordingly, the equivalence ratio at the ejector exit drops from
around 2.4 to 0.98.
η = 13.716D3 − 5.769 (7)
Fig. 15. Negative pressure contours in the X-Y cross section under different
Figs. 12 and 13 respectively illustrate the negative pressure contours mixing section lengths at V = 60 m/s.
and air streamlines in the X–Y cross section under different mixing
section diameters (D3 = 1.38, 1.8 and 2.2 mm) at V = 60 m/s. It can be
seen that with the increase of mixing section diameter, both the area and
7
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
Fig. 16. Comparison of the entrainment ratios of ejectors with dual and four air
inlets under different fuel jet velocities.
8
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
4. Conclusions
(1) The incoming air mixes with the fuel flow through two routes,
which can be divided by whether a streamline is inhaled into the
Fig. 18. Comparison of the entrainment ratio variation of n-butane and DME
former part of the suction section.
with fuel jet velocity. (2) The entrainment ratio increases with an increasing fuel jet ve
locity due to the increased magnitude of negative pressure.
(3) The most effective parameters to promote the entrainment ratio
are the diameters of the nozzle and the mixing section. Specif
ically, reducing the nozzle diameter or increasing the mixing
section diameter can augment the entrainment ratio drastically.
(4) Changing the length of mixing section has a slight effect on the
entrainment ratio of the ejector.
(5) Increasing the total area of air inlets leads to a drastic decrease in
the entrainment ratio.
Acknowledgement
References
9
G. Yang and A. Fan Chemical Engineering and Processing - Process Intensification 160 (2021) 108295
[11] D.G. Norton, D.G. Vlachos, Hydrogen assisted self-ignition of propane/air mixtures [24] S. Li, J. Yan, Z. Liu, Y. Yao, X. Li, N. Wen, G. Zou, Optimization on crucial ejector
in catalytic microburners, Proc. Combust. Inst. 30 (2005) 2473–2480. geometries in a multi-evaporator refrigeration system for tropical region
[12] G. Chen, Y. Chao, C. Chen, Enhancement of hydrogen reaction in a micro-channel refrigerated trucks, Energy 189 (2019) 116347.
by catalyst segmentation, Int. J. Hydrogen Energy 33 (2008) 2586–2595. [25] A.M. Abed, M.A. Alghoul, A.N. Al-Shamani, K. Sopian, Evaluating ejector efficiency
[13] Y. Li, G. Chen, F. Wu, T. Cheng, Y. Chao, Combustion characteristics in a small- working under intermediate pressure of flash tank-absorption cooling cycle:
scale reactor with catalyst segmentation and cavities, Proc. Combust. Inst. 34 parametric study, Chem. Eng. Process.: Process Intensif. 95 (2015) 222–234.
(2013) 2253–2259. [26] H. El-Dessouky, H. Ettouney, I. Alatiqi, G. Al-Nuwaibit, Evaluation of steam jet
[14] W.A. Siringnano, T.K. Pham, D. Dunn-Rankin, Miniature-scale liquid-fuel-film ejectors, Chem. Eng. Process.: Process Intensif. 41 (2002) 551–561.
combustor, Proc. Combust. Inst. 29 (2002) 925–931. [27] W. Presz Jr, G. Reynolds, C. Hunter, Thrust Augmentation With mixer/ejector
[15] Y.H. Li, Y.C. Chao, N.S. Amade, D. Dunn-Rankin, Progress in miniature liquid film Systems, AIAA, 2002, 2002-0230.
combustors: double chamber and central porous fuel inlet designs, Exp. Therm. [28] S. Weber, S. Schaepe, S. Freyer, M. Kopf, C. Dietzsch, Effect of ejector operation on
Fluid Sci. 32 (2008) 1118–1131. the oxygen transfer in a pilot jet loop reactor, Chem. Eng. Process.: Process Intensif.
[16] D.C. Kyritsis, I. Guerrero-Arias, S. Roychoudhury, A. Gomez, Mesoscale power 131 (2018) 43–50.
generation by a catalytic combustor using electrosprayed liquid hydrocarbons, [29] F. Rahman, D.B. Umesh, D. Subbarao, M. Ramasamy, Enhancement of entrainment
Proc. Combust. Inst. 29 (2002) 965–972. rates in liquid-gas ejectors, Chem. Eng. Process.: Process Intensif. 49 (2010)
[17] D.C. Kyritsis, S. Roychoudhury, C.S. Mcenally, L.D. Pfefferle, A. Gomez, Mesoscale 1128–1135.
combustion: a first step towards liquid fueled batteries, Exp. Therm. Fluid Sci. 28 [30] Y. Yang, W. Du, T. Ma, W. Lin, M. Cong, H. Yang, Z. Yu, Numerical studies on
(2004) 763–770. ejector structure optimization and performance prediction based on a novel
[18] D.C. Kyritsis, B. Coriton, F. Faure, S. Roychoudhury, A. Gomez, Optimization of a pressure drop model for proton exchange membrane fuel cell anode, Int. J.
catalytic combustor using electrosprayed liquid hydrocarbons for mesoscale power Hydrogen Energy 45 (2020) 23343–23352.
generation, Combust. Flame 139 (2004) 77–89. [31] W.G. Gardner, I. Wang, J.W. Jaworski, N.A. Brikner, J.M. Protz, Experimental
[19] J. Chen, X.F. Peng, Z.L. Yang, J. Cheng, Characteristics of liquid ethanol diffusion investigation and modeling of scale effects in jet ejectors, J. Micromech. Microeng.
flames from mini tube nozzles, Combust. Flame 156 (2009) 460–466. 20 (2010) 085027.
[20] Y.H. Gan, J.L. Xu, Y.Y. Yan, M. Wang, Y.L. Luo, Z.L. Yang, A comparative study on [32] Y. Zhu, W. Cai, C. Wen, Numerical investigation of geometry parameters for design
free jet and confined jet diffusion flames of liquid ethanol from small nozzles, of high performance ejectors, Appl. Therm. Eng. 29 (2009) 898–905.
Combust. Sci. Technol. 186 (2014) 120–138. [33] D. Satoh, S. Tanaka, K. Yoshida, M. Esashi, Micro-ejector to supply fuel–air mixture
[21] Y.H. Gan, Z.B. Luo, Y.P. Cheng, J.L. Xu, The electro-spraying characteristics of to a micro-combustor, Sens. Actuator A-Phys. 119 (2005) 528–536.
ethanol for application in a small-scale combustor under combined electric field, [34] S. Tanaka, K.S. Chang, K.B. Min, D. Satoh, K. Yoshida, M. Esashi, MEMS-based
App. Therm. Eng. 87 (2015) 595–604. components of a miniature fuel cell/fuel reformer system, Chem. Eng. J. 101
[22] Y.H. Gan, Y. Tong, Y.G. Ju, X. Zhang, H.G. Li, X.W. Chen, Experimental study on (2004) 143–149.
electro-spraying and combustion characteristics in meso-scale combustors, Energy [35] Y. Fan, Y. Suzuki, N. Kasagi, Development of a large-entrainment-ratio
Convers. Manage. 131 (2017) 10–17. axisymmetric supersonic ejector for micro butane combustor, J. Micromech.
[23] B.M. Tashtoush, M.A. Al-Nimr, M.A. Khasawneh, A comprehensive review of Microeng. 16 (2006) 211–219.
ejector design, performance, and applications, Appl. Energy 240 (2019) 138–172. [36] W.G. Gardner, J.W. Jaworski, A.P. Camacho, J.M. Protz, Experimental results for a
microscale ethanol vapor jet ejector, J. Micromech. Microeng. 20 (2010) 045019.
10