- Email: [email protected]
Autoignition and combustion behavior of emulsion droplet under elevated temperature and pressure conditions
Energy 163 (2018) 800e810
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Autoignition and combustion behavior of emulsion droplet under elevated temperature and pressure conditions Jonghan Won a, Seung Wook Baek a, **, Hyemin Kim b, * a
Department of Aerospace Engineering, Korea Advanced Institute of Science and Technology, 373-1 Guseong-dong, Yuseong-gu, Daejeon, South Korea Department of Aeronautical·Mechanical Design Engineering, Korea National University of Transportation, Daehak-ro 50, Chungju City, Chungbuk, South Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 January 2018 Received in revised form 23 August 2018 Accepted 24 August 2018 Available online 25 August 2018
In this study, experiments were conducted to investigate the combustion characteristics of an water-inoil W/O emulsion droplet under elevated temperature and pressure conditions. The base fuel used was ndecane, and total volume ratios of 10, 20, and 30% of distilled water were mixed for producing the emulsion fuel. Span 80 with a volume ratio of 2% was added as a surfactant, and the emulsion fuel was homogeneously mixed via ultrasonication. The combustion process of an emulsion droplet was divided into five stages: droplet heating, classical combustion, puffing, secondary classical combustion, and surfactant combustion. The ignition delay decreased with elevated ambient temperatures, whereas an increase in the ambient pressure and water volume ratio resulted in longer ignition delays. The droplets did not ignite in 500 C or 600 C conditions at 1 bar because of the significant Stefan flow of fuel vapor. After droplet ignition, the droplet combustion process, including classical combustion, puffing, and surfactant combustion, followed. The average burning rate increased with ambient pressure, but it was insensitive to ambient temperatures and water volume ratios. After flame extinction, a secondary flame reappeared because of the combustion of surfactant and residues. © 2018 Elsevier Ltd. All rights reserved.
Keywords: W/O emulsion fuel Droplet Temperature Pressure Water volume ratio Puffing
1. Introduction Numerous researchers dedicated to improving energy efficiencies and decreasing pollutant emissions of conventional fossil fuel. Researches were conducted in total direction because it is urgent matter for sustainable nature and humanity [1e4]. Nevertheless, there were unavoidable limitations to the amount of improvement when conventional fossil fuels were used. Especially, it may be reached the limit in reducing several pollutants matter during combustion such as particulate matter (PM), nitrogen oxide (NOx), and carbon monoxide (CO), which are harmful for environment and human health [5,6]. In the reasons above, emulsion fuel has received particular attention as an alternative fuel for reducing pollutant emissions and improving combustion efficiency. When the emulsion fuel is combusted, superheated water vapor is generated inside the fuel droplet and it explodes at the surface of the droplet; this is known
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (S.W. Baek), [email protected] (H. Kim). https://doi.org/10.1016/j.energy.2018.08.185 0360-5442/© 2018 Elsevier Ltd. All rights reserved.
as “micro-explosion” or “puffing.” Several researchers have reported that the puffing process induces fuel atomization, and therefore, high combustion efficiency and low pollutant emission can be achieved without any additional apparatus or cost [7,8]. For the above reasons, several researchers have conducted studies on emulsion fuel combustion. Ithnin et al. [9], experimentally studied the combustion of water-in-diesel emulsions under various engine load conditions. They reported that emulsion fuel containing 20% water had a maximum cylinder pressure and pressure increase rate comparable to other cases. In addition, emulsion fuel exhibited lower NOx and PM emissions, whereas CO and carbon dioxide emissions increased under low- and high-load conditions compared to the pure diesel case. Ochoterena et al. [10], conducted a study on the spray development and combustion of water-in-diesel emulsions and micro-emulsion fuels. Spray development, droplet break-up, vapor penetration, and combustion processes were observed optically. They reported that an emulsion fuel droplet penetrated further than a conventional fuel droplet, and atomization of emulsion fuel decreased the soot concentration and flame temperature compared to regular diesel fuel. Deng and Zhou [11] experimentally examined
J. Won et al. / Energy 163 (2018) 800e810
the combustion of a coal tar/water emulsion. In their study, a longer ignition delay of the emulsion droplet was observed, but the peak temperature of the droplet was achieved considerably earlier than ordinary coal tar. This indicated the potential of soot reduction and the extent of burnout of cenospheres when using emulsion fuel. Gong and Fu [12] observed ignition delay times for oil-in-water emulsion droplets with various fuel compositions. They reported that the ignition delay time can be decreased by adding more volatile component fuel to the emulsion, and it was notably decreased when the proportion of the volatile component was 10e20%. The effect of this volatile component was not evident at higher concentrations. The distinctive combustion characteristics of emulsion fuels led previous researchers to conduct a number of experimental and theoretical studies, as discussed above. A study of the combustion of a single liquid fuel droplet under high temperature and pressure conditions is the primary approach for investigating the basic phenomena of spray combustion [13,14]. Although several studies dealt with emulsion fuel droplet, no studies have been conducted regarding a single emulsion droplet under high temperature and pressure conditions. Therefore, there is lack of studies and data regarding emulsion fuel droplet combustion under elevated pressure conditions [15,16]. Considering that numerous combustors are operated under high-temperature and high-pressure conditions, observation of the autoignition and combustion behaviors of emulsion fuel under these conditions are essential. This study could provide fundamental knowledge on emulsion fuel during the combustion process. This study focuses on the combustion characteristics of a water/ n-decane emulsion fuel droplet under elevated temperatures and pressures. The water volume ratios in the emulsion fuels were set to 10%, 20%, and 30%, and 2% vol. of Span 80 was added in all cases for obtaining a stable suspension. The combustion characteristics of an emulsion fuel droplet under various temperature and pressure conditions were observed, including ignition delays and burning rates. The results obtained under each experimental condition were compared. 2. Experimental setup The experimental setup comprised a high-pressure combustion chamber, an optical observation system, and measuring sensors. Various ambient pressure and temperature conditions could be investigated with this experimental setup. 2.1. Experimental setup Fig. 1 is a schematic diagram of the experimental apparatus that has been verified and used in previous studies [17,18]. The inner diameter and height of the cylindrical stainless-steel combustion chamber employed in this study were 150 mm and 800 mm, respectively. A single emulsion droplet of 900 ± 100 mm was suspended at the tip of a K-type thermocouple wire (Omega Engineering, Inc.). The shield of the thermocouple was removed, and the two wires in the thermocouple were welded for droplet installation. The diameter of the inner wire was 50 mm and the welded bead size was 100 mm. The droplet temperature could be measured while the droplet was suspended at the tip of the thermocouple during the experiment. This droplet installation method was widely used for several researchers due to its simplicity and accuracy of droplet temperature measurement [19]. Refer to the thermocouple manufacturer (Omega Inc.), the standard error limit of thermocouple was grater of 0.4% The target ambient temperature could be achieved by heating the electric furnace within an error range of ±5 C measured by the
801
K-type thermocouple inside the furnace. The side of the rectangular parallelepiped shape of the electric furnace was 100 mm and its height was 185 mm. The inner part of the electric furnace was encased by a ceramic board for heat insulation and a small 20-mm hole was drilled in the bottom for the thermocouple insertion where a droplet was suspended. When the temperature inside the electric furnace reached the experimental temperature, it was moved to the experimental position, and the experiment was started, as shown in Fig. 1. During the experiment, droplet images were recorded by a high-speed camera at 200 frames per second. A history of the droplet diameter was acquired using an image processing program that was validated in previous studies [20,21]. In the post-processing of the images, errors smaller than 3.12% were found; therefore, the experimental errors were in good agreement with those obtained in previous studies. 2.2. Emulsion fuel Distilled water and n-decane were used for producing water-inoil (W/O) emulsion fuel. The thermophysical properties of water and n-decane are presented in Table 1. Volume ratios of water of 10%, 20%, and 30% were mixed in the n-decane, and 2% vol. of Span 80 was added as a surfactant. As Span 80 has lipophilic characteristics (HLB value: 4.3), it is a representative surfactant that can be used in W/O emulsion. The emulsification process was performed using an ultrasonicator (Sonics & Materials, Inc.). It was operated for 10 min in pulse mode (5 s ON & 5 s OFF) to prevent a rapid increase in the emulsion temperature. The experiment was conducted within an hour of the preparation of the emulsion fuel to prevent errors due to phase separation of the emulsion fuel. The boiling temperature of water and n-decane across the wide pressure range could be calculated as follows:
lnðpkPa Þ ¼ A*lnðTboil Þ þ
B 2 þ C þ D*Tboil Tboil
(1)
The coefficients for water are A ¼ 7.34297, B ¼ 7276.39, C ¼ 67.0245, and D ¼ 4.16191 106, and for n-decane A ¼ 7.76881, B ¼ 8163.33, C ¼ 69.7646, and D ¼ 2.62033 106 [22]. The boiling temperatures at different experimental pressures are presented in Table 2. 2.3. Experimental conditions The experiments were conducted under various temperature (500, 600, and 700 C) and pressure (1, 5, 10, and 15 bar) conditions, with an initial droplet diameter set to 900 ± 100 mm. The experiment was conducted under each temperature and pressure condition with different water/n-decane fuel ratios. Each experiment was repeated a minimum of five times under identical conditions for consistency of the results. The chamber was purged with dry air after each experiment to remove the combustion residue. The ignition delay was the time between the start of the experiment and the droplet ignition. Droplet ignition was defined when a yellow flame first appeared in the image of the droplet, and the droplet burning time was calculated by counting the numbers of frames from droplet ignition to flame extinction. The burning rate constant (Kb) is widely used for examining the combustion phenomena of a single droplet. Burning rate is defined as the temporal change in the squared droplet diameter (d2), and is calculated as follows [23]:
Kb ¼
d 2 d dt
(2)
Unlike droplet evaporation, the observation of a droplet during
802
J. Won et al. / Energy 163 (2018) 800e810
Fig. 1. Experimental setup.
Table 1 Thermophysical properties of n-decane and water.
Specific heat capacity (kJ/kg K) Viscosity (mPa s at 20 C) Density (g/mL) Latent heat (kJ/kg)
water
n-decane
4.19 890 0.997 2257
2.21 920 0.73 263
Table 2 Boling temperature of water and n-decane in various pressure conditions.
1 bar 5 bar 10 bar 15 bar
Water ( C)
n-decane ( C)
100 152 180 198
174 248 291 320
combustion is difficult because of the flame surrounding it. Therefore, it is more beneficial to introduce the average droplet burning rate (Kb,avg), that is the mean change of the square droplet diameter from droplet ignition (dig) to flame extinction (dext). The time between droplet ignition and flame extinction is known as the droplet burning time (tburning). The average droplet burning rate can be calculated as follows [24]:
Kb;ave ¼
d2ig d2ext tburning
(3)
Before proceed the research, validation of experimental setup by comparing the results with previous research was carried out. Fig. 2 shows a comparison of the ignition delay for various n-decane and water volume ratios between a reference study by Jeong et al. [25] and the present study. Although the experimental temperatures differed marginally, the same trends with similar values are observed.
3. Results and discussion 3.1. Typical emulsion droplet combustion behavior Although the detailed combustion phenomena were significantly different for each experimental condition, the typical emulsion droplet combustion characteristics could be observed throughout the experiment. Fig. 3-(a) shows the droplet during the experiment (water volume ratio: 20%, atmospheric temperature: 600 C, and atmospheric pressure: 5 bar), and Fig. 3-(b) shows the droplet diameter and temperature history. As reported in previous studies [26], the combustion process of an emulsion fuel droplet
J. Won et al. / Energy 163 (2018) 800e810
803
Pressure: 1 bar Jeong et al. [25] at 647 Present study at 700
2500
Ignition delay (ms)
2000
1500
1000
500
0 D90+W10
D80+W20
D70+W30
Decane+Water volume ratio Fig. 2. Comparison of ignition delays for different volume ratios between this study and Jeong et al. [25] at 1 bar.
can be divided into several stages, including droplet heating, ignition, classical droplet combustion, puffing, secondary classical combustion, and surfactant combustion. The droplet heating period was initiated immediately after the electric furnace decreased to the experimental position. The primary feature of this period was an increase in the droplet temperature because of the ambient temperature. The droplet temperature gradually increased and approached the boiling temperature of water (152 C) at 5 bar. The droplet diameter increased marginally initially because of thermal expansion, and then started to decrease, as shown in the first period of Fig. 3-(b). Droplet puffing was not observed during this period as the droplet did not heat sufficiently for water vapor generation. The time between the start of experiment and the ignition is the ignition delay. Under the current conditions, at time 0.69 s, the droplet was autoignited in the vicinity of the droplet surface. This is one of the primary characteristics of droplet ignition under high pressure. Under the high-pressure condition, the Stefan flow of fuel vapor from the droplet surface is slow because of an increase in the boiling temperature and a decrease in the diffusion coefficient [27,28]. Under these conditions, ignition occurred close to the droplet surface. As the flame developed around the droplet, the droplet temperature increased rapidly because of the significant heat feedback from the flame. After ignition, droplet combustion without puffing (classical droplet combustion) was observed for a short period of time. This was because the droplet temperature did not reach the water boiling temperature for the puffing process to initiate. The droplet temperature increased until it reached approximately 170 C. The fuel vapor diffusion zone and flame zone are clearly shown in Fig. 3-(a). The structure of the flame close to the droplet changed under different combustion conditions, and is discussed in the next section. Under the 5-bar condition, the puffing process was initiated soon after the classical combustion period. The primary characteristic of this period is distortion of the flame because of puffing. When the droplet temperature exceeded the water boiling temperature, superheated water droplets in the droplet converted to water vapor and exploded from the droplet surface. As can be seen in Fig. 3-(b)-puffing, the flame around the droplet distorted as the
puffing process atomized the fuel droplet. This atomization process can increase the combustion efficiency [29]. The intensity of puffing was dependent on the experimental conditions. During the puffing process, the rate of increase of the droplet temperature was suppressed. In section C in Fig. 3-(b), the droplet temperature was maintained at approximately 180 C during the puffing period despite heat feedback from the flame. In this period, heat addition from the flame and heat loss from the water latent heat was balanced; therefore, the droplet temperature did not change significantly. Typically, an increase in the water volume ratio hindered a droplet temperature increase via a vigorous puffing process [26]. In this study, secondary classical droplet combustion appeared after the puffing period as the water content in the droplet decreased during puffing. A significant increase in the droplet temperature was observed in the late stages of droplet combustion. In the bulk of the experimental cases, the droplet could not be observed because of the luminous flame zone covering the whole droplet. However, in a number of cases, specifically under lowpressure conditions, the diffusion of fuel vapor was sufficient to observe the droplet during the combustion process. The changes in droplet diameter during the combustion process are shown in Fig. 3-(b). This result shows linear regression during droplet combustion. In this result, Kb was 1.65 during the droplet combustion, whereas the calculated Kb,avg was 1.69. This indicated that Kb,avg also represented the combustion rate of the droplet effectively under the present conditions. The flame extinguished approximately 350 ms after droplet ignition. After flame extinction, a secondary flame appeared as shown in Fig. 3-(b)-Surfactant combustion. The surfactant combustion burned low-volatile surfactant and pyrolysis of residues [30]. During this period, the droplet was too small to be optically observed. The temperature during this period increased rapidly and the surfactant combustion lasted less than 40 ms. In the next sections, the effects of ambient temperature, pressure, and water volume ratio on the ignition and combustion of the emulsion fuel droplet are discussed.
804
J. Won et al. / Energy 163 (2018) 800e810
Fig. 3. a) Optical images of droplet at 5 bar, 600 C and water ratio 20% vol., b) History of droplet diameter squared and temperature with stage classification: A (Droplet heating), B (Classical combustion), C (Puffing), D (Secondary classical combustion), and E (Surfactant combustion).
3.2. Effect of ambient temperature The influence of ambient temperature on droplet behavior is discussed in this section. In Fig. 4, the droplet diameter and temperature history at
15 bar at a water volume ratio of 10% is shown. The droplet diameter and time were normalized for the comparison. As mentioned above, the droplet diameter change could not be observed during combustion, as a yellow luminous flame obscured the droplet. In all cases, the droplet diameter increased in the early stages of
Normalized 2 2 diameter (d /d0)
J. Won et al. / Energy 163 (2018) 800e810
1.2
1.0
0.8 0.0
0.5
1.0
1.5
2.0
Ignition
o
Droplet temperature ( C)
300
200
Ambient temperature o 500 C o 600 C o 700 C
100
0 0.0
0.5
1.0
1.5
2.0
2
Normalized time (s/d0)
Igntion delay (ms)
Fig. 4. Time variation of normalized droplet diameter and temperature for different ambient temperature conditions (Ambient pressure: 15 bar, water volume ratio: 10%).
the droplet heating period because of thermal expansion. During this period, the droplet temperature increased more rapidly as the ambient temperature increased. For example, the droplet heating time was approximately 35% faster when the ambient temperature was 700 C compared to 500 C. This was because of the increase in heat feedback under the higher ambient temperature conditions. Referring to previous studies [21], a higher droplet temperature resulted in a higher fuel vapor concentration around the droplet surface, and the fuel/air mixture could be generated relatively quicker for droplet ignition. Therefore, a rise in ambient temperature expedited droplet ignition and eventually reduced the ignition delay time. For all cases, the droplet ignited when the droplet diameter started to decrease. The droplet temperature increased sharply after ignition because of heat feedback from the flame. As the droplet temperature reached approximately 150e170 C, which is marginally smaller than the water boiling temperature, the droplet ignited. Therefore, puffing rarely occurred before droplet ignition. The characteristics of puffing at different ambient pressures are discussed in the next section. As the ambient temperature was smaller, the droplet temperature at ignition was slightly greater. Higher activation energy for ignition was required for a smaller fuel vapor concentration; therefore, the droplet temperature at ignition increased as the ambient temperature
2500
2500
2000
2000
1500
1500
1000
1000
500
500
0
0
5
10
15
805
0
0
5
Pressure (bar)
10
(B)
(A) 2500
2000
Ambient temperature o 500 C o 600 C o 700 C
1500
1000
Water volume ratio A: 10% B: 20% C: 30%
500
0
0
5
10
15
(C) Fig. 5. Ignition delay under different temperature and pressure conditions.
15
806
J. Won et al. / Energy 163 (2018) 800e810
the flame is the dominant heat source for droplet heating and evaporation [21]. However, the luminosity of the flame around the droplet is relatively small for a water volume ratio of 30%. Under these conditions, the water vapor dilutes the fuel vapor concentration around the droplet surface, and thus, the heat feedback from the ambient temperature is smaller than the case with a water volume ratio of 10%. Flow induction from vigorous puffing, as seen in Fig. 7, was another reason for greater differences in burning rates at a water volume ratio of 30%. 3.3. Effect of ambient pressure A change in the ambient pressure significantly affected droplet ignition and combustion behavior. The graph in Fig. 8 shows the results of droplet diameter and temperature at an ambient temperature of 700 C and a water volume ratio of 20%, with changing ambient pressure. For the cases with 5, 10, and 15 bar, the changes in the droplet diameter were similar. The droplet diameter increased during the droplet heating period and started to marginally decrease before droplet ignition. However, the droplet diameter started to decrease more rapidly for the 1 bar case. This indicated that the droplet was sufficiently heated during the droplet heating period and evaporation of the droplet was dominant. In Fig. 8, the droplet temperature at 1 bar rapidly reached the water boiling temperature at a normalized time of 0.4 and maintained a temperature marginally above the water boiling
3.5
3.5
3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.5
2
Average burning rate (mm /s)
decreased [31]. The ignition delay for different ambient temperatures is shown in Fig. 5. The lines in the graphs indicate the ignition delays at different ambient temperatures, and each graph was labelled according to the water volume ratios. The ignition delay was affected by droplet heating time, evaporation, molecular diffusion, and the chemical reaction of the fuel-oxidizer [32]. All of the above, except for the heating time, decreased at low ambient temperatures, resulting in an increase in ignition delay. For all experimental conditions, the results clearly indicated that the ignition delay decreased as the ambient temperature increased. The average burning rate for different ambient temperatures is shown in Fig. 6. In the results, the ambient temperature was not significantly affected by the burning rate for water ratios of 10% and 20%. This was because the flame temperature was significantly higher than the ambient temperature and the bulk of the heat feedback on the droplet was from the flame as opposed to the ambient conditions. Under these conditions, the distance between the flame and the droplet was the primary factor that determines the amount of heat feedback. A marginal difference in the burning rate under ambient temperature conditions compared to the other cases was observed for the 30% water ratio case. The reason for this could be the difference in the flame structure. Fig. 7 shows the droplet flame for a classical droplet combustion period for different water volume ratios at 15 bar and 600 C. At a water ratio of 10%, a yellow flame obscured the droplet. This was because of the high concentration of fuel vapor around the droplet surface. In this case,
1.0
0
5
10
15
Pressure (bar)
1.0
0
5
10
(B)
(A)
3.5
Ambient temperature o 500 C o 600 C o 700 C
3.0
2.5
2.0
Water volume ratio A: 10% B: 20% C: 30%
1.5
1.0
0
5
10
15
(C) Fig. 6. Average burning rate under different temperature and pressure conditions.
15
J. Won et al. / Energy 163 (2018) 800e810
10 %
20 %
807
30%
Fig. 7. Images of classical combustion and droplet puffing at different water ratios at 15 bar and 600 C.
temperature because of the latent heat of water. This result indicated that the droplet at 1 bar was sufficiently heated during the droplet heating period. In this condition, puffing was initiated before droplet ignition. The heating rate of the droplet was higher for the 1 bar case after a normalized time of 0.3 because of puffing. The droplet diameter graph indicates that intensive puffing initiated at a normalized time of 0.3, and this could induce the inner flow of the droplet. Therefore, the heat transfer rate from the droplet surface to the inner droplet would be increased [32,33]. Puffing was rarely observed prior to ignition at elevated pressures as the droplet temperature was significantly smaller than the water boiling temperature. Therefore, droplet heating was relatively slow compared to the 1bar case. The droplet temperature at ignition increased with ambient pressure. The droplet temperature had to undergo a greater increase to generate a sufficient fuel vapor concentration under the higher ambient pressure conditions because of the ClausiuseClapeyron relationship [34]. As shown in Figs. 5 and 8, the ignition delay increased with ambient pressure as a longer heating time was needed for ignition under high-pressure conditions. Interestingly, an increase in the ignition delay was observed at 1bar conditions. It appears that the vigorous puffing process diluted the fuel vapor around the droplet, which negatively affected
the ignition. After ignition, the droplet temperature further increased, and puffing started at the elevated ambient pressure conditions. The most distinctive feature of the flame at different pressures was its transparency. Fig. 9 clearly shows that the flame sheet narrows and becomes opaquer with increasing ambient pressure. Under 1-bar conditions, the flame was generated at a distance from the droplet surface, whereas the flame approached the droplet under higher ambient pressure conditions. In addition, the flame at 10 bar and 15 bar was so dense that the droplet could not be seen. This was because of the difference in diffusion of the fuel vapor. On the one hand, under low ambient pressure conditions, it was easy for fuel vapor to diffuse; therefore, the flame was held back from the droplet surface. On the other hand, increasing the ambient pressure suppressed the Stefan flow by decreasing the fuel vapor diffusion. Therefore, the flame sheet shrank, and the droplet was covered by the flame [21]. Under these conditions, the puffing process assisted the widening of the flame sheet, even under high ambient pressure conditions, by spreading atomized fuel, as seen in Fig. 7. A rise in the average burning rate at higher ambient pressure conditions was clearly observed in Fig. 6. As discussed above, an increase in ambient pressure caused the flame to approach the droplet surface. Therefore, the overall burning rate at elevated ambient pressures increased because of an increase in heat
808
J. Won et al. / Energy 163 (2018) 800e810
Normalized 2 2 diameter (d /d0)
1.5 1.0 0.5 0.5
1.0
1.5
Ambient pressure 1 bar 5 bar 10 bar 15 bar
o
Droplet temperature ( C)
400 0.0
300
200
100
Ignition 0 0.0
0.5
1.0
1.5
2
Normalized time (s/d0) Fig. 8. History of normalized droplet diameters and temperatures for different ambient pressure conditions (Ambient temperature: 700 C, water volume ratio: 20%).
DCT
DCT
DCT
Fig. 9. Flame sheet formation under different pressure conditions at
feedback from the flame. However, the average burning rate increased in the 1-bar cases. The reason for this was that the burning rate did not actually increase, but the measured value of the droplet diameter could have been exaggerated because of droplet inflation due to puffing and the significant amount of liquid content inside the droplet had already evaporated because of the long ignition delay. 3.4. Effect of water volume ratio Fig. 10 shows the normalized droplet diameters and temperatures for different water volume ratios when the ambient temperature and pressure were 600 C and 10 bar, respectively. In the graph, the rate of droplet temperature increasing decreased with increasing water volume ratio. As can be seen in Table 1, water has a higher specific heat capacity than n-decane at the same volume. Therefore, water in the emulsion fuel hindered droplet temperature increases during the droplet heating period. The ignition delay for different water volume ratios is shown in Fig. 11, that is the same graph as in Fig. 5, however, the x-axis was
DCT 700 C.
rearranged to more clearly compare the effect of water volume ratio, and each graph showed different ambient temperatures. For all conditions, the ignition delay increased with water volume ratio because of the delay in droplet heating as discussed above. However, the difference in ignition delay for each water volume ratio decreased with increasing ambient temperature. As the droplet heating time under higher temperature conditions decreased, the ignition delay among the different water volume ratio cases also decreased. After ignition, the droplet temperature increased rapidly over a short period and was maintained at approximately 250 C, as shown in Fig. 10. During this period, heat loss from a vigorous puffing process prevented an increase in droplet temperatures. The puffing period time and intensity increased with increasing water volume ratio. For example, puffing continued until droplet flame extinction for water volume ratios of 30%. Distortion of the flame sheet at high water volume ratios, as shown in Fig. 7, was a typical phenomenon during droplet combustion regardless of the ambient conditions. It was apparent that the puffing process decreased the flame temperature and enhanced the fuel/oxidizer mixing process
2
1.2 1.0 0.8 400 0.0
o
Droplet temperature ( C
Average burning rate (mm/s )
Normalized 2 2 diameter (d /d0)
J. Won et al. / Energy 163 (2018) 800e810
0.5
300
1.0
1.5
2.0
2.5
809
3.5
3.5
3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.5
1.0
0
5
10
15
1.0
0
Pressure (bar)
Ignition
5
10
15
(B)
(A)
3.5
200
3.0
Water volume ratio 10% 20% 30%
100
Water volume ratio 10% 20% 30%
2.5
2.0
1.5
0 0.0
0.5
1.0
1.5
2.0
2.5
1.0
0
2
Normalized time (s/d0)
Ignition delay (ms)
[26]. The average burning rate of emulsion droplets at different water volume ratios is shown in Fig. 12, which is a modified version of
Fig. 6. Despite a number of deviations, the average burning rate was not significantly affected by water volume ratios. The reason for this was the compensation effect of heat feedback from the flame and the puffing process. An increase in the water volume ratio diluted
2500
2000
2000
1500
1500
1000
1000
500
500
5
10
15
Pressure (bar)
0
0
5
10
15
(B)
(A)
2500
15
Fig. 12. Average burning rate under different water volume ratio conditions.
2500
0
10
(C)
Fig. 10. History of normalized droplet diameters and droplet temperatures for different water volume ratios (Ambient temperature: 600 C, pressure: 10 bar).
0
5
Ambient temperature o o o A: 500 C B: 600 C C: 700 C
2000
Water volume ratio 10% 20% 30%
1500
1000
500
0
0
5
10
15
Ambient temperature o o o A: 500 C B: 600 C C: 700 C
(C) Fig. 11. Ignition delay under different water volume ratio conditions.
810
J. Won et al. / Energy 163 (2018) 800e810
the flame because of the water vapor and decreased the heat feedback towards the droplet, as well as the burning rate, as discussed above. However, the puffing process increased the droplet burning rate because of fuel atomization. The compensation effect of these two factors resulted in similar droplet burning rates regardless of water volume ratios. This result implies that emulsion fuel can achieve a similar burning rate to conventional fuel, even at lower flame temperatures. In the above results, the ignition delay and average burning rate of droplets were affected by the ambient temperature, pressure, and the water volume ratio depending on the conditions. These results indicate that a consideration of combustion characteristics during a combustion system design is required to effectively utilize an emulsion fuel combustor. 4. Conclusion The autoignition and combustion characteristics of a single emulsion fuel droplet under various temperature and pressure conditions were examined experimentally. The results are summarized below. 1. Combustion of an emulsion fuel droplet typically comprised several stages, including droplet heating, classical combustion, puffing, secondary classical combustion, and surfactant combustion. 2. Increasing the ambient temperature decreased the ignition delay of droplets by increasing the droplet heating rate. However, it did not have a significant influence on the average burning rate. 3. Ignition delay increased with ambient pressure because of the suppression of fuel vapor generation. However, the average burning rate increased because of the increase in heat feedback from the flame. 4. Increasing the water volume ratio delayed droplet ignition by delaying droplet heating. However, the average burning rate was similar regardless of the water volume ratio because of the compensation effect of heat feedback and puffing. Acknowledgments This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No.2017R1C1B5016130). References [1] Han Y, Zeng Q, Geng Z, Zhu Q. Energy management and optimization modeling based on a novel fuzzy extreme learning machine: case study of complex petrochemical industries. Energy Conserv Manag 2018;165:163e71. [2] Geng ZQ, Qin L, Han YM, Zhu QX. Energy saving and prediction modeling of petrochemical industries: a novel ELM based on FAHP. Energy 2017;122: 350e62. [3] Cui Y, Geng Z, Zhu Q, Han Y. Review: multi-objective optimization methods and application in energy saving. Energy 2017;125:681e704. [4] Geng Z, Li H, Zhu Q, Han Y. Production prediction and energy-saving model based on Extreme Learning Machine integrated ISM-AHP: application in complex chemical processes. Energy 2018;160:898e909. [5] Tan YH, Abdullah MO, Nolasco-Hipolito C, Zauzi NSA, Abdullah GW. Engine performance and emissions characteristics of a diesel engine fueled with diesel-biodiesel-bioethanol emulsions. Energy Convers Manag 2017;132: 54e64.
[6] Geng ZQ, Dong JG, Han YM, Zhu QX. Energy and environment efficiency analysis based on an improved environment DEA cross-model: case study of complex chemical processes. Appl Energy 2017;205:465e76. [7] Nguyen K-, Dan T, Asano I. Effect of double injection on combustion, performance and emissions of Jatropha water emulsion fueled direct-injection diesel engine. Energy 2015;80:746e55. [8] Kichatov B, Korshunov A, Kiverin A, Son E. Experimental study of foamed emulsion combustion: influence of solid microparticles, glycerol and surfactant. Fuel Process Technol 2017;166:77e85. [9] Ithnin AM, Noge H, Kadir HA, Jazair W. An overview of utilizing water-indiesel emulsion fuel in diesel engine and its potential research study. J Energy Inst 2014;87(4):273e88. n M, Andersson S, Denbratt I. Optical studies of spray [10] Ochoterena R, Lif A, Nyde development and combustion of water-in-diesel emulsion and microemulsion fuels. Fuel 2010;89(1):122e32. [11] Deng S, Zhou J. An experimental study of the effect of water content on combustion of coal tar/water emulsion droplets. Energy 2011;36(10):6130e7. [12] Gong J, Fu W. A Study on the effect of more volatile fuel on evaporation and ignition for emulsified oil. Fuel 2001;80(3):437e45. [13] Shinjo J, Xia J. Combustion characteristics of a single decane/ethanol emulsion droplet and a droplet group under puffing conditions. Proc Combust Inst 2017;36(2):2513e21. [14] Watanabe H, Suzuki Y, Harada T, Matsushita Y, Aoki H, Miura T. An experimental investigation of the breakup characteristics of secondary atomization of emulsified fuel droplet. Energy 2010;35(2):806e13. [15] Ma Z, Li Y, Li Z, Du W, Yin Z, Xu S. Evaporation and combustion characteristics of hydrocarbon fuel droplet in sub- and super-critical environments. Fuel 2018;220:763e8. [16] Zhang X, Li T, Wang B, Wei Y. Superheat limit and micro-explosion in droplets of hydrous ethanol-diesel emulsions at atmospheric pressure and diesel-like conditions. Energy 2018;154(1):535e43. [17] Ghassemi H, Baek SW, Khan QS. Experimental study on evaporation of kerosene droplets at elevated pressures and temperatures. Combust Sci Technol 2006;178(9):1669e84. [18] Javed I, Baek SW, Waheed K. Evaporation characteristics of heptane droplets with the addition of aluminum nanoparticles at elevated temperatures. Combust Flame 2013;160(1):170e83. [19] Harada T, Watanabe H, Suzuki Y, Kamata H, Matsushita Y, Aoki H, Miura T. A numerical investigation of evaporation characteristics of a fuel droplet suspended from a thermocouple. Int J Heat Mass Trans 2011;54(1e3): 649e55. [20] Kim H, Baek SW, Chang D. A single n-heptane droplet behavior in rapid compression machine. Int J Heat Mass Tran 2014;69:247e55. [21] Khan QS, Baek SW, Ghassemi H. On the autoignition and combustion characteristics of kerosene droplets at elevated pressure and temperature. Combust Sci Technol 2007;179(12):2437e51. [22] Chemical engineering and materials research information center, https:// www.cheric.org/research/kdb/hcprop/showcoef.php?cmpid¼10&prop¼PVP. [23] Spalding DB. The combustion of liquid fuels. Symp (Int Combust) 1953;4(1): 847e64. [24] Wang C-, Chen J-. An experimental investigation of the burning characteristics of water-oil emulsions. Int Commun Heat Mass Tran 1996;23(6):823e30. [25] Jeong IC, Lee KH. Auto-ignition and micro-explosion behaviors of droplet arrays of water-in-fuel emulsion. Int J Automot Technol 2008;9(6):735e40. [26] Kim H, Baek SW. Combustion of a single emulsion fuel droplet in a rapid compression machine. Energy 2016;106:422e30. [27] Kim H, Baek SW, Chang D. Auto-ignition characteristics of single n-heptane droplet in a rapid compression machine. Combust Sci Technol 2014;186(7): 912e27. [28] Turns S. An introduction to combustion: concepts and application. third ed. New York: Mc Graw Hill; 2012. [29] Lin C-, Wang K-. Diesel engine performance and emission characteristics using three-phase emulsions as fuel. Fuel 2004;83(4e5):537e45. [30] Gan Y, Qiao L. Combustion characteristics of fuel droplets with addition of nano and micron-sized aluminum particles. Combust Flame 2011;158(2): 354e68. [31] Brandes E, Hirsch W, Stolz T. Autoignition temperatures for mixtures of flammable liquids with air at elevated pressures. Proc Eur Combust Meet 2005;218:1e5. [32] Setyawan HY, Zhu M, Zhang Z, Zhang D. An experimental study of effect of water on ignition and combustion characteristics of single droplets of glycerol. Energy Procedia 2015;75:578e83. [33] Mandal DK, Bakshi S. Internal circulation in a single droplet evaporating in a closed chamber. Int J Multiphas Flow 2012;42:42e51. [34] Gogos G, Ayyaswamy PS. A model for the evaporation of a slowly moving droplet. Combust Flame 1988;74(2):111e29.
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Autoignition and combustion behavior of emulsion droplet under elevated temperature and pressure conditions Jonghan Won a, Seung Wook Baek a, **, Hyemin Kim b, * a
Department of Aerospace Engineering, Korea Advanced Institute of Science and Technology, 373-1 Guseong-dong, Yuseong-gu, Daejeon, South Korea Department of Aeronautical·Mechanical Design Engineering, Korea National University of Transportation, Daehak-ro 50, Chungju City, Chungbuk, South Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 January 2018 Received in revised form 23 August 2018 Accepted 24 August 2018 Available online 25 August 2018
In this study, experiments were conducted to investigate the combustion characteristics of an water-inoil W/O emulsion droplet under elevated temperature and pressure conditions. The base fuel used was ndecane, and total volume ratios of 10, 20, and 30% of distilled water were mixed for producing the emulsion fuel. Span 80 with a volume ratio of 2% was added as a surfactant, and the emulsion fuel was homogeneously mixed via ultrasonication. The combustion process of an emulsion droplet was divided into five stages: droplet heating, classical combustion, puffing, secondary classical combustion, and surfactant combustion. The ignition delay decreased with elevated ambient temperatures, whereas an increase in the ambient pressure and water volume ratio resulted in longer ignition delays. The droplets did not ignite in 500 C or 600 C conditions at 1 bar because of the significant Stefan flow of fuel vapor. After droplet ignition, the droplet combustion process, including classical combustion, puffing, and surfactant combustion, followed. The average burning rate increased with ambient pressure, but it was insensitive to ambient temperatures and water volume ratios. After flame extinction, a secondary flame reappeared because of the combustion of surfactant and residues. © 2018 Elsevier Ltd. All rights reserved.
Keywords: W/O emulsion fuel Droplet Temperature Pressure Water volume ratio Puffing
1. Introduction Numerous researchers dedicated to improving energy efficiencies and decreasing pollutant emissions of conventional fossil fuel. Researches were conducted in total direction because it is urgent matter for sustainable nature and humanity [1e4]. Nevertheless, there were unavoidable limitations to the amount of improvement when conventional fossil fuels were used. Especially, it may be reached the limit in reducing several pollutants matter during combustion such as particulate matter (PM), nitrogen oxide (NOx), and carbon monoxide (CO), which are harmful for environment and human health [5,6]. In the reasons above, emulsion fuel has received particular attention as an alternative fuel for reducing pollutant emissions and improving combustion efficiency. When the emulsion fuel is combusted, superheated water vapor is generated inside the fuel droplet and it explodes at the surface of the droplet; this is known
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (S.W. Baek), [email protected] (H. Kim). https://doi.org/10.1016/j.energy.2018.08.185 0360-5442/© 2018 Elsevier Ltd. All rights reserved.
as “micro-explosion” or “puffing.” Several researchers have reported that the puffing process induces fuel atomization, and therefore, high combustion efficiency and low pollutant emission can be achieved without any additional apparatus or cost [7,8]. For the above reasons, several researchers have conducted studies on emulsion fuel combustion. Ithnin et al. [9], experimentally studied the combustion of water-in-diesel emulsions under various engine load conditions. They reported that emulsion fuel containing 20% water had a maximum cylinder pressure and pressure increase rate comparable to other cases. In addition, emulsion fuel exhibited lower NOx and PM emissions, whereas CO and carbon dioxide emissions increased under low- and high-load conditions compared to the pure diesel case. Ochoterena et al. [10], conducted a study on the spray development and combustion of water-in-diesel emulsions and micro-emulsion fuels. Spray development, droplet break-up, vapor penetration, and combustion processes were observed optically. They reported that an emulsion fuel droplet penetrated further than a conventional fuel droplet, and atomization of emulsion fuel decreased the soot concentration and flame temperature compared to regular diesel fuel. Deng and Zhou [11] experimentally examined
J. Won et al. / Energy 163 (2018) 800e810
the combustion of a coal tar/water emulsion. In their study, a longer ignition delay of the emulsion droplet was observed, but the peak temperature of the droplet was achieved considerably earlier than ordinary coal tar. This indicated the potential of soot reduction and the extent of burnout of cenospheres when using emulsion fuel. Gong and Fu [12] observed ignition delay times for oil-in-water emulsion droplets with various fuel compositions. They reported that the ignition delay time can be decreased by adding more volatile component fuel to the emulsion, and it was notably decreased when the proportion of the volatile component was 10e20%. The effect of this volatile component was not evident at higher concentrations. The distinctive combustion characteristics of emulsion fuels led previous researchers to conduct a number of experimental and theoretical studies, as discussed above. A study of the combustion of a single liquid fuel droplet under high temperature and pressure conditions is the primary approach for investigating the basic phenomena of spray combustion [13,14]. Although several studies dealt with emulsion fuel droplet, no studies have been conducted regarding a single emulsion droplet under high temperature and pressure conditions. Therefore, there is lack of studies and data regarding emulsion fuel droplet combustion under elevated pressure conditions [15,16]. Considering that numerous combustors are operated under high-temperature and high-pressure conditions, observation of the autoignition and combustion behaviors of emulsion fuel under these conditions are essential. This study could provide fundamental knowledge on emulsion fuel during the combustion process. This study focuses on the combustion characteristics of a water/ n-decane emulsion fuel droplet under elevated temperatures and pressures. The water volume ratios in the emulsion fuels were set to 10%, 20%, and 30%, and 2% vol. of Span 80 was added in all cases for obtaining a stable suspension. The combustion characteristics of an emulsion fuel droplet under various temperature and pressure conditions were observed, including ignition delays and burning rates. The results obtained under each experimental condition were compared. 2. Experimental setup The experimental setup comprised a high-pressure combustion chamber, an optical observation system, and measuring sensors. Various ambient pressure and temperature conditions could be investigated with this experimental setup. 2.1. Experimental setup Fig. 1 is a schematic diagram of the experimental apparatus that has been verified and used in previous studies [17,18]. The inner diameter and height of the cylindrical stainless-steel combustion chamber employed in this study were 150 mm and 800 mm, respectively. A single emulsion droplet of 900 ± 100 mm was suspended at the tip of a K-type thermocouple wire (Omega Engineering, Inc.). The shield of the thermocouple was removed, and the two wires in the thermocouple were welded for droplet installation. The diameter of the inner wire was 50 mm and the welded bead size was 100 mm. The droplet temperature could be measured while the droplet was suspended at the tip of the thermocouple during the experiment. This droplet installation method was widely used for several researchers due to its simplicity and accuracy of droplet temperature measurement [19]. Refer to the thermocouple manufacturer (Omega Inc.), the standard error limit of thermocouple was grater of 0.4% The target ambient temperature could be achieved by heating the electric furnace within an error range of ±5 C measured by the
801
K-type thermocouple inside the furnace. The side of the rectangular parallelepiped shape of the electric furnace was 100 mm and its height was 185 mm. The inner part of the electric furnace was encased by a ceramic board for heat insulation and a small 20-mm hole was drilled in the bottom for the thermocouple insertion where a droplet was suspended. When the temperature inside the electric furnace reached the experimental temperature, it was moved to the experimental position, and the experiment was started, as shown in Fig. 1. During the experiment, droplet images were recorded by a high-speed camera at 200 frames per second. A history of the droplet diameter was acquired using an image processing program that was validated in previous studies [20,21]. In the post-processing of the images, errors smaller than 3.12% were found; therefore, the experimental errors were in good agreement with those obtained in previous studies. 2.2. Emulsion fuel Distilled water and n-decane were used for producing water-inoil (W/O) emulsion fuel. The thermophysical properties of water and n-decane are presented in Table 1. Volume ratios of water of 10%, 20%, and 30% were mixed in the n-decane, and 2% vol. of Span 80 was added as a surfactant. As Span 80 has lipophilic characteristics (HLB value: 4.3), it is a representative surfactant that can be used in W/O emulsion. The emulsification process was performed using an ultrasonicator (Sonics & Materials, Inc.). It was operated for 10 min in pulse mode (5 s ON & 5 s OFF) to prevent a rapid increase in the emulsion temperature. The experiment was conducted within an hour of the preparation of the emulsion fuel to prevent errors due to phase separation of the emulsion fuel. The boiling temperature of water and n-decane across the wide pressure range could be calculated as follows:
lnðpkPa Þ ¼ A*lnðTboil Þ þ
B 2 þ C þ D*Tboil Tboil
(1)
The coefficients for water are A ¼ 7.34297, B ¼ 7276.39, C ¼ 67.0245, and D ¼ 4.16191 106, and for n-decane A ¼ 7.76881, B ¼ 8163.33, C ¼ 69.7646, and D ¼ 2.62033 106 [22]. The boiling temperatures at different experimental pressures are presented in Table 2. 2.3. Experimental conditions The experiments were conducted under various temperature (500, 600, and 700 C) and pressure (1, 5, 10, and 15 bar) conditions, with an initial droplet diameter set to 900 ± 100 mm. The experiment was conducted under each temperature and pressure condition with different water/n-decane fuel ratios. Each experiment was repeated a minimum of five times under identical conditions for consistency of the results. The chamber was purged with dry air after each experiment to remove the combustion residue. The ignition delay was the time between the start of the experiment and the droplet ignition. Droplet ignition was defined when a yellow flame first appeared in the image of the droplet, and the droplet burning time was calculated by counting the numbers of frames from droplet ignition to flame extinction. The burning rate constant (Kb) is widely used for examining the combustion phenomena of a single droplet. Burning rate is defined as the temporal change in the squared droplet diameter (d2), and is calculated as follows [23]:
Kb ¼
d 2 d dt
(2)
Unlike droplet evaporation, the observation of a droplet during
802
J. Won et al. / Energy 163 (2018) 800e810
Fig. 1. Experimental setup.
Table 1 Thermophysical properties of n-decane and water.
Specific heat capacity (kJ/kg K) Viscosity (mPa s at 20 C) Density (g/mL) Latent heat (kJ/kg)
water
n-decane
4.19 890 0.997 2257
2.21 920 0.73 263
Table 2 Boling temperature of water and n-decane in various pressure conditions.
1 bar 5 bar 10 bar 15 bar
Water ( C)
n-decane ( C)
100 152 180 198
174 248 291 320
combustion is difficult because of the flame surrounding it. Therefore, it is more beneficial to introduce the average droplet burning rate (Kb,avg), that is the mean change of the square droplet diameter from droplet ignition (dig) to flame extinction (dext). The time between droplet ignition and flame extinction is known as the droplet burning time (tburning). The average droplet burning rate can be calculated as follows [24]:
Kb;ave ¼
d2ig d2ext tburning
(3)
Before proceed the research, validation of experimental setup by comparing the results with previous research was carried out. Fig. 2 shows a comparison of the ignition delay for various n-decane and water volume ratios between a reference study by Jeong et al. [25] and the present study. Although the experimental temperatures differed marginally, the same trends with similar values are observed.
3. Results and discussion 3.1. Typical emulsion droplet combustion behavior Although the detailed combustion phenomena were significantly different for each experimental condition, the typical emulsion droplet combustion characteristics could be observed throughout the experiment. Fig. 3-(a) shows the droplet during the experiment (water volume ratio: 20%, atmospheric temperature: 600 C, and atmospheric pressure: 5 bar), and Fig. 3-(b) shows the droplet diameter and temperature history. As reported in previous studies [26], the combustion process of an emulsion fuel droplet
J. Won et al. / Energy 163 (2018) 800e810
803
Pressure: 1 bar Jeong et al. [25] at 647 Present study at 700
2500
Ignition delay (ms)
2000
1500
1000
500
0 D90+W10
D80+W20
D70+W30
Decane+Water volume ratio Fig. 2. Comparison of ignition delays for different volume ratios between this study and Jeong et al. [25] at 1 bar.
can be divided into several stages, including droplet heating, ignition, classical droplet combustion, puffing, secondary classical combustion, and surfactant combustion. The droplet heating period was initiated immediately after the electric furnace decreased to the experimental position. The primary feature of this period was an increase in the droplet temperature because of the ambient temperature. The droplet temperature gradually increased and approached the boiling temperature of water (152 C) at 5 bar. The droplet diameter increased marginally initially because of thermal expansion, and then started to decrease, as shown in the first period of Fig. 3-(b). Droplet puffing was not observed during this period as the droplet did not heat sufficiently for water vapor generation. The time between the start of experiment and the ignition is the ignition delay. Under the current conditions, at time 0.69 s, the droplet was autoignited in the vicinity of the droplet surface. This is one of the primary characteristics of droplet ignition under high pressure. Under the high-pressure condition, the Stefan flow of fuel vapor from the droplet surface is slow because of an increase in the boiling temperature and a decrease in the diffusion coefficient [27,28]. Under these conditions, ignition occurred close to the droplet surface. As the flame developed around the droplet, the droplet temperature increased rapidly because of the significant heat feedback from the flame. After ignition, droplet combustion without puffing (classical droplet combustion) was observed for a short period of time. This was because the droplet temperature did not reach the water boiling temperature for the puffing process to initiate. The droplet temperature increased until it reached approximately 170 C. The fuel vapor diffusion zone and flame zone are clearly shown in Fig. 3-(a). The structure of the flame close to the droplet changed under different combustion conditions, and is discussed in the next section. Under the 5-bar condition, the puffing process was initiated soon after the classical combustion period. The primary characteristic of this period is distortion of the flame because of puffing. When the droplet temperature exceeded the water boiling temperature, superheated water droplets in the droplet converted to water vapor and exploded from the droplet surface. As can be seen in Fig. 3-(b)-puffing, the flame around the droplet distorted as the
puffing process atomized the fuel droplet. This atomization process can increase the combustion efficiency [29]. The intensity of puffing was dependent on the experimental conditions. During the puffing process, the rate of increase of the droplet temperature was suppressed. In section C in Fig. 3-(b), the droplet temperature was maintained at approximately 180 C during the puffing period despite heat feedback from the flame. In this period, heat addition from the flame and heat loss from the water latent heat was balanced; therefore, the droplet temperature did not change significantly. Typically, an increase in the water volume ratio hindered a droplet temperature increase via a vigorous puffing process [26]. In this study, secondary classical droplet combustion appeared after the puffing period as the water content in the droplet decreased during puffing. A significant increase in the droplet temperature was observed in the late stages of droplet combustion. In the bulk of the experimental cases, the droplet could not be observed because of the luminous flame zone covering the whole droplet. However, in a number of cases, specifically under lowpressure conditions, the diffusion of fuel vapor was sufficient to observe the droplet during the combustion process. The changes in droplet diameter during the combustion process are shown in Fig. 3-(b). This result shows linear regression during droplet combustion. In this result, Kb was 1.65 during the droplet combustion, whereas the calculated Kb,avg was 1.69. This indicated that Kb,avg also represented the combustion rate of the droplet effectively under the present conditions. The flame extinguished approximately 350 ms after droplet ignition. After flame extinction, a secondary flame appeared as shown in Fig. 3-(b)-Surfactant combustion. The surfactant combustion burned low-volatile surfactant and pyrolysis of residues [30]. During this period, the droplet was too small to be optically observed. The temperature during this period increased rapidly and the surfactant combustion lasted less than 40 ms. In the next sections, the effects of ambient temperature, pressure, and water volume ratio on the ignition and combustion of the emulsion fuel droplet are discussed.
804
J. Won et al. / Energy 163 (2018) 800e810
Fig. 3. a) Optical images of droplet at 5 bar, 600 C and water ratio 20% vol., b) History of droplet diameter squared and temperature with stage classification: A (Droplet heating), B (Classical combustion), C (Puffing), D (Secondary classical combustion), and E (Surfactant combustion).
3.2. Effect of ambient temperature The influence of ambient temperature on droplet behavior is discussed in this section. In Fig. 4, the droplet diameter and temperature history at
15 bar at a water volume ratio of 10% is shown. The droplet diameter and time were normalized for the comparison. As mentioned above, the droplet diameter change could not be observed during combustion, as a yellow luminous flame obscured the droplet. In all cases, the droplet diameter increased in the early stages of
Normalized 2 2 diameter (d /d0)
J. Won et al. / Energy 163 (2018) 800e810
1.2
1.0
0.8 0.0
0.5
1.0
1.5
2.0
Ignition
o
Droplet temperature ( C)
300
200
Ambient temperature o 500 C o 600 C o 700 C
100
0 0.0
0.5
1.0
1.5
2.0
2
Normalized time (s/d0)
Igntion delay (ms)
Fig. 4. Time variation of normalized droplet diameter and temperature for different ambient temperature conditions (Ambient pressure: 15 bar, water volume ratio: 10%).
the droplet heating period because of thermal expansion. During this period, the droplet temperature increased more rapidly as the ambient temperature increased. For example, the droplet heating time was approximately 35% faster when the ambient temperature was 700 C compared to 500 C. This was because of the increase in heat feedback under the higher ambient temperature conditions. Referring to previous studies [21], a higher droplet temperature resulted in a higher fuel vapor concentration around the droplet surface, and the fuel/air mixture could be generated relatively quicker for droplet ignition. Therefore, a rise in ambient temperature expedited droplet ignition and eventually reduced the ignition delay time. For all cases, the droplet ignited when the droplet diameter started to decrease. The droplet temperature increased sharply after ignition because of heat feedback from the flame. As the droplet temperature reached approximately 150e170 C, which is marginally smaller than the water boiling temperature, the droplet ignited. Therefore, puffing rarely occurred before droplet ignition. The characteristics of puffing at different ambient pressures are discussed in the next section. As the ambient temperature was smaller, the droplet temperature at ignition was slightly greater. Higher activation energy for ignition was required for a smaller fuel vapor concentration; therefore, the droplet temperature at ignition increased as the ambient temperature
2500
2500
2000
2000
1500
1500
1000
1000
500
500
0
0
5
10
15
805
0
0
5
Pressure (bar)
10
(B)
(A) 2500
2000
Ambient temperature o 500 C o 600 C o 700 C
1500
1000
Water volume ratio A: 10% B: 20% C: 30%
500
0
0
5
10
15
(C) Fig. 5. Ignition delay under different temperature and pressure conditions.
15
806
J. Won et al. / Energy 163 (2018) 800e810
the flame is the dominant heat source for droplet heating and evaporation [21]. However, the luminosity of the flame around the droplet is relatively small for a water volume ratio of 30%. Under these conditions, the water vapor dilutes the fuel vapor concentration around the droplet surface, and thus, the heat feedback from the ambient temperature is smaller than the case with a water volume ratio of 10%. Flow induction from vigorous puffing, as seen in Fig. 7, was another reason for greater differences in burning rates at a water volume ratio of 30%. 3.3. Effect of ambient pressure A change in the ambient pressure significantly affected droplet ignition and combustion behavior. The graph in Fig. 8 shows the results of droplet diameter and temperature at an ambient temperature of 700 C and a water volume ratio of 20%, with changing ambient pressure. For the cases with 5, 10, and 15 bar, the changes in the droplet diameter were similar. The droplet diameter increased during the droplet heating period and started to marginally decrease before droplet ignition. However, the droplet diameter started to decrease more rapidly for the 1 bar case. This indicated that the droplet was sufficiently heated during the droplet heating period and evaporation of the droplet was dominant. In Fig. 8, the droplet temperature at 1 bar rapidly reached the water boiling temperature at a normalized time of 0.4 and maintained a temperature marginally above the water boiling
3.5
3.5
3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.5
2
Average burning rate (mm /s)
decreased [31]. The ignition delay for different ambient temperatures is shown in Fig. 5. The lines in the graphs indicate the ignition delays at different ambient temperatures, and each graph was labelled according to the water volume ratios. The ignition delay was affected by droplet heating time, evaporation, molecular diffusion, and the chemical reaction of the fuel-oxidizer [32]. All of the above, except for the heating time, decreased at low ambient temperatures, resulting in an increase in ignition delay. For all experimental conditions, the results clearly indicated that the ignition delay decreased as the ambient temperature increased. The average burning rate for different ambient temperatures is shown in Fig. 6. In the results, the ambient temperature was not significantly affected by the burning rate for water ratios of 10% and 20%. This was because the flame temperature was significantly higher than the ambient temperature and the bulk of the heat feedback on the droplet was from the flame as opposed to the ambient conditions. Under these conditions, the distance between the flame and the droplet was the primary factor that determines the amount of heat feedback. A marginal difference in the burning rate under ambient temperature conditions compared to the other cases was observed for the 30% water ratio case. The reason for this could be the difference in the flame structure. Fig. 7 shows the droplet flame for a classical droplet combustion period for different water volume ratios at 15 bar and 600 C. At a water ratio of 10%, a yellow flame obscured the droplet. This was because of the high concentration of fuel vapor around the droplet surface. In this case,
1.0
0
5
10
15
Pressure (bar)
1.0
0
5
10
(B)
(A)
3.5
Ambient temperature o 500 C o 600 C o 700 C
3.0
2.5
2.0
Water volume ratio A: 10% B: 20% C: 30%
1.5
1.0
0
5
10
15
(C) Fig. 6. Average burning rate under different temperature and pressure conditions.
15
J. Won et al. / Energy 163 (2018) 800e810
10 %
20 %
807
30%
Fig. 7. Images of classical combustion and droplet puffing at different water ratios at 15 bar and 600 C.
temperature because of the latent heat of water. This result indicated that the droplet at 1 bar was sufficiently heated during the droplet heating period. In this condition, puffing was initiated before droplet ignition. The heating rate of the droplet was higher for the 1 bar case after a normalized time of 0.3 because of puffing. The droplet diameter graph indicates that intensive puffing initiated at a normalized time of 0.3, and this could induce the inner flow of the droplet. Therefore, the heat transfer rate from the droplet surface to the inner droplet would be increased [32,33]. Puffing was rarely observed prior to ignition at elevated pressures as the droplet temperature was significantly smaller than the water boiling temperature. Therefore, droplet heating was relatively slow compared to the 1bar case. The droplet temperature at ignition increased with ambient pressure. The droplet temperature had to undergo a greater increase to generate a sufficient fuel vapor concentration under the higher ambient pressure conditions because of the ClausiuseClapeyron relationship [34]. As shown in Figs. 5 and 8, the ignition delay increased with ambient pressure as a longer heating time was needed for ignition under high-pressure conditions. Interestingly, an increase in the ignition delay was observed at 1bar conditions. It appears that the vigorous puffing process diluted the fuel vapor around the droplet, which negatively affected
the ignition. After ignition, the droplet temperature further increased, and puffing started at the elevated ambient pressure conditions. The most distinctive feature of the flame at different pressures was its transparency. Fig. 9 clearly shows that the flame sheet narrows and becomes opaquer with increasing ambient pressure. Under 1-bar conditions, the flame was generated at a distance from the droplet surface, whereas the flame approached the droplet under higher ambient pressure conditions. In addition, the flame at 10 bar and 15 bar was so dense that the droplet could not be seen. This was because of the difference in diffusion of the fuel vapor. On the one hand, under low ambient pressure conditions, it was easy for fuel vapor to diffuse; therefore, the flame was held back from the droplet surface. On the other hand, increasing the ambient pressure suppressed the Stefan flow by decreasing the fuel vapor diffusion. Therefore, the flame sheet shrank, and the droplet was covered by the flame [21]. Under these conditions, the puffing process assisted the widening of the flame sheet, even under high ambient pressure conditions, by spreading atomized fuel, as seen in Fig. 7. A rise in the average burning rate at higher ambient pressure conditions was clearly observed in Fig. 6. As discussed above, an increase in ambient pressure caused the flame to approach the droplet surface. Therefore, the overall burning rate at elevated ambient pressures increased because of an increase in heat
808
J. Won et al. / Energy 163 (2018) 800e810
Normalized 2 2 diameter (d /d0)
1.5 1.0 0.5 0.5
1.0
1.5
Ambient pressure 1 bar 5 bar 10 bar 15 bar
o
Droplet temperature ( C)
400 0.0
300
200
100
Ignition 0 0.0
0.5
1.0
1.5
2
Normalized time (s/d0) Fig. 8. History of normalized droplet diameters and temperatures for different ambient pressure conditions (Ambient temperature: 700 C, water volume ratio: 20%).
DCT
DCT
DCT
Fig. 9. Flame sheet formation under different pressure conditions at
feedback from the flame. However, the average burning rate increased in the 1-bar cases. The reason for this was that the burning rate did not actually increase, but the measured value of the droplet diameter could have been exaggerated because of droplet inflation due to puffing and the significant amount of liquid content inside the droplet had already evaporated because of the long ignition delay. 3.4. Effect of water volume ratio Fig. 10 shows the normalized droplet diameters and temperatures for different water volume ratios when the ambient temperature and pressure were 600 C and 10 bar, respectively. In the graph, the rate of droplet temperature increasing decreased with increasing water volume ratio. As can be seen in Table 1, water has a higher specific heat capacity than n-decane at the same volume. Therefore, water in the emulsion fuel hindered droplet temperature increases during the droplet heating period. The ignition delay for different water volume ratios is shown in Fig. 11, that is the same graph as in Fig. 5, however, the x-axis was
DCT 700 C.
rearranged to more clearly compare the effect of water volume ratio, and each graph showed different ambient temperatures. For all conditions, the ignition delay increased with water volume ratio because of the delay in droplet heating as discussed above. However, the difference in ignition delay for each water volume ratio decreased with increasing ambient temperature. As the droplet heating time under higher temperature conditions decreased, the ignition delay among the different water volume ratio cases also decreased. After ignition, the droplet temperature increased rapidly over a short period and was maintained at approximately 250 C, as shown in Fig. 10. During this period, heat loss from a vigorous puffing process prevented an increase in droplet temperatures. The puffing period time and intensity increased with increasing water volume ratio. For example, puffing continued until droplet flame extinction for water volume ratios of 30%. Distortion of the flame sheet at high water volume ratios, as shown in Fig. 7, was a typical phenomenon during droplet combustion regardless of the ambient conditions. It was apparent that the puffing process decreased the flame temperature and enhanced the fuel/oxidizer mixing process
2
1.2 1.0 0.8 400 0.0
o
Droplet temperature ( C
Average burning rate (mm/s )
Normalized 2 2 diameter (d /d0)
J. Won et al. / Energy 163 (2018) 800e810
0.5
300
1.0
1.5
2.0
2.5
809
3.5
3.5
3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.5
1.0
0
5
10
15
1.0
0
Pressure (bar)
Ignition
5
10
15
(B)
(A)
3.5
200
3.0
Water volume ratio 10% 20% 30%
100
Water volume ratio 10% 20% 30%
2.5
2.0
1.5
0 0.0
0.5
1.0
1.5
2.0
2.5
1.0
0
2
Normalized time (s/d0)
Ignition delay (ms)
[26]. The average burning rate of emulsion droplets at different water volume ratios is shown in Fig. 12, which is a modified version of
Fig. 6. Despite a number of deviations, the average burning rate was not significantly affected by water volume ratios. The reason for this was the compensation effect of heat feedback from the flame and the puffing process. An increase in the water volume ratio diluted
2500
2000
2000
1500
1500
1000
1000
500
500
5
10
15
Pressure (bar)
0
0
5
10
15
(B)
(A)
2500
15
Fig. 12. Average burning rate under different water volume ratio conditions.
2500
0
10
(C)
Fig. 10. History of normalized droplet diameters and droplet temperatures for different water volume ratios (Ambient temperature: 600 C, pressure: 10 bar).
0
5
Ambient temperature o o o A: 500 C B: 600 C C: 700 C
2000
Water volume ratio 10% 20% 30%
1500
1000
500
0
0
5
10
15
Ambient temperature o o o A: 500 C B: 600 C C: 700 C
(C) Fig. 11. Ignition delay under different water volume ratio conditions.
810
J. Won et al. / Energy 163 (2018) 800e810
the flame because of the water vapor and decreased the heat feedback towards the droplet, as well as the burning rate, as discussed above. However, the puffing process increased the droplet burning rate because of fuel atomization. The compensation effect of these two factors resulted in similar droplet burning rates regardless of water volume ratios. This result implies that emulsion fuel can achieve a similar burning rate to conventional fuel, even at lower flame temperatures. In the above results, the ignition delay and average burning rate of droplets were affected by the ambient temperature, pressure, and the water volume ratio depending on the conditions. These results indicate that a consideration of combustion characteristics during a combustion system design is required to effectively utilize an emulsion fuel combustor. 4. Conclusion The autoignition and combustion characteristics of a single emulsion fuel droplet under various temperature and pressure conditions were examined experimentally. The results are summarized below. 1. Combustion of an emulsion fuel droplet typically comprised several stages, including droplet heating, classical combustion, puffing, secondary classical combustion, and surfactant combustion. 2. Increasing the ambient temperature decreased the ignition delay of droplets by increasing the droplet heating rate. However, it did not have a significant influence on the average burning rate. 3. Ignition delay increased with ambient pressure because of the suppression of fuel vapor generation. However, the average burning rate increased because of the increase in heat feedback from the flame. 4. Increasing the water volume ratio delayed droplet ignition by delaying droplet heating. However, the average burning rate was similar regardless of the water volume ratio because of the compensation effect of heat feedback and puffing. Acknowledgments This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No.2017R1C1B5016130). References [1] Han Y, Zeng Q, Geng Z, Zhu Q. Energy management and optimization modeling based on a novel fuzzy extreme learning machine: case study of complex petrochemical industries. Energy Conserv Manag 2018;165:163e71. [2] Geng ZQ, Qin L, Han YM, Zhu QX. Energy saving and prediction modeling of petrochemical industries: a novel ELM based on FAHP. Energy 2017;122: 350e62. [3] Cui Y, Geng Z, Zhu Q, Han Y. Review: multi-objective optimization methods and application in energy saving. Energy 2017;125:681e704. [4] Geng Z, Li H, Zhu Q, Han Y. Production prediction and energy-saving model based on Extreme Learning Machine integrated ISM-AHP: application in complex chemical processes. Energy 2018;160:898e909. [5] Tan YH, Abdullah MO, Nolasco-Hipolito C, Zauzi NSA, Abdullah GW. Engine performance and emissions characteristics of a diesel engine fueled with diesel-biodiesel-bioethanol emulsions. Energy Convers Manag 2017;132: 54e64.
[6] Geng ZQ, Dong JG, Han YM, Zhu QX. Energy and environment efficiency analysis based on an improved environment DEA cross-model: case study of complex chemical processes. Appl Energy 2017;205:465e76. [7] Nguyen K-, Dan T, Asano I. Effect of double injection on combustion, performance and emissions of Jatropha water emulsion fueled direct-injection diesel engine. Energy 2015;80:746e55. [8] Kichatov B, Korshunov A, Kiverin A, Son E. Experimental study of foamed emulsion combustion: influence of solid microparticles, glycerol and surfactant. Fuel Process Technol 2017;166:77e85. [9] Ithnin AM, Noge H, Kadir HA, Jazair W. An overview of utilizing water-indiesel emulsion fuel in diesel engine and its potential research study. J Energy Inst 2014;87(4):273e88. n M, Andersson S, Denbratt I. Optical studies of spray [10] Ochoterena R, Lif A, Nyde development and combustion of water-in-diesel emulsion and microemulsion fuels. Fuel 2010;89(1):122e32. [11] Deng S, Zhou J. An experimental study of the effect of water content on combustion of coal tar/water emulsion droplets. Energy 2011;36(10):6130e7. [12] Gong J, Fu W. A Study on the effect of more volatile fuel on evaporation and ignition for emulsified oil. Fuel 2001;80(3):437e45. [13] Shinjo J, Xia J. Combustion characteristics of a single decane/ethanol emulsion droplet and a droplet group under puffing conditions. Proc Combust Inst 2017;36(2):2513e21. [14] Watanabe H, Suzuki Y, Harada T, Matsushita Y, Aoki H, Miura T. An experimental investigation of the breakup characteristics of secondary atomization of emulsified fuel droplet. Energy 2010;35(2):806e13. [15] Ma Z, Li Y, Li Z, Du W, Yin Z, Xu S. Evaporation and combustion characteristics of hydrocarbon fuel droplet in sub- and super-critical environments. Fuel 2018;220:763e8. [16] Zhang X, Li T, Wang B, Wei Y. Superheat limit and micro-explosion in droplets of hydrous ethanol-diesel emulsions at atmospheric pressure and diesel-like conditions. Energy 2018;154(1):535e43. [17] Ghassemi H, Baek SW, Khan QS. Experimental study on evaporation of kerosene droplets at elevated pressures and temperatures. Combust Sci Technol 2006;178(9):1669e84. [18] Javed I, Baek SW, Waheed K. Evaporation characteristics of heptane droplets with the addition of aluminum nanoparticles at elevated temperatures. Combust Flame 2013;160(1):170e83. [19] Harada T, Watanabe H, Suzuki Y, Kamata H, Matsushita Y, Aoki H, Miura T. A numerical investigation of evaporation characteristics of a fuel droplet suspended from a thermocouple. Int J Heat Mass Trans 2011;54(1e3): 649e55. [20] Kim H, Baek SW, Chang D. A single n-heptane droplet behavior in rapid compression machine. Int J Heat Mass Tran 2014;69:247e55. [21] Khan QS, Baek SW, Ghassemi H. On the autoignition and combustion characteristics of kerosene droplets at elevated pressure and temperature. Combust Sci Technol 2007;179(12):2437e51. [22] Chemical engineering and materials research information center, https:// www.cheric.org/research/kdb/hcprop/showcoef.php?cmpid¼10&prop¼PVP. [23] Spalding DB. The combustion of liquid fuels. Symp (Int Combust) 1953;4(1): 847e64. [24] Wang C-, Chen J-. An experimental investigation of the burning characteristics of water-oil emulsions. Int Commun Heat Mass Tran 1996;23(6):823e30. [25] Jeong IC, Lee KH. Auto-ignition and micro-explosion behaviors of droplet arrays of water-in-fuel emulsion. Int J Automot Technol 2008;9(6):735e40. [26] Kim H, Baek SW. Combustion of a single emulsion fuel droplet in a rapid compression machine. Energy 2016;106:422e30. [27] Kim H, Baek SW, Chang D. Auto-ignition characteristics of single n-heptane droplet in a rapid compression machine. Combust Sci Technol 2014;186(7): 912e27. [28] Turns S. An introduction to combustion: concepts and application. third ed. New York: Mc Graw Hill; 2012. [29] Lin C-, Wang K-. Diesel engine performance and emission characteristics using three-phase emulsions as fuel. Fuel 2004;83(4e5):537e45. [30] Gan Y, Qiao L. Combustion characteristics of fuel droplets with addition of nano and micron-sized aluminum particles. Combust Flame 2011;158(2): 354e68. [31] Brandes E, Hirsch W, Stolz T. Autoignition temperatures for mixtures of flammable liquids with air at elevated pressures. Proc Eur Combust Meet 2005;218:1e5. [32] Setyawan HY, Zhu M, Zhang Z, Zhang D. An experimental study of effect of water on ignition and combustion characteristics of single droplets of glycerol. Energy Procedia 2015;75:578e83. [33] Mandal DK, Bakshi S. Internal circulation in a single droplet evaporating in a closed chamber. Int J Multiphas Flow 2012;42:42e51. [34] Gogos G, Ayyaswamy PS. A model for the evaporation of a slowly moving droplet. Combust Flame 1988;74(2):111e29.