0
Research Papers: Fuel Combustion

# Investigation of A Low Emission Liquid Fueled Reverse-Cross-Flow CombustorPUBLIC ACCESS

[+] Author and Article Information
Preetam Sharma

Department of Aerospace Engineering,
Indian Institute of Technology Kanpur,
Kanpur 208016, UP, India
e-mail: preetams@iitk.ac.in

Naman Jain

Department of Aerospace Engineering,
Indian Institute of Technology Kanpur,
Kanpur 208016, UP, India
e-mail: jnaman@iitk.ac.in

Vaibhav Kumar Arghode

Department of Aerospace Engineering,
Indian Institute of Technology Kanpur,
Kanpur 208016, UP, India
e-mail: varghode@iitk.ac.in

1Corresponding author.

Contributed by the Advanced Energy Systems Division of ASME for publication in the Journal of Energy Resources Technology. Manuscript received January 31, 2019; final manuscript received April 3, 2019; published online April 18, 2019. Assoc. Editor: Ashwani K. Gupta.

J. Energy Resour. Technol 141(10), 102202 (Apr 18, 2019) (9 pages) Paper No: JERT-19-1062; doi: 10.1115/1.4043437 History: Received January 31, 2019; Accepted April 07, 2019

## Abstract

The investigated combustor employs injection of liquid fuel (ethanol) into the strong cross-flow of air using a round tube to achieve effective fuel atomization in non-premixed mode of operation. The reverse-flow configuration (air injection from the exit end) allows effective internal product gas recirculation and stabilization of the reaction zone. This apparently suppresses near-stoichiometric reactions and hot spot regions resulting in low pollutant (NOx and CO) emissions in the non-premixed mode. The combustor was tested at thermal intensity variation from 19 to 39 MW/m3 atm with direct injection (DI) of liquid fuel in cross-flow of air injection with two fuel injection diameters of 0.5 mm (D1) and 0.8 mm (D2). The combustion process was found to be stable with NOx emissions of 8 ppm (for D1) and 9 ppm (for D2), the CO emissions were 90 ppm for D1 and 120 ppm for D2, at an equivalence ratio (ϕ) of 0.7. Macroscopic spray properties of the fuel jet in cross-flow were investigated using high-speed imaging techniques in unconfined and nonreacting conditions. It was found that the fuel jet in smaller fuel injection diameter (D1) case penetrated farther than that in D2 case due to higher fuel injection momentum, thus possibly resulting in a finer spray and better fuel-oxidizer mixing, and in turn leading to lower CO and NOx emissions in the D1 case as compared with the D2 case.

<>

## Introduction and Background

Low pollutant emissions from combustion systems have always been the primary focus area for researchers because of their deleterious effect on human health and the environment. The primary pollutants linked to aircraft gas turbines are oxides of nitrogen (NOx), carbon monoxide (CO), unburned hydrocarbons, and particulates (smoke) [1,2]. The pernicious influences of NOx emissions on ozone depletion, acid rains, the formation of photochemical smog, etc. is a serious cause of concern for regulating agencies across the globe. With the ever-increasing global energy demand, there is a quest for achieving cleaner combustion and more efficient harvesting of energy using renewable resources.

Many strategies to limit NOx and CO emissions have been reported in the literature [3]. The bulk of the research in limiting NOx emissions concentrates on reducing the temperature in the combustion chamber typically below 1800 K as above this threshold; due to thermal Zeldovich mechanism, the NOx formation rate increases significantly [4]. At a temperature of around 1800 K, the amount of thermal NOx produced in a few seconds is comparable with that of NOx produced in just a few milliseconds when the temperature is closer to 2100 K for a given O2/N2 ratio [4]. Several NOx suppression techniques such as lean premixed combustion, exhaust gas recirculation, internal product gas recirculation, and water injection aim to reduce near-stoichiometric hot regions (>1800 K), present in conventional diffusion flames [1,2]. Recently, a new technique known as colorless distributed combustion (CDC) has been successfully demonstrated to achieve low levels of CO and NOx pertaining to gas turbine applications [516].

The CDC technology inherits certain features of high-temperature air combustion (HiTAC) [17,18] such as discrete and direct injection of air and fuel in the combustion chamber to facilitate sufficient product gas recirculation and fast fuel-oxidizer mixing to stabilize distributed reactions and avoid high-temperature thin reaction zones. The requirement of product gas recirculation and fast fuel-oxidizer could be partially met with different configurations of air and fuel injections [12]. It was found that the reverse flow configuration, where the air injection is from the same side of the combustor exit, facilitates favorable residence time characteristics thus leading to lower CO emissions, while injection of fuel (gaseous) in cross-flow of air jet results in better fuel-oxidizer mixing leading to lower NOx emissions [10].

Other similar technologies related to HiTAC are moderate and intense low oxygen dilution (MILD) combustion [19] and flameless oxidation (FLOX) [20].

Even though a substantial amount of work has been done in HiTAC/CDC/MILD/FLOX combustors using gaseous fuels, there is a paucity of literature on the operation of such combustors with liquid fuels. Weber and Smart [21] conducted experiments in a refractory lined furnace (heat load of 0.58 MW) with the preheated combustion air at 1573 K. Light fuel oil produced almost invisible flames with 80 ppm of NOx emissions, whereas the heavy fuel oil burnt with visible flames giving NOx emissions of about 250 ppm at the exit of the furnace. Derudi and Rota [22] focused on the investigation of the sustainability of flameless combustion for liquid hydrocarbons (n-octane) using a dual-nozzle laboratory scale burner, and low NOx (30 ppm) and negligible CO emissions were reported. In a different investigation by Reddy and Kumar [23], a strongly swirling flow field was employed to recirculate large amounts of product gases to achieve flameless combustion using liquid fuels that were injected using an atomizer. Kerosene, diesel, and gasoline fuels were tested at heat release intensity of 5 MW/m3 atm (heat load of about 20 kW), and very low NOx (<10 ppm) and CO (<20 ppm) emissions were reported.

In a number of studies [2427], a stagnation point reverse flow (SPRF) combustor was tested with heptane and Jet-A as fuels. Crane et al. [24] investigated SPRF combustor using a straight-walled and a venturi injector to inject the fuel from the top of the combustor coaxially with the air. It was reported that the combustor showed stable combustion using Jet-A fuel with emissions of less than 1 ppm NOx and 5 ppm CO with pressure losses less than 5% at a heat release intensity of 10 MW/m3 atm.

Biomass derived fuels such as ethanol represents a potential alternative option to fossil fuels for reducing greenhouse gases and can be used in combustion devices for power generation. Arghode et al. [28] employed ethanol fuel in the reverse-cross-flow combustor, where the liquid fuel was directly injected using a round tube, and very low NOx and CO emissions were reported. The NOx emissions of 6.4 ppm in the non-premixed (direct injection) mode and 1.8 ppm in the premixed-prevaporized mode and CO emissions of about 220 ppm for the non-premixed (direct injection) mode and 160 ppm for the premixed-prevaporized (PP) mode were reported. The combustor was operated at a very high thermal intensity (170 MW/m3 atm). The present study is an advancement of the above-mentioned work [28,29]. In the present study, two fuel injection round tubes with internal diameters of 0.5 mm (D1) and 0.8 mm (D2) are used to investigate the influence of fuel injection momentum on the combustion and emission characteristics. Atomization study of the liquid fuel jet breakup in cross-flow is included using high-speed imaging techniques.

Since the present study involves direct injection of the liquid fuel into a strong cross-flow of air, studies related to liquid jets in cross-flow are also reviewed. Liquid jet injected into gaseous cross-flow finds many applications pertaining to the aerospace industry such as gas turbine combustors, afterburners, ramjets, lean premixed prevaporizer (LPP), etc. A plethora of studies have been done in the past on the breakup processes of liquid jets in subsonic uniform cross-flows [3032], where many important parameters such as column and surface breakup, jet width, droplet size, droplet velocity, the jet/spray trajectory (penetration height) of liquid jet in cross-flowing air were found to influence the mixing of the injected liquid with the air. The aerodynamic Weber number (We) and the liquid-to-air momentum flux ratio (q) are the most relevant nondimensional numbers related to the physics of this problem [3032].

The fuel atomization in the liquid fueled reverse-cross-flow combustor investigated here is different from the classical case of a liquid jet in uniform cross-flow. Here, the cylindrical liquid fuel jet is injected into a nonuniform cross-flow of air. This is because the air is injected through a round tube and it emerges in the form of a jet which entrains the surrounding product gases as it propagates toward the bottom of the combustor. The geometry also allows for the large recirculation of the product gases, which adds to the complexity (nonuniformity) of the flow field near liquid fuel injection location.

Depending on the application under consideration, a very few types of nonuniform cross-flow have been investigated in the literature. Becker and Hassa [33] investigated the case of liquid fuel injection into a prototype LPP injection module, where the liquid jet was injected radially outward at elevated temperature and pressure conditions. It was observed that the liquid placement was dependent on the variation of liquid-to-air momentum flux ratio (q). Later, Becker et al. [34] studied the effect of a film ring, introduced between two annular swirling flows, on the jet behavior. Gong et al. [35] studied the water jets injected radially inward into a cylindrical chamber with a swirling cross-flow. It was observed that the breakup process was a strong function of parameters such as q, swirl number, injection diameter, and inclination angle. Tambe et al. [36] and Tambe and Jeng [37] experimentally studied the liquid jets injected transversely into a shear layer generated by merging two codirectional air streams of different velocities. They also studied the transverse injection of liquid jets in swirling cross-flow, and it was shown that the radial penetration increased with increasing q. Lower radial and higher circumferential penetrations were noted for higher swirl angles. Sikroria et al. [38] also studied the jet breakup phenomenon in swirling cross-flow and concluded that the process was significantly influenced by instability column waves. A decrease in penetration height was observed for some cases, contrary to the expected trend due to these disturbances.

In a very recent study, Xia et al. [39] studied the jet characteristics of a water impinging on free-air jet at angles of 30, 45, and 60 deg, by using high-speed photography to identify the spray structures and breakup processes and by using phase Doppler anemometry, measured the droplet size and velocity. They observed that the spray divergence angle (or the jet spread angle) decreases with increasing liquid-to-air momentum flux ratio (q), i.e., a larger Weber number (We) leads to better atomization. Jadidi et al. [40] investigated breakup characteristics of water jets injected into transverse free air jets pertaining to the application in suspension-solution thermal sprays. In another publication, Jadidi et al. [41] numerically studied the primary breakup of round nonturbulent liquid jets in shear-laden gaseous cross-flow. They showed that when the gas velocity profile is linear, the average velocity as well as the gradient of the gas velocity profile controls the breakup process and must be taken into account. The penetration was found to significantly increase with the increase in the ratio of cross-flow velocities at the bottom and top of the inlet boundary.

In order to elucidate the effect of atomization of the liquid jet injected into the cross-flow on the combustion characteristics, atomization study was also done in the present investigation. It should be noted that because of the flow field being severely nonuniform in the present combustor as a result of strong recirculation of the product gases, it is difficult to distinctively classify the nonuniformity and obtain primary breakup characteristics in detail as done in earlier studies [3341]. However, an attempt has been made in the present investigation to report the macroscopic spray properties. The fuel atomization process was captured using a high-speed camera and the images were processed to obtain the primary breakup characteristics and spray parameters such as averaged spray outer-edge and centerline trajectories to describe the spray penetration and its impact on the combustor performance. The study was conducted in an unconfined and nonreacting condition keeping the injection parameters similar to that used in the combustor for all the operating cases. These macroscopic spray properties can provide an overall picture of the atomization process of the liquid jet for the two fuel injection diameters investigated here. The image processing algorithms proposed by Tan et al. [42] were used to extract the spray parameters from the raw images.

## Combustor Geometry

Figure 1 shows a photograph and schematics of the combustor under present investigation. The air inlet and the exit ports are located on the same side to facilitate the reverse flow configuration. The fuel is directly injected into the cross-flow of air jet using a simple small diameter round tube. The combustor is made of a rectangular, high-temperature quartz tube with 5 mm wall thickness to give complete optical access with single fuel injection, air inlet, and exhaust ports. The quartz plates have sufficient optical flatness and transparency for the current investigation. The top and bottom sides of the combustor are sealed with machinable ceramics that facilitate in the fixing of the supply lines and minimize heat loses to the top and bottom steel plates used for structural support. The air injection diameter (dair) is 6 mm (fixed) and the exit port diameter (dexit) is 10 mm (fixed).

The air injection diameter is chosen so as to reduce the total pressure loses (<5%) as well as achieve sufficient recirculation of gases inside the combustor [12]. The liquid fuel is injected using a simple round tube and two cases with an internal diameter of 0.5 mm (D1) and 0.8 mm (D2) are investigated. The injection diameters are chosen such that the injection velocity difference between the two cases is close to half for a given fuel flow rate so as to study the effect of fuel jet momentum and hence penetration on the combustion and emission characteristics for the reverse-cross-flow geometry. The combustor volume is marked in yellow color in Fig. 1(a). It should be noted that the configuration investigated here cannot be started directly with ethanol. So, a pilot flame was used to first stabilize the combustor using liquefied petroleum gas (LPG) and once the combustion is stabilized on the gaseous fuel, ethanol supply was switched ON and the LPG supply was turned OFF gradually for the transition.

## Experimental Setup

###### Details of Experimental Setup for Combustor Performance.

Figure 2(a) shows the schematics of the experimental setup for combustor performance. The combustor was operated at a heat load range of 3.125–6.25 kW with thermal intensity (defined as the heat energy released per unit time in a unit volume of combustor scaled by the operating pressure) variation from 19 to 39 MW/m3 atm. In the present investigation, the fuel flow rate was varied keeping the air flow rate constant to simulate varying heat load operation of a gas turbine combustor. The set of experimental conditions are given in Table 1.

The compressed air was metered to the combustor through a precision stainless steel choked flow orifice. The upstream pressure of the orifice was controlled using a regulator to supply the required mass flow rate of the air and it was ensured that the downstream pressure was lower than that required for choking. The upstream pressure was maintained at ± 0.014 atm of the desired pressure. The liquid fuel, i.e., ethanol was pumped to the combustor from a stainless steel bottle pressurized with nitrogen. The ethanol used in the present study is of laboratory grade and has a minimum of 99.9% ethanol (by volume) and a maximum of 0.1% water (by volume) as impurity.

An LPG fired pilot burner was incorporated to ignite the combustor. The combustor was allowed to run for at least 20 min after ignition to achieve steady conditions before recording the data. A portable electrochemical gas analyzer (MRU OPTIMA-7) was used to collect the exhaust gas samples for pollutants (NOx and CO). The analyzer works by drawing a sample of the flue gases from the duct using a built-in gas pump through a sampling probe. The sample is then cleaned and dried using condensate separator with built-in filter and analyzed using electrochemical sensors. The analyzer was serviced and calibrated before performing the experiments. It was observed that the emission readings stabilized within 3–4 min of operation for any change in operating condition, i.e., change in equivalence ratio for a given diameter test case. The experiments were repeated three times for each test case and the uncertainty was estimated to be ± 0.5 ppm for NOx and ±5% for CO measurements.

Due to perfect mixing of fuel and air prior to combustion in the premixed-prevaporized case, it forms a baseline case for comparing NOx and CO emissions with the non-premixed (DI) cases. Prevaporized ethanol was injected far upstream in the air injection line to ensure proper premixing.

The OH* chemiluminescence intensity distribution in the combustor was obtained using an intensified digital 12-bit charged couple device (ICCD) camera system with a narrow bandpass filter (OH* at 307 nm). For a set of experiments, same exposure time (5 ms), gain, f-stop setting, and same distance of the ICCD camera from the combustor were maintained. The camera was sharply focused to locate the reaction zone in the combustor. Post-processing of the images was done to subtract the electronic noise and the background from the raw images.

A digital SLR camera with a resolution of 1920 × 1080 pixels was used for capturing the global flame images. For a set of images, the same exposure time and f-stop settings were used to compare the light intensity. No post-processing of the images was done for altering the contrast/brightness.

###### Details of Experimental Setup for Atomization Studies.

Figure 2(b) shows a schematic diagram used for studying the atomization characteristics of the liquid fuel jet injected into the cross-flowing air jet in unconfined and nonreacting conditions. The air and fuel supply systems were similar to those mentioned in Sec. 3.1. The injection parameters such as the fuel flow rate and the air flow rate and, hence, the liquid-to-air momentum flux ratio (q) were same to that used in the combustion studies for both the diameter cases. The objective of this study is to understand the influence of fuel injection momentum on macroscopic spray behavior such as jet trajectory and jet spread. The parameters computed in Table 2 are based on the average injection velocities of the fuel and air at the inlet of the combustor. The horizontal and vertical separations between the air injection port and the fuel injection port are marked in the diagram and are same as that used in the combustor (refer to Fig. 1 for more details).

A high-speed (Phantom VEO 640L) CMOS camera was used to capture nearly 1236 images at 1000 frames per second (fps). A maximum resolution of 2560 × 1600 pixels with the exposure time of 10 µs was used. These parameters were kept constant for all the test cases. A 1000 W halogen lamp with a light diffuser was used as a light source. The distance of the light source from the camera was adjusted to give the best quality images in the back-lighting mode.

###### Image Processing Methodology.

The instantaneous images (exposure time of 10 µs) were used to understand the primary breakup process. To obtain the other spray characteristics, a preprocessing algorithm was adopted. The background images (without any liquid fuel being injected) were first subtracted from the raw images. Then from the “Sobel” edge detection filter in matlab (by MathWorks®), edges of all the instantaneous images were obtained. The edges detected were then morphologically corrected and an arithmetic averaging of all the images was performed for a given test case. Averaging of the edge images eliminated the contribution from noise as it was seen that the noise filter did not affect the results. These averaged images were then used to obtain the spray characteristics.

Tan et al. [42] proposed a method to estimate the spray centerline and outer boundaries from the first and second moment of the pixel values making them independent of user interpretation of spray and hence, the same method was employed in the current study.

Figure 3(a) shows an averaged spray image with centerline trajectory and location of the origin (O). As large-size ligaments were present near the liquid and air impingement point, the origin was chosen slightly downstream of it to estimate the trajectories. The origin (O) is located at a distance of about 2dair from the tip of the air injector along the air jet axis. The spray outer edges can be identified in Fig. 3(b). The important nondimensional numbers and injection parameters used in the atomization study are listed in Table 2.

## Results and Discussion

###### A Note on Primary Breakup.

The primary breakup observed in two consecutive images is marked in Fig. 4 for the case of D1 (fuel injection diameter = 0.5 mm) and ϕ = 0.5. The time interval between these two consecutive images is 1 ms. There is a transfer of energy between the two jets leading to the breakup of the liquid jet. It is observed that the liquid jet is almost intact (no stripping of drops from it due to shear) until it impinges the air jet. Upon impingement with the air jet, it bends in the cross-flow direction almost immediately unlike the classical case of a liquid jet in uniform cross-flow where the liquid column bends gradually. The two circles show the ligaments being formed due to the stretching of the liquid column. These ligaments then further disintegrate into droplets due to strong aerodynamic forces. Xia et al. [39] found the primary characteristics of free air-on-water impinging jets similar to that of twin-fluid coaxial atomizers. In the present case, the impingement angle is 90 deg. The liquid structures observed in the near field view of the spray in the present study have some resemblance to those observed by Xia et al. [39]. Figure 5 shows the near field views of spray for the two diameter cases, i.e., D1 and D2 for a range of ϕ corresponding to the thermal loading conditions similar to that of the combustor.

As the air flow rate is fixed in the present case, the increase in the fuel flow rate is seen to promote the atomization as with the increase in fuel flow rate, the liquid-to-air momentum flux ratio (q) also increases. From Fig. 5 for D1 case, the liquid column forms large blobs of liquid which then disintegrate into ligaments for low fuel flow rates corresponding to ϕ = 0.4 and 0.5. For higher flow rates, the creation of a liquid sheet is observed near the impingement point. These sheets under the action of strong aerodynamic force disintegrate into ligaments which later break into drops of varying sizes until the critical condition is reached. For D2 case, the liquid jet did not form for ϕ less than 0.56 because of low injection velocity. Only two cases corresponding to ϕ = 0.7 and 0.8 are shown in Fig. 5 to identify the near-field structures. The sheet formation and its disintegration into ligaments can be observed similar to the previous case. Comparing the two injection diameter cases for ϕ = 0.7 and 0.8, it is seen that the D2 case forms larger near field structures than D1 case. There is a possibility of the formation of large droplets from these structures as compared with D1 case. These large droplets have higher evaporation times that can affect the combustor performance. Thus, there is a decrease in atomization quality as we move from D1 to D2 case. This effect is felt as an increase in emissions from the combustor which will be discussed in the sections to follow (Sec. 4.2.3).

###### Jet Penetration.

The dispersion of fuel within the fuel-air mixer (in gas turbine combustors) typically depends on the penetration of the jet in cross-flow spray-plume. Hence, it is an important parameter governing the mixing of fuel and air inside the combustion chamber. The jet penetration is commonly defined by spray-core centerline or the plume's outer-edge in the literature using a large number of image-processing algorithms. Plenty of trajectory correlations [43] have been proposed by the researchers in the past for uniform cross-flow, and it has been observed that the liquid-to-air momentum flux ratio (q) is an important parameter governing the trajectory of the spray-plume. It should be noted that the value of the parameter q in the present study is less than 1. This is because of the high air injection velocity (∼72 m/s), whereas in typical gas turbine combustors, it is about 30–40 m/s [2]. The high injection velocity of air is an important characteristic to achieve conditions closer to CDC [12]. In the present study, the spray centerline and the outer-edge trajectories have been obtained using the algorithms used in Ref. [42] with certain modifications. The centerline trajectory is estimated by finding out the locus of the center of intensity (CI) points at every downstream location (pixels) from the chosen origin for the averaged images (preprocessing methodology described in Sec. 3.2.1). The CI can be mathematically [42] expressed as given in Eq. (1), Display Formula

(1)
$CI(y)=∑x=0n[xI(x,y)]∑x=0nI(x,y)$
where “x” is the location along the x-axis (i.e., downstream), “I(x, y)” is the intensity of the pixel at the corresponding (x, y) location, and “n” represents the upper limit of “x.” The spray edges were estimated by taking the standard deviation about the centerline trajectory as the jet in cross-flow sprays are asymmetric about the mean. It was found that CI ± σ accurately traced the jet boundaries in the present case and, hence, higher multiples of σ were not necessary to determine the jet boundary. The mathematical representation [42] of the same is given in Eqs. (2) and (3), Display Formula
(2)
$σupper(y)={(∑(x=CI(y))n[I(x,y)])−1∑(x=CI(y))n[(x−CI(y))2I(x,y)]}2$
Display Formula
(3)
$σlower(y)={(∑x=0CI(y)[I(x,y)])−1∑x=0CI(y)[(x−CI(y))2I(x,y)]}2$
where σupper is the standard deviation after the CI(y) location and σlower is the standard deviation before the CI(y) location.

Figure 6 represents a comparison of the centerline trajectories for the two injection diameter cases, obtained by the methodology described above. It can be seen that the jet in D1 case penetrated farther than D2 case. This is attributed to the higher jet momentum (q) in D1 case for a given ϕ. The penetration varies along the length of the spray. However, the penetration near the impingement point is important as it influences the size of ligaments at the onset of formation.

Figure 7 represents the outer edge trajectories of spray for the two injection diameter cases. As the jet penetrates into the cross-flow, the droplets formed travel downstream (x-direction) with the cross-flow momentum and disperse along the transverse direction (y-direction) as the spray entrains the surrounding air. It can be observed that the spray is deflected toward the fuel injector from the axis of the air injector. This deflection was observed for all operating cases. The idea of estimating centerline trajectories and the spray edges in the present study was to show that the smaller injection diameter had a higher penetration and hence, a better fuel dispersion in the combustor resulting in better emission characteristics. However, it should be noted that the flow field inside the combustor (confined and reacting) is much more complex than this simple representation due to the large product gas recirculation within the combustor. The method used to describe the jet penetration and outer edge trajectories in the present study to report macroscopic averaged spray properties is an inexpensive tool to gain useful insights into the physics of the problem.

###### Global Flame Images.

The global flame images for non-premixed cases (D1 and D2) were taken at the same camera settings to identify the global flame features. The photographs were taken with a black background for enhancing the visibility of the reaction zone. Figure 8(a) shows the global flame images for D1 case. It can be seen from the figure that as ϕ is increased, the visible light intensity from the flame increases with flame occupying a large volume of the combustor. A long bluish flame region can be identified extending approximately from the fuel injection location near the top-left of the combustor to the stagnation zone at the bottom. It can also be observed that as ϕ is increased, the flame that is confined to the left half of the combustor starts widening somewhere near the middle of the combustor. This widening is observed to increase with an increase in ϕ. This widening of the flame can be attributed to the increased penetration of the fuel jet (increased momentum) into the air jet as ϕ is increased. Similar global flame characteristics were also observed for D2 case (see Fig. 8(b)). The combustor did not operate for equivalence ratios below 0.56 for D2 case as the fuel flow rate was not enough to form a jet through the tube because of low injection velocity. The same was also observed for the D1 case for ϕ less than 0.45.

###### OH* Chemiluminescence Images.

The OH* chemiluminescence images for D1 and D2 cases are presented in Figs. 9(a) and 9(b), respectively. The OH* intensity is presented in arbitrary units.

It can be observed from the OH* images that for a given ϕ, the reaction zone is wider for D1 case than D2 case. This widening is attributed to the higher penetration (also see Fig. 8(a)) of the fuel jet into the air jet for D1 case. Higher penetration leads to disintegration of the liquid jet into finer droplets, which lead to better dispersion and subsequent mixing inside the combustor. For both the diameter cases, the reaction zone is seen to widen with the increase in ϕ due to increase in liquid-to-air momentum ratio, which leads to higher penetration. It can also be seen that the majority of the heat release happens in the left portion of the combustor in the vicinity of the air jet. As the ϕ is increased keeping the air-flow constant, the fuel penetration and dispersion alter considerably leading to better fuel-air mixing and hence, stable flames.

###### Emission Characteristics.

The emission characteristics, i.e., NOx and CO corrected to 15% O2 for both non-premixed (direct injection (D1 and D2)) and PP cases are presented in Figs. 10 and 11, respectively.

For D2 case, NOx of about 9 ppm is obtained at ϕ = 0.6. As ϕ is increased, NOx emission is found to increase due to an increase in the flame temperature. Peak NOx emission of about 11 ppm is observed for ϕ = 0.8. The NOx emissions for D1 case are found to be lower than that of D2 case at all operating equivalence ratios. This can be attributed to the better atomization process in D1 case (see Figs. 57). The higher penetration in D1 case can possibly lead to finer droplets. These droplets can then undergo rapid evaporation leading to better fuel-air mixing and forming a combustible mixture, which burns at leaner local equivalence ratios thus avoiding near-stoichiometric burning zones. This, in turn, can suppress hot regions in the combustor leading to low emissions. The NOx for D1 case is about 6 ppm at ϕ = 0.5 and the maximum of about 9 ppm is at ϕ = 0.8. For the PP case, NOx emission of 0.5 ppm is observed at ϕ = 0.6 and it increases with an increase in temperature and is about 4 ppm for ϕ = 0.8. The premixed-prevaporized case forms the baseline case with very low NOx emissions. The emissions from smaller fuel injection diameter (D1) case is lower than the D2 case; however, still there is a possibility to further reduce the NOx emissions closer to the premixed-prevaporized case.

CO of about 45 ppm at ϕ = 0.8 is observed in D2 case, whereas it is about 70 ppm at ϕ = 0.8 in case of D1. As ϕ is decreased, the CO emissions shoot up rapidly for both the diameter cases. This might be due to the poor atomization leading to inefficient mixing of the liquid fuel with the air. Also at lower ϕ, the temperatures are lower and hence the conversion of CO to CO2 is suppressed [2]. For the PP case, the CO emissions are almost negligible as compared with the D1 and D2 cases. The minimum CO emission for the PP case is about 5 ppm at ϕ = 0.7 and maximum of 9 ppm at ϕ = 0.8. From Fig. 11, we can see that with a decrease in fuel injection diameter, the CO emissions lie closer to the PP case suggesting smaller diameter would give better trends.

Overall, the results suggest very low NOx and CO emissions using liquid fuel in a reverse-cross-flow configuration. It should be kept in mind that these emissions are obtained without the use of a conventional atomizer by directly injecting the liquid fuel into the cross-flow of air.

Ethanol can be extracted from the biomass, i.e., from the fermentation of sugar, corn, agricultural wastes, etc., and can be used for power generation using state-of-the-art combustion technologies. In this study, ethanol is shown to yield very low emissions and this combustion technique be extended to the real gas turbine fuels such as Jet-A. Proper scaling of the combustor geometry and flow parameters needs to be done to match the actual gas turbine requirements. Arghode and Gupta [12] estimated pressure loss for similar CDC combustors and found it to be less than 5%, which is a necessary requirement for the operation of gas turbine combustors.

###### Lean Operational Limit.

To obtain lean operational limit (shown in Fig. 12), the fuel flow rate was decreased slowly keeping the air flow rate constant until the blowoff occurred for both non-premixed and premixed cases (D1, D2, PP). Generally, wider stability is desired to achieve higher turndown ratio for gas turbine applications. Also, operation at overall leaner equivalence ratios results in low emissions. For the present investigation, D1 case has the lean operational limit of ɸ = 0.45, whereas it is ɸ = 0.56 for D2 case. For the PP case, the lean operational limit is about ɸ = 0.58. The lower lean operational limit for D1 case might be due to higher penetration of the liquid jet into cross-flow. The higher penetration leads to better dispersion of the fuel droplets and sustains combustion at leaner equivalence ratios. For D2 case, as the ϕ is decreased, the fuel injection velocity decreases rapidly and hence the penetration decreases. This results in the formation of ligaments and droplets, which are comparatively large in size. Hence, it is difficult to sustain combustion at very low ϕ for D2 case.

From the above discussion, we see that the lower leaner operational limits could be obtained by utilizing lower diameter tubes for the injection of the liquid fuel. We can also see that lean operational limit is highest for the PP case which is undesirable.

## Conclusions

The current work presents an experimental investigation of a reverse-cross-flow combustor utilizing simple injection of the liquid fuel jet directly into the cross-flowing oxidizer stream without the use of a conventional atomizer. Atomization study using high-speed imaging technique was also performed in unconfined and nonreacting conditions to report averaged spray properties and their impact on combustion characteristics. Combustion and emissions characteristics for two injection diameters were studied. Single-digit NOx and double-digit CO emissions (in ppm) were achieved at ϕ = 0.7 for the non-premixed (direct injection) cases. For the premixed-prevaporized case, single-digit emissions of both NOx and CO were obtained for all operating conditions. Through the high-speed imaging, primary breakup characteristics and averaged jet centerline and outer-edge trajectories were computed and it was seen that the smaller (D1) diameter case had higher penetration into the cross-flow than the D2 case, suggesting that a smaller fuel injection diameter lead to better fuel/oxidizer mixing and resulted in lower NOx and CO emissions. At lower equivalence ratios (lower fuel injection velocity), higher CO emissions suggest that the combustor performance was affected by poor atomization of the issuing liquid fuel jet. Thus, if the diameter of the tube is small enough to form a jet and thus have enough momentum to penetrate the air jet, cleaner combustion could be sustained for leaner equivalence ratios utilizing the reverse-cross-flow geometry. Burning liquid fuels with minimum emissions using novel combustion techniques indicates the potential for its use in various industrial and power generation applications.

## Acknowledgements

The authors acknowledge the financial support received as Early Career Research Award (ECR/2016/001205) from the Science and Engineering Research Board (SERB), Department of Science and Technology (DST), Government of India for undertaking this work. The support extended by Dr. Abhijit Kushari for acquiring the chemiluminescence images is gratefully acknowledged.

## References

Lieuwen, T. C., and Yang, V., 2013, Gas Turbine Emissions, Cambridge University Press, Cambridge.
Lefebvre, A. H., 1999, Gas Turbine Combustion, 2nd ed., Taylor & Francis, London.
Emami, M. D., Shahbazian, H., and Sunden, B., 2018, “Effect of Operational Parameters on Combustion and Emissions in an Industrial Gas Turbine Combustor,” ASME J. Energy Resour. Technol. 141(1), p. 012202.
Miller, J. S., and Bowman, C. T., 1989, “Mechanism and Modeling of Nitrogen Chemistry in Combustion,” Prog. Energy Combust. Sci. 15(4), pp. 287–338.
Scenna, R., and Gupta, A. K., 2015, “Preheats Effects on JP8 Reforming Under Volume Distributed Reaction Conditions,” ASME J. Energy Resour. Technol. 138(3), p. 032202.
Said, A. O., Khalil, A. E. E., and Gupta, A. K., 2016, “Dual-Location Fuel Injection Effects on Emissions and NO*/OH* Chemiluminescence in a High Intensity Combustor,” ASME J. Energy Resour. Technol. 138(4), p. 042208.
Scenna, R., and Gupta, A. K., 2018, “The Influence of the Distributed Reaction Regime on Fuel Reforming Conditions,” ASME J. Energy Resour. Technol. 140(12), p. 122002.
Arghode, V. K., and Gupta, A. K., 2013, “Role of Thermal Intensity on Operational Characteristics of Ultra-Low Emission Colorless Distributed Combustion,” Appl. Energy, 111, pp. 930–956.
Arghode, V. K., Gupta, A. K., and Yu, K. H., 2010, “Investigation of Non-Premixed and Premixed Distributed Combustion for GT Application,” 48th AIAA Aerospace Sciences Meeting Including New Horizons Forum and Aerospace Exposition, Orlando, FL, Jan. 4–7.
Arghode, V. K., and Gupta, A. K., 2011, “Investigation of Reverse Flow Distributed Combustion for Gas Turbine Application,” Appl. Energy, 88(4), pp. 1096–1104.
Arghode, V. K., and Gupta, A. K., 2011, “Investigation of Forward Flow Distributed Combustion for Gas Turbine Application,” Appl. Energy, 88(1), pp. 29–40.
Arghode, V. K., and Gupta, A. K., 2010, “Effect of Flow Field for Colorless Distributed Combustion (CDC) for Gas Turbine Combustion,” Appl. Energy, 87(5), pp. 1631–1640.
Arghode, V. K., and Gupta, A. K., 2009, “Effect of Confinement on Colorless Distributed Combustion for Gas Turbine Engines,” 45th AIAA/ASME/SAE/ASEE Joint Propulsion Conference Exhibit, Denver, CO, Aug. 2–5.
Khalil, A., Arghode, V. K., and Gupta, A. K., 2010, “Distributed Combustion With Swirl for Gas Turbine Application,” 49th AIAA Aerospace Sciences Meeting Including New Horizons Forum and Aerospace Exposition, Orlando, FL, Jan. 4–7.
Kim, H. S., Arghode, V. K., and Gupta, A. K., 2009, “Combustion Characteristics of a Lean Premixed LPG-Air Combustor,” Int. J. Hydrogen Energy, 34(2), pp. 1045–1053.
Khalil, A., and Gupta, A. K., 2016, “On the Flame-Flow Interaction Under Distributed Combustion Conditions,” Fuel, 182, pp. 17–26.
Gupta, A. K., Bolz, S., and Hasegawa, T., 1999, “Effect of Air Preheat Temperature and Oxygen Concentration on Flame Structure and Emission,” ASME J. Energy Resour. Technol. 121(3), pp. 209–216.
Tsuji, H., Gupta, A. K., Hasegawa, T., Katsuki, M., Kishimoto, K., and Morita, M., 2002, High Temperature Air Combustion: From Energy Conservation to Pollution Reduction, CRC Press, Boca Raton.
Cavaliere, A., and de Joannon, M., 2004, “Mild Combustion,” Prog. Energy Combust. Sci. 30(4), pp. 329–366.
Lammel, O., Schmitz, G., Aigner, M., and Krebs, W., 2010, “FLOX® Combustion at High Power Density and High Flame,” ASME J. Eng. Gas Turbines Power, 132(12), p. 121503.
Weber, R., and Smart, J. P., 2005, “On the (MILD) Combustion of Gaseous, Liquid, and Solid Fuels in High Temperature Preheated Air,” Proc. Combust. Inst. 30(2), pp. 2623–2629.
Derudi, M., and Rota, R., 2011, “Experimental Study of the Mild Combustion of Liquid Hydrocarbons,” Proc. Combust. Inst. 33(2), pp. 3325–3332.
Reddy, V. M., and Kumar, S., 2013, “Development of High Intensity Low Emission Combustor for Achieving Flameless Combustion of Liquid Fuels,” Propul. Power Res. 2(2), pp. 139–147.
Crane, J., Neumeier, Y., Jagoda, J., Seitzman, J., and Zinn, B. T., 2006, “Stagnation Point Reverse-Flow Combustor Performance With Liquid Fuel Injection,” ASME Turbo Expo 2006: Power for Land, Sea, and Air, Barcelona, Spain, May 8–11.
Gopalakrishnan, P., Bobba, M. K., Radhakrishnan, A., Neumeier, Y., and Seitzman, J. M., 2007, “Characterization of the Reacting Flowfield in a Liquid-Fueled Stagnation Point Reverse Flow Combustor,” 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, Jan. 8–11.
Bobba, M. K., Gopalakrishnan, P., Periagaram, K., and Seitzman, J. M., 2007, “Flame Structure and Stabilization Mechanisms in a Stagnation Point Reverse Flow Combustor,” ASME Turbo Expo 2007: Power for Land, Sea, and Air, Montreal, Canada, May 14–17.
Bobba, M. K., Gopalakrishnan, P., Radhakrishnan, A., Seitzman, J. M., Neumeier, Y., Zinn, B. T., and Jagoda, J., 2006, “Flame Stabilization and Mixing Studies in a Novel Ultra-Low Emissions Combustor,” 44th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, Jan. 9–12.
Arghode, V. K., Khalil, A., and Gupta, A. K., 2012, “Fuel Dilution and Liquid Fuel Operational Effects on Ultra-High Thermal Intensity Distributed Combustor,” Appl. Energy, 95, pp. 132–138.
Sharma, P., and Arghode, V. K., 2017, “Experimental Investigation of Low Emission Liquid Fuelled Reverse Cross Flow Combustor,” ASME 2017 Gas Turbine India Conference, Bangalore, India, Dec. 7–8.
Sallam, K. A., Aalburg, C., and Faeth, G. M., 2004, “Breakup of Round Nonturbulent Liquid Jets in Gaseous Crossflow,” AIAA J. 42(12), pp. 2529–2540.
Wu, P.-K., Kirkendall, K. A., Fuller, R. P., and Nejad, A. S., 1997, “Breakup Processes of Liquid Jets in Subsonic Crossflows,” J. Propul. Power, 13(1), pp. 64–73.
Broumand, M., and Birouk, M., 2016, “Liquid Jet in a Subsonic Gaseous Crossflow: Recent Progress and Remaining Challenges,” Prog. Energy Combust. Sci., 57, pp. 1–29.
Becker, J., and Hassa, C., 2003, “Liquid Fuel Placement and Mixing of Generic Aeroengine Premix Module at Different Operating Conditions,” ASME J. Eng. Gas Turbines Power, 125(4), pp. 901–908.
Becker, J., Heitz, D., and Hassa, C., 2004, “Spray Dispersion in a Counter-Swirling Double-Annular Air Flow at Gas Turbine Conditions,” Atomization Sprays, 14(1), pp. 15–35.
Gong, X., Choi, K. J., and Cernansky, N. P., 2006, “Lean Direct Wall Injection Mode Atomization of Liquid Jets in Swirling Flow,” J. Propul. Power, 22(1), pp. 209–211.
Tambe, S., Elshamy, O., and Jeng, S. M., 2007, “Liquid Jets Injected Transversely Into a Shear Layer,” 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, Jan. 8–11.
Tambe, S., and Jeng, S. M., 2010, “Three-dimensional Penetration and Velocity Distribution of Liquid Jets Injected Transversely Into a Swirling Crossflow,” 22nd Annual Conference on Liquid Atomization and Spray Systems (ILASS Americas), Cincinnati, OH, May 16–19.
Sikroria, T., Kushari, A., Syed, S., and Lovett, J. A., 2014, “Experimental Investigation of Liquid Jet Breakup in a Crossflow of a Swirling Air Stream,” ASME J. Eng. Gas Turbines Power, 136(6), p. 061501.
Xia, Y., Alshehhi, M., Hardalupas, Y., and Khezzar, L., 2017, “Spray Characteristics of Free Air-on-Water Impinging Jets,” Int. J. Multiphase Flow, 100, pp. 86–103.
Jadidi, M., Moghtadernejad, S., and Dolatabadi, A., 2016, “Penetration and Breakup of Liquid Jet in Transverse Free Air Jet With Application in Suspension-Solution Thermal Sprays,” Mater. Des., 110, pp. 425–435.
Tan, Z. P., Zinn, B. T., Lubarsky, E., Bibik, O., Shcherbik, D., and Shen, L., 2016, “A Moments-Based Algorithm for Optimizing the Information Mined in Post-Processing Spray Images,” Exp. Fluids, 57(19), pp. 1–13.
No, S.-Y., 2015, “A Review on Empirical Correlations for Jet/Spray Trajectory of Liquid Jet in Uniform Cross Flow,” Int. J. Spray Combust. Dyn. 7(4), pp. 283–314.
View article in PDF format.

## References

Lieuwen, T. C., and Yang, V., 2013, Gas Turbine Emissions, Cambridge University Press, Cambridge.
Lefebvre, A. H., 1999, Gas Turbine Combustion, 2nd ed., Taylor & Francis, London.
Emami, M. D., Shahbazian, H., and Sunden, B., 2018, “Effect of Operational Parameters on Combustion and Emissions in an Industrial Gas Turbine Combustor,” ASME J. Energy Resour. Technol. 141(1), p. 012202.
Miller, J. S., and Bowman, C. T., 1989, “Mechanism and Modeling of Nitrogen Chemistry in Combustion,” Prog. Energy Combust. Sci. 15(4), pp. 287–338.
Scenna, R., and Gupta, A. K., 2015, “Preheats Effects on JP8 Reforming Under Volume Distributed Reaction Conditions,” ASME J. Energy Resour. Technol. 138(3), p. 032202.
Said, A. O., Khalil, A. E. E., and Gupta, A. K., 2016, “Dual-Location Fuel Injection Effects on Emissions and NO*/OH* Chemiluminescence in a High Intensity Combustor,” ASME J. Energy Resour. Technol. 138(4), p. 042208.
Scenna, R., and Gupta, A. K., 2018, “The Influence of the Distributed Reaction Regime on Fuel Reforming Conditions,” ASME J. Energy Resour. Technol. 140(12), p. 122002.
Arghode, V. K., and Gupta, A. K., 2013, “Role of Thermal Intensity on Operational Characteristics of Ultra-Low Emission Colorless Distributed Combustion,” Appl. Energy, 111, pp. 930–956.
Arghode, V. K., Gupta, A. K., and Yu, K. H., 2010, “Investigation of Non-Premixed and Premixed Distributed Combustion for GT Application,” 48th AIAA Aerospace Sciences Meeting Including New Horizons Forum and Aerospace Exposition, Orlando, FL, Jan. 4–7.
Arghode, V. K., and Gupta, A. K., 2011, “Investigation of Reverse Flow Distributed Combustion for Gas Turbine Application,” Appl. Energy, 88(4), pp. 1096–1104.
Arghode, V. K., and Gupta, A. K., 2011, “Investigation of Forward Flow Distributed Combustion for Gas Turbine Application,” Appl. Energy, 88(1), pp. 29–40.
Arghode, V. K., and Gupta, A. K., 2010, “Effect of Flow Field for Colorless Distributed Combustion (CDC) for Gas Turbine Combustion,” Appl. Energy, 87(5), pp. 1631–1640.
Arghode, V. K., and Gupta, A. K., 2009, “Effect of Confinement on Colorless Distributed Combustion for Gas Turbine Engines,” 45th AIAA/ASME/SAE/ASEE Joint Propulsion Conference Exhibit, Denver, CO, Aug. 2–5.
Khalil, A., Arghode, V. K., and Gupta, A. K., 2010, “Distributed Combustion With Swirl for Gas Turbine Application,” 49th AIAA Aerospace Sciences Meeting Including New Horizons Forum and Aerospace Exposition, Orlando, FL, Jan. 4–7.
Kim, H. S., Arghode, V. K., and Gupta, A. K., 2009, “Combustion Characteristics of a Lean Premixed LPG-Air Combustor,” Int. J. Hydrogen Energy, 34(2), pp. 1045–1053.
Khalil, A., and Gupta, A. K., 2016, “On the Flame-Flow Interaction Under Distributed Combustion Conditions,” Fuel, 182, pp. 17–26.
Gupta, A. K., Bolz, S., and Hasegawa, T., 1999, “Effect of Air Preheat Temperature and Oxygen Concentration on Flame Structure and Emission,” ASME J. Energy Resour. Technol. 121(3), pp. 209–216.
Tsuji, H., Gupta, A. K., Hasegawa, T., Katsuki, M., Kishimoto, K., and Morita, M., 2002, High Temperature Air Combustion: From Energy Conservation to Pollution Reduction, CRC Press, Boca Raton.
Cavaliere, A., and de Joannon, M., 2004, “Mild Combustion,” Prog. Energy Combust. Sci. 30(4), pp. 329–366.
Lammel, O., Schmitz, G., Aigner, M., and Krebs, W., 2010, “FLOX® Combustion at High Power Density and High Flame,” ASME J. Eng. Gas Turbines Power, 132(12), p. 121503.
Weber, R., and Smart, J. P., 2005, “On the (MILD) Combustion of Gaseous, Liquid, and Solid Fuels in High Temperature Preheated Air,” Proc. Combust. Inst. 30(2), pp. 2623–2629.
Derudi, M., and Rota, R., 2011, “Experimental Study of the Mild Combustion of Liquid Hydrocarbons,” Proc. Combust. Inst. 33(2), pp. 3325–3332.
Reddy, V. M., and Kumar, S., 2013, “Development of High Intensity Low Emission Combustor for Achieving Flameless Combustion of Liquid Fuels,” Propul. Power Res. 2(2), pp. 139–147.
Crane, J., Neumeier, Y., Jagoda, J., Seitzman, J., and Zinn, B. T., 2006, “Stagnation Point Reverse-Flow Combustor Performance With Liquid Fuel Injection,” ASME Turbo Expo 2006: Power for Land, Sea, and Air, Barcelona, Spain, May 8–11.
Gopalakrishnan, P., Bobba, M. K., Radhakrishnan, A., Neumeier, Y., and Seitzman, J. M., 2007, “Characterization of the Reacting Flowfield in a Liquid-Fueled Stagnation Point Reverse Flow Combustor,” 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, Jan. 8–11.
Bobba, M. K., Gopalakrishnan, P., Periagaram, K., and Seitzman, J. M., 2007, “Flame Structure and Stabilization Mechanisms in a Stagnation Point Reverse Flow Combustor,” ASME Turbo Expo 2007: Power for Land, Sea, and Air, Montreal, Canada, May 14–17.
Bobba, M. K., Gopalakrishnan, P., Radhakrishnan, A., Seitzman, J. M., Neumeier, Y., Zinn, B. T., and Jagoda, J., 2006, “Flame Stabilization and Mixing Studies in a Novel Ultra-Low Emissions Combustor,” 44th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, Jan. 9–12.
Arghode, V. K., Khalil, A., and Gupta, A. K., 2012, “Fuel Dilution and Liquid Fuel Operational Effects on Ultra-High Thermal Intensity Distributed Combustor,” Appl. Energy, 95, pp. 132–138.
Sharma, P., and Arghode, V. K., 2017, “Experimental Investigation of Low Emission Liquid Fuelled Reverse Cross Flow Combustor,” ASME 2017 Gas Turbine India Conference, Bangalore, India, Dec. 7–8.
Sallam, K. A., Aalburg, C., and Faeth, G. M., 2004, “Breakup of Round Nonturbulent Liquid Jets in Gaseous Crossflow,” AIAA J. 42(12), pp. 2529–2540.
Wu, P.-K., Kirkendall, K. A., Fuller, R. P., and Nejad, A. S., 1997, “Breakup Processes of Liquid Jets in Subsonic Crossflows,” J. Propul. Power, 13(1), pp. 64–73.
Broumand, M., and Birouk, M., 2016, “Liquid Jet in a Subsonic Gaseous Crossflow: Recent Progress and Remaining Challenges,” Prog. Energy Combust. Sci., 57, pp. 1–29.
Becker, J., and Hassa, C., 2003, “Liquid Fuel Placement and Mixing of Generic Aeroengine Premix Module at Different Operating Conditions,” ASME J. Eng. Gas Turbines Power, 125(4), pp. 901–908.
Becker, J., Heitz, D., and Hassa, C., 2004, “Spray Dispersion in a Counter-Swirling Double-Annular Air Flow at Gas Turbine Conditions,” Atomization Sprays, 14(1), pp. 15–35.
Gong, X., Choi, K. J., and Cernansky, N. P., 2006, “Lean Direct Wall Injection Mode Atomization of Liquid Jets in Swirling Flow,” J. Propul. Power, 22(1), pp. 209–211.
Tambe, S., Elshamy, O., and Jeng, S. M., 2007, “Liquid Jets Injected Transversely Into a Shear Layer,” 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, Jan. 8–11.
Tambe, S., and Jeng, S. M., 2010, “Three-dimensional Penetration and Velocity Distribution of Liquid Jets Injected Transversely Into a Swirling Crossflow,” 22nd Annual Conference on Liquid Atomization and Spray Systems (ILASS Americas), Cincinnati, OH, May 16–19.
Sikroria, T., Kushari, A., Syed, S., and Lovett, J. A., 2014, “Experimental Investigation of Liquid Jet Breakup in a Crossflow of a Swirling Air Stream,” ASME J. Eng. Gas Turbines Power, 136(6), p. 061501.
Xia, Y., Alshehhi, M., Hardalupas, Y., and Khezzar, L., 2017, “Spray Characteristics of Free Air-on-Water Impinging Jets,” Int. J. Multiphase Flow, 100, pp. 86–103.
Jadidi, M., Moghtadernejad, S., and Dolatabadi, A., 2016, “Penetration and Breakup of Liquid Jet in Transverse Free Air Jet With Application in Suspension-Solution Thermal Sprays,” Mater. Des., 110, pp. 425–435.
Tan, Z. P., Zinn, B. T., Lubarsky, E., Bibik, O., Shcherbik, D., and Shen, L., 2016, “A Moments-Based Algorithm for Optimizing the Information Mined in Post-Processing Spray Images,” Exp. Fluids, 57(19), pp. 1–13.
No, S.-Y., 2015, “A Review on Empirical Correlations for Jet/Spray Trajectory of Liquid Jet in Uniform Cross Flow,” Int. J. Spray Combust. Dyn. 7(4), pp. 283–314.

## Figures

Fig. 1

(a) Photograph and (b)–(d) schematics of the reverse cross-flow combustor

Fig. 2

Schematics of the experimental setup for (a) combustor performance and (b) atomization study

Fig. 3

Averaged spray image with (a) origin (O) and jet centerline trajectory and (b) spray outer edges

Fig. 4

Consecutive images depicting ligament formation for D1 case at ϕ = 0.5

Fig. 5

Primary breakup characteristics for D1 and D2 cases at various ϕ cases

Fig. 6

Jet centerline trajectories for (a) ϕ = 0.7 and (b) ϕ = 0.8

Fig. 7

Spray outer boundaries for (a) ϕ = 0.7 and (b) ϕ = 0.8

Fig. 8

Global flame images for (a) D1 and (b) D2 diameter cases at various ϕ

Fig. 9

OH* images for (a) D1 and (b) D2 cases at various ϕ

Fig. 10

NOx emissions

Fig. 11

CO emissions

Fig. 12

Lean operational limit

## Tables

Table 1 Experimental parameters
Table 2 Atomization parameters and experimental conditions (ρG = density of air, ρL = density of ethanol, UG = air injection velocity, σ = surface tension of ethanol, all properties are taken at 300 K and atmospheric pressure). D1 = 0.5 mm, D2 = 0.8 mm.

## Errata

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections