author's personal copy - kaistprocom.kaist.ac.kr/download/ijp/131.pdfexothermic...

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

http://www.elsevier.com/authorsrights

Author's personal copy

Evaporation characteristics of kerosene droplets with diluteconcentrations of ligand-protected aluminum nanoparticles at elevatedtemperatures

Irfan Javed a,1, Seung Wook Baek a,⇑, Khalid Waheed a,1, Ghafar Ali b, Sung Oh Cho b

a Division of Aerospace Engineering, School of Mechanical, Aerospace and Systems Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro,Yuseong-Gu, Daejeon 305-701, Republic of Koreab Department of Nuclear and Quantum Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-Gu, Daejeon 305-701, Republic of Korea

a r t i c l e i n f o

Article history:Received 6 April 2013Received in revised form 17 June 2013Accepted 4 July 2013Available online 27 July 2013

Keywords:Kerosene vaporizationHigh-energy-density fuelsNanofluid fuelsNanoparticlesAluminum

a b s t r a c t

The evaporation characteristics of kerosene droplets containing dilute concentrations (0.1%, 0.5%, and1.0% by weight) of ligand-protected aluminum (Al) nanoparticles (NPs) suspended on silicon carbide fiberwere studied experimentally at different ambient temperatures (400–800 �C) under normal gravity. Theevaporation behavior of pure and stabilized kerosene droplets was also examined for comparison. Theresults show that at relatively low temperatures (400–600 �C), the evaporation behavior of suspendedkerosene droplets containing dilute concentrations of Al NPs was similar to that of pure kerosene dropletsand exhibited two-stage evaporation following the classical d2-law. However, at relatively high temper-atures (700–800 �C), bubble formation and micro-explosions were observed, which were not detected inpure or stabilized kerosene droplets. For all Al NP suspensions, regardless of the concentration, the evap-oration rate remained higher than that of pure and stabilized kerosene droplets in the range 400–800 �C.At relatively low temperatures, the evaporation rate increased slightly. However, at relatively high tem-peratures (700–800 �C), the melting of Al NPs led to substantial enhancement of evaporation. The max-imum increase in the evaporation rate (56.7%) was observed for the 0.5% Al NP suspension at 800 �C.

� 2013 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

1. Introduction

Nanofluid fuels, a new class of nanofluids containing highlyexothermic energetic-material (i.e., Al, B, Fe, and C) nanoparticles(NPs), have received considerable attention in recent years [1–6].These consist of stable suspensions of energetic NPs (10–100 nm)in traditional liquid fuels. Energetic NPs are usually metallic andhave advantages such as higher reactivity, faster burning rates,and lower ignition temperatures compared to micron-sizedparticles in solid propellants and fuels [1]. Metallic NPs have thepotential to enhance the volumetric energy density of liquid fuels.However, despite this fascinating benefit, nanofluid fuels haverarely been studied. Problems such as the low dispersity of metalNPs in liquid hydrocarbons or nonpolar organic solvents and thepotential emissions of metal and metal oxide particles have limitedthe production and application of nanofluid fuels.

In hot-plate experiments, Tyagi et al. [7] observed that with theaddition of small amounts of Al and Al2O3 NPs, the ignition

probability of nanoparticle-laden diesel fuel was significantly high-er than that of pure diesel fuel. Jackson et al. [8] found that theaddition of Al NPs could substantially decrease the ignition delaytime of n-dodecane in a shock tube above 1175 K. Using an aerosolshock tube, Allen et al. [9] observed that the addition of 2% (byweight) of Al NPs to ethanol and JP-8 reduced their ignition delaysby 32% and 50%, respectively. Van Devener and Anderson [10] firstperformed tests of JP-10 catalytic combustion using soluble CeO2

NPs as the catalyst. Their results indicate that the ignition temper-ature of JP-10 was significantly reduced. The group later estab-lished a unique process that produced oxide-free, fuel-soluble,and air-stable boron NPs that were coated with the combustioncatalyst ceria [11,12]. Gan and Qiao [13] studied the effects ofnano- and micro-sized Al particles on the burning characteristicsof n-decane and ethanol fuel droplets. Their results show that forthe same particles and surfactant concentrations, the disruptionand micro-explosion behavior of the micro-sized suspensionsoccurred later with much greater intensity. Gan et al. [14]compared the burning behavior of dilute and dense suspensionsof boron and iron NPs in ethanol and n-decane. Simultaneous burn-ing of both the droplets and the particles was observed for dilutesuspensions, whereas in dense suspensions most particles wereburned as large agglomerates after consumption of the liquid fuel.

0010-2180/$ - see front matter � 2013 The Combustion Institute. Published by Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.combustflame.2013.07.007

⇑ Corresponding author. Fax: +82 42 350 3710.E-mail address: [email protected] (S.W. Baek).

1 Permanent address: Pakistan Institute of Engineering and Applied Sciences(PIEAS), Nilore 45650, Islamabad, Pakistan.

Combustion and Flame 160 (2013) 2955–2963

Contents lists available at SciVerse ScienceDirect

Combustion and Flame

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


Evaporation of liquid fuel droplets is an important processduring combustion. Liquid fuels are injected into a combustor assprayed droplets, and the vaporization and oxidation characteris-tics of these droplets are related to combustion performance. Thedesign and optimization of various practical combustion systems,such as liquid propellant rocket engines, diesel engines, gas tur-bines, and oil-fired furnaces, require accurate knowledge of thevaporization of these liquid fuel droplets. The evaporation behaviorof nanofluid fuel droplets is more complex than that of pure liquidfuel droplets due to their multi-component, multi-scale, and multi-phase nature, which is more difficult to predict. Experimental stud-ies, which are very rare, are required to provide the evaporationcharacteristics of nanofluid fuel droplets.

Chen et al. [15] studied the effect of laponite, Fe2O3, and Ag NPson the evaporation rate of deionized water under natural convec-tion at room temperature. These nanofluid droplets evaporate atdifferent rates from the base fluid. The evaporation rates of variousAg and Fe2O3 nanofluids transition from one constant value to an-other during the evaporation process. The authors explained theeffects of various NPs on evaporation based on the apparent heatof evaporation. Gan and Qiao [16] investigated the effect of AlNPs on the evaporation of ethanol and n-decane fuel droplets un-der natural and weak forced convection at temperatures up to380 K. A deviation from the classical d2-law was observed underthese conditions at 300 and 320 K. Gan and Qiao [17] demon-strated that the evaporation rates of ethanol-based nanofluids con-taining multi-walled carbon nanotubes (MWCNTs) or carbonnanoparticles (CNPs) are both higher than the evaporation rate ofpure ethanol. They have also shown that radiation absorption canbe significantly enhanced by adding a small amount of Al NPs toethanol, which increases the nanofluid droplet temperature andenhances the droplet evaporation rate [18]. More recently, we haveshown [19] that the addition of Al NPs to heptane increases itsevaporation rate at relatively high temperatures (above 400 �C).At relatively low temperatures (below 400 �C), the formation oflarge agglomerates results in a compact shell that suppresses theevaporation of heptane.

In our previous work [19], we used a single-component liquidhydrocarbon (heptane) as a base fuel. In the present study, we usedkerosene, which is a multi-component liquid hydrocarbon.Kerosene was selected because most aviation fuels (e.g., JP-5,JP-7, JP-8, or Jet A/A-1) and liquid hydrocarbon propellants (e.g.,RP-1) can be described generically as kerosene [20]. The evapora-tion and combustion behavior of NP-laden kerosene droplets willprovide us with an improved understanding of the effects of addingNPs to such multi-component hydrocarbon fuels. Surface-modifiedAl NPs were added to the kerosene to suppress the formation oflarge agglomerates (residues). Oleic acid (OA) was selected for sur-face coating of the Al NPs, instead of the previously used Span 85surfactant, because of its lower molecular weight and its boilingpoint (360 �C), which is closer to the boiling point of the base fuel(180–270 �C). Due to this surface modification technique, un-oxidized, ligand-coated, and fuel-soluble NPs were obtained thatwere more stable in suspension and had high active metalcontents [12]. We then experimentally investigated the evapora-tion characteristics of multi-component hydrocarbon-based nano-fluid fuel droplets with the addition of dilute concentrations ofligand-protected Al NPs at elevated temperatures.

2. Experimental methods

2.1. Materials and characterization instruments

Al NPs (99.9%, metal basis, 70 nm) were purchased from USResearch Nanomaterials (Houston, Texas). Kerosene (extra pure,

boiling range 180–270 �C, specific gravity 0.8 at 15 �C) wasobtained from Junsei Chemical Co. (Japan), and OA (C18H34O2,technical grade, 90%) was purchased from Sigma–Aldrich. Siliconcarbide (SiC) fiber (100 lm diameter) was obtained from Goodfel-low (England). All these materials, except the Al NPs, were used intheir as-received form without further treatment.

A planetary ball mill (Retsch PM100, GmbH, Germany) wasused to modify the surfaces of the Al NPs. The morphology of theAl NPs, as received and after coating with OA, was studied usingfield-emission scanning electron microscopy (FESEM, Magellan400, FEI Company Eindhoven, Netherlands). The residues were alsoexamined by the same instrument. Quantitative analysis wascarried out with energy-dispersive X-ray spectroscopy (EDX) incombination with the FESEM.

2.2. Nanofluid fuel preparation

The stable nanofluid fuels were prepared using a two-stepmethod, which is the most economic method to produce varioustypes of nanofluids with various concentrations of NPs, either inthe laboratory or at a large scale, because the metal NP synthesistechniques have already been scaled up to industrial productionlevels [21]. A post-synthesis surface modification concept was ap-plied to the purchased Al NPs to improve their dispersion stabilityin liquid hydrocarbons. Using a ball mill, the surfaces of the NPswere coated by OA to prepare oxide-free and fuel-soluble Al NPs.The procedure adopted here has been described elsewhere [12],except for the following amendments.

A planetary ball mill (PM100) was used with a 12-mL stainlesssteel (SS) milling jar and 3-mm-diameter SS grinding balls. The AlNPs were added to the milling jar along with OA, which was usedto coat the surfaces of the Al NPs. The amount of OA was optimizedto 1:2 (ratio of OA to Al mass) by conducting a series of experi-ments with different ratios (1:2, 1:1, and 2:1) and studying the sta-bility of the suspensions in kerosene. The time of ball milling wasalso optimized to 1 h. The charge ratio (ratio of the ball to the sam-ple mass) was kept low (�2:1) to minimize excessive grinding ofthe Al NPs. The resulting NP/OA paste was dispersed in kerosenethrough vigorous agitation by hand. A homogeneous suspensionwas obtained in this way, which remained stable for 10 h withno obvious sedimentation of the NPs.

Figure 1a shows an SEM image of the Al NPs as received fromthe supplier, which clearly indicates that the NPs were sphericalin shape with smooth surfaces. Most of the NPs were well sepa-rated from each other. An SEM micrograph of the Al NPs coatedwith OA is shown in Fig. 1b. This figure indicates that the Al NPsmaintained their shapes and sizes during the coating process.

2.3. Experimental procedure

An experimental apparatus was previously fabricated and in-stalled to study the evaporation, autoignition, and combustionbehavior of a single suspended liquid-fuel droplet at elevated pres-sures and temperatures, as described in [22–25]. The same expri-mental setup has also been used to study the evaporation ofheptane-based nanofluid fuel droplets [19]. The experimental pro-cedure, and the data reduction and analysis, were also described indetail in that study. The experiments were performed with an iso-lated nanofluid fuel droplet suspended by a fine SiC fiber (100 lmdiameter). The heat loss from the fiber can be neglected duringmost of a droplet’s lifetime for fiber diameters of less than100 lm [26]. The initial average diameter of a droplet was1.0 ± 0.10 mm. The ambient temperature was varied from 400 to800 �C, which is higher than the boiling point of kerosene andthe melting point of the Al NPs, and the ambient pressure was keptconstant at 0.1 MPa. High-temperature ambience was provided by

2956 I. Javed et al. / Combustion and Flame 160 (2013) 2955–2963


a freely falling electric furnace. The evaporation process was re-corded by a high-speed charge-coupled device (CCD) camera. Aflexible image-processing code was developed using Matlab to ob-tain the droplets’ diameters from the captured images. This codewas executed iteratively for each image, resulting in the temporalvariation of droplet diameter during evaporation. The evaporationrate can be expressed as the time derivative of the droplet diame-ter squared, Cv = �d(d2)/dt [27,28]. Thus, the droplet evaporationrate was obtained from the temporal history of droplet diametersquared by measuring the slope of its linear regression.

3. Results and discussion

3.1. Pure kerosene droplet evaporation

The evaporation rate of pure kerosene droplets was evaluatedfirst as a baseline to compare the evaporation rates of stabilizedand kerosene-based nanofluid fuel droplets. Fig. 2 shows the vari-ation in the normalized diameter squared (d2/d2

0) with the normal-ized time (t/d2

0), where d0 is the initial droplet diameter, fordifferent ambient temperatures and at a constant pressure of0.1 MPa for pure kerosene droplets. This normalization parameterwas suggested by classical droplet evaporation theory [27]. Theevaporation of each sample followed the same general behavior.After a finite heating-up period, the variation in the droplet diam-eter squared became approximately linear with time, while follow-ing the d2-law during the last stages of evaporation. Despite the

multi-component nature of kerosene, its evaporation follows thed2-law in the range 400–800 �C. Kerosene is a blend of relativelynonvolatile petroleum fractions. It typically consists of 60% paraf-fins, 32% naphthenes, and 7.7% aromatics by volume [24]. As men-tioned in the materials section, the kerosene used in the presentstudy had a boiling point range of 180–270 �C. Each componentvaporizes at a different rate, depending on its volatility. Therefore,a concentration gradient exists around and within a droplet. High-temperature gradients between the environment and the dropletresult in faster heat diffusion compared to mass diffusion. Thus,the temperature inside a droplet remains fairly uniform duringthe evaporation process so that the mass diffusion due to the con-centration gradient controls the evaporation rate. This explainswhy the evaporation process for a multi-component droplet likekerosene follows the d2-law for elevated temperatures from 400to 800 �C, as shown in Fig. 2, and as discussed in our previous work[23]. An increase in the ambient temperature decreases the totallifetime of the kerosene droplets.

Fig. 1. SEM micrographs of (a) Al NPs as received from supplier and (b) Al NPswhose surfaces are coated with oleic acid in a ratio of 1:2 (ratio of OA to Al mass)and sample taken after 1 h ball milling.

Fuel: KeroseneP= 0.1 MPa

t /d20 (s/mm2)

d2/d2 0

0 1 2 3 4 5 6 70

0.2

0.4

0.6

0.8

1

1.2

T=400 oCT=500 oCT=600 oCT=700 oCT=800 oC

Fig. 2. Variations of normalized diameter squared with the normalized time fordifferent ambient temperatures and at a constant pressure of 0.1 MPa for purekerosene droplets.

Fuel: KeroseneP=0.1MPa

Temperature (oC)

Cv(mm2 /s)

400 500 600 700 800 900 10000.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Ghassemi et al. [23]Present

Fig. 3. Comparison of the evaporation rates of kerosene droplets between presentstudy and previous study by Ghassemi et al. [23] at different ambient temperaturesand constant pressure of 0.1 MPa.

I. Javed et al. / Combustion and Flame 160 (2013) 2955–2963 2957


The kerosene droplet evaporation rates obtained from the lin-ear part of the evaporation data in Fig. 2 are plotted in Fig. 3. Thisfigure clearly shows that the evaporation rate increased with theambient temperature. A simple comparison of kerosene evapora-tion rates between the previous results of our group [23] andour current study at 0.1 MPa was plotted (Fig. 3) to illustrate theconsistency of the experimental apparatus. The evaporation ratesof the kerosene droplets obtained in the present study were lowerthan the values previously obtained because of the different fiberdiameters that we used for the droplet suspensions. Ghassemiet al. [23] suspended droplets with an initial diameter of 1.0–1.2 mm on quartz fibers with diameters of 0.125 and 0.2 mm thatwere rounded at the tips with bead diameters of 0.25 and0.35 mm, respectively. However, in our current research, eachdroplet was suspended on a 0.1-mm SiC fiber with no bead at itstip, and droplets having similar initial diameters ranging from0.954 to 1.0642 mm were selected from the experimental data tominimize the effect of the initial droplet size on the evaporationrate.

3.2. Evaporation of kerosene droplets with the addition of oleic acid

To understand the evaporation behavior of kerosene-basednanofluid fuel droplets at elevated temperatures, it is essential tomeasure their evaporation rates with the addition of variousconcentrations of OA while holding the temperature constant.The evaporation rates of the stabilized kerosene droplets provideanother baseline to distinguish the effect of the addition of NPson droplet evaporation behavior.

Figure 4 shows the normalized temporal histories of the diam-eter squared of kerosene droplets with 0.25% and 0.5% surfactant(OA) at various ambient temperatures (400–800 �C). Here and be-low, mass percent values are used. To minimize the effect of theinitial droplet size on the evaporation rate, droplets having similarinitial diameters ranging from 0.9781 to 1.0796 mm were selectedfrom the experimental data. The general evaporation behavior ofstabilized kerosene droplets was similar to that of pure kerosenedroplets, i.e., after a finite heating-up period, the droplet diametersquared decreased linearly and followed the d2-law. Fig. 4a showsthe lifetime history of stabilized kerosene droplets evaporated at400 and 500 �C. With the addition of OA to the kerosene, the drop-let heating-up period, as well as the linear evaporation period, in-creases at 400 �C. Kerosene has a boiling range of 180–270 �C,whereas OA has a boiling point of 360 �C. Due to the addition ofOA to the kerosene, a monolayer of OA is produced at the surfaceof the droplet, which inhibits the outward diffusion of the kero-sene, resulting in an increase in the evaporation period and areduction in the evaporation rate [29]. This suppression is promi-nent at temperatures close to the boiling point of OA. Fig. 4b showsthe lifetime history of stabilized kerosene droplets evaporated at600, 700, and 800 �C. At 800 �C with 0.5% OA, there were suddenincreases and decreases in droplet diameter (d2/d2

0) during theheating-up period, indicating droplet distortion due to the entrap-ment of high-volatility components. These distortions in droplets’diameters increase the heating-up period and consequently en-hance the evaporation rate slightly at 800 �C.

Figure 5 compares the evaporation rates of stabilized kerosenedroplets containing 0.25% and 0.5% OA with pure kerosene dropletsat various ambient temperatures (400–800 �C). The evaporationrate of stabilized kerosene droplets was equal to the evaporationrate of pure kerosene droplets at all temperatures. Because onlylow concentrations of OA (surfactant) were in the kerosene, theevaporation rate of stabilized kerosene droplets was unaffectedand remained the same as that of pure kerosene droplets at thesetemperatures.

3.3. Evaporation of kerosene-based nanofluid fuel droplets

Figure 6 shows normalized temporal histories of kerosene drop-lets’ diameters squared with the addition of 0.1%, 0.5%, and 1.0% AlNPs whose surfaces were coated with 0.05%, 0.25%, and 0.50% ofOA, respectively, at ambient temperatures ranging from 400 to800 �C. The droplets were selected from the experimental datawithin a narrow range of initial diameters from 0.987 to1.106 mm to reduce the effect of the initial droplet diameter onthe evaporation properties.

At relatively low temperatures (400–600 �C), the evaporationhistory of kerosene-based nanofluid droplets showed a similartrend to that of kerosene, i.e., it started with a finite heating-upperiod, followed by d2-law evaporation, as shown in Fig. 6a. Theonly difference between pure kerosene and kerosene-basednanofluid droplets was that the droplets having Al NPs requiredless time for evaporation. The early dry-out clearly indicates thehigher evaporation rates of droplets containing Al NPs, regardless

P= 0.1 MPa

500°C 400°C

t /d20 (s/mm2)

d2/d2 0

0 1 2 3 4 5 6 7 8

-1 0 1 2 3 4 5 6 7

0

0.2

0.4

0.6

0.8

1

1.2

KeroseneKerosene+0.25% OAKerosene+0.50% OA

P= 0.1 MPa

800°C 700°C 600°C

t /d20 (s/mm2)

d2/d2 0

0 0.5 1 1.5 2 2.5 3 3.5 4

-0.5 0 0.5 1 1.5 2 2.5 3 3.5

-1 -0.5 0 0.5 1 1.5 2 2.5

0

0.2

0.4

0.6

0.8

1

1.2KeroseneKerosene+0.25% OAKerosene+0.50% OA

(a)

(b)Fig. 4. Diameter squared histories of evaporating stabilized kerosene dropletscontaining various concentrations of OA at different ambient temperatures; (a) at400 and 500 �C, (b) at 600, 700 and 800 �C.



of concentration. At 500 and 600 �C, the evaporation behavior ofthe 1.0% Al NP suspension droplets was slightly different. Com-pared to the 0.5% suspension, the droplets containing 1.0% Al NPsevaporated faster initially and then slightly slower during the laterstages of evaporation; due to this slight reduction, the evaporationperiod was longer than that of the 0.5% suspension droplets. Thiswas likely caused by the formation of NP agglomerates in the1.0% Al NP suspension droplets, resulting in less evaporation rateenhancement compared to the 0.5% Al NP suspension droplets atthese temperatures.

Figure 6b shows that at relatively high temperatures(700–800 �C), the general evaporation behavior of nanofluid fueldroplets was different from that of pure kerosene droplets, irre-spective of the Al NP concentration. The heating-up period of theAl NP suspension droplets was shorter, and evaporation startedearlier, compared to pure kerosene droplets. This was followedby a high-evaporation-rate period, which was reduced slightly inthe later stages when the Al NPs were concentrated. Overall, thelifetime of the Al NP suspension droplets was shorter than thatof pure kerosene droplets, which indicates a significant increasein the evaporation rate when Al NPs were added to kerosene. Theother distinguishing feature of the addition of Al NPs to kerosenedroplets was the onset of bubble formation and micro-explosions.Regardless of the Al NP concentration, bubble formation andmicro-explosions were observed at these temperatures. Micro-explosions were identified in the later stages of the evaporationperiod at 700 �C (see Supplementary video) and were also observedin the earlier stages at 800 �C. It is important to note that for micro-explosions to be most effective, they must occur early in the life-time of a droplet [30]. The least-squares method was applied tothe linear part of the experimental data to obtain the evaporationrate of the kerosene-based nanofluid droplets.

As the nanofluid fuel droplets were exposed to relatively hightemperatures (700–800 �C), the Al NPs present at the droplets’ sur-face were heated above the boiling point of the liquid fuel. These AlNPs provided multiple nucleation sites for the surrounding liquid,and evaporation started early in the heating-up period. The Al NPspresent just inside the droplets’ surface may also have attained atemperature higher than the local boiling point of the high-volatil-ity component, creating multiple heterogeneous nucleation sitesthat generated superheated vapors. These superheated vapors

may have been the cause of droplets’ fragmentation in the earlystages of nanofluid droplets’ evaporation. In contrast, the micro-explosions observed during the later stages of evaporation weredue to the agglomeration of NPs at the surface of the droplets. Asthe liquid evaporated, the NPs were concentrated at the droplets’surface and may have caused an increase in the droplets’ surfacetemperature. Thus, the low-volatility-component liquid present in-side the droplets was heated beyond the local boiling point, accu-mulating as a superheated vapor that ruptured the droplets in thelater stages of evaporation.

Figure 7 presents sequential photographs of evaporating kero-sene droplets containing 0.5% and 1.0% Al NPs at 800 �C. Dropletdistortion and fragmentation were clearly observed. For thenanofluid fuel droplet containing 0.5% Al NPs (Fig. 7a), droplet frag-mentation occurred at 0.500 s, followed by droplet distortion, asshown in subsequent frames at 0.505, 0.510, and 0.515 s. Theliquid evaporated up to 1.220 s, until an internal bubble started

P=0.1MPa

Temperature (oC)

Cv(mm2 /s)

400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

1.2

KeroseneKerosene+0.25% OAKerosene+0.50% OA

Fig. 5. A comparison of the evaporation rates of stabilized kerosene dropletscontaining various concentrations of OA with pure kerosene droplets at differentambient temperatures and constant pressure of 0.1 MPa.

P= 0.1 MPa

500°C 400°C600°C

t /d20 (s/mm2)

d2/d2 0

0 1 2 3 4 5 6 7 8 9

-1 0 1 2 3 4 5 6 7 8

-2 -1 0 1 2 3 4 5 6 7

0

0.2

0.4

0.6

0.8

1

1.2KeroseneKerosene+0.05% OA+0.1% Al NPsKerosene+0.25% OA+0.5% Al NPsKerosene+0.50% OA+1.0% Al NPs

800°C 700°C

t /d20 (s/mm2)

d2/d2 0

0 0.5 1 1.5 2 2.5

-0.5 0 0.5 1 1.5 2

0

0.2

0.4

0.6

0.8

1


P= 0.1 MPa

(a)

(b)Fig. 6. Temporal variation of the diameter squared of kerosene droplets with diluteconcentrations (0.1%, 0.5% and 1.0%) of Al NPs (coated with OA) at differentenvironmental temperatures (a) at 400, 500 and 600 �C, (b) at 700 and 800 �C.



to form at 1.225 s, followed by bubble rupture at 1.230 s. Fig. 7bshows that for the 1.0% Al NP suspension, droplet fragmentationfirst occurred at 0.670 s, along with droplet distortion before andafter fragmentation, as shown in the images for 0.665, 0.675,0.680, 0.685, 0.690, 0.695, and 0.700 s. The droplet distortion indi-cated the presence of high-volatility-component vapors inside thedroplet. Then, the liquid evaporated up to 1.165 s until the start ofbubble formation. Eventually, bubble formation was completed at1.255 s, and it ruptured at 1.260 s.

Droplet fragmentation was delayed with increasing Al NP con-centration in the early stages of evaporation, whereas in the later

stages the rupture of droplets occurred earlier and with greaterintensity. This was because the increase in Al NP concentrationcaused the temperature inside droplets to rise slowly at the startof evaporation, but in the later stages of evaporation the more con-centrated Al NPs caused droplet rupture to occur earlier and withgreater intensity.

3.3.1. Effect of temperature on the evaporation rate of kerosene-basednanofluid fuel droplets

Figure 8 compares the evaporation rates of kerosene-basednanofluid fuel droplets containing dilute concentrations (0.1%,

t = 0.000 s t = 0.495 s t = 0.5000 s t = 0.505 s t = 0.510 s

t = 0.515 s t = 1.225 s t = 1.230 s t = 1.235 s t = 1.515 s

(a)

t = 0.000 s t = 0.630 s t = 0.665 s t = 0.670 s t = 0.675 s

t = 0.680 s t = 0.685 s t = 0.690 s t = 0.695 s t = 0.700 s

t = 1.165 s t = 1.255 s t = 1.260 s t = 1.265 s t = 1.315 s

(b)Fig. 7. Sequential photographs of vaporization of kerosene droplets with various concentrations of Al NPs at elevated temperature (a) with 0.5% Al NPs at 800 �C, (b) with 1.0%Al NPs at 800 �C; Camera speed: 200 fps.



0.5%, and 1.0%) of Al NPs with those of pure and stabilized kerosenedroplets under different ambient temperatures. The trend in theevaporation rate curves of the kerosene-based nanofluid fuel drop-lets is similar to those for pure kerosene, i.e., the evaporation rate in-creased monotonically as temperature increased from 400 to 800 �C.However, the increasing trend in the evaporation rate of pure kero-sene was represented by a power-law fit, whereas the evaporationrate of kerosene-based nanofluid fuels increased exponentially withtemperature. This figure also clearly shows that the addition of di-lute concentrations of Al NPs to kerosene enhanced the evaporationrate at all temperatures in the range 400–800 �C. At relatively lowtemperatures (400–600 �C), the evaporation rate of the nanofluidfuel droplets was slightly higher than that of pure kerosene droplets,whereas at relatively high temperatures (700–800 �C), the evapora-tion rate of the nanofluid fuel droplets was considerably higher thanthat of the stabilized or pure kerosene droplets.

It is possible that at relatively low temperatures (400–600 �C),the Al NPs enhanced only the heat transfer inside the droplets byincreasing the liquid effective thermal diffusivity and at the drop-lets’ surface by raising their surface temperature. At relatively hightemperatures (700–800 �C), the Al NPs present at or near the sur-face may have attained temperatures higher than the local liquidboiling point and may have provided multiple heterogeneousnucleation sites for the surrounding liquid, enhancing the evapora-tion rate substantially. It is also possible that the Al NP surfaceswere modified, producing oxide-free, ligand-protected NPs [12].This would have allowed the melting of oxide-free Al NPs presentat the surface of the droplets when exposed to temperatures higherthan the Al NP melting point (660 �C). The fractional melting of AlNPs may have extensively enhanced the evaporation rate of thenanofluid fuel droplets by providing their heat of fusion to the sur-rounding liquid.

The residues obtained at the end of evaporation were collectedto confirm the physical state of the Al NPs present at the surface ofthe droplets and the shell formation of the nanofluid fuel droplets.Residues of the 0.5% Al NP suspension droplets were obtained attwo different temperatures (600 and 800 �C) and studied usingFESEM and EDX. The selected temperature values were belowand above the melting temperature of the Al NPs (660 �C). The res-idues were collected very carefully. Initially, a nanofluid fuel drop-let was exposed to these temperatures, and as the droplet startedto evaporate, the furnace was lifted upward immediately to avoidexposing the residue to the high-temperature ambience. After that,nitrogen was purged to cool the residue to room temperature sothat it could be removed.

Figures 9 and 10 show FESEM images of the residues obtainedafter evaporation at 600 and 800 �C for a suspension of 0.5% AlNPs. The residue obtained at 600 �C was soft and crumbly andbroke during preparation of the SEM samples, as shown inFig. 9a. A magnified view of the residue (Fig. 9b) indicates thatmost of the Al NPs maintained their original spherical shapes anddid not melt. The NPs were also loosely packed, and many poreswere present. When a 0.5% Al NP suspension droplet was exposedto 800 �C, the residue was sticky, and most of it remained on theSiC fiber, as shown in Fig. 10a. The residue also contained largeparticles at its surface. A magnified view of this residue (Fig. 10b)indicates that some of the NPs lost their spherical shape, fusedtogether, and were converted into larger particles and lumps. Thisindicates that the NPs present at the surface of the droplet melted.

The composition of the residues obtained after evaporation at600 and 800 �C was analyzed by EDX. The results indicate thatthe evaporative residues were composed mostly of Al and O, withsmall amounts of C. The O content was due to the oxidation of theAl NPs present at the surface of the residues after vaporization andduring preparation of the samples for FESEM. The O contents were5.3% and 14.6% (by weight) in the residues of the 0.5% NP

P=0.1MPa

Temperature (oC)

Cv(mm2 /s)

400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4KeroseneKerosene+0.05% OA+0.1% Al NPs

P=0.1MPa

Temperature (oC)

Cv(mm2 /s)

400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4KeroseneKerosene+0.25% OAKerosene+0.25% OA+0.5% Al NPs

P=0.1MPa

Temperature (oC)

Cv(mm2 /s)

400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4KeroseneKerosene+0.50% OAKerosene+0.50% OA+1.0% Al NPs

(a)

(b)

(c)Fig. 8. A comparison of the evaporation rates of kerosene-based nanofluid fueldroplets containing dilute concentrations of Al NPs with pure and stabilizedkerosene droplets at different ambient temperatures; (a) 0.1% Al NPs, (b) 0.5% AlNPs, (c) 1.0% Al NPs.



suspension evaporated at 600 and 800 �C, respectively, as shown inFig. 11. This indicates high oxidation of Al NPs to Al2O3. Both of theresidues were cooled down to room temperature in the presence ofnitrogen before being taken out into air; therefore, the high oxygencontent was due to the presence of elemental Al at their surfaces,which was oxidized. The EDX results imply that the residueobtained at 800 �C contained high amounts of elemental Al at itssurface, which was thought to be the melted Al NPs.

These results indicate that the evaporation enhancement wasdue only to the addition of Al NPs to the kerosene. The surface tem-perature of a pure kerosene droplet cannot exceed the boiling pointof the liquid (270 �C), but the presence of Al NPs allows the tem-perature of the NPs present at the droplets’ surface to reach thetemperature of the environment, resulting in local hot spots inthe liquid droplets. After heating beyond the liquid local boilingtemperature, the Al NPs can provide multiple nucleation sites,evaporate the surrounding liquid, and result in increased evapora-tion rates. The observed micro-explosions were also due to thepresence of these heterogeneous nucleation sites in the droplets.Ambient temperatures of 700 and 800 �C greater than the meltingpoint of the Al NPs resulted in fractional melting of the NPs accu-mulated at the droplets’ surface, which supplied additional heatto the liquid droplets by providing their heat of fusion, resultingin extensive enhancement of the evaporation rate. Further studiesfocusing on nanofluid fuel evaporation at high temperatures withdense concentrations of Al NPs will provide valuable insight intothe occurrence of micro-explosions due to the NPs and their effectson the evaporation rate.

3.3.2. Effect of Al NP concentration on the evaporation rateFigure 12 indicates that the evaporation rate of the kerosene-

based nanofluid droplets was also affected by the concentrationof the NPs. As the NP concentration increased from 0.1% to 1.0%,the evaporation rates at relatively low ambient temperatures

Fig. 9. SEM micrographs of the evaporation residue obtained at 600 �C for 0.5% AlNPs suspension containing 0.25% OA (a) overall view and (b) magnified view,magnification = 50,000.

Fig. 10. SEM micrographs of the evaporation residue obtained at 800 �C for 0.5% AlNPs suspension containing 0.25% OA (a) overall view and (b) magnified view,magnification = 50,000.

Al

O

C

Energy (eV)100 200

T=600oCT=800oC

Fig. 11. EDX spectrum of the evaporation residues (agglomerates) obtained at 600and 800 �C for 0.5% Al NPs with 0.25% OA.



(400–600 �C) increased initially and then decreased. The maximumincreasing effect was obtained with 0.5% Al NPs at low ambienttemperatures. The evaporation enhancement for the 1.0% NP sus-pension at these temperatures was lower than that observed withthe 0.5% NP suspension due to the high agglomeration of NPs.

At relatively high ambient temperatures (700 and 800 �C), theevaporation rate also increased with the NP concentration up to0.5% and then decreased, becoming lower than the evaporationrates observed with the 0.1% NP suspension but higher than theevaporation rate of pure kerosene. The maximum increasing effectwas also obtained with the 0.5% Al NP suspension at high ambienttemperatures. When the concentration of ligand-protected NPs in-creased beyond a critical value (0.5% in this case), more heat wastransferred to the Al NPs rather than to liquid, thereby reducingthe evaporation enhancement. The maximum increase in the evap-oration rate of kerosene-based nanofluid droplets with 0.5% Al NPswas 56.7% compared to the pure kerosene droplets’ evaporationrate at 800 �C (corresponding to an increment from 0.782 to1.226 mm2/s).

4. Conclusions

The evaporation characteristics of kerosene-based nanofluiddroplets with dilute concentrations of Al NPs (0.1%, 0.5%, and 1.0%)were studied at different elevated temperatures (400–800 �C) undernormal gravity. Similar to heptane-based nanofluid fuel droplet [19],the kerosene-based nanofluid fuel droplets containing dilute con-centrations of Al NPs were vaporized in the same way as the purefuel droplets and followed d2-law at temperatures from 400 to600 �C. The new findings of this experimental study are summarizedas follows.

(1) Micro-explosions were observed at relatively high tempera-tures (700–800 �C) in the kerosene-based nanofluid fueldroplets. The absence of micro-explosions in pure and stabi-lized kerosene droplets indicates that this phenomenon wasbased solely on the presence of Al NPs. The added Al NPsprovided multiple heterogeneous nucleation sites to pro-mote micro-explosions.

(2) The ambient temperature significantly affected the evapora-tion rates of the kerosene-based nanofluid fuel droplets. TheAl NPs enhanced the evaporation rate of kerosene droplets atall tested temperatures (400–800 �C). At relatively low

temperatures (400–600 �C), the Al NPs in the dropletsincreased the evaporation rate slightly. However, at rela-tively high temperatures (700–800 �C), the evaporation ratewas enhanced substantially due to melting of the NPs pres-ent at the droplets’ surface.

(3) The evaporation rate of nanofluid fuel droplets was alsoaffected by the concentration of the NPs. The evaporation-rate enhancement increased with the NP concentration upto 0.5%, after which it decreased and became lower than thatof the 0.1% NP suspension. The maximum increasing effectwas obtained at 800 �C with the 0.5% NP suspension, wherethe evaporation rate was increased by 56.7%.

Acknowledgment

This work was supported by the Midcareer Researcher Programthrough the NRF grant funded by MEST (2010-0000353).

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.combustflame.2013.07.007.

References

[1] K.K. Kuo, G.A. Risha, B.J. Evans, E. Boyer, Materials Research Society Symposium– Proceedings 800 (2003) 3–14.

[2] G. Young, Metallic Nanoparticles as Fuel Additives in Airbreathing Combustion(2007).

[3] G. Young, K. Sullivan, M.R. Zachariah, K. Yu, Combustion and Flame 156 (2009)322–333.

[4] J. Sabourin, D. Dabbs, R. Yetter, F. Dryer, I. Aksay, ACS Nano 3 (2009) 3945–3954.

[5] R.H. Chen, T.X. Phuoc, D. Martello, International Journal of Heat and MassTransfer 54 (2011) 2459–2466.

[6] S. Acharya, S. Karmakar, K.M. Dooley, Journal of Propulsion and Power 28(2012) 707–718.

[7] H. Tyagi, P.E. Phelan, R. Prasher, R. Peck, T. Lee, J.R. Pacheco, P. Arentzen, NanoLetters 8 (2008) 1410–1416.

[8] D.E. Jackson, D.F. Davidson, R.K. Hanson, in: 44th AIAA/ASME/SAE/ASEE JointPropulsion Conference and Exhibit, Hartford, CT, United States, 2008.

[9] C. Allen, G. Mittal, C.J. Sung, E. Toulson, T. Lee, Proceedings of the CombustionInstitute 33 (2011) 3367–3374.

[10] B. Van Devener, S.L. Anderson, Energy & Fuels 20 (2006) 1886–1894.[11] B. Van Devener, J.P.L. Perez, S.L. Anderson, Journal of Materials Research 24

(2009) 3462–3464.[12] B. Van Devener, J.P.L. Perez, J. Jankovich, S.L. Anderson, Energy & Fuels 23

(2009) 6111–6120.[13] Y. Gan, L. Qiao, Combustion and Flame 158 (2011) 354–368.[14] Y. Gan, Y.S. Lim, L. Qiao, Combustion and Flame 159 (2012) 1732–1740.[15] R.H. Chen, T.X. Phuoc, D. Martello, International Journal of Heat and Mass

Transfer 53 (2010) 3677–3682.[16] Y. Gan, L. Qiao, International Journal of Heat and Mass Transfer 54 (2011)

4913–4922.[17] Y. Gan, L. Qiao, Energy & Fuels 26 (2012) 4224–4230.[18] Y. Gan, L. Qiao, International Journal of Heat and Mass Transfer 55 (2012)

5777–5782.[19] I. Javed, S.W. Baek, K. Waheed, Combustion and Flame 160 (2013) 170–183.[20] T. Edwards, Journal of Propulsion and Power 19 (2003) 1089–1107.[21] W. Yu, H. Xie, Journal of Nanomaterials (2012).[22] H. Ghassemi, S.W. Baek, Q.S. Khan, Combustion Science and Technology 178

(2006) 1031–1053.[23] H. Ghassemi, S.W. Baek, Q.S. Khan, Combustion Science and Technology 178

(2006) 1669–1684.[24] Q.S. Khan, S.W. Baek, H. Ghassemi, Combustion Science and Technology 179

(2007) 2437–2451.[25] H. Ghassemi, S.W. Back, Q.S. Khan, 43rd Aerospace Sciences Meeting and

Exhibit (2005).[26] C.K. Law, Combustion Physics, Cambridge University Press, 2006.[27] S.R. Turns, An Introduction to Combustion: Concepts and Applications,

McGraw-Hill Science, 2006.[28] D.B. Spalding, Some Fundamentals of Combustion, Butterworth, London, 1955.[29] S.C. Wong, A.C. Lin, H.Y. Chi, Symposium (International) on Combustion 23

(1991) 1391–1397.[30] C.K. Law, Progress in Energy and Combustion Science 8 (1982) 171–201.

P=0.1MPa

Temperature (oC)

Cv(mm2 /s)

400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

1.2


Fig. 12. Comparison between the evaporation rates of kerosene-based nanofluidfuel droplets with 0.1%, 0.5% and 1.0% Al NPs under different ambient temperatures.


author's personal copy - kaistprocom.kaist.ac.kr/download/ijp/131.pdfexothermic...

Documents