Flameless combustion with liquid fuel: A review focusing on fundamentals and gas turbine application Academic research paper on "Mechanical engineering"

Contents lists available at ScienceDirect

Applied Energy

journal homepage: www.elsevier.com/locate/apenergy

Flameless combustion with liquid fuel: A review focusing on .■»crossMark

fundamentals and gas turbine application

Fei Xing3'*, Arvind Kumarb, Yue Huang3, Shining Chana, Can Ruana, Sai Guc*, Xiaolei Fand

a School of Aerospace Engineering, Xiamen University, Xiamen 361005, PR China b Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur 208016, India cDepartment of Chemical and Process Engineering, University of Surrey, Guildford GU2 7XH, UK d School of Chemical Engineering and Analytical Science, The University of Manchester, Manchester M13 9PL, UK

HIGHLIGHTS

• Technical and physical challenges in liquid flameless combustion are presented.

• The research status about the different liquid fuels used for flameless combustion are summarized.

• Flameless studies related with gas turbine engines are discussed in details.

• Several research areas needed to improve the flameless combustion's application are provided.

ARTICLE INFO ABSTRACT

Flameless combustion has been developed to reduce emissions while retaining thermal efficiencies in combustion systems. It is characterized with its distinguished features, such as suppressed pollutant emission, homogeneous temperature distribution, reduced noise and thermal stress for burners and less restriction on fuels (since no flame stability is required). Recent research has shown the potential of flameless combustion in the power generation industry such as gas turbines. In spite of its potential, this technology needs further research and development to improve its versatility in using liquid fuels as a source of energy. In this review, progress toward the application of the flameless technique was presented with an emphasis on gas turbines. A systematic analysis of the state-of-the-art flameless combustion and the major technical and physical challenges in operating gas turbines with liquid fuels in a flameless combustion mode was presented. Combustion characteristics of flameless combustion were explained along with a thorough review of the modelling and simulation of the liquid fuel fed flameless combustion. A special focus was given to the relevant research on the applications of flameless combustion to the inner turbine burners. The paper was concluded by highlighting the recent findings and pointing out several further research directions to improve the flameless combustion in gas turbines, including in-depth flow and combustion mechanisms, advanced modelling, developed experimental technology and comprehensive design methods aiming at gas turbine flameless combustors.

© 2017 Elsevier Ltd. All rights reserved.

Article history:

Received 31 August 2016

Received in revised form 2 February 2017

Accepted 4 February 2017

Keywords:

Flameless combustion Liquid fuel

Moderate and Intense Low oxygen Diffusion

(MILD)

Gas turbine

Contents

1. Introduction.......................................................................................................... 29

2. Fundamentals and challenges in flameless combustion research................................................................ 30

2.1. Fundamentals of flameless combustion...............................................................................30

2.2. Technical and physical challenges...................................................................................31

3. State of the art flameless combustion with liquid fuel........................................................................ 33

3.1. Bio fuels........................................................................................................33

3.2. Diesel..........................................................................................................33

* Corresponding authors. E-mail addresses: f.xing@xmu.edu.cn (F. Xing), sai.gu@surrey.ac.uk (S. Gu).

http://dx.doi.org/10.1016/j.apenergy.2017.02.010 0306-2619/® 2017 Elsevier Ltd. All rights reserved.

3.3. Kerosene.......................................................................................................34

3.4. Other liquid hydrocarbons.........................................................................................36

4. Flameless combustion for gas turbines..................................................................................... 36

4.1. FLOX combustor.................................................................................................37

4.2. Flameless burner.................................................................................................39

4.3. Colorless distributed combustion....................................................................................40

4.4. Model flameless combustor........................................................................................40

4.5. Flameless combustion based on trapped vortex........................................................................42

4.5.1. Netherlands and Italy...................................................................................... 42

4.5.2. Germany and USA........................................................................................ 42

4.6. Flameless inner-turbine burner.....................................................................................44

4.7. Summary.......................................................................................................44

5. Knowledge gap and future research....................................................................................... 45

5.1. Working conditions and atomization.................................................................................48

5.2. Simulation and measurement.......................................................................................48

5.3. Design methods and process.......................................................................................49

6. Conclusion remarks.................................................................................................... 49

Acknowledgment...................................................................................................... 50

References........................................................................................................... 50

1. Introduction

Fossil fuels have been used by the society since thousands of years. With rapid expansion and development of technology, the society rapidly reached a point of overconsumption of any kind of natural resources. The facts that these resources are exhaustible, and our utilization rate is high, lead to the concern that in a very near future the fossil fuels will run out.

The concerns about energy first surfaced during the energy crisis of the seventies of the past century. This leads to development of nuclear and solar energy. As can be seen in Fig. 1, fossil fuels still account for over 3/4th of the world's total energy supply today (Mtoe in Y axis label means million tonnes of oil per year) [1]. Although nuclear and solar energy may provide more promise for the future, fossil fuel cannot be quickly replaced for all applications, at least in the near future because of their several advantages including non-radioactivity, safety, matured utilization technologies, high conversion efficiency and cost (the data is from http://www.bp.com/).

However, when one considers the negative impact of fossil fuels, besides the limited reserves, concerns over environmental issues turn out to be quite serious. In 1992, the United Nation Conference on Environment and Development provided global efforts to protect our environment. Then, at the Kyoto protocol in 1997, many developed countries discussed the possibility and requirement of reducing the actual carbon emissions by 7% below

2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 Year

Fig. 1. Global energy consumption by source in 2012 and in future [1].

the 90s level over the next 10 years [2]. During the end of the last century more and more attention was given to the utilization of fossil fuels. Indeed, many problems are directly related to their consumption as fuel, and the pollutant, such as CO, CO2, hydrocarbon, soot and NOx, that they produce during their combustion [3].

At present, gas turbines are a principal source of new powergenerating capacity throughout the world, and the dominant source for air-breathing flight vehicles as well. Gas turbines will continue to be an important combustion-based energy conversion device for many decades to come, for example, for air craft propulsion, ground-based power generation, and mechanical-drive applications [4]. Large aerospace companies continue to push gas turbines to new limits, for example, Pratt and Whitney plans to produce future engines with 15-25% lower CO2, 70-85% lower NOx, CO, and UHC, and have 15-25% lower fuel burn while providing reductions in noise, improved thrust-to-weight ratio, etc. [5].

The impact of the emissions, such as NOx, CO and UHC on the environment is critical now. The gas turbines always work at high operation temperature which leads to the formation of NOx. In this way, the gas turbines are one of the main sources of pollutant in the lower layers of atmosphere [4]. Thermodynamically improvement of combustion efficiency and power output suggests the higher flame temperature (see Fig. 2). However, the long residence time of molecular nitrogen at peak temperature zone with high availability of oxygen (for few seconds above 1800 K or for only milliseconds above 2300 K) will lead to the formation of NOx. They are formed from the oxidation of the free nitrogen in the combustion air or fuel, and are called ''thermal NOx." They are mainly a function of the stoichiometric adiabatic flame temperature of the fuel, which is the temperature reached by burning a theoretically correct mixture of fuel and air in an insulated vessel.

With this big background, besides the high energy demand, every gas turbine designer is looking forward to converting the natural form of energy (gaseous fuels and liquid fuels) to useful work in a green manner with high efficiency. Therefore, many new techniques and new types of gas turbine combustors were developed for reducing the peak temperature and availability of oxygen in order to reduce the NOx and CO. For example, stage combustor, rich-quench-lean (RQL), lean direct injection (LDI), lean premixed pre-vaporized (LPP), etc. [4]. More recently, nitrogen dilution [6], steam-water injection [7] and humidified exhaust recirculation [8] technologies have been also used to reduce NOx and CO emission in turbine engine. In this paper, we will discuss another technique called flameless combustion and its application in gas turbine area.

& 40 o o o

20 18 16 14

12 ¡3

10 20 30 40 50 60 70 Percent of takeoff power

Fig. 2. Relation between the powers output of gas turbine and the amount of emissions [4].

In 1971, an interesting phenomenon was observed at high furnace temperature above auto ignition temperature of the mixture, that with exhaust gas recirculation, no flame could be seen and no UV-signal could be detected. Despite that, the combustion was stable and smooth, the fuel was burnt completely and the NOx emissions were close to zero. This is called flameless combustion [9]. Its acronyms are usually related to the inlet air, such as: Excess Enthalpy Combustion (EEC), High Temperature Air Combustion (HiTAC) in Japan, or Moderate and Intense Low Oxygen Diffusion (MILD) Combustion in Italy, Flameless Oxidation (FLOX) in Germany, Colorless Distributed Combustion (CDC) and Low NOx Injection (LNI) in the USA. No matter how it is called, it implies hot oxidizers with exhaust gas recirculation at high turbulence level, and under typical conditions, it occurs with no flame, very lean and stable.

Flameless combustion has many merits over traditional combustion models, such as low pressure oscillation in combustor which reduces the noise, high combustion efficiency which helps to decrease the consumption of energy, and most importantly, extremely low pollutant exhaust emissions. The low emission property makes flameless combustion a more environmental friendly way of air transportation and combat. After all, the greenhouse exhaust emissions created by aircraft engines are directly emitted into the stratosphere, the ozone layer can be greatly destroyed and the global warming problem will arise.

Developments in gas turbines are primarily focused on improving stability, emission and reducing noise. The aerospace industry continues to strive for higher compression ratios to increase engine efficiency, while burning at leaner conditions to reduce fuel consumption and lower NOx production. A new combustion technology is needed to meet the demands of higher pressure, leaner fuel combustion while meeting increasingly more strict emission standards. So far many topics have been reviewed to develop the fundamental knowledge of combustion in gas turbines and to understand how modern technologies are changing the way in which these engines operate. The uniform temperature distribution characteristics and low rate of emissions of flameless combustion are desirable in gas turbines. Since most of the aircraft engines and industrial combustors are operated with liquid fuels and research on flameless combustion with liquid fuel is limited with only few on-going laboratory works, it is important to review the flameless combustion with liquid fuels. Such review should help to understand the role of flameless combustion on the improvement of gas turbine performance.

The technical review report presented here will first discuss major technical and physical challenges in flameless combustion. Secondly, review of studies on flameless combustion of selective liquid fuels will be presented. Thirdly, flameless combustion studies related with the gas turbines all over the world will be discussed. Thereafter, the authors will also discuss some basic research areas in flameless combustion application in gas turbines that need improvement. The discussion will focus on future directions, flameless combustion, and inner turbine burner.

2. Fundamentals and challenges in flameless combustion research

2.1. Fundamentals of flameless combustion

Flameless combustion is based on exhaust gas recirculation in which NOx formation is suppressed without compromising the thermal efficiency. In flameless combustion, the re-circulated exhaust gas is defined as the exhaust gas that is re-circulated and mixed into combustion air before the reaction. The peak temperature of reaction can be reduced with increased inserts but beyond certain limits it leads to unstable combustion and blow off. Because of distinct benefits of flameless combustion, various efforts have been made to define the condition boundaries for flameless combustion in different systems.

The relationship between the exhaust gas recirculation rate and the furnace temperature in diffusion combustion is established by Wunning and Wunningin [10] for the non-premixed flameless combustion of methane in a laboratory furnace system, in which the relative internal recirculation rate (KV) was defined as in Eq. (1):

Kv = -r

Mf + MA

where ME, MF and MA are mass flow-rates of internal recirculation gas, initial fuel and air jet, respectively. Accordingly, a stability limit diagram was mapped as seen in Fig. 3 showing the boundaries of different combustion modes. Stable combustion (zone A) is achievable over the whole range of operating temperature with a narrow recirculation rate window. For the higher relative recirculation rate, if the operating temperature is lower than the self-ignition temperature, the flame becomes unstable leading to the extinguishment as seen the 'No Reaction' zone in Fig. 3. However, the combination of

high temperature (above the self-ignition temperature), and high exhaust gas recirculation rate (greater than 3) can result in steady and stable reactions without flame (flameless zone C).

At a high recirculation rate, the re-circulated exhaust gas dilutes oxygen concentration in the air stream and preheats the air stream leading to an increase in chemical reaction time and hence a smaller Damkohler number, which is an essential prerequisite for achieving flameless combustion. The uniform thermal field is one of the main characteristics of flameless combustion. The temperature rise in flameless combustion is moderate and dispersed due to distributed heat release during the combustion. In addition, it was observed that the effect of air preheating on NOx emission becomes insignificant with high recirculation rate of exhaust gas.

The map of different combustion regimes based on the inlet temperature (Tin) and temperature increase (AT) coordinate was measured for a methane combustion system (molar fractions: CH4/O2/N2 = 0.1/0.05/0.85), as seen in Fig. 4 [11]. The self-ignition

0 2 4 6

Exhaust of Gas Recirculation

Fig. 3. Stability limits diagram as functions of furnace temperature and relative internal recirculation rate: zone A - stable combustion, zone B - unstable combustion, zone C-flameless zone [10].

temperature (Tsi) of 1000 K, which is estimated based on a numerical computation, is used as the boundaries to map different combustion zones. For regions with higher AT, i.e. feedback and high temperature combustion, the required temperature for sustaining the flame is guaranteed, hence known as the traditional combustion process. When conditions satisfy Tin > Tsi and AT < Tsi, flameless combustion was identified (marked as the Mild Combustion in Fig. 4). The mild combustion process cannot be sustained without preheating the reactants. Though in results shown in Fig. 4 efforts were made to map the boundaries of flameless combustion for specific systems, it is difficult to develop a generic criterion to define flameless combustion for various combustion systems with different fuels. Therefore, the conclusions from the identification of the operating boundaries for flameless combustion are still restricted to the specific fuel, and flame observation is still the most common approach for research on such identification.

Flame is a product of self-propagating exothermic reaction that usually has a luminous reaction zone, in which the temperature increases from the minimum to the maximum, and accordingly the intermediate and product concentration increases. The high temperature in the reaction zone promotes decomposition of fuel molecules leading to the formation of free radicals. Reactions between reactive radical species are extremely exothermic for the product to be formed in excited electronic state with high energy level.

As seen in Fig. 5, when the temperature of the mixture is higher than the self-ignition temperature, as the recirculation rate increases gradually, there is no flame front, no visible flame, no UV or ionization detection, and no noise or roar. A transparent flame combustion called flameless combustion forms. In this case, homogeneous combustion occurs, the reaction zone temperature increases only few hundred degrees, and CO and NOx are abated to very small residual values.

2.2. Technical and physical challenges

According to Li et al. [13] and Noor at al. [14], technical challenges of gas flameless combustion can be summarized with the help of following issues: the way to judge flameless phenomena; physics of flow and combustion including high turbulence inten-

Tsi=1000K

Tin, k

Tsi=1000K

Fig. 4. Distinguished combustion state with inlet temperature and temperature raise (residence time = 1 s, atmospheric pressure) [11].

sity, fuel injection momentum and temperature demand of the oxidant; experimental facilities and test methods; combustor design methods; combustion testing; and technology development tracking. In addition, from the authors' point of view, flameless combustion with liquid fuel is different from the one with gas fuel, therefore, the physical challenges may also include: atomization and evaporation of different liquid fuel; fuel injection and interaction with the high temperature gas; reaction mechanism of flameless combustion with liquid fuels; flow control of gas recirculation; gas entrainment; turbulence and reaction interactions, etc. Overall, in order to develop a practical flameless combustor, each of these aspects must be addressed by either modelling or experimental approaches or combining both.

As previously stated, the design of flameless combustion systems depends largely on accurate estimation of similarity between the liquid fuel and the gas fuel systems, which requires proper understanding of the mechanism of flameless formation. For example, from 2006 to 2009, the researchers from University of Cincinnati [15-20] modified the old version of the EU burner leading to the development of a new burner with high turbulence intensity. The new combustion system uses fuel injectors with air jet flows into the recirculation zone [20]. The design is made such that a very high swirl forms and creates a very strong recirculation zone by turbulent processes. The fuel is then injected directly into the recirculation zone, and mixed with re-circulated exhausted gas to reduce the reaction rate. Chemiluminescence measurements were employed to provide data of species concentrations. Results showed that the reaction mechanism of flameless combustion is still unclear and a multi-pronged approach (e.g., when and how

to realize flameless combustion) is needed to address this challenge. It was hoped that the new, more advanced ''physics-based" methods and the theory for flameless can be applied correctly to this kind of problem.

The flameless combustion occurs when both high levels of turbulence and oxidizer temperature exist for an evenly distributed fuel/air mixture. This kind of combustion has several advantages for application in gas turbine combustors, such as:

• stable operation at low equivalence ratios

• producing low concentrations of NOx and CO

• significant noise reduction

• evenly distributed temperature pattern in the reaction zone and at the outlet.

What mentioned above suggest flameless combustion as a desirable mode of combustion for gas turbines, as lean instabilities become more of a problem for higher pressure ratio engines. The implementation of flameless combustion has been successfully demonstrated in non-adiabatic type combustion systems, such as industrial furnaces [13,21]. However, operational parameters for gas turbine combustors are very different from those of industrial furnaces, which possess several major challenges to applying flameless combustion to gas turbines as follows.

Firstly, the operational range of flameless combustion combus-tors is narrow, and hence cannot satisfy the dynamic requirements for a gas turbine, especially that of an aircraft. The stable flameless combustion can be achieved only when the specific operating conditions are maintained. However, the operation of gas turbine

combustors is dynamic and difficult to maintain at the specific conditions. For example, when the aircrafts climb or take maneuver-able flights, the air flow and fuel injection often change dramatically, leading to the interruption of flameless combustion. Secondly, the O2 concentration and pressure in a gas turbine is higher than in an industrial burner, hence a harsher environment exists in a gas turbine. The gas turbine cycle efficiency increases with Turbine Inlet Temperature (TIT), producing high average operational temperature that is beneficial to the flameless combustion. However, gas turbine combustors operate at a low overall equivalence ratio, i.e., very high O2 concentration in the combustor. Since flameless combustion is a low O2 concentration combustion phenomenon, creating a suitable operational zone may be difficult. Thirdly, the volume required by flameless combustors is large as compared to the conventional combustors, which is an important constraint for aircraft applications. A large amount of combusted gases need to be circulated within the combustor to decrease the concentration of O2. This also results in lower volumetric heat densities as compared to existing conventional gas turbine combus-tors. Finally, it is also challenging to develop sufficient gas recirculation for flameless combustion without adding a mechanical system for creating recirculation. The difficulty remains in controlling the reaction rate in a way in which sufficient mixing is achieved before completion of the reaction.

Closely associated with liquid fuel flameless combustion for gas turbines, an important near term challenge for flameless combustion is to determine the suitable conditions for specific application using various fuels. Systematic fundamental studies will provide optimum design guidelines for a specific application. A few technical objectives are as follows:

• Fundamental understanding of effects of fuel types, high pressure, initial air temperature and exhaust gas recirculation ratio on flameless combustion using liquid fuel.

• Design of flameless combustors, considering geometry, liquid droplet breakup and evaporation, fuel supply and gas matching, control of the equivalence ratio in the combustion zone.

• Development of numerical models for the simulation of flame-less phenomena in gas turbines at working conditions.

• Development correlations between numerical models and ground tests, and validation.

3. State of the art flameless combustion with liquid fuel

Based on the previous flameless combustion investigations with gas fuels [13], the consensus for this type of combustion appears to be:

• Preheating of combustion air with high temperature exhaust recirculation and high-speed injections of air and fuel are the main requirements for achieving flameless combustion.

• Strong entrainment of high-temperature exhaust gases, diluting fuel and air jets, is the key to maintaining flameless combustion.

• Essential environment conditions for the establishment of flameless combustion are: local oxygen concentration < 5-10%, and local temperature > the fuel self-ignition temperature in the reaction zone. Such conditions can be achieved by high dilution of the reactants with the flue gas (N2 and CO2-rich exhaust gas).

• In comparison with conventional combustion, the thermal efficiency of flameless combustion can be increased by more than 30% but the NOx emission can be reduced by more than 70%, when a regenerator is used to recycle the waste heat of flue gases [22].

Flameless combustion has been extensively investigated for systems using gaseous fuels such as hydrogen, methane and etha-

nol. However, to the best of the authors' knowledge, there is no pilot scale or large-scale investigation of flameless combustion using liquid fuels although liquid fuels are essential to gas turbines. Therefore, it becomes imperative to understand the characteristics of flameless combustion with liquid fuels to increase the engine efficiency and reduce NOx and CO emissions. In the following sections of this contribution, the focus is on reviewing the available work on the flameless combustion with liquid fuels.

3.1. Bio fuels

Biomass-derived fuels have assumed an ever-increasing role in the projected sustainable energy supply system of the future, considering the fossil fuel supplies vary with the increasing world energy demand. Some researchers published considerable interest in bio fuels in the flameless combustion systems in order to significantly lower the emission levels.

Flameless combustion with biomass fuel was first investigated with gaseous biomass fuels because of the similar physical properties and combustion mechanisms of gaseous fossil and biomass fuels. Kaneko lab at the University of Tokyo developed a MGT (micro gas turbine) system that use biomass gas as fuel. A trial combustor with three combustion zones and one dilution zone was designed for the MGT system. In this combustor, premixed gases are injected vertically into the main flow and mixed with the burned gas supplied from previous stage. At this moment, flameless combustion takes place and achieved the lower emissions [23]. Thereafter, effort was made to use liquid bio-ethanol as the fuel for the flameless combustion in micro gas turbine [24,25]. A comprehensive analysis is presented for the possibilities that are offered by a bio-fuel supplied to a micro-gas turbine, together with the attempt to a nearly zero-emission operation. It was also found that the efficiency of flameless combustion is close to that of conventional combustion with kerosene fuel if the fuel/ air ratio is maintained as the same. The emission levels can be further decreased by selecting a different location of the pilot injector. All the research works mentioned above are based on CFD without experiments validation. Meanwhile, the lack of species concentration and temperature in the combustors makes it difficult to understand the mechanism of the flameless combustion.

Ellis et al. [26] investigated the combustion process within a Power, Water Extraction, and Refrigeration system, known as PoWER (Fig. 7). Four liquid bio-fuels were tested for exhaust emission and soot behaviour in the PoWER gas turbine system as see in Fig. 8. The exhaust gas recirculation was applied to the system through a semi-closed cycle gas turbine system called the High-Pressure Regenerative Turbine Engine (HPRTE). The major objective in the development of this system is long-term operation on liquid biofuels flameless combustion, in order to demonstrate the potential for coupling such systems in a distributed generation mode to relatively small biofuel processing plants needed to take advantage of biomass resources. It is the first but important step for the liquid bio fuel.

3.2. Diesel

Torresi et al. [27] designed an aerodynamic ally staged swirled burner using the diesel as the fuel. The burner has been experimentally tested and numerically simulated under diluted and highly preheated inlet flow conditions. The staged injection is realized through a double coaxial air inlet with the same swirl orientation. The diesel is injected through a central atomizing nozzle characterized by very high range ability. The air was heated to 673 K and diluted by CO2 and H2O, and the concentration of O2 is 12.59%. It may be noted that there is no information about the diesel atomization and evaporation. The numerical simulation was in

Fig. 7. PoWER system diagram [23].

1.From Recuperator

2.Premixed Gases Injection Slit For Primary Combustion

3.Primary Combustion Zone(Premixed Combustion) 1

Air Air/Fuel Mixture

Fuel Burned Gas

4.Primary Gas Injection Nozzle for Secondary or tertiary Combustion

6.Dilution Air Inlet

7.To Turbine

[j 5.Primary or Tertiary Combustion Zone(Flameless Combustion )

Fig. 8. Schematic view of combustor in PoWER system [26].

a good agreement with experimental results, confirming that the burner, under properly operating conditions, is able to burn the fuel completely without a flame front and with a very uniform temperature field, as seen in Fig. 9.

In China, Professor Lin's research team also realized the flame-less combustion of liquid diesel in 2012 [28,29]. After Li et al. [13] firstly pointed out that flameless combustion occurred not only in high temperature low oxidant concentration but also in high temperature high oxidant concentration, Lin [28,29] also believes that air preheating is not an essential condition to attain flameless mode and that the air injection speed is more important to lower the whole reaction rate.

A cubical combustor was designed with a recirculation structure, as shown in Fig. 10, and flameless combustion experiments were carried out by using 0# diesel. It has been observed that with the increase of the injection momentum of the reactants, the combustion mode is converted from flame to flameless while the recirculation structure does not change. The liquid diesel was supplied by air blast atomizer, however, the performance of the atomizer was not mentioned. It can be concluded from the analysis in the works [28,29] that the air injection velocity and the entrainment and mixing with the high temperature exhaust gas are considered

as the most important factors in controlling the reaction rate and changing the combustion mode, and that the appropriate preheating of the combustion air is an effective way to enhance the combustion stability. The effect of the high temperature was not emphasized, which would reduce the cost of the experiment remarkably. Otherwise, the conclusions are very important and useful.

3.3. Kerosene

Since kerosene is a complex mixture of hydrocarbons, effort has been made to achieve flameless combustion using liquid kerosene after several attempts on achieving flameless combustion with the liquid hydrocarbons [21-35].

Experimental and numerical research has been carried out to use a two-stage combustor to investigate characteristics of flame-less combustion with kerosene. The design of the combustor is based on the injection of fuel and air at ambient conditions, as seen in Fig. 11a. The fuel injection and air injection schemes are believed to have an impact on fuel spray, increasing shear force and resulting in enhancing mixing and evaporation of droplets. The latest research published in 2014 [32] shows that the two-stage combus-

Fig. 9. Contours of temperature for flame (a) and flameless (b) conditions with the relative experimental images [27].

tor was not ideal for establishing flameless combustion. Therefore, a swirl-based combustor with a chamfer at the top of the combus-tor (see Fig. 11b) was considered, aiming at improving the droplet residence times and the recirculation rate. Observations from this study are summarized [32]. Firstly, flameless combustion was stabilized in the base combustor. However, for higher fuel flow rates flameless combustion was not achieved and unburned fuel accumulated in the combustor. Secondly, a chamfer near the exit in the modified combustor configuration helped increase the circulation rate and residence time. The outstanding performance of the burner with very low chemical and acoustic emissions at high heat release rates indicates the potential for various industrial and gas turbine applications.

Simulation and experimental research on flameless combustor based on trapped-vortex [33,34] were also carried out with kerosene. The zero dimensional and three dimensional numerical simulation was performed first. The emissions and temperature profile of the outlet are in good agreements with the cavity configuration. In the experimental work, the emphasis was to analyze the influences of the inlet air temperature, air flow-rate and equivalent ratio. The fuel is supplied by air atomization injectors, when the inlet temperature is above 550 K, and the reference velocity is higher than 10 m/s. The authors considered that the flameless combustion was stabilized by the trapped vortex. But the simulation results was not validated by the experiments (see Fig. 12).

3.4. Other liquid hydrocarbons

To avoid the complex mixture of real liquid hydrocarbon fuels, surrogate fuels are commonly used with two purposes, for designing better reproducible experimental tests, and for obtaining physical insights from well-controlled fundamental and kinetic studies. Since 2006, Gutmark's research team has been focusing on the flameless combustion and the use of gas fuel [15-19]. In 2009, they published a paper about the flameless combustion with several kinds of liquid hydrocarbons [20].

In a collaborative work between Goodrich Aerospace and the University of Cincinnati, a flameless burner was designed and tested [15]. In the first design, the burner was composed of a circle of premixed air/fuel jets directed to mix with the combustion products. The results were globally unsuccessful [15,16]. The burner demonstrated the usual premixed flame instabilities when combustion transits out of steady flameless mode. They found that it is a key factor to achieve the necessarily high turbulence intensity for flameless model.

So after this, researchers from Goodrich Aerospace and the University of Cincinnati modified the design of combustor to gain a more a very strong recirculation zone and to operate at the conditions typical to gas turbines as shown in Fig. 6. In particular, the emphasis was placed on achieving high mixing rate while

maintaining a low pressure drop across the fuel injection. A low-noise high-sensitivity microphone and thermocouples were used to determine temperature uniformity of the flameless model.

The research work has shown that operating the burner at high oxidizer temperature (from 325 °C to 525 °C) and relatively low pressure drop (from 3% to 5%) allows flameless combustion to occur while running very lean (equivalence ratio is from 0.7 to near LBO). Different fuels including propane (liquid), n-butane (liquid), n-pentane, n-hexane, toluene, jet-A and a blend of alkanes/alkenes centered on C9 were tested. All the fuels, with the exception of n-butane, showed very similar characteristics [20]. According to the conclusion of the research, we must take care in designing the compact flameless combustor for aerospace application. High turbulence and fast mixing would tend to produce greater magnitudes of acoustical pressure oscillations, this level of thermo-acoustic stability might cause the fluid and combustion instabilities.

Derudi et al. [36] from Italy also focused on the investigation of the sustainability of flameless combustion for liquid hydrocarbons using a dual-nozzle laboratory-scale burner. Air and gaseous fuel are fed through the bottom of the combustion chamber, as shown in Fig. 13a. After finding that configuration is not suitable for liquid fuels, the apparatus has been modified by implementing the double-nozzle (DN) inlet configuration as shown in Fig. 13c, where the preheated air enters the combustion chamber through the bottom nozzle, while the liquid fuel is injected as a well-dispersed and homogeneous spray through a water-cooled plain jet airblast atomizer. The two jets interact perpendicularly and mix with each other in a high turbulence region. The research has shown that the DN configuration allows to sustain flameless combustion by directly injecting different liquid hydrocarbons under the conditions that are previously established using a gaseous fuel as shown in Fig. 13b. DN configuration provides different results for liquid hydrocarbons, suggesting that the flameless combustion characteristics are probably more influenced by the physical state of fuels than the chain length of the hydrocarbon. For example, the flame-less combustion region in terms of the window Tavg (average combustion chamber temperature) -Kv (gas recirculation rate) space was found enlarged with the liquid hydrocarbons in comparison with the gaseous ones.

Though the chemical reaction mechanism involved in such flameless combustion burner remains unknown and numerical or experimental approaches are required to reveal this, such work does corroborate its potential for flameless combustion applications with liquid fuel.

4. Flameless combustion for gas turbines

The application described in this section is exploratory since many examples from gas turbines can be reported in which partial fulfillment of the flameless condition occurs, though most of them

Start Strut

Fig. 12. Schematic diagram of the flameless combustor [33,34].

use the gaseous fuel. Very few gas turbine systems have been built with the purpose to satisfy flameless combustion conditions, especially with liquid fuels. Nevertheless some of them deserve mention, because they partially fulfill flameless combustion conditions and they are of interest to foresee possible potentials and problems.

4.1. FLOX combustor

Flameless oxidation (FLOX) based on high internal flue gas recirculation for gas turbines was investigated [37-45]. Several liquid fuels were tested under atmospheric conditions [37] and similar results were obtained. Natural gas, as well as mixtures of natural gas and H2 were used as fuel at the second stage of the project [38,39]. The NOx and CO emissions were monitored under different operating conditions and at two air preheat temperatures of 703 K and 873 K, and 20 bar pressure with a maximum thermal power of 475 kW. The imaging of OH* chemiluminescence and planar laser-induced fluorescence (PLIF) of OH were applied in order to characterize the influence of various parameters, such as mixture composition, degree of premixing, velocities on the pollutant emissions, flame zone and the relative temperature distributions (see Fig. 14).

Successful operation of the FLOX® combustor equipped with an optical combustion chamber was demonstrated at high pressure as well. It was shown that the jet exit velocity of the fuel/air mixtures had a strong influence on the mixing process within the combustion chamber, and that the low emission operating range is increased with the increase in jet exit velocity. With adding H2 to the natural gas, the range of stable operation could be extended;

however, the NOx emissions are also increased, probably due to inhomogeneities in the temperature distribution. In this research, instantaneous Chemiluminescence Images with an exposure time of 40 is were successfully used to help to understand the flame structure and the extinction happening.

A reversed flow combustor for small or medium size gas turbines, called optional schemes for an adiabatic flameless oxidation combustor (FLOXCOM) [41,42], was studied. A procedure for calculating the thermodynamic parameters and the gas properties, at every stage of the combustion process, was developed in 2004 [41]. It is shown that the oxygen concentration at the end of the combustion zone increases when the difference between combustion and air inlet temperatures decreases. It is believed that the key to a low NOx flameless gas turbine combustor is to keep a low O2 concentration within the combustion zone and a relatively even temperature distribution within the combustion chamber. In order to meet the above-mentioned stringent demands, the schematic of the combustion methodology is presented as in Fig. 15 [42].

A schematic of the conceptual gas turbine combustor operating on the proposed combustion methodology is illustrated in Fig. 16a. The various junctions and stations described in the thermodynamic model (Fig. 15) are also depicted in the Fig. 16b for better understanding. Salient features of this concept as compared to other proposed combustors for gas turbines are:

• Fuel (CH4) is injected into the deficient O2 and high CO2, H2O concentration recirculation zone, to make sure that the fuel is injected in an optimum environment where flameless combustion can take place.

I-PLIF Measurement Air-cooled plane windows

OH-PLIF and OH detection

Nozzle Configurations

Air-..^ ■ Fuel—> -

Laser at 283nm

dl=0mm Non premixed

dl=24mm premixed

Fig. 14. 3D drawing of the combustion chamber with the FLOX burner (left) and the schematic of the FLOX burner with the OH-PLIF measurement planes marked (right) [39].

Primary Air

! T23=1582K !

02=10.7% 1 C02=5.2%

Junction

| T32=1837K I © Station I 02=8.6% | |_C02=6.2% _ |

I Ti=700K

I 02=21% "

C02=0.0%

Pre-combus tion

| T24=2050K i | 02=5.3% , C02=7.6%

Main Combustion

Secondary Air

Fuel=CH4

Heat Exchanger

Heat extraction=80KJ

T6=1500K

I 06=13.4% I

I C06=3.9% I

T51=765K I 1 02=23% I C02=0.0%

Fig. 15. Schematic of the proposed new flameless combustion methodology [42].

Fuel Inlet

Secondary Air

Primary Air

Composite

Metallic Fins

Fuel * - Injector

Mixing Holes

Heat Transfer Surface

Ï i Up- -U=L J I I ■ Vw*

^uTatiott F ^^ ^

Xo»e ___— —

Combustion . Mixing

----M^- -X5

Zone Zone

Recirculation Zone

Fig. 16. Schematic of a gas turbine combustor operating on the newly proposed cycle [42].

Certain amount of energy is transferred from primary combustion zone to the secondary cooling air, thus reducing combustion temperature and hence limiting the NOx formation. Energy is added in two steps, partially in the pre-combustion region (between ''4" and ''2") of the recirculation zone and partially in the main combustion zone (between ''2" and ''3"). Thus,

it limits the maximum temperature rise below the temperature above which NO formation would have increased exponentially (ffi2000 K).

In 2011, Levy cooperated with Melo et al. from Technical University of Lisbon and developed the schematic of a gas turbine

Combustion Temperarute

300-450mm

Emission Probe

Microphone

Combustion Chamber Fuel Injector

Heat Plenum Air

Fig. 18. Schematic of EU burner (a) and modified combustor (b) [15,20].

Support Flange

combustor operating on the new cycle mentioned in Figs. 15 and 16 [43]. The main characteristic of such design is the formation of a large recirculation vortex, stimulated by the momentum of the incoming air jets and aided by the specific geometry of the combustion chamber, as seen Fig. 17. The air from the compressor enters at station 1 and is split into two streams with identical flow rates. One stream is entrained and its oxygen concentration is diluted by the recirculated combustion products and directed toward point 2. At that point the fuel (methane) is injected and mixed. The combustible mixture ignites at station 3 after a certain ignition delay time. Combustion occurs between points 3 and 4, and thereafter the combustion products are split, partially recirculating with fresh air and partially exiting the combustor while diluting with fresh air (point 5). The two streams mix and exit the combustor at point 6, at the required combustor's exit temperature, typically determined by the performance of the turbine located immediately downstream.

The new combustor was experimentally investigated and found as stable over a relatively wide range of operating conditions. At specific conditions, i.e., air inlet temperature = 425 K, equivalent ratio: 0.24-0.28, the NOx level was measured as lower than 10 ppm. However, the CO level was measured as relatively high, about 700-1200 ppm, showing that there was still space for improving the design to increase combustion efficiency without

hampering the combustor performance. The authors thought that the flameless combustion did not happen in the recent design theme.

4.2. Flameless burner

As previously stated, from 2006 to 2009, the research team from University of Cincinnati [15-20] modified the old version of the EU burner, and developed a new concept combustion using the swirl method to achieve the necessary high turbulence intensity. In the first approach, the burner was tested unsuccessful [15,16]. The new burner was designed to form a very strong recirculation zone and to operate at conditions typical to gas turbines. In particular, the emphasis was placed on achieving high mixing rate while maintaining a low pressure drop across the fuel inject (see Figs. 7 and 18b). Inlet temperature, pressure drop and combustor geometry were varied to determine the boundaries of flameless combustion within the constraints acceptable to current and future aircraft engines. Data were collected starting at equivalence ratio of 0.7 and decreasing to lean-blow-out (LBO) to determine the flammability limits of the flameless burner. Important aspects of flameless combustion include uniform temperature distribution, low emissions, and decoupling between heat transfer, fluid dynamics, and acoustics.

The research work considered the high turbulence intensity, maximized the residence time, and assisted fast fuel/air/combustion product mixing, and suggested high inlet temperature as the most important aspect for the flameless combustor design. The points above are common points as other researchers have also proposed. Except those points, the research work using the gaseous propane as the fuel has also shown that:

• The transition to flameless mode from regular combustion is gradual and a definite transition point cannot be defined clearly.

• High air mass flow rates promote evenly distributed flame, resulting in good mixing, strong reaction, and less NOx formation.

• The increase of the inlet temperature reduces the LBO limit, and expands the range of equivalence ratios where the flameless mode occurs.

4.3. Colorless distributed combustion

From 2010 to 2016 [46-52], the research team from University of Maryland put forward a new concept called colorless distributed combustion (CDC), which is based on the principle of high temperature air combustion (HiTAC) [22].

It is a fact that the major challenge in the CDC (no visible flame signatures) research is about the design of burner for application to gas turbine combustors. Such design determines the effectiveness of the combustion process in the CDC mode, and hence controls the ignition time, temperature uniformity and pollutant emissions especially.

The research team compared the CDC with rich burn-quick quench-lean burn (RQL) combustion and other low emission combustion models, and CDC showed great potential to reduce NOx and CO emissions in addition to improved pattern factor and low noise. Many different combustor models (as shown in Fig. 19) were designed to prove that the mixing between the combustion air and product gases to form hot and diluted oxidant prior to its mixing with the fuel is critical.

Experimental and numerical methods were used to find out that the flow field configurations and significant recirculation of gases had great effect on fuel jet characteristics as well as the fuel/air mixing. Distributed reaction zone, better thermal filed uniformity and lower NO (7 ppm, phi = 0.7) and CO (20 ppm, phi = 0.7)

emissions were observed for non-premixed model [46-48]. In 2012 [49], they achieved ultra-low NOx emissions for both the novel premixed (1 ppm) and non-premixed (4 ppm), low CO emission (30 ppm at phi = 0.5) and very low pressure fluctuations (<0.025%) characteristics of CDC, when Tair = 600 K (see Fig. 20).

Furthermore, in Refs. [50,52], results were represented for both non-swirling "linear" flow and swirling flow. Swirling flow exhibited high velocity region at the core of the combustor to further promote mixing and entrainment of reactive species. All of these research aspects will provide valuable information for significant improvements of gas turbine combustors performance.

4.4. Model flameless combustor

In 2010, researchers from Chinese Academy of Sciences focused on the dynamic characteristics of a flameless model combustor [53-55]. The model combustor is a can-type reverse flow combustor (Fig. 21) with two parts: head and chamber. The head of the combustor comprises the air and the fuel distribution units. All the air is injected into the chamber by co-flow injection through the twelve main nozzles. There is a concave recirculation structure at the head of combustion chamber in which the burned gas recirculates and mixes with fresh reactants. The positional relationship between the nozzles and the concave structure raises the mixture's temperature above the fuel self-ignition temperature, and dilutes the air to reduce the concentration of O2, thereby achieving flame-less combustion. Additionally, there is a pilot nozzle fixed in the center of the concave recirculation structure, which is designed for ignition and maintaining combustion in case that the equivalence ratio is low. Therefore, in the combustor's mixed mode, the fuel is partially premixed. The pure methane and nitrogen-diluted methane (volume ratio of CH4:N2 = 1.0) were used as the main fuel, and propane was used for the pilot.

In this model combustor, there are three working modes (see Fig. 22). These are Pilot-only mode, Mixed mode (the pilot and main nozzles work together), and Flameless mode. The pilot nozzle is shut off, while the main nozzles continue to supply fuel, attaining flameless combustion.

The highlight of the research is the use of the dynamic pressure sensors to detect the dynamic pressure signal. An autoregressive (AR) model was used to estimate the power spectrum. Further-

Fig. 19. Line Diagram of CDC test facility and the different flow field configurations [46].

0.5 0.6 0.7 0.8 Equivalence Ratio

Fig. 20. NO and CO emissions in non-premixed and premixed conditions (p denotes premixed) for the CDC combustor [49].

Gas Cell

Pilot i

Nozzle

Fuel Nozzles

Concave Air CeU Structure

Air Fuel Air

Fig. 21. Cross-section view of flameless combustion model combustor and fuel nozzle [53].

Air Fuel Air

Fig. 22. The combustor ignition sequence (a) Mode I: pilot nozzle only; (b) Mode II: pilot and main nozzles; (c) Mode III: main nozzles only [53,54].

more, there was no dominant oscillation amplitude in flameless combustion. The flameless combustion mode showed lower combustion noise and no thermal-acoustic oscillation problems while achieving ultra-low NOx and CO emissions. However, when the

pilot flame coexisted with the main combustion flame, instability was excited at certain equivalence ratios. The experimental method is very helpful to identify the transformation from the conventional combustion to the flameless mode.

Primary Air Fuel

Main Air

Fig. 23. Sketches of TVC geometry with and without diffuser and their injection strategy [57].

4.5. Flameless combustion based on trapped vortex

The trapped vortex is a design concept for the gas turbine com-bustor. It is currently being developed for the low-emission and high-performance combustion systems in aircrafts and ground power gas turbines [56]. The trapped vortex combustor (TVC) design is based on a fast mixing process of hot combustion products and reactants, and can function appropriately when a vortex is ''trapped" within a cavity where reactants are injected and efficiently mixed. Since part of the combustion occurs within the recirculating (vortex) zone, the reactants mix with the products in a ''typical" flameless regime, thereby, the flameless combustion process may occur in this zone.

4.5.1. Netherlands and Italy

The first group who combines flameless combustion and trapped vortex combustor (TVC) is the research team from Delft University of Technology and University of Rome in 2006 [57]. They agreed that reactants are mixed at high temperature in a TVC by means of a vortex and burn at a low oxidizer concentration and at high recirculation factors like a flameless burner. However, they also pointed out that the amounts of the mainstream flow from these two types of combustion, which were involved in the combustion process, were different.

The first configuration (see Fig. 23a) was proposed as the solution for low-power and single-cavity combustion chambers. In order to stabilize the vortex, the mass flow rate of the primary air injected should be lower than the rate of the main air stream. But the simulation results showed that the combustion primarily occurs next to the primary air injectors and downstream the cavity along the outlet pathway. It means that the combustion is incomplete because of inefficient mixing. The next design of geometry contains the diffuser (see Fig. 23b) which is capable of supplying sufficient oxidizer to the secondary cavity, thus providing a more stable vortex. The second design was found able to generate two independent vortices, each trapped within its cavity, by means of two (almost) independent air streams.

A novel double cavity TVC geometry has been numerically investigated. Critical issues highlighted in this work are the vortex stability and the combination of these two kinds of geometry as shown in Fig. 23. The emissions and pressure drop are well predicted, but the Outlet Temperature Distribution Factor (OTDF)

parameter is still too high. The outflow temperature is found non-uniform, hence the characteristics of the flameless mode are not identified clearly. Therefore, geometry modification was proposed for further improvement.

From 2010 to 2012, Di Nardo et al. developed a burner prototype which promotes flows in cavities to stabilize the flame [58,59]. The right combination of the cavity design and the fluid dynamics inside the cavity enhances the formation of stable vortex, so that it promotes the heat and mass transfer processes with the incoming flow and hence establishes the appropriate conditions for flameless combustion.

TVC is chosen for the annular combustion chamber, which is made up of a single cavity with air introduced tangentially (Fig. 24). The tangential air flow creates a vortex that fills the entire chamber to realize a flameless combustion relying only on internal mixing of reagents and exhaust gases. Use of the high speed jets helps to recycle more products, while the depression recalls a larger quantity of products and accelerates the vortex. It is found that even for fast rotating vortices, mixing is not fast enough to prevent the production of NOx in large amount.

After the first unsatisfactory attempt, Di Nardo et al. developed the new prototype as shown in Fig. 25 [59]. The new design was simplified with a linearized sector of the annular chamber (square section of 190 x 190 mm). The vortex in such a combustion chamber is created by two flows, while other streams of air and syngas fuel are distributed among the tangential flows, feeding the ''vortex heart". The air stream, which is introduced in the middle, provides primary oxidant to the combustion reaction, while the tangential flows provide the excess air to cool the walls and the combustion products.

The optimization principle of the prototype geometry is to balance different flows to establish the vortex that can fill the entire combustion chamber. The large exhaust recirculation and the good mixing is able to assure satisfying the fundamental prerequisites for a flameless combustion regime. A sensitivity analysis allowed determining the optimal operating conditions for which the contemporary reduction of the major pollutants species was achieved.

4.5.2. Germany and USA

In 2007, the cooperation on the combustion noise characteristics of a Flameless Trapped-Vortex Reheat Burner (FTVRB) between University of Technology Berlin and University of Cincinnati

Fig. 25. Burner geometry and path lines from LES simulation [59].

Emission Probe

Thermocouple

FTVR B

Cavity [ j

I j Thermocouple

Photomultiplier

Microphones

Air Supply

Main Fuel Lance .¡2

---------Ij

Loudspeaker

_Axial Cavity Air Fuel Air

Center Body

OOÔOC

Tangential Cavity Air Fuel Air

Tangential Fuel Air

Fig. 26. FTVRB set-up [18,53].

[18,60] were published. In Fig. 26 schematics of the FTVRB com-bustor test rig (left) and the combustor cavity (right) are depicted. In FTVRB burner, a small part of the main flow is deflected into the cavity to establish the cavity circumferential vortex. The vortex is strengthened by axial injection of the secondary cavity air through six circumferential holes (see Fig. 26 right). Additionally, the combination of the axial and tangential air injection with the cavity vortices forms a spiral vortex. The spiral vortex enhances mixing and provides an important controlling factor for the cavity combustion process. The cavity is fuelled in the middle of the cavity

depth. The stabilizing effect of the cavity pilot flame may also amplify the burner's combustion noise. Strong axial air jets stabilized the flame deep within the cavity, intensified the combustion process by enhancing the recirculation of hot combustion products, and thus caused a reduction in CO emissions. For strong tangential air injection, the pilot flame was located in the interaction zone of main and cavity flow, reducing the cavity temperature and the NOx emission.

The researcher [60] obtained much information through the tests with pressure sensors, PIV, thermocouples and flame photos.

The results showed that flameless combustion in the cavity could not be ensured over its entire working range. The combination of the local heat release at the flame front and the mixing dynamics, which also includes coherent structures in the cavity, leaded to local oscillation of the heat release. The cavity air injection pattern and the cavity equivalence ratio were identified as major parameters for controlling and optimizing the FTVRB performance with respect to pollutants and combustion noise. A relation between the noise and heat releases was presented, however, there was still need for digging the mechanism of coupling of heat release and acoustic characteristics in flameless combustion.

4.6. Flameless inner-turbine burner

For ground-based gas turbines, Wang et al. [45] investigated the techno-economic feasibility of applying the technology of flame-less combustion to a simple gas turbine cycle, compared to that of conventional combustion technology. For flameless combustion, the main characteristic is to recirculate internal flue gas into the combustion zone for the dilution of combustion. Due to the high recirculation ratios, the maximum reaction temperature in flame-less oxidation operation is much lower than in conventional combustion, thus reducing NOx formation considerably. This reduces the net power production by 5.38% to 413 MW and lowers the heating valve efficiency from 33.5% to 32.7%. The main environmental change is the 92.3% reduction in NOx emissions from 112 to 8.6 mg/m3 (5% O2).

For aero gas turbines, the introduction of flameless combustion technology in gas turbines will be of great interest because it has been demonstrated as a stable form of combustion yielding simultaneously low concentrations of CO and NOx, intrinsic thermo acoustic stability and uniform temperature distribution within the limits of a gas turbine. Of course, there are still some restrictions of flameless combustion, such as the high inlet temperature, which in a gas turbine would be the compressor discharge temperature. However, there will be some other ways to use the flameless technology, such as ITB (inner-turbine burner) [61]. ITB is a new technology and developed especially for civil aero gas turbine com-bustors without an afterburner. Its target is to increase the thrust-to-weight ratio and to widen the range of engine operation. Combustion would extend from main combustors into the turbine passage, which is troublesome at first sight, because it can lead to an increase in heat transfer challenges. However, a significant benefit can result from augmented burning in the turbine. In Refs. [61,62], the thermodynamic cycle analysis was performed to demonstrate the performance of aero-engines with and without the interstage turbine burners. Results showed that the inner-turbine

burner produces extra thrust, but at the cost of increased fuel consumption for current compressor ratio values. A 10%-20% increase in efficiency can be achieved. At the higher compressor ratio values and high flight Ma number projected for the future, the turbine burner is superior in both thrust and fuel consumption.

The first and only published paper that mentioned the combination of ITB and flameless combustion is reported by Ochrymiuk and Badur [63]. Flameless combustion is applied into the second SEV (Sequential En Vironmental) burner, which may be a perfect place for flameless combustion (see Fig. 27). Fig. 28 shows several stages in the working process of GT 26.

Stage 1: The compressed air is fed into the first EV burner, creating a homogeneous, lean fuel/air mixture and the flow in the burner forming a recirculation zone.

Stage 2: The mixture ignites into a single, low temperature flame ring. The recirculation zone stabilizes the flame within the chamber, avoiding contact with the combustor wall. Stage 3: The hot exhaust gas exits in this first EV chamber, moving through the high pressure turbine stage before entering the second SEV combustor.

Stage 4: Vortex generators in the SEV combustor enhance the SEV mixing process, while carrier air, injected with the fuel at the fuel lance, delays spontaneous ignition. Stage 5: Ignition occurs when the fuel reaches self-ignition temperature in the free space of the SEV combustion chamber. The hot gas then continues its path into the low pressure turbine.

In stage 1, there is no demand on the inlet temperature from compressor since flameless combustion is not need. Stage 3 and 4 will provide the hot gas and enough recirculation gas (depending on the temperature raise and the stoichiometric ratio in the EV burner), which are the requirements for the realization of flameless combustion. So, in stage 5, when the self-ignition temperature of fuel is reached, flameless combustion will occur.

The novel ITB with flameless combustion also would yield several challenges, such as shortening the residence time at a low pressure lost with adequate vaporization of liquid fuel, controlling the mixing and combustion process, enhancing the flow dynamic stability of a stratified flow with a large turning acceleration, and meeting the increased demands for cooling and aerodynamic-force loading on rotor and stator blades.

4.7. Summary

Some key issues that can be concluded from the analyses presented above are that the air injection velocity (high turbulence

Fig. 27. GT26 gas turbine and key working stages [63].

Fig. 28. Design map for the typical flameless combustor.

intensity), the entrainment and mixing with high temperature exhaust gas are considered to be the most important factors. While some Chinese researchers think that the air preheating is not an essential condition to attain flameless model [13,28,29]. The summary of some important information from these groups of researchers are shown in Table 1. From Table 1, it can be concluded that further progress is needed in the flameless combustion technology area. Several of possible aspects of further progress are as following:

• Majority of the reported work on flameless combustion is based on CFD simulation. However, such CFD models need experimental validation. Furthermore, there are still challenges in terms of simulating droplets evaporation, distribution and coupling detailed reaction mechanism for flameless combustion in commercial CFD codes. In-house codes and specialized UDF should be encouraged.

• The experimental techniques for measuring the species concentration and the temperature in the combustors are needed. All the information in the combustion field is very critical for diagnostic and understanding the mechanism of switching in the flame model and validating the simulation results.

• The effects of atomization and evaporation of the droplets on the combustion performance are neglected. Droplets' atomization

and evaporation are the most noticeable characteristics of the liquid fuels, and it is also the key factor for realizing the flame-less combustion model.

• In most of the experiments, the velocity of the air in the flame-less combustors is too slow, even considering the small size of the combustors. The similarity theory should be used to draw the conclusive working parameters to make the experimental results more beneficial to understand the mechanism of the flameless combustion.

With these viewpoints, in the following section, future research directions are suggested for improving the understanding of flameless combustion for the gas turbines with liquid fuels.

5. Knowledge gap and future research

Previous study has been summarized above. Though the physical characteristics of flameless phenomenon has been briefly described, there is no detailed analysis of the flameless issues and problems for gas turbine application because most of the research is based on the industrial furnace. The performance goals for gas turbine include working time and range, weight and size, cost and reliability, and emissions. So the feasible application of

Table 1

Summary of the flameless combustion for gas turbines.

Fuel type

Combustion Working conditions type

Atomiza tion

Biomass gas

Natural gas and H2

Partial premix

premixed

02 mass fraction 0.17-0.23 Oxidant pressure 270 kPa Oxidant temperature 723 K Oxidant rate 90 g/s (O 110 mm) Equivalence ratio 0-1

Inlet air pressure 20 bar Inlet air temperature 600-735 K Inlet air velocity 40-160 m/s

No need

No need

Natural gas and Partial

hydrogen-rich syngas premix

Syngas Premixed

Methane Partial

premix

Methane Non

premixed

Methane Non

premixed

Tangential airflow rate 0-12 g/s No need Axial airflow rate 0-12 g/s Main airflow rate 47.2 g/s Air temperature 298 K Equivalence ratio0.59-0.78

Inlet air pressure 20 bar No need

Inlet air temperature 700 K Inlet air velocity 62-75 m/s Equivalence ratio 0.2-1.4

The initial temperature 1000 K No need The initial pressure 1 bar Reference Velocity 12 m/s Combustor diameter 10 mm Equivalence ratio 0.57-0.8

Inlet air pressure 2.5 atm No need

Inlet air temperature 530 K Inlet air velocity 25-35 m/s Equivalence ratio 0.46

Air temperature 425 K No need

Airflow rate 0.01-0.025 m3/s Fuel flow rate 0.2-0.43 g/s Equivalence ratio 0.24-0.28

Butane and propane

Propane

Partial premix

Premix

02 mass fraction 6.7-7.3% No need

Oxidant pressure 200 kPa Oxidant temperature > 1300 K Oxidant mass rate70-102 g/s (<1> 160 mm)

air excess factor 0.95-1.5

Air temperature 523-823 K No need

Air mass flowratelO-35 g/s (<1> 100 mm)

Equivalence ratio 0.3-0.55

Research methods

Additional remark

Year/reference

Chemkin and experiments A multi-staged flameless combustor design scheme

(without further information) was proposed for micro-gas turbine at atmospheric pressure

Experiment with PLIF, temperature, pressure and emissions measurement

Experiments with PIV, pressure spectra, flame images, and emissions measurement

The FLOX" combustor was firstly operated at high pressure with low emissions for gas turbine. The jet exit velocity of the fuel/air mixtures had a strong influence on the mixing process within the combustor

The cavity air injection pattern and the cavity equivalence ratio were identified as the major parameters for flameless combustion controlling

Chemkin and Fluent (Realize 1<-e, LES and PI, EDC)

Experiments with PIV, OH chemiluminescence images, pressure pulsating, emissions measurement

The preliminary optimization of geometry causes different flows in perfect balance and a vortex filling the entire volume, which would establish the appropriate conditions for flameless combustion

Both frequency and amplitude of the pulsation are specific for each equivalence ratios, which have a strong impact on NOx emissions in MILD combustor for gas turbine applications

Experiments with PIV, pressure spectra, flame images, and emissions measurement

Chemkin (Chemical Reactor Modelling)

Experiments with LDV, gas temperature, and emissions measurement

House CFD code (k-e model, laminar flame let joint PDF) Experiments

In flameless model, the emissions were extreme low and without the thermo- acoustic problems. The air injection pattern and equivalence ratio were identified as major parameters for controlling and optimizing performance

The flameless combustor for gas turbine was experimented and exhibited stable operation over a relatively wide range of operating conditions. Its main design characteristic is the formation of a large recirculation vortex with distributed combustion and nearly uniform temperature values

Validated and supplied CFD code from UK. The modelling method would be based on a combined passive scalar and reaction-progress-variable approach

Experiments with PIV, ICCD for flame images, emissions measurement

The strong central recirculation of exhaust gases is crucial for achieving the flameless mode. High preheat temperature and high air flow rates helped in reducing the combustion instability. The present design can achieve low NOx and CO emissions with low pressure loss

2007/Ref. [23]

2008, 2011/Refs. [38,39]

2010/Ref. [53]

2010, 2012/Refs. [58,59]

2014/Ref. [40]

2007/Ref. [60]

2004, 2007, 2011/Refs. [41-43]

2006/Ref. [44]

2006/Ref. [15]

Table 1 (continued)

Fuel type

Combustion

Working conditions

Atomization

Research methods

Additional remark

Year/reference

Propane (liquid), n-butane (liquid) and Jet A Non premixed Oxidant temperature > 600 K Oxidant flow rate 355 g/s Equivalence ratio 0.25-0.75 Pressure atomizer with 0.8 Flow Number Experiment with SPIV, PLIF, temperature, and noise and emissions measurement Most fuels with the exception of n-butane showed very similar combustion characteristics, so the sensitivity of flameless combustion process was minimized. Flameless combustion only occurs at a minimum temperature 1073 K 2009/Ref. [20]

Liquid bio-diesel Non premixed Reference from Rover IS/60 gas turbine No mention Experiments with outlet gas analyzer No significant difference in exhaust emissions among four fuels. Soot volume fraction and emissions were investigated using two-color spectrometry and infrared absorption gas analysis 2008/Ref. [26]

Liquid bio-fuels Lean premixed O2 mass fraction 0.17-0.23 Oxidant temperature 905 K Oxidant flow rate 0.808 kg/s Fuel mass flow rate 0.0065 kg/s No mention Fluent (Realize k-e and EDC) The relation between the emissions and recirculation ratio was built. The emission can be decreased by selecting a different location of the pilot injector to allow the flameless model to be approached 2011/Ref. [25]

Liquid diesel Non premixed O2 mass fraction 0.23 Oxidant temperature 323 K Oxidant flow rate 0.5 m3/h Equivalence ratio 0.25-0.5 The SMD is between 30 and 60 im Fluent (Realize k-e) Experiments with outlet gas analyzer, thermocouples There was a critical injection momentum from the air blast atomizer for the flameless combustion mode 2012/Refs. [28,29]

Liquid kerosene Non premixed Reference velocity 12-22 m/s Oxidant temperature > 560 K Equivalence ratio 0.2-0.36 No mention Chemkin and Fluent (Realize k-e& DO & PDF). Trapped-vortex concept was used as the gas generator and flame stabilizer. The temperature peak was eliminated and the temperature profile was uniformity 2012, 2014/Refs. [34,35]

Liquid kerosene Non premixed Reactants dilution ratio > 2.71 Fuel mass flow rate 28.67 g/s at 9 bar injection pressure Equivalence ratio 0.6-1 A pressure swirl fuel injector provides SMD 17-23 im Fluent (RSM and P1 and PDF) Experiments with outlet gas analyzer, thermocouples A two stage combustor was designed with the tangential air injection and a chamfer to achieve better recirculation and also higher residence times. The outstanding performance of the burner with very low chemical and acoustic emissions at high heat release rates 2013, 2014/Refs. [3133]

Liquid kerosene and methane Non premixed Velocity of fuel 20 m/s Velocity of inlet air 40-100 m/s Inlet air pressure 30 atm Equivalence ratio 0.5-0.69 No mention Fluent (RANS and LES, EDM and EBU) A double cavity TVC geometry has been investigated. Although the vortex was stable, the outflow temperature is not uniform, and the characteristics of flameless are not obvious 2006/Ref. [57]

flameless combustion technology need to be theoretically summarize clearly, and the knowledge gap and future research should be focused appropriately.

5.1. Working conditions and atomization

For gas turbine application, the combustion inside a gas turbine combustor should be firstly emphasized, it has three main features: a high-Reynolds-number turbulent diffusion flame, a large recirculation zone generated by the bluff body and swirling primary air flow, and the great combustion intensity (rate of thermal energy released per unit volume) with liquid fuels. From the former flameless combustion research, a series of characteristics on the flameless mechanisms have been drawn. But the different working conditions between gas turbine and industrial furnace would also draw attention.

Firstly, the operational velocity of the injection for flameless combustion may be much higher than the reference velocity in an aero gas turbine combustor (about 20-30 m/s), which may cause a significant total pressure loss. The high velocity is preferred because the sufficient momentum of the injection and the entrain-ment of the gases have significant effects on the critical factor, the gas recirculation rate in the combustion chamber. Secondly, the O2 concentration in a normal gas turbine combustor inlet is higher than in an industrial burner. In order to reach the target value of the recirculation rate (KV), large amount of combusted gases need to be circulated inside the combustor to decrease the concentration of O2. This may result in lower volumetric heat densities and smaller temperature raise compared to conventional gas turbine com-bustors, and thus decrease the thrust-weight ratio or power-weight ratio of the engine. Thirdly, the high jet velocity and low O2 concentration would make the ignition and the flame stabilization difficult. In very fuel-lean conditions, flame blowing out becomes an issue for small combustor size (especially for aero engine combustor) because the large ratio of surface area and volume. For ground gas turbine, flame quenching is also a problem for gas fuels with low heating values.

For liquid fuels, the atomization and evaporation in turbulence would be primary factors for flameless combustion in engineering applications, which have been ignored in most of the current literatures. From a practical point of view, deformation and vaporization phenomena of droplets are very important for the spray flows that are used commonly in liquid-fuelled combustion systems. The knowledge from these relatively 'simple' phenomena is possible to help to understand the complex mechanisms of the two-phase spray combustion phenomenon encountered in practical liquid-fuelled flameless combustion systems. There are considerable experimental, theoretical and numerical data in the open literatures on the vaporization and burning of single droplets subjected to isolated and/or coupled conditions such as, gas ambient pressure, gas ambient temperature, liquid properties, and droplet spacing.

5.2. Simulation and measurement

The key parameter in the interaction between the combustion and the turbulence is Damkohler number, which is described by two characteristic time scales and can determine the flameless phenomena. The smaller Damkohler number, the faster the gas and the fuel mix, which means the combustion process would be delayed to approach the flameless conditions. Obviously, the rapid diffusion and mixing on the Kolmogorov scale is mainly depended on the turbulence effect produced by the airflow impingement. Hence, existing turbulence and combustion models for flameless combustion need to be considered in a combined manner in numerical studies to determine the value of Damkohler number.

For simulation work, the appropriate turbulence model and random combustion model should be chosen at first to simulate the turbulent combustion flowfield, then after post-processing calculation, the integral length, turbulence, Kolmogrov length and turbulent Reynolds number of the flowfield would be obtained. Secondly, the diffusion zone thickness and reaction zone thickness under the com-bustor pressure should be calculated with the CHEMKIN software. Finally, the Kolmogrov scale and the reaction zone thickness should be compared for choosing the appropriate combustion model.

When the Kolmogrov scale is greater than reaction zone thickness, EDM [64] and EDC [65] combustion model are more used than other models [25,50-52]. When the researchers use these models, some issues should be also noticed. For example, CHEMKIN Equil package could be used to calculate the dissociation of combustion product to avoid the flame temperature higher than real value with inaccurate specific heat and enthalpy data. If the EDM model is used, the model constant A with the standard value 4 should be with verification and validation. Eleven types of turbulent diffusion flames are simulated with CFD, and the simulation results revealed that for the predicted flame temperature, an optimal A value can be found to match experimental data for each flame [66]. This value depends on fuel, chemistry, and the turbulence strength of combustion flow field, and the A value varies between 0.77 and 25 for the eleven types of flames. A correlation of optimal A value with turbulent Reynolds number is presented in the reference.

Developing chemical-kinetic models that describe combustion of liquid kerosene are of practical importance for developing realistic models for simulation of flameless combustion for aero engine application. Due to the complex nature of kerosene combustion and the difficulty to reveal such reaction mechanism through experimental investigation, recent research is focused on the surrogate fuels [67,68], which have similar physical and chemical properties to the commercial kerosene. Their reference-state ther-modynamic properties are obtained by a combination of the properties of C10H22, C12H26, C8H16 and C8H10 species, and the general form of the molar specific heat at constant pressure inJ/(kmol K) is expressed with five coefficients. Such combination of the properties is described by NASA polynomial parameterization, and is also used by CHEMKIN, COSILAB and CANTERA software, which would be helpful to perform the simulation work.

The velocity and concentration of the chemical species can significantly impact the flow and heat transfer properties in the combustion zone and to the other various phenomena, including droplets deformation, atomization and evaporation. Understanding the precise velocity and the associated chemical and thermodynamic state of the species in the flowfield would help us to find the details and evaluate the performance of the flameless combustors.

To the authors' knowledge, additional experiments on turbulence in hot gases as well as in a wider range of turbulent integral length scale and Reynolds number are needed for assessing the effect of turbulence on droplets deformation and evaporation in forced convective turbulent flows. In all published studies, the suspended droplet technique has been utilized only at ambient room temperature and atmospheric pressure, which does not satisfy the need to develop new correlations for the deformation and evaporation of the droplets. Meanwhile, the new numerical methods such as Lattice Boltzmann Methods and Moving Particle Semi-implicit Method for simulation work should be improved to meet the high resolution requirement in the interface between the gas and the liquid droplets.

There have been many laser diagnosis techniques for measuring the flow field and species concentration in the combustors, such as Mie Scattering (MS), Rayleigh Scattering (RS), Spontaneous Vibra-tional Raman Scattering (VRS), Laser Induced Fluorescence (LIF) and Molecular Tagging Velocimetry (MTV). Combination of these methods would be beneficial for flameless combustion measure-

ment. For example, one might use RS and LIF to take simultaneous images of temperature profiles and NO densities, RS, VRS, and LIF to image a number of species and temperature simultaneously, or MTV and VRS to measure flow velocities, species and temperature simultaneously. Such combination of methods may be easy to realize because different methods use the same hardware arrangement. When a method needs to be changed, the necessary work is often only a few operations on a laser and a filter which change the excitation wavelength or the range of emission wavelengths. Such combined methods can give extensive information about the progress of flameless combustion phenomena.

5.3. Design methods and process

The geometry of the combustor plays a vital role in the flameless formation for inner re-circular model like Figs. 16-19, 21, and 26. A great advantage of the recirculation structure is to entrain and circulate the high temperature exhaust gas in a small combustion chamber with the high speed air jet. This entrainment and circulation increase the residence time of reactants and products, and thus complete a series of coupled processes which reflect the concept of flameless combustion, including endothermic evaporation/heat release, mixing and slow reaction. But it is also challenging to develop sufficient gas recirculation for flameless combustion without adding a mechanical system for creating recirculation. The difficulty remains in controlling the reaction rate in a way in which sufficient mixing is achieved before completion of the reaction.

Current gas turbine combustor configurations do not exhibit sufficient flow performance to attain the desired global performance for the whole engine. High resolution design and optimization methodologies can be employed to accelerate the closure of gas turbine combustor design and to increase/enhance the capture of complex flow field and species concentration effects on new flameless combustion concepts. The following technical challenges need to be addressed:

• Numerical algorithms are needed to efficiently resolve computational fluid dynamics for the flow field and the species concentration in the complex flow fields and thus to guide the design of new concept combustors.

• Rapid design methodologies will be needed to couple modelling of aerodynamics, thermodynamics, and chemical phenomenology in reactive gas flows of flameless combustor in the gas turbine operational environments.

Furthermore, it is also very important to establish a new design procedure for novel combustors, considering that the working conditions are different from the conventional gas turbine combustors. Referring to Refs. [41-43], Fig. 28 shows the design procedure for a typical flameless combustor with relevant research tools and reference pictures.

Briefly, there are four stages in the flameless combustor design procedure: I -Thermodynamic Analysis Stage, II - Concept Design Stage, III - Preliminary Design Stage, and IV - Application Stage. In Thermodynamic Analysis Stage, principal schemes of the internal flow inside of the flameless combustor are considered by CHEMKIN to calculate chemical equilibrium temperature and specie concentrations. If the primary results indicate that the proposed cycle would reduce the formation of NOx and meet the other demand of the engines, the next stage would be proceeded to. Stage II, the concept design stage is generated according to the results of stage I. It will be accomplished by commercial and in-house codes. Fine details of the flow field in the flameless com-bustors will be obtained through the turbulence, the two phase, the radiation and the flame models. Those models will provide sufficient information for optimizing the conceptual design after val-

idated via experiments. The velocity, diameters of the droplets, temperature, pressure and specie concentrations in the flameless combustor will be measured in the experiments. Stage III, the preliminary stage takes several optimization rounds. In stage IV, the final stage, the flameless combustor will be installed in the real engine to replace the conventional combustor, and the whole engine will be tested to get a satisfactory performance.

6. Conclusion remarks

In spite of decades of research devoted to flameless combustion, there are still many challenges in the flameless phenomena analyses and the combustor design. Therefore, many "unknowns" still exist in the field of flameless combustion that even our best experimental or numerical analysis cannot adequately predict. Current experimental approaches are not adequate for capturing the pressure, temperature, mass flow rate and high-enthalpy states in gas turbines, but leave us to extrapolate our test data and to approximate. Uncertainty in the ability to model chemical reactions in computational simulations or to adequately predict flow features leaves additional work to be done before CFD predictions will be fully trustworthy.

Still, many speculative considerations have been presented as follows in order to make the whole framework more consistent and rich with potential for practical gas turbine applications.

• Preliminary understanding about the mechanism of flameless is that O2 concentration in the combustion air decreases quickly, leading to an increase of the characteristic reaction time that becomes comparable with the characteristic mixing time, which, on the contrary, is lowered by the high turbulence generated by the high-velocity reactants jets. Therefore, the reaction zone is uniformly distributed throughout the combustor volume with lower peak flame temperature than that of conventional mode.

• The key technique for a gas turbine to realize the flameless combustion is to organize the flow field in the combustor to form the high temperature gas recirculation and the dilution of fresh reactants. There are three ways: (1) external recirculation where the gas flows outside the gaseous fuel jet, like FLOX Combustor; (2) internal recirculation where the gas flows inside the fuel jet, like EU burner from Cincinnati; (3) cyclic periodical gas flows and mixing with fuels in the center of the flow circle, like flameless combustion based on the trapped vortex.

• The way the fuel and air are injected into the furnace chamber is of primary importance for the distributions of furnace temperature, oxygen, and fuel that thus affects NOx emission and combustion efficiency. A group of parameters including fuel property, droplet distribution, evaporation, mixture formation and subsequent combustion with preheating and dilution of reactants need to be discussed and developed.

• The difficulty in designing a flameless prototype arises from the fact that there are no standard design tools. In spite of the innovative concept, its design and implementation involves the traditional issues of a gas turbine burner such as how to design a component with a suitable geometry in the changing operating conditions, how to ensure the absence of thermo acoustic oscillations and the instability of the burner.

• It is not necessary to diffuse the compressed air to very low inlet velocities because high air inlet velocities can be used to enhance the internal recirculation. Thermal radiation would become a substantial part of the total heat transfer between the recirculation gases and the secondary cooling air due to the distributed flame and uniform temperature. Alternative heat transfer techniques can be used to enhance the heat transfer between the cooling air and the combustor wall.

• Trapped vortex combustor is able for its intrinsic nature of improving mixing of hot combusted gases and fresh mixture that represents a prerequisite for a diluted combustion and at the most a flameless combustion regime. The trapped vortex technology offers several advantages as a gas turbines burner: burning low calorific value fuels, extremely low NOx emissions, and extension of the flammability limits.

A typical flameless combustor appears to have potential to substitute the conventional gas turbine combustor. Its distributed exothermic reaction and consequently, uniform temperature avoids the need for the very high adiabatic flame temperature values associated with high-NOx formation. Therefore, flameless combustion poses itself undoubtedly as a technology combining high efficiencies, and low pollutant emissions. All these aspects make flameless combustion worthy of further investigations and attention.

Acknowledgment

This review paper is written during the time when the author FX has visited School of Chemical Engineering and Analytical Science, The Univeristy of Manchester and School of Engineering, Cranfield University, UK. The financial support from Cranfield University and The Univeristy of Manchester is kindly acknowledged. FX thanks the China Scholarship Council (CSC) for supporting him as a CSC visiting scholar at The University of Manchester (201308350040). In addition, FX would also like to thank Professor Sai Gu and his research team for their inspirational discussions on the presented topic and other research topics throughout the visiting period.

The authors are grateful to National Natural Science Foundation of China (Grant No. 51406171), Natural Science Foundation of Fujian Province (Grant No. 2015J05111), and Fundamental Research Funds for the Central Universities of China (Grant No. 20720150180) for providing the funding support for the research. The authors also would like to acknowledge the financial support by the UK Engineering and Physical Sciences Research Council (EPSRC) project grant: EP/K036548.

References

[1] Chefurka P. World energy and population trends to 2100. Website: http:// www.paulchefurka.ca/WEAP/WEAP.html.

[2] Gupta AK. Flame characteristics and challenges with high temperature air combustion. In: Proceedings of the international joint power generation conference, Miami Beach, Florida; 2000.

[3] Beer JM. Combustion technology development in power generation in response to environmental challenges. Progr Energy Combust Sci 2000;26:301-27.

[4] Lieuwen CT, Yang V. Gas turbine emissions. Cambridge University Press; 2013.

[5] Schweitzer, Jeffrey K. Propulsion technology readiness for next generation transport systems. In: AIAA/ICAS international air and space symposium and exposition: the next 100 years. AIAA; 2003. p. 2787.

[6] Lee IB, Woo IS, Lee MC. Effects of nitrogen dilution on the NOx and CO emission of H2/CO/CH4 syngases in a partially-premixed gas turbine model combustor. Int J Hydrogen Energy 2016:15841-51.

[7] Benini E, Pandolfo S, Zoppellari S. Reduction of NO emissions in a turbojet combustor by direct water/steam injection: numerical and experimental assessment. Appl Therm Eng 2009;29(17):3506-10.

[8] Taimoor AA, Muhammad A, Saleem W. Humidified exhaust recirculation for efficient combined cycle gas turbines. Energy 2016;106:356-66.

[9] Weinberg FJ. Combustion temperature: the future. Nature 1971;233:239-41.

[10] Wunning JA, Wunning JG. Flameless oxidation to reduce thermal no-formation. Prog Energy Combust Sci 1997;23(1):81-93.

[11] Cavaliere A, Joannon M. Mild Combustion. Prog Energy Combust Sci 2004;30 (4):329-66.

[12] Gupta AK. Developments in high temperature air combustion (flameless oxidation) and fuel reforming. In: 44th AIAA aerospace sciences meeting and exhibit. Reno (NV): AIAA; 2006. p. 1498.

[13] Li P, Mi J, Dally BB, Wang F, Wang L, Liu Z, et al. Progress and recent trend in MILD combustion. Sci China Technol Sci 2011;54(2):255-69.

[14] Noor MM, Wandel AP, Yusaf T. MILD combustion: the future for lean and clean combustion technology. Int Rev of Mech Eng 2014;8(1):251-7.

[15] Li G, Gutmark EJ, Stankovic D, Overman N, Cornwell M, Fuchs L, et al. Experimental study of flameless combustion in gas turbine combustors. In:

44th AIAA aerospace sciences meeting and exhibit. Reno (NV): AIAA; 2006. p. 546

[16] Overman N, Cornwell M, Gutmark EJ. Application of flameless combustion in gas turbine engines. In: 42nd AIaA/ASME/SAE/ASEE joint propulsion conference & exhibit. Sacramento (CA): AIAA; 2006. p. 4920.

[17] Overman NR. Flameless combustion and application for gas turbine engines in the aerospace industry. Master thesis. Aerospace Engineering, University of Cincinnati; 2009.

[18] Gutmark EJ, Paschereit CO, Guyot D, Lacarelle A, Moeck J, Schimek S, et al. Combustion noise in a flameless trapped-vortex reheat burner (FTVRB). In: 13th AIAA/CEAS aeroacoustics conference. AIAA; 2007. p. 3697.

[19] Guillou E. Flame characteristics and application of flameless combustion. Master thesis. Aerospace Engineering, University of Cincinnati; 2009.

[20] Guillou E, Cornwell M, Gutmark E. Application of "Flameless" combustion for gas turbine engines. In: 47th AIAA aerospace sciences meeting including the new horizons forum and aerospace exposition. Orlando (FL): AIAA; 2009. p. 225.

[21] Abuelnuor AA, Wahid MA, Mohammed MA, Saat A. Flameless combustion role in the mitigation of NOX emission: a review. Int J Energy Res 2004;38 (7):827-46.

[22] Tsuji H, Gupta A, Hasegawa T, Katsuki M, Kishimoto K, Morita M. High temperature air combustion: from energy conservation to pollution reduction. FL: CRC Press; 2003.

[23] Hiramatsu M, Nakashima Y, Adachi S, Yamasaki Y, Kaneko S. Combustion characteristics of small size gas turbine combustor fueled by biomass gas employing flameless combustion. In: Proceedings of GT2007, ASME Turbo Expo 2007. p. 463-9.

[24] Cameretti Mc, Piazzesi R, Reale F, Tuccillo R CFD analysis of the flameless combustion in a micro-gas turbine. In: Proceedings of ASME turbo expo; 2009. p. GT2009-59750.

[25] Cameretti MC, Tuccillo R, Reale F, Piazzesi R. Liquid bio-fuels in an EGR equipped micro gas turbine. In: Proceedings of ASME turbo expo; 2011. p. GT2011-46341.

[26] Ellis W, Lear W, Singh B, Srinivasan A, Crittenden J, Sherif SA. Flameless combustion of biofuels in a semi-closed cycle gas turbine. In: 46th AIAA aerospace sciences meeting and exhibit. Reno (NV): AIAA; 2008. p. 1140.

[27] Torresi M, Camporeale SM, Fortunato B, Ranaldo S, Mincuzzi M, Saponaro A. Diluted combustion in a aerodynamically staged swirled burner fueled by diesel oil. Processes and technologies for a sustainable energy, Ischia; 2010.

[28] Cui YJ, Lin QZ. Realization of flameless combustion of liquid fuel. Adv Mater Res 2012;512-515:2088-92.

[29] Cui Y, Ye T, Lin Q. Effect of fuel air equivalence ratio on the flameless combustion of liquid hydrocarbon. J Univ Sci Technol China 2012;42 (4):325-34.

[30] Reddy VM, Kumar S. Development of high intensity low emission combustor for achieving flameless combustion of liquid fuels. Propuls Power Res 2013;2 (2):139-47.

[31] Reddy VM, Sawant D, Trivedi D, Kumar S. Studies on a liquid fuel based two stage flameless combustor. Proc Combust Inst 2013;34(2):3319-26.

[32] Reddy VM, Katoch A, Roberts WL, Kumar S. Experimental and numerical analysis for high intensity swirl based ultra-low emission flameless combustor operating with liquid fuels. Proc Combust Inst 2015;35(3):3581-9.

[33] Zang P, Zhang K, Cui Y, Nie C. Numerical simulation of flameless combustor based on trapped-vortex. J Eng Thermophys 2012;33(9):1615-8 [in Chinese].

[34] Wang K, Zang P, Xing S, Fang A, Cui Y, Nie C. Experimental study of flameless combustor based on trapped-vortex. Gas Turbine Technol 2014;27(3):14-9 [in Chinese].

[35] Wang L, Qi D, Sui X, Xie X. Analysis of Re influence on MILD combustion of gas turbine. Energy Power Eng 2013;05(04):92-6.

[36] Derudi M, Rota R. Experimental study of the mild combustion of liquid hydrocarbons. Proc Combust Inst 2011;33(2):3325-32.

[37] Flamme M. New combustion systems for gas turbines (NGT). Appl Therm Eng 2004;24(11-12):1551-9.

[38] Lückerath R, Meier W, Aigner M. FLOX® combustion at high pressure with different fuel compositions. J Eng Gas Turbines Power 2008;130(1). 011505-1-7.

[39] Sadanandan R, Lückerath R, Meier W, Wahl C. Flame characteristics and emissions in flameless combustion under gas turbine relevant conditions. J Propul Power 2011;27(5):970-80.

[40] Varea E, Kruse S, Pitsch H, Albin T, Abel D. Actuation studies for active control of mild combustion for gas turbine application. In: Proceedings of ASME turbo expo 2014: turbine technical conference and exposition, Düsseldorf, Germany; 2014.

[41] Levy Y, Sherbaum V, Arfi P. Basic thermodynamics of FLOXCOM, the low-NOx gas turbines adiabatic combustor. Appl Therm Eng 2004;24(11-12):1593-605.

[42] Levy Y, Rao GA, Sherbaum V. Chemical kinetic and thermodynamics of flameless combustion methodology for gas turbine combustors. In: 43rd AIAA/ ASME/SAE/ASEE joint propulsion conference & exhibit 2007. AIAA; 2007. p. 5629.

[43] Melo MJ, Sousa JMM, Costa M, Levy Y. Flow and combustion characteristics of a low-NOx combustor model for gas turbines. J Propul Power 2011;27 (6):1212-7.

[44] Awosope IO, Kandamby NH, Lockwood FC. Flameless oxidation modelling: on application to gas turbine combustors. J Energy Inst 2006;79(2):75-83.

[45] Wang YD, Huang Y, Mcllveen-Wright D, McmullanJ, Hewitt N, Eames P, et al. A techno-economic analysis of the application of continuous staged-combustion and flameless oxidation to the combustor design in gas turbines. Fuel Process Technol 2006;87(8):727-36.

[46] Arghode VK, Gupta AK. Effect of flow field for colorless distributed combustion (CDC) for gas turbine combustion. Appl Energy 2010;87(5):1631-40.

[47] Arghode VK, Gupta AK. Investigation of forward flow distributed combustion for gas turbine application. Appl Energy 2011;88(1):29-40.

[48] Arghode VK, Gupta AK. Development of high intensity CDC combustor for gas turbine engines. Appl Energy 2011;88(3):963-73.

[49] Arghode VK, Gupta AK, Bryden KM. High intensity colorless distributed combustion for ultra-low emissions and enhanced performance. Appl Energy 2012;92(2):822-30.

[50] Khalil AEE, Gupta AK. Swirling flowfield for colorless distributed combustion. Appl Energy 2014;113(6):208-18.

[51] Khalil AEE, Gupta AK. Impact of pressure on high intensity colorless distributed combustion. Fuel 2015;143(143):334-42.

[52] Khalil AEE, Brooks JM, Gupta AK. Impact of confinement on flowfield of swirl flow burners. Fuel 2016;184:1-9.

[53] Cui Y, Lü X, Xu G, Nie CQ. Dynamic analysis of a flameless combustion model combustor. Sci China Technol Sci 2010;53(8):2291-8.

[54] Lü X. Investigation of flameless combustion technology for hydrogen-rich fuels in micro gas turbine. PhD thesis. Engineering Thermo Physie, Chinese Academy of Sciences; 2010 [in Chinese].

[55] Mao YH, Xu G, Fang A, Hou L. Advances in the study of gas turbine flameless combustion technology. J Eng Therm Energy Power 2011;26(5):7 [in Chinese].

[56] Xing F, Wang P, Zhang S, Zou JF, Zheng Y, Zhang R, et al. Experiment and simulation study on lean blow-out of trapped vortex combustor with various aspect ratios. Aerosp Sci Technol 2012;18(1):48-55.

[57] Losurdo M, Bruno C. Numerical simulations of trapped vortex combustors feasibility study of TVC integration in traditional GT combustion chambers. In: 42nd AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit 2006. AIAA 2006-5140, Sacramento, California.

[58] Di Nardo A, Galletti C, Giammartini S, et al. Design and analysis of a gas turbine trapped vortex combustion. Proc Technol Sustainable Energy 2010:1-6.

[59] Di Nardo A, Calchetti G, Mongiello C. Non-premixed syngas MILD combustion: the trapped-vortex approach. In: Proceedings of the seventh international symposium on turbulence, heat and mass transfer, Palermo, Italy, 2012.

[60] Guyot D, Bothien MR, Moeck JP, Schimek S, Lacarelle A, Paschereit O, et al. Pollutant and noise emissions in a flameless trapped-vortex reheat burner (FTVRB). In: 43rd AIAA/ASME/SAE/ASEE joint propulsion conference & exhibit, 2007. AIAA; 2007. p. 5630.

[61] Sirignano WA, Dunn-Rankin D, Liu F, Colcord B, Puranam S. Turbine burners: performance improvement and challenge of flame holding. AIAA J 2012;50 (8):1645-69.

[62] Fang XY, Xing F, Xu LL, Xing P, Wang PY. Performance analysis of three kinds of aero-engine with inter-stage turbine burner. J Propul Technol 2013;34 (11):1513-9.

[63] Ochrymiuk T, Badur J. Flameless oxidation at the Gt26 gas turbine numerical study via full chemistry. Task Q2001;5(2):239-46.

[64] Magnussen B, Hjertager B. On mathematical modeling of turbulent combustion with special emphasis on soot formation and combustion. Symp Combust 1977;16(1):719-29.

[65] Magnussen B. On the structure of turbulence and a generalized eddy dissipation concept for chemical reaction in turbulent flow. In: 19th American Institute of Aeronautics and Astronautics Aerospace Science meeting, St Louis, Missouri, USA; 1981.

[66] Wang P. The model constant A of the eddy dissipation model. Progr Comput Fluid Dyn IntJ 2016;16(2).

[67] Lindstedt RP, Maurice LQ. Detailed chemical-kinetic model for aviation fuels. J Propul Power 2000;16(2):187-95.

[68] Zheng D, Yu WM, Zhong BJ. RP-3 aviation kerosene surrogate fuel and the chemical reaction kinetic model. Acta Physico-Chimica Sinica 2015;31 (4):636-42.