- Email: [email protected]
Effects of hydrogen addition and Carbone dioxide dilution on the velocity field in non reacting and reacting flows
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Effects of hydrogen addition and Carbone dioxide dilution on the velocity field in non reacting and reacting flows Ibtissem Hraiech a,b, Jean-Charles Sautet b, Mohamed Ali Mergheni a,b,*, Hmaied Ben Ticha a, Hazem Touati a, Abdallah Mhimid a a b
LESTE, Ecole Nationale d'Ingenieurs de Monastir, Tunisia CORIA UMR6614 e CNRS, Universite et INSA de Rouen, Saint Etienne du Rouvray, France
article info
abstract
Article history:
The evolution of anti-pollution standards and the optimization of combustion efficiency
Received 18 July 2014
push the development of new fuels with high energy efficiency. It is necessary to develop
Received in revised form
new alternative fuel to improve the efficiency of conventional systems, reduce emissions
17 September 2014
(NOx, SOx, soot particles) and recover for its materials. A new fuel called bio-hythane, a
Accepted 23 September 2014
mixture of natural gas up to 20% hydrogen and up to 50% Carbone dioxide, from the re-
Available online 16 October 2014
covery of the waste from households and agriculture, via suitable digesters provides a source of renewable energy and usable, is a very interesting solution to improve emission
Keywords: Hythane
standards and optimization of the combustion chambers. This experimental study is led in a non-reacting configuration and in combustion in
Velocity field
order to focus on the effects of hydrogen addition and CO2 dilution in the fuel on the ve-
Hydrogen
locity profiles, the turbulence intensity and the turbulent kinetic energy. From PIV mea-
CO2 dilution
surements the results show that from the velocity and fluctuations profiles the high
Turbulent kinetic energy
diffusivity and the low density of hydrogen allow bio-hythane jet to spread more efficiency in the combustion. The combustion decreases the entrainment of the ambient fluid and raises the viscosity of the flow, leading to an increase of the longitudinal velocity along the bio-hythane jet and a reduction of the turbulence in comparison with non-reacting configuration. The Carbone dioxide addition reduces the fall of the turbulence because the temperature of the flame is less important. Effects of hydrogen and CO2 are also highlighted by analysis of turbulent kinetic energy along the bio-hythane. Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
nieurs de Monastir, Rue Ibn Aljazzar, 5000 Monastir, Tunisia. * Corresponding author. Ecole Nationale d'Inge E-mail addresses: [email protected], [email protected] (M.A. Mergheni). http://dx.doi.org/10.1016/j.ijhydene.2014.09.129 0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Introduction Regulations on protection of the environment lead to develop new technologies of combustion. A new alternative fuel, mixture of natural gas (NG), hydrogen up to 20% in volume and carbon dioxide up to 50% in volume, called bio-hythane, has been created from the recovery of the waste from households and agriculture, via suitable digesters. It provides a source of renewable energy and usable, generates a better thermal efficiency and a reduction of pollutant emissions and is expected to play an important role in future energy production. Due to the properties of hydrogen, specially low density, high molecular diffusivity, wide flammability limits, high flame speed, and low ignition energy (Choudhuri and Gollahalli [1,2]) hydrogen in the fuel permits combustion systems to operate with lean fuel mixtures. Choudhuri and Gollahalli [2] carried out an experimental investigation on turbulent NGeH2 jet diffusion flame and showed a reduction in the soot concentration and emission index of CO (EICO), but an increase in NO and NOx emissions with the addition of hydrogen to the fuel. El Ghafour et al. [3] studied experimentally the effect of hydrogen addition on combustion characteristics of NGeH2 hybrid fuel turbulent diffusion flame at a fixed Reynolds number (4000). They observed that the addition of hydrogen improves the flame stability, reduces the flame length for relatively high hydrogen concentrations and increases the NO and CO concentration. The influence of hydrogen addition to natural gas on the flow dynamics was investigated experimentally in nonreacting flow and in combustion by Yon and Sautet [4]. Their results show that the combustion decreases the entrainment of the ambient fluid, increases the temperature in the fuel jet and consequently the viscosity of the flow, leading to an increase of the longitudinal velocity along the hythane jet and a decrease of the turbulence in comparison with non-reacting configuration. These effects are more and more important with hydrogen addition due to the decrease of the flame liftoff height and the increase of the heat release rate. Takagi et al. [5] carried out an experimental investigation on turbulent jet flows with and without flame, and they found that the presence of flame decreases the entrainment of ambient fluid, which accelerates the flow. The chemical effect of CO2 replacement of N2 in air on the burning velocity of CH4 and H2 flames was studied numerically by Liu et al. [6]. The relative importance of the chemical effect of CO2 on the burning velocity increases as more CO2 is added to replace N2 in air. Previous studies have proved that the soot formation in diffusion flames were decreases by CO2 addition to the coflow air, resulting from the short residence time in the inception region [7e11]. It was shown that CO2 dilution thermally and chemically limits the formation of soot precursors due to the decrease of H radicals consumed in the reaction CO2 þ H / CO þ OH. The effects of different additives to air on the lift-off of a laminar CH4/air diffusion flame have been explored experimentally by Guo et al. [12]. They observed that the addition of CO2 causes flame lift-off due to the dilution, thermal and
19819
chemical effects. The dilution effect being the most significant one, followed by the thermal effect. The three effects tend to reduce combustion intensity and cause flame to be lifted. The radiation and transport property effects are negligible. The effect of CO2 addition to the air on the transition from an attached flame to a lifted flame issued from a coaxial nonpremixed methane-air jet is studied experimentally by Min et al. [13]. Results shows that the CO2 is the best destabilize among the diluents, because the three effects (dilution > thermal > chemistry) induce loss of flame stability (CO2 has a strongest ability to break flame stability, than N2). Dally et al. [14] investigated experimentally and numerically that CO2 addition in a non-premixed methane/air flame lowered flame temperature by decreasing reactant concentration inside the reaction zone. The present study is conducted in non-reacting and reacting flow from a cylindrical burner to investigate the effect of the hydrogen addition and CO2 dilution on the flow aerodynamics. The mole fraction of hydrogen (% H2) in NGeH2 mixture varies from 0% to 20% and the mole fraction of carbon dioxide volume varies from 0% to 50%. The experimental setup consists of a 15 kW burner powered by natural gas added to hydrogen and carbon dioxide. A study of the jet aerodynamic through Particle Image Velocimetry (PIV) in non-reacting flow and in combustion allows characterizing the flow fields. The evolution of the root mean square (RMS) of the two components of velocity is investigated to deduce the effects of CO2 dilution and the hydrogen addition on the turbulence intensity and the turbulent kinetic energy.
Fig. 1 e . Sketch of the burner.
19820
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Table 1 e Gas flow characteristics. Jet Flame mixture Fuel mixture QvNG QvH2 Qvfuel %CO2 QxCO2 Qtot Exit % NG % H2 Jet reynolds pjet Jet Schmidt %NG %H2 (1/min) (1/min) velocity (total (total number number (fuel) (fuel) (1/min) (1/min) (1/min) U0 (m/s) mixture) mixture) 100
0
24.09
0
24.09
95
5
23.73
1.25
24.98
90
10
23.34
2.59
25.94
85
15
22.92
4.046
26.97
80
20
22.47
5.61
28.09
0 10 20 0 10 20 0 10 20 0 10 20 30 0 10 20 30
0 2.6 6 0 2.7 6.2 0 2.8 6.4 0 2.9 6.7 11.5 0 3.1 7 12
24 26.7 30.1 24.9 27.7 31.2 25.9 28.8 32.4 26.9 29.9 33.7 38.5 28.1 31.2 35.1 40.1
Thanks to PIV measurements in non-reacting flow and in combustion, the effect of combustion on the average 2D normalized longitudinal velocity fields Uz =U0 in the XZ-plane, are investigated. The evolution of the turbulence intensity along the bio-hythane jet centerline as a function of the hydrogen volume fraction and CO2 dilution rate is investigated in non-reacting fuel.
Experimental setup and measurement technique
14.2 15.7 17.7 14.7 16.3 18.4 15.3 16.9 19.1 15.9 17.6 19.8 22.7 16.5 18.4 20.7 23.6
6312 9005 10,741 7518 8959 10,769 7422 8914 10,804 7322 8868 10,847 13,486 7205 8822 10,899 13,713
0.83 0.93 1.03 0.79 0.9 1 0.75 0.86 0.97 0.71 0.83 0.95 1.06 0.66 0.78 0.9 1.02
The present study aims to investigate the effects of the hydrogen addition in the fuel and the effect of CO2 dilution on
0 0 0 5 4.5 4 10 9 8 15 13.5 12 10.5 0.2 0.18 0.16 0.14
100 90 80 95 85.5 76 90 81 72 85 76.5 68 59.5 80 72 64 56
the flow aerodynamics. The burner depicted in Fig. 1 consists of a jet of bio-hythane (mixture CH4/H2/CO2). The cylindrical burner, of internal diameter d ¼ 6 mm, brings the mixture CH4/ H2/CO2 into the ambient air. The natural gas has a density of 0.83 kg/m3 and a volume composition of 85% CH4, 9% C2H6, 3% C3H8, 2% N2, 1% CO2, and traces of higher hydrocarbon species. The hydrogen volume fraction (aH2 ) in the fuel mixture (NG þ H2) varies between 0 and 20%. The fuel volumetric flow rate is: Qvfuel ¼ QVGN þ QvH2
Experimental setup
0.8 0.86 0.91 0.75 0.78 0.83 0.71 0.73 0.77 0.67 0.69 0.73 0.79 0.65 0.68 0.7 0.76
(1)
CO2 is introduced in the fuel mixture to study the effect of CO2 dilution. The percentage of CO2 (aCO2 ) in the mixture (NG þ H2 þ CO2) varies from 0% (no dilution) to 50%.
Fig. 2 e Particle image velocimetry experimental setup.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Table 2 e Parameters of PIV setup. Non-Reacting flow Dt (ms) 15 Magnification (mm pixel1) Explored height 0e100 in the jet (mm)
40
100e200
Reacting flow
18 0.0051
45
65
0e100 100e200 200e300
The CO2 flow rate is QvCO2 and the total jet flow rate is: Qtot ¼ Qvfuel þ QvCO2
(2)
The regulation of natural gas flow is realized by a mass flow regulator. Hydrogen and carbon dioxide flow rates are controlled by sonic nozzles and pressure regulator. Table 1 Summarizes the different parameters of the experiment including natural gas, hydrogen, Carbone dioxide and total jet flow rates, the jet exit velocity, U0, the bulk jet Reynolds Number (Re ¼ U0d/n) as well as the Schmidt number, Sc ¼ n/D, where D is the molecular diffusivity.
Measurement technique To study the aerodynamics of the flow according to the hydrogen concentration and the dilution of CO2, the Particle Image Velocimetry (P.I.V) is used. This non-intrusive method, permits to procure 2D images of the flow and the instantaneous two-dimensional velocity fields, the experimental P.I.V setup described in Fig. 2 needs a light source (120 Mj/pulse, pulse duration of 8 ns), in this case a double-pulsed NdeYag laser (Big Sky CFR200, Quantel) with a wavelength of 532 nm and a frequency of 10 Hz.
19821
An optical system consisting of three consecutive lenses creates the laser sheet with 500 mm thickness. A CCD camera (2048*2048 pixel2, Image Pro X Lavision) with a dynamic range of 14 bits is oriented perpendicularly to the laser sheen. A convergent lens (f ¼ 85 mm), placed perpendicularly to the light source, collects the signal of Mie scattering emitted by the seeded particles. Oil particles (mean diameter of 3 mm) and Zirconium Oxide particles (mean diameter of 10 mm) are used respectively in non-reacting and reacting configurations. An interference filter is used to reject the bright luminosity in front of the lens of the PIV setup. The magnification is 0.051 mm pixel1 and the time interval between two images varies according to the investigated configurations (Table 2). From recording 500 instantaneous pairs of images, with a resolution of 2048*2048 pixels2, a statistical processing enables to obtain the mean velocity field, and the fluctuations for each height and configuration. Fig. 3 shows the mean twodimensional longitudinal velocity field in non-reacting flow and the characteristic zones of a flow for aH2 ¼ 20% and aCO2 ¼ 20%. This study focused on the effect of the hydrogen addition and the effect of CO2 dilution on dynamic fields. In order to investigate all the velocity components in the upstream of the flow, it was necessary to used Particle Image Velocimetry. Fig. 4 displays the images of the seeded particles in non reacting flow for a hydrogen volume fraction equals to 0 and Carbone dioxide Carbone volume fraction equals to 0. This figure shows instantaneous, mean and RMS velocity fields. This allows to study the velocity radial component and the longitudinal component of the velocity and RMS radial component and the RMS longitudinal component.
Results and discussion Mean velocity field
Fig. 3 e Mean two-dimensional longitudinal velocity fields in non-reacting flow (aH2 ¼ 20%, aCO2 ¼ 20%).
The longitudinal velocity profiles allow studying the effects of the hydrogen addition and the dilution of CO2 on the dynamics of the flow (Fig. 5). Fig. 5(a) displays the radial profiles of the mean longitudinal jet velocity Uz normalized by the exit velocities U0 in non-reacting flow, without CO2 dilution (aCO2 ¼ 0%), with the hydrogen fraction from aH2 ¼ 0% to aH2 ¼ 20% for three different heights, for Z ¼ 10, 50 and100 mm. This figure shows that with hydrogen addition the maximum of the normalized longitudinal velocity of the =U0 ¼ 1:26 to bio-hythane jet grows by 29% (from Umax z Umax =U ¼ 1:63 with a ¼ 20%) near the jet exit (Z ¼ 10 mm). 0 H2 z =U0 ¼ 0:71 to This difference, at Z ¼ 50 mm, is 18% (from Umax z =U ¼ 0:84 with a ¼ 20%) and 13% downstream Umax 0 H 2 z =U0 ¼ 0:38 to Umax =U0 ¼ 0:43 with (Z ¼ 100 mm) (from Umax z z aH2 ¼ 20%). From this figure it can be seen that the difference =U0 between the configuration of maximum velocity Umax z aH2 ¼ 0% and aH2 ¼ 20% is less important far from the burner due to the high molecular diffusivity and the low density of hydrogen, favoring the entrainment with the ambient. The effect of carbon dioxide dilution on the normalized mean velocity is presented on Fig. 5 (b) for Z ¼ 10, 50 and100 mm in the configuration without H2 addition (aH2 ¼ 0%). This figure shows that, with CO2 dilution, Umax =U0 z
19822
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 4 e a) Images of the seeded particles, b) Normalized mean velocity fields and c) Normalized RMS velocity fields of the component Uz in non-reacting flow for aCO2 ¼ 0% and aH2 ¼ 0%.
decays only by 11% for the three heights in the jet. For =U0 ¼ 0:71 for aCO2 ¼ 0% and 0.65 example, at z ¼ 50 mm, Umax z for aCO2 ¼ 20%. The combined effects of H2 and CO2 is highlighted comparing Fig. 5(b) and (c) The difference of maximum ve=U0 without hydrogen, between the configuration locity Umax z aCO2 ¼ 0% and aCO2 ¼ 20% is less important than with aH2 ¼ 20%, for example, at z ¼ 10 mm, with aH2 ¼ 0% this difference is equals to 11% and 15.4% with aH2 ¼ 20% (Fig. 5(c)). From the averaged velocity profiles in non-reacting flows, the evolution of the mean centerline jet velocity, Ucl, can be deduced. Fig. 6 shows the inverse of Ucl normalized by the jet 0 exit velocity U0 as a function of the reduced coordinate ZZ deff with the % CO2 added at various % H2 where deff is defined by: deff ¼ d
rext exp 0:5 rjet
(3)
From this figure, it can be seen that the rate of decrease of the mean longitudinal velocity Ku increase with the addition of CO2 and decreases when adding the hydrogen this is due to the low density. It can be noted that the rate of decrease of the mean longitudinal velocity along a free jet is inversely proportional to the density of the jet, this result depicts a good agreement with the available data in the existing literature [16]). For example, in the configura(Chassaing [15] and Page tion where 20% of hydrogen is added to air, the rate of decrease of the mean longitudinal velocity passes from Ku ¼ 3.36 for 0% of Carbone dioxide, to Ku ¼ 5.05 when the percentage of Carbone dioxide is 20%. Also, in the configuration where 0% of carbon dioxide is added to air, the rate of decrease of the mean longitudinal
velocity passes from Ku ¼ 5.76 for 0% of hydrogen, to Ku ¼ 3.36 when the percentage of hydrogen is 20%. Fig. 7 compares the average 2D normalized longitudinal velocity fields Uz/U0 in the XZ-plane for aH2 ¼ 0% and aH2 ¼ 20% (aCO2 ¼ 0%) and for aCO2 ¼ 0% and aCO2 ¼ 20% (aH2 ¼ 0%), obtained in non-reacting flow and with combustion. In non-reacting flow and in combustion, Fig. 7(a) shows that the hydrogen addition rises the velocities along the biohythane jet as the normalized longitudinal velocities are lower with aH2 ¼ 0% in comparison with aH2 ¼ 20%. Concerning the fields in combustion, the longitudinal velocity of the bio-hythane jet increases and the dynamic greatly differs from non-reacting flow. The heat release modifies the entrainment, the viscosity and mixing of the flow, and accelerates the bio-hythane jet. In non-reacting flow and in combustion, Fig. 7(b) shows that, without hydrogen, the Carbone dioxide addition decreases the velocities along the bio-hythane jet. Fig. 7(c) shows the combined effects of H2 and CO2. In combustion, for aH2 ¼ 20%, the bio-hythane jet is accelerated and the dynamics differs from non-reacting flow. It can be noted that in combustion the effect of CO2 is offset with the hydrogen addition due to the low density induced by addition of H2.
Longitudinal velocity fluctuations From the PIV measurement, the longitudinal velocity standard deviation U0z is studied in the XZ-plane. Fig. 8 shows the longitudinal velocity fluctuations of the bio-hythane jet in non-reacting flow. The figure displays a minimum
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
19823
Fig. 5 e Radial profiles of normalized mean longitudinal velocity for Z ¼ 10, 50, 100 mm, in non-reacting flow. a-As a function of the hydrogen volume fraction (for aCO2 ¼ 0%). b-As a function of the carbon dioxide volume fraction (for aH2 ¼ 0%). c-As a function of the carbon dioxide volume fraction (for aH2 ¼ 20%).
19824
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 6 e Normalized mean longitudinal velocity along the jet centerline, in non-reacting flow. a-As a function of the hydrogen volume fraction (for aCO2 ¼ 0%). b-As a function of the carbon dioxide volume fraction (for aH2 ¼ 0%). c-As a function of the carbon dioxide volume fraction (for aH2 ¼ 20%).
fluctuations zone at the center of the jet and a maximum in the shear layers. With Hydrogen, this maximum is reached at about Z ¼ 5d to Z ¼ 6d from the nozzle. Low fluctuations zone at the nozzle exit corresponds to the potential core. In the shear layers, the high fluctuations increase with the jet development, until the mixing layers join. With Carbon dioxide addition, this maximum is located about Z ¼ 6d from the nozzle and the maximum level of fluctuations high fluctuations decreases, due to the higher jet density and the decrease of the ambient air entrainment
(Sautet [17]). Also, it can be noted that the normalized longitudinal velocity RMS U0z =U0 in non-reacting flow and the expansion of the jet increase with addition hydrogen due to the low density of hydrogen. However, with aH2 ¼ 20%, the expansion of the jet is lower in the configurations without CO2 in comparison with aCO2 ¼ 20%.The benefic effect of H2 on the mixing is balanced with the CO2 dilution. Fig. 9 shows the evolution of the normalized RMS longitudinal velocity along the jet centerline in non-reacting and reacting flows.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
19825
Fig. 7 e Effect of combustion on the average 2D normalized longitudinal velocity fields UZ =U0 in the XZ-plane, in nonreacting flow and in combustion.
19826
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 8 e Normalized RMS of longitudinal jet velocity U0Z =U0 in non-reacting flow.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 9 e Normalized RMS velocity along the bio-hythane jet centerline in non-reacting and reacting flows.
19827
19828
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 10 e Evolution of the turbulence intensity, U0Z =UZ along the jet centerline.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 11 e Comparison of the two components of the normalized velocity fluctuations along the jet centerline.
19829
19830
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
It can be seen from this figure that for a pure natural gas jet, in the potential core (Z < 5d) the turbulence rates are similar without reaction and in combustion but from Z ¼ 5d, the turbulence is higher is non-reacting configuration than in combustion, in fact, the flame increases the temperature in the fuel jet and consequently the viscosity, leading to a decrease of the normalized velocity fluctuations. With the hydrogen addition, the turbulence is higher is non-reacting configuration than in combustion however from Z ¼ 7d the longitudinal fluctuations along the bio-hythane jet are higher in combustion than in non-reacting configuration. This is due to the high longitudinal velocity in combustion and the low density induced by addition of H2, which remains a high level of turbulence. With the addition of CO2, the turbulence is higher is nonreacting configuration than in combustion due to the high density of CO2, and this fall of the turbulence is reduced.
Turbulence intensity From the mean and RMS longitudinal velocity, it is possible to deduce the evolution of the turbulence intensity along the fuel jet U0z =Uz . Fig. 10 shows the evolution of the turbulence intensity along the bio-hythane jet according to the hydrogen addition and the carbon dioxide addition. At the exit of the nozzle (Z ¼ 0d), the turbulence intensity is around 5%. At Z/d ¼ 16, it reaches 30% in non-reacting configurations and 25% in combustion, values published in the literature [18]. Without hydrogen and Carbone dioxide, the turbulence intensity is similar in the 2 configurations from the exit of the burner to Z ¼ 5d. After this characteristic height corresponding to the end of the potential core, the turbulence intensity is higher in non-reacting flow than with the flame. The evolution of the turbulence intensity in combustion differs from non-reacting flow as and when the flame and the temperature perturb the flow, because of the heat release rate which perturbs the flow. The flame increases the temperature in the fuel jet and consequently the viscosity, leading to a decrease of the turbulence intensity. This fall of the turbulence is intensified with the hydrogen addition because the temperature of the flame and the burnt gases at the flame bottom are more important and this fall is reduced because the temperature of the flame is less important with the carbon dioxide addition.
Turbulent kinetic energy In non-reacting flow, from the measurement of the RMS of the two components of velocity ðU0X and U0Z Þ, it is possible to deduce the evolution of the turbulent kinetic energy k: k ¼ 0:5 U0Z exp 2 þ 2 U0X exp 2
(4)
Fig. 11 displays the evolution of the two components of velocity standards deviation ðU0X ; U0Z Þ normalized by the exit velocity U0 along the bio-hythane jet according to aH2 and aH2 . With hydrogen addition, the velocity and the fluctuation increase in the flow along the jet and decreases with carbon dioxide dilution. The longitudinal fluctuations are more important than radial ones corresponding to the classical behavior of a jet. From the measurement of the RMS of the two components of velocity, the turbulent kinetic energy can be deduced. Fig. 12 shows the turbulent kinetic energy, k, normalized by the squared of exit velocity U20 , along the centerline of the biohythane jet according to the hydrogen volume fraction (aH2 ¼ 0% and aH2 ¼ 20%) and the dilution rate in carbon dioxide (aCO2 ¼ 0%, and aCO2 ¼ 20%). With hydrogen addition, the jet interacts with the surrounding fluid more upstream in the flow and the maximum peak of energy appears earlier in the jet. As an example, without CO2, this maximum peak appears at Z ¼ 5.5d and equals to 3.6% for aH2 ¼ 0%, whereas for aH2 ¼ 20%, the maximum of k appears for Z ¼ 4.5d and equals to 5.7%.The growth of the turbulent kinetic energy with hydrogen addition is due to the increase of the fluctuation of the two components of velocity with hydrogen addition. The study of the turbulent kinetic energy according to the carbon dioxide shows that, for aH2 ¼ 0%, the normalized turbulent kinetic energy decreases with aCO2 As an example, this maximum peak appears for Z ¼ 4.9d and equals to 3.8% for aCO2 ¼ 0%, whereas for aCO2 ¼ 20%, the maximum of k appears for Z ¼ 5.2d and equals to 2.6%. This result depicts a good agreement with the available [16]). They data in the existing literature (Sautet [17]. and Page confirmed that the evolution of the kinetic energy reaches a maximum and the axial position of this maximum is considered as the limit of the potential core and they noted that the maximum peak of energy appears more upstream in the jet development when the jet density decreases.
Fig. 12 e Normalized turbulent kinetic energy k along the bio-hythane jet centerline.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Moreover, it can be seen that the difference of normalized turbulent kinetic energy k along the bio-hythane jet centerline between the configuration aCO2 ¼ 0% and aCO2 ¼ 20% with 0% H2 is less important than with 20%H2, for example, with aH2 ¼ 0% this difference is equals to 31.5% and 37.9% with aH2 ¼ 20%.
Conclusion An experimental study of a non-reacting and reacting jet according to the hydrogen addition and the carbon dioxide dilution has been investigated. PIV technique allows deducing the instantaneous two-dimensional velocity fields. The results of radial profiles of the mean longitudinal velocities show that hydrogen addition effect on the longitudinal component is less and less significant while the height in the flow increases and the carbon dioxide addition reduces the mean longitudinal velocity. The combustion decreases the entrainment of the ambient fluid and raises the viscosity of the flow, leading to an increase of the longitudinal velocity along the bio-hythane jet and a reduction of the turbulence in comparison with non-reacting configuration. These effects are more and more important with hydrogen addition due to the increase of the heat release rate. The turbulence intensity is more important in non-reacting flow than in combustion and is intensified with the hydrogen addition and is reduced with carbon dioxide addition. Concerning the turbulent kinetic energy along the centerline of bio-hythane jet, the peak of maximum energy is higher and occurs for a height less important with hydrogen addition, and is less with CO2 dilution.
Aknowledgments This work has been supported by the international program PHC UTIQUE 12G1125.
references
[1] Choudhuri AR, Gollahalli SR. Combustion characteristics of hydrogenehydrocarbon hybrid fuels. Int J Hydrogen Energy 2000;25:451e62.
19831
[2] Choudhuri AR, Gollahalli SR. Characteristics of hydrogen hydrocarbon composite fuel turbulent jet flames. Int J Hydrogen Energy 2003;28:445e54. [3] El-Ghafour SAA, El-dein AHE, Aref AAR. Combustion characteristics of natural gas-hydrogen hybrid fuel turbulent diffusion flame. Int J Hydrogen Energy 2010;35:2556e65. jets [4] Yon S. Oxy-combustion de l'hythane dans des bruˆleurs a pare s [The se de doctorat]. France: Universite de fortement se Rouen; 2013. [5] Takagi T, Shin HD, Ishio A. Combust. Flame 1981;40:121e40. [6] Liu FS, Guo HS, Smallwood GJ. The chemical effect of CO2 replacement of N2 in air on the burning velocity of CH4 and H2 premixed flames. Combust Flame 2003;133(4):495e7. [7] Du DX, Axelbaum RL, Law CK. Soot formation in strained diffusion flames with gaseous additives. Combust Flame 1995;102:11e20. [8] Zhang C, Atreya A, Lee K. Sooting structure of methane counterflow diffusion flames with preheated reactants and dilution by products of combustion. In: Twenty-fourth symposium (International) on combustion, vol. 24e1; 1992. p. 1049e57. [9] Oh KC, Shin HD. The effect of oxygen and carbon dioxide concentration on soot formation in nonpremixed flames. Fuel 2006;85(5e6):615e24. [10] Liu FS, Guo HS, Smallwood GJ, Gulder OL. The chemical effects of carbon dioxide as an additive in an ethylene diffusion flame: implications for soot and NOx formation. Combust Flame 2006;125:778e87. [11] Guo HS, Smallwood GJ. A numerical study on the influence of CO2 addition on soot formation in an ethylene/air diffusion flame. Combust Sci Technol 2008;180(10e11):1695e708. D, Baillot F. A numerical [12] Guo H, Min J, Galizzi C, Escudie study on the effects of CO2/N2/Ar addition on liftoff of a laminar CH4/air diffusion flame. In: 22nd international colloquium on the dynamics of explosions and reactive systems. Minsk: Belarus; 2009. [13] Min J, Baillot F, Guo H, Domingues E, Talbaut M, PatteRouland B. Impact of CO2, N2 or Ar diluted in air on the length and lifting behavior of a laminar diffusion flame. Proc Combust Inst 2011;33:1071e9. [14] Dally BB, Riesmeier E, Peters N. Effect of fuel mixture on moderate and intense low oxygen dilution combustion. Combust Flame 2004;137:418e31. [15] Chassaing P, Harran G, Joly L. Density fluctuation correlations in free turbulent binary mixing. J Fluid Mech 1994;vol. 279:239e78. J. Contribution a l'e tude des jets turbulents [16] Page triques a masse volumique variable [The se de axisyme d’Orle ans; 1998. doctorat]. France: Universite sur le de veloppement scalaire et [17] Sautet JC. Effets de densite se de doctorat]. France: dynamique des jets turbulents [The de Rouen; 1992. Universite [18] Townsend AA. The structure of turbulent shear flow. University Press; 1976.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Effects of hydrogen addition and Carbone dioxide dilution on the velocity field in non reacting and reacting flows Ibtissem Hraiech a,b, Jean-Charles Sautet b, Mohamed Ali Mergheni a,b,*, Hmaied Ben Ticha a, Hazem Touati a, Abdallah Mhimid a a b
LESTE, Ecole Nationale d'Ingenieurs de Monastir, Tunisia CORIA UMR6614 e CNRS, Universite et INSA de Rouen, Saint Etienne du Rouvray, France
article info
abstract
Article history:
The evolution of anti-pollution standards and the optimization of combustion efficiency
Received 18 July 2014
push the development of new fuels with high energy efficiency. It is necessary to develop
Received in revised form
new alternative fuel to improve the efficiency of conventional systems, reduce emissions
17 September 2014
(NOx, SOx, soot particles) and recover for its materials. A new fuel called bio-hythane, a
Accepted 23 September 2014
mixture of natural gas up to 20% hydrogen and up to 50% Carbone dioxide, from the re-
Available online 16 October 2014
covery of the waste from households and agriculture, via suitable digesters provides a source of renewable energy and usable, is a very interesting solution to improve emission
Keywords: Hythane
standards and optimization of the combustion chambers. This experimental study is led in a non-reacting configuration and in combustion in
Velocity field
order to focus on the effects of hydrogen addition and CO2 dilution in the fuel on the ve-
Hydrogen
locity profiles, the turbulence intensity and the turbulent kinetic energy. From PIV mea-
CO2 dilution
surements the results show that from the velocity and fluctuations profiles the high
Turbulent kinetic energy
diffusivity and the low density of hydrogen allow bio-hythane jet to spread more efficiency in the combustion. The combustion decreases the entrainment of the ambient fluid and raises the viscosity of the flow, leading to an increase of the longitudinal velocity along the bio-hythane jet and a reduction of the turbulence in comparison with non-reacting configuration. The Carbone dioxide addition reduces the fall of the turbulence because the temperature of the flame is less important. Effects of hydrogen and CO2 are also highlighted by analysis of turbulent kinetic energy along the bio-hythane. Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
nieurs de Monastir, Rue Ibn Aljazzar, 5000 Monastir, Tunisia. * Corresponding author. Ecole Nationale d'Inge E-mail addresses: [email protected], [email protected] (M.A. Mergheni). http://dx.doi.org/10.1016/j.ijhydene.2014.09.129 0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Introduction Regulations on protection of the environment lead to develop new technologies of combustion. A new alternative fuel, mixture of natural gas (NG), hydrogen up to 20% in volume and carbon dioxide up to 50% in volume, called bio-hythane, has been created from the recovery of the waste from households and agriculture, via suitable digesters. It provides a source of renewable energy and usable, generates a better thermal efficiency and a reduction of pollutant emissions and is expected to play an important role in future energy production. Due to the properties of hydrogen, specially low density, high molecular diffusivity, wide flammability limits, high flame speed, and low ignition energy (Choudhuri and Gollahalli [1,2]) hydrogen in the fuel permits combustion systems to operate with lean fuel mixtures. Choudhuri and Gollahalli [2] carried out an experimental investigation on turbulent NGeH2 jet diffusion flame and showed a reduction in the soot concentration and emission index of CO (EICO), but an increase in NO and NOx emissions with the addition of hydrogen to the fuel. El Ghafour et al. [3] studied experimentally the effect of hydrogen addition on combustion characteristics of NGeH2 hybrid fuel turbulent diffusion flame at a fixed Reynolds number (4000). They observed that the addition of hydrogen improves the flame stability, reduces the flame length for relatively high hydrogen concentrations and increases the NO and CO concentration. The influence of hydrogen addition to natural gas on the flow dynamics was investigated experimentally in nonreacting flow and in combustion by Yon and Sautet [4]. Their results show that the combustion decreases the entrainment of the ambient fluid, increases the temperature in the fuel jet and consequently the viscosity of the flow, leading to an increase of the longitudinal velocity along the hythane jet and a decrease of the turbulence in comparison with non-reacting configuration. These effects are more and more important with hydrogen addition due to the decrease of the flame liftoff height and the increase of the heat release rate. Takagi et al. [5] carried out an experimental investigation on turbulent jet flows with and without flame, and they found that the presence of flame decreases the entrainment of ambient fluid, which accelerates the flow. The chemical effect of CO2 replacement of N2 in air on the burning velocity of CH4 and H2 flames was studied numerically by Liu et al. [6]. The relative importance of the chemical effect of CO2 on the burning velocity increases as more CO2 is added to replace N2 in air. Previous studies have proved that the soot formation in diffusion flames were decreases by CO2 addition to the coflow air, resulting from the short residence time in the inception region [7e11]. It was shown that CO2 dilution thermally and chemically limits the formation of soot precursors due to the decrease of H radicals consumed in the reaction CO2 þ H / CO þ OH. The effects of different additives to air on the lift-off of a laminar CH4/air diffusion flame have been explored experimentally by Guo et al. [12]. They observed that the addition of CO2 causes flame lift-off due to the dilution, thermal and
19819
chemical effects. The dilution effect being the most significant one, followed by the thermal effect. The three effects tend to reduce combustion intensity and cause flame to be lifted. The radiation and transport property effects are negligible. The effect of CO2 addition to the air on the transition from an attached flame to a lifted flame issued from a coaxial nonpremixed methane-air jet is studied experimentally by Min et al. [13]. Results shows that the CO2 is the best destabilize among the diluents, because the three effects (dilution > thermal > chemistry) induce loss of flame stability (CO2 has a strongest ability to break flame stability, than N2). Dally et al. [14] investigated experimentally and numerically that CO2 addition in a non-premixed methane/air flame lowered flame temperature by decreasing reactant concentration inside the reaction zone. The present study is conducted in non-reacting and reacting flow from a cylindrical burner to investigate the effect of the hydrogen addition and CO2 dilution on the flow aerodynamics. The mole fraction of hydrogen (% H2) in NGeH2 mixture varies from 0% to 20% and the mole fraction of carbon dioxide volume varies from 0% to 50%. The experimental setup consists of a 15 kW burner powered by natural gas added to hydrogen and carbon dioxide. A study of the jet aerodynamic through Particle Image Velocimetry (PIV) in non-reacting flow and in combustion allows characterizing the flow fields. The evolution of the root mean square (RMS) of the two components of velocity is investigated to deduce the effects of CO2 dilution and the hydrogen addition on the turbulence intensity and the turbulent kinetic energy.
Fig. 1 e . Sketch of the burner.
19820
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Table 1 e Gas flow characteristics. Jet Flame mixture Fuel mixture QvNG QvH2 Qvfuel %CO2 QxCO2 Qtot Exit % NG % H2 Jet reynolds pjet Jet Schmidt %NG %H2 (1/min) (1/min) velocity (total (total number number (fuel) (fuel) (1/min) (1/min) (1/min) U0 (m/s) mixture) mixture) 100
0
24.09
0
24.09
95
5
23.73
1.25
24.98
90
10
23.34
2.59
25.94
85
15
22.92
4.046
26.97
80
20
22.47
5.61
28.09
0 10 20 0 10 20 0 10 20 0 10 20 30 0 10 20 30
0 2.6 6 0 2.7 6.2 0 2.8 6.4 0 2.9 6.7 11.5 0 3.1 7 12
24 26.7 30.1 24.9 27.7 31.2 25.9 28.8 32.4 26.9 29.9 33.7 38.5 28.1 31.2 35.1 40.1
Thanks to PIV measurements in non-reacting flow and in combustion, the effect of combustion on the average 2D normalized longitudinal velocity fields Uz =U0 in the XZ-plane, are investigated. The evolution of the turbulence intensity along the bio-hythane jet centerline as a function of the hydrogen volume fraction and CO2 dilution rate is investigated in non-reacting fuel.
Experimental setup and measurement technique
14.2 15.7 17.7 14.7 16.3 18.4 15.3 16.9 19.1 15.9 17.6 19.8 22.7 16.5 18.4 20.7 23.6
6312 9005 10,741 7518 8959 10,769 7422 8914 10,804 7322 8868 10,847 13,486 7205 8822 10,899 13,713
0.83 0.93 1.03 0.79 0.9 1 0.75 0.86 0.97 0.71 0.83 0.95 1.06 0.66 0.78 0.9 1.02
The present study aims to investigate the effects of the hydrogen addition in the fuel and the effect of CO2 dilution on
0 0 0 5 4.5 4 10 9 8 15 13.5 12 10.5 0.2 0.18 0.16 0.14
100 90 80 95 85.5 76 90 81 72 85 76.5 68 59.5 80 72 64 56
the flow aerodynamics. The burner depicted in Fig. 1 consists of a jet of bio-hythane (mixture CH4/H2/CO2). The cylindrical burner, of internal diameter d ¼ 6 mm, brings the mixture CH4/ H2/CO2 into the ambient air. The natural gas has a density of 0.83 kg/m3 and a volume composition of 85% CH4, 9% C2H6, 3% C3H8, 2% N2, 1% CO2, and traces of higher hydrocarbon species. The hydrogen volume fraction (aH2 ) in the fuel mixture (NG þ H2) varies between 0 and 20%. The fuel volumetric flow rate is: Qvfuel ¼ QVGN þ QvH2
Experimental setup
0.8 0.86 0.91 0.75 0.78 0.83 0.71 0.73 0.77 0.67 0.69 0.73 0.79 0.65 0.68 0.7 0.76
(1)
CO2 is introduced in the fuel mixture to study the effect of CO2 dilution. The percentage of CO2 (aCO2 ) in the mixture (NG þ H2 þ CO2) varies from 0% (no dilution) to 50%.
Fig. 2 e Particle image velocimetry experimental setup.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Table 2 e Parameters of PIV setup. Non-Reacting flow Dt (ms) 15 Magnification (mm pixel1) Explored height 0e100 in the jet (mm)
40
100e200
Reacting flow
18 0.0051
45
65
0e100 100e200 200e300
The CO2 flow rate is QvCO2 and the total jet flow rate is: Qtot ¼ Qvfuel þ QvCO2
(2)
The regulation of natural gas flow is realized by a mass flow regulator. Hydrogen and carbon dioxide flow rates are controlled by sonic nozzles and pressure regulator. Table 1 Summarizes the different parameters of the experiment including natural gas, hydrogen, Carbone dioxide and total jet flow rates, the jet exit velocity, U0, the bulk jet Reynolds Number (Re ¼ U0d/n) as well as the Schmidt number, Sc ¼ n/D, where D is the molecular diffusivity.
Measurement technique To study the aerodynamics of the flow according to the hydrogen concentration and the dilution of CO2, the Particle Image Velocimetry (P.I.V) is used. This non-intrusive method, permits to procure 2D images of the flow and the instantaneous two-dimensional velocity fields, the experimental P.I.V setup described in Fig. 2 needs a light source (120 Mj/pulse, pulse duration of 8 ns), in this case a double-pulsed NdeYag laser (Big Sky CFR200, Quantel) with a wavelength of 532 nm and a frequency of 10 Hz.
19821
An optical system consisting of three consecutive lenses creates the laser sheet with 500 mm thickness. A CCD camera (2048*2048 pixel2, Image Pro X Lavision) with a dynamic range of 14 bits is oriented perpendicularly to the laser sheen. A convergent lens (f ¼ 85 mm), placed perpendicularly to the light source, collects the signal of Mie scattering emitted by the seeded particles. Oil particles (mean diameter of 3 mm) and Zirconium Oxide particles (mean diameter of 10 mm) are used respectively in non-reacting and reacting configurations. An interference filter is used to reject the bright luminosity in front of the lens of the PIV setup. The magnification is 0.051 mm pixel1 and the time interval between two images varies according to the investigated configurations (Table 2). From recording 500 instantaneous pairs of images, with a resolution of 2048*2048 pixels2, a statistical processing enables to obtain the mean velocity field, and the fluctuations for each height and configuration. Fig. 3 shows the mean twodimensional longitudinal velocity field in non-reacting flow and the characteristic zones of a flow for aH2 ¼ 20% and aCO2 ¼ 20%. This study focused on the effect of the hydrogen addition and the effect of CO2 dilution on dynamic fields. In order to investigate all the velocity components in the upstream of the flow, it was necessary to used Particle Image Velocimetry. Fig. 4 displays the images of the seeded particles in non reacting flow for a hydrogen volume fraction equals to 0 and Carbone dioxide Carbone volume fraction equals to 0. This figure shows instantaneous, mean and RMS velocity fields. This allows to study the velocity radial component and the longitudinal component of the velocity and RMS radial component and the RMS longitudinal component.
Results and discussion Mean velocity field
Fig. 3 e Mean two-dimensional longitudinal velocity fields in non-reacting flow (aH2 ¼ 20%, aCO2 ¼ 20%).
The longitudinal velocity profiles allow studying the effects of the hydrogen addition and the dilution of CO2 on the dynamics of the flow (Fig. 5). Fig. 5(a) displays the radial profiles of the mean longitudinal jet velocity Uz normalized by the exit velocities U0 in non-reacting flow, without CO2 dilution (aCO2 ¼ 0%), with the hydrogen fraction from aH2 ¼ 0% to aH2 ¼ 20% for three different heights, for Z ¼ 10, 50 and100 mm. This figure shows that with hydrogen addition the maximum of the normalized longitudinal velocity of the =U0 ¼ 1:26 to bio-hythane jet grows by 29% (from Umax z Umax =U ¼ 1:63 with a ¼ 20%) near the jet exit (Z ¼ 10 mm). 0 H2 z =U0 ¼ 0:71 to This difference, at Z ¼ 50 mm, is 18% (from Umax z =U ¼ 0:84 with a ¼ 20%) and 13% downstream Umax 0 H 2 z =U0 ¼ 0:38 to Umax =U0 ¼ 0:43 with (Z ¼ 100 mm) (from Umax z z aH2 ¼ 20%). From this figure it can be seen that the difference =U0 between the configuration of maximum velocity Umax z aH2 ¼ 0% and aH2 ¼ 20% is less important far from the burner due to the high molecular diffusivity and the low density of hydrogen, favoring the entrainment with the ambient. The effect of carbon dioxide dilution on the normalized mean velocity is presented on Fig. 5 (b) for Z ¼ 10, 50 and100 mm in the configuration without H2 addition (aH2 ¼ 0%). This figure shows that, with CO2 dilution, Umax =U0 z
19822
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 4 e a) Images of the seeded particles, b) Normalized mean velocity fields and c) Normalized RMS velocity fields of the component Uz in non-reacting flow for aCO2 ¼ 0% and aH2 ¼ 0%.
decays only by 11% for the three heights in the jet. For =U0 ¼ 0:71 for aCO2 ¼ 0% and 0.65 example, at z ¼ 50 mm, Umax z for aCO2 ¼ 20%. The combined effects of H2 and CO2 is highlighted comparing Fig. 5(b) and (c) The difference of maximum ve=U0 without hydrogen, between the configuration locity Umax z aCO2 ¼ 0% and aCO2 ¼ 20% is less important than with aH2 ¼ 20%, for example, at z ¼ 10 mm, with aH2 ¼ 0% this difference is equals to 11% and 15.4% with aH2 ¼ 20% (Fig. 5(c)). From the averaged velocity profiles in non-reacting flows, the evolution of the mean centerline jet velocity, Ucl, can be deduced. Fig. 6 shows the inverse of Ucl normalized by the jet 0 exit velocity U0 as a function of the reduced coordinate ZZ deff with the % CO2 added at various % H2 where deff is defined by: deff ¼ d
rext exp 0:5 rjet
(3)
From this figure, it can be seen that the rate of decrease of the mean longitudinal velocity Ku increase with the addition of CO2 and decreases when adding the hydrogen this is due to the low density. It can be noted that the rate of decrease of the mean longitudinal velocity along a free jet is inversely proportional to the density of the jet, this result depicts a good agreement with the available data in the existing literature [16]). For example, in the configura(Chassaing [15] and Page tion where 20% of hydrogen is added to air, the rate of decrease of the mean longitudinal velocity passes from Ku ¼ 3.36 for 0% of Carbone dioxide, to Ku ¼ 5.05 when the percentage of Carbone dioxide is 20%. Also, in the configuration where 0% of carbon dioxide is added to air, the rate of decrease of the mean longitudinal
velocity passes from Ku ¼ 5.76 for 0% of hydrogen, to Ku ¼ 3.36 when the percentage of hydrogen is 20%. Fig. 7 compares the average 2D normalized longitudinal velocity fields Uz/U0 in the XZ-plane for aH2 ¼ 0% and aH2 ¼ 20% (aCO2 ¼ 0%) and for aCO2 ¼ 0% and aCO2 ¼ 20% (aH2 ¼ 0%), obtained in non-reacting flow and with combustion. In non-reacting flow and in combustion, Fig. 7(a) shows that the hydrogen addition rises the velocities along the biohythane jet as the normalized longitudinal velocities are lower with aH2 ¼ 0% in comparison with aH2 ¼ 20%. Concerning the fields in combustion, the longitudinal velocity of the bio-hythane jet increases and the dynamic greatly differs from non-reacting flow. The heat release modifies the entrainment, the viscosity and mixing of the flow, and accelerates the bio-hythane jet. In non-reacting flow and in combustion, Fig. 7(b) shows that, without hydrogen, the Carbone dioxide addition decreases the velocities along the bio-hythane jet. Fig. 7(c) shows the combined effects of H2 and CO2. In combustion, for aH2 ¼ 20%, the bio-hythane jet is accelerated and the dynamics differs from non-reacting flow. It can be noted that in combustion the effect of CO2 is offset with the hydrogen addition due to the low density induced by addition of H2.
Longitudinal velocity fluctuations From the PIV measurement, the longitudinal velocity standard deviation U0z is studied in the XZ-plane. Fig. 8 shows the longitudinal velocity fluctuations of the bio-hythane jet in non-reacting flow. The figure displays a minimum
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
19823
Fig. 5 e Radial profiles of normalized mean longitudinal velocity for Z ¼ 10, 50, 100 mm, in non-reacting flow. a-As a function of the hydrogen volume fraction (for aCO2 ¼ 0%). b-As a function of the carbon dioxide volume fraction (for aH2 ¼ 0%). c-As a function of the carbon dioxide volume fraction (for aH2 ¼ 20%).
19824
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 6 e Normalized mean longitudinal velocity along the jet centerline, in non-reacting flow. a-As a function of the hydrogen volume fraction (for aCO2 ¼ 0%). b-As a function of the carbon dioxide volume fraction (for aH2 ¼ 0%). c-As a function of the carbon dioxide volume fraction (for aH2 ¼ 20%).
fluctuations zone at the center of the jet and a maximum in the shear layers. With Hydrogen, this maximum is reached at about Z ¼ 5d to Z ¼ 6d from the nozzle. Low fluctuations zone at the nozzle exit corresponds to the potential core. In the shear layers, the high fluctuations increase with the jet development, until the mixing layers join. With Carbon dioxide addition, this maximum is located about Z ¼ 6d from the nozzle and the maximum level of fluctuations high fluctuations decreases, due to the higher jet density and the decrease of the ambient air entrainment
(Sautet [17]). Also, it can be noted that the normalized longitudinal velocity RMS U0z =U0 in non-reacting flow and the expansion of the jet increase with addition hydrogen due to the low density of hydrogen. However, with aH2 ¼ 20%, the expansion of the jet is lower in the configurations without CO2 in comparison with aCO2 ¼ 20%.The benefic effect of H2 on the mixing is balanced with the CO2 dilution. Fig. 9 shows the evolution of the normalized RMS longitudinal velocity along the jet centerline in non-reacting and reacting flows.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
19825
Fig. 7 e Effect of combustion on the average 2D normalized longitudinal velocity fields UZ =U0 in the XZ-plane, in nonreacting flow and in combustion.
19826
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 8 e Normalized RMS of longitudinal jet velocity U0Z =U0 in non-reacting flow.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 9 e Normalized RMS velocity along the bio-hythane jet centerline in non-reacting and reacting flows.
19827
19828
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 10 e Evolution of the turbulence intensity, U0Z =UZ along the jet centerline.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Fig. 11 e Comparison of the two components of the normalized velocity fluctuations along the jet centerline.
19829
19830
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
It can be seen from this figure that for a pure natural gas jet, in the potential core (Z < 5d) the turbulence rates are similar without reaction and in combustion but from Z ¼ 5d, the turbulence is higher is non-reacting configuration than in combustion, in fact, the flame increases the temperature in the fuel jet and consequently the viscosity, leading to a decrease of the normalized velocity fluctuations. With the hydrogen addition, the turbulence is higher is non-reacting configuration than in combustion however from Z ¼ 7d the longitudinal fluctuations along the bio-hythane jet are higher in combustion than in non-reacting configuration. This is due to the high longitudinal velocity in combustion and the low density induced by addition of H2, which remains a high level of turbulence. With the addition of CO2, the turbulence is higher is nonreacting configuration than in combustion due to the high density of CO2, and this fall of the turbulence is reduced.
Turbulence intensity From the mean and RMS longitudinal velocity, it is possible to deduce the evolution of the turbulence intensity along the fuel jet U0z =Uz . Fig. 10 shows the evolution of the turbulence intensity along the bio-hythane jet according to the hydrogen addition and the carbon dioxide addition. At the exit of the nozzle (Z ¼ 0d), the turbulence intensity is around 5%. At Z/d ¼ 16, it reaches 30% in non-reacting configurations and 25% in combustion, values published in the literature [18]. Without hydrogen and Carbone dioxide, the turbulence intensity is similar in the 2 configurations from the exit of the burner to Z ¼ 5d. After this characteristic height corresponding to the end of the potential core, the turbulence intensity is higher in non-reacting flow than with the flame. The evolution of the turbulence intensity in combustion differs from non-reacting flow as and when the flame and the temperature perturb the flow, because of the heat release rate which perturbs the flow. The flame increases the temperature in the fuel jet and consequently the viscosity, leading to a decrease of the turbulence intensity. This fall of the turbulence is intensified with the hydrogen addition because the temperature of the flame and the burnt gases at the flame bottom are more important and this fall is reduced because the temperature of the flame is less important with the carbon dioxide addition.
Turbulent kinetic energy In non-reacting flow, from the measurement of the RMS of the two components of velocity ðU0X and U0Z Þ, it is possible to deduce the evolution of the turbulent kinetic energy k: k ¼ 0:5 U0Z exp 2 þ 2 U0X exp 2
(4)
Fig. 11 displays the evolution of the two components of velocity standards deviation ðU0X ; U0Z Þ normalized by the exit velocity U0 along the bio-hythane jet according to aH2 and aH2 . With hydrogen addition, the velocity and the fluctuation increase in the flow along the jet and decreases with carbon dioxide dilution. The longitudinal fluctuations are more important than radial ones corresponding to the classical behavior of a jet. From the measurement of the RMS of the two components of velocity, the turbulent kinetic energy can be deduced. Fig. 12 shows the turbulent kinetic energy, k, normalized by the squared of exit velocity U20 , along the centerline of the biohythane jet according to the hydrogen volume fraction (aH2 ¼ 0% and aH2 ¼ 20%) and the dilution rate in carbon dioxide (aCO2 ¼ 0%, and aCO2 ¼ 20%). With hydrogen addition, the jet interacts with the surrounding fluid more upstream in the flow and the maximum peak of energy appears earlier in the jet. As an example, without CO2, this maximum peak appears at Z ¼ 5.5d and equals to 3.6% for aH2 ¼ 0%, whereas for aH2 ¼ 20%, the maximum of k appears for Z ¼ 4.5d and equals to 5.7%.The growth of the turbulent kinetic energy with hydrogen addition is due to the increase of the fluctuation of the two components of velocity with hydrogen addition. The study of the turbulent kinetic energy according to the carbon dioxide shows that, for aH2 ¼ 0%, the normalized turbulent kinetic energy decreases with aCO2 As an example, this maximum peak appears for Z ¼ 4.9d and equals to 3.8% for aCO2 ¼ 0%, whereas for aCO2 ¼ 20%, the maximum of k appears for Z ¼ 5.2d and equals to 2.6%. This result depicts a good agreement with the available [16]). They data in the existing literature (Sautet [17]. and Page confirmed that the evolution of the kinetic energy reaches a maximum and the axial position of this maximum is considered as the limit of the potential core and they noted that the maximum peak of energy appears more upstream in the jet development when the jet density decreases.
Fig. 12 e Normalized turbulent kinetic energy k along the bio-hythane jet centerline.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 9 8 1 8 e1 9 8 3 1
Moreover, it can be seen that the difference of normalized turbulent kinetic energy k along the bio-hythane jet centerline between the configuration aCO2 ¼ 0% and aCO2 ¼ 20% with 0% H2 is less important than with 20%H2, for example, with aH2 ¼ 0% this difference is equals to 31.5% and 37.9% with aH2 ¼ 20%.
Conclusion An experimental study of a non-reacting and reacting jet according to the hydrogen addition and the carbon dioxide dilution has been investigated. PIV technique allows deducing the instantaneous two-dimensional velocity fields. The results of radial profiles of the mean longitudinal velocities show that hydrogen addition effect on the longitudinal component is less and less significant while the height in the flow increases and the carbon dioxide addition reduces the mean longitudinal velocity. The combustion decreases the entrainment of the ambient fluid and raises the viscosity of the flow, leading to an increase of the longitudinal velocity along the bio-hythane jet and a reduction of the turbulence in comparison with non-reacting configuration. These effects are more and more important with hydrogen addition due to the increase of the heat release rate. The turbulence intensity is more important in non-reacting flow than in combustion and is intensified with the hydrogen addition and is reduced with carbon dioxide addition. Concerning the turbulent kinetic energy along the centerline of bio-hythane jet, the peak of maximum energy is higher and occurs for a height less important with hydrogen addition, and is less with CO2 dilution.
Aknowledgments This work has been supported by the international program PHC UTIQUE 12G1125.
references
[1] Choudhuri AR, Gollahalli SR. Combustion characteristics of hydrogenehydrocarbon hybrid fuels. Int J Hydrogen Energy 2000;25:451e62.
19831
[2] Choudhuri AR, Gollahalli SR. Characteristics of hydrogen hydrocarbon composite fuel turbulent jet flames. Int J Hydrogen Energy 2003;28:445e54. [3] El-Ghafour SAA, El-dein AHE, Aref AAR. Combustion characteristics of natural gas-hydrogen hybrid fuel turbulent diffusion flame. Int J Hydrogen Energy 2010;35:2556e65. jets [4] Yon S. Oxy-combustion de l'hythane dans des bruˆleurs a pare s [The se de doctorat]. France: Universite de fortement se Rouen; 2013. [5] Takagi T, Shin HD, Ishio A. Combust. Flame 1981;40:121e40. [6] Liu FS, Guo HS, Smallwood GJ. The chemical effect of CO2 replacement of N2 in air on the burning velocity of CH4 and H2 premixed flames. Combust Flame 2003;133(4):495e7. [7] Du DX, Axelbaum RL, Law CK. Soot formation in strained diffusion flames with gaseous additives. Combust Flame 1995;102:11e20. [8] Zhang C, Atreya A, Lee K. Sooting structure of methane counterflow diffusion flames with preheated reactants and dilution by products of combustion. In: Twenty-fourth symposium (International) on combustion, vol. 24e1; 1992. p. 1049e57. [9] Oh KC, Shin HD. The effect of oxygen and carbon dioxide concentration on soot formation in nonpremixed flames. Fuel 2006;85(5e6):615e24. [10] Liu FS, Guo HS, Smallwood GJ, Gulder OL. The chemical effects of carbon dioxide as an additive in an ethylene diffusion flame: implications for soot and NOx formation. Combust Flame 2006;125:778e87. [11] Guo HS, Smallwood GJ. A numerical study on the influence of CO2 addition on soot formation in an ethylene/air diffusion flame. Combust Sci Technol 2008;180(10e11):1695e708. D, Baillot F. A numerical [12] Guo H, Min J, Galizzi C, Escudie study on the effects of CO2/N2/Ar addition on liftoff of a laminar CH4/air diffusion flame. In: 22nd international colloquium on the dynamics of explosions and reactive systems. Minsk: Belarus; 2009. [13] Min J, Baillot F, Guo H, Domingues E, Talbaut M, PatteRouland B. Impact of CO2, N2 or Ar diluted in air on the length and lifting behavior of a laminar diffusion flame. Proc Combust Inst 2011;33:1071e9. [14] Dally BB, Riesmeier E, Peters N. Effect of fuel mixture on moderate and intense low oxygen dilution combustion. Combust Flame 2004;137:418e31. [15] Chassaing P, Harran G, Joly L. Density fluctuation correlations in free turbulent binary mixing. J Fluid Mech 1994;vol. 279:239e78. J. Contribution a l'e tude des jets turbulents [16] Page triques a masse volumique variable [The se de axisyme d’Orle ans; 1998. doctorat]. France: Universite sur le de veloppement scalaire et [17] Sautet JC. Effets de densite se de doctorat]. France: dynamique des jets turbulents [The de Rouen; 1992. Universite [18] Townsend AA. The structure of turbulent shear flow. University Press; 1976.